The proportion of polypeptide chains which generate native foldsPart 1: analysis of reduced codon set

experiments 3

The proportion of polypeptide chains which generate native foldspart 2: theoretical studies 8

The proportion of polypeptide chains which generate native foldspart 3: designed secondary structures .13

The proportion of polypeptide chains which generate native foldspart 4: reusing existing secondary sequences
...16

The proportion of polypeptide chains which generate native foldspart 5: experimental extraction from random
sequences .17

The proportion of polypeptide chains which generate native foldspart 6: extraction from random sequences .24

Vitamin B12 ..28

Antifreeze protein evolution: turning wrenches into hammers 29

Biological robots will blow your mind!

The non-evolution of apoptosis .32

New DNA repair enzyme discovered ...40

ATP synthase: majestic molecular machine made by a mastermind ..40

The immunoglobulin heavy chain gene family and gene duplication .42

Lipid rafts: evidence of biosyntax and biopragmatics .43

Mitochondriacreated to energize us .47

Serial cell differentiation: intricate system of design ..48

Searching for needles in a haystack .49

Bone building: perfect protein 55

31

VIRUSES

The bacterium that came in from the cold ..57

Does biological advantage imply biological origin? 57

Germ with seven motors in one! 59

The fungus that walks ..59

What about parasites? 61

Swine fluIs it evidence of evolution? .63

Bacteria trapped for millions of years under Antarctic ice ...64

Germs miniature motor has a clutch 65

Even a tiny virus has a powerful mini-motor ... 55

New bacteria show wonder upon wonder ..67

More Sleeping Beauty bacteria 68

The proportion of polypeptide chains which generate native foldsPart 1: analysis of reduced codon set
experiments
by Royal Truman
Creationist scientists and Intelligent Design proponents have drawn attention to the sparseness of native-like folded
proteins among random polypeptide sequences. Contrary to this opinion, it was alleged that protein folds are very
common among random amino acid chains. We found this to be a surprising opinion, and this statement and the
references cited prompted this six-part series. We review the best-known lines of evidence used to claim that random
polypeptides often lead to native-like folds. We conclude that although a good estimate of the proportion of true native-like
folds remains unknown, it is astronomically small. In Part 1 we address protein-folding experiments that are based on
proteins constructed of very few but specified types of amino acids.
Many scientific reasons have been identified for doubting life could have arisen by natural processes. 14 Examples
include: only one enantiomer of amino acids must be used in most proteins; the difficulty of forming long protein chains in
water; the statistical improbability of creating large RNA and DNA chains; the presence of the genetic code; and the need
for many protein sequences which fold into a single stable conformation.
Figure 1. All but one of the
biological l-amino acids either
melt or decompose (d) above
220C. Table 1 clarifies that
most actually decompose and
do not melt at all. Asn has the
lowest melting point, 234C.
All these, and other, hurdles
would need to be met with no
intelligent guidance. If one
prerequisite
seems
insurmountable,
then
the
naturalist
paradigm
is
implausible. In this six-part
series we will evaluate one
claim: that many random
polypeptide sequences fold
into
a
single
stable
conformation. The basis for these claims will be examined in depth in order to introduce some objectivity into the
discussion. Of course, merely folding reliability is a necessary but not sufficient condition for important chemical
processes. A particular fold only offers a stable scaffold upon which a useful geometric and electronic environment may or
may not be available.All known life-forms require a large number of different kinds of folded proteins together at the same
time and location. If very few random sequences fold reliably, then obtaining such an ensemble naturally is very unlikely.
This would indicate that abiogenesis beginning with proteins in the absence of a genetic code would have no scientific
support.Biological proteins are usually classified as fibrous, membrane and globular.5 Fibrous proteins are produced in the
construction of hair, nails, tendons and ligaments. They are genetically expressed by mostly higher life-forms for structure
but not for the fundamental biochemical processes necessary for life.Membrane proteins probably make up the majority of
all proteins found in the cell. 6 They regulate, among many other functions, signal transduction, transport across the
membrane and secretion. The structure of membrane proteins, however, is completely different when embedded in a
membrane as to when in aqueous solution. This makes it very difficult to study and characterize membrane proteins in
their relevant state. An authority pointed out recently that Many proteins are retained within cell membranes and we know
virtually nothing about the structures of these proteins and only slightly more about their functional roles. 7Globular
proteins have been the best studied, being easier to isolate in vitro, separated from other cellular bio-chemical
interactions. Thousands of globular proteins are critically important, and are used as enzymes and for other purposes 8. In
fact, Within all cells every reaction is regulated by the activity of enzymes. 9 To function at all, and reliably, globular
proteins must fold into precise three dimensional structures. A polypeptide which produces a single, lowest energy folded
structure will not thereby automatically provide any biological value, but such a scaffold is one prerequisite for useful
function.
Table 1. Melting or decomposition (d) temperature for biological l-amino acids.

l-Amino acid
C
Amino acid polymerization is strongly disfavoured in water,10 rendering
unreasonable the notion that life originated naturally in an oceanic hot dilute
Ala
297
soup11 or warm pond12. Furthermore, the extreme high melting point of biological l13
amino acids (see table 1 and figure 1) makes it essentially impossible to form
Arg
244
linear chains under dry conditions. In fact, most natural amino acids decompose at
high temperatures and dont melt at all. Temperature in the 200350C range boils
other organic substances and is inimical to life. In addition to a stable fold, globular
Asn
234235
proteins must remain soluble in water and not form tarry-like amorphous clumps.In
this series, we will be discussing the folding of globular proteins. Such folding must
Asp
270271
occur quickly and reliably into precise three-dimensional shapes to assure effective
biochemical physiology. Therefore, if proteins arose naturalistically from random
polypeptide sequences, the proportion of random chains which produce native-like
Cys
260261
folds is expected to be reasonably high. Contra this requirement, finding properly
folded proteins among random sequences is rare.Our evolution-touting
Gln
185186
counterparts are aware of the significance of this sparsity:
This general argument has become of some importance as support for the view
that proteins could not have arisen from natural pre-biotic chemical processes on
Glu
247249
earth and as support for creationism.14
A key piece of evidence offered15 by Denton and co-authors for the view that
Gly
233
properly folded proteins are common among random polypeptides was work
carried out in Professor Sauers lab at MIT:
In libraries of random amino acid sequences, alpha helical proteins displaying
His
287
cooperative thermal denaturation of specific oligomeric states have been
recovered at frequencies of 1%.16
Ile
284
Sauers actual statement deviates in key details from the claim above:
In fact, in libraries composed of random combinations of Leu, Gln, and Arg,
Leu
293295
proteins resistant to intracellular proteolysis were found at frequencies of about
1%. Purification and biochemical studies of several of these proteins revealed
them to be -helical, oligomeric, and to display reversible thermal denaturation.
Lys
224.5
However, even the most native-like of these random proteins differed from natural
proteins in requiring some denaturant for solubility and in having extremely rapid
Met
280282
rates of amide exchange.17
Therefore the problem of random protein folding is not as simple as Denton et
al. have suggested.
Phe
283
Testing a large number of random polypeptide sequences to determine how many
produce native-like folds would be very difficult. One strategy is to build chains
Pro
220222
using only a few of the twenty natural amino acids, and to use smaller chains
which are less likely to offer as many large random regions which interfere with
folding. This reduces the variety of sequences which could be produced.
Ser
228
Knowledge of what affects folding permits the research to be designed intelligently.
For example, one of the strongest driving forces which cause folding is the ability
Thr
255257
to bury hydrophobic side chains inside the protein core, away from the aqueous
environment.A binary pattern which uses a combination of hydrophobic (H)
hydrophilic or polar (P) residues can facilitate the rough initial shaping of the
Trp
289
protein. Analysis of existing -helices in biological proteins shows a high frequency
of residue patterns such as PHPPH and HPPHHH.18 For at least -sheets located
Tyr
342344
on protein surfaces (i.e. not buried into the protein core) a binary pattern such as
18
HPHPH and PHPHP is often found. These ideas provide the insight to build
Val
315
polypeptides based on a reduced set of amino acids which satisfy the binary
pattern as well as possible.
The experiment
The details of key experiments performed at MIT were published in 1994. 19 What was done experimentally and
resulted?

d
d

d
d
d
d
d
d
d
d
d
d
d
d

what

Figure 2. Creation of proteins based on amino acids

QLR (Q=glutamine, L=leucine, and R=arginine). The
plasmid backbone and oligonucleotide cassettes used
to create the library are shown. In each cassette, 19
21 codons use the nucleotides C (cystein) at the first
position; a mixture of nucleotides A,T and G at the
second position; and an equal mixture of A and G at
the last position.In Davidson and Sauers original
work, a library of synthetic genes was prepared by
linking together three or four oligonucleotide cassettes
which consisted of 19 to 21 codons each. Each
cassette was synthesized with a 10-bp (base-pair)
self-annealing sequence at the 3 end which allows
synthesis of the second strand using special
enzymes. This permits all the cassettes to be joined
together as a single synthetic gene in a random manner giving
rise to an exceptional repertoire of polypeptide sequences of
around 80 residues composed predominantly of glutamine,
leucine and arginine or QLR proteins (figure 2).20 All the codons
in the cassettes were based on the following formula: C
(Cytosine) always in the first position; A (Adenine), T (Thymine)
or G (Guanine) in the proportions 50%, 40% and 10%,
respectively, in the second position; and A/G 50:50 in the third

position. According to the rules of the genetic code, this will produce the amino acids glutamine (Q), leucine (L) and
arginine (R) (figure 3). For experimental convenience, a tryptophan codon was added to permit fluorescent studies.
Figure 3. Dipeptide formed by reacting two amino acids. The three side groups used in the experiments, R 1 and R2, are
shown, labelled Q, L and R, according to standard naming conventions for amino acids.
The joined cassettes were ligated to a backbone fragment which contained a promoter to ensure the gene would be
expressed. At the carboxy-terminal tail the epitope tag DYKDDDDK was added; six codons for histidine to allow
separation of the resulting protein by affinity purification; plasmid pBR322 origin of replication; an ampicillin-resistant gene
and the lacIq gene. The backbone for this ensemble, the vector to be introduced into E. coli bacteria, was constructed from
a plasmid (pDW239). The E. coli survivors from exposure to ampicillin permitted identification of those colonies which
possessed the plasmid with the artificial gene, from which the expressed artificial proteins could be extracted.
Note that the authors do not claim in this paper 19 that 1% of these artificial proteins fold properly: the quote above by
Sauer alone was published two years later.17
Evaluation
We shall see that these experiments do not provide an estimate of the proportion of random sequences which would lead
to a reliable native-type fold. What they do permit, however, is to prove that the proportion must be considerably lowerthan
1%! We shall now consider a series of corrective proportions, pn, that must now be applied to the 1% figure.
Correction p1: the proportion of the three QLR AAs (amino acids) is not random
Professor Sauer is an expert in protein chemistry and protein folding 21 and knows which combination of residues, and in
what proportion, would be best able to produce protein secondary structures. The propensity for a given amino acid
residue to be found in -helix22, -strand23 or turn24 had been thoroughly documented25 long before Sauer and his coworkers performed this work. In 1996 he summarized 26the design elements appropriate to construct a small protein based
on alpha helices (which knowledge he and his co-workers applied in the work described above19):
The use of residues with reasonable secondary structure propensities. This is known from statistical studies using
databases of secondary structures.
The choice of an appropriate binary pattern to ensure that polar residues face out to contact the solvent and non-polar
residues face in to form a hydrophobic core.
The use of turn and capping sequences to properly form the bend and terminate the helices.
The use of hydrophobic side chains that allow complementary packing of the protein core.
Perhaps the introduction of buried polar interactions.
Figure 4. Long, rigid QLR proteins with
excessive hydrophobic nature and long
alpha helices might be clumping in various
manners with the other QLR members. The
location for intermolecular helix interactions
may vary for the different conglomerates.
To optimize the chances of producing a
protein with alpha helices, the particular
codons described above in the cassettes
were carefully designed to generate on
average: 50% of the polar glutamine (Q);
40% of the hydrophobic leucine (L); and
10% of the charged residue arginine (R). 20 The resulting polypeptides are referred to in the papers as QLR proteins.
Notice how the carefully designed proteins generate an appropriate proportion between hydrophobic and hydrophilic
residues; the use of 50% hydrophobic side chains neither too large nor too small; and a small amount of polar residues
which could be buried into the protein core.The sequences generated could incorporate Q, L or R randomly, affected only
by the relative proportion of the codons used. This will allow many sequences to be generated which satisfy reasonably
well binary patterning rules for helices. The experimental design also allows sequences to include the opposite, which is
useful, since Marshall and Mayo note that,
Most naturally occurring proteins contain some buried polar and exposed hydrophobic amino acid residues. In some
cases, these residues are necessary for protein stability; for instance, many turns contain buried polar residues which
form hydrogen bonds to main chain amides.27
Of the library generated, three proteins were expressed sufficient for purification and studies. 19 The proportion of Q/(Q+L)
was found to vary between 0.456 and 0.534, as designed by the cassettes (Ibid.). The amount of amino acid R varied
from 7% to 13%. These ranges represent a small fraction of a percent of all sequences which could be generated using
these three amino acids, and most of the excluded regions would not satisfy the intended ideal hydrophobic/hydrophilic
proportions.Therefore, the proportion of alpha helices and other features reported, which they believe have some
semblance to proteins,19 (resistance to proteolysis and cooperative thermal denaturation) among all random QLR chains,
can only be orders of magnitude smaller than the estimate reported. Incidentally, resistance to proteolysis need not imply
native-like folded tertiary structure,28,29 neither must cooperative thermal denaturation.30
Correction p2: the AAs chosen easily form alpha helices
Davidson and Sauer admit that The alpha-helical structure of the QLR proteins is not unexpected, given the high helical
propensities of glutamine, leucine, and arginine. 31 The three proteins the authors isolated revealed extremely high
fractional helicity values: 32%, 60% and 70%, which is not representative of random polypeptides. Hence, the
extrapolation to random sequences using all natural amino acids will require a considerable correction factor.
Correction p3: the proteins were not soluble
A high concentration of chaotropic agents32 was needed to force the proteins which were isolated to remain in solution in
water. Now, to be biologically useful, a protein should not form insoluble clumps as evidenced in this experiment.We are
not arguing that solubility is absolutely necessary to define true folding. However, together with other characteristics
mentioned below, what was reported does not resemble native-like folds, and implies that these proteins are, to a high
degree, amorphous conglomerates. Of all QLR proteins which could be formed, only the subset which displays the kinds
of solubility expected for real, native-like folds (and none were found!) should have been factored into the calculations.
Sticky hydrophobic patches must be avoided and p3 must have a value less than one.
Correction p4: the proteins were much too rigid
The authors point out that None of the proteins showed significant loss of alpha-helical content up to 90C, the highest
temperature tested.28 This is significant, as the authors point out: We know of no natural proteins that retain their
secondary structure in the presence of 6.0 M Gdn-HCl at 90C. 33 Proteins require enough flexibility to interact with other
partners to fulfil various functions. Structures which could be considered evolutionary dead ends should not be counted

as sequences with native-like properties. The significance of this observation will be revisited below. Once again, the
value of p4 must be considerably less than 1.
Correction p5: the proteins lack conformational specificity
Native folds possess discrete conformation.
In order to exhibit conformational specificity, a protein must satisfy three criteria. First, the protein must fold to a unique
tertiary structure rather than exhibiting the conformational heterogeneity that is characteristic of molten globule and
gemisch states. The protein must possess the desired oligomerization state Finally the designed variants must assume
the target fold rather than assuming an alternate fold. 34NMR spectra can be very helpful in determining conformational
states.35 But this was not performed in these experiments.However, gel filtration experiments using the three isolated QLR
proteins revealed that they exist as multimers. In other words, the QLR polymers generated adhere together. Filtration of
artificial protein QLR-1 did not produce a homogeneous entity, and probably forms two or more oligomeric species; QLR-2
was probably a trimer and QLR3 a tetradecamer.36The observations identified under p3p5 suggest that the proteins are
much too hydrophobic to produce native-like folds. Abnormally long alpha helices are formed, which become so stable,
due to intra-molecular and intermolecular hydrophobic interactions, that as soon as one conformation has been reached,
the protein is committed and cannot unfold and search for another slightly lower energy state. In other words, a large
number of similar clumps of protein can form. Once the hydrophobic portions of long patches of alpha coils between two
QLR members interact, somewhere along their chains they will remain strongly bonded and insoluble. This will prevent
them from dissociating and attempt to realign in other, potentially more stable arrangements. The energy necessary to
attain the dissociation transition state would be too high.As a rule, real proteins must not be able to fold and lock into a
multitude of similar conformations. Most of the wrong conformations would not interact correctly with the necessary
chemical partners and could cause much damage by adhering where they should not. There are a few exceptions where
natural proteins have been carefully designed to be able to equilibrate into more than one discrete and well-defined fold
for special reasons, for example to serve as switches. In these cases it must be easy to alternate between intended
conformations.Since the necessary N-cap and C-cap sequences to clearly define where an -helix begins and
terminates are missing, a wide variety of alternative QLR protein conformations could form, contra what defines nativefolded states. Blundell and Zhu37, and Richardson and Richardson38 have shown that special sequences are required to
produce N-caps and C-caps.Marshall and Mayo also point out that sequences that are overly hydrophobic are prone to
aggregation and are predicted to have a smaller energy gap between a target structure and alternate
state.39 Furthermore, aggregates may arise from partially folded states rather than the native state. 40 One purpose of
turns is to ensure that the relevant portions of the protein are held in the correct positions with respect to each other. Since
no sequences which could be classified as turns were found, then a corrective factor much smaller than one is needed.
Correction p6: the data is based on only small proteins
The experiments were designed to produce QLR proteins about 60 or 81 residues long. We saw that these led to high helix content and amorphous structures which are much too stable to generate discrete folds. But the problems for these
small domains, such as excessive stability, will be considerably worse for much larger and typical-sized domains 41 of
about 150 residues (and many folds exist which have more than 500 residues). 42 For example, suppose the alpha coil
were to become 25% larger for these polypeptides. The strength of the intermolecular interactions between multiple
members would grow exponentially, and so would the possible locations along the helices where they could interact,
leading to ever more, and stronger, clumped variants.There are far more QLR varieties which are 150 residues long than
for the short 70 residue chains: 3 150 / 370 = 1038, and in this vastly greater set the proportion able to fold properly must be
considerably smaller than whatever the correct value for the smaller QLR proteins.
Correction p7: the data is based on only single-domain proteins
Most proteins have more than one domain, unlike those reported in the study. 19 Each domain must fold properly and in the
presence of the others.Different domains interact with their intended biochemical partners, which permits the simplest lifeforms to execute the necessary biochemical process with only a few hundred kinds of proteins. 43 No life-forms are known
which use only proteins of size 300 AA or less, and if tiny proteins of 6081 residues are supposed to be the starting point
for an evolutionary processes, then far more of them would be needed than using complex multi-functional versions.
Correction p8: the QLR proteins only generated alpha helices
Estimates vary for how many protein folds exist or could exist, but a value between one thousand and ten thousand is
reasonable. Folds are classified in the CATH 44 and SCOP45 databases, and FSSP/DALI46 classifications are also used to
analyze possible protein structures. Folds are defined by alpha helices and beta sheets plus their geometric relationships
together.None of the folds which require beta sheets, nor mixtures of + , were generated in the experiments reported.
The proportion of random polypeptides which possess different classes of secondary structure ( coils, sheets and
turns) needs to be estimated. The proportion leading to only coils cannot be simply used as representative for the
variety of folds found in nature.
Correction p9: the proteins are not forming native-type folds
A surprising observation is the inconsistent and misleading use of the words protein fold. Based on several facts, it is
apparent that the CATH classification of fold is meant when quoting the supposed 1% proportion of folded to non-folded
random polypeptides. In CATH, this term has a precise meaning and is based on well-defined properties of biological
proteins. For example, the figure shown in a recent paper15 by Dr Denton refers to structural classes of protein folds and
references a book published by Orengo et al. who created and now maintains the CATH system. Furthermore, the same
figure that was shown15 in that paper is used routinely in CATH literature.47Dentons paper quotes Sauers work, which we
are evaluating here, referring to these proteins as being folded. But no CATH fold was obtained from the QLR proteins
and Sauer and his co-authors never claimed so. If millions or billions of times more QLR sequences were to be generated
then a true CATH-like fold might be found. Clearly corrective factor p9 is much smaller than one.
Discussion
Here, we have carefully analyzed whether protein folds are very common among random amino acid chains as it is
claimed in the evolutionary literature. What we have found is quite revealing. Using CATH topology, the standard for
protein folds, none of the randomly formed QLR proteins qualify as protein folds. The work reported by Sauers group was
also very clear about this.18 They use the word folding in a vague, non-technical way. The QLR proteins have one or
more huge alpha helices, as deliberately designed, which then clump together in an amorphous manner. A sheet of
paper can also be crumbled in many ball-like amorphous ways with no resemblance to a protein. Native-like globular
proteins fold into a precisediscrete topology relying on secondary structures and other precise chemical interactions. The
researchers themselves did not claim that the QLR polypeptides do this,18 nor was any CATH fold claimed for any of the
polypeptides generated.Although the above nine corrective terms are not independent, all of them undoubtedly have
values far lower than one. Therefore, to extrapolate from Sauers QLR experiments to real folds based on the twenty
natural amino acids of an average chain length, we would need to make some corrections:Proportion of proteins able to

fold p = 0.01 p1 p2 p3 p4 p5 p6 p7 p8 p9.Although we still cannot make a very good estimate for the
proportion able to fold properly, the value of 1% is too high by many orders of magnitude. Considering the estimated
pi values throughout the text, this proportion is expected to be very small.The author contacted 48 Dr Sauer hoping to
discuss the above corrective factors and included the following comments to him to illustrate how 1% could not possibly
be representative of typical domains nor proteins.Unfortunately, I cannot estimate very reasonably the proportion of
random sequences based on the twenty optically pure amino acids, but lower limits can be placed. For example, all
combinations of binary patterning49 for helices and coils would be covered by a pattern such as: (p/n)AA(p/n)AA where
p=polar; n=non-polar; A=Anything. This implies for a 150 residue domain that only every third position would pose a
constraint. Well pretend any of nine out of twenty residues could be used as p or n. Once a polar or non-polar position is
established, it is obvious from chemical considerations that every combination of AA in the next two positions, as Ive
permitted, would not really be possible (we would permit two and even three prolines next to each other; up to three huge
tryptophans, one being part of the secondary structure, etc.), so our estimate is clearly far too generous. Even for turns,
the pattern (p/n)AA is generous.Therefore, a lower limit everyone can agree to would be no larger than (9/20)51 = 1018,
which is not compatible with the 1% estimate.Note that these generous assumptions would imply a proportion of
(9/20)100 = 1035 for an assumed average size 300 AA protein.This shows Dentons 1% claim is an overestimation by
mega-orders of magnitude. Sauer did not defend the 1% claim, but referred the author to unrelated work by Professor
Szostak. This later work will be addressed in parts five and six of this series. Since Sauers pioneering experiments
provide no guidance to quantitative native-like folding behaviour it is pointless to quote the 1% value, as done by
Denton15, without clarifying its meaning. Watters and Baker also examined Sauers experiment and reached the same
conclusion: Few, if any, of the proteins in these screens were truly native-like, suggesting de novo formation of proteins
may be very difficult.50
Additional requirements to produce native protein folds

Figure 5. Examples of different folded topologies, displayed with RasTop 2.2. Left: CATH topology (fold) 3.40.5 (Ribosomal
Protein L9; domain 1); Domain: 2hbaA00. Right: CATH topology (fold) 3.75.10 (L-arginine/glycine Amidinotransferase;
Chain A) Domain: 1xknA00.
All known folds contain secondary structures: alpha helices and beta sheets, connected by chains called turns. The
structure of the folded proteins can differ dramatically (see figure 5 for examples).5153

Figure 6. Examples of two domains classified in the same topology, CATH fold 3.40.5. Displayed with RasTop 2.2. Left:
Classification: 3.40.5.10; protein: Ribosomal Protein L9 domain 1; domain: 2hbaA00. Right: Classification: 3.40.5.20;
protein: PriA/YqbF domain; domain: 2hjqA01.
The folds take only the backbone features into account, and on this basis commonalities can often be discerned for
protein classified in the same fold (see figure 6).54,55 Although all proteins have been classified into only about a thousand
unique folds in the CATH56 and SCOP57databases, merely possessing helices or sheets is by no means sufficient to
guarantee proper folding. Some additional relevant details includes:The non--helix22 and -sheet23 portions of domains
are involved in helping the folding process in bringing the secondary structures together, which will then form into a
suitable location. In some proteins the proportion of residues found in the turn can exceed 30% and are not random
structures58 Some features in protein turns have been identified and formally classified 24: -turns59; -turns60 (at least eight
forms
have
been
identified)61; -turns62 (there
are
two
forms)63; -turns64;
hairpins65;
and -loops66.
62
The residues located in the turns bend the main-chain and affect the distance separating the secondary structures.
Often specific motifs are found across members of some protein families, like the tyrosine corner in Greek key betabarrel
proteins.62

The need for these special features to form the -helix and -sheet structures is one reason why so many amino acids are
needed for biological proteins, and why the three used to construct the QLR proteins would not suffice.To place -helices
at the correct location and to define their size, certain combinations of residues determine where they begin and
end.67 For example, a four-residue N-cap often terminates the end of helices and, in addition, often the next two residues
which flank the box display typical hydrophobic interactions.Classification of folds is based for the most part on only the
backbone topology, which is a minor portion of the protein mass. The side chains are the key to providing biological
function and also play important roles in determining whether a stable fold will be produced. The common use of the
cartoon representations of secondary structure leads to an over emphasis of the backbone. Figure 7 shows a typical
cartoon-type display (left), in which the details of the side chain are excluded, and the same structure with the sidechains
included
(right).
The side chains, it must be remembered, can interact whether attached to helices or sheet (see figure 7). And the details
of the side chains are fundamental to understanding how a protein folds (and functions). These create microenvironments with precise spatial and electronic details. Therefore, structural and electronic requirements in the protein
core place constraints as to which residues can be used at various residue locations. The side chains must ensure that in
the core, cavities are avoided and strain due to interference between side chains must be minimized.
Only some combinations of amino acids are able to provide the environment necessary, through their side-chains, to
produce stable conformations within a given fold.Although proteins are generally dynamic, flexible structures, they must
usually equilibrate most of the time near the native state. But this means that other conformations, which are energetically
similar and easily attained, must often be prevented for a protein to be useful. Judicious choice of residues at specific
positions can destabilize undesirable competitive conformations, and is one reason for intolerance to substitution by other
residues with side chains which are quite different in size, shape, or charge characteristics.18

Figure 7. Ribosomal Protein L9 domain 1; domain: 2hbaA00, displayed with RasTop 2.2. Left: Usual cartoon
representation of secondary structures, with no side chains shown. Right: The same domain and orientation, including side
chains. The backbone is shown as a solid chain.
Summary
The QLR proteins were expertly designed, knowing that otherwise it would be unlikely to find native-like folds among
random sequences using 20 AAs. Although it is legitimate to first simplify a problem to gain insights, any quantitative
conclusions made for the original problem cannot be based merely on the simplified work without reasonable
extrapolations or experimental calibrations. We have not directly addressed all the papers dealing with QLR and other
simplified proteins, but have shown that the claim that 1% of random proteins would produce native-like folds is
unsupportable based on the QLR experiments.
The proportion of polypeptide chains which generate native foldspart 2: theoretical studies
by Royal Truman
Two decades ago Lau and Dill developed a computer program based on a two-dimensional lattice model. With this
program they showed that between 10-6 to 10-10 random polypeptides, a hundred residues long, would produce stable
protein-like folds. Ever after, their seminal work has been quoted as evidence for the claim that randomly formed
polypeptides readily produce native-like folds and for a naturalistic origin of globular proteins. However, the model has
never been calibrated nor validated against empirical data. Here, we show that their program/model is far too removed
from the realities of protein chemistry to permit any kind of quantitative estimates.
Two decades ago Lau and Dill developed1 a computer program, and the
quantitative claims from the seminal paper are still quoted. 2Backofen
emphasized the importance of the problem being addressed:
The protein structure prediction problem is one of the most (if not the
most) important problem in computational biology. This problem consists
of finding the conformation of a protein with minimal energy. Because of
the complexity of this problem, simplified models like Dills HP-lattice
model have become a major tool for investigating general properties of
protein folding.3
Lau and Dill wrote4 that theoretical estimations predict the fraction of
random protein sequences that fold into stable, native-like structures to lie
between 10-6 and 10-10. The authors point out, correctly, that probabilities
much lower than these would raise doubts whether a functional protein
could arise naturally, and comment:

This general argument has become of some importance as support for the view that proteins could not have arisen from
natural prebiotic chemical processes on earth and as support for creationism.4
We concur that a very low probability would fit better with creationism than materialism, but following good scientific
methods, we should first collect the raw data and then decide about its significance.
The results of the model led Lau and Dill to claim,
On this basis, extrapolation shows that for chains of n = 100 monomer units, the fraction of these 2D sequences which
fold is in the range of 10-6 to 10-10, depending on the strictness of the criterion used to classify a sequence as a folding
molecule.4
The authors conclude the paper with a sweeping statement about what they claim to have shown:
And it suggests how molecules as complex as catalytic globular proteins could have arisen so readily in simple prebiotic
solutions, wherein only a virtually negligible fraction of all possible sequences would have been sampled during the origins
of life.5
The experimentFigure 1. Two-dimensional model for a protein chain
based on two kinds of amino acid: dark = hydrophobic; light = nonhydrophobic. Each amino acid can occupy only one position in the
lattice.6
Proteins are modeled as chains composed of only two elements: H
(hydrophobic) or P (polar, meaning non-hydrophobic). Each element,
which represents an amino acid, can occupy a single square within a
two-dimensional lattice. Thus, each amino acid is connected to its
chain neighbour(s) and can also be in contact with topological
neighbours adjacent in space (figure 1).Every H-H contact between
topological neighbours is assigned a contact energy <0 and every
other interaction between neighbour pairs has energy equal to 0. For
short chains of n = 10 to 25 elements,7 every accessible conformation
was found using an algorithm they programmed. The number of H-H contacts defined the energy of the whole chain, and
sequences with lowest energy represent protein native states.
The reasons for the project were stated in the Abstract:
we address two questions regarding the evolution and origins of globular proteins. (i) How will protein native
structures and stabilities be affected by single and double-site mutations? (ii) What is the probability that a randomly
chosen sequence of amino acids will be compact and globular under folding conditions? 8What is now required, based on
goal (ii) is of paramount importance: a way to map the sequences of Hs and Ps, and their calculated energies, to the
propensity of real proteins to fold, and remain folded, in a discrete native state. Without such calibration, the computer
program is worthless. In addition, one must not forget that for proteins the conformation in the lowest-energy state will not
automatically be a properly folded and soluble protein, based on well-defined alpha helices and beta coils. The protein
could simply produce a tangled mess or be attached to other proteins in some amorphous manner.The criterion used for
whether folding occurred was described as follows:
To decide whether a sequence is a folder or not, we use the compactness (13)n = t/t max, where t is the total number of
topological neighbours in a given conformation and tmax is the maximum possible number of topological neighbours which
could be achieved by any conformation of a given chain length. Below we consider different criteria for globularity, ranging
from the most strict (t = t max) to less strict (t >= t max 2); a conformation is considered folded if it satisfies this criterion
under folding conditions.8
No empirical justification for the (13)n = t/tmax assumption was provided, based on known protein chemistry realities, and
the reference cited9 was not helpful. Essentially, the authors assume that if enough hydrophobic residues (linked amino
acids) are neighbours in space (but not adjacent along the chain sequence) in the two-dimensional lattice, then a nativelike fold will form.
Figure 2. Proportion of random polypeptides with native-like folds
decreases with chain length but the quantitative relationship is
unknown.
The assumption was then used to answer the question posed in
the Abstract:
What fraction of sequence space corresponds to compact
globular molecules under folding conditions? Let N(n) equal the
number of sequences of chain length n in the sequence space,
and let Nf(n) equal the number of sequences which foldi.e.,
those in which the conformations of lowest energy (native states)
are maximally compact or nearly so. The simulations show that
the fraction of sequence space corresponding to folding
molecules diminishes approximately as
Nf(n)/N(n) = kan, (1)
where the constants are k = 2.04 and a = 0.792 for t = t max, k =
4.87 and a = 0.813 for t >= t max 1, and k = 4.06 and a = 0.860
for t = tmax 2.4
Now, everyone agrees that the proportion of polypeptides leading to native-like folds decreases with chain length, but it
remains to be clarified experimentally, or based on sound theoretical principles, how rapidly. Figure 2 illustrates various
possibilities.
The parameters for the equation of form ka n were determined empirically, based only on the speculative criterion for
folding mentioned above. We used the equation with several values of n, leading to the results given in table 1. There we
find for n = 100 the range of 10-6 to 10-10, which the authors claimed4 to be a realistic estimate for the proportion of stable,
native-folded proteins in a random sequence of amino acids that length.
k

kan(n=50)

kan (n=100)

kan (n=150)

kan (n=300)

t = tmax

2.04

0.792

1.8E-05

1.5E-10

1.3E-15

8.5E-31

t >= tmax-1

4.87

0.813

1.6E-04

5.0E-09

1.6E-13

5.2E-27

t >= tmax-2

4.06

0.86

2.2E-03

1.1E-06

6.1E-10

9.1E-20

Table 1. Proportion of stable, native folds, based on an extrapolation from a computer model using the relationship ka n, for
a hypothetical two-dimensional lattice of n amino acids.4
Evaluation
The authors clearly believe they have a useful tool to understand biological proteins, and use the computer program to
make a variety of broad statements about the quantitative effects of single and double mutations; the proportion of lowerenergy degenerate states and when they occur10; and about the nature of the internal protein core. They use phrases like,
the simulations show and the simulation results many times.The reader of their paper is confronted with the obvious
question of the relevance to real proteins. Amazingly, the results from their model were not calibrated in any manner
against real biological data. Although they comment on their model and approach that It has no adjustable parameters or
additionalad hoc assumptions,8 this claim is misleading. For example, the criterion for whether a sequence is a folder or
not has not been verified quantitatively in any manner. As well see below, the authors examined
between n = 10 to 25 points on a lattice. Still, many refer to this model with very little semblance to known protein
chemistry, while extrapolating to n = 70 and higher.2 The readers who have not examined the original source for such
quantitative values are led to believe that the values have a biological basis in reality; when they are purely speculative
models based on minimal assumptions.The authors concluded that globular proteins arose readily in simple pre-biotic
solutions,5 but they did not address any of the chemical and thermodynamic issues which preclude this bold extrapolation.
Additional considerations must include the need for homochiral amino acids; avoidance of side chain reactions; intramolecular peptide bond formation; instability of growing polypeptides in water; the presence of other reactive chemicals;
and so on.To be as generous as possible to the authors model, we shall assume that optically pure amino acids only
were available, react to form only linear proteins, and do not agglomerate with other chains. Nevertheless, the effects of
some Neglected Realities (NR1NR5 below) need to be estimated if any value is to be extracted from their report. Then
an estimate for the proportion of random polypeptides of chain length namino acids which lead to a native, stable fold can
be obtained.
Figure 3. For a chain in two dimensions, an additional monomer can theoretically
assume any of three positions in the lattice.
Neglected Realities (NR) for biological proteins
NR1: folding degrees of freedom
In the computer model, the next member of each chain can theoretically assume
any of three conformations (figure 3).
Those who have played the computer game Pacman 11 should recognize the
principle: one can only move forward, left, or right. Given that at least two residues
are needed for a polypeptide and that the end of any chain has a maximum of
three possibilities at which the next residue could be placed on the lattice, the
maximum number of chains for the computer simulation is easily seen 12 to be
given by:
maximum number of chains, C = 3(n-2) (1)
for n residues, where n is two or greater.
However, a little thought shows that as chain size increases in the two-dimensional lattice, the maximum number of
variants possible becomes far less than implied by formula (1). This is illustrated in figure 4. The quantitative effect is
significant, since every chain end with less than three subsequent alternatives limits the entire branch of possibilities
starting from that point.
Figure
4. As
the
twodimensional lattice becomes
larger, ever more ends of the
chain will have fewer degrees
of freedom for additional
growth. A:
zero
chain
extension possibilities. B: two
degrees of freedom at the end
position. C: one degree of
freedom at the end position.
In real proteins, the potential to assume erroneous conformations is much greater than implied by the lattice model, 13 and
the requirements to ensure funneling into the native fold configuration increases accordingly. There are two ways to
calculate the number of folding conformations of proteins. Using n for the number of amino acids, some researchers 13 use
the formula 32n-2 and others14 use 8n. For most practical purposes it does not too much matter which approach is used,
since the resulting discrepancy calculated rarely affects the conclusions made. 15For an average-sized 150-amino-acid
domain of a protein,16 the potential number of folds is vastly greater
than the two-dimensional model suggests, even ignoring the fact that
eq. (1) overestimates the maximum number of variants by many
orders of magnitude:
32n-2 / 3n-2 = 3298 / 3148 = 4 1071 (2)
This fact is not properly captured by the authors through limiting the
computer runs to between 10 and 25 circles on the two-dimensional
lattice. Note that the computer model only permits exact 90 angles,
whereas protein can adjust angles in three dimensions over several
residues to accommodate crowded folds.In addition, the spatial
relationship between residues involved in alpha helices 17 and beta
sheets18 shows no resemblance to the flat two-dimensional shapes
assumed for the lattice. For example, a regular alpha helix has 3.6
residues per turn19 due to hydrogen bonds formed between residues i
and i+4. A variant helix called 310 is based on hydrogen bonds formed
between residues i and i+3.20 There are even examples of hydrogen
bonds formed between residues i and i+5, known as the pi helix.20

Beta sheets, parallel and anti-parallel, also involve hydrogen bonds.21 In addition, there are many kinds of turns which also
represent secondary structure.22 And all these secondary structures, fundamental to producing native folds, have little to
do with whether the residue is polar or hydrophobic.
Figure 5. Dipeptide formed from the condensation of two amino acids. The twenty natural amino acids each have a
different side group (R1 and R2); these differ considerably in size and chemical characteristics.
NR2: packing geometries
In the model, only two elements are allowed to build the chain, H or P. In reality, the side chains of real amino acids
differ considerably in size, shape and electrical charge. 23 The overall environment of the folded protein brings different
side chains together, and to pack properly various constraints must be met. If the side groups brought together in the
protein core are too small, then cavities are introduced into which water can penetrate, producing an instable
conformation. If some of the side chains are too large, stable packing is also hindered.Figure 5 illustrates how different
amino acids can be. For example, compare the size of the side chain, R, for glycine with that for tryptophan. The residues
classified as H (hydrophobic) can have dramatically different effects on the ability to pack into a native-like fold, depending
on which residues are located nearby.Very different residues, like the three shown in figure 5, were indiscriminately
treated as equivalent H in the computer model. But large side groups, like tryptophan in figure 5, cannot be compressed
near other large side chains in the core. Proline, the third hydrophobic amino acid shown in figure 5, forces a sharp kink
in the protein, severely disturbing the geometry of most protein cores except when the rest of the chain has been
appropriately designed. Salt bridges24 (figure 6) can form with some residues, contributing to the stability of native folds,
and so can disulfide bonds25 formed between the thiol groups of cysteine residues (figure 7).

Figure 6. Example of salt bridges with the amino acids glutamic acid and
lysine.24
Figure 7. Disulfide bond formed between two
cysteine amino acids.25
NR3: minimum energy necessary to remain folded
The proportion of random real polypeptides able to fold, and to remain folded reliably, was not addressed in this paper.
We quoted above what the authors did: they compared the total number of topological neighbours in a given conformation
with the maximum possible number of topological neighbours which could be achieved, and then concluded, on the basis
of two-dimensional hydrophobic interactions, that a large number would be folders. 8The model is too crude to be
quantitatively relevant, even as a rough indicator. The actual distance between the hydrophobic side chains and strength
of the specific interactions will determine the stabilizing contribution. Treating all non-H-H interactions as neutral is wrong,
since electrically charged amino acids like aspartic acid and glutamic acid (negatively charged); and lysine, arginine and
histidine (positively charged) will generally provide a stabilizing contribution if in contact with water instead of in the core.
NR4: conformations with local minimum energies
The authors assume that the most stable conformation will inevitably be found and ignore the critically important question
of how long this may take. In fact, there are far more possible conformations than could ever be examined by trial and
error. This protein-folding problem is well known and referred to as the Levinthal Paradox, 13 which we shall revisit
(later).We saw in eq. (2) that the lattice model understates absurdly the number of incorrect ways a protein could fold.
Finding the native-like fold among random sequences in a time span relevant for biological purposes would be prohibitive.
The fundamental issue is that acceptable sequences must guide the folding process, like a funnel, into the correct
conformation. This means that alternative, low-energy conformations along the path to the correct fold must be prevented.
Locally minimum energy states could be stable enough to hinder unfolding to permit additional trials in the search for a
better conformation. Another perspective is that the protein could spend too much time in countless alternative lowenergy, but functionally worthless, topologies for that sequence to be useful (figure 8).
Figure 8. A protein can remain trapped in a
local energy minimum along the multitude of
paths leading to an absolute lowest energy
conformation.

Figure 9. In a two-dimensional lattice permitting only two kinds of

elements, neighbouring dark ones not adjacent along the chain are
assumed to provide favourable interactions. Once a conformation with a
favourable local minimum energy is formed, it will be difficult to unfold to
permit searching for an absolute best conformation.
One can use the two-dimensional lattice to illustrate the difficulty. During
chain folding, the next circle assumes up to three possible positions. This
will often result in variants with a very stable local topology (figure 9).
Those variants would include a portion locked into a particular
arrangement and cannot unfold easily (especially under biologically
relevant time scales) to search for the best-of-all conformation. A collection
of random sequence will have folded members distributed throughout a
vast range of conformations. But the pathway to getting there is littered
with pitfalls. Therefore, useful real proteins must hinder the formation of
undesired false minima.
NR5: binary assumption
The authors assume that only hydrophobic (H-H) interactions need to be considered to model protein folding. Therefore,
their amino acids are either hydrophobic, H, or not hydrophobic, P. They refer to experimental work done elsewhere
which used optimal proportions of amino acid chains with these characteristics, for example, the experiments of
Rao et al.26 on random polymer sequences of lysine, alanine, and glutamic acid. Lysine is charged, alanine is
hydrophobic and glutamic acid is polar (hydrophilic). These are optimized choices known to generate secondary structure
and stable folds. No attempt was made to calibrate these other kinds of experiments with biologically relevant amino acid
proportions.Similar work has been done in professor Sauers lab at MIT.27 In their case leucine (hydrophobic), glutamine
(polar, hydrophilic), and a small amount of Arginine were used. Their choice was based on the high propensity of leucine
and glutamine to form alpha helices, which should facilitate formation of stable, folded proteins.But none of these others
studies, which tested large numbers of binary sequences, were able to identify a native-like folded protein, as also
reported in this journal.28 Therefore, it is not justified to restrict the parameters which cause folding to only hydrophobicity
considerations.
Levinthal Paradox revisited
The polypeptide backbone chain, figure 5, is represented by sequences N-C -C-N-C-C-, where C refers to the carbonyl
carbon. The next carbon along the side chain attached to C is called C, and so forth. Atoms C-N-C-C define the torsion
angle , whereas N-C-C-N-defines angle . Conformations about these two angles create unfavourable contacts with
neighbouring atoms which limit the conformation space used by native folds. Ramachandran diagrams29 plot angles on
the y-axis and the on the x-axis, both from 180 to 180 degrees. Experimentally observed angles for residues from
biological proteins have been used to generate contours of the regions used in nature, one contour representing strands, the other -helices and a small region of left-handed helices, which are very rare. These contours show that only
about 30% of the space of backbone conformations are used by biological proteins per residue.30 Since most proteins
consist of hundreds of residues, a random sequence is extremely unlikely to find itself in a biologically relevant
conformation.Glycine, with a hydrogen as the side chain, is not subject to this restriction. Turns are also not taken into
account. Since the average protein size is greater than 300 residues3133 (a conservative figure often used by creation
scientists), we can simplify our calculations to obtain a reasonable estimate for the portion of backbone conformational
space used by native-like folds:
(0.3)300 = 10-157
Residues in an acceptable contour region may need to unfold if the true native fold is to be found. For example, a portion
of the backbone which initially would possess angles consistent with -coils may need to rearrange entirely to participate
in a -strand with distant regions of the folding protein. To illustrate, translation occurs at rates of about 20 residues per
second,34 yet many soluble proteins fold entirely within 0.1 seconds. This suggests that in the cell restrictions exist for the
folding process, or, alternatively, there might be a way to reopen the chain to permit the correct final states to be
formed.The lattice model oversimplifies and neglects this reality. Finding the lowest energy conformation would not occur,
even in billions13 of years, for most random polypeptides of average protein length. Therefore, specific amino acid
sequences must be present which funnel the folding into a limited number of paths. And this implies that a very small
number of random sequences would provide the necessary physical details to ensure proper folding in a biologically
relevant time scale.Furthermore, we suggest that some well-defined amino acids must be present at various positions
toprevent incorrect possibilities from being explored. The folded proteins are generally only between 15 and 40 kJ/mole
stabler35 than the open chain,36 a remarkably low value, given how often and how quickly the native state is found. One
way to prevent incorrect conformations may be through the use of well-designed turns. Turns are combinations of amino
acids leading to specific shapes and features, and connect alpha helices and beta coils in proteins. We suggest that
careful analysis and designed mutations of turns may reveal their importance not only in generating the correct final layout
but also in preventing incorrect alternatives during the process of searching for the native state. Turns are good candidate
portions of proteins to test this hypothesis: are undesired conformations often hindered by the sequences present in turns.
It is also possible that undiscovered cellular equipment besides chaperons help guide the folding process. These nanomachines could be located as part of ribosomes or near them. Formation of individual secondary structure in an open
chain is not such an easy matter.
Although helices have regular repeating hydrogen bonds coupled with a uniformity of bond lengths and angles this
periodicity masks their marginal stability. In an isolated state most helices will unfold and only synthetic propyl-alanine
helices are reasonably described as stable. Initiation of helix formation is a slow and unfavourable process. This arises
because five residues must be precisely positioned to define the first hydrogen bond between residues 1 and 5.37
Summary
This paper showed that the program developed by Lau and Dill is severely hampered. As a computer simulation, it has
little, if any, relevance to the subject modelled. More severe, still, is that it has never been tested nor calibrated against
real biological data. This seems to be a general trend in Evoland. Apparently, unproven models and unrealistic simulations
are taken as substitutes for real biology.Although the high proportion claimed for random polypeptides able to produce
native-like folds might be attractive to those favouring a naturalistic origin for globular proteins, modern investigators know
that this widely cited paper does not produce quantitatively meaningful information. Referring specifically to Lau and Dills
work, Backofen and Will comment:

The main motivation for studying simplified protein models is to be able to predict model structures much more quickly
and more accurately than is possible for real proteins. However, up to now there was a dilemma: the algorithmically
tractable, simple protein models can not model real protein structures with good quality and introduce strong artefacts.38
Conversely, the computer expert may confirm that the algorithm to find all conformations for any length n of residues
would be correct for the lattice model, but have no idea whether the model represents anything biologically relevant.It is
imperative that any computer simulation must be tested against real data before making quantitative claims. This has not
been done in this report,1 nor any follow-up work, but then surprisingly, the results were extrapolated to larger lattices with
no knowledge of how real proteins react to increasing chain length. In addition, the authors ignored the time necessary to
find proper folds and just assumed that if they existed, nature would easily find them.The quantitative claims are little more
than speculation and could easily be wrong by a hundred or more orders of magnitude for reasonably sized proteins.
Quoting the quantitative conclusions is irresponsible and to be avoided even if in harmony with ones preconceptions.
Some creation scientists might feel that the lower bound of 10-10 for a hundred residues is acceptable, since 3 or 4 such
domains in a protein, or large domains, would imply for an average size protein a proportion of about 10-35, but this
temptation is to be avoided. 39 Inevitably, others would argue that the smaller domains arose somehow and then nature
only had to link them together.It is regrettable that we dont have a simplifying algorithm, properly calibrated, to calculate
the correct proportion40 of random protein sequences able to generate native-like folds. A computer program able to
perform the calculations in a physically realistic manner to test random sequences exceeds by far the worlds current
computing resources.What is the correct proportion? The data in this six-part series do not provide the answer, except to
confirm that it is lower than the estimates each author suggested. It would be sensible to examine the
facts first and then to reason out an interpretation within our worldview. For example, if large proportions of random
sequences were to fold in energetically almost degenerate states, then so be it. The creationist would then point out the
consequence, that very few would be found at a given time in the biologically relevant conformation and would not remain
folded corrected for very long in that shape.Papers modelling reality should be much more critically scrutinized for their
scientific merit. A reason why they are not receiving adequate critique is the need for readers to cover multiple disciplines.
In the paper mentioned above,1 the person knowledgeable in cell biology might be intimidated by unfamiliar techniques or
phrases, such as a recently developed lower-resolution lattice statistical mechanics model of protein folding. 40 Papers
such as these often also include some very intimidating and unfamiliar mathematics 4 which a biologist may feel
unqualified to challenge. Therefore, creation scientists should not only be specialists in their own fields, but should also be
encouraged to become multidisciplinary.
The proportion of polypeptide chains which generate native foldspart 3: designed secondary structures
by Royal Truman
In this six-part series, we are evaluating those reports which
claim that a high proportion of random polypeptide chains
would produce protein native-like folds. Often claims are made
by others who cite these papers without explaining the
designed elements which went into the experimental protocol.
A thorough understanding of the work actually performed, or
neglected, is necessary before general statements can be
made for truly random protein chains of biologically relevant
length. Here, we discuss a paper published by a research
group at the University of Zurich, which performed one of the
most sophisticated attempts to create new but unspecified
folds through intelligent design. The results show that, even
though all information known about the requirements to
generate secondary structures (binary patterning; length; Ncaps; C-caps; optimal choice of amino acids) went into the
design, under natural aqueous conditions no examples of
native folded proteins were found.

Several research groups13 have attempted to create novel proteins with new polypeptide sequences. Usually a specific
target fold is selected and they attempt to reproduce its shape and characteristics using other combinations of amino
acids. Although such artificial proteins can be intelligently designed, creating novel ones which fold properly has been far
more challenging than expected. None of these studies provided the quantitative data we wished to find to help estimate
the proportion of random sequences between 50 and 350 AA (amino acids) which would produce native-like folds under
biologically relevant conditions. Therefore, those studies which targeted a specific fold, almost always unsuccessfully, will
not be reviewed in this series of papers.Our purpose here is to evaluate the evolution-inspired premise that native-like
folds are easy to generate naturally. To address this, we will discuss research involving a series of expressed cassettes
containing code for semi-random short protein structures. The design of these structures was based on statistical
properties of protein secondary structures, to provide optimal chances of generating artificial proteins with native-live
folds. As we will see, about 10 9 sequences were generated, but none of them were reported to possess properties typical
of true native-like states. Only under abnormal conditions, such as high salt concentration, were some poorly-defined
globular-like properties identified, lacking a discrete tertiary structure.To determine whether many or few polypeptide
sequences would produce native-like folds, the Zrich group4constructed artificial genes based on polar and nonpolar
residue patterns which favour secondary structure 5 (-helices, -strands and turns, i.e. the components of protein folds).
These genes produced several libraries of synthetic proteins which should possess the following characteristics: all helix; all -strand; and -helix plus -strand, having an average length of 100 amino acid residues. As a rule, about 90%
of a biological protein consists of secondary structure.6 Since random sequences will rarely form secondary structures
consistently under biologically relevant temperatures and aqueous conditions, the authors decided to restrict the space of
possibilities to semi-random chains, most likely to form such structures. Otherwise very few, if any, artificial proteins with
properly folded conformations would be identified.
The experiment

The experiment conducted by the Zrich group was carefully designed and was based upon the upfront knowledge we
have on naturally occurring functional proteins. In order to optimize the chance to find random new combinations, they
started with short amino acid patterns known to favour -helix and -strand formation. The designed details include 1)
kinds of residues in tailored environments with other specific amino acids, 2) that -helices in nature average about ten
amino acids and 3) that -strands typically use about five amino acids. In addition, tri-nucleotides were used to code for
the amino acids most likely to generate the secondary structures desired. Short chains were produced and these linked
together to express proteins with a high proportion of stable secondary structures.
Construction of -helices
To construct these common protein structures, blocks of five and ten amino-acids were designed. To define one end (the
N-cap side) of the helix, the researchers benefited 7 from statistical studies8 that show amino acids Asn, Ser, Asp, Thr, or
Gly followed by Pro or Glu occur with highest frequency. They used Ser followed by either Pro or Glu, mixed in equal
proportions. At the other end (the C-cap), they used the statistically preferred residue Gly.8The choice of amino acids used
between the bounding N-cap and C-cap was guided by the studies of West and Hecht. 9 They had created a database of
penta-peptide polar (p) and non-polar (n) patterns found empirically in helices (the binary patterning). The most
frequently occurring patterns, pnnpp, nnppn and nppnp, were used.4 to design the five-residue -helix containing
cassettes. For the ten-residue cassette, the most frequent pattern, ppnnppnnpp, was selected.Using the tri-nucleotide
technology developed by Virneks,10 the five amino acids most frequently occurring in helices 7 were used: Gln, Glu, Lys,
Arg and Ala for the polar ones (p); and Ile, Phe, Leu, Met and Ala for the non-polar residues (n). Ala was allowed in the
experiments at the n and p location, since it is known to have the highest propensity to form helices (figure 1).
Construction of -strands
Binary patterns npnpn and pnpnp are the most commonly found folding patterns in beta-strands, especially in domains
exposed to the exterior of the folded protein.11 To construct these patterns, Arg was also designed into the cassette,
because it is known to be a good N-capping amino acid for -strands.In addition, Val, Tyr, Thr and Ser were chosen 7 to
encode p in the binary pattern. The reason for this is that these amino acids are known to have the highest frequency on
the water-exposed side of the alternating -strand.12 In the buried hydrophobic interior of proteins, the order of preference
is, in contrast, Val, Cys, Phe, Ile and Leu. Therefore, these AA were used to encode the non-polar (n) residues (figure 1).
Cys was excluded7 to avoid the complications which would result from the formation of random disulfide bridges.
Figure 1. Design of the
secondary structure (helix,
strand and turn) cassettes later
linked to form new artificial
genes. Polar and non-polar
amino
acids
are
shown
as p and n, respectively.14
Construction of -turns
-turns are another important
structure
component
of
proteins. They reverse the
peptide chains at the surface of
the protein and permit them to
become globular.13 The most
common, type I turn,13 was
used in the experiment. It
consists of a hydrogen bond
formed between the mainchain position i and i+3. Amino
acids between i and i+3 were included based on their known frequency of occurrence13 in the turn cassette, which was
later bonded to other cassettes. Asp, Ser, Asn, and Thr were chosen for position i; Asp, Ser, Thr, and Pro for positioni+1;
Asp, Ser, Asn, and Ala for position i+2; and Gly for position i+3 (figure 1).The cassettes prepared were amplified using
PCR and digested with restriction enzymes BamHI and BgIII, leading to the overhangs shown in figure 2. Since DNA
base-pairing is between A-T and C-G, the ends of the structural cassettes can be ligated together to form longer
oligonucleotides. Multiple rounds of ligation and digestion with the enzymes were carried out to generate four libraries of
around 100 amino acids in length.
The libraries were constructed as follows:
Library 1 (105t): Using structural modules from 10, 5 and turn cassettes
Library 2 (10t): 10 and turn cassettes
Library 3 (105t): 10, 5, and turn cassettes
Library 4 (t): and turn cassettes.
Figure 2. The structural cassettes
carry the enzymatic restriction
sites Bg III
andBam HI.
This
permits them to be ligated
together. The amino acid pair GlySer connects the cassettes. A, C,
G and T are the DNA four
nucleotides used in the genetic
code.
All libraries were cloned in a
plasmid15 expression
vector16 (derivative of pQE1617) obtained from Qiagen18 with a T5 promoter (which is recognized by E. coli RNA
polymerase) and a strong Shine-Dalgarno sequence 19 (which included a histidine tag at the C-terminal to promote
transcription). The vectors were introduced into E. coli and exposed to ampicillin and chloramphenicol 20 to kill the bacteria
lacking the modified plasmid. The expressed proteins of the seven most promising colonies from the four libraries were
examined using various standard tests.
Results
Library 1 and 2 (105t and 10t)

Three proteins were examined, but the authors did not explain the basis for their selection nor why only three were
chosen. Under normal, biological aqueous conditions the circular dichroism (CD) spectra indicated a substantial
proportion of random coil formation (i.e. absorption having a minimum at 200 nm) and little or no evidence for helices.21 Only under very high concentrations of NaCl did the CD spectra suggest formation of helices. Salts affect
properties of proteins such as stability and solubility. Therefore, the helix formation and the ensuing tertiary structure
formation may be at least partially due to the high concentration of salt added. 23Gel filtration experiments indicated that
the three proteins were larger than expected for native globular proteins, which is typical for molten globule, but atypical
for properly folded proteins. Sedimentation equilibration experiments in (NH 4)2SO4 indicated that one of the proteins
formed as a trimer and the other two as an ill-defined mixture of probably monomer, dimers and trimers. The fact that they
could be isolated, and had not been proteolyzed in E. coli, is likely due to their aggregate states, which protect them. Had
they been degraded by proteases in the bacteria, this could be interpreted as lack of stable folding. But the opposite, lack
of degradation, need not demonstrate stable folding, as suggested in the paper.221,8-Anilinonaphtalene sulfonate (ANS) is
a fluorescent probe that can detect the accessible hydrophobic core area. Detectable binding is consistent with moltenglobule states of a protein, whereas it generally shows only weak binding to native or totally unfolded structures.All three
proteins tested showed no detectable binding to ANS in the absence of salts, but significant binding in 3.5 M NaCl. 23 The
lack of binding in the absence of NaCl indicates lack of secondary or tertiary structure, in agreement with the CD tests.
In high concentrations of NaCl, these three proteins do form helices, based on the CD results and gel filtration
experiments, and binding to ANS is consistent with formation of a molten-globule.Urea equilibrium unfolding experiments
in the presence of both NaCl showed no cooperative unfolding behaviour.Taking all the results together, the experiments
imply that under aqueous conditions as typically found in nature, these three proteins remain in an unfolded state, but in
the presence of high concentrations of NaCl, they form a molten-globule state. 24 The results provide no evidence that the
sequences in this library led to a native folded state. That only three candidates were selected, possibly on the basis of
their greater quantities, implies to us that the authors found this library to be unpromising.
Library 3 and 4 (105t and t)
Two proteins from the 105t library were chosen for further characterization. 24 Why only two were selected, after so much
effort in creating the library, and why these particular two, was not discussed by the authors. Neither bound to Ni-NTA
agarose beads, except in the presence of 6 M GdnHCl. This indicates that the histidine tag is inaccessible and implies
aggregation of the protein. Furthermore, one of the proteins precipitated above 0.5 M NaCl and the other at 1.5 M NaCl.
These facts imply that these proteins have exposed hydrophobic patches due to the sheet. 24CD spectra of both proteins
showed the presence of a random coil conformation and no -helical character, except for a very weak signal in the
presence of high concentration of NaCl. The signals in the absorption spectra were much too weak to determine whether
caused by -helix or -sheet formation.Two proteins from library t were expressed and studied in more detail.
Electrospray mass spectrometry showed they had the expected mass, but they could not be solubilized in high
concentration of denaturants like 8 M urea or in 6 M GdnHCl. Electron micrographs showed clear evidence they were
clumping together, forming amyloid-like fibrils. This effect illustrates the difficulty of generating -sheets which could in
principle participate in native folds built using amino acid patterns of pnpnp and npnpn (figure 1).The failure to identify
properly folded artificial proteins in this library is not entirely surprising. Matsuura et al. point out: West et al. found that a
combinatorial library of -strands, in which each module has alternating polar and non-polar residues, tends to form
soluble amyloid fibrils.25
Evaluation of results
The work described in this paper is one of the most carefully planned experiments designed to produce folded proteins
using semi-random sequences.It is of note that introducing -strand cassettes into the protein made them very
aggregation-prone. This experimental insight is useful and led the authors to conclude that the presence of -strands
seems to require precise topological arrangement to prevent aggregation.26Furthermore, It follows that in all- proteins a
number of mechanisms must be operative to prevent this fibril formation, including topology enforcing turns
(underrepresented in our libraries) and deviations from the regular spacing of the -strands in the sequence.Also, nativelike folded new proteins were not found, in spite of the astuteness which went into the design. Remarkably, here the
authors offer the correct insight: Because of the restricted nature of our cassettes (only 45 amino acids allowed at most
positions) and the absence of any selection applied, we cannot reasonably expect to obtain the finely tuned packing
required for natively folded proteins.In other words, evolution could not have started with far fewer than the twenty natural
amino acids, and built proteins from them.From figure 1 we see that seventeen different amino acids were used in the
polypeptides, placed at the residue position most likely to be useful, but this was not sufficient. The specific residues
which should be used at each position 27 depend on the overall context and design of the whole protein in its final folded
state.
Quantitative conclusions
As scientists, we are naturally disappointed that the authors were unable to identify new kinds of folds. This could have
provided insights into the puzzle of how and why naturally occurring proteins fold so quickly and reliably in the face of so
many alternative conformations and such small difference in energy between the most stable native-fold and the random
state. New folds might also offer possibilities for medical or industrial applications. But our purpose here is to evaluate the
evolution-inspired premise that native-like folds are easy to generate naturally.The library contained about 10 9 different
gene sequences, but the number of plasmids actually introduced into E. coli was not estimated in the paper. Presumably a
correspondingly large number was used. The implication is that a pool of several million different sequences were
available, from which the most promising ones were selected for characterization. A high proportion of the bacteria
generated measurable amounts of protein: If only clones with correct reading frames are considered, 37, 72, 90, and
10% of libraries 10t, 105t, t, and 105, respectively, have detectable protein expression.20
Conclusion
The data reported are important, since they show that even sequences only 100 amino acids long, optimally designed,
have a chance under 109 of producing true protein folds. Even though all information known about the requirements to
generate secondary structures (binary patterning; length; N-caps; C-caps; optimal choice of amino acids) went into the
design, under natural aqueous conditions no examples of native folded proteins were found.The authors believe that a
few of the proteins generated have some kind of molten globule structure. However, they admit,
Although natural proteins, in general, have a distinct global free energy minimum that allows them to fold into one unique
structure, molten-globulelike properties probably lack a distinct global minimum, and thus do not have specific tertiary
structure.26
And a non-specific tertiary structure cannot be used for biological purposes, as an evolutionary starting point, nor be of
value which natural selection could fine-tune.Even a biological protein possesses a multitude of false folded states in a

local minimum energy state. However, a sequence which folds close to the native state is needed as an evolutionary
starting point.
The results imply that the proportion of sequences leading to a proper fold from among truly random polypeptides must be
very small. 109sequences 100 AA long were generated and the most promising ones were analyzed without success.
Random sequences would be orders of magnitude even less likely to produce native folds.
The proportion of polypeptide chains which generate native foldspart 4: reusing existing secondary sequences
by Royal Truman
Various experimental approaches to generate native-like protein folds from scratch have failed. This has prompted originof-life researchers to develop a new strategy. A large research team randomly combined gene sections able to generate
605 helices, 328 strands and 246 loops, which were found in stably folded biological proteins. Not all 108 clones were
examined in detail, but evaluation of the most promising candidates failed to identify any with native-like folds. These
experiments demonstrate the difficulty of generating proteins with all the necessary structural parts present harmoniously
to produce a stable native-like fold, and illustrates the immense difficulty for this to occur by chance.
In Part 3 of this series, we mentioned 1 that several research
groups have attempted to design novel artificial proteins based
on a specific target fold2-4 or binary patterns with remarkably little
success.5 We concluded that native-like folds were not formed,
an opinion shared by others:6
A more recent report details the use of cassettes of binary
patterned residues around known protein motifs that are
ultimately randomly assembled without fitting any designed
folding constraint. However, these latter methods have not yet
yielded stable, novel folds under physiological conditions.7
This prompted a research group6 to run an experiment in which it
semi-randomly combined sequences from real biological genes
that are responsible for the secondary structures found in the
resulting proteins. In this manner portions of proteins known to be
involved in secondary and tertiary structure would be present in
the novel proteins and likely provide the scaffold for novel folds.
The outcome
First, all sequences leading to alpha-helices, beta-strands, and loops found in 190 non-redundant protein present inE.
coli were identified. From these a library consisting of 605 helices, 328 strands and 246 loops was assembled, 8using
optimal stoichiometries. Carefully designed polymerase chain reactions linked these strands and loops together to
produce new genes, to ensure that the same orientation of the original helices and sheets found in E. coli would
result.The PCR fragments formed were submitted to agarose gel electrophoresis to remove lengths <300 bp 10 so that
proteins of 100 or more amino acids would result when expressed in E. coli. On average about seven fragments for each
clone were used in the artificial genes created. These were fused to a sequence which would produce a GFP 9 protein at
one end of the product translated from the gene. 10 The new genes were inserted into a plasmid (EGFP fusion vector
JG1)10 and these transferred into E. coli hosts.The modified E. coli were sorted using fluorescence activated cell sorting
(FACS), retaining those showing a high fluorescence (GFPuv fluorescence > 80 RFU). This fluorescence comes from the
small GFP folding reporter mentioned above, deliberately generated as part of each member in the library of the artificial
genes. In the resulting protein, high fluorescence may indicate this GFP portion is soluble and stable to proteolysis, 10 and
therefore potentially pointing to it being located in a folded environment.Of 1 10 8 cells sorted by fluorescence, 1.6
106 (1.6% of the total) were identified10 as high fluorescence, meaning they might have folded. This was too many for
detailed analysis, so 1,149 of the most promising clones were chosen for further analysis.Of the 1,149 clones, 70 were
selected on the basis of various criteria: 44 with very high fluorescence (>1,000 RFU); 22 with significant solubility
scores,11 of which fourteen were already part of the high fluorescence set; 12 and 18 clones lacking high fluorescence and
significant solubility scores but displaying 12 high Nickel-HRT assay scores (associated with protein solubility).It was
possible to obtain sufficient protein from seven of the 70 clones for further study. One was a duplicate and another was
omitted for further investigation since a nonsense mutation (which leads to a premature stop codon) had occurred.The
dye BisANS13 was added to each of the five clones, since its fluorescence increases upon binding to unfolded and molten
globule proteins. Fluorescence was measured at pH values varying 14 between 3.6 and 9.6. The changes seem to indicate
that the five proteins are undergoing pH induced structural changes, one trait of folded proteins.Next, ultraviolet circular
dichroism spectra were obtained14 for the five clones at wavelengths between 250 nm and 200 nm. This is a method used
to identify the presence of secondary structure, especially alpha helices. In one clone, no helical content was determined.
For the other four clones, a considerable amount of alpha-helical structure was identified, although the values deviated
dramatically from that expected based on the characteristics of the segments they were built from (which were designed
from natural proteins found in E. coli). Also disconcerting, the AGADIR program that predicts alpha-helical structure
predicts these four clones should have little to no alpha-helical character.Thermal denaturation of the four proteins was
determined by monitoring circular dichroism at 222 nm. The temperature was raised slowly and then lowered again
between 20C and 90C11 leading to near super-imposable folding and unfolding profiles for three of the clones (not clone
5.26).Finally, the researchers subjected the proteins of three clones (5.6, 5.26 and 5.31) to NMR analysis, a key method
to determine protein structures.
The chemical shift dispersion of the 1-D NMR spectra of proteins is a qualitative measure of protein folding and tertiary
packing of protein side chains.15
The NMR spectra resemble the results from a reference, unfolded protein (FKBP12), leading the authors to conclude that
The selected clones should therefore not be considered viewed as native-like proteins but rather molten globule like.15
Clone name

CD spectra

Predicted from library content

AGADIR

5.1

40

22

1.2

5.6

29

7.8

5.26

15

23

4.8

5.31

47

17

0.3

Table 1. Percent alpha-helical content predicted using three methods.14

Discussion
In this study, a large number of gene sequences known to produce secondary structures of properly folded biological
proteins were combined to generate novel artificial genes. Surprisingly, none of the carefully constructed and tested
artificial proteins showed the expected protein folds. As usual in science, though, when unexpected results occur, one can
often postulate a good reason, which may or may not be correct. In this case the authors suggest,
The lack of correlation between the fragment origins and the structure they appear to assume in the selected
polypeptides is not too surprising since it is well known that primary sequence is not the sole determinant of secondary
structure formation.15
Nevertheless, the authors stated15 that they have no explanation for the positive CD spectra that indicated the presence of
alpha-helices.The proteins generated by the synthetic/artificially constructed (i.e. designed) genes was only examined for
some of the 108 E. coli isolated. The results reported permit some estimates for the proportion of sequences which would
fold for the carefully optimized original library.We can assume, as the authors do, that of all the modified E. coli created,
those displaying GFPuv fluorescence of less than 80 RFU did not contain a synthetic gene which leads to a native-like
protein. Then, of the original pool, 0.16 are candidates for having folded properly.Of the 1.6 10 6 cells comprising
promising clones, 1,149 clones were selected. We will assume that the selection was random, and that no criteria, such
as strength of fluorescence to pick the best candidates was used. This is not clear from the paper, and would seem like a
foolish decision. Of this set, the 70 which displayed evidence they might contain folded proteins were selected: 70/1,149 =
0.061.Of the seventy best candidates so far, seven were picked. It is not clear how to interpret this: perhaps some of the
remaining sixty three were unsuitable. Possibly abnormal behaviour on the separation (Comassie-stained bands by SDS
PAGE after a single nickel chelating FPLC purification step12) column was observed. Of the seventy potential candidates,
none were native-like folded proteins, leading to a factor ranging between 1/70 = 0.014 and 7/70 = 0.1. In favour of the
naturalist view, well use the more generous value of 0.1.These three terms allow us to state that a library of proteins
carefully designed as described contained a proportion less than 0.16 0.061 0.1 = 1 103 folded proteins. The only
quantitative conclusion possible is that the proportion of native-like folds among random polypeptides must be many
orders of magnitude lower than 103.The popular hypothesis that random gene shuffling produced countless new
functional proteins is not compatible with these results. Based on detailed bio-informatic analysis from an evolutionary
perspective, Conant and Wagner reached a similar conclusion:
We have studied shuffling in genes that are conserved between distantly related species. Specifically, we estimated the
incidence of gene shuffling in ten organisms from the three domains of life: eukaryotes, eubacteria, and archaea,
considering only genes showing significant sequence similarity in pairwise genome comparisons. We found that
successful gene shuffling is very rare among such conserved genes.16
Conclusion
Random shuffling of pre-existing protein-domains appears not to be a mechanism to account for novel functional proteins.
Rather, protein-building elements need to be carefully organized together for a new protein to produce a native-like fold.
This all points in the direction of design, hence a designer.
The proportion of polypeptide chains which generate native foldspart 5: experimental extraction from random
sequences
by Royal Truman
This is the second to last of a six part series which critiques the most widely cited papers claiming that a high proportion of
random polypeptides naturally form native-like folds. Parts 5 and 6 might not seem like exciting reading, but they cover
the only known experiments we are aware of which look at random polypeptides. When Professor Robert Sauer from MIT
was contacted by the author questioning some published claims, no effort was made to defend his own work. Instead,
attention was drawn to the research by Professor Jack Szostak, which is discussed here.We find the claim that one out of
1011 random polypeptides in free nature would produce proteins reliably folded to be unconvincing. The artificial
polypeptides identified were much smaller than average-sized proteins, depend on the presence of zinc and ATP, and
lacked the rich secondary structures characteristic of biological proteins.
Figure 1. Strategy to estimate the proportion of random polypeptides with a
native-like fold. Outer circle covers all possible sequences of length n amino acids
(20npossibilities), a portion of which is sampled (a). Of these some show
indications of folding-related properties (a) from which the best candidates are
selected and mutated to generate new candidates (b). The best sequences
among a and b are members of the sampling space a + b. The best candidates
in b are used to generate better variants c and so on. The estimated number of
sequences of quality c compared to sampling size a + b + c gives an indication of
the expected proportion in the total population of length n (the outer circle).What
proportion of random polypeptide chains based on the twenty natural amino acids
would fold into native-like folds? Ideally, one would examine experimentally a
library of random sequences to answer this question. However, examining a vast
number of random sequences 150 AAs (amino acids) long (the average size of a
domain)1 or 300 AA (the average size of a protein) is not feasible should indeed
the proportion of those folding reliably be very small. Experimental methods such as CD (circular dichroism) spectra,
NMR, denaturating spectra, X-ray crystallography, and so on, are very time-consuming.To our knowledge only the series
of experiments pioneered by Professor Szostak, winner of the 2009 Nobel Prize in Physiology or Medicine, have
addressed this question empirically. During the last decade other researchers had made contributions to this original work,
and some have concluded that a fairly high proportion of random polypeptides can fold stably. Lo Surdo states that
It had been previously suspected that the probability of a stable fold arising at random is extremely small (argument
based on the complexity paradox that the disparity between the many potential protein sequences and the relatively few
different structures known). However, we have confirmed that functionally directed in vitro evolution from random
sequences can generate a novel fold with a tailored function. In addition, such folds may display the key features of

naturally evolving proteins.2We shall examine in this paper two key studies and two more in the next part of this series to
see if this conclusion is warranted.The proportion of native-like folded polypeptides is expected to decrease as chain size
increases, due to undesirable interferences and decreasing solubility. For this reason, these experiments typically work
with small chains of n = 80 AA (or shorter). The potential sequence space therefore covers 20 n = 2080, or 1
10104 alternative chains. Through clever search strategies, shown in figure 1, the researchers iteratively narrow the search
space to promising regions in the sequence space.
First study
A method to create large libraries of mRNA sequences lacking frameshifts, stop codons and internal translation initiating
events has been developed3 and was combined with the experimental technology described under Methods below.
Typically about 6 1012 different sequences are rapidly examined in these experiments based on their ability to bind to
immobilized ATP in a separation column. The researchers reasoned that strong and selective binding to ATP might be a
relevant criteria to identify properly folded proteins. Noteworthy is that ATP binding proteins are found in all major enzyme
classes.4
Figure 2. Sequences very different from
that producing a native-like folded state
might be selectable and steadily
improved (assumption A); or distant
sequences with properties suggesting
folding-state
potential
might
be
evolutionary dead-ends, offering no
opportunity to be selected to produce a
native-like folded state. (Based on work
from Axe5).Since 1012 << 2080, the
maximum
number
of
alternative
sequences 80 AA long, no duplicate
sequences were likely to have been
generated. After removing the sequences
which did not bind to immobilized ATP, a
subset of promising candidates were
isolated (members a in figure 1). Errorprone PCR amplification generated new, similar variants (region b in figure 1) likely not to have been present among those
in sampling space a. From these the best candidates were isolated (sequences b, figure 1). This process was repeated
to select other polypeptides with yet stronger affinity for ATP, sequences c, and so on. In this manner, promising portions
of the overall search space can be efficiently identified.The key insight is that sequences b were the result of an
intelligently directed search, and have a high probability of including sequences able to bind better to ATP than the best
from a could. They are members of a much larger pool of candidates, a + b (figure 1).The search strategy assumes that
sequences very different from those producing native-like folded proteins could be initially identified by natural selection
and thereafter improved upon. This view is illustrated as assumption A in figure 2. However, others 5 believe that chains
significantly different from native folds are useless as an evolutionary starting point, since there are countless energy
minima which have no relevance to the protein folding state (assumption B in figure 2). This matter is not addressed in
these series of studies, although clearly the authors simply assume A to be true.
Figure 3. In vitro selection and
amplification of mRNA-displayed
proteins. The DNA library was
constructed with the following
elements:
a
T7
promoter
sequence, which is not translated
but is required by the T7 RNA
polymerase8 used to perform the
translations; the following TMV
(tobacco mosaic virus) sequence
is also not translated and serves
as a translation enhancer. An
FLAG-tag follows the start codon,
which consists of amino acids
DYKDDDD to permit affinity
chromatography.9 The next block
consists of 80 contiguous random
non-stop codons, followed by a 6
histidine tag,10 used for affinity purification.11 The affinity tags permit the mRNA with frameshifts and internal starting
events to be eliminated. The mRNA transcribed is ligated to a linker which includes 27 adenines and puromycin, which
results in the fused mRNA-puromycin-protein product. Purification and reverse transcription generated a mixed RNA-DNA
chain attached to puromycin-protein. The library was incubated with bound ATP-agarose and placed in a separation
column. Fused proteins which dont bind to ATP were washed away, and then the proteins which did bind were eluted with
free ATP.Now to the specifics. In the seminal paper, Keefe and Szostak 6prepared a library of 6 1012 non-redundant
random proteins 80 AA long, designed to avoid random stop codons so as to generate the maximum variety
possible.7 The random 80 AA region was flanked by short affinity tags to permit subsequent purification of the proteins,
figure 3.The eluted fractions were amplified by polymerase chain reaction (PCR). The DNA produced was used to
generate another library of mRNA-displayed proteins and the scheme repeated for a total of eight rounds (figure 3).
During the first round about 0.1% of the proteins bound to ATP in the column, and after eight rounds of selection the
amount had increased to 6.2%. The low level of ATP-binding of the best sequences isolated so far (6.2%) is probably due
to the conformational heterogeneity of the sequences isolated 7: the same sequence can fold in many non-native-like
manners.The authors cloned and sequenced 24 of the members after the eighth round and found that most could be
grouped into four sequences (figure 4A). These were unrelated to each other and to any biological sequences known.
Representatives of each of the four families (AD) were examined. Between 5% and 15% of these fused mRNA-displayed
proteins bind to immobilized ATP (see last column of figure 4A). Note that far more of the chain binds to ATP if still

attached to its mRNA (second to last column of figure 4A), which was not discussed in the publication. This fact illustrates
that organic molecules such as ATP can easily bind to many classes of bio-chemicals, not just to polypeptides.
In order to optimize binding to ATP, the improved library was mutagenized three times using mutagenic PCR amplification,
with an average rate of 3.7% per amino acid for each round. After six additional rounds of amplification without
mutagenesis and in vitro selection followed by elution with ATP, the proportion which binds to ATP in the column rose to
34%.
Of the 56 clones sequenced, all were derived from a member of the family B (figure 4B).

Figure 4. A) Consensus sequences of initially selected protein families after round 8; B) Sequence data of mutagenized
re-selected proteins after round 18, free protein; C) Truncated versions of DNA-tagged protein 1819; D) Deletion
analysis of clone 1819 fused to MBP protein.12 Click here for a larger image.
Comparing to the original family B revealed that the mutations generated had produced four which bound effectively to
ATP (present more than 39 times in the 56 sequences examined), see second row from the top in figure 4b, labelled
18predom. In addition, 16 other substitutions generated were also selectively enriched (present more than 4 times
among the 56 examined), see second row from the top in figure 4b, labelled 18select. This suggests that different amino
acids distributed throughout the protein sequence might be interacting with various portions of ATP.
After selection in round 18, eight individual proteins were chosen randomly. As free protein, the proportion which binds in
the column (and was then eluted with free ATP) varied from 5% to 40%.12

Figure 5. Dissociation constants, Kd, of

ATP and similar analogs bound to
clone 1819.17 The
innermost
phosphate is designated by a, the
middle one by b and the outermost one
by g. The arrows show the group
removed and the resulting Kd value.
The four proteins that bound best to
ATP were expressed in E. coli as fused
proteins to maltose-binding protein
(MBP), which is often done when the
proteins are not soluble enough to work
with conveniently. The clone which
bound best was labelled 1819 (figure
5) and had a dissociation constant of
Kd = 100 nM for ATP at 4C and 1:1
stoichiometry. However, gel filtration
indicated
that
65%
of 1819 is
monomeric and the rest binds in higherorder aggregates. Only the monomers
bond to ATP.Unsurprisingly, selection
for binding to ATP did not improve
binding to various other bio-chemicals.
Chemical analogs to ATP, such as
CTP13, GTP14, UTP15, and ITP16 bound
far less effectively to protein 18
19 (figure 5). Furthermore, removing
one, two or three phosphate groups led to molecules which bound less effectively to ATP (figure 5). Removing either of
the 2 or 3 hydroxyl groups also reduced the amount of binding.The minimal region for ATP binding of clone 1819 was
explored by deleting portions of the protein. The fraction of the protein which bound to the ATP in the column, relative
to 1819, is shown in the last column of figure 4C. Removing large sections from 1819 improved the binding: those
residues were apparently interfering with binding to ATP. A core domain of 45 amino acids was found to be sufficient to
bind efficiently.12 Significantly, the deletion shown in the fourth row from the top in figure 4C had a ten-fold deleterious
effect on ATP binding. This drew attention to the fact that aCXXC motif, involving two cysteine amino acids, had been
destroyed.In another experiment, variant 1819 was fused to MBP protein and the dissociation constants, K d, were
measured after removing various fragments (figure 4D). Loss of affinity was observed for sections removed in the Nterminal portion (left-hand side of the sequences in figure 4D, note the last three rows). The authors propose that regions
surrounding the important core which interacts with ATP might stabilize its structure or that additional amino acids might
also interact with ATP.12 This observation is consistent with the accepted view that domains require a significant number of
residues to be present. The CATH database of domains gives an average domain length of 159.5 AAs and the three
smallest have 13, 14, and 17 AAs (out of 128,688 entries). These three smallest domains consist of merely a single tiny
alpha coil.18At the large end,
the CATH database revealed
18 domains having over 500
AAs.18
Figure 6. Crystal structure of
clone 1819 with
the
coordinates provided in the
pdb
database.
The
representation was created
with RasTop 2.221 A) Using
protein
identifier
1UW1.22 The small circle
shows the location of the
chelating
zinc
atom. B) Using protein identifier 1UW1. Notice that the researchers reported a dimeric structure in the crystal state. Two
ADP molecules were found in the crystal structure. C) Using protein identifier 3DGN in the pdb database.22 The crystal
was formed in the presence of high concentration ADP to ensure its integration.
The authors concluded, on the basis of their experiments that
The frequency with which ATP-binding proteins occur in sequence space can be estimated from the observed recovery of
four such proteins from a non-redundant library of 6 x 1012 random sequences. On the basis of the average behaviour of
the proteins isolated before mutagenesis, only about 10% of the potentially functional sequences present in the first round
would be expected to generate correctly folded active proteins and thus survive to be amplified We therefore estimate
that roughly 1 in 1011 of all random sequence proteins have ATP-binding activity comparable to the proteins isolated in this
study.Four years after this work 3 was published, the crystal structure of protein 1819 was elucidated2,19 (figures 6AC)
independently by two groups, and it was reported 20 to overlap quite well with a small portion of domain ARG-GAP of a
biological protein (figure 7).It must be emphasized that ADP, and not ATP, was found co-crystallized with clone 1819,
even when the presence of ADP was rigorously prevented.
Second study
In a similar study23 also performed in Dr Szostaks lab (see Methods below), polypeptides able to bind to
streptavidin24 (figure 8) were identified from a semi-random library of 6.7 1012 members.

Figure 7. Crystal structure of 278-aminoacid domain ARG-GAP with which a small

portion of clone 1819 shows similarity. Xray coordinates deposited in the pdb
database under identifier 1DCQ.22 Protein
displayed with RasTop 2.2.21
The library had been designed using short
eleven-amino-acid cassettes which were
then concatamerized together. By design,
44% of the incorporated cassettes encoded
a peptide polar/non-polar pattern compatible
to forming amphipathic alpha-helices and
45% to form beta-strands; 11% were not
patterned. Similar experiments to that
described above were performed6 except
that mutations were not deliberately
generated. After seven rounds of selection
for streptavidin and amplification with PCR,
20 different sequences were observed and
analyzed. All were found to have frameshifted, destroying the intended patterns
which would lead to secondary structure. This experimental misfortune does provide us with another example of random
sequences, however. The mutations and subsequence strong selection were due to the generation of one or more of the
amino acid pattern HPQ,27 which is known to bind strongly to streptavidin.The clone with a single HPQ pattern which
bound most strongly to streptavidin was selected for additional study. Removing more than half of the residues led to a
38-amino-acid species with slightly higher binding affinity, but additional deletions lowered the affinity.
The authors provide two possible explanations for these observations:
The flanking amino acids might stabilize the active conformation of the HPQ tripeptide motif.
Several distinct peptide elements could be interacting with
different parts of the streptavidin.
Figure 8. Secondary structure of a streptavidin monomer
with bound biotin.25 The secondary structure of a streptavidin
monomer is composed of eight anti-parallel -strands, which
fold to give an anti-parallel beta barrel tertiary structure. A
biotin26 binding-site is located at one end of each -barrel,
which has a high affinity as well as a high avidity for biotin.24
Evaluation of the first study
The B-family proteins possess a pair of CXXC motifs
(two Cysteines separated by two amino acids, X)
(highlighted in figures 4AD) and the presence of a single
bound zinc, revealed by atomic absorption spectroscopy.
The proteins with good ATP-binding properties, including the
best-binding
clone 1819, require zinc.
The
authors
suggested that these proteins bind to ATP with a folded
structure nucleated around, or stabilized by, a Zn 2+ion
coordinated to the four invariant cysteines of the CXXC
sequences.17They reported that Mg2+ and other cations
were not found to be suitable. Note that
Zinc was not added to the selection buffer, but is present at
about 10 micromole mammalian blood, from which the
reticulocyte lysate used for mRNA translation is prepared.17
The presence of zinc was accidental and no proteins were
reported which bound well to ATP which lacked this metal.As reiterated in a later paper, 28 the estimate of 1 out of
1011 involved sequences for whichAll of the variants examined required high concentrations of free ATP in order to remain
stably folded and soluble.29They add, the ancestral and binding optimised proteins formed visible precipitates during
the first 24 h and became completely insoluble after three days. Surprisingly, the core domain of the divergent protein 18
19 also formed a visible precipitate after 24 h and was almost completely insoluble by three days. 30In the process of
preparing 1819 for X-ray diffraction, large amounts were expressed in E. coli and then purified to homogeneity.
Apparently the natural mixture of 1819 conformations was not used:
Our structure determination targeted the functional core of ANBP [ANBP is an abbreviation for artificial nucleotide
binding protein]. We aimed to optimise the production of a protein with a single conformation by screening different ANBP
constructs, thus identifying N- or C-terminal residues unnecessary for either folding or ligand binding. 31The implication is
that the secondary structure shown in the crystals generated seem not to represent the conformation formed in the
solvated state at biologically relevant temperatures.Only residues 773 could be identified in the electron density map,
and residues 710 on the N-terminus were poorly defined. 32 The X-ray data revealed considerably smaller and less
secondary structure than observed in CATH biological domains classified as Alpha Beta. 33 The crystals were developed at
100 K, a very low temperature, far below the freezing point of water. These facts suggest that under ambient conditions
very little, if any, secondary structure is actually present. It is well known that the crystal state of a protein sometimes does
not reflect its topology in solution.34 Incidentally, the N-terminal alpha helix had been provided by amino acids present in
the FLAG sequence, and therefore cannot be legitimately considered representative of a random sequence. Removing
this alpha helix from figure 6B leaves still less secondary structure, considerably less than found in virtually all native
proteins.

Figure 9. Human alpha alchohol dehydrogenase, showing zinc

chelated to four cysteine residues (C symbol in the figure). There
is no evidence the zinc atom is producing the secondary
structures.
Figure
generated
using
jmol
from:
www.rcsb.org/pdb/explore/jmol.do?stru
ctureId=1HT0&bionumber=1.
Figure 6A shows how the four cysteines binding to zinc hold the
secondary structures together. However, this feature is not
observed in biological proteins for which a similar zinc motif has
been found (figure 7): the region held together by the zinc atom,
while surely of biological value, is not responsible for the folding
of the large domain. Indeed, the tiny zinc-chelated portion is far
removed from the alpha coils and beta sheets, best seen by
rotating the molecule using a protein viewer such as RasTop.A
well-known protein motif is known as the zinc finger.35 These are
very small structures found on about 500 proteins in humans, 36 which usually consist of two cysteines located in a short
strand or turn region followed by an -helix which contains two His ligands. The key observation is that although zinc is
used with these biological motifs, the zinc fingers are not responsible for the folded structure of the proteins in which they
are found.Figure 9 illustrates another example where zinc is chelated to four cysteins in a representative biological protein
with rich secondary structure. Notice how the alpha coils and beta sheets are very distant from the chelation site, so that
the native fold could not be driven by the presence of zinc.However, in the synthetic protein described, a zinc atom seems
to be entirely responsible for what little structure is present in the protein.
Analysis of the second study
We offer the following comments about this second study:
For the optimized 38-amino acid polypeptide, 13% did not bind even at high concentrations of streptavidin. Real proteins
involved in enzymatic catalysis are highly specific and could not fail to bind to their intended ligands under these
experimental conditions.Removing more than half of the polypeptide led to slightly stronger binding, which reflects the
potential for larger chains to offer steric interference to protein ligand interactions. This reinforces the fact that as chain
size for random polypeptides increases, the probability of deleterious interferences and insolubility increase.Replacing the
HPQ pattern with a similar HGA reduced the extent of binding by a factor of a thousand. 37 The special HPQ pattern was
almost the whole cause for binding to streptavidin. 38 This reinforces the need to be very close to the final and correct
sequence before natural selection can sense the genes existence.
Evaluation of these studies
As reported in a later contribution, Unfortunately, biophysical characterization of the selected ATP binding proteins proved
impossible due to poor solubility.39 For these reasons, laboratory tests necessary to characterize the secondary and
tertiary structure, such as CD spectra and NMR, were not available when the first study was published. 6Analysis of the
data published until 2004 suggests the following interpretation. The fortuitous contamination with zinc has led to a
chelating site, thanks to the presence of CXXC motifs. Some of the isomers were thereby able to offer a small surface
which can be moulded in the presence of ATP to permit some as of yet uncharacterized interactions. Notice how ATP only
seems to interact with a single tiny beta sheet, see figure 6B. Selection for ATP did not generate a microenvironment on
the protein which was also conducive to strong binding of ATP chemical analogs such as CTP, GTP, UTP, and ITP. This is
not particularly surprising, given the selective protocol.Removal of portions of the protein and improved binding upon
mutating various positions need not indicate multiple interactions, the strength of which has been improved, but rather
the removal of disturbing amino acids which can lead to insolubility, entanglement, intermolecular interactions and such
features deleterious to ATP binding.Often the structure of a protein in its dissolved condition under natural conditions and
temperature is very different from the structure elucidated after crystallization using X-ray diffraction. The process of
crystallizing clone 1819, at very cold temperatures, can select the particular conformations which are particularly stable.
We suspect that the minimal secondary structure identified (figure 6B) is an artefact of the crystallization. Chelation with
zinc forces several amino acids into close proximity, facilitating creation of alpha coils and beta sheets under biologically
irrelevant conditions.The limited data available supports this view. In the next part to this series, we draw attention to later
work in which the authors discovered, unexpectedly, that CD spectroscopy at biologically relevant temperatures denied
the presence of secondary structure. In addition, we must draw attention to the fact that in the absence of ATP suitable
crystals were not formed. It appears that not only zinc, but also ATP is responsible for moulding clone 1819, since
otherwise it would not crystallize. The evidence is clear that this protein alone cannot produce a native-like fold under the
above water-freezing temperatures. In a later paper we read,Unlike many naturally occurring proteins, protein 1819
requires high concentrations of free ligand in order to remain stably folded and soluble .40This fact was not mentioned in
the earlier work, nor quotes made subsequently of the estimated proportion of random peptides with native-like folds.We
welcome these kinds of studies to test the plausibility of our models, whether one believes in creation or evolution. We
have some suggestions for those involved in this kind of work.Justify the belief that highly specific binding to a ligand such
as ATP, but not analogs, can permit an estimate of random proteins which produce a native-like fold. Why must a nativelike fold be able to bind to this particular molecule? Might not additional ligands also bind strongly to other members of the
sampled portion? Why must a strong interaction at a judiciously constrained portion of the protein have any relevance to
native protein folding?Generate a large number of variants similar to clone 1819 and then determine whether there is
evidence for secondary and native structure in the absence of ATP and zinc.Repeat the studies based on selection for
ATP or another ligand, using n = 40 200 amino acids, to determine whether the proportion (not absolute number) with a
given level of binding follows a pattern. If larger random-sequence proteins become more chaotic, multimeric and
insoluble, and prevent binding to a ligand, then native-like folds will be less likely.
We conclude that the proteins studied do not offer evidence of native-like folding.
We offer some objections for the suggested proportion of random polypeptides leading to stable folding, 1011, for
evolutionary modelling purposes.The data suggests that ATP is necessary, but once bound to the crystallized clone 18
19 was immediately destroyed (hydrolyzed to ADP). Since ATP has negative charge, it is clear that it binds to the basic
amino acids, as seen in 1819 and other crystal structures. ATP has many hydrogen and oxygen atoms which are able to
form numerous H-bonds. And adenine in ATP can form planar stacking interactions with aromatic residues. No wonder
that it binds somewhere! Zinc can bind to cysteines and other residues41,42 such as the side chains of aspartic acid,
glutamic acid and histidine, leading to local structures able to bind bio-molecules. These facts illustrate the vast number of
destructive possibilities available along an evolutionary path towards forming a properly folded protein. And using an
example which indiscriminately destroys the eminently valuable energy carrier, ATP, is a questionable argument for how

proteins could arise naturalistically. It appears rather to highlight constraints which must be overcome.It would be
extremely rare for a random polypeptide sequence to reliably chelate with zinc in a free ocean. Seawater has a
concentration of only about 30 ppb zinc43 but thousands of times greater (about 75 ppm) in the earths crust.43 Others
report an even lower concentration in seawater, such as 0.65 ppb, and rivers containing between 5 and 10 ppb
zinc.44 The higher concentration in rivers reinforces the fact that the source of seawater zinc was erosion of the crust, and
supposedly three or four billion years ago the concentration of zinc in a primitive ocean would have been far lower still. In
addition, zinc is normally not chemically freely available to chelate with polypeptides, but is found firmly associated with
other base metals such as copper and lead in ores, and binds well with sulphides instead. 41,45
If metal chelation with random polypeptides had played a key role in the origin of proteins, then it would be reasonable for
metals such as aluminium and iron, on average about a thousand times more abundant on Earth 46 than zinc, to dominate
protein chemistry. By similar reasoning, silicon would be expected to play a far more important role than phosphorous,
being
three
hundred
times
more
abundant, 46 but
the
opposite
is
true.
In a standard reference book on proteins we read,
Surprisingly elements such as aluminium and silicon that are very abundant in the Earths crust (8.1 and 25.7 percent by
weight, respectively) do not occur in high concentration within cells. Aluminium is rarely, if ever, found as part of proteins
whilst the role of silicon is confined to biomineralization where it is the core component of shells. 47
This is the opposite of what a naturalistic origin for proteins predicts.
Figure 10. Structure of puromycin.49
The average biological protein is four times larger than 80 AA, posing some
difficulties.
A 2006 study estimated that about 10% of human proteins (2,800) potentially
bind zinc.41 Few, if any, biological proteins are composed of domains, with
about 80 AA of them all containing zinc. The 1011 proportion proposed for
stable folding was based on a very small domain, and we need to extrapolate
to a reasonable-size protein by using the geometric mean of the possibilities.
For example, if in the presence of zinc, 1011 sequences were to produce a
native-like fold, but without it the probability were only 10 15, then the
probability for a full protein with four domains might be more like [(10 11) x
(1015) x (1015) x (1015)]1/4 = 1.0 x 1014per domain, or (1.0 x 1014)4 = 1 x 10
56
for the whole protein.Claiming that life could have initiated with much
smaller proteins, like 80 AAs, than observed throughout nature today
probably does not simplify matters for an evolutionist. Proteins must interact
with several other biochemicals, including other proteins, to be able to
perform useful processes. If a few hundred multifunctional proteins (of size
300 AA on average) are necessary for autonomous life, 48 this number must
be several times greater if the numerous functions per protein need to be
handled by multiple simpler versions. Furthermore, the functional regions of
extant proteins are held together in suitable geometries, facilitating
interaction with the multiple intended partners. This would not be true initially
for simpler disconnected proteins.We conclude that the proteins studied do
not offer evidence of native-like folding and that the claim that one out of
about 1011 random sequences 80 AA long form native-like folds is not well
supported.

Figure 11. Sequence of steps carried out at a ribosome

which lead to protein elongation.50

Figure 12. Reaction forming a peptide bond between the

amino acids at the A and P sites of a ribosome.51
Figure 13. Proposed mechanism for the
RNA-peptide
formation
on
a
ribosome. A) Translation occurs towards the
attached puromycin, labelled P. B) When
the ribosome reaches the end of the mRNA
it comes in contact with the deliberately
attached stretch of 27-adenines which
causes translation to stall. The puromycin
now has time to enter the A site of the
ribosome and accepts the nascent peptide.
The exact mechanism is not yet
known.53 C) Once formed, the mRNApeptide fusion is purified from the mRNA
which lack puromycin and from free
peptides using affinity chromatography.

Slightly under 1% of the input mRNA is converted to the fusion product, 54 producing a library of about 1012 fusions in a 10ml translation reaction.53
Materials and methods used in these studies
Puromycin (figure 10) is an antibiotic that mimics the aminoacyl end of tRNA and terminates translation of mRNA. Figure
11 shows how codons are used by a ribosome to identify the next amino acid to be added to a growing protein. During the
last step the protein is located at the P site of the ribosome, and at that time puromycin enters the A site instead of the
tRNA which corresponds to the next codon.To further clarify, the chain-growing reaction which forms a new peptide bond
between the amino acids at the A and P sites is shown in figure 12. In nature, puromycin interferes at the A site by forming
a stable amide linkage to the nascent protein using its free amino group.In the methodology developed to create mRNA
display proteins,52 the puromycin is covalently attached to the mRNA of interest and remains so after the mRNA and its
newly translated protein is released from the ribosome (figure 13). In this manner the mRNA and resulting protein remain
physically linked.Therefore, isolated proteins which bind to ATP in a column also provide their corresponding mRNA.
Copies of those mRNAs can be made using PCR and they can also be deliberately mutated to generate similar variants in
the hope of finding better proteins.
The proportion of polypeptide chains which generate native foldspart 6: extraction from random sequences
by Royal Truman
A series of experiments designed to estimate the proportion of random polypeptides displaying minimal protein folding are
evaluated. Candidates were isolated through binding to ATP, exposure to a denaturant, rounds of in vitro selection and
error-prone PCR amplification, followed by removal of regions which might interfere with folding. We conclude that the
results from circular dichroism spectras and irreversible melts deny the presence of native-like folding in solution under
ambient conditions, but rather existence of an artificial zinc-nucleated structure lacking canonical alpha helices or beta
strands. There is no reason to believe the polypeptides identified could serve as an initial evolutionary point towards
biological proteins with native-like folds.We point out that although the true proportion of minimally folded proteins among
random polypeptides about 300 amino acids in length remains unknown, the current experiments indicate the upper limit
must be considerably less than one out of 1011.
Figure 1. Relationship between protein
sequence and folding ability. Sea level
represents the minimum stability and other
requirements before natural selection could
improve on an initial conformation.
In the fifth part of this series on estimates of
random polypeptides which will produce
native-like folds, we examined1 some
experiments initiated at Harvard University
by Professor Szostak, winner of the 2009
Nobel Prize in Physiology or Medicine. To
our knowledge, only these experiments
have attempted to determine experimentally
and quantitatively the proportion of random
proteins which fold, beginning with random
sequences built from the twenty natural
amino acids. This work merits careful
analysis.Other studies212 have attempted to
generate de novo proteins based on rational
designs to restrict the size of the search space. These have not permitted a quantitative estimate of the proportion among
random protein sequences.
Study number three
The first two series of studies were presented in part 5 of this series 1 and the first one was re-examined13 by lead author
and Nobel Prize winner Szostak due to some details which had not been communicated before:Unfortunately,
biophysical characterization of the selected ATP binding proteins proved impossible due to poor solubility. 14Although the
conclusion that about one out of 1011 random polypeptides would produce a stable fold had been disseminated through
the earlier publication,15 in this later paper the authors admit that poor solubility, not mentioned before, is a significant
detail:This observation led to the question of whether protein sequences isolated from unconstrained random sequence
libraries could be evolved to adopt a folded state of reasonable stability or whether most such proteins might represent
examples of evolutionary dead ends.13This is a very good insight. Mutating polypeptide sequences with some degree of
thermodynamic stability may often lead to local energy minima and not a true protein native-like fold. Figure 1 illustrates
how the starting point must be reasonably close to the sequence of a native-fold if it is to serve as an evolutionary starting
point. Such sequences are represented as the top peaks in the fitness landscapes in figure 1. Proteins with insufficient
stability and other requirements would be figuratively under water and not visible to natural selection. And irrelevant
candidates could be viewed as isolated from the mountains, separated by water.The experimental details 12 were
explained in part 5 of this series 1 and wont be repeated here. A new experiment was designed 13 to isolate more soluble
proteins by systematically selecting those which bind to ATP in the presence of a denaturant (GuHCl), which would
facilitate opening into a more linear polymer. Clone 1819 in the earlier work had been found to be sensitive to
denaturation (50% loss of ATP binding in 1 M urea),14 which suggests that a stable, native-like fold had not been formed.
The output from round 17 in the study reported earlier 1,15 was now used to create a pool of mRNA-displayed proteins, by
performing six rounds of in vitro selection and amplification, without deliberately creating additional mutational variants.14
The mRNA-displayed proteins were incubated with ATP-agarose beads in the presence of GuHCl, and the denaturant
concentration was increased from 1.5 to 3 M to ensure that less than 10% of the input from each round would remain
attached to the ATP-agarose beads, which became the starting point for the next round of selection. In the absence of
denaturant (GuHCl), 67% of the material after round six bound to the ATP column, compared to about 35% for the starting
sequences in round one.14Ten sequences from the output of round six were analyzed. Of these, six remained soluble even
in the absence of ATP.18 These six were examined in more detail after cloning as maltose-binding protein (MBP) fusion
proteins with a thrombin-cleavable linker which permits removal of the MBP portion. These six proteins did not aggregate

as easily as the sequences isolated from the earlier work,1,15 during which only binding to ATP had been optimized with no
consideration to ensuring they would remain soluble. From the six most soluble sequences, the one binding best to ATP
(i.e. having the lowest dissociation constant) labelled B6 was selected for detailed studies. Unfortunately, gel filtration
analysis showed that 35% of this material formed undesirable higher-order aggregates.19Portions of the protein chain
were removed systematically by the authors from sequence B6, while ensuring that the two CXXC (Cystein-X-X-Cystein,
where X is any residue) portions necessary for chelation with zinc remained intact. The intention was to eliminate
deleterious regions that might nucleate aggregation.19 This effort was successful, since deletion of the residues at
position 7482 resulted in a protein that was over 98% monomeric according to gel filtration analysis. 19 Furthermore, this
sequence led to a two-fold improvement in binding to ATP.Further deletions led to a 62-amino-acid-core ATP binding
sequence, labelled B6-62, with comparable affinity for ATP and an elution profile consistent with monomeric
protein.Analogous to the results from the earlier work,1,15 the MBP:B6-62 fusion protein has almost no affinity for ATP
analogs CTP, GTP, UTP, ITP and 2-chloro-ATP. Sequential removal of gamma, beta and alpha-phosphate groups in ATP
successively raised the dissociation constant,20and so did modification of the 2- or 3-deoxy portions of the ring. These
observations led the authors to believe the artificial protein was generating a folded structure amenable to interaction with
ATP but not to other substances.Species B6-62 was further analyzed. Shifts in tryptophan fluorescence in the presence of
increasing concentrations of denaturing GuHCl, with and without ATP, showed a sigmoidal curve consistent with a single
cooperative transition between the native and denatured states. 20This corresponds to an overall stabilization by 1 nM
ATP of 2.7 Kcal/mol.21Proton NMR spectra was not consistent with a random coil conformation, and implied some kind of
ordering of the protein. The HSQC spectrum 22 revealed about 45 well-resolved peaks, which the authors considered
consistent with a native-like structure. Unlike clone 1819, the folding-optimized version seems to be
monomeric.23However, the results from circular dichroism spectroscopy (CD) were completely unexpected! There was no
evidence for the secondary structure features (alpha helices and beta sheets) used to form all natural, native-like protein
folds.
Analysis of study number three
Is B6-62 truly representative of native-like proteins? The authors provide the same conclusion we reached:Our initial data
already suggest that this protein has an unusual metal nucleated structure lacking canonical alpha helices or beta
strands.24In a later paper the thermal stability of the folding-optimized protein and several variants were monitored by
changes in circular dichorism. They report thatAlthough each of the proteins examined gave sigmoidal curves consistent
with a single transition between the native and denatured states, none of the melts were reversible.25
The CD spectra of clone B6-62 provides the decisive answer:
This spectral signature is not consistent with standard alpha-helix (two negative bands near 222 nm and 208 nm and a
strong positive band near 190 nm), beta-sheet (negative band near 217 nm and positive band near 195 nm), or random
coil (strong negative band at 200 nm). One possible explanation for this unexpected CD spectrum is that the structural
architecture of this protein is defined by loops collapsed around a zinc-nucleated core. CD spectra taken in the presence
of increasing concentrations of ATP did not affect the magnitude or appearance of the 226 nm signal. In contrast, the CD
spectrum of B6-62 denatured with 4 M GuHCl changed dramatically, becoming consistent with a random coil peptide, as
expected for an unfolded protein.21
The CXXCXnCXXC zinc-binding site, without which nucleation cannot occur, is not used by any biological proteins known
which bind nucleotides.26 This renders the relevance of the studies discussed here of questionable use in evaluating the
probability of biological proteins having arisen by chance. It is not surprising that a molecule like ATP offers many binding
interaction possibilities with polypeptides, via numerous H-bonds possibilities, or by sandwiching the adenine nucleobase
between two protein aromatic side-chains26 in a stable pi-stacking interaction. The phosphate groups could coordinate
with any number of polar side-chains.One must not forget where these improved sequences came from. 15 Clone B6 was
not present in the original library of 6 x 1012 random sequences. After eight rounds of selecting the mRNA-displayed
proteins which bind most strongly to ATP, three consecutive rounds of mutagenic PCR amplification with an average
mutagenic rate of 3.7% per amino acid for each round
had been carried out.Recall that B6 had to be
judiciously modified to eliminate interfering portions of
the chain and to create a soluble variant. In arriving at
the denaturing-resistant optimized version B6-62,
twelve amino acids along the 62 residue sequence
had to be substituted. Of the 2062 = 5 1080 possible
sequences of 62 residues, a miniscule proportion
would have this folding strength. Since the larger 80
AA included portions interfering with folding, it is
difficult to justify the claim that 1011 would be able to
fold properly,27 meaning in a manner relevant for
biological purposes. And it is far more difficult to imply
that this proportion is representative of all proteins,
which on average are several times larger than 80
AA.
Figure 2. Crystal structure of clone 1819, displayed with Jmol Viewer,
Study number four
In follow-up studies28,29 a few years after the work summarized in study number one, 1,15 clone 1819 was further optimized
for improved folding stability. The pool of protein 1819 was amplified serially to generate more variants, with error-prone
PCR and a target mutagenesis rate of 3.5% per amino acid position. 30 The experimental methods match those described
in study number three, above, in this paper. Selection for folded variants was performed in the presence of a denaturant
(GuHCl) on an ATP-derivatized affinity resin. After five rounds the amount of protein which bound to ATP agarose beads
had levelled off to about 40%. 30 However, one would expect about 100% of biological ATP-binding enzymes to remain
bound.This work was motivated by the wish to show suitable proteins could arise naturally without being generated by the
genetic code. The first sentence of the paper states,Primordial enzymes presumably evolved from pools of random
sequences, but it is not known how these molecules achieved catalytic function.31We read further down the page, over 3
billion years ago when primordial proteins first appeared on the primitive Earth. These and other authors repeat
elsewhere, Presumably the first proteins arose from pools of random sequences.32 Given the difficulties in the RNA
World Hypothesis,33,34 it is not surprising an alternative model, that the necessary proteins self-assembled before the
genetic code became available, remains popular.

Figure 3. Crystal structure of clone 1819, displayed with Jmol

Viewer,. The circle shows the location of the chelating zinc
atom.
The optimized version of clone 1819 was called DX (double
mutant protein) and differed from the progenitor by two amino
acids.35 The structure, when crystallized, is shown in figure 2
and includes ADP. The crystal structure reported by two
research groups28,36 for 1819 were essentially identical and
also included ADP (figure 3). This was a surprise, since ATP,
and not ADP, had been present with the protein during the
crystallization process (ADP is ATP with a phosphate group
removed).The authors deduced that in the crystal state the phosphate of ATP forms a strong intramolecular hydrogen bond
with the 2-OH on the sugar ring, lowering the energy of
activation to hydrolyze ATP to ADP. The crystal structure of DX
suggested that the -phosphate of ATP is stabilized with
hydrogen bonds to three amino acid side chains40 (Tyr43, Lys34
and Arg41), held in a position which facilitated hydrolysis to
ADP. To test this theory, a variant labelled Y43F was created by
mutating the tyrosine at position 43 to phenylalanine, which led
to a crystal structure which did not hydrolyze ATP (the
coordinates are deposited in the protein database under identifier 3DGO). The structure was identical to 3DGL (figure 2).
Unlike in the crystal state, an experiment with thin layer chromatography (TLC) in solutionrevealed most of the ATP
remained bound to the protein and was not hydrolyzed. This is reminiscent of the contradictory information mentioned
above: although as a crystal structure, secondary structures seemed to be present in clone B6-62, in solution under
ambient aqueousconditions, CD spectra showed these were absent.
Analysis of study number four
However interesting the chemistry reported may be, the structures of biological proteins which catalyze hydrolysis of ATP
in a bent position are very different, requiring participation of divalent metal ions and numerous hydrophobic and
electrostatic contacts to constrain ATP in a bent geometry.40 Any relevance of the experiments just discussed to biological
proteins is only speculation.Once again, no evidence was found for native-like folding in solution under ambient
conditions. A zinc atom chelated with portions of a polypeptide introduces some conformational constraints. Further
moulding of a portion of the protein was performed by ATP. The careful laboratory process of creating the crystal, at a very
low temperature, caused the artificial protein, already constrained by the zinc, to assume some secondary structure. But
these conditions are not expected to be found under natural conditions relevant for evolutionary scenarios.
We believe that the so-called folding behaviour is an artefact, in agreement with what the authors also wrote:
One interpretation of this result is that, in solution, ATP is able to adopt multiple conformations when bound to protein DX,
one of which is the bent conformation whose lifetime is short relative to the time required for hydrolysis. During protein
crystallization, the equilibrium between the different bound states shifts to favour the bent conformation, which is
observed in the crystal structure of protein DX obtained in the presence of saturating amounts of ATP. 41However,
biological enzymes fold into stable structures under ambient conditions, providing optimal geometric and electronics
regions to which their ligands bind reliably.It is possible that the helices are too short to remain in a single, stable
conformation. In one of the leading textbooks on protein chemistry, we read,In the helix, the first four NH groups and
last four CO groups will normally lack backbone hydrogen bonds. For this reason very short helices often have distorted
conformations and form alternative hydrogen bond partners.42The authors interpreted these results as a possible model
for primitive enzymes which could eventually evolve into the sophisticated versions found in extant organisms. The
opposite conclusion seems more reasonable: it illustrates how valuable ATP, essentially impossible to form under natural
conditions on its own, could be indiscriminately destroyed. The useful energy obtained from the hydrolysis of ATP in cells
must be tightly regulated and coupled to biochemical networks to harvest this energy.The authors did not report any data
which would permit a quantitative estimate of the random sampling space covered by their directed evolution, which would
be many orders of magnitude greater than the original library of 6 x 1012 members.
It is not clear why they conclude that
the de novo evolution of this ATP binding protein demonstrates that folded proteins with desired functions occur more
frequently than previously thought.41
On the same page they had just written,
Since the bent conformation is believed to be essential for hydrolysis, it is possible that this reaction is limited to the
crystalline environment where the bent conformation is stabilized by a network of well-ordered water molecules.
The reader has no way of knowing what proportion of random sequences in a natural setting would create the kind of
structure reported for the crystalline state. The latter requires a multitude of identical proteins to be brought together and
thermodynamically equilibrated by slow cooling.
Evaluation of studies three and four
The question we are interested in is the proportion of random proteins which would fold into native-like folds, which would
then provide a stable scaffold and the potential to be useful for some cellular function. No-one questions that ATP, and
other large organic molecules, can interact with a variety of substances, including polypeptides and even amorphous tarlike material.Of fundamental importance in evaluating these studies is to be careful in how the term protein fold is used.
Both the CATH43 and SCOP44protein classification systems use the technical term fold to define precise threedimensional topologies, and the reader could be misled to believe a true fold has been identified under ambient conditions
for the synthetic proteins reported.Intended interactions of biological proteins with ATP are strong and very selective, such
as binding of the adenine ring by protein kinases.45Inspection of the chemical structure of ATP 1 reveals many locations
where accidental non-covalent interactions with polypeptides can occur. Hydrogen bonds can form with the amino,
hydroxyl and phosphate groups and so can Van der Waals interaction with the flat, aromatic adenine ring. Therefore, it is
not surprising that about 0.1% of the random proteins generated were reported to be able to bind to ATP in the seminal
paper.46 Eight rounds of selection, without creating additional variety, increased the fraction which binds to 6.2%. This only
demonstrates that some portions of the protein can interact with the much smaller ATP molecule, an observation
chemically expected.
Some criteria to characterize native-like folded proteins

We can contrast the characteristics of the polypeptides isolated with those of typical folded proteins:Presence of
secondary structures features, such as alpha helices, beta sheets and stabilized connecting coils. Typically more than
60% of a protein folds into alpha helices and beta sheets. 47Self-assembly of a single three-dimensional topology, before
interaction with other bio-chemicals to form the quaternary structure. The amino acid sequence, and not structural
information obtained from some template, determines its folding.48,49 If denatured non-destructively, such as by using
organic solvents (e.g. urea) or detergents (e.g. sodiumdodecyl sulphate), they can usually refold quickly back to the native
state once the denaturant is removed. No cofactors are necessary for the self-assembly, 48 initially nor after removing the
non-destructive50 denaturing reagents by dialysis.Folding into a single, lowest-energy state. This also takes into account
constraints to funnel the folding process into the correct topology by using sequence details which generate impediments
to folding into an incorrect local energy minimum.
Solubility in water.51
Formation of a discrete protein species, usually monomeric. High-order aggregates often result from misfolded proteins 52
54
and underlie many diseases. This implies that sticky hydrophobic patches on the surface must be avoided. Higher
order aggregates must use precise and reliable locations for interactions, and not lead to a chaotic mixture of different
aggregation states nor interaction sites.
Researchers believe that folding often involves four steps:55
The unfolded stage, which actually lacks truly randomised structure.56 This is followed by
Formation of secondary structures. These first elements of secondary structure are only metastable, but it is believed that
when enough interact by random movements, a cumulative stabilizing effect can arise, leading to more stable collections
of secondary structure.57 Soluble proteins are driven by hydrophobic interactions that force non-polar residues together in
the central core of the protein. This condensation is called a hydrophobic collapse. 58 This leads to The molten globule
which is more compact and resembles the native protein in overall shape, but lacks many of the non-covalent side-chain
bonds, covalent disulfide bonds, charge pair interactions and other stabilizing features which characterize the final state.
Molten globules are still quite open and flexible, and the secondary structures may not be entirely formed.59
The final state is the native-like fold.Some proteins do not appear to go through a molten globule intermediate. But more
significantly, in vitro studies reveal the presence of additional intermediates which are not along the pathway towards a
native state. In vivo specific proteins catalyse the elimination of unwanted folds. 60 The existence of other conformational
states, which partially share characteristics with truly native folds, is probably the key to understanding the structure of
clone 1819.
Interpretation of the experimental data
Interaction of several AAs with a ligand does not automatically imply a native protein fold is present. All biological nativelike folds are composed of alpha helices and/or beta strands, unlike the structure of the reported proteins, such as the
optimized B6-62:
Given the unusual CD spectra indicating the absence of significant alpha helix or beta strand contribution we decided to
investigate .61Now, circular dichroism (CD) in the near and far UV range shows different spectra for native, folded and
molten globule states.65 It is also used routinely to identify alpha helices, beta structure motifs, and random coils. 62 We
agree with the authors interpretation of the clone B6-62:However, the results of CD spectroscopy suggest that the folded
protein has an unusual structure that may differ from normal biological protein structures. 63Earlier we proposed1 that the
minimal secondary structure suggested by crystallographic analysis of clone 1819 was probably an artefact and not
representative of the state under ambient aqueous conditions. Crystallographers are aware 64 that the process of
crystallization of pure substances to permit X-ray diffraction can generate a conformation very different from that found
under ambient conditions. For proteins this is especially problematic, due to the only marginal stability of the best
conformation:Although helices have regular repeating hydrogen bonds coupled with a uniformity of bond lengths and
angles this periodicity masks their marginal stability. In an isolated state most helices will unfold. 65Stabilization through
chelation with zinc would create small constrained regions on polypeptides without requiring the kinds of secondary
structures found in biological proteins. If such a structure would remain stable, portions could interact with a variety of
organic substances. To illustrate, a tiny portion of antibodies (which are proteins) are highly variable, permitting interaction
with millions of different ligands (antigens).66,67 However,
Only 510 amino acids in each hypervariable region form the
antigen-binding site. As a result, the size of the antigenic
determinant that an antibody recognizes is generally comparably
small. It can consist of fewer than 25 amino acids on the surface of
a globular protein, for example.68
In a prebiotic world with no genetic code producing many identical
proteins, naturalistic scenarios must take into account the great
variety of different substances which would be present, among
which useful protein sequences are supposed to be found. The
evidence we have examined illustrates that many worthless and
deleterious
chemical
interactions
can
occur.
As
an
example,69 molecules like ATP can also bind to random DNA
sequences, in addition to random proteins, in a fairly high
proportion.Evolutionary scenarios must take into account the need
to prevent deleterious chemical interactions. As pointed out earlier,1,70 clone 1819hydrolyzes ATP, thereby destroying one
of the most important molecules used by cells, in an unregulated manner which would serve no purpose.
Figure 4. Probability of forming soluble, native-like-fold proteins decreases rapidly with chain length. The shape of the
curve has not determined experimentally yet.
B6-62 is almost three times smaller than an average domain. 73 This was formed after the authors judiciously removed
portions of protein B6 to avoid aggregation and insolubility problems. The proportion of native-like folded proteins
decreases as the chain length increases. No plausible native-live candidates were identified from the initial library of 6
1012 sequences, from which the best candidates were amplified and mutated to enhance the chance of finding a suitable
variant. Therefore, the limited data suggests that the proportion of native-like folded domain in a random library could well
be less than 1011 and many orders of magnitude lower for proteins which consist of
multiple domains (which the majority do).Besides the rapidly increasing tendency to
aggregate and become insoluble with increasing chain length (figure 4), there are other
reasons why larger, biologically relevant proteins of average size 300 AA are more
difficult to create using natural processes.One example is that proline residues (figure 5)
need to isomerize to the correct isomer. The cis and trans isomers of the X-Pro peptide

bond (where X represents any amino acid) are nearly equal energetically. The fraction of X-Pro peptide bonds in the cis
isomer form when unconstrained typically ranges from 1040%. Cistrans proline isomerization is a very slow process
that can prevent correct protein folding by trapping one or more proline residues crucial for folding in the non-native
isomer, especially when the native protein requires the cis isomer.71,72 Clone 1819 has only three prolines, but an
average-size domain of 150 AA73 would have about 10 prolines, given that 4/61 codons code for proline, and an averagesize protein of 300 AA would have 20 prolines.
Figure 5. Structure of the amino acid proline.
All organisms have prolyl isomerase enzymes to catalyze this isomerization 72,74 but in a prebiotic scenario the native folds
would often not be produced in a relevant time period.
As another example, the number of disulphide bonds increases with protein size and these could form indiscriminately,
hindering or preventing formation of the native state in a pre-biotic scenario.
Conclusion
We cannot agree that this series of experiments demonstrates that about 10 11 random protein sequences produce nativelike folds. The estimate that 1011 random polypeptides would possess a native fold was based17 on clone1819. Additional
information surfaced after this proposed estimate had been widely disseminated. Recall that the folding results were only
possible in the presence of zinc, that protein 1819 forms an insoluble precipitate after 3 days 75 and that high
concentrations of ATP were necessary for it to remain stably folded. 30 Surely clone 1819 cannot be considered a
representative evolutionary starting point for a new true protein.The DX variant with better solubility was not part of the
original 1013 random sequences but of a larger space of random sequences, and therefore the proportion which folds
properly needs to be calibrated accordingly. That the value is lower than 10 11 is certain, but how much lower? The
secondary structures reported were found in crystals frozen 76 at 90 K but were not found in the CD spectra at
evolutionary-relevant temperatures. DX is also contingent upon the presence of chelating zinc. Furthermore, ATP was
also found77 associated with a -sheet in the crystal structure of DX, suggesting that all or most secondary structure could
be an artefact, caused by zinc and ATP, forming only at temperatures around 100 K (173C).
Final comments
In Part 1 of this series,78 I tried to find a lower limit for protein folding everyone could agree with. I pointed out that all
combinations of binary patterning79 for helices and coils would be covered by a polypeptide pattern such as:
(p/n)AA(p/n)AA,where p = polar; n = non-polar; A = Anything. This is a very generous assumption. This implies for an 80residue domain that only every third position would pose a constraint, leading to a proportion of (9/20) 27 = 1010. Since this
is actually an order of magnitude greater than proposed15 in the studies analyzed here, perhaps others would be willing to
accept the reasonableness of my lower bound. This implies that a proportion of about (9/20) 100 = 1035 random
polypeptides 300 residues long, about an average size protein, would fold properly.
Of great value would be experiments to isolate and characterize properly folded random polypeptides in the absence of
ligands and chelating metals, especially those metals expected to be present in extraordinarily low concentration in a
primitive ocean.
Vitamin B12
Evidence of design
by John McEwan
Figure 1. Chemical structure of vitamin B12.
Vitamins are organic compounds required by organisms in tiny amounts
as nutrients. In other words, an organic chemical compound is called a
vitamin when it cannot be synthesised in sufficient quantities by an
organism, and must be obtained from the diet. Vitamin B12 has a
recommended intake for an adult of a tiny 2 to 3 g (micrograms) per day.
It contains cobalt, and so it is also known as cobalamin, or the red
vitamin, due to its bright red colour.1 It is exclusively synthesized by
bacteria and is found in the diet primarily in meat, eggs and dairy
products.2 Vitamin B12 is necessary for the synthesis of red blood cells,
the maintenance of the nervous system, and growth and development in
children.Since vitamin B12 can be stored in the body, nutritional
deficiency of vitamin B12 is rare. Approximately 25 mg is contained in
the body in adults, around 50% of this in the liver. In people changing to
diets low in B12, including vegans and some vegetarians, it can take over
10 years for deficiency disease to develop. 3The discovery and
identification of the functions of B12 have been particularly fascinating.
B12 deficiency causes anemia. By accident, George Whipple discovered
the cure for B12 deficiency by investigating anemia in dogs. He found
that feeding large amounts of liver seemed to most-rapidly cure the dogs
anemia, and hypothesized in 1920 that liver ingestion be tried for
pernicious anemia in humans (a form of anemia due to inadequate absorption of vitamin B12).After a series of careful
clinical studies, George Minot and William Murphy set out to find the substance in liver that cured anemia in dogs, and
found that it was iron. They found further that the liver-substance that cured pernicious anemia in humans was something
else entirely different. In 1926 vitamin B12 was identified by this coincidence. For their work in pointing the way to a
treatment for anemia, they shared the 1934 Nobel Prize in Physiology or Medicine.A Nobel Prize was also awarded to
Dorothy Crowfoot Hodgkin and her team in 1956 for determining the complicated chemical structure of the molecule
(figure 1).What makes B12 particularly interesting is how it is taken up into the body from food. The absorption/ transport
of B12 from food to cells relies on four successive proteins in mammals; HC (haptocorrin), gastric IF (intrinsic factor), the
IF-receptor and TC (transcobalamin).4 Initial uptake of B12 requires binding in the stomach to salivary HC, which then
serves to protect the B12 from stomach acids. Then in the duodenum (the first part of the small intestine), enzymes digest
the HC and release B12 again. IF, a protein synthesized by gastric cells, next binds the B12 specifically.Cells in the ileum
(the last part of the small intestine) absorb the IFB12 complex by a specific receptor, yet another protein. Inside the cell,
the IFB12 complex is degraded and B12 is transferred to TC, which delivers B12 around the body in the blood. 5 B12
must be attached to IF for it to be absorbed, as ileum cells only recognize the B12-IF complex, not B12 alone. It is more
commonly an inability to produce or absorb IF than a low level of B12 in the diet that leads to B12 deficiency.Therefore,
absorption of food vitamin B12 requires HC, IF, IF-receptor and TC. Also required are an intact and functioning salivary

system, stomach, pancreas and small intestine. Problems with any one of these organs makes vitamin B12 deficiency
possible. Each of the proteins involved in B12 absorption are highly complex molecules with their own specific role, and
the chance of any of these forming by random processes is miniscule! Furthermore, the irreducibly complex combination
of proteins and organs all operating at different stages in B12 absorption defies biological evolution. Since bacteria have
the ability to make B12, why would this have been lost in higher animals and replaced with such a complicated uptake
system (as microbes are supposed to have evolved into molecular biologists)?
Antifreeze protein evolution: turning wrenches into hammers
by Shaun Doyle
Evolutionists have often claimed that gene duplication provides the raw material to produce new functions through
subsequent mutation and natural selection. However, finding gene duplications that have produced new functions hasnt
been easy. Most gene duplications studied have been silenced and subjected to deleterious mutations, rendering them
useless.1 However, a class of proteins called antifreeze proteins (AFPs) appear to have gone against this trend. AFPs are
found in a wide variety of organisms: fish, insects, plants and microbes. They also take as many different forms as there
are organisms that have them, and many are believed to have evolved via gene duplication events. These proteins bind to
the surface of ice crystals and prevent water molecules from binding to the ice crystals, preventing the ice crystals from
growing. This enables organisms to survive in sub-zero temperatures without freezing.However, postulating that gene
duplication and subsequent mutation can result in new functional proteins is not enough. To build a plausible case for neoDarwinism one needs to identify the source of the new gene and outline the major mutations that actually lead to the
change. Researchers have recently posited a detailed evolutionary scenario for the evolution of an antifreeze protein from
such a gene duplication event in a species of Antarctic eelpout (ray-finned fish), Lycodichthys dearborni.2 So do these
AFPs represent a neo-Darwinian mechanism producing a new protein? And if they are, what does this say for the
plausibility of neo-Darwinism forming the mechanistic basis for microbes-to-man evolution?
The edge of eelpout evolution
Lycodichthys dearborni is one of several species of Antarctic eelpout of the family Zoarcidae. It is found in McMurdo
Sound, south of the Ross Sea, off the coast of Antarctica. Zoarcids are one of only two families of fish to possess a
particular class of AFPs, type III. The origin of type III AFPs has been particularly difficult for evolutionists to trace. 3
Figure
1. Molecular
evolution
of AFPIII from SAS-B. One
daughter SAS-B duplicate
(SAS-B) underwent Nterminal domain deletion
(seventh codon of E1
through
E5)
and
neofunctionalization
into AFPIII.
Regions
in SAS-B corresponding
to the regions in the twoexon AFPIII gene
are
indicated with dark grey for
the two genes, with
nucleotide
sequence
identities given. The partly
non-protein-coding signal
peptide (SP) precursor
sequence in SAS-B that
was modified to become a
coding
sequence
for
the AFPIII signal peptide is
shown
at
the
bottom. LdSAS-A lacks the
5 flanking
sequence
homology
(grey
bar)
with LdSAS-B andAFPIII;
thus, it is not the
evolutionary
progenitor
to AFPIII. (From Deng et al., ref. 2, p. 21595.)The researchers proposed that the evolutionary model Escape from
Adaptive Conflict (EAC) fits the evolution of the AFP in the L. dearborni. EAC states that conflict between an old and an
emerging new function within a single gene could preserve a gene duplication, which would allow each duplicate to freely
optimize one of the functions. They proposed that a type III antifreeze protein gene evolved from a duplicated copy of the
sialic acid synthase (SAS) gene called LdSAS-B (Ld stands for Lycodichthys dearborni). They posit that in one duplicate,
the N-terminal SAS domain was deleted and replaced with a new signal peptide (figure 1). This removed conflict between
SAS and ice-binding functions in the duplicate and allowed rapid optimization of the remaining C-terminal domain to
become a secreted AFP.There appears to be one random mutation that enabled the copy of LdSAS-B to translate a
protein that ended up being an effective AFP: a 4.4 kb deletion coding for the SAS N-terminal domain from codon 7 of E1
(exon 1) to the end of E5 (exon 5).4,5 This achieved two fortuitous things that transformed the putative SAS gene into
an AFP gene: it stripped it of any SAS function, and enabled a change in reading frame, which altered the signal peptide,
enabling the new protein to be secreted.I suggest the large deletion mutation in the putative LdSAS-B copy
produced both the increased ice-binding capabilities and enabled the putative AFP to be secreted out of the cell. The
deletion was so large, it rendered the copy gene unrecognizable to the translation machinery as an SAS gene. It also
destroyed the integrity of the first exon, which exposed coding for a signal peptide targeting the protein for secretion that
was previously latent in the 5 flanking region of LdSAS-B.

Figure 2. The structure of a mature type III AFP of Lycodichthys

dearborni.A. The primary structure peptides in bold are thought to form the flat
ice-binding surface typical of type III AFPs (from Deng et al., ref. 2, supporting
information, Figure S6; and Protein Data Base www.rcsb.org/pdb, entry
1UCS.pdb).B. Tertiary structure, positioned to show the ice-binding surface,
with the peptides with an active ice-binding function highlighted in black (from
Protein Data Base www.rcsb.org/pdb, entry 1UCS.pdb, displayed in Protein
Workshop).
As Cheng et al. found:
We discovered a precursor signal peptide coding sequence appropriately
located in the extant LdSAS-B, starting from 54 nt upstream of the translation
start site through the first six codons of LdSAS-B E1 (Fig. 2 and Fig. S2B). An
intragenic deletion from the seventh codon of E1 through E5 of LdSAS-B and
linkage of the new E1, the old I5 [intron 5], and E6 would complete the
formation of the nascent two-exonAFPIII gene encoding the secretory
antifreeze protein.6This latent coding was therefore mistakenly recognized by
the translation machinery as the first exon (which had been truncated at the 3
end due to the deletion mutation) rather than as part of the 5 flanking region,
and it became the signal peptide needed to signal the AFP (largely coded for on
the second exon of AFPIII).Strong selective pressure in the Antarctic waters
would make tandem duplications of AFPIII a likely response to the conditions in
order to increase the amount of AFP manufactured, and would kill off any
eelpout that didnt have the gene. Therefore, in short, the answer isyes, random
mutation and natural selection is a likely mechanism for how this AFP and many others were produced.
From neo-Darwinism to evolution
However, the story doesnt end there. The foundational question remains: is this an example of neo-Darwinism evidence
for molecules-to-man evolution? There is no doubt that this is a new, functional protein; many readers might be convinced
that it answers the creationists demand for evidence for naturalistic evolution. But its not that simple: to possibly stand as
evidence for molecules-to-man evolution, evidence for new proteins has to have at least four characteristics:
they need to be formed by a plausible naturalistic mechanism,
they need to be complex,
they need to be specified, and
they need to be functionally integrated into the organisms biochemical processes.
AFPs generally fill criterion 1: there is good evidence that many, if not most, are formed through random mutation and
natural selection. Moreover, this particular naturalistic scenario for the Antarctic eelpout AFP is plausible. One may wonder
whether designed mechanisms for variation had some role to play, especially given the large number of tandem repeats
and the retrotransposon attached to the 5 end of the gene cluster. The origin of such designed adaptability is hard for
evolution to explain, but it is speculative at best for this particular scenario.Not all AFPs fulfil criterion two: many are very
simple, yet very ordered.7 The Antarctic eelpouts AFP is only 65 peptides long, but is not arranged in a repetitious
sequence.8 Moreover, many short sentences that carry a lot of information are less than 65 characters long. Consider how
the
original
Greek
text
of John
1:1 wouldve
been
written:
That sentence is only 52 characters long (51 if the subscript iota is left out), and there is much more repetition of
characters there than in the AFP. Complexity is therefore not merely a measure of sequence length, but also largely of the
lack of repetition in the sequence. So while short, it is of sufficient complexity to satisfy criterion 2.Criterion three,
specificity, is where AFPs in general run into major problems. As stated previously, AFPs are notorious for having as many
forms as the creatures that possess them. Consider a hammer and a wrench. Each is designed for a different function: a
hammer cant do all that a wrench can, but many things can be used to do the job of a hammer when a hammer isnt
available, including a wrench. However, everyone knows that a wrench was not designed to be a hammer, and if you try to
use it as a hammer too often, it may no longer work as a wrench.The SAS protein is like the wrench, and AFPs are like
hammers. All that AFPs are required to do is bind to ice, and for that they basically need only a hydrophobic end and a
hydrophilic endi.e. they must be amphiphilic. The hydrophilic end binds to the ice crystals, while the hydrophobic end
repels water molecules, stopping them from binding to the ice. Such a situation is found in many proteins with completely
different structures because of the 20 universal amino acids that comprise proteins, half of which are hydrophilic and half
are hydrophobic.9Being an AFP is a very non-specific job that many different random proteins could perform.AFPs also fail
to satisfy the fourth criterion: functional integration into the cells biochemical processes. They are not known to interact in
any other cellular process other than secretion from the cell and binding to ice crystals. 10 Proteins such as AFPs
effectively slow the efficiency of other functions in the cell by cluttering it with degenerate debris. 11 Under normal
circumstances, any mutation that enables the translation of either degenerate DNA or non-coding DNA will merely gum
up the works of the cell and distract the transcription and translation machinery from other tasks, slowing cellular
processes down.12 But again, fortuitous combinations that were already latent in the LdSAS-B gene meant that the
translation of this particular debris was beneficial for survival in a particular environment.
This example may cause some excitement for evolutionists, but it doesnt contradict what we may expect within a young
age scenario. It is a far cry from demonstrating that mutation and natural selection are necessary and sufficient to explain
the history of life. Evolutionists have failed to grasp the evidential burden they have to actually substantiate their just-so
stories on how high-information-content biological structures could form naturalistically. AFPs in general, and this example
in particular, fail as evidence for molecules-to-man evolution.
Conclusion
AFPs are demonstrably simplistic proteins in their function, more akin to beneficial debris than a new complex and
specified protein. Creating an antifreeze protein naturalistically is qualitatively different from creating, for example, the
blood-clotting cascade, cellular differentiation programs, the photosynthetic pathway, or a bacterial flagellum. The
difference is between incidental (and accidental) function and essential biological structure.
Incredible Kinesin!
Biological robots will blow your mind!
by Calvin Smith
Published: 26 June 2012 (GMT+10)

Stand and deliver

Kinesin molecules are motor proteins found inside living things. Known as the workhorse of the cell, they haul vital cargo
along roadways in cells called microtubules. Steven Block (professor of applied physics and of biological sciences at
Stanford University) has described kinesin this way; "Kinesin functions like a locomotive in cells to ferry cargo back and
forth.1
Illustration by Caleb Salisbury
A typical kinesin molecule is a mere 70 billionths of a metre
(three-millionths of an inch) long and has an amazing
likeness to a person! A typical kinesin has two arms on one
end (that hold onto the cargo) and two legs on the other end
that walk along the microtubule, pulling the cargo toward its
final destination. In a sense they are like the postman
delivering mail inside cells.
Biological robots?
Inside all life forms that have nuclei in their cells
(eukaryotes), proteins and other parts need to be delivered to specific places within the cell at specific times. If the needed
part is a protein, a manufacturing plant (called theribosome) receives blueprints for the part from the nucleus (the
information is stored in the nucleus on a strand of DNA, but the blueprint is sent in the form of an RNA copy of that section
of DNA).This is a complex coordinated effort, as something must first access the creatures DNA library, unzip it at the
exact location needed for the specific information required (for whatever part is to be manufactured), create a duplicate of
the information for the part and deliver it to the factory. (See animation, below left.)Then another organelle in the cell
(called the Golgi apparatus) packages the needed part by wrapping it in a bag (called a vesicle) and imprints the address
where the part is to be delivered in the cell onto the outside of the vesicle parcel.Then a kinesin is summoned. It picks up
the parcel and walks along microtubule roadways in the cell and delivers the parcel where it is needed. (Many different
types of kinesin [and kinesin-related proteins] with different specifications and functions have been discovered in various
organisms from yeast to humans. The above example was simply an example of a common task.)
A view from above
Kinesin is the miniscule longshoreman (stevedore) of the cell, toting parcels of cargo on its shoulders as it steps along a
scaffolding of microtubules. Each molecule of ATP fuel that kinesin encounters triggers precisely one 8-nanometer step of
the longshoreman.To grasp the complexity of what scientists are observing kinesin do, we could use the following
hypothetical scenario as an analogy from a more familiar point of view:Joe is working at his job one day when his machine
breaks down. He identifies the broken part and makes a call from his cell phone to a local manufacturer requesting a new
one, giving them the part number.The manufacturer agrees and records Joes address. The manufacturer has a list of all
the part numbers on hand but not the schematic for them so they send an email to another company (that has a copy of
all of the blueprints for every part needed in the industry) requesting the blueprint. A person there makes a photocopy of
the needed section and delivers it to the manufacturer.From the instructions in the blueprints, the factory then
manufactures the part and puts it in a package marked with the postal address from its database.A courier is contacted.
He comes to the factory and picks up the package. Having detailed maps of the city, the courier plots out and travels
along the most convenient route and delivers the package. Mission accomplished!Most would agree that the level of
complexity just described is pretty impressive. The technology and integrated components (such as the specialized
knowledge, communications systems, manufacturing capability, and databases) needed for such intricate procedures are
incredibly sophisticated, and all of these steps were coordinated by intelligent people at every stage. However, the actual
processes involving kinesin are far more complicated than what Joe experienced above.
All in a days work
As astounding as this is, research is showing that kinesins do far more than initially thought. Kinesins are now known to
support mitosis (cell division) and meiosis (cell division in which a nucleus divides into four daughter nuclei to make
reproductive cells). In addition to transport of mundane cellular cargo, kinesins transport the neurotransmitters needed
for neurons to communicate with one another.Certain kinesins can dismantle the microtubules after their journey.
Controlling the length of microtubules is particularly important
during cell division2lack of control can cause chromosomal
instability, which is in turn linked to human cancer.3Professor
Matthias Rief (from the Physics Department of the Munich
University of Technology) says, Our results show that a
molecular motor must take on a large number of functions over
and above simple transport, if it wants to operate successfully in
a cell. It must be possible to switch the motor on and off, and it
must be able to accept a load needed at a specific location and
hand it over at the destination.4
123rf.com/Viktor Gmyria
Fast and efficient
Not only do these incredible kinesin robots perform a variety of
tasks, they also do so with incredible efficiency! Check out these
state of the art features:
PowerNot only is it tiny, but kinesins motor is about 50 percent efficient, which is about twice as good as a gasoline
engine. And pound for pound, kinesin produces nearly 15 times more power than that man-made engine.5
SpeedThe kinesin motor is impressively fast, capable of 100 steps per second. Scaled up to our own dimensions, a
motor with corresponding properties would travel at similar speeds and produce as much horsepower per unit weight as
the jet engines of the Thrust supersonic car6, which recently broke the sound barrier. 7 (This would be proportional to a
person moving 600 meters per second or 1,300 miles per hour!)
Energy efficientKinesins are powered by the universal energy compound known as ATP (which is produced by another
incredible molecular motor called ATP synthasesee animation, below right. Each molecule of ATP fuel that kinesin
encounters triggers precisely one 8-nanometer step of the postman, but kinesins go into sleep mode when cargo isnt
attached to prevent ATP from being wasted. Similar to how modern computers shut down after a period of un-use to
conserve energy, kinesin have a hibernation feature as well. (Although scientists know that the motor folds over in an
autoinhibited8 state when resting, the molecular mechanism remains unclear.)

Team playersKinesin molecules also work together when the going gets tough! If the load needing transport is too
heavy for one postman to handle, there is significant evidence that cargoes in-vivo are transported by multiple
motors.9
They also demonstrate multiple handling of their cargo. Similar to runners in a relay race, kinesins can hand off their
cargo to a fresh bystander after delivering it a certain distance, and the other kinesin will finish the delivery process.
Flexible planningKinesins also have a bypass mode ability that allows them to navigate around obstructions they may
encounter. Similar to a GPS system re-computing, kinesins have demonstrated the remarkable ability to re-route
automatically when needed.
RecyclingThe most ardent champion of the green movement would be jealous of the kinesins conservation and
recycling capability. There is good evidence they are either transported back to the cell center in groups by large transport
units (like mass transit in cities) or alternatively dismantled and their parts recycled when done their tasks.10
Committed to naturalism
Of course such incredibly sophisticated bio-technology screams Design, but does the desiner get the glory in the
scientific literature describing these amazing machines and processes? No, nature does:It is impressive
how nature (emphasis mine) manages to combine all of these functions in one molecule. In this respect it is still far
superior to all the efforts of modern nanotechnology and serves as a great example to us all.11Why is it that at a time
when science is revealing such telling evidence of the designer`s handiwork that intelligent people can see the evidence
and deny the designer? Its because of the atheistic, evolutionary indoctrination that most people in the Western world
receive, of course. Atheism is committed to naturalism, and so as Dr Scott Todd (an immunologist at Kansas State
University) said: Even if all the data point to an intelligent designer, such an hypothesis is excluded from science because
it is not naturalistic. Of course, according to evolutionary theory, eukaryote cells supposedly evolved well over two billion
years ago12. This means evolutionists are willing to believe that such astoundingly sophisticated technology like molecular
motors and their operating systems arose through natural processes (with no intelligence) very early on their imaginary
timeline. But this is technology far superior to anything the most intelligent scientific minds on the planet have ever
produced!
Is evolution the answer to our beginnings?
Motion at the cellular level is a hallmark of being alive, Block has said. A fundamental question is, how did living
organisms figure out how to move? The answer is they developed kinesin and several other very efficient protein motors.
If kinesin were to fail altogether, you wouldnt even make it to the embryo stage, because your cells wouldnt survive. Its
that important.13Evolutionists have no plausible theory on how something as sophisticated as kinesin (and the required
operating and communication systems) could have evolved in a gradual fashion (let alone all of the countless other
functions and features we know of in so called simple living things).However, when we see similar machines and
operating systems (robots, computers, the Internet, etc.) in our everyday life at work or home they are always the result of
intelligent and intentional design. How much more logical to believe that the ultimate mind we are able to conceive
created all of the marvelous machinery within us and the world around us!
The non-evolution of apoptosis
by Philip B. Bell
The phenomenally complicated programme of cellular death, otherwise known as apoptosis, is the chief source of
occupation for tens of thousands of scientific researchers. Creationists happily ascribes praise to the omniscient designer
for the incredible designed complexity that is apparent. Conversely, the person who subscribes to methodological
naturalism faces the significant challenge of accounting for the origin and evolution of apoptosis. The oft-claimed
conservation of various apoptotic components, from the very earliest life-forms, does not suffice as an explanation.
Apoptosis-style demise is now recognised in unicellular eukaryotes and even bacteria and, in recent years, a handful of
evolutionists have published hypotheses in the scientific
literature in which they have attempted to explain the
simultaneous evolution of apoptosis and endosymbiosis. The
latter, itself is an unproven hypothesis for the origin of the first
unicellular eukaryotic cells, including the origin of mitochondria.
An examination of these evolutionary ideas, for a naturalistic
origin of apoptosis, forms the main focus of this paper.

* Items with an asterisk, the first time they are mentioned, are defined in a glossary at the end of the article.
Figure 1. Diagrammatic representation of the hypothesised events during endosymbiosis. The schematic illustrates the
two fundamental events that are envisaged to have contributed to the first eukaryotes. The phagosome (invaginated
engulfing membrane) becomes the outer membrane of the endosymbiont. Theorists differ on the order of events; i.e.
whether a well-developed nucleus evolved prior to the acquisition of mitochondria (1) and chloroplasts (2) or whether
mitochondria predated chloroplasts in serial endosymbiotic episodes.
Apoptosis* or programmed cell death is a ubiquitous cellular phenomenon in living organisms. An earlier paper described
this process in detail, contrasted it with necrotic cell death and provided a framework in which to understand cell death
from a young-earth creationist perspective.1,2 Readers are advised to familiarise themselves with that paper in order to
better appreciate the arguments presented here. Some scientists have recently questioned the distinction between
apoptosis and necrosissince this relates to the authors previous arguments, an appendix includes further discussion.
Apoptotic phenomena have been described in several unicellular eukaryotes*. The alleged evolutionary conservation of

apoptosis, from the earliest eukaryotic cells, would therefore appear to be problematic on theoretical grounds. Not only
must evolutionists explain how apoptosis evolved before the invention of multicellularity, but they are faced with
explaining howsingle cellsthat have acquired a functional apoptotic responsepass on this more advantageous,
advanced genetic complement to their progeny? In what follows, some evolutionist attempts to grapple with apoptotic
origins are reviewed. A major component of these ideas is the hypothesis of endosymbiosis*.
Evolutionists on apoptosisa review
Some examples of the claims by evolutionists, that the widespread occurrence of apoptosis indicates it to be a highly
evolutionary-conserved mechanism, have already been documented by this author. 3 The following assertion typifies this
approach:The conservation of transduction* pathways and functional homology of effector molecules [involved in
apoptosis and differentiation] clearly bear witness that the principles of life established during prokaryotic and eukaryotic
unicellular
evolution,
although
later
diversified,
have
been
unshakably
cast
to
persist
during
metazoan*phylogenesis.4However, such statements are circular and, by definition, explanation-free. The argument,
although not spelled out as such, runs something like this: homology is similarity due to common ancestry; thus the subcellular and biochemical homology observed in apoptotic mechanisms is evidence that these multifarious life forms
sprang from a common ancestor. As Wells has pointed out, such claims for homology are nonsense. 5However, for the
purposes of argument, let us allow the possibility that apoptotic mechanisms have evolved; that is, that their exquisite
design is indeed the product of mutations and natural selection, occurring over millions of years. Have any evolutionists
succeeded in their attempts to give substantive explanations for the origin and evolution of apoptosis purely by reference
to time, chance and natural processes? The answer is a resounding no; as we shall see, such hypotheses are woefully
inadequate. This paper is predominantly a critique of two papers in which the authors speculate on apoptotic origins.6,7
Unicellular apoptosis
From an evolutionary view-point, the fact that apoptotic phenomena have been observed in many species of unicellular
eukaryotes8-17 places the origin of apoptosis before that of metazoan life. However, it is conceded that, The cell death
pathways of protozoans show no homology to those in metazoans, where several death pathways seem to have
evolved in parallel.18One idea is that ancient viral infections transferred key elements of apoptotic pathways to the
nuclear DNA of such early cells 17 but it is difficultif not impossibleto conceptualize the step-wise production of highly
complex apoptotic cascades in single cells. Nevertheless, let us imagine that a miraculous combination of the precise
information-adding mutations occurred that specified for a complex, fully functioning apoptotic mechanism in a unicellular
eukaryote; i.e. hopeful-monster-style! Fitness in evolutionary terms is measured by an organisms survival chances. But,
in the case of this unicell, the true test of its apoptotic mechanism is its demise rather than its survival. It is difficult to
imagine how natural selection could select good apoptotic genes for their survival value. Therefore, an evolutionary
scenario purporting to account for either the origin of apoptosis or its improvement by natural selection has conceptual
difficulties, placing the onus on evolutionists to provide a convincing rationale for these things.For a long time these issues
were ignored, or else were not widely appreciated. In recent years, however, several scientists have put forward
hypotheses that attempt to address this conundrum (see later) and there is now discussion of apoptosis in prokaryotes *.
One web-article says of bioscientists at the University of Melbourne, Australia,
[They] believe apoptosis may have arisen even before the first
multi-cellular organisms, possibly in single-celled bacteria, in
which virus-infected cells suicided to protect their relatives.
Multi-cellular life forms later recruited the apoptosis
mechanism as a way of discarding unwanted cells during
embryogenesis viruses that can inhibit apoptosis are an
obvious hazard, so evolution invented a backup system to
eliminate virusinfected cellsthe cytotoxic T-cell.19
The existence of various apoptotic signatures in the
developmental processes of several species of extant bacteria
has been reported,20,21involving gene activation and the
interaction of various signal transducers and their regulators. In
other words, what has been traditionally termed bacterial
autolysisself-digestion of the cell wall by peptidoglycan
hydrolase enzymes, resulting in the cells disintegrationmay
represent apoptosis. Programmed death in bacteria also
appears to occur in the presence of damaging agents such as
antibiotics,22,23 with some interesting implications for certain
types of antibiotic resistance.24 However, fascinating though
the findings of these all research efforts may be, accounting for
the evolutionary origin of apoptotic mechanisms in early
bacteria is quite another matter.
Endosymbiosis
Figure 2. Kroemers Highly speculative model for the
endosymbiotic evolution of mitochondrial permeability
transition (adapted and redrawn from figure 5 of the authors
paper; see ref. 6). (A) The hypothetical moment of
accommodation of the aerobic bacterial endosymbiont
(possessing both an inner and outer membrane) into the host
cell. Bacterial molecules such as perforins translocate to the
host cell phagosome, allowing diffusion of small molecules
such as ATP. Precursors of ANT (adenine nucleotide
translocator), PBR (peripheral benzodiazepine receptor) and
cyclophilin D are also envisaged to be present, though not in
their contemporary locations. (B) Later in evolution, a true PT
pore complex arises (enclosed by the dashed box). The PTdependent release of molecules such as cytochrome c and protease enzymes may then cause apoptosis.A handful of
evolutionists, pondering unicellular apoptosis, have speculated that endosymbiosis (the hypothesis for the origin of
mitochondria* and chloroplasts in eukaryotic cells) and apoptosis evolved simultaneously. Endosymbiosis theory was first
popularised by Lynn Margulis in the mid 1970s 25 and, with modifications, is now almost universally accepted by
evolutionists. Cellular organisation of eukaryotes is so much more complex than that of prokaryotesincluding

membrane-bound nucleus, mitochondria, chloroplasts, Golgi body, endoplasmic reticulum, 9 2 flagellum/cilium

arrangement, cytoskeleton, diploid stage in life cycle, mitotic and meiotic cell divisionthat their alleged evolutionary
origin is a fundamental question in biology. Of course, from a creationist perspective, each basic kind of prokaryotic and
eukaryotic organism (unicellular and multicellular) is the creation of the designer and no continuum between these
fundamentally different cellular organisations is expected. However, evolutionary theory must account for eukaryotic
origins. The basic idea of endosymbiosis is that aerobic, autotrophic bacteria took up residence inside larger prokaryotes
and became the forerunners of mitochondria. Likewise, chloroplasts are said to be descended from photosynthesising
prokaryotes (e.g. cyanobacteria) that were engulfed by larger prokaryotes (figure 1).Thus, a mutually beneficial
arrangement between host and endosymbiont is imagined; e.g. anaerobic cells would have benefited from aerobic
endosymbionts in environments where oxygen became available. In time, some of the functions of these precursors of
mitochondria and chloroplasts were allegedly transferred to the nucleus of the host cell. Not surprisingly, there are many
problems with such scenarios.26 In spite of the general acceptance of the basic tenets of endosymbiosis, Jerlstrm notes
that current scientific evidence conflicts with the stepwise evolution from prokaryote to primitive eukaryote and then to
eukaryote.27Indeed, very recently, a further problem came to light as researchers confirmed the presence of so-called
mitochondrial remnants in the gut parasite Giardia intestinalis.28 This nucleated unicell (a protozoan) has long been
thought to lack mitochondria and, therefore, has been said to be an intermediate between prokaryotes and eukaryotes; as
such it has been the standard textbook example of a key player in eukaryotic history. 29 It was argued that the nucleus
developed prior to the acquisition of mitochondria. Evolutionists now recognise that the finding of mitosomes
in Giardia argues strongly against it being an intermediate in the endosymbiotic evolution of eukaryotes. 30In spite of these
problems with endosymbiosis, many evolutionists will undoubtedly continue to contend for a simultaneous origin of
endosymbiosis and apoptosis. Ideas of a coupled endosymbiosis/apoptosis origin will now be examined in some detail to
see how they stand up to close scrutiny.
Kroemers hypothesisrole of mitochondrial permeability transition
A French scientist, Guido Kroemer, published the first major attempt at a hypothesis for apoptosis evolution. 6 In spite of
the huge diversity of apoptosis pathways, some features are usually the same in the majority of species/tissues/cell types;
e.g. DNA fragmentation, externalization of the membrane phospholipid, phosphatidylserine, cell shrinkage, production of
ROS* (reactive oxygen species), and activation of proteases. 31 Seemingly, a common pathway exists at the point of no
return, the executioner stage. Kroemer argues therefore, that mitochondria play a key role at this stage and he
speculates on their evolutionary origins. He initially gives a detailed review of the various requirements for the central
executioner of apoptosis, which may be summarized as follows:
It must become activated at the effector stage (i.e. at point of no return);
Its presence should be sufficient to cause apoptosis but vital if apoptosis is to occur at all;
Many triggering pathways should converge onto it;
It should coordinate all nuclear, cytoplasmic and membrane apoptotic manifestations;
It should be ubiquitous as diverse cell types undergo apoptosis;
It must include a function(s) that is(are) essential for cell survival, otherwise mutations could potentially result in a
supremely apoptosis-resistant cell;
It should act like a switch (on/off) so that cells either die or survive.
Apoptosis executioner revealed
Kroemer convincingly argues that the mitochondrial permeability transition step (hereafter MPT) fulfils these criteria.
MPT* involves the movement of solutes across the inner mitochondrial membrane, disrupting the potential of the transmembrane proton pump and resulting in the efflux of soluble proteins from the matrix and inter-membrane space of the
mitochondrion to the cytoplasm.32 This occurs via permeability transition (PT) pores (or mitochondrial
megachannels).33,34 Evidence that MPT constitutes the apoptosis executioner step includes the following: many triggering
pathways do converge on MPT;35 MPT is manifest by disparate cellular effects (nuclear apoptotic changes, production of
ROSwhich themselves can trigger MPT, altering cellular redox potentialsand oxidation of membrane lipids); molecular
components of the PT pore and MPT events are ubiquitous.36 The caspases (ICE/proteases) have been previously
mentioned as executioners as well as their activation by mitochondria; 37 ongoing research is helping to clarify the role of
protease cascades and MPT at this crucial juncture of apoptosis.What seems to be important is the release, via these
pores, of (a) Apoptosis Inducing Factor (AIF)a potent inducer of the nuclear apoptotic changes, culminating in
oligonucleosomal DNA fragmentation, and (b) cytochrome c*an activator of one of the key apoptotic-signature protease
(caspase) enzymes.38 Proteins in the Bcl-2 family, such as Bcl-2 and Bcl-xl, reside in the outer mitochondrial
membrane2,37 and research suggests that they can stop liberation of AIF 39 and cytochrome c40,41 by controlling the MPT,
thereby suppressing apoptosis. Conversely, the Bcl-2 antagonist, Bax, has been shown to promote MPT by disrupting the
trans-membrane potential.42
Apoptosis/endosymbiosis hypothesis and axioms
Refreshingly, Kroemer admits that he takes as a given the widely accepted endosymbiosis hypothesis. 25He states that
this is one of his premises and describes his subsequent hypothesis as speculation.43Additionally, because of his
evolutionary world-view, the existence of apoptotic phenomena in unicellular eukaryotes as well as all metazoa (animal,
plant and fungal cells) constrains him to believe that apoptosis evolved before multicellular life appeared44 indeed he
plumps for its simultaneous origin with endosymbiosis (discussed below). From this alone, we see that no matter how
plausible his ideas might seem, they are not the inescapable conclusion of data from operational science. In other words,
providing a possible evolutionary scenario does not equate to proof for the origin and alleged conservation of apoptosis.
One can equally choose to regard the ubiquity of apoptotic machinery in living cells as testimony to a common design
plan.As evidence for the apoptosis/endosymbiosis hypothesis, Kroemer mentions that certain MPT-like phenomena and
associated molecules (or their homologues; or analogues) have been found in widely disparate cell types, including the
yeast, Saccharomyces cerevisiae, and various bacteria. He states, it appears possible that many of the constituents of
the PT pore and several apoptogenic mitochondrial proteins were already present in the aerobic bacterium from which the
mitochondrion evolved [emphasis added].45 However, although certain mitochondrial cell-death events in eukaryotic cells
seem to have some parallels in bacteria, it has been reported elsewhere that the specifics of MPT-mediated mitochondrial
destruction are not thought to be related to autolysis of bacterial cells. 21 Furthermore, the characteristic nuclear apoptotic
events of multi-cellular eukaryotes are absent from unicellular eukaryotes like yeast. Needless to say, many evolutionists
interpret this to mean that mitochondrial apoptotic phenomena are ancient and phylogenetically conserved, whereas the
nuclear events are a later innovation.Having discussed several additional premises (themselves based on another
authors hypothesis for bacterial apoptosis!20), Kroemer states: it is conceivable that the basic mechanism of apoptosis
became fixed during evolution in the very moment in which endosymbiosis became established [emphasis added].46Of
course, implicit within this statement is the admission that the basic mechanism of apoptosis is irreducibly complex.

However, if one considers the extraordinary complexity of apoptosis,1,2 this statement is seen to be a tremendous leap of
faith, little short of belief in miracles.
Multiplied speculation
Figure 3. Schematic depiction of the key elements of the
endosymbiosis/apoptosis hypothesis of Blackstone and
Green. This is alleged ancient signalling pathway is said
to be the precursor of apoptosis. (A) A rapidly dividing
anaerobic host cell with its newly acquired aerobic
endosymbiont protection from ROS is afforded by the
relationship.(B) Under conditions of low metabolic
demand, host cell division is much reduced, resulting in
the release of ROS and cytochrome c from the
protomitochondria. Cytochrome c acts synergistically with
ROS formation to produce highly reactive free radicals
(the most mutagenic ROS) and a consequent high
mutation rate in the host. This, in turn, is assumed to
trigger sexual recombination, generating novel host cells.
Kroemer goes on to detail his Highly speculative model
for the molecular evolution of mitochondrial permeability
transition (PT)47 which is summarised as follows: The
aerobic bacterium (precursor of the protomitochondrion)
that invaded or was ingested by the potential host cell
(itself a bacterium) is envisaged to possess toxins, which
may or may not have been host-specific. In order to avoid
releasing these harmful chemicals into the host
cytoplasm (thereby killing the host cell) the hosts
bactericidal enzymes (precursors of apoptotic proteases)
had to be sequestered in sub-cellular organelles (e.g. a
lysosome; though how this evolved is not explained)48 or
else maintained as inactive precursors. Thus, a sort of
stand-off was established, where any attempt by the
protomitochondrion to kill the host, or vice versa, was
inhibitedobligating both parties to accept a symbiotic
relationship.From this moment, the two initially
independent organisms are forced to co-evolve. During this co-evolution, large parts of the bacterial genome are
gradually incorporated into the nuclear genome [emphasis added].49How did MPT simultaneously evolve during this
endosymbiosis event? The author describes a scenario whereby certain bacterial molecules (e.g. porins) are imagined to
be precursors of the PT pore complex that forms across the mitochondrial membranes. A basic PT pore was formed at the
moment of endosymbiosis, with porins and other molecules hopping across from bacterial membrane to the phagosome*
later on, other molecules evolved to produce the PT pores seen in mitochondria today (figure 2). After yet more
speculationthe text is littered with words like may, possibly, conceivable, speculativethe reader is told,The
essential role of the PT pore (or its components) in the host-parastie [ sic; parasite] coordination, for instance at the level
of ATP* metabolism or respiratory control, would then account for the fixation of PT throughout eukaryotic evolution. In
other words, the interaction of a few proteins at the host/parasite interface would be neuralgic [painful] for endosymbiosis
but would also lay the evolutionary grounds of apoptotic cell death. 49The significant possibility of host cell rejection of
foreign proteins and nucleic acid is envisaged by Kroemer to be the very facilitator for simultaneously establishing both
apoptosis and endosymbiosis! But, one does not need to be an expert biochemist or cell biologist to realise that this is a
case of story-telling; another example of turning contrary evidence into evidence for evolution, revealing a considerable
faith in veritable biochemical and cellular miracles. With a few crude brush strokes, the on-looker is expected to visualize
a picture in which exquisitely fine detail has also suddenly appeared on the canvas, without questioning where it came
from! Is the admirer of the work to conclude that these intricacies are somehow a property of the paint pigments? Yet
evolutionists must similarly suspend disbelief each time they indulge in these origin scenarios (choosing to overlook the
stupendous biochemical complexity that really exists). Unfortunately for them, even to peep beneath the lid of the Black
Box of the cell is to be confronted with a world of astonishing complexity, the simplest of whose apoptotic molecular
machines (proteins)not to mention their interactionsdemands an explanation, yet whose existence is simply
ignored.50 To say that the evidence of apoptosis points to the creation of a supremely intelligent designer is the most
rationaland this author would add, honestconclusion to which one could come.Kroemer realises that the piece-meal
(slow and gradual) evolution of apoptosis in unicells is conceptuallyeven theoreticallyvery difficult, to say the least:
the existence of PCD* [in unicellular eukaryotes] obviously cannot constitute a direct advantage for Darwinian
selection.51This is precisely why he argues that his endosymbiont hypothesis helps explain why these cells already have
apoptotic capabilities; i.e. apoptosis developed in the very moment that endosymbiosis occurred.
Blackstone and Green hypothesisHost cell manipulation by ATP and ROS
Blackstone and Green, who acknowledge the ideas of Kroemers paper, begin the introduction to their article thus,
Biological and biochemical mechanisms often seem dauntingly complex, suggesting to some that such mechanisms
could not have evolved. While this conclusion need not follow, the complexity nonetheless remains.52
It is highly significant that the first sentence is referenced to Michael Behes book in which he discussed the irreducible
complexity of several biochemical systems. 50 Blackstone himself wrote a scathing review of Behes thesis in a popular
biology journal, charging him with committing the basic logical error of argumentium AD ignorantiumi.e. arguing for the
truth of a proposition on the basis that it has not been proven false, or vice versa. 53 Therefore, it is transparent that this
paper by Blackstone and Green is a case of taking up the gauntlet that Behe threw down when his book was published.
By commencing their paper in this way, the authors are tacitly confirming that the phenomenon of irreducible complexity
applies to apoptosisone of the major conclusions drawn in an earlier paper by this author 1 although they attempt to
provide a rationale for how this might have been circumvented. Incidentally, Behe has ably responded to Blackstones
criticisms in some detail and the interested reader is referred to his paper.54
Axioms

As with Kroemer, these two authors uncritically assume endosymbiosis as fact.55 Hence, mitochondria are presumed to be
descendants of the eubacteria (protomitochondria)56 that were engulfed by primitive host cells. The authors also assume
that,
Caspases may be a relatively recent evolutionary addition to an older signalling pathway. 57
The protomitochondria are assumed to have been aerobes; as such, it is supposed that their oxidative phosphorylation
(using an electron transport chain) would have given them a survival advantage relative to the host cell (discussed below).
In contrast, the primitive host cell is said to have been anaerobic, this despite the overwhelming evidence for an oxygenrich atmosphere from the earliest times of Earth history,58-63 incorporating the Precambrian era during which the putative
first eukaryotes arrived on the scene. It is left to the reader to speculate as to where, exactly, endosymbiosis occurred.
Endosymbiotic role for ATP and ROS
In the presence of oxygen, such protomitochondria were allegedly better equipped to survive than the incipient host cell
due to a more rapid ATP synthesis and a more sophisticated defence against ROS. The endosymbiont theory requires
that, at some point in time, these protomitochondria became dependant for their survival on the host cells; why this
should have been necessary is not explained but one presumes that the author would lean towards Kroemers obligative
endosymbiosis event, described above. Again, the crucial questions surrounding how this endosymbiosis event occurred
are avoided. Instead, all of the authors subsequent ideas (discussed below) are focussed on the stage just afterwards:
the putative new eukaryote. They argue that from this point on,Natural selectionwould favor those protomitochondria that
influenced the hosts phenotype to enhance their own rate of replication.57These protomitochondria achieved this by
supposedly using their ATP and ROS to manipulate the host!Readers are invited to imagine a rapidly dividing host cell
(i.e. oxygen is present) with protomitochondria inside it:The large metabolic demands of the dividing host cells would
trigger a maximal rate of phosphorylation in the protomitochondria as long as supplies of substrate remained adequate,
thus shifting the redox state of the mitochondrial matrix in the direction of oxidation. 57The idea is that the consequent
production of ATP satisfies the host cells energy requirements, and the host also benefits from the fact that little/no
harmful ROS are produced (figure 3A). Thus, the host cells can divide unhindered which also provides more living space
for the protomitochondria.Conversely, low rates of host cell division would mean much less demand for ATP and would
cause the protomitochondria to enter a sort of resting state with oxidative phosphorylation just ticking over. However, the
protomitochondria would produce more ROS, causing lots of genetic mutations in the host cells, followed by their sexual
recombination (figure 3B). This is another example of evolutionists imagining advantageous mutations that
protomitochondrial cells is thought to have generated genetically novel hosts that thereby enhance the survival of the
protomitochondria!
Apoptosis as a vestige of host/endosymbiont conflicts?
Having sought to establish their case for a manipulative role of ATP and ROS, the connection of all the foregoing to
apoptosis is made:
The unexpected role of mitochondrial cytochrome c in programmed cell death may be an evolutionary vestige of levelsof-selection conflicts between protomitochondria and their hosts [emphasis added].65The over-used term vestige, by
evolutionists, immediately sets the alarm bells ringing. 66 However, let us critically examine this idea. In eukaryotic cells,
inhibition of the mitochondrial electron transport chain is known to enhance production of harmful ROS (e.g. superoxide
and hydrogen peroxide). If cytochrome c is released into the cytoplasm from mitochondria (as occurs in mammalian cells,
prior to the caspase activation stage of apoptosis) it catalyses further reactions involving these ROS, forming particularly
mutagenic ROS. Therefore, Blackstone and Green surmise that when protomitochondria were stressed they released
cytochrome c, thereby enhancing ROS formation, which, in turn, led to the mutation and genetic recombination of host
cells. They further speculate that this benefited the protomitochondria by creating a less stressful environment inside the
host! How or why such an outcome logically follows is not explained and is hardly self-evident.As with Kroemers
hypothesis, that of Blackstone and Green lacks pertinent details and barely mentions any of the complex apoptosis
machinery. They skip over these things (as well they might) and simply assert that:Later in the history of symbiosis, with
conflicts between mitochondria and the host cell largely resolved by the transfer of all but a fragment of the mitochondrial
genome to the nucleus, this ancient signaling [sic] pathway may have been co-opted into a new function, that of
programmed cell death in metazoans.65No attempt is made to suggest how the DNA instructions for this signalling
pathway passed from the mitochondrial matrix to the host cells genome (in spite of the many obstacles to their doing so),
or why this should have occurred. More importantly, although this paper purports to present a hypothesis for the evolution
of apoptosis, it is merely stated that this hypothetical metabolic signalling between host cell and protomitochondria was
possibly co-opted as a programme of cell death (Figure 3). Considering the bewildering array of tightly interwoven
components that constitute the apoptotic machinery,67 it seems almost farcical to postulate the interplay of these few biomolecules of protomitochondria and host cell as being the precursor of apoptosis!Moreover, the authors fail to give any
reasons, let alone offer a plausible scenario, for how their hypothetical mechanism for generating sexual recombination in
host unicells (the alleged novel habitats for endosymbionts) became fundamentally involved in the apoptosis
of multicellular organisms. The fact is that all such ideas remain essentially untestable, concerned as they are with events
that are imagined to have happened in deep time, as the authors themselves admit. Since cytochrome c production is a
key part of their hypothesis, it is pertinent that it does not appear to play a role in apoptosis signalling in the nematode
worm, Caenorhabditis elegans,68 unlike the situation in mammals and in yeast.69
Discussion and conclusions
Unicellular apoptosisevolved or designed?
It seems clear that death-programs do exist in unicellular organisms, both prokaryotes and eukaryotes, although even
evolutionists admit that these pathways show little or no homology with true apoptotic cascades described in multicellular
organisms.18 Bacteria are known to be able to take up naked DNA and (if this foreign DNA can be recognized by the host
cell DNA polymerase enzymes) replicate this together with their own DNA; this is a known mechanism for acquiring
antibiotic resistance, for example. Could this possibly explain the origin of apoptotic functionality? The author has been
unable to locate any papers where this case is argued but even a basic apoptosis-type mechanism (being irreducibly
complex70 ) would involve too many components to make this a plausible idea. The host of apoptotic molecular machines
involved in even the simplest eukaryotes (not to mention their pleiotropic interactions) renders any idea of piece-by-piece
addition of components by DNA uptake and transformation extremely improbable. Incorporating just a few of the
components of an irreducibly complex system into the cell would give it no survival benefit. Rather, it would be less fit
because resources would be wasted; natural selection operates to maintain genetic integrity and such transformed cells
would likely be weeded out. In addition, since there would be no selection against mutation in these acquired but unused,
apoptosis-component genes, their DNA sequences would almost certainly become scrambled over time.From a
creationist perspective, just as apoptosis is known to have numerous roles in multicellular creatures, including
humans,71 so the programmed deletion of unicells must be of functional benefitif not to the unicell itself, then to its

surviving clonal siblings. In bacteria at least, since stretches of DNA from damaged cells could conceivably be hazardous
(due to uptake and transformation), bacterial demise might better serve the population as a whole; i.e. such altruism by
the few, benefits the many by preventing potential genetic conflict between genes in the remaining bacteria. When
altruistic behaviour of the minority increases the inclusive fitness of the general population of closely related individuals,
this is termed kin selection. Accepting that apoptotic-style cell-deletion of unicellular organisms (including eukaryotes)
might be an example of kin selection makes perfectly good sense without giving any ground to evolution. An example of
natural selection, it may be, but support for the evolution of apoptosis, it certainly is notrather the implied preprogramming necessary for such apoptotic altruism in unicells is compelling evidence for a teleological view of these
organisms. It might be argued that the persistence phenomenon in bacteria 24 is a form of kin selection in this context,
potentially allowing gradually accrued death-program genes (by uptake and transfection) to be passed on to clonal
siblings even though the majority of cells die. However, unless these had survival value at every one of the dozens of
intermediate steps (towards a fully fledged apoptosis program), maintaining their sequence integrity and place in the
genome would be highly implausible for the reasons given at the end of the previous paragraph.There is even the
intriguing possibility that, from a design perspective, the designer has incorporated carefully regulated death-programs
into certain unicellular organisms in order to facilitate their symbiosis with the host. For instance, in cultured Fibrobacter
succinogenes (bacterial members of the gut flora of ruminant mammals), the lysis rate has been found to be influenced by
extracellular sugar concentration.72 When the sugar level is depleted, the bacteria produce an extracellular proteinase
enzyme which inactivates autolysins, thereby preventing bacterial death. However, in the presence of high sugar
concentration, despite the fact that F. succinogenes exhibit a logarithmic growth rate, many of the ruminal bacteria lyse,
akin to apoptosis. While the impact of bacterial lysis in ruminants is not entirely clear, this autolytic regulation (when sugar
levels are low) seems to decrease the turnover of stationary cells, increasing the availability of microbial protein in the
animals lower gut and thus, benefiting the host. This example of symbiosis, involving an apoptosis-style response and a
switching mechanism to boot, again supports a purposeful design explanation rather than one of random, gradualistic
processes.
Apoptosis falsifies evolution
Two hypotheses for the concurrent evolution of apoptosis and endosymbiosis have been critically reviewed and found
wanting. In a more recent paper, other evolutionists instead argued that the endosymbiotic bacterial ancestors of
mitochondria are unlikely to have contributed to the recent mitochondrial death machinery . 18 However, rather than
provide a rational alternative, they merely speculate that this complex mitochondrial apoptotic machinery derives from
mutated eukaryotic precursors for which they admit that there is no direct evidence! This lack of any direct evidence for
either idea leads to the unavoidableand for the creationist, unsurprisingconclusion that apoptosis evolution is
a belief that ignores empirical science. Rather, the challenge of the irreducibly complex nature of the apoptosis machinery
still stands in defiance of the keenest attempts of scientists to demonstrate otherwise. The very existence of apoptosis
effectively falsifies evolution.I am grateful to my colleague Monty White for reviewing the manuscript and to two
anonymous reviewers, whose comments helped to strengthen the arguments presented. I am particularly indebted to one
of the reviewers, whose suggestions inspired some of my discussion of unicellular death-programs.
Glossary
apoptosis

An active, non-inflammatory process (requiring energy) involving the programmed deletion of scattered
cells by fragmentation into membrane-bound bodies which are ingested by other cells.

ATP

Adenosine triphosphate is the predominant high-energy phosphate compound in all living organisms. It
plays a pivotal role in metabolic reactions and is basically the energy currency of cells.

cytochrome c

An iron-containing protein (with similarities to haemoglobin) that evolutionists consider to be one of the
most ancient biological molecules in living organisms. Cytochromes generally, are components of
electron transport chains in both mitochondria and chloroplasts.

endosymbiosis

Refers to the hypothetical origin of the first eukaryote. It is believed that aerobic bacteria and
photosynthesising bacteria were taken in by another bacterial cell (becoming the precursors of todays
mitochondria and chloroplasts) and established a mutually beneficial relationship.

eukaryote

A cell with a true, membrane-bound nucleus and subcellular, membrane-bound organelles.

metazoa

An old taxonomic word which is still used generically, as in this paper, to describe multi-celled
organisms; i.e. as opposed to prokaryotes and unicellular eukaryotes (including yeasts, protozoa and
others).

mitochondria

Organelles, found in all eukaryotic cells, which are the cells powerhouses. They are bound by a
double membrane, the inner of which is folded into plate-like structures called cristae. Mitochondria
house the machinery of the terminal electron transport chain including the cytochrome enzymes. They
also contain enzymes involved in oxidative phosphorylation.

MPT

Acronym for mitochondrial permeability transition (see text for explanation).

PCD

Programmed cell death; a synonym for apoptosis.

phagosome

The name given to the membrane that forms around any material that is engulfed by a cell (by a
process termed phagocytosis).

prokaryote

Any cell which does not have the diagnostic features of a eukaryote, principally the bacteria, but also
unicellular blue-green algae and other, more obscure unicellular organisms. Instead of a nucleus, there

is a circular duplex of DNA.

ROS

Reactive oxygen species. Also termed reactive oxygen intermediates (ROIs). These are short lived,
energetic and potentially toxic; e.g. the superoxide anion, O2, is harmful and tends to generate other
ROS.

transduction

In the context of biochemical pathways involved in cellular apoptosis or differentiation, this means the
conversion of a signal from one form into another.

AppendixCellular fate: apoptosis or necrosis? Is the distinction valid?

Kroemers paper6 gives examples of pathologies (i.e. not healthy situations) where the distinction between apoptosis and
necrosis is not always clear cut. It is known, for example, that Bcl-2 expression inhibits necrotic death in several
experimental models by inhibiting disruption of the mitochondrial trans-membrane potential. Consequently, the fate of a
stressed cell might depend on whether there is time for proteases to be activated (downstream of MPT) and to target
nuclear and cytoplasmic effectors of apoptosis. If the injurious agent results in very rapid ATP depletion, necrosis is the
outcome. This is compatible with this authors published observations that certain cytotoxic drugs that are used in cancer
chemotherapy induce apoptosis at low concentration but necrosis at higher concentration.77 A recent article by
Tavernarakis inNew Scientist re-examined the distinction between apoptosis and necrosis, 78 suggesting that there might
actually be, a core necrosis program that is activated upon injury and ravages the cell.However, as the article
reveals, investigations into necrosis at the molecular level have not revealed any specific genes or gene-products in
contrast to what we know of apoptosis.79 The author states that the necrosis-trigger culprits are the lysosomes (the oftdescribed suicide bag organelles of the cell) and implies that this fact has emerged recently; rather, this has been known
for many years. Nevertheless, new research has revealed that increased levels of Ca2 ions are what cause the lysosomes
to release their lethal enzyme cocktail and this obviously calls into question the entirely passive image of necrosis.This
and several other lines of evidence are causing researchers to consider whether the long-standing distinction between
apoptosis and necrosis might be too simplisticthey argue that the cells fate should be viewed as on a continuum
between programmed cell death and necrosis (catastrophic) cell death. However, the research findings that have
inspired this rethink have all involved cellular response todamaging chemicals, heat shock etc. There may sometimes be
a continuum between these modes of cell demise (e.g. morphologically), but as Tavernarakis states in the New
Scientist article, concerning necrosis,Unlike programmed cell death, no biochemical processes have evolved specifically
to carry it out.Necrosis is never a good thing \.80 The blurring of apoptosis and necrosis comes from studying the
morphology of cell attrition as a result of injurious agentsnot present in the once-perfect, world.
Nano-scale aligning tool used in the assembly of actin filaments
by Giovanie Adams
Figure 1. The structure of filamentous actin. A. A 14-subunit
filamentous actin as an isosurface, viewed along its
longitudinal axis. B. The structure of 3-subunits of filamentous
actin as an isosurface, viewed from the pointed end. The solid
lines pass through the center of each actin monomer, showing
the position of each actin monomer. The dashed paths travel
from the pointed end to the barbed end and show the position
of the next monomer with respect to the previous
monomer.Eukaryotic cells have systems for maintaining their
shape and all their movement, and for the transport of
molecules within them. Intra-cellular networks of fibers
assemble from actin proteins, and are an important part of
these systems. This network of actin filaments maintains cell
shape by forming a support structurean important
component for cell motilityand provides the paths for the
transport of molecules within the cell. The intra-cellular traffic of
molecules along these paths is also necessary for cell fission
(the division of one parent cell into two daughter cells). Actin
filaments are also a major component of the muscle fibers of
animals and are essential for the contractile apparatus of
muscles.1Actin filaments are made up of two long, twisted
chains consisting chiefly of tens to thousands of monomeric or
globular actin proteins. The assembly and disassembly of actin
filaments is controlled at each step by sets of actin-binding
proteins.1 The initial assembly of a short actin filament is a
rate-limiting step in filament assembly,2 and a subset of actin-binding proteins has been designed to overcome the relative
instability of these short filaments.3 This process of assembling short filaments of two or three subunits from monomeric
actin is called actin nucleation.The spire family of proteins is a family of actin-binding proteins that nucleates a pool of
actin monomers and prepares the product of nucleation for elongation. 4 Spire overcomes the relative instability of short
filaments made up of actin dimers and trimers, by helping to overcome the kinetic barrier to nucleation. 2 Spire also
attaches the newly synthesized actin filament to a membrane and aligns the filament to a nano-scale machine, a dimer of
formin proteins.The formin dimer then proceeds to extend the filament by adding actin monomers, assembling long
double-helical-twisted actin filaments. This complicated mechanism, of actin nucleation and preparation of the nucleated
product for elongation by spire, suggests the work of an intelligent designer who engineered it in an incredibly intricate
manner.
Structure of filamentous actin and spire proteins
Actin filaments appear in electron micrograph images as thin, dense lines of approximately 7 nm in diameter. 1 The
filaments are made up of two chains of monomeric actin with each actin monomer rotated 166.15 from the other. This
produces two helical chains that lie almost back to back and wind into a right-handed, double-helical structure with a very
long pitch with respect to its diameter (figure 1).Filamentous actin is polar and, therefore, the two ends of the filament are
different. Because of this polarity, the filaments often fit into other cellular structures in only one orientation, in a similar
manner to how many of the parts of man-made structures are assembled. Growth, the addition of monomeric actin to the

filament, occurs predominantly in one direction; the plus direction. This end of the filament (where the majority of growth
occurs) is known as the barbed end and the other end is called the pointed end (figure 1).

Figure 2. The domain organization of spire family of proteins. The kinase non-catalytic C-lobe domain (KIND), the four
WASP-homology domains 2 (WH2), the Spir-box (S-box) and the FYVE zinc finger domain of spire.
Comparison of the spire family of proteins to other proteins has identified seven domains (figure 2). At the carboxyl-end of
the protein is a domain called the FYVE zinc finger, 5 and adjacent to this is a domain called the Spir-box (S-box). 6 A
cluster of four, evenly spaced domains, called the WiskottAldrich syndrome protein homology domain 2 (WH2), are
located in the central region of the protein, 7 and a domain called the kinase non-catalytic C-lobe domain (KIND) is at the
amino-end.8 The FYVE zinc finger domain may bind to the membrane, thus, FYVE may anchor the newly assembled actin
filament to a membrane.5 The S-box is a potential binding site for Rab GTPase. 9 The four WH2 are connected by linker
regions; they bind monomeric actin and are sufficient to nucleate actin4. Finally, KIND has a high affinity for the FH2
domain of formin.10
Nucleation of monomeric actin
Figure 3. Proposed mechanism for actin filament
nucleation by spire. From left to right the schematic
shows actin nucleation by spire. Scene one shows
the binding of an actin monomer to the carboxylend WH2 domain of spire and the binding of the
FH2 domains of a formin dimer to KIND of spire.
The FYVE and S-box domains of spire and the
other domains of formin are not shown in these
scenes. Scene two shows a second actin monomer
binding to the second most carboxyl-end WH2
domain. Scene three shows the stabilization of two
actin monomers by the two most carboxyl-end WH2
domains and the intervening linker region. Scene
four shows the stabilization of four actin monomers
into a helical structure by the WH2 domains and the intervening linker regions. Scene five shows the formation of an actin
filament by four actin monomers binding to the single helical structure formed by four actin monomers. Scene six shows
dissociation of KIND from the FH2 dimer and rapid polymerization of the actin filament by the FH2 dimer. All molecules
are represented as isosurfaces.It remains to be established exactly how a pool monomeric actin is nucleated by spire, but
it may occur via the following steps (figure 3). 4,10 The process is initiated by KIND of spire binding to the FH2 domains of a
formin dimer; this step inhibits formin but enhances spire activity. Next, each WH2 domain of spire binds to an actin
monomer. The two WH2 domains closest to the carboxyl-end and the intervening linker region align and stabilize the
bound actin monomers. Actin monomers bound to other WH2 domains are then aligned to the initial structure by the
action of the other linker regions. This results in the formation of a single-stranded helical polymer of four actin monomers.
Following these steps, a second helical polymer consisting of another four actin monomers binds on the other side of the
helix by self-polymerization. Once an actin filament of about eight monomers has formed, KIND of spire dissociates from
the FH2 domains of the formin dimer. This exposes the FH2 domains to catalyze the incorporation of actin monomers into
the filament resulting in rapid polymerization of the barbed end of the actin filament by the formin dimer.Actin filaments are
highly dynamic structures that rapidly assemble and disassemble.2 To prevent disassembly of the newly synthesized
filament from the pointed end, spire caps the pointed end of the filament. 4Although spire performs machine-like functions
during actin nucleation, the main role of spire is probably to act as a nano-scale alignment tool. A broad definition of a tool
is an entity used as an interface between two or more entities, that makes it easier for one entity to act upon the other. In
the case of the function of spire in actin nucleation, spire probably acts as an interface between the actin monomers and
formin dimer entities, facilitating the alignment of these entities. That is, the four WH2 domains of spire align four actin
monomers, and the KIND of spire probably aligns the four monomers to the FH2 domains of a formin dimer. Therefore,
spire helps to overcome the kinetic barrier to actin nucleation, stabilizing actin dimers and trimers.2
Conclusions
Actin filament nucleation by spire is yet another example of the complexity of the automated systems needed to assemble
the nano-scale structures and machines found within living cells. Spire is likely to be an elaborate nano-scale alignment
tool with some machine-like functions. This would make spire far superior to modern tools engineered by humans. All this
attests to the work of an intelligent designer who engineered these structures and machines in an incredible intricate
manner.

New DNA repair enzyme discovered

by Jonathan Sarfati
Published: 13 January 2010(GMT+10)

Figure 1: Bacillus cereus alkylpurine DNA glycosylase alkD bound to

DNA containing a G-T mismatch.
(www.ncbi.nlm.nih.gov/Structure/mmdb/mmdbsrv.cgi?uid=84991)
Our information is stored on the famous DNA double helix molecule.
This is so efficient that just five round pinheads full of DNA could hold
all the information of the earths entire human population. 1 Just one of
these pinheads would have 2 million times the information content of a
2 TB hard drive. And each of our 100 trillion cells has 3 billion DNA
letters (callednucleobases) worth of information.2But chemically, DNA
is actually a very reactive molecule (and RNA is even more so), so its
highly implausible that it could have arisen in a hypothetical primordial
soup.3 Indeed, about a million DNA letters 4 are damaged in a cell on
a good day. One common form of DNA damage is called alkylation
this means a small hydrocarbon group is attached to one of the
letters, and there are many places for the attachment. This changes
the shape enough so it can no longer fit into the double helix. This can
prevent DNA replication or reading the gene.So living creatures must
have elaborate DNA repair machinery. University of Chicago biologist
James Shapiro points out that:all cells from bacteria to man possess a
truly astonishing array of repair systems which serve to remove
accidental and stochastic sources of mutation. Multiple levels of
proofreading mechanisms recognize and remove errors that inevitably
occur during DNA replication. cells protect themselves against
precisely the kinds of accidental genetic change that, according to
conventional theory, are the sources of evolutionary variability. By
virtue of their proofreading and repair systems, living cells are not passive victims of the random forces of chemistry and
physics. They devote large resources to suppressing random genetic variation and have the capacity to set the level of
background localized mutability by adjusting the activity of their repair systems. 5For example, there is base excision
repair: special enzymes called DNA glycosylases run down the DNA molecule, detect the damaged letter, grab it, put it
in a specially shaped pocket, then chop it out. Then other enzymes repair the resulting gap.Scientists at North American
universities have discovered another ingenious repair enzyme in bacteria, called AlkD. 6 This has a very different structure.
It works by flipping a positively charged damaged basehighly unstableand the one its paired with, from the inside to
the outside of the helix. Then they are both detached, and the gap filled. Understanding these enzymes could lead to
more effective chemotherapy.Evolution has a major problem in explaining repair machinery. Natural selection requires that
the information selected for can be reproduced accurately. But without an already functioning repair mechanism, the
information would be degraded quickly. Furthermore, the instructions to build this repair machinery is encoded on the very
molecule it repairs, another vicious circle for evolution.7
ATP synthase: majestic molecular machine made by a mastermind
by Brian Thomas
Figure 1. The whole ATP synthase machine
with individually manufactured protein
subunits each labelled with Greek letters.
H+ ions (protons) flow through a special
tunnel in ATP synthase, as the arrow
indicates. This induces mechanical motion,
forcing the axle and base to spin together
like a turbine. Nearly 100% of the spinning
momentum is converted to chemical energy
in the formation of ATP molecules! Three
ATPs are produced for every 10 protons.
(Adapted from Kanehisa Laboratories,
<www.genome.jp/kegg>)Life depends on an
incredible enzyme called ATP synthase, the
worlds tiniest rotary motor.1 This tiny protein
complex makes an energy-rich compound,
ATP (adenosine triphosphate). Each of the
human bodys 14 trillion cells performs this
reaction about a million times per minute.
Over half a body weight of ATP is made and
consumed every day!All living things need
to make ATP, often called the energy
currency of life. ATP is a small molecule
with a big job: to provide immediately usable
energy for cellular machines. ATP-driven
protein machines power almost everything
that goes on inside living cells, including manufacturing DNA, RNA, and proteins, clean-up of debris, and transporting
chemicals into, out of, and within cells. Other fuel sources will not power these cellular protein machines for the same
reasons that oil, wind, or sunlight will not power a gasoline engine.Logical thinking about an automobile engine leads us to
think that only a clever person (with mind and will) could make a machine that converts energy from one form to another
for the purpose of moving a car.2 The machine shows orderly, non-random proportions and clever use of interdependent
parts that are the right size, shape and strength to work together for an overall purpose. The same inference from
machine back to maker is validly inferred from machines found in nature back to their designer.3 Everyone knows that a
painting comes from a painter, because the painting shows specified complexity, or a complex and recognizable pattern

that is not a property of the paint. That is, the paint molecules do not spontaneously organize themselves into a portrait of
Mona Lisa, for example.4
Figure 2: Ribbon diagram of a top view of the head portion of ATP synthase, called F 1-ATPase. It has six protein
subunits, and consists of three active sites, where three ATP molecules are formed for each full rotation of the axle. The
very top end of the axle is just visible, leaning against the upper right inside wall of F 1. Cellular machinery constructs the
head, after which it self-assembles onto the base.
(www.rcsb.org/pdb/explore/images.do?structureId=2F43)ATP
synthase occurs on the inner membranes of bacterial cells, and the
innermost membranes of both mitochondria and chloroplasts, which
are membrane-bound structures inside animal and plant cells (see
figure 1).ATP synthase manufactures ATP from two smaller
chemicals, ADP and phosphate. ATP synthase is so small that it is
able to manipulate these tiny molecules, one at a time. ATP
synthase must convert some other form of energy into new ATPs.
This energy is in the form of a hydrogen ion (H +) gradient, which is
generated by a different whole protein system to ATP
synthase.5 Hydrogen ions pour through ATP synthase like wind
through a windmill. This comprises a positively charged electric
current, in contrast to our electric motors, which use a negative
current of electrons.ATP synthase is a complex engine and pictures
are necessary to describe it. Scientists use clever techniques to
resolve the exact locations of each of many thousands of atoms
that comprise large molecules like ATP synthase. 6 This protein
complex contains at least 29 separately manufactured subunits that
fit together into two main portions: the head (figure 2) and the base
(figure 3).7 The base is anchored to a flat membrane (figure 1) like
a button on a shirt (except that buttons are fixed in one place,
whereas ATP synthase can migrate anywhere on the plane of its
membrane). The head of ATP synthase forms a tube (figure 2). It
comprises six units, in three pairs. These form three sets of docking
stations, each one of which will hold an ADP and a phosphate. ATP
synthase includes a stator (stationary part), which arcs around the
outside of the structure to help anchor the head to the base (figure
1).
Figure 3: A ribbon diagram (side-view) of the base of ATP
synthase, called the F0 portion. It is made of twelve helical protein
subunits arranged in a circle, forming the wall of a tube. This tube
provides a tunnel across the membrane (not shown) in which it is
anchored .
(www.rcsb.org/pdb/explore/images.do?structureId=2CYD)Now we
can look at the amazing, efficient way that this marvelous micro-machine works. Notice in figure 1 a helical axle labeled
in the middle of the ATP synthase. This axle runs through the center of both
the head and base of ATP synthase like a pencil inside a cardboard toilet paper
tube.Here is the magic: When a stream of tiny hydrogen ions (protons) flows
through the base and out the side of ATP synthase, passing across the
membrane, they force the axle and base to spin. 8 The stiff central axle pushes
against the inside walls of the six head proteins, which become slightly deformed
and reformed alternately.9 Each of your trillions of cells has many thousands of
these machines spinning at over 9,000 rpm.10The spinning axle causes
squeezing motions of the head so as to align an ADP next to a phosphate,
forming ATP in bucket loads. Many other cellular protein machines use ATP,
breaking it down to ADP and phosphate again. This is then recycled back into
ATP by ATP synthase. Lubert Stryer, author of Biochemistry adds,
the enzyme appears to operate near 100% efficiency .11
Figure 4: Overall 3-D molecular structure of ATP synthase rotor by Stock et al 7,
minus the stator.
This motor is incredibly high-tech design in nano-size.
Evolutionary scientists have suggested that the head portion of ATP synthase
evolved from a class of proteins used to unwind DNA during DNA replication.12
However, how could ATP synthase evolve from something that needs ATP,
manufactured by ATP synthase, to function? This bizarre suggestion underlines
the role of our beliefs in how we interpret origins. Evolutionists are often driven by
a bias which they do not admit: methodological naturalism. This is the
assumption that the processes which explain theoperation of phenomena are all
we can use to describe the origin of those phenomena. Creation scientists,
looking at the same ATP synthase phenomenon also have a bias: supernatural
origins are possible in a theistic universe. The big question is: whose bias is
correct? I submit that a creation bias is clearly true because it makes sense
according to the principles of causality.
We might also consider that ATP synthase is made by processes that all need
ATPsuch as the unwinding of the DNA helix with helicase to allow transcription
and then translation of the coded information into the proteins that make up ATP
synthase. And manufacture of the 100 enzymes/machines needed to achieve this needs ATP! And making the
membranes in which ATP synthase sits needs ATP, but without the membranes it would not work. This is a really vicious
circle for evolutionists to explain.
What about the characteristics of the One who designed the amazing abilities of the ATP synthase nano-motor? Keep in
mind that the smaller a machine is, the more ingenious the effort needed to build it.

The immunoglobulin heavy chain gene family and gene duplication

by Yingguang Liu
The immunoglobulin heavy chain gene family is an example of a gene family with irreducibly complex regulation.
Therefore the family could not have been produced by past gene duplication followed by diversification. However, the
tandem array of homologous genes within the locus predisposes the chromosomal region to unequal crossing-over and
polymerase slip, causing copy number polymorphism in the human population. Duplications early in history may account
for some of the paralogues fixed in the human race, but functional complementation between paralogues and deliberate
regulation mechanisms both point to intelligent design.
Figure 1. The immunoglobulin G molecule is made of two longer heavy
chains and two shorter light chains. Each chain has a variable region and
a constant region.
Gene duplication is a specific type of mutation that increases the size of
an organisms genome. Conversely, most genes of advanced organisms
have nonallelic homologues (paralogues) within the same genome,
forming gene families. The degree of sequence homology and functional
similarity between paralogous genes vary from family to family. This paper
focuses on distinguishing between historical duplication and common
design at the time of creation. This review examines the immunoglobulin heavy chain gene family to illustrate which
members of this family may be products of past gene duplication, and which are evidently not.
A family with irreducibly complex regulation cannot be produced by gene duplication
Immunoglobulins, also called antibodies, are proteins consisting of light chains and heavy chains, each having a variable
region and a constant region (figure 1). They are classified according to the constant region of the heavy chains. In
humans, there are five types (isotypes) of immunoglobulins (IgA, IgG, IgD, IgE, and IgM) corresponding to the five types
of heavy chains (, , , and ). Each antibody molecule is encoded by a tandem array of repetitive elements on
chromosome 14 (figure 2a, top). The highly repetitive V, D and J genes encode the variable region. 1One each of the V, D
and J genes joins one of the downstream constant region genes (designated in figure 2 by the lowercase Greek letters).
IgM and IgDwhose heavy chains are encoded by the and genes, respectivelyare expressed on the surface of
nave B cells before any immune response. An immune response first causes the gene to be expressed as secreted
IgM. Later the gene is skipped and one of the downstream g, e or genes are expressed through a process called class
switch recombination. The process occurs on the DNA level in antibody-producing cells (B lymphocytes).Class switch
recombination requires transcription and is accomplished with a unique set of cis-acting elements and trans-acting factors.
In human and mammals, it requires the simultaneous presence of multiple enhancers (E elements) within the locus of the
family, as well as an immunoglobulin (I) promoter and a switch (S) region upstream of each heavy chain gene (figure 2a,
top). Vertebrates lacking the S regions (e.g. catfish)
do not undergo class switch recombination, even
though they may possess linked heavy chain
genes.2Deletion of one of these cis-elements in mice
resulted in impaired class switch or no class switch. 35
The organization of the I promoters, the S regions,
and the structural genes suggests that a unit of I
promoter S region structural genewas duplicated
multiple times to produce the heavy chain gene
family. However, the irreducibly complex nature of
the family precludes its origination as a single unit. It
requires at least two I-S-s units to achieve class
switch recombination (figure 2a and b). Lacking a
function, the cis-regulatory elements would have
degenerated before gene duplication could ever
occur.Of the trans-acting factors involved in class
switch, the key enzyme, activation-induced
deaminase (AID), is B-cell specific and, therefore,
serves only to produce functional antibodies.
Although it is also used in another aspect of antibody
synthesis (somatic hypermutation of V regions),
class switch requires a functional domain of the
enzyme
that
is
not
required
for
hypermutation.6,7 There exists strong evidence that
other class switch-specific enzymes/cofactors are
required to guide , and switches to
achieve
the
appropriate
immune
response.7,8 Therefore multiple genes must be
simultaneously added onto the genome outside of
the heavy chain gene family for class switch
recombination to occur, further underlining the
irreducible complexity of the system.Class switch
also requires multiple enzymes involved in DNA
rearrangement, excision repair and nonhomologous
end-joining (NHEJ) pathways (figure 2 legend).
Although these enzymes are not class switchspecific, and therefore could theoretically have evolved independently, their expression in B cells and coordination with
the class switch-specific factors necessitates careful design. For example, the excision repair enzymes, such as uracil
DNA glycosylase (UNG), are activated in B lymphocytes to cause mutations instead of repairing them, both during class
switch recombination and during somatic hypermutation.Moreover, development of a functional class switch mechanism
must occur concomitantly with structural and functional diversification of the various immunoglobulin isotypes. The five

human immunoglobulin isotypes have different structures, tissue distributions and functions. Switching without
diversification is futile, and diversification without switching would fail to express the downstream genesanother aspect
of irreducible complexity.
Figure 2. (a) Organization of the human immunoglobulin heavy chain locus. V, D, J: variable region genes; , , , , :
constant region genes; I: I promoter; S: switch region; E: enhancers. The region upstream of the heavy chain gene,
including the I promoter, I exon, S region, as well as the 5 end of exon, is expanded to show an RNA transcript which,
after splicing, induces single-stranded loops in the DNA. The loops, stabilized by both the transcript and stems formed
from inverted repeats, are targets of the activation-induced deaminase (AID). (b) AID recognizes the loops, deaminates
cytosine and, in conjunction with uracil DNA glycosylase (UNG) as well as an AP (apyrimidinic site) endonuclease,
introduces double stranded breaks in two S regions. The broken ends from different S regions are subsequently joined by
the nonhomologous end-joining mechanism, eliminating the intervening sequences as a circle.All components of an
irreducibly complex system must be created with their designated functions simultaneously. However, Darwinian
mechanisms require gene duplication and subsequent diversification, with regulatory elements developed either before or
afterwards. Gene families such as the immunoglobulin heavy chain genes are not products of gene duplication because:
the system could never have functioned as a single-copy gene;
elements that regulate the entire family (such as the enhancers of the heavy chain locus) do not exist in lower
organisms;2
trans-acting factors and family-controlling cis-acting elements cannot be produced by duplication of a structural gene with
its linked regulatory elements;
degeneration of duplicates happens faster and with more certainty than meaningful diversification.9
Copy number polymorphism results from duplication or deletion
Homologous gene clustering predisposes a chromosomal region to unequal crossover or polymerase slip, causing
variation in copy numbers within the gene family. Due to the highly repetitive nature of the immunoglobulin heavy chain
genes, this locus displays considerable copy number polymorphism among the human population. For example,
duplication of the heavy chain genes is 22% in Mongoloids, 10% in Caucasians and 5% in Negroids.10 Most of the
duplications encompass multiple genes, such as -2, 1-4, 1-, etc., and some triplications exist, such as of 1-
among the Japanese. Deletions, mostly of 4 or 1, are much rarer (1.53.5% depending on race). Duplications cause no
or slight elevation in serum immunoglobulin levels, while deletions cause more significant lowering of the corresponding
immunoglobulin levels.11 The lower frequency of deletions in the population is presumably due to negative selection, while
the evolutionary significance of duplications is unclear, but presumably not as significant as are deletions.
Beneficial gene repetitions may be produced either by duplication or by design
Figure 2 also demonstrates that the --- cluster is repeated once within the locus. There are functional differences
between corresponding genes in the two clusters. The products of the 2 and 4 genes in the downstream cluster (IgG2
and IgG4) are significantly less potent in activating complement or binding Fc receptors than those of the 1 and 3 genes
in the upstream cluster (IgG1 and IgG3). While the downstream gene encodes all the IgE, the upstream gene is a
pseudogene (i.e. a gene that does not code for a functional protein).There exist apparently healthy individuals with
homozygous deletions of some heavy chain genes, especially of the genes in the downstream cluster. 12,13 It is therefore
possible that the two clusters were produced by duplication of one original cluster. Subsequently, the upstream gene
pseudogenized, while the downstream and genes subfunctionalized (lost parts of their functions due to degenerative
mutation). The inability of IgG2 and IgG4 to activate complements offers the human body a noninflammatory type of
protection, contrary to IgG1 and IgG3 which cause inflammation through complement activation. Further, IgG2 and IgG4
may have a role in regulating the proinflammatory effects of IgG1 and IgG3. This duplication/subfunctionalization is
therefore beneficial to the human race. However, the duplicates could not have fixed in mankind through natural selection
because the increase in fitness was conceivably small and the new duplicate would have quickly drifted out of the
population.14 Although a duplication/subfunctionalization scenario can be speculated for the origin of the two clusters,
there always remains the possibility that they were created in the first people to produce subclasses of immunoglobulins
with diverse functions. In fact, immunoglobulins encoded by the two heavy chain gene clusters are differentially regulated.
While class switch to the upstream cluster is activated by a subset of immune cells called Th1 cells, class switch to the
downstream cluster is activated by another subset of T lymphocytes (Th2 cell). There is increasing evidence pointing to
deliberate regulation mechanisms that facilitate class switch to the downstream heavy chain genes in situations where a
noninflammatory response is more desirable. For example, IgA1-producing B lymphocytes can be induced to secrete IgA2
when they are recruited from the systemic circulation to the mucosa of the colon. 15Likewise, we cannot rule out the
hypothesis that the two linked genes within each --- cluster indicate another duplication event before duplication of
the entire cluster, although all four genes may have been created together to encode functionally different IgG
subclasses.
Conclusion
Just as multiple V, D and J genes are needed to produce the diversity required for adaptive immunity, 1 the immunoglobulin
heavy chain genes were designed as a family to produce various immunoglobulin isotypes for different functions. The
unique regulatory elements of class switch recombination must have been created along with the structural genes.
However, the tandem array of homologous DNA sequences in the locus allowed errors in crossover and/or replication,
producing duplications and deletions during human history, for good or for bad.
Lipid rafts: evidence of biosyntax and biopragmatics
by J.G. Andrew
Information theorist Werner Gitt has proposed that information is best understood within a multidimensional framework.
This framework has five dimensions: the statistical, syntactic, semantic, pragmatic and apobetic. Current evolutionary
views of information and biology are largely restricted to the statistical dimension. Alex Williams has recently argued that
the inheritance of biological information should be understood within this multidimensional Gittian framework and has
emphasised that extra-nuclear cell structure may contribute to the information content of cells. In this article I show that
the emerging concept of lipid rafts in cell membranes represents both a syntactic and a pragmatic information structure
within the cell. As such they illustrate the existence of a higher order of organisation within the cell than traditionally
understood and provide strong support for the Gittian creationist concept of how information is structured and inherited in
biology.
* Terms marked with an asterisk are defined in the Glossary at the end of this article.

Figure 1. Diagram showing some of the lateral

organisation involved in lipid raft domains. The
outer
plasma
membrane
leaflet
has
a
heterogeneous arrangement of cholesterol and
sphingolipids. They appear to congregate into
patches that have distinct biophysical properties.
This enables control of the lateral organisation of
membrane-bound proteins based on their
phosphorylation and activation states and the
presence or absence of specific lipid anchors.
Werner Gitt has described information in terms of
statistics, syntax, semantics, pragmatics and
apobetics.1 Within this theory, the lower levels of
information are hierarchically arranged in order to
express the higher levels: i.e. statistics, syntax,
semantics and pragmatics are arranged in such a
way as to express the apobetics or purpose, within the organism .In an analogy with linguistics, this is equivalent to words,
sentences, paragraphs and context being arranged in such a way that the intended meaning(endowed by the author) may
be conveyed, understood and acted upon.This contrasts sharply with the evolutionary position, which denies that any
purpose exists in biology. In the selfish gene theory, for example, the appearance of purpose is considered to be nothing
more than a consequence of natural selection. Genes are believed to be the descendants of a hypothetical primordial
replicator that originated life and produced all of lifes variety simply by natural selection favouring different combinations
in different environments. Information in this theory is nothing more than a serendipitous accumulation of statistical errors
that happen to have survival value for the genes.Conversely, the Gittian information framework, and creationists in
general, interpret genetics and other biological processes in terms of their roles in enabling purpose to be fulfilled. If Gitts
information framework is applicable to biology then there should be evidence of information operating at the levels of
syntax, semantics and pragmatics within biological systems. Alex Williams has recently described these features within a
key area of genetics.2 He proposes that the nucleotide sequence found in DNA represents statistical information, the
amino acids specified by codons represent a semantic arrangement, the sequence of amino acids represents a syntactic
arrangement, and the function of gene products represents a pragmatic arrangement. This description is appropriate but
is incomplete because gene products do not simply operate pragmatically but also form syntactic, semantic and pragmatic
arrangements within ever more complex processes that make up the whole living organism.In addition, Williams also
highlighted the importance of structural modes of inheritance, including cellular architecture, as providing a basis for the
stasis that is a major characteristic of inheritance and which forms part of the apobetic purpose of reproducing after like
kind.3In this article, I will show that the emerging concept of lipid rafts and their function as cell-signalling platforms can be
understood as both a syntactic and a pragmatic arrangement of proteins and lipids and that this is derived from an
interaction between genetic and epigenetic factors*. As such, they constitute a molecular context or cellular paragraph
within which meaning is conveyed. Within the so-called semiotic triad of object, sign and interpreter, 4,5 lipid rafts form part
of the interpretive machinery of the cell that allows the meaning of the symbol to be defined and connected to an
appropriate response. Understanding the interpretative machinery of the cell is an important step in beginning to describe
the way in which non-sentient entities, such as cells, translate abstract information into action without conscious will. The
interpretative machinery of the cell closely resembles Gitts description of operational information.1 It is therefore apparent
that lipid rafts form part of an operational information structure within the cell.
Lipid rafts
The term lipid raft is used to describe areas of cell membranes that are selectively enriched in cholesterol,
sphingolipids* and signalling proteins* . Since Simons and Ikonen brought attention to the possibility of their existence in
19976 there has been an explosion in efforts to detect and characterise these curious structures. 7 They are believed to
form functional domains on the outer leaflet of cell membranes that selectively include or exclude particular proteins
depending on their levels of phosphorylation* , activation or the presence or absence of important lipid tails. In addition,
lipid rafts are believed to somehow connect to intracellular signalling and structural proteins and influence their activities.
Lipid rafts may thus concentrate members of different signalling cascades* within a discrete area of the cell allowing the
components of the pathway to be in the right place at the right time.7
Do lipid rafts exist?
Although there are several lines of evidence that argue for the existence of lateral arrangements of cholesterol,
sphingolipids and proteins,8some of the techniques used to study them may artificially congregate these elements (e.g.
detergent extraction)9,10 while other techniques have produced mixed results (e.g. fluorescent microscopy). 11
14
Nevertheless, the weight of evidence seems to suggest that lateral organisation of the outer leaflet of the cell membrane
occurs in vivo.
What are lipid rafts comprised of?
The plasma membrane* is composed of two phospholipid layers with the hydrophobic hydrocarbon* chain tails facing
each other and the hydrophilic* polar* head groups facing the internal and external surfaces. Membranes have long been
known to also contain proteins, cholesterol, sphingolipids (including the class of glycosphingolipids known as
gangliosides), and other components. It has only recently been proposed that these various components do not exist
completely randomly or evenly across the plasma membrane surface but that there is a level of lateral organisation. Thus
there is evidence that sphingolipids and cholesterol are relatively concentrated in rafts in the outer leaflet of the
membrane and that proteins may have a preference for either the raft or non-raft regions depending on the biophysical
properties of the protein in question. See figure 1.This lateral arrangement is controlled in part by the levels of cholesterol
and glycolipids, including gangliosides. Gangliosides are glycosphingolipids with one or more sialic acid* residues and are
often used as markers of lipid raft domains. Their biosynthesis and regulation is complex (indeed it constitutes another
cellular process that displays syntax, semantics and pragmatics!) and it is worth noting that they are not directly encoded
in the genome since they are not proteins. They require the cooperative action of several enzyme complexes to construct
them, direct them to their appropriate location and to maintain their presence in the outer plasma leaflet. 1522 The
hydrophobic ceramide* tails, the hydrophilic sugar residues and the high levels of cholesterol mean that lipid raft domains
have distinct physicochemical properties compared with neighbouring non-raft membrane.
What do lipid rafts do?
Lipid rafts have been implicated in a plethora of cellular processes. The common mechanism thought to underlie their
function in most cases is their ability to preferentially concentrate or exclude molecules of cell signalling cascades. 7 This

ability is a result of the different biophysical properties of raft and non-raft membrane and the changes in chemical
properties of proteins upon activation, phosphorylation or other events. For example, immunoglobulin E (IgE) signalling
forms part of the allergic immune response and is activated when IgE binds to Fc receptors*(FcR) present on the plasma
membrane of mast cells* . This probably increases the partitioning of the FcR into lipid rafts with subsequent cross-linking
of FcR-bound IgE leading to phosphorylation of the FcR. This can then initiate secondary downstream signalling events
including the release of histamine* .23 Other signalling pathways are proposed to operate in a similar manner with lipid
rafts acting as concentrating platforms.2426
Lipid rafts as biosyntax
Within written language systems, syntax refers to the set of rules that govern the way in which semantic objects (words)
are positioned with respect to one another, how they may be joined together and how they may be modified while still
retaining the intended meaning. For example the statement: Does John like Rosalyn? differs from the statement John
does like Rosalyn! only in word order and punctuation. However, the change in order of the first two words changes the
meaning (semantics) and purpose (apobetics) of the statement from being a question designed to elicit an answer to
being a statement that answers a question. There is no statistical difference between the information content of the two
statements (since they contain the same letters, words and spaces) but they differ at the level of syntax and consequently
at the semantic, pragmatic and apobetic levels.
Figure 2. Lipid rafts as biosyntax. An example of how
lipid rafts contribute to molecular syntax within cells. The
binding of NRG to ErbB4 does not exert any
physiological changes under conditions in which lipid
rafts are not allowed to form. When lipid raft domains
are allowed to form, NRG-ErbB4 translocates into rafts
along with other key components of the ERK signalling
pathway including Shc and Grb2. This allows the
otherwise meaningless arrangement of receptors and
signalling proteins to exert a physiological function and
thus have semantic and pragmatic meaning. In this
manner lipid rafts function as a component of a
molecular syntax that defines the arrangement and
meaning of symbols within a code.In a similar manner,
lipid rafts can be seen to contribute to a molecular
syntax within the cell. For example, it has been shown
that in cultured cortical neurons* , the activation of the
transmembrane signalling molecule ErbB4 by its ligand
neuregulin (NRG) induces the translocation of ErbB4 and adaptor signalling molecules* into lipid raft domains from nonraft domains.27 Furthermore, this translocation into lipid rafts is required for NRG to exert its downstream effects, including
activation of the protein kinase ERK. When lipid rafts are disrupted, the effects of NRG-induced activation of ErbB4 are
not seen.This situation is analogous to the sentences discussed above where the same words mean something different
when the syntax is altered. The two conditions, ErbB4 in lipid rafts and ErbB4 not in lipid rafts, differ only in the relative
position of the relevant components of the system and not in the presence or absence of these components. The lipid
rafts do not change the statistics of the information in the signalling pathway. Instead, they form a context in which the
arrangements of the members take on defined meanings. Thus in these cultured cortical neurons, lipid rafts provide the
molecular syntax in which NRG-induced activation of ErbB4 can bring about particular effects that do not occur in their
absence (figure 2).
Lipid rafts as biopragmatics
Another example involving ERK signalling illustrates the way in which lipid rafts can act in the role of molecular
pragmatics. In linguistics, semantics deals with the range of possible meanings of words and in biology it refers to the
range of possible functions of molecules. Syntax, as we have seen above, deals with the different ways that components
can be arranged to activate those meanings. Pragmatics, on the other hand, is all about context. An author identifies
a particular intended meaning by the context in which he/she uses a word, and in biology, the particular function of a
molecule is also specified by the context. In this example, the protein molecule CD45 has two quite different cellular
functions depending on its arrangement within or outside of lipid rafts.
Figure 3. Lipid rafts as biopragmaticsdefining the effects of
the protein CD45. Some CD45 is located outside lipid rafts and
is linked to ERK activation and stimulation of IL-2 production.
There is also some CD45 that is found within lipid raft domains
and is linked to antagonism of IL-2 production. The molecular
basis of this remains unclear, but it may be that different sets of
signalling proteins are preferentially linked to raft and non-raft
membrane domains. This arrangement constitutes a molecular
pragmatics that allows a single molecule to take on different
meanings depending on its context.Zhang and colleagues
have shown that membrane compartmentalisation of the
transmembrane protein tyrosine phosphatase CD45 is an
important feature in the regulation of T cell activation* in the
mouse.28 These authors found that a proportion of CD45 is
found within lipid rafts and a proportion is found outside lipid
raft domains. The raft-localised CD45 was found to antagonise
IL-2 production (IL-2 is a cytokine involved in several important
immunological processes) while the CD45 located outside lipid rafts played a role in promoting IL-2 production via ERK
activation.There is evidence that during the stimulation of T cells, raft-associated CD45 is excluded from these domains.
This presumably leads to less inhibition of IL-2 and greater stimulation of IL-2 via ERK. Thus a single molecule can
have opposite effects (stimulate or inhibit IL-2 production) depending on the molecular context within which it operates
(i.e. within or outside of lipid rafts) (figure 3).The mechanisms linking CD45 signalling to IL-2 stimulation or inhibition are
not well understood, but it is likely that the raft and non-raft membrane domains contain different proteins comprising
distinct intracellular signalling cascades. Lipid rafts act as a component of this molecular context and so contribute to the
pragmatics of the cell.

Inheritance of lipid rafts

Lipid rafts are components of cell membranes comprising of lipids and proteins. As such, they are passed on from parent
cell to daughter cell during cell division as part of the cellular architecture that each cell inherits. In addition, the proteins
that contribute to their formation and maintenance (e.g. the enzymes involved in ganglioside biosynthesis) are inherited
both epigenetically as proteins in the lipid bilayer, and genetically as genes on chromosomes in the nucleus. This means
that lipid rafts are passed on fully established and operational in an epigenetic manner, while the relevant information for
their maintenance is passed on in a genetic manner. This illustrates how inheritance of the organised membrane structure
forms part of the cellular basis of stasis of the created kind, as Williams has hypothesised. 3
Cell autonomy
Syntax, semantics and pragmatics are used in language by intelligent beings according to conventions that
allow purpose or apobetics, to operate. This requires an ongoing conscious tie between the language symbols and their
referents. But the control and communication that operates within and between cells goes on autonomously without an
ongoing conscious tie between symbols (e.g. DNA sequence, cell-signalling proteins, etc) and their referents (e.g. protein
sequence, effectors of cell-signalling pathways, etc). Instead, the ties between object, sign and interpreter are established
and maintained by an inherited information-based control system that operates according to biochemical laws.Could such
information-based cellular life have arisen via the selfish gene replicator mechanism? It would seem not, because a gene
can only use its information if it is part of a mechanism that contains transcription, translation and implementation facilities
(i.e. a cell). So when lipid rafts are observed to function as syntactic and pragmatic information structures in cells it points
to an intelligent rather than random original source.
Conclusion
The existence and operation of the lipid rafts in cell membranes can be seen as an implementation of syntax and
pragmatics in order to enable apobetics to operate in the biological world in the absence of an ongoing conscious tie
between object, sign and interpreter by the cell. It therefore supports the notion that the multidimensional framework of
information of Werner Gitt is applicable to biological systems. Since this understanding of information requires an
intelligent source, observing it in biological systems suggests their origin in an intelligent source rather than random,
unintelligent statistical processes.The Gittian view of information emphasises apobetics as the ultimate controlling factor in
information-processing systems. This highlights the need to understand biology from the perspective of apobetics-down
(i.e. the purpose-orientated creationist perspective), rather than from the perspective of statistics-up (i.e. the geneorientated evolutionary perspective). There is a fresh need for a derived apobetic theology of biology which provides a
purpose-oriented framework for understanding biological processes and the nature of bioinformation. Such a theology
would need to define the key purposes of creatures and relate these to biological processes, and would provide a
philosophical foundation for studying the information present in organisms.
Glossary
Adaptor
signallingProteins and other molecules that connect extracellular signals to intracellular signalling
molecules:
cascades.
Ceramide:
an N-acyl sphingosine which forms the lipid portion of glycosphingolipids.
Cortical neurons:
nerve cells from the cortical region of the brain.
Epigenetically:
information inherited via genes is inherited genetically. Information inherited in any other way is
inheritedepigenetically.
Fc receptors:
the portion of an immunoglobulin molecule that initiates its effector functions (e.g. complement
activation, opsinisation, etc).
Histamine:
Derived from an amino acid, histamine is a chemical messenger involved in initiating many of
the events involved in acute inflammation.
Hydrocarbon:
an organic molecule which consists of carbon and hydrogen atoms.
Hydrophilic:
having an affinity for water. By contrast, hydrophobic means lacking an affinity for water.
Mast cells:
an immune cell residing in connective tissue. It releases several substances, including
histamine, involved in acute inflammatory reactions.
Phosphorylation:
the addition of a phosphate group to a protein molecule. This often alters protein structure and
function and acts as a regulatory switch.
Plasma membrane:
the lipid layer which comprises the outer wall of each individual cell.
Polar:
the hydrogen bonding end of the phospholipid molecule, which is hydrophilic.
Sialic acid:
A 9 carbon acidic sugar. It is present on many glycolipids and glycoproteins and is responsible
for much of the negative charge on cell surfaces.
Signalling cascades:
series of reactions which occur as a result of a single stimulus.
Signalling proteins:
proteins which contribute to the transfer of information from cell to cell.
Sphingolipids:
structural lipids derived from sphingosine (a long chain amino alcohol).
T cell activation:
under appropriate conditions, stimulation of inactive immune T-lymphocytes results in them
becoming activated and capable of assisting in the immune response.

Mitochondriacreated to energize us
by David Demick
General
diagram
of
a
mitochondrion.
Mitochondria vary in size and shape, from nearly
spherical to long threadlike filaments.

Mitochondria are small, membrane-bound organelles serving as energy generators in eukaryotic cells. Most cells have
hundred to thousands of them, depending on their energy needs. Mitochondria are very good at what they dothey
generate about 95% of a cells energy in the form of adenosine triphosphate (ATP) by oxidizing pyruvate (a by-product of
anaerobic glycolysis) to CO2 and water. They are ovoid to filamentous in shape, generally ranging from one to seven
micrometers in length (about the same size and shape as small bacteria). Since the discovery that mitochondria possess
their own DNA, it has been frequently theorized that mitochondria evolved from ancient bacteria ingested by larger cells.
This is known as the endosymbiont theory of mitochondrial origin. Sometimes it is stated boldly:More than a billion years
ago, aerobic bacteria colonized primordial eukaryotic cells that lacked the ability to use oxygen metabolically. A symbiotic
relationship developed and became permanent. The bacteria evolved into mitochondria, thus endowing the host cells with
aerobic metabolism, a much more efficient way to produce energy than anaerobic glycolysis. 1Sometimes it is stated more
cautiously:In the endosymbiont theory, the ancestor of the eukaryotic cell (we will call this organism a protoeukaryote)
ispresumed to have been a large, anaerobic, heterotrophic prokaryote that obtained its energy by a glycolytic pathway.
Unlike present-day bacteria, this organism had the ability to take up particulate matter . The endosymbiont
theory postulates that a condition arose in which a large, particularly complex, anaerobicprokaryote took up a small
aerobic prokaryote into its cytoplasm and retained it in a permanent state [emphasis added]. 2Whichever way it is stated, it
is given an aura of authority and certainty by its frequent repetition in writings on cell biology. Many students find it
convincing. However, like many evolutionary ideas, it may look solid from a distance, but gaps appear on close
scrutiny.The evidence for the endosymbiont theory revolves around selected similarities between mitochondria and
bacteria, especially the DNA ring structure. However, these similarities do not prove evolutionary relationship. There is no
clear pathway from any one kind of bacteria to mitochondria, although several types of bacteria share isolated points of
similarity. Indeed, the scattered nature of these similarities has left plenty of room for a less-publicized direct evolution
theory of mitochondrial origin, in which they never had any free-living stage. 3 There is enough diversity among the
mitochondria of protozoa to make evolutionists wonder if endosymbiotic origin of mitochondria occurred more than once.4
Mitochondrial DNA
The endosymbiont theory implies that there should be considerable autonomy for mitochondria. This is not the case.
Mitochondria are far from self-sufficient even in their DNA, which is their most autonomous feature. Mitochondria actually
have most of their proteins coded by nuclear genes, including their DNA synthesis enzymes. For example, human
mitochondria have 83 proteins, but only 13 are coded by mtDNA (mitochondrial DNA). Even those proteins which are
coded by mtDNA often have large subunits that are coded by nuclear DNA. These nuclear-coded mitochondrial proteins
must be labelled and transferred from the cytoplasm across two membranes. This intricate, hand-in-glove working
between mtDNA and nuclear DNA presents a major difficulty for evolutionists. They have yet to propose a reasonable
mechanism by which so many genes could be transferred intact (along with appropriate labelling and control
mechanisms) to the nucleus.Plants and other lower creatures may have more mitochondrial genes than the higher
animals do, but they still fall far short of the number necessary for free-living existence. Plants have also been found to
have much more non-coding mtDNA than the higher animals. Referred to as junk DNA by evolutionists, it is held to have
been eliminated by evolution from the mitochondrial genomes of the higher animals, to the point that humans have
virtually no non-coding mtDNA. Evolution seems to be remarkably unpredictable in its handling of junk DNA, allowing it to
accumulate haphazardly in the nuclear DNA of higher animals and man, but efficiently eliminating it from mtDNA. It
doesnt seem reasonable for evolutionists to have it both ways.There are more important differences between mtDNA and
nuclear or prokaryotic DNA. The main one is that the genetic code for mtDNA differs from the standard DNA code in slight
but significant ways. Why? Evolutionists make much of the universality of the genetic code, saying that it offers strong
support for common descent of all living things. If this is trueif the code is so highly conserved in evolution through over
a billion years and millions of speciesthen even a few exceptions to the rule are hard to explain. (On the other hand,
from a design standpoint the answer may lie in the simpler protein synthetic machinery served by mtDNA, which uses
fewer tRNAs, and is less specific in codon recognition.) Lack of introns is another important difference. The higher
mtDNA has no introns, whereas nuclear DNA and some lower mtDNA do have them. Again, the bacteria from which
mitochondria are supposed to have evolved also lack introns. Thus, were asked to believe that the pre-mitochondrial
bacteria sporadically evolved introns as they became primitive mitochondria, and then lost them again as eukaryotic
evolution ensued. As evolutionists grapple with the biochemical details, the endosymbiont theory becomes more and
more cumbersome and vague.5
Intracellular control
As alluded to earlier, mitochondrial numbers are controlled within each cell by energy needs. They can also travel within
cells on cytoskeletal microtubule rails wherever energy is needed (near the ribosomes in pancreatic zymogen cells, near
the proton pumps in gastric acid-secreting cells, etc.). 6 This complex intracellular control is highlighted by a common
pathological abnormality in which certain body cells become bloated by an oversupply of mitochondria. These cells,
known to medicine as oncocytes, are packed by malformed or malfunctioning mitochondria, in which various mutations
have been detected.7,8 Also, when mutated mitochondria derived from a maternal oocyte populate all of the bodys cells,
the results can be devastating. A whole spectrum of degenerative multisystem diseases associated with mitochondrial
mutations has been described recently, with more being discovered.9,10 Such diseases tend to affect tissues most heavily
dependent on aerobic metabolism, such as neural and muscular tissue. These observable phenomena underscore the
harsh reality that random changes in mitochondria or microbes do not produce complex new structures and regulatory
systems, but rather disease and death.It should also be pointed out that the engulfing of bacteria by larger cells is one of
the commonest phenomena in nature, happening countless times each hour. Yet, nothing really like the formation of
mitochondria has ever been observed. There may be rare modern examples of endosymbiosis between two different
types of cells, such as the Chlorella algae within green paramecia. Also, infecting or parasitic microbes can persist for a
time inside of larger host cells due to encapsulation or other protective factors. Still, these events are far from the radical
biotransformation demanded by the endosymbiont theory, and no one untainted by evolutionary preconceptions would
ever dream of classifying mitochondria as once-separate life forms, as some evolutionists have suggested. It is essentially
an evolutionary miracle, assumed to have happened in the past, but never seen or duplicated in the present.
Chloroplasts
Furthermore, if we accept this naturalistic miracle of mitochondrial origin we are forced to conclude that the same miracle
happened repeatedly. Evolutionists also postulate an endosymbiotic origin for chloroplasts, the organelles of
photosynthesis in higher plants. Chloroplasts have their own DNA, once again with a ring structure. They are similar in
some respects to present-day photosynthetic bacteria. However, because of biochemical variety among chloroplasts (like
the mitochondria), evolutionists are once again forced toward the unlikely conclusion that their endosymbiotic origin
occurred more than once!According to this endosymbiont hypothesis, eucaryotic cells started out as anaerobic cells
without mitochondria or chloroplasts and then established a stable endosymbiotic relationship with a bacterium, whose

oxidative phosphorylation system they subverted to their own use . Plant and algal chloroplasts seem to have been
derived later from an endocytic event involving an oxygen-evolving photosynthetic bacterium. In order to explain the
different pigments and properties found in the chloroplasts of present-day higher plants and algae, it is usually assumed
that at least three different events of this kind occurred [emphasis added].11Although it is correctly admitted here that the
endosymbiont scenario is actually only a hypothesis, it is presented as the only possibility. However, as shown above, the
fine print admits that assumption and speculation are major components of this idea.Why do mitochondria and
chloroplasts have their own DNA? Evolutionists believe that it is a source of cellular inefficiency, and that evolution has
been slowly phasing out cytoplasmic DNA over time. (This raises the obvious question of why there is any mtDNA left at
all, to which the evolutionary response is that the process of elimination is either incomplete or arrested.) However,
viewing mtDNA as inefficient may just be a reflection of our own ignorance of the fine details of mitochondrial function.
Deeper knowledge may show that manufacture of certain mitochondrial protein subunits on-site is very efficient, just as
the energy-harnessing chemistry of the mitochondrial enzymes has been shown to be.
Conclusion
Given the enormous leaps of biochemical and genetic integration which are demanded by the endosymbiont theory,
creationist skepticism is entirely justified. There is no compelling reason to believe it unless one has already decided that
evolution is true. The creationist model, holding that structures may look similar because they were designed to do similar
jobs, is a more reasonable way to view the miracle of mitochondria.
Serial cell differentiation: intricate system of design
by Shaun Doyle
Published: 4 June 2008(GMT+10)
Image Wikipedia
Figure 1. The process of hematopoiesis (the
generation of blood cells) is an example of the serial
cell differentiation process.
Single celled organisms replicate as fully functional
cells, and they maintain cellular integrity through a
system of direct epigenetic inheritance,1 or cell
memory. Some tissues in multicellular organisms
proliferate in the same way. However, the majority of
tissues in adult multicellular organisms dont.Most
tissues in mature multicellular organisms replicate via
a method called serial differentiation.2 Cells go
through a series of differentiation stages as they
duplicate, ending in a fully differentiated cell, which
eventually dies and passes out of the system, or is
recycled by apoptosis (programmed cell death). There
are three different types of cells in this system: stem
cells, a class called transient amplifying cells (TACs)
and fully differentiated cells.
Serial differentiation
Stem cells
The undifferentiated cells are the only ones in this differentiation process that are self-renewing, i.e. they produce
daughter cells that are exactly like the mother cell. These cells have the capacity to divide and change into many different
types of cells. They are also very important during embryonic development, where new cell types are constantly
needed.3 These stem cells are kept relatively few in number, and the cell lines proliferate through the differentiation
process.
Transient amplifying cells
The daughters of stem cells do more than just self-renew; they differentiate into different kinds of cells. However, they
dont change into fully differentiated cells immediately; they change into a class called transient amplifying cells (TACs).
While TACs divide; unlike stem cells, TACs do not self-renew. Rather, the daughter cells of TACs are one stage further
along the differentiation process than the mother cell. These cells amplify the number of cells that will eventually become
fully differentiated from the original stem cell that they started from.
Fully differentiated cells
A particular stem cell goes through a number of cell division events and the differentiation process of the TAC stage to
produce fully differentiated cells. These are the mature cells that carry out the different jobs of the tissues, such as blood
cells (figure 1), reproductive cells and epithelial cells. These cells no longer divide or differentiate, and once they have
served their purpose, they are deleted from the system and their components are recycled. 4
Designed for maintenance
This is a rather elaborate system to conjure up if you just want to maintain tissues! It is also metabolically expensive
because not only do the mature cells require nutrients, but so do the stem cells and TACs. Therefore, youre feeding cells
that dont actually do anything in the body except replicate. So why bother using so much energy?As Pepper et al. point
out, the aim of this process is to separate the self-renewing and active proliferating properties of cells into different
groups.2 This severely limits the number of duplications that any one cell line will undergo, which limits the possibility of
mutational damage taking hold in a particular tissue.This system actively works against natural selection of individual cells
in favour of tissue integrity to suppress somatic evolution, which is the change that the body is subjected to due to
mutation and selection within the bodys cell population. Pepper et al. comment:We hypothesize that this is achieved in
animals by compartmentalizing self-renewing tissues such that one cell population (stem cells) undergoes self-renewal,
while another (TACs) undergoes active proliferation. If no cell population combines both these necessary elements of
somatic evolution, somatic evolution is thereby suppressed.The stem cells are maintained as a small and quiescent
population through slow self-renewal. The maintenance of the self-renewing population at low levels militates against
selection of highly proliferative strains of stem cells.The later stages of the differentiation process are focussed on
proliferation, but they dont self-renew. Each duplication event moves the daughter cells along the next stage of
differentiation, until the cells are shed after they have become fully differentiated.While it would cost less energy to just
have self-renewing mature cells, it would result in the quick death of the organism if something went wrong in comparison
to serial differentiation. Less energy would be used up because the body would not have to support stem cells and TACs,

but only fully differentiated cells. However, there is a much higher chance a mutation that increases the reproductive
success of a particular cell would gain a hold in such a setup when compared to serial differentiation. Therefore, the
benefit of longevity far outweighs the energy cost incurred for maintaining the system.
Evolution of multicellularity and serial differentiation
Pepper et al. also comment on the prospects of serial cellular differentiation aiding the transition from unicellular to
multicellular life:It is believed that multicellular organisms could not have arisen or been evolutionarily stable without
possessing mechanisms to suppress somatic selection among cells within organisms, which would otherwise disrupt
organismal integrity. Here, we propose that one such mechanism is a specific pattern of ongoing cell differentiation
commonly found in metazoans with cell turnover, which we call serial differentiation. 5They believe that this transition
from unicellularity to multicellularity is controlled by epigenetic alterations:Thus, our results support the suggestion that
epigenetic inheritance played a central role in the transition from unicellular to multicellular life by helping to control
selection among the cells of the newly emergent multicellular individual. 5However, both serial differentiation and the
multicellular organism have to be assumed for this to work. At best it suggests how multicellularity persisted, but it does
not suggest its origin.There is a fundamental evolutionary conflict in a multicellular organism: cellular selection vs bodily
integrity. Generally, natural selection at the cellular level will favour those cells that are better at reproductive competition
and survival. However, if those cells are allowed to proliferate in an uncontrolled manner in a multicellular organism, it will
inevitably disrupt the organisms bodily integrity, and harm or kill the organism. 6This inevitably kills these fitter cells too
because they cause the host to die.Cancer is a prime example. A cancer is essentially a mess of excessively proliferating
cells within a multicellular organism. In an environment (the organism) with limited resources, such cells will naturally outcompete normal cells because normal cells generally dont proliferate indefinitely. The cancer cells outstrip the normal
cells for resources and take over the system. However, this leads to malfunction in the organism, and if left untreated, will
inevitably kill the organism.At the organismal level, selection will favour those traits that preserve bodily integrity, which
seeks to control proliferation of cells beyond what is necessary. Pepper et al. confer:Multicellular organisms could not
emerge as functional entities before organism-level selection had led to the evolution of mechanisms to suppress celllevel selection.7However, this leads to a conundrum for the evolutionist: how do multicellular creatures evolve from single
celled organisms when cellular selection is diametrically opposed to organism-level selection? A single cell seeks to
proliferate more than its competitors; the multicellular organism seeks to control such proliferation to what is needed at a
higher level of organisation. This can be seen in the process of apoptosis as well:Even today, apoptosis serves an
essential role in terms of cellular altruism. It helps to ensure that an organisms genetic integrity is not compromised, by
removing some somatic cells that have sustained irreparable, genetic mutations. Crucially, apoptosis also helps to
maintain a species genetic integrity, by eliminating aberrant germ cells that would otherwise carry intact but faulty genes
into the next generation.8The system of serial differentiation is designed to enhance bodily integrity, not reduce it. The
system has to be in place before it can be selected for, yet organism-level selection cannot take over without measures
such as serial differentiation in place. The very existence of this system argues against the evolution of multicellularity.
Conclusion
Serial differentiation is an essential system for the maintenance of mature multicelled organisms. It serves to separate the
self-renewal and proliferative stages of cell division, which limits the effect mutations have on tissues. Evolution cannot
explain the origin of the system, and neither can it explain the origin of multicellularity.
Searching for needles in a haystack
by Royal Truman
The variability of amino acids in polypeptide chains able to perform diverse cellular functions has been shown in many
cases to be surprisingly limited. Some experimental results from the literature are reviewed here. Systematic studies
involving chorismate mutase, TEM-1 lactamase, the lambda repressor, cytochrome c and ubiquitin have been
performed in an attempt to quantify the amount of sequence
variability permitted. Analysis of these sequence clusters has
permitted various authors to calculate what proportion of
polypeptide chains of suitable length would include a protein able
to provide the function under consideration. Until a biologically
minimally functional new protein is coded for by a gene, natural
selection cannot begin an evolutionary process of fine-tuning.
Natural selection cannot favour sequences with a long term goal
in mind, without immediate benefit. An important issue is just how
difficult statistically it would be for mutations to provide such initial
starting points. The studies and calculations reviewed here
assume an origin de novo mainly because no suitable genes of
similar sequence seem available for these to have evolved from.
If these statistical estimates are accepted, then one can reject
evolutionary scenarios which require new proteins to arise from
among random gene sequences.
Proteins are chemically bonded chains of amino acids (AAs) (figure 1). All living organisms on Earth depend on
specialized services these provide. One of 20 different AAs1 can be placed at each residue position of the polypeptide,
offering an immense sequence space of possible variants. Most alternatives are biologically useless.Many scientists,
including several prominent agnostics, are persuaded that Darwinian trial-and-error could not have produced the
necessary genetic infrastructure for life to be possible.2 The fraction of all possible AA chains having any biological value
is miniscule. Requiring hundreds of unrelated combinations of amino acids forming polypeptides, in the right proportion
and same place, for the simplest of autonomous life forms to be possible, is indistinguishable from demanding a miracle.
Additional requirements for other classes of biochemicals found in all cells compounds the improbability. The minimal
requirements for a putative initial evolutionary starting point via naturalist means cannot be justified from what is known
from chemical and thermodynamical principles.
Figure 1. Condensation of amino acids leads to polypeptide polymers.
Biologically functional polypeptides are called proteins. The R group side chains
define the amino acids.
We will limit this discussion to real, biological, genetically based organisms and
exclude speculative constructs such as abstract replicators, 3 RNA-world
arguments4 and chemical hyper-cycles.5 Even if such hypothetical structures

could exist at some point, a transformation to life as we know it, based on the genetic code, would confront us with the
issues discussed here anyway.For a primitive organism to evolve and increase the range of functions performed, many
new kinds of genes are needed. It has been proposed that different genes may have arisen from duplicated copies 6 on the
same genome, which diverged through mutations and ended up coding for novel proteins. I believe this concept has
limited explanatory potential. The number of mutational trials needed to arrive at truly novel genes is prohibitive given the
great differences observed among families of unrelated proteins. Nevertheless, divergence of paralogous genes
(duplicates on the same genome) and lateral gene transfer remain key explanatory notions in the evolutionary toolkit. This
is justified, since we will see here that de novo origin of proteins in living organisms is statistically not plausible. An
analysis of duplicate genes and lateral gene transfer will follow in later papers.Just how difficult would it be for mutations
to generate new valuable genes by chance? It depends for one thing on what fraction of random amino acid chains would
provide new useful functions with enough advantage for natural selection to act upon. The conclusions from several
published studies have been summarized in Table 1. The technical details are discussed in the Appendices as an aid to
those wishing to understand the original literature.Three general approaches are described in the literature to examine the
proportion of sequences able to provide a particular protein function: (1) random chains of amino acids are generated to
see if useful variants appear; (2) existing protein sequences are mutated to see how much change is tolerated; (3)
sequence variability across organisms is compared. Especially interesting are those cases where no, or few, similar
protein classes are available from which the highly restricted version could plausibly have derived. This is an area I and
other non-evolutionists are currently researching.
Testing of random polypeptide sequences
In this approach, many polypeptide sequences are randomly generated and tested for some property related to that of
functional proteins. This literature711 will not be reviewed at this time. I have searched the literature for years without
success for an example in which anything useful for the organism was claimed using this approach. Examples of, for
example, stability to proteolysis 7 or cooperative denaturation,8 even crude catalytic effects,12 are certainly chemically
interesting, but these do not yet provide plausible starting points for Darwinian selection to take place. It is important to
keep in mind that expressed genes cost considerable energy resources, 13 and mere analogy to properties used by real
proteins is not something natural selection can act upon.
Table 1. Probability a random polypeptide of
suitable length would produce various functional
proteins. (Click on table to see larger version)
A new gene which produces a polypeptide serving
no useful function which is merely harder to
degrade, will not provide a selective advantage. In
fact, being unable to degrade and recycle such
building material in a regulated manner14 would be
disadvantageous. Furthermore, it appears that the
potential for interference in existing processes would simply be introduced. Crude enzymes accelerate the forward and
backward reaction by lowering transition state energies (figure 2), and could simply facilitate decomposition of useful
metabolites in the absence of a carefully tailored series of linked biochemical steps (see figure 3). Generally, several
biochemically coupled reactions with multiple enzymes need to be carefully engineered to work together, with regulatory
feedback inhibition, for metabolic processing to work. In this paper we are only considering the difficulty of obtaining a
single protein such as an enzyme, and not that of obtaining whole, functional new networks or gene systems.
Systematic modification of parts of a protein
Figure 2. Enzymes are chemical catalysts which accelerate the rate of a reaction by
lowering an energy barrier. Both the forward and backward reaction is accelerated,
but the proportion of materials which result after equilibration is unchanged.
Figure 3. An enzyme would accelerate decomposition of chemical species B back to
A faster, hindering evolution of a metabolic network able to produce C. Only until
energetically favourable coupled reactions (species B to C) are in place would the
enzyme be useful. But without the enzyme in the first place, the necessary B
materials would generally not be available.
In this approach, a method is needed to efficiently kill off individuals whose mutant
protein is not functional. The sequences (usually the base pairs of the gene) present
in
the survivors are then determined. There are various experimental setups.
In
one approach, the original gene is deactivated and the researcher seeks to generate
an alternative functional sequence. The protein coded for has a key function, such as
being part of a metabolic process to synthesize a necessary biochemical. The
researcher keeps the test organism alive by providing the lacking nutrient. Different
variants of the defective gene are made available, via a plasmid or other vector, and
the nutrient is then denied. Survival indicates a working variant is present.
In another strategy, mutated genes able to protect against a poison or virus are introduced into a host and the sequences
from the survivors are analyzed.
Comparison of sequences across taxa
Gene sequences for similar functions across different organisms can be compared in an effort to estimate how much
variety is tolerated. Patterns can often be identified, such as that only amino acids possessing similar polarity or size are
allowed at a given position on the chain. If the data set is large enough, some rough statement of number of alternatives
should be possible.15Arriving at reliable estimates for a given protein is very difficult. An average protein consists of over
300 AAs, each of which could be affected by mutations with any of 20 possible AAs at each location. Furthermore, one
would have to check which mutations are compatible with other mutations on the same gene. Therefore, it is worthwhile to
examine carefully the assumptions which the authors use in the estimates reported. This is the purpose of the
Appendices. Readers interested in the details are encouraged to read these and to examine the original papers.The
studies discussed in the Appendices explain the basis for the experiments performed to estimate what proportion of amino
acid sequences of a particular length would led to the protein function studied. The published numerical estimates are
summarized in Table 1, which is the take-home message of this paper.The astronomically small values are not the
probabilities of generating a near-optimal protein or gene, but the chances of generating a starting point before the natural
could be invoked. In one paper Dr Heisig and I,16 and in another Drs Scherer and Loewe, 17 independently estimated the
maximum number of polypeptide alternatives which may have been generated using the most optimistic assumptions
possible. The current evolutionary models assume life has existed for about four thousand million years, leading to a large

number of organisms which may have generated new genes. Very short generation times, high mutational rates and huge
populations were assumed16,17 to provide the largest number of mutational attempts possible to favour the evolutionary
scenarios. We estimated that the maximum number of polypeptide variants coded for genetically which could ever have
been generated is about 1046.1046 is the maximum number of attempts available from which the evolutionist must account
for all useful proteins produced. Everyone agrees that the vast majority of random polypeptide sequences would be
biologically worthless, but the open question is roughly what fraction might be useful.Systematic studies of gammaProteobacteria18,19 show that of about 14,158 gene families present, more than half (7,655) are represented by only one
gene. In other words, most of the genes are very different from each other. The common ancestor believed to have lived
over 500 million years ago18 did not provide an ancestral evolutionary starting point for all these gene families according to
these authors.Thousands of genes unrelated to both the gamma-Proteobacteria examined and to all others whose
sequences are found in public databases must be accounted for. Perhaps they evolved in other organisms and were
transferred laterally. However, it was reported19 that 42.5% of the 7,655 single-gene families were unrelated to any other
sequences in all current databases.The general pattern I observe in the literature is that as the number of organisms
sequenced increases, ever more unique genes are discovered which are unrelated to any other known genes. Stover et
al. were unable to find homologs for 32% of the Open Reading Frames (probable genes) identified in the
bacterium Pseudomonas aeruginosa.18,20The general view among evolutionists is that gene duplication has led to new
genes among eukaryotes, but lateral gene transfer (LGT) does this for prokaryotes. The latter represent the vast majority
of organisms. But invoking LGT does not solve the problem. Thousands of novel genes, unrelated to others, must come
from somewhere.This analysis shows that none of the examples summarized in Table 1 can be expected to have arisen
by chance processes and then fixed in a population. 10 46 random attempts are far too few to satisfy the miniscule
probabilities calculated. Furthermore, LGT from unknown taxa is often invoked as the origin of novel genes. This implies
that the value of 1046 is far too great, since we must subtract the statistical contribution during hundreds of millions of
years by the populations assumed to being the LGT recipient. And there are hundreds of additional proteins in all freeliving organisms even less likely to have arisen by chance than indicated by most of the probabilities reported in Table 1. I
intend to quantify more examples in future papers. That all of these actually arose naturalistically is not
reasonable.Systematic genomic comparison studies are leading to the view18,19 among evolutionists that a core of about
100 unrelated genes are present in all organisms. These alone were insufficient to support a free-living cell, but after
countless mutations or gene eliminations all evidence for them has been lost. In any event, it is simply inescapable that at
some point a large number of unrelated genes need to have arisen from among random sequences.I hope Table 1 will
provide a good basis for quantitative discussions as to whether design or natural processes best explain the origin of life
and the complexity observed. Evolutionary theory assumes that a series of genes evolved from preceding ones. Where
the original ones came from fades into the misty zones of speculation. This line of reasoning only makes sense if chains
of successive genes, with novel functions, can be built using statistically plausible jumps. The analysis of sequence
variability reported here suggests that huge statistical gaps often separate islands of functional proteins from potential
starting points.
Appendix A
AroQ chorismate mutase21
Figure
4.
AroQ-type
chorismate
mutase,
entry
1ECM.pdb in the Protein Data
Bank,
<www.rcsb.org/pdb>.
Displayed with RasTop. The
protein is a symmetrical
association of two 93 residue
domains
Figure
5.
AroQ-type
chorismate
mutase,
entry
1ECM.pdb in the Protein Data
Bank,
<www.rcsb.org/pdb>.
Displayed with RasTop. Only one of the 93 residue symmetrical domains is shown.
The probability of obtaining a functional Chorismate mutase from among 95 amino acid chains was reported in Table 1 as
being 1044. The details of this experiment are summarized in this appendix.In the experiment 21 plasmids containing
variants of AroQ chorismate mutase (figures 4 and 5) were introduced into an Escherichia coli strain (KA13). The purpose
of the encoded protein is to catalyze the Claisen rearrangement of chorismate to phephenate (figure 6), 22 which is an
essential step in the biosynthesis of the amino acids tyrosine 23 and phenylalanine.24The DNA sequence was modified in
two regions which code for -helices, engineered in such a manner that only any of eight natural amino acids could
appear in those regions. Specifically, every polar amino acid in the original wild type from Methanococcus jannaschii was
randomly replaced by one of the four polar natural amino acids, and each non-polar position by one of the four non-polar
amino acids. Several positions known to be critical for the enzymatic function were left unchanged.The modified bacteria
were transferred to a minimal medium lacking tyrosine and phenylalanine. The AroQ DNA sequences of the surviving
colonies were analyzed and the number of unique variants determined. The authors then extrapolated to conclude that
the chances of obtaining a minimally functionally AroQ would be about 1 out of 5 x 1023 sequences from among those
generated. In the generated set, however, portions of the three -helices were replaced by only 8 selected amino
acids25 although any of 20 natural amino acids could show up in nature.Professor Scherer pointed out 26 that in these
sequences the hydropathy requirements to produce the folds had already been designed into the experiments. This
improved the chances of obtaining a functional mutant significantly. To take this into account, he proposed an additional
average probability of 0.5 per residue position for the 77 amino acids involved in the -helices. This leads to a more
accurate probability of 0.577 x (2 x 1024) = 1047 of obtaining AroQ functionality from among all random polypeptides of the
same length as the wild type. Another author suggested a value of 10 53.27,28Dr Axe has pointed out21 that most proteins
are near an optimal state and this needs to be taken into account in these kinds of experiments. Typically certain amino
acids must be present and in a very demanding 3-dimensional structure to create an enzymatic active site. Replacing one
of these residues can be deadly. The rest of the protein must provide a stable scaffold, which holds the critical portions of
some amino acids in ideal locations in three dimensions, for the enzyme to work. Modification in the position of some
bonds by a few tenths of an Angstrom is often unacceptable in some regions of a protein.

Figure 6. AroQ chorismate mutase is an enzyme

used during synthesis of amino acids
phenylalanine and tyrosine.2224 (Click on figure to
see larger version)
However, enzymes typically fold reliably into one
of
the
most
thermodynamically
stable
configurations, and this final state is so stable
that alternate amino acids often have little effect.
One could replace a small number of amino acids
at different positions in a large number of
separate experiments and then incorrectly
conclude that all these substitutions are always
permissible in the presence of each other. This is
the error of overlooking context dependence,2931 also discussed32 in this journal. Taking these factors into account
indicates that the estimated proportion of 1047 is probably too large.
Appendix B
TEM-1 penicillinase27
The probability of obtaining a functional TEM-1 penicillinase from among 153 amino acid chains was reported in Table 1
as being 1077. The details of this experiment are summarized in this appendix.-lactamases are enzymes which protect
bacteria from penicillin-like antibiotics. TEM-1 penicillinase is a typical class A -lactamase consisting of 263 residues, and
includes two structural domains. The whole protein, once folded, reveals several features which include nine strands,
twelve helices and three chains. The larger 153 amino acid domain was studied by Axe. 27How many sequences would
provide the enzymatic function? All possible mutations would require 20 153 different genes to be examined, which is not
realistic. Axe shows how careful reasoning does permit extrapolation to a reasonable estimate based on far fewer
mutants.Protein folding is a concerted effort involving multiple portions of the polypeptide concurrently. Interactions
between the side chains of different amino acids bring portions together in an orchestrated order, which leads reliably to
the same three-dimensional, final, stable folded pattern. These considerations imply that the number of distinct folding
patterns is relatively small33 and in the order of 103 to 104. This places constraints on the properties of amino acids which
may be substituted via mutations.Alignment of 44 large-domain sequences from different organisms, obtained from public
databases, allowed each of the 153 positions to be characterized in terms of the properties of the amino acids tolerated
there: hydrophobic, hydrophilic, intermediate, not hydrophobic, not hydrophilic or unconstrained. This defined
the hydropathic signature of this protein folding class (figure 7).
Figure 7. Sequence alignment of the large-domain
portion of 44 -lactamases from different organisms.
The physico-chemical properties of the various amino
acids found in the same column are informative as to
the design constraints found at that location. A
hydropathic signature, defined in ref. 28, allows one
to summarize which amino acids are permitted at
each column: hydrophobic, hydrophilic, intermediate,
not hydrophobic, not hydrophilic, or unconstrained.
(Click on figure to see larger version)Proteins often
continue to function in spite of mutations due to
excess robustness built in. Portions of the folded chain are held together near optimally, under thermodynamical
considerations, through a large number of interactions. Therefore, sub-optimality through a few mutations will often not
lead to discernable loss in function. Therefore, one cannot conclude that mutations which individually seem harmless
would be acceptable when present concurrently. The optimized proteins have a kind of buffering effect. Demonstrating
that alternative amino acids are acceptable, by inducing mutations on a near optimal wild type, does not permit an
estimate of the number of acceptable sequences with minimal functionality. To make a reasonable estimate would require
actually generating the variants with multiple mutations to identify which alternatives would really work.The design of Axes
experiment27 reflects how natural selection would have to go about fine-tuning a novel enzyme. A minimally useful
sequence must first exist upon which natural selection could act. He generated a large number of TEM-1 variants by
mutating 49 positions, introduced the plasmids in an E. coli strain by electroporation, and isolated a colony having 33
substitutions (relative to the original sequence). Exposure to a low concentration of ampicillin permits selection of those
bacteria with a functioning enzyme. The candidate starting sequence for the subsequent experiments showed resistance
at 10 g/ml, but none at 20 g/ml concentrations at 25C. 34 This enzyme provided 0.3% of the wild type activity at 25C,
and only 0.01% at 37C.34 Since the enzymatic reactive site was not mutated, the loss in activity probably reflects lower
ability to hold the protein together in a suitable three-dimensional geometry.A large number of bacteria with the above
less-than-optimal candidate starting sequences were grown. From these sequences, mutant plasmids where engineered
in a manner to optimize the proportion satisfying the hydropathic signature. Four sets of random mutations, each involving
ten residue positions, were performed. The number of mutants satisfying the hydropathic signature was calculated (on
average over 85% of the mutants generated), and those surviving ampicillin poisoning were sequenced. The geometric
mean calculated from the pass rates of the four experiments led to an upper-bound estimate of 0.38 per position. This is
the probability that a random mutation at a residue position which meets the
hydropathic signature constraints will be acceptable.The value of 0.38 is
generous for several reasons. In one of the four experiments no acceptable
mutants were obtained (of 54,000 mutants generated which still satisfied the
hydropathic signature!), but a probability of 0.002 was used anyway.
Furthermore, acceptable mutations within sets of ten residue positions will
certainly not be permissible in the presence of all other acceptable mutations
for the remaining 15310 positions.
Figure 8. Large domain of TEM-1 penicillinase includes many structural
components (loops, helices, and strands). All residues not between 62214
were removed from entry 1ERM.pdb in the Protein Data Bank,
<www.rcsb.org/pdb>. Displayed with RasTop
For the whole large domain (figure 8) the proportion of acceptable mutants
which are signature compliant would thus be less than 0.38 153 = 1064.35 The

number of open reading frames (here only a portion of a gene) leading to the signature under study, based on which
codons code for which amino acids, is 10 33. In conclusion, among random polypeptides a proportion of less than 10 64 x
1033 = 1097 would provide a working large domain -lactamase enzyme using the same fold characteristics. 35It is possible
other protein folding families could offer a necessary stable framework for enzymatic activity. If a species has about a
million different protein variants and a thousand or so fold types, then about 0.1% of the fold types on average would be
suitable for a particular function. Based on other work, 36 at most 1 out of 10 10 random sequences would fold to a stable
pattern based on hydropathic constraints alone. These considerations led to an estimate 28 that about 1 out
of 1077 sequences of 153 amino acids could perform the function under study.
Appendix C
Sequence analysis of the lambda repressor fold36
The probability of obtaining a functional lambda repressor from among 92 amino acid chains was reported in Table 1 as
being 1063. The details of this experiment are summarized in this appendix.
Background
Bacteriophage lambda, probably the most extensively studied bacterial virus, 37 has a genome of about 50 genes, 38 and
under suitable conditions can become integrated into DNA of bacteria such as E. coli. Within the host there are two modes
of replication.39 (1) Once integrated into the host genome it can be replicated along with the rest of the DNA. A key
component of this prophage state is the lambda repressor protein (cI protein), which occupies the operator, blocking the
alternative reproductive pathway, and also activates its own transcription. (2) In the lytic state, whereby the virus is not
inserted into the host chromosome, the cro protein attaches to a different site in the operator, preventing synthesis of the
repressor protein and permitting its own synthesis.In the prophage state most of the virus genes are not transcribed. In
the lytic state the virus DNA is extensively transcribed and organized into new bacteriophage, then released by rupturing
the host cells outer membrane. This kills the cell, of course.An infecting virus usually adopts the prophage stage. But
when the host is badly stressed or damaged, an integrated virus converts to the lytic state. For this to be possible, the
repressor protein needs to be inactivated.The lambda repressor protein is an example of helix-turn-helix proteins which
bind to specific DNA sequences.40 Other examples include tryptophan repressor, lambda cro and CAP.41 These kinds of
proteins often exist as symmetric dimers, able to bind to two DNA stretches per protein (e.g. on opposite strands of
complementary DNA), which doubles the number of contacts and squares the affinity constant. 41A mutant variant of
lambda repressor protein not able to function properly is easy to monitor experimentally, since the virus in the lytic state
kills the host.
The experiments
Sauer at MIT examined42 mutants at 25 residue positions (823 and 7583) in two -helical regions of the -repressor
distributed along positions 1 to 92 of the N-terminal end of the protein. The whole protein usually contains about 237
residues.4345 Plasmids were engineered which contained an ampicillin-resistant gene and an origin of replication which
allows production of single-stranded DNA for sequencing purposes.Oligonucleotide cassettes 46 were synthesized for
several experiments. At each position where amino acid mutations in the protein are to appear, codons of type NNG/C
were prepared in equal proportions, where the N indicates any of the four bases (A, C, T or G). Thus only 32 of the
possible 64 genetic code alternatives were needed to generate all possible natural amino acids. Between one and three
positions were allowed to differ from the wild type.The modified plasmids contained special restriction sites which
permitted the cassettes to be ligated at predetermined positions, ensuring the desired mutant proteins would be coded for.
The plasmids were transformed into E. coli K-12 strain X90. Exposure to ampicillin killed off the E. coli lacking inserted
plasmid (since the bacteria lacks the ampicillin-resistant gene provided via the plasmids). The phages cI then destroy the
cells lacking a suitable -repressor, since the virus only had the option of entering the lytic state. Surviving E. coli colonies
thus have functional repressor protein present in the plasmid. At least 510% of wild-type activity was necessary to
survive.
Figure 9. Functional -repressor proteins identified
after mutating several residues between positions 1
and 92 of the N-terminal end, using oligonucleotide
cassettes.42 (Click on figure to see larger version)
Survivors were analyzed and the alternative amino
acids at each residue position were reported. The 25
positions mutated were supplemented with the results
from an earlier study42 in which positions 8491 had
been mutated in three separate experiments involving
three and four residue positions at a time. The
alternative amino acids found at each residue
position are shown in figure 9.
The available data gives some indication as to the
tolerable variability. By multiplying the number of
alternatives at each position shown in figure 9 the
authors concluded that about 4 x 1022 different
sequences would be functional over the 33
positions.47 Extrapolating to the 92 positions of the
domain under consideration indicates that a
proportion of about 1063 would be functional.

I believe this estimate is still too large due to context dependence: a tolerated mutation at one position will often be
deactivating when multiple other otherwise acceptable mutations are present. At one point they write,However, in general
there appears to be no dependence of a change at one position on a change at another, as most changes were recovered
in several different mutant backgrounds.48This is a surprising statement for several reasons. What is needed are
experiments in which only mutations at a particular residue are generated, followed by additional tests for which
this and additional residues are modified.
Figure 10. Context dependence of mutations in -repressor proteins. All mutations reported in
positions 14, 15/16, and 1417 using oligonucleotide cassettes.42 (Click on figure to see larger
version)
Figure 11. Context dependence of mutations in -repressor proteins. All mutations reported in
positions 83, 81/82, and 8183 using oligonucleotide cassettes.42 (Click on figure to see larger
version)
In the reported data43 only two such series of experiments were
performed, generating at most three mutations with respect to
the wild type. This permits us to determine whether the same
mutations at one position affect the probability of additional ones
being acceptable elsewhere.(i) All possible mutations were
generated in positions 14, 15/16 and 1417. The results are
shown in figure 10. Unfortunately, no variability was found in
position 14, so this is uninformative. Experiment 1417
produced the wild type
sequence and one mutation (RK) at position 17. Experiment
15/16 also produced the
wild type sequence and four other amino acids were tolerated at
position 16. But why
were none of these four alternatives at that position identified in
experiment 1417? The
extra mutation, (RK), probably hindered this!(ii) All possible
mutations were generated
in positions 83, 81/82 and 8183. The results are shown in
figure 11. Experiment 83
generated the wild type plus ten other amino acids at that
position. Experiment 81
83, however, failed to produce six of these at the same position
83. This may well be due
to the presence of additional mutations at the other two
locations. It is also true,
that experiment 8183 did not produce any functional variants
with
an amino acid
missing at the corresponding site of experiment 83.
In most cases an A is
found at residue 81 for experiments 81/82 and 8183. Lets
examine these cases.
Experiment 81/82 displayed, in addition to the wild type R at
position
82,
another 11 alternative amino acids. Of these, experiment 83
produced only 1 (E)! It
would appear, once again, that mutations in the third position
limited the number of
possibilities at other residues.33(iii) Experiments from the same
laboratory were reported
earlier in which seven positions of the same protein were
mutated. I pointed out in
this journal32 that the number of variants generated increased
greatly with number of
mutational differences from the wild type. Nevertheless, the
number
of
functional
alternatives identified was precisely in the opposite order.
Clearly,
on
average,
existing mutations prevented otherwise acceptable mutations
from
producing
a
functional protein.This fact is surely known to these researchers,
since in the earlier paper
the authors wrote:In separate experiments, five of the seven
core
positions
were
altered individually. Only one to three amino-acid substitutions at
each position yield a fully
functional protein as is common for buried positions. 49The
authors of the later
paper36 are aware that the context of mutations are an issue,
and pointed out with
respect to the estimated number of acceptable sequences,On
the
one
hand,
this
calculation overestimates the number of functional sequences,
since
changes
at
individual positions are less likely to be independent of one
another as more positions
are allowed to vary. Moreover, combining changes at several
positions, each of which
individually decreases the activity of the proteins slightly, may
result in a protein that is
essentially non-functional. On the other hand, some changes
which are not allowed
when positions are randomized individually may be tolerated in
other
sequence
contexts.50However, the magnitude of both effects is surely
dramatically different and
hardly compensate significantly. What examples for novel
compensatory
multiple
mutations are found in the data reported? 42 At most only one. In
experiment 8183 amino
acids SA in the first two positions led to a functional protein, but
this mutant was not found in experiment 81/82 (an example was obtained with SR). On the other hand, the authors
pointed out51 that there is a 58% chance that not all tolerated amino acids were identified at position 82, making likely that
a larger data set for experiment 81/82 may well display the missing amino acid.Whether introducing simultaneously
multiple mutations which compensate for each other is actually realistic to evolutionary theory, is questionable. For
example, it is possible that using the two largest hydrophobic and a single smallest hydrophobic residue would work in
some context, but whether theoretical intermediates (e.g. one of the largest hydrophobic amino acids only) might work is
not assured. Such solutions would often require an all-or-nothing set of circumstances.In contrast, an overly generous
assumption of mutational context independence can have a dramatic effect. Let us reconsider the data in figure 11 and
neglect the few sequences for which an A was not found in position 81. Experiment 83 produced ten alternatives, and
experiment 81/82 generated eleven functional alternatives at position 82. We see that this simplification reflects closely
the assumptions made in figure 9 regarding residues 82 and 83. Then the assumption of context independence, as
proposed by the authors, predicts about 10 x 11 = 110 variants from experiment 8183 (with a wild type A in position 81),
or 1101/2 = 10.49 per position properly weighted. However, only 4 were actually found, 41/2 = 2 on average. Whether one
assumes (10.49)n or (2)n over n residue positions, leads to dramatic different estimates for the number of acceptable
variants.Testing all mutations at a large number of positions is experimentally not feasible, given the huge number of
possibilities 20n for n residues positions. Simplifying approaches are needed leading to large doubts in the estimates. The
proposal of about 1057 functional alternatives50seems to be too high, since for this to be possible up to 67 of the 92
positions of this portion of the protein must be mutable at the same time and in all combinations based on the data from
figure 9. (In the 33 residues studied 9 positions were invariant (see figure 9), so all but 9 x 92/33 positions on the full
domain must tolerate for all the combinations of mutations to arrive at the authors number of polypeptide alternatives,

1063).A more realistic estimate might be guesstimated as follows. We shall assume that any of the mutations shown in
figure 9 are acceptable and also concurrent. The authors used a Monte Carlo simulation to identify the probability that not
enough plasmids had been generated to identify all acceptable alternatives at each position. For all those residues51 we
shall assume two more amino acids would be acceptable (Table 2), and shall further assume that all these additional
mutations would be mutually compatible. We neglect the possibility of a handful of variants involving multiple
compensatory mutations. As shown in Table 2, about 3.1 x 1021 alternatives were estimated.

Table 2. Functional -repressor proteins identified after mutating several residues between positions 1 and 92 of the Nterminal end, using oligonucleotide cassettes.42Two more amino acids are assumed for all residues where additional
amino acids might be tolerated. (Click on figure to see larger version)The number of alternatives will be extrapolated by a
simple factor of 92/33 to cover the whole domain, i.e. about three separate sections. As a partial compensation for the
above assumptions, we will say that acceptable mutations are limited to each of the roughly three sections, but not
between them. This leads to an estimate of (92/33) x (3.1 10 21) = 8.6 1021.The resulting proportion of functional
variants, 8.6 x 1021 / (20)92 (ca. 2 1098) is considerably smaller than what the authors suggested, 1063.
Appendix D
Cytochrome c proteins52
The probability of obtaining a functional cytochrome c from among amino acid chains of suitable length was reported in
Table 1 as being 1044in one case and 10112 on another. The details of this experiment are summarized in this appendix.
Yockey52 collected a list of all known cytochrome c protein sequences and lined up 110 residue positions after taking into
account apparent mutational deletions. He then expanded the list of known sequences generously, using a
model53 developed by Borstnik and Hofacker,54,55assuming many other sequences might also be tolerated, as already
discussed in this journal.16 We reported56 that a fraction in the order of 2.0 x 10 44 of the 110-residue chains would offer a
starting point for natural selection to begin fine-tuning a cytochrome c sequence. Incidentally, the information theory basis
for these calculations assumes context independence:32 all individually acceptable amino acids substitutions would
supposedly lead to a functional cytochrome c as in the presence of other mutations. The true proportion of functional
alternatives is surely many orders of magnitude smaller, a mathematical issue in the use of information theory I have
brought to Yockeys attention.Yockeys latest calculations57 suggested that the proportion of polypeptides leading to
functional cytochrome c is actually much lower: 1.6 x 10112.
Appendix E
Ubiquitin58
This protein is present in all examined eukaryotes type cells. Current evolutionary thinking is that the first eukaryote cell
lived about 2.7 Ga ago. 48 Since all plants, animals and fungi possess ubiquitin (UB), unlike prokaryotes, this gene must
have arisen virtually instantaneously under evolutionary assumptions. 58I collected all known and reliable sequences for
UB and calculated the number of alternatives using information theory. About 60% of UB residues seem to tolerate no
mutations at all, and in 17 other positions a single alternate amino acid was occasionally found. In almost all the latter
cases this exception was found in only a single organism, and some of these sequences may simply be incorrectly
reported data.I estimated58 that a proportion of about 4 x 1083 polypeptides, 76 residues long, would produce a functional
UB. Several things need to be considered in this estimate.
(i) Not all eukaryotes have been examined. On the other hand, many sequences were identical for species not especially
close according to current evolutionary theory. Therefore, not too much more variety is to be expected.
(ii) There are at least three families of ubiquitin distinct for animals, plants and fungi. It is possible these alternatives are
not interchangeable, in which case the amount of acceptable variability for the putative initial ancestor would be restricted.
(iii) My estimate assumes that all mutations present at any residue would be compatible with all other mutations at other
locations.
Bone building: perfect protein
by Jonathan Sarfati
Bones are an amazing example of design, present in all vertebrates. They have a huge advantage over man-made
girders, in that they are constantly rebuilding and redesigning themselves to cope with changing stress directions.1
This involves a fine balance of the activity of bone-depositing cells (osteoblasts) and bone resorbing cells (osteoclasts).
Its been recently shown that thyroid-stimulating hormone (TSH), best known for what its name saysstimulating the
production of hormones in the thyroid glandhas an important role. It oversees both types of cellswithout it, bones
have osteoporosis in some parts (too little bone, so very weak), and are too dense in other patches.2
Osteocalcin and hydroxyapatite
The strength of bones mainly comes from the hexagonal mineral hydroxyapatite (HA, formula Ca 5(PO4)3OH).3 But this
must be built up in the right patterns. In vertebrate bones, this is built up with a special protein called osteocalcin (OC). It
is a small protein, 49 amino acids long (5.8 kDa), and is highly conserved, meaning that its sequence is almost identical
among vertebrates. Human OC has the sequence Tyr Leu Tyr Gln Trp Leu Gly Ala Pro Val Pro Tyr Pro Asp Pro Leu Gla
Pro Arg Arg Gla Val Cys Gla Leu Asn Pro Asp Cys Asp Glu Leu Ala Asp His Ile Gly Phe Gln Glu Ala Tyr Arg Arg Phe Tyr
Gly Pro Val. 4Like all proteins, the instructions for OC are in the DNA, 5 but there is more to its manufacture than simply
decoding / translating the code and synthesizing the OC on a ribosome. Firstly, the transcription (DNAmRNA) is
regulated by 1,25-dihydroxy-Vitamin D3, one reason that Vitamin D is so important for healthy bones. It is then first
decoded (translated) as a preproosteocalcin, which is 98 amino acids long. This comprises three parts: a 23-residue
signal protein that is cleaved during translation,5 a 26-residue target propeptide, and the 49-residue mature protein.6

Model of osteocalcin (OC) engaging an

hydroxyapatite (HA) crystal based on a
Ca2+ lattice
match.
The
OC-bound
Ca2+ (small circles in big ones) and HA
Ca2+ (small circles) on the crystal surface
align perfectly (from Hoang et al.).7Even this
does not complete the process; this requires
another
vitaminK.
Vitamin
K1
or
phylloquinone, best known for its vital role in
the blood clotting cascade, is an essential
co-factor in -carboxylation. That is, the
specific glutamyl residues (Glu, from the
amino acid glutamic acid) at positions 17, 21
and 24 have a second carboxyl group (
COOH) added to form -carboxyglutamyl
residues (Gla). This changes the structure
and stabilizes the -helical portion of the
protein.6Even now, the OC protein is fairly
shapeless. But when OC meets calcium
ions, it folds to a special structure.8The two
1
2+
carboxyl groups on the -carboxyglutamyl residues chelate to the Ca ions, as shown by Fourier-Transform Infrared
spectroscopy (FTIR).9There was no spectral change when Ca 2+ was added to decarboxylated OC (i.e. as it would be
before converted by Vitamin K), showing that there is no binding without the carboxylation. 9 Amazingly (for
uniformitarians), enough osteocalcin to produce an immune reaction was found in bones of an Iguanodon dated to 120
Ma,10 yet proteins could not last for millions of years. And the fact that its a bone protein shows it cant be contamination
from outside.
Osteocalcins crystal structure
Now, pig OCs crystal structure has been
discovered, using a type of X-ray diffraction
called the iterative single anomalous scattering
method. This provides new insights into how
finely designed it must be to work.7 The active
site has a negatively charged region that binds
the positively charged Ca2+ ions. Five Ca2+ ions
are coordinated by three special Gla residues
and an Asp at position 30. But not in just any old
wayfive calcium ions are bound in the same
arrangement as in the exposed face of a HA
crystal, parallel to the c axis. So the OC can dock
on the HA and add the calcium, and thus grow
the crystal, making the bone grow in the area
needed.Profile of docking OC (top) on HA crystal
surface (from Hoang et al.).7To do this, OCs
building blocks, the amino acids, must be in a
very precise sequence. For example, there is a
tightly packed core involving the hydrophobic
residues Leu 16, Leu 32, Phe 38, Ala 41, Tyr 42, Phe 45 and Tyr 46. There is also hydrogen bonding to stabilize the
connection between different -helices, Asn 26 in the helix 12 linker and Tyr 46 in 3. The helices 1 and 2 form a Vshaped arrangement stabilized by a disulphide bridge between Cys 23 and Cys 29.Thus the very precise sequence of
OC, as well as the metabolism to form the essential -carboxyglutamyl residues, seems to be yet another example of
irreducible complexity, a hallmark of design. 11 This means this is yet another component that must be exactly right for the
alleged transition from invertebrate to vertebrate. So it is not surprising that evolutionists have no fossil evidence for how
the transition occurredthis protein alone shows it could never have happened.12
Summary
Bones are dynamic supports, constantly rebuilding to cope with changing stresses.
Bone shape is a delicate balance of bone deposition and resorption.
Bone strength is largely from the mineral hydroxyapatite (HA).
One vital component for bone growth is the small, highly conserved protein osteocalcin (OC).
Large fragments have been found in dinosaur bones dated over 100 million years old, although measured rates of
breakdown mean that nothing should have survived that long.
Vitamin K is essential to modify three amino acid residues of OC, otherwise it cant bind calcium at all.
Recent discovery of OCs crystal structure shows that it binds calcium in exactly the right geometry to add to a certain
crystal face of HA.Thus bone construction is irreducibly complex.Which explains why there is not only no fossil
intermediate between invertebrate and vertebrate, but why it could not exist .
VIRUSES
The bacterium that came in from the cold
by Jonathan OBrien
Some animals live for 200 years or more,
and there are trees which live much longer
still.1,2 But heres a story about an organism
which, some scientists say, trumps them all.
There have been several recent claims that
scientists have found living bacteria many
millions of years old,3 and now another
similar claim has been made by researchers

from Newcastle University, UK.While studying seafloor sediments off the Norwegian archipelago of Svalbard, researchers
were surprised to find dormant spores of a thermophilic (heat-loving) bacterium. These germinate when temperatures
reach a very warm 50C, similar to microbes which thrive deep in the oceanic crust, and in sub-surface oil reservoirs.
Given the frigid conditions at Svalbard, the scientists could not account for the presence of a bacterium that relies on a
much warmer environment to complete its life-cycle. 4Dr Casey Hubert, one of the researchers, proposed that the bacteria
leak out of the ocean crust in rising hot fluids, ending up in the icy arctic seawater where they become dormant due to the
cold. So far, so good, as hydrothermal fluids have been observed emerging from the seafloor in other locations. But
Huberts theory then goes that, in order for the bacteria to complete their life-cycle, they must wait to become deeply
buried in very large quantities of sediment again, until conditions become warm enough for them to germinate. And going
by observable rates of sea floor sedimentation occurring today, Hubert claims it would take 100 million years to deposit
such deep quantities of sediment. So he is using this sedimentation rate to justify the claim that the bacterium has a 100
million-year life-cycle.According to other experts, the DNA of bacteria would be destroyed within 100 thousand years due
to the effects of molecular motion and background radiation, consistent with the second law of thermodynamics, 5 so
Hubert cannot be right. In any case, his claim is observationally unverifiable, as no known test can confirm that a dormant
spore can live anything like as long as he claims.It is drawing a very long bow indeed for the researchers who found the
bacterium off Svalbard. In fact, they dont yet know how, when or even if the dormant spores will germinate. They used the
secular geological principle ofuniformitarianism, the idea that present conditions are the key to the past, when they
accepted that todays (observable) sedimentation rate applied for the alleged (and unobserved) previous 100 million
years.But we know that the whole world suffered a massive flood , during which sedimentation rates would have been
vastly greater than what is observed today. The scientists are speculating, in the dark and without evidence, when they
talk of a bacterium that takes an unobserved, and very long, 100 million years to come in from the cold.
Does biological advantage imply biological origin?
by Shaun Doyle
The origins of sexual dimorphism and multicellularity are two of the greatest
mysteries to evolution. For either of them to evolve requires massive restructuring
of the biological system from the molecular to the organismal levels. Moreover,
there are massive selection and energetic barriers that must be crossed to get from
unicellular to multicellular life and to evolve sexual dimorphism. Two recent news
articles have claimed that certain biological advantages in sexual dimorphism 1 and
multicellularity2 provide a reason why they evolved in the first place.
Intra-cell communication and sexual dimorphism
The first study discusses the question: why are there two sexes?3 In terms of
evolution, its not the best number of mating types because it only allows us to mate
with half of the population. However, researchers have proposed that inheriting
mitochondrial DNA (mtDNA) from just one parent instead of both may serve to
offset this disadvantage. Most sexually reproducing creatures only receive nuclear
DNA from their father but get the other half of their nuclear DNA plus their entire
cellular structure, including mtDNA, from their mother. The researchers proposed
that because this setup only passed one set of mtDNA to offspring, it allowed for
more efficient synchronization between the nucleus and mitochondria, and
between mitochondria, than would be possible if mitochondria were inherited from
both parents. According to their modelling, they were correctuniparental
inheritance of mitochondria (UIM) produced fitter offspring than biparental
inheritance of mitochondria (BIM) under most realistic selection constraints.But the
researchers also explore this question: Could uniparental inheritance of
mitochondria have arisen to facilitate better co-adaptation of mitochondrial and nuclear genes, and so explain the
evolution of two sexes?,3 to which they ultimately give a positive answer. However, this misplaces the important question
for the evolution of any new traithow it arose in the first place. Essentially, they perform a cost-benefit analysis between
UIM and BIM, determine that UIM is the better system, and then conclude that UIM must have evolved from BIM. But this
skips over the succession of events that supposedly led to the evolution of UIM from BIM because the researchers
assume that since UIM and sexual dimorphism exist, they must have evolved. This is clearly begging the question of
evolution, but its worse. Evolution is taken as so incontrovertible that questions of how (the succession of evolutionary
events) are deemed superfluous, and all that matters is why UIM evolved. However, all they have established is that UIM
provides the functional grounds for sexual dimorphism in eukaryotes, and the origin of that function is the very question
which the fact of its functionality does not directly address.
Kinship and the evolution of multicellularity
Another recent study showed how high-relatedness between cells is a necessary prerequisite for multicellularity. 4The
researchers ran two experiments on the amoeba Dictyostelium discoideum, one where they tested the effects of lowrelatedness on Dictyosteliums ability to form multicellular fruiting bodies, and the other tested the effects of mutation
accumulation in a single clonal line. The researchers found that when different lines were mixed, it didnt take long for
cheater mutants to take advantage of the fruiting bodies and propagate ahead of the non-cheaters, to the point where
there were so many cheaters that many lines were unable to form fruiting bodies at all by the end of the experiment. In
contrast, fruiting ability was never lost in the mutation accumulation experiment where high-relatedness was maintained,
as per conditions in the wild. As a result, the researchers concluded that high-relatedness was necessary and sufficient to
maintain the viability of the multicellular stage in Dictyosteliums life cycle. The researchers then applied their findings to
multicellular life in general:Thus, we conclude that the single-cell bottleneck is a powerful stabilizer of cellular cooperation
in multicellular organisms.5This is a fair application of their research. It highlights a necessary prerequisite for functional
multicellularity, and it doesnt extend all the conclusions about high-relatedness in Dictyostelium to all multicellular life. But
compare this to the questions the news article says this research answers:How could the extreme degree of cooperation
multicellular existence requires ever evolve? Why arent all creatures unicellular individualists determined to pass on their
own genes?2This presupposes that functional multicellularity evolved, and proposes that the mere existence of highrelatedness among cells is the reason why it evolved. However, creationists can also presuppose the necessity of highrelatedness among cells for functional multicellularity to be possible without an appeal to evolution. 6 We have here
confusion between the functional grounds of a trait and the historical cause of a trait. If multicellularity evolved, then it
must have evolved from a population of clones, but this tells us very little about the succession of events that led from a

unicellular ancestor to the first metazoan or plant. Therefore, it is not a helpful explanation of the evolution of
multicellularity.
How useful is Dictyostelium for studying the evolution of multicellularity?
Figure 1. Multicellular fruiting bodies ofDictyostelium discoideum.
But is Dictyostelium a model organism for studying how the evolution of
multicellularity might proceed? The researchers point out that cheaters
are not a problem for Dictyostelium colonies even if they were the size of
a blue whale.7 However, even the news article admits that a blue-whalesizedDictyostelium colony is not the same thing as a blue whale. 2 But it
fails to describe why. It is one thing to grow a colony to the size of a blue
whale, but it is a different thing to maintain the colony at that size in a
multicelled state over a period of decades. The researchers admit that
little cell division occurs in the multicelled phase of Dictyostelium.7 In a
multicellular stage that has little cell division, there is obviously no need
for strategies such as serial differentiation 8 to maintain the multicelled
state. Thus it is not surprising that the multicelled stage
in Dictyostelium does not last very longa day or less.Moreover, 80% of
the individual Dictyostelium cells survive the multicelled phase, and then
go on to reproduce as unicellular organisms. Even volvocine
algae,9 which do not possess the separation of totipotency and cellular
immortality (e.g. reduced mitotic capacity or multipotency) that is the
hallmark true differentiated multicellularity,10 sacrifice thousands of
somatic cells in their multicellular phase to produce perhaps a dozen or
so germ cells. This proportion drastically increases again out of
necessity when the organism possesses a functional cellular
differentiation program designed to structure and maintain the
multicelled state.8 While there is an analogy to unicellular bottlenecking
in the dispersal of the spores at the end of Dictyosteliums multicelled
phase, Dictyostelium has nothing like the proportion of cellular sacrifice
that occurs in multicelled life. Volvocine algae are likely the closest that
life capable of free-living unicellularity can ever come to true differentiated multicellularity, and Dictyostelium doesnt even
approach this level of multicellular coordination, let alone what is found in plants, animals, and fungi.
Therefore, Dictyostelium can only tell us so much about multicellular life, and it can provide little information on the
historical sequence necessary to evolve true differentiated multicellularity.
Does an advantage provide an origins narrative?
Both of these studies have been said to answer some important questions about evolution. Both have been said to
provide a reason why this or that trait evolved because these traits have been demonstrated to convey a certain selective
advantage above the presumed ancestral condition, or the acquisition of a new trait has proved impossible without a
certain feature.The reasoning basically goes like this: Why did x structure evolve? Because it conveyed y advantage.
There is a major fallacy here: note that this doesnt deal with how it evolved because key causal links between the
ancestral and the descendant traits have been ignored. The question of how involves discussions of evolutionary
mechanisms, such as specific mutations, specific regulatory and developmental changes, their effects, and the selection
mechanisms that have contributed to the preservation of these changes. Explaining the advantages of a rewired system
does not explain the rewiring sequence that took place to change the old configuration into something completely new,
and it does not directly explain how the wiring got there in the first place. In fact, the soundest inference from such
parameters as proper wiring and fitted function is to an intentional creation, not unintentional nature.Rarely is this sort of
detailed, step-by-step narrative ever provided for even the smallest evolutionary events, let alone the massive
morphological restructuring involved in turning (for example) a free-living unicellular organism into one with differentiated
multicellularity, or even turning a sarcopterygian ancestor into a tetrapod descendant. On the few occasions that such
plausible narratives are constructed, they are only for incidental (and accidental) biological function, not essential
biological structure.11For any fruitful debate to proceed regarding the plausibility of evolution, the right questions need to
be asked and answered. Asking why one structure has an advantage over another is the wrong question with which to
seek origins answers. A serious researcher would instead inquire into the feasibility of each step along a hypothetical
evolutionary path, like from unicellularity to multicellularity.12 However, the science media typically obfuscates these
matters, and even researchers fail to appreciate the depths of explanation that evolutionary theory needs to provide a truly
compelling narrative for the history of biology capable of outweighing the design explanation with which it competes.
Germ with seven motors in one!
by Jonathan Sarfati
Published: 15 January 2013 (GMT+10)
Over the last two decades, scientists have uncovered
some of the amazing machinery in microscopic living
cells. These include germs with a miniature motor that
generates waves in a tiny tail that allows germs to swim
the bacterial flagellum.1 This even turns out to have a
clutch to disconnect the motor from the tail.2 Even more
miniaturized is the tiniest motor in the universe, ATP
synthase, which makes the vital energy molecule ATP
(adenosine triphosphate).3 Remarkably, a virus has a
tiny motor used to wind up DNA into tight
packages.4Some germs have more than one flagellum.
Sometimes they work individually but still the germ
manages to coordinate the motors. Other germs have
the tails loosely bundled. But the marine bacterium MO1 is different again. Here, seven flagella are tightly

bundled in a sheath.The mystery was how they could all rotate in the same direction without interfering with each other.
Now a research team from French and Japanese universities5 has worked out how. They produced a series of 2dimensional images of cross sections to build up a 3-dimensional picture (electron cryotomographylike a CAT scan, but
with an electron microscope and very cold temperatures).The seven flagella are actually surrounded by 24 fibrils (tiny
fibres), in a hexagonal array. And these fibrils rotate in the opposite direction to the flagella, allowing them to rotate freely.
The researchers diagram shows the flagella as large gear wheels with the fibrils as smaller gear wheels. These gears or
bearings enable the flagella to spin very fastso the germ can swim about 300 m/s, or 10 times faster than E.
coliand Salmonella.
www.pnas.org
Schematic model of 7 flagella and 24 fibrils rotating in a tight bundle
smoothly within the sheath by the counter rotation of neighboring
flagella
and
fibrils.
Click here to view an animation.
The researchers evidently had no use for evolution in their
research. Instead, they referred to complex and exquisite
architecture, and said:This design must be playing an essential
role in the fast, smooth rotation of the flagellar apparatus that
allows the rapid swimming of MO-1.But in the last paragraph, the
researchers paid the obligatory fact-free homage to goo-to-you
evolution:Taken together, these features of the MO-1 flagellar
apparatus represent an advanced level of evolution of a motility
apparatus. It is also intriguing that the same pattern of an
intertwined hexagonal array in two evolutionary distant systems:
the basal bodies of flagella and fibrils of the MO-1 flagellar
apparatus, and the thick and thin filaments in vertebrate skeletal
muscle. Similar architectures of filamentous structures presumably
evolved independently in prokaryotes and eukaryotes to fulfill the
requirements for two very distinct mechanisms to generate motion:
counter rotation and axial sliding.This is yet another example of
appealing to convergence: the same design feature allegedly evolved not just once but twice. But more to the point: in
the late 1940s, the famous evolutionist J.B.S. Haldane predicted that we would find no wheels or magnets in living
creatures.6 This is because these would not work unless fully formed. Thus natural selection could not have produced
them step by small step, each an improvement over the previous one. Such motors thus falsify evolution by Haldanes
own words. MO-1 also senses magnetism, 7 following Earths magnetic north pole in a helical path. So MO-1 provides two
strikes against evolution.
The fungus that walks
by Rodney McQueen and David Catchpoole
Despite intense scientific scrutiny, these what-are-they? organisms continue to
baffle.Even after many decades of research, the experts still debate how best to
classify them.1,2 For many years the organisms pictured here were labelled
fungus, because part of their life cycle is like that of many fungi. Mostly, though,
they move and feed like one-celled animals. Yet they make their own cellulose,
like a plant. And incredibly, as we shall see, there are times when they look and
move like multi-celled animals. Part of their life cycle is also very similar to certain
bacteria. Just what are they?These are the so-called slime moulds, which are tiny
(about 0.02 mm long, or 1/1000th inch), and despite their repulsive-sounding
name, can look very beautiful. They are found all over the world, living in rich damp
soil, leaf litter, or animal manure. They come in two broad types: plasmodial and
cellular.3 We will mainly consider the cellular slime moulds, one of
which, Dictyostelium discoideum, has been (and continues to be) the subject of
intensive scientific scrutiny.4
Soil refiners
When conditions are favourable, slime moulds are free-living, single-celled, amebalike micro-organisms. They are not actually amebae as scientists would normally understand amebae to be (i.e. a type of
protist, a one-celled organism). But they are so like them in appearance and form of motion that scientists usually call
them by that name since no other suitable term is available. (This highlights the dilemma faced by scientists who try to
label these creatures within an evolutionary framework.)Slime moulds creep about in or on the soil, leaf litter, logs or
manure, consuming bacteria as they go. With their capacity for self-cloning by dividing themselves into two new daughter
amebae, slime moulds can rapidly multiply their numbers so as to exploit favourable environments. Each of these new
amebae creeps off on its own as a completely separate entity. This can continue for as long as environmental conditions
stay favourable. That is, as long as the ground is moist enough, and they can find enough bacteria for food.But should
conditions begin to take a nasty turn (usually a dwindling food supply), something utterly remarkable happens. Thousands
of individual amebae from the same vicinity begin to congregate together. This they do by responding to a substance that
one (or more) of them starts to secrete. It has been shown that about one in every 2,000 amebae has the ability to start
producing this substance under unfavourable conditions. Other nearby amebae start to stream towards the source of this
chemical. Then they too begin to secrete the same substance, which spreads out in wave-like pulses and attracts more
amebae, and so on.5
Bizarre behaviour
Within hours, up to 500,000 amebae have clumped together in a multi-celled mass. 6 This they are able to do by secreting
a sticky substance which glues them together. Each ameba is still, however, an entirely independent individual. But the
clump of thousands of one-celled individuals now begins to behave as if it were a single slug-like creature.And, in some
species such as Dictyostelium, something else very strange happens. The amebae pile up one on top of the other to form
a steep-sided mound. Then the mound topples onto one side, and may slither away like a garden slug (up to 23 mm
long, or 1/10th inch), with a clearly defined head end. Scientists are still figuring out the mechanisms that determine why
some amebae become leaders, while others become followers, meekly falling into line behind the leaders. Amazingly, this

self-redistribution of the amebae into leaders and followers during the slug stage can occur even when the slug is cut up
into smaller segments, with tiny fragments capable of forming themselves into a midget body (and able to proceed
through the rest of the life cycle quite normally). 6Slugs of some species then migrate considerable distances, depositing a
thick slime sheath as they travel, which collapses behind them. It has been found that these slugs, still consisting of notquite-so-independent amebae that are now acting in unison, are remarkably sensitive to light and heat. They will move
towards a weak source of heat or a light as dim as the faintly luminescent hands of a wrist watch.

The mound of clustered amebae topples The migrating slug leaves a trail of slimeWhoa! Heres a good site for a launching
over and begins to move away as a behind.
pad. The slug slows and enters the
slug.
standing finger stage.

As one slug slows, another begins to liftThe amebae at the tip push down A forest of slime mould fruiting bodies,
itself higher.
through the mass to form a stalk, lifting containing masses of spores ready to fly
the others up into the breeze.
to fresh hunting grounds.
Shortly after the slug has chosen the stage for acting out the final scene, it lifts its front tip up into the air. Somehow it
keeps doing this until once again the slug is standing upright, reaching for the sky like a miniature sky-scraper. The
amebae in the tip now swell with water, and then secrete around themselves a cellulose wall. Here it begins to take on a
plant-like characteristic as cellulose is a key component in plant cell walls.Then another strange thing happens. Imagine a
sausage standing up on end. Now imagine that you poke down on the top end with your finger, in such a way that you
push the tip downwards into the sausage. Imagine now that this indented tip keeps on going right down through the centre
of the sausage until it reaches the bottom. In other words, the whole mass is turning itself inside out. While the tip is
thrusting itself down through the centre of the sausage, the lower and outer parts of the sausage are making their way to
the top up the outside. This is exactly what happens to the slug.As cells at the tip turn to move downwards they secrete
cellulose. Thus they turn into a long, slender cellulose stalk (up to about 13 mm, or inch). 6 The very last group of
amebae to reach the top do not move on down the inside, but sit there like the king of the castle. Thus, the long, slender
stalk now has a rounded blob on the top. This structure is called the fruiting body, and looks very much like the fruiting
body of many fungi, which is the primary reason that slime moulds were grouped with the fungi for many years.The
amebae that end up perched on top of the stalk (the fruiting body) become the spores. This transformation from amebae
into spores is a complex process, but basically, each ameba at the top produces, and then wraps itself in, a cellulose cell
wall, which acts as a weather-resistant overcoat. Now it is a spore, lying dormant and ready to be carried off by the first
breeze, or by sticking to some passing creature. The purpose of this whole inside-out manoeuvre appears to be to form a
rigid stalk high enough for the wind, etc., to spread the spores. (About 70% of the amebae in the original slug become the
thick-walled spores of the fruiting body, with the remaining 30% sacrificing themselves to form the supporting stalk.)1
The spores can travel enormous distances before landing. Every time you inhale, you are probably breathing in numbers
of such spores. If a spore happens to land where conditions are favourable (e.g. moist ground with lots of bacteria), it will
germinate and take on the adult form. Thus the cycle of life continues.
Major problem for evolutionists
So what exactly are slime moulds? Despite the similarities, they cant readily be classified as fungi, plants, multi-celled
animals, bacteria, or protists.7 But the core of the problem is not which group to put slime moulds in, but why bother doing
so? Evolutionists operate on the premise that all life is related, believing that we all have a primeval slime single-celled
organism as our ancestor. Consequently, they classify species by assigning them to positions in their evolutionary family
tree according to how closely related they appear to be.8Trying to support their theory of lifes origins, and somewhat
challenged by the staggering complexity of single-celled organisms like Dictyostelium, evolutionists struggle to find a
place for the slime moulds. In fact, experts recently acknowledged, People within and outside our field dont know where
to put Dictyostelium.2 As more is learnt of these creatures, the more difficult it becomes. For example, the cellular slime
moulds have been divided into two sub-groups which, as one evolutionist writes, may not be closely related to each
other!9 Other evolutionists have made the revealing observation about cellular slime moulds that it is apparent that
they do not exist to fit into neat categories!1 Clearly, the existence of slime moulds is a conundrum for evolutionists.
What about parasites?
by Robert Gurney

If everything was
very good, why are there harmful
parasites?When I worked as a doctor in a bush hospital in
Tanzania, East Africa, I treated many patients for a disease
called Bilharzia, or Schistosomiasis. This is a parasitic disease
caused by several species of fluke, or flatworm, of the
genus Schistosoma. The particular form of the disease which I
treated was the one caused by the species Schistosoma
haematobium.1This worm, like so many other parasites, has a
bizarre and complex life-cycle which defies an evolutionary
explanation.It assumes seven different forms as it progresses
through its life-cycle, and part of that cycle takes place within
an intermediate host, a freshwater snail. The complexity of
such organisms points to a designer; but the existence of
disease-causing parasites and microbes does also raise
questions concerning the very good creation and the global
Flood.
Evidence for creation
As far as their physical structure is concerned, all living creaturesincluding the most basic and simple forms of life
(one-celled organisms)are stupendously complex, sophisticated, computerized machines. Molecular biologist Michael
Denton has estimated that if we knew how to build a machine as complex as the cell, it would take at leastone million
years to build one cellworking day and night, and churning out the parts on a mass production basis. 2One kind of
complexity is the way in which some animals exist in several completely different forms during their life-cycles. The
butterfly, for example, exists as an egg, a caterpillar, a chrysalis and a butterfly. Within the chrysalis, the internal organs of
the caterpillar dissolve into a soup, and then this soup organizes itself into a beautiful, incredibly complex, adult butterfly!
The whole process, from egg to butterfly, is coded in the genes, and is a marvel which defies any kind of evolutionary
explanation.3The humble Schistosoma may not excite our admiration in the same way; but during its life-cycle it exists
in seven different forms (including the egg)! Again, it is a marvel which defies an evolutionary explanation.
The origin of harmful parasites
\ If disease-causing parasites did not exist in the original creation, how did they come to be here now? We do not know
the answer for certain, but we can reasonably speculate. These parasites must have been benign and beneficial in their
original form. Perhaps some were independent and free-living, and others had beneficial symbiotic relationships with
animals or humans. These once-harmless creatures degenerated, and became parasitic and harmful. So there is no
compelling reason why the designer should not have caused these organisms to become pathogenic (disease-causing)
miraculously. However, there are other methods He could have used.Perhaps some became parasitic as a result of
mutations. They degenerated, lost their ability to live independently or symbiotically, and became harmfully dependent on
their hosts. Note that mutations overwhelmingly cause loss of genetic information. Such degenerative changes are
evident in disease-causing microbes like a Mycoplasma that causes a type of pneumonia and the germs that cause
leprosy and cholera.5Other kinds of genetic change may have been involved too. For example, microbes can swap genes.
The bacterium that causes bubonic plague probably resulted from more than one kind of change. 6In many cases,
however, the life-cycle of the parasite is so complex that new genetic information may have been needed. Mutations do
not provide new genetic information; so the information may have been there from the beginning. However, it was in a
switched off mode. Another possibility is that some parasites had symbiotic relationships with animals, and they became
parasitic and harmful only when they invaded humans. Also, degeneration in the human immune system could have
contributed to our susceptibility to parasitism by formerly benign organisms.If some of the people who survived the global
Flood, how could they have carried all the disease-causing parasites
and microbes which can survive only in humans? Possible answers are
not difficult to find, however, if one is prepared to look for them.The most
obvious answer is that most, if not all, present-day pathogenic
organisms that can dwell only in humans did not exist in their present
form at the time of the Flood. There has been plenty of time for them to
develop since the Flood. Many pathogenic organisms which existed in
the pre-Flood world may have perished in the Flood, but others have
taken their place.Skin vesicles on the forearm, created by the
penetration of Schistosoma.One suggestion is that some pathogens
which can survive only in humans today were much less specialized at
the time of the Flood.7 At that time they were able to flourish in a wide
variety of animals; but since then, they have become entirely dependent
on human hosts. Common viral diseases of humans today may well have derived from animal diseases. A New
Scientist report states:Just as historians such as William McNeill, of the University of Chicago, and other researchers
trace smallpox back to cowpox, so measles probably evolved from rinderpest or canine distemper, and influenza from hog
diseases.8Some of Dr Wielands suggestions 7 involve organisms which were already pathogenic (usually in animals)
when the Flood began. Asymptomatic carriers could have carried some diseases. All the organisms which became
pathogenic however, had to be derived from organisms which were entirely benign.

Technical information: Schistosomiasis life cycle

Wikipedia

Life-cycle
of Schistosoma
haematobium
(Click for larger image)
The parasite begins life as an egg, which is shed into the ureter or the bladder of the human host. It is passed out of the
body with the urine, and on contact with fresh water, it hatches to release a free-swimming miracidium. This is a ciliated
organism, swimming by means of its many hair-like cilia. It finds a fresh-water snail and penetrates the snails foot. Here it
transforms into a primary sporocyst. Germ cells within the primary sporocyst then begin to divide and produce secondary
sporocysts. These migrate to the snails hepatopancreas. Germ cells within each secondary sporocyst then begin to
divide again, this time producing thousands of new parasites, called cercariae, which emerge daily from the snail host.
These have tails, and are highly mobile. They are the larvae capable of infecting humans. Penetration of the human skin
occurs after the cercaria has attached to and explored the skin. The parasite secretes enzymes that break down the skins
protein to enable penetration of the cercarial head through the skin. As it penetrates the skin, it loses its tail and transforms
into the migrating schistosomulum stage.The schistosomulum enters the blood stream and travels to the lungs, where it
undergoes further developmental changes before migrating to the liver. In the liver, it begins to feed on red blood cells, and
it is there that the nearly mature worms pair, the longer female residing in the gynaecophoric channel of the shorter male.
The adult worms are about 10 mm long. The worms ultimately migrate from the liver to the venous plexus of the bladder,
ureters and kidneys. When the parasites reach maturity, they begin to produce eggs, and these pass through the ureteral
or bladder wall into the urine.1

Swine fluIs it evidence of evolution?

Just as with AIDS and avian influenza (bird flu), the latest
virus alert is being hailed as proof that evolution is true.
In some places (e.g. Egypt), pigs were slaughtered in an effort
to guard against swine flu. But with human-to-human
transmission of the virus now confirmed, authorities in
countries like Australia are looking to increased quarantine
measures, including school closures and travel restrictions, to
limit its spread. Photo: stockxpert

by Carl Wieland and David Catchpoole

Published: 2 June 2009(GMT+10)
Those who say that the creation/evolution debate is really only of academic interest, with little relevance to our daily lives
(or to anything else1), must surely have been confronted by swine flus continuing domination of worldwide news
headlines.2,3 And of course, the secularists couldnt resist linking these tragedies with evolutionary propaganda.
Swine flu is evolution in action, shouted LiveScience in its headline,4 with these very confrontational opening
sentences as a challenge to any readers doubting evolution:
Anyone who thinks evolution is for the birds should not be afraid of swine flu. Because if theres no such thing as
evolution, then theres no such thing as a new strain of swine flu infecting people.For the rest of the population, concern is
justified.LiveSciences editorial director Robert Roy Britt went on to hammer the line that swine flu is an excellent
example of evolution at work. His article quoted Michael Deem, a bioengineer at Rice University in Texas: Yes, this is
definitely evolution. It was certainly clear from Britts words that he very much had creationists in mind as a target for
mocking:While much of the modern controversy over evolution centers around whether humans evolved from non-human
primates (scientists overwhelmingly agree this is the case), some people still try to poke holes in the theory of evolution,
one of the most solid theories in science. In addition to evidence from ancient fossils and modern DNA studies, one of the
many lines of evidence supporting evolution is that it can quite simply be seen in action among some species that evolve
particularly rapidly, such as fruit flies.But on no stage does evolution unfold more quickly or with more potentially sickening
or lethal consequences for humans than among viruses. It is, to pass on a scary phrase used among scientists and
marketers, viral evolution. And you could be the star host of this all-too-often deadly show.Other anti-creation voices, too,
were quick to jump on the creationist-mocking bandwagon. Typical was this blogger who posted on the Huffington
Post website:How many creationists are running to their doctors right now to get flu medication? If they are doing that,
theyve voided the concept of creationism. The swine flu is a virus that EVOLVED right in front of our eyes! It found a new
way to survive. You would not need a new flu vaccine every winter if creationism was real. Heck, even when a new flu
vaccine comes out, its already obsolete, because 23 more strains have already EVOLVED from the base flu strain the
vaccine is based on. The whole idea of creationism is "I reject your reality and substitute my own." The war on science
and knowledge is over folks, theres a new sheriff in town! 5And this offering in vulgarity came out of the mouth of well
known comedian Bill Maher, in his HBO television seriesReal Time with Bill Maher:You cant c--p all over Darwin and
stem cell research and global warming, then come crawling back to science when you want Tamiflu.6,7
Change in viruses is not evolutionary change
These claims that the H1N1 swine flu demonstrates evolution
are not the first to have been made about a virus. But as our
earlier articles on AIDS and bird flu make clear, and as we
shall again revisit shortly, changes in viruses certainly
represent change, but such changes are notthe sort of
changes that molecules-to-man evolution requires.But what
about the jump from pigs to humans, some readers might ask?
Well, the ability of an animal virus to jump the species barrier
has been previously discussed in creationist circlesthe idea
of a virus changing hosts turns out to be important for a
consistent creation model. Today there are viruses which infect
only humans. But if animal viruses could have later become
harmful to humans as well, then there is no such problem. In
fact, measles is believed to have originally come from a virus
(canine distemper) which normally only infects dogs.The AIDS
virus is actually believed to have very likely jumped from the green monkey population in Africa to establish a new host in
humans. And the 191819 Spanish flu outbreak, which swept the world and killed more than 20 million people (more
even than the just finished WWI, and more than the Black Death in 14 th century Europe), is believed to have possibly
started in birds and spread to humans..
So, how do we make sense of this?
Such changes are in no way evolution as normally understood, as this sort of change is simply not capable in principle of
generating even one small step along the assumed path of vent-mud-to-virologist biological change.Actually, no informed
evolutionist will try to argue that a virus represents a simple life form analogous to the beginnings of the evolutionary
process on Earth. The reason is obviousfor a virus to exist there must first be a full-blown self-reproducing organism. A
virus cant reproduce without the complex machinery 8 of a truly living creaturehence viruses are generally considered to
be not living. Since viruses are parasitic on cellular life, the first life could not have been anything like a
virus.Interestingly, Britt in his LiveScience article is clearly aware of this, and cleverly tried to counter it in advance under
the sub-heading But are they alive?:One of the little hole-poking exercises used by critics of evolution is to argue that
viruses are not alive. Tell that to the host.Viruses may be living or non-living, depending on the definition of life, Deem
explained in an email interview with LiveScience. Viruses + the host (pig or human) are definitely alive. So, this for sure is
an example of evolution in the living system of the virus + pig + human.But then Britt admits that David Schaffer, a
professor of chemical engineering and bioengineering at the University of California at Berkeley, takes a slightly different
view:"Viruses are not alive, in that they do not have the ability to replicate themselves independently, without infecting
and relying upon a cell to do so," Schaffer said. "That said, biological entities need not be alive in order to evolve."
Notice that although Deem and Schaffer are at odds with each other over whether viruses are living, neither doubts that
viruses evolve! But viruses, do not evolve. Rather, like actual living things, they do mutate (the term is properly applied)
and change. Readers, however, should not be misled by the evolutionary barrage over swine flu (and other viruses) into
thinking that such change represents evolution.In fact, Britts own words make clear that the swine flu virus (at least in its
present form), has likely become that way not by the generation of new genetic information (something essential for
pondscum-to-pigs evolution to have occurred), but from existing genetic information:
Viruses steal DNA that they find useful to their success.
Many viruses can easily incorporate ready-made genes from other viruses into their genomes, as explained at [the
website9] Understanding Evolution. This is a possibility anytime a host is infected with two different viral strains.
Thats likely whats happened with swine flu.
It appears the H1N1 swine flu may be a reassortment of the H (hemagglutinin) gene from typical North American pigs
with the N (neuraminidase) and M (matrix) genes from European pigs, Deems said. If so, this new virus is an example of

the importance of recombination in evolution. That is, evolution proceeds not only by small mutations of individual DNA or
RNA bases, but also by transmission of large pieces of genetic material from one individual to another.
So, despite the evolutionary handwaving there, no evolution! Rearrangement of already existing information doesnt
explain the newencyclopaedic information of more complex living creatures. Britt goes on to explain that if you contract a
run-of-the-mill flu that causes only mild symptoms but then also contract a really deadly influenza virus that heretofore
was only transmitted between pigs, then youre at risk. Of whatviral evolution? He says that the two viruses inside you
swap genes, and now youre the host of a newly evolved swine flu virus that can infect your whole family, your colleagues
at work, some people at the airport you later fly out of who touch the same armrest you held, and then some folks in the
country you fly to. Voila, pandemic!Newly evolved? No, as theres no new genetic information, just already-existing
genetic information reshuffled, in this case.Weve repeatedly shown how demonstrating change is not enough to
demonstrate a supposed evolutionary history of life on earth. (See Muddy Waters and Beetle bloopers). In fact, when the
biological changes generally used to argue for evolution are looked at in detail, they turn out to be the precise opposite of
what such a process would require (see The evolution trains a-comin (Sorry, a-goin in the wrong direction)). Bacteria can
change to become antibiotic resistant, for instance, but such changes result from a loss of information (see Superbugs not
super after all and Anthrax and antibiotics: Is evolution relevant?). This is hardly a recipe for heading upwards along the
presumed evolutionary path alleged to have turned amebas into alert airport screening staff capable of intercepting
passengers with swine flu.Incidentally, for the benefit of Messrs Britt, Maher, and the Huffington Post blogger, believers in
creation do NOT come crawling back to science for its undoubted benefits to mankind (including, e.g. Tamiflu).
Creationists not only readily acknowledge the important contribution of experimental science and medicine in
compassionately seeking to ameliorate the debilitating impacts of the Curse on peoples lives, but many creation-believing
scientists were responsible for those advances.It is perhaps especially ironic that Maher, of all people, should be accusing
creationists of ignoring demonstrated facts from experimental science. This is because, despite the incontrovertible
evidence for the germ theory of disease (first put forward by a creationist, Louis Pasteur), Maher has publically said that
he doubts it!10.
Bacteria trapped for millions of years under Antarctic ice
by Carl Wieland
Published: 23 April 2009(GMT+10)
Perito Moreno Glacier, Argentina
Science news sites are abuzz with the discovery of
an ecosystem trapped under an Antarctic glacier for
an alleged time of at least 1.5 million years. 1It was
not that long ago that the world of science believed
that nothing could survive in an environment that was
not only near freezing, but totally dark and lacking
oxygen. Since then, a range of creatures have been
found that thrive in the most amazing conditions. We
wrote about a number of such extremophiles
in Creation magazine of December 2001see Life
at the extremes.The latest discovery involves a pool
of iron-rich water 400 metres (1300 feet) underneath
the Taylor glacier in Antarctica. As long ago as 1911
Arctic explorers were drawn to a frozen waterfall-like
feature at the glaciers edge. This became known as
Blood Falls because of its striking bright-red
colour,2 which was thought to be due to red algae.
However, subsequent analysis showed that the
colour was due to rust (iron oxide) in the water that
flowed out from time to time from beneath the glacier before freezing rapidly.Due to the unpredictable timing of these
leakages from underneath the glacier, head researcher Jill Micucki, of Dartmouth University, took several years before
she was able to analyze the salty water, which was probably once part of the sea. Interestingly, it contained absolutely no
oxygen. In it were thriving colonies of bacteria that make a living without either oxygen or sunlight. They do this by
chemically transforming iron and sulfur compounds.The pool under the ice is estimated to be about 5 km (3 miles) wide,
and it was probably trapped, for instance in some fjord, when the Taylor Glacier reached sufficient size. The excitement
about the find is not because of the strange metabolism of these creatures, because this was known already from bacteria
in similar light-and-oxygen-free environments. The intense interest seems to have come from two main factors.One factor
is the desire to find life elsewhere in the solar system, most parts of which involve very inhospitable environments. This
would reinforce the evolutionary belief that life just happens by chance chemistry.3The other is the fascination at finding a
time capsule of things believed to have been living in total isolation for millions of years. But the facts fit perfectly well
with these bacteria having lived there only since they were trapped during the great post-Flood Ice Age. In fact, the latter
explains it somewhat better, because bacteria have extremely short generation times, sometimes even minutes. That
means that a huge number of generations would have occurred over even a million years. So these bacteria would have
had the equivalent of hundreds of millions or even billions of years of human generations. This is sufficient to
evolve radically different formsif the large timescale were real and the neoDarwinian mechanism of mutations and
natural selection were adequate to turn microbes into mammoths.Ann Pearson, a Professor of Earth and Planetary
Sciences at Harvard, says that the species living there are similar to contemporary organisms, and yet quite different.
Not that different, however. The main species discovered from the Blood Falls outflow is apparently Thiomicrospora
arcticathe same species already known and described years earlier. Its species name derives from its discovery under
ice at the opposite end of the worldthe Arctic. With these trapped Antarctic bacteria remaining the same species as
those not entombed with them, it makes much more sense for the entrapment to have been thousands, not millions of
years ago.
Germs miniature motor has a clutch
by Jonathan Sarfati
Published: 8 July 2008 (GMT+10)
Bacterial flagellum: powered by an electric motor

Figure 1. The bacterial flagellum (with rotary motor) has many

features which people recognize as design features, such as a
clutch.
Many bacteria are powered by real electrical outboard motors,
only 45 nm in diameter.1 These motors connect to long, thin,
whip-like helical filaments several times as long as the germ,
via a universal joint. This converts the rotary motion of the
motor into wavelike motions in the filament. The motor
comprises a stator, rotor, drive shaft and bushing that guides
the driveshaft out through the cell wall. The assemblage of
motor and filament is called a flagellum.1Bacteria often have
several flagella, and their concerted motion enables the cell to
swim at 35 cell lengths per second.1While our electrical motors
are powered by a negatively charged current (electron flow in
wires), the flagellar motor is powered by positively charged
current. This is a flow of hydrogen ions (protons, H +), from the
outside to the inside of the cell (except for marine bacteria and
bacteria that live in very alkaline conditions (i.e. low
concentration of H+), where sodium ions are used instead). The
proton movement is driven by either an electrical or pH
gradient, and the energy to generate this gradient comes from
the oxidation of its food. The proton flow changes the shape of
one of the stator proteins, which exerts a force on one of the rotor proteins, thereby driving the rotor.1 A recent article said:
The flagellum is one of natures smallest and most powerful motorsones like those produced by B. subtilis can rotate
more than 200 times per second, driven by 1,400 piconewton-nanometers of torque. Thats quite a bit of (miniature)
horsepower for a machine whose width stretches only a few dozen nanometers.2
A clutch
This same article reported on another astounding discovery:
that this motor even has a clutch to disconnect the motor from
Bacterial flagellum with rotary motor, with the
the filament. Scientists from Indiana University Bloomington (IU)
following features
and Harvard University actually discovered this by accident
when researching biofilms.3
These are slimy sheets a fraction of a millimetre thick that form
on any surface that has a supply of nutrients and water,
including teeth and pipes.4
IU biologist Daniel Kearns, the project leader, explains:
We were trying to get at how the bacteriums ability to move
and biofilm formation are balanced. We were looking for the
genes that affected whether the cells are mobile or stationary.
Although B. subtilis is harmless, biofilms are often associated
with infections by pathogenic bacteria. Understanding biofilm
formation may eventually prove useful in combating bacterial
infections.2,5That is, the fast and furious motions of the bacteria
might disrupt the formation of biofilms, so the bacteria need
some means of stopping it quickly. The researchers discovered
that a protein called EpsE was responsible somehow. But how
did it work? There were two possibilities: One possibility is a
brake, locking up the motor so preventing it spinning; another is
Self assembly and repair
simply to disconnect the motor from the filament, just as a clutch
Water-cooled rotary engine
in a car disconnects the drive wheels from the engine.To decide
Proton motive force drive system
between these options, the researchers attached the filaments
Forward and reverse gears
to a glass slide, and observed the bacterium. The flagellar motor
Operating speeds of up to 100,000 rpm
was powerful enough to turn the entire germ once every five
Direction reversing capability within 1/4 of a turn
seconds, without EpsE. If EpsE were a brake, then the
Hard-wired signal transduction system with shortbacterium would also be unable to turn, like the wheels on a
term memory
braked car; if it were a clutch, then the bacteria would be free to
Clutch to disconnect filament from motor when
rotate if powered by another source, like the wheels of a car
required
coasting down a hill on neutral, powered by gravity. It turned
[from Bacterial Flagella: Paradigm for Design,
out that the bacteria with the protein present could indeed rotate
video,
passively, powered by the random collisions of molecules
<www.arn.org/arnproducts/videos/v021.htm>, and
(Brownian motion6). In other words the filament freewheels.This
the update in this article for the last one]
molecular clutch, EpsE, is thought to dock on the flagellums
rotor, a donut-shaped structure at the base of the flagellum.
There, EpsE interacts with one of the rotor proteins, called FliG,
which changes the rotors shape so that it disengages from the engine. Or as described in the perspective:
Motile cells are powered by interaction of the FliG protein with the MotA/B complex (which generates torque). The protein
EpsE acts as a molecular clutch to disengage the rotary flagellar motor, leaving the flagellum intact but unpowered. This
shuts down motility and facilitates biofilm formation. 4This clutch mechanism is very efficient: it means that the germ needs
to make only one protein to halt the powered filament motion, and this takes only 15 minutes. It also preserves the motor
intact, so it could reactivate if necessary, rather than needing to be rebuilt from scratch. There also may be an advantage
to building biofilms if the filaments were free to rotate in neutral rather than stopped rigidly.4
Design or evolution?
Project leader Daniel Kearns made the obligatory vacuous homage to evolution (cf. these chameleon researchers):
We think its pretty cool that evolving bacteria and human engineers arrived at a similar solution to the same problem:
How do you temporarily stop a motor once it gets going? 2It would make more sense to say:We think its pretty cool that
human engineers solved the problem: How do you temporarily stop a motor once it gets going? with a clutch, while the
Designer of the bacterial flagellum had anticipated that solution.

Even a tiny virus has a powerful mini-motor

by Jonathan Sarfati
Viruses are particles so tiny that they cant be seen by an ordinary light
microscope, but only under an electron microscope. They are not living
organisms because they cannot carry out the necessary internal metabolism
to sustain life, nor can they reproduce themselves. They are infectious
particles, made up of DNA (or RNA) and protein, and can reproduce
themselves only by hijacking the machinery of an infected living cell. The
infected cell produces multiple copies of the virus, then bursts to release the
new viruses so the cycle can repeat.Viruses come in many different sizes,
shapes and designs, and they operate in quite diverse ways. One of the most
common types is the bacteriophage (or simply phage) which infects bacteria.
It consists of an infectious tailpiece made of protein, and a head capsule
(capsid) made of protein and containing DNA packaged at such high pressure
that when released, the pressure forces the DNA into the infected host
cell.How does the virus manage to assemble this long information molecule at
high pressure inside such a small package, especially when the negatively
charged phosphate groups repel each other? It has a special packaging
motor, more powerful than any molecular motor yet discovered, even those in
muscles. Douglas Smith, an assistant professor of physics at UCSD,
explained the challenge:The genome is about 1,000 times longer than the diameter of the virus. It is the equivalent of
reeling in and packing 100 yards of fishing line into a coffee cup, but the virus is able to package its DNA in under five
minutes.1Dr Smith and some colleagues at UCSD joined researchers from the American Catholic University (Washington,
DC) to solve the problem.2 They analysed the bacteriophage T4 (above right)a virus that infects E. coli bacteria, the
type that inhabit human intestinesusing laser tweezers to hold onto a single molecule of DNA, and measure the force
on it by the viruss packaging motor.They showed this motor exerts a force of >60 piconewtons. This sounds small (610 11
N), but for its size, its twice as powerful as a car engine. So the motor, a terminase enzyme complex, can capture and
begin packaging a target DNA molecule within a few seconds. 2 Such a motor must use a lot of energy, since the phospate
groups are negatively charged so repel each other. So in one second, this one goes through over 300 units of lifes energy
currency. This energy currency is the molecule ATP (adenosine triphosphate),3 and this itself is generated by a remarkable
molecular motor, ATP synthase.4 The virus has a complementary motor-enzyme, ATPase, built into its packaging engine,
to release the energy of the ATP.And not only is the packing motor powerful, it can change its speed as if it had gears. The
researchers say that this is important, because the DNA fed to it from the cell is likely not a straightforward untangled
thread. Dr Smith said:
Just as it is good for a car to have brakes and gears, rather than only being able to go 60 miles per hour, the DNApackaging motor may need to slow down, or stop and wait if it encounters an obstruction.Doug Smith, researcher.Just
as it is good for a car to have brakes and gears, rather than only being able to go 60 miles per hour, the DNA-packaging
motor may need to slow down, or stop and wait if it encounters an obstruction.1
A report said:
It may permit DNA repair, transcription or recombinationthe swapping of bits of DNA to enhance genetic diversityto
take place before the genetic material is packaged within the viral capsid.1
Other vital machines
This motor is just another example of the complexity required even for sub-life forms such as viruses to exist, let alone
real life. Since life requires long molecules to store information and pass it on to the next generation, there must also be
machinery just to deal with its awkward physical properties before life can even get started by chemical evolution.Here are
two more machines just to deal with the long thready properties of DNA, so that life can function.
Separating the double helix
For replication, the two strands must be separated so a copy can be made. The strands are separated by a molecular
motor called helicase. This is a ring-shaped molecule that lies on the replication fork, where the two strands separate.
Helicase pulls one strand through its hole, while the other strand is shuttled away. 5Helicase also doesnt need to wait
passively for the fork to widen; rather, researchers from Cornell University showed that it opens the forkactively.6 One of
them, Michelle Wang, said, Basically, it is an active unwinding motor. 7 However, the unwinding is much faster in cells
than in the test tube, so Dr Wang suggested, accessory proteins are helping the helicase out by destabilizing the fork
junction.Since replication is vital for life, helicases are vital to all living organisms. Dr Wangs colleague Smita Patel
pointed out also, Helicases are involved in practically all DNA and RNA metabolic processes. Further, as Dr Patel
explained, Defects in helicases are associated with many human diseases, ranging from predisposition to cancer to
premature aging. So the origin of such elaborate machinery and the energy source is just one more problem for chemical
evolution to solve.
Transcription and the scrunching machine
Even the copying of the right part of the DNA to mRNA so that its information can be read requires intricate machinery.
This involves an enzyme called RNA polymerase, comprising four protein chains. And another protein tells the RNA
polymerase where to start reading the DNA template. Then the enzyme complex moves along the DNA strand, adding the
matching RNA letters one at a time, then stops in the right place.Richard Ebright and his team from Rutgers University
have discovered more intricacies in this process of transcription,8,9,10 Indeed, it is this mRNA that is translated into proteins
in the complex machines known as ribosomes.DNA is double stranded and only one strand is copied, so it must be
unwound for copying. The copying machine, called RNA polymerase (RNAP), first locks on to the start of the gene (i.e.,
protein-coding sequence). The anchored RNAP then reels in the DNAa process termedscrunching.11 This unwinds the
double strand so the mRNA copy can be formed off one of them. Also, the unwinding process stores energy, just like
winding the rubber band of a rubber-powered airplane. And just like the toy plane, this energy is eventually released, with
the machine then breaking free of its starting point and shooting forward. This also rewinds the unwound DNA
(unscrunching) which then escapes from the back of the machine.12
Conclusion
Life depends upon the long double-thread information molecule DNA, and it could not function without machines capable
of dealing with such long double-threaded molecules. Yet the information for these machines is coded on the threads!
These machines require the ATP synthase motor to generate and use their energy, yet this motor is also coded on the

DNA. The code needs the machines, and the machines need the code. Life presents us with many such chicken-andegg problems for which naturalistic theorists have no answer.
New bacteria show wonder upon wonder
by David Demick
A recently discovered new type of bacteria is raising eyebrows among microbiologists, organic chemists and
environmental scientists. Their unique properties are creating research challenges in nitrogen chemistry and membrane
structure. They efficiently perform a chemical operation that was thought to be impossibleconversion of ammonia to
pure nitrogen in the absence of oxygen (see figures 1 and 2). These bacteria are named after their distinctive metabolism
as anammox, which is short for anaerobic ammonia oxidation. It is being suggested that they show how life evolved in
the early earth by providing a link between prokaryotes and eukaryotes. However, on closer inspection, their unique
features make them fit in better with a creationist view of the microbial world.
Figure 1. Anaerobic ammonium oxidation (anammox) carried
out by anammox bacteria, where ammonium and nitrite are
converted
into
dinitrogen
gas.
Click here for larger view
Waste tanks and deep waters
The story of their discovery begins with different but related
observations. In two very different environments (deep anoxic
areas in natural water bodies, and sealed oxygen-free organic
waste tanks), ammonia content was noticed to be less than
expected. In addition, the waste tanks also showed production
of nitrogen gas. This anaerobic ammonia consumption coupled
with nitrogen production was suspected to the work of bacteria,
but the chemical reactions involved were unknown in nature.
There was considerable skepticism until a Dutch
microbiologist, Gijs Kuenen, succeeded in isolating the
responsible organism from the waste tanks. It was verified that
the new bacteria perform the hypothesized anammox reaction,
converting ammonia and nitrite to nitrogen and water under
anaerobic conditions. However, this was only the beginning of
surprises to come from these newly recognized organisms.
A rocket-science puzzle
The new bacteria were provisionally named Brocadia
anammoxidans, after their bright red colour and distinctive chemistry. Electron microscopy revealed something else
unexpectedunlike most other bacteria, these had an internal membrane-bound organelle. Furthermore, this organelle
was unique. It appeared to be crucial for the nitrogen-metabolizing process, and was accordingly named an
anammoxosome. For still another surprise, it was found to contain hydrazine, a highly toxic and explosive nitrogen
compound used in rocket fuel. Hydrazines hydrophobic structure is such that it should diffuse easily through most
biological membranes, and hence it should kill the surrounding bacterium. How could these little organisms contain and
store such a dangerous substance?
A new molecule that looks like a ladder
Figure 2. The stoichiometry (quantities of reactants and
products)
of
anammox
(from
Strous et
al).5
Click here for larger view
The answer was found to lie in a novel organelle membrane
structure, which provoked such praise from organic chemists
as extraordinary and even impossible. The membrane turned out to owe its remarkable impermeability to a tight, ladderlike, carbon-rich, multi-ring structure made of linked cyclobutanes, and descriptively termed ladderane (see figure 3). It is
still not clear just how the microbes make this remarkable and unique membrane material, or why they use it to contain
hydrazine. It is thought that the hydrazine may function as a nitrogen-containing intermediate form, and that its high
energy content may drive the necessary nitrogen-forming reactions. It is clear that the anammox bacteria will keep
researchers busy for some time to come. In the meantime, a synthetic process for making ladderane has been patented,
and it is hoped to eventually have applications in the electronics field.1
Evolutionary headache
What of the origin of these strange organisms? The anammox bacteria have been classified with the bacterial phylum
Planctomycetes. This is a recently recognized group of bacteria which have internal membrane structure, unlike most
prokaryotes. Some have an internal membrane that surrounds the genome, termed a nuclear body or nucleoid. Other
internal structures, such as the anammoxosome, seem to be primitive organelles. This gives them features similar to
eukaryotes (although the nucleoid is not a true nucleus). They also have a protein-rich cell membrane that is free of
peptidoglycan, a feature which they share with the Archaea (a different group of bacteria-like organisms). On the other
hand, their genome structure is closest to the true bacteria, so they have some features in common with three great
domains of living things. Hence, evolutionists hope that some of the Planctomycetes, including their newest anammox
members, will qualify as links between hypothetical earliest forms of life and subsequent diversifying organisms.
However, attempts to fit these novel prokaryotes into an evolutionary scheme are proving elusive. 2Similarities are
overshadowed by complex and crucial differences. For example, the anammoxosome is completely unique, both in
membrane structure and metabolic chemistry. It cant be called a stepping stone toward or from other organelles. The
anammox microbes fit better into a creationist view of bacteria, in which these tiny creatures are a multitude of distinct and
unique organisms, without evolutionary relationships, created to reproduce after their own kind. Their -given purpose is to
maintain a chemically livable biospherespecifically, in this case, to help regulate nitrogen balance and availability.

Figure 3. Ladderanes X and Y found in anammox lipids

(above), and an example of the chemical structure of
ladderane
Y
(below).
Click here for larger view
Millions of yearsor a superintelligent designer?
Origins questions aside, scientists are scrambling to exploit
these bacteria in sewage-treatment facilities and similar sites
for detoxifying ammonia-rich waste, where they are expected
to greatly improve efficiency and lower toxicity. They are also
being found to be much more widespread in the biosphere
than previously thought. They appear to play a previously
unsuspected and important role in the global nitrogen cycle,
especially in the oceans.3This eagerness to exploit the
anammox bacteria is part of a growing trend in technology to
look to nature for efficient solutions. Scientists are realizing that
living things are full of high-tech surprises, just waiting to be
copied and used. In areas as diverse as weak or strong adhesives (gecko feet and mussels) and streamlining of motor
vehicles (boxfish design), engineers are finding unexpected but highly useful inventions ready-made for them. Janine
Benyus, co-founder of the Biomimicry Guild, comments:
If you have a design problem, natures probably solved it already. After all, its had 3.8 billion years to come up with
solutions . The truth is, natural organisms have managed to do everything we want to do without guzzling fossil fuels,
polluting the planet, or mortgaging the future.4
Sadly, as this comment illustrates, millions of years often takes the credit for these ingenious natural designs away from
the designer.4
More Sleeping Beauty bacteria
Once again, bacteria have been revived from Antarctic ice said to be eight million years old.
by David Catchpoole
The glacial ice samples claimed to be eight million years old
were collected from Antarcticas upper Beacon Valley (not
shown here). But after millions of years, even in the freezer,
the microbes should long ago have disintegrated. The fact that
intact microbes were resuscitated is actually powerful
evidence for what the young age model says is
their actual maximum age.Our earlier article Sleeping Beauty
bacteria commented on news reports that scientists had
revived eight-million-year-old bacteria from frozen Antarctic
soil.Well, its happened againthat is, other researchers have
similarly been able to revive bacteria found in blocks of
Antarctic ice, dated as being eight million years
old.1,2,3However, just as happened last time, other scientists are
urging caution, saying that as with most claims of ancient
microbes being revived, they would consider that
contamination of samples with modern microbes was a distinct
possibility.One can certainly understand their scepticism, given their acceptance of the age assigned to the ice. For
bacteria and other microbes to have been sitting dormant in ice for eight million years defies reason, given that the
organisms should have long ago fallen apart. The fragility of complex biological molecules (e.g. DNA) is such that, even
when protected from moisture, heat and other forms of energy such as radiation, they eventually disintegrate. This is from
the random effects of molecular motion and background radiation, consistent with the Second Law of Thermodynamics.
After all, experts have said that there shouldnt even be any DNA remaining after 100,000 years, let alone the complete,
intact machinery of a living cell.4Yet one cannot ignore the persistent reports of ancient microbes being revived from
frozen water and sediments,5and even from salt crystals said to have formed 250 million years ago!6 Note, too, that in
each case, the researchers were meticulous in their efforts to prevent contamination.
So whats the answer?
The problem is with the millions-of-years dating, which is based upon evolutionary and uniformitarian assumptions. Thus
the revival of bacteria from ice, frozen sediments, and salt crystals all dated as being millions of years old is instead
powerful evidence of the young age timeframei.e. the samples are much younger than evolutionists have claimed.
So the resuscitation of microbes is not that surprising, as the ice is no older than about 4,500 years (when the entire earth
was flooded for many months.