Beruflich Dokumente
Kultur Dokumente
Content
I
1
II
3
10
11
12
13
14
III
15
16
17
18
19
20
21
22
23
24
25
26
27
Molecular medicine:
genetics, genomics and
proteomics for
diagnosis and therapy
To d a y
To d a y
A
A
B
B
D
D
C
C
To m o r r o w
To m o r r o w
Diagnostics
Diagnostics
A
A
D
D
B
B
C
C
G
G
G
G
G
G
C
C
C
C
C
C
A
A
A
A
A
A
A
A
A
A
A
A
T
T
T
T
T
T
C
C
A
C
C
A
T
T
T
T
T
T
A
A
A
A
A
A
Pharmacogenomics:
genes and drug response
19
Proteomics:
seeing through the
undergrowth
31
Targets
for medicine
47
PCR:
an outstanding method
65
SNPs:
the great importance
of small differences
83
DNA chips:
choosy fish hooks
95
Basic conditions:
ethics, law and society
115
Prospects:
more knowledge for
medical science
127
A brief glossary
of terms
137
I
1
II
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
III
27
Molecular medicine: genetics, genomics and proteomics for diagnosis and therapy
chemical
synthesis
improvement
by computer
target molecule
potential
drugs
binding assay
biological
target search
rational
drug design
high-throughputscreening
10
best candidate
efficacy studies
biological
evaluation
computer so as to interact in a highly specific way with a certain part of a target molecule; only afterwards would they be
produced by chemists. This approach has not yet brought the
success that was hoped for.
High-throughput screening New, automated methods
then made it possible to test large numbers of substances in
hundreds or even thousands of miniaturised assays for certain biological, chemical and physical properties. Experiments
such as binding to a target, which previously had to be performed individually in molecular biology laboratories, could
now be performed in this way. The desired properties of suitable molecules can then be improved in a further step. Highthroughput screening has already resulted in the development of a number of particularly effective drugs.
Molecular medicine: genetics, genomics and proteomics for diagnosis and therapy
11
A multiplicity of possible
causes
12
cell membrane
nucleotidebinding
region
regulatory
region
chloride
Defect on chromosome 7.
Molecular medicine: genetics, genomics and proteomics for diagnosis and therapy
13
14
environment
genes
nutrition
disease
lifestyle
Molecular medicine: genetics, genomics and proteomics for diagnosis and therapy
15
The importance of looking for the causes of disease has not changed at all since Morgagni propounded his organ-based pathology: only if we
truly understand a disease can we treat it correctly. Nowadays,
1a, 1b,
2a, 2b,
3a
1b, 2a,
2b, 2c,
3a
4
4
5a
Over the past few years it has become increasingly clear that
even infectious diseases are subject to a complex set of
genetic influences. This is true not just in regard to the
causative pathogens whose genetic background is often
well known but also in regard to the host. Genetic differences between individuals make some people more, and
others less, susceptible to particular infections. In addition,
our genes influence the way our body deals with all types of
drug, including anti-infective agents. In particular, drugs for
use against viruses, which evolve rapidly, often show unsatisfactory efficacy and troublesome side effects. Therefore, if
better drugs are to be developed, both the genome of the
pathogen and that of the patient must be taken into account.
An example of an infectious disease that shows this kind of
16
complexity is hepatitis C,
which is expected to become
far more common over the
next few decades. Untreated,
this disease leads to cirrhosis
of the liver in about 20% of
those infected and to liver
1b
cancer in a smaller proportion
2a
of patients. The responsible
1b,
pathogen, hepatitis C virus
1b, 3a
6
(HCV), occurs in at least six
different types, plus subtypes,
3b
that occur with different frequency in different parts of the
1b, 3a
world (see figure). The existence of these variants has a
major influence on the effectiveness of standard therapies.
For example, interferon, the
most important agent for use in this disease, is relatively
ineffective against HCV type 1, the predominant type in
Europe and North America. Moreover, this type of HCV
generally causes more severe disease than other types.
In addition, standard interferon preparations usually cause
severe side effects, all the more so because they generally
have to be taken three times weekly. Administration at such
short intervals is necessary because interferon is broken
down very rapidly in the body and therefore acts for only a
few hours. Patients are thus obliged to put up with a constant ebb and flow of side effects. Improved drugs such as
pegylated interferon have been available for some time
now. Used in combination with ribavirin, pegylated interferon significantly increases the efficacy of treatment.
Molecular structure of Pegasys with (right) and without (left) PEG coating
chronic course, be mild or severe, and undergo spontaneous cure or lead to liver cancer, the direction taken by the
disease in an individual being largely determined by that
individuals genome. At present, however, the genes
responsible for these differences are largely unknown.
These genes are therefore another important object of HCV
research, since depending on the genetic predisposition of
the patient (and the virus), treatment may be necessary or
unnecessary, a particular drug may be suitable or unsuitable, and cure may be possible or unlikely. Therefore, the
more is known about the relationships between the genome
of the pathogen and that of the host, the more specific will
be the drugs that can be developed for use in this disease.
Molecular medicine: genetics, genomics and proteomics for diagnosis and therapy
17
At the same time, these new methods and discoveries are a continuation of the revolution that began several hundred years ago
less emphasis on signs and symptoms, more investigation of
causes. And despite all the progress that has been made, the possibilities offered by medicine remain limited: the interactions
between our genes are so complex, our body is so adaptable, and
the influence of our environment and lifestyle is so great that we
cannot expect to find definitive answers and one-hundred-percent effective therapies in the very near future. Rather, genetics,
genomics and proteomics will first help doctors to avoid ineffective or even dangerous therapies after all, there is no such
thing as a panacea. But even this is a major step forward.
References
Lindpaintner K: Pharmacogenetics and the future of medical practice: conceptual considerations.
Pharmacogenomics 1: 23-26, 2001
Bundesministerium fr Bildung und Forschung (ed.): Das nationale Genomforschungsnetz.
Bonn, 2003
Geschftsstelle des Wissenschaftlichen Koordinierungskomitees des Deutschen Humangenomprojekts (ed.): Das Humangenomprojekt 1st and 2nd edition
Healthnet-Services GmbH: (M)Eine Geschichte der Pathologie, Teil 1-7:
http://hns.pvs-bw.de/mod.php?mod=userpage&page_id=30
18
Pharmacogenomics:
genes and drug response
To d a y
To m o r r o w
Diagnostics
B
C
People are all different thats obvious at first glance. There are
extra-long beds, creams for sensitive skin, height-adjustable seat
belts, tailored shirts and three dozen standard shoe sizes. Each
of us has different strengths and weaknesses, abilities and needs.
Environmental factors, chance and above all the small differences in our genomes make each of us unique. But if one person
finds a standard bed to be too short and another finds a standard
shoe size too big, why should we assume that everyone responds
to drugs in the same way?
Terms
In fact, it has been known for
some time that the efficacy
Pharmacogenetics describes the influence of genes on the
and tolerability of drugs
efficacy and side effects of drugs.
Pharmacogenomics studies interactions between drugs and
vary from one person to the
the genome.
next. Thus, some patients
Pharmacokinetics investigates the uptake, conversion and
need a lot more or a lot less
breakdown of drugs in the body over time. Environmental factors,
diet and genetic predisposition all play a role.
of a given drug than most
Pharmacodynamics deals with the influence of genes on the
people; side effects keep ocinteractions between drugs and their molecular targets.
curring unexpectedly; and
sometimes a drug that is usually highly effective does not work at all. Our uniqueness is
reflected in our bodys response to drugs. Because of this, personalised medicine has emerged as a hot new topic of discussion. Future drugs, it is hoped, will be better adapted to our genetic diversity and dissimilar life circumstances and will be
more efficient, more specific and safer. And they will be supported by a battery of fast, simple genetic tests that will enable
doctors to select the right drug for their patients specific needs.
First example
Herceptin
20
As early as 1958 the German pediatrician Friedrich Vogel suspected that our genes play an
important role in determining our response to
drugs. He even proposed a name for the branch of science that
investigates this phenomenon: pharmacogenetics, the study of
the influence of genes on drug effects. The new approach quickly led to the first application of genetics in medicine. Over 100
relevant genes are now known, and many more will follow as scientists rapidly refine the field of pharmacogenetics with the help
Pharmacogenetics
as a research discipline
21
22
effect:
target molecule(s)
etiological or palliative
potential side effects
conversion:
not always
necessary
cleavage of a
protective group
uptake:
e.g. via specific
receptors or channels
breakdown:
cleavage of
protective groups
often by modification
and/or cleavage
new compounds
may be formed
heart
gut
excretion:
often possible only after
conversion or breakdown
: (inactive) drug
: (inactive) drug in the bloodstream
: protective groups
: active drug
ureter
: cleaved, water-soluble
degradation product
or prevent the action of a drug. Pharmacodynamics describes these underlying etiological differences.
If a drug does not act on the cause of a disease but
rather on its manifestations, many genes may be involved in and interfere with its effects. These palliative
differences are also the subject of pharmacodynamic
research.
23
Prominent example:
cytochrome P450
24
Safety Can adverse reactions to a specific drug be predicted? Can the patient be switched to another drug that
he/she is more likely to tolerate? Are there alternatives? Can
provisions be made in advance for medical supportive measures?
Prevention If the cause of the disease is related to genetic factors, the disease can be diagnosed early by means of
tests and possibly avoided by initiating specific measures such
as diet or exercise.
As with all proteins in the body, this variable activity ultimately depends on our genome, which,
in turn, is affected by external factors, including
drugs. Cytochrome P450 is a good example of this phenomenon
as well. The 50-plus genes that code for this family of enzymes
can be activated or suppressed by drugs. In this way drugs may
affect each other, intensifying or cancelling each others effects,
even if they have completely unrelated targets.
Thus, the genome and drugs form a complex network of dependencies and uncertainties, of effects and side effects, which
ultimately makes our bodys individual response to drugs unique, especially when pharmacodynamic factors are also taken
into consideration. Pharmacogenomics can elucidate this interplay and help doctors find out whether, at what dose and with
what risk a drug can be administered to a given patient before
problems arise in the first place.
25
26
10,000 experiments at
once: DNA chips
27
28
References
Lindpaintner K: Pharmacogenetics and the future of medical practice: conceptual considerations.
Pharmacogenomics 1: 23-26, 2001
Lindpaintner K: Herausforderungen und Verheiungen einer individuell zugeschnittenen
Behandlung komplexer Krankheiten. Roche, 2000
Frobse R, Albrecht H: Die ganz persnliche Pille. DIE ZEIT, 15/2002
Ma MK et al.: Genetic basis of drug metabolism. Am J Health Syst Pharm 59(21): 2061-2069,
2002
Kroll W, Hartwig W: Pharmakogenomik. Nachrichten aus der Chemie 50, March 2002
Lifescience.de: Pharmacogenomik Die Suche nach den idealen Pillen.
http://www.lifescience.de/ratgeber/mitte/index2.html
Human Genome Project Website: http://www.ornl.gov/
Human Genome Project Pharmacogenomics:
http://www.ornl.gov/hgmis/medicine/pharma.html
29
Proteomics:
seeing through the undergrowth
The importance of proteins lies in the multiplicity of the tasks that they perform. They play a central role in almost all the processes involved in the
life of an organism or viewed on a smaller scale a cell:
Diverse structures
and functions
32
}
}
primary structure
secondary
structure
tertiary structure
A chain of up to twenty different amino acids (primary structure the variable regions are indicated by the squares of different colours) arranges itself into three-dimensional structures. Among these, helical and planar regions are particularly
common. The position of these secondary structures in relation to one another determines the shape of the protein, i.e.
its tertiary structure. Often, a number of proteins form functional complexes with quaternary structures; only when
arranged in this way can they perform their intended functions. When purifying proteins, it is extremely difficult to retain
such protein complexes in their original form.
quaternary
structure
33
amount and complexity of the relevant data and knowledge already available or yet to be obtained, attempts to intervene in the
world of proteins e.g. by means of drugs have until now been
based largely on the principle of trial and error. The new discipline of proteomics aims to change this.
Protein catalogues
to provide order
and perspective
Proteome research as a
link between disciplines
34
Strategy: standardisation
and automation
35
clinical sample
tissue
protein separation
(2D gel electrophoresis)
isoelectric point (pl)
blood
tissue
protein extraction
detail
spot picking
(each spot contains
a protein)
bioinformatics
protein identification by
database search
mass spectrometry
(of each spot)
The first step is always that of separating the mixture of proteins in a sample. The most important
method used to achieve this both in classical
protein analysis and in proteomics is two-dimensional (2-D)
gel electrophoresis. This technique has been used routinely since
the 1980s, and was in fact the subject of the conference in Siena
at which the term proteomics was first used.
In 2-D gel electrophoresis, the proteins in a sample are applied
to a rectangular piece of synthetic gel. Within this gel the proteins are separated firstly according to their charge and then at
Separation by charge
and size
36
Alzheimers disease
molecular weight
SDS-Page
IEF
tein reaches its isoelectric point, i.e. once its net charge is
zero, it stops.
In the next step, the separated proteins are further sorted
according to size. This occurs at a right angle to the direction of the first separation, i.e. in a second dimension. The
detergent sodium dodecyl sulphate (SDS) is added for this
purpose. The molecules of this substance bind to the proteins to an extent that depends upon the size of the protein molecules. Once again, an electric field is applied,
however in this step the rate of migration is determined by
the charge of the SDS, which in turn is determined by the
size of the protein molecules. This type of separation is
known as SDS-PAGE, or sodium dodecyl sulphate-polyacrylamide gel electrophoresis.
Proteins that are very similar (e.g. modified forms of the
same molecule) are often extremely difficult to separate in
conventional gels. In such cases use is made of narrowrange gels with an extremely gentle gradient within the pH
range being examined. In this way even minimal differences
in charge can be detected.
37
Presorting increases
resolution
In order to identify the cleaved protein, the peptide mixture is fed into a mass spectrometer. As its
name suggests, this device is able to measure the
mass (and thus the weight) of molecules. A peptide mixture to
be analysed is embedded in a carrier material a matrix and
subjected to a laser impulse. The matrix transfers the energy of
the laser to the peptides, which are thereby ionised, i.e. charged,
and vaporised. The charged peptides are now accelerated by a
powerful electric field and fly through a flight tube. Small peptides fly faster than large peptides. The time that they take to
reach the end of the tube therefore indicates the size of the pro-
38
UV laser
powerful
electric field
laser beam
+
++
+ +
+
amplifier
data analysis
in computer
matrix peptides
ionisation
acceleration
In MALDI-TOF MS (matrix-assisted laser desorption ionisation time-of-flight mass spectrometry), the protein sample to be investigated is digested with a specific protease
and the resulting peptide mixture is embedded in a matrix
in a mass spectrometer. The energy of a laser impulse
applied to the matrix is transferred to the peptides, which
are thereby ionised and vaporised. The charged peptides
time of flight
detector
measurement
analysis
39
3000
3500
4000
4500 5000
3000
3500
4000
4500 5000
1500
3000
3500
4000
4500 5000
3000
3500
4000
4500 5000
2000
2500
3000
3500
4000
40
5000
A peculiarity of mass spectra is the broadening of the signal that occurs as a result of the natural occurrence to an
extent of about 1% - of the heavy carbon isotope 13C. With
larger peptides there is a high probability that at least one
carbon atom in the molecule will have this greater mass.
The presence of this isotope broadens the signal, and the
resulting strong, broad signal can obscure a weaker signal.
Nevertheless, the resolution of the technique can be considerably increased by use of mathematical methods and
in particular by use of extremely sensitive spectrometers.
The accuracy of modern MALDI-TOF mass spectrometers
is rated at about 10 ppm (parts per million).
and subjected to mass spectrometry in virtual form in a computer. To remain with the analogy of forensic science, this is equivalent to a suspects fingerprint being worked out from his photograph a possibility that would presumably make every police
officer in the world green with envy!
4500
13C
Automated structural
analysis
41
42
43
Dozens of signalling pathways are involved in the development of the various forms of cancer. Each newly discovered
pathway provides further potential targets for medical intervention. Proteomics can help in the elucidation of such signalling pathways.
control
ICE3
ras-transformed
mouse lymphocytes
prevents the cell from undergoing transformation into a cancer cell. In order to be able to exert this control function, ICE3
is split in healthy cells by an enzyme known as granzyme B
to form apopain. The 2-D gel of the cancer cells shows a
large amount of unsplit, i.e. inactive, ICE3. From a separate
experiment with gene chips it is known that the genetic
instructions for the production of granzyme B are absent
from the cancer cells. From this combination of findings the
signalling pathway that operates here can be deduced:
Signalling pathway diagram
ICE3
HMG2
ras
BTF3a
The 2D gel on the left shows a subset of the proteome of a
normal cell; that on the right the same subset of the proteome of a cancer cell. The differences are readily apparent.
- A protein referred to as ICE3 is present in greater quantity in the cancer cell.
- The amount of the high mobility group protein HMG2
is reduced in the cancer cell.
- The transcription factor BTF3a is formed in greater
amounts in the cancer cell.
ICE3 plays an important role in programmed cell death, or
apoptosis. This suicide of body cells occurs when the
genetic material of a cell is severely damaged; in this way it
44
granzyme B
apopain
apoptose
uncontrolled
cell
growth
Proteomics is nevertheless confronted by problems. One of the most important of these is a direct consequence of the central characteristic of
the proteome, namely its complexity. The total number of proteins in the human body is now known to be many times greater
than the number of genes. There are estimated to be about
100,000 different proteins per cell type in our body; some of
these are present in almost all cells, whereas others occur only in
a small number of cells. The Roche database, for example, currently contains over 150,000 mass spectra, corresponding to
around 4500 proteins. From this it is clear that a great deal of
work remains to be done. It is also likely that many proteins, and
in particular many modifications, i.e. subsequent alterations to
proteins, will prove to be extremely rare. Unlike the situation
with the human genome, it will be difficult ever to claim to have
catalogued the human protein completely. In fact, this will in all
likelihood be impossible, since failure to find any new proteins
in a particular period most certainly does not mean that no proteins remain to be discovered (for example, our body is constantly producing new antibodies in response to antigens). This
applies even more to protein interactions with other proteins
and with other components of the body such as genes. Only
potent interactions of this kind are easy to find.
Another problem that confronts proteome research is already
well known to its big brother, genome research: the problem of
patenting. As with genome research, it is to be expected that
courts of law will eventually determine what can be protected,
and how. For as long as uncertainty prevails in this area, however, every publication constitutes a risk for a researching company. The fact that this problem is well known from genome
research will at least prove helpful in this regard.
References
Langen H: Proteomics as a new field in biology: applications and potentials. Presentation at Roche Roundtable on Genetics and Genomics, May 2000
Screening Trends in Drug Discovery: Future Trends in Proteomics. Interview with Hanno Langen in:
Screening, 2/2001
Brauckmann B: Protein analysis helps in the evaluation of new drugs. Roche Facets No. 14, 2000
Human Proteome Organisation Website: http://www.hupo.org/
45
They are the stars of medicine. Everywhere they are sought after, pursued and adored. Whenever they are found, they become
an object of feverish research. Scarcely any other area of biological research is being pursued with such financial backing and
intellectual effort as the search for targets for new drugs. This
is only to be expected, since it is in drug targets that pharmaceutical research has placed its greatest hopes for new, safer,
more efficient and more effective therapies.
In its broadest sense, a pharmaceutical target is any molecular
site within the body that is potentially susceptible to attack by
drugs. Most such targets are proteins, though other biomolecules such as DNA, RNA, sugars and fats also have potential as
targets for drug action. Common to all such molecules is their
key role in metabolic processes and thus their importance for
bodily function and malfunction. Expressed in another way,
targets often play a role in the development of disease. And that
is what makes them such interesting objects of research.
DNA, the substance that bears our genetic information, controls the vast majority of bodily processes and lays down the framework for the
bodys reactions to the environment. Many diseases are due to
genes, though in most cases external factors also play a role. Malfunction of a gene can be due to an alteration in the sequence of
Important cause
of disease: DNA
48
49
insulin
extracellular
intracellular
p91
p60PIK
*
*
PI 3'kinase
p85 p 110
??
ISSK
*
*
IRS-1
ISK
RAS
dynamin
Shc
SHPTP2
(SYP)
SOS
B-RAF
GRB 2
*14-3-3
RAF-1
Jun
Fos
p70s6k
??
MAPK
*
p90Rsk
p70s6k
GSK3
*
PP-1
phosphorylase
kinase
inactivation
glucose
transport
glycogen
synthase
activation
CRE's
PLO
*
*
protein
synthesis
PKC
PKC
40S
cGI-PDE
(type III)
translocation
PKa
MEK
NFr B translocation
transcription
factor
activation
GLUT4
translocation
??
glycogen
deposition
Molecules that play key roles in metabolism are potential drug targets.
Improve or invent
50
y
NSF
SNAP
RABs
ARFs
Ga, y
SNARE`s
7TM
z
z
in a disease process and in this way lead to the development of new, and possibly more effective or better
tolerated, drugs;
be closer to the molecular cause of a disease than
presently known targets and thereby provide a more
specific and surer method of attack;
play a role in production of the symptoms, rather than
the cause, of a disease and therefore also have potential for use as a therapeutic target in other diseases with
similar symptoms;
be used as targets in diseases that have hitherto been
regarded as essentially untreatable due to a lack of any
suitable target; these include various types of cancer,
since cancer is often due to a variety of causes that need
to be treated individually.
51
Successful search
at many levels
52
proteomics
databases
SNPs
gene chips
PCR
transg. mice
unknown
molecule
e.g. protein
proteomics
databases
structural
investigation
protein
structure
possibly
prediction of
structure
databases
SNPs
gene chips
PCR
transg. mice
genetic
background
associated
gene
transcript
regulation
proteomics
databases
gene chips
HTS
properties
physical
chemical
biological
proteomics
databases
SNPs
gene chips
transg. mice
proteomics
databases
SNPs
gene chips
transg. mice
HTS
function
e.g. enzyme,
hormone,
structural
protein
molecular
environment
associated
metabolic pathway
complexes with
other proteins
when, where,
how, how much
proteomics
databases
SNPs
gene chips
transg. mice
proteomics
databases
transg. mice
HTS
selection
drug
finding
attack which
target where?
chemical
synthesis
test physical,
chemical and
biological
properties
databases
transg. mice
drug
evaluation
preclinical and
clinical studies
marketing
authorisation
procedure
yield useful findings only when their results are considered together. Most techniques are used in more
or less modified forms at various levels.
53
Single nucleotide polymorphisms, or SNPs (pronounced snips) have become increasingly important in recent years. These randomly occurring variations in single DNA subunits are transmitted from
generation to generation and are considered to be responsible
for many medically important phenomena (see chapter on
SNPs) including intolerance, side effects and variations in the
effectiveness of drugs. They also play a role in the development
of many diseases. Thus, the consistent occurrence of specific
SNPs in patients with certain signs or symptoms suggests that
these SNPs interfere with the function of disease-related genes.
Those genes can then serve as potential targets for drugs. As
with proteomics and genomics, the systematic search for SNPs
is now giving rise to extensive databases.
54
In theory, the DNA chip technique is also suitable for the investigation of proteins, since in principle any molecule can be investigated on the silicon surface of chips. For this purpose, however, proteins like DNA have to be attached to, or ideally
synthesised on, the substrate, and this is vastly more difficult to
achieve with sensitive, highly complex molecules such as proteins than with a relatively stable and structurally simple molecule such as DNA.
Vanishingly small amounts of DNA can be rendered visible by means of the polymerase chain
reaction (PCR). With the aid of the DNA-extending enzyme polymerase, a single DNA molecule can be copied in
a chain reaction (amplified) as many times as desired and thus
made available for investigation. This technique has made it possible, among other things, to detect the presence of minute
amounts of the genetic material of HIV, the AIDS virus, and in
this way identify variants of this deadly infectious disease. If a
drug is to be used against one of the few, but highly variable,
products of viral RNA, PCR can be used to identify the specific
variant of this target molecule present in a patient and then
select the most appropriate drug. In future PCR will therefore
55
High-throughput
screening (HTS)
56
Animal models
57
Transgenic mice as a
model
58
Microinjection of DNA
DNA
embryonic stem cells
unfertilised ovum
DNAmicroinjection
donor blastocyst
fertilised ovum
reimplantation in mouse
prepared for pregnancy
reimplantation
mouse prepared
for pregnancy
chimeric mouse
breeding
transgenic offspring
59
60
search now that the existence of global markets and public funding has made the development of drugs for use against them economically viable for private companies.
The decision as to whether, when and how a potential target
should be further developed depends on a number of factors.
First among these is medical benefit: what benefit will a new
drug bring to victims of the disease it is intended to treat? In the
case of treatments for previously untreatable diseases (or forms
of disease) the answer is obvious, however even benefits such as
substantially improved tolerability, greater potency or avoidance
of side effects can justify the development of a new drug.
Another important factor influencing the decision for or against
a drug target is a scientific analysis of anticipated expenditure
versus risks. In other words, what is the probability that an agent
for use against the target can actually be developed, and what expenditure will be required in order to develop such an agent? Before these questions can be answered, certain details of the structure and function of the target must already be known after all,
it is pointless to direct efforts and resources to targets that are
difficult to influence if the therapeutic objective might be more
easily achievable by influencing other targets.
A third important factor influencing the evaluation of drug targets is economics. Every privately
owned pharmaceutical company has to finance
its research spending with income obtained from the sale of its
products without money there can be no research. Though
this is fundamentally true of any company that develops new
products, some particular considerations apply to the field of
medicine.
z Probably the most important of these see above is a duty
of care towards patients. The high degree of responsibility
borne by the healthcare industry in this regard inevitably influences economic decision-making. This gives rise to conflicts whenever a medically (or socially or politically) correct
decision is clearly economically incorrect in other words,
when it is clear from the outset that the cost of developing a
medically worthwhile drug can never be recovered via sales
of the drug. Some such conflicts can be resolved via orphan
disease programmes in which the state provides financial
and legal support for research into particularly rare diseases
and the development of drugs for use against them.
61
search is the composition of the electoral college that participates in the decision-making process. This includes not
just the company itself in the form of its management, research, finance and marketing directors, employees, investors
and financial analysts, but also patients in the form of patient
associations, legislators in the form of regulatory authorities,
and the public in the form of the press and a variety of associations.
z An additional peculiarity of the pharmaceutical industry is
the sheer unpredictability of biology. All scientific and technical advances notwithstanding, interventions in biological
systems are always subject to a high degree of uncertainty.
Evaluation of targets is therefore often difficult. Even the fact
that the total number of potential targets is unknown is important in that it has a substantial impact on the economic
significance of each and every target. A newly discovered biomolecule can turn out to be useless after prolonged research,
while years later it may suddenly acquire considerable value.
The same is true of new drugs: whatever precautions are taken, a new drug may prove to be unsuitable or even dangerous;
conversely, a completely new set of indications for a drug may
suddenly be found. Since the biological system that forms the
subject of medicine is the human body, such risks and uncertainties are of the greatest importance and the cost of researching and developing new drugs is correspondingly high.
z Undoubtedly one of the most-discussed aspects of the economics of pharmaceutical research is the question of patents.
While patent protection is essential for the survival of any
economic sector that undertakes research and development,
biological patents are subject to particular problems. Prominent among these is the fact that the distinction between a
(nonpatentable) discovery and a (patentable) invention is often difficult to make in biology. It is also argued that overgenerous patent protection can inhibit further research, and
thus medical progress. This view is propounded not just as
always by competitors, but also by a variety of community
groups, politicians and basic researchers representing to
some extent conflicting interests. After some heated initial
discussions and a certain amount of legal toing and froing,
the legal and practical arrangements that have now been
made, though still somewhat woolly, provide companies
with a sensible degree of security for forward planning.
62
research director
medical
profession
financial analysts
target selection
customer
advocates
market group
drug hunter
regulatory agency
Many different constituencies with partly overlapping and conflicting interests influence target selection in pharmaceutical research.
63
position in
drug class
indication
original sales
forecast
Tamoxifen (Novaldex)
1st
breast cancer
100,000
Captopril (Capoten)
1st
USD 20 million
Cimetidine (Tagamet)
1st
Fluoxetine (Prozac)
Atorvostatin (Lipitor)
peptic ulcer
700,000
nd
depression
th
hypercholesterolemia
2
5
that can improve our understanding of the disease process in patients. We must therefore hope that the pop stars of medicine
continue to make headlines.
References
Lindpaintner K: Pharmacogenetics and the future of medical practice: conceptual considerations. Pharmacogenomics 1: 2326, 2001
Knowles J, Gromo G: Target selection in drug discovery. Nature Rev, Vol 2, January 2003
Brauckmann B: From basic research with genetically modified mice to new forms of medical
therapy. Roche Facets No. 14, May 2000
Ebeling M: Mit eigener Bio-IT am Forschungspuls. Internal Roche publication
Human Genome Project Website: http://www.ornl.gov/hgmis/
Geschftsstelle des Wissenschaftlichen Koordinierungskomitees des Deutschen
Humangenomprojekts (ed.): Das Humangenomprojekt 1st and 2nd edition
64
66
denaturing
polymerase
Zielsequenz
extension
Polymerase-Kettenreaktion: PCR
DNS
getrennte
DNS-Strnge
(denaturiert)
hybridisation
1
2
repeat cycles
4
5
67
Copies of copies
of copies
68
Soon after its discovery the PCR method was refined in several ways. One of the first modifications of the original protocol concerned the polymerases used.
Like all enzymes, polymerases function best at the body temperature of the organism in which they originate 37C in the
case of polymerases isolated from humans. Below this temperature the enzymes activity declines steeply, above this temperature it is quickly destroyed. In PCR, however, the two strands of
the DNA molecule must be separated in order to permit the primers to anneal to them. This is done by raising the temperature
to around 95. At such temperatures the polymerases of the vast
majority of organisms are permanently destroyed. As a result,
new enzyme had to be added in the first reaction step of each cycle a time-consuming and expensive proposition.
A solution was found in hot
springs. Certain microorganisms thrive in such hot
pools under the most inhospitable conditions, at temperatures that can reach 100C
and in some cases in the presence of extreme salt or acid
concentrations. The polymerases of these organisms
are adapted to high temperatures and are therefore
ideal for use in PCR. Today
Hot spring: Thermus aquaticus. Nearly all PCR techthe polymerases used in
niques in the world now use the Taq polymerases isolatnearly all PCR methods the
ed from the bacterium Thermus aquaticus, a microorworld over are derived from
ganism that dwells in hot springs at about 70C. The first
such microorganisms. This
T. aquaticus strain from which polymerases for PCR were
prominent bacterium goes
obtained was found in Yellowstone National Park in the
by the name of Thermus USA.
aquaticus, and its heat-stable
polymerase, called Taq polymerase, supports an entire industry. The organism was originally discovered in a 70C spring near Great Fountain Geyser in Yellowstone National Park in the USA. Employees of Cetus, who
Kary Mullis was working for at the time of his discovery, isolated the first samples from the hot spring and then cultivated in
the laboratory one of the most useful bacterial strains known today. Meanwhile Thermus aquaticus has been found in similar
hot springs all over the world.
69
The introduction of Taq polymerase has certainly not been the only modification to the PCR
method. This was helped by the fact that Mullis
published his discovery relatively early though not without some difficulty. Both Science and Nature, the two most
renowned scientific journals, failed to recognise the significance
of PCR and rejected the paper describing the method. Moreover,
despite global patent protection, the use of the PCR technique is
still free and unrestricted for basic researchers thanks to Roche,
which owns the rights to the method. In 1991 Roche obtained
an exclusive license from Mulliss former employer Cetus for
300 million dollars. Scientists from all over the world have modified the PCR method in many ways and adapted it for routine
diagnostic testing and molecular research. At the same time,
more and more new applications are emerging.
Further developments
around the world
In the 1990s biology was faced with one overriding preoccupation: the unravelling of the genome.
Thanks to huge technical and organisational efforts, first viruses and bacteria, then yeasts, plants and animals
relinquished the secrets of their genetic material. This accomplishment would have been unthinkable without PCR, which
made it possible to prepare large amounts of DNA within a short
time. The simple cloning of DNA has therefore remained one of
the main uses of the method. Thus PCR is used whenever the exact sequence of DNA building blocks needs to be determined:
e.g. in other genome sequencing projects, in gene research, in the
investigation of genomic changes, in the search for targets, etc.
An important topic in the field of genomics today is SNPs (pronounced snips), single nucleotide changes in the genome which
appear to account for a large proportion of the genetic differences between individuals (see chapter on SNPs). Among
other things, SNPs are responsible for disease susceptibility and
for differences in the way patients respond to drugs. In order to
detect such hereditary and often widespread variations, scientists have to sequence the genome of many different people in
parallel. Genes with SNPs are also potential targets for new
drugs. PCR therefore plays a key role in this important area of
drug research.
Forerunner of genomics:
DNA sequencing
70
adenine
thymine
guanine
cytosine
Sensitive determination:
qualitative PCR
71
Quantitative PCR provides additional information beyond mere detection of DNA. It indicates
not just whether a specific DNA segment is present in a sample, but also how much of it is there. This information is required in a number of applications ranging from medical diagnostic testing through target searches to basic research.
Consequently, although quantitative PCR was not described until the 1990s, the method already exists in a number of variants
and protocols to meet a broad range of requirements. Theoretically it is possible to calculate the amount of DNA originally
present in a sample directly from the amount found at the end
of a PCR run. If, for example, there were not one but two double strands at the start of the reaction, exactly twice as much will
be present after each cycle. However, this simple approach
founders on the fact that conditions for the polymerases are not
optimal at the start or end of PCR. At the start the performance
of the enzymes is limited by the small amount of template present, while in the final cycles the enzymes activity declines as a
72
Target research
73
forward
primer
forward
primer
probe
probe
R
probe
forward
primer
74
Example: STEP
75
colon normal
level
colon cancer
The values obtained are plotted and compared. The example below is a comparative diagram showing the expression
profile of various tissues, including samples from the colon
of a healthy subject and of a patient with cancer of the colon
(circles). The gene in question was suspected of being more
active in the presence of colon cancer. However, the STEP
procedure showed that the colon-specific gene is equally
active in healthy and ill individuals.
ured and plotted. In many cases this reveals whether a target really is associated with a disease.
In addition to target evaluation, the STEP method like other
quantitative PCR techniques is also used in other areas of biological science, e.g. in the development of model systems (for
example new cell lines) and in basic research. The establishment
of whether, when and how a gene is expressed provides important information on the role of the corresponding gene product
in the bodys molecular network. This approach has shed light
on many biological processes, e.g. the bodys response to external factors such as administration of drugs.
76
choose the right treatment doctors often need to know the actual concentration of pathogens present. PCR is one of the few
techniques available today that is able to measure the pathogen
load. This is an important parameter, for example, in viral infections, which often follow a chronic course and produce no
clinical effects for some time, despite infection and in some
cases ongoing physical damage. In this case the viral load in
the bloodstream can provide an indication of how the disease
is progressing.
In addition, quantitative determinations serve to monitor the
success of treatment. If a drug works in a patient, his/her pathogen load will decline sharply. However, some viruses change so
rapidly that they cannot be completely eradicated by the drugs
used: they become resistant to them. This often occurs, for
example, in viral hepatitis C infection. The hepatitis C virus
(HCV) often causes chronic inflammation of the liver, leading
to liver cirrhosis or even cancer in a substantial proportion of
those affected. Damage to the liver often accumulates over
decades without being directly detectable. As mentioned earlier,
qualitative PCR has been used for some years now for diagnosing HCV infection. But the quantitative form of the technique
opens up whole new perspectives for the treatment of the disease. With the help of this method it is possible to monitor the
success of treatment as well as determine how rapidly the disease
is progressing, if at all. Unlike with conventional methods, it is
also possible to ascertain whether the disease has been eradicated or has become chronic.
Hepatitis C viruses.
77
Example: HIV
Another important application in which quantitative PCR is used in the field of infectious diseases is AIDS. Those infected with the causative agent of the disease, the human immunodeficiency virus (HIV), have to take a
cocktail of usually three drugs indefinitely, this being the only
way to keep the virus in check, if only for a while. HIV mutates
extremely rapidly, quickly becoming resistant to drugs. Viruses
that are resistant to all three drugs at the same time sometimes
even occur, requiring a new cocktail to be used.
In order that this moment, known as a viral breakthrough,
should not pass unnoticed rapid proliferation of the viruses
could cause the disease to flare up again the quantity of viral
particles in the blood must be measured periodically. A rise in
the viral load indicates that the drugs being used are losing their
efficacy. Quantitative PCR permits such monitoring and helps
doctors adjust the treatment optimally.
Genetic factors are always involved in the development of cancer. Their contribution varies greatly depending on the type of cancer. Genes not
only help to determine progression of the disease but can also
have a substantial influence on the effectiveness of the available
treatments. Identifying the genes that play a role in the development of cancer is therefore an important step towards improving treatment. Both qualitative and quantitative PCR play a crucial role in the fight against cancer. PCR can identify genes that
have been implicated in the development of cancer. Often the
genes exist in a number of variants with significantly different
effects. One example is the gene known as p53, whose product is
a central monitor of cellular division. If the function of this
monitor is disrupted in a cell, the cell can become cancerous relatively easily. Variants of p53 and similar genes can be detected
by qualitative PCR, giving doctors and patients an indication of
their personal risk of developing cancer or if the patient already has cancer how aggressively it can be expected to progress. Because multiple changes have usually accumulated before cancer actually develops, a reliable test must examine a large
number of gene variants. For this reason DNA chips are being
increasingly used to screen people for genetic changes (see chapter on DNA chips).
Applications
in cancer therapy
78
Meanwhile, quantitative PCR also is gaining importance in the fight against cancer. This is the
case where a cancer has a genetic basis but is not
due to an altered gene. In some cancers the genes that control
cellular division are intact but have been switched off. This can
occur through a process known as promoter methylation. In the
DNA region containing the start information for reading the
downstream gene (the promoter) the cell attaches small molecules (methyl groups) to specific building blocks of the DNA
(the cytosine bases). As a result, polymerases that normally read
the genes and produce working copies of them are no longer
able to dock to the start region. The gene therefore remains silent and no gene product is formed.
Only in recent years has it emerged that this mechanism of promoter methylation shuts down vital genes in many cancer cells.
Methylation status is therefore of crucial importance because
it provides information on the chance of a tumour becoming
malignant and giving rise to metastases. A simple and reliable
method used for detecting these crucial DNA changes is methylation-specific PCR (MSP). In this relatively new technique cellular DNA is first treated with sodium bisulphite, which converts normal cytosine to the RNA building block uracil but
leaves methylated cytosine intact. This results in different products depending on the methylation status of the DNA (see box).
Specific primers for those products are used in the subsequent
PCR procedure. In this way it can be determined if the original
DNA was methylated or not.
Example:
promoter methylation
A host of other
applications
79
unmethylated DNA
methylated DNA
CH3 CH3
ATC
CGCGTCG
CH3 CH3
ATC
ATC
primer 2
CH3 CH3
CH3
CGCGTCG
GCGCAGC
ATC
primer 1
ATC
primer 2
GCGCAGC
primer 2
no product
3.
PCR product
only with primer 1
80
CH3
CGCGTCG
primer 1
2.
UGUGTUG
ACACAAC
ATC
NaHSO3
1.
UGUGTUG
CH3
CGCGTCG
ACACAAC
primer 1
no product
PCR
PCR product
only with primer 2
cytosines, by contrast, remain unchanged (1). The subsequent PCR procedure then uses specific primers for the
various products formed (2). Hence, either the original
methylated or the unmethylated DNA is copied. The original DNA was therefore methylated or not (3) depending on
the primer used to obtain a product.
References
Mullis KB, Faloona FA: Specific synthesis of DNA in vitro via a polymerase-catalyzed chain
reaction. Methods Enzymol 155: 335350, 1987
Higuchi R, Dollinger G, Walsh PS, Griffith R: Simultaneous amplification and detection of
specific DNA sequences. Bio/Technology 10: 413417, 1992
Wilfingseder D, Stoiber H: Quantifizierung von PCR-Produktmengen durch real-time PCRVerfahren. Antibiotika Monitor, Heft 1/2/2002
Reidhaar-Olson JF, Hammer J: The impact of genomics tools on target discovery. Curr Drug
Discovery, April 2001
Brock TD: Life at high temperatures. Bacteriology 303: Procaryotic Microbiology, 1994
http://www.bact.wisc.edu/Bact303/b27
Mullis KB: The polymerase chain reaction. Nobel Lecture, December 8, 1993
http://www.nobel.se/chemistry/laureates/1993/mullis-lecture.html
Kary Mullis Website: http://www.karymullis.com
Deutsches Hepatitis C Forum e.V. Homepage Qualitativer HCV-RNA Nachweis:
http://hepatitis-c.de/pcr1.htm
81
G
G
G
G
G
G
C
C
C
C
C
C
A
A
A
A
A
A
A
A
A
A
A
A
T
T
T
T
T
T
C
C
A
C
C
A
T
T
T
T
T
T
A
A
A
A
A
A
SNPs: common,
hereditary, stable
84
DNA sequence
C
C
C
C
C
C
A
A
A
A
A
A
A
A
A
A
A
A
T
T
T
T
T
T
C
C
A
C
C
A
T
T
T
T
T
T
A
A
A
A
A
A
M
M
SNP C
SNP A
individuals with
different SNPs
In the following theoretical example, replacement of a cytidine (C) with a guanine (G) in the gene results in formation
of an amino acid with completely different properties:
instead of the large, basic amino acid glutamine (Gln), the
small, neutral amino acid glycine (Gly) is formed.
SNP
unchanged
Gene
AAG-CGA-ATT-AGG AAG-GGA-ATT-AGG
Protein Lys -Gln -Ile -Arg Lys -Gly -Ile -Arg
85
chromosome
gene
ex in
pr
i.e. the sequences of a gene that are translated into the gene
product the protein. SNPs that are present in exons can
have a major influence on the function of the protein concerned if they result in incorporation of an alternative amino
acid.
z pSNPs (phenotype-relevant SNPs): Both gSNPs and cSNPs
can influence a persons phenotype: the former primarily via
the amount, and the latter usually via the form, of the protein for which the gene codes. pSNPs are the most important
type of SNP from the point of view of medicine. They form
one of the foundations of pharmacogenetics, the branch of
science concerned with the influence of gene variation on the
effectiveness and tolerability of drugs.
SNP analysis is now performed on a number of mature technology platforms with very high accuracy and reproducibility.
86
Medical effect at
two levels
87
100
80
cure rate in %
gastric ulcer
peptic ulcer
60
40
20
0
mut/mut
wt/mut
cyp2c19 genotype
z Molecular target. SNPs in the gene that codes for the target
cyp2c19 in
stomach ulcers
88
wt/wt
Another example of pharmacogenetically important SNPs is the gene for the beta2-adrenergic receptor. This molecule performs a number of
functions in the body (see box). Activation of it in the lungs
relaxes the smooth muscle of the airways. Some anti-asthma
Beta2-adrenergic
receptors
27
NH2
164
cell membrane
COOH
89
90
Although SNP research on a large scale has therefore been possible only for a relatively short period of time, a large number of these variations in the human
91
92
93
logical, but very challenging, step in the use of SNPs. This approach has the distinct advantage that it is not limited to the
still small number of genes about which we know enough to
declare them candidate genes. At present, however, such tests
are still too costly and time-consuming for routine use, and
not a single example has been published. Also, because of a
number of unresolved issues, the feasibility of genome-wide
SNP association studies remains quite controversial.
From the perspective of patients, SNPs are a major step towards a more personalised approach to
treatment that takes genetic differences between
people into account. From the perspective of the pharmaceutical industry, SNPs have become important parameters to be
considered in drug research and development. SNPs have the
potential to lead to the discovery of new targets and thus eventually to the development of new and better drugs, and they bear
the promise of leading to more effective, safer and better tolerated forms of treatment. Our understanding of the importance
of these small but potentially crucial differences is growing all
the time. Finding those that matter for healthcare has therefore
become an important new aspect of pharmaceutical research
and development.
References
Foernzler D: SNPs kleine Unterschiede mit groer Wirkung. BioWorld, June 2000
Stoneking M: Single nucleotide polymorphisms: from the evolutionary past Nature
409: 821-822, 2001
Chakravarti A: Single nucleotide polymorphisms: to a future of genetic medicine.
Nature 409: 822-823, 2001
The SNP Consortium Website: http://www.ncbi.nlm.nih.gov/SNP
TSC data on the CYP2C19 gene:
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId=1557
Holden AL: The SNP Consortium: summary of a private consortium effort to develop
an applied map of the human genome. BioTechniques 32: S22-S26, 2002
Weiner MP, Hudson TJ: Introduction to SNPs: discovery of markers for disease.
BioTechniques 32: S4-S13, 2002
Abraham J, Wilson DE: Roche scientists exceed expectations of genetic discoveries
more than 18,000 mouse SNPs identified in 27 months. Roche press release, Palo
Alto and Pleasanton, Calif., 5th November 2002
94
96
DNA chip
fluorescently labelled
RNA (probe)
50m
these differences, almost all DNA chips exploit the same biological principle, that of hybridisation.
The four bases that are the building blocks of DNA always pair
with the same partner. Our genetic material therefore consists
of two strands of DNA arranged in the form of a twisted rope
ladder, or double helix. The two strands are complementary in
the sense that the sequence of one can be deduced from that of
the other. This joining together, or hybridisation, of two nucleic acid chains to form a double-stranded structure has been exploited by biologists for decades. For example, labelled short
segments of single-stranded DNA (oligonucleotides) can be used
to search for the presence in our genome of oligonucleotides
with the complementary base sequence.
DNA chips do basically the same thing. DNA fragments tethered
to the chip bind to complementary base sequences in the solution being studied. The difference is that DNA chips make it
possible to perform many such experiments at once: millions of
copies of each of several hundred thousand different oligonucleotides can now be accommodated on a chip measuring just
one square centimeter. Conversely, such a chip can be used to
search for tens of thousands of different DNA segments in a solution and it is precisely this possibility that forms the basis of
entirely new applications in biological research and medicine.
97
fluorescent
labelling
probe
5'
3'
T
target DNA
T
5'
G
3'
hybridisation
G
C
Field of application:
gene research
98
A
A
G
G
AG
A T CG
AGAC
CGTC
T AG T
One of the most important DNA chip technologies was developed by the Californian company
Affymetrix. The name of this companys best-known product,
GeneChip, is often used synonymously with the term DNA chip
to refer to any such product. (In addition to DNA chip and
gene chip, other terms including microarray, genome chip
and gene array are in common use.) Affymetrix manufactures
its GeneChips using the principle of photolithography, just as in
the manufacture of computer chips. In this technique a light
source, special masks and photosensitive protector molecules
are used to deposit billions of oligonucleotides with (at present)
up to 700,000 different base sequences alongside each other in
tiny cells (spots) on a chip (see box on page 100).
99
total RNA
biotin-labelled
cRNA
cDNA
reverse
transcription
AAAA
in vitro
transcription
B
B
AAAA
B
AAAA
B
fragmentation
GeneChip
expression
array
B
B
fragmented,
biotin-labelled
cRNA
hybridisation
B
wash and
stain
scan and
quantitate
B
B
100
Competing techniques:
design and function of DNA chips
The tasks performed using DNA chips are many and varied, and the design of such chips is correspondingly
diverse. Affymetrixs GeneChip is a commonly used DNA
chip, however various other other manufacturers are
offering a variety of techniques aimed at winning over
customers. Points of difference include not just chip design,
but also the way in which the experiments are performed
and the way in which the results are analysed.
z Probe material The most commonly used DNA chips
use short oligonucleotide chains as probes, however
RNA, cDNA, genes and even whole chromosomes can
be attached to chips.
z Manufacture DNA can be attached to chips in various
ways. These include photolithography, a technique borrowed from the computer chip industry. Other techniques include application by pipette, dropping and
electronic methods, e.g. in a manner similar to the
operation of an inkjet printer.
z Target molecules The probe and target molecules are
dependent on each other. Depending on what type of
target molecule is present on it, a chip may be suitable
for the study of other types of DNA, such as oligonucleotides, RNA, cDNA, genes, chromosomes or whole
genomes.
Reaction Hybridisation is not the only reaction that can
occur on a DNA chip. DNA molecules can also be
bound by ligases or via chemical or photochemical reactions. Another possibility is to combine PCR (polymerase chain reaction, see chapter on PCR) with a chip.
Detection Different reactions on the chip require
different methods of detection, and hybridisation can be
detected in various ways. In addition to fluorescence,
mass spectrometry (MS, see chapter on proteomics),
in particular, and also conductivity and electronic
methods, can be used for detection.
Analysis DNA chip experiments generate enormous
quantities of data that would be impossible to evaluate
without computer assistance. Of importance in this regard are not just suitably sophisticated programs, but
also automatic control of experiments, image analysis,
databases, Internet links and platforms and visualisation
of results.
Variety on a chip
101
Growing number
of applications
Now routine:
basic research in biology
102
A time of upheaval:
DNA chips in medicine
103
DNA chips also have potential for use in consumer protection. An example of this is in the
field of green gene technology, i.e. the use of gene
technology in agriculture. The fact that in most industrialised
countries a proportion of the population takes a sceptical view
of this technology has led to the introduction of a variety of regulations including compulsory labelling. Given, however, that
the vast majority of foods produced in this way do not differ visibly from conventional products, it is often only by means of an
examination of the genome of the plant concerned that adherence to such regulations can be effectively checked. DNA chips
are well suited for use in such tests.
Checking up on green
gene technology
Chip instead of a
magnifying glass:
ecology and taxonomy
104
DNA chips can make such distinctions more rapidly, more simply and above all more reliably. This ability creates applications
for DNA chips in all situations in which closely related species
need to be studied. These include ecology, taxonomy, anthropology and research into evolution.
DNA chips could also have a bright future in forensic medicine. Prominent in this field of application is the ability of DNA chips to detect differences between individual genomes and thereby to identify
people. This can be required for identification of victims and already plays a crucial role in the search for and conviction of criminals. Many countries, e.g. the Netherlands, also allow their
police to use the genetic profile of people they are seeking in order to draw conclusions as to the external appearance of the person concerned. So far this applies mostly to determination of
sex, however the discovery of more genes could make it possible
also to determine a persons hair colour, eye colour and ethnic
origin. Though at present such forensic tests are performed almost exclusively using PCR methods, DNA chips can play a useful complementary role in many such tasks and in future may be
able to perform such tasks more rapidly and simply than PCR,
thereby supplanting it in this application.
Sure identification:
forensic medicine
Examples now exist of all these fields of application of DNA chips. In most cases commercial
products are already available, though in some
applications DNA chips are still in the developmental phase. As
mentioned above, most interest is focused on basic research in
biology and on medical research and drug development. In the
latter field, use of DNA chips could initiate a change of direction
towards a more personalised medicine that exploits the small
but significant genetic differences that exist between people in
order to develop new, more effective and safer drugs, especially
for specific subpopulations. DNA chips that provide high resolution at a low price form the basis for the kind of rapid and simple genetic test that is essential for personalised medicine. They
are thus important not just for finding genes responsible for
diseases, but also for the development of new drugs, for correct
diagnosis and for the choice of the most appropriate treatment
for the individual patient.
Focus on medical
applications
105
Another medical application of DNA chips is their use in infectious diseases. In this application attention is focused not just on
the genes of the patient, but more specifically on those of the pathogens. Many viruses (e.g. HIV), in particular, along with various other kinds of pathogen, develop resistance to important
drugs extraordinarily rapidly via mutations in their genome.
DNA chips can be used to examine the genome of such pathogens rapidly and reliably so that treatment that is optimal for the
individual patient can be chosen.
106
lar medicine). DNA chips could here again, along with PCR
become the most important means of distinguishing between
these variants. A more recent development is a DNA chip to detect human papillomavirus (HPV). The two dozen variants of
this virus cause genital warts, which are generally similar to the
more familiar common wart and are similarly harmless. Nevertheless, three of the twenty or more variants of HPV can cause
cervical cancer in women. Precise identification of the particular variant present in affected women is therefore of great importance. Up to now the condition of the neck of the womb has
been assessed by means of a diagnostic smear so that any altered
tissue can be removed. In extreme cases the entire uterus needs
to be removed. A DNA chip now makes it possible to identify papillomavirus present in the patients blood and thus to estimate
the risk of cancer more precisely. Women at high risk can thus
adjust their family planning to their increased risk.
Important application:
cancer diagnosis
107
genetic damage
other signals
mdm2
+
activation and
synthesis
autoregulatory
loop
-
DNA repair
direct
involvement
p53
indirect
involvement
other
genes
other
functions
+
bax etc.
apoptosis
p21
The tumour suppressor gene p53 is one of the most important genes involved in the development of cancer. Its gene
product (P53, written with a capital P) plays a central role
in the growth and division of somatic cells. It is active especially when genetic damage is present in the cell, but it can
also be activated by external signals. It is known to have at
least three functions:
1. Control of the cell cycle. Cells divide in a regular pattern
of events known as the cell cycle. If the genetic material
of a cell is damaged, P53 holds the cycle in the G1 phase
a sort of resting phase in order to permit repair of
the DNA. The signal for this is transmitted via a number
of proteins including P21.
2. Apoptosis. If the damage to a cells DNA is too great,
P53 induces the cell to commit suicide. This process,
known as apoptosis, stops occurring in cancer cells,
with the result that they replicate in an uncontrolled
fashion. P53 appears to be able to initiate apoptosis in
various ways, including induction of the bax gene.
108
The decision as to whether such a drug will be useful therefore depends on the genetic profile of the patient concerned. Techniques that can identify each of the variants of
p53 and of its genetic environment are therefore an important precondition for specific, rapidly-acting and effective
therapy. The p53 GeneChip offers this possibility. On this
chip are several thousand short DNA segments with which
the p53 gene variants can be identified. The illustration
shows the fluorescence pattern that results from such a
test.
109
Use in preventive
medicine
First pharmacogenomic
product:
the AmpliChip CYP450
110
As in the case of their use in relation to cytochrome P450, DNA chips will in future find uses
in many areas of diagnosis namely wherever
genes play a role in the genesis or development of a disease.
Four such areas can be identified:
z the metabolism i.e. the absorption, conversion and breakdown of drugs; the genes of the cytochrome P450 family of
enzymes fall into this category;
z the action of different genotypes in cancer, i.e. the genetic
changes that play a role in cancer; the variants of p53 fall into
this category;
z a group of genes that play a role in the reaction of the body
to infection;
z genes that influence individual susceptibility to pathogens;
the extremely rare cases in which a gene variant can prevent
HIV infection are a well known example of this.
In all these areas DNA chips can permit precise diagnosis of the
genetic basis of a disease. With increasing knowledge of the
111
There are at least four areas in which DNA chips have a potential role in
medical diagnosis: drug metabolism, genotypes of cancer, infection-related genes and susceptibility to pathogens. The most important genes
currently known to play a role in these areas are indicated.
112
References
Erlich H: Diagnostic applications of genomics. Talk given at Roche R&D Media Day,
Munich/Penzberg, April 2002
Pedrocchi M: Multiprobe array systems for the analysis of human genes
Certa U et al.: Biosensors in biomedical research: development and applications of gene
chips. Chimia 53: 5761, 1999
Sinclair B: Everything's great when it sits on a chip. The Scientist 13[11]:18, May 24, 1999
Baron D: Genomics und Proteomics mit Gen-Chips und Protein-Arrays. Pharmazeutische
Zeitung 31/2001
DNA Microarray Website of Leming, Shi: http://www.gene-chips.com/
Affymetrix Website: http://www.affymetrix.com
InformationsSekretariat Biotechnologie Website: Genexpressionsanalyse
http://www.i-s-b.org/wissen/broschuere/chip.htm
113
Basic conditions:
ethics, law and society
The first, and most important, of the various interest groups involved is that of patients. Without
their acceptance, no change is possible. Their expectations of medicine are based on the elementary need for
preservation or restoration of individual health a need that
medicine has traditionally sought to satisfy with conventional
remedies and is now attempting to satisfy with the fruits of molecular research. Newly acquired understanding of the molecular basis of diseases is being used to combat diseases earlier
and more effectively and in some cases to prevent them from
occurring in the first place. Molecular biology thus has the
potential to be of great practical benefit to patients.
116
The aim must therefore be to exploit the opportunities, while identifying and limiting the risks,
that arise from newly acquired medical knowledge. At the core of more personalised medicine, for example, is
the need to know and analyse the individual characteristics of
patients. Only on the basis of this knowledge is it possible, for
example, to develop medicines whose use is based on genetic
criteria and which therefore are markedly more effective and
better tolerated in certain patients than are currently used medicines (see chapter on pharmacogenomics). At the same time,
however, this knowledge inevitably reveals to the medical staff
involved, to the dispensing pharmacy, to the patients healthcare
insurance fund, etc. an intimate detail of the genome of the
patient concerned, namely the genetic variation upon which
prescription of the drug in question is based. There is nothing
fundamentally new about this: for example, the fact that a person gets regular prescriptions for insulin identifies that person
unmistakeably as a diabetic. Many other forms of treatment
likewise reveal peoples diseases.
Exploiting opportunities,
limiting risks
117
The mere fact that information on the diseases present in an individual has to be acquired and to some extent passed on to third
parties is therefore no great problem in itself; much more prob-
lematic in this regard is the nature and quality of the information that is passed on, and in particular the possible predictive
value of the results of genetic tests. In practice, however, any predictions of this kind will in the vast majority of cases be limited
to a statement as to whether a particular patient is likely to respond to a particular drug or group of drugs. More detailed or
specific information is generally not to be expected from such
tests.
The situation is different, however, in the case of tests that can
determine a persons likelihood of developing certain rare
hereditary diseases long before the actual onset of illness. This
predictive ability is most certainly highly desirable in the case of
diseases for which drugs will in future be developed that can not
only cure the disease, but also prevent, or at least delay the onset of, the disease in individuals identified as being at risk.
Preventive medicine should, must and will become increasingly
important in the future. Nevertheless, in some cases, the implications of genetic testing for the patient as well as for family
members may be profound as for example in some rare familial
diseases, such as Huntingtons disease, cystic fibrosis or hemophilia.
In these cases, individuals and their families may perceive the resultant information as stigmatising.
118
Therefore, if the individual patient is to be convinced of the value of more personalised medicine
with all its implications, a series of conditions must be satisfied:
z New, better targeted drugs should represent a definite therapeutic advance, especially in the case of diseases whose treatment has hitherto been unsatisfactory or nonexistent. Personalised medicines should be more effective and better
tolerated than presently available drugs. These two requirements apply mostly to research.
z New treatments must be marketed at a realistic price, i.e. a
price that is reasonable and appropriate to the medical value
of the drug concerned, and must be available to patients via
their regular healthcare services.
z People must be protected against any form of discrimination
arising from genetic or molecular tests especially as the
terms health and illness have no absolute meaning (see
box). The legal basis for such protection must be established
by politicians in representation of society as a whole.
Varied requirements
119
The requirements of individual patients for customised medicine differ from the interests of society as a whole. In many cases the interests of
these two sides appear, at least at first glance, to be in direct conflict, e.g. in relation to the sharing of burdens and risks between
individual patients and society. Like the question of confidentiality of patient data, however, this is not a fundamentally new
issue, and in most industrialised countries the problem is regulated by means of a greater or lesser degree of redistribution of
medical costs from the individual to the public purse. In terms
of economics, therefore, molecular medicine is simply a therapeutic innovation whose costs and benefits have to be weighed
against each other and whose impact on the balance between
public and private funding of medicine needs to be considered.
More problematic, however, is the question of how acquired data
should be handled, i.e. how patients can be protected against
misuse of their personal genetic information, in particular. Current data protection legislation is mostly aimed at minimising
the amount of data that are collected, stored, distributed and
Society as a whole:
the public interest
120
In the crossfire of conflicting interests in healthcare, criticism is often levelled at industry and
commerce. This is due above all to the fact that
companies operating in the healthcare sector seek to make a
profit out of health, a treasured asset both of individuals and of
society. However, this sweeping criticism overlooks the central
role played by commercial enterprises in satisfying the requirements of the various interest groups. After all, the pharmaceutical and diagnostics industry strives to meet the need of patients
for better targeted and safer treatments, and at the same time
makes an important contribution to human society and culture
via the research that it undertakes. And because it is highly innovative, it has to accept a high degree of responsibility.
Nevertheless, the changes that are about to occur in medicine
certainly raise questions for the affected industrial sectors and
do so precisely because of the special responsibility that these
sectors bear towards patients. Because whether and at what price
a medicine can eventually be offered for sale depends on a num-
121
ber of very complex factors. In the first place, the process of developing, testing and obtaining marketing authorisation for new
drugs is set to become increasingly protracted and expensive.
The research that underpins the development of new drugs is
also expensive. Even though the drug regulatory authorities of
many countries are presently working on ways of simplifying
drug registration procedures, at least for particularly important
and innovative new drugs, it is becoming increasingly difficult
for companies to recoup the development costs of new drugs.
Pharmaceutical companies are also subject to multiple constraints in relation to the pricing of new medicines firstly as a
result of their need to recoup the development costs of the drugs
concerned, and secondly as a result of the price controls that are
imposed in many countries. Therefore, if the development of
more personalised forms of therapy for necessarily smaller target
groups is to be made economically viable, it may be necessary to
adopt new patterns of thinking and devise new forms of cooperation between industry and healthcare funding authorities.
Research:
laying foundations and
informing the public
122
tific knowledge is axiomatic to scientists, but by no means obvious to patients. Here again, frank and objective education of the
public is required.
Genetic testing and molecular diagnostics therefore can and must be accompanied by better education of the public. Education alone, however, will not convince the public of the value and purpose of the new possibilities
in medicine. Of more use in this regard are instruments that
allow to define their potential risks and benefits.
Such instruments have in fact already been available for a long
time. This too illustrates the point that molecular and pharmacogenetic tests do not differ fundamentally from the standard
techniques used in medicine today: they too can be assessed in
terms of parameters such as sensitivity and specificity that provide an objective measure of the informational content of test
results.
For example, a genetic test whose purpose is to spare patients
from a life-threatening side effect of an important drug must
satisfy the following criteria:
1. Specificity: The test must reliably indicate that a person
found to have the observed gene variant(s) will develop the
side effect in question. The less specific the test (i.e. the high-
123
124
References
Schreiber H-P: Humangenomforschung, Gentechnik und Gesellschaft
Schreiber H-P: Human-Genom-Forschung und die Notwendigkeit eines sozialvertrglichen Umgangs mit genetischem Wissen
Lindpaintner K: The importance of being modest, or: How good is good enough?
Reflections on the pharmacogenetics of abacavir. Pharmacogenomics 3: 835838,
2002
Genetics in Discovery and Development. Roches Ethical Principles including Roche
Charter on Genetics. Roche, 2000
Roche Biomarker Program: Economic impact of genetics, 2004. Internal document
125
Prospects:
more knowledge for medical science
No two illnesses
are the same
128
Consequence:
a new role for diagnostics
129
target
sequence
DNA strand
double helix
coiled
DNA
chromosome
No two treatments
are the same
130
two treatments are the same. A drug might be right for one patient but wrong for another, even though both patients have the
same illness, because drugs can vary in their efficacy and tolerability in different individuals. The field of pharmacogenetics
has been investigating the reasons underlying this phenomenon
for over a hundred years now, but only recently have molecular
genetic techniques made it possible to apply these insights to
clinical medicine. Pharmacogenetics is now threatening to upset the second half of the dogma of one disease one treatment.
In future the choice of the right treatment will depend not just
on the disease diagnosed, but also on the way in which each patients body deals with the drugs in question. To make this kind
of choice possible, two closely related factors need to be taken
into account:
z Genetic factors: Pharmacogenetics is concerned with the relationship between the gene variations and the bodys
response to drugs. Genetic differences can cause drugs to be
absorbed, metabolised or excreted too rapidly or too slowly.
Or they can prevent sufficient drug from reaching the target
site. Or they can give rise to adverse or even dangerous side
effects. Ruling out such genetically caused uncertainties
relating to the efficacy and safety of drugs will be one of the
major challenges facing pharmaceutical researchers in the
coming decades.
z Environmental factors: External factors are at least as important as genetic factors in determining the efficacy and safety
of drugs. Prominent among these factors is diet. Elements of
our diet can interact with drugs, accelerating or preventing
their uptake and affecting their excretion and utilisation. The
same applies to interactions between different drugs, which
can enhance or reduce each others effects and exacerbate
each others side effects. External stress factors such as physical and mental fitness, environmental toxins, radiation,
temperature and so forth can also influence the efficacy and
safety of drugs. In practice, the environmental influences to
which a patient is exposed cannot be exhaustively determined; also, they vary over time though this means that they
can be influenced. This is not true of gene variants. It is therefore all the more imperative to recognise how environmental
factors influence the way the body interacts with drugs.
131
If future therapies are to be based on genetic factors, medicine will inevitably become more personalised. However, the term personal in this
context does not mean that at some time in the
future patients will have their own tailor-made therapy. Rather,
it means that a far broader range of therapeutic options will be
offered from which doctors can select the one most suited to
their individual patients. Of course, such choices are already
available, at least for some diseases, however the number of such
choices will increase and so too hopefully will the success of
therapy. As an inevitable consequence of this development, the
target groups for drugs will become smaller. The indications for
new drugs will be determined not only by the molecular causes
of the diseases being treated, but also by the pharmacogenetic
profile of the individual patients. This is unexplored territory in
pharmacology.
In future, therefore, patients will be able to expect that a drug
that is prescribed for them is more likely to be truly suited to
them than at present. The effects of almost all currently used
drugs can vary to a greater or lesser extent, and in extreme cases
lack of efficacy is even the rule (see box). The safety of many currently used drugs is similarly unsatisfactory; for example, some
three patients per thousand die each year from severe side effects
of major drugs. This figure needs to be reduced, since even the
occasional occurrence of severe side effects can be acceptable
only if the disease concerned is relatively rare and unresearched
and therapeutic options and experience are correspondingly
limited.
Consequence:
the personalisation of
medicine
drug group
poor efficacy
AT2 antagonists
10 25%
SSRIs
10 25%
ACE inhibitors
10 30%
Beta-blockers
15 25%
132
Statins
30 70%
Beta 2 antagonists
40 70%
The efficacy of drugs is often unsatisfactory and may fluctuate in the extreme. The table gives the incidence of poor
efficacy for several major drug groups.
Angiotensin II (AT2) antagonists, like ACE (angiotensinconverting enzyme) inhibitors, are antihypertensive drugs
which are widely used in the treatment of heart failure. The
same applies to beta-blockers. Selective serotonin reuptake inhibitors (SSRIs) are used as psychotropic drugs,
especially in obsessive-compulsive disorders and, like tricyclic antidepressants, for depression. Statins, also known
as HMG-CoA reductase inhibitors, lower cholesterol levels.
Beta 2 antagonists are important antiasthma drugs.
Consequence:
integrated healthcare
133
For the pharmaceutical industry, these developments impose the need for a continuous rethink.
A new order will prevail in the healthcare market,
where the changes are already in full swing. New
strategies, alliances and competitions are emerging:
1. Integration of diagnosis and treatment: The more finely differences between individuals are distinguished and considered, the more difficult it is to separate these two poles. Close
cooperation is required here: drugs whose prescription depends upon pharmacogenomic considerations will be prescribable only if a corresponding means of testing is available.
Thus, a specific genetic variation first has to be identified in
the patient so that a drug geared to this variation can be sensibly used. And because the development of diagnostic tests
and therapy are to some extent interdependent, companies
with expertise in both these areas will find themselves at an advantage. Expertise therefore needs to be gathered together
either within a single company or else by means of close alliances between companies. The traditional boundaries between
diagnostics and therapy will therefore largely disappear.
2. Greater development risk: The fact that the available options
for treating most of the major common diseases are still unsatisfactory means one thing above all: Pharmaceutical companies will have to be more willing to take the risks associated with the development of new drugs with new mechanisms
of action. Certainly, the future will continue to hold the occasional surprise, as when a well-established drug is found to
possess previously unsuspected beneficial properties. But for
the most part, medical progress will depend on the exploration of new avenues particularly via new target molecules, which are already the most hotly contested objects in
medical research. Above all, new diagnoses, new targets and
new drug groups mean considerably stepped-up research
and development efforts with an undiminished risk of failure. Nevertheless, the effort may well be worthwhile. Successful developments that address unmet medical needs have
a huge sales potential.
3. Smaller target groups: The advent of more personalised
medical care inevitably means that a new drug can only be
sensibly used in a limited number of patients. This limits sales
possibilities, thus making it more difficult to recover the costs
of research and development. However, the development of
such drugs also has advantages. For example, drugs developed
Consequence:
upheaval in the pharmaceutical industry
134
135
Nevertheless, progress is opening up far-reaching new opportunities for medicine at the scientific and technical levels. Personalised diagnosis and treatment promise to be substantially more
effective with substantially fewer side effects. At the same time
they can tackle the causes of diseases whose treatment has until
now been only symptomatic and often inadequate. Notwithstanding all the commercial and ethical imponderables, in a certain sense the new possibilities also impose a moral obligation
to apply the new findings of molecular medicine for the practical benefit of patients.
136
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
137
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
fi
138
fi
fi
fi
Cover picture
Conceptual computer artwork. Part of a DNA molecule, chromosomes, a DNA
autoradiogram and the triplets of nucleotide bases that code for amino acids
in a protein.
Source: Mehau Kulyk, Science Photo Library
Published by
F. Hoffmann-La Roche Ltd
Corporate Communications
CH-4070 Basel, Switzerland
2007
Third edition
All trademarks mentioned enjoy legal protection.
Any part of this work may be reproduced, but the source should be cited in full.
This brochure is published in German (original language) and English.
Printers:
7000632-2
140