Gene expression can be controlled at any of several stages, which we divide

broadly into transcription, processing, and translation:

Transcription often is controlled at the stage of initiation. Transcription is not
usually controlled at elongation, but may be controlled at termination to
determine whether RNA polymerase is allowed to proceed past a terminator to the
gene(s) beyond.
In bacteria, an mRNA is typically available for translation while it is being
synthesized; this is called coupled transcription/translation. (In eukaryotic cells,
transcription takes place in the nucleus while translation takes place in the
cytoplasm).
Translation in bacteria may be directly regulated, but more commonly it is
passively modulated. The coding portion or open reading frame of a gene can be
assembled either with common or rare codons, which correspond to common or
rare tRNAs. mRNAs containing a number of rare codons take longer to translate.
The basic concept for the way transcription is controlled in bacteria is called the
operon model and was proposed by Francis Jacob and Jacques Monod in 1961.
They distinguished between two types of sequences in DNA: sequences that code
for trans- acting products (usually proteins) and cis-acting DNA sequences. Gene
activity is regulated by the specific interactions of the trans-acting products with
the cis-acting sequences. In more formal terms:
A gene is a sequence of DNA that codes for a diffusible product, either RNA or a
protein. The crucial feature is that the product diffuses away from its site of
synthesis to act elsewhere. Any gene product that is free to diffuse to find its target
is described as trans-acting.
The description cis-acting applies to any sequence of DNA that functions
exclusively as a DNA sequence, affecting only the DNA to which it is physically
linked.

To help distinguish between the components of regulatory circuits and the genes
that they regulate, we sometimes use the terms structural gene and regulator gene.
A structural gene is simply any gene that codes for a protein (or RNA) product.
Protein structural genes represent an enormous variety of structures and functions,
including structural proteins, enzymes with catalytic activities, and regulatory
proteins. A type of structural gene is a regulator gene, which simply describes a
gene that codes for a protein or RNA involved in regulating the expression of other
genes.

The sites of transcription regulation on DNA are usually (but not exclusively)
located just upstream of the target gene. The sequences that mark the beginning
and end of the transcription unit-the promoter and terminator-are examples of cisacting sites. A promoter serves to initiate transcription only of the gene(s)
physically connected to it on the same stretch of DNA. In the same way, a
terminator can terminate transcription only by an RNA polymerase that has
traversed the preceding gene(s). In their simplest forms, promoters and terminators
are cis-acting elements that are recognized by the same trans-acting species; that
is, by RNA polymerase (although other factors also participate at each site).
Additional cis-acting regulatory sites are often combined with the promoter.
A bacterial promoter may have one or more such sites located close by; that is, in
the immediate vicinity of the start point. A eukaryotic promoter is likely to have a
greater number of sites that are spread out over a longer distance.
A classic mode of transcription control in bacteria is negative control: A
repressor protein prevents a gene from being expressed. In the absence of the
negative regulator, the gene is expressed. Close to the promoter is another cisacting site called the operator; which is the binding site for the repressor protein.
When the repressor binds to the operator, RNA polymerase is prevented from
initiating transcription, and the expression is therefore turned off. An alternative
mode of control is positive control. This is used in bacteria (probably) with about
equal frequency to negative control, and it is the most common mode of control in
eukaryotes. A transcription factor is required to assist RNA polymerase in
initiation at the promoter. In the absence of the positive regulator, the gene is
inactive: RNA polymerase cannot by itself initiate transcription at the promoter.
In addition to negative and positive control, a gene that encodes an enzyme
may be regulated by the concentration of its substrate or product (or a chemical
derivative of either). Bacteria need to respond swiftly to changes in their
environment. Fluctuations in the supply of nutrients (such as the sugars glucose or
lactose) can occur at any time, and survival depends on the ability to switch from
metabolizing one substrate to another. Yet economy is important, too: A bacterium
that indulges in energetically expensive ways to meet the demands of the
environment is likely to be at a disadvantage. Thus, a bacterium avoids
synthesizing the enzymes of a pathway in the absence of the substrate, but is ready
to produce the enzymes if the substrate should appear. The synthesis of enzymes in
response to the appearance of a specific substrate is called induction and the gene
is an inducible gene.

The opposite of induction is repression, where the repressible gene is

controlled by the amount of the product made by the enzyme. For example, E. coli
synthesizes the amino acid tryptophan through the actions of an enzyme complex
containing tryptophan synthetase and four other enzymes. If, however,
tryptophan is provided in the medium on which the bacteria are growing, the
production of the enzyme is immediately halted. This allows the bacterium to
avoid devoting its resources to unnecessary synthetic activities.
Induction and repression represent similar phenomena. In one case the
bacterium adjusts its ability to use a given substrate (such as lactose) for growth; in
the other it adjusts its ability to synthesize a particular metabolic intermediate
(such as an essential amino acid). The trigger for either type of adjustment is a
small molecule that is the substrate (or related to the substrate) for the enzyme, or
the product of the enzyme activity, respectively.
Small molecules that cause the production of enzymes that are able to
metabolize them (or their analogues) are called inducers. Those that prevent the
production of enzymes that are able to synthesize them are called corepressors.
These two ways of looking at regulation negative versus positive control and
inducible versus repressible control - are typically combined to give four
different patterns of gene regulation: negative inducible, negative repressible,
positive inducible and positive repressible. This enables a bacterium to perform
the ultimate in inventory control of its metabolism to allow survival in rapidly
changing environments.
The unifying theme is that regulatory proteins are trans-acting factors that
recognize cis-acting elements (usually) upstream of the gene. The consequences
of this recognition are either to activate or to repress the gene, depending on the
individual type of regulatory protein. A typical feature is that the protein functions
by recognizing a very short sequence in DNA, usually < I0 bp in length, although
the protein actually binds over a somewhat greater distance of DNA. The bacterial
promoter is an example: RNA polymerase covers >70 bp of DNA at initiation, but
the crucial sequences that it recognizes are the hexamers centered at -35 and -10.
A significant difference in gene organization between prokaryotes and
eukaryotes is that structural genes in bacteria are organized in operons that are
coordinately controlled by means of interactions at a single regulator. In contrast,
genes in eukaryotes are usually controlled individually. As a result, an entire
related set of bacterial genes is either transcribed or not transcribed.

Structural Gene Clusters Are Coordinately Controlled

Bacterial genes are often organized into operons that include genes coding
for proteins whose functions are related. The genes coding for the enzymes of a
metabolic pathway are commonly organized into such a cluster. In addition to the
enzymes actually involved in the pathway, other related activities may be included
in the unit of coordinated control, such as the protein responsible for transporting
the small molecule substrate into the cell.
The cluster of the lac operon containing the three lac structural genes, lacZ,
lacY, and lacA, is typical. The key feature is that the structural gene cluster is

transcribed into a single polycistronic mRNA from a promoter where initiation of

transcription is regulated.

The lac Operon Is Negative Inducible

Transcription of the lacZYA genes is controlled by a regulator protein encoded by the lacI gene.
Although adjacent to the structural genes, lacI comprises an independent transcription unit with
its own promoter and terminator. In principle, lacI need not be located near the structural genes
because it specifies a diffusible product. The lacI gene can function equally well if moved
elsewhere, or can be carried on a separate DNA molecule (the classic test for a trans-acting
regulator).

The lacZYA genes are negatively regulated: They are transcribed unless turned off by the
regulator protein. Note that repression is not an absolute phenomenon; turning off a gene is not
like turning off a light bulb. Repression can often be a reduction in transcription by five-fold or

100-fold. A mutation that inactivates the regulator causes the structural genes to be continually
expressed, a condition called constitutive expression. The product of lacI is called the lac
repressor, because its function is to prevent the expression of the lacZYA structural genes.
The repressor is a tetramer of identical subunits of 38 kD each. A wild-type cell contains ~10
tetramers. The repressor gene is not controlled; it is an unregulated gene. It is transcribed into a
monocistronic mRNA at a rate that appears to be governed simply by the affinity of its (poor)
promoter for RNA polymerase. In addition, lacI is transcribed into a poor mRNA. This is a
common way to restrict the amount of protein made. In this case, the mRNA has virtually no 5 '
UTR, which restricts the ability of a ribosome to start translation. These two features account for
the low abundance of lac repressor protein in the cell.
The repressor functions by binding to an operator (formally denoted O lac) at the start of the
lacZYA cluster. The sequence of the operator includes an inverted repeat. The operator lies
between the promoter (Plac) and the structural genes (lacZYA). When the repressor binds at the
operator, it prevents RNA polymerase from initiating transcription at the promoter. When cells of
E. coli are grown in the absence of a -galactoside they have no need for -galactosidase, and
they contain very few molecules of the enzyme, about five per cell. When a suitable substrate is
added, the enzyme activity appears very rapidly in the bacteria. Within 2 to 3 minutes some
enzyme is present, and soon each bacterium accumulates ~5000 molecules of enzyme. (Under
suitable conditions, -galactosidase can account for 5%-10% of the total soluble protein of the
bacterium.) If the substrate is removed from the medium, the synthesis of the enzyme stops as
rapidly as it started.

Lac Repressor Is Controlled by a Small-Molecule Inducer

The ability to act as inducer or corepressor is highly specific. Only the substrate/product of
the regulated enzymes or a closely related molecule can serve this function. In most cases,
though, the activity of the small molecule does not depend on its interaction with the target
enzyme. For the lac system the natural inducer is not lactose, but a byproduct of the LacZ
enzyme, allolactose (lactose is converted into allolactose by -galactosidase). Allolactose is
also a substrate of the LacZ enzyme, so it does not persist in the cell. Some inducers resemble the
natural inducers of the lac operon but cannot be metabolized by the enzyme. The example par
excellence is isopropylthiogalactoside (IPTG), one of several thiogalactosides with this
property. IPTG is not metabolized by -galactosidase; even so, it is a very efficient inducer of the
lac genes.
Molecules that induce enzyme synthesis but are not metabolized are called gratuitous
inducers. The existence of gratuitous inducers reveals an important point. The system must
possess some component, distinct from the target enzyme, that recognizes the appropriate
substrate, and its ability to recognize related potential substrates is different from that of the
enzyme. The separate component that represses the lac operon is the lac repressor protein,
which is encoded by the lacI gene. The lac repressor protein is induced by allolactose and IPTG
to allow expression of lacZYA . The LacZ enzyme (-galactosidase) utilizes allolactose and
lactose as substrates. lacI is not induced by lactose, and the LacZ enzyme does not metabolize
IPTG.

The component that responds to the inducer is the repressor protein encoded by lacI. Its
target, the lacZYA structural genes, is transcribed into a single mRNA from the promoter just
upstream of lacZ. The state of the repressor determines whether this promoter is turned off or on.

The crucial features of the control circuit reside in the dual properties of the repressor: It
can prevent transcription, and it can recognize the small-molecule inducer. The repressor has two
types of binding site: one type for the operator DNA and one type for the inducer. When the
inducer binds at its site, it changes the structure of the protein in such a way as to influence the
activity of the operator-binding site. The ability of one site in the protein to control the activity of
another is called allosteric control.
Induction accomplishes a coordinate regulation: All the genes are expressed (or not
expressed) in unison. The mRNA is translated sequentially from its 5' end, which explains why
induction always causes the appearance of -galactosidase, -galactoside permease, and galactoside transacetylase, in that order. Translation of a common mRNA explains why the
relative amounts of the three enzymes always remain the same under varying conditions of
induction. Usually, the most important enzyme is first in the operon.
The constitution of the lac operon has several potential paradoxes. First, the lac operon
contains the structural gene (lacZ) coding for the -galactosidase activity needed to metabolize
the sugar; it also includes the gene (lacY) that codes for the protein needed to transport the
substrate into the cell. If the operon is in a Repressed state, how does the inducer enter the cell to
start the process of induction? The second paradox is that -galactosidase (encoded by lacZ) is
required to make the inducer allolactose to induce the synthesis of -galactosidase. How is
allolactose synthesized to allow induction of the gene? (An operon with a mutant lacZ gene
cannot be induced.)
Two features ensure induction of the Lac operon. First, the operon has a basal level of
expression, ensuring that a minimal amount of LacZ and LacY proteins are present in the cellenough to start the process. Even when the lac operon is not induced, it is expressed at a residual
level (0. 1 % of the induced level). In addition, some inducer enters the cell via another uptake
system. The basal level of -galactosidase then converts some lactose to allolactose, leading to
induction of the Lac operon.

The Lac Operon Has a Second Layer of Control: Catabolite

Repression
The E. coli lac operon is negative inducible. Transcription is turned on by the presence of lactose
by removing the lac repressor. This operon, however, is also under a second layer of control and
cannot be turned on by lactose if the bacterium has a sufficient supply of glucose. The rationale
for this is that glucose is a better energy source than lactose, so there is no need to turn on the
operon if there is glucose available. This system is part of a global network called catabolite
repression that affects about 20 operons in E. coli. Catabolite repression is exerted through a
second messenger called cyclic AMP (cAMP) and the positive regulator protein called the
catabolite repressor protein (CRP) (CRP can also stand for cAMP receptor protein and is also
called catabolite activator protein, or CAP). The lac operon is therefore under dual control.

Some promoters, do not allow RNA polymerase to initiate transcription without assistance from
an ancillary protein. Such proteins are positive regulators, because their presence is necessary to
switch on the transcription unit. Typically, the activator overcomes a deficiency in the promoterfor example, a poor consensus sequence at -35 or -10 or both. One of the most widely acting
activators is CRP. This protein is a positive regulator whose presence is necessary to initiate
transcription at dependent promoters. CRP is active only when bound to cAMP, which behaves
as a classic small-molecule inducer for positive control.

cAMP is synthesized by the enzyme adenylate cyclase. Adenylate cyclase activity is repressed
by high glucose. Thus, the level of cAMP is inversely related to the level of glucose. Only with
low levels of glucose is the enzyme active and able to synthesize cAMP. In turn, cAMP binding
is required for CRP to bind DNA and activate transcription. Thus, transcription activation by
CRP only occurs when cellular glucose levels are low.

The Tryptophan Operon: Negative Transcriptional Control of

Repressible Genes

The tryptophan (trp) operon of E. coli consists of five structural genes that encode enzymes
needed for synthesis of the amino acid tryptophan. It is regulated by the trp repressor, which is
encoded by the trpR gene. Because the enzymes encoded by the trp operon function in a

biosynthetic pathway, it is wasteful to make the enzymes needed for tryptophan synthesis when
tryptophan is readily available. Therefore, the operon functions only when tryptophan is not
present and must be made de novo from precursor molecules. To accomplish this regulatory goal,
the trp repressor is synthesized in an inactive form that cannot bind the trp operator as long as
tryptophan levels are low. When tryptophan levels increase, tryptophan acts as a corepressor,
binding the repressor and activating it. The repressor-corepressor complex then binds the
operator, blocking transcription initiation.
Like the lac operon, the trp operon is subject to another layer of regulation. In addition to being
controlled at the level of transcription initiation by the trp repressor, expression of the trp operon
is also controlled at the level of transcription elongation by a process called attenuation. This
additional level of control serves to adjust levels of transcription in a more subtle fashion. When
the repressor is not active, RNA polymerase begins transcription of the leader region but it often
does not progress to the first structural gene in the operon. Instead, transcription is terminated
within the leader region; this is called attenuation. The ability to attenuate transcription is based
on the nucleotide sequences in the leader region and on the fact that in procaryotes, transcription
is coupled with translation. The leader of the trp operon mRNA is unusual in that it is translated.
The product, which has never been isolated, is called the leader peptide. In addition to encoding
the leader peptide, the leader contains attenuator sequences. When transcribed, these sequences
form stem loop secondary structures in the newly formed mRNA. We define these sequences
numerically (regions 1, 2, 3, and 4). When regions 1 and 2 pair with one another (1:2), they form
a secondary structure called the pause loop, which causes RNA polymerase to slow down. The
pause loop forms just prior to the formation of a second structure called the terminator loop,
which is made when regions 3 and 4 base pair (3:4). A poly(U) sequence follows the 3:4
terminator loop, just as it does in other rho-independent transcriptional terminators. However, in
this case, the terminator is in the leader rather than at the end of the gene. Another stem-loop
structure can be formed in the leader region by the pairing of regions 2 and 3 (2:3). The
formation of this antiterminator loop prevents the generation of both the 1:2 pause and 3:4
terminator loops.
How do these various loops control transcription termination? Three scenarios describe the
process. In the first, translation is not coupled to transcription because protein synthesis is not
occuring. In other words, no ribosome is associated with the mRNA. In this scenario, the pause
and terminator loops form, stopping transcription before RNA polymerase reaches the trpE gene.
In the next two scenarios, translation and transcription are coupled; that is, a ribosome associates
with the leader mRNA as the rest of the mRNA is being synthesized. The interaction between
RNA polymerase and the nearest ribosome determines which stem-loop structures are formed.
As a ribosome translates the mRNA, it will follow the RNA polymerase. Among the first several
nucleotides of region 1 are two tryptophan (trp) codons; this is unusual because normally there is
only one trp per 100 amino acids in E. coli proteins. If tryptophan levels are low, the ribosome
will stall when it encounters the two trp codons. It stalls because the paucity of charged tRNAtrp
molecules delays the filling of the Asite of the ribosome. Meanwhile the RNA polymerase
continues to transcribe mRNA, moving away from the stalled ribosome. The presence of the
ribosome on region 1 will prevent it from base pairing with region 2. As RNA polymerase
continues, region 3 is transcribed, enabling the formation of the 2:3 antiterminator loop. This
prevents the formation of the 3:4 terminator loop. Because the terminator loop is not formed,

RNA polymerase is not ejected from the DNA and transcription continues into the trp
biosynthetic genes. If, on the other hand, there is plenty of tryptophan in the cell, there will be an
abundance of charged tRNAtrp and the ribosome will translate these two trp codons in the leader
peptide sequence without hesitation. Thus the ribosome remains close to the RNA polymerase.As
RNA polymerase and the ribosome continue through the leader region, regions 1 and 2 are
transcribed and readily form a pause loop. Then regions 3 and 4 are transcribed, the terminator
loop forms, and RNA polymerase is ejected from the DNA template. Finally, the presence of a
UGA stop codon between regions 1 and 2 will cause early termination of translation. Although
the leader peptide will be synthesized, it appears to be rapidly degraded.

Attenuations usefulness is apparent. If the bacterium is deficient in an amino acid other than
tryptophan, protein synthesis will slow and tryptophanyl-tRNA will accumulate. Transcription of
the tryptophan operon will be inhibited by attenuation. When the bacterium begins to synthesize
protein rapidly, tryptophan may be scarce and the concentration of tryptophanyl-tRNA may be
low. This would reduce attenuation activity and stimulate operon transcription, resulting in larger
quantities of the tryptophan biosynthetic enzymes. Acting together, repression and attenuation
can coordinate the rate of synthesis of amino acid biosynthetic enzymes with the availability of
amino acid end products and with the overall rate of protein synthesis. When tryptophan is
present at high concentrations, any RNA polymerases not blocked by the activated repressor
protein probably will not get past the attenuator sequence. Repression decreases transcription
about seventyfold and attenuation slows it another eight- to tenfold; when both mechanisms
operate together, transcription can be slowed about 600-fold.
Attenuation is important in regulating at least five other operons that include amino acid
biosynthetic enzymes. In all cases, the leader peptide sequences resemble the tryptophan system
in organization. For example, the leader peptide sequence of the histidine operon codes for seven
histidines in a row and is followed by an attenuator that is a terminator sequence.

The Arabinose Operon: Transcriptional Control by a Protein that

Acts Both Positively and Negatively