|Bacterial Growth and Regulation
Above: a section through a dividing Bacteroides fragilis cell (as seen with a transmission electron
microscope). The peripheral ribosome-rich region (riboplasm) and the more central and more
translucent nucleoid (nucleoplasm) can be clearly seen (lighter areas). The cell has been chemically
fixed, which causes the DNA fibrils to clump, giving the nucleoplasm the appearance of largely empty
space occasionally criss-crossed by random fibrils of DNA. In cryofixation the DNA is more evenly
dispersed as a fine fibril mesh throughout the nucleoid (which is more likely to be the natural state in life).
This cell has also been grown in a medium with elevated salts (in this case bile salts, which occur
naturally in its environment) which causes two changes: 1) more clumping or precipitation of DNA,
occasionally into very thick dark strands, seen in chemical fixation; 2) The nucleoid often becomes more
compacted, occupying less of the cell volume (seen in both chemical and cryofixation) although in these
cells it still appears quite dispersed, it becomes more condensed as salt concentration is increased,
although the cells remain viable.
The rod-shaped cell is dividing by binary fission - it is splitting into two new daughter cells as the cell
envelope grows inwards around the cell equator. During this process, the daughter nucleoids
disentangle and each daughter cell receives a complete nucleoid and at least one copy of the genome.
Binary fission gives rise to exponential growth and the rate at which bacteria can grow is like something
from science fiction! Some bacteria can double every 10 or 20 minutes under optimum conditions
(adequate oxygen, plentiful nutrients, an optimum temperature). A quick calculation shows that at this
rate, starting with one cell, after 24 to 48 hours the total mass of the cell's progeny (the population of its
descendants) would outweigh the Earth! Fortunately for Earthlings, Nature does not allow this as
conditions do not remain optimum for long - the cells eventually run out of food, or poison themselves
with their own waste products and predation keeps numbers in check (bacteria have many natural
predators, including protozoa, rotifers and sponges).
Structure of the Nucleoid
Bacteria (with few exceptions) have no nuclear membrane or envelope to enclose the DNA, as do
eukaryotes (like animals, fungi and plants). Instead the DNA sits naked in the cytoplasm. In bacteria, the
nucleoid is a circle of double-stranded DNA (dsDNA) that is a little over 1 mm long (1.1 mm) and yet is
enclosed in a cell only 1-2 micrometres (0.001 to 0.002 mm) long! Each bacterial cell generally contains
one copy of the genome which consists of a single large circular DNA molecule (however, see plasmids
and cell morphology below). The DNA of Escherichia coli is about 3.8 Mbp long (3800 kbp or 3 800 000
bp, bp = base pair) and contains about 1000 genes. Most bacteria have only one chromosome
9although multiple copies may be present in a single cell under certain conditions).
The DNA is packaged in two ways:
1) It is supercoiled, meaning that the double helix of DNA is further wound into a supercoiled helix. It is
a bit like taking a rubber band, representing the DNA, and twisting it, to represent the double helix and
holding the ends together in a circle; you will find that it tends to form a series of criss-crossing loops or
supercoils rather than a simple circle because it will likely be overwound (positive supercoiling) or
underwound (negative supercoiling). Only if wound by precisely the right amount will the DNA form a
relaxed circle. DNA is a right-handed helix (meaning that looking down the axis of the DNA, each strand
follows a clockwise path). Turning the DNA a few times in the opposite, antoclockwise, direction will begin
to unwind it and if the two ends are then fastened together one has negatively supercoiled DNA as found
Enzymes called DNA topoisomerases control the form, shape or topology of the DNA. Type I
isomerases cut only a single strand of DNA, whilst type II isomerases cut both strands. DNA gyrase is a
type II isomerase essential for DNA replication and transcription. It cuts both strands of the
negatively-supercoiled DNA simultaneously, causing it to relax and the two strands to separate in the
targeted region, opening up the DNA double helix (forming on open reading frame or ORF) so that
other enzymes can read the code contained on one or both single strands so as to replicate the DNA or
produce a mRNA transcript (see ribosomes).
2) Proteins fold the supercoiled DNA into larger loops. Some loops are held closed by the proteins in a
supercoiled state, compacted and held in toward the centre of the cell in the ribosome-free region of the
central nucleoid. Other loops are opened by isomerase enzymes and extend out into the riboplasm
where ribosomes can read the open strands and produce mRNA transcripts.
In Escherichia coli, a well-studied example, the loops of DNA emanate radially from a central core. The
DNA is thus randomly divided into domains, which are dynamic, changing over time. The length of
individual supercoiled domains follows an exponential distribution, with a mean of 10 kb (kb = kilobase =
1000 bases) and there are about 500 domains per chromosome.
The role of DNA-binding proteins called histones in packaging and compacting DNA in eukaryotic
chromosomes is well known. These histones form spools around which DNA is wound. For a long term
such proteins were thought not to occur in bacteria, since they have less of a need to package their
DNA, and indeed appear to package their DNA less tightly, nevertheless similar proteins do occur in
bacteria. The protein H-NS is a major component of the nucleoid in gram negative bacteria and functions
as a global transcription regulator, affecting the transcription of a large number of genes (possibly by
affecting the shape or uncoiling of the DNA). This protein can form DNA to DNA bridges and is involved
in compacting DNA. When compacted, the transcription of DNA is reduced or prevented and the more
such nucleoprotein complexes there are, the slower the global rate of transcription. In other words, H-NS
appears to package DNA and hence switch-off transcription and gene expression. Histone-like proteins
have been found in Escherichia coli, such as HU. Generally, however, although the chromosome is tightly
organised and regulated, the degree of organisation is not as great as it is in a eukaryotic nucleus.
The Cell Cycle - DNA synthesis and segregation
When one considers the large size of the bacterial chromosome, relative to the cell, and its complex
topology then it is remarkable that the DNA can be synthesised and moved around so rapidly as cells
may divide once every 20 minutes or less, under optimum conditions. (Eukaryotes, of course, have even
larger genomes which typically consist of several chromosomes, and have ways of packing their DNA
into an extremely dense state).
Eukaryotic cells have a well-defined cell-cycle. Prokaryotes also follow a cell cycle. In Escherichia coli
growing at an optimum temperature of 37 C it takes 40 minutes to duplicate the parent chromosome and
then the cell requires a further 20 minutes, after the completion of DNA synthesis, to complete cell
division. These time periods are fixed regardless of nutrient levels and how fast cells are growing. This is
fine when the cells are dividing every 60 minutes or longer, but under optimum growth conditions the
cells may divide once every 20 minutes. However, DNA synthesis still takes 40 minutes, since this is how
long it takes the pair of replication forks to traverse the 1.1 mm chromosome - they can not speed-up!
So how do they achieve this paradoxical feat?
DNA is synthesised by an enzyme called DNA polymerase (specifically DNA-dependent DNA
polymerase) along with assisting enzymes that combine into an enzyme complex or machine called the
replisome. One DNA polymerase reads one single DNA strand and another copy reads the other DNA
strand. Each polymerase reads the bases on the DNA strand and matches them up with their
complementary bases as it synthesises the complementary strand. Thus each daughter DNA molecule
(daughter chromosome) is dsDNA in which one strand is an older parental strand and the other strand
newly synthesised (so-called semi-conservative replication, since half the old DNA is saved in each
daughter DNA). DNA polymerase
In slow growing cells, doubling every 60 minutes or longer, each daughter cell receives a single
chromosome that has been replicated in advance. In rapidly growing cells, doubling every 40 or so
minutes, each daughter receives a chromosome whose replication began 60 minutes earlier in the
grandmother cell and this chromosome is already replicating - it contains a pair of replication forks that
are already producing the granddaughter chromosomes.
In very rapidly growing cells, doubling every 20 minutes, the daughter chromosomes also begin to
replicate before they themselves are fully formed - a pair of replication forks open up on each daughter
chromosome as it itself is being synthesised. Thus, daughter granddaughter and great granddaughter
chromosomes are being synthesised simultaneously. Thus each daughter cell receives a chromosome
that is producing granddaughter and great granddaughter chromosomes - with two forks on each
daughter chromosome making two granddaughter strands, each of which has its own pair of replication
forks making the great granddaughter chromosomes (making 6 replication forks in total). Thus, multiple
copies of the chromosome are synthesised in parallel (but out of phase) overcoming the finite speed of
DNA polymerase. By making three generations of chromosomes in parallel, it is possible for the cells to
divide every 20 minutes!
Of course, if nutrients are in short supply, it may take much longer than one hour for cells to divide. It
has been estimated that in very nutrient-poor environments, such as deep inside rocks beneath the
Earth's surface, that starving bacteria may divide only once on average every 100 years.
In both Escherichia coli (Gram negative) and Bacillus subtilus (Gram positive) new-born cells have a
single chromosome (which may be replicating already as explained above, but let us assume it is not for
simplicity, i.e. assume slow growth). The origin and terminus are near the cell poles, opposite to
one-another. During replication, the origin moves to the replisome at the centre of the cell. Daughter
origins move rapidly to opposite poles. The chromosomes segregate at 0.1 to 0.3 micrometres per
minute (about 100 x faster than cell elongation during growth). This process involves the attachment of
each daughter chromosome to the inside of the cell envelope, via motor proteins which move the
chromosomes and are part of the bacterial cytoskeleton.
The Cell Cycle - wall growth and cell division
In rod-shaped bacteria (bacilli) new wall material is inserted at various points along the side-walls if little if
any new material added to the hemispherical end-walls or caps. Thus, the cell elongates as it grows, but
maintains a constant diameter (keeping the cell thin like this reduces diffusion distances for materials
entering and leaving the cell and so increases nutrient absorption and growth rate). However, more
material is apparently added at the middle of the rod where the wall grows inwards as a circular septum
which eventually closes off, completing cell division.
In cocci, new cell wall material is added mostly, if not exclusively, along an equatorial wall band, which
invaginates as the cell elongates and new wall material is added beneath the wall band, forming a wall
notch which eventually becomes a dividing septum (cross-wall), separating the two daughter cells.
Role of the Cytoskeleton in Cell Elongation and Division
For a long time it was assumed that bacteria lacked a cytoskeleton. After all, they are usually much
smaller than eukaryotic cells and have tough cell walls and they generally lack internal membranous
organelles to support and move around. However, electron microscopists occasionally glimpsed
filamentous and tubular proteins and eventually it was discovered that the cytoskeleton plays important
roles in bacteria, even if it is much simpler than the eukaryotic cell cytoskeleton.
In bacilli, there are helices of filamentous cytoskeletal proteins just inside the cell envelope (attached to
the cell membrane?) parallel to the long axis of the cell, rather like coiled springs in Bacillus subtilis and
Escherichia coli (in Rhodobacter sphaeroides they form rings). In Escherichia coli, these consist of the
protein MreB (similar to the protein actin in eukaryotes, MreB is an actin homologue) which has roles in
maintaining cell shape and in positioning the chromosomes and cell polarity. These cytoskeletal fibres
form tracks for the enzymes involved in peptidoglycan synthesis. The same fibres also seem to form
tracks for the daughter chromosomes to move along when they segregate (chromokinesis), with the
replication origins moving apart.
In one model the origins (or a region close to them) attach to the cell membrane, possibly to the MreB
tracks or to a motor protein attached to these tracks, with the origins on each of the two parental strands
attaching at separate adjacent positions near the middle of the cell and then moving apart as replication
proceeds. In older models, cell wall (and cell membrane) growth in-between the attached origins causes
them to separate, since cell wall material is deposited preferentially in this equatorial region. It is now
thought that the molecular motors moving the chromosomes along the MreB tracks accounts for the
initial segregation, with perhaps cell wall growth accounting for much of the subsequent separation. DNA
condensation also pulls on the DNA, contracting it and this may help segregation too. Several other
proteins are involved in chromosome segregation, such as ParA and ParB and the migS region of the
DNA molecule (close to the replication ori) is also necessary (and may be the site of chromosome
attachment to the cell membrane or to the cytoskeleton)..
When replication of the chromosome is complete and the two replication forks meet at the terminus, the
daughter strands remain joined and are separated by the enzyme topoisomerase IV. The protein Ftsk
is involved in activating topoisomerase IV and septation (formation of the septum or cross-wall that
separates the daughter cells) and the formation of the divisisome.
In the gram positive Bacillus subtilis, there are several actin homologues: MreB, Mbl (a mreB
homologue)and MreBH. Mbl is involved in cell elongation, forming helices lining the inside of the cell
membrane and arranged parallel to the long axis of the cell. These filaments cause helices of
peptidoglycan to be inserted into the cell wall, maintaining the rod shape of the cell. As the cell grows
and elongates, more Mbl is added at intervals along the helix, lengthening it and new longer helices of
peptidoglycan are inserted into the wall. The older helices of peptidoglycan stretch out as the cell
elongates, twisting the ends of the cell as they slowly unwind and straighten. In Bacillus subtilis, MreB
seems to have a role in regulating cell width.
The Z Ring
Tubulin homologues (similar to the tubulin that forms microtubules in eukaryotic cells) also occur in
bacteria. One is the FtsZ protein, which occurs in most eubacteria and archeabacteria (archaea) and
some chloroplasts and mitochondria. There are about 32000 to 15000 FtsZ molecules per cell in
Escherichia coli, 5000 in each Bacillus subtilis cell. FtsZ monomers are thought to polymerise and,
perhaps along with other proteins, form the Z ring in cell division. The Z ring spans the cell circumference
and contains about 30% of the cell's FtsZ. (The proteins FtsA, ZipA, ZapA, EzrA, Noc, SlmA, minC and
SulA affect FtsZ assembly). The Z ring is involved in cell division and forms a cell division ring in
Escherichia coli, possibly actively constricting the cell as the septum forms (in a manner similar to the
actin ring in cytokinesis in animal cells).
DNA Replication in Bacteria
The circular double-stranded DNA chromosome of bacteria is duplicated by bidirectional synthesis. It
begins near a special site on the chromosome, called the origin (ori, O) where the DNA opens up as
hydrogen bonds between the bases in the complementary strands are enzymically broken and the
strands separate, forming a replication 'bubble' at which the DNA forks as the two strands separate on
either side. The DNA continues unzipping at these forks, in both directions simultaneously, as the
replisome (an enzyme complex containing the enzyme DNA polymerase which synthesises new DNA
and enzymes responsible for unzipping the double helix and other associated operations). These
so-called replication forks move around the chromosome in opposite directions as the DNA is unzipped
and duplicated (hence bidirectional synthesis). DNA polymerase reads the base sequence of the
unzipped single strands of parental template DNA (blue circles below) and matches the sequence of
bases with new bases which are polymerised (bonded together end-to-end in a molecular chain) to
create the complementary daughter strands (red dashed circles below). This continues until the
replication forks reach the terminus (T) at the far-side of the chromosome, opposite the origin.
The end result is two daughter DNA duplexes (double-stranded DNA) in which one strand (the solid blue
circle in the diagram above) is old parental DNA, the other (the red dashed circle) newly synthesised
daughter DNA. Since each new molecule preserves half of the original DNA, this mode of synthesis is
called semi-conservative. (In the hypothetical alternative, conservative synthesis, both strands are
newly synthesised in one DNA molecule, whilst the other retains all the parental DNA).
Plasmids and Megaplasmids
In addition to the single main chromosome, bacteria may possess one or more dispensable,
self-replicating and small circular dsDNA molecules called plasmids. These are not essential and not
every individual cell will possess plasmids. Essentially they are infectious agents that typically benefit
their host, for example, they may contain genes that help the cell cope with certain stresses such as
genes encoding resistance to an antibiotic (such plasmids are called R factors or resistance factors).
Some give the host pathogenic abilities, or allow them to synthesise antibiotics or pigments. Some have
no apparent function, other than their own replication, and are called cryptic plasmids (and may be
somewhat parasitic, simply hitching a ride inside the host cell and using host nutrients to replicate).
Some plasmids also carry genes that encode for mechanisms that enable them to 'infect' other bacterial
cells with copies of themselves, such as by forming sex pili and bridges during conjugation. Plasmids
may remain free in the cytoplasm or insert reversibly into the host chromosome.
Some plasmids are incompatible with others, in which case the incompatible types cannot reliably
replicate when present together in the same cell. Plasmids vary in size from less than 10 kbp to over 100
The actin homologue ParM (with about 15 000 to 18 000 molecules per cell) appears to form filaments
stretching from one cell pole to the other, and playing a role in plasmid replication and partitioning
plasmids between the two daughter cells.
Vibrio cholerae is one of the minority of bacterial species that have more than one chromosome in their
genome, it has two. Chromosome I is the main chromosome and the smaller chromosome II may be
descended from a megaplasmid (an exceptionally large plasmid). During cell division, chromosome one
attaches its ori close to one cell pole and then the daughter chromosomes separate rapidly, with one
remaining fixed whilst the other moves to the opposite pole. Chromosome II attaches its ori near the cell
centre and segregates in the usual way.
Cell division and DNA synthesis are usually
tightly coupled. Changes to the cell cycle
may decouple DNA synthesis from cell
division. When this happens cell shape may
change dramatically. For example, treating
Escherichia coli with DNA-synthesis inhibiting
drugs prevents cell division, resulting in
greatly elongated cells, but without
increasing the DNA content. This is not what
has happened to the Bacteroides fragilis cell
on the left. It is greatly elongated, but the
presence of nuclear material throughout the
length of the cell shows that DNA synthesis
has been taking place and keeping pace with
cell growth, but that septa have not formed.
(The cell also contains dark granules,
possibly of stored reserve materials and the
cytoplasm is also different in appearance to
that of many of the neighbouring cells). Some
bacteria can form long multinucleoid
filamentous cells in response to various
stimuli, such as the detection of chemicals
from a potential predator, in which case the
elongated cells may be too long for the
predator to ingest. All sorts of factors, both
physiological and pathological, may affect
bacterial cell size and shape. I don't know
what caused the anomaly in this case.
Regulating Cell Composition
The number of ribosomes in a bacterial cell (and hence the extent of the riboplasm and size of the cell)
are well regulated. Very slow growing cells, with a scarce supply of nutrients, have excess ribosomes,
not all are being utilised. (Note that a given ribosome works at a more-or-less constant rate when it is
translating mRNA unless amino acids are in very short supply and then the ribosome must wait for the
correct amino acids to arrive). If the cell then encounters more nutrients (such as by inoculation into
nutrient-rich medium in a shift-up experiment) it has a spare capacity of ribosomes (and also mRNA)
with which to resume rapid growth. If nutrient concentrations are high enough, then all the ribosomes will
be fully engaged and working at full capacity. Now the cell will synthesise more ribosomes and enlarge
its cytoplasmic volume. Faster growing cells have higher RNA to protein ratios and higher RNA to DNA
ratios due to the increase in ribosomes (which contain ribosomal RNA). In optimum growth conditions,
ribosomes account for about 40% of the cell's dry mass. If a well nourished cell is suddenly deprived of
nutrients (such as by inoculation into nutrient-poor medium, a shift-down experiment) there is no
further net RNA and protein synthesis - new RNA and enzymes are manufactured from recycled
components made available by degradation of some of the surplus ribosomes. DNA synthesis and cell
division continue (unless nutrients are very severely limiting) and results in smaller cells (with fewer
Growth of Bacterial Populations
The doubling of bacterial numbers when bacteria multiply by binary fission is an example of exponential
growth. The exponential function is a mathematical function whose value at a future time (or other
dependent variable depending on the context) depends on its value at the present time. Thus, if we start
with 2 cells then after one more cell cycle we have 4 cells, if we start with 120, then we get 240. When
bacteria are transferred into a new culture with different nutrients, there is typically a lag phase, during
which little or no cell division occurs but which the cells prepare for cell division. They may have to
synthesise new enzymes to cope with the change in nutrients, or they may have been previously starved
and so need to repair cell damage before resuming cell division. Once the lag phase has passed,
however, then they resume exponential growth.
This exponential growth continues whilst food lasts, but it cannot continue forever. Eventually the
conditions become less favourable for growth as the nutrients become depleted and/or toxic waste
products accumulate in the environment. Cell division slows as population growth enters the
deceleration phase and eventually the net growth rate becomes zero and the populations enters what is
called stationary phase, so-called because during this phase viable cell numbers remain more-or-less
constant. As old worn-out cells die, they release enzymes that break open the dead cells and release their
nutrients which are recycled and some new cells are born.
Stationary Phase Physiology
Some very interesting things happen to bacteria in stationary phase. Some of these events are regulated
responses, as cells try to make the best of a bad situation, 'hoping' to survive long enough to encounter
fresh nutrients and repair themselves in a new lag phase before resuming growth. Growth during
stationary phase is unbalanced growth, the various components of the cell do not keep pace with
one-another. DNA and protein synthesis decline, whilst ribosome synthesis increases. In this way the cell
has lots of ribosomes ready to rapidly resume growth should the chance arise. (At these low growth rates
not all the ribosomes are functioning). In stationary phase, cell maintenance takes an increasing share of
the cell's resources, as growth declines - the cell's place less emphasis on trying to grow and more
emphasis on maintaining vital systems, such as the integrity of their DNA. Cell shape usually changes and
in the case of rod-shaped cells, continued division outstrips cell growth and a population of smaller cells
results. This is an active response and probably serves to increase the surface area to volume ratio of
the cells for rapid uptake of nutrients, should the chance arise in future. The cells are also typically
spherical / coccoid which is known to enhance their passive dispersion by Brownian motion. Whilst in this
dormant or semi-dormant state cells may also be resistant to many stresses which effect rapidly growing
and metabolising cells. Stationary phase can last a variable length of time, depending on species and
conditions. Eventually the number of viable cells will decline as the population dies and enters decline or
Batch and Continuous Culture
The graph above illustrates the phases we have discussed, which occur in a closed system (one in which
fresh nutrients do not enter the system (gases, however, are free to enter or leave). Such a system might
be a batch culture, in which bacteria are grown in liquid nutrient broth in a sterile flat-bottomed flask on
a shaker (mixing in air via a sterile cotton wool plug to ensure oxygen availability) in an incubator at 37 C.
In contrast, continuous culture involves passing in a continuous or periodic fresh supply of nutrient
broth whilst flushing out the old broth and its waste products and might be achieved using a chemostat.
This maintains fairly rapid population growth, but by flushing through the broth the bacterial numbers are
Quorum Sensing and Biofilm Formation
Many bacteria sense their cell density! They secrete chemical messengers (autoinducers) and the
concentration of these messengers (i.e. the strength of the signal received) is an indication of cell
density, a phenomenon called quorum sensing (QS). This allows bacteria to alter their behaviour
accordingly - genes switch on or off depending on the density of cells. In particular, when the density is
sufficiently high quorum-sensing activates genes responsible for group behaviour, such as biofilm
differentiation or virulence. Biofilms are a normal part of the life-cycle of most bacteria (see the article on
slime cities). When in biofilms bacteria are also much more resistant to antimicrobial agents, such as
bleach, detergents and antibiotics.
Q. What happens if we grow bacteria in batch culture in the presence of a limited amount of a
carbon/energy source, such as glucose and a second carbon/energy source like lactose?
A. The population may begin growing slowly with a lag phase of variable length, depending on the state of
the initial innoculum. Then they will grow exponentially, using glucose as their carbon/energy source,
since it can be metabolised more rapidly and efficiently than lactose and so is the preferred substrate.
Until the glucose begins to run out and then they will enter deceleration and stationary phase. Assuming
that waste products do not limit growth, the cells will then enter a second lag phase. Previously the cells
were only metabolising glucose and will have synthesised enzymes for that purpose and so during the lag
phase they must synthesise the necessary enzymes to metabolise lactose, before growth can resume.
They will then enter a second phase of exponential growth before entering a second stationary phase
and subsequent decline phase when all carbon/energy sources are exhausted.
Such an experiment produces a diauxic growth curve:
The diauxic growth experiment shows how well regulated bacteria are - when given a choice of nutrients
they will feed preferentially on the one that is most easily utilised and exhaust its supply before
synthesising the enzymes needed to metabolise the less preferred nutrient. This strategy optimises
growth. This phenomenon also necessitates sensing - the cells must sense what nutrients are available
in their environment and respond accordingly.
The lac Operon
Lactose is a disaccharide (double sugar - two sugar units bonded together) of glucose and galactose
and must be split into glucose and galactose before it can be metabolised, an additional step not needed
with glucose and so bacteria generally prefer to metabolise glucose. However, as we have seen, if
glucose is not available but lactose is, then the cells must gear up to metabolise lactose by synthesising
the enzyme beta-galactosidase which breaks the lactose up into glucose and galactose and also
transporters to import lactose across the cell membrane. The genes necessary for these tasks are
clustered together side-by-side on the chromosome beside control genes that switch the gens on or off.
Such a unit is called an operon and this one is the lactose or lac operon.
Notice that RNA polymerase reads and transcribes the lac operon, as shown above, from left to right.
The names of genes are generally italicised and often in lower case, whilst the proteins they produce
are non-italicised and capitalised, so the gene lacZ produces the protein LacZ which is also called
beta-galactosidase (an alternative name for lactose is glucose-beta-D-galactoside).
The lac operon consists of two principle parts - the control genes: lacI (which is technically outside the
operon), lacP and lacO and the structural genes (those that produce the useful end-products): lacZ,
lacY and lacA.
The gene lacP, between lacI and lacZ, is the promoter - it contains the RNA polymerase binding site
where transcription begins. It also contains a CAP site, just before (uspstream, that is to the left of) the
RNA polymerase binding site, to which a protein called CAP (catabolite activating protein or CRP, cyclic
AMP receptor protein) can bind. Switching the promoter on or off controls transcription.
The gene lacO is the operator, that is the on/off switch, and is positioned between lacP and lacZ. Note
the small size of the promoter and operator.
Negative Control of the lac Operon
When there is no lactose in the medium the operon is switched off. Also, when lactose is present along
with a more easily metabolised carbon/energy source, such as glucose, it is also switched off as the
glucose is used instead. When glucose runs out or is otherwise absent and lactose is present, then the
operon switches on and lactose is metabolised. This is an example of metabolic control. In negative
control, the gene is 'actively' switched off, not by directly consuming ATP, but by the cell actively
synthesising the repressor protein.
Metabolism is the sum of the chemical reactions occurring in the cell and is divided into catabolism
(dissimilation), those reactions which break down more complex molecules into simpler components,
and anabolism (assimilation), which are those reactions which build more complex molecules from
simpler components. We are dealing here with a catabolic reaction - lactose is a carbon and energy
source and is first broken down before the carbon and energy provided are used in assimilation.
Switching the operon off. The lac operon is under negative control, meaning it is actively switched off
and passively switched on. The gene lacI produces the LacI protein which is the repressor responsible
for controlling gene induction. This protein is a tetramer (for LacI polypeptides bound and folded
together) and it binds the lacO operator and overlaps the promoter as it does so, preventing RNA
polymerase from binding and reducing transcription - the repressor switches the operon off. This is an
example of competetive inhibition - sometimes RNA polymerase binds first, preventing the repressor
from binding, but as the level of repressor increases the polymerase binds less and less. Thus this is
not a simple binary on/off switch for there are different levels at which the operon can operate. This off
state is the deafult state, that is when lactose is absent and/or glucose is present.
Switching the operon on. When lactose is present in the medium it is imported into the cell at a low basal
(constitutive) rate and binds to the repressor. The lactose is converted into allolactose which acts as an
inducer and this inducer-repressor complex will not bind lacO and the repressor detaches from the
operator and RNA polymerase can bind freely and the operon is switched on. Transcription proceeds
and mRNA is produced and translated by the ribosomes, producing the lacY permease which allows
rapid import of lactose into the cell and the beta-galactosidase enzyme which breaks-up the lactose
into glucose and galactose which are then catabolised. Only teh structural genes are transcribed, the
operator and promoter are not.
Positive Control of the lac Operon
In positive control, the cell produces a protein that switches the promoter on or turns it up, making it a
stronger promoter to which RNA polymerase binds more easily. The crp gene is situated elsewhere on
the chromosome, remote from the lac operon and is transcribed to produce the protein CAP (catabolite
activating protein or CRP, cyclic AMP receptor protein) which forms a dimer (two CAP polypeptides
bonded and folded together). This is inactive, but becomes activated when cyclic AMP (cAMP) binds to
it. In its active state it binds the lacP promoter (and the promoters of a number of other genes for
catabolism, causing a general increase in catabolic activity) converting it from a weak promoter to a
strong promoter, increasing transcription.
The cAMP is acting as a state sensor and internal cell messenger and is an example of an alarmone,
when the glucose concentration inside the cell is high, the cell synthesises less cAMP and the cAMP
concentration drops. As a result when glucose is abundant the synthesis of other catabolic enzymes
under CAP control is reduced.
However, during starvation or in the presence of less favourable metabolites, such as succinate, the cell
synthesises more cAMP and the concentration of cAMP increases, activating CAP and switching on the
transcription of catabolic enzymes.
This phenomenon was historically called 'catabolite repression' as it was assumed to be a form of
negative control before it was realised to be positive. This term is still in widespread use, however we
shall not use it in this context. A better term is: catabolism induction.
It seems that although both negative and positive control may be found in both prokaryotes and
eukaryotes, that negative control is more common in prokaryotes whilst positive control is more common
in eukaryotes. This makes sense, bacteria have fewer genes and so it often makes more sense to
switch off those that are not needed and keep the majority on. Eukaryotes have many more genes and
in multicellular organisms, such as plants and animals, only those required by each differentiated cell
type will be switched on, so it is easy to switch on those that are needed.
Mechanisms of Cell Regulation - Control Mechanisms
The negative and positive control of the lac operon are two examples of metabolic control. Other
mechanisms also exist. All these mechanisms change the levels of key proteins/enzymes within the cell
and can be classified as follows:
1) Changing Gene Structure
For example, Salmonella typhimurium exhibits flagellin phase variation. Flagellin is a key protein
component of the flagella and when Salmonella typhimurium infects a host the flagella is a major target
for host antibodies. To counter this the bacterium can switch its flagellin structure between two different
forms. A small segment of DNA inserted into the flagellin gene causes the switch - the sequence of DNA
in the gene is changed. Each cell will produce only one or other of the two flagellins at any given time,
but within a population about half will have one type and half the other, such that if an antibody is raised
against one of the flagellin types only half the population will be affected.
2) Transcription Control
This is the most frequently used control mechanism. Enzymes are either induced (e.g. the lac operon)
or repressed. Thus transcription of the gene is switched on or off (or turned up and down). Another
method of translational control involves changing the efficiency of transcription, that is how well RNA
polymerase binds to the promoter. RNA polymerase is a multi-component enzyme and one component,
called the sigma factor, can be readily changed. Eukaryotic cells have many different sigma factors,
and each is best at recognising certain promoters. Prokaryotes also utilise this method, though
apparently less so. For example, Bacillus subtilis has four different sigma-factors which activate different
genes involved in different stages of sporulation (endospore formation).
3) Translation Control
This modifies the translation of mRNA. An example is the production of ribosome proteins. If ribosomes
are partially assembled (that is if the cell is making more) then ribosome proteins will preferentially bind
to the assembling ribosomes and complete assembly. However, if they find no assembling ribosome they
will instead bind to ribosomal protein mRNA and block translation. In this way the cell has a store of
ready-made RNA so that it can rapidly synthesise ribosomes when needed - a useful advantage since
ribosomes are essential for cell growth. Another example, unique to prokaryotes, is attenuation (see
4) Post-translational Modification of Proteins
If mRNA has been transcribed and translated and the protein synthesised, then it may still be possible to
switch the protein on or off. This is common in eukaryotes, for example many enzymes are produced as
inactive proenzymes which have to be chemically modified before they become active (allowing safe
storage of enzymes without 'eating' the cell!). There are some examples in prokaryotes, for example the
regulation of glutamine synthetase, a key enzyme in controlling nitrogen metabolism and which
synthesises glutamine from glutamate. The enzyme is deactivated by an enzyme which adenylates it
(adds an adenyl group to it) which occurs when ammonia concentrations are high. This happens
because glutamine synthetase works well at low ammonia concentrations but requires ATP. When
ammonia levels are high an alternative route for glutamate metabolism can function, which does not
require ATP and so switching the enzyme off saves ATP, which is the cell's precious store of
immediately utilisable energy. Other examples include the phosphorylation of enzymes and proteins
by enzymes called kinases. This plays a crucial role in cell signalling in both eukaryotes and
5) Allosteric Regulation
This is a method of switching proteins on or off that does not involve their covalent modification.
Enzymes regulating metabolic pathways are often turned up by their substrate and turned down by the
end-product of their metabolic pathway. A metabolic pathway is a chain of chemical reactions that lead
to an end-product via several intermediate chemicals. The intermediates may also be useful as other
metabolic pathways may branch from them. An example of such a pathway in prokaryotes is shown
Attenuation is an ingenious mechanism of translational control unique to prokaryotes. It involves
different folding patterns in mRNA, one of which will allow transcription, whilst the other blocks it. Recall
that RNA is not simply a single strand, though it largely is it does partially pair with itself in certain
regions (as bases pair by hydrogen-bonding to complementary bases on the same nucleotide strand)
forming some double-stranded regions.
One example is the trp operon which codes genes for the synthesis of the amino acid tryptophan. In
bacteria, transcription and translation are concurrent, that is as RNA polymerase produces mRNA
transcripts, ribosomes begin to read and translate these transcripts before they are finished, such that a
forming mRNA has ribosomes with forming polypeptides moving along it. This is possible in prokaryotes
because no nuclear envelope separates transcription from translation as it does in eukaryotes.
Attenuation requires this concurrency and so is only possible in prokaryotes.
There are two scenarios:
1) The cell has plenty of tryptophan and so does not need to synthesise any. In this case it does not
want the ribosomes to translate the mRNA transcript. transcription is only one step ahead of translation,
but once transcribed, regions 3 and 4 pair to one-another, forming a tight double-stranded loop, called
a terminator hairpin because the ribosome cannot pass this tight loop and detaches once it reaches
the STOP codon (UGA) in region 2 and translation is not complete, instead only a short popypeptide
(consisting of region 1 and part of region 2) is produced and this is rapidly degraded by nuclease
2) The cell is starved of tryptophan and needs translation of the trp mRNA to complete. The first
segment of the mRNA (region 1 in the diagram below) has two consecutive tryptophan (trp) codons,
UGG-UGG. If tryptophan is in short supply then this slows down the ribosome as it is waiting for
tryptophan to arrive, bound to its tRNA carrier. Whilst the ribosome is paused, the RNA polymerase
continues to synthesise the mRNA, synthesising regions 2, 3, and 4. Regions 2 and 3 are synthesised
before the ribosome can get there and region 2 pairs with region 3. This prevents region 3 from pairing
with region 4 to form the terminator hairpin. The loop formed by pairing between regions 2 and 3 is wide
and can be easily negotiated by the ribosome. The ribosome passes the STOP codon without
detaching, it may pause briefly, but is presumably able to move quickly on before proteins called
terminating factors are able to arrive and terminate translation. Translation is complete.
These situations are summarised in teh diagram below:
This is part of the nitrogen metabolism pathway. Nitrogen is an essential element for the synthesis of
nucleic acids and their nucleotide building blocks and proteins and their amino acid building blocks. The
abbreviations are: ATP, adenosine trisphosphate; CTP, cytidine trisphosphate; UMP; UDP; UTP. Red
(dashed) arrows indicate inhibition and the blue (dotted) arrows activation of the controlling enzyme.
The end-product CTP inhibits the action of the enzyme ATCase (aspartate transcarbamylase) by
negative feedback (specifically end-product inhibition). This means that as the concentration of
CTP increases, the rate of its production slows, which makes perfect sense if its is surplus to
requirements since then nitrogen can be used to make other products. The synthesis of CTP uses lots
of valuable ATP, so CTP synthesis increases when ATP is abundant as ATP stimulates ATCase activity
(at least in vitro) which is an example of positive feedforward.
The activators and inhibitors shown in the diagram are all allosteric activators/inhibitors or allosteric
effectors and the process by which they regulate enzyme function is allosteric regulation. Allosteric
proteins are literally 'differently shaped' proteins and the term refers to proteins or enzymes that can
change shape. When these shape changes alter the shape of the active site of the enzyme then they
alter its rate of action. Enzymes catalyse reactions when the reactants or substrates bind to the
enzyme's active site, where the reaction is catalysed and one or more products are produced and
released from the active site. For example, ATCase acts on the substrate carbamyl phosphate and turns
it into the product carbamyl aspartate.
Allosteric effectors, like CTP, bind their target enzymes at sites remote from the active site and induce
the enzyme to change shape and become more or less active. If the enzyme has several such sites and
can change shape through a graded series of forms then it is possible to inhibit or activate the enzyme
by varying degrees. This occurs in enzymes made up of several polypeptide units, each with its own
allosteric effector binding site, its own active site and each capable of changing shape. Consider, for
example, an enzyme made up of 4 polypeptide units. Binding of one allosteric inhibitor switches off 1
unit, but 3 remain functional. Binding of a second inhibitor switches off a second active site, and so on.
Such enzymes often show cooperativity, in that binding of the substrate to the active site of one unit
also induces a shape change which also induces shape-changes in the other units, increasing their
affinity for the substrate. Binding of a second substrate induces another change, further increasing the
affinity of the two remaining vacant sites, and so on. Allosteric effectors can also bind cooperatively, for
example binding of an inhibitor to one unit may increase the affinity of binding of the other units to other
molecules of the inhibitor, making the enzyme more responsive to further inhibition.