GENE EXPRESSION

GSND 5113Q
LEARNING OBJECTIVES
Recognize start and stop signals for prokaryotic transcription
Understand regulation of the lac operon (& other related operons)
Become acquainted with some mechanisms of epigenetic regulation
Learn principles underlying plasmid maintenance ("addiction") systems
Appreciate why T7 expression systems are widely used in biotechnology
Emanuel Goldman, Ph.D.
New Jersey Medical School
Rutgers Biomedical Health Sciences (RBHS)
Rutgers, The State University of New Jersey
Department of Microbiology, Biochemistry & Molecular Genetics
International Center for Public Health (ICPH)
225 Warren Street, room E450T
Newark, NJ 07103 USA
Telephone: 973-972-4367
Fax: 973-972-3644 or -8982
Email: [email protected]
 TRANSCRIPTION
RNA polymerase: "Core" = α2ββ’; additional subunit ω not essential but increases efficiency
"Holoenzyme" = Core + Sigma factor (σ)
Promoter: DNA sequence usually upstream of gene;
(Initiation) σ70: -10 region (TATA Box) and -35 region
Elongation: Factors GreA and GreB:
relieve transcription pausing during elongation via nucleolytic activity
Termination: Intrinsic: Stem-Loop followed by a run of U’s; or
rho dependent: rho factor protein binding

“TATA box” RNA start

-35 region -10 region +1
5’ TTGACA TATAAT Pu 3’ σ70

examples of σ70 controlled genes:

tRNATyr : TTTACAGCGGCGCGTCATTTGA•TATGATGCGCCCC•G (tRNA)

trp : TTGACAATTAATCATCGAACTAGTTAACTAGTACGC•A (tryptophan biosynthesis)

rrnD1 : TTGTGCAAAAAATTGGGATCCC•TATAATGCGCCTCCG (rRNA)

 Model for ‘intrinsic’
transcription termination
TSS = transcript stabilization site
TBS = upstream tight product binding site
LBS = 3' proximal loose product binding site
o = nucleotide in the template DNA strand
 = nucleotide in the transcribed RNA
shaded areas = domains of RNA polymerase

Stem-loop detaches transcript from TSS;

consecutive U's weaken binding of RNA
product to template. The coincidence of
these two features with the appropriate
spacing between them detaches the transcript
from polymerase, terminating transcription.

Terminated
Model for λ N-mediated antitermination

A complex of cellular proteins including NusA, NusB, NusG, and

S10 (of the 30S ribosomal subunit), along with the λ-encoded N
protein, assembles at signals near the 5’ end of the mRNA called
“nut” sites, consisting of stem-loop ‘box B’ and a primary consensus
sequence ‘box A’. This complex interacts with and modifies the
behavior of RNA polymerase to resist both intrinsic and rho-dependent
transcription termination signals.
Transcriptional Regulation 1 Overview:

1. RNA polymerase recognition/binding of the promoter

a. Repressor-operator model for controlling initiation of transcription.

Operon: set of genes on one transcription unit, controlled coordinately.

b. Induction of the lac operon, constitutive expression,

cis- and trans-acting elements.

c. Polarity

d. cyclic AMP plus its receptor protein (catabolite repression)

e. Phosphotransferase system

f. DNA bending (gal and ara operons)

g. Autoregulation: product of a gene is transcriptional repressor

of mRNA for that gene.
Transcriptional regulation 1 overview continued:

2. Epigenetic control of expression

a. DNA rearrangements: Salmonella phase variation

b. Methylation control: pathogenic pili phase variation

3. Toxin-antitoxin systems for plasmid maintenance

a. Antisense RNA: hok-sok system for plasmid R1

b. Proteolysis: ccdA and ccdB for F factor

4. T7 expression systems

a. Specificity of T7 RNA polymerase

b. Widely used for expression of recombinant genes

 Regulation of the lac operon - I

• Structural genes: β-galactosidase (z), permease (y), transacetylase (a).

• Substrate is lactose, a disaccharide composed of galactose and glucose monosaccharides.
• Promoter (P), operator (O); active repressor (I) inactivated by inducer (lactose or IPTG);
allosteric changes. Repressor binds as tetramer. The analog IPTG bypasses need for permease.
• Phenotypes and genotypes of mutants; cis-trans tests.
• Necessity for basal (leaky) expression for substrate import, and to convert lactose to allolactose.
• Polarity and polarity suppressors (SuA).
Regulation of the lac operon - II
 lac OPERON MUTANTS
SELECTED lac OPERON GENOTYPES AND PHENOTYPES
OC operator constitutive:
No inducer Induced
operator does not recognize β-galactosidase permease β-galactosidase permease
repressor; cis dominant activity
mRNA activity mRNA activity activity
i+P+O+z+y+ - - - + + +
i- repressor mutant:
repressor does not recognize operator; -
i+P+O+z+y- - - + + -
also constitutive, trans dominant
i+P+O+z-y+ - - - + - +
P- promoter mutant:
RNA polymerase cannot transcribe +
i+P+OCz+y- + - + + -
operon; non-inducible, cis dominant
i-P+O+z+y- + + - + + -
iS super-repressor:
repressor does not recognize inducer; -
i+P-O+z+y- - - - - -
non-inducible, trans dominant

These examples were set up this way to allow us to iSP+O+z+y- - - - - - -

determine whether expression was from the
chromosomal operon or the plasmid operon. For
example, for an operator constitutive chromosomal CIS-TRANS TESTS
mutation, the structural genes are always i+P+OCz+y-/F'i+P+O+z-y+:
expressed, even if you add a plasmid with the wild wild type iPO loci on F' don't change chromosomal phenotype; cis dominant;
type regulatory elements. However, the plasmid
phenotype: z product (β-galactosidase) always made (constitutive); y product (permease) inducible
genes in the operon are still repressed and
inducible. This is because the chromosomal i-P+O+z+y-/F'i+P+O+z-y+:
operator is cis-acting. But you wouldn't be able to wild type i gene on F' changes chromosomal phenotype; trans dominant;
observe this if both z and y were + on both phenotype: β-galactosidase always made in absence of F', repressed & inducible in presence of F‘
operons. By having z+, y- on the chromosome , the i+P-O+z+y-/F'i+P+O+z-y+:
operator constitutive mutant only produces the z
wild type iPO loci on F' don't change chromosomal phenotype; cis dominant;
product, no active y product. By having z-, y+ on
the plasmid, the wild type regulatory elements on phenotype: z product (β-galactosidase) never made (non-inducible); y product (permease) inducible
the plasmid will allow no y product without iSP+O+z+y-/F'i+P+O+z-y+:
inducer, but y product will be made with inducer. mutant chromosomal i gene changes F' expression; trans dominant;
So when you look at the cell + plasmid together, phenotype: β-galactosidase never made; permease never made in this host (inducible in other hosts)
you can observe z product all the time but y
i+P+O+z+y-/F'iSP+O+z-y+:
product only upon induction. This proves that the
operator constitutive mutation on the chromosome iS gene on F' prevents β-galactosidase induction; trans dominant;
has no effect on the plasmid copy of the operon. phenotype: β-galactosidase never made, permease never made with this F' present
cis & trans acting elements _ +
O z y a
P
i F'
repressor

repressor

repressors can bind to operators anywhere in cell = trans acting

operators can only affect adjacent genes = cis acting

P O z+ y _
i
a

chromosome
 Cis- and Trans- acting elements in control of the lac operon
Cis-acting elements are generally targets in DNA for binding of trans-acting factors, which are generally proteins. A cis-
acting element can only regulate genes physically linked to it, i.e., adjacent genes on the same DNA. A trans-acting factor, on
the other hand, is a diffusible product that can affect expression of target genes anywhere in the cell, whether on a
chromosome or plasmid.

In the case of the lac operon, a cis-acting operator on the chromosome only regulates the structural genes (beta-galactosidase,
permease, etc.) on the same DNA. Thus, an operator mutation may affect expression of the structural genes for the operon on
the chromosome, but will have no effect on any genes on a separate piece of DNA, like a plasmid.

Also in the lac operon, a trans-acting repressor regulates the structural genes of the operon regardless of the location of those
genes (or the location of the gene encoding the repressor). Thus, a repressor protein when it is synthesized can find its target
binding site (i.e., operator) whether the binding site is on the chromosome or on a plasmid, and may therefore affect
expression of the structural genes for the operon whether they reside on the chromosome and/or on a plasmid.

As an example, if a cell bears a chromosomal mutation in the lac operator that does not recognize (i.e., bind) the repressor,
there will be no repression of the structural genes on the chromosome that are physically linked to that operator, and
expression will be constitutive for those genes. However, if there is another copy of the lac operon on a plasmid (i.e., in a
merodiploid), the chromosomal operator mutation will have no effect on expression of the structural genes that reside on the
plasmid copy of the operon. Thus, the structural genes on the plasmid will still be repressed by the repressor protein (which is
wild-type) binding to the non-mutant (i.e, wild-type) operator in the plasmid copy of the operon. The operator is considered
"cis-acting" because it only can affect expression of genes physically linked to it on the same DNA.

Conversely, if a cell bears a chromosomal mutation in the gene for the lac repressor such that the mutant repressor protein
cannot recognize a wild-type operator, any wild-type operator will not be repressed by that mutant repressor, and expression
will again be constitutive. However, if a wild-type repressor gene is present on a plasmid copy of the lac operon (i.e., in a
merodiploid), that wild-type repressor will be able to bind to the wild-type operator in the chromosomal copy of the lac
operon (as well as to the wild-type operator in the plasmid copy of the operon). This will change the phenotype from
constitutive to repressed and inducible. The repressor is considered "trans-acting" because the protein made in the cell can act
on any operator in the cell, whether on a chromosome or on a plasmid.
 POLARITY
Nonsense mutation Nonsense mutation
Wild type near 3' end of gene 1 near 5' end of gene 1
Gene 1 Gene 2 Gene 1 Gene 2 Gene 1 Gene 2
DNA X X
mRNA 5'- -3'

Protein 1 Protein 2 Fragment 1 Protein 2 Fragment 1 No product

(active) (active) (inactive) (active) (inactive) (inactive)

Nonsense mutation Nonsense mutation

SuA (rho mutant)
near 3' end of gene 1 near 5' end of gene 1
Gene 1 Gene 2 Gene 1 Gene 2 Gene 1 Gene 2
DNA X X
mRNA 5'- -3'

Protein 1 Protein 2 Fragment 1 Protein 2 Fragment 1 Protein 2

(active) (active) (inactive) (active) (inactive) (active)

Premature termination of translation due to stop codon near 5' end of gene 1 mRNA leads to rho-dependent transcription termination before gene 2 is
transcribed, a result of coupling of transcription and translation. A rho mutant (SuA) allows transcription to continue.
For wild type Rho factor to cause transcription termination, there has to be 90-100 nucleotides of unstructured, unbound RNA for Rho to be able to bind, as a
hexamer. This signals RNA Polymerase to terminate transcription. When a coding sequence contains a nonsense codon, that often provides the conditions (90-
100 nt unstructured unbound RNA) for Rho to bind and terminate transcription. The Rho mutant, SuA, does not terminate transcription even when the 90-100 nt
of unstructured unbound RNA is present.
A nonsense mutation near the 3' end of the upstream gene may not fulfill the required conditions for Rho to act. There may be less than the required 90-100
nt before the next start codon, for example.
Also, these events are probabilities, not absolutes. So while a nonsense mutation 25 codons before the normal termination won't affect the downstream gene,
if the mutation were 50 codons upstream, there may be some effect on the downstream gene. 75 codons upstream will produce an even greater effect on the
downstream gene. This is what is meant by a "gradient of polarity".
Catabolite repression
• Cyclic AMP (cAMP) levels low in glucose,
high in absence of glucose.
• cAMP binds to CAP protein (also called CRP)
which binds to consensus sequence present in
DNA of many promoters to activate transcription;
interacts with α subunit of RNA polymerase.
• Absence of glucose: phosphorylation of factor IIGlc
Inability to make cAMP activates adenyl cyclase, cAMP synthesis.
inhibits CAP binding to CAP • Presence of glucose: non-phosphorylated factor IIGlc
site on DNA and no
blocks transport of some sugars; called
activation of target operons
"inducer exclusion."

A model to account for the ability of glucose to

interfere with the synthesis of cAMP, an action
leading to one component of glucose-mediated
catabolite repression. (A) High-energy phosphate
from phosphoenolpyruvate (PEP) is transferred
through the components of the glucose specific
phosphotransferase system (PTS), a reaction
resulting in the transport and phosphorylation of a
In the absence of glucose, molecule of glucose by the last component, protein
cAMP complexes with CAP, IIGlc. During this process, the amount of the
binds to control regions phosphorylated form of IIGlc is very low, because
of DNA and activates
transcription of target of the rapid phosphorylation of glucose residues; in
operons addition, adenylate cyclase is inhibited by IIGlc; the
cell's inability to make cAMP leads to the
secondary inability to activate catabolite-
repressible operons. (B) The pathways used when
glucose runs out. Adenylate cyclase becomes active
because of the removal of IIGlc; cAMP is formed
and complexes with CAP protein, which can then
bind to the control regions of catabolite-repressible
operons, thereby activating their transcription.
 Phosphotransferase System and Inducer Exclusion
─Glucose +

In absence of glucose, In the presence of

phosphorylated IIGlc glucose, factor IIGlc
activates adenylate LacZ
phosphorylates
cyclase to make cAMP,
glucose to glucose-6-
which complexes with
phosphate (G6P),
the cyclic-AMP receptor
protein (CRP) to activate which is then further
transcription of several metabolized to
catabolic operons, phosphoenolpyruvate
including the lac operon. and the Krebs cycle
Lactose is now able to (not shown). Non-
enter the cell, is phosphorylated IIGlc
converted to the true blocks entry of
inducer, allolactose, by several other sugars
β-galactosidase, which including lactose; this
then inactivates the lacI is called "Inducer
repressor to turn on Exclusion".
transcription of the
operon.

Modified from Figure 5 in "Kazuyuki Shimizu, 'Metabolic Regulation of a Bacterial Cell System with Emphasis on Escherichia coli
Metabolism,' ISRN Biochemistry, vol. 2013, Article ID 645983, 47 pages, 2013. doi:10.1155/2013/645983“.
 E. coli galactose operon
• Dual operators
• Overlapping promoters
• DNA bending & repression

Note: the lac operon was also

subsequently found to have a
secondary operator that
facilitates DNA bending (via
repressor subunit interactions),
to further limit RNA polymerase
access to the promoter under
repressed conditions.
 The Galactose Operon
The galactose operon has two overlapping promoters, P1 and P2, and three structural genes, galE, galT, and galK.
P1 is activated by the catabolite activation system, while P2 does not require cAMP activation. This allows for
enzymes to be present to utilize galactose even in the presence of glucose, and to increase those enzymes in the
absence of glucose (as long as galactose is also present). Both promoters are repressed by a repressor protein that
binds as two dimers to two operator sites on the DNA, one upstream of the transcription start, and one
downstream, within the first coding sequence, of the galE structural gene of the operon. This downstream
operator was discovered when constitutive mutants were found that mapped in the galE gene, which encoded an
altered GalE protein that was still enzymatically active for galactose utilization. In other words, part of the
sequence of the galE gene is doing double-duty, encoding an operator as well as encoding amino acids that are
part of the GalE protein (but not required for the active site or proper folding of the protein).

The purpose of having dual promoters has to do with the cell's need for these enzymes. Galactose enzymes are
required even in the absence of galactose as a carbon source, since galactose is a component of cell wall
synthesis. Therefore, one of the promoters, P2, provides for basal expression of the operon. One of the enzymes
in the gal operon (the galE product) converts UDP-glucose to UDP-galactose for use in cell wall synthesis. This
is why the enzymes need to be made even in the absence of cAMP activation. However, when galactose is
utilized as a carbon source, the other promoter, P1, is activated to provide higher levels of the enzymes.

The two repressor dimers, bound to their respective operators, join to form a tetramer that bends the DNA in a
loop, blocking RNA polymerase access and repressing transcription. This was demonstrated by changing the
spacing between the two operators by 5 base pairs, about a half turn of the DNA helix, which prevented tetramer
association between the two repressor dimers bound to their operators because they were now on opposite faces
of the DNA helix, resulting in a constitutive phenotype.
Arabinose operon of E. coli

• AraC protein is multifunctional: both activator

and/or repressor of arabinose operon
• araC expression autoregulates, i.e., AraC protein
represses transcription of it’s own mRNA
• Arabinose operon expression requires catabolite
activation, i.e., cAMP + CRP (aka CAP), as well
as inducer (arabinose)
• Arabinose operon repression results from
association of AraC protein binding at two sites,
araI and araO2, forming a DNA loop
 Example of autoregulation:
Escherichia coli alanyl-tRNA synthetase (alaRS) gene
AlaRS

AlaRS AlaRS In presence of excess AlaRS,

transcription of alaRS gene
AlaRS
AlaRS is repressed
AlaRS
AlaRS

P O alaRS gene

In absence of excess AlaRS,

AlaRS
transcription of alaRS gene
is active
RNA polymerase mRNA

P O alaRS gene
P = promoter, O = operator

Autoregulation
• Product of a gene acts as a repressor of transcription for the message from that same gene.
• Therefore, when a particular product is in excess, it will repress further synthesis of itself.
• Examples: araC (previous slide); alanyl-tRNA synthetase; lexA (next lecture) and many others.
 Phase variation in Salmonella flagella

Regulation via DNA rearrangements: phase variation in Salmonella flagella.

• Site-specific inversion mechanism; occurs at a frequency of about 1 in 104 per cell division.
• Expression of one of two different flagellar antigens, called H1 and H2.
• Gene for H2 adjacent to 970 bp segment of DNA that can exist in either of 2 opposite orientations.
• In one orientation, promoter for H2 is active; mRNA for H2 and a downstream gene made.
• The downstream gene encodes a repressor protein which represses transcription of the H1 gene;
thus, when the 970 bp region is in this orientation, H2 is made, and H1 is not made (due to
repression of the H1 promoter).
• In the opposite orientation, the promoter for H2 is not active, and mRNA for H2 and the
downstream repressor protein is not made.
• In the absence of repression, the H1 promoter is now active, and H1 protein is produced.
• The orientation of the 970 bp segment is controlled by a flip-flop mechanism similar to
transposition by transposable elements.
• On the 970 bp segment is a gene called hin, the product of which catalyzes site-specific
recombination between the ends of the segment.
• The ends are a pair of inverted complementary repeat sequences, and recombination thereby inverts
the intervening DNA.
 Phase variation in pathogenic pili synthesis Epigenetic regulation
Pap (pyelonephritis-associated pili) of
Regulatory region for pili synthesis uropathogenic E. coli are responsible for
adhesion to the urinary tract. Pap pili
synthesis is controlled by an epigenetic
mechanism that switches between the
presence (ON) or absence (OFF) of pili. The
preferred state is OFF (by >100 fold),
however the pili are necessary for invasion
of the urinary tract.
The pili are encoded by the papBA
operon; transcription is activated by cAMP
DNA methylation and by methylation of the GATC-proximal
site (sequence recognized by Dam).
states and Methylation prevents Lrp from binding to
Lrp binding DNA at sites 1-3, thereby allowing RNA
polymerase to transcribe the operon.
Lrp = Leucine-responsive Pap phase variation model. A model for
regulatory protein transition from phase OFF to phase ON is
shown in panels A–F. In A, the OFF state is
shown in which Lrp binds to sites 1–3,
blocking methylation of GATCprox and
inhibiting papBA transcription. Non-
methylated pap GATC sites are depicted as
open circles, and methylated sites are
depicted as closed circles. The binding of
PAP Lrp at sites 1–3 reduces the affinity of Lrp
phase for sites 4–6 (“mutual exclusion”), and is
depicted by an arrow with a negative sign.
variation Transition to the phase ON state is
model hypothesized to require DNA replication,
which dissociates Lrp from pap DNA
generating two daughter duplexes of which
one is shown (B). PapI and Dam work
together to increase affinity of Lrp for sites
4–6 (C). Activation of pap transcription
requires cAMP-CAP (D), which stimulates
papB transcription (E). Because PapB
activates papI transcription, and PapI
Dam = DNA activates papB transcription, the phase ON
adenine methylase state is self-perpetuating (F). Transition to
OFF could follow further DNA replication.
Adapted from Hernday et al. (2002) Proc Natl Acad Sci USA 99:16470-6
 Plasmid Maintenance Systems (Toxin-Antitoxin "Addiction")

Plasmid addiction. When a plasmid

containing a functional toxin-antitoxin
operon is introduced into a bacterial
cell, both toxin and antidote are
expressed at low levels. Owing to
complex formation between the toxin
and antidote, the cell is protected
against the action of the toxin.
Transcription (or translation) of the
toxin is regulated by the toxin–antidote
complex, which acts as a repressor. As
long as the cell retains at least one
copy of the plasmid, this situation
remains stable. If the plasmid is lost,
however, the system is activated. The
antidote (or antidote RNA) is
continuously degraded by a specific
protease (or nucleases), while the toxin
(or toxin mRNA) remains stable.
Because neither fresh toxin nor fresh
antidote can be produced in the
plasmid-free cell, the toxin is freed. As
a result, the toxin can attack its target
in cells that have lost the plasmid,
thereby inhibiting cell growth and
ultimately killing the cell.

Buts et al. (2005) Trends Biochem Sci 30:672-679

 Plasmid R1 maintenance by antisense RNA

Excerpted from Brantl S (2002) Antisense RNAs in plasmids: control of replication and maintenance. Plasmid 48:165-73.

• The hok-sok system for maintaining plasmid R1: plasmid codes for a poison and an antidote.
• hok gene product is a poison which affects membranes and will kill the host if expressed.
• The mRNA for hok is very stable (persists for at least 40 min).
• sok gene RNA is antisense to hok gene, prevents translation of hok product; thus, an antidote.
• sok RNA is very unstable; if a cell divides and loses the plasmid, it will die because it needs the plasmid to make the sok antidote;
progeny retain the stable killer mRNA for the hok poison.
• Upstream of hok gene is another gene, mok (mediator of killing); mok product is irrelevant.
• Promoter is upstream of mok, and hok can only be made by translational coupling from mok.
• Nonsense mutation in mok destroys the system; but system still works with a frameshift mutation in mok (providing the mutation
doesn't generate a premature stop codon).
• Secondary/tertiary structure of mok-hok mRNA inhibits translation, so expression of hok poison is ordinarily very low; 3' end
processing of the mRNA needed to get any expression.
 Plasmid F maintenance by proteolysis

De Jonge N, Garcia-Pino A, Buts L, Haesaerts S, Charlier D, Zangger K, Wyns L, De Greve H, Loris R. (2009) Rejuvenation of CcdB-poisoned gyrase by an intrinsically disordered protein domain. Mol Cell 35:154-63.
Interaction Diagram for the ccd System. Solid shapes represent structured protein domains, while the thinner lines represent domains in an intrinsically disordered state.
CcdB (blue) converts between two conformations (1), alternately forming the CcdA- or the GyrA-binding surface. The equilibrium between both CcdB conformations is
dictated by the relative affinities of CcdB for GyrA (orange hues) and CcdA (green) and by the concentration of CcdA. When CcdB poisons gyrase (2), a large portion of the
CcdA37-72-binding surface is shielded and only the CcdA65-72 segment can access its binding surface (3). The initial interaction of this segment induces a conformational
change in CcdB, releasing it from gyrase (4). After or coincident with CcdB release, the CcdA37-63 segment locks onto CcdB (5). The resulting CcdB:CcdA complex is the
building block of the repressor complex (6) and forms a chain of alternating CcdA2 and CcdB2 dimers. In this complex, both the high-affinity binding site (H) and the partially
overlapping low-affinity binding site (L) are occupied. Upon depletion of CcdA by Lon-dependent degradation, excess CcdB resolves the repressor complex to form a soluble
hexameric, nonrepressing complex (7), enabling de novo synthesis of CcdA. This results in lower CcdB:CcdA ratios and consequently in rerecruitment of the low-affinity
binding site (8), leading to repression of the ccd operon. When Lon-mediated CcdA degradation outpaces CcdA synthesis (9), CcdB is eventually freed to poison gyrase.

• The ccdA and B system for maintaining F factor: plasmid codes for a poison and an antidote.
• ccdB gene product a poison which targets DNA gyrase; DNA cleaved, host killed if expressed
• ccdB protein is stable, passed on to progeny.
• ccdA gene product blocks action of ccdB poison; thus, an antidote.
• ccdA protein unstable in lon+ cells; lon codes for a protease; if cell divides and loses plasmid, it dies because plasmid needed to make ccdA
antidote; progeny retain stable killer ccdB poison.
• In lon- cells, ccdA product is stable, so plasmid maintenance impaired.
• ccdA and B on one mRNA; both protein products are transcriptional repressors of operon.
 T7 Expression Systems
• Bacteriophage T7 has a highly specific RNA polymerase that has facilitated its use in
commercially available expression systems for cloned genes in recombinant DNA.

• Phage T7 has an ~40,000 bp linear double-stranded DNA which contains

contains 56 genes (encoding 59 proteins). 92% of the genome is coding.

• Genes transferred into the cell in order from 0.3 to 19.5. DNA pulled into the cell by
transcription, slowly initially by host transcription, faster later by T7 transcription.

Direction of DNA entry into the host

Rho-dependent terminator T7 terminator

Early T7 promoters T7 genes

promoters in green numbered
in blue

T7 genome

RNase III cleavage sites shown below genome

1st gene 0.3 inhibits class 1 restriction enzymes
Promoters for T7 mRNAs
 Mechanism of T7 Expression Systems

cloned gene expressed

following induction of
T7 RNA polymerase cloned gene

T7 promoter
(23 nucleotides)
plasmid containing
T7 RNA polymerase recombinant gene

inducible by IPTG ampicillin-

resistance
gene

Lac repressor

host chromosome
 Notes on the mechanism of T7 expression systems
• The gene for T7 RNA polymerase is integrated into the chromosome of an E. coli host cell

• The T7 RNA polymerase gene (coding region only) is under control of the lac promoter

• Therefore, the T7 RNA polymerase is only made in large quantities following induction (by IPTG)

• The recombinant gene for a protein of interest is cloned on a plasmid under control of a T7 promoter

• The plasmid encodes an ampicillin drug-resistance gene to allow selection and retention of plasmid

• The cloned gene is only expressed when T7 RNA polymerase has been synthesized

• Because T7 RNA polymerase is so specific (23 nt recognition), there is no expression of other genes

• The only leakiness in the system comes from basal transcription of lac promoter to make T7 RNA polymerase

• This can be minimized by adding a compatible plasmid to the cell containing the gene for T7 lysozyme

• T7 lysozyme has been found to inhibit the activity of T7 RNA polymerase (not shown on the previous slide)

• Therefore, any basal (leaky) expression of T7 RNA polymerase can be inhibited by the lysozyme

• Only under robust induction with IPTG is there enough active T7 RNA polymerase to express the cloned gene

• Extremely large quantities of a protein of interest can be synthesized in this system

A full description of T7 expression systems can be found in Studier et al. (1990) Methods in Enzymology 185:60-89