Module 5

Associated papers:

Gardner et al. (2000), “Construction of a genetic toggle switch in E. coli,” Nature

403:399

Elowitz and Leibler (2000), “A synthetic oscillatory network of transcriptional

regulators,” Nature 403:335
 What is synthetic biology?

• Bioengineering by a different name.

• Bioengineering enabled by large-scale gene DNA synthesis.

• Bioengineering enabled by large-scale gene DNA synthesis

and a modular approach to design.

• Chemical biology.
 What is Moore’s Law?
"Moore's law" is the observation that, over the history of
computing hardware, the number of transistors in a dense
integrated circuit doubles approximately every two years.
The observation is named after Gordon E. Moore, co-
founder of the Intel Corporation, who first described the
trend in a 1965 paper[1][2][2][3] and formulated its current
statement in 1975. His prediction has proven to be accurate,
in part because the law now is used in the semiconductor
industry to guide long-term planning and to set targets for
research and development.[4]

http://www.synthesis.cc/2013/04/updated-
dna-cost-and-productivity-curves-plus-a-few-
more-thoughts-on-moores-law.html
Maybe a more intuitive way to look at it
http://www.synthesis.cc/cgi-bin/mt/mt-
search.cgi?blog_id=1&tag=Carlson%20Curves
And here’s what that might mean to you ….
 Synthetic Biology

Module functionality in tectons

Ref: Bromley EHC, Channon K,

Moutevelis E et al. Peptide and
protein building blocks for
synthetic biology: from
programming biomolecules to
self-organized biomolecular
Synthetic biology space systems. ACS Chem. Bio. 3(1): 38-
50
The extreme version on the previous slide
suggests that everything can be engineered
rationally; that we can basically design from
molecule to ecosystem.

Suggest a limitation to this dream? Why won’t

biological pieces necessarily fit together in a
rational way?
However, I like this view better:
It boils down to this:
How do you make things?

Synthetic biology is basically Lego

What might be aspects of modularity
in bioengineering? What can be used in a
modular way?

• Part -- What are examples?

• Circuit – What are examples?

• Chassis – What are examples?

• Environment – What are examples?

What are parts for a transcription circuit?

How do you program these parts?

What other parts?

Why is a terminator useful?

 Overall structural architecture of T7 RNA polymerase

- A polymerase domain and an N-terminal domain.

The polymerase domain resembles a right hand with
palm, thumb and finger sub domains.
v Thumb domain: Very flexible, functions to stabilize
the ternary complex during transcription by wrapping
around the template.
v Palm domain: Trio of b-strands (conserved in
all nucleic acid polymerases) Contains the catalytically
most critical amino acids (Asp 537 and Asp 812)
v Fingers domain: Specificity loop (unique to T7 RNAP)

N-terminal domain: Gives the characteristic concave shape for the enzyme.
Involved in promoter recognition and DNA melting during transcription initiation.

Active site: Located in the deep pocket bound by fingers, thumb, palm and N-terminal domain.
Like most everything else in biology, the job of T7 RNA
polymerase is to help its gene / genome (the bacteriophage)
survive. Incidentally, this means that it actually *doesn’t*
want to talk to the cell, which it is in the process of blowing
up. Here’s what the genome of the T7 bacteriophage looks
like:

Promoters

Proteins
This set of promoters actually allows the T7 bacteriophage to do
an ordered ‘dance’ of transcription that results in just the right
genes being made and translated at just the right times in the
viral life cycle … without any interference by its prey, E. coli:
 Promoter specificity

*Cheetham, G.M.; Jeruzalmi, D.; Steitz, T. Nature 1999, 399, 80-83.

In general, how do proteins recognize nucleic acids?

• Is recognition primarily in the major groove, or the minor? Why?

• What is an ‘arginine fork?’

• Draw a picture of how glutamine would recognize adenosine

 Residues in T7 RNAP important for sugar recognition
2' OH discriminators

Tyr 639
BASEN BASE(N+1)
H O His 784
OH O
O O H N
N
O_
5' RNA O O
OH Amino acids
transcript O
2+
Pa O
O
incoming
ribonucleotide
R425, G542 ,Y639, H784
Mg O P
O 2+ O
Mg O Pyrophosphate
O leaving group
Asp 537 O O
P
O O
O_

Asp 812

Comparison of T7 RNAP and T7 DNAP

active sites

in blue/red:- T7 RNAP residues

In black:- T7 DNAP residues
To overcome the idiosyncracies of gene expression, it is sometimes
useful to have a *universal* means of producing proteins, a method
whereby you can be reasonably sure that the protein will get made
no matter what organism you put it in. There are currently no
universal methods that work across the broad domains of life, but
there are some methods that will largely work within a domain
(eubacteria, eukaryotes).

It is also useful if the methods used are *orthogonal* to the host in

question. By this, we mean that the machinery will interact with
itself, but not so much with other aspects of metabolism in the
organism into which it is introduced. We try to avoid *cross-talk*
between the introduced machinery and the host machinery, except
where such cross-talk is desired.
You’ve seen this before, in Module 4, when we looked at the ability
to alter translation and the genetic code. For example, the M.
janaschii tyrosyl tRNA synthetase was introduced into E. coli
precisely because it didn’t know how to use any E. coli tRNAs (for the
most part); conversely, the M. janaschii tyrosyl tRNA cannot be
charged by E. coli tRNA synthetases (for the most part). Thus, the M.
janaschii tyrosyl tRNA synthetase:tRNA pair was *orthogonal* to the
remainder of E. coli’s second genetic code. This allowed it to be
commandeered for the incorporation of unnatural amino acids
across from stop codons (by changing both the tRNA synthetase to
charge unnatural amino acids, and the tRNA to be a suppressor
tRNA).

That said, keep in mind that the orthogonal tRNA synthetase:tRNA

pair is *not* orthogonal to the first genetic code: the cell’s mRNAs,
ribosome, and so forth will still be used.
Now, the question becomes, how can we make a universal, but
orthogonal, transcription apparatus? To think about this, answer the
following questions for yourself:

Should the introduced machinery be able to transcribe the cell’s

genes?

Should the genes for the introduced machinery be able to be

transcribed by the cell’s machinery?

Should the products of transcription still be able to be used by the

translation apparatus of the cell?
What do you think you’ll need for universal, orthogonal transcription
machinery?

A new polymerase? Yes!

And this polymerase should interact with … a new promoter

sequence!

And perhaps new, modified nucleotides, like those we saw in

Module 3?

No! Because then the ribosome would not know what to do with
the mRNAs that were made (although an entirely orthogonal first
*and* second genetic code based on modified nucleotides would be
totally cool!).
Parts aren’t’ just what nature gives us, they can be engineered: Letter

pubs.acs.org/synthbio
Meyer et al. (2015)
Directed Evolution of a Panel of Orthogonal T7 RNA Polymerase
Variants for in Vivo or in Vitro Synthetic Circuitry
Adam J. Meyer,† Jared W. Ellefson,† and Andrew D. Ellington*,†,‡,§
†
Institute for Cellular and Molecular Biology, ‡Department of Chemistry and Biochemistry, and §Center for Systems & Synthetic
Biology, University of Texas at Austin, Austin, Texas 78712, United States
*
S Supporting Information

ABSTRACT: T7 RNA polymerase is the foundation of

synthetic biological circuitry both in vivo and in vitro due to
its robust and specific control of transcription from its cognate
promoter. Here we present the directed evolution of a panel of
orthogonal T7 RNA polymerase:promoter pairs that each
specifically recognizes a synthetic promoter. These newly
described pairs can be used to independently control up to six
circuits in parallel.

KEYWORDS: directed evolution, transcription, genetic circuits

T
What are they (we) trying to do? Create orthogonal
he field of synthetic biology relies on designed, synthetic
circuits made from genetic parts that function robustly
T7 RNAP specifically recognizes its promoter primarily
through the specificity loop (residues 739 to 770), especially by
polymerase:promoter combinations (new enzymes:new DNA
and orthogonally to the cell’s operating system. All too often,
synthetic parts are evolved, engineered, or mined to perform a
interactions made between residues R746, N748, R756, and
Q758 with the major groove from −7 to −11.18 Mutations to

substrates) that could be used for orthogonal transcription (as we

particular function, but are ultimately unusable due to their
weak activity or promiscuity. As parts are combined to form
this critical part of the promoter usually results in substantial
loss of promoter recognition.19 Promoters with single

discussed several slides back)

complex genetic circuits, each suboptimal part undermines the substitution that are weakly recognized by T7 RNAP can be
whole. To generate parts that behave as expected, it may be recognized by T7 RNAPs with single mutations to N748,20
advisible to start with a well-studied and well-understood part, R756,19 or Q758.21
and tailor its function to the needs of the community.1−4 Previous attempts to create T7 RNAP mutants capable of
How are we going to create new polymerase:promoter
combinations?

The promoter is easy, we just change the basic sequence of the T7

RNA polymerase promoter. We just make up our minds and design it
and synthesize it.

But then we have to change the polymerase to recognize this new

promoter! That’s the hard part.

We’ll use what’s called directed evolution, laboratory evolution

methods. We’ll randomize the portions of the polymerase
responsible for recognition of the promoter (see: a few slides back)
and *select* for those that recognize the new promoter.
Natural selection and directed evolution are really the same thing:
a battle between genotypes for dominance. Or, in more scientific
parlance: evolution is the change in frequency of alleles over
time. That’s it. It doesn’t particularly matter what drives the
change in frequency of alleles, but when it can be coupled to
phenotype, then we get the more familiar: survival of the fittest.

(Note that there are amusing subtleties between: “change in the

frequency of alleles over time” and “survival of the fittest”).

To have survival of the fittest you have to have individuals; or to

go full Mad Max Beyond Thunderdome on you: “Two DNAs in,
one DNA out.”

We’re actually going to do many DNAs in, some DNAs out.

ACS Synthetic Biology Letter

region. T7 RNAP variants capable of recognizing the synthetic

promoter should therefore selectively produce Taq DNAP
protein and, in turn, be amplified by the Taq DNAP during
PCR.
We• firstDifferent
sought to alterpolymerase
the molecular variants
recognition of(different
T7
sequences of amino acids, in the
RNAP to drive transcription from the novel promoter PCTGA.
This promoter was chosen because it is different from the wild-
recognition
type T7 promoter at all four loop) are
of the most transformed
critical base pairs (−11 into
to −8) and also different from other known orthogonal
promoters a (Pcell and expressed.
CGG, PT3, PK1F, and PN4) at two or more of those
positions.
We generated a library of T7 RNAP variants in which the six
amino Each
• acids mostcell expresses
responsible a different
for promoter
18−21
recognition (R746,
L747, N748, R756, L757, and Q758) were completely
randomizedpolymerase (cloning
by using oligonucleotides is a ‘single
synthesized using a mix
of trimermolecule’
phosphoramidites. technique).
24
The resulting library (theoret-
ically 20 or 6.4 × 10 in complexity) was transformed into E.
6 7

coli at about 2-fold coverage. Transformed cells were then

grown, induced, and emulsified. Upon thermal cycling of the
• The
emulsified cells
reaction, cellsare
wereensconced
lysed and emulsion (look it up)
PCR led to in a
selective water-in-oil emulsion.
amplification of functional If you
specificity don’t
determining
regions. The emulsion was broken with chloroform, and
know
biotinylated what
amplicons werethis is, goby make
recovered some
streptavidin bead- salad

by overlapdressing
PCR, agarose from oil and
gel-purified, water.
based purification. Full-length T7 RNAP genes were assembled
digested with restriction
endonucleases, and ligated into an expression plasmid for
further rounds of selection. The library began to converge on
The polymerases
• sequences
active after four rounds of in cellsand
selection, in after
turn drive
seven
Figure 1. Compartmentalized partnered replication selection scheme.
(Top) E. coli cells containing the Taq DNA polymerase gene under the production of Taq polymerase, a
rounds the library was dominated by a single variant (termed
CTGA-R7-1, Supplementary Table 2) that had the sequence
the control of a synthetic T7 promoter variant are transformed with a
saturation (or error-prone) mutagenesis library of T7 RNAP mutants. thermostable
L747I, N748T, R756T, and Q758K DNA polymerase,
(Supplementary under
Figure 1).
In order to further ascertain the functionality of CTGA-R7-1,
Variants capable of recognizing the promoter produce Taq DNA
polymerase protein. (Middle) Whole E. coli cells are compartmental- the control
we transformed an expressionofconstruct
a novel for theT7polymerase
RNA into
ized in a water-in-oil emulsion. The aqueous droplets also contain an E. colipolymerase promoter.
strain that also carried a GFP gene coupled to the
primers, dNTPs, and Taq DNA polymerase buffer. (Bottom) PCTGA promoter (Supplementary Figure 1). CTGA-R7-1
Emulsions are thermal cycled, leading to E. coli cell lysis and in vitro produced about 1% as much GFP from PCTGA as the wild-
PCR amplification of the T7 RNAP genes that led to the production of type T7 RNAP produced from PT7. This weak activity was in
Taq DNA polymerase during the in vivo expression step. PCR product line with the generally low efficiencies seen by others for
is recovered and prepared for the next round of selection.
specificity variants.24 We hypothesized, however, that further
mutations in the regions flanking the promoter specificity
ACS Synthetic Biology Letter

region. T7 RNAP variants capable of recognizing the synthetic

promoter should therefore selectively produce Taq DNAP
protein and, in turn, be amplified by the Taq DNAP during
• So … individual cells do or don’t express
PCR.
We first sought to alter the molecular recognition of T7
RNAP to Taq driveDNA polymerase
transcription from the novelbased
promoter onPCTGA
how .
goodwas
This promoter the T7because
chosen RNAPit isvariant is at
different from the wild-
type T7 promoter at all four of the most critical base pairs (−11
to −8)recognizing
and also different a promoter.
from other known orthogonal
promoters (PCGG, PT3, PK1F, and PN4) at two or more of those
positions.
Emulsions
•We generated a librarycan
of T7 be
RNAP put through
variants PCR.
in which the six The
amino acids most responsible for promoter recognition (R746,
emulsion
L747, N748, R756, L757,‘bubble’
and Q758) stays
18−21 intact … but the
were completely
cell inside … blows up …library
releasing
randomized by using oligonucleotides synthesized using a mix
of trimer phosphoramidites. 24
The resulting (theoret-… the
ically 20thermostable
6 7
polymerase!
or 6.4 × 10 in complexity) was transformed into E.
coli at about 2-fold coverage. Transformed cells were then
grown, induced, and emulsified. Upon thermal cycling of the
• If the emulsion (again, think shaking up
emulsified reaction, cells were lysed and emulsion PCR led to
selective amplification of functional specificity determining
regions.oil Theand water
emulsion was for a salad)
broken includes
with chloroform, and
biotinylated amplicons were recovered by streptavidin bead-
primersFull-length
based purification. for amplifying
T7 RNAP genes thewereT7 RNA
assembled
polymerase
by overlap gene …digested with restriction
PCR, agarose gel-purified,
endonucleases, and ligated into an expression plasmid for
further rounds of selection. The library began to converge on
active sequences after four rounds of selection, and after seven
Figure 1. Compartmentalized partnered replication selection scheme. • …
rounds the then each
library was cell /bybubble
dominated either
a single variant (termed
(Top) E. coli cells containing the Taq DNA polymerase gene under
the control of a synthetic T7 promoter variant are transformed with a amplifies, or doesn’t, its own T7 RNA
CTGA-R7-1, Supplementary Table 2) that had the sequence
L747I, N748T, R756T, and Q758K (Supplementary Figure 1).
saturation (or error-prone) mutagenesis library of T7 RNAP mutants.
Variants capable of recognizing the promoter produce Taq DNA polymerase
In order variant
to further ascertain geneofbased
the functionality on how
CTGA-R7-1,
we transformed an expression construct for the polymerase into
polymerase protein. (Middle) Whole E. coli cells are compartmental-
ized in a water-in-oil emulsion. The aqueous droplets also contain
well
an E. coli strainthe
that T7
also RNA
carried polymerase
a GFP gene coupled variant
to the
primers, dNTPs, and Taq DNA polymerase buffer. (Bottom) PCTGA protein worked in transcription!
promoter (Supplementary Figure 1). CTGA-R7-1
Emulsions are thermal cycled, leading to E. coli cell lysis and in vitro produced about 1% as much GFP from PCTGA as the wild-
PCR amplification of the T7 RNAP genes that led to the production of type T7 RNAP produced from PT7. This weak activity was in
Taq DNA polymerase during the in vivo expression step. PCR product line with the generally low efficiencies seen by others for
is recovered and prepared for the next round of selection.
specificity variants.24 We hypothesized, however, that further
mutations in the regions flanking the promoter specificity
 To remind you: this is PCR

Note that the two primers are *different.* You need one for each strand. You’re going to be
designing your own primers soon. The ‘in between’ here would have been the T7 RNA
polymerase gene (or a portion thereof that could been inserted into a partial T7 RNA
polymerase gene
 further rounds of selection. The library began to converge on
active sequences after four rounds of selection, and after seven
Figure 1. Compartmentalized partnered replication selection scheme. rounds the library was dominated by a single variant (termed
• The amplified DNA can be re-cloned into cells, the cells re-
(Top) E. coli cells containing the Taq DNA polymerase gene under
the control of a synthetic T7 promoter variant are transformed with a
CTGA-R7-1, Supplementary Table 2) that had the sequence
L747I, N748T, R756T, and Q758K (Supplementary Figure 1).
saturation (or error-prone) mutagenesis library of T7 RNAP mutants.
emulsified … and the competition between ‘good’ DNAs (with
Variants capable of recognizing the promoter produce Taq DNA
polymerase protein. (Middle) Whole E. coli cells are compartmental-
In order to further ascertain the functionality of CTGA-R7-1,
we transformed an expression construct for the polymerase into
good polymerase genes) and ‘bad’ DNAs (with less good
ized in a water-in-oil emulsion. The aqueous droplets also contain
primers, dNTPs, and Taq DNA polymerase buffer. (Bottom)
an E. coli strain that also carried a GFP gene coupled to the
PCTGA promoter (Supplementary Figure 1). CTGA-R7-1
polymerase genes) can continue.
Emulsions are thermal cycled, leading to E. coli cell lysis and in vitro
PCR amplification of the T7 RNAP genes that led to the production of
produced about 1% as much GFP from PCTGA as the wild-
type T7 RNAP produced from PT7. This weak activity was in
Taq DNA polymerase during the in vivo expression step. PCR product line with the generally low efficiencies seen by others for
is recovered and prepared for the next round of selection.
specificity variants.24 We hypothesized, however, that further
• After multiple cycles of selection and amplification, you can find
of T7 RNAP we generated a strain of E. coli in which the
mutations in the regions flanking the promoter specificity
determinants would provide substantive improvements in
individual protein variants (out of libraries of millions) that are
production of Taq DNAP was dependent on the use of a
synthetic T7 promoter. Individual cells were compartmental-
activity. We therefore subjected a larger region encompassing
most of the “fingers” domain18 (amino acid residues 663−793)
really good at the new promoters:
ized in a water-in-oil emulsion with 5′-biotinylated primers for
PCR amplification of the T7 RNAP specificity determining
to error-prone PCR (at a rate of two mutations per kb) and to
CPR optimization. After six further rounds of selection (with

Table 1. Summary of the Most Active T7 RNAP Mutantsa

activity on
mutant sequence cognate promoter cognate
WT WT PT7
TAATACGACTCACTATA
100.0
1
CGG-R12-KIRV Q744K, L747V, N748H, L749I, R756E, L757M, H772R, E775V PCGG
TAATACCGGTCACTATA
50.5
2
CTGA-R13-
AKSRV
V725A, Q744K, L747I, N748S, R756T, Q758K, H772R, E775V PCTGA
TAATACCTGACACTATA
42.9
3
T3-R5-RRVH T745K, N748D, L749M, M750I, G753R, H772R, E775V, Q786H PT3
TAATAACCCTCACTATA
160.3 4
K1F-R5-KIKR Q744K, L749I, M750K, Q754S, R756N, I761V, H772R PK1F
TAATAACTATCACTATA
76.2 5
N4-R5-YRNRV N671Y, L747I, N748D, L749C, M750V, F751I, Q754T, F755R, L757M, Q758A, P759L,
D770N, H772R, E775V
PN4
TAATAACCCACACTATA
25.5 6
a
A summary of the most active variant for each promoter. The “activity on cognate” value is the average of three different experiments on three
different days (each in triplicate). All values are normalized to wild-type (WT) T7 RNAP activity as 100.

1071 dx.doi.org/10.1021/sb500299c | ACS Synth. Biol. 2015, 4, 1070−1076

Special question: could you imagine some way in which you could set
up an entirely separate, entirely orthogonal translation apparatus, so
that only the RNAs produced by your introduced transcription
machines would be used by your introduced translation machines?
How would that work?
For circuit (as opposed to part) engineering, you need to know
just a bit more about the T7 RNA polymerase promoter, and how
it works with its corresponding polymerase:

Note that the polymerase needs to initiate (efficiently) with

several guanosines, which effectively become part of the
promoter.
 Keep in mind reaction coordinate diagrams! Because that’s what
again underlies transcription initiation. So, just for fun:

• Initial contact is called the ‘closed complex’

• Then the polymerase opens a DNA loop, leading to the ‘open complex’
• The open complex then initiates transcription, with those guanosines
• The open complex must segue to an elongation complex that can read a whole gene;
this is inefficient
• Imagine what this would look like on a reaction coordinate diagram (may show up on
the final)
 How do you regulate a transcription circuit?

Promoter Operator

What is the function of the operator?

What binds to it? How is it regulated?

Polymerase Repressor

How do you program the operator?

Repressors & Activators
What about translation, how do you program that?

Shine-Delgarno
sequence

How do you program translation initiation?

Why are the complementary sequences different?

What aspect of transcription initiation does this remind you of?

What are the parts here?
What will this circuit do if put into a cell?

T7 RNA T7 RNA polymerase

polymerase
promoter

What is wrong with this explanation?

What context is this occurring in?

What is a plasmid?
What are the essential elements here?
Are these elements modular?
What would be the difference between the function
of the autogene on a plasmid versus in the genome?
Traditional cloning with restriction enzymes
Or … overlap any two (or more) fragments with homologous ends
via so-called Gibson assembly
With multiple overlaps, Gibson assembly can be used to quickly
create inserts for plasmids
Or, you can do the best of both worlds: a restriction system
that allows you to quickly assemble virtually any ends.

This looks like restriction cloning … what’s the difference?

You need a special kind of restriction site: so-called Type IIS
restriction sites:

Unlike normal restriction enzymes (Type II) that recognize a

particular sequence (often a palindrome, why?) and cleave
within it, Type IIS enzymes recognize a sequence and cut
whatever sequence is adjacent (in a particular, polar
direction), no matter what it is. Recognize near, cleave far.
 Here are some examples of Type IIS restriction enzymes:

For the sequence 5’ GAAGACGGTCCATGA, what happens when this is

cut with BbsI?
 The advantage of the so-called “Golden Gate” cloning strategy is it
allows you to have libraries of ‘parts’ with slightly differing
functions. Each of the parts will have a particular ‘sticky end’ for
cloning, allowing them toResearch
come Article together in an organized way:

lay
and

ogy
the What constitutes an
ade
ast- expression unit, in this
tep
ign diagram? What is
to
any minimally needed in order
For
set to express a protein?
na
oin
the
wly
an
tly,
in
of
heir
to
ple
zes
n a
nce
This modularity has in turn allowed researchers to construct useful
(and usable!) ‘toolkits’ of parts, that can be further built into
circuits.
ACS Synthetic Biology Research Article

This is the *yeast*

toolkit.

What parts might be

different for a
bacterial toolkit?
Definition of Part Types
There are eight primary part Types in our assembly standard and three of those have options
to split into Subtypes. It should be noted that Types are technically defined only by their flank- Here’s how everything
ing overhangs, and the contents need not necessarily match the biologically-defined functions
we describe (Supplementary Figure S6). However, in order to ensure compatibility between
might fit together in a
all parts designed by all researchers, we recommend that new parts be designed to match the plasmid; note that the
Types defined here.
‘sticky’ ends connecting
modular parts are
different from each other.

This means that the

pieces can be mixed
together all at once and
ligated! Very easy multi-
part constructions!

Supplementary Figure S6
The different parts for any of these cloning methods can
come pre-made (ala the Golden Gate kit), or you can add
in your own parts via PCR with appropriate
oligonucleotides.
Your gene

Similarly, you can put in a

new RBS; you’d just have
Your
Your gene to add it to your primer …
promoter
and start downstream of
the native RBS

Red regions: your type

Your IIS restriction site for
promoter Your gene
Golden Gate cloning –
not just the sticky end,
but everything needed
for amplification and
cleavage
Now, on to actual circuits! We’ll be going over two papers:

Gardner et al. (2000), “Construction of a genetic toggle switch in

E. coli,” Nature 403:399

and

Elowitz and Leibler (2000), “A synthetic oscillatory network of

transcriptional regulators,” Nature 403:335

To a first approximation, these are the ‘founding’ papers for the

field of synthetic biology.
 What will be the function of this circuit?

Gardner et al., Nature 403, 339-342(20 January 2000)

How is this configuration similar to what
you saw in a linear format? What is R1?
What are the
parts here?

* *
Examine the previous slide.

There are four basic types of repressors:

Which type is the

Lac repressor?
How is this like LacI or not like LacI?
The master control (sort of)
What does this tell us about the repressor / cro binding site?
How does the repressor / cro recognize its binding sequence?
Think about how this somewhat complex circuit must work.

The transcription unit (gene) for cI is to the left.

The transcription unit for Cro (gene) is to the right.

They are essentially battling one another for control, via

competing for inhibiting one another’s transcription and
expression.

But keep in mind things are dynamic, always changing in the cell!
The circuit *maintains* itself, but if perturbed can reach a new set
state (for example … UV light)
a b
PflMI -35 -10 SD
A AGGAGGAAAAAAATG
Ptrc-2 CCATCGAATGGCTGAAATGAGCTGTTGACAATTAATCATCCGGCTCGTATAATGTGTGGAATTGTGAGCGGATAACAATTTCACACAGGA
B AGGAATTTAAATG
Olac
C AGGAAACAGACCATG
SphI -35 -10 SD
D AGGAAACCGGTTCGATG
PL-s1con GCATGCACAGATAACCATCTGCGGTGATAAATTATCTCTGGCGGTGTTGACATAAATACCACTGGCGGTtATAaTGAGCACATCAGCAGG//GTATGCAAAGGA
E AGGAAACCGGTTATG
OL3 OL2 OL1
F AGGACGGTTCGATG
SphI -35 -10 SD G AGGAAAGGCCTCGATG
PLtet0-1 GCATGCTCCCTATCAGTGATAGAGATTGACATCCCTATCAGTGATAGAGATACTGAGCACATCAGCAGGACGCACTGACCAGGA
H AGGACGGCCGGATG
Otet2 Otet2

What is different between these promoter ‘parts?’

 “Although all of the toggle plasmids contain the same
configuration of elements, one plasmid, pIKE105, does
not exhibit bistability. This result is probably due to the
reduced efficiency of the Tet repressor relative to the
lambda repressor (cI). To maintain bistability, the reduced
efficiency requires a corresponding decrease in the strength
of the pLtetO-1 promoter relative to the pLsIcon promoter.”
a b
PflMI -35 -10 SD
A AGGAGGAAAAAAATG
Ptrc-2 CCATCGAATGGCTGAAATGAGCTGTTGACAATTAATCATCCGGCTCGTATAATGTGTGGAATTGTGAGCGGATAACAATTTCACACAGGA
B AGGAATTTAAATG
Olac
C AGGAAACAGACCATG
SphI -35 -10 SD
D AGGAAACCGGTTCGATG
PL-s1con GCATGCACAGATAACCATCTGCGGTGATAAATTATCTCTGGCGGTGTTGACATAAATACCACTGGCGGTtATAaTGAGCACATCAGCAGG//GTATGCAAAGGA
E AGGAAACCGGTTATG
OL3 OL2 OL1
F AGGACGGTTCGATG
SphI -35 -10 SD G AGGAAAGGCCTCGATG
PLtet0-1 GCATGCTCCCTATCAGTGATAGAGATTGACATCCCTATCAGTGATAGAGATACTGAGCACATCAGCAGGACGCACTGACCAGGA
H AGGACGGCCGGATG
Otet2 Otet2

Basically, in a repressor competition, there has to be a way for both

repressors to be clear winners. How might cooperativity help?
 The bistability of a toggle is its feature; but it’s not guaranteed

In this Slide, I’m

asking you to
interpret
diagrams that
you have no idea
of, except for
what you can
read. Can you
figure these out
for yourself?
What are
these
values?
Read the
paper

What might cause the monostable states (see also next slide)?
 a
Let’s focus on pTAK117, from 1 IPTG 42°C pTAK117

Normalized GFP expression

pTAK130
pTAK131

the previous Table.

pTAK132
0
1 pTAK102
(control)

Redraw pTAK117 in the style 1 pTAK106

(control)

of Slide #23, but linear (see 0

0 5 10 15 20

also your exercise sheet). For b

1 IPTG
Hours
aTc pIKE107

Normalized GFP Expression

pTAK117, what is R1? What is 0
pIKE105

P1? 1 pTAK102
(control)
!
*%
0 '$
1 pIKE108 &*
Adjacent, look at the
(control)
&'
0 ,
performance of pTAK117: it c
0 5 10
Hours
15 20
2
0
toggles ‘on’ and stays ‘on’ in IPTG

GFP expression
1,600
8
,
the presence of IPTG; it 800
&'
No IPTG
toggles ‘off’ and stays ‘off’ in
46
0
'1

the presence of heat (normal 0 10

Hours
20 30 "
'(

growth temperature is 37 !"#$%& ' !"#$%&'()'*$% $+ ,*&'),*-*'./ 01" 2(". &1)3*%2 *%3*4)'"& 5"(*$3& $+ 41"#*4)- $(
'1"(#)- *%364'*$%/ 01" -*%"& *% ( )%3 )7 81*41 )(" )55($9*#)'*$%& $+ '1" &8*'41*%2
-$
1
degrees C.). Why (see 3.%)#*4&7 )(" *%4-63"3 +$( 4-)(*'./ (7 50:; '$22-" 5-)&#*3& )%3 4$%'($-&/ )7 5<;= '$22-"
5-)&#*3& )%3 4$%'($-&/ *7 !"#$%&'()'*$% $+ '1" -$%2>'"(# &'),*-*'. $+ '1" &"5)()'"
(
("
worksheet)? "95("&&*$% &')'"& *% 50:;??@/ 3
Red dots = pTAK117;
Blue dots = pTAK102

What is going on
Here/?

What is FACS/?

How is the
information
obtained by
FACS different?
 How does
this differ
from Type IV?
P1 Ptrc-2 P1 Ptrc-2
RBS1 rbs E RBS1 rbs E
GFPmut3
3/ R1
lacI
GFPmut3

T1T2 Type II T1T2 T1T2 Type III T1T2

*
Design a circuit to express GFP only when there was *no*
IPTG in the media.
 How does RecA
impact cI?

Based on
this figure,
what do you
suppose
TraA does?

What is happening in this circuit?

Kobayashi et al. (2004) PNAS 101:8414
What is this circuit
doing?

What controls GFP

production?

What is the role of

AHL/?
*

What kind of protein

is LuxR/?

What does the thermal Why does the toggle decline in the
shift do? absence of inducer (*)? Isn’t that
counter to what it’s supposed to do/?
Why would such a system be useful?
“The partial decrease in fluorescence observed 12 h
after removal of IPTG reflected the relaxation from a
state where LacR is completely inactive to a stable state
where cI is the dominant repressor, but LacR still has
some basal activity. The stability of distinct expression
states was confirmed in a separate control experiment
where stable expression was observed for up to 50 h
(correpsonding to 50-60 generations) after the removal
of the inducing factor.”
 What will this circuit do?
!"##"$% #& '(#)$"
a Repressilator Reporter -, !<2 "5 =8
9:=:3@8> 18
PLlac01
K6435@L8A2
ampR
tetR-lite #D8 5@;8
PLtet01 M@C2 .2 #8;
kanR :B874== @39
TetR TetR
pSC101 gfp-aav ?Q, ;@3658
λPR
origin 5@;82 #D8 4
λ cI LacI GFP =8B8=> :E S
8VD@R@5 :>9
lacI-lite
ColE1 A8587;@38A
λ cI-lite 743C8 :E N8
PLtet01 @3587B4=>I @
SMT =8B8=>
:38 43:5D8
434=H>@> :E
4B874C8 D4=
b =:3C87 5D43
9:3A@5@:3>
β

steady state
 =@G8 5: A@>5@3C6@>D >69D >5:9D4>5@9 8EE895> E7:; N:>>@R=8 @357@3>@94==H
9:;N=8V AH34;@9> O>69D 4> @3587;@558398 :7 9D4:5@9 R8D4B@:67P2
OnM675D87
a cell-by-cell basis,5:it@A835@EH
>56A@8> 478 388A8A oscillates. This is5D8
43A 9D4749587@`8 actually
>:6798> what in
:E J695645@:3> @3 5D8 78N78>>@=45:7 43A :5D87 A8>@C38A 3851:7G>2 ^3
EEN475@96=47I
terms is=:3C87
called a ring oscillator.
8VN87@;835> N87E:7;8A 63A87 Why is EE terminology
9D8;:>545@9 9:3F
good for>D:6=A
A@5@:3> synthetic biology?
834R=8 ;:78 9:;N=858Why might
>545@>5@94= it be limiting?
9D4749587@`45@:3 :E

0 1,000
Time (min)
60 140 250 300 390 450 550 600
a
0 1000
min)
b

52, &,3&,$$)0.%"& 120

"1 %2&,, &,3&,$$"& c
%2, (,#%&, "1 %2, 100
(arbitrary units)
Fluorescence

2%08 &,3&,$$);0, 80
%2, &)-2% 3&"/"%,&
60
,+'(,+ ;8 %2,
30.$/)+ C&)-2%E 40
.$/)+$* %&.#$(&)3: 20
"/ %2, ,- #.!& //01
,3&,$$)0.%"& /"+,0 0
0 100 200 300 400 500 600
.+8 $%.%, )$ $%.;0, Time (min)
.&),$ ;,%6,,# %2,
Represent the function of the repressilator …
via FACS. What would this look like? Why?

Could you draw the repressilator such that

each repressor had its own ‘output,’ in the
sense of being a different colored reporter?
What would that circuit look like? What would
that circuit look like via FACS?
 tion must be exercised when making simplifying assumptions in the (
i
i
a b s
lacI c
pLlacO-1
c
− IPTG f
l

yemGFP M
0 60 120 180 T
pLlacO-1 Time (min)
y
Figure 3 | An oscillator with no positive feedback loop. a, Network diagram t
of What does feedback
the negative this circuit do? This oscillator is similar to the dual-
oscillator. d
feedback oscillator except that the hybrid promoter pLlacO-1 (ref. 14) gives t
expression of lacI and yemGFP in the absence of LacI or in the presence of a
How
IPTG does requiring
without it work?an activator. b, Single-cell density map trajectories s
for cells containing this oscillator (see Supplementary Movie 11 and p
What controls
Supplementary Fig. the
5). amplitude and period of oscillation? o
518
Stricker et al. (2008), Nature 456:516 ©2 0 0 8 Macmillan Publis hers Lim
 un-
ach, a + Arabinose b
12

Fluorescence (a.u.)
been
nt11 araC yemGFP 8
ered
per- − IPTG 4
The
work lacI 0
0 60 120 180
back Time (min)
-cell c d
and What does araC do/?
cles.
e of How does this circuit modify the previous one?
ited
tion Why? What control does it provide?
ucer
del-
As
0 IPTG60 or arabinose
120 are
180increased,
0 what
60 will happen
120 to the
180
ng a e oscillations? f
oop,
pro-
The
Is AraC inhibition
cooperative?
 , we found

Nor
Mol

con
(×
obtained 0 0
0 80 160 0 30 60
Time (min) Time (min)
n of inducer d 50 e
begins with 60
moters,
Period (min)

Period (min)
rt delay
30
uicker than
s occurs for
umber 10
ator 10
0 10 20 30 0 1 2 3
d, assuming IPTG (mM) Arabinose (%)
of both
abundant Figure 4 | Modelling the genetic oscillator. a, Intermediate processes are
ce as many explicitly modelled in the refined oscillator model. b, c, Simulation results
AraC levels Does
fromthe impact
Gillespie of IPTG
simulations concentration
(b) or also(cmake
deterministic modelling sense?
) at 0.7%
e to the arabinose and 2 mM IPTG. AraC dimers (green), LacI tetramers (red) and
sufficient lacI mRNA (black) are shown. d, e, Comparison of modelling and
ff and the Forexperiment
these diagrams toperiod
for oscillation be true,
at 0.7%which part
arabinose (d) oris2controlling
mM IPTG
(e). Values from deterministic modelling (blue curve), stochastic
yed, the
egins anew.
thesimulations
circuit?(grey symbols, Supplementary Fig. 18), and microscopy (red
me required diamonds) or flow cytometry (green circles) are shown. Lower and upper
on the rate error bars in d represent the 16th and 84th percentiles, respectively, of the
urst. stochastic data, corresponding to 61 s.d. for a normal distribution.
tion
s roughly
design of engineered gene circuits. We found that a full model of the
system that takes into account intermediate steps such as multimeri-
 What does this circuit do?

Break it down: what are LacI and IPTG doing?

What is bla doing? How does it impact murein?

Sohka et al. (2009), PNAS 106:10135

 Overall, there is a dynamic balance between synthesis and degradation

Which step does the beta lactam inhibit?

aM-Pp = aM pentapeptide

Overall, in the presence of beta-lactamase, what builds up?

How does this affect AmpR?
 What does this circuit do?

As IPTG increases, what happens to bla?

As bla increases, what happens to aM-Pp?
What in turn happens to ampR?
And finally what happens to GFP and tetC?

What would happen to a bacteria grown in the

pesence of both Amp and Tet?
This is what is called a bandpass filter. At too high a concentration
of Amp, the cell dies, because beta-lactamse can’t keep up. At
too low a concentration of Amp, the cell dies, because not enough
murein breakdown product (AM-Pp) accumulates to up-regulate
tetC production.

Where does IPTG come in?

MICs were ‘tuned’
by IPTG.

Note that no matter

Where the activity This is in solution.
is tuned to, what the what would this
circuit relies on is look like on a plate,
the residual Amp with a single source
making Am-Pp. of ampicillin?
Here’s an experiment where bacteria are grown on tetracycline, but with a disc of
ampicillin in the middle, diffusing outwards

What happens to the circles as Amp increases?

How would IPTG impact this?

Fig. 3. Bacterial pattern formation using the band-pass filter circuit. Liquid c
plates containing Tet and/or IPTG. Ninety minutes after plating, filter paper di
Amp, IPTG, or Tet. Plates were incubated at 37 °C. Visible light and fluorescen
S4 and Fig. S5. (A) Cells on IPTG plates with different concentrations of Tet and
concentrations in the center. (C) Cells on Tet plates with different concentratio
RH06 cells (strains with different levels of !-lactamase activity) on Tet/IPTG pla
of 2 compounds. The black line comprising all points equidistant from the 2 poi
is a constant. (F) Cells on Tet plates with discs of Amp, IPTG, or Tet at the ind
 Placement of sources and
sinks allows the production
of patterns

How did the

Tet dots in the
‘H’ influence
pattern?