Team:SCU China/Repressilator/qs
Quorum sensing, or the
control of gene expression in response to cell density, is used by both
gram-negative and gram-positive bacteria to regulate a variety of physiological and
chemical functions. In all cases, quorum sensing involves the production and
detection of extracellular signalling molecules called autoinducers. While universal
signalling themes exist, variations in the design of the extracellular signals, the
signal detection apparatuses, and the biochemical mechanisms of signal relay have
allowed quorum sensing systems to be exquisitely adapted for their varied uses,
enabling bacteria to architect complex community structures. [2]
Quorum sensing is first integrated into our
circuit to solve three main problems:
1. The plasmid (pLPT107) we get is totally 9800bp three reporters (CFP/YFP/RFP)
contained and if we insert the four melatonin synthases (5800bp) into pLPT107, the
whole plasmid will ultimately reach 15600bp which is too big to transform
efficiently into bacteria.
2. Three repressors, four enzymes and at least one reporter give the cell such a
heavy protein synthesis pressure. And it may greatly influence the growth rate of
E.coli. So it is better to divide two system into two cells and combine each other
with inner-related QS.
3. A more regular and precise oscillating pattern still could not be obtained in our
previous part. So QS have been settled on as a cell-cell communication pathway to
urge the bacteria synchronize in a coherent oscillatory state, according to a
computational modeling work in 2000[1].
It is predicted that diverse and noisy community of such genetic oscillators
interacting through a quorum-sensing mechanism could self-synchronize in a robust
way, leading to a substantially improved global rhythmicity in the system.
To illustrate the QS function in synchronizing
repressilator network, we used three GIF to help understand this process and
relative logic.
As shown in the curves, the concentration of three repressors varies alternatively and the peak for each one is followed by the peak of the other one that repress it. So in the time scale if taking TetR into account at the very beginning, the adjacent peak must belong to LacI because just LacI inhibits the expression of TetR and once TetR protein concentration decreases, it indicts that increasing LacI protein level has exceeded the pLac promoter threshold and tetR gene has been repressed. In other words TetR decrease because LacI increase and this logical order are able to be transferred to the concentration peak order in time scale. This logic is applicable for all three repressors because they are logically symmetry.
The ultimate QS circuit is pCI LuxI-J23106 LuxR-pTet/Lux pR Hybrid promoter CI as shown in part2. And LuxI (controlled by pCI promoter) is expressed when CI lambda repressor concentration is low enough (threshold) and this situation only appear when cellular TetR concentration is over threshold/TetR dominant in a oscillating period. Then LuxI will catalyze AHL synthesis and AHL can defuse freely in and out of the cell as a small nonpolar molecule.
Part 3 displays the impact of quorum sensing on repressilator. LuxR is constitutively expressed in the cell and in the early nonsynchronous stage different cells may produce AHL at different phase so Quorum Sensing especially in some limited area could initiate at any time corresponding to any repressilator phase and may have various impact. According to our design pTet/Lux pR hybrid promoter should have a basic property that once TetR protein concentration exceeds threshold downstream CI transcription will be inhibited no matter QS happen or not. This property remove the QS impact on repressilator’s TetR dominant stage. As for CI dominant phase the actual effective period is between TetR concentration early increasing state to peak (accurately shortly before peak because central dogma lead to a time delay, which means although gene is repressed but protein concentration still increase for the mRNA remaining) and if additive CI is expressed the whole CI dominant phase will not elongate for the reason that CI protein half-life is determined by itself.
Summarily, the QS have the most obvious impact on LacI rapid increasing phase in one oscillating period. Because at this period of time if QS activate additive CI expression, LacI gene will be repressed simultaneously and quickly ‘jump’ to CI dominant phase after a short period of time. Comparing with normal repressilator circuit, QS inserted one give the coherent a only LacI phase shift every QS happens, which render those bacteria ‘chase’ to achieve a much more synchronous stage.
Based on this modelling work, two circuit named
4p0＆3p1 are designed to provide a theoretically feasible solution and achieve the
ultimate macroscopic, population-scale oscillation. Both two plans comprise two
parts—cell A and cell B which include repressilator plus quorum sensing system and
synthases system respectively. The Major difference between our design and that in
the paper is we choose CI-lambda to control the expression of LuxI (AHL synthase)
and pTet-LuxpR hybrid promoter (BBa_K176000) for QS CI repressor, because pLac
(LacO1 promoter) exists leak expression phenomenon.
In “4p0” system, luxI gene of the “enzyme cell” is replaced by COMT (rate-controlling enzyme of melatonin synthesis pathway) gene. Therefore, with high level of repressor TetR in “repressilator cells”, COMT will be expressed in “enzyme cells”. In this way, QS circuit can transfer and modulate the rhythmic signal from “repressilator cells” to “enzyme cells” via AHL concentration fluctuation and then ‘enzyme cells’ receive the signal and modulate the COMT expression with periodical pattern. In this system QS functions not only as ‘synchronizor’ but also ‘modulator’ assist the whole working steadily and efficiently.
Cell A-Repressilator cell
For “repressilator cells”, circuit can be seen as two system (QS and Repressilator)
inner-related with each other via pCI controlled LuxI and additive hybrid promoter
induced CI lamba. pCI-LuxI is thought as the Repressilator output/QS input, matching
with QS output/Repressilator input to form a bistable system, and simultaneously
negative feedback pattern in QS circuit increases the potential balance or synchrony
of whole system.
Cell B-Enzyme cell
Two similar gene circuit consists of following parts-constitutively expressed luxR,
tetR (+LVA tag target for its quick degradation) controlled by luxpL promoter,
fluorescent protein mOrange controlled by pTet promoter (These three are identical
for both) and luxI under the control of pTet promoter just for 3p1 comparing to COMT
just for 4p0. The whole circuit is encoded on pSB1C3 plasmid in Escherichia coli
strain BL21/DHL708 which are used to express enzymes in melatonin synthesis pathway.
When the level of repressor TetR is high, the
expression of CI is inhibited, which
contribute to the expression of LuxI catalyzing the synthesis of acyl-homoserine
lactone (AHL), a signaling molecules in bacterial quorum sensing. Subsequently, AHL
starts to defuse and accumulate in bacterial population as a small molecule. When
the amount of AHL reach the threshold, bacterial quorum sensing in both cells are
initiated rapidly. In this circumstance, constitutively expressed LuxR can interact
with AHL and then have conformational changes for binding to and inhibit/activate
luxpL/R promoter. As a consequence, in the cell B the amount of TetR (+LVA tagged)
repressors expressed will fall down quickly and then the following pTet promoter
will be activated, while in the cell A this impact on the pTet/LuxpR hybrid promoter
are not determined by effective LuxR (binding AHL) but dominantly by inner cell TetR
repressor’s concentration. In other words, although quorum sensing has initiated if
the number of the TetR repressors are still high enough (over the threshold) the
hybrid promoter will ultimately maintain in a repressed state.
In “3p1” system, this gene circuit is used for accumulate AHL in bacterial
population more quickly. Mathematical and computational predictions have suggested
that the oscillation is much more stable and regular if the amount of AHL is large
enough. When the level of repressor TetR is high in “repressilator cells”, LuxI will
be expressed in “enzyme cells” which contribute to the synthesis of AHL. On this
occasion, both “repressilator cells” and “enzyme cells” produce AHL in bacterial
population. As a result, it will take less time to reach stable stage, with
fluorescent protein mOrange indicting whether the circuit works as expected. And we
rename QS system ‘synchronizor’.
To confirm QS function
in architect cell-cell communication we build LuxpR-GFP-constitutive (J23106)
LuxR-pSB1C3, pTet-LuxI/pCI-LuxI-pSB1C3 and pTet/LuxpR hybrid promoter-J23106
LuxR-J23106 TetR-pSB1C3 via co-culture and additive AHL. pTet-LuxI/pCI-LuxI-pSB1C3
are exclusively built and tested because they can constitutive or semi-constitutive
express LuxI(AHL synthase) when the TetR/CI-Lambda repressor are not added in and
both are key components for QS modulator system in 3p1/4p0 circuit. All circuits
referring to are transformed into E.coli DHL708 and BL21 simultaneously for the
reason that four melatonin synthase are designed to express in BL21 or DHL708 and
pTet/LuxpR hybrid promoter-Lambda as the QS output versus repressilator input have
to be confirmed in DHL708. Function of LuxR transcriptional activator (Lux pR) is
confirmed in both GFP reporter and hybrid promoter-RFP circuits after co-culture
with DHL708/BL21 pTet-LuxI/pCI-LuxI for about 13 hours. Several nutrition and
culture conditions (M9/LB/artificial intestinal fluid) are set to explore the QS
threshold in different milieu, especially artificial intestinal fluid are used to
mimic the QS in the intestine. pCI-LuxI and Hybrid promoter-J23106 LuxR-J23106 TetR
(known as UHP) are co-cultured in same wells of a 96-well plate with formula 8uL
pCI-LuxI DHL708 bacteria liquid culture/ 8uL UHP bacteria liquid culture/ 134 uL
corresponding blank culture medium and UHP DHL708 used in this section is
conditioned with 80 ng/mL ATc 4 hours before measurement( aTc+ indicted in the
figure). Continuous incubation in plate reader for 37 degree Celsius and measure RFP
fluorescence (580/630 nm) every 15 mins. Supplementary blank mediums are added in
the measurement process because culture medium is evaporating in 37℃ incubation and
the bacteria are diluted after this, shown as black arrow in the figure. As Figure
3a,b, indicts, the OD is about 0.1 at the 0 time point, while the fluorescence/OD
(means the relative RFP protein concentration in cell unit) start at a low level
about 8000 A.U. per OD. It is obvious that fluorescence for all four groups begin
increase exponentially at about 60-75mins latter with the OD600 reaching 0.2. As for
pTet-LuxI/UHP-aTc+ groups pTet-LuxI BL21 are also designed to co-culture with
UHP-aTc+ DHL708 because pTet-LuxI is necessary in 3p1 cell B. Slightly different
from pCI-LuxI/UHP-aTc+ groups, data exhibits that mainly after 135mins groups begin
early exponential phase, with OD600 reaching 0.25-0.3.
According to the computational modelling results, we find QS system function like iPTG synchronize in the early state (video), which all accelerate the oscillating period for those in LacI dominant period but not in the rest (TetR＆CI). So during a few following generations this pattern will assist the coherent to achieve self-sychrony. And to imitate this process different concentration(100uM/250uM/500uM/1000uM) of iPTG is added in the culture M9 medium and use to incubate the DHL708+pLPT107. We measure this oscillation pattern and period during whole project totally for three times and all shown in Figure 5. At the first time, we dilute the bacteria every 60-90 minutes to maintain the coherent OD600 into the 0.2-0.5 interval to keep it in growth exponential phase and we get the oscillating period for about 2 hours, seriously it is really a little bit short and not so convincing because we do not valuate the influence the dilution work to the data accuracy and maybe the 2-hour-period result from the dilution. For the second time we continuous 37℃ incubate the bacteria into plate reader for 10 hours and acquire each data point per 30mins, but as indicts the so called period become much more irregular. More data points are needed to do further analysis, so we have our third measurement—continuous incubate 13 hours and gain each point for 15mins, however, still similar pattern. In our modelling work, we analyze this phenomenon and give a reasonable explanation that the coherent are not well synchronized just via iPTG so we hope after QS system is inserted into the repressilator, the oscillation could be urged into a much more steadily synchronous stage.
1. Repeat DHL708ΔclpXP and MC4100 repressilator measurement for more than 30 hours to
confirm oscillation.
2. Insert QS circuit (pCI LuxI-J23106 LuxR-pTet/LuxpR hybrid promoter Lambda) into
repressiltor pLPT107 by replacing mCFP and mKate2 reporter as indicted in fig2 and
measure the oscillation pattern of this optimized one (period ＆ regular or not).
3. Repeat pTet/LuxpR hybrid promoter confirmation work.
1. Garcia-Ojalvo J, Elowitz M B, Strogatz S H. Modeling a synthetic multicellular clock:
Repressilators coupled by quorum sensing[J]. Proceedings of the National Academy of
Sciences of the United States of America, 2004, 101(30):10955-60.
2. Bassler B L. How bacteria talk to each other: regulation of gene expression by quorum
sensing[J]. Current Opinion in Microbiology, 1999, 2(6):582.
