<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1 General idea
Tsinghua iGEM 2012 presents the idea of “Domino Effect in E. coli” as a potential approach to allow a whole E. coli community to perform exactly the same with only one weak initial signal and carry out information processing. This goal can be achieved by a set of quorum sensing-based design, including basic modules of signal reception and release, signal amplification and complex modules of logic gating. Furthermore, the population of bacteria community is a huge advantage for information processing compared to existing single cell-based computing circuitries.
Figure 1A. This figure shows how conventional domino effect
works, as the tile in front fall, they topple the tiles behind them
as a result – signal transmission and
Figure 1B. This figure demonstrates our basic concept of
E. coli dominos, inspired by the conventional domino effect.
From the figures above, if we look at it from another perspective, the domino effect can be viewed as a form of information processing, with the possibility to realize signal amplification and logic gating. We attempted to use biological elements to replace the mechanical elements in mechanical domino effect: each bacterium represents a domino tile; the action of topple here refers to a set of information transmission process, and the mechanical force is replaced by chemical signaling molecules; these elements are the key to building a set of “E. coli dominos”, like the one we showed in figure 1B.
Quorum sensing in prokaryotes allows bacteria to communicate through chemical signals like AHLs, and coordinate behaviors among the whole bacterial community. This mechanism is ideal to serve as a basis for our project.
In the basic module, a small proportion of E. coli in the community will receive and recognize an external signal, then send it forward to bacteria nearby using AHL signals, causing the same reaction in cells nearby. Since each bacterium may affect tens or even hundreds of surrounding bacteria, the signal amplifies exponentially. Thus, the signal will be transmitted and amplified step by step.
As an upgrade of the basic modules, a multi-signal system is designed, starting with two distinct AHL signals working independently, so as to achieve the logic gating and calculation functions. Specifically speaking, distinct information processing functions, or logic gates, such as ‘AND’, ‘AND NOT’, ‘XOR’ can be realized by unique biological circuitries, and the assembly of these modules shows promise for obtaining complex computation functions.
Taken together, our project aims to establish a novel information processing system in bacterial community, based on the logic of domino effect and quorum sensing system, which possesses the capability of information transmission, amplification and logic gating.
The designs of our project are illustrated as followed.
3.1 Basic modules
3.1.1 Signal reception and output:
The basic function of signal reception and release is realized by a simple pathway including AHL as signal molecule, LuxR as signal reception part and RFP as a sign of signal reception and subsequent release. The AHL molecules may come from two sources: external (i.e. AHL solution we drop into the system) and internal (i.e. AHL from neighboring stimulated cells).
Here we used two kinds of AHL molecules from two species, with the same functioning mechanism and no cross-linking: 3OC6HSL from V. fischeri used in Lux system and 3OC12HSL from Pseudomonas aeruginosa used in Las system. The binding of AHL to LuxR will form “the dimer of dimer”, which consists of two AHL molecules and two LuxR protein molecules. This complex will activate the transcription from the promoter PLuxR, leading to subsequent expression of RFP. RFP, or other such proteins, serves as an output of the input information. In this part, two signal pathways using two different kinds of AHL molecules are constructed.
One property of quorum sensing signals is that the chemical molecules are able to freely move across membranes, which is also critical for our project in the signal transmission module. The concept of positive feedback is also applied in this module, making it possible to intensify signal strength and avoid attenuation.
We constructed our own positive feedback loop in E. coli with several different parts. LuxI, with the function of expressing AHL synthase, was inserted downstream of RFP in the aforementioned basic module. Thus, once external AHL is received and RFP expressed, AHL would be further synthetized in the same bacterium. As a result, AHL molecules synthesized after RFP could affect the E. coli itself, synthesizing even more AHLs and completing the positive feedback loop; alternatively, it could travel across membranes to other bacteria nearby, and start the activation of identical modules in those cells. In other words, information received by the initial bacterium is transmitted to its neighbors, and they would respond the same way and pass the signal on to other cells, just as domino tiles would pass the mechanical force on to other tiles. Furthermore, the expression of RFP would spread to the whole community over time, and accumulate in a certain time period. Both the positive feedback loop and the enormous population of bacteria could serve for the purpose of signal amplification.
When different signals meet in the same cell, the state of the cell would shift from “transmission” to “processing”. The conventional dominoes can be arranged to function as logic gates, which inspires us to design similar information processing circuitries with our system.
Figure2. An abstract of domino tiles to complete logic gate-like functions, which are also realizable with our biological system.
Here we designed a simple part to achieve a logic gate-like information processing. When only signal B (3OC12HSL) is present, RFP is expressed and can be seen. But if signal A (3OC6HSL) is added with signal B, cI repressor is expressed and then inhibit the expression of RFP.
We also designed a system to accomplish the information processing. In this system, there are some fixed parts such as Plac, Plux, and some alterable parts like R1 to R4 (protein coding genes) and PA, PB (promoters). In order to make one logic gate available, we can just change the R1 to R4 or PA, PB.
When R1, R2, R3, R4 and PA, PB in figure above represent different protein coding gene or different promoter, as shown in the table, various logic gates are realized.
This complex pathway achieves the 'AND-gate' mentioned in logic gate design. When there is no signal input, cI is expressed constantly. If we add only signal B (3OC12HSL, which can activate Plas promoter), Hybrid promoter Plas/cI is repressed by cI, thus no RFP can be observed. With only signal A (3OC6HSL, which can activate Plux promoter), Plas/cI promoter cannot be activated without signal B. When signal A and B are present together, because of the activation of Plux by signal A, lacI is expressed and will then inhibit the expression of cI, Plas/cI will be activated by signal B without cI, then RFP can be seen.
We thank Prof. SUN Zhirong, Prof. CHEN Guoqiang and Prof. DAI Junbiao for discussion and analysis. We thank LI Teng and FU Xiaozhi for suggestions and instructions during our experiment. We thank Peking iGEM team for supplying AHL solution. Our team and project are supported by grants from Tsinghua university.
- Michael B. Elowitz & Stanislas Leibler. A synthetic oscillatory network of transcriptional regulators.Nature. (2000) Jan 20;403(6767):335-8.
- Chenli Liu et al. Sequential Establishment of Stripe Patterns in an Expanding Cell Population. Science. (2011) Oct 14;334(6053):238-41.
- Wang WD, Chen ZT, Kang BG, Li R. Construction of an artificial intercellular communication network using the nitric oxide signaling elements in mammalian cells. Exp Cell Res. (2008) Feb 15;314(4):699-706.
- Dasika MS, Burgard A, Maranas CD. A computational framework for the topological analysis and targeted disruption of signal transduction networks. Biophys J. (2006) Jul 1;91(1):382-98.
- Ladbury JE, Arold ST. Noise in cellular signaling pathways: causes and effects. Trends Biochem Sci. (2012) May;37(5):173-8.
- Ausländer S, Ausländer D, Müller M, Wieland M, Fussenegger M. Programmable single-cell mammalian biocomputers. Nature. (2012) Jul 5;487(7405):123-7.