The basic design of Comput-E.coli dictates that colonies of different cell lines, containing the same general gene regulatory network (GRN), must reside in a system within close proximity of each other. Additionally, these colonies should be able to communicate with each other effectively while using a process that allows them to differentiate between themselves and each of their neighboring cells (left and right).Based on these vital parameters; the method decided upon for this critical task was Acyl-Homoserine Lactone Quorum Sensing.

Quorum sensing is the process by which many species of bacteria, and some other organisms, regulate important secondary metabolic processes/outputs as a function of being part of a large group of individuals (Waters 2005). In essence, quorum sensing allows a population of bacterial cells to act in unison as a symbiotic whole; resembling a multicellular organism (Miller 2001).

Gram-negative bacteria achieve this form of communication via small diffusible hormone-like chemical signals referred to as autoinducers (Waters 2005). Autoinducers are acylated-homoserine lactones; composed of a lactone ring and a variable acyl side chain (Galloway 2011). The autoinducer-provoked modification is executed through transcriptional control; either repression or activation. A well characterized example of this transcriptional change, which ultimately leads to a symbiotic relationship between organisms, can be found in the Hawaiian Squid Euprymna scolopes.

In this squid species, and other eukaryotic hosts, a pure culture of the bacteria V.fischeri colonizes a specialized light organ. This light organ provides a nutrient rich environment for the bacteria to grow, and in return the bacteria attain high cell densities and emit bioluminescence; which can be used as a benefit to the host (Miller 2001). The unanimous transcriptional activation of the V.fischeri bioluminescent pathway, within the culture, is achieved through AHL signaling. The development of this signal, and its subsequent effects, requires the presence of two critical AHL genes; Lux I and Lux R. After many decades of research, it is now evident that Lux I and Lux R homologues exist in almost all the cases of AHL signaling (Galloway 2011).

AHL synthase

The first AHL gene required for the completion of AHL signaling pathway is; the ‘I’ type gene. This gene encodes an AHL synthase which is the basis of the AHL signal formation. AHL synthases require two cytoplasmic substrates in order to create an AHL; S-Adenosylmethionine (SAM) and Acyl carrier protein (ACP) (Boyer 2009).

The chemical reaction responsible for AHL creation, and catalyzed by the synthase, is the formation of an amide bond between SAM and the acyl side chain of ACP. This acylation is then followed by a lactonization, yielding the final products; the AHL signal, a holo-acyl carrier protein, and the remaining 5’-methy-thioadenosine group from SAM (Galloway 2011). Each AHL synthase acquires its specificity, in terms of creating its proper AHL signal, by a variable C domain used to recognize a particular ACP side chain (Boyer 2009).

AHL receptor

Additionally, the ‘R’ type gene is reqiured. This gene encodes an AHL receptor which is crucial for AHL recognition as well as DNA transcriptional modification. AHL receptors bind their exclusive AHL using the variable N domain, and also specifically bind their target DNA sequence using the conserved C domain (Boyer 2009).

Receptor-AHL binding is triggered by a threshold accumulation of AHL inside the cell; while its subsequent specificity is regulated by the chemical structure of the variable side chain found on the AHL signal. Crosstalk can sometimes arise when AHL signals are structurally or chemically related; that is to say they have similar carbon chain lengths and oxidation levels (Boyer 2009).

Receptor-DNA binding is mediated by receptor-AHL binding, and is specifically executed using a unique target DNA sequence; termed the ‘lux-box-like sequence’. This sequence is found in the promoter region of AHL-receptor targeted genes (Greenberg 2000). These target sequences are conserved in terms of DNA binding reception but are also variable enough to provoke specificity dictated by the structure of the particular receptor protein being bound. In some cases, such as the LuxI/R AHL signaling pathway, these target sequences can be found in the promoter of the AHL synthase gene itself, creating a positive feedback loop (Boyer 2009).

More often the not, receptors act as a transcriptional activator of their target DNA sequence(s). The classic example of this is seen in the LuxI/R bioluminescent pathway; where the AHL-bound receptor targets the activation of the luxICDABE operon producing bioluminescence (Waters 2005). However, exceptions do exist where the receptor can act as a repressor by binding DNA, and blocking transcription, in the absence of its specific AHL (Boyer 2009).


Activation of the AHL signaling pathway amongst neighboring cells, relies heavily on one crucial step; the diffusion of the AHL molecules through the cell membrane and into the cell cytoplasm. AHL signals are said to be able to diffuse freely across the membrane since they are amphipathic molecules which theoretically should easily overcome the obstacles of crossing a phospholipid bilayer. Another important key factor is they ability of AHLs to diffuse locally in media. This allows cell-cell communication to occur even between cells that are not immediately adjacent to one another.

Comput-E.coli; AHLs

When choosing which AHL’s to use for Comput-E. coli we had to keep in mind all the limitations touched upon previously. In essence, we had to ensure that the AHL’s used:

  • Were not found endogenously in E.coli
  • Were not similar enough to induce potential cross talk
  • Did not diffuse slowly or not diffuse at all (they could not be long)
  • Had similar potential diffusion speeds based on structure
  • And were part of differing AHL signaling pathways which produced different AHL signals.

The three AHL signals chosen were done so because they respected all of the stipulations stated above. Three AHL’s were required in order to establish the directionality within the system (Left, Center, Right). As you can see; the AHL’s are of relatively similar lengths so they should diffuse at similar speeds, they differ slightly from each other based on oxidation and carbon chain length so there should be little to no cross talk, and they all use different synthase/receptor genes from different organisms which are not E.coli.

Circuit Design

AHL Synthesis Cassette

The ASC is composed of a promoter (PTetR), a cis-repressor that binds the ribosome binding sequence (RBS) at the mRNA level, preventing its translation into protein, and one of three AHL synthases. Only one of these three cassettes, exists in any given cell type at any time. The synthase found in that cell represents its own state (Center) so it can produce its own distinctive AHL as a directional signal for its neighbors (Right and Left). The active AHL’s are produced from their respective synthases only when all of the following occur in the same generation:

  • The Gas Clock input (TetR) is not produced: TetR is made only when Ethylene gas surpasses the threshold, activating its promoter PompC. Synthesized TetR then goes and binds the promoter of the ASC and represses transcription.
  • The Bistable Memory input (taR12) is produced: taR12 is only produced when the state of the cell is “ON”. taR12 can then go and bind to the inactive ASC gene transcript (cis-repressor + AHL) cleaving the cis-repressor and freeing the RBS for subsequent translation.

The active AHL’s can only perform their expected signal process when the Interface input (Signal Processing Cassette) has been synthesized: As we now know, the AHL’s require their respective receptors in order to alter DNA transcription. The receptor cassette is only synthesized when Ethylene gas surpasses the threshold since it also has the promoter PompC. This therefore separates signal formation and signal processing temporally.

Signal-processing Cassette (Receptors)

The signal-processing cassette consists of all three Receptors present in all cell types. This cassette is regulated temporally as a function of gas concentration, as explained previously. All cell types require all three receptors since each given cell will be receiving all three AHL types as an input. For example; Center- its own signal OHL, Right-its right neighbor BHL, and Left- its left neighbor OHHL. Since all three AHL’s will eventually be present in the cell the colony must be able to ‘process’ all three when needed.

Temporal Regulation of AHLs (Graph)

Below is a simulation of the gas clock and how each informational step (synthesis and processing) is regulated via the gas clock. Note how AHL synthesis and AHL processing are out of phase and never truly overlap significantly. This allows for the reliable processing of all AHLs generated in one generation and one generation alone. To achieve this pattern of expression, we needed to alter the promoter strengths of the Ompc promoter controlling AHL Synthase genes. This will be verified in future work through site directed mutagenesis of the promoter regions.

Future Considerations

Although AHL’s appear to be an effective system for synthetic cell-to-cell communication they are not without problems. One issue, that poses a critical problem for Comput-E. coli, is the AHL half-life. Even in conditions that favor degradation the most, AHL half-life is around four hours, and at that, these conditions do not favor cell viability. This lengthy half-life forces the gas clock to greater than 4 hours so as to contain all AHL from one generation in their respective generations. Ideally, a cellular automaton would have the shortest possible generation time, limited only by cellular processes (i.e. transcription and translation). Therefore, AHL’s produced must be degraded before every new generation and must also be done in a temporally coherent manner.
This problem could be solved using generalized AHL quenchers (degrading enzymes), of which there are two kinds; AHL lactonases and AHL acylases (Boyer 2009). AHL lactonases hydrolyze the ester bond of lactone ring effectively deactivating the AHL; an example of an AHL lactonase is AiiA found in many Bacillus species (Boyer 2009). On the other hand, AHL acylases hydrolyze the amide bond of the AHL releasing the acyl side group from the lactone ring; an example of an AHL acylases is AiiD found in V.paradoxus. Introducing one of these enzymes in the confines of the gas clock at a reasonable time could allow for timely degradation of AHL signals and reduce the generation time by several hours


Boyer, Mickael, and Florence Wisniewski-Dye. "Cell-cell Signaling in Bacteria; Not Simply a Matter of Quorum." Microbiology Ecology 70 (2009): 1-19. FEMS. Web. Aug. 2013.

Galloway, Warren R. J. D., James T. Hodgkinson, Steven D. Bowden, Martin Welch, and David R. Spring. "Quorum Sensing in Gram-Negative Bacteria: Small-Molecule Modulation of AHL and AI-2 Quorum Sensing Pathways." Chemical Reviews 111.1 (2011): 28-67. Chemical Reviews. Web. Aug. 2013

Miller, Melissa B., and Bonnie L. Bassler. "Quorum Sensing in Bacteria." Microbiology 55.1 (2001): 165-99. Annual Reviews. Web. July 2013

Parsek, Matthew R., and Peter E. Greenberg. "Acyl-homoserine Lactone Quorum Sensing in Gram-negative Bacteria: A Signaling Mechanism Involved in Associations with Higher Organisms." Proceedings of the National Academy of Sciences 97.16 (200): 8789-793. PNAS. Web. Aug. 2013.

Sitnikov, Dmitry M., Jeffrey B. Schineller, and Thomas O. Baldwin. "Control of Cell Division in Escherichia coli: Regulation of Transcription of StsQA Involves Both RopS and SidA-mediated Autoinduction." Microbiology 93 (1996): 336-41. NCBI. Web. Aug. 2013.

Waters, Christopher M., and Bonnie L. Bassler. "Quorum Sensing; Cell-to-Cell Communication in Bacteria." Cell and Developmental Biology 21 (2005): 319-46.Annual Reviews. Web. July 2013.