SURF 2019: Engineering a synthetic high-bandwidth intercellular communication system through plasmid conjugation

2019 SURF project description

• Mentor: Richard Murray
• Co-mentor: John Marken

Background

In order to realize many of synthetic biology's long-term goals, it is necessary to engineer populations of interacting cells rather than focusing on engineering individual cells alone. So far, work in this area has almost exclusively used quorum sensing systems to transfer information between cells, due to these systems' simplicity and modularity. However, because a quorum sensing system encodes its information in the concentration of a single signaling molecule, and because different systems exhibit a high degree of cross-talk with each other, quorum sensing is a low-bandwidth information transmission medium [1]. This makes quorum sensing a poor choice for implementing engineered populations with a high degree of complexity.

Plasmid conjugation is a common type of bacterial horizontal gene transfer where a plasmid in one cell transfers a copy of itself into another adjacent cell. Because entire synthetic circuits can be contained on the transferred plasmid, conjugation is an extremely high-bandwidth channel for intercellular information transfer. Furthermore, synthetic conjugation systems have already been developed where 'helper plasmids' encode all of the genes required for plasmid transfer but lack the ability to transfer themselves, thereby conferring on its host cell the ability to modularly send any plasmid that contains the appropriate 'origin of transfer' (oriT) sequence [2,3]. Such systems allow distinctions to be made between 'sender' and 'receiver' strains, which is an important prerequisite in the development of a flexible intercellular communication system.

Despite the advantages of plasmid conjugation, so far there has been no published work which uses it as the basis for a general synthetic intercellular communication system. To give just one example of the new capabilities which are made available by conjugation-based signaling, I am currently developing a 'message routing' framework that allows messages to be transferred only to designated recipients within a population (Fig. 1). In this system, the cells express Cas9 and a guide RNA which serves to identify their strain. The signal plasmid contains an 'address' region which includes binding sites for the guide RNAs. The signal plasmid can only be transferred to a strain which does not have a site on the address, as otherwise the plasmid will be immediately cleaved by Cas9 upon receipt. Integrase attachment sites flank the guide RNA binding sites in the address, allowing the cells to use integrase-mediated cassette exchange to dynamically swap out guide RNA binding sites within the address, editing the list of 'allowed' recipients of the signal plasmid. In this way one can construct a defined path of information flow through a population, such as the linear path shown in Fig. 1.

Fig. 1: (a) A 3-node linear signal path built with two guide RNAs and one integrase. The signal plasmid originally starts in Cell A, and cannot be sent to Cell C as the address contains the C gRNA binding site. However, once the signal plasmid enters Cell B, the TP901 integrase swaps the C binding site for an A binding site. The signal plasmid is now unable to return to Cell A, but is now able to proceed to cell C. (b) A 4-node linear signal path built with three guide RNAs and two integrases.

Project Goals

The aim of this SURF project is to make experimental and/or theoretical contributions to the development of a conjugation-based intercellular communication system. These contributions can encompass anything from improving the fundamental properties of the conjugation system itself, to implementing a new population circuit that uses conjugation, to exploring broader concepts about what can be enabled by high-bandwidth intercellular communication. Some examples for project directions are given below.

Fruitful avenues for experimental work include:

• Creating helper plasmids with externally-controllable 'dials' to dynamically tune the rate of conjugation.
  • The regulatory architecture governing the genes associated with conjugation is well-understood [4]. By introducing inducible copies of the master transcription factors regulating transfer onto the helper plasmid, we should be able to modify existing synthetic conjugation systems to allow for more versatile control of the transfer dynamics.
• Implementing systems that allow cells to dynamically switch (and be switched) between sending and receiving states.
  • Currently, synthetic conjugation systems distinguish between Senders and Receivers by having cells containing the helper plasmid be Senders and all other cells being Receivers. However, because the helper plasmid cannot transfer itself, there is no way to dynamically alter these distinctions without re-engineering the whole population. By deleting the master regulators of the transfer genes on the F plasmid and replacing them with inducible versions, it should be possible to create basally inert versions of the helper plasmid that can be switched on and off by external signals (and even by the cells themselves).
• Creating spatially-defined population-level circuits.
  • One of the interesting properties of conjugation is that it can only occur between adjacent cells. This allows spatial circuits to be defined at the length scale of cell-cell contacts, rather than by the diffusion constant of a molecule. Can we build a spatial circuit that harnesses this distinction for a functional role?

Interesting questions to address theoretically/computationally include:

• Are there new types of population-level circuits which can be built when plasmids (instead of quorum sensing signals) are the medium of communication?
  • In addition to the high bandwidth and the fact that signaling can only occur between adjacent cells, conjugation-based communication has many other properties that distinguish it from quorum-based communication. These include the fact that the messages (plasmids) can replicate semi-independently of their hosts and the fact that the message can propagate through different cells and update its content based on the conditions within each cell. Are these extra properties enough to distinguish conjugation-based communication as a qualitatively distinct form of communication from quorum signaling? If so, there should be population circuits which can be designed with conjugation to perform functions that equivalent quorum-based circuits could not perform. Do these circuits exist? If so, what do they look like? For further reading along these lines, see [5,6,7].
• Can we think of general schemes for information representation in plasmids that allow for the encoding and interpretation of arbitrary messages? What are the advantages and disadvantages of different schemes?
  • Although an intercellular communication channel based on quorum sensing can only transfer information encodable into a single scalar variable, it does have the advantage of being able to easily map a continuous-valued input signal into a continuous-valued message. Because conjugation-based messaging is driven by the receipt of a plasmid into a cell, the simplest implementation of signaling is in representing a multi-dimensional binary message (where components of that message might either be present or absent from the signal plasmid). Although there are many ways to transfer continuous-valued information across a conjugation-based communication channel, it is not clear a priori whether there is an optimal molecular implementation (and if so, what it is).

Prerequisite Skills

For Experimental Work

• Familiarity with standard BSL-1 wetlab procedures
• Familiarity with molecular cloning techniques
• Familiarity with synthetic biology and genetic circuit design

For Computational Work

• Coding experience, preferably in Python
• Familiarity with differential equations
• Familiarity with synthetic biology and genetic circuit design

References

1. Davis RM, Muller RY, Haynes KA. Can the natural diversity of quorum-sensing advance synthetic biology?. Frontiers in bioengineering and biotechnology. 2015 Mar 10;3:30.
2. Dimitriu T, Lotton C, Bénard-Capelle J, Misevic D, Brown SP, Lindner AB, Taddei F. Genetic information transfer promotes cooperation in bacteria. Proceedings of the National Academy of Sciences. 2014 Jul 29;111(30):11103-8.
3. Strand TA, Lale R, Degnes KF, Lando M, Valla S. A new and improved host-independent plasmid system for RK2-based conjugal transfer. PloS one. 2014 Mar 3;9(3):e90372.
4. Zatyka M, Thomas CM. Control of genes for conjugative transfer of plasmids and other mobile elements. FEMS microbiology reviews. 1998 Feb 1;21(4):291-319.
5. Goñi-Moreno A, Amos M, de la Cruz F. Multicellular computing using conjugation for wiring. PLoS One. 2013 Jun 20;8(6):e65986.
6. Beneš D, Rodríguez-Patón A, Sosík P. Directed evolution of biocircuits using conjugative plasmids and CRISPR-Cas9: design and in silico experiments. Natural Computing. 2017 Sep 1;16(3):497-505.
7. Goni-Moreno A, de la Cruz F, Rodriguez-Paton A, Amos M. Dynamical Task Switching in Cellular Computers. bioRxiv. 2018 Jan 1:479998.