Escherichia coli

For safety reasons, the bacteria gram negative E. coli was chosen to mimic V. cholerae. E. coli is an easy organism to deal with, especially as it is well documented, easy to transform with exogenous DNA and easy to culture. The strain K-12 MG1655 was transformed with a plasmid allowing expression of the protein CqsA from V. cholerae, the enzyme responsible for the synthesis of CAI-1. However, as a proof of concept, we also transformed our E. coli strain with the gene coding for the CqsA of V. harveyi, a non-pathogen strain, producing the molecule C8-CAI-1 (an analogue of the V. cholerae CAI-1)^1,2. C8-CAI1 is a carbohydrate chain based displaying an hydroxyl group on carbon 3 and ketone function on carbon 4. The CqsA synthetase from V. harveyi produce C8-CAI-1 from endogenous E. coli (S)-adenosylmethionine (SAM) and octanoyl-coenzyme. cqsA from both V. harveyi and V. cholerae were placed under the pLac promoter and we used plasmid pSB1C3 to maintain compatibility with the iGEM registry.

V. harveyi

V. harveyi is a gram negative bacteria, well studied for its quorum sensing system. This bacteria displays its own pathway for the detection of C8-CAI-1. The gene cqsS encodes for the sensor C8-CAI-1 and a single point mutation in its sequence allows V. harveyi to detect both C8-CAI-1 and CAI-1 from V. cholerae. To avoid auto-activation of V. harveyi, we used the JMH626 strain, in which the cqsA gene, coding the enzyme involved in the production C8-CAI-1, has been deleted. Furthermore, additional genes luxS and luxS coding for key enzymes involved in the expression of other quorum sensing molecules have been deleted. All these mutations make the strain JMH626 specific for detecting non-endogenous C8-CAI-1. V. harveyi is also able to regulate the activation of genes under the control of the promoter pQRR4, in a C8-CAI-1 concentration dependent manner ¹. At high C8-CAI-1 concentration, the promoter is inactivated. Thus we added the inverter tetR/pTet to activate a gene of interest in presence of C8-CAI-1. The gene of interest is als that encodes for the acetolactate synthase (Als). This enzyme synthetized diacetyl from pyruvate³. Diacetyl is our ransmitter molecule (Figure 1).

The pBBR1MCS-4⁴, a broad host range plasmid, was chosen to allow the transfer of the system into V. harveyi by conjugation (i.e. this is the only way to modify the V. harveyi chassis).

In conclusion, we designed a V. harveyi strain enable to detect both exogenous CAI-1 or C8-CAI-1, and to produce diacetyl as a molecular response.

P. pastoris

V. harveyi cannot not be used as the effector since production of antimicrobial peptides (AMPs) is lethal for Vibrio species. P. pastoris is a yeast commonly used in academic laboratories and industry for its high potential to produce protein. In addition, yeasts were previously described to produce a wide range of AMPs ^5,6. Finally, a system allowing efficient communication between yeast and prokaryotes has already been decribed i.e. the diacetyl-dependant Odr-10 receptor system⁷. This system allows the expression of targets genes under the control of pFUS1 via the Ste12 pathway (Figure 2). For all these reason, we thus chose P. pastoris. We used the constitutive pGAP promoter to express the receptor Odr-10 in P. pastoris

To kill V. cholerae, we looked for a new and innovative antibiotic solution to limit the risk of acquired-resistance. We decided to use AMPs, that are small membrane disrupting molecules toxic for a large panel of microorganisms⁸. Here we selected AMPs from crocodiles. Crocodiles live in harsh environment and are known to possess an impressive defence system, that allows them to catch very few disease and antimicrobial peptides are part of it⁹. We focused on 3 different AMPs described to have the best efficiency against V. cholerae. Those AMPs are Leucrocin I¹⁰, D-NY15¹¹ and cOT2¹². Leucrocine I possess one cationic charge for 7 amino acids. D-NY15 is its optimized counterpart with 4 cationic charges and a sequence of 15 amino-acid long. Finally, cOT2 is 29 amino acid long and possesses 6 cationic charges. These AMPs were placed under control of the pGAP constitutive promoter for preliminary tests and under pFUS1 promoter to promote their expression in response to diacetyl. The genetic constructions were inserted into the integrative pPICZα plasmid i.e. a good plasmid for protein production. The signal peptide α-factor was fused to the AMPs to allow for secretion of the peptides.

The action of these AMPs is the last event of our synthetic consortium.