Design and construction of synthetic adhesins driving the specific attachment of Escherichia coli to target surfaces, cells, and tumors

Resumen

One of the aims of genetic engineering and synthetic biology is the design of microorganisms with novel capabilities that could be interesting for the development of new vaccines, diagnostic sensors, and therapeutic interventions for major diseases such as cancer. In this regard, the availability of genetic elements able to program the adhesion of the engineered bacterium to different targets would be extremely useful. This work reports the development of synthetic adhesins (SAs) that enable to precisely program the adhesion properties of Escherichia coli (E. coli) bacteria to different target surfaces, including tumor cells. The structural organization of these novel SAs is defined by a domain that anchors the SA into the bacterial outer membrane, derived from the N-terminal fragment of Intimin comprising residues 1-659, and an adhesive domain based on the smallest antibody fragments known to date, termed VHHs. This modular organization allows the modification of the binding specificity of the SA by exchange of the VHH sequence. We demonstrate that SAs are efficiently displayed on the surface of E. coli and are able to drive bacterial adhesion to antigen-coated abiotic surfaces as well as to target tumor cells expressing on their surface the antigen recognized by SAs. SAs were constitutively and stably expressed from the chromosome of an engineered E. coli strain lacking a conserved set of natural adhesins (i.e. type 1 fimbriae, Antigen 43 and mat fimbriae) and constitutively expressing the lux operon as bioluminescent reporter. Using tumor xenograft mouse models we have demonstrated that engineered E. coli strains carrying SAs colonize efficiently solid tumors expressing the cognate antigen recognized by the SA using lower doses (ca. two order of magnitude) of systemically administered bacteria, compared to control strains with SAs binding an unrelated antigen or the wild type E. coli strain. In addition, we observed that the engineered strains were cleared faster from non-target organs (e.g. liver and spleen) probably due to the deletion of natural adhesins. The fast and specific adhesion mediated by SAs was also employed to characterize the influence of both, flagella and YcgR protein in the adhesion process, as well as to investigate the short-term transcriptional response of E. coli upon adhesion to tumor cells. Our results indicate that, whereas active bacterial motility mediated by flagella is important for an efficient adhesion to target cells, the lack of YcgR protein does not affect the ability of bacteria to adhere to target cells, suggesting that the arrest of bacterial motility upon adhesion is independent of YcgR protein. In addition, we analyzed by RNAseq the global transcriptional response of E. coli bacteria upon adhesion (15 min) to target tumor cells. Our results indicate a common transcriptional response upon adhesion regardless of the cellular receptor recognized. Genes involved in sulfur uptake and its metabolism were upregulated upon adhesion, whereas genes involved in the transport of intermediates of the tricarboxylic acid cycle and tryptophan synthesis were downregulated. We found that the activity of the gene fusion between the chromosomal yeeE promoter region and the mCherry reporter gene was upregulated in response to bacterial adhesion. Lastly, we have also demonstrated that SAs can drive the specific adhesion of non-live bacterial derived nanoparticles towards target tumor cells expressing a therapeutically relevant cell surface receptor (i.e. EGFR).