Logo: to the web site of Uppsala University

Contact‐dependent growth inhibition (CDI) allows bacteria to recognize kin cells in mixed bacterial populations. In Escherichia coli, CDI mediated effector delivery has been shown to be species‐specific, with a preference for the own strain over others. This specificity is achieved through an interaction between a receptor‐binding domain in the CdiA protein and its cognate receptor protein on the target cell. But how conserved this specificity is has not previously been investigated in detail. Here, we show that class II CdiA receptor‐binding domains and their Enterobacter cloacae analog are highly promiscuous, and can allow for efficient effector delivery into several different Enterobacteriaceae species, including Escherichia, Enterobacter, Klebsiella and Salmonella spp. In addition, although we observe a preference for the own receptors over others for two of the receptor‐binding domains, this did not limit cross‐species effector delivery in all experimental conditions. These results suggest that class II CdiA proteins could allow for broad‐range and cross‐species growth inhibition in mixed bacterial populations.

The C-terminal (CT) toxin domains of contact-dependent growth inhibition (CDI) CdiA proteins target Gram-negative bacteria and must breach both the outer and inner membranes of target cells to exert growth inhibitory activity. Here, we examine two CdiA-CT toxins that exploit the bacterial general protein secretion machinery after delivery into the periplasm. A Ser281Phe amino acid substitution in transmembrane segment 7 of SecY, the universally conserved channel-forming subunit of the Sec translocon, decreases the cytotoxicity of the membrane depolarizing orphan10 toxin from enterohemorrhagic Escherichia coli EC869. Target cells expressing secY(S281F) and lacking either PpiD or YfgM, two SecY auxiliary factors, are fully protected from CDI-mediated inhibition either by CdiA-CTo10EC869 or by CdiA-CTGN05224, the latter being an EndoU RNase CdiA toxin from Klebsiella aerogenes GN05224 that has a related cytoplasm entry domain. RNase activity of CdiA-CTGN05224 was reduced in secY(S281F) target cells and absent in secY(S281F) Delta ppiD or secY(S281F) Delta yfgM target cells during competition co-cultures. Importantly, an allele-specific mutation in secY (secY(G313W)) renders DppiD or Delta yfgM target cells specifically resistant to CdiA-CTGN05224 but not to CdiA-CTo10EC869, further suggesting a direct interaction between SecY and the CDI toxins. Our results provide genetic evidence of a unique confluence between the primary cellular export route for unfolded polypeptides and the import pathways of two CDI toxins. IMPORTANCE Many bacterial species interact via direct cell-to-cell contact using CDI systems, which provide a mechanism to inject toxins that inhibit bacterial growth into one another. Here, we find that two CDI toxins, one that depolarizes membranes and another that degrades RNA, exploit the universally conserved SecY translocon machinery used to export proteins for target cell entry. Mutations in genes coding for members of the Sec translocon render cells resistant to these CDI toxins by blocking their movement into and through target cell membranes. This work lays the foundation for understanding how CDI toxins interact with the protein export machinery and has direct relevance to development of new antibiotics that can penetrate bacterial cell envelopes.

Antimicrobial resistance (AMR) is an increasing problem worldwide, and new treatment options for bacterial infections are direly needed. Engineered probiotics show strong potential in treating or preventing bacterial infections. However, one concern with the use of live bacteria is the risk of the bacteria acquiring genes encoding for AMR or virulence factors through horizontal gene transfer (HGT), and the transformation of the probiotic into a superbug. Therefore, we developed an engineered CRISPR-Cas9 system that protects bacteria from horizontal gene transfer. We synthesized a CRISPR locus targeting eight AMR genes and cloned this with the Cas9 and transacting tracrRNA on a medium copy plasmid. We next evaluated the efficiency of the system to block HGT through transformation, transduction, and conjugation. Our results show that expression of the CRISPR-Cas9 system successfully protects E. coli MG1655 from acquiring the targeted resistance genes by transformation or transduction with 2–3 logs of protection depending on the system for transfer and the target gene. Furthermore, we show that the system blocks conjugation of a set of clinical plasmids, and that the system is also able to protect the probiotic bacterium E. coli Nissle 1917 from acquiring AMR genes.