. Bacteria on and within the body (the microbiota) influence human physiology, therapeutic responses, and dis- ease states. Cyclomodulins are bacterial toxins and effectors that modulate eukaryotic cell cycle progression, proliferation, differentiation, or apoptosis, and may be genotoxic. Certain strains of E. coli in the human gut contain a gene cluster (referred to as ?clb?) that encodes small molecule cyclomodulins known as precolibac- tins. Evidence suggests precolibactins are prodrugs that are converted to cytotoxins (colibactins) by a dedi- cated peptidase (colibactin peptidase, ClbP). clb+ E. coli induce DNA double-strand breaks in mammalian cells in vitro and in vivo, suggesting these molecules are trafficked (by an unknown mechanism) to eukaryotic cells, and initiate tumor formation in colitis-susceptible mice treated with azoxymethane. Several independent stud- ies have demonstrated that the clb cluster is epidemiologically correlated with colorectal cancer in humans. As colibactins are unstable, all isolation efforts have employed clbP deletion strains to facilitate accumulation of the more stable precolibactins. We developed convergent high-yielding syntheses of linear precolibactin bio- synthetic precursors and showed they transform to unsaturated imines after ClbP deacylation; these imines alkylate DNA by nucleotide addition to an electrophilic cyclopropane. Structure?function studies established distinct DNA recognition and prodrug domains. Of equal significance, our data indicate that the use of clbP deletion strains results in the production of alternative, non-genotoxic structures, such as precolibactins A?C. Precolibactin-886 is the most complex clb isolate known and is the first that contains an ?-aminomalonate resi- due, which is believed to be important for cytopathic effects. We hypothesize that the unusual macrocyclic structure of precolibactin-886 also derives from employment of a clbP deletion strain. To test this we will pre- pare precolibactin-886 and key synthetic derivatives/biosynthetic precursors and elucidate their chemistry. We will determine if deacylation of the linear precursor to precolibactin-886 leads to production of similar electro- philic imines. We will evaluate the potency, cell cycle effects, and DNA-damaging abilities of synthetic colibac- tins and controls in a zebrafish model. Using enzymology, genetic deletion studies, and X-ray crystallography, we will elucidate the roles of the enzymes ClbL, ClbO, ClbM and ClbS, which are encoded in the clb cluster but do not have well-defined functional roles. The latter two enzymes phenotypically contribute to colibactin re- sistance and their study may illuminate methods to inhibit clb+ E. coli-associated colorectal cancer. This grant employs four investigators with non-overlapping expertise in chemical synthesis, natural products biosynthesis and isolation, preclinical studies of clb+ E. coli in vitro and in vivo, and enzymology and protein crystallography. This work will establish a mechanistic model that accounts for all known precolibactins, define the molecular mechanisms by which certain E. coli induce carcinogenesis, and inform strategies to inhibit clb+ E. coli-driven tumorigenesis. These studies will provide insights into the functional roles of non-proteiogenic cyclomodulins.
. Select strains of Escherichia coli in the human gut produce small molecules known as precolibactins, which are precursors to tumorigenic colibactins. This grant presents a plan to synthesize and study the most advanced (pre)colibactins in vitro and in vivo, and to elucidate the molecular mechanisms by which these E. coli drive tumor formation. This work may ultimately illuminate new therapeutic strategies to inhibit bacteria-associated colorectal cancers.