Deciphering the Role of Colicins during Colonization of the Mammalian Gut by Commensal E. coli

Abstract

Colicins are specific and potent toxins produced by Enterobacteriaceae that result in the rapid elimination of sensitive cells. Colicin production is commonly found throughout microbial populations, suggesting its potential importance for bacterial survival in complex microbial environments. Nonetheless, as colicin biology has been predominately studied using synthetic models, it remains unclear how colicin production contributes to survival and fitness of a colicin-producing commensal strain in a natural environment. To address this gap, we took advantage of MP1, an E. coli strain that harbors a colicinogenic plasmid and is a natural colonizer of the murine gut. Using this model, we validated that MP1 is competent for colicin production and then directly interrogated the importance of colicin production and immunity for MP1 survival in the murine gut. We showed that colicin production is dispensable for sustained colonization in the unperturbed gut. A strain lacking colicin production or immunity shows minimal fitness defects and can resist displacement by colicin producers. This report extends our understanding of the role that colicin production may play for E. coli during gut colonization and suggests that colicin production is not essential for a commensal to persist in its physiologic niche in the absence of exogenous challenges.

Keywords: DNA damage response; colicin; colonization; commensal E. coli.

Figure 1

The MP1 colicin is SOS (bacterial DNA damage stress response pathway) regulated and active. (A) Sequence analysis revealed two overlapping SOS boxes and an IscR binding site upstream of the start codon. (B) GFP expression from the colY promoter depends upon DNA damage. The kinetics of GFP expression mirror another SOS-controlled gene, dinB. Time-dependent induction of GFP is represented as fluorescence intensity normalized to optical density at 595 nm at the start of the experiment. The error bands show the standard deviations of results from four independent biological replicates for each condition. (C) Shown is the inhibition halo of E. coli K12 (MG1655) by cell lysates of wild-type MP1 generated either with or without UV exposure. Dilutions of cell lysates were made and indicated. Dashed circles indicate location of spotted lysate.

Figure 2

Colicin-deficient strain can establish colonization in healthy mouse gastrointestinal tract. (A) Shown is the inhibition halo of ∆colY strain exposed to 50 J/m² and extracted lysate spot plated on a lawn of E. coli MG1655-K12 and inhibition halo of wild-type strain exposed to 50 J/m² and extracted lysate spot plated on a lawn of ∆colY strain. Dilutions of cell lysates were made and indicated. Circles indicate location of spotted lysate. (B) Groups of 6- to 8-week-old C57BL/6 male mice were orally inoculated with either the wild-type strain (black squares) or the ∆colY strain (purple squares). On the days indicated, fecal samples were collected and CFU counts were determined. Each square represents one mouse and the limit of detection was 10³. Significant p values are noted (NS, not significant) and were calculated using a two-tailed unpaired Student’s t test. If colonies were too numerous to count on a plate, the corresponding animal was excluded from the data on that day.

Figure 3

Competition of ∆colY in the adult mouse gastrointestinal tract. (A) Schematic of the competition protocol. Mice were inoculated in a 1:1 ratio of either strain and fecal samples were collected at various days post-inoculation. Fecal samples were serially diluted and plated onto LB/tetracycline plates. A representative plate is shown. (B,C) Groups of 6- to 8-week-old C57BL/6 mice were co-inoculated. (B) Log competitive index (CI) was calculated by taking the ratio of the output colonies normalized to the input ratio. Each circle represents one specific animal. (C) CFU counts per gram of feces for the competition experiment. Each square represents the mean and standard deviation from all individual mice. Limit of detection was 10². Significant p values are noted (** p > 0.005) and were calculated using a one sample t-test. If colonies were too numerous to count on a plate, the corresponding animal was excluded from the data for that day.

Figure 4

Characterization and colonization of the adult mouse gut by a colicin-sensitive strain. (A) Inhibition halo of wild-type strain exposed to 50 J/m² and extracted lysate spot plated on a lawn of ∆colY ∆cyi strain. Dilutions of cell lysates were made and indicated above. Circles indicate location of spotted lysate. (B) Groups of 6- to 8-week-old C57BL/6 mice were inoculated with the colicin wild-type strain or the ∆colY ∆cyi strain. Significant p values are noted (NS, not significant. * p > 0.05) and were calculated using a two-tailed unpaired Student’s t test. (C,D) Groups of 6- to 8-week-old C57BL/6 mice were co-inoculated. (C) Log competitive index (CI) was calculated by taking the ratio of the output colonies normalized to the input ratio. Each circle represents one specific animal. Significant p values are noted (NS, not significant, ** p > 0.005) and were calculated using a one sample t-test. (D) CFU counts per gram of feces for the competition experiment. Each square represents the mean and standard deviation from all individual mice. Limit of detection was 10³. If colonies were too numerous to count on a plate, the corresponding animal was excluded from the data for that day.

Figure 5

Commensal strain invasion. (A,B) Groups of 6- to 8-week-old C57BL/6 mice were inoculated with either the wild-type strain or the ∆colY ∆cyi strain. On day 9, as represented by the down arrow, the mice were inoculated with the wild-type strain (A) or the ∆colY ∆cyi strain (B). Each square represents the mean and standard deviation from the mean of all individual mice. Limit of detection was 10². If colonies were too numerous to count on a plate, the corresponding animal was excluded from the data for that day. + indicates the wild-type strain was not recoverable from one mouse; ++ indicates the ∆colY ∆cyi strain was not detected from two mice by the termination of the experiment. Significant p values are noted (* p > 0.05, ** p > 0.005) and were calculated using a two-tailed unpaired Student’s t test.