Experimental genital tract infection demonstrates Neisseria gonorrhoeae MtrCDE efflux pump is not required for in vivo human infection and identifies gonococcal colonization bottleneck

Abstract

The MtrCDE efflux pump of Neisseria gonorrhoeae exports a wide range of antimicrobial compounds that the gonococcus encounters at mucosal surfaces during colonization and infection and is a known gonococcal virulence factor. Here, we evaluate the role of this efflux pump system in strain FA1090 during in vivo human male urethral infection with N. gonorrhoeae using a controlled human infection model. With the strategy of competitive infections initiated with mixtures of wild-type FA1090 and an isogenic mutant FA1090 strain that does not contain a functional MtrCDE pump, we found that the presence of the efflux pump is not required for an infection to be established in the human male urethra. This finding contrasts with previous studies of in vivo infection in the lower genital tract of female mice, which demonstrated that mutant gonococci of a different strain (FA19) lacking a functional MtrCDE pump had a significantly reduced fitness compared to their wild-type parental FA19 strain. To determine if these conflicting results are due to strain or human vs. mouse differences, we conducted a series of systematic competitive infections in female mice with the same FA1090 strains as in humans, and with FA19 strains, including mutants that do not assemble a functional MtrCDE efflux pump. Our results indicate the fitness advantage provided by the MtrCDE efflux pump during infection of mice is strain dependent. Owing to the equal fitness of the two FA1090 strains in men, our experiments also demonstrated the presence of a colonization bottleneck of N. gonorrhoeae in the human male urethra, which may open a new area of inquiry into N. gonorrhoeae infection dynamics and control. TRIAL REGISTRATION. Clinicaltrials.gov NCT03840811.

Fig 1. Results of competitive infections in men and mice.

Panel A shows competitive infections of men with mixtures of FA1090 and FA1090ΔmtrD. Twelve men were competitively inoculated with mixtures of FA1090 and FA1090ΔmtrD. Participants were observed for signs and symptoms of clinical urethritis up to 5 days post-inoculation and received antibiotics as soon as urethral discharge developed or on the last day of study follow-up (day 5 post inoculation). First void urines were obtained daily after inoculation until the final study day and quantitatively cultured to determine infectivity. Inocula were also quantitively cultured. Ten men became infected and developed gonococcal urethritis with urethral discharge and their data were included in final analyses presented here. Colonies from the quantitative cultures of inocula and daily urine specimens were enumerated and up to 96 single colony isolates per culture per day were picked and stored until strain determination by real-time PCR. The colony real-time PCR results from inocula and daily urine were used to calculate daily competitive indices using the formula: CI = mutant cfu (output)/wild-type cfu (output) ÷ mutant cfu (input)/wild-type cfu (input). Output refers to the number of wild-type and mutant cfu enumerated from cultures of urine sediment or from cultures of mouse vaginal swab suspension. Input refers to the number of cfu enumerated from culture of bacterial suspension used to inoculate that cohort/group of men or mice. Thus, CIs reported for each participant and mouse were calculated with the exact proportion of strains identified in the inoculum of the group/cohort they were a part of. CIs are graphed on the logarithmic scale. Log₁₀CI greater than 0 indicated the mutant was favored and Log₁₀CI less than 0 indicates that the wild-type was favored. Horizontal bars represent median CIs. White squares refer to men from whom only the wild-type was recovered. Grey squares denote men from whom a mix of the two strains was recovered. Black squares denote men from whom only the mutant strain was recovered. White, grey, or black squares with red borders refer to the treatment day results when the participant became symptomatic (clinical urethritis with urethral discharge) and/or was treated with antibiotics because it was the day of post-inoculation observation (day 5). Panels B, C, and D show the results of the mouse competitive infections. Eleven mice were inoculated with mixtures of FA1090 and FA1090ΔmtrD (B), 7 mice were inoculated with mixtures of FA19 and FA19ΔmtrD (C), and 6 mice were inoculated with mixtures of FA19 and FA19mtrD::kan (D). Mouse CIs were calculated using the same formula as for the men and are also graphed on the logarithmic scale with horizontal bars denoting the median CIs. For mice inoculated with mixtures containing deletion strains (B and C), a total of 48 single colonies recovered from mouse vaginal swabs collected on day 1 after inoculation and then every other day (day 1, day 3, and day 5) until the last day of positive cultures (all mice cleared the infection by day 7; last day of positive cultures for all mice was day 5) were picked and individually screen by real-time colony PCR to determine the strain for each colony. For competitive infections with FA19 and FA19mtrD::kan, strain determination was done with quantitative selective culture by culturing inocula and vaginal swab suspensions on GC agar containing streptomycin (to recover and enumerate total gonococci) and GC agar containing streptomycin and kanamycin (to recover and enumerate mutant gonococci). We divided the number of kanamycin resistant cfu by the total number of gonococci recovered. Longitudinal strain composition data are available from all mouse competitive infections, except for 1 of the 2 cohorts of mice inoculated with FA1090 and FA1090ΔmtrD (n = 4 mice, mice 1–4, Panel B), for whom only 96 single colonies from the last day of positive cultures were saved for real-time colony PCR screening; for the remaining 7 mice inoculated with FA1090 and FA1090ΔmtrD (mice 5–11) day 1, day 3, and day 5 colony real-time PCR data are available. White squares refer to mice from whom only the wild-type was recovered on last day of positive cultures. Grey squares denote mice from whom a mix of the two strains was recovered. Black squares denote mice from whom only the mutant strain was recovered. White, grey, or black squares with red borders refer to the results from the final day of positive cultures before the mouse cleared the bacteria. Note the different y axis scale for Panel D, imposed by the lower limit of detection of quantitative selective culture on GC agar for the FA19+FA19 mtrD::kan competitive infections.

Fig 2. Gonococcal strain dynamics during the early course of infection.

Panel A. Theoretical infection outcomes with respect to strain composition for infections initiated from 50:50 mixtures of two strains, shown in black and white, that differ in their competitive advantage in scenarios of bottleneck present early in the infection or no bottleneck (adapted with permission from [31]. Each pie chart represents a gonococcal population containing predicted proportions of the two theoretical strains. When the two strains have equal fitness and a bottleneck is present early in the course of infection that results in a population size restriction, we predict that equal numbers of outcomes favoring each strain will be observed. Panel B. Empirically observed outcomes in the current study for infections initiated with mixed inocula. Two to four men were inoculated with the same inoculum preparation in each of the four study cohorts. A total of 10 evaluable participants were available for final analysis (281, 291, 292, 300, 301, 302, 303, 311, 313, 314). The pie chart representing the average inoculum from the four cohorts refers to the mean strain composition of the four independent inocula. Each participant pie chart represents the daily mixture of gonococcal strains recovered from the urine over the 1- to 5-day period after inoculation. Last pie chart in the series for each participant refers to the study treatment day when clinical urethritis was apparent and/or when antibiotic treatment was administered.