Penile Anaerobic Dysbiosis as a Risk Factor for HIV Infection

Abstract

Sexual transmission of HIV requires exposure to the virus and infection of activated mucosal immune cells, specifically CD4⁺ T cells or dendritic cells. The foreskin is a major site of viral entry in heterosexual transmission of HIV. Although the probability of acquiring HIV from a sexual encounter is low, the risk varies even after adjusting for known HIV risk factors. The genital microbiome may account for some of the variability in risk by interacting with the host immune system to trigger inflammatory responses that mediate the infection of mucosal immune cells. We conducted a case-control study of uncircumcised participants nested within a randomized-controlled trial of male circumcision in Rakai, Uganda. Using penile (coronal sulcus) swabs collected by study personnel at trial enrollment, we characterized the penile microbiome by sequencing and real-time PCR and cytokine levels by electrochemiluminescence assays. The absolute abundances of penile anaerobes at enrollment were associated with later risk of HIV seroconversion, with a 10-fold increase in Prevotella, Dialister, Finegoldia, and Peptoniphilus increasing the odds of HIV acquisition by 54 to 63%, after controlling for other known HIV risk factors. Increased abundances of anaerobic bacteria were also correlated with increased cytokines, including interleukin-8, which can trigger an inflammatory response that recruits susceptible immune cells, suggesting a mechanism underlying the increased risk. These same anaerobic genera can be shared between heterosexual partners and are associated with increased HIV acquisition in women, pointing to anaerobic dysbiosis in the genital microbiome and an accompanying inflammatory response as a novel, independent, and transmissible risk factor for HIV infection.IMPORTANCE We found that uncircumcised men who became infected by HIV during a 2-year clinical trial had higher levels of penile anaerobes than uncircumcised men who remained HIV negative. We also found that having higher levels of penile anaerobes was also associated with higher production of immune factors that recruit HIV target cells to the foreskin, suggesting that anaerobes may modify HIV risk by triggering inflammation. These anaerobes are known to be shared by heterosexual partners and are associated with HIV risk in women. Therefore, penile anaerobes may be a sexually transmissible risk factor for HIV, and modifying the penile microbiome could potentially reduce HIV acquisition in both men and women.

Keywords: anaerobes; cytokines; foreskin inflammation; human immunodeficiency virus; penile microbiome; susceptibility.

FIG 1

Relationship between the absolute abundances of anaerobic bacteria and risk of HIV seroconversion in the case-control study. At study baseline, each 10-fold increase in the absolute abundance (measured in the log₁₀-transformed number of 16S rRNA gene copies per swab) of anaerobic bacteria Prevotella, Dialister, Finegoldia, Peptoniphilus, and Peptostreptococcus increased the odds of HIV infection. The risk for seroconversion further increased after adjustment for age, marital status, number of extramarital sexual partners, condom use, and genital discharge symptoms. Under each genus, the adjusted odds ratio (i.e., the mean adjusted odds of HIV seroconversion per 10-fold increase in absolute abundance of bacteria) and 95% confidence interval (in parentheses) are shown. Graphically, the 95% confidence interval for the unadjusted odds ratio is shown as purple bar for each genus, and the mean and 95% confidence interval of the adjusted odds ratio is shown as a whisker plot. Detailed results can be found in Table S2 in the supplemental material.

FIG 2

Relationship between absolute abundance of penile anaerobes and concentration of IL-8 in the case-control study. The relationship between anaerobe abundance, measured in the number of 16S rRNA gene copies per swab (log₁₀) and IL-8 concentration, measured in picograms per milliliter (log₁₀) is similar between men who seroconverted (cases) (red circles) and men who remained HIV negative (controls) (gray circles). The lines describe best-fit models based on AIC selection, which supported spline functions for all anaerobic bacteria except two anaerobic bacteria. Significant relationships (P < 0.05) are shown by solid red lines, and nonsignificant relationships are shown by solid black lines. To demonstrate the underlying data trend, polynomial spline models are shown by dashed black lines. Under the name of each genus, c represents the breakpoint used in the piecewise spline model, and pre- and post-c slopes are the slope before and after c, respectively; slopes that are significantly different from zero are indicated by asterisks as follows: *, P < 0.05; **, P < 0.01. Detailed results can be found in Table S3 in the supplemental material.