Background: Adult male circumcision (MC) reduces HIV acquisition by >50% in heterosexual men, but the mechanism of this protection is not completely understood. Recent evidence showed that MC reduces penile pro-inflammatory anaerobes; and since HIV preferentially infects activated CD4 T cells, this may represent one protective mechanism of MC. We hypothesized that MC reduces activated CD4 T cells in penile tissue; however, penile tissue cannot be biopsied post circumcision. We therefore first identified cytokines in penile swabs that correlate with the number of activated CD4 T cells in foreskin tissue. We measured these cytokines in swabs collected in a large, randomized trial of MC pre- and post-circumcision as a surrogate of changes in activated CD4 T cell density following circumcision. Methodology: Foreskin tissue and coronal sulcus swabs were collected in a cross-sectional study of 89 men undergoing elective MC. Tissue density of HIV-target cells was determined with a combination of flow cytometry and immunofluorescence; levels of 14 cytokines and chemokines were measured in swabs with an electrochemiluminescent multiplex ELISA. We then measured cytokine and chemokine levels in swabs collected from 111 men randomized to circumcision and 107 uncircumcised controls during a clinical trial of MC in Rakai, Uganda. Swabs were collected at enrolment and at 6, 12 and 24 months, and cytokine levels were compared between control and intervention men using generalized linear modeling. Results: In the cross-sectional study, we found that levels of IL-8 in coronal sulcus swabs correlated with the tissue density of highly HIV-susceptible CD4 T cells, including Th17, Th1 and CCR5+ CD4 T cells (all p<0.05), but not Tregs. In the longitudinal study, pre-circumcision cytokine levels were similar between control and intervention men, but after 6 months MC was associated with a reduction in IL-8 relative to controls (adjusted PRR 0.67, 95% CI: 0.52, 0.87), and this differential increased over the 2 years of follow-up. No cytokine changes were observed in uncircumcised men, and MC was not associated with changes in vaginal cytokines among female sexual partners. Conclusions: Levels of IL-8 on the foreskin surface correlates with the density of highly HIV-susceptible CD4 T cell populations in the underlying foreskin tissue. Male circumcision is associated with reduced IL-8 on the coronal sulcus, suggesting reduced HIV acquisition after MC may be partly due to a reduction in penile inflammation, and thus HIV-susceptible cells, in remaining penile tissue.