Current epidemiological data showing that the use of the injectable contraceptive progestin Depotmedroxyprogesterone acetate (DMPA) is associated with increased HIV-1 acquisition is controversial. However, animal and ex vivo data reveal plausible biological mechanisms whereby MPA may increase HIV-1 acquisition. Relatively high levels of endogenous progesterone (P4) found in the luteal phase of the menstrual cycle have also been linked to increased HIV-1 acquisition in animal, clinical and ex vivo models. One of the central hypotheses of the present study was that the mechanism of MPA-induced increase in HIV-1 infection occurs via a different mechanism to that of the luteal phase. Furthermore, MPA has been shown to activate both the glucocorticoid receptor (GR) and its target, the progesterone receptor (PR) isoform B (PR-B), which are both transcription factors and regulate genes involved in immune function. Both the GR and PR are expressed in the cervix, the primary site of heterosexual HIV-1 infection. PR is regulated by endogenous estrogen (E2), of which the concentrations fluctuate throughout the menstrual cycle, and GR expression also varies in response to stress hormones, leading to conditions of varied relative levels of GR/PR. The immune-related consequences of changing the relative levels of GR and PR-B are not well understood. Therefore another hypothesis of this study was that changing the relative levels of GR/PR-B modulates HIV-1 infection and immunomodulatory gene expression in response to the GR/PR agonist, MPA. Since GR and PR-B recognize similar DNA target sequences and may regulate the same genes at the same time, the final hypothesis of the present study was that GR and PR-B reciprocally modulate each other’s activity, through possible association. To investigate the effects of exogenous hormones on HIV-1 infection and mechanisms thereof, peripheral blood mononuclear cells (PBMCs) and TZM-bl cervical cells were used as model systems for HIV-1 infection. These cells were stimulated with P4 and E2 at concentrations mimicking the menstrual cycle phases or with levels of MPA at the upper range of peak serum levels detected in DMPA users. Cells were infected with the R-tropic HIV-1 infectious molecular clone, HIV-1Bal_Renilla and luciferase assays were used to measure HIV-1 infection. Levels of HIV-1 CD4 receptor and CCR5 co-receptor protein or mRNA were measured by flow cytometry or qPCR, respectively, while activation of CD4+ T cells using the activation marker CD69 was measured by flow cytometry in PBMCs. To investigate the effects of changing GR/PR-B levels on HIV-1 infection and immune gene regulation, GR/PR levels were altered in End1/E6E7 immortalized endocervical and HeLa/TZM-bl cervical carcinoma cells by GR siRNA knockdown with or without the simultaneous over-expression of PR-B, and cells were stimulated with MPA or the GR agonist Dexamethasone. mRNA expression iii of key immunomodulatory genes in End1/E6E7 and HeLa cells was measured by qPCR. The modulation of GR activity by PR-B was assessed by promoter-reporter assay in COS1 and U2OS cells over-expressing GR and PR and stimulated with GR- and/or PR-specific ligands. Association of GR and PR-B was measured by co-immunoprecipitation in COS1 and MCF-7 cells, while co-localization of GR and PR-B was measured by confocal microscopy and super-resolution structured illumination microscopy in COS1 cells. MPA significantly increased HIV-1 infection in both PBMCs and TZM-bl cells, while luteal phase hormones did so to a lesser extent. However, MPA but not luteal phase hormones increased the ratio of CD4+/CD8+ T cells in PBMCs. MPA but not luteal phase hormones also increased CCR5 protein expression on CD4+ T cells in PBMCs and total CCR5 mRNA expression in TZM-bl cells. In addition, MPA but not luteal phase hormones increased activation of CD4+ T cells in PBMCs. Using a GR antagonist or GR siRNA, it was shown that the GR but not PR-B is required for MPA-, but not luteal phase hormone-induced increased HIV-1 infection in PBMCs and TZM-bls. The presence of PR-B altered the anti-inflammatory, GR-mediated regulation of some key immunomodulatory genes, including GILZ and IL-6, in End1/E6E7 and HeLa cells in response to MPA. In general, basal (unliganded) expression of immunomodulatory genes exhibited a pro-inflammatory profile in the presence of PR-B. Co-immunoprecipitation assays showed that GR and PR-B appeared to associate. Confocal microscopy suggested GR and PR co-localized in the nucleus in response to GR- and/or PRspecific ligands, while super-resolution microscopy showed that co-localization occurred in select regions within the nucleus. Taken together, MPA increases HIV-1 infection in a manner different from that of luteal phase hormones, most likely involving increased CD4+ T cell frequency (CD4+/CD8+ ratio), activation and increased expression of CCR5 on CD4+ T cells, and requiring the GR. Furthermore, PR-B modulates GR-mediated immune function gene regulation, via potential association and region-specific nuclear co-localization. This suggests that the relative levels of GR/PR may play an important role in determining the inflammatory and immune responses and HIV-1 infection in HIV-1 target cells, both in DMPA users and women not using hormonal contraception.

Objectif : Étudier l'association entre l’utilisation de contraceptifs hormonaux et le risque d'acquisition du VIH-1 chez les femmes au Malawi, en Afrique du Sud, en Zambie et au Zimbabwe. Devis : Analyses secondaires de 2887 femmes âgées de 17-55 ans ayant participé à l’étude HPTN 035, une étude de phase II/IIb sur l’efficacité de deux gels microbicides pour prévenir la transmission du VIH chez les femmes à risque. Méthodes : L'association entre l'utilisation de contraceptifs hormonaux et le risque d'acquisition du VIH-1 a été évaluée en utilisant des modèles de Cox. Des risques relatifs sont estimés où le groupe de référence est celui des femmes qui n’utilisent pas de contraceptifs hormonaux. De plus, un modèle multivarié de Cox est utilisé afin de contrôler pour les facteurs potentiellement confondants. Résultats : Les contraceptifs injectables ont été utilisés par 52,1% des femmes, alors que les contraceptifs oraux ont été utilisés par 20,7% de celles-ci. Pendant l'étude, il y a eu 192 séroconversions. L'incidence observée du VIH était de 2,28; 4,19 et 4,69 pour 100 personne-années pour les contraceptifs oraux, injectables et non hormonaux, respectivement. Lors de l’analyse multivariée, nous n'avons trouvé aucune association significative entre l’usage des contraceptifs hormonaux et l’acquisition du VIH-1. Le risque relatif ajusté (RRa) pour les contraceptifs oraux est de 0,573 (IC de 95% : [0,31-1,06]) et 0,981 (IC de 95% : [0,69 ; 1,39]) pour les contraceptifs injectables. Conclusions : Bien que cette étude ne démontre pas d’association entre l’usage des contraceptifs hormonaux et le VIH-1, nous concluons toutefois que ces méthodes de contraception ne protègent pas contre le VIH-1, et il est ainsi recommandé aux femmes utilisant des hormones contraceptives de toujours utiliser le condom pour prévenir l'infection au VIH-1. / Objective: To investigate the association between the use of hormonal contraceptive and the risk of acquiring HIV-1 infection in women from Malawi, South Africa, Zambia and Zimbabwe. Design: Secondary analyses of 2887 women aged 17-55 years who participated in the HPTN 035 trial, a Phase II/IIb trial on the efficacy of two microbicide gels to prevent HIV transmission in women at risk in Africa. Methods: The association between the use of hormonal contraceptive and the risk of acquiring HIV-1 was evaluated using Cox proportionnal models. Relative risks of exposed women were estimated using as a reference group the women who do not use hormonal contraceptives. In addition, a multivariate Cox model was used to control for potentially confounding factors. Results: Injectable contraceptives were used by 52.1 % of women, while oral contraceptives were used by 20.7% of them. During the study, there were 192 seroconversions. The observed HIV-1 incidence was 2.28, 4.19 and 4.69 per 100 woman-years for oral, injectable and non-hormonal contraceptive users, respectively. In multivariate analysis, we found no significant association between the use of hormonal contraceptives and HIV-1 acquisition. The adjusted relative risk (aRR) for oral contraceptives was 0.573 (95% CI: [0.31 to 1.06]) and 0.981 (95% CI: [0.69, 1.39]) for injectable contraceptives. Conclusions: Although this study did not demonstrate an association between hormonal contraceptive use and the risk of HIV-1 infection, we conclude, however, that these methods of contraception do not protect against HIV-1, and it is thus recommended that women using contraceptive hormones always use condoms to prevent HIV-1.