Hormonal Contraception and HIV-1 Acquisition: Biological Mechanisms

Abstract

Access to effective affordable contraception is critical for individual and public health. A wide range of hormonal contraceptives (HCs), which differ in composition, concentration of the progestin component, frequency of dosage, and method of administration, is currently available globally. However, the options are rather limited in settings with restricted economic resources that frequently overlap with areas of high HIV-1 prevalence. The predominant contraceptive used in sub-Saharan Africa is the progestin-only three-monthly injectable depot medroxyprogesterone acetate. Determination of whether HCs affect HIV-1 acquisition has been hampered by behavioral differences potentially confounding clinical observational data. Meta-analysis of these studies shows a significant association between depot medroxyprogesterone acetate use and increased risk of HIV-1 acquisition, raising important concerns. No association was found for combined oral contraceptives containing levonorgestrel, nor for the two-monthly injectable contraceptive norethisterone enanthate, although data for norethisterone enanthate are limited. Susceptibility to HIV-1 and other sexually transmitted infections may, however, be dependent on the type of progestin present in the formulation. Several underlying biological mechanisms that may mediate the effect of HCs on HIV-1 and other sexually transmitted infection acquisition have been identified in clinical, animal, and ex vivo studies. A substantial gap exists in the translation of basic research into clinical practice and public health policy. To bridge this gap, we review the current knowledge of underlying mechanisms and biological effects of commonly used progestins. The review sheds light on issues critical for an informed choice of progestins for the identification of safe, effective, acceptable, and affordable contraceptive methods.

Figure 1.

Schematic diagram illustrating steroid receptor selectivity of progestins and intracellular mechanism of regulation of transcription by steroid receptors. (a) Binding and transcriptional activity. Solid black lines indicate high-affinity binding similar to or greater than that of the endogenous cognate agonist ligands. These are progesterone (Prog) for the PR, cortisol (Cort) for the GR, dihydrotestosterone (DHT) for the AR, and aldosterone (Ald) for the MR. Stippled black line indicates lower affinity (∼10-fold lower), whereas dotted black lines indicate very low relative affinity (at least 100-fold lower) compared with the endogenous agonist ligands. Plus signs (+) indicate agonist or partial agonist activity, where four plus signs (++++) indicate full agonist activity similar to endogenous agonist ligands and fewer plus signs indicate relative partial agonist activity. Minus signs (−) indicate antagonist activity toward the endogenous agonist, where four minus signs (−−−−) indicate full antagonist activity and fewer minus signs indicate relative partial antagonist activity. Data sources include references listed in Tables 1 and 2 and references discussed in the text. Note that there is considerable uncertainly and/or inconsistency in some of the data as discussed in the text, especially for affinities and transcriptional activities of LNG and ETG for GR, AR, and MR. Additionally, steroid receptor activity can be gene and cell specific, and many of the activities in animal preclinical models may not represent activities in human tissues. (b) Mechanisms of transcriptional regulation by MPA acting via the GR. Other progestin agonists and partial agonists can both transactivate and transrepress genes via their cognate receptors using similar mechanisms. Green arrows indicate increase in transcription of target genes (transactivation by direct binding of receptor to DNA) whereas red stop lines indicate inhibition (transrepression by tethering of receptor to other transcription factors), both taking place in the nucleus. Chromatin effects are not indicated. Antagonist activity is not depicted but involves recruitment of corepressors instead of coactivators by DNA-bound receptors, due to the conformation induced by the antagonist bound to the receptor. Additional mechanisms of gene regulation may be involved. Ald, aldosterone; Cort, cortisol; DHT, dihydrotestosterone; Prog, progesterone. With data from Stanczyk et al. (24) and Africander et al. (25).

Figure 2.

Typical sigmoidal dose–response curves for progestins acting via steroid receptors. (a) Curves depicting a stimulatory response (such as transactivation) for a full agonist (100% efficacy) and a partial agonist (60% efficacy). The potency, or the concentration of steroid for half-maximal response (EC₅₀ value), of 1 nM is indicated for the full agonist by a dotted orange line. (b) Curves depict three separate stimulatory responses with potencies or EC₅₀ values of 0.1, 1, or 10 nM (a, b, and c, respectively). An orange arrow illustrates how the curves shift to the left as the EC₅₀ decreases and potency increases, which occurs with increased progestin steroid receptor concentration. (c) The curve shows a stimulatory response for an agonist with orange dotted lines to indicate how the percentage point response changes as the concentration of progestin changes by 0.3-fold, for different parts of the curve. (a–c) The slopes of the sigmoidal curves are set at 1, assuming one steroid ligand binding reversibly to one receptor, the absence of cooperative effects on ligand binding, and no spare receptors.

Figure 3.

Established biological mechanisms of MPA potentially affecting acquisition of HIV-1 in the FGT. The figure depicts the mechanisms of MPA action for which experimental evidence has been obtained (Tables 3–5). Multiple effects of MPA are consistent with actions via steroid receptors that regulate multiple genes and biological functions. (1) Decreased levels of soluble protective mediators on the mucosal surface, including antimicrobial and antiviral factors and antibodies. (2) Increased secretion of inflammatory mediators that may recruit and activate HIV-1 target cells and immune function cells involved in defense. Decreased expression of some mediators that activate immune function, causing immunosuppression. (3) Changes in the composition of vaginal microbiota and production of soluble factors by bacteria. (4) Increased infection with other viruses such as HSV-2. (5) Increased frequency of HIV-1 target cells in vaginal lumen, including CCR5⁺CD4⁺ T cells. (6) Barrier integrity and epithelial protective function. Decreased expression of proteins involved in barrier integrity, increased barrier permeability. (7) Transcytosis. Increased uptake and transcytosis of HIV-1 via single-layered columnar epithelium of the endocervix and upper FGT. (8) Increased frequency of tissue-resident target cells and elevated CCR5⁺ expression on CD4⁺ T cells. Increased frequency of macrophages and decreased frequency of regulatory T cells in endometrium. (9) Innate immune response. Decreased early innate immune responses in FGT mediated by pDCs. (10) Adaptive immune response. Decreased antigen presentation by professional antigen-presenting cells and decreased induction and maintenance of T and B cell responses. (11) Increased proliferation of HIV-1 in target T cells.

Figure 4.

Mean circulating estradiol levels in premenopausal women on HCs. Mean circulating estradiol in (a) normal premenopausal contracepting women (N = 78); (b–f) premenopausal women using progestin-only contraceptives (b, Mirena, N = 8625; c, Implanon, N = 1126; d, Jadelle, N = 8827; e, NET-EN, N = 7323; f, DMPA, N = 31); and normal postmenopausal women (g, N = 1446). The graphs represent mean values with lines showing 95% confidence intervals (225). Reproduced with permission from Wolters Kluwer Health.