Increased HIV-1 infection in PBMCs treated in vitro with menstrual cycle phase hormones or medroxyprogesterone acetate likely occurs via different mechanisms

Abstract

Problem: Both luteal phase progesterone (P4) levels and use of the intramuscular (IM) injectable progestin-only contraceptive depo-medroxyprogesterone acetate (DMPA-IM) have been linked to increased S/HIV acquisition in animal, clinical and in vitro models. Several plausible mechanisms could explain MPA-induced HIV-1 acquisition while those for the luteal phase are underexplored.

Method of study: Peripheral blood mononuclear cells (PBMCs) were treated with P4 and estrogen at concentrations mimicking the luteal phase, follicular phase or with levels of MPA mimicking peak serum levels in DMPA-IM users. Cells were infected with an R5-tropic infectious molecular clone and HIV-1 infection was measured. A role for the glucocorticoid receptor (GR) was investigated using the GR/PR antagonist RU486. CCR5 protein levels and activation status, assessed by levels of the activation marker CD69, were measured by flow cytometry after treatment in vitro and in PBMCs from naturally-cycling women or DMPA-IM users.

Results: Both MPA and luteal phase hormones significantly increased HIV-1 infection in vitro. However, MPA but not luteal phase hormones increased the CD4+/CD8+ T cell ratio, CCR5 protein expression on CD4+ T cells and increased expression of the activation marker CD69. The GR is involved in MPA-induced, but not luteal phase hormone-induced increased HIV-1 infection. In DMPA-IM users, the frequency of CCR5-expressing CD3+ and CD8+ cells was higher than for women in the luteal phase.

Conclusions: MPA increases HIV-1 infection in a manner different from that of luteal phase hormones, most likely involving the GR and at least in part changes in the frequency and/or expression of CCR5 and CD69.

Keywords: CCR5; HIV-1; PBMCs; glucocorticoid receptor; luteal phase; medroxyprogesterone acetate; menstrual cycle.

FIGURE 1

HIV‐1 replication in blood bank peripheral blood mononuclear cells (bbPBMCs) increases upon treatment with luteal phase estrogen (E2) + progesterone (P4) and medroxyprogesterone acetate (MPA). bbPBMCs were treated for 2 or 7 days, then infected with R5‐tropic HIV‐1 for 2 h. After 5 days in ligand‐free media, cells were harvested for Renilla luciferase (infection) or for cell viability (MTT). Pooled data are from at least three independent experiments, with two to three PBMC donors each, for a total of seven independent donors. Infection was plotted as mean + standard error of the mean (SEM) relative to each day's vehicle control in the presence of virus set to 100%. Statistical comparisons were carried out using a two‐way ANOVA with Dunnett's multiple comparisons post‐test. Stars above bars indicate significance compared to own vehicle control in the presence of virus, with **** indicating p < .0001

FIGURE 2

MPA‐increased but not luteal phase‐increased HIV‐1 infection is mediated by the glucocorticoid receptor (GR). (A) bbPBMCs were treated for 7 days with the indicated hormones, either in the absence or presence of 100 nM RU486, then infected as in Figure 1. The results show pooled data from at least four independent experiments, with two to three bbPBMC donors each, for a total of 18 independent donors for –RU486, including n = 7 from Figure 1 to 11 independent donors for +RU486, performed in parallel. Infection was plotted relative to the vehicle controls in the presence of virus set to 100%. (B) A representative western blot indicating GR and GAPDH protein levels from matched PBMC samples at Day 0, 2, 7, and 12 post‐treatment with the indicated hormones. (C) Western blots were scanned and quantified for relative GR levels by calculating integrated density values from n = 7 independent donors. Statistical comparisons were carried out using a two‐way ANOVA with multiple comparisons post‐tests, indicated by stars (*), or a Wilcoxon test, indicated by letters (a–d). Stars or letters above bars indicate significance for the comparison indicated by lines, with ****/a, ***/b, **/c and */d indicating p < .0001, p < .001, p < .01, and p < .05 respectively

FIGURE 3

MPA, unlike luteal or follicular phase E2+P4, increases expression of CCR5 on CD3+ and CD4+ T cells and expression of CD69 on CD4+ T cells. bbPBMCs were treated for 7 days with the indicated hormones then stained with conjugated antibodies specific for the relevant markers, fixed and then sorted by flow cytometry. Samples were gated as shown in Supplementary Figure S1. The results are pooled from 2 independent experiments, with three to four donors each, for a total of seven to eight independent donors. Data are plotted as frequency (A, B) or median fluorescent intensity (MFI) (C, D) relative to the vehicle control set to 1. Statistical comparisons were carried out using a parametric one‐way ANOVA with Dunnett's multiple comparisons post‐test or paired t‐tests or a non‐parametric Kruskal‐Wallis test with Dunn's multiple comparisons post‐test or paired t‐tests. Stars above bars indicate significance compared to own vehicle control, unless between groups as shown by lines, with ****, ***, ** and * indicating p < .0001, p < .001, p < .01, and p < .05 respectively. Some data (Vehicle vs. MPA) has previously been published in a different context but are included here in the context of comparison to luteal and follicular phase data

FIGURE 4

MPA but not luteal or follicular phase E2+P4 increases the CD4/CD8 ratio in CD3+, CD69+ and CCR5+ cells. bbPBMCs were treated and stained as indicated in Figure 3. Data are plotted as MFI (A), frequency (B) or CD4/CD8 ratio in CD3+ (C), CD69+ (D) or CCR5+ (E) cells relative to the vehicle control set to 1. Statistical comparisons were carried out using a parametric one‐way ANOVA with Dunnett's multiple comparisons post‐test or a non‐parametric Kruskal‐Wallis test with Dunn's multiple comparisons post‐test. Stars above bars indicate significance compared to own vehicle control, with ****, ***, ** and * indicating p < .0001, p < .001, p < .01 and p < .05 respectively. Some data (Vehicle vs. MPA) has previously been published in a different context but are included here in the context of comparison to luteal and follicular phase data

FIGURE 5

The frequency of CCR5‐expressing cells is higher in CD3+ and CD8+ T cells of intramuscular depo‐medroxyprogesterone acetate (DMPA‐IM) users compared to women in the luteal phase. Archived ex vivo PBMCs taken at time of hysterectomy (hPBMCs) from women in the luteal phase (n = 7) or follicular phase (n = 5) or DMPA‐IM users (n = 9) were thawed and stained for the indicated markers by flow cytometry. Data are plotted as non‐normalized raw values of frequency (A, C) or MFI (B, D) of the indicated marker/s. Statistical comparisons were carried out using a one‐way ANOVA with Tukey's multiple comparisons post‐test or unpaired t tests or Mann Whitney t tests. Stars show significance between the indicated groups, with ** and * indicating p < .01 and p < .05 respectively