Soluble CD52 mediates immune suppression by human seminal fluid

Abstract

Seminal fluid provides for the carriage and nutrition of sperm, but also modulates immunity to prevent allo-rejection of sperm by the female. Immune suppression by seminal fluid has been associated with extracellular vesicles, originally termed prostasomes, which contain CD52, a glycosylated glycophosphoinositol-anchored peptide released from testicular epithelial cells. Previously, we reported that human T cell-derived CD52, bound to the danger-associated molecular pattern protein, high mobility group box 1 (HMGB1), suppresses T cell function via the inhibitory sialic acid-binding immunoglobulin-like lectin-10 (Siglec-10) receptor. Here we show that human seminal fluid contains high concentrations of CD52 complexed with HMGB1, which mediates T cell suppression indirectly via Siglec-7 on antigen-presenting cells. Proliferation of natural killer (NK) cells, which express Siglec-7 and play a key role in the immune defence of the uterus, was directly suppressed by seminal fluid CD52. These findings elucidate a critical function of seminal fluid to suppress cellular immunity and facilitate reproduction.

Keywords: CD52; HMGB1; NK cell; Siglec-7; T cell; seminal fluid.

Figure 1

Seminal fluid contains high concentrations of CD52. Individual seminal fluid samples (n=25) were diluted as shown and assayed for CD52 by ELISA. Inset: titration of recombinant human CD52-Fc.

Figure 2

Seminal fluid suppresses T-cell proliferation and function. (A) Proliferation of CD4⁺ T cells in response to tetanus in the absence or presence of 26 individual, serially diluted seminal fluid samples. PBMCs were labelled with CFSE dye and incubated for 7 days in the absence or presence of tetanus toxoid (10 LFU/ml) and seminal fluid samples. After staining for CD4, divided (CFSE^dim) CD4⁺ cells were expressed as the cell division index. (B) IFN-γ ELISpots in PBMCs in response to tetanus in the absence and presence of individual seminal fluid samples. PBMCs were incubated in triplicate in an ELISpot plate coated with anti-IFN-γ antibody, for 18 h at 37°C in 5% CO₂ air, in the presence of tetanus (10 LFU/ml) and seminal fluid samples. CD52-Fc (10 μg/ml) was included as a positive control. Data are mean ± SD.

Figure 3

Blocking or depletion of CD52 prevents T-cell suppression by seminal fluid. (A) CD4⁺ T-cell proliferation in PBMCs measured by CFSE dye dilution (see Figure 2A ) in response to tetanus in the presence of seminal fluid and either IgG3 isotype control or CF1D12 (10 μg/ml) anti-CD52 glycan antibody. (B) Proliferation of Jurkat T cells incubated in triplicate for 48 h in the presence of seminal fluid (final dilution 1:40) ‘depleted’ by either agarose-bound human control IgG or humanized anti-CD52 antibody, alemtuzumab. ³H-thymidine was added for the last 16 h of incubation, and the cells then washed and analysed by scintillation counting. Data are mean ± SD. (C) IFN-γ ELISpots in PBMCs in response to tetanus (see Methods) in the presence of seminal fluid (final dilution 1:40) ‘depleted’ by either agarose-bound human control IgG or alemtuzumab anti-CD52 antibody. Data are mean ± SD. Individual seminal fluid samples are numbered.

Figure 4

Siglec-7 mediates immune suppression by seminal fluid. (A) Raji human B-cell lymphoma cells (1x10⁵ per flat bottom well) were incubated in triplicate with CD52-Fc (10 μg/ml) or seminal fluid samples (1/40 dilution) +/- anti-Siglec 10, -9 or -7 antibodies (10 μg/ml) for 9 days. ³H thymidine was added to wells during the last 16 hours of incubation; the cells were then collected on glass fibre filters, washed, dried and counted in scintillant in a beta-counter. (B) PBMCs were incubated in triplicate with tetanus (10 LFU/ml) and seminal fluid (1/40 dilution) +/- anti-Siglec-7, -9 or -10 antibody (10 μg/ml) in ELISpot plates pre-coated with IFN-γ antibody, for 24 h at 37°C in 5% CO₂ air, before development of IFN-γ spots. Data are mean ± SD.

Figure 5

Seminal fluid does not suppress the function of purified CD4⁺ T cells. CD4⁺ T cells (5x10^3/well), >96% pure by negative immunomagnetic selection, were incubated with anti-CD3/CD28 Dynabeads (1 bead/cell) in the presence of seminal fluid (final dilution 1:20) in ELISpot plate wells pre-coated with antibody to IFN-γ for 24 h, before development of IFN-γ spots.

Figure 6

Depletion of NK cells from PBMCs markedly decreases IFN-γ expression in response to tetanus. NK cells in PBMCs were stained with FITC antibody to CD56 and depleted by flow sorting. Post-sort analysis demonstrated 100% depletion. PBMCs depleted or not of CD56⁺ cells were then analysed by IFN-γ ELISpot in response to tetanus (10Lfu/ml) or anti-CD3/28 Dynabeads as described in Methods.

Figure 7

Seminal fluid CD52 inhibits expansion of primary human NK cells. Untouched NK cells were purified by negative immunomagnetic selection from freshly prepared PBMCs and suspended in NK MACS medium containing 5% human AB serum. Cells were seeded at 15,000 cells/well in round-bottom 96-well plates in a total volume of 100 μl in the presence of IL-15 (final 20ng/ml) and medium only, or seminal fluid at 1:10 dilution depleted or not depleted of CD52. After 96 h, viable cells were enumerated against cell counting beads in a flow cytometer.