Potent and rapid activation of tropomyosin-receptor kinase A in endometrial stromal fibroblasts by seminal plasma

Abstract

Seminal plasma (SP), the liquid fraction of semen, is not mandatory for conception, but clinical studies suggest that SP improves implantation rates. Prior in vitro studies examining the effects of SP on the endometrium, the site of implantation, surprisingly revealed that SP induces transcriptional profiles associated with neurogenesis. We investigated the presence and activity of neurogenesis pathways in the endometrium, focusing on TrkA, one of the canonical receptors associated with neurotrophic signaling. We demonstrate that TrkA is expressed in the endometrium. To determine if SP activates TrkA signaling, we isolated the two most abundant endometrial cell types—endometrial epithelial cells (eEC) and endometrial stromal fibroblasts (eSF)—and examined TrkA activity in these cells after SP exposure. While SP only moderately activated TrkA in eEC, it potently and rapidly activated TrkA in eSF. This activation occurred in both non-decidualized and decidualized eSF. Blocking this pathway resulted in dysregulation of SP-induced cytokine production by eSF. Surprisingly, while the canonical TrkA agonist nerve growth factor was detected in SP, TrkA activation was principally induced by a 30–100-kDa protein whose identity remains to be established. Our results show that TrkA signaling is highly active in eSF and is rapidly induced by SP.

TrkA signaling is highly active in endometrial stromal fibroblasts and is rapidly induced by seminal plasma.

Abbreviations

eEC

Endometrial epithelial cells

eSF

Endometrial stromal fibroblasts

FRT

Female reproductive tract

IL

Interleukin

LIFf200

Leukemia inhibitory factor

NGF

Nerve growth factor

pTrkA

Phospho-TrkA

SP

Seminal plasma

Introduction

Seminal plasma (SP) is a complex, lipid- and protein-rich fluid produced by the male accessory sex organs [1, 2]. While SP has long been thought to serve primarily as a nutrient source and transport medium for sperm [3], more recent studies in humans and animal models suggest that SP participates in other accessory functions that promote reproductive success and offspring development [4–8]. For example, SP contains high levels of ions, including Ca²⁺, Mg²⁺, and Zn²⁺, which maintain osmotic balance and serve as essential cofactors for enzymatic reactions, including regulation of sperm motility [9–11]. Lipids in SP, such as prostaglandin E2, are important for immunomodulation in the female reproductive tract (FRT). Seminal plasma also contains a variety of cytokines and other proteins that elicit important signal transduction pathways in the FRT. For example, transforming growth factor beta (TGF-β) is thought to help establish tolerance to sperm and fetal antigens [12–14].

Interestingly, SP-derived neurotrophins are also important in signaling. In induced ovulators (species that require an external signal to ovulate, such as rabbits and camelids), seminal nerve growth factor (NGF) serves as the ovulation-inducing factor (OIF) [15]. NGF is a neurotrophin most studied for regulating neuron growth and survival, but its role as an OIF suggests broader signaling effects. NGF is also present in human SP [16], although its function is unclear because humans are cyclical ovulators.

When exposed to SP, cells from the FRT undergo sharp changes in transcription. For example, in both cervical and vaginal cells, SP induces the production of proinflammatory cytokines, including interleukin (IL)-6 and granulocyte-macrophage colony-stimulating factor (GM-CSF) [17]. The endometrium of the upper FRT also responds to SP exposure [18–20]. Although semen is initially deposited into the lower FRT, components from the lower FRT gain rapid access to the upper FRT as a consequence of peristaltic contractions [21, 22]. We and others found that the two major cell types of the endometrium—the endometrial epithelial cells (eEC) which line the uterine lumen and endometrial glands and the endometrial stromal fibroblasts (eSF) which are the major cell type of the underlying stroma—respond to SP exposure by upregulating expression of inflammatory cytokines, including TNF-α, IL-1β, IL-6, and leukemia inhibitory factor (LIF) [18–20]. Some of these cytokines, in particular IL-6 and LIF, regulate uterine receptivity for embryo implantation and immune tolerance at the implantation site [23–25]. Using global gene analysis, we found that in addition to inflammation, gene pathways associated with cellular proliferation, viability, and neurogenesis are activated in eEC and eSF in response to SP exposure [20]. Because these processes are all associated with NGF signaling, we set out to investigate the role of seminal NGF in the response of eEC and eSF to SP exposure. We focused on the ligand–receptor interaction between NGF and its high-affinity receptor, tropomyosin-receptor kinase (TrkA), a receptor associated with cell proliferation, survival, and differentiation in neuronal and nonneuronal tissues [26, 27]. NGF-TrkA signaling may have a role in promoting a healthy pregnancy, since a balanced, threshold level of NGF is required to ensure appropriate maternal tolerance and normal pregnancy progression [28]. Given observations that SP promotes implantation and/or proper development during pregnancy [4, 6-8], we sought to determine if SP activates TrkA signaling in eEC and eSF, and if so, the contribution of NGF in this process. We demonstrate here the rapid activation of TrkA in the endometrium by a proteinaceous component of SP that surprisingly does not appear to be NGF.

Materials and methods

Endometrial epithelial and stromal cells

All research proceedings were performed in accordance with the Declaration of Helsinki. Primary human endometrial tissue was obtained from the UCSF Endometrial Tissue Bank (ETB), and the Cooperative Human Tissue Network (CHTN). All donors provided written, informed consent, and were confirmed to not be pregnant at the time of tissue collection. Information regarding the age, cycle phase, sampling method, and clinical diagnosis (if available) of each donor is included in Table 1.

Table 1.

Patient information for endometrial samples used in this study.

Donor	Cells used	Diagnosis	Phase	Age (y)	Sampling method
A	Stromal	Oocyte donor	ESE	31	EMB
B	Stromal	Oocyte donor	ESE	27	EMB
C	Stromal	Oocyte donor	ESE	21	EMB
D	Stromal	Oocyte donor	ESE	24	EMB
E	Stromal	Oocyte donor	ESE	26	EMB
F	Stromal	Oocyte donor	ESE	25	EMB
G	Stromal	Pelvic pain	MSE	49	EMB
H	Stromal	Fibroids	Atrophic	60	Hysterectomy
I	Stromal	Fibroids	Atrophic	45	Hysterectomy
J	Epithelial	Pelvic pain (E)	P	39	EMB
K	Epithelial	Menorrhagia (E)	SE	33	EMB
M	Stromal and Epithelial	NA	MSE	42	Hysterectomy

P, proliferative; ESE, early secretory; MSE, mid secretory; SE, secretory; EMB, endometrial biopsy; (E), endometriosis; NA, not available.

Samples from the UCSF ETB were processed the day of tissue collection; samples from the CHTN were processed identically but 1 day after tissue collection since delivery of these samples (stored in saline, at 4°C) required overnight shipment. Endometrial EC and eSF were isolated as described [20]. Briefly, endometrial tissue was digested in Hank balanced salt solution with Ca⁺⁺ and Mg⁺⁺ (VWR Scientific) containing 6.4 mg/ml collagenase type I (Worthington Biochemical) and 100 U/ml hyaluronidase (Sigma-Aldrich). Digested cells were passed through a Falcon 40-μm filter (Fisher Scientific) to isolate single cells from luminal epithelial sheets and glandular fragments. Endometrial SF were further isolated and purified using selective attachment as described [20, 29].

After isolation, selectively attached eSF were cultured in serum-containing fibroblast growth medium (SCM: 75% phenol red-free Dulbecco's modified Eagle's medium (Life Technologies) and 25% MCDB-105 (Sigma-Aldrich), supplemented with 10% fetal bovine serum (FBS, BenchMark cat 100–106 from Gemini), 1 mM sodium pyruvate (Sigma-Aldrich) and 5 μg/ml insulin) until confluency. The cells were then serially passaged as described [30], using Trypsin EDTA 1X (Corning) to detach the cells between passages. The isolated eEC were plated in 24-well plates coated with Matrigel (VWR Scientific), and cultured in defined keratinocyte serum-free medium 1X (Life Technologies) as described [29]. Endometrial EC cultures were routinely monitored for epithelial morphology and the absence of eSF contamination, and were considered ready for treatments when an epithelial monolayer with dome-like folding structures was visible [29]. All experiments used eSF and eEC cultures from individual donors; in no instances were cells from different donors combined in culture.

Seminal plasma processing

De-identified semen samples were obtained from the UCSF Center for Reproductive Health (IRB # 14–15361). Semen samples were liquefied at room temperature for 2 h and then frozen at –80°C. When 20 samples were collected, all samples were thawed simultaneously, pooled, and centrifuged at 828 × g for 10 min to remove spermatozoa and other cells. The supernatant containing SP was collected, aliquoted, and frozen at –80°C, and served as the SP stock. All experiments used this stock of pooled SP, except for those comparing SP samples from different individuals (Figure 5C and Table 2). To compare SP from different individuals, 17 de-identified semen samples were obtained from the UCSF Center for Reproductive Health and processed into SP as described above. For quantifying NGF levels, the individual SP samples were vortexed for 10 s, diluted 1:1 in phosphate-buffered saline (PBS) to reduce viscosity, and then quantified for NGF levels using the Beta Nerve Growth Factor Human ELISA kit (Abcam). The reported NGF levels correspond to that present in 100% SP.

Figure 5.

Seminal plasma-mediated activation of TrkA in eSF is not mediated by NGF. (A) Endometrial SF from a woman without uterine pathologies were treated with NGF for the indicated times at the indicated concentrations. Endometrial SF were also treated in parallel with SP for 5 min. After treatments, lysates were collected and probed for TrkA pY785 levels. (B) Seminal plasma was treated with 0.25, 1, or 5 μg/ml of α-NGF neutralizing antibody for 1 h. The treated SP was then added to eSF from the same donor as in panel A at a concentration of 1% or 10%, and assessed for TrkA pY785 levels. (C) Three samples with low levels of NGF (156, 193, and 251 pg/ml) and three samples with high levels of NGF (1878, 1951, and 2788 pg/ml) were added to eSF from the same donor as in panel A at final concentrations of 1% and 10% for 5 min, after which lysates were blotted for TrkA pY785 levels.

Table 2.

NGF levels in semen samples from 17 different individuals.

NGF samples	Concentration (pg/ml) in undiluted sample
1	156
2	193
3	193
4	251
5	267
6	332
7	350
8	361
9	378
10	390
11	627
12	717
13	805
14	1163
15	1878
16	1961
17	2788

Average concentration = 753 ± 185 pg/ml.

Cell treatments

All eSF cultures were used between passage 2 (P2) and passage 4 (P4). Endometrial SF were seeded at a density of 5 × 10⁵ cells per well in six-well tissue culture plates (VWR Scientific) or 1 × 10⁵ per well in 24-well tissue culture plates (VWR Scientific), after which cells were fed every 48 h with SCM until confluent. For treatment of cells with SP, an aliquot of SP was first thawed and vortexed to ensure homogeneity. A solution of SCM containing 0, 1%, or 10% SP was then prepared, and added to the cells for 5 min, 1 h, 2 h, 4 h, or 18 h as indicated. These conditions were chosen based on prior studies defining the effects of SP on cells of the FRT [17, 31, 32], and prior reports that the in vivo effects of SP occur within 24 h [31, 33]. All cells were kept at 37°C for the duration of their treatments. Following SP treatment, cells were lysed and then analyzed by western blot as described below. To exclude the possibility that SP itself was responsible for the observed phenotype independent of eSF, an empty well containing no cells was equivalently treated with 10% SP and denoted the “no cell control.” For assessing the role of purified factors on TrkA activation, 1, 10, or 100 ng/ml NGF (R&D Systems) and 20, 200, or 2000 μM ZnCl₂ (Sigma-Aldrich) were added to the cells in place of SP. For NGF neutralization, anti-NGF (Abcam) was added at 0.25, 1, or 5 μg/ml to 10% SP and thoroughly mixed for 1 h, and then the mixtures were added to the eSF cultures at final SP concentrations of 0%, 1%, or 10%. Where indicated, eSF were treated with 50 nM K252a (Sigma-Aldrich) for 1 h, followed by 0%, 1%, or 10% SP for 5 min.

All eEC were grown in 24-well Matrigel-coated plates and used at passage 1. When eEC were approximately 75% confluent or when dome-shaped growths characteristic of cultured eEC [20] were observed, the cells were treated with SP similar to methods described above for eSF.

Immunohistochemistry

Immunohistochemistry was conducted using formalin-fixed, paraffin-embedded endometrial tissue samples. Samples were deparaffinized in Xylene (Sigma-Aldrich), washed with decreasing concentrations of ethanol (100%, 95%, 90%, 80%, and 70%), and rehydrated in PBS. Thereafter, the slides were blocked with 10% normal goat serum (Thermo-Fisher) in PBS for 45 min. Sections were then incubated with the primary antibody overnight at 4°C. The primary antibody for TrkA detection was a polyclonal goat anti-human antibody (Santa Cruz Biotechnology). In negative control slides, the primary antibody was replaced with IgG isotype (Thermo-Fisher). Slides for TrkA detection were incubated with goat anti-rabbit secondary antibodies (Vector Laboratories) for 45 min at room temperature. After 30 min incubation with ABC complex (Vectastain Elite ABC immunoperoxidase detection kit; Vector Laboratories), freshly prepared diaminobenzidine-hydrogen peroxide solution (DAB kit; Vector Laboratories) was added to the slides, which were thereafter rinsed with distilled water. The slides were counterstained with hematoxylin (Vector Laboratories), followed by exposure to tap water for 5 min. Slides were then dehydrated with increasing concentrations of ethanol, permeabilized with xylene, and mounted with Clarion mounting medium (Sigma-Aldrich). Slides were viewed on a Leica DM 5000 microscope (Leica Microsystems).

Western blotting

After SP, NGF, or Zn treatment, cells were washed once with PBS. Cells were then lysed in ice-cold lysis buffer (50 mM Hepes, pH 7.4, 125 mM NaCl, 0.2% NP-40, 0.1 mM PMSF, added to 1X complete EDTA-free protease inhibitor (Roche) and 1X PhosSTOP phosphatase inhibitor (Roche)). Lysates were agitated at 4°C for 20 min and centrifuged for 10 min at 8161 × g at 4°C. Supernatants were collected, diluted in 6X Laemmli's sample buffer (0.375 M Tris, pH 6.8, 60% glycerol, 12% SDS, 0.06% bromophenol blue, in Millipore H2O, with 1/8 β-mercaptoethanol by volume), and boiled at 95°C for 5 min. For SDS-PAGE analysis, equal volumes (20 μl) of lysate per sample were loaded onto 7.5% Criterion Tris-HCl gels (Bio-Rad). After electrophoresis, proteins were transferred to polyvinylidene difluoride membranes (Bio-Rad) for 3 h at 60 V at 4°C, and blocked overnight in 3% BSA in Tris-buffered saline-Tween 20 (TBS-T; 50 mM Tris, 150 mM NaCl, 0.1% Tween 20) at 4°C. The membranes were washed with TBS-T and incubated overnight at 4°C with primary antibodies. Primary antibodies used for immunoblotting were anti-TrkA (Cell Signaling Technology, rabbit polyclonal used at 1:1000), anti-phospho-TrkA pTyr785 (Cell Signaling Technology, rabbit monoclonal used at 1:1000), anti-phospho-TrkA pY751 (Life Technologies, rabbit polyclonal used at 1:1000), anti-phospho-TrkA pTyr490 (Sigma-Aldrich, rabbit polyclonal used at 1:1000), and anti-β-actin (Sigma-Aldrich, mouse monoclonal used at 1:5000). After primary antibody incubation, membranes were washed in TBS-T and incubated for 2 h at 4°C with the appropriate secondary antibodies. The secondary antibodies used were 1:5000 donkey anti-rabbit HRP (GE Healthcare) and 1:5000 sheep anti-mouse HRP (GE Healthcare). Membranes were washed in TBS-T and developed using Western lighting ECL Pro (Perkin Elmer), and visualized in a dark room using Hyperfilm ECL (Fisher Scientific) or with the FluorChem M system (ProteinSimple).

Immunofluorescence

Seminal plasma-treated eSF and eEC were fixed in ice-cold 100% methanol, permeabilized with 0.1% Triton X-100 (Sigma Aldrich), blocked with 10% normal goat serum (Sigma Aldrich) as reported [34], and incubated overnight at 4°C with the following primary antibodies at 1:200 dilution: Phospho-TrkA pY785 (Cell Signaling Technology), mouse anti–CDH1 (Abcam), and rabbit anti–human vimentin (Abcam). Cells were then washed three times with PBS and 0.1% Tween-20 buffer and incubated for 1 h at room temperature with the corresponding Alexafluor 488–conjugated goat anti-mouse or goat anti-rabbit secondary antibodies. The cells were subsequently washed 3 × 5 min with PBS, and then counterstained with Prolong Gold Antifade Reagent with 4΄,6-diamidino-2-phenylindole (DAPI; Life Technologies), and visualized with a Zeiss Axio Observer Z1 inverted microscope (Zeiss). As a control for any background caused by SP itself, 10% SP was added to a well of eSF after the cells had been fixed with methanol. This control served to demonstrate that the response of the cells to SP, rather than SP autofluorescence, was responsible for the observed signal.

Cycle phase

To assess the impact of cycle phase on TrkA activation, eSF were cultured in low-serum media (SFM, composed of SCM with 2% instead of 10% FBS and no insulin) for 14 days and fed every 48 h with SFM containing 10 nM estrogen (E2; β-estradiol, Sigma-Aldrich), 10 nM estrogen with 1 μM progesterone (E2P4, P4: progesterone, Sigma-Aldrich), or the vehicle control (ethanol, or EtOH). On the 14th day, conditioned media from the three conditions (EtOH, E2, E2P4) were collected and probed for insulin-like growth factor binding protein-1 (IGFBP1), a marker of decidualization, using the IGFBP1 ELISA kit (Alpha Diagnostics International). Decidualization of E2P4-treated samples was confirmed by IGFBP1 levels >125 ng/ml per 10⁵ cells [35]. Across all tested samples, the vehicle and E2 treatments yielded undetectable levels of IGFBP1.

Interleukin 6 and leukemia inhibitory factor ELISA

Endometrial SF were pretreated with 50 nM of K252a or dimethyl sulfoxide (DMSO; Sigma-Aldrich) for 1 h, and exposed to 1% SP. After 6 h, conditioned media were collected and analyzed for the levels of human IL-6 and LIF by ELISA, following the manufacturer's protocols (Life Technologies). Promega CellTiter-Glo reagent (Fisher Scientific) was used to assess cellular viability of eSF treated with SP and K252a.

Nerve growth factor functional assay

NGF activity was assessed by monitoring the ability of recombinant NGF or endogenous NGF in SP to induce proliferation of the TF-1 cell line [36]. NGF activity was monitored using methods similar to those previously described [37]. A total of 15,000 TF-1 cells were seeded per well of a 96-well plate, and cells were cultured in the presence of RP10 media (RPMI-1640 supplemented with 10% FBS) alone, or RP10 in the presence of the indicated concentration of recombinant NGF (R&D Systems). Alternatively, cells were cultured in RP10 supplemented with 0.1% SP. Higher concentrations of SP were not used because such treatment conditions were cytotoxic to TF-1 cells. Where indicated, neutralizing anti-NGF antibody (5 μg/ml, from Abcam) was added. After 3 days of culture, cellular proliferation was monitored by quantitating ATP levels using the CellTiter-Glo assay (Promega). Luminescence, the output of the assay that reflects cellular ATP levels, was quantitated using an Enspire luminometer (Perkin Elmer).

Filtration

Microcon centrifugal filters (EMD Millipore) with defined membrane pore sizes (YM-3, YM-10, YM-30, YM-100) were used to narrow down the molecular weight range of the active factor in SP. Seminal plasma was thawed and diluted 1:10 in SCM. Diluted SP (500 μl) was added to each of the four filters placed in microfuge tubes. Tubes were then centrifuged at 14,000 × g as recommended by the manufacturer (YM-3: 100 min; YM-10: 30 min; YM-30 and YM-100: 12 min). After the initial spin, the retentate (unfiltered material) was re-centrifuged at 14,000 × g using a new filter and tube for the same amount of time as in the first spin. The filtered volumes from the first and second spins were combined and used to treat eSF.

Proteinase K treatment

To confirm the degradation of seminal proteins by proteinase K, 200 μg/ml of proteinase K (Sigma-Aldrich) was added to 100% SP for 18 h at 37°C. After this incubation, 5 mM PMSF (Sigma-Aldrich) was added to the tube to inactivate proteinase K. 6X Laemmli buffer was added to the sample, which was then boiled for 5 min at 95°C. Seminal plasma incubated for 18 h at 37°C in the absence of proteinase K and a freshly thawed aliquot of SP were included as controls. These controls were similarly treated with PMSF and Laemmli buffer and boiled. Boiled proteinase K–treated and control samples were then further diluted 1:400 into 1X Laemmli buffer, and equal volumes of the samples were loaded onto a 15% Tris-HCl gel (Bio-Rad). After electrophoresis, protein degradation was visualized using the Pierce Silver Stain kit (Fisher Scientific), following the manufacturer's protocol.

Statistical analysis

The Student t-test was used for analysis. Mean and standard errors of the mean were reported. Statistical significance was established at P < 0.05 (*), and all tests were two-tailed.

Results

TrkA is expressed in the endometrium

To determine the relevance of SP signaling via the TrkA pathway in the upper FRT, we first set out to assess whether TrkA is expressed in the endometrium. Endometrial tissues were isolated from women and assessed for TrkA expression by immunohistochemistry. TrkA expression was observed in the epithelial and stromal layers of the endometrium (Figure 1a). In contrast, limited TrkA expression was detected in the myometrium. Expression of TrkA in eSF was confirmed by western blot (Figure 1b).

Figure 1.

TrkA is expressed in the endometrium. (A) Endometrial tissue sections from a woman diagnosed with uterine fibroids were stained via immunohistochemistry with an anti-TrkA antibody. Anti-TrkA staining of the myometrium and control goat IgG staining of the endometrium were performed as controls. Antibody staining is visualized by the brown coloring online and dark grey in print. (B) Endometrial stromal cells from a woman without uterine pathologies (no fibroids, endometriosis, adenomyosis, polyps, or abnormal uterine bleeding) were treated with 1% or 10% SP for 5 min, or 1, 2, and 4 h. Cell lysates were blotted using anti-TrkA and anti-β-actin antibodies. This figure is available in color at Biology of Reproduction online.

Seminal plasma induces TrkA activation in the endometrium

To determine if SP activates endometrial TrkA, eSF were treated with SP for 5 min, and 1, 2, and 4 h. Expression of total TrkA was not altered after treatment with 1% or 10% SP (Figure 1b). Strikingly, however, potent activation of TrkA, as assessed by phosphorylation of residue 785 (TrkA pY785) [38, 39], was observed after only 5 min of treatment with 1% or 10% SP, and this activation was sustained for at least 4 h (Figure 2a). An empty well similarly exposed to 10% SP gave no pY785 signal, excluding the possibility that SP itself was responsible for the observed signal (Figure 2a).

Figure 2.

Seminal plasma treatment activates TrkA in eSF. (A) Endometrial SF from a woman without uterine pathologies were treated with medium, 1% SP, or 10% SP. At each indicated time point (5 min and 1, 2, and 4 h), lysates were blotted with an anti-pTrkA pY785 antibody to detect phosphorylation at TrkA residue 785, an indicator of TrkA activation. The same lysates were similarly blotted with anti-pTrkA pY751 and anti-pTrkA pY490 antibodies to assess phosphorylation at TrkA residues 751 and 490. The no cell control indicates treatment of an empty well containing medium with 10% SP. The content in these wells was collected and blotted in parallel with the other samples. Data shown are representative of seven independent donors. (B) Endometrial SF from a woman diagnosed with uterine fibroids were treated with 1% and 10% SP for 5 min. The eSF were stained with anti-pTrkA pY785 antibody (shown in green online and as white striated staining patterns in print) to assess TrkA activation by immunofluorescence. DAPI counterstaining is shown in blue online and as light circular shapes (corresponding to individual nuclei) in print. The 10% SP postfixation control corresponds to eSF that were fixed prior to treatment with 10% SP. Data shown are representative of three independent donors. This figure is available in color at Biology of Reproduction online.

In addition to Y785, residues Y490, Y670/Y674/Y675, and Y751 of TrkA are also phosphorylated upon TrkA activation [38, 40]. Exposure of eSF to 1% or 10% SP efficiently induced both pY751 and pY490 phosphorylation (Figure 2a). At the later times, two separate TrkA pY751 bands were observed, likely representing antibody reactivity to different isoforms of TrkA [41]. Interestingly, the pY490 signal was decreased by 1 h post-treatment, suggesting that TrkA pY490 induction is more transient than induction of pY785 and pY751.

We next performed immunofluorescence studies to confirm SP-mediated activation of TrkA in eSF. Endometrial SF were treated with 1% or 10% SP for 5 min and then stained for pY785 TrkA expression. An increase in pY785 phosphorylation was induced by both 1% and 10% SP (Figure 2b). A “post-fix control” in which eSF were fixed with paraformaldehyde before treating with 10% SP was performed to show that the observed pY785 induction was due to a response of the eSF, and not nonspecific signal from SP. Therefore, by both western blotting and immunofluorescence, we found that SP potently and rapidly induces TrkA activation in eSF.

TrkA activation in endometrial epithelial cells

To determine if SP also rapidly induces TrkA activation in eEC, the other major cell type of the endometrium, we treated eEC with 1% or 10% SP and then performed western blotting for TrkA pY785 levels. Although 10% SP increased TrkA pY785 levels, this induction was not as dramatic as that observed in eSF (Supplementary Figure S1a). Treatment of eEC with 1% SP did not induce TrkA activation above the 0% SP basal level. By immunofluorescence, 10% SP but not 1% SP increased TrkA pY785 levels (Supplementary Figure S1b). These data suggest that SP also activate TrkA in eEC, but to a lesser extent than in eSF. We therefore focused on eSF for further characterization of SP-induced TrkA activation in the endometrium.

The effects of ovarian hormones on TrkA activation by seminal plasma

Endometrial SF are particularly responsive to the effects of the ovarian hormones estradiol (E2) and progesterone (P4) and, in response to E2P4, undergo decidualization, a differentiation process that renders the endometrium receptive for implantation. We next assessed whether SP-induced TrkA activation occurs after a 14-day pretreatment with E2 or E2P4, which mimics the proliferative and secretory endometrial phases, respectively. SP treatment induced phosphorylation of Y785 TrkA after treatment with vehicle, E2, or E2P4 (Figure 3). Among eSF from different donors, there was variability in the extent to which SP induced pY785 in E2P4-treated eSF relative to the other conditions. In some donors, E2P4-treated eSF responded more potently to SP, but in others, it responded less potently. Regardless, the data demonstrate that activation of TrkA by SP occurs under hormone conditions reflecting both the proliferative and secretory phases.

Figure 3.

Seminal plasma treatment activates TrkA in both decidualized and nondecidualized cells. Endometrial SF from three women without uterine pathologies were treated for 14 days with ethanol (EtOH), estradiol (E2), or estradiol with progesterone (E2P4) to induce decidualization (see Methods section). These cells were then exposed to 1% or 10% SP for 5 min and probed for TrkA phosphorylation at Y785 by western blot. Data from three different eSF donors are shown.

Blocking TrkA activation with K252a deregulates seminal plasma-induced interleukin-6 and leukemia inhibitory factor

We next investigated the downstream effects of blocking TrkA activation in eSF. K252a is a tyrosine kinase inhibitor that is commonly used to assess the effects of blocking TrkA signaling [42]. Both the baseline level of pY785 and the elevated levels induced by SP were decreased in the presence of K252a (Figure 4a), demonstrating that K252a is an inhibitor of TrkA activation in eSF. Because SP rapidly and potently induces IL-6 in eSF [20], a cytokine important for uterine receptivity [23–25], we assessed how K252a affects SP-mediated induction of IL-6 and its closely related family member LIF that also plays important roles in uterine receptivity [43, 44]. Under noncytotoxic conditions (Supplementary Figure S2), SP-exposed cells treated with K252a secreted higher levels of IL-6 relative to cells not exposed to the inhibitor (Figure 4b). In comparison, K252a blocked SP-mediated induction of LIF (Figure 4c). These results suggest that in eSF, blocking TrkA activation with K252a de-represses SP-mediated IL-6 induction, but has the opposite effect on LIF production.

Figure 4.

The TrkA inhibitor K252a treatment increases SP-induced IL-6 but suppresses SP-induced LIF secretion in eSF. (A) Endometrial SF from a woman without uterine pathologies were treated with the 50 nM K252a or DMSO for 1 h, then exposed to 1% or 10% SP for 5 min, and probed by western blot for TrkA phosphorylation at residue Y785. (B) Endometrial SF from a woman without uterine pathologies were pretreated with 50 nM K252a or DMSO for 1 h and exposed to 1% SP. IL-6 levels in culture supernatants were quantitated by ELISA. Data shown are representative of five independent experiments. *P < 0.05 as determined using a two-tailed t-test. (C) Culture supernatants from cells described in panel B were assessed for levels of LIF by ELISA. Data are representative of three independent experiments. **P < 0.01 as determined using a two-tailed t-test.

Nerve growth factor is not the primary factor in seminal plasma responsible for inducing TrkA activation

Given that NGF is present in semen and is a known inducer of TrkA, we next investigated whether seminal NGF was responsible for the rapid activation of TrkA by SP. Analysis of 17 semen samples from different donors revealed the average semen NGF concentration to be 753 ± 185 pg/ml (Table 2). To test if NGF activates TrkA, eSF were treated with 1 ng/ml, approximating the levels of NGF in 100% SP. Unlike SP which potently induced pY785 after 5 min, exogenous NGF did not activate TrkA (Figure 5a). To see whether higher, supraphysiological levels of NGF may induce TrkA activation, we also tested 10 and 100 ng/ml NGF, and found that none of these treatments activated TrkA. Endometrial SF treated for 1 or 18 h with NGF also did not induce TrkA activation, demonstrating that the lack of induction was not simply due to slower kinetics (Figure 5a). To confirm that the exogenous NGF used was functional, we demonstrated that it could increase proliferation of the NGF-responsive cell line TF-1 [37] (Supplementary Figure S3). Furthermore, antibody-mediated neutralization of NGF in SP—which we confirmed had NGF activity that could be diminished by anti-NGF treatment (Supplementary Figure S3)—did not diminish SP-induced TrkA activation in eSF (Figure 5b).

To provide further evidence that NGF is unlikely to be the main factor responsible for the rapid activation of TrkA by SP, we compared different SP samples containing different endogenous levels of NGF (Table 2). The three SP samples with lowest and highest levels of NGF were added to eSF and assessed for TrkA activation. Although there was donor-dependent variability, samples containing high endogenous levels of NGF did not induce greater TrkA activation than those samples containing low levels of NGF (Figure 5c), further supporting the notion that activation of TrkA by SP is likely not mediated by NGF.

Zn is not the TrkA inducing factor

Given that zinc (Zn) is present at high concentrations (∼2 mM) in SP and has been associated with Trk induction [45], we next tested whether Zn might be the SP factor inducing TrkA activation. Endometrial SF were treated with 20 and 200 μM Zn, corresponding to concentrations typically present in 1% and 10% SP. Zn treatments at these concentrations did not induce TrkA activation in eSF (Supplementary Figure S4). Although 2 mM Zn did induce a low level of TrkA activation (Supplementary Figure S4), given that this level of Zn corresponds to a concentration of SP that is 10× higher than that used in our assays, we believe that Zn is not responsible for the TrkA activation observed in our system.

TrkA-activating factor in seminal plasma is proteinaceous and has a molecular weight of 30–100 kDa

We then set out to further characterize the factors in SP responsible for TrkA activation. To determine if the TrkA activator in SP is proteinaceous, we first treated SP for 18 h with proteinase K, which degraded most seminal proteins (Supplementary Figure S5). Endometrial SF exposed to proteinase K–treated SP did not induce TrkA pY785 (Figure 6a), suggesting that the responsible factor is a protein.

Figure 6.

The factor(s) in SP responsible for TrkA activation is degraded by proteinase K and has a molecular weight of 30–100 kDa. (A) Seminal plasma was incubated in the presence or absence of 200 μg/ml proteinase K for 18 h at 37°C, and then spiked with PMSF to inhibit further proteolytic activity. A final concentration of 1% or 10% SP (control, mock-treated, or proteinase-K treated) was exposed for 5 min to eSF from a woman without uterine pathologies, and then probed for pTrkA pY785 levels. (B) Seminal plasma was centrifuged using filter units with nominal molecular weight limits of 3, 10, 30, and 100 kDa to generate size fractions <3 kDa (YM-3), <10 kDa (YM-10), <30 kDa (YM-30), or <100 kDa (YM-100). The filtrates, or unfractionated SP, were then added to eSF at final concentrations of 1% or 10%. After 5 min, lysates were generated and blotted for pTrkA pY785 levels. The eSF used in these experiments originated from a woman without uterine pathologies.

To assess the size of the inducing factor, size filtration chromatography was performed. Four different sized-based fractions of SP were isolated using centrifugal filters. SP filtered to contain only factors <3 kDa (YM-3) or <10 kDa (YM-10) did not induce TrkA. While SP filtered to contain proteins <30 kDa partially activated TrkA, that filtered to contain proteins <100 kDa had potent activity. These data suggest that the factor in SP that induces TrkA activation in eSF is proteinaceous and has a molecular weight between 30 and 100 kDa.

Discussion

While the success of in vitro fertilization (IVF) and assisted fertilization techniques demonstrate that SP is not inherently necessary for conception, studies both in vitro and in vivo have demonstrated that SP can have beneficial effects for pregnancy. In humans, exposure to vaginal capsules containing SP was associated with enhanced implantation rates in women with infertility [6]. Furthermore, exposure of the reproductive tract to semen in women undergoing IVF was also associated with increased implantation rates and embryo viability [7, 8]. Studies in animals additionally support the beneficial effects of SP on pregnancy outcomes and offspring development. Bromfield and colleagues found poorer growth trajectories and metabolic phenotypes in offspring of mice conceived in the absence of SP than those conceived in its presence [4, 14]. SP is known to induce paternal antigen-specific regulatory T cells—which are necessary for implantation—and elicits embryotrophic cytokines that support development of implantation-competent blastocysts (reviewed recently in [46]). In addition to these effects, SP may promote implantation through additional mechanisms. We previously reported using an ex vivo system that SP induces a potent and rapid transcriptional response in endometrial cells characterized by increased cell migration, proliferation, and viability, the latter including neurogenesis-associated cellular pathways [20]. These responses likely contribute to the ability of SP to promote endometrial receptivity.

In this study, we sought to expand on our prior finding that neurogenesis pathways are induced upon exposure of eEC/eSF to SP, by investigating the effects of SP on TrkA, one of the canonical receptors associated with neurogenesis [26]. Our main finding is that SP rapidly induces the activation of TrkA in eSF. Endometrial SF reside in the stroma, underneath the single layer of endometrial epithelium. Although eSF do not directly interface with the uterine lumen, they are essential for embryo attachment, and hormone conditions of the secretory phase can affect epithelial junctional protein expression and diminish epithelial barrier function, thereby allowing passage of luminal contents to the underlying stroma where eSF reside [47–49]. Such luminal contents can include SP, which can conceivably gain access to the endometrium through peristaltic contractions of the uterine cavity, although this has not yet been directly demonstrated in vivo in women.

We found that the TrkA signaling induced by SP is very rapid, occurring within 5 min of SP exposure. Since SP components are quickly cleared from the FRT by infiltrating neutrophils and macrophages [33], the ability of SP to quickly activate TrkA suggests that this signaling pathway may be a common postcoital response. TrkA signaling in response to SP also appears to be sustained at least for several hours, suggesting the induction of TrkA by SP may exert long-term effects on signaling in endometrial cells. Given the key role of eSF in implantation, the ability of SP to so rapidly and potently induce TrkA activation could conceivably influence endometrial receptivity.

What are the consequences of TrkA signaling?

In neuronal tissues, NGF-TrkA signaling is implicated in neuronal survival, proliferation, and function [26]. In nonneuronal tissues, TrkA has similarly been implicated in cell proliferation, but also serves additional functions, depending on anatomical location. In the FRT, TrkA signaling has been associated with folliculogenesis and the onset of first ovulation [50–52], ovarian angiogenesis [53, 54] and epithelial ovarian cancer [55]. While there is less known about the function of TrkA within the endometrium, treatment with K252a suppresses growth of human endometrial cancer cells, suggesting that TrkA has a role in endometrial cell proliferation [56]. Given the associations of TrkA with cell growth and cancer, activation of TrkA by SP in eSF may serve a similar role in signaling the proliferation of endometrial stromal cells, perhaps in preparation for embryo implantation. Interestingly, SP promotes decidualization, a differentiation program essential for implantation [57]; whether this effect is mediated through TrkA signaling remains to be determined.

Additionally, given TrkA’s function in angiogenesis, TrkA signaling by SP may also affect endometrial vascular remodeling. Endometrial spiral artery remodeling is necessary for construction of the vascular architecture at the feto-maternal interface [58]. Preeclampsia and intrauterine growth restriction are pregnancy complications whose pathophysiologies have been linked to inadequate remodeling of spiral arteries and decreased placental perfusion [59]. Because IVF and intrauterine insemination, which are typically administrated in the absence of SP exposure, may lead to increased incidence of preeclampsia [60, 61], SP, by inducing TrkA activation, may help protect against the development of preeclampsia and other pregnancy complications associated with incomplete endometrial vascular remodeling.

TrkA and the immune response

Interestingly, we found that inhibiting TrkA signaling with the kinase inhibitor K252a promoted SP-mediated induction of IL-6 but inhibited SP-mediated induction of LIF, suggesting that TrkA signaling may also have regulatory roles in cytokine production in the endometrium. IL-6 and LIF are both members of the IL-6 family of cytokines. IL-6 is induced by SP [18] and functions as a key regulator of uterine receptivity for embryo implantation and inflammatory events at the implantation site [23]. Elevated IL-6 may be associated with unexplained infertility, recurrent miscarriage, preeclampsia, and preterm delivery [62]. One proposed mechanism involves excess IL-6 inhibiting the generation of regulatory T cells necessary for pregnancy tolerance. Our data that TrkA normally limits SP-induced IL-6 suggest that TrkA signaling in the endometrium may promote beneficial pregnancy outcomes by regulating the development of immune tolerance.

In contrast to IL-6, LIF induction by SP is inhibited by K252a. LIF plays a role in promoting in embryo implantation by enabling uterine receptivity to blastocyst adhesion and subsequent trophoblast invasion [63, 64]. As endometria of infertile women are associated with lower levels of LIF [63, 65], SP-induced TrkA signaling in eSF may promote LIF induction and limit IL-6 secretion, which together ensure improved implantation and pregnancy outcomes.

What is the TrkA-activating factor in seminal plasma?

Surprisingly, our data suggest that NGF is likely not the major factor in SP responsible for the activation of TrkA in eSF. Three lines of evidence support this notion. First, treatment of eSF with exogenous NGF at concentrations matching those present in SP did not replicate the TrkA activation after SP treatment. Second, anti-NGF treatment of SP prior to addition to eSF did not abrogate TrkA signaling. Finally, treatment of eSF with semen samples containing low vs. high endogenous NGF levels did not correlate with low and high levels of TrkA activation, respectively. While the factor responsible for TrkA activation is likely not NGF, it is presumably proteinaceous, as treatment of SP with proteinase K abrogated this activity. From size-filtration chromatography experiments, we found that the inducing factor has a molecular weight of 30–100 kDa. Given that the molecular weight of NGF is less than 30 kDa (27 kDa for NGF, 13.5 kDa for β-NGF) and is resistant to enzymatic digestion by proteinase K [66], these findings lend further credence to the notion that the inducing factor is unlikely to be NGF.

A potential candidate for the TrkA activating factor in SP is luteinizing hormone (LH). Human LH is the hormone released by the anterior pituitary responsible for inducing ovulation in human females via the LH surge, and has a molecular weight of 32 kDa. It is also present in human semen, at levels higher than that in serum, and is associated with sperm motility and metabolism of fructose and glucose [67]. There is also evidence that LH functions with NGF in the ovulation induction of pre-pubertal heifers [68]. Furthermore, LH is implicated in the induction of TrkA receptors in bovine ovarian follicles [51]. Future studies should determine whether LH, or the numerous other proteins in SP many of whose functions remain unclear [69], serve as the factor in SP responsible for the rapid and potent activation of TrkA in eSF.

Limitations

While our present work highlights the rapid activation of TrkA signaling in eSF by SP, there are limitations to our in vitro design. Most of our studies were conducted with eSF as targets, since induction of TrkA in eEC after SP exposure was not as potent. However, eEC are likely more readily exposed to SP than eSF, given their interface with the endometrial lumen. Endometrial EC require paracrine interactions with eSF to respond to both estrogen and progesterone signaling [70, 71]. Future studies under more physiological conditions where eEC are co-cultured with patient-matched eSF in the presence of ovarian hormones, although technically challenging, would provide a powerful platform to study how SP affects TrkA signaling pathways in the two dominant cell types of the endometrium. A second limitation of this study is our use of samples from a heterogeneous group of donors, including those isolated from women with endometrial diseases (fibroids or endometriosis), which could influence outcomes. However, the consistency in responses suggests that use of specimens from different donors was not a major confounder. Furthermore, biopsies were obtained at different phases of the cycle (Table 1), although we have found that once eSF and eEC are isolated, cultured, and passaged, their behavior does not depend on the cycle phase in which they were originally obtained [72]. We were not powered sufficiently to determine whether eSF from women with endometrial pathologies respond differently to SP-induced TrkA signaling, or whether cycle phase at time of biopsy affects results. Future studies incorporating larger numbers of donor samples can address these issues.

Main conclusions

TrkA is expressed in the endometrium and is rapidly activated in eSF, and to a lesser extent in eEC, after exposure to SP. The factor(s) in semen responsible for TrkA activation is a protein with a molecular weight of 30–100 kDa.

Supplementary data

Supplementary Figure S1. Seminal plasma treatment of eEC activates TrkA to a lower extent than that observed in eSF. (A) Endometrial EC from a woman diagnosed with pelvic pain were treated with the indicated concentrations of SP for 5 min and then assessed for TrkA phosphorylation at residue Y785 by western blot. (B) Endometrial EC isolated from a woman undergoing hysterectomy (benign indication) were treated with the indicated concentrations of SP for 5 min. The cells were stained with an anti-pTrkA pY785 antibody to assess TrkA activation. Data representative of three eEC donors are shown.

Supplementary Figure S2. Treatment of eSF with the TrkA inhibitor K252a treatment does not diminish cellular viability. Endometrial SF from the same donor as that used in Figure 4B and C were treated with 50 nM K252a or 0.1% DMSO and assessed for viability using by monitoring ATP levels using the CellTiter-Glo viability kit.

Supplementary Figure S3. Confirmation of functional activity of NGF. (A) TF-1 cells (15,000/well) were cultured for 3 days in the absence of NGF, in the presence of the indicated concentration of recombinant NGF, or in the presence of 0.1% SP. NGF activity was assayed by monitoring proliferation rates through quantitation of ATP levels by measuring luminescence using the CellTiter-Glo assay. ****, P < 0.0001 (by one-way analysis of variance with a Bonferroni post test) for each sample versus the no-NGF control. n.s. non-significant. (B) TF-1 cells were cultured as described in (A), in the absence of NGF, in the presence of 10 ng/ml NGF, or in the presence of 100 ng/ml NGF + 5 μg/ml anti-NGF (top). Alternatively, cells were cultured in the absence of NGF, in the presence of 0.1% SP, or in the presence of 0.1% SP + 5 μg/ml anti-NGF (bottom). NGF activity was assayed by monitoring proliferation rates through quantitation of ATP levels by measuring luminescence using the CellTiter-Glo assay. * P < 0.05, and ** P < 0.01 as determined using a two-tailed t-test.

Supplementary Figure S4. Zn added at concentrations in SP does not activate TrkA in eSF. 20 μM, 200 μM, and 2 mM Zn (corresponding to Zn equivalent in 1%, 10%, and 100% SP, respectively) were incubated for 5 min with eSF isolated from a woman without uterine pathologies. Cells were then assessed for TrkA activation by assessing TrkA pY785 levels.

Supplementary Figure S5. Proteinase K treatment degrades seminal proteins. Seminal plasma was incubated for 18 h at 37°C in the presence or absence of 200 μg/ml proteinase K. Total protein content in the SP samples was assessed by silver stain. A freshly thawed sample of SP was included as a positive control.

Notes

Edited by Dr. Melissa E. Pepling, PhD, Syracuse University