Neurotrophins and Their Receptors, Novel Therapeutic Targets for Pelvic Pain in Endometriosis, Are Coordinately Regulated by IL-1β via the JNK Signaling Pathway

Abstract

Pelvic pain in women with endometriosis is attributed to neuroinflammation and afferent nociceptor nerves in ectopic and eutopic endometrium. The hypothesis that uterine nociception is activated by IL-1β, a prominent cytokine in endometriosis, was tested herein. Immunofluorescence histochemistry confirmed the presence of neurons in human endometrial tissue. Expression of nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) and their receptors in endometrial tissue and cells was validated by immunohistochemistry and Western blotting. Isolated endometrial stromal cells (ESCs) were subjected to dose-response and time-course experiments with IL-1β and kinase inhibitors to characterize in vitro biomarkers. Neural biomarkers were co-localized in endometrial nerve fibers. NGF, BDNF, and their receptors tropomyosin receptor kinase (Trk) A, TrkB, and p75 neurotrophin receptor were all expressed in primary ESCs. IL-1β stimulated higher TrkA/B expression in ESCs derived from endometriosis cases (2.8- ± 0.2-fold) than cells from controls (1.5- ± 0.3-fold, t-test, P < 0.01), effects that were mediated via the c-Jun N-terminal kinase (JNK) pathway. BDNF concentrations trended higher in peritoneal fluid of endometriosis cases but were not statistically different from controls (P = 0.16). The results support the hypothesis that NGF and BDNF and their corresponding receptors orchestrate innervation of the endometrium, which is augmented by IL-1β. We postulate that JNK inhibitors, such as SP600125, have the potential to reduce neuroinflammation in women with endometriosis.

Graphical abstract

Figure 1

Immunofluorescence histochemistry (IFH) shows neural cell adhesion molecule (NCAM)-L1 (green) fibers coursing through the stroma between epithelial glands of the endometrium functionalis layer. A: Faint background staining of glandular lumina is noted (A, upper left panel). To validate that these fibers were indeed neurons, double IFH was performed with anti-neurofilament (NFL) (red) (A, upper right panel). DAPI-positive (blue) nuclei are shown in the lower left panel. Merged image shows bright yellow nerves in the lower right panel. B: Microtubule associated protein (MAP)-2 (green)–positive fibers (upper left panel) and NFL (red) (upper right panel) show co-located fibers (merged image, lowerright panel). C: NFL (green)–positive fibers (upper left panel) and β-actin (red) (upper right panel) identify the same fibers as well as a fainter, generalized stromal background (upper right panel). DAPI-positive (blue) nuclei are shown in the lower left panel and highlight gland and, less so, stroma nuclei. The merged image shows bright yellow nerves in the right lower panel. Similar findings were confirmed in two women without and two women with endometriosis. Scale bars = 200 μm.

Figure 2

A and B: Immunoperoxidase histochemistry was used to localize nerve growth factor (NGF) (A) and brain-derived neurotrophic factor (BDNF) (B) in endometrium. NGF was localized in stroma and epithelia, whereas BDNF was predominantly stromal. C: Tropomyosin receptor kinase (Trk) A/B epitopes showed strong stromal and weaker epithelial staining. D: p75 neurotrophin receptor (NTR) was strongly expressed in linear fibers coursing through the stroma. E: IL-1 type 1 membrane receptor (IL-1R1) predominated in the cytoplasm of endometrial stromal cells, sparing blood vessels and glands. F: A negative control slide using nonimmune serum IgG revealed no immunoperoxidase staining. Similar findings were confirmed in two women without and two women with endometriosis. Scale bars = 50 μm.

Figure 3

A and B: Endometrial stromal cells (ESCs) were derived from eight independent participants with (E1 to E4) or without (C1 to C4) endometriosis. Nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), tropomyosin receptor kinase (Trk) A/B, p75 neurotrophin receptor (NTR), IL-1β, and IL-1 type 1 membrane receptor (IL-1R1) (respectively, panels 1 to 6) were evaluated by Western blotting, before (−) or after (+) 24 hours of incubation with 0.1 nmol/L IL-1β. In nearly every case, co-incubation with IL-1β increased accumulation of the corresponding neural or cytokine protein. Proinflammatory IL-1β (37 kDa) was strongly autoregulated by recombinant IL-1β treatment (A and B, panel 5). By contrast, connexin 43 (Cx43) was down-regulated by IL-1β treatment (A and B, panel 7), and β-actin expression was unaffected (A and B, panel 8). Molecular masses of each band are indicated at the right. C: Scattergram of ESC neural protein responses to IL-1β stimulation (fold increases relative to control cultures without IL-1β). Results from ESCs derived from controls (C, blue symbols) and patients with endometriosis (E, red symbols) are shown. n = 4 controls and 4 case patients.

Figure 4

Dose-response experiments with the c-Jun N-terminal kinase (JNK) inhibitor SP600125 (SP) over a range of 0 to 50 μM were performed in IL-1β–treated and control endometrial stromal cells (ESCs). Tropomyosin receptor kinase (Trk) A and B and phosphorylated TrkA/B (pTrkA/B) were all modestly stimulated by IL-1β, whereas SP dose dependently reduced neurotrophin receptor proteins with an apparent half maximal inhibitor concentration of approximately 30 μM. Treatment with SP alone (lane 8) reduced the neurotrophin receptor proteins below control levels (lane 1). β-Actin expression was unaffected (panel 4).

Figure 5

A and B: Recombinant IL-1β treatment of endometrial stromal cells (ESCs) for 24 to 72 hours increased, whereas co-incubation with 30 μmol/L SP600125 (SP) at each time point decreased neurotrophin and inflammatory proteins (panels 1 to 7). Proinflammatory IL-1β and IL-1 type 1 membrane receptor (IL-1R1) production responded similarly (panels 8 and 9). Even in ESCs without IL-1β treatment, SP modestly inhibited most of the neurotrophin-related proteins (lane 8) relative to control (lane 1). β-Actin expression was unaffected (panel 10). B: Analysis of variance using laser densitometry quantification of each of the 10 major protein bands, with the exception of β-actin (not shown), revealed significant differences among the three treatment groups with post hoc Scheffé tests revealing intergroup differences between control and IL-1β–exposed ESCs and IL-1β–exposed and IL-1β plus SP–treated cells (B). Note the logarithmic scale to accommodate the wide range of gel band densities compared. Qualitatively similar findings were observed in two women without and two women with endometriosis. n = 3. P < .01 (B). BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; NTR, neurotrophin receptor; pTrkA/B, phosphorylated tropomyosin receptor kinase A/B; TrkA/B, tropomyosin receptor kinase A/B.

Figure 6

A and B: Nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) were measured by enzyme-linked immunosorbent assay (ELISA) in endometrial stromal cell (ESC) supernatants before (Control) and after (IL-1) cytokine stimulation for 24 to 72 hours, without or with SP600125 (SP). C: IL-6 was measured by ELISA in culture supernatants before (Control) and after (IL-1) cytokine stimulation for 24 hours in ESCs derived from two women without (C1 and C2) and two women with (E1 and E2) endometriosis. IL-1β–stimulated IL-6 was reduced 49% ± 5% when ESCs were co-incubated with 30 μmol/L SP600125 (SP) compared with IL-1β stimulation alone (P = 0.01). n = 4.

Figure 7

Soluble concentrations of nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) were measured in the peritoneal fluid of 39 or 40 women. A–C: NGF (A) was not statistically different between controls (C) and patients with endometriosis (E) (P = 0.76), but BDNF levels (B) trended 2.3-fold higher in the latter cases (P = 0.16). (C) IL-6, used as a positive control, confirmed 2.2-fold higher concentrations in patients with endometriosis (E) compared with controls (C) (P = 0.03). n = 40 patients (B); n = 26 controls (A and C); n = 13 patients with endometriosis (A and C).

Figure 8

Diagram of neuroinflammation in endometriosis. We propose that IL-1β activates IL-1 type 1 membrane receptor (IL-1R1) in endometrial stromal cells, predominantly via the c-Jun N-terminal kinase (JNK) pathway, to promote neurotrophin and neurotrophin receptor (NTR) production and signaling, which in turn activate neurogenesis and pain. Parallel pathways promote IL-6 production and IL-1β autoregulation, potentially creating a vicious cycle of inflammation and pelvic pain. BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; SP, SP600125; TrkA/B, tropomyosin receptor kinase A/B.