The dysregulation of leukemia inhibitory factor and its implications for endometriosis pathophysiology

Abstract

Endometriosis is an estrogen dominant, chronic inflammatory disease characterized by the growth of endometrial-like tissue outside of the uterus. The most common symptoms experienced by patients include manifestations of chronic pelvic pain- such as pain with urination, menstruation, or defecation, and infertility. Alterations to Leukemia Inhibitory Factor (LIF), a cytokine produced by the luminal and glandular epithelium of the endometrium that is imperative for successful pregnancy, have been postulated to contribute to infertility. Conditions such as recurrent implantation failure, unexplained infertility, and infertility associated diseases such as adenomyosis and endometriosis, have demonstrated reduced LIF production in the endometrium of infertile patients compared to fertile counterparts. While this highlights the potential involvement of LIF in infertility, LIF is a multifaceted cytokine which plays additional roles in the maintenance of cell stemness and immunomodulation. Thus, we sought to explore the implications of LIF production within ectopic lesions on endometriosis pathophysiology. Through immunohistochemistry of an endometrioma tissue microarray and ELISA of tissue protein extract and peritoneal fluid samples, we identify LIF protein expression in the ectopic lesion microenvironment. Targeted RT qPCR for LIF and associated signaling transcripts, identify LIF to be significantly downregulated in the ectopic tissue compared to eutopic and control while its receptor, LIFR, is upregulated, highlighting a discordance in ectopic protein and mRNA LIF expression. In vitro treatment of endometriosis representative cell lines (12Z and hESC) with LIF increased production of immune-recruiting cytokines (MCP-1, MCP-3) and the angiogenic factor, VEGF, as well as stimulated tube formation in human umbilical vein endothelial cells (HUVECs). Finally, LIF treatment in a syngeneic mouse model of endometriosis induced both local and peripheral alterations to immune cell phenotypes, ultimately reducing immunoregulatory CD206⁺ small peritoneal macrophages and T regulatory cells. These findings suggest that LIF is present in the ectopic lesions of endometriosis patients and could be contributing to lesion vascularization and immunomodulation.

Keywords: cytokine; endometriosis; immunomodulation; infertility; leukemia inhibitory factor (LIF); vascularization.

Figure 1

LIF is present in the ectopic lesion microenvironment of endometriosis patients. Ectopic and eutopic tissues (A) and PF (B) from endometriosis patients contain LIF as detected through ELISA. No significant differences in LIF values were seen across tissue type. Analysis performed as unpaired Student’s T-test. (C–E), Endometrioma TMA of matched endometriosis (eutopic and ectopic; n=19) and control endometrium (n=22) was stained with an anti-LIF antibody. Area quantification of percent positive stain was calculated for the total core area (F–H) and luminal and glandular epithelium (I–K) respectively. Patients were stratified by menstrual phase- proliferative (G, J) and secretory (H, K), for both area quantifications. Patient samples used in (A–C, E) reflect 3 separate patient cohorts. Magnification provided at 4x and 20x; scale bar 100µm. Analysis performed as one-way ANOVA with Tukey post-hoc, **P<0.01, ****P<0.0001.

Figure 2

LIF gene expression is significantly downregulated in ectopic tissue compared to eutopic and control. Volcano plots showing differentially expressed genes in the LIF signaling pathway between (A) ectopic (n=9) and control (n=10), (B) ectopic and eutopic (n=9), (C) eutopic and control tissue samples. Vertical dashed lines indicate a fold change of +/-1 and horizontal blue line denotes a significance value of P=0.05.

Figure 3

LIF treatment in vitro did not alter proliferation or apoptosis in endometriosis representative cell lines but stimulated the release of growth factors and immune recruiting cytokines. WST-1 (A–C) and Caspase (D–F) assays were conducted in endometriosis representative cell lines- 12Zs (white bars), HUVECs (red bars), and hESCs (dashed bars), to detect LIF influence on proliferation and apoptosis respectively. (G–M), Cell lines were treated for 24hrs with PBS or varying rhLIF concentrations (1, 20, 100ng/mL) and supernatant analyzed for 48 cytokines pertaining to angiogenesis, inflammation, and cell growth (HD48-Multi-plex Analysis, EveTech). Analysis performed as one-way ANOVA with Tukey post-hoc, *P<0.05, **P<0.01.

Figure 4

LIF treatment promotes tubulogenesis in HUVEC cell line. HUVEC were treated with VEGF, PBS, or rhLIF (1, 20, 100ng/mL) and incubated for 4hrs before image acquisition- representative images provided for each treatment condition (A–E). Images were analyzed by WIMASIS Software to determine metrics of tube formation including (F) tube length and (G) total branching points. Analysis performed as one-way ANOVA with Tukey post-hoc, *P<0.05. Scale bar 100µm.

Figure 5

LIF treatment in a mouse model of endometriosis alters the local peritoneal immune response. (A) Gating strategy for flow cytometric analysis of myeloid (B–F) and lymphoid (G–J) markers on immune cells from the PF of mice injected i.p with PBS (white bars) or rmLIF (grey bars; 300ng, 1µg) for 14 days. LPM gated as: single cells, live, SSC^low, CD11b⁺, F4/80^hi, MHCII^low. SPM gated as: single cells, live, SSC^low, CD11b⁺, F4/80^mid, MHCII^hi. Tregs gated as: single cells, live, SSC^low, CD11b^-, F4/80^-, MHCII^-, CD3⁺, CD4⁺, CD25⁺, FOXP3⁺. Results reflect duplicate experiments- one per rmLIF dosage. Analysis performed as unpaired Student’s T-test, *P<0.05, **P<0.01. LPM, large peritoneal macrophages, SPM, small peritoneal macrophages.

Figure 6

LIF treatment in a mouse model of endometriosis alters the peripheral immune response. (A) Gating strategy for flow cytometric analysis of myeloid (B) and lymphoid markers (C–F) on immune cells from the spleen of mice injected i.p with PBS (white bars) or rmLIF (grey bars; 300ng, 1µg) for 14 days. Tregs gated as: single cells, live, SSC^low, CD11b^-, F4/80^-, MHCII^-, CD3⁺, CD4⁺, CD25⁺, FOXP3⁺. Results reflect duplicate experiments- one per rmLIF dosage. Analysis performed as unpaired Student’s T-test, *P<0.05, **P<0.01.

Figure 7

LIF treatment did not alter lesion growth or proliferation in a mouse model of endometriosis. I.p injections of PBS (white bars) or rmLIF (grey bars; 300ng, 1µg) were administered to C57BL/6 mice (n=6 for all groups) for 14 days, one week after endometriosis inducing surgery. Endometriosis-like lesions were collected upon sacrifice and subjected to IHC for markers of proliferation-Ki67 (A, B) and angiogenesis-CD31 (E, F), as well as LIF (I, J) and LIFR (M, N). Representative stain analysis provided for each marker from both the PBS and LIF treatment groups. Ki67 was analyzed as percent of cells expressing Ki67 over the total cell number as detected by a cytonuclear algorithm (C, D), while all other stains (CD31, LIF, LIFR) were analyzed by percent area quantification of stain (G, H, K, L, O, P). No statistical differences were detected across the four stains. Analysis performed as unpaired Student’s T-test. Scale bar 100µm.