Pharmacological blockage of the CXCR4-CXCL12 axis in endometriosis leads to contrasting effects in proliferation, migration, and invasion

Pharmacological blockage of the CXCR4-CXCL12 axis in endometriosis leads to contrasting effects... Abstract High levels of inflammatory factors including chemokines have been reported in peritoneal fluid and blood of women with endometriosis. CXCL12 mediates its action by interaction with its specific receptor, CXCR4, reported to be elevated in human endometriosis lesions and in the rat model of endometriosis. Activation of the CXCR4-CXCL12 axis increases cell proliferation, migration, and invasion of cancer cells. To obtain insights into the CXCR4 expression profile in lesions and endometrium, as well as functionality of the CXCR4-CXCL12 axis in endometriosis, we analyzed the expression of CXCR4 in tissues on a human tissue array and studied CXCL12-mediated activation of proliferation, invasion, and migration in vitro. We observed differences in levels of nuclear CXCR4 expression among lesion types, being higher in ovarian lesions. Endometriotic cell lines (12Z) showed higher levels of CXCR4, proliferative and migratory potential, and AKT phosphorylation/kinase activity compared to untreated control cells (endometrial epithelial cells). CXCL12 and endometriotic stromal cell-enriched media increased proliferation of non-endometriotic epithelial cells. CXCL12 caused a significant increase in 12Z cell invasion but had no effect on migration; AMD3100, a CXCR4-specific inhibitor, significantly increased invasion of 12Z cells but decreased their migration. However, treatment with CXCL12 plus AMD3100 significantly decreased invasion and migration of 12Z cells. In conclusion, the CXCR4-CXCL12 axis is functional in endometriosis cells, but the expression of CXCR4 varies among lesions. CXCL12 promoted proliferation, migration, and invasion of endometriotic cells, while inducing AKT phosphorylation and activity, but pharmacologically blocking this axis in the absence of the ligand induced their invasiveness. Introduction The pathologic mechanisms underlying endometriosis development and progression remain poorly understood, although genetic, epigenetic, environmental, and immune/inflammatory factors have been implicated. Most of the available treatments for endometriosis are hormones that cause undesired side effects while not being curative [1, 2]. Thus, a greater understanding of this complex and multifactorial disease and dissection of the biological and signal transduction mechanisms that contribute to its pathophysiology is needed in order to facilitate the identification of novel non-hormonal candidate therapeutic targets. When CXCL12 binds its specific receptor, the G protein coupled receptor CXCR4, it activates mechanisms related to normal reproductive biology (e.g., menstruation, ovulation, trophoblast implantation, and parturition [3, 4]) as well as immunity (e.g., natural killer cells and T cells [5]). Hyperactivation of the CXCR4-CXCR12 axis results in disease processes such as tumorigenesis, invasion, and metastasis [6]. This axis is involved in various cancer cell behaviors that also characterize endometriosis (e.g., increased basal proliferation rates [7–9]; and increased invasiveness [10]). Although few, there are some studies providing support for the involvement of the CXCR4-CXCL12 axis in endometriosis: (i) there are high levels of inflammatory factors including chemokines in the peritoneal fluid and serum of women with endometriosis, including chemokines such as CXCL12 [11–14]; CXCL12 is overexpressed in endometriosis lesions compared to eutopic endometrium from patients and controls [15–17]; (ii) CXCR4 protein expression is highest in both ovarian endometriosis and endometrial carcinoma when compared to normal ovary [18]; (iii) global gene profiling of endometriotic lesions using laser-captured microdissected epithelial cells identified a 20-fold increased expression of CXCR4 when compared to normal endometrial epithelial cells (EEC) [19]; and (iv) lesions from the rat model of endometriosis have higher levels of CXCR4 gene expression when compared to endometrium from controls [20]. Based on these data, we hypothesized that the CXCR4-CXCL12 axis could play a role in the development or progression of endometriosis. Thus, the aim of this study was to obtain insights into the profile of expression across different lesions types and into the functionality of the CXCR4-CXCL12 axis in endometriosis, as a first step to explore the possibility of targeting this axis as a therapy for this incapacitating disease. We show here that there are functional interactions between CXCL12 and CXCR4 in endometriotic epithelial cells that leads to increased proliferation, invasion, and migration, as well as phosphorylation and activation of serine-threonine protein kinase (AKT), which may facilitate their ectopic survival and growth. However, we also show that there were unforeseen effects of the CXCR4 inhibitor that could preclude its clinical use in endometriosis as a single agent. Materials and methods Immunohistochemistry of CXCR4 in human endometriotic lesions on a tissue array The endometriosis-focused tissue array used for this study has been previously described and used for other investigations [21–23]. All studies involving human tissues were approved by the Internal Review Board of the Ponce Health Sciences University (PHSU). A total of 164 deidentified endometriotic tissue blocks were used to construct a tissue array in collaboration with the PHSU–Moffitt Cancer Center Partnership. To account for the heterogeneity of endometriosis, this tissue array includes lesions localized in the ovaries (n = 29), fallopian tubes (n = 16), peritoneum (n = 34), skin (umbilical region) (n = 4), and gastrointestinal tract (n = 7). Additionally, the tissue array includes endometrial samples obtained from patients with endometriosis and controls (women undergoing hysterectomy for benign uterine disorders) in the proliferative and secretory phases of the menstrual cycle. All blocks were analyzed by two pathologists to document endometriosis (glands and stroma) and determine the menstrual cycle phase in the case of endometrium according to Noyes criteria [24]. The tissue array block was cut in 10 μm for immunostaining. Immunohistochemistry was conducted as described by Ruiz et al. [25]. Primary antibody against CXCR4 was used at a 1:300 dilution (R4 monoclonal-clone 44716) (R&D Systems Minneapolis, MN) (Table S1). Negative control for the immunostaining was done with normal serum instead of primary antibody in control tissue slides. Immunohistochemical staining (IHC) intensity was evaluated in glands and stroma using a semiquantitative method. Staining intensity was blindly scored as absent (0), weak (1), moderate (2), or intense (3) by two independent, blinded observers. Mean IHC intensity score and interquartile range were calculated per compartment per sample for statistical analysis. Cell cultures Human endometrial stromal cells (HESC), highly invasive and immortalized with human Telomerase reverse transcriptase (hTert), were purchased from American Type Culture Collection (#CRL 4003). Noninvasive epithelial cells (EEC, which have been shown to have MCF7 characteristics by DNA fingerprinting [26]), and human endometriotic epithelial cells (12Z), SV40 T antigen-transformed and derived from a peritoneal lesion, were obtained as part of a collaboration with Dr Fazleabas and Dr Starzinski [27]. The 12Z were grown Dulbecco's Modified Eagle's Medium (DMEM)/F12, 10% Fetal Bovine Serum (FBS); HESC were grown in phenol-free DMEM, charcoal-treated 10% FBS, 1% insulin-transferrin-selenium (ITS); the medium for EEC cells was DMEM/F12, 10% FBS, 160 ng/ml insulin as previously described [28, 29]. Endometriotic stromal cells derived from a peritoneal endometriotic lesion and grown at low passage numbers (Primary Endometriotic Cells (PED)) (developed in Dr Idhaliz Flores laboratory; Bello et al. unpublished data) were cultured in complete DMEM supplemented with 10−8 mol/L estradiol and 1% ITS. Estradiol and ITS were removed from media 24 h prior to experiments. Enzyme-linked immunosorbent assay for CXCL12 The cell lines 12Z, PED, HESC, and EEC were cultured as previously described [21, 22]. The cells were cultured during 24, 48, and 72 h. Cell culture supernatant was collected, centrifuged, and stored in aliquots at –20°C. Levels of human CXCL12 were determined using a Quantikine enzyme-linked immunosorbent assay (ELISA) (R&D systems, Minneapolis, MN) following manufacturer's instructions and as described before [30]. Briefly, the assay diluent was added to samples (in triplicate) and to the standards (serially diluted positive control), and incubated for 2 h at room temperature (RT). CXCL12-specific antibody was added to each well and incubated at RT for 2 h followed by washes. The substrate solution was then added to each well and incubated in the dark. Finally, the optical density was determined using a microplate reader (MRX Revelation, Dynex Technologies, Inc., Chantilly, VA) set to 450 nm with correction wavelength set at 540 nm. Western blotting for CXCR4 Total protein extraction from the cell cultures, quantification, and western blot analysis have previously been described in detailed [31]. Dilutions of anti-CXCR4 (monoclonal-clone 44716) (R&D Systems Minneapolis, MN) and GAPDH (monoclonal-DM1A) (Calbiochem, Darmstadt, Germany) used were 1:300 and 1:1000, respectively (Table S1). Experiments were repeated three times. Modulation of in vitro cell proliferation by CXCL12 or stromal cell-enriched media The BrdU cell proliferation assay kit was used to study 12Z and EEC cell proliferation in response to CXCL12, primary endometriotic stromal cell (PED)-enriched media, or complete media as control, following the manufacturer's instructions (Millipore, Temecula, CA). The amount of the incorporated BrdU, indicative of newly synthesized DNA, was detected using a microplate reader (MRX Revelation, Dynex Technologies, Inc., Chantilly, VA) set at dual wavelengths of 450/550 nm. As blank we used cell media with no cells and as background cells without the BrdU reagent. To measure profilerative responses to CXCL12, epithelial cells (12Z or EEC) were seeded at 4000 and 2000 densities in 100 μl media in a flat bottom 96-well plate overnight. CXCL12 (hrCXCL12 alpha, R&D Systems, Inc., Minneapolis, MN) was added to the media to a final concentration of 10 and 100 ng/ml, and proliferation rates were measured by BrdU incorporation. These doses were based on a published study by Laird et al. [32]. To determine the proliferative effects of the stromal cell-enriched media on the epithelial cells (12Z and EEC), PED (primary endometriotic) and HESC (hTERT immortalized stromal endometrial) cells were cultured at a density of 1.0 × 106 cells in 2 ml per well in a 6-well plate without estrogen and ITS. After 24 h of culture, conditioned media was collected, filtered, and added to 12Z and EEC cells for 24 h. Proliferation rates were measured by BrdU incorporation following manufacturer's protocol. Experiments were repeated at least three times. Modulation of phospho-AKT by CXCL12 Cells were grown and incubated with media containing 10 or 100 ng/ml of human recombinant (hr) CXCL12 alpha, or complete media as control, for 10 s, 1, 5, 10, 15, and 30 min. Cells were lysed with RIPA lysis buffer supplemented with protease and phosphatase inhibitors (Roche Diagnostics, Indianapolis, IN). Protein concentration was determined using a standard curve generated by a Bradford protein assay. Prior to Western Blot (WB), total AKT was immunoprecipitated. Briefly, proteins were diluted in lysis buffer containing protease and phosphatase inhibitors and 50 μl of Sepharose protein A (Rockland Gilbertsville, PA) was added, followed by addition of anti-AKT antibody or phospho-AKT (Cell Signaling Technology, Danvers, MA), and incubation overnight at 4°C with gentle rocking. After centrifugation and washes, membranes were incubated with horseradish peroxidase-conjugated secondary antibody, and signals were detected with an automated image documentation system (ChemiDoc, Bio-Rad, Hercules). Band density was analyzed using Quantity One software (Bio-Rad, Hercules, CA). Experiments were repeated three times. Modulation of AKT kinase activity by CXCL12 Cell lines (12Z and EEC) were cultured as previously described. Media containing 10 or 100 ng/ml CXCL12, or complete media as control, were added during 1, 10, 30 min. Cellular lysates were analyzed using a nonradioactive immunoprecipitation-kinase assay to detect total phospho-AKT at serine 473 and phosho-glycogen synthase kinase 3 (GSK-3)α/β(ser21/9) (AKT Kinase Activity kit, Cell Signaling Technology, Danvers, MA) following the manufacturer's protocol. In brief, after treatments, cells were lysed and sonicated 10 min on ice, followed by centrifugation. Cell lysates were incubated with beads/immobilized antibody overnight with gentle rocking at 4°C. Cell lysate containing immobilized antibody was centrifuged, and after washes the pellet was suspended in 50 μl in kinase buffer supplemented with 1 μl of 10 nM ATP and the kinase substrate, GSK-3α/β fusion protein. Samples were incubated during 30 min at 30°C. The reaction was terminated by adding 25 μl of 3× sodium dodecyl sulfate (SDS) buffer, vortex, and microcentrifugation. WB analyses were performed as previously described to detect phospho-AKT and phospho-GSK-3α/β. Experiments were performed in triplicate. Modulation of in vitro migration and invasion by CXCL12 and/or AMD3100 Cell migration and invasion of 12Z and EEC in response to CXCL12 plus/minus the CXCR4-specific inhibitor AMD3100 (Sigma-Aldrich, St. Louis, MO) were quantified using BD BioCoat Matrigel Invasion Chamber following standard protocols (BD Biosciences, Bedford, MA). In brief, epithelial cells were grown in the upper chambers in complete media or media with 100 nM AMD3100. Complete media supplemented with 10 or 100 ng/ml CXCL12 was added to the lower chamber in 24-well plates. Wells with Matrigel-coated inserts or control inserts were used to differentiate invasion from migration, respectively [33, 34]. After 24 h, the cells were fixed and stained with Diff-Quik stain (Dade Behring, Inc. Newark, DE) following the manufacturer's instructions. The inserts were allowed to dry and mounted in microscope slides for counting and visualization of the invading and migrating cells using a light microscope. Experiments were repeated three times. Statistical analyses Data from cell culture experiments (e.g., proliferation, AKT phosphorylation, AKT function, migration and invasion) were analyzed by Mann–Whitney, non-parametric Student t-test, compared to basal or untreated conditions. Analysis of variance (ANOVA) using Tukey and Dunn multiple comparison tests was used to analyze tissue array immunostaining differences. All analyses were conducted using SPSS 15.0 (Chicago, IL, USA) and GraphPad Prizm v5 was used for graphical representation of the data. Statistical significance was set at P < 0.05. Results Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array We analyzed by IHC the protein expression of the chemokine receptor CXCR4 in human endometriosis lesions from five different anatomical sites (ovaries, peritoneum, fallopian tubes, skin, and gastrointestinal tract) as well as eutopic endometrium from women with endometriosis and controls included in a custom-made endometriosis-focused tissue array. From 164 samples, 137 core biopsies (84%) could be analyzed; biopsies that did not include stroma and glands were excluded from the analysis. Nuclear CXCR4 (nCXCR4) expression was in general higher in stroma compared to glands, and significantly higher in the stroma of ovarian endometriosis compared to fallopian tube lesions and proliferative endometrium from controls. In that respect, the proliferative endometrium from patients showed a similar expression of nCXCR4 than endometriotic lesions (Figure 1). nCXCR4 expression in glands was highest in ovarian compared to both fallopian lesions and proliferative endometrium from cases and controls. Cytoplasmic CXCR4 expression in stroma was not significantly different among the tissues analyzed, although lesions showed a slightly increased level of cCXCR4 expression compared to endometrial tissues (P = 0.0343). Figure 1. View largeDownload slide Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array. (A) A total of 164 formalin-fixed paraffin-embedded endometrial and endometriotic human on a tissue array were analyzed by immunohistochemistry. The immunostaining intensity of CXCR4 in nuclear and cytoplasmic compartment of the stromal and glands cells was evaluated and shown graphically. The data were analyzed by ANOVA and with Dunn Multiple Comparison post hoc test, the statistical significance level among them are indicated by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. (B) Representative pictures showing immunostaining in different lesion types are shown. Figure 1. View largeDownload slide Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array. (A) A total of 164 formalin-fixed paraffin-embedded endometrial and endometriotic human on a tissue array were analyzed by immunohistochemistry. The immunostaining intensity of CXCR4 in nuclear and cytoplasmic compartment of the stromal and glands cells was evaluated and shown graphically. The data were analyzed by ANOVA and with Dunn Multiple Comparison post hoc test, the statistical significance level among them are indicated by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. (B) Representative pictures showing immunostaining in different lesion types are shown. In vitro CXCR4 and CXCL12 expression CXCR4 protein expression was analyzed in 12Z, HESC, and EEC by WB. We showed higher levels of CXCR4 expression in 12Z cells compared to EEC (Figure 2). HESC, an endometrial stromal cells also expressed CXCR4. ELISA results showed that none of the cell lines studied (12Z, HESC, EEC, PED) expressed CXCL12 alpha, or its expression was below detection levels (data not shown). This contrasts with the findings of increased levels of CXCL12 in human endometriotic tissues. It is possible that other CXCL12 isoforms (from beta to gamma) not measured here are involved. Others have previously shown that endometrial stromal cell lines do not express CXCL12, contrary to what is seen in whole tissues [35]. Figure 2. View largeDownload slide CXCR4 protein analysis by western blot of endometrial and endometriotic cell lines. Endometrial epithelial (EEC), human endometrial stromal (HESC), and endometriotic epithelial (12Z) cells lines were cultured in complete media. Total protein was extracted, quantified, and separated by electrophoresis. Levels of CXCR4 were analyzed by immunoblotting, and GAPDH was used as loading control. At least three experiments in three different passage numbers were conducted. Differences in the levels of GAPDH can be explained by the different cell types (epithelial vs. stromal), which was normalized by densitometric analysis. Figure 2. View largeDownload slide CXCR4 protein analysis by western blot of endometrial and endometriotic cell lines. Endometrial epithelial (EEC), human endometrial stromal (HESC), and endometriotic epithelial (12Z) cells lines were cultured in complete media. Total protein was extracted, quantified, and separated by electrophoresis. Levels of CXCR4 were analyzed by immunoblotting, and GAPDH was used as loading control. At least three experiments in three different passage numbers were conducted. Differences in the levels of GAPDH can be explained by the different cell types (epithelial vs. stromal), which was normalized by densitometric analysis. Effects of CXCL12 or stromal cell-conditioned media on in vitro proliferation rates We next studied the proliferation rate of 12Z, HESC, and EEC using the BrdU proliferation assay. Our results confirmed that endometriotic epithelial cells, 12Z, had a statistically significant higher basal level proliferation rate compared to the control cell lines HESC and EEC (Figure 3A). Next, we studied the in vitro proliferative effects of CXCL12 in the studied cells lines using the BrdU proliferation assay. Treatment with CXCL12 for 24 h induced a significant increase in EEC cells proliferation rates by 33% (100 ng/ml) and 41% (200 ng/ml); despite expressing CXCR4 neither 12Z nor HESC cells responded to CXCL12 in this manner (Figure 3B). Figure 3. View largeDownload slide Effects of CXCL12 or stromal cells-conditioned media on in vitro cell proliferation. (A) Cells were cultured in a 96-well plate for 48 h and proliferation was measured by BrdU incorporation. Absorbance values were measured at 450/550 nm after 24 h of incubation with BrdU. Endometriotic epithelial (12Z) cells are represented by filled diamonds, noninvasive epithelial cells (EEC) by filled squares, and human EEC by filled triangles. Results represent data from two experiments in triplicate, and are reported as means ± standard error of the mean. (B) Cells were treated with 10 or 100 ng/ml CXCL12 (hrCXCL12 alpha). Basal condition consisted of cells growing in complete media. (C) Cells were cultured in filtered conditioned media from endometrial stromal cells (HESC) or endometriotic stromal cells (PED) that were seeded at a density of 1 × 106 cells and incubated for 24 h in a fixed volume of 2 ml, in the presence or absence of neutralizing antibodies against CXCR4 or CXCL12α. Absorbance value was measured at 450/550 nm after 24 h of incubation with BrdU. Results represent data from three experiments in triplicate. Results represent data from six values per bar (*P ≤ 0.05). Figure 3. View largeDownload slide Effects of CXCL12 or stromal cells-conditioned media on in vitro cell proliferation. (A) Cells were cultured in a 96-well plate for 48 h and proliferation was measured by BrdU incorporation. Absorbance values were measured at 450/550 nm after 24 h of incubation with BrdU. Endometriotic epithelial (12Z) cells are represented by filled diamonds, noninvasive epithelial cells (EEC) by filled squares, and human EEC by filled triangles. Results represent data from two experiments in triplicate, and are reported as means ± standard error of the mean. (B) Cells were treated with 10 or 100 ng/ml CXCL12 (hrCXCL12 alpha). Basal condition consisted of cells growing in complete media. (C) Cells were cultured in filtered conditioned media from endometrial stromal cells (HESC) or endometriotic stromal cells (PED) that were seeded at a density of 1 × 106 cells and incubated for 24 h in a fixed volume of 2 ml, in the presence or absence of neutralizing antibodies against CXCR4 or CXCL12α. Absorbance value was measured at 450/550 nm after 24 h of incubation with BrdU. Results represent data from three experiments in triplicate. Results represent data from six values per bar (*P ≤ 0.05). The epithelial cell lines 12Z and EEC were cultured in conditioned media from the stromal cell lines PED (endometriotic) and HESC (normal) to determine whether there are secreted factors that can induce proliferation of epithelial cells. As controls, neutralizing antibodies to CXCR4 or CXCL12 were added to the culture medium. 12Z cells cultured in media conditioned by HESC cells showed an increase in proliferation. However, this increase was not statistically significant nor was it reduced by the neutralizing antibodies to CXCR4 or CXCL12. EEC cells cultured with media conditioned by PED cells showed a statistically significant proliferation increase of 68% (Figure 3C). The proliferation increased observed in EEC cells cultured in conditioned media was not decreased by treatment with neutralizing antibodies to CXCR4 or CXCL12. Effects of CXCL12 on AKT Serine 473 phosphorylation and AKT kinase activity After normalization against total AKT and relative to basal expression, our data show that the levels of AKT phosphorylation were higher in the endometriotic 12Z cells compared to EEC cells at both CXCL12 doses tested (Figure 4). The higher dose of CXCL12 (100 ng/ml) significantly increased AKT phosphorylation compared to the lower dose (10 ng/ml) (P < 0.05 at 1 min). Also, we observed significant increases in AKT kinase activity in 12Z cells induced by CXCL12 (10 ng/ml) at 1, 10, and 30 min of treatment compared to no treatment (Figure 5). At the higher dose (100 ng/ml) and longer experimental time (30 min), CXCL12 treatment significantly decreased AKT activity in 12Z cells. In contrast, AKT activity was not significantly changed in EEC cells after treatment with 10 and 100 ng/ml CXCL12 at any of the times studied. Figure 4. View largeDownload slide Effects of CXCL12 on AKT Serine 473 phosphorylation. Total protein concentration was determined and subjected to immunoprecipitation against total AKT and immunoblotting against total AKT and AKT phosphorylated at serine 473. (A) Representative western blot of the phosphorylation of AKT in endometriotic epithelial cells (12Z) or noninvasive epithelial cells (EEC) treated with CXCL12. Cell were cultured and treated with either 10 or 100 ng/ml of CXCL12 for 0, 10 s, 1, 5, 10, 15, and 30 min. Basal condition consisted of cells growing in their respective complete media. (B) Graphical representation of densitometry of at least three experiments per cell line. Data were analyzed by nonparametric Student t-test using GraphPad Prizm v5. Significance was set at P < 0.05. Figure 4. View largeDownload slide Effects of CXCL12 on AKT Serine 473 phosphorylation. Total protein concentration was determined and subjected to immunoprecipitation against total AKT and immunoblotting against total AKT and AKT phosphorylated at serine 473. (A) Representative western blot of the phosphorylation of AKT in endometriotic epithelial cells (12Z) or noninvasive epithelial cells (EEC) treated with CXCL12. Cell were cultured and treated with either 10 or 100 ng/ml of CXCL12 for 0, 10 s, 1, 5, 10, 15, and 30 min. Basal condition consisted of cells growing in their respective complete media. (B) Graphical representation of densitometry of at least three experiments per cell line. Data were analyzed by nonparametric Student t-test using GraphPad Prizm v5. Significance was set at P < 0.05. Figure 5. View largeDownload slide AKT kinase activity assay. Endometrial epithelial (EEC) and endometriotic epithelial (12Z) cells were cultured and treated with 10 or 100 ng/ml CXCL12. Basal condition consisted of cells growing in their respective complete media. Total protein was isolated, quantified, and phospho AKT at serine 473 was immunoprecipitated and incubated with ATP and a GSK3 fusion protein. A western blot was performed to identify phospho AKT and phosphorylation state of the fusion protein. (A) Representative results of endometriotic 12Z cells and EEC. (B) Graphical representation of the quantification of phosphorylation by densitometry analysis of triplicate experiments. *P < 0.05. Figure 5. View largeDownload slide AKT kinase activity assay. Endometrial epithelial (EEC) and endometriotic epithelial (12Z) cells were cultured and treated with 10 or 100 ng/ml CXCL12. Basal condition consisted of cells growing in their respective complete media. Total protein was isolated, quantified, and phospho AKT at serine 473 was immunoprecipitated and incubated with ATP and a GSK3 fusion protein. A western blot was performed to identify phospho AKT and phosphorylation state of the fusion protein. (A) Representative results of endometriotic 12Z cells and EEC. (B) Graphical representation of the quantification of phosphorylation by densitometry analysis of triplicate experiments. *P < 0.05. Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration in vitro We then determined the effects in invasion and migration of CXCL12 and/or the CXCR4 inhibitor AMD3100 in the epithelial cell lines (12Z or EEC) using the boyder chamber assay. CXCL12 alone (10 and 100 ng/ml) caused a significant increase in migration of 12Z cells; CXCL12 alone (10 ng/ml only) increased their invasion. AMD3100 alone significantly decreased migration of 12Z cells by 0.42-fold (P ≤ 0.0001) but increased their invasion capacity by 2.4-fold (P ≤ 0.0001) (Figure 6). Treatment with CXCL12 (10 ng/ml) plus AMD3100 significantly decreased invasion (P ≤ 0.01) and migration (P ≤ 0.0001) of 12Z cells. On the other hand, the migration capacity of EEC cells was significantly increased by CXCL12 alone (100 ng/ml) (P ≤ 0.01), but decreased by AMD3100 alone (P ≤ 0.001), and by CXCL12 plus AMD3100 (P ≤ 0.001). Invasiveness of EEC was decreased by AMD3100 alone or in combination with CXCL12, but this effect did not reach statistical significance. Figure 6. View largeDownload slide Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration. 12Z cells or EEC cells were cultured at the upper chamber at a 25 000 cell density; the bottom chamber media contained 10 or 100 ng/ml CXCL12, the CXCR4 inhibitor AMD3100 (100nM), or a combination of CXCL12 (100 ng/mL) and AMD3100 (100 nM). Cells were cultured for 24 h, followed by fixation, staining, and visualization of the cells under microscope to count the cells. (A and C) Number of cells counted in the control insert (migration). (B and D) Number of cells that invaded through the Matrigel matrix (invasion). Data were analyzed by comparing each treatment to media only using nonparametric Student t-test in GraphPad Prizm v5. Significance was set at P < 0.05 and represented by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. Figure 6. View largeDownload slide Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration. 12Z cells or EEC cells were cultured at the upper chamber at a 25 000 cell density; the bottom chamber media contained 10 or 100 ng/ml CXCL12, the CXCR4 inhibitor AMD3100 (100nM), or a combination of CXCL12 (100 ng/mL) and AMD3100 (100 nM). Cells were cultured for 24 h, followed by fixation, staining, and visualization of the cells under microscope to count the cells. (A and C) Number of cells counted in the control insert (migration). (B and D) Number of cells that invaded through the Matrigel matrix (invasion). Data were analyzed by comparing each treatment to media only using nonparametric Student t-test in GraphPad Prizm v5. Significance was set at P < 0.05 and represented by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. Discussion There are many reports of high levels of CXCR4 and its ligand, CXCL12, in endometriosis [18, 19, 31, 36–38]. Thus, it has been hypothesized that the CXCR4-CXCL12 axis might be an effective nonhormonal target for endometriosis, since blocking its action could interfere with activation of key disease mechanisms. To address this hypothesis, we assessed the protein expression profile of CXCR4 in glands and stroma of different types of endometriotic lesions, and studied cellular responses (proliferative, migratory, invasive) mediated by activation and blocking of the CXCR4-CXCL12 axis in vitro. Our study showed that pharmacologically blocking the CXCR4 receptor has contrasting effects depending on the presence of the ligand, and possibly on the concomitant expression of other receptors. These data support additional research on the mechanisms involved when this axis is targeted in the context of endometriosis, in order to support (or not) further evaluation of CXCR4-blocking drugs in preclinical studies and clinical trials. Immunohistochemical analysis of lesions on an endometriosis-focused tissue array showed that nuclear CXCR4 (nCXCR4) protein expression levels were generally higher in ectopic lesions compared to proliferative endometrium from controls, as shown by others [37, 38]. Our study of CXCR4 expression using a tissue array allowed us to document differences in the expression levels in different lesion types/localizations, and by cell type (glands vs. stroma). In general, expression of nCXCR4 protein was higher in the stromal compartment, and significantly higher in ovarian compared to fallopian tube lesions, which may be explained by different pathophysiological mechanisms underlying the various lesion types observed in patients. Notably, the expression of nCXCR4 in proliferative endometrium of controls was lower compared to secretory endometrium, while eutopic endometrium from patients had a level of nCXCR4 expression similar to that of ectopic lesions. Cytoplasmic expression of CXCR4, on the other hand, was higher in the glandular compartment, in accord with our own findings and those of others [39], but we did not observe significant differences among tissue types. Nuclear localization of CXCR4 indicates activation of the receptor possibly leading to the endometriotic phenotypes (i.e., proliferation, migration, invasion) that we observed here to be activated in vitro. It is important to highlight that activation of the CXCR4 receptor varied depending on lesion type/localization, being higher in ovarian lesions, which may impact response to therapy with CXCR4 blockers. For example, this treatment may be more effective in women with ovarian endometriosis in particular. CXCR4 is also expressed in eutopic endometrium from patients. This is important to document since the efficacy of pharmacological interventions for endometriosis via blocking this axis will likely depend on the levels of and the tissue expression profile of CXCR4. Also of interest is that CXCR4 is highly expressed in many cancers including ovarian, and that ovarian lesions have been associated with malignant transformation to ovarian cancers (endometrioid and clear cell types) [40, 41]. Whether CXCR4 could constitute a biomarker of transformation potential, alone or in combination with other markers (e.g., Ki67), needs to be studied further. We also show here that the endometriotic epithelial cell line 12Z expresses higher basal levels of CXCR4 compared to the control cell lines, HESC and EEC. 12Z cells also had a higher proliferation when compared to the control cells, thus confirming previous observations [42]. CXCL12 did not increase the proliferation rate of 12Z cells despite them expressing the CXCR4 receptor. It is possible that 12Z cells already have a high basal proliferation rate that cannot be induced further by CXCR4-CXCL12 axis activation. In contrast, the epithelial cell line EEC did respond to CXCL12 by increasing its proliferation rate, showing that the ligand-receptor interaction used in these experiments is functional, and suggesting that this effect may be mediated by the CXCR4 receptor. The endometrial stromal cells HESC expressed CXCR4, as shown here, but no increase in proliferation was observed in response to treatment with CXCL12. It is possible that the receptor is nonfunctional because the interaction, function, or activities of G proteins are known to be cell context-dependent (i.e., epithelial vs stromal) [43]. In addition, other cellular responses could have been activated by the ligand through this receptor in the stromal cells, such as autophagy, that we did not assess in the present study [44]. The proliferation of the epithelial control cells EEC, but not of 12Z cells, was induced by stromal cell-conditioned media (HESC and PED) cells compared to control media. The proliferative effect of the PED-conditioned media on EEC proliferation rates was not reduced by blocking CXCR4 or CXCL12 with neutralizing antibodies, suggesting that other secreted factors might be mediating this effect. In the context of endometriosis, it can be argued that the shed endometrium tissue reaching the peritoneal cavity could initiate an inflammatory reaction, and factors secreted by inflammatory cells (and also by the stromal component of the refluxed endometrium) could induce proliferation of the EEC. In addition, other relevant processes could be activated by numerous factors secreted into the peritoneal fluid, including degradation of extracellular matrix, adhesion, and angiogenesis leading to the development of ectopic lesions. It would be important that follow-up studies are conducted to identify the stromal-derived factor(s) that may be mediating the observed growth-promoting effects on epithelial cells. It is well documented that the CXCR4-CXCL12 axis activates the AKT pathway; therefore, in order to assess its functionality in the cell lines studied, we conducted AKT phosphorylation and kinase activity assays. We observed that in the control epithelial cells (EEC), a transient increase in AKT phosphorylation was observed at the lowest dose of CXCL12. In contrast, there was an increase in AKT phosphorylation associated with the higher CXCL12, which was sustained for up to 30 min. Accordingly, we also observed sustained AKT kinase activity only in the endometriotic 12Z cells. Taken together, these results suggest that in endometriotic cells the AKT signaling pathway is more CXCL12-responsive (i.e., is differentially modulated by both incubation time and ligand concentration) than in the control cells, although in both cell lines this signaling pathway is functional (as demonstrated by GSK substrate phosphorylation). The observed higher levels of AKT activity induced by 10 ng/ml of CXCL12 in 12Z cells may explain in part the increased invasiveness induced by CXCL12, an effect that was not seen in the control cells at any of the dose studied. It is important to take into account that there are other pathways that can be activated by the CXCR4-CXCL12 axis, including phosphoinositol-3-kinase, mitogen-activated protein kinase, Src kinase, and focal adhesion (PyK2, FAK) pathways, which were not studied here [45]. Therefore, more studies are necessary to dissect the signal transduction mechanisms activated in response to CXCL12 in order to understand the contribution of chemokine to cellular functions relevant to endometriosis. This study also demonstrated that the endometriotic epithelial cells 12Z have a higher basal migratory potential compared to the control epithelial cell line EEC. Treatment of 12Z cells with the CXCR4 antagonist, AMD3100, significantly decreased migration but, intriguingly, increased their invasion capacity. Combining the drug with CXCL12 counteracted this effect (Figure 7). Recently, an additional receptor, CXCR7, has been reported to bind to CXCL12 [46]. While limited, there is some evidence that CXCR7 is expressed by endometriotic lesions, endometrial cells, and endometrial cancer [47, 48]. AMD3100 is an allosteric agonist for CXCR7 [49], known to bind and activate CXCR7, but the molecular effects resulting from this interaction are not well understood. We hypothesize that the CXCR4:CXCL12 interaction is primarily responsible for inducing migration of 12Z cells, while the CXCR7:CXCL12 interaction leads to invasion, a possibility that needs to be experimentally proven. AMD3100 by itself significantly decreased EEC migration, but none of the treatments (ligand, inhibitor, or their combination) induced their invasion. Taken together, these data suggest that the endometriotic epithelial cells migrate more in response to low concentrations of the ligand; these cells do not produce CXCL12 and can be more readily sensitive to exogenous sources of CXCL12 in the peritoneal environment. While their migration can be reduced by the CXCR4 inhibitor, treatment with this drug results in pleiotropic responses leading to increases in invasion, limiting its possible use as a candidate nonhormonal treatment for endometriosis unless it is given in combination with CXCL12. Figure 7. View largeDownload slide Summary of studied CXCR4-CXCL12 axis in endometriosis. CXCR4-CXCL12 axis activation produces a rapid and transient effects of AKT phosphorylation. This phosphorylation increased AKT activity and may induce cell proliferation migration and invasion. Treatment of 12Z with AMD3100 decreased migration and increased invasion suggesting the activation of different biological mechanism. Figure 7. View largeDownload slide Summary of studied CXCR4-CXCL12 axis in endometriosis. CXCR4-CXCL12 axis activation produces a rapid and transient effects of AKT phosphorylation. This phosphorylation increased AKT activity and may induce cell proliferation migration and invasion. Treatment of 12Z with AMD3100 decreased migration and increased invasion suggesting the activation of different biological mechanism. In conclusion, our results provide evidence that CXCR4 is differentially expressed across lesion types, being highest in ovarian endometriosis. In addition, endometriotic epithelial cells (12Z) express high levels of CXCR4, are highly migratory, and have a functional CXCR4-CXCL12 axis. Finally, our data suggest that not all cancer-targeted drugs may also be effective in endometriosis despite both diseases sharing the same cellular behaviors. Therefore, before targeting this pathway therapeutically, additional experiments including preclinical investigations in animal models are needed to dissect the molecular pathways activated by CXCL12 alone or in combination with receptor blocking drugs. Supplementary data Supplementary data are available at BIOLRE online. Supplementary Table S1. Antibodies Table. Acknowledgments We acknowledge the role of Dr Miosotis García in the selection and pathological analysis of samples used in the development of the endometriosis tissue array, pathology consultation by Dr Adalberto Mendoza of Southern Pathology Services, Inc., and technical help provided by Perla Báez, PhD. We thank Dr. Virgilio A. Salvo for discussions in experimental design and data interpretation as well as providing key reactives. Conflict of Interest: The authors have declared that no conflict of interest exists. † Grant Support: This study was supported by the National Institutes of Health–National Institute of Child Health and Human Development grants # R01-HD050559 (IF), F31HD072594 (AR), 3R01-HD050559-01A1S1 (LR), and R01-HD082373 (ATF); by the National Institutes of Health–National Institute of General Medical Sciences grants # S06-GM08239-20 (IF), R25-GM082406 (PHSU RISE Program) (AR and MCC), and R25-GM096955 (MCC and BJTC); by National Institutes of Health–National Institute on Minority Health and Health Disparities grants # G12-MD007579 (Molecular Biology and Genomics Core); and by National Institutes of Health–National Institute of Cancer grant 3 U56-CA126379 (construction of the tissue array). Edited by Dr. Romana Nowak, PhD, University of Illinois Urbana-Champaign. References 1. Vercellini P et al.   Endometriosis: pathogenesis and treatment. Nat Rev Endocrinol  2014; 10: 261– 275. Google Scholar CrossRef Search ADS PubMed  2. Macer ML , Taylor HS. Endometriosis and Infertility. Obstet Gynecol Clin North Am  2012; 39( 4): 535– 549. Google Scholar CrossRef Search ADS PubMed  3. Park DW , Yang KM. Hormonal regulation of uterine chemokines and immune cells. Clin Exp Reprod Med  2011; 38( 4): 179– 185. Google Scholar CrossRef Search ADS PubMed  4. Piao HL et al.   CXCL12/CXCR4 signal involved in the regulation of trophoblasts on peripheral NK cells leading to Th2 bias at the maternal-fetal interface. Eur Rev Med Pharmacol Sci  2015; 19( 12): 2153– 2161. Google Scholar PubMed  5. Noda M et al.   CXCL12-CXCR4 chemokine signaling is essential for NK-cell development in adult mice. Blood  2011; 117( 2): 451– 458. Google Scholar CrossRef Search ADS PubMed  6. Domanska UM et al.   A review on CXCR4/CXCL12 axis in oncology: no place to hide. Eur J Cancer  2013; 49( 1): 219– 230. Google Scholar CrossRef Search ADS PubMed  7. Li SF et al.   The number of proliferating cell nuclear antigen positive cells in endometriotic lesions differs from that in the endometrium. Vichows Archiv A Pathol Anat  1993; 423( 4): 257– 263. Google Scholar CrossRef Search ADS   8. Nisolle M , Casanas-Roux F, Donnez J. Immunohistochemical analysis of proliferative activity and steroid receptor expression in peritoneal and ovarian endometriosis. Fertil Steril  1997; 68( 5): 912– 919. Google Scholar CrossRef Search ADS PubMed  9. Meresman GF et al.   Oral contraceptives suppress cell proliferation and enhance apoptosis of eutopic endometrial tissue from patients with endometriosis. Fertil Steril  2002; 77( 6): 1141– 1147. Google Scholar CrossRef Search ADS PubMed  10. Matsuzaki S , Darcha C. In vitro effects of a small-molecule antagonist of the tcf/ss-catenin complex on endometrial and endometriotic cells of patients with endometriosis. PLoS One  2013; 8( 4): e61690. Google Scholar CrossRef Search ADS PubMed  11. Valles CS , Dominguez F. Embryo-endometrial interaction. Chang Gung Med J  2006; 29( 1): 9– 14. Google Scholar PubMed  12. Cocchi F et al.   Identification of RANTES, MIP-1alpha, and MIP-1beta as the Major HIV-Suppressive Factors Produced by CD8+ T Cells. Science  1995; 270( 5243): 1811– 1815. Google Scholar CrossRef Search ADS PubMed  13. Simon C et al.   Potential implications of chemokines in reproductive function: an attractive idea. J Reprod Immunol  1998; 38( 2): 169– 193. Google Scholar CrossRef Search ADS PubMed  14. Herington JL et al.   Immune interactions in endometriosis. Expert Rev Clin Immunol 2011; 7( 5): 611– 626. 15. Bellelis P et al.   Transcriptional changes in the expression of chemokines related to natural killer and T-regulatory cells in patients with deep infiltrative endometriosis. Fertil Steril  2013; 99( 7): 1987– 1993. Google Scholar CrossRef Search ADS PubMed  16. Kitaya K et al.   Spatial and temporal expression of ligands for CXCR3 and CXCR4 in human endometrium. J Clin Endocrinol Metab  2004; 89( 5): 2470– 2476. Google Scholar CrossRef Search ADS PubMed  17. Leconte M et al.   Role of the CXCL12-CXCR4 axis in the development of deep rectal endometriosis. J Reprod Immunol  2014; 103: 45– 52. Google Scholar CrossRef Search ADS PubMed  18. Furuya M et al.   Up-regulation of CXC chemokines and their receptors: implications for proinflammatory microenvironments of ovarian carcinomas and endometriosis. Hum Pathol  2007; 38( 11): 1676– 1687. Google Scholar CrossRef Search ADS PubMed  19. Chand AL et al.   Laser capture microdissection and cDNA array analysis of endometrium identify CCL16 and CCL21 as epithelial-derived inflammatory mediators associated with endometriosis. Reprod Biol Endocrinol  2007; 5( 1): 18. Google Scholar CrossRef Search ADS PubMed  20. Flores I et al.   Molecular profiling of experimental endometriosis identified gene expression patterns in common with human disease. Fertil Steril  2007; 87( 5): 1180– 1199. Google Scholar CrossRef Search ADS PubMed  21. Colon-Diaz M et al.   HDAC1 and HDAC2 are differentially expressed in endometriosis. Reprod Sci  2012; 19( 5): 483– 492. Google Scholar CrossRef Search ADS PubMed  22. Ruiz LA et al.   Dysregulation of lysyl oxidase expression in lesions and endometrium of women with endometriosis. Reprod Sci  2015; 22( 12): 1496– 1508. Google Scholar CrossRef Search ADS PubMed  23. Colon-Caraballo M , Monteiro JB, Flores I. H3K27me3 is an epigenetic mark of relevance in endometriosis. Reprod Sci  2015; 22( 9): 1134– 1142. Google Scholar CrossRef Search ADS PubMed  24. Noyes RW , Hertig AT, Rock J. Dating the endometrial biopsy. Am J Obstet Gynecol  1975; 122( 2): 262– 263. Google Scholar CrossRef Search ADS PubMed  25. Ruiz A et al.   Basal and steroid hormone-regulated expression of CXCR4 in human endometrium and endometriosis. Reprod Sci  2010; 17( 10): 894– 903. Google Scholar CrossRef Search ADS PubMed  26. Korch C et al.   DNA profiling analysis of endometrial and ovarian cell lines reveals misidentification, redundancy and contamination. Gynecol Oncol  2012; 127( 1): 241– 248. Google Scholar CrossRef Search ADS PubMed  27. Zeitvogel A , Baumann R, Starzinski-Powitz A. Identification of an invasive, N-cadherin-expressing epithelial cell type in endometriosis using a new cell culture model. Am J Pathol  2001; 159( 5): 1839– 1852. Google Scholar CrossRef Search ADS PubMed  28. Krikun G et al.   A novel immortalized human endometrial stromal cell line with normal progestational response. Endocrinology  2004; 145( 5): 2291– 2296. Google Scholar CrossRef Search ADS PubMed  29. Hombach-Klonisch S et al.   Regulation of functional steroid receptors and ligand-induced responses in telomerase-immortalized human endometrial epithelial cells. J Mol Endocrinol  2005; 34( 2): 517– 534. Google Scholar CrossRef Search ADS PubMed  30. Tsutsumi A et al.   Estrogen induces stromal cell-derived factor 1 (SDF-1/CXCL12) production in human endometrial stromal cells: a possible role of endometrial epithelial cell growth. Fertil Steril  2011; 95: 444– 447. Google Scholar CrossRef Search ADS PubMed  31. Ruiz A et al.   Basal and steroid hormone-regulated expression of CXCR4 in human endometrium and endometriosis. Reprod Sci  2010; 17( 10): 894– 903. Google Scholar CrossRef Search ADS PubMed  32. Laird SM et al.   Expression of CXCL12 and CXCR4 in human endometrium; effects of CXCL12 on MMP production by human endometrial cells. Hum Reprod  2011; 26( 5): 1144– 1152. Google Scholar CrossRef Search ADS PubMed  33. Oberlin E et al.   The CXC chemokine SDF-1 is the ligand for LESTR/fusin and prevents infection by T-cell-line-adapted HIV-1. Nature  1996; 382( 6594): 833– 835. Google Scholar CrossRef Search ADS PubMed  34. Yu J et al.   Combination of estrogen and dioxin is involved in the pathogenesis of endometriosis by promoting chemokine secretion and invasion of endometrial stromal cells. Hum Reprod  2008; 23( 7): 1614– 1626. Google Scholar CrossRef Search ADS PubMed  35. Kitaya K et al.   Spatial and temporal expression of ligands for CXCR3 and CXCR4 in human endometrium. J Clin Endocrinol Metab  2004; 89( 5): 2470– 2476. Google Scholar CrossRef Search ADS PubMed  36. Borrelli GM et al.   Immunohistochemical investigation of metastasis-related chemokines in deep-infiltrating endometriosis and compromised pelvic sentinel lymph nodes. Reprod Sci  2015; 22( 12): 1632– 1642. Google Scholar CrossRef Search ADS PubMed  37. Leconte M et al.   Role of the CXCL12-CXCR4 axis in the development of deep rectal endometriosis. J Reprod Immunol  2014; 103: 45– 52. Google Scholar CrossRef Search ADS PubMed  38. Moridi I et al.   Bone marrow stem cell chemotactic activity is induced by elevated CXCL12 in endometriosis. Reprod Sci  2017; 24: 526– 533. Google Scholar CrossRef Search ADS PubMed  39. Laird SM et al.   Expression of CXCL12 and CXCR4 in human endometrium; effects of CXCL12 on MMP production by human endometrial cells. Hum Reprod  2011; 26( 5): 1144– 1152. Google Scholar CrossRef Search ADS PubMed  40. Pearce CL et al.   Association between endometriosis and risk of histological subtypes of ovarian cancer: a pooled analysis of case-control studies. Lancet Oncol  2012; 13( 4): 385– 394. Google Scholar CrossRef Search ADS PubMed  41. Lee AW et al.   Evidence of a genetic link between endometriosis and ovarian cancer. Fertil Steril  2016; 105( 1): 35– 43. e1–10. Google Scholar CrossRef Search ADS PubMed  42. Kahyaoglu I et al.   Comparison of Ki-67 proliferative index between eutopic and ectopic endometrium: a case control study. Taiwan J Obstet Gynecol  2012; 51( 3): 393– 396. Google Scholar CrossRef Search ADS PubMed  43. Busillo JM , Benovic JL. Regulation of CXCR4 signaling. Biochim Biophys Acta  2007; 1768( 4): 952– 963. Google Scholar CrossRef Search ADS PubMed  44. Mei J et al.   Estrogen promotes the survival of human secretory phase endometrial stromal cells via CXCL12/CXCR4 up-regulation-mediated autophagy inhibition. Hum Reprod  2015; 30( 7): 1677– 1689. Google Scholar CrossRef Search ADS PubMed  45. Arcaro A , Guerreiro AS. The phosphoinositide 3-kinase pathway in human cancer: genetic alterations and therapeutic implications. Curr Genomics  2007; 8( 5): 271– 306. Google Scholar CrossRef Search ADS PubMed  46. Balabanian K et al.   The chemokine SDF-1/CXCL12 binds to and signals through the orphan receptor RDC1 in T lymphocytes. J Biol Chem  2005; 280( 42): 35760– 35766. Google Scholar CrossRef Search ADS PubMed  47. Zhou WH et al.   Co-expression of CXCR4 and CXCR7 in human endometrial stromal cells is modulated by steroid hormones. Int J Clin Exp Pathol  2015; 8( 3): 2449– 2460. Google Scholar PubMed  48. Gu HQ et al.   The role of the SDF-1/ CXCR7 axis on the growth and invasion ability of endometrial cancer cells. Arch Gynecol Obstet  2017; 295( 4): 987– 995. Google Scholar CrossRef Search ADS PubMed  49. Kalatskaya I et al.   AMD3100 is a CXCR7 ligand with allosteric agonist properties. Mol Pharmacol  2009; 75( 5): 1240– 1247. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biology of Reproduction Oxford University Press

Pharmacological blockage of the CXCR4-CXCL12 axis in endometriosis leads to contrasting effects in proliferation, migration, and invasion

Loading next page...
 
/lp/ou_press/pharmacological-blockage-of-the-cxcr4-cxcl12-axis-in-endometriosis-RnvbSIaZwV
Publisher
Society for the Study of Reproduction
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0006-3363
eISSN
1529-7268
D.O.I.
10.1093/biolre/iox152
Publisher site
See Article on Publisher Site

Abstract

Abstract High levels of inflammatory factors including chemokines have been reported in peritoneal fluid and blood of women with endometriosis. CXCL12 mediates its action by interaction with its specific receptor, CXCR4, reported to be elevated in human endometriosis lesions and in the rat model of endometriosis. Activation of the CXCR4-CXCL12 axis increases cell proliferation, migration, and invasion of cancer cells. To obtain insights into the CXCR4 expression profile in lesions and endometrium, as well as functionality of the CXCR4-CXCL12 axis in endometriosis, we analyzed the expression of CXCR4 in tissues on a human tissue array and studied CXCL12-mediated activation of proliferation, invasion, and migration in vitro. We observed differences in levels of nuclear CXCR4 expression among lesion types, being higher in ovarian lesions. Endometriotic cell lines (12Z) showed higher levels of CXCR4, proliferative and migratory potential, and AKT phosphorylation/kinase activity compared to untreated control cells (endometrial epithelial cells). CXCL12 and endometriotic stromal cell-enriched media increased proliferation of non-endometriotic epithelial cells. CXCL12 caused a significant increase in 12Z cell invasion but had no effect on migration; AMD3100, a CXCR4-specific inhibitor, significantly increased invasion of 12Z cells but decreased their migration. However, treatment with CXCL12 plus AMD3100 significantly decreased invasion and migration of 12Z cells. In conclusion, the CXCR4-CXCL12 axis is functional in endometriosis cells, but the expression of CXCR4 varies among lesions. CXCL12 promoted proliferation, migration, and invasion of endometriotic cells, while inducing AKT phosphorylation and activity, but pharmacologically blocking this axis in the absence of the ligand induced their invasiveness. Introduction The pathologic mechanisms underlying endometriosis development and progression remain poorly understood, although genetic, epigenetic, environmental, and immune/inflammatory factors have been implicated. Most of the available treatments for endometriosis are hormones that cause undesired side effects while not being curative [1, 2]. Thus, a greater understanding of this complex and multifactorial disease and dissection of the biological and signal transduction mechanisms that contribute to its pathophysiology is needed in order to facilitate the identification of novel non-hormonal candidate therapeutic targets. When CXCL12 binds its specific receptor, the G protein coupled receptor CXCR4, it activates mechanisms related to normal reproductive biology (e.g., menstruation, ovulation, trophoblast implantation, and parturition [3, 4]) as well as immunity (e.g., natural killer cells and T cells [5]). Hyperactivation of the CXCR4-CXCR12 axis results in disease processes such as tumorigenesis, invasion, and metastasis [6]. This axis is involved in various cancer cell behaviors that also characterize endometriosis (e.g., increased basal proliferation rates [7–9]; and increased invasiveness [10]). Although few, there are some studies providing support for the involvement of the CXCR4-CXCL12 axis in endometriosis: (i) there are high levels of inflammatory factors including chemokines in the peritoneal fluid and serum of women with endometriosis, including chemokines such as CXCL12 [11–14]; CXCL12 is overexpressed in endometriosis lesions compared to eutopic endometrium from patients and controls [15–17]; (ii) CXCR4 protein expression is highest in both ovarian endometriosis and endometrial carcinoma when compared to normal ovary [18]; (iii) global gene profiling of endometriotic lesions using laser-captured microdissected epithelial cells identified a 20-fold increased expression of CXCR4 when compared to normal endometrial epithelial cells (EEC) [19]; and (iv) lesions from the rat model of endometriosis have higher levels of CXCR4 gene expression when compared to endometrium from controls [20]. Based on these data, we hypothesized that the CXCR4-CXCL12 axis could play a role in the development or progression of endometriosis. Thus, the aim of this study was to obtain insights into the profile of expression across different lesions types and into the functionality of the CXCR4-CXCL12 axis in endometriosis, as a first step to explore the possibility of targeting this axis as a therapy for this incapacitating disease. We show here that there are functional interactions between CXCL12 and CXCR4 in endometriotic epithelial cells that leads to increased proliferation, invasion, and migration, as well as phosphorylation and activation of serine-threonine protein kinase (AKT), which may facilitate their ectopic survival and growth. However, we also show that there were unforeseen effects of the CXCR4 inhibitor that could preclude its clinical use in endometriosis as a single agent. Materials and methods Immunohistochemistry of CXCR4 in human endometriotic lesions on a tissue array The endometriosis-focused tissue array used for this study has been previously described and used for other investigations [21–23]. All studies involving human tissues were approved by the Internal Review Board of the Ponce Health Sciences University (PHSU). A total of 164 deidentified endometriotic tissue blocks were used to construct a tissue array in collaboration with the PHSU–Moffitt Cancer Center Partnership. To account for the heterogeneity of endometriosis, this tissue array includes lesions localized in the ovaries (n = 29), fallopian tubes (n = 16), peritoneum (n = 34), skin (umbilical region) (n = 4), and gastrointestinal tract (n = 7). Additionally, the tissue array includes endometrial samples obtained from patients with endometriosis and controls (women undergoing hysterectomy for benign uterine disorders) in the proliferative and secretory phases of the menstrual cycle. All blocks were analyzed by two pathologists to document endometriosis (glands and stroma) and determine the menstrual cycle phase in the case of endometrium according to Noyes criteria [24]. The tissue array block was cut in 10 μm for immunostaining. Immunohistochemistry was conducted as described by Ruiz et al. [25]. Primary antibody against CXCR4 was used at a 1:300 dilution (R4 monoclonal-clone 44716) (R&D Systems Minneapolis, MN) (Table S1). Negative control for the immunostaining was done with normal serum instead of primary antibody in control tissue slides. Immunohistochemical staining (IHC) intensity was evaluated in glands and stroma using a semiquantitative method. Staining intensity was blindly scored as absent (0), weak (1), moderate (2), or intense (3) by two independent, blinded observers. Mean IHC intensity score and interquartile range were calculated per compartment per sample for statistical analysis. Cell cultures Human endometrial stromal cells (HESC), highly invasive and immortalized with human Telomerase reverse transcriptase (hTert), were purchased from American Type Culture Collection (#CRL 4003). Noninvasive epithelial cells (EEC, which have been shown to have MCF7 characteristics by DNA fingerprinting [26]), and human endometriotic epithelial cells (12Z), SV40 T antigen-transformed and derived from a peritoneal lesion, were obtained as part of a collaboration with Dr Fazleabas and Dr Starzinski [27]. The 12Z were grown Dulbecco's Modified Eagle's Medium (DMEM)/F12, 10% Fetal Bovine Serum (FBS); HESC were grown in phenol-free DMEM, charcoal-treated 10% FBS, 1% insulin-transferrin-selenium (ITS); the medium for EEC cells was DMEM/F12, 10% FBS, 160 ng/ml insulin as previously described [28, 29]. Endometriotic stromal cells derived from a peritoneal endometriotic lesion and grown at low passage numbers (Primary Endometriotic Cells (PED)) (developed in Dr Idhaliz Flores laboratory; Bello et al. unpublished data) were cultured in complete DMEM supplemented with 10−8 mol/L estradiol and 1% ITS. Estradiol and ITS were removed from media 24 h prior to experiments. Enzyme-linked immunosorbent assay for CXCL12 The cell lines 12Z, PED, HESC, and EEC were cultured as previously described [21, 22]. The cells were cultured during 24, 48, and 72 h. Cell culture supernatant was collected, centrifuged, and stored in aliquots at –20°C. Levels of human CXCL12 were determined using a Quantikine enzyme-linked immunosorbent assay (ELISA) (R&D systems, Minneapolis, MN) following manufacturer's instructions and as described before [30]. Briefly, the assay diluent was added to samples (in triplicate) and to the standards (serially diluted positive control), and incubated for 2 h at room temperature (RT). CXCL12-specific antibody was added to each well and incubated at RT for 2 h followed by washes. The substrate solution was then added to each well and incubated in the dark. Finally, the optical density was determined using a microplate reader (MRX Revelation, Dynex Technologies, Inc., Chantilly, VA) set to 450 nm with correction wavelength set at 540 nm. Western blotting for CXCR4 Total protein extraction from the cell cultures, quantification, and western blot analysis have previously been described in detailed [31]. Dilutions of anti-CXCR4 (monoclonal-clone 44716) (R&D Systems Minneapolis, MN) and GAPDH (monoclonal-DM1A) (Calbiochem, Darmstadt, Germany) used were 1:300 and 1:1000, respectively (Table S1). Experiments were repeated three times. Modulation of in vitro cell proliferation by CXCL12 or stromal cell-enriched media The BrdU cell proliferation assay kit was used to study 12Z and EEC cell proliferation in response to CXCL12, primary endometriotic stromal cell (PED)-enriched media, or complete media as control, following the manufacturer's instructions (Millipore, Temecula, CA). The amount of the incorporated BrdU, indicative of newly synthesized DNA, was detected using a microplate reader (MRX Revelation, Dynex Technologies, Inc., Chantilly, VA) set at dual wavelengths of 450/550 nm. As blank we used cell media with no cells and as background cells without the BrdU reagent. To measure profilerative responses to CXCL12, epithelial cells (12Z or EEC) were seeded at 4000 and 2000 densities in 100 μl media in a flat bottom 96-well plate overnight. CXCL12 (hrCXCL12 alpha, R&D Systems, Inc., Minneapolis, MN) was added to the media to a final concentration of 10 and 100 ng/ml, and proliferation rates were measured by BrdU incorporation. These doses were based on a published study by Laird et al. [32]. To determine the proliferative effects of the stromal cell-enriched media on the epithelial cells (12Z and EEC), PED (primary endometriotic) and HESC (hTERT immortalized stromal endometrial) cells were cultured at a density of 1.0 × 106 cells in 2 ml per well in a 6-well plate without estrogen and ITS. After 24 h of culture, conditioned media was collected, filtered, and added to 12Z and EEC cells for 24 h. Proliferation rates were measured by BrdU incorporation following manufacturer's protocol. Experiments were repeated at least three times. Modulation of phospho-AKT by CXCL12 Cells were grown and incubated with media containing 10 or 100 ng/ml of human recombinant (hr) CXCL12 alpha, or complete media as control, for 10 s, 1, 5, 10, 15, and 30 min. Cells were lysed with RIPA lysis buffer supplemented with protease and phosphatase inhibitors (Roche Diagnostics, Indianapolis, IN). Protein concentration was determined using a standard curve generated by a Bradford protein assay. Prior to Western Blot (WB), total AKT was immunoprecipitated. Briefly, proteins were diluted in lysis buffer containing protease and phosphatase inhibitors and 50 μl of Sepharose protein A (Rockland Gilbertsville, PA) was added, followed by addition of anti-AKT antibody or phospho-AKT (Cell Signaling Technology, Danvers, MA), and incubation overnight at 4°C with gentle rocking. After centrifugation and washes, membranes were incubated with horseradish peroxidase-conjugated secondary antibody, and signals were detected with an automated image documentation system (ChemiDoc, Bio-Rad, Hercules). Band density was analyzed using Quantity One software (Bio-Rad, Hercules, CA). Experiments were repeated three times. Modulation of AKT kinase activity by CXCL12 Cell lines (12Z and EEC) were cultured as previously described. Media containing 10 or 100 ng/ml CXCL12, or complete media as control, were added during 1, 10, 30 min. Cellular lysates were analyzed using a nonradioactive immunoprecipitation-kinase assay to detect total phospho-AKT at serine 473 and phosho-glycogen synthase kinase 3 (GSK-3)α/β(ser21/9) (AKT Kinase Activity kit, Cell Signaling Technology, Danvers, MA) following the manufacturer's protocol. In brief, after treatments, cells were lysed and sonicated 10 min on ice, followed by centrifugation. Cell lysates were incubated with beads/immobilized antibody overnight with gentle rocking at 4°C. Cell lysate containing immobilized antibody was centrifuged, and after washes the pellet was suspended in 50 μl in kinase buffer supplemented with 1 μl of 10 nM ATP and the kinase substrate, GSK-3α/β fusion protein. Samples were incubated during 30 min at 30°C. The reaction was terminated by adding 25 μl of 3× sodium dodecyl sulfate (SDS) buffer, vortex, and microcentrifugation. WB analyses were performed as previously described to detect phospho-AKT and phospho-GSK-3α/β. Experiments were performed in triplicate. Modulation of in vitro migration and invasion by CXCL12 and/or AMD3100 Cell migration and invasion of 12Z and EEC in response to CXCL12 plus/minus the CXCR4-specific inhibitor AMD3100 (Sigma-Aldrich, St. Louis, MO) were quantified using BD BioCoat Matrigel Invasion Chamber following standard protocols (BD Biosciences, Bedford, MA). In brief, epithelial cells were grown in the upper chambers in complete media or media with 100 nM AMD3100. Complete media supplemented with 10 or 100 ng/ml CXCL12 was added to the lower chamber in 24-well plates. Wells with Matrigel-coated inserts or control inserts were used to differentiate invasion from migration, respectively [33, 34]. After 24 h, the cells were fixed and stained with Diff-Quik stain (Dade Behring, Inc. Newark, DE) following the manufacturer's instructions. The inserts were allowed to dry and mounted in microscope slides for counting and visualization of the invading and migrating cells using a light microscope. Experiments were repeated three times. Statistical analyses Data from cell culture experiments (e.g., proliferation, AKT phosphorylation, AKT function, migration and invasion) were analyzed by Mann–Whitney, non-parametric Student t-test, compared to basal or untreated conditions. Analysis of variance (ANOVA) using Tukey and Dunn multiple comparison tests was used to analyze tissue array immunostaining differences. All analyses were conducted using SPSS 15.0 (Chicago, IL, USA) and GraphPad Prizm v5 was used for graphical representation of the data. Statistical significance was set at P < 0.05. Results Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array We analyzed by IHC the protein expression of the chemokine receptor CXCR4 in human endometriosis lesions from five different anatomical sites (ovaries, peritoneum, fallopian tubes, skin, and gastrointestinal tract) as well as eutopic endometrium from women with endometriosis and controls included in a custom-made endometriosis-focused tissue array. From 164 samples, 137 core biopsies (84%) could be analyzed; biopsies that did not include stroma and glands were excluded from the analysis. Nuclear CXCR4 (nCXCR4) expression was in general higher in stroma compared to glands, and significantly higher in the stroma of ovarian endometriosis compared to fallopian tube lesions and proliferative endometrium from controls. In that respect, the proliferative endometrium from patients showed a similar expression of nCXCR4 than endometriotic lesions (Figure 1). nCXCR4 expression in glands was highest in ovarian compared to both fallopian lesions and proliferative endometrium from cases and controls. Cytoplasmic CXCR4 expression in stroma was not significantly different among the tissues analyzed, although lesions showed a slightly increased level of cCXCR4 expression compared to endometrial tissues (P = 0.0343). Figure 1. View largeDownload slide Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array. (A) A total of 164 formalin-fixed paraffin-embedded endometrial and endometriotic human on a tissue array were analyzed by immunohistochemistry. The immunostaining intensity of CXCR4 in nuclear and cytoplasmic compartment of the stromal and glands cells was evaluated and shown graphically. The data were analyzed by ANOVA and with Dunn Multiple Comparison post hoc test, the statistical significance level among them are indicated by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. (B) Representative pictures showing immunostaining in different lesion types are shown. Figure 1. View largeDownload slide Immunostaining of CXCR4 in human endometrium and endometriotic lesions on a tissue array. (A) A total of 164 formalin-fixed paraffin-embedded endometrial and endometriotic human on a tissue array were analyzed by immunohistochemistry. The immunostaining intensity of CXCR4 in nuclear and cytoplasmic compartment of the stromal and glands cells was evaluated and shown graphically. The data were analyzed by ANOVA and with Dunn Multiple Comparison post hoc test, the statistical significance level among them are indicated by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. (B) Representative pictures showing immunostaining in different lesion types are shown. In vitro CXCR4 and CXCL12 expression CXCR4 protein expression was analyzed in 12Z, HESC, and EEC by WB. We showed higher levels of CXCR4 expression in 12Z cells compared to EEC (Figure 2). HESC, an endometrial stromal cells also expressed CXCR4. ELISA results showed that none of the cell lines studied (12Z, HESC, EEC, PED) expressed CXCL12 alpha, or its expression was below detection levels (data not shown). This contrasts with the findings of increased levels of CXCL12 in human endometriotic tissues. It is possible that other CXCL12 isoforms (from beta to gamma) not measured here are involved. Others have previously shown that endometrial stromal cell lines do not express CXCL12, contrary to what is seen in whole tissues [35]. Figure 2. View largeDownload slide CXCR4 protein analysis by western blot of endometrial and endometriotic cell lines. Endometrial epithelial (EEC), human endometrial stromal (HESC), and endometriotic epithelial (12Z) cells lines were cultured in complete media. Total protein was extracted, quantified, and separated by electrophoresis. Levels of CXCR4 were analyzed by immunoblotting, and GAPDH was used as loading control. At least three experiments in three different passage numbers were conducted. Differences in the levels of GAPDH can be explained by the different cell types (epithelial vs. stromal), which was normalized by densitometric analysis. Figure 2. View largeDownload slide CXCR4 protein analysis by western blot of endometrial and endometriotic cell lines. Endometrial epithelial (EEC), human endometrial stromal (HESC), and endometriotic epithelial (12Z) cells lines were cultured in complete media. Total protein was extracted, quantified, and separated by electrophoresis. Levels of CXCR4 were analyzed by immunoblotting, and GAPDH was used as loading control. At least three experiments in three different passage numbers were conducted. Differences in the levels of GAPDH can be explained by the different cell types (epithelial vs. stromal), which was normalized by densitometric analysis. Effects of CXCL12 or stromal cell-conditioned media on in vitro proliferation rates We next studied the proliferation rate of 12Z, HESC, and EEC using the BrdU proliferation assay. Our results confirmed that endometriotic epithelial cells, 12Z, had a statistically significant higher basal level proliferation rate compared to the control cell lines HESC and EEC (Figure 3A). Next, we studied the in vitro proliferative effects of CXCL12 in the studied cells lines using the BrdU proliferation assay. Treatment with CXCL12 for 24 h induced a significant increase in EEC cells proliferation rates by 33% (100 ng/ml) and 41% (200 ng/ml); despite expressing CXCR4 neither 12Z nor HESC cells responded to CXCL12 in this manner (Figure 3B). Figure 3. View largeDownload slide Effects of CXCL12 or stromal cells-conditioned media on in vitro cell proliferation. (A) Cells were cultured in a 96-well plate for 48 h and proliferation was measured by BrdU incorporation. Absorbance values were measured at 450/550 nm after 24 h of incubation with BrdU. Endometriotic epithelial (12Z) cells are represented by filled diamonds, noninvasive epithelial cells (EEC) by filled squares, and human EEC by filled triangles. Results represent data from two experiments in triplicate, and are reported as means ± standard error of the mean. (B) Cells were treated with 10 or 100 ng/ml CXCL12 (hrCXCL12 alpha). Basal condition consisted of cells growing in complete media. (C) Cells were cultured in filtered conditioned media from endometrial stromal cells (HESC) or endometriotic stromal cells (PED) that were seeded at a density of 1 × 106 cells and incubated for 24 h in a fixed volume of 2 ml, in the presence or absence of neutralizing antibodies against CXCR4 or CXCL12α. Absorbance value was measured at 450/550 nm after 24 h of incubation with BrdU. Results represent data from three experiments in triplicate. Results represent data from six values per bar (*P ≤ 0.05). Figure 3. View largeDownload slide Effects of CXCL12 or stromal cells-conditioned media on in vitro cell proliferation. (A) Cells were cultured in a 96-well plate for 48 h and proliferation was measured by BrdU incorporation. Absorbance values were measured at 450/550 nm after 24 h of incubation with BrdU. Endometriotic epithelial (12Z) cells are represented by filled diamonds, noninvasive epithelial cells (EEC) by filled squares, and human EEC by filled triangles. Results represent data from two experiments in triplicate, and are reported as means ± standard error of the mean. (B) Cells were treated with 10 or 100 ng/ml CXCL12 (hrCXCL12 alpha). Basal condition consisted of cells growing in complete media. (C) Cells were cultured in filtered conditioned media from endometrial stromal cells (HESC) or endometriotic stromal cells (PED) that were seeded at a density of 1 × 106 cells and incubated for 24 h in a fixed volume of 2 ml, in the presence or absence of neutralizing antibodies against CXCR4 or CXCL12α. Absorbance value was measured at 450/550 nm after 24 h of incubation with BrdU. Results represent data from three experiments in triplicate. Results represent data from six values per bar (*P ≤ 0.05). The epithelial cell lines 12Z and EEC were cultured in conditioned media from the stromal cell lines PED (endometriotic) and HESC (normal) to determine whether there are secreted factors that can induce proliferation of epithelial cells. As controls, neutralizing antibodies to CXCR4 or CXCL12 were added to the culture medium. 12Z cells cultured in media conditioned by HESC cells showed an increase in proliferation. However, this increase was not statistically significant nor was it reduced by the neutralizing antibodies to CXCR4 or CXCL12. EEC cells cultured with media conditioned by PED cells showed a statistically significant proliferation increase of 68% (Figure 3C). The proliferation increased observed in EEC cells cultured in conditioned media was not decreased by treatment with neutralizing antibodies to CXCR4 or CXCL12. Effects of CXCL12 on AKT Serine 473 phosphorylation and AKT kinase activity After normalization against total AKT and relative to basal expression, our data show that the levels of AKT phosphorylation were higher in the endometriotic 12Z cells compared to EEC cells at both CXCL12 doses tested (Figure 4). The higher dose of CXCL12 (100 ng/ml) significantly increased AKT phosphorylation compared to the lower dose (10 ng/ml) (P < 0.05 at 1 min). Also, we observed significant increases in AKT kinase activity in 12Z cells induced by CXCL12 (10 ng/ml) at 1, 10, and 30 min of treatment compared to no treatment (Figure 5). At the higher dose (100 ng/ml) and longer experimental time (30 min), CXCL12 treatment significantly decreased AKT activity in 12Z cells. In contrast, AKT activity was not significantly changed in EEC cells after treatment with 10 and 100 ng/ml CXCL12 at any of the times studied. Figure 4. View largeDownload slide Effects of CXCL12 on AKT Serine 473 phosphorylation. Total protein concentration was determined and subjected to immunoprecipitation against total AKT and immunoblotting against total AKT and AKT phosphorylated at serine 473. (A) Representative western blot of the phosphorylation of AKT in endometriotic epithelial cells (12Z) or noninvasive epithelial cells (EEC) treated with CXCL12. Cell were cultured and treated with either 10 or 100 ng/ml of CXCL12 for 0, 10 s, 1, 5, 10, 15, and 30 min. Basal condition consisted of cells growing in their respective complete media. (B) Graphical representation of densitometry of at least three experiments per cell line. Data were analyzed by nonparametric Student t-test using GraphPad Prizm v5. Significance was set at P < 0.05. Figure 4. View largeDownload slide Effects of CXCL12 on AKT Serine 473 phosphorylation. Total protein concentration was determined and subjected to immunoprecipitation against total AKT and immunoblotting against total AKT and AKT phosphorylated at serine 473. (A) Representative western blot of the phosphorylation of AKT in endometriotic epithelial cells (12Z) or noninvasive epithelial cells (EEC) treated with CXCL12. Cell were cultured and treated with either 10 or 100 ng/ml of CXCL12 for 0, 10 s, 1, 5, 10, 15, and 30 min. Basal condition consisted of cells growing in their respective complete media. (B) Graphical representation of densitometry of at least three experiments per cell line. Data were analyzed by nonparametric Student t-test using GraphPad Prizm v5. Significance was set at P < 0.05. Figure 5. View largeDownload slide AKT kinase activity assay. Endometrial epithelial (EEC) and endometriotic epithelial (12Z) cells were cultured and treated with 10 or 100 ng/ml CXCL12. Basal condition consisted of cells growing in their respective complete media. Total protein was isolated, quantified, and phospho AKT at serine 473 was immunoprecipitated and incubated with ATP and a GSK3 fusion protein. A western blot was performed to identify phospho AKT and phosphorylation state of the fusion protein. (A) Representative results of endometriotic 12Z cells and EEC. (B) Graphical representation of the quantification of phosphorylation by densitometry analysis of triplicate experiments. *P < 0.05. Figure 5. View largeDownload slide AKT kinase activity assay. Endometrial epithelial (EEC) and endometriotic epithelial (12Z) cells were cultured and treated with 10 or 100 ng/ml CXCL12. Basal condition consisted of cells growing in their respective complete media. Total protein was isolated, quantified, and phospho AKT at serine 473 was immunoprecipitated and incubated with ATP and a GSK3 fusion protein. A western blot was performed to identify phospho AKT and phosphorylation state of the fusion protein. (A) Representative results of endometriotic 12Z cells and EEC. (B) Graphical representation of the quantification of phosphorylation by densitometry analysis of triplicate experiments. *P < 0.05. Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration in vitro We then determined the effects in invasion and migration of CXCL12 and/or the CXCR4 inhibitor AMD3100 in the epithelial cell lines (12Z or EEC) using the boyder chamber assay. CXCL12 alone (10 and 100 ng/ml) caused a significant increase in migration of 12Z cells; CXCL12 alone (10 ng/ml only) increased their invasion. AMD3100 alone significantly decreased migration of 12Z cells by 0.42-fold (P ≤ 0.0001) but increased their invasion capacity by 2.4-fold (P ≤ 0.0001) (Figure 6). Treatment with CXCL12 (10 ng/ml) plus AMD3100 significantly decreased invasion (P ≤ 0.01) and migration (P ≤ 0.0001) of 12Z cells. On the other hand, the migration capacity of EEC cells was significantly increased by CXCL12 alone (100 ng/ml) (P ≤ 0.01), but decreased by AMD3100 alone (P ≤ 0.001), and by CXCL12 plus AMD3100 (P ≤ 0.001). Invasiveness of EEC was decreased by AMD3100 alone or in combination with CXCL12, but this effect did not reach statistical significance. Figure 6. View largeDownload slide Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration. 12Z cells or EEC cells were cultured at the upper chamber at a 25 000 cell density; the bottom chamber media contained 10 or 100 ng/ml CXCL12, the CXCR4 inhibitor AMD3100 (100nM), or a combination of CXCL12 (100 ng/mL) and AMD3100 (100 nM). Cells were cultured for 24 h, followed by fixation, staining, and visualization of the cells under microscope to count the cells. (A and C) Number of cells counted in the control insert (migration). (B and D) Number of cells that invaded through the Matrigel matrix (invasion). Data were analyzed by comparing each treatment to media only using nonparametric Student t-test in GraphPad Prizm v5. Significance was set at P < 0.05 and represented by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. Figure 6. View largeDownload slide Effects of CXCL12 and/or the CXCR4 inhibitor AMD3100 in invasion and migration. 12Z cells or EEC cells were cultured at the upper chamber at a 25 000 cell density; the bottom chamber media contained 10 or 100 ng/ml CXCL12, the CXCR4 inhibitor AMD3100 (100nM), or a combination of CXCL12 (100 ng/mL) and AMD3100 (100 nM). Cells were cultured for 24 h, followed by fixation, staining, and visualization of the cells under microscope to count the cells. (A and C) Number of cells counted in the control insert (migration). (B and D) Number of cells that invaded through the Matrigel matrix (invasion). Data were analyzed by comparing each treatment to media only using nonparametric Student t-test in GraphPad Prizm v5. Significance was set at P < 0.05 and represented by *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. Discussion There are many reports of high levels of CXCR4 and its ligand, CXCL12, in endometriosis [18, 19, 31, 36–38]. Thus, it has been hypothesized that the CXCR4-CXCL12 axis might be an effective nonhormonal target for endometriosis, since blocking its action could interfere with activation of key disease mechanisms. To address this hypothesis, we assessed the protein expression profile of CXCR4 in glands and stroma of different types of endometriotic lesions, and studied cellular responses (proliferative, migratory, invasive) mediated by activation and blocking of the CXCR4-CXCL12 axis in vitro. Our study showed that pharmacologically blocking the CXCR4 receptor has contrasting effects depending on the presence of the ligand, and possibly on the concomitant expression of other receptors. These data support additional research on the mechanisms involved when this axis is targeted in the context of endometriosis, in order to support (or not) further evaluation of CXCR4-blocking drugs in preclinical studies and clinical trials. Immunohistochemical analysis of lesions on an endometriosis-focused tissue array showed that nuclear CXCR4 (nCXCR4) protein expression levels were generally higher in ectopic lesions compared to proliferative endometrium from controls, as shown by others [37, 38]. Our study of CXCR4 expression using a tissue array allowed us to document differences in the expression levels in different lesion types/localizations, and by cell type (glands vs. stroma). In general, expression of nCXCR4 protein was higher in the stromal compartment, and significantly higher in ovarian compared to fallopian tube lesions, which may be explained by different pathophysiological mechanisms underlying the various lesion types observed in patients. Notably, the expression of nCXCR4 in proliferative endometrium of controls was lower compared to secretory endometrium, while eutopic endometrium from patients had a level of nCXCR4 expression similar to that of ectopic lesions. Cytoplasmic expression of CXCR4, on the other hand, was higher in the glandular compartment, in accord with our own findings and those of others [39], but we did not observe significant differences among tissue types. Nuclear localization of CXCR4 indicates activation of the receptor possibly leading to the endometriotic phenotypes (i.e., proliferation, migration, invasion) that we observed here to be activated in vitro. It is important to highlight that activation of the CXCR4 receptor varied depending on lesion type/localization, being higher in ovarian lesions, which may impact response to therapy with CXCR4 blockers. For example, this treatment may be more effective in women with ovarian endometriosis in particular. CXCR4 is also expressed in eutopic endometrium from patients. This is important to document since the efficacy of pharmacological interventions for endometriosis via blocking this axis will likely depend on the levels of and the tissue expression profile of CXCR4. Also of interest is that CXCR4 is highly expressed in many cancers including ovarian, and that ovarian lesions have been associated with malignant transformation to ovarian cancers (endometrioid and clear cell types) [40, 41]. Whether CXCR4 could constitute a biomarker of transformation potential, alone or in combination with other markers (e.g., Ki67), needs to be studied further. We also show here that the endometriotic epithelial cell line 12Z expresses higher basal levels of CXCR4 compared to the control cell lines, HESC and EEC. 12Z cells also had a higher proliferation when compared to the control cells, thus confirming previous observations [42]. CXCL12 did not increase the proliferation rate of 12Z cells despite them expressing the CXCR4 receptor. It is possible that 12Z cells already have a high basal proliferation rate that cannot be induced further by CXCR4-CXCL12 axis activation. In contrast, the epithelial cell line EEC did respond to CXCL12 by increasing its proliferation rate, showing that the ligand-receptor interaction used in these experiments is functional, and suggesting that this effect may be mediated by the CXCR4 receptor. The endometrial stromal cells HESC expressed CXCR4, as shown here, but no increase in proliferation was observed in response to treatment with CXCL12. It is possible that the receptor is nonfunctional because the interaction, function, or activities of G proteins are known to be cell context-dependent (i.e., epithelial vs stromal) [43]. In addition, other cellular responses could have been activated by the ligand through this receptor in the stromal cells, such as autophagy, that we did not assess in the present study [44]. The proliferation of the epithelial control cells EEC, but not of 12Z cells, was induced by stromal cell-conditioned media (HESC and PED) cells compared to control media. The proliferative effect of the PED-conditioned media on EEC proliferation rates was not reduced by blocking CXCR4 or CXCL12 with neutralizing antibodies, suggesting that other secreted factors might be mediating this effect. In the context of endometriosis, it can be argued that the shed endometrium tissue reaching the peritoneal cavity could initiate an inflammatory reaction, and factors secreted by inflammatory cells (and also by the stromal component of the refluxed endometrium) could induce proliferation of the EEC. In addition, other relevant processes could be activated by numerous factors secreted into the peritoneal fluid, including degradation of extracellular matrix, adhesion, and angiogenesis leading to the development of ectopic lesions. It would be important that follow-up studies are conducted to identify the stromal-derived factor(s) that may be mediating the observed growth-promoting effects on epithelial cells. It is well documented that the CXCR4-CXCL12 axis activates the AKT pathway; therefore, in order to assess its functionality in the cell lines studied, we conducted AKT phosphorylation and kinase activity assays. We observed that in the control epithelial cells (EEC), a transient increase in AKT phosphorylation was observed at the lowest dose of CXCL12. In contrast, there was an increase in AKT phosphorylation associated with the higher CXCL12, which was sustained for up to 30 min. Accordingly, we also observed sustained AKT kinase activity only in the endometriotic 12Z cells. Taken together, these results suggest that in endometriotic cells the AKT signaling pathway is more CXCL12-responsive (i.e., is differentially modulated by both incubation time and ligand concentration) than in the control cells, although in both cell lines this signaling pathway is functional (as demonstrated by GSK substrate phosphorylation). The observed higher levels of AKT activity induced by 10 ng/ml of CXCL12 in 12Z cells may explain in part the increased invasiveness induced by CXCL12, an effect that was not seen in the control cells at any of the dose studied. It is important to take into account that there are other pathways that can be activated by the CXCR4-CXCL12 axis, including phosphoinositol-3-kinase, mitogen-activated protein kinase, Src kinase, and focal adhesion (PyK2, FAK) pathways, which were not studied here [45]. Therefore, more studies are necessary to dissect the signal transduction mechanisms activated in response to CXCL12 in order to understand the contribution of chemokine to cellular functions relevant to endometriosis. This study also demonstrated that the endometriotic epithelial cells 12Z have a higher basal migratory potential compared to the control epithelial cell line EEC. Treatment of 12Z cells with the CXCR4 antagonist, AMD3100, significantly decreased migration but, intriguingly, increased their invasion capacity. Combining the drug with CXCL12 counteracted this effect (Figure 7). Recently, an additional receptor, CXCR7, has been reported to bind to CXCL12 [46]. While limited, there is some evidence that CXCR7 is expressed by endometriotic lesions, endometrial cells, and endometrial cancer [47, 48]. AMD3100 is an allosteric agonist for CXCR7 [49], known to bind and activate CXCR7, but the molecular effects resulting from this interaction are not well understood. We hypothesize that the CXCR4:CXCL12 interaction is primarily responsible for inducing migration of 12Z cells, while the CXCR7:CXCL12 interaction leads to invasion, a possibility that needs to be experimentally proven. AMD3100 by itself significantly decreased EEC migration, but none of the treatments (ligand, inhibitor, or their combination) induced their invasion. Taken together, these data suggest that the endometriotic epithelial cells migrate more in response to low concentrations of the ligand; these cells do not produce CXCL12 and can be more readily sensitive to exogenous sources of CXCL12 in the peritoneal environment. While their migration can be reduced by the CXCR4 inhibitor, treatment with this drug results in pleiotropic responses leading to increases in invasion, limiting its possible use as a candidate nonhormonal treatment for endometriosis unless it is given in combination with CXCL12. Figure 7. View largeDownload slide Summary of studied CXCR4-CXCL12 axis in endometriosis. CXCR4-CXCL12 axis activation produces a rapid and transient effects of AKT phosphorylation. This phosphorylation increased AKT activity and may induce cell proliferation migration and invasion. Treatment of 12Z with AMD3100 decreased migration and increased invasion suggesting the activation of different biological mechanism. Figure 7. View largeDownload slide Summary of studied CXCR4-CXCL12 axis in endometriosis. CXCR4-CXCL12 axis activation produces a rapid and transient effects of AKT phosphorylation. This phosphorylation increased AKT activity and may induce cell proliferation migration and invasion. Treatment of 12Z with AMD3100 decreased migration and increased invasion suggesting the activation of different biological mechanism. In conclusion, our results provide evidence that CXCR4 is differentially expressed across lesion types, being highest in ovarian endometriosis. In addition, endometriotic epithelial cells (12Z) express high levels of CXCR4, are highly migratory, and have a functional CXCR4-CXCL12 axis. Finally, our data suggest that not all cancer-targeted drugs may also be effective in endometriosis despite both diseases sharing the same cellular behaviors. Therefore, before targeting this pathway therapeutically, additional experiments including preclinical investigations in animal models are needed to dissect the molecular pathways activated by CXCL12 alone or in combination with receptor blocking drugs. Supplementary data Supplementary data are available at BIOLRE online. Supplementary Table S1. Antibodies Table. Acknowledgments We acknowledge the role of Dr Miosotis García in the selection and pathological analysis of samples used in the development of the endometriosis tissue array, pathology consultation by Dr Adalberto Mendoza of Southern Pathology Services, Inc., and technical help provided by Perla Báez, PhD. We thank Dr. Virgilio A. Salvo for discussions in experimental design and data interpretation as well as providing key reactives. Conflict of Interest: The authors have declared that no conflict of interest exists. † Grant Support: This study was supported by the National Institutes of Health–National Institute of Child Health and Human Development grants # R01-HD050559 (IF), F31HD072594 (AR), 3R01-HD050559-01A1S1 (LR), and R01-HD082373 (ATF); by the National Institutes of Health–National Institute of General Medical Sciences grants # S06-GM08239-20 (IF), R25-GM082406 (PHSU RISE Program) (AR and MCC), and R25-GM096955 (MCC and BJTC); by National Institutes of Health–National Institute on Minority Health and Health Disparities grants # G12-MD007579 (Molecular Biology and Genomics Core); and by National Institutes of Health–National Institute of Cancer grant 3 U56-CA126379 (construction of the tissue array). Edited by Dr. Romana Nowak, PhD, University of Illinois Urbana-Champaign. References 1. Vercellini P et al.   Endometriosis: pathogenesis and treatment. Nat Rev Endocrinol  2014; 10: 261– 275. Google Scholar CrossRef Search ADS PubMed  2. Macer ML , Taylor HS. Endometriosis and Infertility. Obstet Gynecol Clin North Am  2012; 39( 4): 535– 549. Google Scholar CrossRef Search ADS PubMed  3. Park DW , Yang KM. Hormonal regulation of uterine chemokines and immune cells. Clin Exp Reprod Med  2011; 38( 4): 179– 185. Google Scholar CrossRef Search ADS PubMed  4. Piao HL et al.   CXCL12/CXCR4 signal involved in the regulation of trophoblasts on peripheral NK cells leading to Th2 bias at the maternal-fetal interface. Eur Rev Med Pharmacol Sci  2015; 19( 12): 2153– 2161. Google Scholar PubMed  5. Noda M et al.   CXCL12-CXCR4 chemokine signaling is essential for NK-cell development in adult mice. Blood  2011; 117( 2): 451– 458. Google Scholar CrossRef Search ADS PubMed  6. Domanska UM et al.   A review on CXCR4/CXCL12 axis in oncology: no place to hide. Eur J Cancer  2013; 49( 1): 219– 230. Google Scholar CrossRef Search ADS PubMed  7. Li SF et al.   The number of proliferating cell nuclear antigen positive cells in endometriotic lesions differs from that in the endometrium. Vichows Archiv A Pathol Anat  1993; 423( 4): 257– 263. Google Scholar CrossRef Search ADS   8. Nisolle M , Casanas-Roux F, Donnez J. Immunohistochemical analysis of proliferative activity and steroid receptor expression in peritoneal and ovarian endometriosis. Fertil Steril  1997; 68( 5): 912– 919. Google Scholar CrossRef Search ADS PubMed  9. Meresman GF et al.   Oral contraceptives suppress cell proliferation and enhance apoptosis of eutopic endometrial tissue from patients with endometriosis. Fertil Steril  2002; 77( 6): 1141– 1147. Google Scholar CrossRef Search ADS PubMed  10. Matsuzaki S , Darcha C. In vitro effects of a small-molecule antagonist of the tcf/ss-catenin complex on endometrial and endometriotic cells of patients with endometriosis. PLoS One  2013; 8( 4): e61690. Google Scholar CrossRef Search ADS PubMed  11. Valles CS , Dominguez F. Embryo-endometrial interaction. Chang Gung Med J  2006; 29( 1): 9– 14. Google Scholar PubMed  12. Cocchi F et al.   Identification of RANTES, MIP-1alpha, and MIP-1beta as the Major HIV-Suppressive Factors Produced by CD8+ T Cells. Science  1995; 270( 5243): 1811– 1815. Google Scholar CrossRef Search ADS PubMed  13. Simon C et al.   Potential implications of chemokines in reproductive function: an attractive idea. J Reprod Immunol  1998; 38( 2): 169– 193. Google Scholar CrossRef Search ADS PubMed  14. Herington JL et al.   Immune interactions in endometriosis. Expert Rev Clin Immunol 2011; 7( 5): 611– 626. 15. Bellelis P et al.   Transcriptional changes in the expression of chemokines related to natural killer and T-regulatory cells in patients with deep infiltrative endometriosis. Fertil Steril  2013; 99( 7): 1987– 1993. Google Scholar CrossRef Search ADS PubMed  16. Kitaya K et al.   Spatial and temporal expression of ligands for CXCR3 and CXCR4 in human endometrium. J Clin Endocrinol Metab  2004; 89( 5): 2470– 2476. Google Scholar CrossRef Search ADS PubMed  17. Leconte M et al.   Role of the CXCL12-CXCR4 axis in the development of deep rectal endometriosis. J Reprod Immunol  2014; 103: 45– 52. Google Scholar CrossRef Search ADS PubMed  18. Furuya M et al.   Up-regulation of CXC chemokines and their receptors: implications for proinflammatory microenvironments of ovarian carcinomas and endometriosis. Hum Pathol  2007; 38( 11): 1676– 1687. Google Scholar CrossRef Search ADS PubMed  19. Chand AL et al.   Laser capture microdissection and cDNA array analysis of endometrium identify CCL16 and CCL21 as epithelial-derived inflammatory mediators associated with endometriosis. Reprod Biol Endocrinol  2007; 5( 1): 18. Google Scholar CrossRef Search ADS PubMed  20. Flores I et al.   Molecular profiling of experimental endometriosis identified gene expression patterns in common with human disease. Fertil Steril  2007; 87( 5): 1180– 1199. Google Scholar CrossRef Search ADS PubMed  21. Colon-Diaz M et al.   HDAC1 and HDAC2 are differentially expressed in endometriosis. Reprod Sci  2012; 19( 5): 483– 492. Google Scholar CrossRef Search ADS PubMed  22. Ruiz LA et al.   Dysregulation of lysyl oxidase expression in lesions and endometrium of women with endometriosis. Reprod Sci  2015; 22( 12): 1496– 1508. Google Scholar CrossRef Search ADS PubMed  23. Colon-Caraballo M , Monteiro JB, Flores I. H3K27me3 is an epigenetic mark of relevance in endometriosis. Reprod Sci  2015; 22( 9): 1134– 1142. Google Scholar CrossRef Search ADS PubMed  24. Noyes RW , Hertig AT, Rock J. Dating the endometrial biopsy. Am J Obstet Gynecol  1975; 122( 2): 262– 263. Google Scholar CrossRef Search ADS PubMed  25. Ruiz A et al.   Basal and steroid hormone-regulated expression of CXCR4 in human endometrium and endometriosis. Reprod Sci  2010; 17( 10): 894– 903. Google Scholar CrossRef Search ADS PubMed  26. Korch C et al.   DNA profiling analysis of endometrial and ovarian cell lines reveals misidentification, redundancy and contamination. Gynecol Oncol  2012; 127( 1): 241– 248. Google Scholar CrossRef Search ADS PubMed  27. Zeitvogel A , Baumann R, Starzinski-Powitz A. Identification of an invasive, N-cadherin-expressing epithelial cell type in endometriosis using a new cell culture model. Am J Pathol  2001; 159( 5): 1839– 1852. Google Scholar CrossRef Search ADS PubMed  28. Krikun G et al.   A novel immortalized human endometrial stromal cell line with normal progestational response. Endocrinology  2004; 145( 5): 2291– 2296. Google Scholar CrossRef Search ADS PubMed  29. Hombach-Klonisch S et al.   Regulation of functional steroid receptors and ligand-induced responses in telomerase-immortalized human endometrial epithelial cells. J Mol Endocrinol  2005; 34( 2): 517– 534. Google Scholar CrossRef Search ADS PubMed  30. Tsutsumi A et al.   Estrogen induces stromal cell-derived factor 1 (SDF-1/CXCL12) production in human endometrial stromal cells: a possible role of endometrial epithelial cell growth. Fertil Steril  2011; 95: 444– 447. Google Scholar CrossRef Search ADS PubMed  31. Ruiz A et al.   Basal and steroid hormone-regulated expression of CXCR4 in human endometrium and endometriosis. Reprod Sci  2010; 17( 10): 894– 903. Google Scholar CrossRef Search ADS PubMed  32. Laird SM et al.   Expression of CXCL12 and CXCR4 in human endometrium; effects of CXCL12 on MMP production by human endometrial cells. Hum Reprod  2011; 26( 5): 1144– 1152. Google Scholar CrossRef Search ADS PubMed  33. Oberlin E et al.   The CXC chemokine SDF-1 is the ligand for LESTR/fusin and prevents infection by T-cell-line-adapted HIV-1. Nature  1996; 382( 6594): 833– 835. Google Scholar CrossRef Search ADS PubMed  34. Yu J et al.   Combination of estrogen and dioxin is involved in the pathogenesis of endometriosis by promoting chemokine secretion and invasion of endometrial stromal cells. Hum Reprod  2008; 23( 7): 1614– 1626. Google Scholar CrossRef Search ADS PubMed  35. Kitaya K et al.   Spatial and temporal expression of ligands for CXCR3 and CXCR4 in human endometrium. J Clin Endocrinol Metab  2004; 89( 5): 2470– 2476. Google Scholar CrossRef Search ADS PubMed  36. Borrelli GM et al.   Immunohistochemical investigation of metastasis-related chemokines in deep-infiltrating endometriosis and compromised pelvic sentinel lymph nodes. Reprod Sci  2015; 22( 12): 1632– 1642. Google Scholar CrossRef Search ADS PubMed  37. Leconte M et al.   Role of the CXCL12-CXCR4 axis in the development of deep rectal endometriosis. J Reprod Immunol  2014; 103: 45– 52. Google Scholar CrossRef Search ADS PubMed  38. Moridi I et al.   Bone marrow stem cell chemotactic activity is induced by elevated CXCL12 in endometriosis. Reprod Sci  2017; 24: 526– 533. Google Scholar CrossRef Search ADS PubMed  39. Laird SM et al.   Expression of CXCL12 and CXCR4 in human endometrium; effects of CXCL12 on MMP production by human endometrial cells. Hum Reprod  2011; 26( 5): 1144– 1152. Google Scholar CrossRef Search ADS PubMed  40. Pearce CL et al.   Association between endometriosis and risk of histological subtypes of ovarian cancer: a pooled analysis of case-control studies. Lancet Oncol  2012; 13( 4): 385– 394. Google Scholar CrossRef Search ADS PubMed  41. Lee AW et al.   Evidence of a genetic link between endometriosis and ovarian cancer. Fertil Steril  2016; 105( 1): 35– 43. e1–10. Google Scholar CrossRef Search ADS PubMed  42. Kahyaoglu I et al.   Comparison of Ki-67 proliferative index between eutopic and ectopic endometrium: a case control study. Taiwan J Obstet Gynecol  2012; 51( 3): 393– 396. Google Scholar CrossRef Search ADS PubMed  43. Busillo JM , Benovic JL. Regulation of CXCR4 signaling. Biochim Biophys Acta  2007; 1768( 4): 952– 963. Google Scholar CrossRef Search ADS PubMed  44. Mei J et al.   Estrogen promotes the survival of human secretory phase endometrial stromal cells via CXCL12/CXCR4 up-regulation-mediated autophagy inhibition. Hum Reprod  2015; 30( 7): 1677– 1689. Google Scholar CrossRef Search ADS PubMed  45. Arcaro A , Guerreiro AS. The phosphoinositide 3-kinase pathway in human cancer: genetic alterations and therapeutic implications. Curr Genomics  2007; 8( 5): 271– 306. Google Scholar CrossRef Search ADS PubMed  46. Balabanian K et al.   The chemokine SDF-1/CXCL12 binds to and signals through the orphan receptor RDC1 in T lymphocytes. J Biol Chem  2005; 280( 42): 35760– 35766. Google Scholar CrossRef Search ADS PubMed  47. Zhou WH et al.   Co-expression of CXCR4 and CXCR7 in human endometrial stromal cells is modulated by steroid hormones. Int J Clin Exp Pathol  2015; 8( 3): 2449– 2460. Google Scholar PubMed  48. Gu HQ et al.   The role of the SDF-1/ CXCR7 axis on the growth and invasion ability of endometrial cancer cells. Arch Gynecol Obstet  2017; 295( 4): 987– 995. Google Scholar CrossRef Search ADS PubMed  49. Kalatskaya I et al.   AMD3100 is a CXCR7 ligand with allosteric agonist properties. Mol Pharmacol  2009; 75( 5): 1240– 1247. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Biology of ReproductionOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial