Niclosamide suppresses macrophage-induced inflammation in endometriosis

Abstract

Endometriosis is a common gynecological disease, which causes chronic pelvic pain and infertility in women of reproductive age. Due to limited efficacy of current treatment options, a critical need exists to develop new and effective treatments for endometriosis. Niclosamide is an efficacious and FDA-approved drug for the treatment of helminthosis in humans that has been used for decades. We have reported that niclosamide reduces growth and progression of endometriosis-like lesions via targeting STAT3 and NFĸB signaling in a mouse model of endometriosis. To examine the effects of niclosamide on macrophage-induced inflammation in endometriosis, a total of 29 stage III–IV endometrioma samples were used to isolate human endometriotic stromal cells (hESCs). M1 or M2 macrophages were isolated and differentiated from fresh human peripheral blood samples. Then, hESCs were cultured in conditioned media (CM) from macrophages with/without niclosamide. Niclosamide dose dependently reduced cell viability and the activity of STAT3 and NFκB signaling in hESCs. While macrophage CM stimulated cell viability in hESCs, niclosamide inhibited this stimulation. Macrophage CM stimulated the secretion of proinflammatory cytokines and chemokines from hESCs. Most of these secreted factors were inhibited by niclosamide. These results indicate that niclosamide is able to reduce macrophage-induced cell viability and cytokine/chemokine secretion in hESCs by inhibiting inflammatory mechanisms via STAT3 and/or NFκB signaling.

Niclosamide is able to inhibit the inflammatory mechanisms in primary endometriotic stromal cells stimulated by macrophages via STAT3 and NFκB signaling.

Introduction

Endometriosis is a chronic inflammatory disease characterized by the persistence and growth of endometrial tissue outside the uterine cavity, primarily on the pelvic peritoneum and ovaries [1, 2]. Although endometriosis is a non-malignant disorder, the majority of affected women experience severe chronic pelvic pain and/or infertility [3, 4]. Because endometriosis is an estrogen-dependent disorder, the most widely used medical treatments are GnRH agonists/antagonists, which suppress ovarian function, as well as oral contraceptives, which exert a predominant “progestational” effect on the endometrium, subsequently reducing endometriosis-associated symptoms [2, 5–8]. However, these approaches alongside surgical removal of lesions only temporarily relieve the symptoms, often cause numerous unwanted side effects, and have a high incidence of relapse [8–11]. Thus, it is necessary to identify new therapeutic targets and efficient drug(s) that improve current treatment options.

Endometriosis is known as an estrogen-dependent inflammatory disease associated with dysregulation of tissue-, peritoneal-, and peripheral cytokines [1]. Increased macrophages, prostaglandins, cytokines and chemokines have been observed in the peritoneal fluids from endometriosis patients, resulting in the establishment of an inflammatory microenvironment that encourages endometrial cell attachment, invasion, and vasculogenesis [12–14]. Chemokines play a major role in the recruitment of macrophages at the site of endometrial tissue engraftment in the peritoneal cavity, a critical step for neuroangiogenesis in the endometriotic lesions [15, 16]. Thus, the altered inflammatory environment further enhances inflammation and consequently promotes endometriotic cell survival and growth.

Recently, we have identified a small molecule, niclosamide, which could be a potential new effective therapy for endometriosis [17]. Niclosamide has been approved by the Food and Drug Administration (FDA) since 1982 for the treatment of helminthosis as an efficacious and affordable drug [18, 19] and is listed among the World Health Organization’s essential medicines [20]. In the past several years, mounting evidence has accumulated from our group and others that niclosamide is a multifunctional drug that is able to target multiple signaling pathways [21–34], indicating that niclosamide can be developed for novel treatments beyond helminthosis [35]. Our results have demonstrated that niclosamide reduces growth and progression of endometriosis-like lesions (ELLs), and inhibits inflammatory signaling such as STAT3 and NFκB in ELLs using a mouse model of endometriosis [17]. RNA-sequencing analysis indicated that transcripts associated with inflammatory responses were reduced by niclosamide. Thus, we hypothesized that niclosamide could be an inhibitor of endometriosis progression by blocking inflammatory pathways. In our recent study, we have reported that niclosamide inhibits macrophage-dependent cell viability, STAT3 and NFκB activity, and secretion of cytokines and chemokines in an immortalized endometriotic epithelial cell line, 12Z [36]. In the present study, we report that niclosamide is also an effective inhibitor to inflammatory signaling mechanisms stimulated by macrophages using primary human endometriotic stromal cells (hESCs).

Materials and methods

Human tissue collection

Collection of endometriosis tissue samples was approved by the Institutional Review Board at the University of Tokyo and Southern Illinois University School of Medicine (SIU-SOM). Written informed consent was obtained from all participating subjects. A total of 29 stage III-IV endometrioma samples, based on the revised American Society for Reproductive Medicine Classification, were obtained from subjects who underwent laparoscopic surgery at the University of Tokyo (18 samples) or SIU-SOM (11 samples). Patient demographics are shown in Table 1.

Table 1.

Patient demographics

						rASRM
Patient	Age	Gravida	Parity	Infertility*	Stage 3	Stage 4
1	46	0	0	No		+
2	31	0	0	Yes		+
3	30	0	0	Yes	+
4	33	1	1	No	+
5	45	0	0	Yes		+
6	29	0	0	No		+
7	36	0	0	No		+
8	44	0	0	No	+
9	25	0	0	No		+
10	43	0	0	No		+
11	33	0	0	Yes	+
12	43	2	2	No		+
13	32	0	0	Yes	+
14	33	1	0	Yes	+
15	43	0	0	No		+
16	29	0	0	No		+
17	45	1	1	Yes		+
18	40	1	1	No	+
27	32	0	0	Yes		+
32	29	1	0	Yes		+
36	34	0	0	Yes		+
37	21	0	0	No		+
39	23	1	0	Yes		+
40	40	6	3	No		+
42	23	1	0	No	+
44	27	0	0	Yes		+
51	38	0	0	Yes		+
54	30	0	0	Yes		+
61	29	0	0	Yes		+

^*Infertility was defined as the inability to conceive after 12 months of unprotected intercourse.

Isolation and culture of hESCs

Primary hESCs were isolated following established protocols with minor modification [37]. Briefly, tissue was minced into small pieces, incubated with 0.5% collagenase type I (Gibco, USA), 0.02% DNase I (Sigma-Aldrich, USA), 0.1% hyaluronidase (Sigma-Aldrich, USA), and/or 0.1% pronase (Millipore, USA) for 20 min × 3 times in a shaking water bath at 37 °C, and filtered through nylon cell strainers with apertures of 100, 70, and then 40 μm (Falcon, USA). The isolated hESCs were cultured in RPMI1640 with 10% charcoal-dextran treated FBS and antibiotic-antimycotic (Gibco, USA) at 37 °C in a humidified 5% CO₂ incubator for a few days. Once the cells reached approximately 80% confluency, they were trypsinized and used for the experiments. We did not use any cells from multiple passages or frozen cells for this study.

Cell viability

Cell viability was performed following our previously described methods [28, 36]. Cells (8 × 10⁴/well) were seeded in 24-well plates, and then dose dependently treated with niclosamide on the next day (18–20 h later). Cell viability was assessed after 48 h of treatment using a Countess II FL Automated Cell Counter (Thermo Fisher, USA) with trypan blue exclusion to determine viable cell percentage.

Macrophage differentiation and study design

Fresh human peripheral blood samples were purchased from Research Blood Components (Boston, MA, USA). Peripheral blood mononuclear cells were isolated using SepMate-50 and Lymphoprep (Stem Cell Technologies, USA) following the manufacturer’s instruction. Using monocyte attachment media (Promo Cell, Germany), monocytes attached to the cell culture plates were used for the study and differentiated into M1 or M2 macrophages with M1- or M2-macrophage generation media DXF (Promo Cell, Germany) supplemented with 10 ng/ml of lipopolysaccharide (LPS, Sigma-Aldrich, USA) and 50 ng/ml of IFN-γ (Millipore, USA) for M1 macrophages, or 20 ng/ml of IL4 (Millipore, USA) and IL10 (Millipore, USA) for M2 macrophages following the manufacturer’s instruction. Macrophage activation was confirmed by analyzing relative mRNA expression levels of PTGS2 and TNF for M1, and IL10 for M2 described previously [38, 39] and are shown in Figure 1A.

Figure 3.

Effects of factors derived from macrophages and/or niclosamide on cell viability in hESCs (n = 11). Cell viability was characterized after treatment with either GM, M1 or M2 CM with/without niclosamide (1 μM). Cell numbers are expressed as the mean ± SEM percentage of control (GM-treated without niclosamide which was set to 100). Significant reductions in cell number following treatment with niclosamide in each group are indicated, ^*P < 0.05, ^***P < 0.001.

Figure 1.

(A) Confirmation of macrophage activation. Relative mRNA expression levels of PTGS2, TNF, and IL10 were analyzed in monocytes (Mo), M1 and M2 macrophages (n = 6–9). ^** or ^*** vs Mo; P < 0.01 or 0.001, respectively. (B) Experimental design. Monocytes were differentiated into M1 or M2 macrophages with M1- or M2-macrophage generation media DXF (Promo Cell, Germany) supplemented with LPS (10 ng/ml) and IFN-γ (50 ng/ml) for M1 macrophages, or IL4 (20 ng/ml) and IL10 (20 ng/ml) for M2 macrophages. The CM were then used for the treatment of hESCs, with or without niclosamide. Legend: CM, conditioned media; hESCs, primary human endometriotic stromal cells; and Mo, monocytes.

Figure 2.

Effects of niclosamide on cell viability, and STAT3 and NFκB signaling in hESCs. Niclosamide was dose dependently treated to examine (A) cell viability (n = 24), (B) activation of STAT3 (n = 7), and (C) p65 (n = 4). Representative western blots and the results of densitometry from western blots are shown. ^*, ^**, or ^*** vs vehicle control; P < 0.05, 0.01, or 0.001, respectively.

Conditioned media (CM) from M1 or M2 macrophages were collected 48 h after the addition of fresh media. The hESCs were serum starved for 48 h and then cultured with control (growth media, GM) or 100% CM from M1 or M2 with or without niclosamide (1 μM, Sigma-Aldrich) for 48 h, washed/added fresh media for 24 h, and collected cells for cell viability and CM from hESCs for further analysis (Figure 1B).

CM from hESCs were analyzed using the Proteome Profiler Human XL Cytokine (ARY022B) and Chemokine (ARY017) Array (R&D Systems, USA) according to the manufacturer’s instructions. Quantitative analysis of density was performed using Image J [imagej.nih.gov, [40]]. ANG, CCL22, CXCL12, MDK, RARRES2, and VCAM1 concentrations in the CM were analyzed by human ANG (IQH-ANG-1), MDC (CCL22, IQH-MDC-1), SDF-1 (CXCL12, IQH-SDF1a-1), MDK (IQH-MDK-1), Chemerin (RARRES2, IQH-Chemerin-1), and VCAM-1 (IQH-VCAM1–1) IQELISA (Ray Biotech, USA), respectively, according to the manufacturer’s instructions.

Western blot

Ten micrograms of total protein from whole cell lysates were separated on NuPage Bis-Tris gels (Invitrogen, USA) and transferred to nitrocellulose membranes (Millipore, USA). Membranes were blocked and incubated overnight with primary antibodies: anti-phospho-STAT3 (1:1000 dilution, 9145, Cell Signaling, USA), anti-STAT3 (1:2000 dilution, 4904, Cell Signaling, USA), anti-phospho-p65 (1:500 dilution, 3033, Cell Signaling, USA), anti-p65 (1:1000 dilution, 8242, Cell Signaling, USA), or anti-actin (1:5000 dilution, 3700, Cell Signaling, USA). Immunoreactivity was visualized with IRDye680 (1:10000 dilution, 926–32222, Li-COR, USA) or IRDye800 (1:10000 dilution, 926–32211, Li-COR, USA) conjugated affinity-purified secondary antibodies using the Odyssey infrared imaging system (Li-COR, USA).

Quantitative real-time PCR analysis

Total RNA was isolated from cells, and cDNA was synthesized from total RNA using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher, USA). Relative gene expression was determined by SYBR green (Bio Rad, USA) incorporation using a Bio-Rad CFX as described previously [41]. Primer sequences were determined using NCBI’s design tools and are provided in Supplementary Table 1.

Statistical analysis

Data were subjected to one-way ANOVA, and Dunnett or Tukey multiple-comparison post-test was used to identify differences between individual means using Prism software (Ver. 5.0, GraphPad, USA). All data met necessary criteria for ANOVA analysis including equal variance as determined by Bartlett’s test. All experimental data are presented as mean with standard error of the mean (SEM). Unless otherwise indicated, a P value less than 0.05 was considered to be statistically significant.

Results

Niclosamide inhibits hESCs functions

Our previous study has demonstrated that niclosamide reduces the size of ELLs with inhibition of cell proliferation and inflammatory signaling using an in vivo mouse model of endometriosis [17]. Niclosamide also inhibits proliferation and function of 12Z endometriotic epithelial cells [36]. In the present study, to determine whether niclosamide has inhibitory effects on primary isolated cells from the endometriotic lesions in patients, hESCs were used to assess cell viability (n = 24) and the activity of STAT3 (n = 7) and NFκB (n = 4) signaling (Figure 2). We found that niclosamide dose dependently decreased cell viability (Figure 2A). Phosphorylation of STAT3 and p65, one of the common subunits of the NFκB transcription factor complex, was also reduced by niclosamide treatment in hESCs (Figure 2B and C).

Figure 4.

Screening of secreted cytokines and chemokines in hESCs. CM collected from hESCs primed with GM, M1 or M2 CM with/without niclosamide (1 μM) were analyzed for proteasome profiles of cytokine and chemokine array or IQELISA (n = 5). (A) Relative fold change of a total of 13 cytokines and chemokines selected from protein array are shown (GM-treated without niclosamide which was set to 1). (B) Verified CCL22 and RARRES2 concentrations in the CM by IQELISA are shown. Significant changes following treatment with niclosamide in each group are indicated, ^*P < 0.05, ^**P < 0.01.

Macrophages promote hESCs viability and inflammation that are inhibited by niclosamide

Endometriosis is associated with dysregulation of inflammatory factors. Increased macrophages, cytokines, and chemokines have been observed in the peritoneal fluid from endometriosis patients [12–14]. Thus, we hypothesize that the macrophage’s effect on hESCs contributes to the establishment of inflammatory microenvironment and promotes endometriotic progression in the peritoneum. We therefore examined whether the macrophages were able to stimulate hESC viability and inflammation. We observed that cell viability (n = 11) was stimulated in hESCs primed in CM by M2 macrophages (Figure 3). Niclosamide not only inhibited cell viability in hESCs (GM and M1), but also blocked M2 macrophage-induced hESC viability (Figure 3). Moreover, cell viability induced by M2 macrophages returned to that of GM from cells treated with niclosamide alone.

Next, we performed protein array analysis (n = 5) to identify proteomic profiles of cytokines and chemokines in hESCs that were stimulated by macrophages and targeted by niclosamide (Figure 4 and Supplementary Figure 1. Fold changes were calculated by dividing the relative protein secretion in hESCs compared with those of GM. Representative film is shown in Supplementary Figure 1, and relative fold change of a total of 13 cytokines and chemokines that were significantly altered by macrophages and/or niclosamide from protein array are shown in Figure 4A. ANG, ANGPT2, BDNF, CCL1, CCL7, CCL17, CCL22, CXCL12, GC, MDK, PLAUR, RARRES2, and VCAM1 were stimulated by CM from M1 and/or M2 macrophages. More specifically, ANG, BDNF, CCL22, CXCL12, GC, MDK, PLAUR, RARRES2, and VCAM1 were stimulated by CM from both M1 and M2 macrophages. On the other hand, ANGPT2 was only stimulated by M1 CM. CCL1, CCL7 and CCL17 were stimulated by M2 CM. Niclosamide inhibited secretion of BDNF in hESCs stimulated by both M1 and M2 CM. The secretion of ANG, ANGPT2, CCL1, CCL7, CCL17, CCL22, CXCL12, MDK, PLAUR, RARRES2, and VCAM1 stimulated by M2 CM was suppressed by niclosamide. ANG, CCL22, CXCL12, MDK, RARRES2, and VCAM1 concentrations in the CM were further quantitated by IQELISA. While two of identified chemokines, CCL22 and RARRES2 concentrations in the CM, were verified and exhibited similar pattern to the results obtained from the protein array. The other four cytokines and chemokines were not able to be verified due to limited sensitivity of the IQELISA kits.

Discussion

We have recently demonstrated that an FDA-approved drug, niclosamide, inhibits growth and progression of ELLs using an established mouse model [17] and reduces cell viability in 12Z cells [36]. Our results further suggest that niclosamide targets inflammatory mechanisms: inhibition of STAT3 and NFκB signaling activity, as well as reduction of macrophage-induced cellular functions [17, 36]. The present study, using primary isolated hESCs from endometriosis patients, further confirmed that niclosamide is able to effectively reduce cell viability and activity of STAT3 and NFκB signaling, as well as hESC functions including cytokine and chemokine secretion stimulated by macrophages.

The contribution of macrophages has been indicated in the establishment of an inflammatory environment and the progression of endometriosis [42, 43]. The present study demonstrates that M2 macrophages stimulate cell viability in hESCs. Bacci et al. have reported that M2 macrophages enhance the lesion growth and vascularization in a mouse model of endometriosis, whereas M1 macrophages reduce the size of lesions [42]. While a mixture of different types and origins of macrophages secrete numerous pro-inflammatory factors to initiate and develop the inflammatory environment and accelerate disease progression [44], our results further support the contribution of M2 macrophages in proliferation and growth of endometriotic cells. On the other hand, M1 and/or M2 macrophages contribute to the secretion of cytokines and chemokines from hESCs. Most identified cytokines and chemokines were stimulated by the factors from both M1 and M2 macrophages. Traditionally, M1 macrophages are known to produce pro-inflammatory cytokines and initiate an immune response, while M2 macrophages are involved in promotion of tissue remodeling and tumor progression [45]. Elevated levels of ANG [46, 47], ANGPT2 [48, 49], BDNF [50–53], CCL7 [54], CCL17 [55], CCL22 [54], CXCL12 [55], MDK [56], PLAUR [57, 58], and RARRES2 [59] have been reported in the endometriotic lesions, peritoneal fluid, and/or serum of women with endometriosis. Thus, both M1 and M2 macrophages are likely to be critical to establish and exacerbate inflammation in endometriotic lesions.

ANG [60], ANGPT2 [61], and BDNF [62] are involved in stimulating angiogenesis. ANG activates capillary endothelial cells [60]. ANGPT2 is one of the key regulators of vascular quiescence facilitating angiogenesis with VEGF [63, 64]. BDNF is a family member of neurotrophins and has been recognized as a regulator in the formation and maintenance of chronic pain in various chronic disorders [65–67] including endometriosis [50, 68]. Increased production of BDNF, caused by estrogen, stimulates neurite outgrowth from dorsal root ganglia (DRG), resulting in increased neuroangiogenesis [69]. BDNF polymorphism might contribute to the increased susceptibility of advanced endometriosis and endometriosis-related infertility [70]. VCAM1 is responsible for monocytic cell adhesion [71], and it is a pro-inflammatory cytokine that is regulated by NFκB [72]. In endometriotic stromal cells, VCAM1 is mainly induced by TNFα, and its suppression leads to reduced proliferation of hESCs [73, 74]. Our results showed that secretion of ANG, ANGPT2, BDNF, and VCAM1 stimulated by macrophages factors was then inhibited by niclosamide, suggesting that niclosamide can target and reduce angiogenic and/or neuroangiogenic stimulation in hESCs.

MDK is reported to promote the formation of intraperitoneal adhesions [75]. CXCL12, aka SDF-1, has been demonstrated in the establishment, growth, and dissemination of endometriosis. CXCL12, via its ligand receptor CXCR4, promotes the invasion and engraftment of stem cells at the endometriotic lesions throughout the peritoneum [76, 77]. Its chemoattractive function is enhanced by steroid hormones [78]. By suppressing CXCL12, niclosamide may be able to disrupt endometriotic stem cell functions via inhibiting CXCL12-CXCR4 signaling, although it was not directly examined in the present study. Supporting this hypothesis, it has been shown that niclosamide can target chemoresistant stem-like cancer cells [34, 79].

CCL17 and CCL22 are known to act as strong chemotaxis on regulatory T cells, Tregs [80]. Production of CCL17 and CCL22 is stimulated in endometrial stromal cells depending on estradiol or progesterone [80]. Treg functions are enhanced by factors from endometrial stromal cells co-cultured with macrophages and contribute to endometriotic immunotolerance by stimulating cell proliferation and invasion [81, 82]. CCL17-CCR4 axis also leads to excessive IL6 production in macrophages by activating NFκB leading to increased inflammation and migration of endometriotic epithelial cells [83]. Thus, elevated CCL17 and CCL22 are able to further recruit and/or stimulate other immune cell functions to establish an inflammatory milieu. The present study demonstrates the importance of macrophages in hESC functions; furthermore, their effects may be inhibited by niclosamide. With the suppression of CCL17 and CCL22, niclosamide has potential to further reduce the activation of Tregs and macrophages.

Collectively, the results of the present study indicate that hESCs were stimulated to secrete cytokines and chemokines by factors from macrophages to enhance/further advance the inflammatory microenvironment. Niclosamide could be an effective drug for endometriosis inhibiting inflammatory mechanisms via STAT3 and/or NFκB signaling. The findings reported in the present study were produced by primary human cell-based assays that have generated new hypotheses and confirmed our pre-existing hypotheses for pathways to target for improved endometriosis treatment. On the other hand, the limitations in this study are whether niclosamide directly affects macrophage functions, and that experiments using normal endometrial stromal cells could not be performed. While side effects of niclosamide include nausea and abdominal pain [84], niclosamide is an efficacious and FDA-approved drug for the treatment of helminthosis in humans that has been used in patients for this purpose for decades [85–87]. Thus, there is potential to employ drug repurposing of niclosamide for the treatment of endometriosis. Specifically, targeting inflammatory dysfunction in endometriosis by niclosamide is a new strategy that contrasts current hormonal treatments. It would be important to know further cellular and molecular inhibitory mechanisms of niclosamide, and how niclosamide improves aberrant inflammatory condition and endometriosis-associated symptoms such as pain.

Author roles

NS and KH designed the studies; NS, AW, TT, MS, and KH performed the experiments; YH and RLM provided ovarian endometrioma samples, KG, PDS, and TW recruited patients; NS, AW, TT, JAM, and KH analyzed data; and NS and KH wrote the paper.

Conflict of interest

The authors declare that no conflict of interest exists.