A novel senotherapeutic strategy with azithromycin for preventing endometriosis progression

To determine whether cellular senescence is involved in ovarian endometriosis, we assessed the prevalence of cellular senescence characteristics in nESCs, eESCs, and CSCs. We first analyzed senescence-specific morphological features (nuclear swelling and cytoplasmic expansion). These morphological features were more pronounced in CSCs than in nESCs or eESCs (nucleus: nESCs; 36.5 ± 2.0 μm 2 , eESCs; 35.5 ± 2.6 μm 2 , CSCs; 52.9 ± 1.9 μm 2 , P  < 0.001, cytoplasm: nESCs; 3415.8 ± 33.5 μm 2 , eESCs; 2574.7 ± 183.5 μm 2 , CSCs; 12181.8 ± 2684.5 μm 2 , P  < 0.01) (Fig.  2 A-D). We next investigated SA-β-Gal expression via SA-β-Gal staining and fluorescent staining with the high-cell-permeant SPiDER-βGal. The proportion of SA-β-Gal staining positive cells in CSCs was significantly higher than that in nESCs or eESCs (nESCs: 26.3 ± 7.0%, eESCs: 31.9 ± 7.6%, CSCs: 65.6 ± 3.7%, P  < 0.001) (Fig.  2 E and F). SPiDER-βGal fluorescence, which quantifies SA-β-Gal activity, was elevated in CSCs compared to those in nESCs or eESCs (nESCs; 0.081 ± 0.011, eESCs; 0.091 ± 0.020, CSCs; 0.36 ± 0.066, P  < 0.01) (Fig.  2 G). Finally, we evaluated the other senescence markers, p16 INK4a , LaminB1, and IL-6, using immunocytochemistry. Proportions of p16 INK4a and IL-6 positive cells were higher and LaminB1 positive cells were lower in CSCs compared to those in nESCs (p16 INK4a : nESCs; 68.9 ± 14.8%, CSCs; 87.0 ± 8.2%, P  < 0.05, IL-6: nESCs; 76.5 ± 15.5%, CSCs; 97.8 ± 3.8%, P  < 0.01, LaminB1: nESCs; 71.4 ± 15.2%, CSCs; 45.2 ± 15.6%, P  < 0.05) (Fig.  2 H and I). IL-6 supernatant concentrations were significantly higher in CSCs than in nESCs (nESCs; 3.2 ± 1.8 pg/mL, CSCs; 677.8 ± 203.6 pg/mL, P  < 0.05) (Fig.  2 J). Fig. 2 CSCs display characteristics of cellular senescence. ( A ) Representative images of nuclear swelling. Scale bar: 20 μm. ( B ) Representative images of cytoplasmic expansion. Scale bar: 200 μm. ( C ) The results of nuclear area quantitative analyses are shown in a box plot ( n  = 5 in each group). ( D ) The results of cytoplasmic area quantitative analyses are shown in a box plot ( n  = 5 in each group). ( E ) Representative SA-β-Gal staining images. Scale bar: 200 μm. ( F ) The proportion of SA-β-Gal staining positive cells in each group is shown (nESCs; n  = 9, eESCs; n  = 8, CSCs; n  = 16). ( G ) Quantitation of SPiDER-β-Gal intensity is shown ( n  = 5 in each group). ( H ) Representative immunocytochemistry images of senescence markers; p16 INK4a , laminB1, and IL-6. Scale bar: 20 μm. Quantitative analysis is shown in ( I ) ( n  = 5 in each group). ( J ) Supernatant IL-6 concentration is shown in the box plot ( n  = 5 in each group). nESCs, normal endometrial stromal cells; eESCs, endometrial stromal cells with ovarian endometriosis (OE); CSCs, OE (chocolate cyst)-derived stromal cells; SA-β-Gal, senescence-associated β-galactosidase. (C, D and J) Horizontal lines centered within the boxes indicate median values. The bottom and top of the box indicate the 25th and 75th percentiles, respectively. The whisker ends indicate the respective minimum and maximum for all data. (F, G and I) Data are shown as means ± SEM. Data were analyzed by a one-way ANOVA (F and G) and the Student’s t-test (I). ∗ P  < 0.05; ∗∗ P  < 0.01; ∗∗∗ P  < 0.001; n.s., not significant CSCs display characteristics of cellular senescence. ( A ) Representative images of nuclear swelling. Scale bar: 20 μm. ( B ) Representative images of cytoplasmic expansion. Scale bar: 200 μm. ( C ) The results of nuclear area quantitative analyses are shown in a box plot ( n  = 5 in each group). ( D ) The results of cytoplasmic area quantitative analyses are shown in a box plot ( n  = 5 in each group). ( E ) Representative SA-β-Gal staining images. Scale bar: 200 μm. ( F ) The proportion of SA-β-Gal staining positive cells in each group is shown (nESCs; n  = 9, eESCs; n  = 8, CSCs; n  = 16). ( G ) Quantitation of SPiDER-β-Gal intensity is shown ( n  = 5 in each group). ( H ) Representative immunocytochemistry images of senescence markers; p16 INK4a , laminB1, and IL-6. Scale bar: 20 μm. Quantitative analysis is shown in ( I ) ( n  = 5 in each group). ( J ) Supernatant IL-6 concentration is shown in the box plot ( n  = 5 in each group). nESCs, normal endometrial stromal cells; eESCs, endometrial stromal cells with ovarian endometriosis (OE); CSCs, OE (chocolate cyst)-derived stromal cells; SA-β-Gal, senescence-associated β-galactosidase. (C, D and J) Horizontal lines centered within the boxes indicate median values. The bottom and top of the box indicate the 25th and 75th percentiles, respectively. The whisker ends indicate the respective minimum and maximum for all data. (F, G and I) Data are shown as means ± SEM. Data were analyzed by a one-way ANOVA (F and G) and the Student’s t-test (I). ∗ P  < 0.05; ∗∗ P  < 0.01; ∗∗∗ P  < 0.001; n.s., not significant To evaluate the effects of senolytic drugs, such as AZM and ABT263, we added these drugs to cultured nESCs, eESCs and CSCs. Treatment with 50 μM AZM for 72 h specifically reduced the surviving fraction in CSCs compared to 0 μM AZM treatment (nESCs: 0.89 ± 0.22, P  = 0.42, eESCs: 1.06 ± 0.17, P  = 0.99, CSCs: 0.37 ± 0.04, P  < 0.001) (Fig.  3 A). Treatment with 1 μM ABT263 for 24 h reduced the surviving fraction in CSCs compared to 0 μM ABT263 treatment (nESCs: 0.62 ± 0.12, P  = 0.11, eESCs: 0.36 ± 0.20, P  = 0.20, CSCs: 0.31 ± 0.08, P  < 0.01) (Fig.  3 B). Although the difference was insignificant, ABT263 reduced the surviving fraction of nESCs and eESCs, whereas AZM had minimal effects on nESCs and eESCs (Fig.  3 A and B). The number of SA-β-Gal positive cells decreased by approximately half after treatment with AZM or ABT263 (nESC: AZM, 0.47 ± 0.04, P  < 0.001, ABT263, 0.43 ± 0.08, P  < 0.001, CSC: AZM, 0.45 ± 0.07, P  < 0.001, ABT263, 0.48 ± 0.06, P  < 0.001) (Fig. S1 A and B). Fig. 3 Effects of AZM and ABT263 on cell viability and IL-6 secretion in CSCs. ( A ) AZM treatment reduces the viable cell fraction in CSCs ( n  = 5 in each group). ( B ) ABT263 treatment reduces the viable cell fraction in CSCs ( n  = 5 in each group). ( C ) Representative cytokine array ( n  = 4 in each group) images (upper panels), quantification of intensity of each spot (middle table). ( D ) The ratio of supernatant IL-6 concentrations before vs. after AZM treatment is shown in box plot form ( n  = 5 in each group). ( E ) The ratio of IL-6 concentrations before vs. after AZM treatment is shown in box plot form ( n  = 5 in each group). nESCs, normal endometrial stromal cells; eESCs, endometrial stromal cells with ovarian endometriosis (OE); CSCs, OE (chocolate cyst)-derived stromal cells; AZM, azithromycin; ABT263, navitoclax. (A and B) Data are shown as means ± SEM. (D and E) The horizontal line centered inside each box indicates the median. The bottom and top of the box indicate the 25th and 75th percentiles, respectively. The ends of the whiskers indicate the minimum and maximum of all the data. Data were analyzed by the Student’s t test. ∗ P  < 0.05; ∗∗ P  < 0.01; n.s., not significant Effects of AZM and ABT263 on cell viability and IL-6 secretion in CSCs. ( A ) AZM treatment reduces the viable cell fraction in CSCs ( n  = 5 in each group). ( B ) ABT263 treatment reduces the viable cell fraction in CSCs ( n  = 5 in each group). ( C ) Representative cytokine array ( n  = 4 in each group) images (upper panels), quantification of intensity of each spot (middle table). ( D ) The ratio of supernatant IL-6 concentrations before vs. after AZM treatment is shown in box plot form ( n  = 5 in each group). ( E ) The ratio of IL-6 concentrations before vs. after AZM treatment is shown in box plot form ( n  = 5 in each group). nESCs, normal endometrial stromal cells; eESCs, endometrial stromal cells with ovarian endometriosis (OE); CSCs, OE (chocolate cyst)-derived stromal cells; AZM, azithromycin; ABT263, navitoclax. (A and B) Data are shown as means ± SEM. (D and E) The horizontal line centered inside each box indicates the median. The bottom and top of the box indicate the 25th and 75th percentiles, respectively. The ends of the whiskers indicate the minimum and maximum of all the data. Data were analyzed by the Student’s t test. ∗ P  < 0.05; ∗∗ P  < 0.01; n.s., not significant To evaluate SASP factors in nESCs and CSCs, we performed a cytokine array with the respective culture supernatants. Representative images of the cytokine array are shown in Fig.  3 C. The integrated densities of IL-6, CXC motif ligand 1 (CXCL1), GRO a/b/g, and Interleukin-8 (IL-8) were significantly higher in CSCs than in nESCs (IL-6; P  < 0.0001; CXCL1; P  < 0.001; GRO a/b/g, P  < 0.05; and IL-8; P  < 0.05). Among these cytokines, IL-6 levels were significantly reduced by AZM treatment ( P  = 0.056) (Fig.  3 C). The relative supernatant concentration of IL-6 was significantly reduced in CSCs compared to that in nESCs with AZM treatment (nESCs: 0.69 ± 0.10, CSCs: 0.36 ± 0.078, P  < 0.05) (Fig.  3 D). By contrast, ABT263 did not impact the relative supernatant concentration of IL-6 (nESCs: 0.37 ± 0.095, CSCs: 0.59 ± 0.14, P  = 0.24) (Fig.  3 E). The experimental protocol using our murine model is shown in Fig.  1 A. AZM-treated and vehicle groups were euthanized 8 weeks after transplantation (at 17-week-old), and endometriotic lesions were evaluated. Multiple cystic lesions were observed in the abdominal cavity (Fig. S2 A). The number of cysts was comparable between the vehicle- and AZM-treated groups (Fig.  1 B and C). However, treatment with AZM for 4 weeks significantly reduced the volume per cyst compared to that in the vehicle group (81.9 ± 9.7 vs. 66.2 ± 18.5 mm 3 , P  < 0.05) (Fig.  1 B and D). Histological analysis of endometriotic lesions was performed by identifying the epithelial component via the pair-box (PAX) 8 expression [ 29 ] and the stromal component via progesterone receptor (PgR) expression (Fig. S2 B). To evaluate proliferative activity within endometriotic lesions, the percentage of Ki67-positive cells was calculated. The percentage of Ki67-positive cells in endometriotic lesions decreased significantly in both epithelial and stromal cells (epithelial; 21.9 ± 8.2 vs. 7.4 ± 4.6%, P  < 0.01, stromal; 3.2 ± 0.9 vs. 0.7 ± 0.3%, P  < 0.01) after AZM treatment (Fig.  1 E and F), and fibrotic lesions in endometriotic cysts were significantly reduced after AZM treatment (36.9 ± 2.8 vs. 13.7 ± 2.6%, P  < 0.0001) (Fig.  1 G and H). We evaluated the effects of AZM on IL-6 expression in murine endometriotic cysts. IL-6-positive areas in epithelial cells were comparable with or without treatment, while in stromal cells, IL-6-positive areas significantly decreased after AZM treatment (epithelial; 30.4 ± 2.2% vs. 28.7 ± 3.4%, P  = 0.66, stromal; 20.0 ± 2.1% vs. 3.3 ± 0.8%, P  < 0.0001) (Fig.  1 I and J). No mice died during the experiment. They were able to take standard amounts of chow and water, and normal feces were observed.

Patient samples were used for stromal cell isolation. All samples were obtained from the Nagoya University Hospital between July 2018 and November 2021. The Ethics Committee of Nagoya University Graduate School of Medicine (2017 − 0503) approved this study. Written informed consent was obtained from each patient who enrolled in this study. Samples were collected from 31 patients: OE samples from 22 patients, endometrial tissues with OE from 13 patients, and endometrial tissues without OE from 9 patients who diagnosed with mature cystic teratomas, leiomyomas, uterine anomalies, or cervical dysplasia, but no history or surgical findings of OE. Normal endometrial stromal cells (nESCs), endometrial stromal cells with OE (eESCs), and OE (chocolate cyst)-derived stromal cells (CSCs) were isolated from patient tissue and cultured as previously described [ 19 ]. Tissue were chopped in Dulbecco’s modified Eagle’s medium (DMEM; Nacalai Tesque, Kyoto, Japan) and incubated with collagenase solution (1 mg/mL; FUJIFILM Wako Pure Chemical Corporation Osaka, Japan) for 30 min at 37 ℃. The cell suspension was filtered through 70 μm filter membranes, followed by centrifugation to obtain a stromal cell pellet. nESCs, eESCs and CSCs were then resuspended in fresh DMEM containing 10% fetal bovine serum (Cosmo Bio, Tokyo, Japan), 100 IU/mL penicillin, 100 mg/L streptomycin, and 25 mg/L amphotericin B. The next day, media containing unattached cells were transferred to a second dish before the media was removed and discarded. The cells were routinely maintained at 37℃ until they reached 90% confluence and were then seeded for experimental purposes, as details below. Depending on their growth, cells were used for sequential experiments in passages 2–7. SA-β-Gal staining of cultured cells was performed to identify senescent cells. Staining was done according to a protocol modified from the previously described [ 27 ] using Senescence beta-Galactosidase Staining Kit (Cell Signaling Technology, Inc., Danvers, MA, USA). Cells were seeded into each well of a 6-well plate and allowed to adhere for 24 h. Adherent cells were washed once with phosphate-buffered saline (PBS) and then fixed in a solution containing 2% formaldehyde and 0.2% glutaraldehyde for 15 min at room temperature. After another two PBS washes, the cells were incubated in a staining solution (150 mM NaCl, 2 mM MgCl 2 , 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2.45 mM [1 mg/mL] 5-bromo-4-chloro-3-indolyl-β-D-thiogalactopyranoside [X-Gal] in phosphate-citrate buffer, pH 6.0) at 37 °C without CO 2 overnight. The percentage of SA-β-Gal-positive cells was determined using an inverted microscope (Olympus IX51, Tokyo, Japan) by counting cells in five random fields of view at a magnification of 200×. Images were captured using a Canon 1100 digital camera (Tokyo, Japan). Staining intensity was quantified using ImageJ software (threshold 110). SA-β-Gal activity was quantitated with a Cellular Senescence Plate Assay Kit using SPiDER-βGal, a fluorogenic substrate for β-galactosidase (SG05, Dojindo, Kumamoto, Japan), following the manufacturer’s instructions. The optimal cell numbers in each well were adjusted using a Cell Count Normalization Kit (C544, Dojindo, Kumamoto, Japan), following the manufacturer’s instructions. Cells were lysed with Lysis Buffer for 10 min at room temperature, then incubated with SPiDER-βGal working solution at 37 °C for 30 min. After adding a stop solution, using an excitation wavelength of 535 nm, emission at 580 nm was recorded using a multimode microplate reader (SpectraMax iD5, Molecular Devices, San Jose, CA, USA). Cells were cultured on coverslips for 24 h, fixed in methanol for 2 min at room temperature, and permeabilized with 0.5% Triton X-100 for 1 min. After blocking with 1% bovine serum albumin for 1 h at room temperature, cells were incubated with primary antibodies against P16 INK4a (ab108349, 1:100, Abcam, Cambridge, UK), LaminB1 (MA1-06103, 1:100, Thermo Fisher Scientific, Waltham, MA, USA), and IL-6 (MAB2061, Bio-Techne, MN, USA) or 2 h at room temperature. Cells were then incubated with goat anti-rabbit (Alexa Fluor® 568, 1:500, Thermo Fisher Scientific, Waltham, MA, USA, for anti-P16 INK4a antibody) or goat anti-mouse (Alexa Fluor® 488, 1:500, Thermo Fisher Scientific, Waltham, MA, USA, for anti-LaminB1 antibody and anti-IL-6 antibody) secondary antibodies for 1 h at room temperature. Nuclear staining was performed using 4’,6-diamidino2-phenylindole (4083, 1:1000; Cell Signaling Technology, Danvers, MA, USA). Visualization was performed using a confocal laser-scanning microscope (BZ9000, Keyence, Osaka, Japan). The effect of AZM and ABT263 on cell viability were determined using cell counting. Cells (1.0 × 10 5 ) were seeded into each well of a 6-well plate and allowed to adhere for 24 h. The cells were cultured in serum-free media for 24 h, followed by treatments with AZM (azithromycin, 50 μM, PZ0007 Sigma Aldrich, St. Louis, MO, USA) for 72 h or ABT263 (navitoclax, 1 μM, HY-10087, MedChemExpress, Monmouth Junction, NJ, USA) for 24 h. Concentrations of AZM and ABT263 were applied in accordance with previous reports [ 25 , 26 ]. Finally, 1 mL of 0.25% trypsin was added to each well, and cells were incubated with it for 5 min at 37 °C. Detached cells were collected and centrifuged, and the number of viable cells was counted. nESCs and CSCs were treated for 72 h with 50 μM AZM. Cytokines secreted by the cells were detected in conditioned medium using RayBiotech Human Cytokine Arrays C3 according to the manufacturer’s instructions (RayBiotech Inc., Peachtree Corners, GA, USA). A LAS4000 CCD Imaging System (Fujifilm Co. Ltd., Tokyo, Japan) was used to detect proteins. The signal intensity of each spot, which represented the secreted cytokines, was evaluated by subtracting the background and normalizing to positive controls using ImageJ software. nESCs and CSCs were treated for 72 h with 50 μM AZM or for 24 h with 1 μM. Interleukin-6 (IL-6) concentrations in the culture supernatants were measured using the CLEIA method (FUJIREBIO, Tokyo, Japan) at an outsourced laboratory (H.U. Frontier, Inc., Tokyo, Japan). A murine endometriosis model was used as previously described [ 28 ]. The experimental protocol is summarized in Fig.  1 A. Female mice (6-week-old, BALB/c) were purchased from Japan SLC (Shizuoka, Japan). All animal experiments were approved by the Animal Experimental Committee of the Nagoya University Graduate School of Medicine (M230117-002 and M230297-002). Before starting the experiments, animals were acclimated for 7 days at 23–25 °C with a 12 h/12 h dark/light cycle and were given standard chow (CE-2; CLEA Japan, Tokyo, Japan) and water. Cage linings were changed every week. Fig. 1 AZM prevents endometriosis progression in murine models. ( A ) Our experimental protocol is shown. Four weeks after uterine tissue transplantation, recipient mice were treated with 50 mg/kg AZM twice weekly. ( B-D ) The numbers of endometriosis lesions were comparable in both groups, while volume per endometriosis lesion was reduced in the AZM group. ( E ) Representative Ki67 immunohistochemical images of endometriosis lesions. Scale bar: 100 μm. ( F ) Percentages of Ki67-positive cells in endometriosis lesions decreased in the AZM group in both epithelial and stromal cells. ( G ) Representative Masson trichrome staining analysis results of endometriosis lesions. Scale bar: 50 μm. ( H ) Fibrotic lesions within endometriosis cysts were significantly reduced in the AZM group. ( I ) Representative images of IL-6 immunohistochemistry in endometriosis lesions. ( J ) IL-6-positive areas in stromal cell lesions were significantly decreased in the AZM group ( n  = 6 in each group). AZM, azithromycin. Data are shown as means ± SEM. Data were analyzed by the Student’s t test. * P  < 0.05; ** P  < 0.01; *** P  < 0.001; n.s., not significant AZM prevents endometriosis progression in murine models. ( A ) Our experimental protocol is shown. Four weeks after uterine tissue transplantation, recipient mice were treated with 50 mg/kg AZM twice weekly. ( B-D ) The numbers of endometriosis lesions were comparable in both groups, while volume per endometriosis lesion was reduced in the AZM group. ( E ) Representative Ki67 immunohistochemical images of endometriosis lesions. Scale bar: 100 μm. ( F ) Percentages of Ki67-positive cells in endometriosis lesions decreased in the AZM group in both epithelial and stromal cells. ( G ) Representative Masson trichrome staining analysis results of endometriosis lesions. Scale bar: 50 μm. ( H ) Fibrotic lesions within endometriosis cysts were significantly reduced in the AZM group. ( I ) Representative images of IL-6 immunohistochemistry in endometriosis lesions. ( J ) IL-6-positive areas in stromal cell lesions were significantly decreased in the AZM group ( n  = 6 in each group). AZM, azithromycin. Data are shown as means ± SEM. Data were analyzed by the Student’s t test. * P  < 0.05; ** P  < 0.01; *** P  < 0.001; n.s., not significant Donor mice were ovariectomized and injected subcutaneously with 17b-estradiol (E2, 100 μg/kg per mouse once a week; Fuji Pharma, Tokyo, Japan) in olive oil. After two weeks, donor uteri were removed after euthanasia. The uteri were slit longitudinally with a linear incision and minced using fine scissors. Twelve recipient mice (9-week-old) were anesthetized with isoflurane and ovariectomized. The minced donor uterine tissue, equivalent to half of a uterus dispersed in 500 μL saline, was injected into the peritoneal cavity. Recipient mice were injected subcutaneously with E2 (100 μg/kg) in olive oil once a week for eight weeks until the end of the experiment (17-week-old), when autopsies were performed. Four weeks after transplantation, 12 recipient mice were divided into 2 groups, AZM or vehicle group. Six mice were treated with AZM (50 mg/kg, A9834, LKT, St. Paul, MN, USA) and the other 6 mice were treated with 0.5% carboxymethyl cellulose (CMC, C4888, Sigma Aldrich, Saint Louis, MO, USA) orally twice a week. The first dose of AZM was administered the day after the transplantation of the donor uterine tissue. All twelve mice were euthanized for analysis 24 h after the last administration. Identified endometriotic lesions were excised as single masses. The number of lesions were recorded, and the diameter of each lesion was measured. Lesion volume was calculated by measuring the width (α), length (β), and height (γ) and applying the formula for the volume of an ellipse (V = 4/3 π abc [mm 3 ]; a = 1/2α, b = 1/2β, c = 1/2γ) as previously described [ 19 ]. The endometriotic lesions were fixed with 10% formalin solution for 24 h, replaced with 70% ethanol, embedded in paraffin, cut into sections of 3 μm thickness, and examined by immunohistochemical analysis. Immunohistochemical staining was performed using a Leica Bond Max automated system (Leica, Bannockburn, IL, USA) according to the manufacturer’s instruction. Primary antibodies were Ki67 (ab9260, 1:500, Abcam, Cambridge, UK), IL-6 (bs-0782R, 1:200, Bioss Antibodies, Woburn, MA, USA), PAX8 (ab97477, 1:2000, Abcam, Cambridge, UK), and PgR (ab16661, Abcam, Cambridge, UK) diluted to appropriate concentrations using BOND Primary Antibody Diluent (AR9352). Immunostaining was performed with either BOND Epitope Retrieval Solution 1 (AR9961) or 2 (AR9640), depending on the primary antibody used. AR9961 was used for anti-Ki67 antibody, anti-IL-6 antibody and anti-PgR antibody. AR9640 was used for anti-PAX8 antibody. The stained sections were observed under a microscope (Axio Imager 2, Zeiss, Oberkochen, Germany). Staining intensity was quantified using ImageJ software (threshold Ki67/180, Masson trichome/200, IL-6/180). Statistical analyses were performed using Student’s t-tests or one-way analyses of variance (ANOVA) using Microsoft Excel and Prism 8 software (GraphPad, San Diego, CA, USA). The significance of the difference ( P  value) was calculated and defined as either ∗ P  < 0.05, ∗∗ P  < 0.01, ∗∗∗ P  < 0.001, or no significance (n.s.). Data are reported as means ± standard error of the mean (SEM), unless specified otherwise.

Our study showed that CSCs contain more senescent cells compared to nESCs or eESCs. CSCs secreted IL-6 at the highest levels among other SASP factors examined. AZM and ABT263 reduced cell viability and AZM effectively reduced IL-6 secretion in CSCs. Furthermore, AZM suppressed IL-6 expression and inhibited endometriosis progression in a murine model. To our knowledge, this is the first report to suggest that AZM may inhibit endometriosis progression. The involvement of cellular senescence in endometriosis has recently been reported [ 21 – 24 ]. Our findings are in line with previous reports in that we observed an increase in several senescence markers expressions in endometriosis. While senotherapy for cancer and chronic inflammatory diseases have been actively investigated [ 3 , 10 , 11 ], no reports currently describe its efficacy in treating endometriosis. We thus assessed two senotherapeutics, the BCL-2 protein inhibitor ABT263, and the antibiotic AZM, which are known to exert highly specific cytotoxicity toward senescent fibroblasts [ 25 , 26 ]. Although a clinical trial of ABT263 was conducted for cancer treatment in combination with anticancer agents, the trial was terminated due to its strong side effects such as grade 3 or 4 thrombocytopenia, and lymphocytopenia, increased in aminotransferases [ 30 ]. AZM is a frequently used antibiotic in obstetrics and gynecology with mostly tolerable side effects (including diarrhea) [ 31 ]. In this study, ABT263 induced cell death in nESC, eESC and CSC, while AZM specifically reduced cell viability in CSC, with less effects on nESC and eESC. Furthermore, although ABT263 was not effective, AZM inhibited IL-6 secretion in CSCs. Considering the side effects of ABT263 and the CSC-specific effects of AZM, AZM was selected for an in vivo study with a murine model. AZM reduced endometriotic lesion volume, epithelial and stromal proliferation, fibrotic lesions, and IL-6 expression. Our results suggest that AZM is a potential non-hormonal therapeutic for endometriosis via suppressing IL-6 expression. Chronic inflammation is a key characteristic of endometriosis [ 1 , 12 ]. However, the pathogenesis of chronic inflammation in endometriosis remains unclear. We hypothesized that cellular senescence might be a contributing factor of chronic inflammation, and we especially focused on SASP, a phenomenon in which senescent cells secrete bioactive substances, including cytokines. CSCs secreted IL-6 at the highest levels relative to all examined SASP factors, including IL-8, CXCL1, and GRO a/b/g. IL-6 is a multifunctional inflammatory cytokine produced by immune cells and fibroblasts [ 18 ]. Reports have associated it with increased ascites and serum levels in patients with endometriosis [ 13 ], suppression of early embryonic development, and reproductive disorders, including decreased sperm motility and suppression of estrogen production by granulosa cells [ 15 , 16 ]. IL-8 promotes endometrial cell adhesion during endometriosis [ 32 ] and is associated with endometriotic infertility, as is IL-6 [ 17 ]. Chemokines such as CXCL1 have been implicated in the etiology of endometriosis [ 33 , 34 ] and are commonly reported to be involved in fibrosis [ 35 ]. The senescent cells in endometriosis may be responsible for the increased secretion of these inflammatory factors via SASP. In turn, senescent cells in endometriosis may arise from the high oxidative stress levels associated with endometriosis. High levels of oxidative stress contribute to the excessive endometrial cell proliferation, characteristic of endometriosis, as well as inflammation and angiogenesis [ 20 , 36 ]. Oxidative stress may cause SIPS, leading to the formation of senescent cells, although further investigation is required for confirmation. This study has some limitations. First, we did not investigate each specific menstrual phase. Cellular senescence is involved in endometrial dedifferentiation [ 37 ]; IL-6 expression is low during the proliferative phase and high during the secretory and menstrual phases [ 38 ]. In this study, we observed that the proportion of senescent cells varied among patients with endometriosis. We believe there is room for further investigation of these factors (age, disease history, course, etc.) in patients with a high fraction of senescent cells who experience different treatment efficacies. Second, it is yet unclear whether the presented AZM inhibitory effects on endometriosis are not due to its antimicrobial effects, but solely due to its senotherapeutic effects. Metronidazole has been reported to reduce endometriosis progression in mice [ 39 ]; Antibiotic treatment targeting Fusobacterium in the endometrium has been reported as a therapeutic option for patients with endometriosis [ 40 ]. The current study did not examine the antimicrobial effects of AZM in endometriosis. Finally, the effects of AZM on ovarian functions, including follicle development, oocyte quality, and fertility were not evaluated due to the use of a peritoneal endometriosis mouse model in this study [ 28 ]. Further investigation in an OE mouse model is required to fully evaluate the effects of AZM on ovarian function [ 41 ].

Cellular senescence is involved in the pathogenesis of endometriosis and AZM may be useful for preventing endometriosis progression by suppressing IL-6 secretion of a SASP. This study shows that AZM may serve as a potential non-hormonal therapeutic treatment for endometriosis. A new focus on cellular senescence may reveal novel therapeutic approaches for endometriosis.

Endometriosis is a chronic inflammatory disease in which ectopic endometrium engrafts and progresses, which exacerbates in an estrogen-dependent manner [ 1 ]. It affects approximately 10% of women of reproductive age and up to 50% of women with infertility [ 1 ]. Ovarian endometriosis (OE) is the most common type of endometriosis and is one of the main causes of menstrual pain and infertility [ 1 ]. Existing pharmacotherapies for endometriosis suppress follicle development and ovulation, thus preventing the patient from conceiving [ 1 ]. Therefore, non-hormonal drugs that can prevent endometriosis progression and contribute to resolving endometriotic infertility are urgently required [ 1 ]. Cellular senescence is a state of irreversible cell cycle arrest [ 2 , 3 ]. Despite growth arrest, senescent cells display metabolic activity and secrete factors, including inflammatory cytokines, chemokines, growth factors, and matrix metalloproteinases, referred to as the senescence-associated secretory phenotype (SASP) [ 2 – 4 ]. There are two types of senescence; replicative senescence resulting from telomere shortening, and stress-induced premature senescence (SIPS) [ 2 ]. One of the major causes of SIPS is oxidative stress [ 5 ]. Senescent cells positively affect tissue homeostasis, normal development, wound healing, and carcinogenesis inhibition [ 2 , 3 , 6 ], while chronic SASP is implicated in various pathologies, including cancer, chronic inflammatory diseases, and age-related diseases [ 2 , 3 , 6 ]. Senescent cells impact their microenvironments in an autocrine and paracrine manner via SASP to promote senescence and inflammation [ 7 – 9 ]. Senotherapy is the pharmacotherapy to induce cell death (senolytics) or inhibit SASP (senomorphics) by targeting unique features of senescent cells [ 3 , 10 ]. Senotherapy has attracted significant interest in treating cancer and chronic inflammatory diseases [ 10 ]. Efficacies of senotherapeutics, including anticancer agents, flavonoids, and antibiotics, have been reported, and a few are undergoing clinical trials [ 11 ]. Inflammation and oxidative stress are the key factors involved in the pathogenesis of endometriosis [ 12 ]. Inflammatory cytokines regulate endometrial proliferation, fibrosis, and infertility [ 13 – 18 ]. Murakami et al. reported the non-hormonal anti-inflammatory effects of reducing interleukin-1 beta (IL-1b) expression and oxidative stress using the nucleotide-binding oligomerization domain, leucine-rich repeat, and pyrin domain-containing (NLRP) 3 inflammasome inhibitors in OE [ 19 ]. Oxidative stress is known to modulate cellular proliferation in endometriosis, thereby leading to its development and progression [ 20 ]. However, the mechanisms underlying inflammation in endometriosis remain unclear. Although the involvement of cellular senescence in endometriosis has recently been reported [ 21 – 24 ], no studies have examined senotherapy in endometriosis. We thus focused on cellular senescence in endometriosis and senotherapeutics such as B-cell lymphoma-2 (BCL-2) protein inhibitor navitoclax (ABT263) and the antibiotic azithromycin (AZM), which have been reported to be highly specific against senescent fibroblasts [ 25 , 26 ]. We hypothesized that senescent cells in endometriosis may cause chronic inflammation that exacerbates lesions through SASP, and that AZM or ABT263 may offer novel non-hormonal strategies for treating endometriosis. In this study, we aimed to investigate the involvement of cellular senescence in endometriosis stroma and assess the effects of senotherapy to develop a novel non-hormonal therapeutic strategy for endometriosis.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2