Luminescent mouse model of endometriosis: three-dimensional morphology of lesions and cytokine profiles

Endometriosis is a gynecological disorder characterized by the presence of endometrial-like tissue outside the uterine cavity. It affects approximately 30–50% of women with infertility [ 1 ]. One of the major challenges associated with endometriosis is the delay in both diagnosis and treatment. An estimated 6 out of 10 patients with endometriosis are not correctly diagnosed, leading to over half of endometriosis patients not receiving the appropriate treatment. On average, women experience a diagnostic delay of 7 to 11 years from the onset of symptoms to an accurate diagnosis [ 2 ]. These diagnostic inaccuracies and delays represent a significant global health concern, highlighting the inadequate recognition of clinical symptoms in affected individuals and the lack of understanding of the disease mechanism [ 3 ]. The better understanding of the pathophysiology of early-stage endometriosis is crucial for developing new therapeutic strategies. Previous reports have shown that multiple factors such as genetic, immunological, hormonal, and environmental factors contribute to the onset of the disease [ 4 , 5 , 6 ]. Among these multiple factors, one of the major challenges is to investigate endometriosis as chronic inflammatory disease [ 7 ]. However, monitoring cytokine levels during the early stages of the disease in humans is difficult, as these stages are often asymptomatic [ 8 ]. In endometriosis, which is a chronic inflammatory disease and a debilitating condition accompanied by pelvic pain, there has been increasing attention on the involvement of pro-inflammatory and anti-inflammatory cytokines in the pathogenesis and their potential as diagnostic biomarkers [ 9 ]. These cytokines have been reported to play important roles in the invasion, angiogenesis, survival, and growth of endometriotic lesions [ 9 , 10 , 11 ]. Multiple studies have shown that pro-inflammatory cytokines such as IL-2, IL-6, TNF-α, and IFN-γ promote the establishment and progression of endometriosis. On the other hand, IL-10, classified as an anti-inflammatory cytokine, is known to suppress the production of pro-inflammatory cytokines such as IL-2, IL-6, TNF-α, and IFN-γ [ 12 ]. To accurately evaluate the interaction between pro-inflammatory and anti-inflammatory cytokines and their role in the pathogenesis of endometriosis, longitudinal profile studies from the early stages of lesions are essential. Various animal models of endometriosis have been reported using mice, rats, and non-human primates, providing valuable insights into endometriosis research [ 13 , 14 , 15 ]. Mouse models have been most frequently reported due to their ease of husbandry, cost-effectiveness, and high potential for genetic modifications [ 13 ]. Recently, a mouse model was reported that allows non-invasive observation of endometriotic lesion formation by transplanting uterine tissue fragments of inbred transgenic mice carrying a high-luminescent reporter gene into the peritoneal cavity of syngeneic mice using bioluminescence imaging [ 16 ]. Such luminescent mouse models enabling non-invasive imaging are considered useful for studying the molecular mechanisms of disease onset and progression, including cytokine dynamics associated with the growth of endometriotic lesions, as well as for therapeutic research. In this study, we aimed to establish a mouse model capable of measuring cytokine dynamics associated with endometriotic lesion growth. Following previous reports, we used C57BL/6 mice expressing Emerald luciferase (ELuc) which is stabler and brighter than firefly luciferase [ 16 , 17 ]. As recipients, we utilized a C57BL/6 congenic strain of albino ( c/c ) and hairless ( hr/hr ) for efficient luminescent imaging [ 18 ]. Induced endometriotic tissues were analysed using X-ray micro-computed tomography ( μ CT) imaging to understand comprehensive spatial relationship between transplanted and recipient tissues in three-dimension. Furthermore, by performing simultaneous multiplex analysis of serum samples from each recipient, we were able to capture characteristics patterns of temporal changes in the levels of seven inflammation-related cytokines (IL-1β, IL-2, IL-6, IL-10, IL-12p70, IFN-γ, and TNF-α).

The whole genome scanning using SNP markers was conducted to clarify the genetic background of the recipient B6.Cg- c/c - hr/hr strain. The results revealed that the genotype matching rate of the recipient strain with C57BL/6JJmsSlc was 91% (164 out of 180 loci), with all SNP markers on chromosome 17, containing the major histocompatibility complex H2 region, being of B6-genotype. The remaining 9% of non-B6-genotype SNP loci were distributed across nine chromosomes 3, 5, 8, 10, 11, 13, 14, 16 and 18 ( Supplementary Table 2 ). Among these chromosomal segments, the minor histocompatibility 8 locus ( H8 ) is known to colocalize with hr gene on chromosome 14 [ 19 , 21 ]. Uterine tissue fragments from B6-CAG-ELuc transgenic mice expressing Emerald luciferase were transplanted into the peritoneal cavity of B6.Cg- c/c - hr/hr recipients, and E2 was administered (E2(+)) to induce endometriotic lesions. As shown in Fig. 2A Fig. 2. Longitudinal observation of endometriosis mouse model by non-invasive in vivo imaging. The ELuc light emission from the lesion in the abdominal cavity was detected by IVIS on 0, 14, 28, 42 d.p.t. in the same recipient mouse administered with E2 (A) and without E2 (B). The exposure time was 30 seconds. (C) Temporal changes of the total luminescent flux (photon/second) with E2 (A) and without E2 (B) at four time points. E2(+): n=10 at each time point, E2(−): n=6. Bars: mean ± SEM. Significant P -values by Welch’s t -test are indicated: * P <0.05, ** P <0.01, *** P <0.001. , we were able to detect non-invasively the location and growth progression of the transplanted tissue fragments in the peritoneal cavity through bioluminescent signals immediately after transplantation (day 0) and up to 14, 28, and 42 d.p.t. In non-E2-administered (E2(−)) recipient mice, an increase in luminescence was observed from day 0 to day 28, but reached a plateau between days 28 and 42 ( Fig. 2B ). Longitudinal observation of endometriosis mouse model by non-invasive in vivo imaging. The ELuc light emission from the lesion in the abdominal cavity was detected by IVIS on 0, 14, 28, 42 d.p.t. in the same recipient mouse administered with E2 (A) and without E2 (B). The exposure time was 30 seconds. (C) Temporal changes of the total luminescent flux (photon/second) with E2 (A) and without E2 (B) at four time points. E2(+): n=10 at each time point, E2(−): n=6. Bars: mean ± SEM. Significant P -values by Welch’s t -test are indicated: * P <0.05, ** P <0.01, *** P <0.001. When comparing the luminescence of transplanted tissue fragments between E2(+) and E2(−) mice, as shown in Fig. 2C, the E2(+) mice showed significantly higher luminescence compared to the E2(−) mice after 14 d.p.t., with a dramatic increase observed by 42 d.p.t. These results indicate that the tissue transplants successfully engrafted and either grew or survived for 42 days regardless of E2 administration, with particularly notable growth observed in the E2(+) mice. The tissue engraftment sites were monitored by recording bioluminescent signals followed by autopsy examination on 14, 28, and 42 d.p.t. Engrafted lesions were observed around the surgical suture sites in all recipients at all time points. In addition, consistent engraftment was observed around the pancreatic tissues throughout the experimental period, though at a lower frequency than the suture sites, while a few lesions adhering to inguinal adipose tissues were observed in later stages on 28 and 42 d.p.t. ( Supplementary Fig. 1 ). Post-mortem endometriotic lesion tissues were observed under a dissecting microscope, focusing on bioluminescent regions detected by IVIS imaging, to confirm the morphology of engrafted lesion tissues. Endometriotic tissue masses were confirmed to be successfully engrafted at all luminescent sites. At 14 d.p.t., a dome-shaped tissue mass was found to be engrafted near the suture site around the median peritoneum. Blood capillaries branching from the recipient’s bilateral inferior epigastric vessels (arrowheads) were observed extending to the base where the dome-shaped tissue mass closely adhered to the peritoneum, reaching the dome-shaped mass ( Fig. 3A Fig. 3. Representative images of endometriotic lesions by dissecting microscopy. (A) A dome-shaped tissue mass mainly consisted of two cystic structures (*) engrafted at the suture site on the peritoneum on 14 d.p.t. The recipient’s epigastric vessels are indicated with arrowheads. White arrow: a part of black suture thread is faintly visible through the tissue. (B) An engrafted tissue mass at the suture site (white arrow) on the peritoneum on 28 d.p.t. The engrafted tissue consisted of multiple cysts containing pale yellow to transparent fluid (*). (C) An expanded cystic mass (*) found near the pancreas. The engrafted tissue was partly covered with the peripancreatic adipose tissue ( pat ). ). At 28 d.p.t., the engrafted tissue mass near the suture site on the peritoneum had formed a cluster of multiple cysts containing pale yellow to transparent fluid. Multiple blood vessels had invaded the base of the engrafted tissue, and the entire surface of the cystic clusters was covered by a thin membranous tissue with a network of capillaries ( Fig. 3B ). At 42 d.p.t., larger cystic clusters were observed around the suture site. Additionally, cystic masses covered by a thin membrane, an adipose tissue and associated blood capillaries were observed near the pancreas ( Fig. 3C ). Representative images of endometriotic lesions by dissecting microscopy. (A) A dome-shaped tissue mass mainly consisted of two cystic structures (*) engrafted at the suture site on the peritoneum on 14 d.p.t. The recipient’s epigastric vessels are indicated with arrowheads. White arrow: a part of black suture thread is faintly visible through the tissue. (B) An engrafted tissue mass at the suture site (white arrow) on the peritoneum on 28 d.p.t. The engrafted tissue consisted of multiple cysts containing pale yellow to transparent fluid (*). (C) An expanded cystic mass (*) found near the pancreas. The engrafted tissue was partly covered with the peripancreatic adipose tissue ( pat ). After observing endometriotic lesions under a dissecting microscope, the lesion tissues were fixed with PFA and treated by phosphotungstic acid for contrast staining of soft tissues. Using μ CT imaging, three-dimensional images were generated to assess how the engrafted lesion tissue adhered to and formed lesions on either the host’s peritoneum or pancreatic adipose tissues ( Figs. 4 A–C Fig. 4. Representative images of endometriotic lesions by 3D mCT imaging and conventional histology. (A1) A 3D reconstructed mCT image of a dome-shape engrafted tissue on 14 d.p.t. (see Supplementary Movie 1A for 360° view). (A2) A digital cross-section of A1 to visualize the internal structure of the cyst. (B1) A cystic engrafted lesion associated with a pancreatic tissue (see Supplementary Movie 1B for 360° view) and (B2) a digital cross-section of the cystic engrafted lesion (*) and the surrounding capillaries on 28 d.p.t. pnc : pancreatic tissue. (C1) An expanded cystic lesion on the peritoneum with a dense capillary network extended on the surface of the cystic lesion on 42 d.p.t. (see Supplementary Movie 1C for 360° view). (C2) Two tissue cross-sections 1 and 2 cut at different positions and angles, created from three-dimensional μ CT data. Expanded cysts (*) fused to form the lesion. (D) H&E-stained tissue section on 42 d.p.t. prepared by embedding in standard paraffin blocks after mCT image acquisition. An expanded cyst (*) with a fluid-filled luminal space was lined with single-layered epithelium and surrounded by uterine gland-like ( gl ) structures and interstitial stroma. ). We obtained three-dimensional voxel information for repeatative observations from any angle and created movies for the overall view (Figs. 4A1, B1, and C1, Supplementary Movies 1A–C ). At 14 d.p.t., the lesion tissue adhered tightly to the peritoneum, penetrating the muscular layer to such an extent that the adhesion site could be confirmed from the outer side of the muscular layer (behind the peritoneum) ( Supplementary Movie 1A ). Furthermore, observing tissue sections at arbitrary angles revealed that many lesions had developed into fused lesions consisting of multiple cysts of approximately 2–3 mm in diameter (Figs. 4A2, B2, and C2). The lesion tissue near the pancreas did not directly adhere to the pancreas itself but became engrafted in the surrounding pancreatic adipose tissue, where the lesion tissue was enveloped by a reticular network of capillaries extending from either the adipose tissue or pancreatic tissue. In the lesion tissue that had adhered to the peritoneum at 42 d.p.t., we could observe an intricate vascular network surrounding the base of the lesion tissue (Figs. 4C1 and C2, Supplementary Movie 1C ). After μ CT analysis, cyst-containing lesion tissues were embedded in paraffin, sectioned, and observed with H&E staining. The lesion tissue consisted of large cysts lined by columnar epithelium along with glandular tissues and interstitial stroma ( Fig. 4D ). Representative images of endometriotic lesions by 3D mCT imaging and conventional histology. (A1) A 3D reconstructed mCT image of a dome-shape engrafted tissue on 14 d.p.t. (see Supplementary Movie 1A for 360° view). (A2) A digital cross-section of A1 to visualize the internal structure of the cyst. (B1) A cystic engrafted lesion associated with a pancreatic tissue (see Supplementary Movie 1B for 360° view) and (B2) a digital cross-section of the cystic engrafted lesion (*) and the surrounding capillaries on 28 d.p.t. pnc : pancreatic tissue. (C1) An expanded cystic lesion on the peritoneum with a dense capillary network extended on the surface of the cystic lesion on 42 d.p.t. (see Supplementary Movie 1C for 360° view). (C2) Two tissue cross-sections 1 and 2 cut at different positions and angles, created from three-dimensional μ CT data. Expanded cysts (*) fused to form the lesion. (D) H&E-stained tissue section on 42 d.p.t. prepared by embedding in standard paraffin blocks after mCT image acquisition. An expanded cyst (*) with a fluid-filled luminal space was lined with single-layered epithelium and surrounded by uterine gland-like ( gl ) structures and interstitial stroma. We measured the serum level of inflammation-related cytokines such as IL-1β, IL-2, IL-6, IL-10, IL-12p70, TNF-α, and IFN-γ on 14, 28 and 42 d.p.t. in the endometriosis mouse model using a multiplex magnetic immunobeads assay. As shown in Figs. 5 A–G Fig. 5. Change in serum level of cytokines overtime. Serum levels of IFN-γ (A), TNF-α (B), IL-12p70 (C), IL-1β (D), IL-6 (E), IL-2 (F), IL-10 (G) in the recipient mice with endometriotic lesions (n=10 per group) and non-treated control as 0 d.p.t. (n=5). Bars: mean ± SEM. Significant P -values are indicated as follows: * P <0.05, ** P <0.01, *** P <0.001, and **** P <0.0001. , examining statistically the graphs of each cytokine concentration revealed that the cytokines could be classified into four groups with characteristic patterns of temporal changes ( Supplementary Table 2 ). IFN-γ ( Fig. 5A ) and TNF-α ( Fig. 5B ) showed significant increases at each time point, demonstrating stable increases ( P <0.01). IL-12p70 ( Fig. 5C ) and IL-1β ( Fig. 5D ) showed gradual increases in the earlier stages ( P <0.05) followed by marked elevation in the later stages ( P 0.05) but showed dramatic increases between 28 and 42 d.p.t. ( P <0.001). IL-10 ( Fig. 5G ) uniquely showed a transient increase in the early stage ( P <0.01) followed by gradual decreases from 14 to 42 d.p.t. ( P <0.01), displaying a pattern different from other cytokines. These distinct patterns suggest that each cytokine plays a distinct role in inflammation and immune response. Change in serum level of cytokines overtime. Serum levels of IFN-γ (A), TNF-α (B), IL-12p70 (C), IL-1β (D), IL-6 (E), IL-2 (F), IL-10 (G) in the recipient mice with endometriotic lesions (n=10 per group) and non-treated control as 0 d.p.t. (n=5). Bars: mean ± SEM. Significant P -values are indicated as follows: * P <0.05, ** P <0.01, *** P <0.001, and **** P <0.0001. Furthermore, when the correlation between temporal changes of each cytokine level and temporal changes of luminescence values indicating the lesion growth of E2(+) recipients was analysed by calculating Pearson correlation coefficients ( Supplementary Table 3 ), IL-6 and IL-2 showed the strongest positive correlations with luminescence/tissue growth (IL-6: r=0.967, IL-2: r=0.945, P <0.001). IFN-γ, TNF-α, IL-1β, and IL-12p70 also showed significant positive correlations. IL-10 was the only cytokine showing a negative correlation with luminescence, though without statistical significance. Particularly during 28–42 d.p.t., the increase in IL-6 and IL-2 showed the strongest association with the luminescence increase/lesion growth.

In this study, we developed an improved endometriosis mouse model optimized for inflammatory cytokine quantification by minimizing experimental invasiveness ( Supplementary Table 1 ). The protocol utilized B6-CAG-ELuc transgenic mice [ 16 ] as donors and B6.Cg- c/c - hr/hr mice as recipients, with modifications aimed at reducing confounding factors. The use of B6.Cg- c/c - hr/hr recipients enhanced luminescent singal detection by eliminating interference from skin pigmentation and fur [ 18 ], while avoiding inflammation from repeated shaving [ 13 ]. For synchronizing estrus cycles, we relied solely on vaginal appearance observation and donors’ uterine edematous changes as indicators, avoiding surgical ovariectomy and hormone administration. By limiting surgical invasion to a single abdominal incision, we achieved consistent lesion establishment while minimizing invasive procedures that could affect cytokine measurements. Following transplantation, endometriotic lesions were established using E2 administration alone, which promotes both endometrial proliferation and angiogenesis [ 16 , 22 ]. The successful engraftment over 40 days demonstrated acceptable histocompatibility between the strains, despite the recipients retaining approximately 9% non-B6-genotype regions potentially including the region containing the H8 on chromosome 14, one of the classical minor histocompatibility loci [ 21 ]. A key advantage of our model is that we successfully induced endometriotic lesions with E2 administration while preserving recipients’ ovaries intact without ovariectomy. This feature makes our model particularly suitable for research on endometriosis-related infertility [ 1 , 26 ]. The cystic lesions were primarily composed of fluid-filled cysts, uterine gland-like epithelium, and the interstitial stroma as reported in previous mouse models [ 13 , 16 , 18 ]. These morphological features were similar to the pathological findings observed in human endometriosis [ 27 ]. However, it has been reported that in experimental mouse models, morphological characteristics of lesion tissues can vary depending on genetically different strains and estrous cycles of donors and recipients [ 28 ]. Therefore, careful attention must be paid to these variables when designing and interpreting experimental studies. Two prominent sites were observed for high-frequency and stable engraftment of uterine tissue transplants: the post-surgical suture sites and the pancreatic region ( Supplementary Fig. 1 ). The similar adhesion sites reported in other mouse experiments [ 16 ], which aligns with clinical observations where endometriosis can develop in cesarean section scars [ 29 ]. By our morphological observations, sites undergoing angiogenesis and tissue repair processes are thought to provide favorable conditions for endometrial tissue engraftment. The preferential adhesion of the endometriotic lesions to the pancreatic region may be attributed to the presence of peripancreatic adipose tissue (PAT), a metabolically active tissue containing small adipocytes and high amounts of stromal vascular cells [ 30 ]. Unlike other adipose tissues, PAT’s unique properties in promoting inflammatory responses [ 30 ] suggest PAT provides more favorable conditions for endometriotic lesion engraftment. The significant vascularization observed around the engrafted lesion tissues in our 3D images further supports this hypothesis. Additionally, since E2 is known to promote vascular endothelial growth factor (VEGF) expression in blood vessels [ 22 ], E2 administration may have also contributed to the rapid growth of lesion tissue via enhanced angiogenesis in these preferential sites. Our longitudinal analysis of cytokine profiles in the mouse model revealed dynamic changes that closely parallel observations reported in endometriosis patients [ 31 , 32 , 33 ]. We observed significantly elevated levels of pro-inflammatory cytokines (IL-1β, IL-2, IL-6, IL-12p70, TNF-α, and IFN-γ) that increased in a time-dependent manner, peaking on 42 d.p.t., coinciding with the period of rapid lesion growth. Notably, IFN-γ and TNF-α showed consistent increases throughout the observation period, suggesting their crucial role in disease progression, which aligns with clinical findings of elevated IFN-γ and TNF-α levels in the peritoneal fluid of patients [ 34 , 35 ]. Interestingly, anti-inflammatory cytokine IL-10 showed a contrasting pattern, with initial elevation followed by decline, reflecting the inflammatory imbalance characterized by the predominance of pro-inflammatory cytokines [ 12 ]. The sharp increase in IL-6 and IL-2 levels on 42 d.p.t. further supports the establishment of an activated inflammatory state, consistent with clinical observations [ 36 , 37 ]. These temporal changes in our model provide valuable insights into the early inflammatory dynamics of endometriosis, offering an experimental foundation for understanding disease progression and potential therapeutic interventions. Rahmawati et al. (2023) reported elevated serum levels of IL-1β, IL-6, IL-8, and IL-12p70 in a study examining women with endometriosis and infertility, particularly noting that IL-12p70 correlated with pain assessment scores in endometriosis [ 33 ]. Since a significant increase in IL-12p70 was also confirmed in the later stage of lesion growth in this mouse model ( Fig. 5C ), combining the model with appropriate methods to evaluate pain, such as measuring changes in spontaneous behavior [ 38 ], may be useful in developing treatments for pain associated with endometriosis. In conclusion, our study has yielded novel insights that contribute to the understanding of endometriosis through the development of an improved mouse model that effectively reflects human disease characteristics. In addition to the successful utilization of B6.Cg- c/c - hr/hr mice as recipients, the combination of advanced imaging techniques and comprehensive cytokine profiling has enabled us to demonstrate new resources and fundamental technologies that promote molecular and cellular level research into the pathophysiology of endometriosis.

All animal experiments and use of genetically modified mice were approved by the Institutional Animal Care and Use Committee (T2023-EP001 and T2024-EP001) and the Genetic Recombination Experiments Committee (T2022-02-17) of the RIKEN Tsukuba Branch, respectively. All mice were cared for and used humanely in accordance with the Committee’s guiding principles. A total of 77 female mice were used in this study, consisting of 36 hemizygous C57BL/6JJmsSlc-Tg(CAG-ELuc)1Nkzom transgenic mice expressing ELuc (B6-CAG-ELuc, RBRC11951) [ 16 ] at 8 weeks old and 41 congenic B6.Cg- Tyr c-2J /Tyr c-2J -Hr hr / Hr hr (B6.Cg- c/c - hr/hr , RBRC05798) mice [ 19 ] at 10–12 weeks old. Female B6-CAG-ELuc mice and B6.Cg- c/c - hr/hr mice were obtained from the Experimental Animal Division, RIKEN BRC through the National BioResource Project of the MEXT, Japan. Mice were acclimatized for more than seven days before the start of experiments. All mice were maintained as specific pathogen-free at a temperature of 23–25°C under a 12-h light/dark cycle. The mice were fed with standard food (CE-2; CLEA Japan, Inc., Tokyo, Japan) and water ad libitum . The genome scanning by SNPs markers was used to investigate the genetic background of B6.Cg- c/c - hr/hr mice. PCR templates for SNP analysis were prepared from tail tissues of two parental breeder pairs of adult B6.Cg- c/c - hr/+ females and B6.Cg- c/c - hr/hr males which produced B6.Cg- c/c - hr/hr recipient females. The genotyping was performed using custom TaqMan SNP Genotyping Assays designed for a set of validated SNPs at 180 polymorphic loci across all chromosomes [ 20 ]. The samples’ genotype was compared with those of various C57BL/6 substrains including C57BL/6JJcl, C57BL/6JCrlj, C57BL/6JJmsSlc, C57BL/6NCrlCrlj, C57BL/6NCrl, C57BL/6NJcl, C57BL/6NSlc, C57BL/6NTac, as well as DBA/2JJcl and CBA/J strains. The evaluation of congenic status primarily focused on the genotype matching rate with the donor strain C57BL/6JJmsSlc, while also taking into consideration the matching of genotypes in the major histocompatibility complex H2 region and minor histocompatibility loci [ 21 ]. Endometriotic lesions were induced by transplantation of the uterine tissues as previously described [ 16 ] with modifications to minimize invasive procedures for the cytokine analysis ( Supplementary Table 1 ). The B6-CAG-ELuc female mice of 8 weeks old as donors and congenic B6.Cg- c/c - hr/hr female mice of 10–12 weeks old as recipients were used for this study. Both donor and recipient females were examined for vaginal appearance, and those not showing characteristics of either proestrus or estrus (vaginal swelling or cornification) were selected for use in this study. The donor females were euthanised by cervical dislocation under inhalation anesthesia with isoflurane (Viatris, Inc., Tokyo, Japan) and both uterine horns, confirmed not to be edematous as typically seen in estrus, were dissected out into sterile saline (Otsuka Pharmaceutical Factory, Inc., Tokushima, Japan). The uterine horns were cut into rounds at approximately 1 mm interval, and minced into smaller fragments of approximately 1 mm × 1 mm on a paraffin wax plate (GC Corp., Tokyo, Japan) using a carbon steel razor blade of 0.1 mm thickness (FA-10, FEATHER Safety Razor Co., Ltd., Osaka, Japan). Small incision of approximately 3 mm was made in the abdominal skin and muscle layer of each recipient anesthetized with a mixture of medetomidine (Domitor, Nippon Zenyaku Kogyo Co., Ltd., Fukushima, Japan), midazolam (Dormicum, Maruishi Pharmaceutical Co., Ltd., Osaka, Japan) and butorphanol (Vetorphale, Meiji Seika Pharma Co., Ltd., Tokyo, Japan). Approximately 30 uterine tissue fragments of one donor were suspended in 200 µ l of sterile saline and transferred into the peritoneal cavity of each recipient. Surgical incisions in the muscle layer were closed using 6-0 silk suture (Natsume Seisakusho, Tokyo, Japan), and incisions in skin were clipped (9 mm MikRon autoclip, Becton Dickinson and Co., Franklin Lakes, NJ, USA). The recipient was injected subcutaneously with 0.5 µ g/mouse of 17b estradiol (E2) (Progynon ® -Depot, Fuji Pharma Co., Ltd., Toyama, Japan) in corn oil (FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) after the operation and subsequently once a week to promote the growth of lesions and angiogenesis [ 16 , 22 ]. The schema of experimental procedures is shown in Fig. 1 Fig. 1. Generation of the endometriosis mouse. (A) An experimental work flow for the preparation of donor tissue fragments, surgical transplantation of the donor uterine tissues to the recipient, and observation of luminescent signals by IVIS. (B) Experimental time schedule. Recipient mouse was injected subcutaneously with 0.5 µ g/mouse of 17β estradiol (E2) weekly. Endometriotic lesions were collected on 14, 28 and 42 days post transplantation (d.p.t.). n=10 mice per group in each time point. . Generation of the endometriosis mouse. (A) An experimental work flow for the preparation of donor tissue fragments, surgical transplantation of the donor uterine tissues to the recipient, and observation of luminescent signals by IVIS. (B) Experimental time schedule. Recipient mouse was injected subcutaneously with 0.5 µ g/mouse of 17β estradiol (E2) weekly. Endometriotic lesions were collected on 14, 28 and 42 days post transplantation (d.p.t.). n=10 mice per group in each time point. The growth of endometriotic lesions in the recipient mice was observed on 14, 28 and 42 days post transplantation (abbreviated as d.p.t.). Each recipient mouse was intraperitoneally injected with 150 mg/kg D-luciferin (FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) under inhalation anesthesia with isoflurane. Ten minutes after D-luciferin administration, bioluminescence was visualized and measured by using the IVIS Lumina XRMS Series III imaging system (PerkinElmer Inc., Waltham, MA, USA) with an exposure time of 30 s. The recipient and control mice were deeply anesthetized with a mixture of medetomidine, midazolam, and butorphanol for blood collection via cardiac puncture on 14, 28 and 42 d.p.t. Age-matched untreated B6.Cg- c/c - hr/hr mice (n=5) served as controls on 0 d.p.t. The mice were euthanized after blood collection. Collected blood samples were incubated for 30 min at RT, and centrifuged at 700× g in 4°C for 10 min. The serum was collected and stored at −80°C freezer until further analysis. Tissue samples of endometriotic lesion were collected for morphological observation and recording under a dissecting microscope (Stereozoom S9i, Leica Microsystems, Ltd., Heerbrugg, Switzerland) and IVIS imaging system. Subsequently, the lesion tissues were fixed with 4% paraformaldehyde (PFA) in PBS (pH 7.4) for further morphological analysis. The three-dimensional (3D) µ CT imaging [ 23 ] was conducted with a Comscan-Xmate E90S system (Comscantecno Co., Ltd., Kanagawa, Japan). The fixed tissues in PFA were rinsed with PBS three times, each for 5 min. The tissues were post-fixed in Bouin’s fixative solution for 24 h at RT and stored in 70% ethanol (EtOH) at RT until further processing. Prior to µ CT scanning, the fixed tissues were immersed overnight in a contrast reagent consisting of 1% phosphotungstic acid (PTA) in 70% EtOH at RT to enhance the contrast of the image of soft tissues [ 24 ]. The contrast-enhanced tissues were scanned using a tube peak voltage of 40 kV and a tube current of 100 µ A. The tissue samples were rotated 360° in 0.3° steps, generating 1,200 projection images. The acquired images were reconstructed and analysed to generate three-dimensional tomographic visualizations using OsiriX software (Pixmeo, Geneva, Switzerland). The fixed tissues after µ CT imaging were processed by automated tissue-processor (Tissue-Tek ® VIP TM , Sakura Finetek, Tokyo, Japan) and embedded in paraffin block. The embedded tissues were sectioned by a sliding microtome (Yamato Kohki, Saitama, Japan) at 4 µ m thickness on glass slides, deparaffinized and stained with hematoxylin and eosin for histological observation. The tissue section was observed by using an upright Olympus BX51 microscope (Evident, Tokyo, Japan). Levels of serum cytokines including IL-1β, IL-2, IL-6, IL-10, IL-12p70, TNF-α, and IFN-γ were quantified using multiplex magnetic immunobeads assay system [ 25 ] and relevant reagents (Mouse Bio-Plex Pro, Catalog No. L6000004C6, Bio-Rad, Hercules, CA, USA). Each serum sample of 30 µ l was used for the assay according to the manufacturer’s instructions. Cytokine levels were expressed as pg/ml. Data were analysed using GraphPad Prism 9.0 (GraphPad Software Inc., La Jolla, CA, USA). Welch’s t -test was used to analyse total luminescence data. The temporal patterns of cytokine levels were analysed by the two-way repeated measures ANOVA with factor 1 (type of cytokine: 7 cytokines) and factor 2 (Time course: 0, 14, 28, 42 d.p.t.), followed by Tukey’s HSD test for post-hoc analysis. The correlation between each cytokine concentration and luminescence value (log 10 [total flux]) at each time point was analysed by Pearson’s correlation coefficient analysis. The temporal luminescence and cytokine level data were presented as the mean ± SEM.