Candida albicans-induced activation of the TGF-β/Smad pathway and upregulation of IL-6 may contribute to intrauterine adhesion

Iatrogenic injury to endometrial tissue is the main cause of intrauterine adhesions (IUA) and infection can also damage the endometrium. The microbiota plays an important role in the health of the female reproductive tract. However, the mechanism is still unclear. In total, 908 patients with IUA and 11,389 healthy individuals were retrospectively selected for this clinical study. Participant information including vaginal microecological results and human papillomavirus (HPV) status were collected. Univariate and multivariate logistic regression analyses were used to identify the factors related to IUA. Next, animal experiments were performed in a curettage-induced IUA rat model. After the procedure, rats in the experimental group received a vaginal infusion of a Candida albicans (C. albicans) fungal solution. On days 3, 7, and 14 after curettage and infusion, the expression levels of IL-6, fibrotic pathway-related factors (TGF-β1, Smad 2, and COL1), and estrogen receptor (ER) and progesterone receptor (PR) in rat endometrial tissues were assessed. Fungal infection of the reproductive tract was found to be an independent risk factor for IUA (P < 0.05). The inflammatory response and degree of fibrosis were greater in rats infected with C. albicans than in the controls. The levels of IL-6, TGF-β1, Smad 2, and COL1 expression in endometrial tissues were significantly higher in the experimental group than in the control group (P < 0.05). However, the ER and PR levels were lower in the IUA group than in the non-IUA group (P < 0.05). C. albicans infection may be related to IUA. C. albicans elicits a strong inflammatory response that can lead to more severe endometrial fibrosis.

and personal lives 8,9 . Harnessing 16S and ITS2 rDNA sequencing analysis, our previous study found that the fungal genera were enriched in the lower genital tract of IUA patients 10 . Moreover, site-specific fungal-bacterial correlation networks were discovered in those patients 10 . Vaginal candidiasis is the most common vaginal fungal infection 11 . The main pathogenic agent of vaginal candidiasis is the polymorphic fungal pathogen Candida albicans (C. albicans) 12 . Only 30% of recurrent vulvovaginal candidiasis cases are caused by non-C. albicans infections 13 .
The specific host immune response to C. albicans is generally considered to play a key role in protection by preventing symbiotic Candida from turning into conditional pathogens at the mucus membrane-Candida interface. Transforming growth factor (TGF)-β may be involved in the immune response to fungal infection in the reproductive tract. Taylor et al. 14 found that TGF-β1 expression was significantly increased in the vaginal tissue of mice with candidiasis. Furthermore, TGF-β1 was found to be dominant in draining lymph nodes, but not in non-draining lymph nodes. These findings suggest that TGF-β1 may be involved in the immune regulation of vaginal Candida infections 14 . Other studies have found that TGF-β plays a specific role in the process of fungal morphology recognition by dendritic cells 15 . Drug therapy has been reported to downregulate the ratio of proinflammatory cytokine interleukin (IL)-1β to anti-inflammatory cytokine TGF-β, thereby reducing the number of neutrophils in the vagina 16 . TGF-β is a key cytokine in human tissue and animal models of fibrotic diseases, such as IUA [17][18][19] . It has been shown to have a significant effect on fibroblast proliferation 20,21 and extracellular matrix production 22,23 in vitro. The interplay between Smad proteins and other proteins in TGF-β signaling pathway mediate regulatory signals that control expression of target genes 24 . So they are mediators and essential components of TGF-β signaling pathway. We speculated that C. albicans infection may lead to IUA and that the immune regulation mechanism may be related to TGF-β/Smad signaling pathway activation by C. albicans.
As a fibrosis-promoting inflammatory cytokine 25 , IL-6 has been detected at significantly increased levels in patients with pulmonary fibrosis 26 , as well as in a fibrosis mouse model 27 . Recent studies have shown that IL-6 can promote fibrosis by mediating chronic inflammation 28 and activating the TGF-β/Smad pathway [29][30][31][32][33] . Studies have further suggested that Smad 2/3 may be the target of IL-6 trans-signal transduction 29,34,35 . The present study took IL-6 as a starting point for further exploration of the mechanism by which C. albicans activates the TGF-β signal transduction pathway.
We speculated that some patients with IUA who have "asymptomatic" genital tract infections may have abnormal changes in reproductive tract microorganisms and that this vaginal microecological imbalance may be involved in the formation of fibrosis after endometrial injury. Furthermore, we posited that the imbalance of immune regulation caused by C. albicans infection may be an important link in the mechanism of IUA and that IL-6 may be a key regulatory factor. In this research we first retrospectively analyzed the relationship between vaginal microecology and IUA, and preliminarily investigated the etiology of IUA using logistic regression analysis. To further explore the mechanism of endometrial injury caused by fungi, C. albicans was used to stimulate primary intimal cells in a pregnant rat uterine curettage model, and the levels of IL-6, TGF-β1, Smad 2, collagen 1 (COLl), estrogen receptor (ER), and progesterone receptor (PR) were subsequently evaluated. By exploring the mechanism underlying C. albicans-mediated IL-6 upregulation in the reproductive tract and its role in IUA formation, we aimed to reveal the various roles of C. albicans in the progression of IUA and provide new ideas and targets for the discovery of effective preventative measures and treatments for this condition.

Methods
Study population. In this study, 908 patients with IUA who were diagnosed in the Gynecology Clinic of Xiangya Third Hospital of Central South University between August 2017 and September 2018 were retrospectively enrolled as the study group. A further 11,389 healthy people who attended the Health Management Center of Xiangya Third Hospital of Central South University during the same period were retrospectively selected as the control group. The clinical data of the study participants were carefully recorded and reviewed by two research assistants.
All 12,297 study participants provided written informed consent, and the study was approved by the Ethics Committee of Xiangya Third Hospital of Central South University (ethics no. I 22031). All methods were performed in accordance with the relevant guidelines and regulations of the Ethics Committee.
The inclusion criteria for participants were as follows: (i) had not used drugs or other means to treat vaginitis in the 3 months before sampling; (ii) had not engaged in sexual intercourse or other vaginal operations in the 1 day before sampling; (iii) had the ability to understand the advantages and disadvantages of participating in this study and willingness to cooperate with the sampling and testing work required; and (iv) had provided written informed consent. Participants were excluded from the study if they had not been sexually active before, had symptoms or signs of reproductive tract infection, were pregnant or lactating, were unable to cooperate with examinations, or withdrew from the study.
Collection of clinical samples. Samples of vaginal secretions were obtained by gynecologists trained to collect secretions from the upper third of the vaginal sidewall. The swabs with vaginal secretions were inserted into aseptic test tubes and sealed. Then, the cervical mucus was wiped away with a large cotton swab, and a human papillomavirus (HPV) sampling brush was extended into the cervical tube and gently rotated clockwise five times. The front segment of the HPV sampling brush was then snapped off and stored in preservation solution. A test barcode was affixed to the specimen bottle, which was then immediately sent to the laboratory.
Testing of vaginal secretions. The pH of vaginal secretions was detected using pH 3.8-5.4 precision pH test paper, as described previously 36  HPV test. HPV DNA was amplified by polymerase chain reaction (PCR) and detected using an HPV gene chip detection kit (HybriBio, Guangdong, China). Twenty-one types of HPV were detected, of which 14 were high risk (16,18,31,33,35,39,45,51,52,56, 58, 59, 66, and 68), 5 were low risk (6, 11, 42, 43 and 44), and 2 were of unknown risk (53 and CP8304) 38 . If any of these HPV types were detected, a sample was considered HPV positive,if none of these HPV types were detected, a sample was considered HPV negative.

Diagnostic information.
Vaginal secretions with a pH of < 4.5 were regarded as normal, whereas those with a pH of ≥ 4.5 were regarded as abnormal 36   Establishing an IUA model in pregnant rats. The rats were purchased at 13 days of gestation, and the experiment was carried out after 2-3 days of adaptive feeding in the laboratory. The rats were anesthetized by intraperitoneal injection of 40 mg/kg of 2% pentobarbital sodium. A median straight incision was made in the lower abdomen, and a 2-3 cm incision was made through the skin, subcutaneous tissue, rectus abdominis, and peritoneum to enter the abdominal cavity. The Y-shaped uterus was located and gently removed from the abdominal cavity and spread on the surgical hole towel. A 0.5-cm longitudinal incision was made at the proximal end of the bilateral uterus; the fetus, placenta, and other gestational tissues were removed in turn. Curettage was performed using a custom-made curette. The length of the curettage was approximately 5-10 cm from the incision. The strength of the curette was tailored to allow a large amount of tissue to be scraped out and the uterine cavity to become rough, while avoiding penetration of the uterus. The uterus was then sutured, and the abdomen closed with 5-0 silk thread.

C. albicans infection assay.
On day 1 after the procedure, the rats were randomly divided into two groups: the experimental group and the control group (n = 6 in each group). In the experimental group, 50 μL C. albicans culture fluid (with a fungal content of ~ 3 × 10 7 ) was injected into the vagina of each rat using a gastric perfusion needle inserted approximately 2 cm inside the vagina (approximately at the junction of the cervical canal and the uterus). Then the C. albicans migrated from the vagina to the uterine cavity. In the control group, 50 μL phosphate-buffered saline (PBS) solution was administered into the vagina of each rat. There was no difference in the timing of administration between the two groups. Uterine samples were obtained from rats in each group on postoperative days 3, 7, and 14. Specimens were stored in a refrigerator at 4 °C for 24 h, then frozen at -80 °C until use. For histological analysis, specimens were stained with H&E and Masson's trichrome. IHC was used to detect the levels of IL-6, TGF-β1, Smad 2, COLl, ER, and PR in the endometrium, and the levels of IL-6 and TGF-β1 were measured by ELISA. www.nature.com/scientificreports/ Histological examination with H&E staining. For H&E staining, three uterus sections were randomly selected from each rat. The samples were first dewaxed and hydrated and then stained with hematoxylin for 10 min. Next, the samples were rinsed under running water for 2 min, differentiated in 1% saline alcohol for 2 s, rinsed again for 15 min, washed with distilled water for 1-2 s, and then stained with eosin for 10 min. The samples were next differentiated in an 80% ethanol solution according to their color before being dehydrated, first with 85% ethanol for 5 min, then with 95% ethanol for 5 min, and finally with anhydrous ethanol for 10 min, twice. Xylene was used to achieve transparency. After sealing with neutral gum, morphological changes in the endometrium were observed under an inverted microscope (40×).
Masson's trichrome staining. Samples were first dewaxed and hydrated, and then stained with a preprepared Weigert's hematoxylin staining solution for 8 min. Next, the samples were differentiated with 1% acidic ethanol differentiation buffer for 10 s, before being washed with water and transferred to Masson's blue solution for 5 min to be blued. After that, they were dyed with Masson's magenta dye for 5 min, washed for 1 min, and then transferred directly to aniline blue dye for 2 min. Finally, the specimens were dehydrated, rendered transparent, and sealed in neutral resins.
IHC staining. Samples were dewaxed and hydrated. After high-temperature and high-pressure antigen repair, the samples were incubated with 3% H 2 O 2 at room temperature for 10 min to block endogenous peroxidase activity. Then, primary antibodies for IL-6, TGF-β1, Smad 2, COIL1, ER, and PR were added at a dilution of 1:200, and the samples were incubated at 4 °C overnight. PBS was used as the negative control solution.
Following overnight incubation in a wet box, the samples were rewarmed to 37 °C for 30 min. Horseradish peroxidase-labeled polymer (ready-to-use; P009IH; Auragene, Hunan, China) was added to the samples, which were subsequently subjected to incubation at room temperature for 30 min, followed by washing with PBS. Four high-power visual fields were selected in each section, and there were no blank areas in any field. Images of the IHC-stained endometrial sections were captured. Brown areas or granules in the endometrium were taken as positive. The integrated optical density (IOD) of each image was measured using Image-ProPlus 6.0 image analysis software (Media Cybernetics, Bethesda, MD, USA). The IOD values of the endometrial proteins were observed, and the mean IOD value was taken as the IOD value for each uterine specimen. The higher the IOD value, the stronger the positive expression of the proteins.
ELISA. The enzyme-linked immunosorbent assay (ELISA) of endometrial tissues in the two groups were conducted according to the IL-6 and TGF-β1 assay kits. Absorbance was measured at 450 nm with an enzymelabeling instrument, and the levels of IL-6 and TGF-β1 were calculated using a standard curve.
Evaluation of the number of endometrial glands and the fibrotic area in rats. In each H&Estained section of rat endometrial tissue, four visual fields were randomly selected for analysis under a microscope (50×). The number of endometrial glands in each visual field was counted, and the mean was calculated. Four visual fields (100×) in each Masson's trichrome-stained section were also randomly selected for microscopic analysis. The area of endometrial interstitial fibrosis, the area of the endometrial stroma, and the glandular density in each visual field were calculated using Image-ProPlus 6.0. The fibrosis ratio was calculated as the mean of the area of endometrial interstitial fibrosis divided by the total area of the endometrial stroma and glands in each visual field. Statistical analysis. Statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC, USA).
Numerical variables were expressed as the mean ± SD. Data were compared using t-tests for count data and one-way analysis of variance (ANOVA) or Fisher's exact probability method for measurement data. Logistic regression analysis was used to determine the dominant variables for establishing IUA prediction models. A two-sided P < 0.05 was considered significant.

Results
Clinical baseline characteristics of the IUA and non-IUA groups. In total, 12,297 participants were included in this study including 908 patients with IUA (7.38%; the IUA group) and 11,389 healthy controls (92.62%; the non-IUA group). Clinical baseline parameters of the study population are presented in Table 1. Six cases (0.7%) in the IUA group were HPV positive, whereas none of the 11,389 cases in the non-IUA group had HPV infection. The proportion of patients with a large population of Lactobacilli was significantly higher in the IUA group than in the non-IUA group (39.6% vs. 0.2%, respectively; P = 0.0000). Compared to the non-IUA group, the IUA group had a higher proportion of cases with abnormal microbiota (56.4% vs. 30%, respectively; P = 0.0000) and poorer vaginal cleanliness (P = 0.0000). Interestingly, the proportion of cases with fungal infection was significantly higher in the IUA group than in the non-IUA group (18.2% vs. 9.1%, respectively; P = 0.0000), as was the proportion of cases with pH > 4.6 (43.2% vs. 3.2%, respectively; P = 0.0000). There was no significant difference in the dominant bacteria or acetylglucosaminidase between the two groups (P > 0.05).

Univariate and multivariate logistic regression analysis of factors influencing IUA occurrence.
To identify factors that influence IUA, we first conducted univariate logistic regression analysis ( General appearance of uterine samples after C. albicans perfusion in rats. The uterine samples from the C. albicans and control groups were not fully contracted on the 3rd day postoperatively. However, the inflammatory edema in the C. albicans group was slightly more severe than that in the control group (Fig. 1).

Changes in inflammation and fibrosis of the reproductive tract after C. albicans perfusion.
In the control group, on postoperative day 3, large areas of necrosis, coagulative necrosis, suppurative necrosis, bleeding foci, and local edema were observed. A small number of glandular structures and more collagen fibers could also be seen in local tissues ( Fig. 2A). On postoperative day 7, a small number of glands, a small number of www.nature.com/scientificreports/ neutrophils in the lamina propria, and more collagen fiber hyperplasia were observed (Fig. 2B). On day 14 after the procedure, there were no obvious glandular structures, but we did observe a small amount of cytoplasmic loss and nuclear pyknosis in the uterine luminal epithelium, scattered neutrophils in the lamina propria, and more collagen fibers (Fig. 2C).
In the C. albicans group, we observed a small number of glandular structures in local tissues, dilated glands, and neutrophils in some glandular cavities on postoperative day 3, as well as more neutrophil infiltration in the lamina propria, necrotic foci and eosinophilia in tissues, lamina propria edema, and multiple local hemorrhagic foci (Fig. 2D). On postoperative day 7, we observed a small number of glandular structures in the lamina propria, necrosis and nuclear fragmentation in some glandular epithelial cells, local uterine epithelial cytoplasmic vacuolation, a small number of lymphocytes, and more collagen fiber hyperplasia in the lamina propria (Fig. 2E). Furthermore, a hematoma had formed in the local tissue, which was surrounded by connective tissue, forming www.nature.com/scientificreports/ a capsule, and there was phagocytosis of hemosiderin-laden macrophages in the outer layer (Fig. 2E). On day 14 after the procedure, we observed inflammatory exfoliation and deletion of the uterine epithelium, exfoliated epithelial cells, necrotic tissue fragments, neutrophils, hyperemia, dilatation of the subepithelial capillaries, no obvious glandular structures, more inflammatory cell infiltration, and local fibroblast hyperplasia in the lamina propria (Fig. 2F).

Changes in inflammation and fibrosis of the reproductive tract after C. albicans perfusion.
In the control group, mild to moderate collagen fiber hyperplasia was observed on day 3 (Fig. 3A), and there was slight collagen fiber proliferation on days 7 (Fig. 3B) and 14 (Fig. 3C).  www.nature.com/scientificreports/ In the C. albicans group, severe collagen fiber hyperplasia was observed on postoperative day 3 (Fig. 3D). On day 7, extremely severe collagen fiber hyperplasia was observed, with some of the fibers forming strips (Fig. 3E). On day 14, there was severe collagen fiber proliferation, with some of the fibers forming strips (Fig. 3F).
The mean number of endometrial glands in the control and C. albicans groups was 36.230 ± 20.630 and 19.320 ± 12.780 /mm 2 , respectively. The endometrial gland density in the C. albicans group was significantly lower than that in the control group (P < 0.001, Table 3). On postoperative days 3, 7, and 14, the ratio of the endometrial fibrosis area to the the total endometrial area in the control group was 0.128 ± 0.058, 0.106 ± 0.038, and 0.069 ± 0.024, respectively; in the C. albicans group, it was 0.211 ± 0.142, 0.371 ± 0.059, and 0.413 ± 0.131, respectively. These results show that the degree of endometrial fibrosis stimulated by C. albicans in the IUA model rats was significantly greater than that in the control group at three different time points (postoperative days 3, 7, and 14) (P < 0.001; Table 3).
Expression of IL-6, TGF-β1, Smad 2, COLl, ER, and PR in the endometrium of IUA model rats. The expression levels of the IL-6, TGF-β1, Smad 2, COLl, ER and PR proteins in the endometrium, as assessed by IHC, were significantly higher in the C. albicans group than in the control group, whereas the levels of ER and PR were significantly lower in the C. albicans group than in the control group (Fig. 4, Table 4). The expression levels of IL-6 and TGF-β1 in the endometrium, as detected by ELISA, were found to be significantly higher in the C. albicans group than in the control group (Table 5).  Table 3. Ratio of endometrial fibrosis area to total area and endometrial gland density in the control and Candida albicans groups. Unless indicated otherwise, data are given as the mean ± SD. POD postoperative day.

Control group
Candida albicans group P-value

Discussion
The reproductive tract microbiome is an extensive and diverse system of pathogenic and non-pathogenic microorganisms that play an important role in female reproductive health 40,41 . In one study, endometrial samples from 46 patients with IUA and 21 infertile patients without IUA were analyzed. The results showed that the proportions of Klebsiella, Shewanella, and Lactobacillus at the genus level were higher in patients with IUA than those without IUA, whereas the proportion of Acinetobacter was significantly lower in patients with IUA 42 . Liu et al. 43    www.nature.com/scientificreports/ microbiological evaluation. In their study, patients with IUA had a significantly lower percentage of Firmicutes and a higher percentage of Actinobacteria at the phylum level than did women in the healthy vaginal secretion group 43 . At the genus level, nearly 50% of patients with IUA were found to have markedly reduced levels of the probiotic Lactobacillus, accompanied by an excess of pathogenic Gardnerella and Prevotella 43 . Their study also pointed out that patients with IUA lacked some normal bacteria, which may impair the endometrium's repair functions 43 . Through a retrospective analysis of a large clinical sample, we found that the fungal infection rate in patients with IUA was higher than that in healthy women and was accompanied by an increased vaginal pH, and fungal infection was further revealed to be an independent risk factor for IUA. It was also found that vaginal fungal infection in patients with IUA occurred simultaneously with an increase in Lactobacillus. This finding is in contrast with the results of a previous study which compared the vaginal microbiota of 50 patients with IUA and 30 healthy women and found that the amount of Lactobacillus was significantly decreased in approximately 50% of the IUA group 43 . It has been confirmed that vaginal fungal infections can be accompanied by an increase in pH 44 . We speculate that the reason for this increase may be that fungal infection elevates the pH of the vagina, which cannot be offset by Lactobacillus. It remains unclear whether this serves as a compensatory mechanism by which the vaginal microecosystem in patients with IUA can maintain a normal vaginal pH so as to resist fungal infection, and whether the increase in vaginal Lactobacillus in patients with IUA results from fungal infection. Further investigation is required to address these issues.
In recent years, fungal vaginitis has been considered to be an immunopathological reaction 45 . A study has shown that vaginal epithelial cells can distinguish colonized yeast from invasive hyphae by activating pathways such as the oxidative stress pathway 46 . Furthermore, invasive hyphae have been associated with cell damage 46 . It has also been shown that C. albicans with mycelial deletion strains can colonize as well as, or even better than, wild-type strains but cannot induce immunopathological markers or mucosal damage 46 . Fungal infection of the reproductive tract is not necessarily accompanied by clinical symptoms 47 . These results suggest that in fungal vaginitis, the transition from spores to hyphae or downstream mycelium-related effectors may be the key to tissue damage and subsequent immunopathological processes 47 .
It has been reported that C. albicans can cause a strong proinflammatory response, upregulate the expression of cytokines, and activate the nuclear factor (NF)-κB signaling pathway 48 . The increase in inflammatory cytokines actively promotes the occurrence and development of IUA 49 . The NF-κB pathway plays an important role in the production of cytokines and cell survival, and is an important part of the immune response, having been found to actively promote the pathogenesis of many inflammatory diseases 50 . Researchers have found significantly increased mRNA and protein levels of NF-κB in patients with IUA and in IUA rat models 51 . Through IHC studies, activation of the NF-κB signal in IUA has been further confirmed, and a close relationship has been detected between the NF-κB signaling pathway and the pathogenesis of IUA, as well as cytokines involved in the pathogenesis of IUA, such as TGF-β1, tumor necrosis factor (TNF)-α, and IL-6 51 . Research has also shown that the activation and expression of the NF-κB signaling pathway in IUA tissue plays an important role in the pathogenesis of the condition 51 .
IL-6 can promote fibrosis and activate the TGF-β/Smad pathway. The IL-6 receptor (IL-6R) is usually membrane bound, although it also exists in a soluble form (sIL-6R). When IL-6 binds to sIL-6R, a complex is formed which activates membrane-bound glycoprotein 130. This complex is constitutively expressed in most cell types. IL-6 binding to SIL-6R also leads to the activation of signal transducers and activators of transcription (STAT) signaling pathways, known as IL-6 trans-signals 48 . Importantly, unlike other soluble cytokine receptors, sIL-6R does not antagonize the activity of IL-6,instead, it has the opposite effect. While sIL-6R can be formed by limited proteolysis of membrane-bound receptors, it can also be secreted directly from cells through selective mRNA splicing 29 . Fibrotic diseases are characterized by the activation of fibroblasts. In the present study, we found that fibroblast activation was attributable to increases in the levels of the inflammatory cytokines TNF-α and IL-6. We also found that IL-6 upregulated the factors related to the TGF-β signaling pathway.
As a key mediator of tissue and organ fibrosis, TGF-β1 has a significant effect on fibroblast proliferation and the production of extracellular matrix in vitro 52 . A major process in TGF-β1 signal transduction is the recognition of Smad proteins 22 . Receptor-Smads (R-Smads,including Smad 2 and Smad 3) are the main downstream effectors of the TGF-β signaling pathway 53 . Considering the key role of IL-6 in the production of TGF-β1, we speculate that IL-6 may activate the Smad signaling pathway. After C. albicans perfusion, we observed that the expression of Smad 2 in endometrial sections was upregulated relative to that in the control group, which indicates that the activation of Smad 2 may be one of the intracellular mechanisms of IL-6-mediated endometrial fibrosis. Recent data have provided new evidence that the Smad protein is mainly activated by the classical TGF-β1 trigger mechanism 22 . Our results also show that IL-6-induced Smad activation was mediated by TGF-β1 in our rat fibrosis model. Table 5. Mean concentration of IL-6 and TGF-β1 in the endometrium of rats in the C. albicans and control groups, as determined by ELISA. ELISA enzyme-linked immunosorbent assay, IL-6 interleukin-6, TGF-β1 transforming growth factor-β1. *P < 0.05, **P < 0.01 compared with control. IL-6 (pg/mL) TGF-β1 (pg/mL) www.nature.com/scientificreports/ The current study has several limitations that should be noted. First, we identified a problem based on retrospective clinical study of a large sample, and then conducted preliminary verification through an animal experiment. We identified an C. albicans-TGF-β/Smad-IL-6 axis, but did not confirm its regulatory relationship through further experiments. The C. albicans was migrated from vagina to uterine cavity, but not colonized in the uterine cavity directly. Therefore, inflammation in the uterus may be affected by inflammation in the vagina, and the increase in inflammatory cytokines might be associated to the occurrence and development of IUA. A more effective and valuable animal model of IUA needs further exploration. Despite these limitations, this is the first study to explore the pathogenesis of IUA from the perspective of reproductive tract fungal infection. Our findings provide a new insight for understanding the pathogenesis of IUA and a new direction for the treatment and prevention of this condition. In future research, the regulation of C. albicans-TGF-β/Smad-IL-6 axis needs to be verified in vivo or in vitro.
In conclusion, this study has found that vaginal microecological disorders exist in patients with IUA, and that vaginal fungal infection may be one of the risk factors for IUA. C. albicans may cause a strong inflammatory response involving IL-6 upregulation and COL1 deposition, which aggravates endometrial fibrosis. C. albicans infection may therefore be related to IUA.

Data availability
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.