Fusobacterium nucleatum is not significantly present in eutopic endometrium from patients with minimal-mild and moderate-severe endometriosis

Genomic DNA was extracted from 40 endometrial biopsies, comprising 20 samples from patients with minimal-mild and moderate-severe endometriosis and 20 samples from controls. The relative abundance of the bacterial genus Fusobacterium was measured using the ΔCt method, which calculates the difference between the mean Ct value obtained with Eubacteria 16S primers and the mean Ct value obtained with Fusobacterium spp. primers for each sample through quantitative PCR. Out of the 40 endometrial biopsies analyzed, only two endometriosis samples ( Figure 1A ) and three control samples ( Figure 1B ) showed positive detection of Fusobacterium spp., based on the difference in Ct values compared to blank controls. In the control group, Fusobacterium spp. had a mean relative value of 703.5 ± 2,214, while the endometriosis group exhibited a mean relative value of 728.0 ± 3,227 ( Figure 1C ). A non-parametric Mann-Whitney test was applied to assess differences between the two groups, resulting in a p-value of 0.6980. These findings suggest no significant difference in the presence of Fusobacterium spp. in the eutopic endometrium of patients with endometriosis compared to controls. To further explore the potential association between Fusobacterium and endometriosis, we conducted genus- and species-specific detection of Fusobacterium nucleatum using an extended cohort of 92 endometrial biopsies (55 from patients with endometriosis and 37 from control subjects). The methodology and primers used for detection are described in the Materials and Methods section. Relative abundance was calculated using the ΔCt method, comparing the mean Ct values obtained with Eubacteria 16S primers to those obtained with either F. nucleatum or Fusobacterium spp. primers for each sample. This approach allowed for the measurement of the relative abundance at both the species level and the genus level ( Figure 2A and B ). At the species level, the mean relative abundance of F. nucleatum in endometriosis cases was 0.3951 ± 0.9897, compared to 0.1914 ± 0.2942 in controls, with no significant difference observed ( p = 0.7381) ( Figure 2C ). At the genus level, the mean relative abundance of Fusobacterium spp. was 0.0011 ± 0.0023 in cases and 0.0016 ± 0.0053 in controls, with no significant difference between the groups ( p = 0.2582) ( Figure 2D ). To investigate whether Fusobacterium was associated with the severity of endometriosis, we performed a subgroup analysis comparing MMD versus MSD patients. The mean relative abundance of Fusobacterium spp. in the MMD group was 0.0010 ± 0.0024, compared to 0.0012 ± 0.0016 in the MSD group ( p = 0.1465, Mann-Whitney test) ( Figure 3A ). At the species level, the mean relative abundance of F. nucleatum in the MMD group was 0.4622 ± 1.119, compared to 0.1785 ± 0.2450 in the MSD group ( p = 0.2936) ( Figure 3B ). Overall, these results demonstrate no significant association between the presence of F. nucleatum at the species or genus level and the occurrence or severity of endometriosis.

This study analyzed a total of 92 eutopic endometrial samples from 55 patients undergoing laparoscopic diagnosis of endometriosis, classified as minimal-mild disease (MMD) (n=42) or moderate-severe disease (MSD) (n=13) based on the rASRM classification of endometriosis ( 4 ) Additionally, 37 endometrial samples were obtained from controls undergoing surgery for benign conditions ( Supplemental Figure 1 ). Fusobacterium spp. genus was investigated following the methodology described by Muraoka et al. ( 22 ) in a randomly selected subset of 40 samples (20 patients with endometriosis and 20 controls). Fusobacterium nucleatum species was assessed across the entire cohort using quantitative polymerase chain reaction (qPCR). A total of 92 patients, recruited between July, 2016 and August, 2019 at 5 sites of the Reproductive Medicine Network: Augusta University, Penn State University, University of Pennsylvania, Wayne State University and Yale University, were included in this study. Endometrial samples were obtained at the time of scheduled gynecologic endoscopic procedures due to either endometriosis (cases) or other benign conditions such as fibroids, ovarian cysts, menorrhagia, or chronic pelvic pain (controls). The participants were aged 18 to 44 years, and those who had taken prescribed antibiotics within two months prior to sample collection were excluded to avoid potential interference with the analysis. Epidemiological and clinical data were collected and are compared in Table 1 to ensure the absence of selection bias. The study was approved by the University of Pennsylvania Institutional Review Board (register number 821891). Informed consent was obtained from all participants prior to sample collection. Endometrial tissue samples were collected using a standardized ‘no-touch’ technique to minimize contamination from the lower genital tract. With the patient in the lithotomy position, the vagina and cervix were first cleaned using a dry sterile cotton swab to remove mucus and debris. A disinfected speculum was then introduced, and a Pipelle catheter was inserted into the uterine cavity without contacting the vaginal walls. Suction was applied while gently rotating the Pipelle to aspirate endometrial tissue in a spiral motion from the fundus to the mid-uterine segment. After collection, the external surface of the Pipelle was cleaned with sterile gauze, and the biopsy sample (approximately 50–70 mg) was transferred to a 1.5 mL conical tube containing RNA stabilization solution. The sample was then placed in cryovials, snap frozen in liquid nitrogen within 15 minutes, and transferred to –80 °C freezers for long-term storage. All personnel involved in the sampling procedure were trained at an in-person session to ensure protocol consistency and sample integrity. For detecting Fusobacterium spp., total DNA from eutopic endometria was extracted using the QIAamp DNA Microbiome Kit (Qiagen), following the protocol used by Muraoka et al. ( 22 ). Host DNA depletion was achieved by incubating samples with benzonase at 37°C for 30 minutes. For detecting F. nucleatum and Fusobacterium spp., total DNA from endometrial biopsies was extracted using the IndiSpin ® Pathogen Kit (Indical Bioscience). Briefly, 25 mg of endometrial tissue was cut into small pieces and treated with proteinase K at 56°C for 3 hours under agitation or until completely lysed. The samples were then mixed with ATL buffer (Qiagen) and disrupted mechanically using a TissueLyser LT (Qiagen) at 50 Hz for 5 minutes with stainless-steel beads. After these pretreatment steps, bacterial nucleic acids were purified according to the manufacturer’s instructions. The specificity of primers for the qPCR assays was verified using genomic DNA from Fusobacterium nucleatum (DSM 15643) and a selection of other bacterial species available in our laboratory: Enterococcus faecalis (DSM 20380) , Neisseria gonorrhoeae (DSM 9188) , Chlamydia trachomatis (DSM 19102) , Escherichia coli (DSM 301) , Lacticaseibacillus rhamnosus (DSM 20021) , Staphylococcus epidermidis (DSM 1798), Klebsiella pneumoniae (DSM 30104), Streptococcus agalactiae (DSM 2134) , Staphylococcus haemolyticus (DSM 20228) , Staphylococcus hominis (DSM 20329) , Streptococcus gallolyticus (DSM 16831) , Lactobacillus iners (DSM 13335) , Lactobacillus crispatus (DSM 20584) and Metamycoplasma hominis (DSMZ 19104). All of them were sourced from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ) collection. The limit of detection of F. nucleatum was determined using serial dilutions with concentrations ranging from 10 ng/μL to 1 fg/μL. Primers for Fusobacterium spp. and Eubacteria 16S were employed following Muraoka et al. ( 22 ) ( Supplemental Table 1 ). For species-specific F. nucleatum primers, a thorough literature search identified candidates with the highest alignment accuracy to the bacterial genome. Three pairs of primers were tested, and the pair demonstrating the highest specificity (F. nucleatum 3) was selected for further use ( Supplemental Table 1 ). The primers were evaluated using Basic Local Alignment Search Tool (BLAST) searches on the NCBI genomic database. Primer specificity was confirmed through qPCR using F. nucleatum DNA as a template. The resulting Ct values were compared with those obtained when amplifying the DNA of the other previously mentioned bacteria (see “ Bacterial DNA template ” section) or using a non-template negative control. As expected, F. nucleatum DNA exhibited lower Ct values indicative of primer specificity, while higher Ct values were observed for negative controls ( Supplemental Figure 2A ). The detection limit was validated using pure F. nucleatum DNA in decreasing amounts, from 10 ng to 1 fg ( Supplemental Figure 2B ), confirming specific amplification at femtogram levels. To further validate the ability of the selected primers to specifically detect F. nucleatum in biological samples, qPCR was performed on saliva and endometrial fluid samples in which the presence of this bacterium had been previously confirmed by 16S rRNA sequencing ( 23 ). The results confirmed the specificity of the primers and showed lower relative detection in endometrial samples, likely due to their low microbial biomass. Negative controls, including two extraction blanks and RNAse-free ultrapure water, yielded no amplification, while pure genomic DNA from F. nucleatum served as a positive control ( Supplemental Figure 3 ). Quantitative PCR reactions were performed on a QuantStudio5 (Applied Biosystems by Thermo Fisher Scientific). Each reaction contained 20–40 ng of DNA for Fusobacterium spp. or 100–200 ng for F. nucleatum , 10 μmol/L forward and reverse primers, and 5 μL of KAPA SYBR FAST qPCR Master Mix (2X) (Merck), which included an integrated hot start, SYBR Green I dye, MgCl2, dNTPs, and stabilizers, in a total reaction volume of 10 μL. The cycling conditions were: initial denaturation: 50°C for 2 minutes, followed by 95°C for 10 minutes; amplification: 40 cycles of 15 seconds at 95°C and 1 minute at 60°C; melting curve analysis: 15 seconds at 95°C, 1 minute at 60°C, and 15 seconds at 95°C. Each experiment included negative (no DNA) and positive controls (commercial purified bacterial DNA) to detect potential contamination or non-specific amplification. A melting curve analysis distinguished specific from non-specific PCR products. All primers were purchased from Integrated DNA Technologies (Coralville, IA). Numerical variables were described using means and standard deviations. Comparisons between two non-normally distributed groups were performed using the Mann-Whitney test, while comparisons between more than two groups used the Kruskal-Wallis Rank Sum Test. Pairwise comparisons were conducted using Dunn’s multiple comparisons test when significant differences were detected. Categorical variables were compared using Fisher’s exact test and presented as counts (n) and percentages. Graphs and statistical analyses were generated using Microsoft Excel, GraphPad, and R. The statistical significance thresholds were defined as follows: non-significant (NS) for P > 0.05, * P < 0.05, ** P < 0.01, and *** P < 0.001. Each figure legend includes the specific statistical methods applied.

Various reports have suggested that antibiotics may be useful in the treatment of endometriosis. In a mouse model, treatment with broad-spectrum antibiotics (vancomycin, neomycin, metronidazole and ampicillin) reduced endometriotic lesion growth, proliferation and inflammation in the gut compared to vehicle-treated mice ( 24 ). This reduction was observed regardless of whether treatment commenced before or after lesion formation. Notably, mice treated exclusively with metronidazole, but not those receiving neomycin, also exhibited a significant reduction in ectopic lesion volume, suggesting a potential role of gut bacteria in modulating endometriosis progression ( 24 ). Also, anecdotical clinical reports suggest that antibiotics may play a role in alleviating clinical symptoms and endometriosis extension for some patients ( 25 ). Species within the Fusobacterium genus are common members of the oral and gastrointestinal microbiota, generally maintaining a symbiotic relationship with their host. Although these bacteria were traditionally considered opportunistic pathogens, recent evidence has demonstrated that certain species, such as Fusobacterium nucleatum , play a critical role not only in periodontitis but also in carcinogenesis ( 20 ). F. nucleatum has been linked to gastric and colorectal cancers through the induction of pro-inflammatory cytokines, including interleukin (IL)-6, IL-8, and tumor necrosis factor (TNF) in human colonoid cultures ( 26 – 28 ). The potential involvement of F. nucleatum in the pathogenesis of endometriosis has garnered significant attention following the study by Muraoka et al. ( 22 ) which reported a notable association between the presence and abundance of Fusobacterium spp. and endometriosis. Their findings raised the possibility of a role for this bacterium in disease development. Our study aimed to replicate and expand on these results using targeted qPCR with a larger sample size and refined methodologies. However, we found no significant differences in the abundance of Fusobacterium spp. or F. nucleatum between the eutopic endometrium of women with and without endometriosis. One limitation of the Muraoka et al. ( 22 ) study was their analysis at the genus level, which used primers that may have lacked specificity, potentially leading to non-specific amplification of Fusobacterium DNA. To address this, we employed species-specific primers targeting F. nucleatum , thereby improving detection accuracy. Additionally, while Muraoka et al. ( 22 ) analyzed a smaller cohort of 20 cases and 20 controls, our study included 55 cases and 37 controls, enhancing statistical power to detect potential associations. Several alternative explanations may account for the differences in findings. One key factor could be the differences in patient cohorts. The Muraoka et al.( 22 ) study exclusively involved Japanese patients with ovarian endometriosis, whereas our study included American patients with diverse endometriosis phenotypes, including peritoneal, ovarian, deep infiltrating, and mixed forms. This heterogeneity may have diluted any specific association between Fusobacterium spp. and ovarian endometriosis. Furthermore, only 7.3% of our patients presented with ovarian endometriomas, compared to 100% in the Muraoka cohort( 22 ). This disparity could significantly impact bacterial colonization patterns, as ovarian endometriomas may provide a more favorable environment for F. nucleatum proliferation. Additionally, the ethnic composition of the two cohorts differed markedly. While all participants in the Muraoka study were of Asian ethnicity ( 22 ), our cohort was ethnically diverse, including Asian, Black, White, mixed-race, and other individuals. These demographic differences could influence microbiome composition and may further contribute to the discrepancy in findings between studies. Another factor that complicates cross-study comparisons is the variation in disease severity. The Muraoka cohort exclusively included patients with advanced-stage endometriosis (stages III-IV) ( 22 ), while most of our patients (76.4%) had minimal-mild disease, with only 23.6% classified as moderate-severe. This discrepancy may influence the extent of F. nucleatum colonization, which could be more prominent in advanced disease stages. For all these reasons, our results are not necessarily contradictory to those reported by Muraoka et al., as their findings were specific to patients with ovarian endometriomas, a subgroup underrepresented in our cohort. Rather, our data suggest that the association with F. nucleatum may not extend to all clinical subtypes of endometriosis, highlighting the need for phenotype-specific analyses. Another important difference lies in the type of sample analyzed. Muraoka et al. included both eutopic and ectopic endometrial samples ( 22 ), whereas our study was restricted to eutopic endometrial tissue. Ectopic samples may provide more direct insights into bacterial colonization of endometriotic lesions, which we were unable to assess. However, it is worth noting that Muraoka et al. reported significant differences in Fusobacterium presence and abundance even when analyzing only eutopic endometrial tissue. Using fluorescence in situ hybridization (FISH), they observed Fusobacterium infiltration in 64% of endometriosis cases versus 7% of controls, and these results were further supported by qPCR showing increased relative bacterial abundance in eutopic endometria from cases compared to controls ( 22 ). These findings suggest a potential association between Fusobacterium presence/abundance and endometriosis within the same tissue type analyzed in our study. In contrast, we did not replicate this observation, underscoring the need for further research across diverse cohorts and with standardized methodologies. It is also important to highlight that, although a significant difference in menstrual cycle phase was observed between patients with endometriosis and controls—specifically in the proportion of individuals who were menstruating at the time of biopsy—this is unlikely to have influenced our findings. Hormonal modulation of Fusobacterium has been primarily associated with elevated estrogen and progesterone levels, which are lowest during menstruation. Furthermore, we did not detect significant differences in Fusobacterium abundance between groups, suggesting that the imbalance in menstrual status does not account for the negative findings of our analysis. Our qPCR methodology provided reliable and sensitive detection of bacterial DNA in endometrial tissues. However, future studies could benefit from complementary techniques such as FISH or metagenomic sequencing. These methods could offer insights into the spatial distribution of F. nucleatum and its interactions within the broader microbial ecosystem. Taken together our findings suggest that the reported association between F. nucleatum and endometriosis may be context-dependent, influenced by specific phenotypes, disease stages, or environmental factors that were not fully captured in our study.

In conclusion, while our study does not support any association between F. nucleatum and endometriosis, our findings underscore the importance of methodological rigor, appropriate sample selection, and patient heterogeneity when investigating the microbiome’s role in complex diseases. The discrepancies between our results and those of Muraoka et al.( 22 ) emphasizes the need for further research to elucidate the potential role of Fusobacterium spp. in endometriosis, particularly in relation to disease progression and subtype-specific differences. This will be essential to assess its potential utility as a diagnostic or prognostic biomarker.

Endometriosis is a chronic inflammatory disease of unknown origin, characterized by the presence of endometrial-like glands and stroma outside the uterus, leading to inflammation, scarring, and adhesions ( 1 ). This condition affects approximately 10% of women of reproductive age and 30–50% of those experiencing infertility ( 2 ). Clinically, endometriosis can manifest in three distinct phenotypes: superficial peritoneal, ovarian, and deep infiltrating endometriosis (DIE) ( 3 ). To account for the wide range of presentations of endometriosis, the American Fertility Society (AFS) established a four-stage classification system. This system was later revised and adopted as the revised American Society for Reproductive Medicine (rASRM) classification in 1997 ( 4 ), which scores endometriosis based on the extent of the disease and the location of the lesions. The pathogenesis of endometriosis remains poorly understood, involving a multifactorial interplay of genetic, hormonal, and environmental influences ( 5 ). However, recent studies have proposed a potential microbial origin ( 6 – 8 ).The human endometrial microbiome plays a critical role in reproductive and gynecological health. A microbiome dominated by Lactobacillus is generally associated with favorable reproductive outcomes, while dysbiosis has been linked to infertility ( 9 – 11 ), pregnancy complications such as preeclampsia ( 12 – 14 ), and an increased risk of endometrial cancer ( 15 ). Emerging evidence suggests that microbial imbalances may be a key factor in the development and progression of endometriosis. Microbiome-based models can differentiate between individuals with and without endometriosis ( 16 ). Specifically, endometrial samples from women with endometriosis showed an increased presence of Gardnerella, α-Streptococci, Enterococci and Escherichia coli compared to those from control women ( 17 ). Similarly, 16S metagenomic analyses detected a significantly higher percentage of Streptococcaceae and Staphylococcaceae in the cystic fluid of women with ovarian endometriomas compared to fluid from non-endometriotic cysts ( 18 ). In addition, a potential link between chronic endometritis and endometriosis has been suggested ( 19 ). Fusobacterium , a genus of anaerobic gram-negative bacteria implicated in inflammatory diseases, has emerged as a potential contributor to the pathogenesis of endometriosis ( 20 ). Different reports have implicated F. nucleatum in various gynecological conditions, including polycystic ovary syndrome (PCOS), endometriosis, and pelvic inflammatory disease (PID) ( 21 ).A recent study from Japan detected Fusobacterium in the eutopic endometrium of 64% of patients with ovarian endometriosis, compared to only 7% in control subjects ( 22 ). Moreover, Fusobacterium was shown to induce myofibroblast activation and promote endometrial lesion proliferation in a murine model, an effect mitigated by antibiotic treatment ( 22 ) suggesting a potential role for Fusobacterium in the growth and maintenance of endometriotic lesions. Given the relevance of these findings, this study aims to investigate the presence of Fusobacterium spp., particularly F. nucleatum , in the eutopic endometrium of patients with minimal-mild to moderate-severe endometriosis classified according to the rASRM staging system using targeted quantitative PCR (qPCR).