The investigation began with a systematic characterization of the fecal microbiome in mice with surgically induced endometriosis. Using 16S rDNA sequencing of the V3–V4 hypervariable region on an Illumina NovaSeq 6000 platform, the researchers profiled the gut microbiota of endometriosis and control mice during diestrus at day 42 post-model establishment. After quality filtering, 700 amplicon sequence variants were shared between groups, with 583 unique to controls and 536 unique to endometriosis mice. Alpha-diversity indices, including Chao1, Shannon, Pielou’s evenness, and Good’s coverage, did not differ significantly between groups, suggesting that overall species richness and evenness remained intact. However, beta-diversity analysis revealed clear structural separation between groups, confirmed by unweighted UniFrac-based principal coordinates analysis, hierarchical clustering, and analysis of similarities. At the phylum level, endometriosis mice showed reduced Firmicutes abundance and increased Bacteroidota, with a significantly altered Firmicutes-to-Bacteroidota ratio compared to controls. Linear discriminant analysis effect size analysis further delineated taxon-level compositional differences, and indicator species analysis nominated g_Ruminococcus, g_Coprococcus, g_Muribaculum, and g_Duncaniella as candidate endometriosis-associated biomarkers. Functional prediction using PICRUSt2 against the KEGG database identified enrichment of sphingolipid biosynthesis and metabolism pathways in endometriosis mice, hinting that lipid metabolism might be a key functional consequence of the dysbiosis.

To translate these compositional shifts into metabolic consequences, the team performed non-targeted liquid chromatography-mass spectrometry metabolomics on peritoneal lavage fluid from endometriosis and control mice. Partial least-squares discriminant analysis revealed clear separation between the metabolic profiles of the two groups, and a 200-permutation test confirmed model robustness without overfitting. Applying thresholds of fold change greater than or equal to 1.2 or less than or equal to 1/1.2, p less than 0.05, and variable importance in projection greater than or equal to 1, the researchers identified 26 upregulated and 2 downregulated metabolites in the endometriosis group. Among these, LysoPC(16:0) exhibited one of the largest fold-changes, highlighted in a heatmap of the top 28 differential metabolites. KEGG pathway enrichment of the full differential metabolite set identified glycerophospholipid metabolism, choline metabolism, linoleic acid metabolism, and unsaturated fatty acid biosynthesis among the top 20 enriched pathways, placing LysoPC(16:0) within a broader lipid metabolic reprogramming of the peritoneal cavity. Spearman’s correlation analysis between significantly altered genera and differential metabolites revealed that multiple genera, including g_Muribaculum, g_Mucispirillum, g_Ruminococcus, and g_Lactiplantibacillus, showed significant positive correlations with peritoneal LysoPC(16:0), suggesting a potential microbiota-metabolite link that demanded causal testing.

The researchers then turned to fecal microbiota transplantation to establish causality. Mice were rendered pseudo-germ-free by a seven-day antibiotic cocktail regimen containing vancomycin, ampicillin, metronidazole, and neomycin sulfate before endometriosis model establishment. Beginning on day 1 after model establishment and repeated every three days, mice were gavaged with sterile phosphate-buffered saline, fecal suspension from control mice, or fecal suspension from endometriosis mice. Peritoneal LysoPC(16:0) was significantly higher in recipients of endometriosis-donor feces than in either control group. These mice also developed heavier lesions and showed a greater proportion of CD206-positive macrophages, a canonical M2 marker, in peritoneal lavage fluid compared with control recipients. These results demonstrated that endometriosis-associated gut microbiota is sufficient to elevate peritoneal LysoPC(16:0) and promote peritoneal M2 macrophage accumulation, thereby facilitating lesion growth. The finding represents a critical advance because it moves beyond correlation to demonstrate that the altered microbiota itself carries the pro-endometriotic signal, with the lipid metabolite serving as the functional mediator between gut dysbiosis and pelvic immune remodeling.

To dissect the cellular mechanism, the team isolated bone marrow-derived macrophages from the femurs and tibias of eight-week-old C57BL/6N mice, cultured them with macrophage colony-stimulating factor for seven days, and stimulated them under multiple conditions. Flow cytometry showed that 50 micromolar LysoPC(16:0) significantly augmented IL-4-induced CD206 expression, confirming that the metabolite promotes M2 polarization in vitro. In a Transwell co-culture assay, mouse endometrial stromal cells placed in the upper chamber migrated significantly more when co-cultured with IL-4 plus LysoPC(16:0)-stimulated macrophages in the lower chamber than with either negative control or IL-4-alone conditions. Additionally, a phagocytosis assay using phycoerythrin-conjugated microspheres demonstrated that LysoPC(16:0)-treated macrophages exhibited enhanced particle uptake over two hours, consistent with prior reports that lysophosphatidylcholine promotes macrophage phagocytic activity. These in vitro findings established that LysoPC(16:0) directly reshapes macrophage phenotype and function, with downstream consequences for endometrial stromal cell migration, a pivotal step in ectopic lesion formation.

In vivo, the researchers administered daily intraperitoneal injections of LysoPC(16:0) to endometriosis mice for two weeks. Mice treated with the metabolite developed larger lesions than vehicle controls, whereas mice treated with AACOCF3, a phospholipase A2 inhibitor that suppresses endogenous LysoPC(16:0) production, showed smaller lesions and a lower proportion of peritoneal CD206-positive macrophages relative to endometriosis controls. To determine whether PPARγ signaling underlies the M2-polarizing activity of LysoPC(16:0), the team used GW9662, a selective PPARγ antagonist. In bone marrow-derived macrophages, LysoPC(16:0) significantly augmented IL-4-induced CD206-positive polarization, and this was substantially reduced by GW9662 pretreatment. In vivo, LysoPC(16:0)-t mice developed heavier lesions and showed a higher proportion of peritoneal CD206-positive macrophages than vehicle controls, and GW9662 co-administration significantly reduced both parameters. To assess whether these findings extend beyond a single genetic background, the researchers repeated the experiment in BALB/c mice using a four-group design, and the directional results were consistent with those in C57BL/6N mice across all outcome measures, confirming that the LysoPC(16:0)–PPARγ–M2 axis is not strain-specific but operates across genetically distinct backgrounds.

A key mechanistic advance of the study is the identification of IL-10 as the downstream effector through which LysoPC(16:0)-polarized M2 macrophages promote lesion progression. Reverse transcription quantitative polymerase chain reaction demonstrated that LysoPC(16:0) significantly upregulated Il10 messenger RNA in IL-4-primed macrophages in a PPARγ-dependent manner, placing IL-10 transcription within the LysoPC(16:0)–PPARγ cascade. The functional importance of IL-10 was confirmed in vivo: neutralizing IL-10 in LysoPC(16:0)-supplemented endometriosis mice with an anti-mouse IL-10 neutralizing antibody significantly reduced lesion weight compared with isotype controls. To formally establish that peritoneal macrophages are causally required rather than incidental bystanders, the researchers co-administered clodronate liposomes to deplete macrophages in LysoPC(16:0)-treated endometriosis mice. At day 14, clodronate-treated mice showed markedly fewer F4/80-positive cells in peritoneal lavage fluid and significantly lower lesion weights than empty liposome controls, despite equivalent LysoPC(16:0) exposure. Macrophage depletion therefore abolished the pro-endometriotic effect of LysoPC(16:0), demonstrating that peritoneal macrophages are required mediators of the metabolite-driven lesion growth.

To provide preliminary cell-based evidence relevant to human endometriosis, the researchers employed two complementary human models. THP-1 monocytes were differentiated into macrophages by phorbol 12-myristate 13-acetate treatment, rested for 24 hours, and stimulated under four conditions: negative control, IL-4/IL-13, IL-4/IL-13 plus LysoPC(16:0), and IL-4/IL-13 plus LysoPC(16:0) plus GW9662. LysoPC(16:0) significantly increased the proportion of CD206-positive cells compared with IL-4/IL-13 alone, and GW9662 reversed this effect, recapitulating the PPARγ-dependent polarization observed in murine macrophages. Conditioned medium was then collected from THP-1 macrophages stimulated under six conditions, including IL-10 neutralization with an anti-human IL-10 antibody or a human IgG1 isotype control. Primary endometrial stromal cells were isolated from ectopic lesions of five patients with surgically and histologically confirmed endometriosis, aged 26 to 37 years with a median age of 33, all with rAFS stage III–IV disease. Conditioned medium from LysoPC(16:0)-stimulated macrophages promoted significantly greater stromal cell migration than conditioned medium from IL-4/IL-13-alone conditions across all five patient samples, and this enhancement was substantially reduced by GW9662 or IL-10 neutralization but not by the isotype control. The consistency of the conditioned medium effect across five independent patient specimens, each tested under all six conditions in parallel, provides preliminary evidence that the axis identified in murine models is also operative in human cells.

The researchers acknowledge several limitations that place practical bounds on these conclusions. The bacterial taxa ultimately responsible for LysoPC(16:0) production were not isolated at the species level, as several candidate genera are obligate anaerobes whose cultivation requires dedicated infrastructure. Confirming a direct strain-metabolite relationship will require single-strain gnotobiotic colonization. The link between LysoPC(16:0) and PPARγ was established pharmacologically with GW9662, consistent with earlier reports that structurally related lysophospholipids can act as PPARγ ligands, but direct biophysical evidence of binding, such as co-crystallography or surface plasmon resonance, was not obtained. Likewise, IL-10 was identified as the principal downstream effector, but transforming growth factor beta and other M2-associated cytokines were not systematically profiled, and how IL-10 signaling drives endometrial stromal cell migration, whether through JAK-STAT3, PI3K-AKT, or another route, remains to be worked out. The translational component relied on a monocyte-derived cell line and primary stromal cells from five patients, a cohort that cannot capture the clinical heterogeneity of endometriosis across disease stage, lesion subtype, or symptom severity. Despite these limitations, the findings carry potential clinical relevance that warrants further exploration. Elevated peritoneal LysoPC(16:0) and altered microbiota composition could be evaluated as candidate biomarkers for endometriosis in future clinical studies, and the phospholipase A2–LysoPC(16:0)–PPARγ–IL-10 cascade offers multiple candidate intervention points for preclinical exploration. As the work is primarily based on murine models and cell-based human systems, its findings should be interpreted as preclinical mechanistic evidence, with clinical applicability to be established in future patient studies.

Subject of Research: Intestinal microbiota-associated LysoPC(16:0) promotes endometriosis via PPARγ-mediated M2 macrophage polarization and IL-10 secretion

Article Title: Intestinal microbiota-associated LysoPC(16:0) promotes endometriosis via PPARγ-mediated M2 macrophage polarization and IL-10 secretion

Article References: Li, N., Yuan, M., Jin, X., Jiao, X., Wang, S., Zhong, Z., Guan, H., Le, X., Ren, Q., Liu, Y., Dong, X., Li, Q., Wang, M., & Wang, G. (2026). Intestinal microbiota-associated LysoPC(16:0) promotes endometriosis via PPARγ-mediated M2 macrophage polarization and IL-10 secretion. iScience, 29(11), Article 117917. https://doi.org/10.1016/j.isci.2026.117917

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117917

Keywords: Intestinal, microbiota-associated, LysoPC, promotes, endometriosis, PPAR, mediated, macrophage, polarization, IL-10, secretion, scientific research