GLP-1R Expression in Ectopic Endometrium: NF-κB Inhibition, Prostaglandin Reduction, and Anti-Inflammatory Mechanisms

[Image: Immunohistochemistry panel showing GLP-1R expression in eutopic and ectopic endometrial tissue samples]

Immunohistochemical and RT-PCR analysis confirmed GLP-1R protein and mRNA expression in human endometrial stromal cells, epithelial cells, and endothelial cells in both eutopic (normally positioned) and ectopic endometrial tissue samples. The UK group (2025) demonstrated GLP-1R expression in 23 of 26 ectopic lesion samples, with expression levels comparable to eutopic tissue in 19 samples. The Korean group confirmed receptor expression and additionally characterized GLP-1R co-localization with the NF-κB p65 subunit in ectopic lesion stromal cells, providing the first immunohistochemical evidence linking GLP-1R distribution to the dominant pro-inflammatory transcription factor in endometriosis.

GLP-1R is a Gs-coupled class B GPCR; agonist binding in endometrial stromal cells activates adenylyl cyclase, increasing intracellular cAMP and activating protein kinase A (PKA). Elevated cAMP/PKA signaling exerts inhibitory effects on IκB kinase (IKK) complex activity — specifically phosphorylation of IKKβ — preventing IκB degradation and thereby maintaining IκB-mediated sequestration of NF-κB in the cytoplasm. This is the molecular link between GLP-1R activation and NF-κB suppression in endometrial tissue.

The clinical significance of receptor expression in ectopic versus eutopic tissue is that pharmacological GLP-1R engagement would be expected to affect both compartments. Differential expression levels or signaling coupling efficiency between eutopic and ectopic tissue could theoretically produce differential effects — potentially greater suppression of inflammatory gene expression in the ectopic compartment if ectopic lesions show altered downstream signaling compared to eutopic tissue. This remains to be characterized in human studies.

[Image: NF-κB to COX-2 to PGE2 pathway diagram with GLP-1R and NSAID intervention points marked]

NF-κB (nuclear factor kappa-light-chain-enhancer of activated B cells) drives the expression of COX-2 (cyclooxygenase-2, encoded by PTGS2) in endometrial stromal cells. COX-2 catalyzes the conversion of arachidonic acid to prostaglandin H2 (PGH2), which is subsequently converted by PGE2 synthase (PTGES/mPGES-1) to prostaglandin E2 (PGE2). In endometriosis, this pathway is pathologically upregulated: ectopic lesions express COX-2 at 5-10-fold higher levels than eutopic endometrium from women without endometriosis, driven by IL-1β, TNF-α, and estrogen-mediated NF-κB activation.

PGE2 exerts multiple pathological effects in endometriosis through cognate EP receptors (EP1-4): EP2/EP4-mediated cAMP elevation in macrophages and NK cells suppresses peritoneal immune surveillance (a critical immune evasion mechanism); EP2/EP4 signaling in stromal cells stimulates aromatase (CYP19A1) expression, creating a local estrogen amplification loop within lesions independent of systemic estrogen levels; and EP1/EP3 signaling mediates pain sensitization through afferent sensory neuron activation. GLP-1R-driven NF-κB suppression reduces COX-2 transcription, thereby reducing PGE2 synthesis across all of these downstream pathological effects simultaneously.

The mechanism is upstream of but distinct from NSAID action: NSAIDs inhibit COX-2 catalytic activity (irreversible for aspirin, competitive-reversible for ibuprofen/naproxen) without affecting NF-κB transcription. GLP-1RA suppression of NF-κB reduces COX-2 protein expression itself, which would theoretically produce a more durable effect per treatment cycle. Additionally, NF-κB inhibition would reduce not only COX-2 but other NF-κB target genes including IL-6, VEGF, and MMP-9, all of which contribute to endometriosis lesion angiogenesis, invasion, and persistence.

[Image: Bar chart of mouse endometriosis model results showing lesion area, IL-1β, and PGE2 reductions with semaglutide]

The 2025 murine studies used a standard syngeneic endometriosis model: uterine fragments from donor mice were transplanted to the peritoneal surface of recipient mice, and after lesion establishment, semaglutide was administered at weight-adjusted doses comparable to human therapeutic levels. Endpoint measurements at four weeks included lesion area by planimetry, peritoneal lavage fluid cytokine quantification by multiplex ELISA, and immunohistochemical assessment of lesion vascularity and proliferative activity.

Key findings: lesion area was reduced 34% (p < 0.01) compared to vehicle control. Peritoneal IL-1β, one of the dominant cytokines responsible for peritoneal immune dysregulation and NK cell suppression in endometriosis, was reduced >50%. Peritoneal PGE2 was reduced >50%, consistent with the expected downstream consequence of NF-κB/COX-2 pathway inhibition. Lesion microvessel density was reduced 22%, suggesting attenuation of VEGF-driven angiogenesis — consistent with NF-κB-mediated VEGF downregulation. Proliferating cell nuclear antigen (PCNA) staining in lesion stromal cells showed a 28% reduction, indicating reduced lesion proliferative activity.

Critical limitations of the murine model: (1) Mice do not spontaneously menstruate; surgically induced ectopic lesion behavior differs from menstruation-driven retrograde implantation. (2) Murine peritoneal immune environment differs substantially from human — NK cell and macrophage proportions, cytokine repertoire, and complement activity are not directly translatable. (3) Semaglutide dosing in rodents required to achieve comparable GLP-1R occupancy may exceed what achieves equivalent receptor engagement in humans per unit body weight. Animal-to-human translational attrition of effect size is historically substantial — typical 40-60% attrition is common in inflammatory disease models.

[Image: Endometriosis lesion autocrine estrogen-prostaglandin loop with peripheral VAT aromatase contribution pathway]

Endometriosis is stringently estrogen-dependent at the lesion level through several mechanisms. Ectopic lesions express ERα and ERβ, with estrogen driving lesion cell proliferation, survival, and resistance to apoptosis. Critically, lesions also contain aromatase (CYP19A1) at high levels — substantially higher than in eutopic endometrium — creating a local autocrine estrogen production loop: PGE2 (via EP2 receptors) upregulates CYP19A1 expression in lesion stromal cells, and locally produced estrogen (predominantly estradiol from lesion aromatase) drives further PGE2 production through an ER-mediated COX-2 induction. This constitutes a self-sustaining, estrogen-prostaglandin amplification circuit within individual lesions.

GLP-1RA-mediated VAT reduction attenuates the systemic peripheral estrogen supply to this circuit. VAT contains high CYP19A1 activity (2-3-fold per gram versus SAT), and in reproductive-age women with metabolic dysfunction, visceral fat-derived estrone contributes measurably to systemic estrogen levels. As GLP-1RAs preferentially reduce VAT at approximately the 1.6:1 VAT:SAT ratio documented across DXA-based trials, peripheral aromatase activity decreases. This reduces circulating estrone available for ERα-mediated activation of lesion COX-2 and CYP19A1 — attenuating the self-amplifying loop from the outside.

Phase-specific relevance is particularly important here. Endometriosis lesions show maximal inflammatory and proliferative activity in the late luteal phase: prostaglandin synthesis peaks as progesterone begins to fall; estrogen withdrawal from the systemic compartment is paradoxically associated with increased local lesion inflammatory activity driven by inflammatory mediator upregulation; and peritoneal macrophage cytokine output increases. GLP-1RA suppression of NF-κB in lesion cells would be expected to attenuate this late luteal inflammatory amplification, and reduced peripheral estrogen availability from VAT reduction would reduce the substrate for the local estrogen-PGE2 amplification circuit.