EXTL3-interacting endometriosis-specific serum factors induce colony formation of endometrial stromal cells

Abstract

Endometriosis is a benign chronic condition characterized by the existence of endometrial-like stroma and glandular tissue in extrauterine locations. The molecular mechanisms of its pathogenesis have not been elucidated. We have studied the role of EXTL3 (exostosin-like 3) in endometriosis and found that it is expressed in endometrial tissue as well as endometriosis lesions. We have found that serum from endometriosis patients contains a factor or factors, which interact with EXTL3 resulting in strongly increased colony formation in regenerating cell culture. We also found increased anti-EXTL3 antibodies in endometriosis patients' sera. EXTL3 is an N-acetyl glucosamine (GlcNAc) transferase, performing a key step in heparan sulfate (HS) glucosaminoglycan synthesis. Many viruses replicate in regenerating epithelial cells and use HS as a receptor for cell entry. We measured antibody titres to viruses, which use HS as a receptor for cell entry, and found rarely increased titres for these viruses in endometriosis sera, whereas titres to viruses using other receptors were equally distributed in study groups. The data indicate that perturbation of HS metabolism is associated with endometriosis.

Figure 1

EXTL3 PCR products of 535 bp were produced from all cDNAs from eutopic endometrium (E) and endometriosis lesions (L). K04-K13 – healthy controls, E213-E47 – endometriosis patients, “−” – no template, gDNA – genomic DNA. Marker – Fermentas ZipRuler Express II.

Figure 2

Fluorescently tagged protein expression in live transfected endometrial stromal cells (A) EXTL3-EGFP (B) control EGFP.

Figure 3

Numbers of regenerating colonies in wells transfected with EXTL3 or control vector and grown with sera from endometriosis patients or endometriosis-free patients (mean ± SD). *p = 0.002, ***p = 9.5 × 10⁻⁵, one-tailed t-test.

Figure 4

Regenerating colonies and glandular structures in culture. Column 1 – Transfected with EGFP, cultured with sera from an endometriosis patient. Column 2 – Transfected with EXTL3-EGFP, cultured with sera from an endometriosis patient. Column 3 – Transfected with EXTL3-EGFP, cultured with serum from a patient without endometriosis. Row 1 – composite image of bright field, DAPI (blue), CD9 (green), and SUSD2-W5C5 (red). The red signal is considerably weaker compared to the others. Row 2 – DAPI, row 3 – CD9, and row 4 – SUSD2 W5C5.

Figure 5

(A) Serum antibody titers to EXTL3, as measured by ELISA, are increased in endometriosis. * - Student’s two-tailed t-test p = 0.042, E: MEAN = 24.1; SEM = 5.42; n = 15 vs. N: MEAN = 13.21; SEM = 2.53; n = 29. (B) titers to PGRMC1 do not differ between groups. Student’s two-tailed t-test p = 0.37, E: MEAN = 2.39; SEM = 0.80; n = 9 vs. N: MEAN = 3.61; SEM = 0.84; n = 16. E – endometriosis, N – Endometriosis-free.

Figure 6

Serum antibody titers to viral antigens, as measured by ELISA, depend on the virus entry receptor and whether the patient has endometriosis. In endometriosis group (E) the titers are rarely increased against viruses binding to heparan sulfate (HPV16, HSV1, CMV), whereas in Endometriosis-free (N) group the titers are higher (* - q < 0.05). Titers against viruses binding sialic acid receptors (Echovirus and Bocavirus) are not significantly different between the groups. (A) HPV16-E6, p = 0.0064, FDRq = 0.0346; (B) HPV155-L1, p = 0.0012, FDRq = 0.0270; (C) HPV19-L1, p = 0.0096, FDRq = 0.0346; (D) HSV1, p = 0.01, FDRq = 0.0346; (E) CMV (HHV5), p = 0.018, FDRq = 0.0436; (F) VZV, p = 0.08, FDRq = 0.0685; (G) human polyomavirus 7 (HPyV7), p = 0.013, FDRq = 0.0388; (H) Echovirus VP1, p = 0.64, FDRq = 0.3697; and (I) Human bocavirus (HBoV) VP1, p = 0.94, FDRq = 0.4867 (Student’s two-tailed t-test, E (n = 9) vs. N (n = 16)).