KRAS Activation and over-expression of SIRT1/BCL6 Contributes to the Pathogenesis of Endometriosis and Progesterone Resistance

Abstract

Endometriosis is an inflammatory condition that is associated with progesterone resistance and cell proliferation, resulting in pain, infertility and pregnancy loss. We previously demonstrated phosphorylation of STAT3 in eutopic endometrium of infertile women with this disorder leading to over-expression of the oncogene BCL6 and stabilization of hypoxia-induced factor 1 alpha (HIF-1α). Here we report coordinated activation of KRAS and over-expression of Sirtuin 1 (SIRT1), a histone deacetylase and gene silencer, in the eutopic endometrium from women with endometriosis throughout the menstrual cycle. The mice with conditional activation of KRAS in the PGR positive cells reveal an increase of SIRT1 expression in the endometrium compared to control mice. The expression of progesterone receptor target genes including the Indian Hedgehog pathway genes are significantly down-regulated in the mutant mice. SIRT1 co-localizes with BCL6 in the nuclei of affected individuals and both proteins bind to and suppress the promoter of GLI1, a critical mediator of progesterone action in the Indian Hedgehog pathway, by ChIP analysis. In eutopic endometrium, GLI1 expression is reduced in women with endometriosis. Together, these data suggest that KRAS, SIRT1 and BCL6 are coordinately over-expressed in eutopic endometrium of women with endometriosis and likely participate in the pathogenesis of endometriosis.

Figure 1

Endometriosis is associated with elevated serum inflammatory cytokines including (a) IL-1a, (b) IL-6 and (c) IL-17, compared to normal controls.

Figure 2

Correlation of between KRAS and SIRT1 in human endometrium with endometriosis. (a) Western blot analysis of SIRT1 and BCL6 proteins in proliferative and secretory phases of human endometrium with endometriosis. β-actin was used as sample-loading control. Representative blots have been cropped to reduce unnecessary area. Full-length blots are presented in Supplementary Fig. S2. (b) Densitometric analysis of KRAS and SIRT1 protein levels by Western blot analysis in eutopic endometrium from proliferative and secretory phase in women with and without endometriosis. (c) Correlation between SIRT1 and KRAS in women with endometriosis throughout the menstrual cycle phases based on Western blot analysis (correlation coefficient = 0.4641, p = 0.0009). (d and e) H-score of KRAS (d) and SIRT1 (e) expression in endometrium from women with and without endometriosis and representative photomicrograph of immunohistochemical staining of KRAS in endometrium from women without and with endometriosis. The results represent the mean ± SEM. ***p < 0.001.

Figure 3

Correlation of between SIRT1 and BCL6. (a) Correlation analysis between SIRT1 and BCL6 in human endometrium with endometriosis. (b) Immunoprecipitation (IP) analysis between SIRT1 and BCL6 in Ishikawa cells and Human endometrium with endometriosis. Representative blots have been cropped to reduce unnecessary area. Full-length blots are presented in Supplementary Fig. S4. (c) Co-localization of SIRT1 and BCL6 in the human endometrium without and with endometriosis by immunofluorescence analysis.

Figure 4

Levels of SIRT1 and BCL6 proteins during progression of endometriosis in a baboon model. (a and b) H-score of SIRT1 (a) and BCL6 (b) expression in endometriosis baboon model induced by intraperitoneal inoculation of menstrual endometrium during progression of endometriosis. The results represent the mean ± SEM. *p < 0.05 and ***p < 0.001. (c and d) Representative photomicrograph of immunohistochemical staining of SIRT1 (c) and BCL6 (d) in the baboon endometrium of pre-inoculation and 3, 9 and 15 months post-inoculation during endometriosis progression.

Figure 5

Levels of SIRT1 in the KRAS activation mouse model. (a) Representative photomicrograph of immunohistochemical staining of SIRT1 in the control and KRAS activation mouse. (b) The mRNA expression level of P4 target genes in the uterus from control and KRAS activation mice (n = 9). The results represent the mean ± SEM. *p < 0.05 and ***p < 0.001.

Figure 6

Regulation of GLI1 gene expression by SIRT1 and BCL6 proteins. (a) Western blot analysis of BCL6 and SIRT1 in Ishikawa cells treated with E2 + MPA for 0, 30 min, 6, 12, and 24 hours. β-actin was used as sample-loading control. Representative blots have been cropped to reduce unnecessary area. Full-length blots are presented in Supplementary Fig. S5. (b) Quantitative real time PCR analysis of GLI1 gene expression in Ishikawa cells treated with E2 + MPA for 0, 6, 12, and 24 hours. (c) Map of BCL6 binding site on the GLI1 promoter (Gray boxes). Negative control (N.C.) region on the GLI1 gene was used as negative control of ChIP assay. Primers used in ChIP assay are presented by arrows. (d) ChIP assay using anti-SIRT1 antibody on GLI1 promoter in Ishikawa cells treated with or without E2 + MPA for 24 hours. The results represent the mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 7

Levels of GLI1 in endometrium from women with and without endometriosis. (a) H-score of GLI1 expression in endometrium from women with and without endometriosis. The results represent the mean ± SEM. *p < 0.05. (b) Representative photomicrograph of immunohistochemical staining of GLI1 in endometrium from women without and with endometriosis.