Treatment of constitutive resistant carcinomas with a demethylating agent restores PR A expression and their antiprogestin responsiveness.

WARGON V; FERNÁNDEZ, S; LANARI C

San Diego, CA. USA

Congreso; AACR; 2009

AACR

Mouse mammary carcinomas induced by medroxyprogesterone acetate (MPA) are maintained by syngeneic transplantation and tumor variants with different hormone responsiveness have been generated. Tumors which grow without the MPA supply were defined as progestin or hormone independent (HI). Most of these tumors regress after antiprogestin (mifepristone, MIF) administration (responsive). Eventually, some of them became resistant to antiprogestins during therapy and they were refered as acquired resistant tumors. Other variants show constitutive resistance to hormone therapy. We have previously demonstrated that responsive tumors show higher levels of progesterone receptor A (PRA) than PRB, while the opposite occurs in both types of hormone resistant tumors, indicating that PR isoform ratio might be a predictor of antiprogestin responsiveness. We have also shown that acquired resistance may be reverted suggesting the involvement of epigenetic mechanisms regulating PRA expression. Thus, we evaluated whether PR A was silenced by methylation in acquired and constitutive resistant tumors. Interestingly, we found that only in constitutive resistant tumors CpG islands of PR A promoter are highly methylated, being responsible of low PR A expression levels. Moreover, we found that treatment of cell cultures of constitutive resistant tumors with the demethylating agent 5-aza-2´-deoxycytidine (AZA) induced PR A re-expression. The aim of this study was a) to investigate whether the in vivo treatment of two constitutive resistant tumors from the MPA breast cancer model (C4-2-HI, 59-HI) with AZA could restore PR A expression and mifepristone sensitivity and b) to study the expression levels of the DNA methyltransferases Dnmt1 (maintenance methylase), Dnmt3a and Dnmt3b (de novo methylases). C4-2-HI and 59-HI tumors were implanted sc in BALB/c mice and when tumors reached a size of 4 mm2 they were treated with a) AZA (0.75mg/kg/every other day for 12 days; ip), b) MIF (12 mg/kg/daily; sc), c) AZA plus MIF or d) vehicle. Neither MIF nor AZA independently inhibited tumor growth whereas the combination of both agents reduced significantly tumor growth (p<0.001).  No effect was observed in similarly treated acquired resistant tumors, indicating that MIF inhibition would be by PR A re expression in constitutive resistant ones. The expression of PR was evaluated in control and AZA-treated tumors and, as expected, an increase in PR A expression was observed by western blots and by immunohistochemistry. The expression levels of Dnmt 1, 3a and 3b were studied by Western blot and immunofluorescence in two responsive tumors and two constitutive resistant tumors. Higher levels of Dnmts were observed in the constitutive resistant tumors (p<0.001) indicating a close relationship between PR-A methylation and methyltransferases levels. In conclusion, we report for the first time that a) constitutive resistant mammary carcinomas have high levels of Dnmts which might be responsible for methylating PR-A promoter, b) PR-A can be re-expressed by treatment with demethylating agents only in constitutive resistant tumors, c) AZA-treated constitutive resistant tumors acquire antiprogestin responsiveness. These results support our hypothesis that tumors with high levels of PR-A are those which respond to antiprogestins treatment.