The novel, small-molecule DNA methylation inhibitor SGI-110 as an ovarian cancer chemosensitizer

Abstract

Purpose: To investigate SGI-110 as a "chemosensitizer" in ovarian cancer and to assess its effects on tumor suppressor genes (TSG) and chemoresponsiveness-associated genes silenced by DNA methylation in ovarian cancer.

Experimental design: Several ovarian cancer cell lines were used for in vitro and in vivo platinum resensitization studies. Changes in DNA methylation and expression levels of TSG and other cancer-related genes in response to SGI-110 were measured by pyrosequencing and RT-PCR.

Results: We demonstrate in vitro that SGI-110 resensitized a range of platinum-resistant ovarian cancer cells to cisplatin (CDDP) and induced significant demethylation and reexpression of TSG, differentiation-associated genes, and putative drivers of ovarian cancer cisplatin resistance. In vivo, SGI-110 alone or in combination with CDDP was well tolerated and induced antitumor effects in ovarian cancer xenografts. Pyrosequencing analyses confirmed that SGI-110 caused both global (LINE1) and gene-specific hypomethylation in vivo, including TSGs (RASSF1A), proposed drivers of ovarian cancer cisplatin resistance (MLH1 and ZIC1), differentiation-associated genes (HOXA10 and HOXA11), and transcription factors (STAT5B). Furthermore, DNA damage induced by CDDP in ovarian cancer cells was increased by SGI-110, as measured by inductively coupled plasma-mass spectrometry analysis of DNA adduct formation and repair of cisplatin-induced DNA damage.

Conclusions: These results strongly support further investigation of hypomethylating strategies in platinum-resistant ovarian cancer. Specifically, SGI-110 in combination with conventional and/or targeted therapeutics warrants further development in this setting.

Figure 1. 5-aza-dC and SGI-110 treatment modulates CDDP sensitivity and alters DNA methylation and gene expression in vitro

(A). Comparison of cell growth rates of parental A2780 cells, A2780-CDDP resistant cells, and CP70 CDDP resistant cells treated with 5µM SGI-110, or vehicle (DMSO 1:2000) for 48 hours followed by CDDP ranging from 0–50µM CDDP. Mean values ± S.E.M. of 8 independent experiments in duplicate are reported. All treatments were significantly different, at P < 0.05, than vehicle control cells. IC₅₀ values are listed in Supplementary Table S3. (B). RT-PCR analysis of MLH1 RNA levels in A2780 cells. Fold change in RNA levels were calculated as 2^(–ΔΔCT) relative to DMSO-treated cells. (C and D). RASSF1A and HOXA11 were significantly demethylated by SGI-110 and the mRNA expression was upregulated in A2780-CDDP resistant cells. All changes are significant (P<0.05). (E). Cells were treated for 3-consecutive days with 0.1 µM SGI-110, 0.3 µM SGI-110 or DMSO (control). LINE1 methylation status was determined by pyrosequencing analysis and expressed as % of DMSO-treated cells. Data shown represent mean values ± S.E.M. from triplicate experiments.

Figure 2. SGI-110 induces the expression of potential CDDP-resistance biomarkers in panel of OC cell lines

(A, B) Fold-change in mRNA expression of DOK2 and ZIC1 in OC cell lines following 3-consecutive day treatment with 0.1µM SGI-110, 0.3µM SGI-110, 1µM SGI-110 or vehicle (DMSO) quantified by qRT-PCR. Data shown represent mean values ± S.E.M. from triplicate samples. (C) Pyrosequencing analysis of basal ZIC1 methylation levels in untreated OC cell lines. Data shown represent mean values ± S.E.M. from triplicate samples. (D) Pyrosequencing analysis of ZIC1 methylation levels in OAW28 cells, following 3-consecutive day treatment with SGI-110. Methylation levels expressed as % of DMSO-treated cells. Data shown represent mean values ± S.E.M. from triplicate experiments.

Figure 3. SGI-110 and CDDP in the QD5 and biweekly treatment regimens decreases A2780-CDDP resistant-derived xenograft tumor growth treated with

CDDP-resistant A2780 xenograft tumor volume was compared among single agent and combination treatment to vehicle control in two treatment schedules (*: P<0.05, **: P<0.01, ***: P<0.001). (A). QD5 treatment. (D). Biweekly treatment. Data shown represent mean values ± S.E.M. from 5 tumor samples.

Figure 4. QD5 and biweekly SGI-110 treatments induce changes in LINE1 methylation in PBMCs and tumor samples from mice with parental A2780 or CDDP-resistant xenografts

Tumor bearing mice were treated with SGI-110 and CDDP according to biweekly or QD5 schedule. Blood samples were collected (biweekly: on days 1, 8, 15, 22, and end of study (EOS); QD5: on days 1, 8, and EOS). Tumors were collected at (EOS) after the mice were sacrificed. DNAs were extracted from PBMCs and tumor and subjected to bisulfite conversion and pyrosequencing for LINE1 methylation (*: P<0.05, **: P<0.01, ***: P<0.001). (A). PBMC LINE1. Left panel- parental A2780 xenograft mice with QD5 regimen, right panel- parental A2780 xenograft mice with biweekly regimen. (B). Tumor LINE1. Left panel- parental A2780 xenograft mice with QD5 regimen, right panel- parental A2780 xenograft mice with biweekly regimen. (C). PBMC LINE1. Left panel- A2780-CDDP resistant xenograft mice with QD5 regimen, right panel- A2780-CDDP resistant xenograft mice with biweekly regimen. (D). Tumor LINE1. Left panel- A2780-CDDP resistant xenograft mice with QD5 regimen, right panel- A2780-CDDP resistant xenograft mice with biweekly regimen. Data shown represent mean values ± S.E.M. from 5 xenograft samples.

Figure 5. qRT-PCR and pyrosequencing analysis of specific genes in QD5 and biweekly schedule mice with parental and CDDP-resistant A2780 xenografts

Selected specific genes showed significantly demethylated and upregulated from A2780 xenografts in the 2 treatment schedules. (A). Parental A2780 xenografts from mice treated with QD5 schedule. (B). Parental A2780 xenografts from mice treated with biweekly schedule. (C). A2780-CDDP resistant xenografts from mice treated with QD5 schedule. (D). A2780-CDDP resistant xenografts from mice treated with biweekly schedule. Data shown represent mean values ± S.E.M. from 5 xenograft samples.

Figure 6. SGI-110 increased CDDP DNA adducts and alters epigenetic marks in vitro

Parental A2780 and A2780 CDDP-resistant cells were treated in triplicate with either no SGI-110 or 5µM SGI-110 in fresh RPMI and grown for 48 hours. Media was removed and replaced with fresh RPMI containing 10µM CDDP for 2hr and were allowed to repair for 0, 2, 4 and 24 hours DNA was extracted from cells analyzed by ICP-mass spectrometry. (A). SGI-110 increased CDDP DNA adducts in vitro. (B and C). Western blot and quantification of protein from parental A2780 and A2780 CDDP-resistant cells treated with 5µM SGI-110 for 48hr and blotted with rabbit anti-β-tubulin, mouse anti-histone H3, and anti-acetyl-histone H4 (B), or anti-trimethyl-histone-H3 (C). Data shown represent mean values ± S.E.M. from triplicate experiments.