Will Epigenetics Help Overcome Cisplatin Resistance in Ovarian Cancer?

Thu, 03/29/2012 - 12:24pm
Nicole Kelesoglu

Epigenetics researchers apply sophisticated techniques and rational multidimensional analyses to discover biomarkers for ovarian cancer.

Why are ovarian cancer survival rates lagging behind survival rates for other cancers? Sadly, the dynamic and highly individualized nature of ovarian tumors makes it a particularly difficult disease to defeat. Fortunately, epigenetics researchers are applying sophisticated techniques and rational multidimensional analyses to discover biologically relevant, specific biomarkers. Recently, a couple of groups have made further progress, zeroing in on the core problem of ovarian cancer recurrence—cisplatin chemotherapy resistance. Their labs’ combinational epigenomic approaches aim to beat ovarian cancer by identifying tumor signature profiles that can track the dynamic nature of chemo-resistance.

Epigenetics is the study of how gene expression is systematically regulated, how environmental cues feedback into those multidimensional systems, and how that information is passed on to future generations. The three main areas of epigenetics include: DNA methylation, which directly silences genes; chromatin and histone protein modifications, which control physical exposure of genes for their expression; and non-coding RNA systems, which apply an additional layer of expression control and establish other epigenetic marks.

DNA methylatransferase methylates cytosines
Epigenetic marks control gene expression. Here a DNA methylatransferase methylates cytosines to silence gene expression, as modified histones within nucleosomes control 3-demensional accessibility of DNA sequences. (Source: New England Biolabs)

Epigenetics and Cancer Research
The vast majority of epigenetic studies involve cancer research. Carcinogenesis is characterized by global loss of DNA cytosine methylation, with concurrent hyper-methylation of the promoters of tumor suppressor genes. These epigenetic “signatures” of DNA methylation, are uniquely tissue and disease specific. Other epigenetic aberration patterns, noncoding RNAs and histone modifications, and chromatin remodeling factors are also advancing our knowledge of cancers. Epigenetic researchers already have “proof of principle” in the approved epigenetic drugs for leukemia. Exciting results from ongoing research and/or clinical trials in lymphoma, lung cancer, breast cancer, and colorectal cancer have everyone on the edge of their seats. These drugs appear to be working synergistically to make chemo-therapy effective or to otherwise reverse cancer progression.

Ovarian cancer represents a special translational research opportunity because it is theorized that the critical problem of chemo-resistance is driven by epigenetic events. Right now, oncologists use trial and error tactics to manage patient care. Because ovarian cancer has subtle symptoms, diagnosis is usually made at an advanced stage and involves invasive laparoscopic surgery. Tumorgenesis in the body is difficult to study experimentally, due to the scarcity of primary early stage tumor tissue samples. 90% of cases are the deadly, invasive ovarian cancer type. Following surgery, women undergo chemotherapies, such as cisplatin (cis-diamminedi-chloroplatinum[II]), and sometimes radiation therapy. Individual tumor profiles are highly variable and dynamic over time, so recurrence is rampant. Recent evidence has shown that a sub-population of ovarian cancer cells can undergo a reversible state of chemo-resistance to cisplatin via epigenetic mechanisms—these cells have been termed “drug-tolerant persisters.”1 It has been suggested that this resistance is based on genetic changes and aberrant epigenetic marks. However, the exact mechanisms of resistance to cisplatin are unknown. Maradeo ME and Cairns P. (2011) provide a great review of epigenetics ovarian cancer translational research.2

human ovary cell
Microscopic photo of a professionally prepared slide demonstrating the structure of a human ovary cell.

Combinational Approaches
The latest news comes from an Oncogene publication by C Zeller et al.3 describing a combinational approach—considering both methylation and expression—to successfully identify (as few as) 13 key gene candidates as causative drivers of cisplatin-resistance in ovarian cancer samples. The profiles were assembled first with BeadChIP array-based methylation profiling, analyzed by the rank products method,4 followed by array-based expression profiling of differential cisplatin sensitivity and then analyzed by rank products analysis. The research began with differential analysis on cisplatin-sensitive/resistant ovarian cancer cell lines. Results were replicated using the approach with in vivo primary cell lines and tumors derived from ovarian cancer patients, at pre-chemotherapy and then when chemo-resistance developed. Next, correlated changes in methylation analysis with expression changes were targeted for drivers. The combined methods analysis was repeated on ovarian cancer cell lines treated with the epigenetic modifiers, DAC and Belinostat. These epigenetic-based drugs resensitize drug-tolerant persisters cells to cisplatin. The data confirmed the gene candidates as cisplatin sensitivity biomarkers for ovarian cancer clinical trials.

The authors point out that their epigenome-wide differential methylation data conflicts with previous studies reporting a lack of methylation of genes in ovarian cancer samples, responsive to HDACi and partially to DNMTi treatment.5 They attribute this difference to the fact that their use of rank products method analysis both avoids fold changes that exceeds an arbitrary threshold...and can identify subtle changes missed by other methods. Their small number of potential biomarkers results represent a filter down to less than 1% of the total methylome changes occurring in cisplatin re-sensitization.

A second group, Wei Yu et al. has reported the first MBD-seq analysis global data set of cisplatin resistance in an ovarian cancer cell model.6 From this data they found several epigenes to target for further research.

The methylated DNA of cisplatin-sensitive and cisplatin resistant ovarian cancer cell lines were enriched by precipitation using Methyl-CpG binding domain protein, followed by next generation sequencing. Many genes within the set were mapped to functional signaling pathways, within the differentially methylated regions (DMRs). Methylation states for each were validated by Methylation-Specific PCR and bisulfite sequencing. Interestingly, their data suggests lower global CpG methylation in the cisplatin resistant cell lines. They treated the resistant cells with a demethylator, 5-aza-29-deoxycytidine (5-aza-dC), and restored expression of these genes. However, there was no data reported on reversing their state to cisplatin sensitivity.

The authors note that the DMRs, between the cisplatin resistant and sensitive cell lines had similarities, but were of a greater number than reported by another paper using differential methylation hybridization (DMH) type analysis.7 DMH separates methylated DNA by methylation sensitive/insensitive restriction enzyme digests, amplification, and labeling for oligonucleotide array. However, this difference may have been due to the derivative cell lines used in each study.

From these new reports, we can imagine these causative genes functioning as the “drivers” of chemo-resistance, while ovarian cancer epigene biomarkers, act as spotters do on auto race teams. Spotters have the key job as the “eyes of the driver,” providing alerts about changes on the track, even arranging several cars to gang up, and improve positioning to win the race. By giving clinicians access to the tactical guidance provided by spotters, appropriate epigenetic drugs could be synchronized to enable expression of those driver genes that reverse cisplatin resistance. With the right biomarker profile input, ovarian cancer management becomes a winnable race.

It’s been said that genes are our destiny. But we are learning that our fate can be altered depending on what happens to genes—in the growth of the fetus, for example, or in response to hormones, environmental stress or nutritional choices. This is epigenetics: the range of studies, the complexity of detail accumulated, the research being conducted into diseases and their treatments is so vast and growing so quickly, it is almost impossible to keep pace with all the advances in research and innovations in method. Social networks such as E3 (Engaging Epigenetics Experts), a microsite developed and supported by New England Biolabs, are indispensable open research communities that promote discussion and collaboration in the epigenetics community. Information found there and on other social media sites will almost certainly become essential to progress in our understanding of ovarian cancer, autism, low birth weights, and much more.

About the Author
Nicole Kelesoglu is a scientific technical writer and social media manager for, which is supported by New England Biolabs.

1. Sharma SV, et al., A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. (2010) Cell. 41(1):69-80.
2. Maradeo ME and Cairns P. Translational application of epigenetic alterations: Ovarian cancer as a model. (2011) FEBS Lett. 585(13): 2112-20.
3. C Zeller et al. Candidate DNA methylation driers of acquired cisplatin resistance in ovarian cancer identified by methylome and expression profiling. (2012) Oncogene 1-10
4. Breitling R et al. Rank products: a simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments. (2004) FEBS Lett, 573(1-3):83-92.
5. Suzuki et al. A genomic screen for genes upregulated by demethylation and histone deacetylase inhibition in human colorectal cancer. (2002) Nat Genet., 31: 141-149.
6. Wei Yu et al. Global analysis of DNA methylation by Methyl-Capture sequencing reveals epigenetic control of cisplatin resistance in ovarian cancer cell. (Dec. 2011) PloS One, 6(12):e29450.
7. Yan PS, Wei SH, Huang TH. Differential methylation hybridization using CpG island arrays. (2002) Methods Mol Biol 200: 87–100.


Share this Story

You may login with either your assigned username or your e-mail address.
The password field is case sensitive.