Abstract
Epigenetic phenomena are known to be at the root of many common diseases. To date, the FDA has approved four epigenetic therapies that show promising results, prolonging lives of terminal cancer patients. However, there is a relative lack of knowledge about epigenetic effects in the long-term and across generations, so epigenetic therapies may have unforeseeable risks if they are used on younger people with non-lethal epigenetic diseases. In this report, we propose a heightening of standards for epigenetic therapy: therapies should be targeted to specific genes in specific cells, patients’ epigenomes should be sequenced before and after treatment, and germline effects should be unacceptable. Moreover, more research should be performed to answer questions about transgenerational epigenetic effects, the effects of altered epigenomes in the long term, and to develop superior assays for screening epigenomes.

Epigenetics Background
Introduction. The code of life is more than a sequence of A’s, C’s, T’s, and G’s. Muscle cells in the human heart contain the same DNA as skin cells in the foot, yet these two cell types behave in radically different ways. Both contain the DNA for every one of over 20,000 human genes but express, only the ones needed for their own form and function. These differences in gene expression are a result of epigenetic controls. Epigenetics refers to any chemical modification on DNA that does not alter the genetic sequence. In addition to governing cellular differentiation, epigenetic mechanisms facilitate the proper functioning of a cell. Neurodevelopmental disorders, immunodeficiency, cancer, and other illnesses can result when these mechanisms go awry.

Methylation. DNA methylation is one of the most prominent and powerful mechanisms of epigenetic control. In humans, enzymes called methyltransferases add methyl groups to CpG sites, short DNA sequences which are abundant throughout the genome. Methyl groups block transcription factors (gene activators) from binding to DNA and performing their normal function. In short DNA methylation has been shown to be able to silence gene expression (Feinberg 2004). Although epigenetic factors do not affect the actual sequence of DNA, they do affect phenotype, the observable characteristics resulting from patterns of gene expression. Specific patterns of methylation are necessary for a cell to modulate the level of expression of each of its genes, controlling which genes are turned on and off.

Epigenetic Disease
Cancer. DNA methylation has been referred to as the “hallmark of cancer” (Szyf 2004). Abnormal methylation patterns throughout the genome that cause blockage of tumor suppressor genes, and overexpression of oncogenes, have been linked to many types of cancer. For instance, breast cancer cases generally exhibit inactivation of a gene called BRCA1. In sporadic (i.e. non-familial) cases, this suppression is usually caused by hypermethylation, rather than mutation, of the gene (Rice 2000). On the other hand, hypomethylation, the absence of methylation, causes overexpression of the flap endonuclease 1 gene in some breast cancer patients. (Singh 2008). In fact, hypomethylation is associated with most types of cancer (Feinberg 2004).

Neurological. Methylation abnormalities have been linked to a wide range of diseases. Fragile X syndrome, one of the leading genetic causes of intellectual disability, is characterized by an excess amount of methylated CpG sites in the FMR1 gene on the X-chromosome.  High levels of methylation in the FMR1 gene disrupt the production of protein necessary for normal brain development. Patients suffering from this disorder are at risk for autism, ADHD, decreased IQ, infertility (in females), and distorted facial features. (Jacquemont 2011)

                Psychological. Epigenetic mechanisms control cellular function throughout the body, but also impact psychological states. In an animal study, rat pups that received better maternal care soon after birth - in the form of licking, grooming, and arched-back nursing - had lower levels of methylation at the glucocorticoid receptor gene. These rats displayed less intense responses to stressful situations than their counterparts who received poor maternal care. The researchers were able to eliminate these differences via epigenetic interference. This study demonstrates the great extent to which epigenetic modifications affect an individual’s psychological state. (Weaver 2004)

Epigenetic Therapy
Fundamental Advantages. The aforementioned epigenetic roots of disease are attractive targets for therapy. Aberrant DNA methylation patterns are inherently reversible, as opposed to permanent genetic mutations. In the case of cancer, epigenetic therapy coaxes tumor cells to return to a healthy state, rewiring their methylation patterns so the cells express genes that halt their cancerous uncontrolled growth. Traditional chemotherapy strategies, on the other hand, aim to kill cancer cells and are fundamentally more toxic to patients.

Recent Successes. There have been some exciting clinical successes with the first generation of epigenetic therapies. To date, four epigenetics treatments have been approved by the FDA for use in cancer patients: Zolinza, Istodax, Vidaza, and Dacogen (Claus 2005). The latter two are relevant to DNA methylation, and work by inhibiting DNA methylation in myelodysplastics syndromes (MDS). The drugs effectively reverse cancerous hypermethylation and return standard function to cell-cycle genes, restoring normal growth rates. They do not work for everyone, but fewer than 10% of patients do experience a complete response, which means a total reversal from diseased to healthy bone marrow and blood (Silverman 2002).

                 Problems. The results of first generation epigenetic therapies are promising, but these drugs should not be seen as the ideal model for future developing therapies, especially if doctors want to treat younger patients with non-lethal epigenetic diseases. First generation drugs fail to satisfactorily address many issues. For one, they all inhibit epigenetic processes; that is, they only work in one direction. As we discussed, many cancers are actually caused by epigenetic inhibition, ie. low methylation levels, and would need to be treated with drugs that restore normal methylation levels (Feinberg 2004). Moreover, the current drugs work by blindly affecting all genes in the genome of all the cells they encounter. One of the scientists behind the Dacogen studies noted there is a “potential for harm,” but so far the adverse events, including red blood cell suppression, diarrhea, anorexia, and others, have been deemed acceptable by the FDA (http://www.pbs.org/wgbh/nova/body/epigenetic-therapy.html). It is possible this potential for harm will be realized if epigenetic therapies are used on patients with a much longer expected life span than the current patient population.

Ethical Questions

Long Term Risk. We do not yet fully understand all the effects of DNA methylation (Rothstein 2009). Perhaps the un-targeted nature of these drugs really does not affect patients, but how can we be sure we are not setting patients up for adverse events later in life? The approved trials were primarily done with elderly, very sick cancer patients; but we know epigenetic modifications can affect us on a longer, even transgenerational, time-scale (Rothstein 2009). Of course, clinical trials are not normally designed to detect side-effects ten, twenty, or thirty years after treatment. If these treatments provide disease alleviation in the short term but end up causing unforeseeable epigenetic abnormalities in the long term, are they acceptable for younger patients with non-terminal disease? Most would argue this depends on the gravity of the adverse effects - a risk/benefit analysis similar to that performed for any drug approval. Can we design clinical trials that take these risks into account? Should they last 10 years? Should the follow-ups last even longer? If the drug does show an immediate benefit in the trial subjects, is it fair to keep it off the shelf for that long? Certainly, a trial of this duration with these looming questions would be a non-starter with the pharmaceutical investment community. At a minimum, researchers should monitor “off-target” epigenetic effects by performing methylation-sensitive sequencing on parts of their patient’s genomes in multiple cell types. This is difficult and expensive with existing technologies, which is another problem worth addressing (Laird 2010).

                Trans-generational Risk. These are not the only potential issues with first generation epigenetic therapies. They act blindly, not only on a genomic level, but also in terms of affected cell type. We cannot exclude the possibility that they will affect the epigenomes of germline cells. In fact, recent studies demonstrate that Dacogen could harm the fetus if given to a pregnant woman or a father planning on having children (Ding 2012). Currently, the drug is marketed alongside strong warnings against having children while being treated (www.dacogen.com). There is evidence that epigenetic modifications can be inherited by children and even grandchildren, so doctors need to keep the health of these unborn people in mind if they prescribe epigenetic drugs to reproductive-age patients (Grossniklaus 2013). The effect of epigenetic therapies on germline cells should be measured not only while the drug is being administered, but also in follow-ups after the treatment is finished. It is critical that more basic research is performed to determine the extent of transgenerational epigenetic effects, especially now that the question has become clinically relevant. If it turns out that these epigenetic treatments will preclude patients from safely having children, we must consider the consequences for the younger people with diseases like Down syndrome, autism, fragile X syndrome, and ICF, which could potentially be epigenetically treated (Kondo 2000 and Kerkel 2010). As aberrant DNA methylation is increasingly correlated with more common diseases like heart disease and obesity, it should be noted that the epigenetic patient pool is not trivial in size, so these questions will only become more pressing in time as they become relevant to a larger portion of the younger population (Campion 2010). If we acknowledge that some patients will have children regardless of any warnings, we must consider if it is acceptable to have synthetically altered epigenomes in the population pool.

Second Generation Epigenetic Therapies

                Higher Standards. If we continue developing first generation epigenetic therapies that affect whole genomes and any cell type, we should more openly acknowledge that we are inviting an exceptional amount of potential for future harm in unforeseeable (and difficult to detect) ways when they are used on non-terminal patients. So far, epigenetic therapy has been held to the same safety standards of other cancer therapies. However, as these drugs affect gene expression, the basis of our existence, it is more appropriate to hold them to the standards of gene therapies. That entails an expectation to target only the genes, pathways, and cells relevant to the disease (www.fda.gov). Germline transmission of epigenetic modifications should be unacceptable. Any MDS patient is eligible for Dacogen, but individuals have different epigenomes – this may explain the low response rates. Patients’ epigenomes should be screened, so doctors can be sure the proposed therapy is relevant to the individual, just as they would sequence a patient’s genome to be sure of a genetic disease. Genome wide epigenetic sequencing has recently become feasible, although it is still difficult, and somewhat error prone (Laird 2010). There is a need for further development of this technology to enable the promise of personalized epigenetic treatment.

                Our Work. These more stringent safety standards demand the development of second-generation epigenetic therapeutics that are properly targeted at the genomic and cellular level, so we can begin treating non-lethal epigenetic disease in younger patients. Our research team, Penn iGEM, took initial steps by designing a novel enzyme that selectively restores methylation at specific DNA sites. We can now achieve this site-specific methylation with greater specificity, easier customization, and 70 times lower costs than the previously reported technologies (https://2013.igem.org/Team:Penn). We can achieve these cost savings by using a TALE system, easily customizable with a $425 kit from Addgene, as opposed to a zinc-finger system, which can cost $30,000 to have constructed by Sigma (addgene.com and protomag.com). Moreover, we have developed an alternative DNA methylation assay, called MaGellin, which is significantly simpler, faster, and cheaper than methylation-sensitive sequencing for applications like ours. MaGellin will accelerate the laborious development process that stands between today and a future with safer and smarter second generation epigenetic therapies.

                Conclusion. Epigenetic phenomena have a significant impact on the way we live and grow, and can be responsible for the way we die. The need for epigenetic therapies is clear, the initial successes are promising for cancer patients, but the model for future developments is not yet set in stone. If doctors want to treat younger patients with non-lethal epigenetic diseases, the consideration of risks must include the long term, and the decisions in the clinic must be based on data from researchers asking fundamental questions. Epigenetics is still an emerging field (Rothstein 2011). To crudely quantify this, we performed a simple search on Web of Knowledge and retrieved only 559 “epigenetic therapy” publications in 2012 as opposed to 36,328 for “chemotherapy” (wokinfo.com). The proposed second generation epigenetic therapies could overcome the hurdles of restoring methylation as opposed to only inhibiting methylation, and targeting specific genes as opposed to the entire genome. However, the issues of their delivery and cellular targeting still loom. It will require significant effort from basic researchers to determine the relevant mechanisms, and translational researchers to optimize the clinical strategies, but the path forward is promising.

Works Cited

http://www.pbs.org/wgbh/nova/body/epigenetic-therapy.html
http://www.dacogen.com/Content/Documents/Dacogen_PI.pdf
http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/CellularandGeneTherapy/default.htm
https://2013.igem.org/Team:Penn
www.wokinfo.com
http://protomag.com/assets/zinc-fingers-entry-fee
http://www.addgene.org/TALeffector/goldengateV2/

Campion, J., Milagro, F., & Martinez, J. A. (2010). Epigenetics and obesity. Prog Mol Biol Transl Sci, 94, 291-347.

Claus, R., Almstedt, M., & Lubbert, M. (2005). Epigenetic treatment of hematopoietic malignancies: in vivo targets of demethylating agents. Semin Oncol, 32(5), 511-520.

Ding, Y. B., Long, C. L., Liu, X. Q., Chen, X. M., Guo, L. R., Xia, Y. Y., et al. (2012). 5-aza-2'-deoxycytidine leads to reduced embryo implantation and reduced expression of DNA methyltransferases and essential endometrial genes. PLoS One, 7(9), e45364.

Feinberg, A. P., & Tycko, B. (2004). The history of cancer epigenetics. Nat Rev Cancer, 4(2), 143-153.

Grossniklaus, U., Kelly, B., Ferguson-Smith, A. C., Pembrey, M., & Lindquist, S. (2013). Transgenerational epigenetic inheritance: how important is it? Nat Rev Genet, 14(3), 228-235.

Jacquemont, S., Curie, A., des Portes, V., Torrioli, M. G., Berry-Kravis, E., Hagerman, R. J., et al. (2011). Epigenetic modification of the FMR1 gene in fragile X syndrome is associated with differential response to the mGluR5 antagonist AFQ056. Sci Transl Med, 3(64), 64ra61.

Kerkel, K., Schupf, N., Hatta, K., Pang, D., Salas, M., Kratz, A., et al. (2010). Altered DNA methylation in leukocytes with trisomy 21. PLoS Genet, 6(11), e1001212.

Kondo, T., Bobek, M. P., Kuick, R., Lamb, B., Zhu, X., Narayan, A., et al. (2000). Whole-genome methylation scan in ICF syndrome: hypomethylation of non-satellite DNA repeats D4Z4 and NBL2. Hum Mol Genet, 9(4), 597-604.

Laird, P. W. (2010). Principles and challenges of genomewide DNA methylation analysis. Nat Rev Genet, 11(3), 191-203.

Rice, J. C., Ozcelik, H., Maxeiner, P., Andrulis, I., & Futscher, B. W. (2000). Methylation of the BRCA1 promoter is associated with decreased BRCA1 mRNA levels in clinical breast cancer specimens. Carcinogenesis, 21(9), 1761-1765.

Rothstein, M. A., Cai, Y., & Marchant, G. E. (2009). The ghost in our genes: legal and ethical implications of epigenetics. Health Matrix Clevel, 19(1), 1-62.

Silverman, L. R., Demakos, E. P., Peterson, B. L., Kornblith, A. B., Holland, J. C., Odchimar-Reissig, R., et al. (2002). Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol, 20(10), 2429-2440.

Singh, P., Yang, M., Dai, H., Yu, D., Huang, Q., Tan, W., et al. (2008). Overexpression and hypomethylation of flap endonuclease 1 gene in breast and other cancers. Mol Cancer Res, 6(11), 1710-1717.

Szyf, M., Pakneshan, P., & Rabbani, S. A. (2004). DNA methylation and breast cancer. Biochem Pharmacol, 68(6), 1187-1197.

Weaver, I. C., Cervoni, N., Champagne, F. A., D'Alessio, A. C., Sharma, S., Seckl, J. R., et al. (2004). Epigenetic programming by maternal behavior. Nat Neurosci, 7(8), 847-854.