January 15, 2015 (Vol. 35, No. 2)

Richard A. A. Stein M.D., Ph.D.

The better we know the genome’s regulatory elements, the better we can design drugs that quell or quicken gene expression, as needed.

In the words of Nobel Prize winner Albert Szent-Györgyi, “Discovery consists of looking at the same thing as everyone else and thinking something different.” Could discovery as understood by Szent-Györgyi be happening in studies of the human genome?

For sure, lots of investigators are looking at the genome. It remains to be seen, however, whether some of these investigators are thinking uncommon (and uncommonly fruitful) thoughts.

Since the human genome was first sequenced, more and more scientists have shifted their focus from protein-encoding genes, which occupy about 2% of the genome, to potential regulatory elements. By focusing on protein-encoding genes, scientists improved their understanding of how mutations in proteins and changes in their expression levels are linked to the pathogenesis of many diseases.

Still, what about the noncoding 98% of the genome, the part once referred to as “junk DNA”? It is now known to contain key cis-regulatory elements, such as enhancers, which regulate the pattern of gene expression in a tissue-specific manner.

Noncoding DNA, once overlooked, now seizes the attention of discovery-minded scientists. They are investigating how heritable DNA that does not code for protein can regulate gene activity. This is the realm of epigenetics.

Epigenetics is fundamentally reshaping our insights into the biology of human disease. The field’s most inspired investigators, some of whom appear in this article, are developing novel concepts not only in basic research but also in drug design and development. These concepts are catalyzing some of the most dynamic and vibrant transformations in biology and medicine.


Epigenetics externally modifies DNA by turning genes “on” or “off.” The actual DNA sequence remains unchanged. [Alexey Rotanov/Shutterstock]

Enhancer Malfunctions in Cancer

“For the past 20 years, my lab has been studying the mixed lineage leukemia (MLL) gene’s translocation into different chromosomes,” says Ali Shilatifard, Ph.D., chairman of the department of biochemistry and molecular genetics at the Northwestern University Feinberg School of Medicine. This translocation process creates chimeric proteins as a result of fusions between the MLL proteins and partners sharing little sequence similarity.

Several of these fusion partners are transcriptional elongation factors. They have long been linked to carcinogenesis due to their involvement in transcriptional control.

Enter Dr. Shilatifard’s group, which has revealed that in Saccharomyces cerevisiae, Drosophila melanogaster, and mice, MLL-related proteins form complexes that can modify chromatin methylation. For example, the group showed that the mammalian MLL3/4-COMPASS-like complexes, which are frequently mutated in cancer, are responsible for the monomethylation of chromatin on enhancers, which is found within the 98% of the genome that does not code for proteins.

”Since mutations within the MLL3/4-COMPASS-like complexes are associated with a large number of human cancers, our studies indicate that maybe there is an enhancer malfunction in cancer,” explains Dr. Shilatifard. As enhancers are responsible for the tissue-specific regulation of gene expression, this could explain the involvement of certain point mutations in the tissue-specific development of malignancies.


Illustration of a DNA molecule that is methylated at the two center cytosines. [Christoph Bock (Max Planck Institute for Informatics) (Own work) [CC BY-SA 3.0 (http://creativecommons.org/licenses/by-sa/3.0)] via Wikimedia Commons Design]

Therapeutic Uses of Noncoding RNAs

“We been developing technologies to change epigenetic regulation in aberrant states, cancer being one of them, and we have a longstanding interest in other developmental disorders,” says Jeannie T. Lee, M.D., Ph.D., professor of genetics and pathology at Harvard Medical School and investigator at the Howard Hughes Medical Institute.

Among noncoding RNA species in mammals, long noncoding RNA, which exists in much greater numbers than proteins, has attracted considerable attention. Even though knowledge about long noncoding RNAs has significantly expanded over the past five years, their functions are still incompletely understood. “We suspect that most long noncoding transcripts regulate the epigenetic state of nearby genes,” speculates Dr. Lee.

In 2011, Dr. Lee co-founded a company, RaNA Therapeutics, to develop epigenetic therapies that upregulate and activate specific proteins that are downregulated in disease states. “The idea is to harness some of the long noncoding RNAs to treat human disease, and what is unique about this approach is that as opposed to RNAi, we are upregulating gene expression,” explains Dr. Lee.

A few years ago, Dr. Lee and colleagues discovered that a noncoding RNA encoded on the murine X chromosome targets the polycomb repressive complex PRC2, which represses gene expression at several thousands of genomic sites, with Ezh2 serving as the RNA-binding subunit during this interaction. This process contributes to the silencing induced by Xist, a noncoding gene from the murine X-inactivation center.

“Our idea was to interfere with the recruitment of PRC2 to this long noncoding RNA and increase the expression of tumor suppressors and other genes,” points out Dr. Lee. Long noncoding RNA molecules that interact with PRC2 are tethered to the chromosome through RNA polymerase II and, as a result, individual mRNA molecules in the vicinity are silenced.

“At RaNA Therapeutics, we are trying to change the expression of genes of interest by targeting long noncoding RNAs,” states Dr. Lee. She adds that doing so may prevent the recruitment of chromatin complexes to the disease genes of interest.

By using synthetic oligonucleotides to specifically block the binding between long noncoding RNA molecules and PRC2, scientists at RaNA are seeking to de-repress these mRNA molecules to modulate aberrant gene expression patterns. This technology platform is based on a proprietary database of tens of thousands of long noncoding RNA and other noncoding RNA species.

Unlocked Nucleic Acid Technologies

“What sets us apart is our capability to control gene expression in both directions, as our technologies allow us to knock down diseased genes with small interfering RNA or to upregulate gene expression with messenger RNA,” says Joseph E. Payne, president and CEO of Arcturus Therapeutics. One of the epigenetic technologies developed by the company’s investigators is unlocked nucleic acid (UNA) oligonucleotides. “Incorporating UNA into an RNA molecule captures very important and significant pharmaceutical benefits,” asserts Payne.

Another Arcturus technology, Lipid-enabled and Unlocked Nucleic Acid modified RNA delivery technology (LUNAR™), uses a biodegradable and biocompatible lipid-enabled delivery technology that was validated in primates for single- and double-stranded RNA molecules. “We combine our LUNAR delivery technology with our UNA Oligomer™ chemistry to produce outstanding drug products for orphan diseases,” continues Payne. “In addition to pursing rare diseases internally, we license our technologies to investigators pursuing other diseases.”

One of the company’s lead therapeutic compounds is a LUNAR-formulated UNA Oligomer siRNA. It targets transthyretin and is being developed for transthyretin-mediated amyloidosis.

Arcturus has licensed its UNA Oligomer technology to two companies, Arrowhead and Tekmira, for treating alpha 1-antitrypsin deficiency-associated liver disease and hepatitis B, respectively. “Tekmira and Arrowhead are strong companies in the RNA community,” states Payne. “As they effectively drive their UNA programs forward, they will further validate our UNA Oligomer chemistry platform.”

Arcturus is also emerging at the forefront of epigenetic research for another development. “We are currently a world leader in allelic selectivity,” says Payne. Unlike most current approaches in the small interfering RNA space, which involve knocking down the alleles from both parents, Arcturus has the ability to selectively reduce the expression of the diseased allele, while retaining the expression of the healthy allele. “We have the ability to achieve greater than 100-fold allelic selectivity,” insists Payne. “This is one of the ways we differentiate ourselves from others in the RNA community.”

These technologies allow the modulation of different types of RNA molecules or of multiple types of RNA species concomitantly. Delivering the technology, however, promises further challenges. “The technologies that we currently have can be delivered intravenously, subcutaneously, intramuscularly, and intrathecally, but what we have not seen a lot of is oral delivery,” admits Payne.

Delivering molecules into the central nervous system is currently performed by intrathecal injections. “It would be useful,” notes Payne, “to have an intravenous formulation that is engineered to cross the blood-brain barrier.”

Small Molecule Inhibitors

“We are dedicated to small molecule research mainly in oncology, and we focus on epigenetics,” says Patrick Trojer, Ph.D., senior director and head of biology at Constellation Pharmaceuticals. Dr. Trojer oversees work on three classes of epigenetic regulators: demetylases, methyl transferases, and chromatin binders, particularly bromodomain-containing proteins, a family of histone acetyl lysine recognition binders.

“As part of these efforts, we are trying to understand how small molecule inhibitors in the epigenetic space translate into transcriptional changes and efficacy in in vitro and in vivo cancer models,” elaborates Dr. Trojer. The first BET inhibitor is currently in a Phase I clinical study, and several EZH2 inhibitors are in preclinical studies.

Recently, Dr. Trojer’s team and collaborators from Promega reported the identification of highly potent and selective EZH2 inhibitors that were able, at pharmacologically relevant doses, to selectively affect the turnover of trimethylated, but not monomethylated, H3K27 residues. The efficacy of these small molecule inhibitors was not limited to lymphomas containing mutant EZH2, but also extended to non-Hodgkin lymphoma models harboring the wild-type protein.

Dr. Trojer indicates that a key challenge is to identify the next best targets behind the existing first-generation epigenetic targets, and he adds that “there is a need for a lot more research to unveil additional, new targets.”

“Recent evidence has shown that genetic mutation of chromatin regulators is quite common in cancer, and understanding these perturbations will open the potential for therapeutically manipulating them,” says Charles W. Roberts, M.D., Ph.D., associate professor of pediatrics at Harvard Medical School.

Studies in Dr. Roberts’ group are focusing on the SWI/SNF chromatin remodeling/tumor suppressor complex. “We had been interested in trying to understand both the normal function of this complex and the mechanisms by which mutations in its subunits drive cancer formation,” recalls Dr. Roberts. About 15 years ago, a subunit of this complex was found to be mutated in malignant rhabdoid tumors, a very aggressive form of childhood malignancy.

As the first known link between an ATP-dependent chromatin remodeling and cancer, this finding about the SWI/SNF complex offered the potential to reveal novel mechanistic and therapeutic insights into malignancy. “In addition, history has taught us that genes that are mutated in early-onset pediatric cancers are often important in other cancers,” observes Dr. Roberts.

In the first studies on this complex, Dr. Roberts and colleagues revealed that 100% of conditional knockout mice developed cancer, with a median onset of 11 weeks. “This was the fastest reported tumor onset following inactivation of a single gene,” says Dr. Roberts. “Cancer development occurred twice as fast as with p53 inactivation.”

Approximately 20% of all pediatric and adult human cancers are now known to harbor mutations in the SWI/SNF subunit. “This complex is the most frequently mutated chromatin regulator,” states Dr. Roberts.

Dr. Roberts and colleagues showed that, despite the very rapid onset of the human cancers in the SNF5 mutation background, these malignancies are remarkably genomically simple. “Even though we are beginning to understand how perturbations occur in the chromatin structure,” adds Dr. Robers, “we are still at the dawn of understanding how chromatin alterations perturb a cell and how to intervene.”

Identifying Chromatin Regulators

“One major area of epigenetics research concerns how chromatin regulators are involved in the pathogenesis of cancer,” says Christopher Vakoc, M.D., Ph.D., assistant professor at the Cold Spring Harbor Laboratory. As part of their work on understanding how chromatin regulators shape cancer pathogenesis, Dr. Vakoc and colleagues found that inhibiting BRD4, a BET bromodomain protein, with a small molecule, JQ1, represents a promising therapeutic strategy in acute myeloid leukemia.

In cells from patients with acute myeloid leukemia, including patients with relapsed or refractory disease, JQ1 was able to inhibit growth at submicromolar concentrations. “A major advance that will happen over the next few years will be the emergence of a large number of next-generation epigenetic therapies that target chromatin regulators as their mechanism of action,” insists Dr. Vakoc.

In parallel with these new developments, it is anticipated that the emergence of resistance to therapy will be an obstacle in the clinic. “While resistance will inevitably emerge, the rules of resistance for epigenetic therapies might not be the same as the rules for other targeted cancer therapies,” warns Dr. Vakoc.

Previous articleShire Acquires NPS Pharma for $5.2B
Next articleBiogen Idec Snaps Up Convergence for Up-to-$675M