Cancer-aiding elements begin illuminating the genome’s “dark matter”

Abstract

When researchers set out to identify every protein-encoding gene in the human genome, initial estimates suggested that they might find up to 100,000. In reality, scientists have uncovered fewer than 20,000, which account for only 2% of the 3.1 billion letters of DNA.¹

The remaining 98% has been referred to as “junk DNA” or “dark matter,” which reflects the once-common assumption that it was little more than genomic filler. An estimated 14,000 pseudogenes, or defective copies of genes, may fall into that category, albeit with some exceptions. Recent studies, however, have cast a new spotlight on a menagerie of other molecules, including some with a lineage far older than the human species, that may play significant roles in the development and progression of cancer and other diseases.

Martin Taylor, MD, PhD, assistant professor of pathology and laboratory medicine at the Brown University Center on the Biology of Aging and the Legorreta Cancer Center in Providence, Rhode Island, began studying one kind of genomic dark matter, LINE-1 retrotransposons, approximately 15 years ago. “The overwhelming evidence says that these are just parasitic sequences that we’ve been evolving with for billions of years,” he says of the mobile elements. “They’re actually older than multicellular organisms.” The virus-like, self-copying sequences, in fact, may have given rise to viruses.

Astoundingly, LINE-1 has effectively written at least one-third of the human genome through its copy-and-paste mechanism. Most of its own 500,000 copies are “fossils” from when humans mutated and defeated the once full-length retrotransposons. However, an estimated 6000 elements are more or less still intact; of those, maybe 100–150 are still functional and capable of hopping around and inserting themselves into other sequences, Dr Taylor says. They are repressed in healthy tissues but can become activated in the event of diseases such as cancer and autoimmune diseases.

By sequencing cancers, researchers have found that LINE-1, on rare occasion, can insert itself into a tumor suppressor gene such as APC or P10 and promote cancer. Work by Dr Taylor and others suggests that LINE-1 also may contribute to cancer in other ways. Those cancer-abetting mechanisms, he says, “have to do with the fact that when LINE-1 gets turned on, the cell thinks it’s infected with a virus—or at least has a virus-like response—and that causes inflammation.” The defensive response, research suggests, can manipulate the tumor microenvironment and alter cell signaling pathways to promote carcinogenesis.

In addition, Dr Taylor says, “The transposons cause a huge amount of DNA damage, and the thinking in the field now is that they may contribute to the chromosomal instability that we see as a hallmark of cancer.” He sees LINE-1 not as a classic cancer driver but rather as an accelerator that more broadly contributes to cancer development and progression, though the extent of that contribution in humans is not yet known.

The evolutionary link to viruses fortuitously suggests a tantalizing target for which drugs already exist. Scientists found that one of the two proteins encoded by intact LINE-1 (ORF2p) is a reverse transcriptase enzyme related to the one in HIV-1. Sure enough, Dr Taylor and his colleagues showed that some reverse transcriptase inhibitors used as HIV-1 antivirals can partially block LINE-1 in cell cultures; this means that they may help to tamp down LINE-1–associated cancer activity.² The researchers now are hoping to assess whether HIV reverse transcriptase inhibitors might inhibit LINE-1 and slow cancer development and progression in high-risk patients.

Early indications of at least a biochemical response to such inhibitors have helped to fuel commercial ventures aimed at developing small molecule inhibitors that target mobile elements activated in cancer, autoimmunity, and other diseases. “The research has gone from being a curiosity to something that has gotten a lot of interest in a very short amount of time, which has been very exciting for us,” Dr Taylor says.

Another collaborative research effort has shown that LINE-1 expression is pervasive in cancer tissue but not in normal tissue, and a single-molecule test can detect the transposon’s other encoded protein (ORF1p) in the blood of patients with cancer—especially those with ovarian, colon, or gastroesophageal malignancies.³ The proof-of-concept study was bolstered by an independent discovery that LINE-1 activation in lung, pancreatic, ovarian, liver, and esophageal cancers provokes the immune system early on to produce telltale antibodies targeting both of the retrotransposon’s proteins.⁴ If the findings bear out in larger patient cohorts with early-stage cancers, Dr Taylor believes that the research could open the door to an ultrasensitive, pancarcinoma early detection test.

In early results, Dr Taylor’s laboratory found that the relative ORF1p levels also correctly predicted whether patients with gastroesophageal cancer would respond positively to chemotherapy, radiation, or immunotherapy and correlated well with the disease prognosis.

Beyond the iconic cancer-associated mutations in genes such as p53 and KRAS, Dr Johnson says that evidence is accumulating for the idea that lower frequency mutations in some lncRNAs may promote tumor growth as well. A far greater number of lncRNAs, however, appear to change their activity levels in response to the widespread cellular disruptions accompanying tumor formation, and this shift results in patterns that are characteristic of different tumors. “Some lncRNAs are rather universal across all tumors, and other ones are very specific to different tumor types in their activity,” he says.

In cancer, multiple elements can be dysregulated simultaneously. “The big challenge is to say amongst all these things which we know are changing in cancer, which of those are changing in a way that’s actually causative to the disease rather than simply a consequence of it?” he says. “What you really want to do when you’re looking for therapy targets is to find things that are disease-causing agents.”

The arrival of the powerful gene-editing tool known as CRISPR-Cas9 gave Dr Johnson’s laboratory a new way to separate correlation from causation. The tool, he says, “enables us to perturb these lncRNA genes in cancer cells and directly ask the question, ‘Are these lncRNAs positive in the growth or the survival of cancer cells?’”

Using the system, the laboratory has precisely deleted various lncRNA genes in dishes of cultured tumor cells, in which each starting cell harbors a different deletion. Unique genetic labels attached to the lncRNAs can point out which ones the cancer cells depend upon the most because a deletion that kills the host cell will knock the corresponding genetic tag from the pooled DNA sequences of the cancer cells that remain. By its absence in the final census, the lncRNA gene signifies its importance to tumor survival.

When the laboratory and their collaborators applied the approach to non–small cell lung cancer, they identified dozens of potential therapeutic targets.⁵ Dr Johnson and his colleagues have begun focusing on two of the most promising and are exploring the use of antisense oligonucleotides, or designer genetic sequences that behave like a drug by binding to and degrading the target lncRNAs. So far, initial tests suggest that the targeted deactivation provokes a strong inhibitory effect on a variety of lung cancer cells; as with LINE-1–based therapeutics, pharmaceutical companies are beginning to take notice and fund cancer drug design efforts.

If the initial ventures pan out, the applications could continue to grow. Given their ability to disrupt DNA sequences across the genome, for example, Dr Taylor says that transposable elements could be linked to a much wider repertoire of anatomical variations, defects, and diseases than previously believed. “Quite a bit of normal biology may in fact be due to LINE-1 or the other parts of our genome that we don’t yet understand,” he says.