Yale researchers have identified an unusual molecular process in normal tissues that causes RNA molecules produced from separate genes to be clipped and stitched together. The discovery that these rearranged products exist in normal as well as cancerous cells potentially complicates the diagnosis of some cancers and raises the possibility that anti-cancer drugs like Gleevec could have predictable side effects.The work is reported in the journal Science.

“Our findings are surprising because we identified in normal cells certain types of gene products— so called chimeric RNAs and proteins—thought to be found only in cancerous cells or in cells on their way to becoming cancerous,” said Jeffrey Sklar, professor of pathology and laboratory medicine at Yale School of Medicine, and senior researcher on the study.

Chimeric proteins are considered to be key factors that drive many forms of cancer. They arise from chromosome abnormalities in which segments of the chromosomes are rearranged. At the sites where chromosome segments reattach, genes fuse giving rise to chimeric RNA, which in turn is used to construct the chimeric protein. Gleevec, a highly successful new anti-cancer drug, was developed to target the chimeric protein product of one such gene fusion.

Sklar’s group earlier discovered that a particular gene fusion, with its associated chimeric RNA and protein, is the probable cause of certain endometrial cancers. Unexpectedly, they also found the same chimeric RNA and protein in healthy uterine tissue — where the chromosomes and genes showed no abnormalities.

“Extensive experiments on the normal tissues and cultured cells from those tissues indicated to us that a previously little-known process, the direct splicing together of two RNAs from separate genes—or trans-splicing—is responsible for producing the chimeras,” said Sklar.

They also found that level of the chimeric molecules in normal cells was decreased by elevated estrogen and increased by reduced oxygen — conditions that control the synchronized cyclic behavior of normal cells that line the inside of the uterus.

These observations suggest that trans-splicing between the RNAs might be common in other normal tissues, because gene fusions have been identified in cancers that arise in many tissues.

“These findings may bring new insights into how cancers operate. It seems that rather than scrambling chromosomes to invent new genes, cancers mimic normal cellular processes, but in an exaggerated and unregulated fashion. You might say that cancers are clever but not very original,” said Yale Research scientist Hui Li, lead author of the paper.

According to the researchers, these results indicate that caution should be exercised in using chimeric gene products as markers for cancer, as is widely done now in cancer diagnosis. Additionally, cancer drugs that target products of chromosomal abnormalities may have varying degrees of toxicity because those same targets may be present in normal cells due to the trans-splicing of RNA.

Expanding on prior research performed at the University of Pennsylvania, Penn biologists have determined that faulty RNA, the blueprint that creates mutated, toxic proteins, contributes to a family of neurodegenerative disorders in humans.

Nancy Bonini, professor in the Department of Biology at Penn and an investigator of the Howard Hughes Medical Institute, and her team previously showed that the gene that codes for the ataxin-3 protein, responsible for the inherited neurodegenerative disorder Spinocerebellar ataxia type 3, or SCA3, can cause the disease in the model organism Drosophila. SCA3 is one of a class of human diseases known as polyglutamine repeat diseases, which includes Huntington’s disease. Previous studies had suggested that the disease is caused largely by the toxic polyglutamine protein encoded by the gene.

The current study, which appears in the journal Nature, demonstrates that faulty RNA, the blueprint for the toxic polyglutamine protein, also assists in the onset and progression of disease in fruit fly models.

“The challenge for many researchers is coupling the power of a simple genetic model, in this case the fruit fly, to the enormous problem of human neurodegenerative disease,” Bonini said. “By recreating in the fly various human diseases, we have found that, while the mutated protein is a toxic entity, toxicity is also going on at the RNA level to contribute to the disease.”

To identify potential contributors to ataxin-3 pathogenesis, Bonini and her team performed a genetic screen with the fruit fly model of ataxin-3 to find genes that could change the toxicity. The study produced one new gene that dramatically enhanced neurodegeneration. Molecular analysis showed that the gene affected was muscleblind, a gene previously implicated as a modifier of toxicity in a different class of human disease due to a toxic RNA. These results suggested the possibility that RNA toxicity may also occur in the polyglutamine disease situation.

The findings indicated that an RNA containing a long CAG repeat, which encodes the polyglutamine stretch in the toxic polyglutamine protein, may contribute to neurodegeneration beyond being the blueprint for that protein. This raised the possibility that expression of the RNA alone may be damaging.

Long CAG repeat sequences can bind together to form hairpins, dangerous molecular shapes. The researchers therefore tested the role of the RNA by altering the CAG repeat sequence to be an interrupted CAACAG repeat that could no longer form a hairpin. Such an RNA strand, however, would still be a blueprint for an identical protein. The researchers found that this altered gene caused dramatically reduced neurodegeneration, indicating that altering the RNA structure mitigated toxicity. To further implicate the RNA in the disease progression, the researchers then expressed just a toxic RNA alone, one that was unable to code for a protein at all. This also caused neuronal degeneration. These findings revealed a toxic role for the RNA in polyglutamine disease, highlighting common components between different types of human triplet repeat expansion diseases. Such diseases include not only the polyglutamine diseases but also diseases like myotonic dystrophy and fragile X.

The family of diseases called polyglutamine repeat disorders arise when the genetic code of a CAG repeat for the amino acid glutamine stutters like a broken record within the gene, becoming very long. This leads to an RNA — the blueprint for the protein — with a similar long run of CAG. During protein synthesis, the long run of CAG repeats are translated into a long uninterrupted run of glutamine residues, forming what is known as a polyglutamine tract. The expanded polyglutamine tract causes the errant protein to fold improperly, leading to a glut of misfolded protein collecting in cells of the nervous system, much like what occurs in Alzheimer’s and Parkinson’s diseases.

Polyglutamine disorders are genetically inherited ataxias, neurodegenerative disorders marked by a gradual decay of muscle coordination, typically appearing in adulthood. They are progressive diseases, with a correlation between the number of CAG repeats within the gene, the severity of disease and age at onset.

In addition to Bonini, researchers whose work contributed to this study are Ling-Bo Li, formerly in the Department of Biology at Penn and now with the Department of Biochemistry at the University of Utah, and Zhenming Yu and Xiuyin Teng of the Department of Biology at Penn and the Howard Hughes Medical Institute.

Funding for this study was provided by the National Institute of Neurological Disorders and Stroke.