RNA modification influences thousands of genes: Revolutionizes understanding of gene expression

Now, researchers at Weill Cornell Medical College have made a discovery that once again forces us to rewrite our textbooks. This time, however, the findings pertain to RNA, which like DNA carries information about our genes and how they are expressed. The researchers have identified a novel base modification in RNA which they say will revolutionize our understanding of gene expression.

Their report, published May 17 in the journal Cell, shows that messenger RNA (mRNA), long thought to be a simple blueprint for protein production, is often chemically modified by addition of a methyl group to one of its bases, adenine. Although mRNA was thought to contain only four nucleobases, their discovery shows that a fifth base, N6-methyladenosine (m6A), pervades the transcriptome. The researchers found that up to 20 percent of human mRNA is routinely methylated. Over 5,000 different mRNA molecules contain m6A, which means that this modification is likely to have widespread effects on how genes are expressed.

"This finding rewrites fundamental concepts of the composition of mRNA because, for 50 years, no one thought mRNA contained internal modifications that control function," says the study's senior investigator, Dr. Samie R. Jaffrey, an associate professor of pharmacology at Weill Cornell Medical College.

"We know that DNA and proteins are routinely modified by chemical switches that have profound effects on their function in both health and disease. But biologists believed mRNA was simply an intermediate between DNA and protein," he says. "Now we know mRNA is much more complex, and defects in RNA methylation can lead to disease."

Indeed, as part of the study, the researchers demonstrated that the obesity risk gene, FTO (fat mass and obesity-associated), encodes an enzyme capable of reversing this modification, converting m6A residues in mRNA back to regular adenosine. Humans with FTO mutations have an overactive FTO enzyme, which results in low levels of m6A and causes abnormalities in food intake and metabolism that lead to obesity.

The researchers uncovered links between m6A and other diseases as well.

"We found that m6A is present in many mRNAs encoded by genes linked to human diseases, including cancer as well as several brain disorders, such as autism, Alzheimer's disease, and schizophrenia," says the study's lead investigator, Dr. Kate Meyer, a postdoctoral researcher in Dr. Jaffrey's laboratory.

"Methylation in RNA is a reversible modification that appears to be a central step in a wide variety of biological pathways and physiological processes," she says.

The first time that m6A was detected in mRNA was in 1975, but at the time scientists were unsure whether this finding was a result of contamination by other RNA molecules, Dr. Jaffrey says. Over 90 percent of RNA is either transfer RNA (tRNA) or ribosomal RNA (rRNA), cellular workhorses that are routinely modified.

But Dr. Jaffrey says he has always been interested in the idea that mRNA may be modified -- "DNA, proteins, other forms of RNA are modified, so why not mRNA?" he says -- so he and investigators in his laboratory developed a technique to help them uncover methylation in mRNA taken from both mouse and human samples.

They used two different antibodies that recognize and bind to m6A in mRNA in order to selectively isolate the mRNAs that contain m6A. By subjecting these mRNAs to next-generation sequencing, they were able to identify the sequence of each individual mRNA they had isolated. Co-authors Dr. Christopher Mason and Dr. Olivier Elemento, assistant professors from the Department of Physiology and Biophysics and Computational Genomics in Computational Biomedicine at Weill Cornell Medical College, then developed computational algorithms to reveal the identity of each of these methylated mRNAs.

The Weill Cornell researchers don't know how the thousands of m6As they detected in humans work to control the function of mRNAs, but they do note that the m6As are located near "stop codons" in mRNA sequences. These areas signal the end of translation of the mRNA, suggesting that m6A might influence ribosomal function. "But we really don't know yet," says Dr. Mason, a co-lead investigator on the study. "It may allow other proteins to bind to mRNA, or subject these mRNAs to a whole new regulatory pathway. Our bioinformatics analyses are providing several hints about the possible impact of methylation on RNA function."

Indeed, in their study, the investigators have already found that m6A sites frequently occur in regions of mRNA that are highly conserved across several species of vertebrates. "This shows that m6A sites are not just important for humans, but rather are maintained under selection across hundreds of millions of years of evolution, and thus are likely of critical importance for all animals," Dr. Mason says.

"This is the first demonstration of an epitranscriptomic modification -- alterations in RNA function that are not due to changes in the underlying sequence," he adds.

"These findings are very, very exciting, and amazing, really, when you consider that mRNA has been around for so long and that nobody realized, in all this time, that they were being regulated in this way," Dr. Jaffrey says. "It was right under our noses."

In addition to investigating how m6A regulates mRNAs within cells, the researchers are now focused on identifying the enzymes and pathways that control mRNA methylation.

Their study already demonstrates that FTO is capable of reversing adenosine methylation and suggests that it acts on a large proportion of cellular mRNA. "FTO mutations are estimated to occur in one billion people worldwide and are a leading cause of obesity and type 2 diabetes. Our studies link m6A levels in mRNA to these major health problems and identify for the first time the mRNAs which are potentially targeted by FTO," Dr. Meyer says.

The investigators are currently working to understand how defective regulation of m6A in patients with FTO mutations causes obesity and metabolic disorders, and they are also developing tests to rapidly identify compounds that inhibit FTO activity. These compounds are expected to inhibit the overactive FTO found in humans, potentially leading to novel therapeutics for diabetes and obesity.

Other study co-authors are Yogesh Saletore and Paul Zumbo, members of Dr. Mason's Integrative Functional Genomics laboratory, in the Department of Physiology and Biophysics at Weill Cornell.