UMass Amherst Chemists Publish ‘Breakthrough’ Article on RNA Sequencing

Craig Martin
Craig Martin

AMHERST, Mass. – For years, researchers who make large quantities of RNA for biomedical and biotech studies have run into a problem where their RNA synthesis produces sequences longer than expected, known as “nontemplated additions,” says chemist Craig Martin at the University of Massachusetts Amherst, an RNA polymerase expert with 30 years of experience in the use of a model system known as T7 RNA polymerase.

Now he and colleagues including first author and doctoral student Yasaman Gholamalipour report that they have figured out what’s happening to cause RNA sequences to replicate inaccurately in high-yield situations.

Details of the work supported by the National Science Foundation appear in Nucleic Acids Research, where journal editors tagged the paper as a “breakthrough article” that presents “high-impact studies answering long-standing questions in the field of nucleic acids research and/or opening up new areas and mechanistic hypotheses for investigation,” and representing “the very best papers published at NAR.”

Martin says, “What we’ve done is really interesting to a lot of people making RNA strands to deliver to cells to turn genes on and off in biomedical and biotechnology applications. To get the large amounts they need for their studies, they engineer DNA to tell RNA polymerase to start copying at a certain spot, but instead of getting a perfectly copied sequence of, let’s say, the 100mer they asked for, they were getting longer sequences, say 110 or 105. And the more they tried to make, the more trouble they would have, and no one knew why.”

The phrase “100mer” is short for a 100-base oligomer, a bit of RNA manufactured as a single-stranded molecule, he explains. Some researchers can use RNA of this length or shorter, but other approaches need RNAs thousands of bases long.

Regardless of the length, Martin says he, Gholamalipur and colleagues discovered that as the RNA polymerase releases many copies of the correct sequence in reactions lasting many hours, one of those sequences may loop back on itself like the tip of a whip and re-bind with any available RNA polymerase, with the polymerase now adding extra bases using the RNA itself as a template. This creates a short, double-stranded RNA segment, and the process can repeat to create longer double-stranded segments, he adds.

“We used to think that if the RNA polymerase came to the end of the DNA sequence it was copying, it might sit there and start adding stuff randomly in a non-templative manner, making extra sequences that were longer than expected, but until recently there were no tools to explore this,” Martin notes. “Then about 20 years ago a new genome sequencing tool came out of the Human Genome Project that can sequence millions of bases at the same time, and now it’s available for use with RNA.”

He says, “Our work with this new tool shows that the correct RNA sequences are getting made, but as they accumulate, the system will start building up the second reaction, the incorrect one. It’s a statistical probability related to pushing the system to make large amounts of RNA strands.”

His lab has traditionally used five-minute reactions, for example, and “if you started out to make small amounts, it wasn’t a problem,” he adds. “We show that high-yield conditions are part of the problem, which itself points toward ways to get around it. So, based on this work we are now developing ways to make large amounts of accurately copied RNA. I think this is one reason the journal labeled this work as a breakthrough, because our work suggests a way to solve the problem.”

The incorrect “self priming” that results in double-stranded, impure RNA fragments has gained a lot of attention, the chemist explains, because many biomedical researchers are attempting to treat genetic diseases by delivering corrective RNA sequences to cells, and the impurities posed by double-stranded segments risk triggering an immune response.

Martin says, “Even if they make it 98 percent correct, the 2 percent might cause trouble because delivering such improper RNA sequences to cells as part of a new therapy treatment could cause the immune response. For some people, this might cause just a fever or a rash, but for others it might be lethal, and they can’t use that drug. If you have a genetic disease and you will be taking that drug for a lifetime, the problem is not trivial.”

He adds, “At my lab we have always been interested in what we call fundamental mechanistic enzymology. It’s biology that won’t light the world on fire now, but we set things up for others to make advances in the future. Maybe now, we’re doing a bit of both.”