Testing thousands of RNA enzymes helps find first 'twister ribozyme' in mammals

The research team tested the activity of over 2,600 different RNA sequences predicted to belong to a class of RNA enzymes called "twister ribozymes," which have the ability to cut themselves in two. Approximately 94% of the tested ribozymes were active, and the study revealed that their function can persist even when their structure contains slight imperfections. The research team also identified the first example of a twister ribozyme in mammals, specifically in the genome of the bottlenose dolphin.

A paper describing the study appeared online today, Nov. 5, in the journal Nucleic Acids Research.

"Whereas DNA is a double-stranded molecule that typically forms a simple helical structure, RNA is single stranded and can fold back on itself, forming diverse structures, including loops, bulges and helixes," said Phil Bevilacqua, distinguished professor of chemistry and of biochemistry and molecular biology in the Eberly College of Science at Penn State and the leader of the research team. "The function of RNA enzymes is based on these structures and they have been categorized into several different classes. We chose to focus on so-called 'twister ribozymes' because one of their functions is to cleave themselves in two, which we can see by determining their genetic sequence."

Prior to this study, about 1,600 twister ribozymes had been proposed based on their genomic sequence and predictions of their structure, but only a handful had been experimentally validated. The team developed an experimental pipeline that allowed them to assess the self-cleaving activity of thousands of these ribozymes in a single experiment, which they call a "Cleavage High-Throughput Assay," or CHiTA. They also identified approximately 1,000 additional twister ribozyme candidates by meticulously hand-searching the genomic context around a short, highly conserved sequence shared by many of the ribozymes in the genomes of 1,000s of organisms.

CHiTA relies on two key factors. One is a recently developed technology called "massively parallel oligonucleotide synthesis," or MPOS. MPOS gives the research team the ability to design and then purchase thousands of diverse ribozyme sequences in the form of small pieces of DNA, all in a single vial. Each of the sequences they design has as its core one of the 2,600 predicted ribozyme sequences. The researchers then add short bits of DNA at either end that allow them to make copies of the DNA and transcribe it into RNA to test its activity.

"With MPOS, we can simply create a spreadsheet with the sequences that we are interested in, send it to a commercial vendor, and they send us back a tube that contains a small amount of each sequence," said Lauren McKinley, a graduate student at Penn State at the time of the research, who recently earned a doctorate, and first author of the paper. "For CHiTA, we need lots of each sequence, so we add bits of DNA to each end of the sequences that allows us to make millions of copies of each using a technique called PCR, but these additional bits could impact our ability to test the ribozymes' functions."

The second key factor for CHiTA helps to overcome this hurdle by removing these added bits of sequence using a protein -- called a restriction enzyme -- that cuts DNA at specific short sequences called recognition sites. However, most restriction enzymes cut the DNA somewhere near the middle of their recognition sites, leaving a portion of the recognition site sequence attached to the two resulting DNA fragments that could still impact ribozyme structure and function.

"We found a restriction enzyme that cuts the DNA a short distance from its recognition site, so we could design our sequences such that it would cut leaving no trace of additional DNA," McKinley said. "This way we could ensure that we were assessing the function of the precise sequence of the ribozymes."

In the lab, the team first makes additional copies of the sequences ordered through MPOS, trims any additional DNA with the restriction enzyme, and then they can transcribe RNA from the DNA sequences. If any of the sequences codes for active ribozymes, as the RNA is produced, they will quickly fold into their functional structure and cleave themselves. The researchers can then collect the RNA and convert it back to DNA -- called cDNA -- so that they can read its sequence to see if it is full length or if it's been cleaved.

"When we sequence the cDNA, we can see how much of the RNA, if any, has been cleaved as an indicator of ribozyme activity," McKinley said. "For about 94% of the sequences we tested, a significant portion of the RNA was cleaved. In fact, the percentage of each active ribozyme that was cleaved is quite similar to earlier efforts that tested ribozymes individually."

The team then looked at predicted structures of the sequences they tested and saw that many of them had slight variations or imperfections compared to the canonical twister ribozyme structure, yet they still self-cleaved. This suggested to the researchers that the function of the ribozymes is very tolerant to slight structural changes, meaning they could still operate even if not perfectly formed.

Tolerance of imperfections suggests that there may be more twisters hidden in nature that wouldn't be found using the original search parameters, according to the team. In fact, new descriptors based on sequences in this study led the research team to discover the first mammalian twister ribozyme, which was identified in the genome of the bottlenose dolphin.

"Understanding these types of tolerances to sequence and structural variation in ribozymes will help us to design new and rigorous ways to identify them," Bevilacqua said. "Our current knowledge of ribozyme function is mostly based on chemistry, and we are only just beginning to learn about their role in biology. Being able to test their activity in a large-scale assay like CHiTA will hopefully accelerate our ability to find new ribozymes and better understand the role they play in the cell. All of this can also help us to work backwards in time to see what could have been possible when RNA's ability to do it all might have been a key to jumpstarting life on the early Earth."

In addition to Bevilacqua and McKinley, the research team at Penn State included McCauley O. Meyer, Aswathy Sebastian and Kyle J. Messina, all of whom were graduate students at the time of the research and have since completed their doctorates; former undergraduate student Benjamin K. Chang; and Research Professor of Bioinformatics Istvan Albert. The National Institutes of Health and the Penn State Huck Institutes of the Life Sciences funded this research.