Writing the playbook for RNA-modifying enzymes and their roles in regulating translation of the genetic code
Picture your favorite chocolate chip cookie. Is it soft and chewy? Thin and crispy? Would you still eat it if it was slightly burnt around the edges? While some of us happily accept any cookie, others may snub the slightly overbaked batch or reject oatmeal raisin when expecting chocolate chip. Enzymes do the same thing in cells.
Enzymes accelerate biochemical reactions by interacting with the molecules they modify within a catalytic pocket. Like a fickle cookie connoisseur, some enzymes are very choosy about the types of molecules they will accept, and the reasons for their selectivity are not always obvious. In a new study published in Cell Genomics, Nicole Martinez, assistant professor of chemical and systems biology and of developmental biology in the School of Medicine and institute scholar at Sarafan ChEM-H, asked why a particular enzyme, called a pseudouridine synthase, readily modifies some molecules while ignoring others that look very similar.
Pseudouridine synthases chemically modify RNA molecules, the molecular messengers that get copied from the genetic code and then used to create proteins that carry out the functions of life within your cells. The locations of these chemical modifications impact how the RNA message is interpreted and, therefore, how a cell behaves. When pseudouridine synthases malfunction, it can lead to certain cancers and neurodevelopmental disorders; but pseudouridine synthases are picky eaters. Decoding why they choose to modify some RNAs and not others could shed light on how these chemical alterations to RNA affect gene regulation in health and disease and could lead to new therapeutic opportunities.
“There are a plethora of options for where the RNA-modifying enzyme can modify within an RNA molecule, but how does it choose that site?” asked lead study author and chemical and systems biology PhD
student Becca Rodell. “We developed a targeted approach to look at specific sites, which allowed us to make quantitative comparisons about the sites that the enzyme really likes to modify — the enzyme’s favorite chocolate chip cookie, versus a site that the enzyme is only going to eat if it’s the only cookie around.”
Modifying RNA molecules of all shapes and sizes
To better understand the features that make certain RNA sites more likely to be modified, the team focused on one member of the pseudouridine synthase family, PUS7. Like DNA, RNA is composed of a four-letter genetic alphabet of molecules that are strung together in a particular sequence. One letter, specific to the RNA alphabet, is called uridine. Pseudouridine synthases, like PUS7, rearrange the chemical bonds between the atoms of uridine to create a modified letter called pseudouridine. Prior studies showed that PUS7 preferentially adds pseudouridines within a specific five-letter RNA sequence motif, but only a small fraction of these sequences across all RNAs are modified, suggesting that additional features, such as an RNA’s 3-D shape, must also influence enzyme activity. Unlike the characteristic DNA double helix, RNAs are single-stranded and can fold to adopt a variety of shapes that can be hard to predict.
To examine this more closely, Martinez and her team selected a subset of PUS7 modification sites containing the five-letter RNA sequence motif it preferentially modifies. They then developed a sequencing method to quantitatively measure what fraction of these sites were modified by the enzyme under different experimental conditions. What they found was surprising. Unlike some RNA-modifying enzymes that always choose RNAs with a particular folded shape, PUS7 modifies a diversity of RNA shapes.
“When we look at the predicted RNA structures of RNAs that were modified by PUS7, visually, we couldn’t see any unifying features. It took a lot of creative thinking to think about how we could look at this data in different ways to extract unifying features about structure,” said Rodell.
The unifying feature turned out to be the patterns of 2-D structure near the modification site and how exposed a pseudouridine modification site is when the RNA molecule is folded up on itself. Uridines that are easier for the enzyme to access were modified more frequently. Using molecules called antisense oligos that interfere with gene expression during the early stages of RNA processing, the team was able to leverage these newly learned rules to disrupt the structures of individual RNA sequences and prevent pseudouridine modification in places where it would normally occur. Antisense oligos have been clinically approved for the treatment of spinal muscular atrophy, but they have not yet been applied therapeutically for disrupting pseudouridine modification, offering a potential new approach to treat diseases driven by PUS7 dysregulation, something that the team is excited to continue exploring.
“Now that we’ve figured out the rules, we really want to show that we can manipulate or inhibit the modification of a specific mRNA in cells and see how that affects the cellular output,” said Martinez.
A possible role for pseudouridine in cell-type-specific genetic regulation
The second surprise came from the comparison of identical RNA molecules isolated in a test tube versus in their natural cellular environments. The team found that the modifications that happen in a test tube do not happen at the same frequency in cells, and different cell types have unique patterns of RNA modification across identical sets of RNA molecules with the same 3-D shape. This suggests that yet-unknown factors, beyond RNA sequence and shape, regulate PUS7 activity within specific cellular contexts.
“It’s exciting to see these differences in chemical modifications across cell types,” said Martinez. “Before, it wasn’t clear whether these are regulatory marks to control gene expression in different environments that the cell might encounter, but perhaps this is a way that our bodies use to tune what genes are expressed in response to specific cellular environments or stages of development.”
The biology surrounding RNA modifications is only beginning to be uncovered. Martinez and her team hope to generalize their technology to be able to discover the modification rules for other pseudouridine synthases across the genome, to explore how pseudouridine modifications at specific sites and in different cell types influence cell behavior, and to harness these discoveries for new therapeutic applications.
“There are thirteen different pseudouridine synthases in humans,” said Martinez. “And so far what we and others have learned is that they all recognize different sequences and different shapes. These tools can start to help us figure out the rules for these different enzymes when it’s not necessarily obvious from the sequence. Now that we understand at least some of the rules for how this particular enzyme works, we may be able to start predicting new modification sites in certain contexts, as well as understanding how these enzymes are regulated and dysregulated in disease.”
Nicole Martinez is also a member of Bio-X, the Stanford Cancer Institute, and the Wu Tsai Neurosciences Institute. Other Stanford co-authors include Ronit Jain, Hossein Shenasa, Matias Montes, and Nicolas Robalin. This research is supported by the National Institutes of Health, the Rita Allen Foundation, the David and Lucile Packard Foundation, and a VPUE Developmental Biology Undergraduate Research Grant.
