Specialized RNA molecules could counter ALS neurodegeneration

Misshapen proteins cause a mess of trouble—particularly in neurodegenerative diseases. But a new study suggests it’s possible that giving them a little bit of extra support could keep them working correctly, and even reverse the damage they have caused.

The new research focuses on one such aberrant protein, TDP-43, which binds to RNA in the cell’s nucleus and is responsible for regulating thousands of human genes. If TDP-43 turns from a healthy, liquid-like phase into diseased, fibrous solid-like aggregates, its presence can be fatal. This protein is one of the key drivers of the diseases amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD)—a discovery first made by pioneering Penn Medicine scientists Virginia M.-Y. Lee, PhD, MBA and the late John Trojanowski, MD, PhD.

There are currently no cures for ALS or FTD, but that could change. In a new study published today in Science, researchers at the Perelman School of Medicine at the University of Pennsylvania reported short RNA molecules that could reverse TDP-43 aggregation and restore its function, an important advance toward RNA-based treatments for ALS and FTD.

“In these diseases, you're really fighting against two things: this nuclear loss of TDP-43 function—disrupting RNA splicing and processing—and a cytoplasmic gain of toxic function through protein aggregation,” said James Shorter, PhD, a professor of Biochemistry and Biophysics at the Perelman School of Medicine at Penn and a senior author of the study. For nearly two decades, concurrent with and following on Lee and Trojanowski’s discoveries, Shorter has studied the causes and mechanisms of TDP-43’s misfolding and sought methods to prevent and reverse it.

When TDP-43 misfolds and collects in the cytoplasm, it no longer performs its normal function regulating the RNA in the nucleus and forms toxic aggregates. But Shorter and his colleagues found that short RNA chaperones could reverse TDP-43 aggregation and restore its function.

Short RNA chaperones are specialized nucleotide sequences that can stabilize the area on a protein structure that binds RNA, also known as its RNA-recognition motifs (RRMs). By doing so, the chaperone helps the protein to fold into its correct shape and restore its correct function.

In an earlier study, Shorter and his colleagues showed that a short RNA sequence, called Clip34, binds to TDP-43’s RRMs and keeps the protein in its healthy, soluble form. But the mechanism underlying that process was unclear, said Katie Copley, PhD, a former graduate student in the Shorter Lab and the lead author of the new Science study.

To find out the mechanism of action, Copley used a highly sensitive technique called hydrogen-deuterium exchange mass spectrometry to map the structure of TDP-43 as it interacts with other molecules. By measuring the rate at which the hydrogen atoms on the TDP-43 protein exchange with the deuterium atoms in the solvent, she could infer properties such as how stable the proteins were and where the molecules bind.

“It's really the ideal technique to get a sense of structural dynamics of the whole protein,” Shorter said. If the exchange happens quickly, it means that the proteins are relatively unstable. If the process is slow, it suggests a more stable structure.

The researchers found that in binding to TDP-43, the Clip34 RNA stabilizes the site where the protein usually engages with RNAs—the RRMs—while destabilizing another area called the prion-like domain. This secondary area is known to drive protein misfolding that causes neurodegeneration.

“We think that the way the short RNA binding affects the structure of that prion-like domain is important for keeping TDP-43 soluble,” Copley said.

Right shape, right size—then the right molecule?

The Clip34 RNA chaperone consists of 34 nucleotides derived from the sequence of an mRNA that encodes the TDP-43 protein. “We knew it was a tight—but not too tight—binding RNA,” Shorter said. This balance appears critical: Clip34 stabilizes TDP-43’s structure without locking it up, so the protein can still bind its natural RNA targets and carry out its normal functions in the nucleus.

The size of Clip34 is also favorable for drug delivery, Shorter added. He points to an example of an FDA-approved antisense oligonucleotide drug, called Nusinersen, which has an 18-nucleotide sequence and can be administered to patients through spinal cord injection. “We were inspired by the antisense oligo approaches, which provide a clinical roadmap for us,” Shorter said.

Once they established Clip34’s mechanism of action, the researchers expanded their search for other short RNAs that could prevent and even reverse toxic TDP‑43 behaviors. They tested a panel of 14 TDP-43 variants, including the normal protein and multiple disease‑linked forms, against different short RNAs. From this work, they identified Malat1_start, another short RNA that effectively chaperones or assists all tested versions of TDP-43 to fold properly.

“We’re very excited about it because it is more broadly effective against diverse TDP-43 variants than Clip34,” Copley said.

Working with Chris Donnelly, PhD, at the University of Pittsburgh, the team used microscopy to show that Malat1_start reduces TDP-43 aggregates in the cytoplasm of human cells. Both Clip34 and Malat1_start also restored the ratio of the protein in the cytoplasm and nucleus of patient-derived neurons to the levels found in healthy controls. More importantly, once TDP-43 returns to the nucleus, both Malat1_start and Clip34 do not interfere with its RNA-processing function.

The potential to fully restore misfolded proteins to functioning

Other researchers are developing antisense oligonucleotides that correct specific RNA processing errors caused by TDP-43's absence from the nucleus, but those approaches only fix one or two issues and do not directly repair the misfolded TDP-43 protein itself—the upstream source of the problem.

“By getting the cytoplasmic TDP-43 back into the nucleus, we should be able to restore all of its functions. That's the attractive feature of our strategy,” Shorter said.

With the promise they saw in patient-derived neurons, the team, working with Brigid Jensen, PhD, at Thomas Jefferson University, proceeded with testing Malat1_start in a mouse model of ALS.

With one dose of the Malat1_start short RNA, the researchers saw a marked reduction in TDP-43 aggregation in the mouse. More surprisingly, Malat1_start largely prevented further loss of motor neurons driven by the disease. “That was pretty impressive,” Copley said.

The short RNAs not only prevented the toxic effect of TDP-43 but also reversed aggregation and TDP-43 dysfunction.

“As far as we know, nothing else has been shown to do this in a mouse model,” Shorter said.

The researchers are now testing other doses of the short RNAs and expanding their investigation into other ALS mouse models.

“The strength of our study is that we now have both the mechanistic and therapeutic framework for these short RNAs, all the way from their effects on pure proteins, to cell models and patient-derived neurons, and even mouse models,” Shorter said.

Protein misfolding is a feature of many other neurodegenerative diseases, such as Alzheimer’s and Parkinson’s diseases. 

“We have shown that this RNA‑chaperone approach also works for FUS protein in ALS/FTD, and we suspect it can be extended to other RNA-binding proteins that are getting into trouble, such as Tau in Alzheimer’s disease cases,” Shorter said.

As for Malat1_start, the team is now putting plans in place to advance this lead toward the clinic. “We’re really excited to advance these further,” Shorter said.