Subscribe to eTOC

Scientists Used Optogenetic Light to Create Aggregated TDP-43 Protein in ALS.
Then They Stopped the Toxic Protein from Accumulating.

Article In Brief

Researchers created a cellular model of TAR DNA-binding protein 43, the toxic protein that is the hallmark in amyotrophic lateral sclerosis and frontotemporal dementia and developed a technique to stop the accumulation of the toxic protein in the cell culture.

Scientists at the University of Pittsburgh School of Medicine have developed the first cellular model to study the aggregation of TAR DNA-binding protein 43 (TDP-43), the toxic protein that is the hallmark in amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD).

Using light pulses, an optogenetic tool, they were able to induce TDP-43 aggregation and to learn more about what makes the protein become toxic and accumulate in these diseases.

Normally, TDP-43 is involved with RNA metabolism (splicing and processing) but many studies have showed that the protein moves out of the nucleus and into the cytoplasm, where it clumps and becomes toxic to neurons. No one knows why and how this leads to the very different phenotypes of ALS and FTD.

In the experiments described in the February 27 online edition of Neuron, the investigators discovered that toxic forms of the protein do not have RNA, which is critical for cell survival. Once they understood this new molecular twist, the team developed an oligonucleotide designed to target TDP-43 specific RNA and stop the accumulation of the toxic protein in the cell culture.

The first study author Jacob Mann, a doctoral student, has been working on the experiments in the laboratory of Christopher J. Donnelly, PhD, assistant professor of neurobiology at University of Pittsburgh, and senior author of the paper. The Donnelly lab studies the molecular pathogenesis of ALS, FTD, and other neurodegenerative diseases.

Mann and Dr. Donnelly theorized that a trafficking problem is the common thread leading to protein clumps in many of these conditions and that treatments to rescue these defects might work in reducing symptoms.

Findings from the study could have implications for understanding these so-called TDP-43 proteinopathies and designing treatments that might help patients, the investigators told Neurology Today.

Study Methods, Findings

To do that, the researchers wanted to improve existing TDP-43 models and find a way to stop the toxic protein from accumulating. Mann began with a light-sensitive protein found in a mustard plant and tagged it to the TDP-43 protein. When introduced into human cells and then exposed to light in the dish, the team members were able to make TDP-43 bind to itself. (They also used florescent tags to monitor TDP-43.) The molecules started clumping together, forming TDP-43 inclusions, and reproducing in response to light.

Mann then started dissecting the newly aggregated proteins. He would break off pieces of the TDP-43 protein and study one part, then another. He would leave the toxic protein in the cell culture for days and study what happens.

He observed something that no one else had ever described in the literature: The only time TDP formed into these clumps was when RNA binding partners were missing. No one had ever suspected that this toxic form of TDP-43 did not have RNA in it. The question remained: Why was the RNA not there?

Mann and Dr. Donnelly conducted other studies to understand why RNA binding prevents TDP-43 from forming these inclusions. They observed that when TDP-43 was in the presence of RNA-containing structures in the cell, such as the nucleus where it normally resides or in stress granules in the cytoplasm, it was protected from clumping. This was a surprise since stress granules have been thought to be a problem in the disease and it suggests that these stress granules might protect TDP-43 from clumping since they have a lot of RNA, Dr. Donnelly explained.


“In essence, the added RNA acts like a decoy, mopping up extra TDP-43 and preventing it from causing toxicity.”



“They have reproduced features of TDP-43 pathology and survival of cells. This has phenotypic consequences.”


To investigate which drugs could stop this process, Mann turned to RNA oligonucleotides comprised of molecular sequences that TDP-43 interacts with. The oligonucleotides are made to mimic TDP-43's natural RNA binding partners. The idea is that the oligonucleotide is providing TDP-43 with a binding partner so it cannot interact with itself.

The investigators added the oligonucleotides to the dish at the same time they added the paired mustard seed and TDP-43 molecules. They turned the light onto the human neuronal cells, and nothing happened. The cells did not bind to themselves and there was no clumping. The cells did not die.

Dr. Donnelly said they refer to these TDP-43 targeting oligonucleotides as “bait-oligonucleotides” because they leave the bait there for the extra protein to keep it from clumping together.”

This novel technique allowed them to show that RNA is no longer present in TDP-43 and may offer clues to the molecular events that are important in solving the puzzle of these diseases, said Dr. Donnelly. Now, they are looking at other models to see if they can replicate their findings.


Dr. Christopher Donnelly (left to right), Elizabeth Daly, and Jacob Mann discuss the project at the 2018 Robert Packard Center for ALS Research Annual Symposium (2018).

They are starting to create animal models to use optogenetics to see how TDP-43 spreads. They have evidence that when TDP-43 starts to clump the normal TDP-43 sitting in the nucleus gets pulled in. (Many other labs have shown this phenomenon of ‘seeding’ in neurodegenerative conditions.) They are now doing studies to see whether the RNA-oligonucleotides would help after this toxic seeding has started.

Expert Commentary

“The authors noted a striking redistribution of TDP-43 from the nucleus to the cytosol upon light-induced aggregation of TDP-43, and the formation of cytosolic aggregates,” said Sami Barmada, MD, PhD, the Angela Dobson and Lyndon Welch professor of neurology at the University of Michigan Health System. “Additionally, the cells that form aggregates seemed to die at a faster rate.”

What was surprising, he added, was that cytosolic TDP-43 often did not form stress granules, and the larger TDP-43 aggregates were devoid of RNA. “These are important clues, indicating that stress granule assembly may occur independent of TDP-43 related toxicity,” he said.

Dr. Barmada said the finding raises the question of whether TDP-43 can be “deactivated” by adding extra RNA that is recognized by the protein. This would complement prior studies by Aaron Gitler and his colleagues at Stanford University, who first reported in Nature Genetics in 2012 that RNA lariats protect cells from overexpression of the potentially toxic TDP-43. “In essence,” he said, “the added RNA acts like a decoy, mopping up extra TDP-43 and preventing it from causing toxicity.”

There were limitations of the current study that will have to be addressed, he added. The TDP-43 hybrid that they made with the Cry2 protein from the mustard seed plant was three times larger than the TDP-43 protein.

“It is possible that some of the aggregation dynamics could be related to the fusion protein or Cry2, rather than TDP-43,” he said. “In addition, despite showing that aggregation is dose-dependent, the authors did not take into account the abundance of the fusion protein.”

The work by these investigators provides a “powerful model to induce TDP-43 protein aggregation,” said Clifford Brangwynne, PhD, a Howard Hughes Medical Institute Investigator and associate professor in the chemical and biological engineering department at Princeton University.

The Pittsburgh scientists used the optoDroplet technique—applying light to control protein-protein interactions—that was developed by Dr. Brangwynne's group. “The research team has reproduced features of TDP-43 pathology and survival of cells. This has phenotypic consequences.”

Dr. Brangwynne studies protein aggregation and was also intrigued that TDP-43 aggregation was happening outside of the presence of stress granules. “Another key message is that the presence of RNA is a good thing for TDP-43,” he said. “RNA keeps TDP-43 from forming irreversible assemblies. It is less prone to aggregation.”


Dr. Donnelly and Mann are inventors of the intellectual property described in the study and submitted it as provisional patents by the University of Pittsburgh.

Link Up for More Information

• Mann JR, Gleixner AM, Mauna JC, et al. RNA binding antagonizes neurotoxic phase transitions of TDP-43 Neuron 2019; Epub Feb 27.
    • Armakola M, Higgins MJ, Figley MD, et al. Inhibition of RNA lariat debranching enzyme suppresses TDP-43 toxicity in ALS disease models Nat Genet 2012; 44(12):1302–1309.