We have a parts list for mitochondria, but we don't know what many of the parts do.
Within every human cell, mitochondria sustain life by converting energy — yet for roughly one in 4,300 people, something in that machinery breaks in ways medicine has long been unable to name. Researchers at Washington University and the University of Wisconsin–Madison have now used CRISPR technology to systematically probe 116 poorly understood mitochondrial proteins, building a map where there was once only darkness. Three rare diseases have been genetically solved through this work, and twenty more potential therapeutic targets have been identified and made publicly available. It is the kind of science that does not cure anyone today but makes the curing of someone tomorrow imaginable.
- Mitochondrial diseases can devastate every high-energy system in the body simultaneously — heart, brain, and muscle — yet most of the 1,300 proteins inside mitochondria remain functionally unknown to science.
- Researchers deleted genes one by one across 116 cell lines and measured the downstream collapse in proteins, lipids, and metabolites, generating a dataset vast enough to require its own purpose-built analysis tool.
- Three real patients — including a child with cerebrofaciothoracic dysplasia and a family carrying cleft lip mutations — had their genetic mysteries resolved directly through this mapping approach.
- Twenty additional proteins have been flagged as likely therapeutic targets, their hypothesized roles documented and released publicly so the broader scientific community can pursue them.
- The MITOMICS database now stands as an open resource, transforming what was a single research effort into an ongoing infrastructure for rare disease discovery.
Mitochondria power nearly every cell in the human body, and when they fail, the consequences are devastating — seizures, developmental delays, vision loss, and profound fatigue can all strike at once. The difficulty has always been that mitochondria contain roughly 1,300 proteins, and the function of most remains unknown. When a patient falls ill, doctors frequently cannot identify which protein is responsible, leaving treatment out of reach.
To change this, researchers at Washington University School of Medicine and the University of Wisconsin–Madison used CRISPR-Cas9 to delete genes one at a time across 116 cell lines — 50 proteins of unknown function and 66 already understood, as a control. For each modified line, they measured changes in cell growth and quantified thousands of proteins, lipids, and metabolites. The resulting data was enormous, so the team built MITOMICS, an application capable of identifying which biological processes collapsed when a specific protein disappeared.
Three genuine disease mysteries were solved in the process. A patient in the United Kingdom with clear signs of mitochondrial dysfunction carried no mutations in any of the usual suspect genes — until the researchers identified a new gene in the relevant pathway and found the patient's mutation within it. In a second case, deleting the gene RAB5IF caused the loss of a protein tied to cerebrofaciothoracic dysplasia, a condition involving severe intellectual disability and distinctive facial features. Collaborators in Turkey then traced a RAB5IF mutation to one case of the dysplasia and two cases of cleft lip within a single family. A third disrupted gene triggered abnormal sugar storage linked to a fatal autoinflammatory syndrome.
Beyond these solved cases, the team identified roughly 20 additional proteins connected to specific cellular pathways — leads they could not fully pursue in a single study but documented and made public. The MITOMICS app has been released openly so other researchers can mine the data for their own discoveries. For a class of diseases too rare to attract large research investments yet severe enough to reshape entire lives, a more complete map now exists — and with it, a clearer path toward intervention.
Mitochondria are the cell's power plants, and when they malfunction, the damage spreads everywhere at once. A person might experience stunted growth, crushing fatigue, seizures, cognitive delays, or progressive vision loss—sometimes all of these together. The problem is that mitochondria contain roughly 1,300 different proteins, and scientists understand the function of only a fraction of them. When something breaks, doctors often cannot identify which protein failed, making treatment nearly impossible.
Researchers at Washington University School of Medicine and the University of Wisconsin–Madison decided to map the unknown territory. They used CRISPR-Cas9 technology to systematically delete genes from human cells, creating a library of cell lines each missing a single gene. In total, they removed genes for 50 mitochondrial proteins of unknown function and 66 with known functions—a controlled experiment designed to reveal what each protein actually does.
For each modified cell line, the team measured what changed. They tracked cell growth rates and quantified the abundance of 8,433 proteins, 3,563 lipids, and 218 metabolites. The resulting dataset was enormous and unwieldy until they built MITOMICS, an app that could sift through the numbers and identify which biological processes collapsed when a specific protein went missing. After validating the method against proteins whose functions were already known, they turned it loose on the mysteries.
Three discoveries emerged with enough clarity to solve real cases of disease. In one instance, a patient in the United Kingdom showed all the signs of a mitochondrial disorder affecting the cell's main energy-producing pathway, yet carried no mutations in any of the genes doctors usually suspect. The researchers identified a new gene in that pathway and found the patient carried a mutation in it—the missing piece. In another case, researchers noticed that removing one gene, RAB5IF, caused the loss of a protein called TMCO1, which had been linked to cerebrofaciothoracic dysplasia, a condition marked by distinctive facial features and severe intellectual disability. Working with collaborators in Turkey, they traced a mutation in RAB5IF to one case of the dysplasia and two cases of cleft lip within a single family. A third gene, when disrupted, triggered problems with sugar storage that contributed to a fatal autoinflammatory syndrome.
Beyond these three solved cases, the researchers identified patterns connecting roughly 20 other proteins to specific biological pathways or cellular processes. They could not pursue every lead in a single paper, but they documented their hypotheses and made them public. The team also released the MITOMICS app itself, complete with user-friendly analysis tools, so that other scientists could explore the data and hunt for their own discoveries.
Mitochondrial diseases are rare—affecting roughly one in every 4,300 people—but their rarity does not diminish their severity. Because mitochondria power nearly every cell in the body, defects tend to strike hardest in tissues with the highest energy demands: the heart, brain, and muscles. The researchers see their work as the beginning of a longer project. Each newly understood protein function opens a potential therapeutic target. The ultimate goal is to understand mitochondria deeply enough to intervene when they fail, something medicine cannot yet do. For now, the map is more complete, and the path forward is clearer.
Citações Notáveis
Every time we discover a function of a new protein, it gives us a new opportunity to target a pathway therapeutically.— David J. Pagliarini, Washington University School of Medicine
It is very exciting to see how our mass spectrometry technology platform can generate data on this scale but more importantly, data that can directly help us to understand human disease.— Joshua J. Coon, University of Wisconsin–Madison
A Conversa do Hearth Outra perspectiva sobre a história
Why is it so hard to diagnose mitochondrial diseases in the first place?
Because the symptoms are scattered across the body—seizures, muscle weakness, vision problems—and they can mimic dozens of other conditions. But the real barrier is that we don't know what most mitochondrial proteins do. You can't fix something when you don't understand how it works.
So CRISPR let you delete genes one at a time and watch what broke?
Exactly. We created cell lines missing each gene and measured what changed—growth rates, protein levels, metabolites. It's like removing one instrument from an orchestra and listening to what sounds different.
And you found three actual patients whose diseases you could now explain?
Yes. One had a mutation in a gene no one had connected to mitochondrial disease before. Another family had mutations in a gene that affects facial development and intellectual ability. The third case involved a fatal inflammatory condition. These weren't theoretical—they were real people we could finally help understand what was happening to them.
What about the other 20 proteins you found patterns for but didn't fully solve?
Those are invitations for future work. We documented the hypotheses and released all the data publicly. Other labs might have patients with those mutations, or they might approach the problem from a different angle. Science moves faster when you share what you know.
Do you think this approach will work for other rare genetic diseases?
The method is general enough that it should. The real bottleneck has always been understanding what proteins do. Once you have that map, diagnosis and treatment become possible.