Researchers have mapped thousands of tiny, previously uncataloged proteins in the human brain, revealing that many of these molecules are altered in Alzheimer’s disease. The study, published in Nature Aging, provides a detailed catalog that links specific short proteins to brain cell energy production and genetic regulation. These discoveries suggest that traditional gene databases capture only a fraction of the brain’s functional molecules.
Genetic databases act as reference dictionaries for biologists, categorizing the DNA sequences that cells translate into functional proteins. For decades, standard scientific practice excluded sequences that code for very short proteins, often called microproteins. These molecules typically contain fewer than 150 amino acids, the basic building blocks of proteins. Because of their short length, they are difficult to detect using standard laboratory equipment.
Microproteins originate from small open reading frames, which are continuous stretches of genetic code capable of being translated into proteins. Many of these regions are located within or near larger, well-known genes. As a result, standard genetic sequencing often attributes any activity in these areas to the primary gene, missing the activity of the smaller microprotein.
To understand the role of these hidden molecules in brain aging and neurodegeneration, Salk Institute for Biological Studies researchers Brendan Miller and Alan Saghatelian led a team to search for them in human brain tissue. They aimed to create a comprehensive atlas of microproteins in the frontal cortex, a brain region heavily impacted by Alzheimer’s disease.
The researchers first constructed a custom genetic database. They used advanced RNA sequencing on tissue from 12 postmortem human brains to document the specific genetic instructions present in the frontal cortex. They appended these instructions to the standard reference database of known human proteins.
Using this updated reference, the team analyzed mass spectrometry data from 610 brain samples. Mass spectrometry is a technique that identifies proteins by breaking them apart and measuring the mass of the fragments. This initial search yielded 4,321 microproteins. More than 3,000 of these were considered unreviewed, meaning they had not been cataloged in standard, manually curated protein databases.
To verify that these unreviewed microproteins were genuine, the researchers cross-referenced their data with ribosome profiling. This technique maps where ribosomes, the cellular machines that build proteins, actively attach to genetic material. They also applied a deep learning tool called PROSIT to evaluate the quality of the mass spectrometry fragments.
This deep learning program predicts what a true protein fragment should look like, allowing the researchers to filter out false positives. This filtering process identified 1,067 unreviewed microproteins with strong, highly confident evidence of existing in the brain.
Next, the team investigated whether these microproteins change in relation to Alzheimer’s disease. They compared mass spectrometry data from individuals without the disease to data from individuals with symptomatic Alzheimer’s disease.
The researchers found dozens of microproteins with altered levels in the brains of individuals with Alzheimer’s disease. For example, a microprotein derived from the BCL3 gene was less abundant in symptomatic individuals. Conversely, a microprotein from the RAB3C gene was more abundant. When the team introduced these specific microproteins into laboratory-grown human microglia, the brain’s primary immune cells, they found that the proteins remained stable and occupied specific areas inside the cells.
The researchers then explored how these microproteins are regulated. Because many microproteins are encoded by DNA located near larger, well-documented genes, the team examined whether the two are produced together. They compared the RNA levels of the microprotein sections with the RNA levels of the primary genes.
They discovered that about half of the microproteins were regulated independently of their larger parent genes. In many cases, the amount of microprotein RNA changed in Alzheimer’s disease while the RNA for the main gene did not. This independent regulation was confirmed in a separate dataset of brain tissue from another cohort of patients.
The team determined that this separation likely results from alternative splicing, a process where cells cut and paste RNA segments differently to create distinct protein products. They identified specific patterns near these cut sites that attract RNA-binding proteins, suggesting a regulatory mechanism that goes awry during neurodegeneration.
The researchers focused deeply on a microprotein encoded by the MKKS gene. Standard databases list the primary product of MKKS as a large protein linked to a rare developmental disorder. The team wanted to verify if this held true in the adult brain.
Through their analysis, they found that the small microprotein, rather than the large primary protein, was the dominant product of the MKKS gene in the frontal cortex. The mass spectrometry data showed thousands of fragments matching the microprotein but almost none matching the large protein. The levels of this microprotein were also lower in the brains of people with symptomatic Alzheimer’s disease.
To understand its function, the researchers mapped out which other genes were active at the same time as the MKKS microprotein. They found a strong overlap with genes involved in the function of mitochondria, the energy-producing structures within cells.
The team then conducted experiments in laboratory microglia to test this mitochondrial connection. Using a gene-editing tool called CRISPR, they deleted the specific DNA sequence that produces the MKKS microprotein without altering the main gene. They then ran a stress test on these modified cells to measure their oxygen consumption rate. Microglia lacking this microprotein showed a reduced ability to consume oxygen and produce energy, confirming its role in cellular respiration.
Finally, the researchers investigated pseudogenes, which are mutated genetic sequences generally considered to be inactive or broken. They reviewed the RNA and ribosome data to see if these regions might actually produce microproteins.
They found evidence that multiple pseudogenes are actively translated into microproteins. One such microprotein, derived from a pseudogene called ACTBP7, is related to the proteins that form the structural framework of a cell. When the researchers added the ACTBP7 microprotein to laboratory microglia, it altered the distribution of the cells’ structural fibers.
While this atlas identifies many previously unknown brain molecules, the lack of mass spectrometry detection for a specific sequence does not mean the protein is absent from the brain. Some microproteins may simply fall below the detection limits of current laboratory equipment. In addition, the exact genetic pathways that produce many of these short proteins still need to be fully mapped.
Finding a microprotein with altered levels in an Alzheimer’s disease brain does not mean the molecule causes the disease. These changes might merely reflect a cell’s response to damage or altered regulatory processes. Future research will need to manipulate these individual microproteins in living systems to distinguish between those that drive neurodegeneration and those that are simply markers of cellular stress.
The study, “A microprotein atlas of the human frontal cortex in Alzheimer’s disease,” was authored by Brendan Miller, Eduardo Vieira de Souza, Calvin Lau, Joan M. Vaughan, Victor J. Pai, Servando Giraldez, Andréa Rocha, Jolene K. Diedrich, Clodagh C. O’Shea, David A. Bennett, and Alan Saghatelian.