We all lose brain cells as we get older. In people with neurodegenerative diseases such as Alzheimer's, Parkinson's and Huntington's, neurons shrivel and die at alarming rates—perhaps three to four times faster than usual in Alzheimer's, for example. Currently, no known drugs reliably halt or reverse such staggering cell death in people, although some drugs are thought to protect neurons from degradation.

An alternative to saving dying neurons—or perhaps a future supplemental therapy—is creating brand new neurons. One way to accomplish this is transforming non-neuronal brain cells into functional neurons. On a cellular level, the brain is as diverse as a rainforest populated by many different species of trees. The human brain contains approximately 170 billion cells, 86 billion of which are neurons and 84 billion of which are glial cells—non-firing cells that assist neurons in various ways. Star-shaped cells known as astrocytes are perhaps the best-studied of the many various glial cells and researchers have had some success converting astroyctes into neurons. Many of these studies, however, have used cells from very young rodent brains.

A study published this week suggests that it's possible to turn at least one class of adult human brain cells known as pericytes into functional neurons. The fact that pericytes help defend and heal the brain—and may retain some of the plasticity of stem cells—makes them all the more appealing as candidate replacements for damaged and dying neurons.

Benedikt Berninger of Ludwig-Maximilians University Munich and his colleagues began their research project with the intent to study astrocytes, just as they have done many times before. They acquired 30 samples of brain tissue from people who were undergoing surgery for disorders such as epilepsy. Sometimes, in order to remove or treat a damaged or malfunctioning brain region, neurosurgeons cannot avoid slicing through healthy brain tissue. Surgeons routinely provide sections of such healthy tissue to researchers studying the brain.

In the lab, Berninger and his teammates grew cultures of brain cells from the tissue samples and searched for astrocytes nestled among the tiny neural gardens. As it turned out, the cultures Berninger and his colleagues grew were mostly devoid of astrocytes. Instead, their Petri dish gardens were rife with pericytes—non-neuronal brain cells that wrap themselves around the brain's delicate blood vessels, regulate blood flow to neurons and help maintain the blood-brain barrier, which protects neurons from bacteria and other pathogens. Pericytes are also known to proliferate in response to injury. Researchers recently showed, for example, that pericytes are essential for the formation of scar tissue in an injured spinal cord. Some evidence even suggests that certain kinds of pericytes boast the same flexibility as mesenchymal stem cells—they can turn into bone cells, fat cells or cartilage cells. Perhaps, Berninger and his colleagues reasoned, the plasticity of pericytes—coupled with their role in healing—might make them especially useful in future treatments for neurodegenerative diseases. So they decided to try changing pericytes into neurons by reprogramming their genomes.

Using viruses, Berninger and his team infected the pericytes in their cultures with two transcription factors—proteins that alter gene expression by binding to segments of DNA and making certain genes more or less accessible to other cellular machinery. One of the transcription factors, Mash1, is known to guide the development of the nervous system. We all begin life as a hollow ball of embryonic stem cells that eventually become the many different kinds of cells in the human body. All somatic cells in your body have the same DNA, but distinct types of cells express very different sets of genes—just as different piano songs are unique combinations of notes played on the exact same set of keys. MASH 1 is like a tiny composer inside embryonic stem cells, making sure they turn on the right combination of genes to become neurons. The second transcription factor Berninger and his colleagues introduced into pericytes was Sox2, which is highly active in stem cells and thought to make DNA more amenable to manipulation by loosening the chemical bonds between DNA and the protein scaffolding that keeps it tightly wound in a bundle called chromatin.

The scientists successfully converted between 10 and 30 percent of the pericytes in various cultures into neurons; the overall success rate was 19 percent. Out of 17 successfully converted neurons selected for further testing, 12 generated electrical impulses. Berninger and his colleagues replicated these results with brain cells from adult mice. The results appear in Cell Stem Cell.

Treating neurodegenerative diseases by genetically reprogramming brain cells is a potential avenue for therapy that researchers have just started to navigate—and they will have to scale plenty of hurdles along the way. Scientists must ensure that the viruses they use to ferry genes or transcription factors into brain cells are harmless.* And they would likely have to perform risky invasive surgery to get the viruses into exactly the right region of the brain. In recent years, however, gene therapy has safely restored vision to the blind. Not only do studies like Berninger's suggest that gene therapy for the brain has similar potential, they also confirm that the fates of some adult cells are not written in stone—rather, they are written in highly editable DNA.

*Editor's note: this sentence was edited for accuracy and clarity