Scientists have known for 15 years that mutations in a single gene lead to Rett syndrome, a severe neurological disorder that affects girls around their first birthdays. In the years since the MECP2 gene was pinpointed, researchers have struggled to understand how it functions in the brain in Rett syndrome.
Now the enigma of Rett syndrome and perhaps other disorders on the autism spectrum could be one step closer to being solved.
A Harvard Medical School team has discovered that when MECP2 is mutated in Rett syndrome, the brain loses its ability to regulate genes that are unusually long. Their finding suggests new ways to consider reversing the intellectual and physical debilitation this disruption causes with a drug that could potentially target this error. The team, led by Michael Greenberg, reported its findings in Nature.
“The longer the gene, the more disrupted it becomes when you lose MECP2,” said Greenberg, the Nathan Marsh Pusey Professor of Neurobiology at HMS. “Rett syndrome may be a defect in this process of fine-tuning the expression of long genes.”
Scientists, including Greenberg, have figured out over the last 10 years that MECP2 plays a role in sculpting the connections between neurons in the developing brain. These synapses are refined by exposure to sensory experiences, just the sort of stimulation a one-year-old would encounter as she learns to walk and talk.
MECP2 is present in all cells in the body, but when the brain is forming and maturing its synapses in response to sensory input, MECP2 levels in the brain are almost 10 times as high as in other parts of the body. The new study connects MECP2 mutations to long genes, which may be more prone to errors simply because their length leaves more room for mistakes.
“Normally, MECP2 may act like a speed bump, fine-tuning long genes by slowing down the machinery that transcribes long genes,” said Harrison Gabel, a postdoctoral fellow in the Greenberg lab and co-first author of the Nature paper. In transcription, the information in a strand of DNA is copied onto a new molecule of messenger RNA, which is then turned into a protein. “Without MECP2, the machinery may be moving too fast, making too much mRNA from these genes, resulting in problems for the neurons.”
Finding this effect of MECP2 on long genes was no small feat. In a typical search for the mechanism behind a genetic mutation, mice are engineered to lack the normal gene so that its absence reveals how it functions. However, work in many different labs has shown that knocking out MECP2 had only subtle effects when analyzed across the genome. The changes in gene expression were inconsistent, small and, using Gabel’s word, “fuzzy.”
Gabel took another approach, querying massive genomic databases such as ENCODE to ask a simple question: What do genes that are affected by mutated MECP2 have in common?
Answer: They are long. Most of them are at least five times longer than the average gene, with many of them more than 50 times longer than the average. It is important to note that the genes identified across dozens of data sets were very long, giving the researchers a common finding where previous conclusions from these data sets had lacked a common theme.
Harrison and co-first author Benyam Kinde, an MD-PhD student in the Greenberg lab, found the long-gene misregulation in multiple mouse models of Rett syndrome and confirmed it in the brain tissue of deceased Rett patients.
For MECP2 to function normally as a speed bump, it binds to a form of methylated DNA found in long genes in the brain. Methyl groups are chemical modifiers of gene activity, and in other parts of the body MECP2 binds methylated CG sites on genes. The methylation pattern that appears to be important for MECP2 in regulating long genes is known as methylated CA, and there appears to be a special mechanism operating as synapses are forming.
“It seems that evolution has used MECP2 and methylated CA to put in place this speed bump so that the expression of long genes is restrained in the brain,” Greenberg said. “As far as Rett syndrome, the thought is now that this subtle but widespread overexpression of long genes might be contributing to the disorder.”
The scientists can’t be sure of what these overexpressed long genes do, but many of them appear to be very important to the function of the brain. This suggests that if they could correct the defect in long-gene expression, they might be able to reverse at least some of the symptoms of Rett syndrome. As a first attempt at a corrective strategy, the researchers selected a cancer drug called topotecan because it blocks an enzyme known to be important for long-gene transcription.
In a lab dish, they added topotecan to neurons lacking MECP2. The drug reversed the long-gene misregulation, suggesting that restoring normal long-gene expression might be a way to correct neurological dysfunction in Rett syndrome and in other autism spectrum disorders with long genes, such as fragile X syndrome. Topotecan, a chemotherapeutic agent, is too toxic, Greenberg said, but derivatives of topotecan might be a worthwhile avenue to pursue.
“We think this issue of long-gene misregulation may be more generally occurring in other disorders of human cognition,” Greenberg said. “The potential is pretty significant because one now has a common regulatory mechanism to target with drugs.”
This work was supported by grants from the Rett Syndrome Research Trust and the National Institutes of Health (1RO1NS048276 and T32GM007753), the Damon Runyon Cancer Research Foundation (DRG-2048-10), the William Randolf Hearst fund and the Howard Hughes Medical Institute.