By Sharon Begley
January 27, 2016
Schizophrenia may be caused by a gene that sends the immune system into overdrive, according to a surprising finding reported Wednesday that outside scientists hailed as “remarkable,” “transformational,” and a “tour de force.”
The study, in the journal Nature, suggests a way to treat the devastating mental illness, which for decades has flummoxed scientists seeking its genetic underpinnings. The field of psychiatric genetics — the needle-in-a-haystack search for DNA variants that cause mental illness — is littered with red herrings and refuted claims. But researchers said this discovery might be different because it may explain one of the only solid observations about the brain pathology of schizophrenia: deficient numbers of connections, or synapses, between neurons in the brain, perhaps due to excessive “pruning” of synapses.
“If they’ve got it right — and I think they have — the reason this is different from other genes that have been associated with psychiatric conditions is that they connected [a gene] to a very specific biological process,” said David Goldstein, director of the Institute for Genomic Medicine at Columbia University Medical Center, who was not involved in the study.
Anytime a gene is biologically plausible — that is, when its function jibes with something biologists know about a disease — scientists have more confidence they have found something real and important, not a spurious association. That’s the good news.
The bad news is that if schizophrenia is indeed caused by too much pruning of synapses, finding a therapy will be “challenging,” said Goldstein. That’s because the brain likely has a small tolerance for the wrong amount of pruning: too much may lead to schizophrenia, but too little could be crippling.
A drug that reduces pruning would therefore have to be exquisitely precise, Goldstein said: “Pruning is at the opposite end of the spectrum of what you’d consider an easy target” for a drug.
Existing drugs for schizophrenia target the hallucinations and delusions of the disease, which strikes some 1 percent of the US population and typically begins in the late teens or early 20s. But nothing helps other symptoms, including trouble with attention, thinking, and memory, let alone underlying causes.
For the study, scientists led by molecular biologist Steven McCarroll of the Stanley Center for Psychiatric Research at the Broad Institute in Cambridge, Mass., analyzed the genomes of 64,785 people from 22 countries, with or without schizophrenia.
They focused on a gene called C4, which — surprisingly for a disorder long viewed as a disease of the nervous system — is involved in the immune system. The gene makes a molecule that tags invading cells such as bacteria for elimination.
The C4 gene comes in long and short forms, and different people have different numbers of copies. After boatloads of genetic, molecular, and cellular tests, McCarroll and his colleagues found that people with the long form of C4, and with lots of copies of the gene, made more of the “destroy me” tag. They were also more likely to have schizophrenia. The more C4, the greater the likelihood.
In earlier research, neuroscientist Beth Stevens of Boston Children’s Hospital and her colleagues found that a relative of C4 plays a role in brain development, causing synapse pruning in mouse brains. In particular, it slaps a molecular tag onto synapses. The tag acts like a note to a molecular arborist: “prune me.”
In the new study, the scientists found that C4 slaps this tag onto synapses. The more C4 activity, the more mouse synapses are pruned. Overpruning may explain why the brains of people with schizophrenia tend to have an abnormally thin cortex and fewer synapses. There has been debate over whether that is the cause or consequence of schizophrenia; this study suggests the former.
That a synapse-pruning gene raises the risk of schizophrenia may explain why the disease shows up in late adolescence. That’s when the brain undergoes a wave of pruning, including in regions involved in higher-order thinking. With too much C4, “cortical circuitry might become too thinned out to perform basic cognitive functions,” said McCarroll.
Not everyone with the schizophrenia-linked form of C4 develops the disease: One copy of the gene raises the risk by 27 percent, while two copies (one from each parent) raises the risk 50 percent to 60 percent, said McCarroll. That makes the absolute risk of developing schizophrenia still less than 2 percent for people with two copies of the C4 variant. Environmental factors or other genes likely can override the schizophrenia-causing effects of C4, or perhaps people need more than one of the 108 genes linked to schizophrenia to develop the disease.
“Individuals with [this version of C4] might never develop schizophrenia,” said Goldstein, and the risk is so low it’s probably not worth testing children or prospective parents for it. It’s not clear if people with a family history of schizophrenia would benefit from genetic screening, but there is abundant evidence that genes are not omnipotent: if one identical twin develops schizophrenia, the other, despite having identical genes, does so in only about half the cases.
Nevertheless, if C4-driven overpruning is a significant cause of schizophrenia, it suggests one way to treat or possibly prevent the disease: find a molecule that hides some of the extra “prune me” tags. Because there is likely a fine line between too much pruning and too little, finding a Goldilocks drug that gets it just right “is going to be hard, no question about it,” said Goldstein. “But at least we have a process to focus on.”
If a drug gets synaptic pruning just right, it might one day avert schizophrenia in young adults who show early signs of the disease. “Many schizophrenia patients endure an agonizing cognitive decline for years after their initial diagnosis,” said McCarroll. That means the right drug would have “a very broad window in which patients might be able to benefit.”