By Alla Katsnelson

Gene therapy delivered to a specific part of the brain reverses symptoms of depression in a mouse model of the disease--potentially laying the groundwork for a new approach to treating severe cases of human depression in which drugs are ineffective. But the invasive nature of the treatment, and the notorious difficulty in translating neuropsychiatric research from animal models to humans, could complicate its path to the clinic.

Many researchers believe that poor signaling of the neurotransmitter serotonin is responsible for causing depression, and common antidepressants act by increasing serotonin's concentration. Research published in the October 21 issue of Science Translational Medicine uses a virus to deliver an extra dose of the gene p11 to the adult mouse brain. The protein expressed by the gene is thought to bind to serotonin receptor molecules and ferry them to the cell surface, positioning them to receive serotonin's signals from neighboring cells.

"I think it awakens the possibility of gene therapy for neuropsychiatric diseases," says Husseini Manji, a senior investigator at Johnson & Johnson Pharmaceutical Research & Development in Titusville, N.J., who was not involved in the study. But, he adds, "thinking about delivering a gene to the brain poses all sorts of challenges."

The p11 gene was only recently linked to depression. In 2006, Paul Greengard, a neuroscientist at the Rockefeller University in New York, and his colleagues created mutant mice lacking the gene, and found that the animals developed depression-like behaviours.

Reversing depression

Michael Kaplitt, a neurosurgeon at Weill Cornell Medical College in New York, whose lab develops gene therapies for brain disorders, teamed up with Greengard and other colleagues in the new study.

The idea was "to identify the areas of the brain in which p11 is particularly important, in order to find targets" for new therapies for depression, says Kaplitt.

The researchers first used a technique called RNA interference, which turns genes off, to block the expression of p11 in two mouse brain areas linked to depression. When they delivered the blocker to one of these areas, known as the nucleus accumbens, the treated animals struggled less than controls when held by the tail, and also swam less actively than controls when released into a container of water--two tests routinely used to determine whether antidepressants are working in animal models.

Next, they injected a viral vector carrying the p11 gene directly into the nucleus accumbens of the mutant mice lacking the gene. The p11 boost in this brain area was enough to undo the mutants' usual depression-like symptoms.

Finally, the researchers turned to humans, and compared postmortem brains of 17 individuals who had depression during their lives, with those of individuals who had not. The nucleus accumbens of people with depression had much lower levels of p11 than that of their non-depressed counterparts.

The findings suggest, says Kaplitt, "that if we can reverse that low level of p11 in this area of the brain we can reduce depressive symptoms."

Treatment hurdles

The findings are interesting, says René Hen, a neuroscientist who studies depression at Columbia University in New York, but "animal models of depression are very imperfect." The behavioral tests used here modeled one dimension of the disease--an inability to experience pleasure from normal activities--but not others, such as stress and anxiety, and probably tap into different brain mechanisms in mice than in humans, he says. "To be convinced that we have something here that will be useful for depression will require a wider panel of tests" in mice.

Also, because the treatment is invasive--requiring brain surgeons to drill into the skull and deliver the therapy to the right spot in the brain--it should only be used by severely afflicted patients that don't respond to other drugs. "But that's really not what the study tested," says Manji, whose commentary on the study was published in the same issue of Science Translational Medicine. The mice in the study might well have responded to Prozac, and it is not clear whether the treatment does treat the severest forms of the disorder.

The long-term effects of the treatment are also unknown, says Manji. For example, would the p11-boosting therapy--which incorporates into recipients' genetic code--continue to function even if a patient's depression has subsided after treatment? That is a big wild card, says Manji, because no one knows the effect of putting the gene into permanent overdrive.

"Obviously, we have to be careful and do this right, to make sure we don't set the field back dramatically by moving too quickly," says Kaplitt. But he contends that other gene therapies already in clinical trials suggest that the treatment could work.

He and his lab, for example, along with the company Neurologix based in Fort Lee, N.J., are conducting a clinical trial to test a similar gene-therapy treatment for Parkinson's disease. (The company, which Kaplitt founded in 1999, also supported the current study.) Another clinical trial, which Kaplitt is not involved in, is testing the effect on depression of a deep brain stimulator implanted into the same brain area.

"We believe the template for this kind of treatment is already there," he says.

Kaplitt and his group are working with researchers at the National Institute of Mental Health in Bethesda, Md., to test the p11 gene therapy in non-human primates. Showing that the therapy is safe and identifying a marker that can track its effects in a living brain, he says, "is the key final piece that would be necessary to go forward in a human trial."