A study that peered into live mouse brains suggests for nearly 70 years we've been targeting the wrong neurons in our design of antipsychotic drugs.

Untangling the vast web of brain cells and determining how drugs work upon them is a tough task. Using a miniature microscope and fluorescent tags, a team of researchers led by Northwestern University neuroscientist Seongsik Yun discovered that effective antipsychotic drugs cling to a different type of brain cell than scientists originally thought.

Just like research suggesting depression might not be a chemical imbalance in serotonin levels, our understanding of schizophrenia treatments may need a rethink if widely-used antipsychotics are targeting different neurons than expected.

"There is an urgent need to understand the neural circuits that drive psychosis and how they are affected by antipsychotic drugs," Yun and colleagues write in their published paper.

As neuroscientist and study senior author Jones Parker told Wired's Max G. Levy, most antipsychotic medications – including the first one approved in 1954, chlorpromazine – have been discovered serendipitously. "So we don't know what they actually do to the brain."

Upon discovering them, scientists noticed drugs that suppressed common symptoms of schizophrenia such as mania, hallucinations, and delusions, seemed to act on the brain's dopamine system.

These antipsychotic drugs, they learned, stifled dopamine transmission between brain cells, and the most potent ones were compatible with a particular type of dopamine receptor labelled D2.

Brain cells called spiny projection neurons, which pack into and extend out of the striatum of the brain, express either D1 or D2 receptors. Unlike D1 receptors, which excite the brain's dopamine system, D2 receptors quieten it.

Linking the potency of antipsychotic drugs to D2 receptors gave rise to the idea that in schizophrenia, the striatum was awash with dopamine, a chemical imbalance that antipsychotics help correct.

But new drugs specifically designed to target D2 receptors did little to alleviate psychosis. And no one had actually ever tested in animal models of psychosis whether decades-old antipsychotic drugs preferentially bound to D2 receptors, so their exact mechanism of action remained unclear.

To investigate, Yun, Parker and their team injected mice with one of four drugs used to treat psychotic illness, and watched how the animals behaved and how their brain cells responded.

They found that haloperidol and olanzapine, two older but efficacious antipsychotics, had some effect on D2 spiny neurons, but their interactions were mostly happening at D1 neurons.

Clozapine, a newer, powerful antipsychotic, steered clear of D2 neurons, and overwhelmingly suppressed D1 cells, which could somehow "explain its clinical superiority, particularly for treatment-resistant schizophrenia," the researchers say.

Meanwhile, MP-10, a drug candidate that failed in clinical trials for schizophrenia, stayed glued to D2 neurons. In fact, MP-10 even made abnormal D1 activity worse.

In other words, a drug's clinical efficacy was closely related to its interaction with D1 neurons; those which normalized overactive D1 neurons relieved psychosis best – a finding which flips our understanding of these drugs on its head.

"These findings provide a novel explanation for antipsychotic drug efficacy," the researchers write. They suggest D1 spiny neurons, not D2-expressing cells, "may be a key driver of psychosis" and that normalizing their activity may be "a key indicator of antipsychotic effectiveness."

While the findings come as a blow to decades of research, it helps explain why some antipsychotic drugs like clozapine work when others don't. Though we should bear in mind dopamine isn't the only neurotransmitter linked to psychosis.

The findings also offer a glimmer of hope that researchers can correct course and use these new insights to design much-improved treatments for schizophrenia. Treatments that can't come soon enough.

The study has been published in Nature Neuroscience.