Angel Dust Paves a New Path to Schizophrenia Treatment

SAN DIEGO —Drugs of abuse have a bad reputation, but many of them can be used to shed light on the workings of the brain in both healthy persons and in those with a neuropsychiatric disorder. Angel dust, the street name for phencyclidine (PCP), is one such drug. Through studying the neurobiology of PCP and its derivative, ketamine, John Krystal, MD, has developed a model to better understand schizophrenia, capturing, as he says, “a bit of the angel in angel dust.” Dr. Krystal, Professor of Clinical Pharmacology at the Yale University School of Medicine in New Haven, Connecticut, gave his presentation at the 160th Annual Meeting of the American Psychiatric Association.

“We have an urgent need to develop new treatments for schizophrenia,” Dr. Krystal said. “It is perhaps the most disabling disorder in psychiatry. Unfortunately, recent large treatment trials with the new generation of medications, that we’ve all had such high hopes for, have produced rather limited benefits compared to the typical antipsychotics.” Current medications are also least effective for cognitive symptoms, which for many patients are the most functionally disabling aspects of the disease.

New treatments must come from a deeper understanding of the mechanisms involved in schizophrenia, Dr. Krystal emphasized. His work has begun to reveal the effects of deficits in N-methyl-d-aspartate (NMDA) glutamate receptors, leading to a model in which those deficits prevent the brain from fully optimizing its ability to function in multiple realms, especially cognitive tasks.

Schizophrenia research initially focused on dopamine, and dopamine D2 blockade remains the only mechanism currently available for treating patients with psychosis. However, hyperactivity of limbic dopamine systems is only present in some patients, some of the time, primarily those patients with recent exacerbations of their illness, said Dr. Krystal. “Other patients have normal or even reduced activation of limbic systems compared to healthy people,” he said. “So it’s not surprising that drugs that block dopamine D2 receptors aren’t that great,” and this intermittent dopamine hyperactivity cannot account for the sustained nature of the symptoms of schizophrenia.

Glutamate is the dominant excitatory neurotransmitter in the brain, acting in about half of all excitatory synapses in the cortex, versus only a few percent for dopamine. Glutamate facilitates communication from the cortex to all other brain regions. Among the various glutamate receptor subtypes, the NMDA receptor is involved in adapting and coordinating glutamate responses throughout the brain. Although early schizophrenia research focused on subcortical dopamine influences, as the field progressed, the spotlight shifted to cortical processes, in which glutamate receptors are most important.

Enter phencyclidine. As early as the 1950s, researchers noted that PCP given to healthy persons produced something that resembled schizophrenia, but they had no idea what the underlying mechanism might be. Then, in the 1980s, researchers determined that PCP and ketamine block NMDA receptors. At the same time, the NMDA receptor was shown to be involved in neural adaptation and network function. Based on this, Dr. Krystal’s group began to study ketamine for insights into schizophrenia.

The investigators showed that in healthy subjects, ketamine could reproduce the positive symptoms of schizophrenia, including delusions, hallucinations, and thought disorder, as well as the negative symptoms. “That was very important, because the other psychopharmacologic models of psychosis did not produce the negative symptoms,” he said. Subjects on ketamine became more distractible, disorganized, and had memory deficits. “But, and extremely importantly, they were still able to do a variety of cognitive tasks they had learned already, but they just couldn’t learn them while they were on ketamine.

“Ketamine was the first agent where one could rigorously show that it produced the full profile of symptoms that resembled schizophrenia,” Dr. Krystal said. Furthermore, it could not be blocked by the standard antipsychotic medications, indicating that it was acting through nondopamine pathways, strongly implicating NMDA receptor dysfunction in schizophrenia symptomatology.

Dr. Krystal believes that there is solid evidence for involvement of the NMDA system. For example, animal studies have shown that gene knockdown can produce some of the symptoms of schizophrenia, including cognitive symptoms. Genes that influence development of glutamate neurotransmission have been identified in human studies of schizophrenia risk factors, and postmortem studies, showing reduced dendritic spines in glutamatergic neurons, also implicate this system. “NMDA receptors do their job of coordinating network function on these dendritic spines,” Dr. Krystal said. “Fewer dendritic spines means fewer places for the NMDA receptors to help coordinate and allow experience to remodel the architecture.” PET studies in humans with schizophrenia show reduction in NMDA receptor density, particularly in the hippocampus.

How might decreased NMDA activity cause impairment in prefrontal working memory? In monkeys, neural recordings in the prefrontal cortex have shown that persistent neuronal activity is important for maintaining information in working memory. NMDA glutamate receptors and dopamine D1 receptors are involved in this process.

“The natural question is where in this process do patients with schizophrenia diverge from healthy subjects? Our data suggest it is not in the encoding of information but during the maintenance of prefrontal cortical activity,” Dr. Krystal said.

“If deficits in NMDA and dopamine D1 receptor activity contribute to cognitive dysfunction in schizophrenia, we ought to be able to enhance these functions by facilitating these neural mechanisms,” he noted. A variety of drugs, including the amino acid lysine, can raise the level of glutamate in the synapse and partially overcome ketamine’s deleterious effect on working memory. A selective dopamine D1 receptor agonist does not yet exist, but amphetamines can improve ketamine-induced impairment.

Dr. Krystal’s most recent work implicates yet another neurotransmitter system, GABA, in the cognitive dysfunction in schizophrenia, through its interaction with the NMDA and dopamine D1 systems. The metaphor for brain network activity, Dr. Krystal suggested, is that of a well-tuned symphony. “Impaired prefrontal cortical function in schizophrenia may reflect failure to tune cortical memory activities as a consequence of the interplay of deficits in glutamate, the main excitatory neurotransmitter in the cortex, and GABA, the main inhibitory neurotransmitter,” he said. “What we are trying to do is optimize the activity of these neural systems in the brain. Under normal conditions, as healthy people, we are pretty close to optimal.”

To keep a working memory pattern firmly in place, the brain must be able to narrow its focus and avoid distraction. When GABA is knocked out, this ability deteriorates because of elevated nontarget activity. “You lose the functional selectivity, the tuning, of the brain,” Dr. Krystal said. Cortical GABA abnormalities have been identified in schizophrenic brains, particularly in “chandelier cells.” These synapse onto the initial output axons of glutamate neurons, providing inhibition.

Just as GABA cells impinge on NMDA cells, NMDA cells, specifically those bearing the NR2A subunit within the receptor, provide one of the main inputs to GABA cells. This suggests that an intrinsic deficit in NMDA glutamate receptor dysfunction would disable the tuning function of the GABA neuronal population. Glutamate neurons then could become hyperactive, through action of non-NMDA glutamate receptors, leading to “noise-amplifying abnormalities.” The result can be described as “chaotic prefrontal cortical activity,” he said.

Lamotrigine inhibits calcium influx and therefore reduces glutamate release. Initial work shows that pretreatment with lamotrigine reduces the disorganizing effects of ketamine. Lamotrigine has also been shown to augment clozapine’s ability to reduce schizophrenic symptoms.

“But even more exciting is a new treatment target in the brain, the metabotropic glutamate receptors type 2,” or mGluR2. These are one of the main feedback inhibition mechanisms for glutamate neurotransmission in the brain, according to Dr. Krystal. Stimulating these receptors turns off glutamate neurons in several brain circuits.

While PCP increases prefrontal glutamate release, pretreatment with an mGlu2R agonist blocks this release. This treatment reduces disruption of working memory caused by either PCP or ketamine in animals and in human subjects. New data also suggest that it may work in patients with schizophrenia. Glutamate reduction with an experimental drug is almost as effective on its own as olanzapine, as shown last year in a large clinical trial in Europe. “This is potentially the first nondopamine D2 receptor-blocking agent that might be effective for treating psychosis,” Dr. Krystal said.

“We’ve been able to go from basic neurophysiology, to human laboratory studies, to the clinic,” he said. “By tracking the mechanism across the different levels of studies, we are beginning to capture a bit of the angel in the angel dust.”         

—Richard Robinson

Return to table of contents