This paper is one of the first that I published during my independent research career. I became interested in the phencyclidine (PCP) model for schizophrenia during residency and published a paper on this topic in the Hillside Journal of Clinical Psychiatry in 1987 ("Negative schizophrenic symptomatology and the PCP [phencyclidine] model of schizophrenia," Hillside Journal of Clinical Psychiatry 9[1]: 12-35, 1987). Based upon my studies during residency, I was able to obtain an NIMH Physician Scientist Award while I was still a 4^th-year resident. That award permitted me to devote essentially full time to laboratory research during my fellowship period and to obtain Ph.D.-level training in neuroscience. This paper represents the culmination of the studies I conducted during research training, along with a synthesis of PCP literature to that time.

PCP psychosis was first described in the late 1950s. By the mid-1960s, it was well established that PCP produced symptoms and cognitive deficits in normal volunteers that closely resembled those of schizophrenia. In particular, PCP-treated subjects developed symptoms such as emotional withdrawal and motor retardation that resembled what came to be known as the "negative" symptoms of schizophrenia, and developed schizophrenia-like concreteness of thought. The close resemblance between PCP psychosis and schizophrenia was reinforced in the 1970s when PCP became a major drug of abuse and the cases of PCP psychosis were routinely misdiagnosed as acute schizophrenia. However, the number of studies investigating PCP effects remained relatively small throughout the 1960s and 1970s, and PCP’s mechanism of action remained largely unknown. In the early 1980s, there was a substantial upturn in interest driven by two major developments. First, in 1979, several groups were able to demonstrate the existence of specific, high-affinity PCP binding sites in mammalian brain. Second, in the early 1980s it was demonstrated that the PCP "receptor" actually represented a binding site located within the ion channel formed by a larger receptor complex termed the N-methyl-d-aspartate (NMDA) receptor.

NMDA receptors are a type of receptor for the excitatory neurotransmitter glutamate and were known to play a prominent role in cognitive processing. It therefore made sense that the behavioral disturbances induced by PCP might be due to its interaction with NMDA-associated PCP receptors. However, there were several competing theories regarding the mechanism of action that had accumulated over the years. At the time the paper was written, it was difficult for researchers not intimately involved in PCP research to sort out the various claims that were being made. There were two theories that were particularly popular. The first claimed that PCP induced psychosis primarily by inhibiting dopamine transport. Since dopamine is the neurotransmitter most associated with schizophrenia, it was easy for individuals to assume the effects of PCP were mediated by its effects at dopamine transporters. The second theory proposed that PCP induced psychosis by binding to the sigma "opiate" receptor. That theory was particularly popular with drug companies that wanted to develop PCP-like agents as neuroprotectants and were hoping that the NMDA-blocking properties of such agents could be dissociated from their psychotomimetic effects.

The paper cited here is primarily a review paper. Its main contribution is that it pulled together clinical literature that had accumulated over several decades and combined it with newer information concerning the molecular targets of PCP. The main "result" section is a graph showing the serum concentrations associated with specific behavioral effects of PCP relative to the concentrations at which PCP was known to interact with specific molecular targets. This graph makes clear that many of the interactions that had been reported for PCP occurred at concentrations well above those that would occur during clinical intoxication. A related graph illustrates the fact that several PCP-like agents, such as ketamine or dizocilpine, induce PCP-like psychotic effects at doses commensurate with their binding affinity at the NMDA-associated PCP receptor. These agents do not interact with either dopamine transporters or sigma sites, so that blockade of those sites by PCP could not account for the behavioral effects of PCP.

I think there are several reasons why the paper has been widely cited. First, it pulls together a complex literature. While the information in the paper had all been published previously, this was the first time it had been reviewed from a clinical perspective in a neuropsychiatric journal. Second, the paper put forth a series of clear and testable hypotheses concerning both the etiology and treatment of schizophrenia. Regarding etiology, the paper predicted that dopaminergic dysregulation, negative symptoms, and cognitive dysfunction might all be attributable to a single underlying deficit in NMDA receptor-mediated neurotransmission. The most salient prediction was that NMDA agonists, such as glycine, d-cycloserine, or d-serine, would significantly ameliorate persistent negative symptoms of schizophrenia. Several small-scale clinical trials have supported this hypothesis, and a large-scale multicenter trial of glycine vs. d-cycloserine vs. placebo is currently underway. This trial may be seen as the ultimate test of the PCP/NMDA model. Finally, it is clear to almost everyone working in schizophrenia research that dopaminergic dysfunction, of itself, cannot account for all the signs and symptoms of schizophrenia. However, there are few papers that say so as explicitly as this one.

Daniel C. Javitt, M.D., Ph.D.
New York University School of Medicine
Program in Cognitive Neuroscience and Schizophrenia
Orangeburg, NY, USA