To the Editor: Hisham M. Ibrahim, M.D., et al. (1) carried out a postmortem study of schizophrenia brains in which they found low subregional N-methyl-d-aspartic acid (NMDA) receptor (NR) NR₁ and NR₂ gene expression in the thalami of their patients. Dr. Ibrahim et al. discounted the possibility that long-term antipsychotic treatment before death could have caused the observed lowering of NMDA, AMPA, and kainate mRNA and receptor expression by citing a single study that indicated that haloperidol did not cause decreased NR₁ binding (2). We note that Dr. Ibrahim et al. did not refer to our own two articles (3, 4), which examined the effects of the antipsychotic flupenthixol on NMDA, AMPA, and kainate receptors in the rat brain.

We used a multiprobe oligonucleotide solution hybridization technique to examine the regulation of gene expression of NR₁, NR_2A, NR_2B, NR_2C, and NR_2D in the left rat brain after treatment with the optical isomers of flupenthixol. At a dose of 0.2 mg/kg per day over 1, 2, 4, 8, 12, and 24 weeks, we found that both isomers downregulated the expression of NR₁ mRNA in most regions of the brain. NR_2A, NR_2B, and NR_2C showed significantly decreased expression from 12 to 24 weeks of treatment, but after 2 weeks NR_2B, NR_2C, and NR_2D expression had increased in several brain regions. NR₁ immunoreactivity in the right brain after 4 and 24 weeks of drug treatment was also examined with Western blotting. Both cis- and transflupenthixol significantly decreased NR₁ immunoreactivity in the right cerebellum after 24 weeks of treatment. Expression of the GluR1–7, KA1, and KA2 glutamate receptor mRNAs in the left rat brain were also studied. Neither cis- nor transflupenthixol was found to alter the gene expression of any of the nine non–NMDA glutamate receptor subunits. On the other hand, we found a nearly twofold increase in gene expression of the dopamine D₂ receptor in specific brain regions.

Our results suggest that the observed mRNA changes in NR₁ and NR₂ found by Dr. Ibrahim et al. in the postmortem brain are likely to have been caused by antipsychotic treatment and are not related to the disease process. However, the lowered expression of AMPA and kainate receptor RNAs found by Dr. Ibrahim et al. could indeed be related to schizophrenia itself rather than be an effect of treatment. Findings of previous postmortem studies of glutamate receptors by ligand binding in the schizophrenic brain have been contradictory. Now that all the human genes expressed in the brain have been identified and because genetic linkages on specific chromosomes in schizophrenia are beginning to be well replicated, it seems most likely that progress in this field will primarily result from the identification of genes with mutations that increase susceptibility to schizophrenia. Following this, the various secondary and/or treatment pathways that interact with these primary genetic causes will then become identifiable.