Technical Comments

Response to Comment on "Diverse Psychotomimetics Act Through a Common Signaling Pathway"

See allHide authors and affiliations

Science  09 Jul 2004:
Vol. 305, Issue 5681, pp. 180
DOI: 10.1126/science.1097564

Seeman's remarks (1) supporting the dopamine hypothesis of schizophrenia are based primarily on in vitro receptor-binding studies that cannot be found in his cited references and that differ markedly from previously published data. Seeman's laboratory recently reported (2) that phencyclidine (PCP) binds to D2 receptors, as assessed by [3H]raclopride binding in striatal homogenates under hypotonic sodium concentrations, with a Ki (inhibition constant) value of 37 μM. That suggests a 10,000-times difference in affinity of PCP to D2 receptors between striatal homogenates (2) and cloned D2 receptors (1), which is indeed remarkable. The Ki value of d-lysergic acid diethylamide (LSD) at cloned D2 receptors (0.8 nM) reported in (1) is more than 200 times lower than that found in the NIMH-sponsored large-scale ligand screening program (180 nM) (3, 4). These studies differed in choice of radioligand and sodium concentration ([3H]raclopride versus [3H]spiperone; hypotonic versus isotonic). Still, Seeman's in vitro data raise an important conceptual question, especially since we came to a very different conclusion from in vivo studies designed to elucidate the mechanism of action of diverse pharmacological compounds. We recently studied (5) different classes of psychotomimetic substances (6)—including d-amphetamine, LSD, and PCP—which act primarily through perturbation of dopamine, serotonin, and glutamate signaling pathways, respectively. We provided evidence that these psychotomimetics cause a similar pattern of changes in DARPP-32 phosphorylation that includes increased levels of phosphorylated Thr34- and Ser130-DARPP-32 and decreased or unaltered levels of phosphorylated Thr75-DARPP-32 [DARPP-32 is an adenosine 3′,5′-monophosphate (cAMP)-regulated phosphoprotein of 32 kilodaltons]. Moreover, using DARPP-32 knockout mice and mice with point mutations in the phosphorylatable residues, we demonstrated the importance of phosphorylation of Thr34 and Ser130 for eliciting psychotomimetic behavioral responses to these drugs.

It is widely accepted that d-amphetamine, LSD, and PCP directly or indirectly affect multiple neurotransmitter receptors and transporters. Amphetamine, owing to its dopamine-releasing properties, indirectly activates postsynaptic D1 and D2 receptors. However, D1 and D2 receptor agonists have opposing actions on DARPP-32 phosphorylation: We reported that D2 receptor agonists decreased phospho-Thr34-DARPP-32 and increased phospho-Thr75-DARPP-32 (7-9). Seeman's proposition that our results can be explained solely by D2 receptor agonism is inconsistent with the current data and predicts results opposite to those found in (5).

LSD has 50- to 100-fold higher affinity for several 5-HT receptors as compared to D2 receptors (4). The D2 receptor antagonist haloperidol has no effect against LSD-induced disruption of prepulse inhibition (PPI) (10), but blocks the disruption of PPI induced by the D2 agonist apomorphine (11). There seems to be a consensus that the PPI deficits induced by LSD and its related compound DOI are not mediated by D2 receptor agonism (10, 11). Numerous drug-discrimination studies have provided indirect support for this notion, showing that the discriminative stimulus cue of LSD is dissimilar to that of apomorphine and cannot be blocked by haloperidol (12, 13). In fact, the psychotomimetic actions of LSD are predominantly mediated by its interaction with serotonergic receptors—primarily 5-HT2A (14)—although other serotonin receptors such as 5-HT5A, 5-HT6, and 5-HT7 also appear to play a role (4, 15, 16; Fig. 1).

Fig. 1.

Schematic illustration of possible dopamine receptor-independent routes to psychosis. MK-801 and PCP inhibit a signaling cascade by which NMDA receptors increase calcium influx, which stimulates calcineurin (PP-2B) activity and reduces phospho-Thr34-DARPP-32. LSD stimulates 5-HT2 receptor-mediated activation of phosholipase C (PLC), resulting in increased casein kinase 1 (CK1) activity and increased phospho-Ser130-DARPP-32. Phospho-Ser130-DARPP-32 reduces the ablility of PP-2B to dephosphorylate phospho-Thr34-DARPP-32. LSD also activates 5-HT6 and 5-HT7 receptors, which results in increased PKA activity and increased phosphorylation of Thr34-DARPP-32.

Emerging pharmacological and genetic evidence suggest that the etiology of schizophrenia in some individuals is associated with a hypofunction of N-methyl-d-aspartate (NMDA) receptor-mediated signaling (17). In general, the deficits in PPI produced by NMDA receptor antagonists, such as PCP and MK-801, appear to be more sensitive to clozapine-like atypical antipsychotics than to typical antipsychotics, which are potent D2 receptor antagonists (11); the ED50 (median effective dose) values of drugs necessary to reverse PCP-induced PPI disruption do not correlate with their affinity for D2 receptors (18). Likewise, MK-801 disrupts PPI in D2 receptor knockout mice, further suggesting that the schizophrenia-like effects of NMDA receptor antagonists are D2 receptor-independent (19). Moreover, in monoamine-depleted rodents, PCP and MK-801 cause stimulation of locomotion that escapes blockade of D2 receptors with haloperidol (20, 21). Similarly, mutant mice carrying targeted point mutations in the glycine binding site of the NMDA receptor exhibit a hyperactivity that is insensitive to antipsychotics (22). In the clinic, typical antipsychotic drugs, including haloperidol, either do not affect PCP-induced psychotic episodes (23) or prevent certain symptoms of PCP intoxication through pharmacologically unselective actions, similar to diazepam and physostigmine (24). In general, it is contended that PCP- and MK-801-induced psychotomimetic effects are associated selectively with noncompetitive inhibition of NMDA receptor-mediated neurotransmission (25). We showed that MK-801 has effects on DARPP-32 phosphorylation virtually identical to those of PCP [table S1 in (5)]. Since PCP and MK-801 differ by more than 1,000-fold in their affinity for D2 receptors (2, 26), but have fairly similar affinities for NMDA receptors (26, 27) and similar potencies in altering DARPP-32 phosphorylation (5), the rational interpretation of our data is that PCP and MK-801 act primarily on NMDA receptors to regulate DARPP-32 phosphorylation. The pattern of DARPP-32 phosphorylation, induced by PCP and MK-801, is consistent with a blockade of the NMDA/Ca2+/calcineurin (PP-2B) signaling pathway (8, 28; Fig. 1). In agreement with our data, recent work (29, 30) suggests that a diminished calcineurin function contributes to schizophrenia. Studies of a genetic mouse model that exhibits schizophrenia-like behaviors (31) provide even stronger evidence that neurotransmitters other than dopamine can, through a DARPP-32 pathway, contribute to psychosis-like symptomatology. In these mice, the levels of the NR1 subunit of the NMDA complex is strongly reduced and the mice exhibit a pattern on DARPP-32 phosphorylation (i.e., increased phosphorylation of Thr34- and Ser130-DARPP-32 and unaltered phosphorylation of Thr75-DARPP-32) (32), similar to that observed following administration of the NMDA receptor antagonists PCP and MK-801 (5).

In conclusion, it appears that perturbations in signaling in any of several neurotransmitter pathways might produce psychotic symptomatology. Our data on DARPP-32 phosphorylation, together with several other observations (17, 22, 29), provide strong support for the notion that neurobiological mechanisms other than hyperactivity of the dopaminergic D2 pathway can lead to psychosis. Attacking these targets may therefore constitute innovative therapeutic approaches to the pharmacological treatment of schizophrenia.


Navigate This Article