Neuroscience 2008—Noncholinergic Approaches to Cognitive Deficits in Schizophrenia
In his previous SfN meeting report, our correspondent Tony Altar assessed cholinergic approaches to treating aspects of schizophrenia psychopathology. In this report, he turns to approaches with H3 histaminergic, 5-HT6 serotonergic, α2 adrenergic, and 5-HT2 receptor antagonists.
15 December 2008. Given the increased emphasis on improving cognitive dysfunction in schizophrenia, dementia, and other illnesses, a variety of mechanisms were emphasized this year for treating cognitive deficits. While probably not sufficient to treat schizophrenia by themselves, selective drugs that work through these mechanisms may be effective adjuncts to typical antipsychotics, and if so, these mechanisms can become desirable components of atypical and "third-generation" (aripiprazole-like) antipsychotic actions. A challenge at this year’s SfN meeting was to show this in co-administration studies, or to engineer these actions into new compounds. Interestingly, as the astute scholar of SfN posters may have noticed, histaminergic, adrenergic, and serotonergic mechanisms share the ability to promote the release of acetylcholine (ACh) and glutamate, augment hippocampal activity, and, at least in rodent behavioral models, improve cognitive functions. Maybe there's a common intracellular mechanism for this, but that may be a topic for next year's meeting.
H3 histaminergic receptor research
Abbott scientists including Holly Robb and Richard Radek reported on ABT-239, an H3 receptor antagonist that improves sensory gating in DBA/2 mice (abstract 291.29). Like other H3 antagonists, ABT-239 probably works by blocking autoreceptors on histaminergic cell bodies, an action which has been proposed for many years to augment histamine release, and, through histamine release, the release of other neurotransmitters including ACh. In an impressive application of electrical recording from CA3 hippocampal neurons, electrodes were implanted in freely moving rats. Robb and colleagues first measured the positive (P50) and negative electrical (N40) deflections produced by an auditory click. This measure is a rodentia analogue of P50/N100 recording in humans. It would certainly be interesting to compare in rats the effects of drugs active in this neuron model with cognition tests based on habituation, which is also a domain of the hippocampus. Habituation models were clearly in vogue at this year’s meeting, including the easy-to-do novel object recognition test (NOR) in rodents, P50/N100 recording in humans, and that old standby, the prepulse inhibition model in rats and humans with psychoses.
Robb and Radek inferred about learning, or "sensory gating,” from the lowering of the overall deflection between the P50 and N40 responses measured in the first click (control) to a second auditory click (test) presented 500 msec after the initial click. The test/control ratio was 0.88 in vehicle-treated rats, and 0.53 in those treated with ABT-239. That result suggests that ABT-239 increases habituation. Other cognition-enhancing drugs, like α7 nicotinic agonists, AChE inhibitors including donepezil, and H3 blockers also decrease this ratio. The muscarinic antagonist scopolamine blocked the ABT-239 enhancement of sensory gating, while neither mecamylamine (a non-selective nicotinic blocker) nor an H1 blocker did. Thus, the improved sensory gating by ABT-239 is due to muscarinic receptor agonism by an H3 antagonism-induced increase in ACh release, and not by a greater stimulation of post-synaptic H1 or H2 receptors, which, like muscarinic receptors, also reside on post-synaptic neurons. Thus, muscarinic agonism promotes hippocampal habituation at the electrophysiological level as a result of H3 antagonism.
To improve "information processing and recognition memory," Alvin V. Terry from the Medical College of Georgia described a drug with M2 muscarinic antagonism (60 nM affinity), H3 antagonism ("sub μM affinity") and low μM potency in blocking AChE in the pharmacology. This "prototype” compound JWS-USC-75-IX (abstract 340.15) is an H2 blocker derived from the structure of ranitidine, itself a Zantac analogue. JWS-USC-75-IX reversed the impairment in NOR produced by the sub-maximal dose of 0.3 mg/kg scopolamine; the AChE inhibitor donepezil was a positive comparator in this study. Like risperidol, JWS-USC-75-IX blocked apomorphine decreases in prepulse inhibition, but unlike risperidol, JWS-USC-75-IX did not reverse MK801 decreases in PPI. Apparently, new drugs are being made in this project area.
5-HT6 serotonergic receptor research
The role of 5-HT6 serotonergic receptor antagonism in memory was elucidated by Wayne Rowe and David Lowe of Memory Pharmaceuticals (abstract 685.9). 5-HT6 receptors are found only in the CNS, where they are most dense in hippocampus, neocortex, hypothalamus, and caudate-putamen. This localization predicts diminished peripheral side effects, and is one of the attractive aspects of targeting 5-HT6 receptor antagonism for psychiatric disease. The greatly decreased number of 5-HT6 binding sites in the tau 301L mutant mouse model of Alzheimer's disease and ability of 5-HT6 antagonists to decrease GABA release and increase acetylcholine and glutamate release in the hippocampus and frontal cortex suggest preclinical effects across multiple cognitive domains. To test whether 5-HT6 antagonist administration could restore cognitive deficits associated with schizophrenia or mild cognitive impairment (MCI), the Lowe and Rowe team of researchers treated aged F344 rats with their clinical candidate, MEM 68626. They observed improved performance in spatial memory using the 8 arm radial maze. MEM 68626 given orally at 3 mg/kg showed a prolonged (~3 hr) t1/2 in rat blood, comparable to the competitor, SKB 742457, and achieved brain concentrations "sufficient for once a day dosing." While that remains to be determined in humans, MEM 68626 also improved novel object recognition in young rats following a 48-hour delay (natural forgetting was complete at one, two, and three days) and spatial navigation memory deficits of aged F344 rats in the Morris water maze. Development of this compound continues while, sadly, Memory Pharmaceuticals is closing its research doors, hopefully not permanently.
Suven Life Sciences scientists from India, as reported by Gopinadh Bhyrapuneni and Nageswararao Muddana, found that SUVN-502 is an orally available, CNS-penetrant 5-HT6 antagonist with 2 nM potency and 100-fold selectivity over other GPCRs that they tested (abstracts 739.9 and 739.13). Microdialysis in the hippocampus and frontal cortex of freely moving rats showed that at behaviorally effective doses, SUVN-502 produces variable increases in ACh release in both regions and increases glutamate in frontal cortex with little change in the hippocampus. SUVN-502 was also active in the NOR and Morris water maze tasks, as seen with other 5-HT6 antagonists (I hope you still remember from the last paragraph [hint: MEM 68626]) and reversed MK801-induced behavioral changes. Having satisfied safety profiling criteria for clinical trials, SUVN-502 has, since June 2008, been in Phase 1 trials. Phase 2 is anticipated to be in Alzheimer's disease and possibly schizophrenia after that.
A similar clinical path awaits A-964324, a 0.5 nM 5-HT6 receptor antagonist at human and rat receptors from Abbott, as reported by Karsten Wicke (abstract 160.2). A-964324 showed at least 500-fold selectivity over 80 other receptors or targets, and is thus also a competitor with the Suven and Memory compounds mentioned above, and GSK-742457 and SB-271046, which bind to 5-HT6 at 0.1 nM and 0.4 nM, respectively. A-964324 increases cortical ACh release in rats, improves social recognition memory as effectively as nicotine, and shows efficacy in a cognitive "flexibility" test.
Further evidence that 5-HT6 receptor antagonism may mediate weight gain increases was provided by David Heal of RenaSci Consulting, United Kingdom, working with Esteve Laboratories, Barcelona. Heal showed that the 13 percent weight gain produced by olanzapine could be blocked by co-administration of the 5-HT6 receptor agonist e-6837 (abstract 584.12). But e-6837 lowered food intake and body weight by itself, so this effect may not so clearly be specific to the 5-HT6 receptor.
α2 adrenergic receptor research
α2 adrenergic receptors are an old target for schizophrenia because so many atypical antipsychotics antagonize them with moderate potency. Continuing from findings first reported in high-profile publications, Torgny Svensson and colleagues from the Karolinska Institute proposed that adjunctive α2 adrenoceptor blockade used in combination with risperidone may provide a more advantageous effect profile in schizophrenia and improved efficacy due to the reduced need for D2 receptor occupancy and reduction of extrapyramidal side effects (abstract 55.17). The addition of the α2 antagonist idazoxan to haloperidol or olanzapine enhanced suppression of avoidances in the conditioned avoidance response (CAR), but idazoxan by itself had no effect. This result is consistent with the high predictive value of CAR for D2 antagonist approaches, and suggests that idazoxan by itself would not be expected to treat psychosis. But is there an inverse relationship between CAR potency and cognition, since CAR potency is associated with less conditioned responses? Idazoxan produced small increases in dopamine release in the nucleus accumbens (not so good, maybe) and frontal cortex (that's the ticket), and augmented the increase produced by risperidol. Oddly, though, risperidol also blocks α2 adrenoceptors, so this augmentation may be due to additional α2 antagonism. Idazoxan facilitated the increases in NMDA-induced current flow in pyramidal neurons of the medial prefrontal cortex to the level seen with clozapine and reversed memory disruption by NMDA blockade, all in rats. Thus, when added to antipsychotics, idazoxan promotes efficacy in the CAR model, in DA release, and in NMDA transmission models, suggestive of the role of α2 adrenoceptor antagonism in augmenting antipsychotic and in the cognitive effects of atypical antipsychotics.
5-HT2A/C receptor research
Herb Meltzer of Vanderbilt University and colleagues, in another important poster at this meeting, provided more support to the view we proposed over two decades ago (see Altar et al., 1986), that potent 5-HT2 receptor antagonism relative to D2 dopamine receptor antagonism contributes to the favorable profile of atypical antipsychotic drugs (abstract 760.13). Nowadays, those receptors are better described by 5-HT2A/C and D2/3. In a direct evaluation of this serotonin-dopamine receptor (SDA) hypothesis, they showed that increasing the "5-HT2A/C portion of the relationship benefits both efficacy and side effects." Pimavanserin (ACP-103) is a 5-HT2A/C inverse agonist that occupies 60 percent of 5-HT2A/C sites in vivo and 30 percent of D2 sites when given to rats at the impressively low dose of 0.1 mg/kg. In a six-week, double blind placebo-controlled study with 423 patients, antipsychotic efficacy measured by PANSS and the CGI scores showed that the efficacy of a low, 2 mg dose of risperidone was potentiated by 10 mg of pimavanserin, a dose that occupies ~80 percent of 5-HT2A receptors. Like we just noted with α2 antagonism, risperidone is also a 5-HT2A receptor blocker, so this effect of pimavanserin may also be an additive effect on top of risperidone. This is consistent with the lack of pimavanserin potentiation of high-dose risperidone. Interestingly, however, the efficacy of 2 mg haloperidol was not potentiated by pimavanserin, which may be a challenge to the SDA hypothesis. The pimavanserin + risperidone combination also diminished extrapyramidal side effects and body weight gain obtained with the low or high (6 mg) doses of risperidone, and did not elevate serum prolactin at the end of the study. Pimavanserin was never given by itself to patients; too bad, as that might have been an important test of whether 5-HT2A/C antagonism by itself is sufficient to treat schizophrenia and, if so, to what degree. That's a hot topic that will soon get a lot of attention, as intracellular interactions between the 5-HT2A receptor and metabotropic glutamate receptors are on the horizon.—C. Anthony Altar.