Perspectives in Pharmacology Time-Dependent Cognitive Deficits Associated with First and Second Generation Antipsychotics: Cholinergic Dysregulation as a Potential Mechanism

Although cognitive dysfunction is considered one of the more debilitating symptoms of schizophrenia, there is a fundamental gap in our knowledge of how the primary pharmacologic treatments of this disease, firstand second-generation antipsychotics (FGAs and SGAs, respectively), affect cognition, particularly over extended periods of time. Moreover, it has been known for decades that chronic treatment with FGAs can lead to imbalances in cholinergic function in the striatum that result in movement disorders; however, there is a growing body of evidence to suggest that both FGAs and SGAs can lead to cholinergic alterations in brain areas more traditionally considered as memory-related, such as cortical and hippocampal regions. Data from our laboratories in rodents indicate that some SGAs (if administered for sufficient periods of time) can be associated with impairments in memory-related task performance as well as alterations in the cholinergic enzyme choline acetyltransferase, the vesicular acetylcholine transporter, and nicotinic ( 7) and muscarinic (M2) acetylcholine receptors. Given the well documented importance of central cholinergic function to information processing and cognitive function, it is important that the mechanisms for such chronic antipsychotic effects be identified. In this review, two potential mechanisms for long-term antipsychotic-related cholinergic alterations in the central nervous system are discussed: 1) antipsychotic antagonist activity at dopaminergic-D2 receptors on cholinergic neurons and 2) antipsychotic effects on neurotrophins that support cholinergic neurons, such as nerve growth factor and brain derived growth factor. Novel strategies to optimize the therapeutics of schizophrenia and maintain cognitive function via adjunctive cholinergic compounds and antipsychotic crossover approaches are also discussed. Schizophrenia is a chronic and disabling illness characterized by severe behavioral symptoms that commonly require life-long therapeutic intervention. Fortunately, the pharmacologic treatments for schizophrenia, which include firstand second-generation antipsychotics (FGAs and SGAs, respectively), have been demonstrated in numerous clinical trials to improve the behavioral symptoms in most schizophrenic patients. It has been widely held since their advent in the late 1980s that SGAs (as a class of compounds) have many advantages over FGAs, including greater improvements in negative symptoms, prevention of relapse, increased functional capacity, fewer movement-related side effects, and superior effects on cognition (reviewed in Miyamoto et al., 2005). Accordingly, SGAs presently account for nearly all of the new This work was supported by the National Institute of Mental Health (Grant MH 066233 to A.V.T.). We have also provided consultation or performed research (relevant to this article) either contractually or in collaboration with a several pharmaceutical companies, including Abbott Laboratories, Janssen Pharmaceutica, Lilly, Memory Pharmaceuticals, Merck, Pfizer, Roche Palo Alto, and Targacept, Inc. Article, publication date, and citation information can be found at http://jpet.aspetjournals.org. doi:10.1124/jpet.106.106047. ABBREVIATIONS: FGA, first-generation antipsychotic; SGA, second-generation antipsychotic; AUC, area under the curve; AChE, acetylcholinesterase; BDNF, brain-derived growth factor; ChAT, choline acetyltransferase; ELISA, enzyme-linked immunosorbent assay; mAChR, muscarinic acetylcholine receptor; NGF, nerve growth factor; nAChR, nicotinic acetylcholine receptor; TD, tardive dyskinesia; VAChT, vesicular acetylcholine transporter. 0022-3565/07/3203-961–968 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 320, No. 3 U.S. Government work not protected by U.S. copyright 106047/3153883 JPET 320:961–968, 2007 Printed in U.S.A. 961 at A PE T Jornals on Sptem er 5, 2017 jpet.asjournals.org D ow nladed from prescriptions written for neuroleptics in the United Sates. Whereas only specifically United States Food and Drug Administration approved for treating psychotic symptoms in schizophrenia and bipolar disease, SGAs are now routinely used “off-label” for the behavioral symptoms of other illnesses in patients ranging in age from the very young (e.g., in autism and attention deficit hyperactivity disorder) to the very old (e.g., Alzheimer’s disease). This widespread off-label drug use is beginning to cause controversy now that several long-term side effects have emerged with SGAs, such as abnormal weight gain, development of diabetes mellitus, hyperlipidemias, etc. (reviewed in Gardner et al., 2005). Further concern has been fueled by the first phase of results (Lieberman et al., 2005) of the Clinical Antipsychotic Trials of Intervention Effectiveness. In this large prospective clinical study, the SGAs, olanzapine, quetiapine, risperidone, and ziprasidone, seemed to offer few significant advantages over the FGA perphenazine when dropout rates (ostensibly due to intolerable side effects, lack of efficacy, etc.) were considered. Given the well documented challenges associated with noncompliance to therapeutic regimens in schizophrenia, there could be several possible interpretations of the Clinical Antipsychotic Trials of Intervention Effectiveness study; however, collectively, the information summarized above would strongly support the conclusion that a more critical look at SGAs is warranted, particularly because they are markedly more expensive than conventional antipsychotics. Antipsychotics and Cognitive Function If SGAs (compared with FGAs) indeed have superior effects on cognition as has been reported (see Keefe et al., 2006), the continued preference for SGAs in the treatment of schizophrenia would be warranted. Neuropsychological evaluations of schizophrenic patients consistently indicate cognitive impairments approaching two standard deviations below the mean of control subjects (Wilk et al., 2004). Such deficits are evident at the onset of the illness, and in contrast to the positive and negative symptoms (which tend to be episodic), cognitive decrements in schizophrenia generally persist throughout the illness (Nuechterlein et al., 1992). In addition, compared with other symptom domains, cognitive dysfunction in schizophrenia (particularly in the domains of memory, executive function, and vigilance) appears to have the greatest impact on key indicators of functional outcome, which include the acquisition of social skills, success in rehabilitation programs, and vocational achievements (reviewed in Green et al., 2000). As noted above, studies focused on the cognitive effects of antipsychotics have commonly reported superior effects of SGAs over FGAs, and several studies have further suggested that SGAs improve cognitive function (reviewed in Harvey et al., 2004). However, a parsimonious interpretation of the currently available literature on this subject would have to take into consideration methodological weaknesses and other limitations of many of the studies. For example, in earlier studies, confounding variables included relatively wide study group heterogeneity, cognitive testing while florid psychotic symptoms were present, and extensive prior antipsychotic exposure by the study subjects. Other limitations included polypharmacy at the time of cognitive testing and the concomitant use of anticholinergic drugs (i.e., drugs known to impair cognition) by those treated with FGAs to control extrapyramidal symptoms (Sharma and Mockler, 1998). A common limitation of both older and newer studies is the large, potentially inappropriate doses of the FGAs used in the comparisons (Carpenter and Gold, 2002). Although some of the aforementioned study design flaws have been addressed in more recent investigations, many of the newer studies have been retrospective or open label in design, and virtually all the studies have been of relatively short duration (i.e., they have rarely exceeded a few months to a year in length, even though most schizophrenic patients require decades of antipsychotic therapy). Although there are a few exceptions (e.g., Wolff and Leander, 2003), the results of most animal studies conducted to date suggest that SGAs and FGAs are either inactive or that they exert negative effects on cognition. In acute studies, the FGA haloperidol (Ploeger et al., 1992) and the SGAs clozapine, olanzapine, and risperidone impaired place navigation in the Morris Water Maze (Skarsfeldt, 1996). Haloperidol (Beatty and Rush, 1983), clozapine (Addy and Levin 2002), and olanzapine (Levin et al., 2005) impaired spatial working memory in radial arm maze tests, and haloperidol, clozapine, and risperidone impaired performance of a delayed nonmatch to position task (Didriksen, 1995). In chronic studies (which are much more relevant to the more common clinical use of antipsychotics in schizophrenia), haloperidol, clozapine, and risperidone impaired acquisition in a radial arm maze task in rats whereas olanzapine had no measurable adverse effects (Rosengarten and Quartermain, 2002). In another chronic study, haloperidol was found to disrupt radial arm maze choice accuracy in rats, but only during the first week of administration (Levin et al., 1987). Most recently, Didriksen et al. (Didriksen et al., 2006) found that acute administration of clozapine and olanzapine impaired water maze performance in rats (replicating earlier studies), whereas in chronically treated animals, the impairments abated in the clozapine-treated animals but were exacerbated in the olanzapine-treated animals. Interestingly, the SGA sertindole remained without effect on water maze performance following acute or chronic treatment in these studies. The results of studies in our laboratories to date also indicate that, like FGAs, representative SGAs (e.g., olanzapine, risperidone, ziprasidone) can impair some cognitive tasks if they are administered for sufficiently long periods of time. Figure 1 provides a summary of six different chronic studies conducted in our laboratory that support such a conclusion. Rats were administered either vehicle alone (very dilute acetic acid in distilled water), 2.0 mg/kg/day haloperidol, 2.5 mg/kg/day risperidone, 10.0 mg/kg/day chlorpromazine, or 10.0 mg/kg/day olanzapine in drinking water for periods ranging from 45 to 180 days. After the treatment period, they were given a drug-free washout period and then tested in a water maze task. To assess the effects of the length of time of drug administration across the studies, the area under the latency learning curve for each animal was calculated, and then group performances were compared across the studies statistically (two-way ANOVA for effects of treatment and time of administration). This method of comparing latency area under the curve (AUC) in water maze studies to provide a more simple comparison between groups has been published previously (Youngblood et al., 1997). As shown in Fig. 962 Terry and Mahadik at A PE T Jornals on Sptem er 5, 2017 jpet.asjournals.org D ow nladed from