, Volume 206, Issue 4, pp 631–640

Reversal of cognitive deficits by an ampakine (CX516) and sertindole in two animal models of schizophrenia—sub-chronic and early postnatal PCP treatment in attentional set-shifting


    • Department of In Vivo NeurobiologyH. Lundbeck A/S
    • Center for Neuropsychiatric Schizophrenia Research, Faculty of Health Sciences, Psychiatric Center GlostrupUniversity of Copenhagen
  • Birte Yding Glenthøj
    • Center for Neuropsychiatric Schizophrenia Research, Faculty of Health Sciences, Psychiatric Center GlostrupUniversity of Copenhagen
  • Rebecca Dias
    • Department of PsychopharmacologyH. Lundbeck A/S
  • Dorrit Bjerg Larsen
    • Department of Discovery ADMEH. Lundbeck A/S
  • Christina Kurre Olsen
    • Department of In Vivo NeurobiologyH. Lundbeck A/S
Original Investigation

DOI: 10.1007/s00213-009-1540-5

Cite this article as:
Broberg, B.V., Glenthøj, B.Y., Dias, R. et al. Psychopharmacology (2009) 206: 631. doi:10.1007/s00213-009-1540-5



Therapies treating cognitive impairments in schizophrenia especially deficits in executive functioning are not available at present.


The current study evaluated the effect of ampakine CX516 in reversing deficits in executive functioning as represented in two animal models of schizophrenia and assessed by a rodent analog of the intradimensional–extradimensional (ID–ED) attentional set-shifting task. The second generation antipsychotic, sertindole, provided further validation of the schizophrenia-like disease models.


Animals were subjected to (a) sub-chronic or (b) early postnatal phencyclidine (PCP) treatment regimes: (a) Administration of either saline or PCP (5 mg/kg, intraperitonally b.i.d. for 7 days) followed by a 7-day washout period and testing on day 8. (b) On postnatal days (PNDs) 7, 9, and 11, rats were subjected to administration of either saline or PCP (20 mg/kg, subcutaneously (s.c.)) and tested on PNDs 56–95, after reaching adulthood. The single test session required rats to dig for food rewards in a series of discriminations following acute administration of either vehicle, or CX516 (5–40 mg/kg, s.c.), or sertindole (1.25 mg/kg, perorally).


The specific extradimensional deficits produced by sub-chronic or early postnatal PCP treatment were significantly attenuated by sertindole and dose-dependently by CX516.


Findings here further establish PCP treatment as model of executive functioning deficits related to schizophrenia and provide evidence that direct glutamatergic interventions could improve these, when assessed in the ID–ED attentional set-shifting task.


SchizophreniaCognitionPhencyclidineRatAttentional set-shiftingID–EDAmpakineCX516Sertindole


Schizophrenia is a severe brain disease with complex pathogenetic and pathophysiological mechanisms and inadequate treatments (Green 1996; Hyman and Fenton 2003) that affects the young and often stays with them for life. The disease is described by positive and negative symptom clusters and cognitive deficits. One important decisive factor of a patient’s functional outcome is impairments within the cognitive domain (Green 1996; Matsui et al. 2008), including attention, working memory, and executive functioning. The deficits in executive functioning seen in schizophrenia can be modeled in rats, by selective lesions to the medial prefrontal cortex (mPFC; Birrell and Brown 2000) and assessed by a rodent analog of the Wisconsin Card Sorting Test or the computerized intradimensional–extradimensional (ID–ED) test. Both assess a subject’s ability to form, maintain, and shift an attentional set. The selective ED shift deficits observed in nonhuman primates with lesions to the dorsolateral prefrontal cortex (Dias et al. 1996a), in humans with damage to the prefrontal cortex (Owen et al. 1991), and in first-episode schizophrenia patients (Fagerlund et al. 2004; Pantelis et al. 1999) are also mimicked in the rat by lesions to the mPFC (Birrell and Brown 2000).

Cognitive symptoms induced by the psychotomimetic noncompetitive N-methyl-d-aspartate receptor antagonist phencyclidine (PCP) in healthy humans (Javitt and Zukin 1991; Krystal et al. 1994; Luby et al. 1959) and animals (du Bois and Huang 2007; Jentsch and Roth 1999; Morris et al. 2005) closely resemble those observed in schizophrenia patients. Specifically, several studies have shown that PCP administered to healthy volunteers induces schizophrenia-like symptoms, by affecting prefrontal areas relevant to deficits in executive functioning (Carlsson et al. 2000; Goff and Coyle 2001; Javitt and Zukin 1991). Moreover, PCP administration affected the ability to perform an attentional set-shifting task (Javitt and Zukin 1991), linking this to first-episode schizophrenia patients (Fagerlund et al. 2004; Pantelis et al. 1999). Among PCP-induced animal models of schizophrenia, many have been shown to be sensitive to the ID–ED testing paradigm (Featherstone et al. 2007; Egerton et al. 2005; Broberg et al. 2008; Rodefer et al. 2005; McLean et al. 2008). However, two models are of particular interest—both models present phenotypes specifically impaired at the ED shift performance, namely, the early postnatal PCP treatment model (Broberg et al. 2008; Wang et al. 2001) and adult sub-chronic PCP administration followed by a washout period model (Jentsch et al. 1997; Rodefer et al. 2005). Early postnatal treatment of rats with PCP on postnatal days (PNDs) 7, 9, and 11 was proposed as a neurodevelopmental model of schizophrenia (Wang et al. 2001). This treatment paradigm produces widespread neurodegeneration in brain areas relevant to the cognitive deficits observed in schizophrenia patients, such as the hippocampus and frontal cortex (Ikonomidou et al. 1999; Wang and Johnson 2005). Likewise, the adult sub-chronic PCP administration followed by a washout period paradigm has been shown to affect hippocampal and fronto-cortical areas, thereby establishing a hypofunction of the prefrontal cortex which resembles the schizophrenia pathology (Cochran et al. 2003).

The glutamate hypothesis of schizophrenia is now widely accepted and clinical evidence showing the effect of drugs acting directly on the glutamate system has also started to emerge (Patil et al. 2007; Weinberger 2007). For example, the use of the ampakine (i.e., positive alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor modulator) CX516 as an add-on treatment to clozapine was reported to exert a beneficial effect on cognitive functioning (Goff et al. 2001). Although this result was later challenged in a follow-up study (Goff et al. 2008), the rational for investigating ampakines as potential therapeutics in a psychiatric disease like schizophrenia still holds promise (Lynch 2006). Hence, the purpose of the present study was to evaluate whether the reversal of cognitive deficits observed in the clinic would translate to the preclinical setting in models resembling the schizophrenia pathology.

We studied the ability of CX516 to reverse attentional set-shifting performance deficits induced by (a) early postnatal and (b) adult sub-chronic PCP treatments, using the rodent ID–ED set-shifting paradigm. The second generation antipsychotic drug, sertindole, was also included as positive control. Sertindole has previously been shown to be effective in reversal of sub-chronic PCP-induced deficits in executive functioning (Goetghebeur and Dias 2009; Rodefer et al. 2008), whereas the effect in the early postnatal PCP model remained to be studied.

Materials and methods


Animals (except from pregnant dams) were housed two rats per cage under controlled conditions (12 h of light starting at 06:00; 20 ± 2°C; 30–70% humidity) in Macrolon (type III) cages with standard sawdust bedding and environmental enrichment (plastic house and wooden chew blocks).


Early postnatal PCP paradigm, previously described in (Broberg et al. 2008): Timed pregnant Lister Hooded rats were obtained at gestational days 14/15 from Charles River (Germany) and were housed individually, in standard conditions as described above, including nesting material, until delivery. The day of parturition was counted as postnatal day (PND) 0. On PND 7, pups were cross-fostered and randomly assigned to a lactating dam, from which they were weaned on PND 25. Animals were tested after reaching adulthood (Pignatelli et al. 2006) on PNDs 56–95.

Adult sub-chronic PCP paradigm: Male Lister Hooded rats (56–63 PND at the time of testing) were obtained from Charles River (Germany) and housed under standard conditions. Animals were acclimated for at least 5 days before the start of sub-chronic dosing and were always tested on day 8 after the last PCP dose.

Animals were given access to water and food ad libitum; however, during training and test (ID–ED test), food was restricted (11 g food per animal per day). During weekends, animals had access to food ad libitum until 12–16 h before the next testing/training day. A total number of 202 animals were used in this experiment.


PCP (molecular weight (MW) 243.4 g/mol, H. Lundbeck A/S) dissolved in isotonic (0.9% w/v) saline. CX516 (piperidin-1-yl-quinoxalin-6-yl-methanone, MW 241.29 g/mol, H. Lundbeck A/S) was dissolved in isotonic water (d-glucose) and pH adjusted to 7 using methansulfonic acid. Sertindole (1-[2-[4-[5-Chloro-1-(4-fluorophenyl)-1H-indol-3-yl]-1-piperidinyl]ethyl]-2-imidazolidinone, MW 441 g/mol, H. Lundbeck A/S) was diluted in 5% β-cyclodextrin at a concentration of 1.25 mg/kg.

The early postnatal PCP administration procedure was adopted from (Wang et al. 2001) and applied, namely subcutaneous (s.c.) dosing of PCP on PNDs 7, 9, and 11 in a 10-ml/kg dose volume. The choice of these specific dosing time points within the first 2 weeks of the rats’ postnatal life builds on a suggested correspondence to the second late trimester in human pregnancy in terms of neurodevelopmental changes (Bayer et al. 1993; Clancy et al. 2001). In addition, the timing is in accordance with a prior hypothesis stating that exposure to PCP during this trimester increases the probability for the progeny to develop schizophrenia (Deutsch et al. 1998; Green et al. 1994). Based on previous results (Broberg et al. 2008), an early postnatal PCP dose of 20 mg/kg was applied during ID–ED testing in the current study.

For the sub-chronic PCP administration procedure, rats were administered 5 mg/kg, intraperitonally (i.p.), 1 ml/kg, b.i.d. for 7 days at 7 am and 7 pm. This was followed by a 7-day washout period prior to behavioral testing. This treatment regime is in accordance with previous methods (Jentsch et al. 1997; Rodefer et al. 2005, 2008; Goetghebeur and Dias 2009).

For comparative evaluation of the effect on deficits in executive functioning, sertindole was dosed 120 min prior to presentation of the first discrimination problem (1.25 mg/kg, perorally (p.o.)), and this dose was adapted from Goetghebeur and Dias (2009) and Rodefer et al. (2008). Based on exposure data (see Table 1), it was determined to dose CX516 (5, 10, 20, and 40 mg/kg, s.c.) at two time points, first 30 min prior to presentation of the first discrimination and second approximately 60 min later (before starting the intradimensional shift 2 reversal discrimination), meaning that all animals were dosed at the same stage of the ID–ED task. Control animals (acute vehicle injections) were counterbalanced for s.c. and p.o. administrations, and the different vehicles were applied.
Table 1

Plasma and brain exposure following s.c. acute administration of CX516 to sub-chronic PCP (columns I and III) and early postnatal PCP (II and IV) treated animals


Plasma (µM)

Brain (µM)





CX516 was administered at 40 mg/kg and exposure was measured at time points (30, 60, and 90 min)

30 min

68.8 ± 7.4

59.8 ± 5.0

32.8 ± 1.9

29.0 ± 4.0

60 min

20.9 ± 2.1

27.1 ± 0.6*

13.6 ± 0.7

15.9 ± 0.8

90 min

5.1 ± 1.0

4.6 ± 0.8

3.0 ± 0.6

2.6 ± 0.4

CX516 (5, 10, 20, or 40 mg/kg) was administered at time 0 min and again at 60 min

5 mg/kg

7.3 ± 0.2

7.1 ± 0.7

3.3 ± 0.3

3.1 ± 0.3

10 mg/kg

14.2 ± 1.2

17.1 ± 0.4

10.1 ± 2.3

10.1 ± 0.8

20 mg/kg

27.6 ± 3.7

39.7 ± 1.6*

15.0 ± 1.9

21.8 ± 1.0**

40 mg/kg

82.0 ± 6.4

82.2 ± 6.5

33.0 ± 3.4

39.6 ± 4.1

All measures were sampled at 90 min. Mean ± SEM (n = 4)

*p < 0.05; **p < 0.001


The test apparatus used was a 30 × 39.5 × 60-cm opaque black test box, divided into three equal-sized areas (a starting–holding area separated by a sliding solid door from a choice area subdivided into two). The floor of the test box was punctured with rows of holes and the whole apparatus was placed on a downflow table to minimize the spread of allergens and mixing of odors. In the choice area, two 11-cm-diameter terracotta pots were placed recessed into the test box floor (with a 2-cm pot lip remaining above the floor) and separated by a divider. Medium cues were added to the pots; odor cues (oils from The Body Shop, Denmark) were applied around the rim of each pot, and all were novel to the rats. Oils were applied the evening before testing to allow full absorption into the pot and thus avoiding the rats’ taking the scent away on their whiskers or fur during the test sessions.

Behavioral testing

The ID–ED test procedure was identical to that reported previously (Goetghebeur and Dias 2009). Briefly, rats were required to locate a food reward (1/2 Honey Nut Cheerio, General Mills, Minneapolis, USA) on the basis of digging media or odor as the relevant perceptual dimension. Prior to the test, rats were trained to dig for a food reward in terracotta pots filled with standard bedding media. For 3 days, rats were habituated to the test pots filled with cage bedding and food rewards. On day 4, all rats were then presented with two different media and, thereafter, two different odors, and were required to distinguish which of the two media or two odors were associated with the food reward.

Subsequently (on day 5), rats were subjected to the following discriminations: simple discrimination (SD), compound discrimination (CD), intradimensional shift 1 (ID1), intradimensional shift 2 (ID2), intradimensional shift 2 reversal (ID2R), extradimensional shift (ED), and extradimensional shift reversal (EDR). The SD test required the rat to discriminate between two different cues within the same dimension (media only, no odors). In the CD, the pots contained the same media as in the SD, but with odors now added to them. New sets of media and odors were employed in ID1, ID2, and ED. In the ID2R and EDR, the incorrect stimuli from ID2 and ED were now correct and vice versa (see Table 2). Animals were required to make six consecutive correct trials in order to advance to the next discrimination. Rats were allowed one “discovery” trial at the start of each training or test discrimination problem in which they were allowed to self-correct a dig in the wrong pot. The discovery trial was counted towards the number of trials to performance criterion if the rat dug first in the correct pot, but not if the rat self-corrected an initial incorrect dig. The number of trials to reach the criterion level of performance at each test stage was recorded. Omissions were defined as an animal’s refusal to participate in the task (sniff or dig in either pot) for greater than 15 consecutive minutes. After an omission, animals were returned to the home cage to rest for approximately 30 min before testing was resumed, continuing from the trial the animal had reached prior to its omission break. A cut off time of 6 h or 100 trials (whichever came sooner) was imposed, after which the animal was not entered in the study. For the present study, a media to odor switch only was given to all animals to reduce variation.
Table 2

Order of discriminations and examples of possible combinations of stimulus pairs



Examples of combinations





Simple (SD)







Compound (CD)








Intradimensional shift 1 (ID1)
















ID2 reversal








Extradimensional shift (ED)








ED reversal








Finally, a quasirandom sequence-generating Excel program was applied to predetermine the order and position in which the individual pots appeared across the trials. Rats were counterbalanced randomly according the drug treatment they received (n = 8–14 per group), and the experimenter testing the rats remained blind to drug condition.


Exposure data of CX516 were collected in two separate experiments. Firstly, the pharmacokinetic profile of CX516 was assessed for a dose of 40 mg/kg (s.c), with collections at 30, 60, and 90 min (n = 4), in both the sub-chronic and early postnatal animal disease models. Secondly, in order to investigate the relationship between CX516 plasma and brain concentrations and the effect on reversal of the PCP-induced ED shift performance deficit (pharmacodynamics), CX516 was dosed (5, 10, 20, or 40 mg/kg) twice, first at 30 min prior to the presentation of the first discrimination and again approximately 60 min later. Samples were collected at 90 min (n = 4).

Plasma and brain contents were determined by liquid chromatography/tandem mass spectrometry on a Sciex API 4000 (Applied Biosystems). Parent > daughter molecular masses were 241.9 > 157.1. Half of each brain was homogenized in four volumes of MeCN/H2O followed by centrifugation and isolation of the supernatant. Plasma and brain samples were spiked 1:1 with internal standard. On-line sample preparation and liquid chromatography was performed with turbulent flow chromatography (Cohesive Technologies), using a dual column configuration. The calibration curves were between 1 and 1,000 ng/ml.

Statistical analysis

The data for all experiments are presented as mean values + standard error of the mean (SEM). ID–ED data were analyzed using a two-way analysis of variance (ANOVA) of repeated measures (animal), with task and treatment as the two factors. Tukey post hoc analyses were used to test differences in means between treatment groups following significant F values. Plasma and brain exposure data presented in Table 1 were subjected to logarithmic transformation to ensure normal distribution, before applying Student’s t test to compare sub-chronically to early postnatally PCP-treated animals for each concentration or time point. P < 0.05 was considered to be the level of significance in all statistical tests.


ID–ED test procedure

The number of trials to reach criterion are presented for each discrimination for the sub-chronic PCP (Fig. 1) and the early postnatal PCP (Fig. 2) paradigms. ANOVA revealed a significant overall treatment × discrimination interaction in both the sub-chronic PCP (F(36, 372) = 2.56; p = 0.001; Fig. 1) and the early postnatal PCP (F(36, 432) = 2.49; p = 0.001; Fig. 2) treatment regimes. Post hoc analysis also confirmed the formation of an attentional set via comparison of performance (trials to criterion) at the ID2 vs. ED test stage for vehicle-treated control animals (p < 0.001 following both sub-chronic PCP and early postnatal PCP treatments), thus verifying that the control animals were “locked” into the relevant perceptual dimension. No differences between treatment groups in any other test stages (i.e., SD, CD, ID1, ID2, IDR, and EDR) were significant (p > 0.5 for all), implying that both the sub-chronic PCP and the early postnatal PCP treatments specifically affected ED shift performance only.
Fig. 1

Bar graph showing the number of trials to reach criterion of six consecutive correct trials for each discrimination following sub-chronic saline (VEHICLE; 1 ml/kg, b.i.d., 7 days), or PCP (5 mg/kg i.p., b.i.d., 7 days plus 7-day washout), or acute sertindole (1.25 mg/kg p.o.), or CX516 l (5, 10, 20, and 40 mg/kg, s.c.) prior to behavioral testing during a single test session conducted day 8 of the washout period. The extradimensional (ED) shift required significantly more trials to complete than an intradimensional (ID2) shift (#p < 0.001), for vehicle-treated control rats. Sub-chronic PCP treatment selectively impaired ED performance (***p < 0.001). Sertindole and CX516 (10 and 20 mg/kg doses only) significantly reversed the selective ED set-shift impairments induced by sub-chronic PCP. $p < 0.001 in comparison with sub-chronic PCP group at the ED discrimination. Error bars represent one standard error of the mean (SEM)
Fig. 2

Bar graph showing the number of trials to reach criterion of six consecutive correct trials for each discrimination following early postnatal saline (VEHICLE; 10 ml/kg, on PNDs 7, 9, and 11, s.c.), or PCP (20 mg/kg on PNDs 7, 9, and 11, s.c.), or acute sertindole (1.25 mg/kg, p.o.), or CX516 l (5, 10, 20, and 40 mg/kg, s.c.) prior to behavioral testing during a single test session conducted after PNDs 56–95. The extradimensional (ED) shift required significantly more trials to complete than an intradimensional (ID2) shift (#p < 0.001), for vehicle-treated control rats. Early postnatal PCP treatment selectively impaired ED performance (***p < 0.001). CX516 (i.e., 5, 10, and 20 mg/kg doses only) and sertindole significantly reversed the selective ED set-shift impairments induced by early postnatal PCP. *p = 0.027 and $p < 0.001 in comparison with early postnatal PCP group at the ED discrimination. Error bars represent one standard error of the mean (SEM)

With respect to the effect of acute pharmacological challenges, sertindole (1.25 mg/kg, p.o.) was effective in reversing the ED shift performance deficit induced by the sub-chronic PCP (p < 0.001) and the early postnatal PCP (p < 0.001), both compared to acute saline challenge. Similarly, two doses of CX516 (10 and 20 mg/kg, s.c.) were highly significant (p ≤ 0.001) in improving the ED shift deficit induced by either treatment regimes. Interestingly, CX516 administered at 5 and 40 mg/kg, s.c., was ineffective at reversing the ED impairment induced by the sub-chronic PCP treatment regime. In direct contrast, the CX516 dose of 5 mg/kg was able to reverse the ED deficit induced by early postnatal PCP (p < 0.05). Among treatments effective in reversing ED shift deficits (sertindole and CX516, 5 (only early postnatal PCP), 10, and 20 mg/kg in both models), neither of these groups differed significantly from the vehicle-treated control groups, with respect to their ED discrimination score (p’s > 0.05).


The pharmacokinetic profile of CX516 was assessed using a 40-mg/kg (s.c.) dose in both the sub-chronic PCP (columns I and III in Table 1), and the early postnatal PCP (columns II and IV in Table 1) schizophrenia-like disease models. As seen in Table 1, the plasma and brain concentration measurements are very similar between the two disease models, meaning that absorption and elimination are likely to be very similar. As reported earlier (Hampson et al. 1998b), the current experiment confirms that CX516 readily crosses the blood–brain barrier. The small difference found at 60-min post injection is significant, but not reflected in the corresponding brain concentration measurement. Using the formula T1/2 (half life) = ln(2)/k (elimination rate constant), the biological half life of CX516 in the rat can be roughly calculated to ~16 min.

Table 1 shows plasma and brain CX516 concentration values following dosing of all concentrations tested in the ID–ED test procedure (5, 10, 20, and 40 mg/kg). Dosing was carried out twice at time zero and again 60 min later. This dosing regimen was used to mimic the situation in the ID–ED test procedure and account for a short half-life of CX516 (see above findings and discussion). As in Table 1, it is also clear from Table 1 that plasma concentration measurements of CX516 are similar between the two disease models; therefore, the absorption and elimination are also likely to be similar. The only discrepancy between the two is at the 20-mg/kg dosing, which is reflected in both plasma (p < 0.05) and brain (p < 0.001) concentration measurements.


The present study evaluated the reversal of cognitive deficits by an ampakine (CX516) and sertindole in two schizophrenia disease-like animal models, namely the sub-chronic PCP and the early postnatal PCP treatment regimes. In accordance with previous publications (Broberg et al. 2008; Rodefer et al. 2005; Goetghebeur and Dias 2009), the current study demonstrated clearly that both preclinical models under investigation (sub-chronic and early postnatal PCP) were characterized by an impaired ability to shift attentional set. This was evident by a significant increase in the number of trials taken to reach criterion performance level at the ED shift stage of the task selectively.

A key strength of the current study, and present in both animal models, is the observation of a statistically significant difference between performance scores at the ID2 vs. ED test stages in control vehicle-treated animals, confirming the formation of an attentional set prior to shifting. Furthermore, the procedure applied here contains multiple ID discriminations (i.e., ID1 and ID2) and fewer reversal learning stages, in contrast to earlier studies (Barense et al. 2002; Birrell and Brown 2000). Such a regime appears more successful in inducing the formation of an attentional set and, possibly, helps to minimize the apparent frustrations induced by reversal learning. Overall, the results of the current study are congruent with findings showing ED shift performance impairment in rats with medial prefrontal cortex damage (Birrell and Brown 2000), and in monkeys with dorsolateral prefrontal cortex damage (Dias et al. 1996b). Comparing the current study to other studies, it should be taken into account that we here only test the media to odor shift. However, we are confident from the initial odor discrimination training data (data not shown) and from the data obtained in previous studies (Broberg et al. 2008; Goetghebeur and Dias 2009) that both vehicle and PCP-treated animals are able to detect and use odors, and therefore the deficit at the ED stage of the task is a genuine impairment in attentional set-shifting performance.

The current study aimed to evaluate the potential of the ampakine CX516 to reverse cognitive deficits in two potential preclinical rodent models of schizophrenia. More specifically, a dose range (5–40 mg/kg) of CX516 was evaluated for its potential to reverse deficits in executive function, as assessed using the ID–ED attentional set-shifting task to test rats treated with either sub-chronic PCP or early postnatal PCP. Interestingly, this study provides evidence that CX516 can, in a dose-dependent manner, reverse the deficits in executive functioning modeled by two animal models of schizophrenia. More specifically, doses of 10 and 20 mg/kg (s.c.) were effective in reversing PCP-induced deficits at the ED shift step for both the sub-chronic and the early postnatal dosing regimen, whereas the 5 mg/kg (s.c.) dose of CX516 was only effective in reversing the ED shift deficit, when applied to animals treated with early postnatal PCP, while a dose of 40 mg/kg (s.c.) could not reverse the deficit in any of the two disease models.

From exposure data in Table 1, it is clear that absorption and elimination of the ampakine CX516 are very similar for the two disease models applied here. The only major difference is observed at a dose of 20 mg/kg (Table 1); however, this difference is not reflected in the behavioral ID–ED test and is probably due to a larger standard error of the mean within the sub-chronic PCP treatment group. Comparing the collected behavioral data (i.e., trials to criterion at the ED step) to exposure levels listed in Table 1, it is seen that plasma and brain concentrations of doses effective in reversing ED shift deficits (5 mg/kg excepted) convert to a plasma concentration range of 14.2–39.7 µM at the ED step (modeled in the exposure study). This dose range correlates well with other rat studies, in which doses of 12.5 to 35 mg/kg improved performance in various animal behavioral studies (Granger et al. 1993; Hampson et al. 1998a; Larson et al. 1995; Rogan et al. 1997). The exposure data also confirm what is found in other literature on CX516, namely a rather short half-life of approximately 20 min (~16 min in the current study) in rats (Hampson et al. 1998a) and 1 h in humans (Goff et al. 2008). We tried to compensate for this by using a repeated dosing regimen in the present study. As seen from exposure data in Table 1, the peak plasma concentration (Cmax) is reached around 30 min postdrug administration or earlier; thus, in the current study, we first injected animals 30 min pre-testing and secondly after approximately 60 min (before starting the ID2R discrimination), to compensate for the short half-life and reach a second Cmax around the time of the extradimensional shift. Thus, the present data support the potential of ampakines, such as CX516, in relieving the cognitive deficits associated with schizophrenia, but success seems highly dependent on thorough consideration of the pharmacokinetic/pharmacodynamic relationship.

In a clinical setting, a preliminary study showed a positive clinical effect of CX516 when given as an add-on therapy to clozapine (Goff et al. 2001). Unfortunately, these findings could not be replicated in a recent follow-up study (Goff et al. 2008), despite good statistical power and a low attrition. It is well known that CX516 is a low potency agent (for review see Arai and Kessler 2007; Black 2005); hence, a sub-optimal exposure might underlie the lack of effect in clinical studies. And although few clinical studies have reported plasma concentration levels, a commonly applied clinical dose of 900 mg converts to ~13 mg/kg (p.o.) in a person of average weight and corresponds to a plasma concentration of about 8 µM (Lynch et al. 1997; Ingvar et al. 1997). Even though the clearance is lower in humans, this comparably low plasma level might explain the negative clinical findings. This is supported, in part, by data in the current study, where a dose of 5 mg/kg CX516 did not improve rat ID–ED test performance in the sub-chronic PCP treatment regimen and showed only minimal significance in the early postnatal PCP treatment regimen. Even though the no effect limit, was not found with reversal of attentional set-shifting deficits by CX516 in the early postnatal PCP treatment regimen, it appears that a “U-shape” dose–response relationship exists in both animal models of schizophrenia. An absent or diminished effect at the lower dose (5 mg/kg) could be explained simply by the lower plasma concentration, and thus lower brain exposure, as could be the case for the sub-chronically PCP-treated animals. For animals treated with early postnatal PCP, however, the effect of the 5-mg/kg dose might suggest a higher sensitivity towards CX516 exposure in this animal model.

In another clinical study, CX516 was administered as a single agent, but was ineffective in improving psychosis and cognition measures in schizophrenia patients; however, this preliminary study was both underpowered and suffered from patient dropout, especially at the higher doses of CX516 (Marenco et al. 2002). In the present study, a “wash-out” of the CX516 reversal effect at the 40-mg/kg dose was seen in both animal disease models. Altered behavior at higher doses was observed in another animal study, where a suppression of exploratory behavior was observed (Granger et al. 1993). In the current study, no apparent adverse effects were observed for CX516 at doses from 5 to 40 mg/kg; however, in a pilot study testing a dose of 80 mg/kg, we observed that the rats lost interest in digging for food rewards. Instead they showed an overt licking behavior, licking silicone connections in the test box. Although speculative, we suggest that this could be due to the excitotoxic action of CX516 caused by stimulation of the glutamatergic system.

The literature contains numerous studies describing CX516 function in brain slices and animal models (for review see Arai and Kessler 2007; Black 2005). In general, the effect of CX516 arises from the ability to augment AMPA receptor transmission and promote long-term potentiation (LTP) formation, when investigated in hippocampal slices (Granger et al. 1993; Staubli et al. 1994; Arai et al. 1994). Since the results from in vivo studies also demonstrated a promotion of LTP by CX516, it was subsequently investigated for its positive effect in a number of animal models of memory. CX516 administration has been proven to promote long-term reference memory, as well as short-term and working memory (for review see Black 2005). Interestingly, in two later studies looking at AMPA function, it was first shown that CX516 did not produce an enhancement in LTP in hippocampal slices (only a lowering of stimulations threshold), whereas it did in slices from the PFC (Black et al. 2000). Furthermore, CX516 was shown to enhance AMPA response in the PFC, both in vitro and in vivo (Baumbarger et al. 2001). Although the discrepancy between these data sets is unclear, the latter setting together with the present data would, in theory, make CX516 ideal for the treatment of the PFC-mediated executive functioning deficits seen in schizophrenia.

Finally, and in agreement with earlier findings (Goetghebeur and Dias 2009; Rodefer et al. 2008), acute administration of the antipsychotic sertindole clearly reversed the selective ED performance deficits induced by sub-chronic PCP treatment, in the present study. Here, we also demonstrated, for the first time, that acute sertindole treatment could also reverse deficits induced by early postnatal PCP treatment. Similarly, sertindole had a positive effect on executive functioning performance, when measured in the Wisconsion Card Sorting Test in humans (Gallhofer et al. 2007). As discussed by Goetghebeur and Dias (2009), this represents a translational link to the rat ID–ED task and adds to the predictive validity of the attentional set-shifting assay, and thereby also supporting a potential future for CX516 or related drugs in the treatment of cognitive deficits associated with schizophrenia, especially disturbances in executive functioning. To further evaluate the validity of the early postnatal PCP model, assessment in the ID–ED task at an earlier time point (e.g. PND 35) would be valuable. This would give an indication as to how early the cognitive deficits are present in the disease model, and thereby how closely it mimics schizophrenia (Jones et al. 1994).


In summary, schizophrenia-related cognitive impairments are poorly treated by existing therapies, possibly due to an exclusive focus on dopamine D2 receptor antagonism, as being the main predictor of drug efficiency (Weinberger 2007). The current study represents a different approach building on the glutamate hypothesis of schizophrenia, suggesting disturbances in the glutamate homeostasis as an underlying cause of the executive functioning deficits seen in schizophrenia (Owen et al. 1991). The two animal models of schizophrenia (i.e., sub-chronic and early postnatal PCP models) tested here represents different ways to model the schizophrenia disease pattern. And although similar in their detection of effects in the ID–ED task when treated with CX516 and sertindole, there are important differences, like the time required to prepare animals and consistency of the induced deficit. Indeed, the early postnatal model takes longer to prepare, but then offers a larger window for applying chronic antipsychotic drug treatment to mimic the clinical situation more closely. This study emphasizes the important role of glutamate homeostasis in the brain and increases our insight into possible novel treatment regimens for cognitive disturbances affecting schizophrenia patients. Interestingly, attempts to reverse the PCP-induced deficits in attentional set-shifting by the use of the ampakine CX516 and the second generation antipsychotic sertindole were similar in these theoretically different animal models of schizophrenia. These data add further to support the potential of both approaches in relieving the cognitive deficits associated with schizophrenia, thereby improving the translational aspect of the ID–ED test paradigm.


Funding from H. Lundbeck A/S, the Faculty of Health Sciences, University of Copenhagen, and the Graduate School of In Vivo Pharmacology, University of Copenhagen supported this research. The authors specifically thank Mr. Pascal Goetghebeur, Mrs. Tanja Bruun, and Mr. Christian S. Pedersen for help and training in the attentional set-shifting task procedure. Additionally, Mr. Anders Sylvest is thanked for his help in preparing the animals for the final test.

Ethical standards

All animal procedures were carried out in compliance with the European Commission Directive 86/609/EEC and with Danish law regulating experiments on animals.

Copyright information

© Springer-Verlag 2009