antipsychotics affect multiple calcium calmodulin

Neuroscience 161 (2009) 877– 886

ANTIPSYCHOTICS AFFECT MULTIPLE CALCIUM CALMODULIN DEPENDENT PROTEINS W. J. RUSHLOW,a,b* C. SEAH,a L. P. SUTTON,b A. BJELICAb AND N. RAJAKUMARa,b

Recent genetics studies have identified several genes that are associated with schizophrenia and may confer increased susceptibility to developing the disorder. One such gene, calcineurin, was found to be altered in the genomes of some schizophrenia patients and reduced mRNA levels have been reported in postmortem schizophrenic brains (Gerber et al., 2003; Eastwood et al., 2005). In addition, calcineurin knockout mice display behavioral changes consistent with the symptoms of schizophrenia (Miyakawa et al., 2003). Recently, we demonstrated that repeated administration of the antipsychotics haloperidol, risperidone and clozapine decreased the protein and mRNA levels but increased the phosphatase activity of calcineurin in the striatum of rats. The changes were specific to antipsychotics and potentially mediated by D2/D3 dopamine receptors, the receptor thought to be essential for the ability of antipsychotics to ameliorate the positive symptoms of schizophrenia (Kapur and Mamo, 2003; Rushlow et al., 2005). The increased calcineurin phosphatase activity observed following antipsychotic treatment is functionally opposite to what is expected in schizophrenia or following repeated treatment of rats with amphetamine (Gerber et al., 2003; Eastwood et al., 2005; Rushlow et al., 2005). Consequently, there is evidence to suggest calcineurin may be involved in the manifestation and amelioration of schizophrenic symptoms. In addition to calcineurin, previous studies have demonstrated that calmodulin (CaM) protein levels (increased) and distribution and calmodulin-dependent protein kinase II␣ (CaMKII␣) protein levels (decreased) and kinase activity (increased) are altered by haloperidol treatment in rats (Lau and Gnegy, 1982; Meshul and Tan, 1994). Our gene screening data (unpublished observations) and a recently published gene array study suggested that additional CaM dependent proteins may be also increased or decreased by antipsychotics including calmodulin-dependent protein kinase IV (CaMKIV) and calmodulin-dependent protein kinase kinase’s (CaMKK) (MacDonald et al., 2005). The purpose of the current study was to examine the effects of haloperidol, risperidone and clozapine on CaM and CaM-related proteins, particularly CaMKIV, CaMKK␣ and CaMKK␤. The effects of lithium, valproic acid and fluoxetine were also examined to determine if the response elicited by antipsychotics is specific. Finally, raclopride (D2/D3 dopamine receptor antagonist) and amphetamine were investigated to determine if any increases/decreases identified may be attributed to dopamine receptors (raclopride) and are reversed in an animal model of the positive symptoms of schizophrenia (amphetamine).

a

Department of Psychiatry, London Health Sciences Centre-University Campus, 339 Windermere Road, London, Ontario, Canada, N6A 5A5 b Department of Anatomy & Cell Biology, The University of Western Ontario, 1151 Richmond Street, London, Ontario, Canada, N6A 5C1

Abstract—Calcineurin is a calmodulin (CaM) dependent protein phosphatase recently found to be altered in the brains of patients suffering from schizophrenia and by repeated antipsychotic treatment in rats. Some data suggest, however, that antipsychotics and schizophrenia may have a more widespread effect on the CaM signaling axis than calcineurin alone. In the current study, the effects of selected psychoactive drugs were investigated using Western blotting, in situ hybridization and immunocytochemistry to determine if they target CaM, calmodulin-dependent protein kinases (CaMK) or calcineurin. Results indicated that repeated treatment with haloperidol, clozapine or risperidone increased CaM protein and CaMII mRNA levels but decreased calmodulin-dependent protein kinase II␣ (CaMKII␣) IV (CaMKIV), kinase ␣ (CaMKK␣), kinase ␤ (CaMKK␤) and calcineurin protein levels in the striatum of Sprague-Dawley rats (Rattus Norvegicus). Closer examination of CaMKIV, CaMKK␣ and CaMKK␤ revealed that the observed decreases in protein levels were short-lived following antipsychotic treatment and reversed (i.e. upregulated) 24 h post-treatment similar to what was previously reported for calcineurin. The D2/D3dopamine receptor antagonist raclopride mimicked the decreases in CaMKIV, CaMKK␣, CaMKK␤ and calcineurin observed following antipsychotic treatment whereas increases in these proteins were observed in an amphetamine model of the positive symptoms of schizophrenia. Mood stabilizers such as lithium and valproic acid or the antidepressant fluoxetine had no effect on CaMKIV, CaMKK␣, CaMKK␤ and calcineurin with the exception of an increase in CaMKK␤ following lithium treatment. The results collectively suggest that antipsychotic specifically target several proteins associated with CaM signaling. © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: antipsychotic, schizophrenia, calmodulin, striatum, haloperidol, clozapine. *Correspondence to: W. J. Rushlow, Department of Psychiatry, London Health Sciences Centre-University Campus, Room B9-144, 339 Windermere Road, London, Ontario, Canada N6A 5A5. Tel: ⫹1-519-6858500⫻34935; fax: ⫹1-519-663-3935. E-mail address: [email protected] (W. J. Rushlow). Abbreviations: ANOVA, analysis of variance; CaM, calmodulin; CaMKII␣, calmodulin-dependent protein kinase 2␣; CaMKII␤, calmodulin-dependent protein kinase II␤; CaMKIV, calmodulin-dependent protein kinase IV; CaMKK␣, calmodulin-dependent protein kinase ␣; CaMKK␤, calmodulindependent protein kinase ␤; CREB, cAMP response element binding protein; DARPP-32, dopamine and cAMP responsive phosphoprotein 32 kDa; GSK-3, glycogen synthase kinase-3; PBS, phosphobuffered saline; PKA, protein kinase A.

0306-4522/09 $ - see front matter © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.03.011

877

878

W. J. Rushlow et al. / Neuroscience 161 (2009) 877– 886

EXPERIMENTAL PROCEDURES

were housed in cages with free access to food and water on a 12-h light/dark cycle.

Drug treatment Adult male Sprague–Dawley rats (Charles River Laboratories, Saint Constant, QC, Canada) 14⫹ weeks of age were used throughout the study (n⫽4 – 8/experimental group). For the initial experiments, rats were injected daily with haloperidol (0.5 mg/kg), risperidone (0.5 mg/kg), clozapine (25 mg/kg) or appropriate vehicles for 28 days and sacrificed 2 h following the final injection. Vehicles for the antipsychotics were 50 mM HCl (pH 6.0) for haloperidol, 100 mM HCl for clozapine (pH 6.0) or 15% ethanol for risperidone. For the time series experiments, rats were injected daily with haloperidol (0.5 mg/kg), risperidone (0.5 mg/kg) or vehicle for 14 days and sacrificed 2, 4, 8 or 24 h post-final injection. Rats were also injected with clozapine (25 mg/kg) for 14 days and sacrificed 2 or 24 h post-injection. The original experiments were conducted using chronically treated rats (28 days) to ensure no changes would be missed due to timing. Later studies were conducted using shorter treatment periods after empirical testing revealed that 14 days was sufficient to achieve similar results. Haloperidol was obtained from Sigma-Aldrich Canada (Mississauga, ON) while risperidone and clozapine were obtained from Tocris Bioscience (Ellisville, MI, USA). To determine if the responses observed were specific to antipsychotics, rats were treated with lithium, valproic acid or fluoxetine. For fluoxetine treatment, rats were injected daily with either fluoxetine (5 mg/kg) or saline (vehicle) for 14 days and sacrificed 2 or 4 h post–final injection. Treatment with valproic acid (28 days) was achieved by adding sodium valproate directly to the drinking water (12 g/L sodium valproate supplemented with 300 mg/ml saccharin). Control rats received water containing 300 mg/ml saccharin alone. Alternatively, rats were treated with valproic acid through daily injections (300 mg/kg) for 14 days. Both methods have been used to treat animals with valproic acid in the literature and hence both methods were used to ensure changes were not potentially missed due to the method of administration. Control rats were injected with saline (vehicle) alone. Valproicinjected rats were sacrificed 2 or 4 h post–final injection. Rats were treated with lithium using rat chow supplemented with 0.25% LiCl for 21 days. Control rats were fed an identical rat chow mixture without the LiCl added. Finally, to determine if changes identified following antipsychotic treatment may be due to D2 dopamine receptor antagonism or found in a putative animal model of the positive symptoms of schizophrenia, the D2/D3 dopamine receptor antagonist raclopride (3 mg/kg) or amphetamine (5.0 mg/kg) was used respectively. Both drugs were injected daily for 14 days and the rats sacrificed 4 h post–final injection. Saline was used as the vehicle for both amphetamine and raclopride. Fluoxetine, raclopride, saccharin and valproic acid were obtained from Sigma-Aldrich (Mississauga, ON, Canada), amphetamine was obtained from Sigma-Aldrich (Gillingham, Dorset, UK) and LiCl containing rat chow along with control chow was obtained from Charles River Laboratories (Wilmington, MA, USA). The drug doses selected for the current study are routinely used in the literature and have been used by us in the past with good results (Rushlow et al., 2005; Alimohamad et al., 2005a; Sutton et al., 2007). In the case of the drugs used to treat psychiatric disorders, the doses selected are also thought to approximate therapeutic levels found in humans following treatment. Treatment duration and post-injection intervals were selected empirically based on our previous studies showing robust and reproducible changes in a variety of signaling proteins (Rushlow et al., 2005; Alimohamad et al., 2005a; Sutton et al., 2007). All experiments were conducted in accordance with Canadian Council on Animal Care guidelines and conformed to international standards on the ethical use of animals. Every effort was made to reduce the number of rats used in the study and minimize suffering. Rats

In situ hybridization Rats treated for 28 days with haloperidol, risperidone, clozapine or saline (control) (n⫽5/treatment group) were perfused using saline followed by 4% freshly depolymerized paraformaldehyde in phosphobuffered saline solution (PBS, pH 7.4). The brains were removed and cryoprotected using an 18% sucrose solution (sucrose in PBS) before sections (10 ␮m) of the striatum were cut using a cryostat (Microm). The sections were stored desiccated in sealed slide boxes at ⫺80 °C until needed. Three separate genes encoding CaM have been identified and designated CaMI, II and III. The length of the CaM mRNA transcribed from the three genes differs although they all produce a protein with the same amino acid sequence. Consequently, to examine the effects of antipsychotics on the different isoforms of CaM using in situ hybridization, cDNAs were generated using polymerase chain reaction (PCR) and primers directed against the variable non-coding regions of CaMI (AGTACCTTCTGTCCACACAC, GAAGAGGACTGTCCTACATC), II (CCATGTTGCATGTGGCTTCC, ACAGTCCACGCTCACAACTC) and III (TGATGACTGCGAAGTGAAGG, ATTGGCGTTTGCTAGAACCG) the cDNA fragments were cloned using the pGEM-T easy system (Promega, Madison, WI, USA) and selected clones sequenced to verify orientation and identity of the cDNA insert. Both T7 sense and antisense constructs were created and used to generate sense (control) and antisense (test) riboprobes. 35S-UTP labeled riboprobes for CaMI, II and III were transcribed, purified and quantified using T7 polymerase as outlined previously (Rushlow et al., 2005). In situ hybridization was conducted using CaMI, II and III riboprobes as outlined previously (Rushlow et al., 1999). Autoradiography and silver grain quantification were also performed as detailed previously (Rushlow et al., 2000). The data were analyzed statistically using an analysis of variance (ANOVA) followed by Tukey’s multiple comparison test.

Western blotting Protein was isolated from the striatum of treated and control rats using protein lysis buffer supplemented with protease inhibitors (Roche Diagnostics, Laval, QC, Canada). The protein was quantified, mixed with loading buffer, heat denatured, run on polyacrylamide gels and transferred to nitrocellulose using Mini-Protean III’s (Bio-Rad Laboratories, Mississauga, ON, Canada). The Western blot procedure and composition of the buffers are outlined elsewhere (Alimohamad et al., 2005a). Detection of the protein of interest was accomplished using target specific antibodies, appropriate horseradish peroxidase– conjugated secondary antibodies (Thermo Scientific, Fisher Canada, Nepean, ON) chemiluminescence (Thermo Scientific, Fisher Canada) and Kodak (Kodak Canada, Toronto, ON) X-ray film. In all cases, ␣-tubulin was used as the loading control. The blots were quantified using a grey scale– calibrated scanner and Kodak molecular imaging software. Densitometry data were analyzed statistically using an ANOVA followed by Tukey’s multiple comparison test or Student’s t-test as appropriate. The source and dilution of the primary antibodies used in the study along with the apparent molecular weights of the band that appeared on the Western blot are outlined in Table 1. All antibodies were obtained from Thermo Scientific, Sigma-Aldrich Canada, Zymed Laboratories (San Francisco, CA, USA) or Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Immunocytochemistry Rats (n⫽3/treatment group) treated with haloperidol, lithium, valproic acid, fluoxetine or appropriate vehicles (described above)


879

Table 1. Source and dilution of antibodies used for Western blotting Epitope

Source

Dilution (WB)

Approximate size (kDa)

CaM CaMKII␣ CaMKII␤ CaMKIV CaMKK␣ CaMKK␤ Calretinin Calbindin D28K ␣-Tubulin Horseradish peroxidase–conjugated anti-mouse, -rabbit or -goat

Zymed Laboratories Sigma-Aldrich Zymed Laboratories Sigma-Aldrich Santa Cruz Biotechnology Santa Cruz Biotechnology Santa Cruz Biotechnology Sigma-Aldrich (IC)/Santa Cruz Biotechnology (WB) Sigma-Aldrich Thermo Scientific

1:1000 1:1000 1:1000 1:10,000 1:100 1:100 1:250 1:250 1:120,000 1:20,000

17 50 50 65 65 67 (Doublet) 29 28 50 N/A

were perfused with saline followed by 4% formaldehyde. Haloperidol and fluoxetine rats were treated daily for 14 days and sacrificed 2 h post–final injection. Valproic and lithium animals were treated via inclusion in the drinking water or rat chow respectively as described above. The brains were removed, cryoprotected and cut (40 ␮m) using a freezing microtome. Sections from the striatum were stained for CaM dependent protein kinase IV, calretinin or calbindin using an avidin– biotin horseradish peroxidase labeling system (Vector Laboratories, Burlingame, CA, USA) as outlined previously (Rushlow et al., 1994). Stained sections were qualitatively examined for regional anatomical changes using a Zeiss Axiovert microscope and images captured using a Sony 3CCD video camera and Northern Eclipse software (Empix Imaging, Mississauga, ON, Canada). Immunocytochemistry for other proteins, such as CaMKK␣ and CaMKK␤, was not successful. Calcineurin in situ hybridization and immunocytochemistry were investigated previously following antipsychotic treatment and not repeated.

RESULTS CaM and CaM dependent proteins Since it is not possible to distinguish CaMI, II and III at the protein level, in situ hybridization was used to examine the different CaM isoforms following antipsychotic treatment. CaMI, II and III antisense riboprobes labeled neurons throughout the striatum while sense riboprobes only resulted in background labeling. CaMII appeared to be upregulated in the striatum following chronic antipsychotic treatment (Fig. 1A), whereas CaMI and III were unaltered. Regional quantification of silver grains in the medial striatum, lateral striatum and nucleus accumbens core confirmed that CaMII mRNA was significantly upregulated (n⫽4 – 8, P⬍0.05) in the striatum whereas CaMI and III mRNA levels were unchanged (Fig. 1B). These three regions were selected for analysis since they showed significant decreases in calcineurin mRNA levels previously and have also been implicated in schizophrenia (Eastwood et al., 2005; Rushlow et al., 2005). Western blotting confirmed that CaM protein levels were elevated in the striatum following repeated treatment with haloperidol, risperidone or clozapine paralleling increases in CaMII mRNA. Calcineurin and CaMKII␣ were included as positive controls since they have been shown to be affected by antipsychotics previously (Meshul and Tan, 1994; Rushlow et al., 2005) and responded as expected in the current study. Western blotting also revealed that CaMKIV, CaMKK␣ and CaMKK␤ were affected by 14 and 28 days of antipsychotic treatment (Figs. 2 and 4 respectively).

Although Western blotting revealed changes in the levels of several proteins following antipsychotic treatment, it was unclear if the changes were uniform or confined to different anatomical subdivisions of the neostriatum. Immunocytochemistry was used to qualitatively examine regional specificity. Immunocytochemistry for CaMKIV showed reduced levels of nuclear staining in the medial and lateral caudateputamen as well as the nucleus accumbens core following 14 days of haloperidol treatment. Representative images from control and haloperidol-treated rats are shown in Fig. 3. Lithium and valproic acid (oral) had no observable effect on the protein levels or distribution of CaMKIV consistent with the results of the Western blots (data not shown). Immunocytochemistry was attempted for the other proteins of interest but did not yield satisfactory labeling. Time course analysis The results of the Western blotting analysis indicate that the protein levels of CaMKIV are reduced following antipsychotic treatment. However, a recent gene array study suggested that CaMKIV mRNA levels are increased following antipsychotic treatment 24 h post–final injection (Macdonald et al., 2005). To clarify, a time course analysis was conducted for CaMKIV as well as CaMKK␣ and CaMKK␤ using different post-injection intervals (2, 4, 8 and 24 h) for risperidone and haloperidol. Clozapine was also tested but only at 2 and 24 h post-treatment. Results indicated that CaMKK␣, CaMKK␤ and CaMKIV levels are significantly reduced in the striatum soon after antipsychotic treatment for all drugs tested (n⫽4, P⬍0.05). However, by 24 h post-treatment, the protein levels of CaMKK␣, CaMKK␤ and CaMKIV were significantly increased in the striatum with the exception of CaMKK␤ and risperidone (Fig. 4). The same pattern was previously reported for calcineurin (Rushlow et al., 2005). Specificity With the exception of CaMKII␤, all of the calcium sensitive proteins tested showed significant changes in protein levels following antipsychotic treatment. To determine if calcium signaling may be globally affected by antipsychotics or if the response may be confined to a specific subset of calcium regulated proteins, Western blotting and immuno-

880


Fig. 1. (A) Darkfield images obtained from sections labeled for CaMII using in situ hybridization. Increased silver grain density is apparent in the medial striatum (MStr), lateral striatum (LStr) and nucleus accumbens (NAc) core following 28 days of treatment with haloperidol (Hal.), risperidone (Risp.) or clozapine (Cloz.) compared to control (Cont.). (B) Quantification of silver grain density in the striatum of sections labeled using in situ hybridization for CaMI, II or III following chronic treatment with haloperidol, risperidone or clozapine compared to control. Only CaMII revealed a significant change in mRNA levels relative to control (* n⫽8, P⬍0.01). Scale bar⫽10 ␮m. Images were obtained from coronal sections approximately 1.20 mm from Bregma.

cytochemistry were conducted for the calcium sensitive proteins calbindin D28K and calretinin. No significant changes in calbindin D28K or calretinin were detected following repeated administration of haloperidol, risperidone or clozapine (n⫽8, P⬎0.05, data not shown). To further evaluate specificity, rats were also treated with lithium, valproic acid or fluoxetine (n⫽8/treatment). These compounds are used to treat patients suffering from bipolar disorder and depression but have no effect on the

core symptoms of schizophrenia. Fluoxetine and lithium had no significant effect on the proteins of interest with the exception of lithium and CaMKK␤ (Fig. 5). Lithium caused a significant (n⫽8, P⬍0.05) upregulation in CaMKK␤ protein levels in the striatum following treatment. Valproic acid also had no effect on the CaM dependent proteins regardless of route of administration (Fig. 5, oral results shown). Although lithium, valproic acid and fluoxetine generally produced negative results with the proteins studied, the


881

Fig. 2. (A) Representative Western blots showing changes in CaM and CaM dependent protein levels in the striatum including calcineurin (CN) following haloperidol (Hal.), risperidone (Risp.) or clozapine (Cloz.) treatment for 28 days relative to vehicle (Veh.). Control samples are shown in the left lane, experimental samples on the right. Alpha-tubulin (␣-Tub) was used as a loading control. (B) Graph showing densitometry measurements obtained from the Western blots. Significant changes (* n⫽8, P⬍0.01) in CaM, CaMKII␣, CaMKII␤, CaMKK␣, CaMKK␤, CaMKIV and calcineurin were detected.

drugs had an effect on the brain since significant changes in phosphorylated glycogen synthase kinase-3 (GSK-3) (positive control) were identified as has been previously reported by other investigators following treatment (Beaulieu et al., 2004; Li et al., 2004). The beneficial effects of antipsychotics are thought to be mediated through D2 dopamine receptor (Kapur and Mamo, 2003), consequently rats were also treated with the D2/D3 specific dopamine receptor antagonist raclopride. Western blotting revealed that repeated treatment of raclopride significantly reduced the protein levels of CaMKIV, CaMKK’s and calcineurin in the striatum. Amphetamine, a drug used to generate an animal model of the positive symptoms of schizophrenia, was found to have the opposite effect on CaMKIV, CaMKK␣ and calcineurin by Western blotting compared to antipsychotics and raclopride (Table 2).

DISCUSSION Antipsychotics, CaM and CaM dependent proteins The results of the current study show that antipsychotics target the CaM second messenger system leading to alterations in the protein levels of CaM and most CaM de-

pendent protein kinases as well as the phosphatase calcineurin. The results are consistent with previous studies showing increased CaM and decreased CaMKII␣ protein levels due to haloperidol treatment (Gnegy et al., 1994; Meshul and Tan, 1994) but extend these findings to show that risperidone and clozapine (atypical antipsychotics) also target CaM and CaMKII␣. However, while the effects of antipsychotics on the CaM signaling axis appear to be extensive they are not complete. CaMKII␤ protein levels were unaffected by antipsychotics as were CaMI and CaMIII mRNA levels. Only CaMII mRNA levels were upregulated suggesting that the CaMII gene is responsible for the increased CaM protein levels observed following antipsychotic treatment. The mRNA data are consistent with previous findings showing that CaMI, II and III can be differentially regulated by pharmacological agents such as amphetamine (Michelhaugh and Gnegy, 2000). While antipsychotics have a widespread effect on CaM and CaM dependent proteins, they do not have an effect on other calcium sensitive proteins such as calbindin D28K and calretinin although there are reports that calbindin D28K and calretinin may be affected in schizophrenia (Reynolds, 1989; Eyles et al., 2002; Tooney and Chahl,

882


response in the striatum following clozapine treatment. For example, clozapine does not elicit a c-Fos response in the dorsal striatum nor does it cause changes in Wnt-related proteins although both haloperidol and risperidone have significant effects on both pathways (Robertson and Fibiger, 1992; Robertson et al., 1994; Alimohamad et al., 2005a). Therefore, changes in calcineurin and other CaM dependent proteins in the striatum following clozapine treatment are somewhat unusual and may be induced by mechanisms other than D2. The indirect do-

Fig. 3. Brightfield images obtained from sections stained for CaMKIV using immunocytochemistry. Reduced CaMKIV staining is apparent in the medial (MCPu) and lateral caudate-putamen (LCPu) (A, B and C, D respectively) and nucleus accumbens core (NAc) (E, F) following repeated haloperidol (Hal.) treatment (B, D, F) relative to vehicle (Veh.) (A, C, E). Scale bar⫽20 ␮M. Images were obtained from coronal sections approximately 1.20 mm from Bregma.

2004). Collectively, the results suggest antipsychotics may selectively target a specific subset of Ca2⫹ sensitive signaling proteins. The identified changes in CaM and CaM dependent proteins were also specific to antipsychotics and not mimicked in the striatum by drugs used to treat conditions such as bipolar disorder (lithium, valproic acid) or depression (fluoxetine). It should be noted that changes in CaM dependent proteins have been reported in other regions of the brain following treatment with mood stabilizers and anti-depressants. For example, changes in CaMKIV have been described in the rat hippocampus following fluoxetine and lithium treatment (Tiraboschi et al., 2004; Tardito et al., 2007). Understanding antipsychotic specificity is an important issue since drugs used to treat other disorders do not alleviate the symptoms of schizophrenia but there is often overlap in signaling proteins affected by the various drugs. D2 dopamine receptor antagonism may be responsible for the changes observed in CaMKIV, CaMKK␣, CaMKK␤ and calcineurin since the D2/D3 receptor antagonist raclopride had the same effects as antipsychotics. However, the raclopride results do not exclude the possibility that other receptors or mechanisms may cause or contribute to the changes seen following antipsychotic treatment. This may be particularly true for clozapine, a relatively weak D2 dopamine receptor antagonist compared to haloperidol or risperidone. Several previous studies have failed to show a

Fig. 4. Graphs showing changes in the protein levels of CaMKK␣, CaMKK␤ and CaMKIV (top, middle and bottom respectively) in the striatum following repeated administration of haloperidol, risperidone or clozapine for 14 days at different post-injection intervals. Statistically significant changes relative to vehicle are denoted by the asterisks (n⫽8, P⬍0.05).


883

Fig. 5. (A) Representative Western blot showing protein levels of CaMKK␣, CaMKK␤, CaMKIV and calcineurin (CN) following chronic treatment with lithium, valproic acid (oral) or fluoxetine in the striatum relative to vehicle alone. The control samples are shown on the left, the experimental samples on the right. (B) Graph showing the densitometry measurements obtained from the Western blots. With the exception of CaMKK␤ and lithium, lithium, valproic acid and fluoxetine had no effect on CaM-related proteins. Statistically significant changes relative to vehicle are denoted by the asterisk (n⫽8, P⬍0.05).

pamine receptor agonist amphetamine resulted in increased CaMKIV, CaMKK␣ and calcineurin protein levels. A previous study also reported that amphetamine treatment reduces CaMI and CaMIII mRNA levels in the dorsal striatum (Michelhaugh and Gnegy, 2000). Therefore, amphetamine has essentially the opposite effect relative to antipsychotics. A recent gene array study found that antipsychotics have an effect on CaM dependent proteins but reported increased as opposed to decreased mRNA levels that would be predicted based on the protein data presented in the current study (MacDonald et al., 2005). The differences noted may be due to the regions sampled (cortex versus striatum) or measuring mRNA versus protein levels. However, our results suggest that differences in the post-injection survival times (2 versus 24 h) may be critical. Two

hours post-injection, repeated haloperidol and risperidone treatment caused significant decreases in CaMKK␣, CaMKK␤, CaMKIV and calcineurin protein levels but by 24 h post-injection most proteins showed significant increases above baseline. Therefore the changes in protein levels are not static but change dynamically depending on the post-injection survival period prior to sampling of the tissue. The swing in protein levels as the final post-injection interval increases may be related to D2 dopamine receptor occupancy in the striatum but additional experiments will be needed to clarify which phase of the response (if any) may be more relevant to the ability of antipsychotics to alleviate psychosis. It is tempting to speculate that the increases seen 24 h post-injection are the more interesting since they fit with the available findings from schizophrenia.

884


Table 2. Changes in CaMKIV, CaMKK␣, CaMKK␤ and calcineurin protein levels following repeated treatment with raclopride or amphetamine Treatment protein Raclopride CaMKIV CaMKK␣ CaMKK␤ Calcineurin Amphetamine CaMKIV CaMKK␣ CaMKK␤ Calcineurin

Protein levels relative to control

64%⫾4.1%* 66%⫾2.6%* 64.6⫾5.7* 61%⫾8.7%* 168%⫾8.6%* 157%⫾4.7%* 112%⫾6.6% 151%⫾3.6%*

Values listed are % changes relative the control levels (100%) ⫾SEM. * Denotes statistically significant changes (n⫽6 – 8, P⬍0.05).

CaM and D2 dopamine receptors D2 dopamine receptors have classically been described as opposing protein kinase A (PKA) activation and calcium signaling (Sidhu and Niznik, 2000; Neve et al., 2004). The observation that CaM and CaM dependent protein levels may be regulated by antipsychotics and raclopride (D2 dopamine receptor antagonists) or amphetamine (indirect DA receptor agonist), therefore, is consistent with previously reported mechanisms of D2 dopamine receptor signaling. However, given that antipsychotics had no effect on other Ca2⫹ signaling proteins such as calbindin and calretinin suggests a certain degree of specificity. The confounding issue is that despite decreased protein levels of calcineurin and CaMKII␣ following antipsychotic treatment, CaMKII␣ kinase and calcineurin phosphatase activities were reported to be increased following antipsychotic treatment (Meshul and Tan, 1994; Rushlow et al., 2005). Likewise, amphetamine was found to increase calcineurin protein levels but decrease calcineurin activity (Rushlow et al., 2005). These results suggest that downregulation of CaM-associated kinases and calcineurin may be compensatory to adjust for elevated CaM and calcium activity within D2 dopamine receptor containing neurons. Such a hypothesis would be consistent with the predicted consequences of antagonizing D2 dopamine receptors and observations from calcineurin knockout mice (a putative model of schizophrenia) and schizophrenic patients suggesting decreased calcineurin activity (Miyakawa et al., 2003; Eastwood et al., 2005). CaM and schizophrenia There is accumulating evidence to suggest that the CaM second messenger system is disturbed in schizophrenia and targeted by antipsychotics. For example, CaMKK␣ and CaMKK␤ regulate CaMKIV and CaMKIV phosphorylates cAMP response element binding protein (CREB) (Soderling, 1999). Changes in CaMKK␣, CaMKK␤ and CaMKIV are reported in the current study following antipsychotic treatment while changes in CREB or phosphorylated CREB have been reported following antipsychotic treatment and in the brains of schizophrenic patients previously (Kyosseva et al., 2000; Turalba et al., 2004; Yang

et al., 2004). Likewise antipsychotics target CaMKII␣ (decrease) and increases in CaMKII␣ and CaMKII␤ mRNA levels have been reported in bipolar disorder and schizophrenia respectively (Novak et al., 2006). Antipsychotics alter CaM protein levels and changes in CaM have been found in schizophrenic patients although results are somewhat mixed (Vargas and Guidotti, 1980; Schreiber et al., 1981, 1982; Kluge and Kuhne, 1985; Gnegy et al., 1994). Repeated treatment with amphetamine, a psychostimulant used to generate a partial model of schizophrenia, also elicits changes in CaM, calcineurin and CaMKII (RobertsLewis et al., 1986; Shimizu et al., 1997; Rushlow et al., 2005). The inverse effect of antipsychotics and amphetamine on CaMKIV, CaMKK␣ and calcineurin is particularly interesting give that both calcineurin and CaMKIV knockout mice show behavioral changes consistent with putative animal models of schizophrenia and that calcineurin has been identified as a potential schizophrenia susceptibility gene (Gerber et al., 2003; Miyakawa et al., 2003; Shum et al., 2005). Collectively, the data suggest that CaM and CaM dependent proteins are disturbed in schizophrenia and targeted by antipsychotics and may be associated with the manifestation and amelioration of schizophrenic symptoms. Consequences of altering CaM-related proteins While antipsychotics alter proteins associated with the CaM second messenger system, it is unclear what the precise consequences might be. However, there are a number of possibilities. For example, altered CaMKIV activity could lead to changes in gene transcription via regulatory factors such as CREB (Soderling, 1999). Calcineurin also regulates transcriptional factors nuclear factor of activated Tcells (NF-AT) and NF␬␤ (Beals et al., 1997; Lilienbaum and Israel, 2003). Interestingly, NF-AT and NF␬␤ are regulated by CaM and calcineurin in conjunction with GSK-3 and protein kinase B (AkT). Akt and GSK-3 have been shown to be affected by schizophrenia and targeted by antipsychotics (Emamian et al., 2004; Alimohamad et al., 2005a,b). Furthermore, CaMKII and calcineurin are known to play important roles regulating glutamate and dopamine receptors and both glutamate and dopamine have been strongly implicated in schizophrenia and antipsychotic drug action (Kitamura et al., 1993; Tong et al., 1995; Takeuchi et al., 2002; Adlersberg et al., 2004). CaMKK’s and CaMKIV can regulate PKA, Akt, and mitogen activated protein kinase (MAPK), signaling pathways that have shown to be affected by haloperidol and/or clozapine (Dwivedi and Pandey, 1999; Soderling, 1999; Pozzi et al., 2003; Emamian et al., 2004). Calcineurin is also involved in the regulation of dopamine and cAMP responsive protein phosphatase 32 kDa (DARPP-32) and protein phosphatase 1 (PP1) (Nishi et al., 1999). DARPP-32 is regulated by both D1 and D2 dopamine receptors and has been shown to be decreased in schizophrenia and affected by antipsychotic treatment (Nishi et al., 1997; Svenningsson et al., 2000; Albert et al., 2002). Therefore, CaM and CaM-related proteins regulate a variety of signaling mechanisms that have been implicated in the patho-


physiology of schizophrenia and antipsychotic drug action and changes in CaM and CaM related proteins may help to explain the interrelationship between several diverse signaling pathways.

REFERENCES Adlersberg M, Hsiung SC, Glickstein SB, Liu KP, Tamir H, Schmauss C (2004) Regulation of dopamine D-receptor activation in vivo by protein phosphatase 2B (calcineurin). J Neurochem 90:865– 873. Albert KA, Hemmings HC Jr, Adamo AI, Potkin SG, Akbarian S, Sandman CA, Cotman CW, Bunney WE Jr, Greengard P (2002) Evidence for decreased DARPP-32 in the prefrontal cortex of patients with schizophrenia. Arch Gen Psychiatry 59:705–712. Alimohamad H, Rajakumar N, Seah YH, Rushlow W (2005a) Antipsychotics alter the protein expression levels of beta-catenin and GSK-3 in the rat medial prefrontal cortex and striatum. Biol Psychiatry 57:533–542. Alimohamad H, Sutton L, Mouyal J, Rajakumar N, Rushlow WJ (2005b) The effects of antipsychotics on beta-catenin, glycogen synthase kinase-3 and dishevelled in the ventral midbrain of rats. J Neurochem 95:513–525. Beals CR, Sheridan CM, Turck CW, Gardner P, Crabtree GR (1997) Nuclear export of NF-ATc enhanced by glycogen synthase kinase-3. Science 275:1930 –1934. Beaulieu JM, Sotnikova TD, Yao WD, Kockeritz L, Woodgett JR, Gainetdinov RR, Caron MG (2004) Lithium antagonizes dopamine-dependent behaviors mediated by an AKT/glycogen synthase kinase 3 signaling cascade. Proc Natl Acad Sci U S A 101:5099 –5104. Dwivedi Y, Pandey GN (1999) Effects of treatment with haloperidol, chlorpromazine, and clozapine on protein kinase C (PKC) and phosphoinositide-specific phospholipase C (PI-PLC) activity and on mRNA and protein expression of PKC and PLC isozymes in rat brain. J Pharmacol Exp Ther 291:688 –704. Eastwood SL, Burnet PW, Harrison PJ (2005) Decreased hippocampal expression of the susceptibility gene PPP3CC and other calcineurin subunits in schizophrenia. Biol Psychiatry 57:702–710. Emamian ES, Hall D, Birnbaum MJ, Karayiorgou M, Gogos JA (2004) Convergent evidence for impaired AKT1-GSK3beta signaling in schizophrenia. Nat Genet 36:131–137. Eyles DW, McGrath JJ, Reynolds GP (2002) Neuronal calcium-binding proteins and schizophrenia. Schizophr Res 57:27–34. Gerber DJ, Hall D, Miyakawa T, Demars S, Gogos JA, Karayiorgou M, Tonegawa S (2003) Evidence for association of schizophrenia with genetic variation in the 8p21.3 gene, PPP3CC, encoding the calcineurin gamma subunit. Proc Natl Acad Sci U S A 100:8993– 8998. Gnegy ME, Agrawal A, Hewlett K, Yeung E, Yee S (1994) Repeated haloperidol increases both calmodulin and a calmodulin-binding protein in rat striatum. Brain Res Mol Brain Res 27:195–204. Kapur S, Mamo D (2003) Half a century of antipsychotics and still a central role for dopamine D2 receptors. Prog Neuropsychopharmacol Biol Psychiatry 27:1081–1090. Kitamura Y, Miyazaki A, Yamanaka Y, Nomura Y (1993) Stimulatory effects of protein kinase C and calmodulin kinase II on N-methylD-aspartate receptor/channels in the postsynaptic density of rat brain. J Neurochem 61:100 –109. Kluge H, Kuhne GE (1985) Preliminary findings on calmodulin-stimulated Ca2⫹-ATPase of erythrocyte ghosts in psychotic patients. Eur Arch Psychiatry Neurol Sci 235:57–59. Kyosseva SV, Elbein AD, Hutton TL, Griffin ST, Mrak RE, Sturner WQ, Karson CN (2000) Increased levels of transcription factors Elk-1, cyclic adenosine monophosphate response element-binding protein, and activating transcription factor 2 in the cerebellar vermis of schizophrenic patients. Arch Gen Psychiatry 57:685– 691. Lau YS, Gnegy ME (1982) Chronic haloperidol treatment increased calcium-dependent phosphorylation in rat striatum. Life Sci 30:21–28.

885

Li X, Zhu W, Roh MS, Friedman AB, Rosborough K, Jope RS (2004) In vivo regulation of glycogen synthase kinase-3beta (GSK3beta) by serotonergic activity in mouse brain. Neuropsychopharmacology 29:1426 –1431. Lilienbaum A, Israel A (2003) From calcium to NF-kappa B signaling pathways in neurons. Mol Cell Biol 23:2680 –2698. MacDonald ML, Eaton ME, Dudman JT, Konradi C (2005) Antipsychotic drugs elevate mRNA levels of presynaptic proteins in the frontal cortex of the rat. Biol Psychiatry 57:1041–1051. Meshul CK, Tan SE (1994) Haloperidol-induced morphological alterations are associated with changes in calcium/calmodulin kinase II activity and glutamate immunoreactivity. Synapse 18:205–217. Michelhaugh SK, Gnegy ME (2000) Differential regulation of calmodulin content and calmodulin messenger RNA levels by acute and repeated, intermittent amphetamine in dopaminergic terminal and midbrain areas. Neuroscience 98:275–285. Miyakawa T, Leiter LM, Gerber DJ, Gainetdinov RR, Sotnikova TD, Zeng H, Caron MG, Tonegawa S (2003) Conditional calcineurin knockout mice exhibit multiple abnormal behaviors related to schizophrenia. Proc Natl Acad Sci U S A 100:8987– 8992. Neve KA, Seamans JK, Trantham-Davidson H (2004) Dopamine receptor signaling. J Recept Signal Transduct Res 24:165–205. Nishi A, Snyder GL, Greengard P (1997) Bidirectional regulation of DARPP-32 phosphorylation by dopamine. J Neurol Sci 17:8147– 8155. Nishi A, Snyder GL, Nairn AC, Greengard P (1999) Role of calcineurin and protein phosphatase-2A in the regulation of DARPP-32 dephosphorylation in neostriatal neurons. J Neurochem 72:2015– 2021. Novak G, Seeman P, Tallerico T (2006) Increased expression of calcium/calmodulin-dependent protein kinase IIbeta in frontal cortex in schizophrenia and depression. Synapse 59:61– 68. Pozzi L, Hakansson K, Usiello A, Borgkvist A, Lindskog M, Greengard P, Fisone G (2003) Opposite regulation by typical and atypical anti-psychotics of ERK1/2, CREB and Elk-1 phosphorylation in mouse dorsal striatum. J Neurochem 86:451– 459. Reynolds GP (1989) Beyond the dopamine hypothesis. The neurochemical pathology of schizophrenia. Br J Psychiatry 155:305– 316. Roberts-Lewis JM, Welsh MJ, Gnegy ME (1986) Chronic amphetamine treatment increases striatal calmodulin in rats. Brain Res 384:383–386. Robertson GS, Fibiger HC (1992) Neuroleptics increase c-fos expression in the forebrain: contrasting effects of haloperidol and clozapine. Neuroscience 46:315–328. Robertson GS, Matsumura H, Fibiger HC (1994) Induction patterns of Fos-like immunoreactivity in the forebrain as predictors of atypical antipsychotic activity. J Pharmacol Exp Ther 271:1058 –1066. Rushlow W, Naus CC, Flumerfelt BA (1994) Colocalization of prosomatostatin-derived peptides in the caudate-putamen of the rat. J Comp Neurol 349:583–595. Rushlow WJ, Rajakumar B, Flumerfelt BA, Naus CC, Rajakumar N (2000) Changes in CArG-binding protein A expression levels following injection(s) of the D1-dopamine agonist SKF-82958 in the intact and 6-hydroxydopamine-lesioned rat. Neuroscience 98:69 –78. Rushlow WJ, Rajakumar N, Flumerfelt BA, Naus CC (1999) Characterization of CArG-binding protein A initially identified by differential display. Neuroscience 94:637– 649. Rushlow WJ, Seah YH, Belliveau DJ, Rajakumar N (2005) Changes in calcineurin expression induced in the rat brain by the administration of antipsychotics. J Neurochem 94:587–596. Schreiber WE, Gentry R, Fischer EH (1981) Calmodulin and cyclic nucleotide phosphodiesterase activities in erythrocytes from normal and schizophrenic subjects. Biochem Biophys Res Commun 100:1415–1421. Schreiber WE, Gentry R, Fischer EH (1982) Erythrocyte calmodulin in chronic schizophrenia. Am J Psychiatry 139:106 –108.

886


Shimizu Y, Akiyama K, Kodama M, Ishihara T, Hamamura T, Kuroda S (1997) Alterations of calmodulin and its mRNA in rat brain after acute and chronic administration of methamphetamine. Brain Res 765:247–258. Shum FW, Ko SW, Lee YS, Kaang BK, Zhuo M (2005) Genetic alteration of anxiety and stress-like behavior in mice lacking CaMKIV. Mol Pain 1:22. Sidhu A, Niznik HB (2000) Coupling of dopamine receptor subtypes to multiple and diverse G proteins. Int J Dev Neurosci 18:669 – 677. Soderling TR (1999) The Ca-calmodulin-dependent protein kinase cascade. Trends Biochem Sci 24:232–236. Sutton LP, Honardous D, Mouyal J, Rajakumar N, Rushlow WJ (2007) Activation of the canonical Wnt pathway by the antipsychotics haloperidol and clozapine involves dishevelled-3. J Neurochem 102:153–169. Svenningsson P, Lindskog M, Ledent C, Parmentier M, Greengard P, Fredholm BB, Fisone G (2000) Regulation of the phosphorylation of the dopamine- and cAMP-regulated phosphoprotein of 32 kDa in vivo by dopamine D1, dopamine D2, and adenosine A2A receptors. Proc Natl Acad Sci U S A 97:1856 –1860. Takeuchi Y, Miyamoto E, Fukunaga K (2002) Activation of the rat dopamine D2 receptor promoter by mitogen-activated protein kinase and Ca2⫹/calmodulin-dependent protein kinase II pathways. J Neurochem 83:784 –796.

Tardito D, Tiraboschi E, Kasahara J, Racagni G, Popoli M (2007) Reduced CREB phosphorylation after chronic lithium treatment is associated with down-regulation of CaM kinase IV in rat hippocampus. Int J Neuropsychopharmacol 10:491– 496. Tiraboschi E, Tardito D, Kasahara J, Moraschi S, Pruneri P, Gennarelli M, Racagni G, Popoli M (2004) Selective phosphorylation of nuclear CREB by fluoxetine is linked to activation of CaM kinase IV and MAP kinase cascades. Neuropsychopharmacology 29:1831– 1840. Tong G, Shepherd D, Jahr CE (1995) Synaptic desensitization of NMDA receptors by calcineurin. Science 267:1510 –1512. Tooney PA, Chahl LA (2004) Neurons expressing calcium-binding proteins in the prefrontal cortex in schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 28:273–278. Turalba AV, Leite-Morris KA, Kaplan GB (2004) Antipsychotics regulate cyclic AMP-dependent protein kinase and phosphorylated cyclic AMP response element-binding protein in striatal and cortical brain regions in mice. Neurosci Lett 357:53–57. Vargas F, Guidotti A (1980) Calmodulin in brain of schizophrenics. Neurochem Res 5:673– 681. Yang BH, Son H, Kim SH, Nam JH, Choi JH, Lee JS (2004) Phosphorylation of ERK and CREB in cultured hippocampal neurons after haloperidol and risperidone administration. Psychiatry Clin Neurosci 58:262–267.

(Accepted 9 March 2009) (Available online 13 March 2009)