Original Article | Published:

Decreased cortical muscarinic receptors define a subgroup of subjects with schizophrenia

Molecular Psychiatry 14, 1017–1023 (2009) | Download Citation

Subjects

Abstract

Schizophrenia is widely acknowledged as being a syndrome, consisting of an undefined number of diseases probably with differing pathologies. Although studying a syndrome makes the identification of an underlying pathology more difficult; neuroimaging, neuropsychopharmacological and post-mortem brain studies all implicate muscarinic acetylcholine receptors (CHRM) in the pathology of the disorder. We have established that the CHRM1 is selectively decreased in the dorsolateral prefrontal cortex of subjects with schizophrenia. To expand this finding, we wanted to ascertain whether decreased cortical CHRMs might (1) define a subgroup of schizophrenia and/or (2) be related to CHRM1 genotype. We assessed cortical [3H]pirenzepine binding and sequenced the CHRM1 in 80 subjects with schizophrenia and 74 age sex-matched control subjects. Kernel density estimation showed that [3H]pirenzepine binding in BA9 divided the schizophrenia, but not control, cohort into two distinct populations. One of the schizophrenia cohorts, comprising 26% of all subjects with the disorder, had a 74% reduction in mean cortical [3H]pirenzepine binding compared to controls. We suggest that these individuals make up ‘muscarinic receptor-deficit schizophrenia’ (MRDS). The MRDS could not be separated from other subjects with schizophrenia by CHRM1 sequence, gender, age, suicide, duration of illness or any particular drug treatment. Being able to define a subgroup within schizophrenia using a central biological parameter is a pivotal step towards understanding the biochemistry underlying at least one form of the disorder and may represent a biomarker that can be used in neuroimaging.

Introduction

Schizophrenia currently defines a disease syndrome defined by commonality in symptom profile.1 It is probable that schizophrenia encompasses a number of illnesses with different pathologies, all of which share the defining symptoms of the disorder. Therefore, as with other syndromes,2, 3 a major research objective should be to identify individual pathologies of the diseases within the syndrome of schizophrenia.4

There have been many efforts to differentiate different disorders within the syndrome of schizophrenia. Thus, symptom analyses suggest three symptom clusters (positive symptoms (an excess or distortion of normal functions), negative symptoms (the diminution or loss of normal functions) and cognitive symptoms (deficits in attention, concentration and memory)5 within the syndrome but no specific pathological mechanisms have been identified that differentiate these clusters. In addition, cluster analysis of measurable parameters, other than the core symptoms associated with the disorder, suggests there is a cognitively impaired endophenotype within the syndrome of schizophrenia.4 As yet, there is no specific pathology that identifies the subjects which form this particular endophenotype within schizophrenia. Thus, to date, there appears to be no biochemical marker that can be used to separate individuals in the syndrome of schizophrenia into defined subgroups. Significantly, a potential ‘biochemical marker’ for a subgroup within the syndrome of schizophrenia will have to have a strong enough ‘signal’ to be identifiable through the diverse pathological ‘noise’ emanating from the total schizophrenia syndrome.

Neuropsychopharmacological,6, 7 neuroimaging8 and post-mortem central nervous system (CNS) studies9, 10, 11 have consistently implicated a role for muscarinic acetylcholine receptors (CHRM) in the pathology and treatment of schizophrenia.12 Furthermore, the availability of more selective tools has made it possible to show that it is the cortical CHRM1 that is selectively decreased in schizophrenia.13, 14, 15 Moreover, a recent finding that the c.267C/C genotype at the c.267A/C polymorphism in the CHRM1 is associated with more severe cognitive deficits in schizophrenia,16 suggests that variation in CHRM1 gene sequence and/or expression might be associated with a key symptom of the disorder. Together, these data suggest that abnormalities in CHRMs may significantly contribute to the pathology of the disorder. Finally, the ability to measure deficits in CHRM via neuroimaging8 raises the possibility of using this measure as a biomarker for schizophrenia.

The potential for the CHRM1 gene to be associated with the severity of cognitive deficits in schizophrenia and the possibility that levels of CHRM1 could be a biomarker for the disorder led us to determine (1) the extent of decreased CHRM1 levels in post-mortem CNS from a relatively large cohort of subjects with schizophrenia and (2) whether cortical CHRM1 expression is governed by genotype. These issues were addressed by measuring [3H]pirenzepine binding in BA9 and sequencing the CHRM1 gene using post-mortem CNS of 80 subjects with schizophrenia and 74 subjects with no history of psychiatric illness (controls).

Materials and methods

[N-methyl-3H]pirenzepine was obtained from NEN Life Sciences Product Inc., Boston, MA, USA. The chemicals used in genotyping were obtained from Applied Biosytems (Foster City, CA, USA) and Invitrogen Australia Pty, Ltd (Mount Waverley, Victoria, Australia). All other chemicals were obtained from Sigma-Aldrich Inc., Castle Hill, NSW, Australia. [3H]micro-scales were obtained from GE Healthcare, Biosciences, Castle Hill, NSW, Australia. BAS-TR2025 Fuji Imaging Plates were obtained from Berthold Australia Pty Ltd, Bundoora, VIC, Australia.

Accession Numbers: For DNA sequencing of CHRM1, NM_000738.

Tissue collection

Approval for this study was obtained from both the Ethics Committee of the Victorian Institute of Forensic Medicine and the Mental Health Research and Ethics Committee of Melbourne Health.

Blocks of cerebellum (DNA extraction) and Brodmann's area 9 (BA9; autoradiography) were excised from 80 subjects who were listed as having schizophrenia in reports of death. Tissue was also collected from the same regions from 74 subjects with no history of psychiatric illness (controls). In all cases, cadavers were refrigerated within 5 h of being found. Within 30 min of removal at autopsy, CNS tissue was rapidly frozen to −70 °C. In cases where death was witnessed, the time between death and autopsy was taken as the post-mortem interval (PMI). When death was not witnessed, PMI was taken as the interval mid-way between the donor last being seen alive and being found dead. The pH of the CNS tissue was measured as described previously.17

Diagnostic evaluation

Case history reviews were completed using the Diagnostic Instrument for Brain Studies,18 enabling a psychologist and psychiatrist to reach diagnostic consensus using DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, fourth edition) criteria.19 Following the case history review, duration of illness (DOI) was calculated as the time from first hospital admission to death. A comprehensive medication history was obtained for each subject and the most recent prescribed dose of antipsychotic drugs was converted to a standardized drug dose.20

Sequencing

A total of 100 ng of genomic DNA, which had been extracted using an established technique,21 was amplified (primers: M1-F1 5′-ggtgatgactttcccctgag and M1-R1 5′-ctgggaatagcgaagtctgg (400 nM), constructed using the CHRM1 sequence (GenBank accession number NM_000738)) in the presence of two units of Platinum Taq polymerase (Invitrogen), 1.5 mM MgCl2 and 200 μM deoxyribonucleotide triphosphate, with denaturation at 94 °C for 7′ and 30 cycles of 94 °C × 30 min, 57 °C × 30 min, 72 °C × 2 min. The PCR product, consisting of the single coding exon and flanking intronic sequence of CHRM1, was sequenced on an ABI3100 genetic analyser with POP6 matrix using the PCR primers and a third internal primer (M1-F2 5′-ggagacagagaaccgagcac) allowing the genotype at each polymorphic site to be determined.

[3H]Pirenzepine binding

In this study [3H]pirenzepine binding is reported in BA9 from 154 cases. These data are made up of results from 77 cases published in two previous studies (10: n=37; 13: n=40) to which we have added 77 new cases for this study. The use of the same batch of [3H]micro-scales, correcting for radioactive decay, across our studies has ensured uniform quantification of [3H]pirenzepine binding. This reflected in the fact that a two-way ANOVA (analysis of variance), with experiment and diagnoses as the variables, shows a variance in radioligand binding with diagnoses (F=37.5, d.f.=1, 148, P<0.0001) but no variance with experiment (F=0.2, d.f.=2, 148, P=0.84) and no interaction between the variables (F=0.18, d.f.=2, 148, P=0.83).

Levels of [3H]pirenzepine binding were measured using a single-point saturation method as described previously,22 giving a good approximation of the density of radioligand binding sites in tissue sections.23 Thus, five tissue sections (20 μm) were cut from the BA9 of each subject and incubated with [3H]pirenzepine (15 nM) in the presence (non-specific binding: two sections) or absence (total binding: three sections) of 1.0 μM quinuclidinyl xanthene-9-carboxylate hemioxalate in 10 mM sodium-potassium phosphate buffer (10 mM KH2PO4, 10 mM Na2HPO4, pH 7.4) at room temperature for 30 min. All sections were washed twice for 2 min in ice-cold 10 mM sodium-potassium phosphate buffer, dipped in ice-cold water and thoroughly dried prior to being fixed overnight in paraformaldehyde fumes in a desiccator. The sections, and a set of [3H]micro-scales, were apposed to a BAS-TR2025 imaging plate until an image of appropriate intensity was obtained for scanning in the BAS 5000 high-resolution phosphoimager (3 days). The intensity of the phosphoimages were measured by comparison to the intensity of the segments of radioactivity on the [3H]micro-scales using AIS image analysis software. Results were expressed as total minus non-specific binding in d.p.m. mg−1 estimated wet weight tissue equivalents (ETE) and then converted to fmol mg−1 ETE.

Statistical analyses

Statistical analyses on genotype were performed using Arlequin ver 2.000: A software for population genetics and data analysis (S Schneider, D Roessli, L Excoffier, University of Geneva, Switzerland, Genetics and Biometry Laboratory, 2000). The genotype frequency data at each single nucleotide polymorphism (SNP) was analysed using a χ2 bivariate analysis.

For radioligand binding data, the data sets were first analysed using kernel density estimation, which is a means of applying smoothing to a population frequency histogram, to estimate a probability distribution of the data sets. These were then fitted using different distribution parameters (GenStat release 9/Windows; VSN International Ltd, UK) to determine the nature of the populations they represented. Next, the data were analysed to determine whether they followed a Gaussian distribution. In addition, Grubb's test was used to determine whether any of the data sets contained outlying points.

Age and brain pH for the three diagnostic cohorts were not normally distributed and so were compared using the Kruskal–Wallis test, while PMI and brain weights were normally distributed and were compared using one-way ANOVAs. DOI and chlorpromazine equivalents for the two groups within schizophrenia were not normally distributed and were compared using the Mann–Whitney test. For [3H]pirenzepine binding density, the impact of diagnosis, genotype and the demographic variables such as donor age, PMI, brain pH, DOI, final recorded antipsychotic drug dose, gender, suicide and prescription of anticholinergic medication or benzodiazepines on the experimental measures were determined using an ANCOVA (analysis of covariance; Mintab release 13.31, Minitab Inc. State College, PA, USA). Fishers' exact test was used to determine differences in the frequency of confounding variables between the two schizophrenia cohorts. Unless otherwise specified, all analyses were carried out using GraphPad Prism version 4.03 for Windows (GraphPad Software Inc., San Diego, CA, USA; for more information see www.graphpad.com).

Results

Demographics

Non-parametric analyses of age, brain pH, DOI and chlorpromazine equivalents and parametric analyses of PMI and brain weight found no significant differences in any of these variables (Table 1).

Table 1: A summary of the demographic data (mean±s.e.m.) for the cohort used in sequencing of the CHRM1 gene and measuring [3H]pirenzepine binding
Full size table

[3H]Pirenzepine binding

The distribution of [3H]pirenzepine binding was homogeneous across the cortical layers of BA9 (Figure 1a), therefore, an integrated measure was taken across cortical laminae.

Figure 1
Figure 1

(a) A representative autoradiograph of [3H]pirenzepine binding in Brodmann's area 9 (BA9), the insert shows non-specific binding. Distribution of the radioligand binding is shown graphically: (b) Data from control subjects. (c) Data from subjects with schizophrenia.

Full size image

Kernel density estimation showed that the probability distribution for [3H]pirenzepine binding in control subjects was approximately symmetrical (Figure 1b) and fitted a normal distribution (Kolmogorov–Smirnov, KS distance=0.08866, P>0.1; D'Agostino–Pearson omnibus normality test, K2=3.768, P>0.1; Shapiro–Wilk normality test, W=0.9774, P>0.1). By contrast, [3H]pirenzepine binding in subjects with schizophrenia had an asymmetrical distribution (Figure 1c), which did not follow a normal distribution (Kolmogorov–Smirnov, KS distance=0.1146, P<0.05; D'Agostino–Pearson omnibus normality test, K2=5.103, P=0.07; Shapiro–Wilk normality test, W=0.9407, P<0.005). The best fit for data from the subjects with schizophrenia was into two groups (P=0.580; representing a good fit). The population represented by the larger of the two groups could be fitted by a normal distribution (Kolmogorov–Smirnov, KS distance=0.07867, P>0.1; D'Agostino–Pearson omnibus normality test, K2=3.784, P>0.1; Shapiro–Wilk normality test, W=0.9619, P>0.05). However, the smaller group could not be adequately fitted to a normal distribution (Kolmogorov–Smirnov, KS distance=0.2045, P<0.05; D'Agostino–Pearson omnibus normality test, K2=4.025, P>0.1; Shapiro–Wilk normality test, W=0.8927, P<0.05).

Following the separation of the schizophrenia cohort into two groups, each group of subjects with schizophrenia was treated as a separate cohort. It was shown that significant variance occurred in [3H]pirenzepine binding with diagnosis (F=127.4, d.f.=2,151, P<0.0001). Post hoc Bonferroni's multiple comparison test revealed that this variance was due to marked decreases in [3H]pirenzepine binding in the smaller cohort of subjects with schizophrenia (mean±s.e.m.: 44.3±6.9 fmol mg−1 ETE) compared to both control (mean±s.e.m.: 182.7±4.5 fmol mg−1 ETE; t=15.72, P<0.001) and schizophrenia (mean±s.e.m.:167.8±4.5 fmol mg−1 ETE; t=13.60, P<0.001) cohorts. There was no difference in [3H]pirenzepine binding between controls and the larger cohort of subjects with schizophrenia (t=2.343, P>0.05; Figure 2).

Figure 2
Figure 2

Results for [3H]pirenzepine binding are shown graphically; data from control subjects shown with open circles, that from subjects with schizophrenia with closed circles and that from subjects with muscarinic receptor-deficit schizophrenia (MRDS) by gray stars. The mean of each group is represented by a solid black line, ***P<0.0001.

Full size image

CHRM1 sequencing

Five SNPs were identified (c.267C>A, c.783C>T, c.1044G>A, c.1221C>T and c.1353C>T) within the amplified region of CHRM1. These could be resolved into seven distinct combinations (Supplementary Table 1). Linkage disequilibrium was demonstrated between c.267C>A and c.1353C>T and between c.267C>A, c.783C>T and c.1353C>T. One genotype (7: c.267C>A+c.783C>T+c.1353C>T) was only detected in subjects with schizophrenia (6.1%) and therefore had a significantly higher frequency in subjects with the disorder (χ2=5.0, d.f.=1, P=0.025). Another genotype (5: c.267C>A and c.1353C>T) occurred at a lower frequency in schizophrenia (χ2=9.50, d.f.=1, P=0.002). Importantly, no individual genotype was associated with decreased levels of [3H]pirenzepine binding in schizophrenia as a whole (F=1.28, d.f.=1,109, P=0.260) and therefore had no strong association with either cohort of subjects with schizophrenia.

Potential confounding variables

The ANCOVA showed that none of the demographic variables; age (F=2.70, d.f.=1, 109, P=0.103), PMI (F=1.73, d.f.=1, 109, P=0.192), brain pH (F=1.23, d.f.=1, 109, P=0.270), gender (F=0.01, d.f.=1, 109, P=0.943), hemisphere (F=2.85, d.f.=1, 109, P=0.094) and brain weight (F=2.04, d.f.=1, 109, P=0.156) had any influence on the levels of [3H]pirenzepine binding. Furthermore, within the two cohorts of subjects with schizophrenia, there was no effect of either DOI (F=0.00, d.f.=1, 109, P=0.970) or final recorded dose of antipsychotic medication, expressed as chlorpromazine equivalents (F=0.78, d.f.=1, 109, P=0.380).

Further analysis was completed on potential confounding factors that may have contributed to the bimodal distribution of [3H]pirenzepine binding in subjects with schizophrenia. Given the nature of our study, the prescription of anticholinergic medication was a potential confound. Fishers' exact test showed there was no significant variation in the number of subjects who were prescribed anticholinergic drugs in either the large (16 of 58) or smaller (12 of 22; P=0.16) cohort of subjects with schizophrenia. There was no effect of receiving such medication on [3H]pirenzepine binding (F=1.36, d.f.=1,109, P=0.247). Similarly, there was no significant difference in the number of subjects who had received benzodiazepines in the two groups of subjects with schizophrenia (large; 16 of 58, small; 10 of 22: P=0.33). Nor did being prescribed benzodiazepines significantly affect levels of [3H]pirenzepine binding (F=0.01, d.f.=1,109, P=0.929). Finally, there was no significant difference in rates of suicide between the two populations of subjects with schizophrenia (large=32/58, small=8/22; P=0.50). ANCOVA showed that suicide did not impact on the levels of radioligand binding (F=0.18, d.f.=1,109, P=0.668).

In the entire schizophrenia cohort, eight people were not prescribed antipsychotic medication at the time of death. While these subjects were all in the larger of the two cohorts, this did not represent a significant difference in the frequency of prescribing antipsychotic drugs between the two cohorts (P=0.19). Furthermore, there were no differences in the rates of the prescription of butyrophenones (large=19, small=2; P=0.15), phenolic phenothiazines (large=26, small=13; P=0.53), aliphatic phenothiazines (large=11, small=4; P=1.00), diphenylbutylpiperidine (large=2, small=1; P=0.19) or atypical antipsychotics (large=3, small=2; P=0.61) (Supplementary Figure 1).

Discussion

In this study, we have shown that CHRM levels, measured using a single-point saturation analysis of [3H]pirenzepine binding, in BA9 from subjects with schizophrenia separates subjects with the disorder into two subgroups, one of which has a marked decrease in cortical CHRMs. Technically, the single-point saturation method can give an apparent decrease in levels of available binding sites, if there is a marked change in the affinity of the radioligand used for the receptor of interest.24 However, in another cortical region we have shown there is no marked change in the affinity for [3H]pirenzepine in subjects with schizophrenia.25 This supports the argument that the changes in radioligand binding in this study are due to a reduced receptor density. Furthermore, our data show that low levels of radioligand binding are not associated with any particular variation in the CHRM1 sequence or any other confounding factor examined in this study. Thus, it would appear that low levels of CHRM differentiate a group of subjects with schizophrenia, which we term muscarinic receptor-deficit schizophrenia (MRDS). Importantly, at least three studies (two post-mortem,11, 26 one neuroimaging8) support the hypothesis that there is an MRDS. These studies show their data in a manner that allowed us to identify a proportion of subjects with schizophrenia who have marked decreases in levels of CHRMs. Thus, it is probable that MRDS has contributed to previous reports of lower levels of CHRMs in schizophrenia but cohort sizes prevented the application of population statistics. Thus, our study is significant as it is the first with cohort sizes of sufficient magnitude to identify MRDS with statistical certainty.

We defined the MRDS using [3H]pirenzepine binding, which predominantly binds to CHRM1/CHRM4.27 In a recent study we suggested that changes in CHRM4 expression may be the dominant change in hippocampal CHRMs in subjects with schizophrenia.28 However, as (1) the CHRM1 constitutes approximately 70% of the total CHRMs in the frontal cortex29 and (2) we have shown that only CHRM1 mRNA and protein are decreased in BA9 from subjects with schizophrenia,13, 14 our current data strongly suggest that we have identified a subset of people with greatly reduced cortical CHRM1. It remains to be determined if this group also has CHRM1 deficits in multiple regions, deficits in other CHRMS and whether other CHRMs are affected in the non-MRDS subjects.

Significantly, in this study we have shown that particular variants of CHRM1 sequence occurred at different frequencies in schizophrenia. However, none of the variants are associated with either levels of [3H]pirenzepine binding or the MRDS. Importantly, while we detected only 5 of the 11 CHRM1 SNPs previously reported,30 the SNPs we failed to detect occur mainly in African Americans and Hispanic populations,30 which are not represented in our cohorts. Moreover, we have confirmed a previous finding16 of complete linkage between the c.267C>A and c.1353C>T SNPs. Although our data identify possible changes in SNP frequencies in schizophrenia, there is the potential for a type I error in our study due to the relatively small sample size. Therefore independent confirmation of our findings on c.267C>A and c.1353C.T are required in much larger populations. However, overall, our data support a previous study,15 which suggested that CHRM1 sequence does not govern levels of CHRM1 expression in the human CNS or schizophrenia.

While our data suggest that potential confounding factors such as age, sex, suicide or DOI do not differ in MRDS when compared to other subjects with schizophrenia, we cannot directly address whether smoking or drug treatments may contribute to the MRDS. Although increased levels of smoking are reported in subjects with schizophrenia,31 we could find no clear evidence that any sub-population of subjects with schizophrenia showed increased levels of smoking. This is perhaps not surprising given that a neuroimaging study of CHRMs reported no significant difference in CHRM availability between smokers and non-smokers with schizophrenia.8 Indeed, the only evidence implicating smoking as a modulator of human CHRM expression is that CHRM1 mRNA is decreased in foetal brainstem and cerebellum32 when mothers smoked during pregnancy. Thus, it seems unlikely that smoking is the cause of decreased CHRMs in the cortex of subjects in the MRDS. The other potential confounding factor in any post-mortem study in schizophrenia is the possible effects of the polypharmacies received by the tissue donors during life, particularly those of antipsychotic drugs. In our study there was no clear difference in rate of prescription of any drug or in the class or levels of antipsychotic drug received by MRDS and non-MRDS subjects. Moreover, animal studies have shown that treatment with antipsychotic drugs10, 33, 34 or muscarinic receptor antagonists35 do not cause marked decreases in cortical CHRMs. Together, these data suggest that decreases in cortical CHRMs in MRDS are not simply due to a subject receiving antipsychotic drugs prior to death.

When considering the potential consequences of decreased CHRM1 in schizophrenia, it is noteworthy that both the CHRM136 and BA937 play important roles in cognition, with CHRM1 being particularly important in working memory.36 These findings underpin the suggestion that decreased cortical CHRM1 is associated with cognitive impairments in schizophrenia and that treatment with CHRM1 agonists could reverse such deficits.38, 39 This hypothesis has more direct support from a study showing that xanomeline (an M1/M4 agonist) improves cognitive function in patients with schizophrenia40 and that the best predictor of cognitive improvement with clozapine is the ratio of plasma N-desmethylclozapine (clozapine's major metabolite; an M1 agonist) to clozapine.7

It should also be noted that the neuroimaging study of CHRMs in schizophrenia found a correlation between the positive symptom sub-scale score on positive and negative syndrome scale and muscarinic receptor occupancy. However, the ligand used binds with high affinity to all muscarinic receptors, therefore these data partially support the argument that agonists at the CHRM4 could have antipsychotic efficacy in subjects with schizophrenia,41 it having been shown that CHRM4 regulate dopamine levels in the nucleus accumbens.42 Importantly, with regards to the therapeutic potential of CHRM agonists, our latest data imply that the MRDS could show the most marked response to CHRM1 agonists.

In conclusion, we have identified a subgroup of subjects with schizophrenia, the MRDS, which we propose have marked deficits in cortical CHRM1. There is now a clear need to fully differentiate the underlying pathology and symptom profile of subjects with MRDS. While it is possible to better define the pathology of MRDS with further post-mortem tissue studies on other CNS markers; we are unable, at this point, to identify a specific symptom profile for this form of the disorder. This is because our tissue collection and case history review occurs ad hoc after death, thus we do not get symptom ratings from case histories. However, the magnitude of reduced cortical CHRM1 in the MRDS explains why subjects with the deficit can be identified in the Single Photon Emission Computed Tomography (SPECT) study that has used a pan-muscarinic receptor ligand.8 This pilot neuroimaging study strongly suggests that the use of Positron Emission Tomography (PET) with a non-selective cholinergic marker, such as [11C]NMPB,43 would identify subjects with MRDS. Using such an approach, it will be possible to obtain accurate symptom ratings and discover if MRDS responds well to treatment with currently available antipsychotic drugs. Nevertheless, based on post-mortem, genetic and pre-clinical data we would predict that subjects with MRDS will present with significant cognitive deficits, which may be responsive to CHRM1 agonists.

Accessions

GenBank/EMBL/DDBJ

References

  1. 1.

    . Subtyping schizophrenia: implications for genetic research. Mol Psychiatry 2006; 11: 815–836.

  2. 2.

    , . Histopathologic diagnosis of celiac disease in children without clinical evidence of malabsorption. Int J Surg Pathol 2007; 15: 354–357.

  3. 3.

    , , , , . Collagen genes and proteins in osteogenesis imperfecta. J Med Genet 1985; 22: 466–478.

  4. 4.

    , , , , , et al. Genetic evidence for a distinct subtype of schizophrenia characterized by pervasive cognitive deficit. Am J Hum Genet 2005; 77: 468–476.

  5. 5.

    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn (text revision). American Psychiatric Association: Washington, DC, 2000.

  6. 6.

    , , . Xanomeline and the antipsychotic potential of muscarinic receptor subtype selective agonists. CNS Drug Rev 2003; 9: 159–186.

  7. 7.

    , , , , , et al. The role of M1 muscarinic receptor agonism of N-desmethylclozapine in the unique clinical effects of clozapine. Psychopharmacology (Berl) 2004; 177: 207–216.

  8. 8.

    , , , , , et al. In vivo determination of muscarinic acetylcholine receptor availability in schizophrenia. Am J Psychiatry 2003; 160: 118–127.

  9. 9.

    , , , . Decreased muscarinic receptor binding in subjects with schizophrenia: a study of the human hippocampal formation. Biol Psychiatry 2000; 48: 381–388.

  10. 10.

    , , , . Low muscarinic receptor binding in prefrontal cortex from subjects with schizophrenia: a study of Brodmann's areas 8, 9, 10, and 46 and the effects of neuroleptic drug treatment. Am J Psychiatry 2001; 158: 918–925.

  11. 11.

    , , , . Investigation of m1/m4 muscarinic receptors in the anterior cingulate cortex in schizophrenia, bipolar disorder, and major depression disorder. Neuropsychopharmacology 2004; 29: 619–625.

  12. 12.

    , , , , . Towards a muscarinic hypothesis of schizophrenia. Mol Psychiatry 2006; 12: 232–246.

  13. 13.

    , , , , . Decreased muscarinic(1) receptors in the dorsolateral prefrontal cortex of subjects with schizophrenia. Mol Psychiatry 2002; 7: 1083–1091.

  14. 14.

    , , , . No change in cortical muscarinic M2, M3 receptors or [(35)S]GTPgammaS binding in schizophrenia. Life Sci 2006; 78: 1231–1237.

  15. 15.

    , , , . Reduced expression of the muscarinic 1 receptor cortical subtype in schizophrenia. Am J Med Genet B Neuropsychiatr Genet 2003; 119: 2–6.

  16. 16.

    , , , , , et al. Association of muscarinic m1 receptor genetic polymorphisms with psychiatric symptoms and cognitive function in schizophrenic patients. Neuropsychobiology 2003; 48: 72–76.

  17. 17.

    , , , , , et al. Tissue pH as an indicator of mRNA preservation in human post-mortem brain. Brain Res Mol Brain Res 1995; 28: 311–318.

  18. 18.

    , , , , , . Diagnostic Instrument for Brain Studies. Mental Health Research Institute: Melbourne, 1999.

  19. 19.

    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn American Psychiatric Association: Washington, DC, 1994.

  20. 20.

    . Antipsychotics (Neuroleptics). In Bezchlibnyk-Butler KZ and Jeffries JJ (eds). Clincal Handbook of Psychotropic Drugs. Hogrefe & Huber: Seattle, Toronto, Göttingen, Bern, 1999, pp 55–84.

  21. 21.

    . Preparation of Genomic DNA from mammalian tissue. In: Crawley JN, Gerfen CR, Rogawski MA, Sibley DR, Skolnick P, and Way S (eds). Current Protocols in Neuroscience. John Wiley & Sons, Inc.: Hoboken, 1999, pp A.1H.1–A.1H.3.

  22. 22.

    , , , , , . The density of muscarinic M1 receptors is decreased in the caudate-putamen of subjects with schizophrenia. Mol Psychiatry 1996; 1: 54–58.

  23. 23.

    . Mathematics and statistics of ligand assays. In: Langan J and Clapp JJ (eds). Ligand Assay: Analysis of International Developments on Isotopic and Nonisotopic Immunoassay. Masson: New York, 1981, pp 55–101.

  24. 24.

    , , . [3H]raclopride binding to brain tissue from subjects with schizophrenia: methodological aspects. Neuropharmacology 1997; 36: 779–786.

  25. 25.

    , , , . Changed cholinergic markers in Brodmann's area 6 from subjects with schizophrenia. Neuroscience Meeting Planner San Diego, CA: Society for Neuroscience, 2007 CD-ROM. 118.8; abstract.

  26. 26.

    , , , , , . Neurotransmitter receptors in frontal cortex of schizophrenics. Arch Gen Psychiatry 1979; 36: 927–934.

  27. 27.

    , , , , , . Affinity profiles of various muscarinic antagonists for cloned human muscarinic acetylcholine receptor (mAChR) subtypes and mAChRs in rat heart and submandibular gland. Life Sci 1999; 64: 2351–2358.

  28. 28.

    , , , . Altered hippocampal muscarinic M4, but not M1, receptor expression from subjects with schizophrenia. Biol Psychiatry 2007; 61: 1161–1170.

  29. 29.

    , , , . Differential regulation of molecular subtypes of muscarinic receptors in Alzheimer's disease. J Neurochem 1995; 64: 1888–1891.

  30. 30.

    , , . Single nucleotide polymorphisms of the human M1 muscarinic acetylcholine receptor gene. AAPS PharmSci 2001; 3: E31.

  31. 31.

    , . How to help psychiatric patients stop smoking. Psychiatr Serv 1995; 46: 435–436.

  32. 32.

    , , , , . Smoking during early pregnancy affects the expression pattern of both nicotinic and muscarinic acetylcholine receptors in human first trimester brainstem and cerebellum. Neuroscience 2005; 132: 389–397.

  33. 33.

    , , , . Effects of typical and atypical antipsychotic drugs on rat brain muscarinic receptors. Neurochem Res 2007; 32: 525–532.

  34. 34.

    , , , , . Effects of antipsychotic medication on muscarinic M1 receptor mRNA expression in the rat brain. J Neurosci Res 2007; 86: 457–464, e-pub ahead of print, doi: 10.1002/jnr.21491.

  35. 35.

    , . Regulation of muscarinic receptor subtypes and their responsiveness in rat brain following chronic atropine administration. Mol Pharmacol 1989; 36: 749–757.

  36. 36.

    , , , , , et al. Selective cognitive dysfunction in acetylcholine M1 muscarinic receptor mutant mice. Nat Neurosci 2003; 6: 51–58.

  37. 37.

    , , , . Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science 2000; 288: 1835–1838.

  38. 38.

    , . Antipsychotic drugs: evolving mechanisms of action with improved therapeutic benefits. Curr Drug Targets CNS Neurol Disord 2004; 3: 217–225.

  39. 39.

    , . Molecular targets for treating cognitive dysfunction in schizophrenia. Schizophr Bull 2007; 33: 1100–1119.

  40. 40.

    , , , , , et al. Antipsychotic activity of muscarinic agonists in animals and humans. Int J Neuropsychopharmacol 2004; 7(suppl 1): S16–S17. abstract.

  41. 41.

    , , , . Muscarinic receptors as a target for drugs treating schizophrenia. Curr Drug Targets CNS and Neurol Disord 2002; 1: 163–181.

  42. 42.

    , , , , , et al. M4 muscarinic receptors regulate the dynamics of cholinergic and dopaminergic neurotransmission: relevance to the pathophysiology and treatment of related CNS pathologies. FASEB J 2004; 18: 1410–1412.

  43. 43.

    , , , , , et al. Assessment of muscarinic receptor concentrations in aging and Alzheimer disease with [11C]NMPB and PET. Synapse 2001; 39: 275–287.

Download references

Acknowledgements

We thank Mrs Jacyln Bartlett, Ms Suzette Sheppard and Mr Geoff Pavey for technical assistance; Dr Kenneth Opeskin, Ms Christine Hill, Professor Nick Keks and Professor David Copolov contributed towards collecting the tissue and clinical information relating to this study.

This study was supported in part by grants-in-aid from the National Health and Medical Research Council (Project Grant no. 350344), The Rebecca L Cooper Medical Research Foundation and the Wood's Family Research Program. ES was the Ronald Philip Griffiths Research Fellow and the recipient of a NARSAD 2005 Young Investigator award, She is now the Royce Abbey Post-Doctoral Fellow. BD is an NH&MRC Senior Research Fellow (no.; 400016).

Author information

Affiliations

  1. Centre for Neuroscience, The University of Melbourne, Parkville, VIC, Australia

    • E Scarr

  2. Rebecca L Cooper Research Laboratories, Mental Health Research Institute of Victoria, Parkville, VIC, Australia

    • E Scarr
    •  & B Dean

  3. Department of Pathology, The University of Melbourne, Parkville, VIC, Australia

    • T F Cowie
    • , S Kanellakis
    •  & B Dean

  4. NeuroProteomics and NeuroGenomics Platform, The National Neuroscience Facility, Parkville, VIC, Australia

    • T F Cowie
    • , S Kanellakis
    •  & B Dean

  5. Department of Psychiatry, The University of Melbourne, Parkville, VIC, Australia

    • S Sundram
    • , C Pantelis
    •  & B Dean

  6. Molecular Psychopharmacology, Mental Health Research Institute, Parkville, VIC, Australia

    • S Sundram

  7. Northern Psychiatry Research Centre, The Northern Hospital, Epping, VIC, Australia

    • S Sundram

  8. Melbourne Neuropsychiatry Centre, The University of Melbourne, Parkville, VIC, Australia

    • C Pantelis

  9. Department of Psychological Medicine, Monash University, Clayton, VIC, Australia

    • B Dean

Authors

  1. Search for E Scarr in:

  2. Search for T F Cowie in:

  3. Search for S Kanellakis in:

  4. Search for S Sundram in:

  5. Search for C Pantelis in:

  6. Search for B Dean in:

Corresponding author

Correspondence to E Scarr.

Supplementary information

Image files

  1. 1.

    Supplementary Figure

Word documents

  1. 1.

    Supplementary Legend

  2. 2.

    Supplementary Table

To obtain permission to re-use content from this article visit RightsLink.

About this article

Publication history

Received

Revised

Accepted

Published

DOI

https://doi.org/10.1038/mp.2008.28

Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)