American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 144B:496 –501 (2007) Mitochondrial DNA Haplogroups and Age at Onset of Schizophrenia Chiara Magri,1 Rita Gardella,1 Stefano Davide Barlati,2 Paolo Valsecchi,2,3 Emilio Sacchetti,2,3,4,5 and Sergio Barlati1,5* 1 Division of Biology and Genetics, Department of Biomedical Sciences and Biotechnology, Brescia University School of Medicine, Brescia, Italy 2 Department of Mental Health, Brescia Spedali Civili, Brescia, Italy 3 University Psychiatric Unit, Brescia University School of Medicine, Brescia, Italy 4 Chair of Psychiatry, Brescia University School of Medicine, Brescia, Italy 5 Centre on Behavioural and Neurodegenerative Disorders, Brescia University and EULO, Brescia, Italy A number of studies support a possible link between mitochondrial dysfunction and schizophrenia. To test the hypothesis of a direct contribution of mitochondrial DNA (mt-DNA) in susceptibility to DSM-IV-TR-schizophrenia, we looked for differences in the frequency distribution of the major European haplogroups (hgs) in 142 patients and 190 controls both of Italian origin. A subgroup of patients (N ¼ 37) and healthy counterparts (N ¼ 41) was also analyzed for possible differences in the relative amount of mt-DNA versus nuclear-DNA in blood cells. Patients and controls were comparable for hg frequency distribution and the relative levels of mt-DNA even after stratification by gender and schizophrenia subtype. However, patients harboring the hg J-T showed an anticipated onset of the disorder. These results indicate that the J-T hg of mt-DNA may have a modulator effect on deeper determinants of schizophrenia. ß 2007 Wiley-Liss, Inc. KEY WORDS: mitochondrial DNA; schizophrenia; mt-haplogroups; mt-DNA levels; age at onset Please cite this article as follows: Magri C, Gardella R, Barlati SD, Valsecchi P, Sacchetti E, Barlati S. 2007. Mitochondrial DNA Haplogroups and Age at Onset of Schizophrenia. Am J Med Genet Part B 144B:496–501. INTRODUCTION Several recent studies support a possible link between mitochondria dysfunction and schizophrenia. Schizophrenialike symptoms have been described in patients with MELAS, a mitochondrial encephalopatia [Suzuki et al., 1990; Prayson and Wang, 1998; Thomeer et al., 1998]. Furthermore, a decrease in ATP levels in the basal ganglia and temporal lobe [Kegeles et al., 1998], a subnormal number of mitochondria in caudate and putamen axonal terminals [Kung and Roberts, 1999], and abnormal activity and expression of respiratory chain components in platelets and brains post-mortem were observed in patients with schizophrenia [Prince et al., 1999; Maurer et al., 2001; Dror et al., 2002]. Schizophrenia patients were also found to have altered expression of genes related to mitochondrial function in different areas of the brain [Middleton et al., 2002; Prabakaran et al., 2004; Altar et al., 2005; Iwamoto et al., 2005]. Finally, some new mitochondrial DNA (mt-DNA) variants were occasionally reported in patients with maternal inheritance of schizophrenia [Martorell et al., 2006]. To date, the question of whether mitochondria dysfunction is a primary cause of schizophrenia, a cellular response to alterations in other metabolic pathways, or an effect mediated by pharmacological treatment has remained unanswered. In order to test the hypothesis of a direct contribution of mt-DNA to susceptibility for schizophrenia, we planned a case–control study, which compared DSM-IV-TR schizophrenia patients and healthy controls of Italian origin for possible differences in haplogroups (hgs) distribution. Indeed, the analysis of haplogroup distribution may be a valid approach to reveal the presence of mt-DNA polymorphisms, associated with specific haplogroups, which could be involved in susceptibility to a complex disease like schizophrenia. Moreover, in order to verify a possible relationship between the quantity of mt-DNA and schizophrenia, the relative amount of mt-DNA versus nuclear DNA was assessed. An altered level of mt-DNA could reflect either a depletion of mt-DNA copy number or a reduction of the number of mitochondria; both alterations could generate a decline of mitochondrial respiratory functions, which could contribute to a complex disease like schizophrenia. MATERIALS AND METHODS Samples Grant sponsor: Health Authority of the Lombardia Region; Grant number: Project 153; Grant sponsor: Centre on Behavioural and Neurodegenerative Disorders; Grant sponsor: IDET Centre of Excellence (MIUR). *Correspondence to: Prof. Sergio Barlati, Division of Biology and Genetics, Department of Biomedical Sciences and Biotechnology, Brescia University School of Medicine, Viale Europa 11, 25123 Brescia, Italy. E-mail: email@example.com Received 23 August 2006; Accepted 19 December 2006 DOI 10.1002/ajmg.b.30496 ß 2007 Wiley-Liss, Inc. The study included 142 patients and 190 healthy controls of both sexes, provided that they gave written informed consent, were unrelated to each others, fulfilled predefined groupspecific inclusion and exclusion criteria, and referred, by oral interview, to have Italian ancestry of at least two generations and to live in northern Italy. The written consent form supplied a concise but unequivocal explanation about the aims of the study. The invited participants also received an explicit guarantee of anonymity as a unique number linked all the individual data. mt-DNA and Schizophrenia For the patients, the inclusion criteria were a DSM-IV-TR diagnosis of schizophrenia [American Psychiatric Association, 2000] and a level of understanding and attention judged to be sufficient to give true informed consent; lifetime comorbidity with other DSM-IV-TR axis I disorders, nicotine and caffeine abuse apart, was an exclusion criterion. Patients were also evaluated for schizophrenia subtype, schizophrenia severity, and age at onset. Schizophrenia severity was evaluated by the Positive and Negative Syndrome Scale (PANSS) [Kay et al., 1987]. Concerning the age at onset, the appearance of the first psychotic symptom represented the preidentified operational cut-off. In order to reach as accurate estimate as possible, direct information from the patients was systematically retrieved along with data obtained from at least one close relative, plus, when available, previous medical records. For the controls, the prerequisites for enrollment were the absence during their lifespan of any DSM-IV-TR axis I disorder, once again nicotine and caffeine abuse apart, and a negative family history for psychoses and mood disorders. The procedures for the enrollment of participants and the collection of any supplementary information have been described in detail in another study [Magri et al., 2006]. Briefly, detailed clinical interviews were implemented, when required, by ad hoc questionnaires and DSM-IV-TR adjusted versions of the Structural Clinical Interview for DSM-IV Axis I Disorders, Clinician Version [First et al., 1996] for patients and the Diagnostic Interview for Genetic Studies [Nurnberger et al., 1994] for controls. Two qualified psychiatrists were in charge of the entire data collection, after they were trained on the procedures included in the protocol and had shown valuable reliability. In any case, discordance between the evaluators precluded the recruitment or the acquisition of specific information. The patients were enrolled from those voluntarily admitted to the Brescia University and Spedali Civili Psychiatric Unit. The controls were volunteers enrolled among consenting doctors, nurses, employers, attendants of Brescia Spedali Civili, and students of Brescia University or their relatives. Patients and controls did not differ in sex distribution (86/ 56 vs. 111/79 male/female ratio; P ¼ 0.69) and age (39 12 vs. 37 14 years; P ¼ 0.14). A paranoid subtype of schizophrenia was diagnosed in 64% of the patients. For 135 patients information about age at onset was also available. These subjects were aged 26.1 7.6 years at the onset of their disorders, with an earlier onset in males than females (24.3 6.7 vs. 28.8 8.2; P ¼ 0.001). DNA Extraction and Quantification Total DNA was purified from 2 ml of fresh whole blood using the Puregene Kit (Gentra Systems, Inc, Minneapolis, MN) according to the manufacturer’s instructions. Exact quantification of the DNA content was carried out with the NanoDrop spectrophotometer (Celbio NanoDrop Technologies, Wilmington, DE). Haplogroup Analysis To classify the most common European mt-hgs H, U, K, J, T, W, and I, a total of eight single nucleotide polymorphisms (SNPs) were studied by restriction fragment length polymorphism (RFLP) analysis in a hierarchical way (Fig. 1). The SNPs and the restriction enzymes used for their characterization are those reported by De Benedictis et al. . The rare hgs were not directly analyzed and were grouped in the Other (10394) class, or in the Other (þ10394) class according to the absence or the presence of the 10394 DdeI restriction site. Primers and PCR conditions are available upon request. Patient–control differences in the distribution of hgs were evaluated by the w2 test, combining the less frequent hgs W and 497 I into one class, since they share the 8250 HaeIII SNP, and the not better defined ‘‘Others (10394)’’ and ‘‘Others (þ10394)’’ into another class. A full logistic regression analysis with groups (patients vs. controls) as dependent variable and mt-hgs (hg H vs. remaining hgs), gender, and hgs by gender interaction as independent variables was utilized to evaluate a putatively different effect of the hgs in the two sexes. A multinomial logistic regression analysis with illness status (control ¼ 1, non-paranoid schizophrenia ¼ 0, paranoid schizophrenia ¼ 1) as dependent variable and mt-hgs, gender, and hgs by gender interaction as predictive variables was performed to define possible specific effects of hgs on defined schizophrenia subtypes. To avoid the reduction of the test power due to patient sub-classification, the hgs were collapsed together according to their phylogeny. The sister hgs U and K as well as J and T were collapsed into two classes since they share the mutation þ12308 HinfI and 4216, respectively, whereas the remaining hgs were clustered into the ‘‘all others’’ class, with the exception of the most frequent hg H, which acted as the reference category. To test the influence of hgs, gender, and schizophrenia subtypes on age of onset, the same hg grouping entered in a factorial analysis of variance on the logtransformed age, to improve its approximation to a Gaussian distribution. Both the full factorial and the main effect model were considered. The correlation between mt-hgs and age at onset of schizophrenia was also studied by the Kaplan–Meier method and the log-rank test for analyses of survival. The influence of hgs on the severity of schizophrenia was tested using a factorial analysis of variance on PANSS outcomes. All analyses were performed with the software SPSS (version 12.0). The power of the w2 test for global frequency hg differences between cases and controls was evaluated by the G* Power software [Faul and Erdfelder, 1992]. Quantitative Analysis of mt-DNA The mt-DNA levels were analyzed by comparing the ratio between mt-DNA and nuclear DNA (n-DNA) in a real-time polymerase chain reaction (PCR) analysis, using a BioRad iCycler iQ (BioRad Laboratories, Hercules, CA) thermocycler. The TaqMan1 RNase P Detection Reagents Kit (Applied Biosystems, Foster City, CA) was used to quantify the nuclear DNA; this kit detects and quantitates copies of the human RNase P gene, which encodes the RNA moiety for the RNase P enzyme. In the absence of a ‘‘ready to use’’ assay, the mt-DNA quantification was carried out utilizing a Custom TaqMan1 Gene Expression Assay (Applied Biosystems), and selecting the 16,081–16,569 bp mitochondrial sequence as the target fragment. For each sample, n-DNA and mt-DNA were amplified in two independent PCR reactions, simultaneously (in the same plate) and in triplicate. Quantitative-PCR (Q-PCR) amplification was performed in 20 ml containing 1 TaqMan1 Universal PCR Master Mix (Applied Biosystems), 1 mix of primer and labeled probe FAM dye, and 20 ng of total genomic DNA. The thermal cycling conditions were 958C for 10 min, 40 cycles of 958C for 15 sec and 608C for 1 min. Determination of mt-DNA Relative Levels The raw data on the fluorescence intensity were analyzed by the iCycler iQ software according to the ‘‘PCR baseline subtracted curve fit’’ analysis mode. The threshold cycles (Ct) were evaluated using the fluorescence thresholds indicated by the thermal cycler. The relative quantification was performed by comparing DCt (difference in threshold cycle) of the reference and the target gene in each sample. The following equation was used: 498 Magri et al. mt-DNA ¼ EDCt ðmean n-DNAmean mt-DNAÞ n-DNA ð1Þ where E represents the efficiency of the reaction, which in our study was equal to 2 since Applied Biosystems reported 100% efficiency for their Taqman assay (Applied Biosystems Application Note). Mean n-DNA and mean mt-DNA are the average Ct values of the triplicate assay for each sample. Group differences were estimated by the Pair Wise Fixed Reallocation Randomisation Testß implemented in the RESTß software [Pfaffl et al., 2002]. In order to avoid any assumption about distribution normality, a non-parametric test was preferred to the parametric T-test. Correlations between mt-DNA level and age at onset, age at collection and the quality of DNA were performed by the Spearman’s test, whereas the influence of hgs on the mt-DNA levels was studied by a factorial univariate analysis of variance. Both the analyses were performed with the software SPSS (version 12.0). RESULTS Haplogroup Analysis The hgs H, U, K, T, W, I, and J accounted for more than 85% of the total variability in both cases and controls. As shown in Table I, patients and healthy comparisons shared similar hg frequencies, which did not differ from those previously reported for the Italian population [Ghezzi et al., 2005]. The full logistic regression analysis demonstrated an absence of direct and interactive gender effects on hgs distribution and confirmed that none of the investigated hgs was associated with risk of schizophrenia. The multinomial logistic regression analysis showed that healthy controls, paranoid and non-paranoid schizophrenia patients did not differ from each other in mt-hg distribution, even when gender and hgs by gender interaction were taken into account. The preliminary analysis of the mean age at onset stratified by hgs (Table II) as well as the Kaplan–Meier method (Fig. 2) highlighted an earlier age at schizophrenia onset for patients harboring the J-T hg (Log Rank w2 ¼ 10.298, P ¼ 0.016). Almost all the patients belonging to hg J-T (92%) had an onset before 28 years, in contrast to 58% observed in those carrying the remaining hgs (Supplementary Fig. 1). After the subdivision of patients in two groups according to their age at onset (<28 years vs. onset 28 years), a binary logistic regression analysis was performed, using as fixed factor mt-hgs (hg J-T vs. remaining hgs). This analysis revealed that patients harboring hg J-T had a 7.8 (95% CI 1.7–34.7) fold increased risk of developing schizophrenia before 28 years. The factorial analysis of variance confirmed the significant effect of both gender (Fdf1 ¼ 12.671, P ¼ 0.001), and mt-hgs (Fdf3 ¼ 3.65, P ¼ 0.014) on schizophrenia onset and excluded an effect of schizophrenia subtypes (Supplementary Tables I and II). In particular, the pairwise comparison of the estimated marginal means showed that patients with the hg J-T had earlier onset than those with the hgs H (P ¼ 0.003), UK (P ¼ 0.033), or ‘‘all others’’ class (P ¼ 0.004) (Supplementary Table III). The factorial analysis of variance performed on the PANSS outcomes did not evidence any correlation between the severity of the disease and the mt-hgs (Fdf3 ¼ 0.997, P ¼ 0.396). The early age at onset in subjects harboring hg J-T did not correlate, therefore, with the severity of the disorder. Quantification of mt-DNA The quantitative analysis of mt-DNA level by Q-PCR assay in blood samples involved a subgroup of 37 patients (19 females and 18 males) and 41 controls (21 females and 20 males). These two subpopulations were representative of the original total sample; indeed, when compared with the remaining participants in the study, they did not differ as to sex, age at collection, mt-hg frequency distribution, and, for patients only, schizophrenia subtypes and age at onset of the disorder. Patients and controls gave comparable results for the relative amount of mt-DNA in the blood cells (P ¼ 0.190) (Fig. 3). Dichotomization for gender and schizophrenia subtype (Fig. 3) did not generate any difference in relative amounts of mt-DNA within and between the groups (male controls vs. male patients, P ¼ 0.700; female controls vs. female patients, P ¼ 0.147; female patients vs. male patients, P ¼ 0.579; female controls vs. male controls, P ¼ 0.655; paranoid subtype vs. nonparanoid subtype, P ¼ 0.100). The analysis of variance excluded a possible influence of mthgs on the mt-DNA levels in patients (Fdf3 ¼ 2.179, P ¼ 0.109), in controls (Fdf3 ¼ 0.221, P ¼ 0.881) and in the whole sample (Fdf3 ¼ 1.513, P ¼ 0.219) (Supplementary Fig. 2). The Spearman’s test showed that there was no correlation between mt-DNA levels and age at onset (P ¼ 0.601), age at collection (P ¼ 0.24), and DNA quality (P ¼ 0.20). DISCUSSION To our knowledge, this is the first study comparing DSMIV-TR schizophrenia patients and healthy controls for the most common European mt-hgs and mt-DNA levels in blood cells. TABLE I. Frequencies of the mt-hgs Analyzed in Our Sample and in Another Sample of the Italian Population Haplogroups H K U J T I W Other (10394) Other (þ10394) Patients (%) Controls (%) w2 P-values Other Italians (%)a 62 (0.437) 10 (0.070) 17 (0.120) 11 (0.077) 14 (0.099) 5 (0.035) 2 (0.014) 17 (0.120) 4 (0.028) 77 (0.405) 13 (0.068) 18 (0.095) 19 (0.100) 26 (0.137) 4 (0.021) 7 (0.037) 19 (0.100) 7 (0.037) 0.57 0.94 0.46 0.48 0.29 0.73b 646 (0.435) 114 (0.077) 186 (0.125) 126 (0.085) 143 (0.096) 36 (0.024) 235 (0.158)d 0.78c a This Italian sample is the CT1 group reported by Ghezzi et al. ; globally this sample did not differ significantly neither from our controls (P ¼ 0.50) nor from our patients (P ¼ 0.99). b 2 w test was evaluated combining the rare hgs W and I into one class. c 2 w test was evaluated combining the hgs ‘‘Others (10394)’’ and ‘‘Others (þ10394)’’ into one class. d This frequency was obtained summing the ‘‘Other’’ and ‘‘L-M’’ class frequencies of Ghezzi et al. . This class includes our W, Other (10394) and Other (þ10394) classes. mt-DNA and Schizophrenia 499 Fig. 1. Phylogenetic tree of the studied mt-DNA hgs. The diagnostic RFLP markers used for hg classification and their cut-site positions, according to the mt-DNA revised Cambridge Reference Sequence (rCRS) [Andrews et al., 1999], are reported along the tree branches. A plus indicates the presence of the restriction site, a minus its absence. The 4,216 mutation has not been investigated. The þ10394 DdeI polymorphism is underlined since recurrent, that means the þ10394 DdeI mutation observed on hg K, I, and J is the result of three independent mutational events. In the haplogroup analysis, no hg contributed to schizophrenia susceptibility and patients and controls were homogeneous in their matrilineal ancestry. Some evidence supports the robustness of this result. Power analysis of the haplogroup data indicated that the sample size was adequate to detect differences of heuristic relevance (effect size ¼ 0.3, power > 0.9) [Cohen, 1988]. Furthermore, stratifications by gender and schizophrenia subtype did not have any effect on hg frequency distribution. Finally, the hg distributions of patients and controls were similar to those reported for a large, representative sample of the general Italian population [Ghezzi et al., 2005]. Even if mt-hg did not act as a risk factor for schizophrenia, subjects harboring the hg J-T presented earlier onset of schizophrenia as compared to patients with different hg profiles. Indeed, in the J-T group the mean age at onset was 22.08 in respect to 26.98 of all the other hgs. It would be interesting as prospective to confirm the result in a larger dataset and verify if particular sub-lineages of J and T accumulate in the cases with anticipated onset. However, our preliminary results suggest that distinct entities with their own onset curves coexist under the common rubric of schizophrenia. Thus, mt-hgs may be a candidate tool for the division of schizophrenia patients into different, clinically relevant endophenotypes. Any comment about the mechanisms by which the hg J-T could anticipate the onset of schizophrenia is obviously purely speculative. Our results only excluded a correlation between mt-hgs and mt-DNA level. However, some evidence in the literature suggests a possible biological explanation. The hg J has been proposed to harbor mt-DNA missense mutations that partially uncouple the mitochondrial oxidative phosphorylation system (OXPHOS) and thus reduce ATP production [Ruiz-Pesini et al., 2004]. Furthermore, reduced ATP levels have been reported to occur in some defined areas of the brain in schizophrenia patients [Fukuzako et al., 1996; Kegeles et al., 1998; Jensen et al., 2006]. Bridging together these two indications, it may be reasonable to hypothesize that a partially uncoupled mt-OXPHOS associated with hg J could anticipate the appearance of schizophrenia, exacerbating an otherwise underlying sub-threshold ATP deficit. The relationship between uncoupled mt-OXPHOS and schizophrenia is also strengthened by a recent article of Martorell et al.  on mt-DNA analysis of six schizophrenia patients with an apparent maternal transmission. Interestingly, the most relevant result was the presence in 5 out of 6 patients of a new non-synonymous heteroplasmic mt-variant at position 12096, which was not found in 95 controls. According to the authors, this mutation, which falls in the TABLE II. Mean Age at Onset of Schizophrenia Patients Stratified by mt-hgs mt-hgs H Female (N ¼ 21) Male (N ¼ 38) U-K Female (N ¼ 16) Male (N ¼ 10) J-T Female (N ¼ 9) Male (N ¼ 15) All other hgs Female (N ¼ 9) Male (N ¼ 17) Mean age at onset 95% CI 26.97 31.14 24.66 26.81 27.94 25.00 22.08 23.00 21.53 27.19 30.44 25.47 24.76–29.18 26.77–35.52 22.39–26.92 24.29–29.32 24.39–31.49 21.23–28.77 19.42–24.74 18.99–27.01 17.67–25.40 24.35–30.04 24.74–36.15 22.18–28.76 Fig. 2. A Kaplan–Meier plot showing the earlier age at onset in schizophrenic patients carrying the mt-hg J-T (Log Rank w2 ¼ 10.298, P ¼ 0.016). 500 Magri et al. Fig. 3. Relative amount of mt-DNA in controls, schizophrenia patients and in the same subjects stratified for gender and paranoid non-paranoid subtypes. Each dot indicates the relative quantity of mt-DNA normalized with RNaseP gene, calculated as in Equation (1). ND4 complex, could interfere with the Hþ pumping activity of the NADH-ubiquinone oxidoreductase and might trigger a reduction in the rate of ATP production. However, the analysis of all our patients with an apparent maternal transmission (N ¼ 13) as well as of the 24 patients with hg J-T, failed to reveal the 12096T > A mutation. These data suggest that, in our patients, this mutation is not related with the schizophrenia phenotype or with the anticipated onset observed in hg J-T patients. For the relative mt-DNA levels in blood cells, our analysis seems to exclude the influence of mt-hgs on mt-DNA levels and, more generally, indicates that the mitochondrial dysfunctions reported in schizophrenia patients should not be due to a constitutive quantitative alteration in mt-DNA quantity. This conclusion agrees with a recent report on the post-mortem brain [Kakiuchi et al., 2005], where the sole quantitative difference reported was a marginal excess of the amount of mtDNA in the frontal cortex of female schizophrenia patients in comparison with female controls and male patients, but the significance disappeared following Bonferroni correction. In conclusion, our results point to an association of the hg J-T with an anticipated onset of the disorder, which is congruent with the possibility of some modulator effect of mt-DNA on deeper determinants of schizophrenia. ACKNOWLEDGMENTS The authors thank Dr. Silvia Bonomi and Dr. Alessandra Mosca for their valuable support with clinical evaluation, Dr. Cristian Bonvicini for his assistance with the statistics, and Prof. Massimo Gennarelli for his helpful suggestions. This study was supported by a grant (project 153) from the Health Authority of the Lombardia Region together with grants from the Centre on Behavioural and Neurodegenerative Disorders and the IDET Centre of Excellence (MIUR). REFERENCES Altar CA, Jurata LW, Charles V, Lemire A, Liu P, Bukhman Y, Young TA, Bullard J, Yokoe H, Webster MJ, Knable MB, Brockman JA. 2005. Deficient hippocampal neuron expression of proteasome, ubiquitin, and mitochondrial genes in multiple schizophrenia cohorts. Biol Psychiatry 58:85–96. American Psychiatric Association. 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th edition. Washington, DC: American Psychiatric Association. Andrews RM, Kubacka I, Chinnery PF, Lightowlers RN, Turnbull DM, Howell N. 1999. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nat Genet 23: 147. Cohen J. 1988. Statistical power analysis for the behavioral sciences, 2nd edition. Hillsdale, NJ: Erlbaum. De Benedictis G, Rose G, Carrieri G, De Luca M, Falcone E, Passarino G, Bonafe M, Monti D, Baggio G, Bertolini S, Mari D, Mattace R, Franceschi C. 1999. Mitochondrial DNA inherited variants are associated with successful aging and longevity in humans. FASEB J 13: 1532–15336. Dror N, Klein E, Karry R, Sheinkman A, Kirsh Z, Mazor M, Tzukerman M, Ben-Shachar D. 2002. State-dependent alterations in mitochondrial complex I activity in platelets: A potential peripheral marker for schizophrenia. Mol Psychiatry 7:995–1001. Faul F, Erdfelder E. 1992. GPOWER: A priori, post-hoc, and compromise power analyses for MS-DOS [Computer Program]. Bonn: Bonn University, Department of Psychology. First MB, Spitzer RL, Gibbon M, Williams JBW. 1996. Structural Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-DV). Washington, DC: American Psychiatric Press. Fukuzako H, Fukuzako T, Takeuchi K, Ohbo Y, Ueyama K, Takigawa M, Fujimoto T. 1996. Phosphorus magnetic resonance spectroscopy in schizophrenia: Correlation between membrane phospholipid metabolism in the temporal lobe and positive symptoms. Prog Neuropsychopharmacol Biol Psychiatry 20:629–640. Ghezzi D, Marelli C, Achilli A, Goldwurm S, Pezzoli G, Barone P, Pellecchia MT, Stanzione P, Brusa L, Bentivoglio AR, Bonuccelli U, Petrozzi L, Abbruzzese G, Marchese R, Cortelli P, Grimaldi D, Martinelli P, Ferrarese C, Garavaglia B, Sangiorgi S, Carelli V, Torroni A, Albanese A, Zeviani M. 2005. Mitochondrial DNA hg K is associated with a lower risk of Parkinson’s disease in Italians. Eur J Hum Genet 13:748– 752. Iwamoto K, Bundo M, Kato T. 2005. Altered expression of mitochondriarelated genes in postmortem brains of patients with bipolar disorder or schizophrenia, as revealed by large-scale DNA microarray analysis. Hum Mol Genet 14:241–253. Jensen JE, Miller J, Williamson PC, Neufeld RW, Menon RS, Malla A, Manchanda R, Schaefer B, Densmore M, Drost DJ. 2006. Grey and white matter differences in brain energy metabolism in first episode schizophrenia: (31)P-MRS chemical shift imaging at 4 Tesla. Psychiatry Res 146:127–135. Kakiuchi C, Ishiwata M, Kametani M, Nelson C, Iwamoto K, Kato T. 2005. Quantitative analysis of mitochondrial DNA deletions in the brains of patients with bipolar disorder and schizophrenia. Int J Neuropsychopharmacol 8:512–522. Kay SR, Fiszbein A, Opler LA. 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 13:261–276. Kegeles LS, Humaran TJ, Mann JJ. 1998. In vivo neurochemistry of the brain in schizophrenia as revealed by magnetic resonance spectroscopy. Biol Psychiatry 44:382–398. Kung L, Roberts RC. 1999. Mitochondrial pathology in human schizophrenic striatum: A postmortem ultrastructural study. Synapse 31:67–75. Magri C, Gardella R, Barlati SD, Podavini D, Iatropoulos P, Bonomi S, Valsecchi P, Sacchetti E, Barlati S. 2006. Glutamate AMPA receptor subunit 1 gene (GRIA1) and DSM-IV-TR schizophrenia: A pilot casecontrol association study in an Italian sample. Am J Med Genet Part B 141B:287–293. Martorell L, Segues T, Folch G, Valero J, Joven J, Labad A, Vilella E. 2006. New variants in the mitochondrial genomes of schizophrenic patients. Eur J Hum Genet 14:520–528. mt-DNA and Schizophrenia Maurer I, Zierz S, Moller H. 2001. Evidence for a mitochondrial oxidative phosphorylation defect in brains from patients with schizophrenia. Schizophr Res 48:125–136. Middleton FA, Mirnics K, Pierri JN, Lewis DA, Levitt P. 2002. Gene expression profiling reveals alterations of specific metabolic pathways in schizophrenia. J Neurosci 22:2718–2729. Nurnberger JI Jr, Blehar MC, Kaufmann CA, York-Cooler C, Simpson SG, Harkavy-Friedman J, Severe JB, Malaspina D, Reich T. 1994. Diagnostic interview for genetic studies. Rationale, unique features, and training. NIMH Genetics Initiative. Arch Gen Psychiatry 51:849– 859. Pfaffl MW, Horgan GW, Dempfle L. 2002. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30:36. Prabakaran S, Swatton JE, Ryan MM, Huffaker SJ, Huang JT, Griffin JL, Wayland M, Freeman T, Dudbridge F, Lilley KS, Karp NA, Hester S, Tkachev D, Mimmack ML, Yolken RH, Webster MJ, Torrey EF, Bahn S. 2004. Mitochondrial dysfunction in schizophrenia: Evidence for 501 compromised brain metabolism and oxidative stress. Mol Psychiatry 9:684–697 , 643. Prayson RA, Wang N. 1998. Mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes (MELAS) syndrome: An autopsy report. Arch Pathol Lab Med 122:978–981. Prince JA, Blennow K, Gottfries CG, Karlsson I, Oreland L. 1999. Mitochondrial function is differentially altered in the basal ganglia of chronic schizophrenics. Neuropsychopharmacology 21:372–379. Ruiz-Pesini E, Mishmar D, Brandon M, Procaccio V, Wallace DC. 2004. Effects of purifying and adaptive selection on regional variation in human mtDNA. Science 303:223–226. Suzuki T, Koizumi J, Shiraishi H, Ishikawa N, Ofuku K, Sasaki M, Hori T, Ohkoshi N, Anno I. 1990. Mitochondrial encephalopathy (MELAS) with mental disorder. CT, MRI, and SPECT findings. Neuroradiology 32:74– 76. Thomeer EC, Verhoeven WM, van de Vlasakker CJ, Klompenhouwer JL. 1998. Psychiatric symptoms in MELAS; a case report. J Neurol Neurosurg Psychiatry 64:692–693.