The World Journal of Biological Psychiatry

ISSN: 1562-2975 (Print) 1814-1412 (Online) Journal homepage: http://www.tandfonline.com/loi/iwbp20

Gene expression study of mitochondrial complex

I in schizophrenia and paranoid personality
disorder

Arvin Haghighatfard, Sarah Andalib, Mozhdeh Amini Faskhodi, Soha

Sadeghi, Amir Hossein Ghaderi, Shadi Moradkhani, Jalal Rostampour, Zeinab
Tabrizi, Ali Mahmoodi, Talie Karimi & Zakieh Ghadimi

To cite this article: Arvin Haghighatfard, Sarah Andalib, Mozhdeh Amini Faskhodi, Soha Sadeghi,
Amir Hossein Ghaderi, Shadi Moradkhani, Jalal Rostampour, Zeinab Tabrizi, Ali Mahmoodi,
Talie Karimi & Zakieh Ghadimi (2017): Gene expression study of mitochondrial complex I in
schizophrenia and paranoid personality disorder, The World Journal of Biological Psychiatry, DOI:
10.1080/15622975.2017.1282171

To link to this article: http://dx.doi.org/10.1080/15622975.2017.1282171

View supplementary material

Published online: 20 Feb 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=iwbp20

Download by: [Fudan University] Date: 20 February 2017, At: 04:56

THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY, 2017
http://dx.doi.org/10.1080/15622975.2017.1282171

ORIGINAL INVESTIGATION

Gene expression study of mitochondrial complex I in schizophrenia and

paranoid personality disorder
Arvin Haghighatfarda, Sarah Andalibb, Mozhdeh Amini Faskhodic, Soha Sadeghid, Amir Hossein Ghaderie,
Shadi Moradkhanif, Jalal Rostampourg, Zeinab Tabrizih, Ali Mahmoodia, Talie Karimii and Zakieh Ghadimij
a
Department of Biology, Science and Research Branch, Islamic Azad University, Tehran, Iran; bInstitute for Brain and Cognitive
Science, Shahid Beheshti University, Tehran, Iran; cDepartment of Biology, Tehran Medical Branch, Islamic Azad University, Tehran,
Iran; dLaboratory of Medical Genetics, National Institute of Genetic Engineering and Biotechnology (NIGEB), Tehran, Iran; eCognitive
Neuroscience Lab, Department of Psychology, University of Tabriz, Tabriz, Iran; fDepartment of Physics, Amirkabir University of
Technology, Tehran, Iran; gDepartment of Cell & Molecular Biology, School of Biology, College of Science, University of Tehran,
Tehran, Iran; hDepartment of Medical Immunology, Shahid Sadoughi University of Medical Sciences and Health Services, Yazd, Iran;
i
Medical Biotechnology Research Center, Ashkezar Branch, Islamic Azad University, Ashkezar, Iran; jDepartment of Biology, Qom
Branch, Islamic Azad University, Qom, Iran

ABSTRACT ARTICLE HISTORY

Objectives: The aetiology and molecular mechanisms of schizophrenia (SCZ) and paranoid Received 7 July 2016
personality disorder (PPD) are not yet clarified. The present study aimed to assess the role of Revised 2 December 2016
mitochondrial complex I and cell bioenergetic pathways in the aetiology and characteristics of Accepted 2 January 2017
SCZ and PPD.
Methods: mRNA levels of all genomic and mitochondrial genes which encode mitochondrial KEYWORDS
complex I subunits (44 genes) were assessed in blood in 634 SCZ, 340 PPD patients and 528 Schizophrenia; paranoid
non-psychiatric subjects using quantitative real-time PCR, and associated comprehensive psychi- personality disorder;
atric, neurological and biochemical assessments. mitochondrial complex I;
Results: Significant expression changes of 18 genes in SCZ patients and 11 genes in PPD patients gene expression; NDUFS1
were detected in mitochondrial complex I. Most of these genes were novel candidate genes for
SCZ and PPD. Several correlations between mRNA levels and severity of symptoms, drug response,
deficits in attention, working memory, executive functions and brain activities were found.
Conclusions: Deregulations of both core and supernumerary subunits of complex I are involved
in the aetiology of SCZ and PPD. These deregulations have effects on brain activity as well as
disorder characteristics.

Introduction depression and mannerisms are frequently seen in

Schizophrenia (SCZ) is a chronic neuropsychiatric dis-
order with 1% prevalence worldwide and no certain
known aetiology. SCZ has great heterogeneity across
individuals and variation in severity of positive and
negative symptoms (Ben-Shachar & Karry 2007; Clay
et al. 2011). Recent hypotheses have addressed both
neurochemical and neurodevelopmental components
in the pathogenesis of SCZ (Dror et al. 2002; Karry
et al. 2004; Mehler-Wex et al. 2006).
Paranoid personality disorder (PPD) is a mental dis-
order characterised by paranoia and a pervasive, long-
standing suspiciousness and generalised mistrust of
others with 0.5–2.5% occurrence in the general popu-
lation and no clear aetiology (Waldinger 1997).
Paranoid thoughts are the main shared symptoms
between PPD and SCZ, but hostility, social avoidance,
both disorders as well. Hallucinations and delusions
are two common symptoms of SCZ that are not pre-
sent in PPD patients. Several studies reported a spe-
cific familial association between SCZ-related
personality disorders such as PPD, and SCZ patients
(Webb & Levinson 1993). While a series of twin studies
found significant heritability for PPD in the range of
35–60%, genetic and molecular bases of PPD remain
unclear (Reichborn-Kjennerud 2010). As the aetiology
of SCZ and PPD remains unclear and their diagnosis is
based on clinical interviews, reliable peripheral bio-
logical markers for these disorders are needed. Whole-
genome gene expression profiling in SCZ revealed
gene expression alteration in several genes with differ-
ent functions such as neurotransmission, gene regula-
tion, cell cycle progression, differentiation, apoptosis,

CONTACT Arvin Haghighatfard Arvinland@yahoo.com Department of Biology, Science and Research Branch, Islamic Azad University, Tehran, Iran
Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/15622975.2017.1282171.
ß 2017 Informa UK Limited, trading as Taylor & Francis Group
2 A. HAGHIGHATFARD ET AL.

Figure 1. Schematic representation of human complex I subunits and their putative topology within the complex. All core subu-
nits along with 30 supernumerary mammalian subunits and their positions in complex I are presented. Bovine homologues are
shown in grey and human subunits in which mutations are found are marked with an asterisk (Ugalde et al. 2004).

microRNA processing and immunity. This functional
diversity in SCZ candidate genes could be related to
pathophysiological heterogeneity of complex psychi-
atric disorders (Sanders et al. 2013).
Mitochondrial complex I (NADH: ubiquinone oxidor-
eductase) has a necessary role in respiration in many
aerobic organisms. Complex I oxidise NADH from the
tricarboxylic acid cycle and b-oxidation, reduces ubi-
quinone, and transports protons across the inner
membrane, contributing to the proton-motive force in
mitochondria. It is also a major contributor to cellular
production of reactive oxygen species (Hirst 2013).
Complex I, which is one of the largest and most
complicated enzymes in cells, contains 44 different
subunits (nuclear and mitochondrial genomes) and
nine redox cofactors. Fourteen subunits are the ‘core’
subunits that are conserved in complex I in all species.
Core subunits contain all the mechanistically critical
cofactors and structural elements, and are sufficient
for catalysis. The other 30 ‘supernumerary’ subunits
are mammalian specific. The supernumerary subunits
are important in assembly, regulation, stability and
protection against oxidative stress. There is a lack of
high-resolution structure-limited knowledge about the
role of 30 supernumerary mammalian subunits in the
assembly and activity of complex I (Figure 1). In add-
ition, the abnormalities or mutations in complex I sub-
units which cause human diseases are not clear
(Vinothkumar et al. 2014).
Impaired cellular energy state, which is a conse-
quence of impaired mitochondrial function, is an inter-
esting hypothesis for explaining the pathophysiology
of SCZ. An abnormal cellular energy state can lead to
alterations in neuronal function, plasticity and brain
circuitry, and thereby to the cognitive and behavioural
aberrations characteristic of SCZ (Ben-Shachar 2002).
 THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 3

There is evidence which indicates that SCZ has
pathological components that could be attributed to
abnormalities in mitochondrial function (Mehler-Wex
et al. 2006). Also, it has been reported that mito-
chondrial dysfunctions negatively affect neuronal
plasticity, and cause cognitive deficits, behavioural
abnormalities and executive dysfunction in several
neurological and neuropsychiatric disorders including
Parkinsonism, bipolar disorder and SCZ (Akarsu et al.
2014).
Ben-Shachar et al. (1999) reported the enzymatic
activity of complex I, an increase in peripheral blood
of patients with SCZ. Dror et al. (2002) reported over-
activity of mitochondrial complex I in patients with
SCZ, which was associated with positive and negative
psychiatric symptoms. A study of mitochondrial
complex I genes has been repeated in European popu-
lation by Mehler-Wex et al. (2006) and detected
over-expression of NDUFS1 in 10 schizophrenic
patients (encoding greatest subunit of mitochondrial
complex I) as a marker for SCZ. Akarsu et al. (2014)
studied the correlation between the expression of four
genes in mitochondrial complex I and III in blood sam-
ples of schizophrenic patients with psychiatric results
(Akarsu et al. 2014). They confirmed the over-expres-
sion of NDUFS1, NDUFV1 and NDUFV2 and a positive
correlation of NDUFV2 over-expression using the Brief
Psychiatric Rating Scale (BPRS) and simplified acute
physiology score in first-episode SCZ.
Similar positive symptoms in SCZ and PPD led us to
evaluate the same pattern of genes in both disorders
to better understand the shared genetic bases of SCZ
and other psychiatric and personality disorders, espe-
cially in bioenergetic pathways. Previous studies of
mitochondrial complex I activity in SCZ was focussed
in large subunits. The present study aimed to compre-
hensively investigate transcript levels of all genomic
and mitochondrial genes which encode mitochondrial
complex I subunits, and the correlation of mRNA levels
with psychiatric symptoms, intelligence, working mem-
ory, cognition, attention, executive functions, neuro-
logical situation, antipsychotic treatment responses
and the biochemistry of SCZ and PPD in a large num-
ber of samples (Tables 1 and 2).

Table 1. Clinical characteristics in SCZ, PPD and non-psychiatric subjects.

Variables SCZ PPD Non-psychiatric
Age 31 (28–35) 33 (27–36) 30 (27–32)
Age of onset (years) 22.5 (18–27) 28.5 (19–30) –
Duration of illness (month) 39.7 (28–41) 16.4 (12–19) –
Duration of hospitalization (day) 16.6 (15–17) 15.4 (14–18) –
Estimated antipsychotic medications (mg/day) Haloperidol: 2 Haloperidol: 3 –
Clozapine: 600 Risperidone: 2
Clonazepam: 0.5 diazepam: 250
Risperidone: 2
Olanzapine: 20
Fluoxetine: 50
Number of antipsychotic resistant 41 6 –
PANSS total 74.42 (64–79) – –
General symptoms 31.32 (27–34)
Positive score 27.38 (24–32) – –
Negative score 14.76 (12–18) – –
BPRS total 56.29 (41to 59) – –
withdrawal/Retardation 6.1 (5–7) – –
Anxious/Depression 8.45 (8–9) – –
Hostile/Suspiciousness 12.19 (10–13) – –
Thinking disturbance 14.34 (13–16) – –
IQ total score 85.3 (82–106) 95.44 (93–112) 98.23 (90–118)
Neutral stimuli Stroop test score 815.9 (728–840) 763.3 (684–785) 684(634–712)
Incongruent stimuli Stroop test score 967.6 (829–983) 914.3 (781–946) 827 (775–855)
Congruent stimuli Stroop test score 852.5 (811–874) 720.8 (634–736) 682(613–707)
WSCT correct responses 59.34 (42–68) 67.32(51–77) 78.13 (73–80)
WSCT completed groups 2.1 (1–4) 2.55 (1–4) 3.86 (2–4)
TMT-A 64.23 (48–68) 52.6 (43–62) 37.4 (29–42)
TMT-B 148.56 (122–156) 111.49 (102–133) 82.36 (66–142)
SOD(U/ml) 0.11 (0.9–0.14) 0.28 (0.24–0.33) 0.35 (0.24–0.41)
EEG-ERP P300-auditory: 593 (508–602) ms P300-auditory: 521 (487–532) ms P300-auditory: 408 (390–432) ms
P300-visual: 542 (494–563) ms P300-visual: 515 (502–568) ms P300-visual: 411 (367–422) ms
MMN standard tone: MMN standard tone: 1881 (1824 to1897) MMN standard tone:
1893 (1811–1924) MMN frequency deviant: 203 (196–211) 1729 (1683–1836)
MMN frequency deviant: MMN duration deviant: 198.4 (182–205) MMN frequency deviant:
223 (206–234) 196.4 (188–202)
MMN duration deviant: MMN duration deviant:
201 (192–207) 197 (191–204)
PANSS: Positive and Negative Syndrome Scale; BPRS: Brief Psychiatric Rating Scale; WSCT: Wisconsin Card Sorting Test; TMT: Trail Making Test; SOD: super-
oxide dismutase; EEG-ERP: electroencephalography–event-related potential; MMN: mismatch negativity.
4 A. HAGHIGHATFARD ET AL.

Table 2. Gene mRNA levels of the cases; ratio (mean ratio of gene mRNA levels in patient groups in comparison with non-psy-
chiatric subjects) and statistical results (group comparisons for mRNA levels of 44 genes of interest versus four housekeeping
genes, due to 0.05/44 ¼ 0.001042)
SCZ vs PPD vs Paranoid SCZ vs
non-psychiatric non-psychiatric SCZ vs non-psychiatric other SCZ patients PPD vs non-psychiatric
Genes Subunits subjects (Ratio) subjects (Ratio) subjects (P and F values) (P and F values) subjects (P and F values)
MT-ND1 core subunit 1 0.98 ± 0.66 0.96 ± 0.72 P¼0.12 P¼0.72 P¼0.24
F¼2.21 F¼1.08 F¼1.44
MT-ND2 core subunit 2 1.02 ± 0.23 0.92 ± 0.41 P¼0.28 P¼0.54 P¼0.16
F¼1.32 F¼1.09 F¼2.11
MT-ND3 core subunit 3 0.95 ± 0.46 0.93 ± 0.51 P¼0.58 P¼0.88 P¼0.74
F¼1.19 F¼0.78 F¼0.68
MT-ND4 core subunit 4 0.83 ± 0.29 1.08 ± 0.27 P¼0.35 P¼0.67 P¼0.56
F¼1.64 F¼0.87 F¼1.53
MT-ND4L core subunit 4L 1.12 ± 0.34 1.03 ± 0.55 P¼0.3 P¼0.94 P¼0.56
F¼1.55 F¼0.61 F¼1.03
MT-ND5 core subunit 5 0.92 ± 0.4 0.87 ± 0.18 P¼0.43 P¼0.82 P¼0.62
F¼1.13 F¼0.38 F¼0.94
MT-ND6 core subunit 6 0.88 ± 0.25 1.06 ± 0.19 P¼0.22 P¼0.46 P¼0.29
F¼1.83 F¼1.76 F¼2.02
NDUFA1 supernumerary A1 0.63 ± 0.35 0.87 ± 0.48 P¼0.0002 P¼0.23 P¼0.11
F¼21.18 F¼2.17 F¼2.24
*down-regulated
NDUFA2 supernumerary A2 0.58 ± 0.62 0.8 ± 0.28 P¼0.0003 P¼0.34 P¼0.08
F¼19. 26 F¼1.13 F¼7.12
*down-regulated
NDUFA3 supernumerary A3 0.85 ± 0.36 0.91 ± 0.67 P¼0.51 P¼0.84 P¼0.72
F¼2.17 F¼1.35 F¼1.27
NDUFA5 supernumerary A5 1.44 ± 0.27 1.32 ± 0.53 P¼0.0004 P¼0.13 P¼0.0004
F¼16.76 F¼2.13 F¼18.21
*up-regulated *up-regulated
NDUFA6 supernumerary A6 0.98 ± 0.04 0.94.± 0.11 P¼0.32 P¼0.4 P¼0.41
F¼1.24 F¼2.11 F¼1.33
NDUFA7 supernumerary A7 0.96 ± 0.24 0.89 ± 0.57 P¼0.6 P¼1.2 P¼0.81
F¼1.19 F¼0.27 F¼0.17

NDUFA8 supernumerary A8 1.74 ± 0.79 1.67 ± 0.26 P¼0.0008 P¼0.18 P¼0.0006

F¼14.66 F¼2.94 F¼17.74
*up-regulated *up-regulated
NDUFA9 supernumerary A9 1.07 ± 0.47 1.05 ± 0.39 P¼0.65 P¼0.52 P¼0.39
F¼0. 82 F¼0.8 F¼1.54
NDUFA10 supernumerary A10 0.85 ± 0.17 0.95 ± 0.42 P¼0.32 P¼0.69 P¼0.44
F¼1.15 F¼0. 66 F¼1.05

NDUFA11 supernumerary A11 1.06 ± 0.24 0.99 ± 0.33 P¼0.8 P¼0.65 P¼0.73
F¼0.32 F¼0.93 F¼0.79

NDUFA12 supernumerary A12 0.94 ± 0.11 1.03 ± 0.35 P¼0.31 P¼0.44 P¼0.26
F¼2.04 F¼1.38 F¼1.54

NDUFA13 supernumerary A13 1.69 ± 0.55 1.73 ± 0.49 P¼0.0008 P¼0.29 P¼0.0005
F¼18.36 F¼2.24 F¼16.19
*up-regulated *up-regulated
NDUFAB1 supernumerary AB1 0.47 ± 0.23 0.87 ± 0.31 P¼0.0004 P¼0.54 P¼0.37
F¼15.71 F¼0.86 F¼1.32
*down-regulated
NDUFB1 supernumerary B1 0.95 ± 0.24 0.93 ± 0.54 P¼0.58 P¼0.48 P¼0.34
F¼1.22 F¼1.32 F¼1.46

NDUFB2 supernumerary B2 1.48 ± 0.37 0.86 ± 0.22 P¼0.0002 P¼0.56 P¼0.42

F¼21.17 F¼1.04 F¼1.28
*up-regulated
NDUFB3 supernumerary B3 1.02 ± 0.56 0.97 ± 0.36 P¼0.21 P¼0.72 P¼1.06
F¼1.33 F¼0.49 F¼0.08

NDUFB4 supernumerary B4 1.1 ± 0.26 0.95 ± 0.28 P¼0.38 P¼0.57 P¼0.49

F¼1.28 F¼0.92 F¼1.19

NDUFB5 supernumerary B5 1.58 ± 0.62 1.52 ± 0.34 P¼0.0003 P¼0.17 P¼0.0002

F¼18.23 F¼3.05 F¼19.16
*up-regulated *up-regulated
NDUFB6 supernumerary B6 0.95 ± 0.46 0.9 ± 0.31 P¼0.52 P¼0.78 P¼0.44
F¼2.14 F¼1.55 F¼1.65
(continued)
 THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 5

Table 2. Continued
SCZ vs PPD vs Paranoid SCZ vs
non-psychiatric non-psychiatric SCZ vs non-psychiatric other SCZ patients PPD vs non-psychiatric
Genes Subunits subjects (Ratio) subjects (Ratio) subjects (P and F values) (P and F values) subjects (P and F values)

NDUFB7 supernumerary B7 1.12 ± 0.18 1.01 ± 0.53 P¼0.27 P¼0.54 P¼0.46

F¼3.12 F¼1. 6 F¼1.95

NDUFB8 supernumerary B8 1.03 ± 0.39 1.09 ± 0.24 P¼0.38 P¼1.19 P¼0.24

F¼2.2 F¼0.45 F¼2.39

NDUFB9 supernumerary B9 1.68 ± 0.46 1.79 ± 0.61 P¼0.0007 P¼0.86 P¼0.0004

F¼17.23 F¼0.13 F¼19.1
*up-regulated *up-regulated
NDUFB10 supernumerary B10 0.48 ± 0.78 0.84 ± 0.52 P¼0.0003 P¼0.83 P¼0.04
F¼20.14 F¼0.9 F¼7.06
*down-regulated
NDUFB11 supernumerary B11 0.56 ± 0.42 0.78 ± 0.59 P¼0.0008 P¼0.18 P¼0.0003
F¼16.24 F¼2.43 F¼19.38
*down-regulated *down-regulated
NDUFC1 supernumerary C1 1.54 ± 0.53 1.13 ± 0.16 P¼0.0004 P¼0.79 P¼0.06
F¼18.21 F¼1.17 F¼5.28
*down-regulated
NDUFC2 supernumerary C2 1.05 ± 0.36 1.08 ± 0.54 P¼1.02 P¼0.51 P¼0.89
F¼0.94 F¼1.63 F¼1.03

NDUFS1 core subunit S1 1.89 ± 0.73 1.74 ± 0.28 P¼0.0001 P¼0.0006 P¼0.0002
F¼22.36 F¼17.36 F¼19.84
*up-regulated *up-regulated *up-regulated
NDUFS2 core subunit S2 0.84 ± 0.7 0.96 ± 0.65 P¼0.87 P¼0.59 P¼0.86
F¼1.0 F¼1.94 F¼1.02

NDUFS3 core subunit S3 1.04 ± 0.58 0.94 ± 0.66 P¼0.24 P¼0.71 P¼0.9
F¼2.15 F¼1.3 F¼0.58

NDUFS4 supernumerary S4 1.62 ± 0.46 1.05 ± 0.74 P¼0.0005 P¼0.83 P¼0.85

F¼17.32 F¼1.12 F¼1.01
*up-regulated

NDUFS5 supernumerary S5 1.2 ± 0.4 0.97 ± 0.86 P¼0.52 P¼0.49 P¼0.73

F¼1.21 F¼1.53 F¼0.96

NDUFS6 supernumerary S6 0.96 ± 0.55 1.03 ± 0.37 P¼0.09 P¼0.64 P¼0.12

F¼4.27 F¼1.12 F¼3.08

NDUFS7 core subunit S7 0.75 ± 0.54 0.68 ± 0.66 P¼0.0008 P¼0.44 P¼0.0003
F¼16.21 F¼2.2 F¼18.42
*down-regulated *down-regulated
NDUFS8 core subunit S8 0.69 ± 0.58 0.54 ± 0.62 P¼0.0006 P¼0.72 P¼0.0007
F¼15.57 F¼1.24 F¼13.09
*down-regulated *down-regulated
NDUFV1 core subunit V1 1.67 ± 0.8 1.71 ± 0.65 P¼0.0004 P¼0.09 P¼0.0006
F¼16.18 F¼2.32 F¼15.63
*up-regulated *up-regulated
NDUFV2 core subunit V2 1.85 ± 0.63 1.73 ± 0.78 P¼0.0007 P¼0.12 P¼0.0002
F¼16.22 F¼2.07 F¼18.3
*up-regulated *up-regulated
NDUFV3 supernumerary V3 1 ± 0.32 1.04 ± 0.73 P¼0.67 P¼0.74 P¼0.35
F¼1.12 F¼1.14 F¼2.17
Significant changes/non-significant changes.

Materials and methods medications, in the 18–24-month period before sam-

Subject selection
The study included 634 unrelated SCZ patients
(414 male, 220 female) and 340 unrelated PPD patients
(204 male, 136 female), diagnosed by two independ-
ent senior psychiatrists based on extended clinical
interview and patients’ DSM-V chart review. All of the
patients had started treatment, including antipsychotic
pling and were considered as chronic patients.
Patients were selected from psychiatric hospitals and
clinics of 22 out of the 31 provinces of Iran. Patients
with schizoaffective or schizotypal illness were
excluded. Chlorpromazine (CPZ) equivalents used to
give the estimated average daily dose; and estimated
lifetime CPZ equivalents considering of duration of
illness. The non-psychiatric group included 528
6 A. HAGHIGHATFARD ET AL.
non-psychiatric subjects (325 male, 203 female) with
matched sex, age, race, socioeconomic situation, famil-
ial situation and education to the patient groups (non-
significant P values in the basic demographic situation
are presented in Supplementary Table 1, available
online) and with no history of any psychological prob-
lems, no current or history of severe medical condi-
tions, neurological disorder, history of head trauma
with loss of consciousness, no psycho-stimulant or opi-
oid drug abuse and alcohol or nicotine dependence.
Subjects were given an explanation on the purpose of
the study and then written informed consent was pro-
vided according to the Declaration of Helsinki. The
study was approved by central ethical committee of
Islamic Azad University in Tehran.
Analysis of clinical data and psychiatric
assessment
Clinical information including duration of illness, sub-
types of SCZ (297 paranoid, 33 catatonic, 53 disorgan-
ised, 168 undifferentiated and 83 residual),
antipsychotic resistance and familial history of psych-
osis were extracted from patient's files. Two trained
senior psychiatrists performed the ratings independ-
ently and a consensus rating was obtained.
Symptom severity was measured with the Positive
and Negative Syndrome Scale (PANSS), a 30-item semi-
structured interview designed to assess three symp-
tom categories associated with SCZ: positive, negative
and general symptoms (APA 1994).
The 24-item version of the BPRS is a rating scale for
psychiatric symptoms such as depression, anxiety, hal-
lucinations and unusual behaviour (Ventura et al.
2000).
Neuropsychological assessment
The Wisconsin Card Sorting Test (WCST) is a neuro-
psychological test of ‘set-shifting’, which use to assess
the following ‘frontal’ lobe functions and study of
executive functions (Drake & Lewis 2003). A number of
stimulus cards are presented to the participant. The
participant is told to match the cards, but not how to
match; but the participant is told whether a particular
match is right or wrong (Kaufman & Lichtenberger
2005).
The Wechsler Adult Intelligence Scale (WAIS) is a
test designed to measure intelligence in adults and
older adolescents. Digit span and Dot span are two
subtests of WAIS which examine the verbal and
Spatial working memory (Kaufman & Lichtenberger
2005).
The Stroop test is a neuropsychological test which
is used to measure sustained attention. This test is
based on the Stroop effect, which is a demonstration
of interference in the reaction time of a task and using
the name of a colour which is printed in a colour not
denoted by the name (Carter et al. 1997).
The Trail Making Test is a neuropsychological test
of visual attention and task switching. It is also use to
examine cognitive and executive functioning
(Gonzalez-Liencres et al. 2014).
Neurological assessment
Electroencephalography (EEG) was recorded using cus-
tom-designed electrode caps with a 64-channel
BioSemi Active-Two system (BioSemi BV) and data were
processed by the Brain Vision Analyser package, version
2.0 (Brain Products GmbH, Munich, Germany). Three
event-related potentials were measured: mismatch
negativity (MMN), amplitude of P300 auditory oddball
task, and P300 visual amplitude. The methods and data
analysis processes were operated according to previous
studies (Silver et al. 2003; Del Re et al. 2014).
Blood sampling and real-time PCR
Blood samples(15 ml) were collected from the cubital
vein without tourniquet using PAXgene blood RNA
tubes (Cat No762165) between 08.00 and 09.00 h in
the hospital laboratory sampling rooms and RNA
extraction started immediately after sampling. Total
RNA was extracted from peripheral blood samples
according to the standard protocols of the RNA
Purification kit (GeneJETTM RNA Purification Kit#K0732,
Fermentas, Latvia). Total RNA was treated with DNase
for the removal of contaminating genomic DNA using
DNase Treatment & Removal Reagents (DNase I,
RNase-free (#EN0521) Fermentas, Latvia), according to
the manufacturers protocol. The quality and integrity
of extracted RNA was examined by gel electrophoresis
and UV spectroscopy, and sampling repeated for sub-
jects with low-quality RNA in the first sample.
The cDNA was synthesised with a Transcription First
Strand cDNA Synthesis Kit (RevertAid Premium First
Strand cDNA Synthesis Kit #K1652, Fermentas, Latvia)
according to the manufacturer’s protocol. Primers for
all genes designed by ‘Gene runner version 3.05’ and
‘oligo7’ software and blasted on the NCBI website (pri-
mer sequences are presented in Supplementary Table
2). An average of four housekeeping genes (b-actin,
PGK1, EMC7 and CHMP2A) were used for normalisation
as endogenous reference genes (CT values of house-
keeping genes in all subjects mean ± SD are presented

in Supplementary Table 3). PCR and agarose gel elec-
trophoresis were used to verify the predicted size of
PCR amplicons for all 48 genes. Standard curves for
each gene were prepared using serial dilutions (1:4) of
pooled cDNA from total RNA extracted from blood
samples of 10 non-psychiatric subjects. In each experi-
ment, the R2 value of the standard curve was more
than 0.99, and no-template control assays resulted in
no detectable signal. Quantitative real-time PCR (qPCR)
was performed using syber green (Thermo Scientific
Maxima SYBR Green/ROX qPCR Master Mix (2X)
#K0221, Fermentas, Latvia). A triplicate method was
performed for quantitative real-time PCR using a
7900HT Fast Real-Time PCR System with a Fast 96-Well
Block Module (Applied Biosystems, Foster City, CA,
USA). PCR data were obtained by Sequence Detector
Software (SDS version 2.3 Rev C Patch, Applied
Biosystems) and quantified by the standard curve
method. This software plotted the real-time fluores-
cence intensity and selected the threshold within the
linear phase of the amplicon profile. The software plot-
ted a standard curve of the cycle at threshold versus
extracted RNA quantity. Samples were measured in
one plate for one target gene and their Ct values were
in the linear range of the standard curve. In qPCR
tests, outliers or sample failures were repeated for
each gene. The ratio was calculated using the pfafle
formula.
Biochemical assessment
Several studies suggest that oxidative stress might
serve as a potential biomarker in the aetiology and
clinical course of SCZ (Oribe et al. 2014). Subjects were
asked to sleep normally the previous night and refrain
from eating, using drugs, drinking alcohol or doing
physical exercise 8 h before the sampling. All persons
with any smoking history were excluded from the
study. Blood samples were collected from the partici-
pants at admission time (08:00 to 09:00 h). Plasma and
serum were obtained by centrifugation and, together
with the whole-blood sample, were kept at 80  C
until analysis. Superoxide dismutase (SOD) was ana-
lysed from serum. SOD activity was determined using
enzyme-linked immunosorbent assay, which measures
the three types of SOD (Cu/Zn, Mn and Fe), according
to the manufacturer’s instructions (Cayman Chemical
Company, MI, USA). Absorbance was read at 450 nm.
Statistical analysis
Descriptive data are expressed as mean ± SD (range)
and level of statistical significance was set at P < 0.05.
THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 7
Compliance with normal distribution for continuous
variables was assessed via the Kolmogorov-Smirnov
test. For multiple group comparisons, statistical differ-
ences were calculated by one-way ANOVA followed by
independent Student’s t-test. Pearson correlation ana-
lysis was conducted to determine the relationship
between variables. Bonferroni correction was used for
multiple comparisons corrections. To control of any
potential confounds, body mass index (BMI), medica-
tion classes, RNA quality and concentration, cDNA syn-
thesis quality, plates/runs of qPCR and primer
efficiency were added as covariates and persistence of
the significant difference in main effect between
groups was assessed by ANCOVA. Statistical analysis
conducted by using SPSS version23.
Results
The clinical and demographic data of the non-psy-
chiatric subjects and patients, including medications,
are given in parts A and B of Table 1. ANCOVA
examinations showed BMI (P ¼ 0.45, f ¼ 0.32), medi-
cation classes (P ¼ 0.27, f ¼ 2.04), RNA quality and
concentration (P ¼ 0.23, f ¼ 2.13), cDNA synthesis
quality (P ¼ 0.39, f ¼ 1.08), plates/runs of qPCR
(P ¼ 0.21, f ¼ 2.21) and primer efficiency (P ¼ 0.42,
f ¼0.87), which were added as covariates and had no
effect on the statistical significance of the differences
between groups.
Gene expression results
In present study, expression changes of 18 genes in
SCZ patients and 11 genes in PPD patients were
detected in mitochondrial complex I. Expression levels
of NDUFS1, NDUFS4, NDUFV1, NDUFV2, NDUFB2,
NDUFB5, NDUFB9, NDUFA5, NDUFA8, NDUFA13 and
NDUFC1 were up-regulated and expression levels of
NDUFAB1, NDUFB10, NDUFB11, NDUFA1, NDUFA2,
NDUFS7 and NDUFS8 were down-regulated in chronic
schizophrenic patients in comparison with non-psychi-
atric individuals. Also, expression levels of NDUFS1,
NDUFV1, NDUFV2, NDUFB5, NDUFB9, NDUFA13, NDUFA8
and NDUFA5 were up-regulated and NDUFB11, NDUFS7
and NDUFS8 were down-regulated in the PPD group
in comparison with the non-psychiatric group.
Significant over-expression of NDUFS1 was found in
paranoid schizophrenic (n ¼ 297) and PPD patients
(n ¼ 340) in comparison with other subtypes of
SCZ (total n ¼ 337). Gene expression results are pre-
sented in Table 2 and Figures 2 and 3, which show a
heat map of gene expression levels in group
comparisons.
8 A. HAGHIGHATFARD ET AL.

Figure 2. Volcano plot visualisation of gene expression level differences in SCZ versus non-psychiatric groups. From 44 genes, sig-
nificantly over-expressed genes (NDUFS1, NDUFS4, NDUFV1, NDUFV2, NDUFB2, NDUFB5, NDUFB9, NDUFA5, NDUFA8, NDUFA13
and NDUFC1) are illustrated in the top left of the plot, and significantly under-expressed genes (NDUFAB1, NDUFB10, NDUFB11,
NDUFA1, NDUFA2, NDUFS7 and NDUFS8) are illustrated on the top right of the plot. Genes with no significant alteration are illus-
trated under the horizontal line.

Clinical characteristics (P ¼ 0.0009), NDUFB9 (P ¼ 0.0008), NDUFV1 (P ¼ 0.0006)

A summary of all the different correlation data, including
correlation coefficient r values for each gene, is provided
in Supplementary Table 3 (sheet 1 for SCZ and sheet 2
for PPD). In addition, two heat maps presented in
Figures 4 and 5 illustrate all the correlations between
gene expression levels and demographic and clinical
data in SCZ and PPD, respectively. Because of the poten-
tial effect of age on gene mRNA levels, patients of a par-
ticular age range were selected. No significant correlation
between expression results of any genes with gender,
age of onset and antipsychotic dosages were observed
in the two patient groups. Duration of illness in the SCZ
group was positively correlated with over-expression of
NDUFS1 (P ¼ 0.0007) or NDUFV1 (P ¼ 0.0009), and under-
expression of NDUFA1 (P ¼ 0.0007). There was a direct
correlation between antipsychotic resistance in patients
(41 subjects listed as drug resistant in SCZ and six sub-
jects in PPD) and over-expression of NDUFS1 (P ¼ 0.0005)
and NDUFV1 (P ¼ 0.0008) and under-expression of
NDUFB11 (P ¼ 0.0004) and NDUFA1 (P ¼ 0.0003).
Psychiatric tests results
There was a significant positive correlation between
over-expression of NDUFS1 (P ¼ 0.0002), NDUFS4
and NDUFV2 (P ¼ 0.0009) and under-expression of
NDUFB11 (P ¼ 0.0004) and NDUFA1 (P ¼ 0.0006) with
total scores of PANSS and BPRS in schizophrenic
patients. In addition, a direct correlation was found
between positive symptom scores and NDUFS1
(P ¼ 0.0006) and NDUFB9 (P ¼ 0.0009) expression, and a
direct correlation between negative symptom scores
and NDUFB11 (P ¼ 0.001) and NDUFA1 (P ¼ 0.0007) and
under-expression in PANSS. In BPRS subscales, with-
drawal/Retardation and Anxious/Depression scores
were positively correlated to under-expression of
NDUFB11 (P ¼ 0.0008) and NDUFA1 (P ¼ 0.001), and
Hostile/Suspiciousness score was positively correlated
to over-expression of NDUFS1 (P ¼ 0.0002) and NDUFB9
(P ¼ 0.001). No significant correlation observed for
Thinking Disturbance score.
Neuropsychological tests results
A significant deficit in working memory, attention, cog-
nition and executive functions was detected in SCZ
and PPD groups in comparison with non-psychiatric
subjects. There was a significant negatively correlation
between over-expression of NDUFV2 (P ¼ 0.0004),
NDUFB9 (P ¼ 0.001) and NDUFS1 (P ¼ 0.0004) and
 THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 9

Figure 3. Volcano plot visualisation of gene expression level differences in PPD versus non-psychiatric groups. From 44 genes, sig-
nificantly over-expressed genes (NDUFS1, NDUFV1, NDUFV2, NDUFB5, NDUFB9, NDUFA13, NDUFA8 and NDUFA5) are illustrated on
the top left of the plot, and significantly under-expressed genes (NDUFB11, NDUFS7 and NDUFS8) are illustrated on the top right
of the plot. Genes with no significant alteration are illustrated under the horizontal line.

under-expression of NDUFA1 (P ¼ 0.001) and WCST cor-
rect responses in both patient groups.No significant
correlation was observed for the number of completed
groups. No significant correlation was observed for
total IQ score in WAIS-IV, but there were correlations
with digit and dot span subtests. Negative correlations
between under-expression of NDUFA1 (P ¼ 0.001),
NDUFB10 (P ¼ 0.001) and NDUFB11 (P ¼ 0.001) and digit
span forward scores and negative correlations
between under-expression of NDUFA1 (P ¼ 0.0005) and
NDUFB11 (P ¼ 0.001) and over-expression of NDUFS1
(P ¼ 0.0008) with digit span backward scores observed
in schizophrenic and PPD patients. No significant cor-
relation was observed between gene expressions and
dot span scores in patients. There was a significant
positive correlation between congruent stimuli score
and under-expression of NDUFA1 (P ¼ 0.001) and
NDUFB11 (P ¼ 0.0009) in the Stroop test in SCZ and
PPD. In the TMT test there was a negative correlation
between under-expression of NDUFB11 (P ¼ 0.001) and
TMT-A and TMT-B second results in SCZ and PPD
groups.
Neurological tests results
In the auditory oddball task and visual p300 we
observed a latency and significant reduction in
amplitude in SCZ and PPD cases in comparison with
non-psychiatric subjects. Also, a deficit in MMN ampli-
tude was revealed in SCZ patients. There was a correl-
ation between p300 visual latency in the task
performance and NDUFS1 (P ¼ 0.001) and NDUFB9
(P ¼ 0.0006) over-expression and NDUFAB11
(P ¼ 0.0008) under-expression in SCZ and PPD groups.
There was a significant correlation between MMN def-
icit (in standard tone, duration and frequency deviant)
with NDUFS1 (P ¼ 0.0009), NDUFV1 (P ¼ 0.001), NDUFV2
(P ¼ 0.001) and NDUFB9 (P ¼ 0.001) over-expression
and NDUFA1 (P ¼ 0.001) and NDUFB11 (P ¼ 0.0007)
under-expression observed in both patient groups. No
correlation was observed in the p300 auditory oddball
task abnormalities and gene expression.
Biochemical test results
A significant reduction of the three types of SOD (Cu/
Zn, Mn and Fe) was detected in SCZ and PPD groups
in comparison with non-psychiatric subjects. SOD
reduction in SCZ was reported in previous studies
(Potluri et al. 2009; Gonzalez-Liencres et al. 2014), but
this is the first report of SOD reduction in PPD
patients. A significant correlation between SOD reduc-
tion and positive symptoms in PANSS was revealed in
SCZ patients. There was a significant correlation
10 A. HAGHIGHATFARD ET AL.
Figure 4. Heat map visualisation of R values between gene expression levels versus demographic data and clinical tests scores in
schizophrenic patients. Black arrays show lower R values, while brighter arrays correspond to higher R values. Green represents an
indirect relation and red represents a direct relation.
between NDUFS1 (P ¼ 0.0003), NDUFV1 (P ¼ 0.0005),
NDUFV2 (P ¼ 0.001), NDUFB9 (P ¼ 0.001) and NDUFC1
(P ¼ 0.001) over-expression and reduction of SOD in
the two patient groups.
Discussion
Novel marker genes in SCZ, paranoid subtype and
PPD
Gene expression studies of mitochondrial complex I
help to clarify the pattern of bioenergetics effects in
behaviour and psychiatric disorders. Several studies
which considered the gene expression of the largest
subunits of mitochondrial complex I presented it as
potential peripheral biomarker for SCZ (Taurines et al.
2010; Akarsu et al. 2014). The present study examined
gene expression of whole subunits in complex I in two
psychiatric disorders for the first time, and detected a
pattern of gene expression alterations including up-
and down-regulations showing the importance of
small subunits as well. This is the first gene expression
study in PPD. Shared gene expression patterns of both
disorders support the hypothesis of a similarity in
pathophysiology of PPD and SCZ. Eighteen candidate
genes were detected in SCZ and 11 genes in PPD; all
of the candidate genes found in PPD were among the
18 differentially expressed genes found in SCZ and
their alteration directions were the same in both disor-
ders. These findings confirmed role of genomic core
subunits in SCZ. Down-regulation of NDUFS7 and
NDUFS8 along with up-regulation of NDUFS1, NDUFV1
and NDUFV2 were detected in both SCZ and PPD
patients. Deregulation of these five subunits from 14
core subunits may cause abnormal structured complex
I in neural cells, which leads to abnormal energetic
state and altered neural activity. None of the seven
mitochondrial genes showed a change in expression,
which may reduce the importance of a maternal pat-
tern in the heritance of SCZ and PPD. Over-expression
of several supernumerary subunits was detected in
SCZ and PPD patients. As oxidative stress is increased

Figure 5. Heat map visualisation of R values between gene expression levels versus demographic data and clinical tests scores in
PPD patients. Black arrays show lower R value, while brighter arrays correspond to higher R values. Green represents an indirect
relation and red represents a direct relation.
in psychiatric disorders, it may relate to the role of
supernumerary subunits in protection against oxidative
stress. Under-expression of five supernumerary subu-
nits was detected in SCZ. Under-expression of
NDUFB11 (which is only under-expressed in super-
numerary subunits in PPD patients) was reported in
brain tissue samples from SCZ patients (Huang et al.
2014). Also, down-regulation of the NDUFA1 gene was
reported in progressive neurodegenerative diseases,
including Alzheimer’s disease (Fernandez-Moreira et al.
2007).
The results reveal NDUFS1 as a specific marker for
the paranoid subtype of SCZ with significant over-
expression in comparison with other subtypes. These
results may introduce NDUFS1 as the first subtype-
specific biomarker in SCZ and revive the hypothesis
about the reclassification of SCZ and the separation of
the paranoid subtype from SCZ. Over-expression of
NDUFS1 in PPD patients alos supports this hypothesis.
Significant over-expression of NDUFS1 in PPD and the
paranoid subtype could reveal it as a candidate gene
THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 11
in the molecular mechanisms of paranoid behaviour.
Medication effects on gene expression are one of the
most important obstacles which make it hard to inter-
pret whether alterations of the mRNA levels of a gene
are related to the disease rather than the medications.
Also, complex I genes are not the direct target of anti-
psychotics, and the significant expression changes in
genes in this study are stronger than that could be
related to antipsychotic usage. In addition, there was
no relation between medication classes and gene
expression that would reduce the importance of medi-
cation in complex I gene expression.
Antipsychotic resistance was associated with gene
expression
The target pathways of antipsychotics are dopamin-
ergic and serotonergic. Previous studies showed that
mitochondrial function is modulated by dopamine
inhibition of respiratory complex I activity (Huang
et al. 2014). Therefore, it is not surprising that gene
12 A. HAGHIGHATFARD ET AL.
expression alternation in mitochondrial complexes
affects pharmacodynamic responses. Regardless of the
limited sample size, the correlation of NDUFS1,
NDUFV1, NDUFB11, and NDUFA1 mRNA levels with
antipsychotic resistance could be related to neurode-
generation of dopaminergic neurons due to the over-
activity of complex I.
Association of clinical symptoms with gene
expression changes in SCZ
The correlations between total score of BPRS and
PANSS and mRNA levels of genes indicated a role of
complex I in the behavioural pathology of SCZ.
Correlation of the positive symptoms score of PANSS,
and the Hostile/Suspiciousness score in BPRS, with
over-expression of NDUFS1 and NDUFB9 genes indi-
cates that these genes are strong markers for paranoid
thoughts, paraphrenia and delusional disorder. Also,
correlations between the negative symptoms score of
PANSS, the withdrawal/Retardation score and the
Anxious/Depression score of BPRS with the down-
regulation of NDUFB11 and NDUFA1 suggest these
genes as potential markers for reduced social engage-
ment, emotional expression and lack of motivation.
Alteration of the expression of complex I genes
induces neuropsychological changes
Previous twin studies mentioned executive functions
as most heritable psychological traits (Friedman et al.
2008). Correlation of executive functions deficit
(showed as a WCST result) and NDUFV2, NDUFS1,
NDUFB9 and NDUFA1 expression shows an influence
of complex I activity on the executive functions of SCZ
and PPD patients. They also revealed effects of com-
plex I activity in frontal lobe functions as the main
region of executive functions. There is a supported
hypothesis that working memory is a core deficit in
SCZ; and, at a component level, ROBO1 is the only
related gene for the phonological loop function of
working memory (Bates et al. 2011). Results have
shown correlations between NDUFA1, NDUFB10,
NDUFB11 and NDUFS1 expression changes and a lack
of function in working memory, especially in verbal
working memory. The association of memory deficit
and NDUFA1, NDUFB10 and NDUFB11 down-regula-
tion could be related to electron transport chain prob-
lems in the prefrontal cortex of SCZ and PPD patients.
This suggests the involvement of the energetic state
of neurons on memory situation and memory disor-
ders. Attention deficits are a prominent aspect of
cognitive dysfunction in SCZ (Carter et al. 1997).
The correlation of attention deficit with NDUFA1 and
NDUFB11 under-expression may suggest the potential
effect of complex I activity in symptoms of attention
disorders, such as SCZ or ADHD. The correlation of
TMT results with the under-expression of NDUFB11
could imply a role of this gene in cognition deficiency
in SCZ, PPD and other cognitive disorders like delirium,
dementia and amnesia.
EEG-ERP abnormalities are related to gene
expression changes
MMN, P300 visual and auditory abnormalities have
been reported in SCZ (Frantseva et al. 2014).
Neurological abnormalities could indicate that endo-
phenotypes are involved in psychiatric disorders.
Correlation of genetic variants or expression patterns
with neurological aberrants in SCZ and PPD could clar-
ify the role of the genes in neural functions as well as
in disorders. MMN and WSCT findings, and digit span
results, indicate a deficit in prefrontal function which is
related to gene expression changes, especially in
NDUFS1 and NDUFB9. P300 visual were associated
with the same gene expression pattern (NDUFS1,
NDUFB9 and NDUFAB11) as well, which implies the
pleiotropic role of complex I genes on a number of
neural functions at the same time (Del Re et al. 2014).
Correlation of complex I genes and ERP abnormalities
could support previous reports about behavioural
symptoms, abnormal neuronal transmission and synap-
tic plasticity in subjects with over-activated mitochon-
drial complexes (Del Re et al. 2014).
Association of SOD reduction with gene
expression and positive symptoms
SOD reduction is a marker for neurodegeneration in
several disorders including SCZ (Halliwell 1992;
Craddock et al. 2007). The correlation between SOD
reduction and over-expression of genes that were cor-
related with positive symptoms in PANSS and BPRS
strongly supports the behavioural effects of SOD
reduction, and could reveal the mechanism of these
effects in SCZ and PPD, which may relate to complex I
activity. The SOD reduction detected in PPD patients
could reveal more details about the pathophysiology
of the disorder. Mitochondrial complex deficiency is
considered to be a cause of neurodegeneration in SCZ
(Konradi et al. 2004; Washizuka et al. 2006; Clay et al.
2011; Park & Park 2012), and these results show that it
may be considered as a marker in PPD as well. Several
similar biological, neurological and psychological bases

of the two different psychiatric disorders, SCZ and
PPD, were determined in the present study.
Limitation and strengths of study
Peripheral blood transcript levels may not accurately
represent expression alterations in the brain, and a
number of studies reported tissue-specific expression
of mitochondrial complex I genes, which is the main
limitation of the present study (Wirtz et al. 2011).
However, as the genetic background of the subjects is
not clear, potential genetic confounding between
cases and controls may affect the results. The present
study also has several strengths. The large sample size
increases the validity of data analysis. In SCZ, this is
the first gene expression study of whole complex I
genes which showed a role of supernumerary subu-
nits, and newly assessed, core subunits in the aeti-
ology of SCZ, and completed the partial patterns that
had been reported in previous studies. This is the first
molecular genetics study on PPD patients that high-
lights the molecular mechanisms, especially in the bio-
energetic pathways, in the aetiology of this personality
disorder. The systematic clinical and biochemical
evaluation, which is a major strength of this study,
helps in the understanding of the comprehensive
effects of complex I deregulation in the psychiatric
and psychological situation in patients.
Conclusion
Our study aimed to uncover the whole pattern of
complex I activities in psychiatric disorders. The find-
ings of this study support the hypothesis of consider-
ing abnormalities in the energy state of neural cells as
a cause of psychiatric disorders such as SCZ and PPD.
Theses abnormalities in the energy state may be
because of the pattern of deregulations in mitochon-
drial complex I activity observed in this study. Up- and
down-regulation of mRNA levels of both core and
supernumerary subunits may lead to abnormal struc-
ture, assembly or functions in complex I. This abnor-
mal complex I may lead to neurodegeneration and a
wide range of psychiatric disorders. Several significant
correlations between demographic, psychiatric, psy-
chological and neurological results with mRNA levels,
highlights the importance of complex I and the bio-
energetics pathway in the aetiology of SCZ and PPD
as well as the mechanisms of brain functions like
memory and attention, and behaviours such as
depression and paranoid thoughts.
The present study was performed in 22 of the 31
provinces of Iran including North, North West, North
THE WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY 13
East and South West of the country. The study included
Iranian patients from the Persian majority and different
minority tribes such as Arabs, Kurds and Armenians.
Acknowledgements
We thank the Presidency of the Young Researchers and
Elites Club, Prof. Karim Zaree and former Vice-Presidency of
Research and Technology in Islamic Azad University, Science
and Research Branch, and Professor Mansor Bayat for all of
their support. This study is supported by the Young
Researchers and Elites Club under annual grants of special
talents members.
Disclosure statement
The authors report no conflicts of interest. The authors alone
are responsible for the content and writing of this article.
Funding
This study is supported by the Young Researchers and Elites
Club under annual grants of special talents members.
References
Akarsu S, Torun D, Bolu A, Erdem M, Kozan S, Ak M, Akar H,
Uzun O.€ 2014. Mitochondrial complex I and III gene mRNA
levels in schizophrenia, and their relationship with clinical
features. J Mol Psychiatry. 2:1.
American Psychiatric Association (APA). 1994. Diagnostic and
statistical manual of mental disorders (DSM). Washington
(DC): American Psychiatric Association; p. 143–147.
Bates TC, Luciano M, Medland SE, Montgomery GW, Wright MJ,
Martin NG. 2011. Genetic variance in a component of the lan-
guage acquisition device: ROBO1 polymorphisms associated
with phonological buffer deficits. Behav Genet. 41:50–57.
Ben-Shachar D, Karry R. 2007. Sp1 expression is disrupted in
schizophrenia; a possible mechanism for the abnormal
expression of mitochondrial complex I genes, NDUFV1
and NDUFV2. PLoS One. 2:e817.
Ben-Shachar D, Zuk R, Gazawi H, Reshef A, Sheinkman A,
Klein E. 1999. Increased mitochondrial complex I activity
in platelets of schizophrenic patients. Int J
Neuropsychopharmacol. 2:245–253.
Ben-Shachar D. 2002. Mitochondrial dysfunction in schizo-
phrenia: a possible linkage to dopamine. J Neurochem.
83:1241–1251.
Carter CS, Mintun M, Nichols T, Cohen JD. 1997. Anterior cin-
gulate gyrus dysfunction and selective attention deficits in
schizophrenia: [15O] H2O PET study during single-trial
Stroop task performance. Am J Psychiatry. 154:1670–1675.
Clay HB, Sillivan S, Konradi C. 2011. Mitochondrial dysfunc-
tion and pathology in bipolar disorder and schizophrenia.
Int J Develop Neurosci. 29:311–324.
Craddock RM, Lockstone HE, Rider DA, Wayland MT, Harris
LJ, McKenna PJ, Bahn S. 2007. Altered T-cell function in
schizophrenia: a cellular model to investigate molecular
disease mechanisms. PLoS One. 2:e692.
14 A. HAGHIGHATFARD ET AL.
Del Re E, Bergen S, Mesholam-Gately R, Niznikiewicz M,
Goldstein J, Woo T, Shenton ME, Seidman L, McCarley R,
Petryshen T. 2014. Analysis of schizophrenia-related genes
and electrophysiological measures reveals ZNF804A associ-
ation with amplitude of P300b elicited by novel sounds.
Translat Psychiatry. 4:e346.
Drake RJ, Lewis SW. 2003. Insight and neurocognition in
schizophrenia. Schizophre Res. 62:165–173.
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.
Fernandez-Moreira D, Ugalde C, Smeets R, Rodenburg RJ,
Lopez-Laso E, Ruiz-Falco ML, Briones P, Martin MA,
Smeitink JA, Arenas J. 2007. X-linked NDUFA1 gene muta-
tions associated with mitochondrial encephalomyopathy.
Ann Neurol. 61:73–83.
Frantseva M, Cui J, Farzan F, Chinta LV, Velazquez JLP,
Daskalakis ZJ. 2014. Disrupted cortical conductivity in
schizophrenia: TMS-EEG study. Cereb Cortex. 24:211–221.
Friedman NP, Miyake A, Young SE, DeFries JC, Corley RP,
Hewitt JK. 2008. Individual differences in executive func-
tions are almost entirely genetic in origin. J Exp Psychol.
137:201.
Gonzalez-Liencres C, Tas C, Brown EC, Erdin S, Onur E,
Cubukcoglu Z, Aydemir O, Esen-Danaci A, Br€ une M. 2014.
Oxidative stress in schizophrenia: a case-control study on
the effects on social cognition and neurocognition. BMC
Psychiatry. 14:268.
Halliwell B. 1992. Reactive oxygen species and the central
nervous system. J Neurochem. 59:1609–1623.
Hirst J. 2013. Mitochondrial complex I. Annu Rev Biochem.
82:551–575.
Huang KC, Yang KC, Lin H, Tsao TTH, Lee SA. 2014.
Transcriptome alterations of mitochondrial and coagula-
tion function in schizophrenia by cortical sequencing ana-
lysis. BMC Genom. 15:1.
Karry R, Klein E, Shachar DB. 2004. Mitochondrial complex I
subunits expression is altered in schizophrenia: a postmor-
tem study. Biol Psychiatry. 55:676–684.
Kaufman AS, Lichtenberger EO. 2005. Assessing adolescent
and adult intelligence. Hoboken (NJ): John Wiley & Sons.
Konradi C, Eaton M, MacDonald ML, Walsh J, Benes FM,
Heckers S. 2004. Molecular evidence for mitochondrial
dysfunction in bipolar disorder. Arch Gen Psychiatry.
61:300–308.
Mehler-Wex C, Duvigneau JC, Hartl RT, Ben-Shachar D,
Warnke A, Gerlach M. 2006. Increased mRNA levels of the
mitochondrial complex I 75-kDa subunit. Eur Child
Adolesc Psychiatry. 15:504–507.
Oribe N, Hirano Y, Kanba S, del Re E, Seidman L, Mesholam-
Gately R, Goldstein JM, Shenton M, Spencer KM, McCarley
RW. 2014. Progressive reduction of visual P300 amplitude
in patients with first-episode schizophrenia: an ERP study.
Schizophr Bull. 41:460–470.
Park C, Park SK. 2012. Molecular links between mitochon-
drial dysfunctions and schizophrenia. Mol Cells.
33:105–110.
Potluri P, Davila A, Ruiz-Pesini E, Mishmar D, O’Hearn S,
Hancock S, Simon M, Scheffler IE, Wallace DC, Procaccio V.
2009. A novel NDUFA1 mutation leads to a progressive
mitochondrial complex I-specific neurodegenerative dis-
ease. Mol Genet Metabol. 96:189–195.
Reichborn-Kjennerud T. 2010. The genetic epidemiology of
personality disorders. Dialogues Clin Neurosci.
12:103–114.
Sanders AR, Go €ring HH, Duan J, Drigalenko EI, Moy W, Freda
J, He D, Shi J, Gejman PV. 2013. Transcriptome study of
differential expression in schizophrenia. Hum Mol Genet.
22:5001–5014.
Silver H, Feldman P, Bilker W, Gur RC. 2003. Working memory
deficit as a core neuropsychological dysfunction in schizo-
phrenia. Am J Psychiatry. 160:1809–1816.
Taurines R, Thome J, Duvigneau JC, Forbes-Robertson S,
Yang L, Klampfl K, Romanos J, M€ uller S, Gerlach M,
Mehler-Wex C. 2010. Expression analyses of the mitochon-
drial complex I 75-kDa subunit in early onset schizophre-
nia and autism spectrum disorder: increased levels as a
potential biomarker for early onset schizophrenia. Eur
Child Adolesc Psychiatry. 19:441–448.
Ugalde C, Vogel R, Huijbens R, van den Heuvel B, Smeitink J,
Nijtmans L. 2004. Human mitochondrial complex I assem-
bles through the combination of evolutionary conserved
modules: a framework to interpret complex I deficiencies.
Hum Mol Genet. 13:2461–2472.
Ventura J, Nuechterlein KH, Subotnik KL, Gutkind D, Gilbert
EA. 2000. Symptom dimensions in recent-onset schizo-
phrenia and mania: a principal components analysis of the
24-item Brief Psychiatric Rating Scale. Psychiatry Res.
97:129–135.
Vinothkumar KR, Zhu J, Hirst J. 2014. Architecture of mam-
malian respiratory complex I. Nature. 515:80–84.
Waldinger RJ. 1997. Psychiatry for medical students.
Arlington (VA): American Psychiatric Pub.
Washizuka S, Kametani M, Sasaki T, Tochigi M, Umekage T,
Kohda K, Kato T. 2006. Association of mitochondrial com-
plex I subunit gene NDUFV2 at 18p11 with schizophrenia
in the Japanese population. Am J Med Genet Part B.
141:301–304.
Webb CT, Levinson DF. 1993. Schizotypal and paranoid per-
sonality disorder in the relatives of patients with schizo-
phrenia and affective disorders: a review. Schizophre Res.
11:81–92.
Wirtz S, Schuelke M. 2011. Region-specific expression of
mitochondrial complex I genes during murine brain devel-
opment. PLoS One. 6:e18897.