We have identified idiopathic carbonyl stress in a subpopulation of schizophrenic patients. We first identified a patient with a mutation in GLO1 (glyoxalase I) who showed increased AGE (advanced glycation end-product) levels and decreased vitamin B6 levels. By applying the observations from this rare case to the general schizophrenic population, we were able to identify a subset of patients (20%) for whom carbonyl stress may represent a causative pathophysiological process. Genetic defects in GLO1 increase the risk of carbonyl stress 5-fold, and the resulting increased AGE levels correlate significantly with PANSS (Positive and Negative Syndrome Scale) scored negative symptoms. Pyridoxamine, an active form of vitamin B6 and scavenger for carbonyl stress, could represent a novel and efficacious therapeutic agent for these treatment-resistant symptoms. In the present article, we describe a unique research approach to identify the causative process in the pathophysiology of a subset of schizophrenia. Our findings could form the basis of a schizophrenia subtype classification within this very heterogeneous disease and ultimately lead to better targeted therapy.

Introduction

Schizophrenia is a debilitating and complex mental disorder with a worldwide prevalence of approximately 1%. Despite extensive research, the full pathophysiology of this disease remains unclear [1,2]. Current therapies focus on treating symptoms using anti-psychotic agents which are unfortunately highly prone to inducing significant side effects. This has led to intensive research into the identification of new disease mechanisms, with the aim of developing novel classes of therapeutic agents, which are both efficacious and free of debilitating adverse effects. There is a strong body of evidence from biochemical and pharmacological studies, using human samples and animal models, which suggests that oxidative or carbonyl stress contributes to the pathophysiology of schizophrenia [36]. Oxidative stress is a central mediator of AGE (advanced glycation end-product) formation, and pyridoxamine, an active form of vitamin B6, detoxifies RCOs (reactive carbonyl compounds) via carbonyl-amine chemistry. The cellular removal of AGEs is largely dependent upon activity of the zinc metalloenzyme Glo1 (glyoxalase I) [7]. The glyoxalase detoxification system is ubiquitous in human tissues, including the brain. This cascade interacts with other metabolizing cascades, including putative gene products involved in the aetiology of schizophrenia, such as glutathione, homocysteine and folic acid metabolites [815].

Studies have revealed that Glo1 dysfunction is involved not only in systemic diseases such as diabetes mellitus [16] and vascular injury [17], but also in neuropsychiatric illnesses, such as mood disorders [18], autism [19,20], anxiety disorders [21] and alcoholism [22]. In mice, altered levels of Glo1 expression are associated with anxiety-like behavioural phenotypes [2325]. Interestingly, GLO1 maps to chromosome 6p21, a linkage region for schizophrenia [2628]. Lending further support to this association is the report of a missense Glu111/Ala111 polymorphism in two multiplex Caucasian pedigrees with schizophrenia spectrum disorders [29].

The present article focuses on idiopathic carbonyl stress in a subpopulation of schizophrenic patients with no history of diabetes mellitus or renal dysfunction. In the patients studied who showed high plasma AGE levels, a proportion of them also harboured a genetic defect in the GLO1 gene, suggesting that, in these schizophrenics, carbonyl stress may be the result of dysfunctional Glo1 detoxification.

GLO1 mutation associated with a schizophrenia pedigree

In this section, we discuss an interesting report from our laboratory, of a schizophrenic with a deficiency in Glo1 [30]. The case was a 60-year-old male who suffered from severe schizophrenia. Within his family were two affected brothers, one of whom committed suicide and the other who was long-term hospitalized due to his illness. He also had two maternal uncles with schizophrenia. Both the case and his brother exhibited treatment-resistant disease, despite intensive therapy. DNA from the case showed a novel mutation, consisting of an adenine insertion at nucleotide 79 in exon 1, introducing a frameshift at codon 27 and a premature termination codon, after aberrant translation of 15 amino acid residues (T27NfsX15) [30]. In addition, expression of Glo1 mRNA and protein was decreased by 50% in the lymphocytes of our patient (Figures 1a and 1b). In the red blood cells of our case, enzymatic activity levels of Glo1 were approximately half that of wild-type controls (Figure 1c). We postulated that this loss-of-function mutation accompanied by AGE accumulation would be shared by affected individuals in this family, although we were unable to access the genetic material to test our theory.

GLO1 and a frameshift mutation

Figure 1
GLO1 and a frameshift mutation

(a) mRNA expression. (b) Results of Western blotting. (c) Enzymatic activity of Glo1. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RBC, red blood cell.

Figure 1
GLO1 and a frameshift mutation

(a) mRNA expression. (b) Results of Western blotting. (c) Enzymatic activity of Glo1. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RBC, red blood cell.

Mutations of GLO1 and increased levels of plasma AGEs in schizophrenics

In a new study, we resequenced the GLO1 gene from 1761 schizophrenics and 1921 control subjects, and identified another novel frameshift mutation [30]. Individuals carrying frameshift mutations also showed a 40–50% reduction in Glo1 activity from red blood cells, a result that supports the finding that genetic mutations in the GLO1 gene typically occur in one allele. As we predicted, schizophrenic carriers of the GLO1 frameshift mutations displayed significantly increased levels of plasma AGEs. Since these patients had no history of diabetes mellitus or renal failure, it is most likely that these increased levels of plasma AGEs are due to reduced Glo1 function. In addition, patients with elevated AGE levels showed a clear concomitant decrease in vitamin B6 levels. In the two GLO1 frameshift mutation carriers, we found markedly reduced vitamin B6 levels. This is probably a reflection of increased vitamin B6 consumption in the detoxification of α-oxoaldehydes, by-products of carbonyl stress. Additionally, in this study, we detected a single nucleotide polymorphism which resulted in an amino acid residue change from the wild-type Glu111 to alanine. Schizophrenics homozygous for the Ala111 variant displayed a 16% reduction in enzymatic activity, accompanied by a significant increase in plasma AGEs [30]. This reduced Glo1 enzymatic activity in Ala111 variants was determined by comparing GFP-fused Glo1 constructs carrying Ala111 and Glu111, using in vitro assays. Interestingly, Glo1 variations were not confined to patients. We found non-psychiatric controls carrying frameshift mutations or homozygous Ala111 alleles, all of whom exhibited normal plasma AGEs and vitamin B6 levels, in contrast with mutation-carrying schizophrenics. This is all highly suggestive of active compensatory mechanisms in non-schizophrenics. In addition, plasma glutathione and zinc levels were significantly higher in non-psychiatric individuals carrying the mutation compared with schizophrenics with a frameshift mutation (glutathione, P=0.03; zinc, P=0.002). These results imply that, in non-psychotic subjects, additional environmental factors compensate for defects in Glo1 activity, possibly involving zinc and glutathione, as both are necessary for Glo1-mediated detoxification of carbonyl compounds.

In support of our hypothesis for idiopathic carbonyl stress being a causative state in schizophrenia, we expanded our study from a small population carrying genetic defects in GLO1, to 178 schizophrenia and 76 control subjects, with no evidence of diabetes mellitus or renal dysfunction, and measured plasma AGEs and vitamin B6 levels. We found significantly higher concentrations of AGEs in patients compared with controls (P<0.0001), whereas vitamin B6 levels were significantly lower in schizophrenics relative to control subjects (P<0.0001) (Figure 2). If we defined high circulating AGEs as plasma levels greater than the mean±2S.D. of controls, 66 of the patients (37.5%) and three of the controls (3.9%) showed high circulating plasma AGEs (distributions were significantly different between schizophrenia and control subjects: χ2=30.05, P<0.00001, odds ratio=14.6, 95% confidence interval=4.42–48.21). For vitamin B6 levels, 39.9% of schizophrenics showed low levels of plasma vitamin B6 compared with 7.4% of control subjects (χ2=20.9, P<0.0001, odds ratio=5.99, 95% confidence interval=2.59–13.83). Low levels were described as being under 6 ng/ml in males and 4 ng/ml in females. As these patients had received anti-psychotic therapy previously, we could not rule out an effect of medication on the induction of carbonyl stress in this cohort. To address this issue, we examined a drug-naive patient with prodromal-phase schizophrenia, who exhibited enhanced carbonyl stress with high plasma pentosidine levels [31]. Plasma pentosidine levels are typically used as a marker for carbonyl stress. Biochemical analysis demonstrated no abnormalities indicative of disease, such as diabetes mellitus or chronic kidney diseases. Crucially, this case suggests that the idiopathic carbonyl stress seen in schizophrenics may not be a result of anti-psychotic medication.

Plasma AGE and vitamin B6 levels in schizophrenics and controls

Figure 2
Plasma AGE and vitamin B6 levels in schizophrenics and controls

(a) AGE levels. (b) Vitamin B6 levels.

Figure 2
Plasma AGE and vitamin B6 levels in schizophrenics and controls

(a) AGE levels. (b) Vitamin B6 levels.

In order to examine the genetic contribution of GLO1 to carbonyl stress in schizophrenia, we tested the correlation between genotypes and plasma AGE levels, by performing χ2 tests (Table 1). The frameshift mutations and homozygous Ala111 variants are significantly associated with high carbonyl stress (χ2=7.72, P=0.005, odds ratio=5.63, 95% confidence interval=1.46–21.64).

Table 1
Association between the GLO1 genotype and carbonyl stress

χ2=7.727; degrees of freedom=1; P=0.0054; odds ratio=5.632; 95% confidence interval=1.466–2164.

 AGEs 
GLO1 genotype High Normal 
Frameshift, Ala/Ala 9 (4.5%) 3 (1.5%) 
Glu/Ala, Glu/Glu 57 (28.5%) 107 (53.5%) 
 AGEs 
GLO1 genotype High Normal 
Frameshift, Ala/Ala 9 (4.5%) 3 (1.5%) 
Glu/Ala, Glu/Glu 57 (28.5%) 107 (53.5%) 

Schizophrenia has long been considered a heterogeneous syndrome, with Kraepelin establishing the conceptual disease category as dementia precox, at the end of the 19th Century. Although current studies suggest that carbonyl stress is not applicable to the pathophysiology of all schizophrenia, approximately 37% of schizophrenics studied showed increased AGEs, suggesting a link between stress and subtypes of the disease. Twin and adoption studies led the way in defining schizophrenia as a disease with a large heritable component. On 23 December 2011, the SchizGene database listed 1727 studies identifying 8788 polymorphisms from 1008 genes thought to be associated with schizophrenia (http://www.szgene.org/). Yet, no causal gene with a large effect has been identified, despite this huge genetic research effort. Indeed, many studies conflict with each other. This lack of consensus is unsurprising considering the known heterogeneity of the disease. Recent genome-wide association studies have been expanding in sample size, in some cases to over 10000 case and control samples [3236], yet only genetic variations with a small effect size have been identified. It is plausible that, in schizophrenia, expanding cohort sizes dilutes the aetiological effect of multiple individual variations. In our studies, we were able to identify a relatively small subgroup of schizophrenics for which idiopathic carbonyl stress is a probable causative factor, from the heterogeneous schizophrenia population. This was achieved by applying our findings from a prototypic case carrying a frameshift mutation of GLO1 and showing extremely high plasma AGEs, along with markedly low serum vitamin B6 levels, to general patients. Within our hypothesis, elevated plasma AGEs and concomitant low vitamin B6 levels could represent the most cogent and easily measurable ‘biomarkers’ in schizophrenia, and ultimately form the basis for the classification of heterogeneous types of schizophrenia based on biological causes.

As an individual marker, depleted vitamin B6 might reflect elevated carbonyl stress induced by Glo1 defects and other as yet unknown cascades in schizophrenics. If carbonyl stress induces the development of schizophrenia and its symptoms, it is logical to assume that agents able to inhibit AGE formation or entrap carbonyl compounds could provide therapeutic benefit to this subset of patients. There are currently a number of AGE-inhibitory compounds available for clinical use, such as angiotensin receptor blockers. In addition, compounds such as pyridoxamine and TM2002 possess potent ability to entrap toxic carbonyl compounds and prevent toxicity. In particular, schizophrenic patients with markedly lowered vitamin B6 and high pentosidine levels may well benefit from treatment with pyridoxamine, a non-toxic water-soluble form of vitamin B6. As mentioned above, non-psychotic subjects carrying a frameshift mutation in GLO1 showed normal plasma AGEs and vitamin B6 levels. They also exhibited significantly higher plasma glutathione and zinc concentrations compared with schizophrenics harbouring the mutation. This finding highlights both of these substrates as natural and safe therapeutic agents for patients with idiopathic carbonyl stress.

The symptoms of schizophrenia are usually classified into one of two categories: positive symptoms which represent a change in behaviour or thoughts, typified by hallucinations or delusions, and negative symptoms which represent a withdrawal or absence of the functions usually expected in healthy subjects. These patients often appear emotionless, flat and apathetic. Most anti-psychotic medications are effective for positive, but not negative, symptoms. We assessed the symptom severity of schizophrenics showing idiopathic carbonyl stress, using PANSS (Positive and Negative Syndrome Scale) [37]. We compared PANSS with plasma levels of AGEs. Although positive symptoms showed no correlation with AGEs, negative symptoms correlated significantly with AGE levels. It is therefore highly plausible that therapeutic inhibition of AGEs may improve negative symptoms. As a suppressor of idiopathic carbonyl stress, pyridoxamine could represent a very attractive medication for improving negative symptoms in this subset of schizophrenics.

Conclusion

Idiopathic carbonyl stress, typified by elevated AGE levels, occurs in a subpopulation of schizophrenics who show no evidence of diabetes mellitus and renal dysfunction. In the present review, we have described findings which suggest that Glo1 deficits and carbonyl stress are linked to the development of a subset of schizophrenia, and show a correlation with the severity of negative symptoms. Elevated plasma pentosidine and concomitant lowered vitamin B6 levels may provide the most cogent and easily measurable ‘biomarkers’ in schizophrenia. They could also form the basis for classifying heterogeneous types of schizophrenia, according to their biological causes.

Glyoxalase Centennial: 100 Years of Glyoxalase Research and Emergence of Dicarbonyl Stress: A Biochemical Society Focused Meeting held at the University of Warwick, U.K., 27–29 November 2013. Organized and Edited by Naila Rabbani and Paul Thornalley (University of Warwick, U.K.).

Abbreviations

     
  • AGE

    advanced glycation end-product

  •  
  • Glo1

    glyoxalase I

  •  
  • PANSS

    Positive and Negative Syndrome Scale

Funding

Our work is supported by Japan Society for the Promotion of Science [KAKENHI grant number 20249054].

References

References
1
Sullivan
P.F.
Kendler
K.S.
Neale
M.C.
Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies
Arch. Gen. Psychiatry
2003
, vol. 
60
 (pg. 
1187
-
1192
)
2
Sullivan
P.F.
The genetics of schizophrenia
PLoS Med.
2005
, vol. 
2
 pg. 
e212
 
3
Schulz
J.B.
Lindenau
J.
Seyfried
J.
Dichgans
J.
Glutathione, oxidative stress and neurodegeneration
Eur. J. Biochem.
2000
, vol. 
267
 (pg. 
4904
-
4911
)
4
Tosic
M.
Ott
J.
Barral
S.
Bovet
P.
Deppen
P.
Gheorghita
F.
Matthey
M.L.
Parnas
J.
Preisig
M.
Saraga
M.
, et al. 
Schizophrenia and oxidative stress: glutamate cysteine ligase modifier as a susceptibility gene
Am. J. Hum. Genet.
2006
, vol. 
79
 (pg. 
586
-
592
)
5
Young
J.
McKinney
S.B.
Ross
B.M.
Wahle
K.W.
Boyle
S.P.
Biomarkers of oxidative stress in schizophrenic and control subjects
Prostaglandins Leukot. Essent. Fatty Acids
2007
, vol. 
76
 (pg. 
73
-
85
)
6
Ng
F.
Berk
M.
Dean
O.
Bush
A.I.
Oxidative stress in psychiatric disorders: evidence base and therapeutic implications
Int. J. Neuropsychopharmacol.
2008
, vol. 
11
 (pg. 
851
-
876
)
7
Thornalley
P.J.
The glyoxalase system in health and disease
Mol. Aspects Med.
1993
, vol. 
14
 (pg. 
287
-
371
)
8
Brown
A.S.
Bottiglieri
T.
Schaefer
C.A.
Quesenberry
C.P.
Jr
Liu
L.
Bresnahan
M.
Susser
E.S.
Elevated prenatal homocysteine levels as a risk factor for schizophrenia
Arch. Gen. Psychiatry
2007
, vol. 
64
 (pg. 
31
-
39
)
9
Frankenburg
F.R.
The role of one-carbon metabolism in schizophrenia and depression
Harv. Rev. Psychiatry
2007
, vol. 
15
 (pg. 
146
-
160
)
10
Gilbody
S.
Lewis
S.
Lightfoot
T.
Methylenetetrahydrofolate reductase (MTHFR) genetic polymorphisms and psychiatric disorders: a HuGE review
Am. J. Epidemiol.
2007
, vol. 
165
 (pg. 
1
-
13
)
11
Gysin
R.
Kraftsik
R.
Sandell
J.
Bovet
P.
Chappuis
C.
Conus
P.
Deppen
P.
Preisig
M.
Ruiz
V.
Steullet
P.
, et al. 
Impaired glutathione synthesis in schizophrenia: convergent genetic and functional evidence
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
16621
-
16626
)
12
Haidemenos
A.
Kontis
D.
Gazi
A.
Kallai
E.
Allin
M.
Lucia
B.
Plasma homocysteine, folate and B12 in chronic schizophrenia
Prog. Neuropsychopharmacol. Biol. Psychiatry
2007
, vol. 
31
 (pg. 
1289
-
1296
)
13
Levine
J.
Stahl
Z.
Sela
B.A.
Ruderman
V.
Shumaico
O.
Babushkin
I.
Osher
Y.
Bersudsky
Y.
Belmaker
R.H.
Homocysteine-reducing strategies improve symptoms in chronic schizophrenic patients with hyperhomocysteinemia
Biol. Psychiatry
2006
, vol. 
60
 (pg. 
265
-
269
)
14
Saadat
M.
Mobayen
F.
Farrashbandi
H.
Genetic polymorphism of glutathione S-transferase T1: a candidate genetic modifier of individual susceptibility to schizophrenia
Psychiatry Res.
2007
, vol. 
153
 (pg. 
87
-
91
)
15
Yao
J.K.
Leonard
S.
Reddy
R.
Altered glutathione redox state in schizophrenia
Dis. Markers
2006
, vol. 
22
 (pg. 
83
-
93
)
16
Kirk
R.L.
Theophilus
J.
Whitehouse
S.
Court
J.
Zimmet
P.
Genetic susceptibility to diabetes mellitus: the distribution of properdin factor B (Bf) and glyoxalase (GLO) phenotypes
Diabetes
1979
, vol. 
28
 (pg. 
949
-
951
)
17
Miyata
T.
van Ypersele de Strihou
C.
Imasawa
T.
Yoshino
A.
Ueda
Y.
Ogura
H.
Kominami
K.
Onogi
H.
Inagi
R.
Nangaku
M.
Kurokawa
K.
Glyoxalase I deficiency is associated with an unusual level of advanced glycation end products in a hemodialysis patient
Kidney Int.
2001
, vol. 
60
 (pg. 
2351
-
2359
)
18
Fujimoto
M.
Uchida
S.
Watanuki
T.
Wakabayashi
Y.
Otsuki
K.
Matsubara
T.
Suetsugi
M.
Funato
H.
Watanabe
Y.
Reduced expression of glyoxalase-1 mRNA in mood disorder patients
Neurosci. Lett.
2008
, vol. 
438
 (pg. 
196
-
199
)
19
Junaid
M.A.
Kowal
D.
Barua
M.
Pullarkat
P.S.
Sklower Brooks
S.
Pullarkat
R.K.
Proteomic studies identified a single nucleotide polymorphism in glyoxalase I as autism susceptibility factor
Am. J. Med. Genet. A
2004
, vol. 
131
 (pg. 
11
-
17
)
20
Sacco
R.
Papaleo
V.
Hager
J.
Rousseau
F.
Moessner
R.
Militerni
R.
Bravaccio
C.
Trillo
S.
Schneider
C.
Melmed
R.
, et al. 
Case-control and family-based association studies of candidate genes in autistic disorder and its endophenotypes: TPH2 and GLO1
BMC Med. Genet.
2007
, vol. 
8
 pg. 
11
 
21
Politi
P.
Minoretti
P.
Falcone
C.
Martinelli
V.
Emanuele
E.
Association analysis of the functional Ala111Glu polymorphism of the glyoxalase I gene in panic disorder
Neurosci. Lett.
2006
, vol. 
396
 (pg. 
163
-
166
)
22
Ledig
M.
Doffoel
M.
Ziessel
M.
Kopp
P.
Charrault
A.
Tongio
M.M.
Mayer
S.
Bockel
R.
Mandel
P.
Frequencies of glyoxalase I phenotypes as biological markers in chronic alcoholism
Alcohol
1986
, vol. 
3
 (pg. 
11
-
14
)
23
Ditzen
C.
Jastorff
A.M.
Kessler
M.S.
Bunck
M.
Teplytska
L.
Erhardt
A.
Krömer
S.A.
Varadarajulu
J.
Targosz
B.S.
Sayan-Ayata
E.F.
, et al. 
Protein biomarkers in a mouse model of extremes in trait anxiety
Mol. Cell. Proteomics
2006
, vol. 
5
 (pg. 
1914
-
1920
)
24
Hovatta
I.
Tennant
R.S.
Helton
R.
Marr
R.A.
Singer
O.
Redwine
J.M.
Ellison
J.A.
Schadt
E.E.
Verma
I.M.
Lockhart
D.J.
Barlow
C.
Glyoxalase 1 and glutathione reductase 1 regulate anxiety in mice
Nature
2005
, vol. 
438
 (pg. 
662
-
666
)
25
Krömer
S.A.
Kessler
M.S.
Milfay
D.
Birg
I.N.
Bunck
M.
Czibere
L.
Panhuysen
M.
Pütz
B.
Deussing
J.M.
Holsboer
F.
, et al. 
Identification of glyoxalase-I as a protein marker in a mouse model of extremes in trait anxiety
J. Neurosci.
2005
, vol. 
25
 (pg. 
4375
-
4384
)
26
Arolt
V.
Lencer
R.
Nolte
A.
Müller-Myhsok
B.
Purmann
S.
Schürmann
M.
Leutelt
J.
Pinnow
M.
Schwinger
E.
Eye tracking dysfunction is a putative phenotypic susceptibility marker of schizophrenia and maps to a locus on chromosome 6p in families with multiple occurrence of the disease
Am. J. Med. Genet.
1996
, vol. 
67
 (pg. 
564
-
579
)
27
Brzustowicz
L.M.
Honer
W.G.
Chow
E.W.
Hogan
J.
Hodgkinson
K.
Bassett
A.S.
Use of a quantitative trait to map a locus associated with severity of positive symptoms in familial schizophrenia to chromosome 6p
Am. J. Hum. Genet.
1997
, vol. 
61
 (pg. 
1388
-
1396
)
28
Nurnberger
J.I.
Jr
Foroud
T.
Chromosome 6 workshop report
Am. J. Med. Genet.
1999
, vol. 
88
 (pg. 
233
-
238
)
29
Turner
W.J.
Genetic markers for schizotaxia
Biol. Psychiatry
1979
, vol. 
14
 (pg. 
177
-
206
)
30
Arai
M.
Yuzawa
H.
Nohara
I.
Ohnishi
T.
Obata
N.
Iwayama
Y.
Haga
S.
Toyota
T.
Ujike
H.
Arai
M.
, et al. 
Enhanced carbonyl stress in a subpopulation of schizophrenia
Arch. Gen. Psychiatry
2010
, vol. 
67
 (pg. 
589
-
597
)
31
Arai
M.
Koike
S.
Oshima
N.
Takizawa
R.
Araki
T.
Miyashita
M.
Nishida
A.
Miyata
T.
Kasai
K.
Itokawa
M.
Idiopathic carbonyl stress in a drug-naive case of at-risk mental state
Psychiatry Clin. Neurosci.
2011
, vol. 
65
 (pg. 
606
-
607
)
32
Steinberg
S.
de Jong
S.
Mattheisen
M.
Costas
J.
Demontis
D.
Jamain
S.
Pietiläinen
O.P.
Lin
K.
Papiol
S.
Huttenlocher
J.
, et al. 
Common variant at 16p11.2 conferring risk of psychosis
Mol. Psychiatry
2014
, vol. 
19
 (pg. 
108
-
114
)
33
Shi
J.
Levinson
D.F.
Duan
J.
Sanders
A.R.
Zheng
Y.
Pe'er
I.
Dudbridge
F.
Holmans
P.A.
Whittemore
A.S.
Mowry
B.J.
, et al. 
Common variants on chromosome 6p22.1 are associated with schizophrenia
Nature
2009
, vol. 
460
 (pg. 
753
-
757
)
34
Stefansson
H.
Ophoff
R.A.
Steinberg
S.
Andreassen
O.A.
Cichon
S.
Rujescu
D.
Werge
T.
Pietiläinen
O.P.
Mors
O.
Mortensen
P.B.
, et al. 
Common variants conferring risk of schizophrenia
Nature
2009
, vol. 
460
 (pg. 
744
-
747
)
35
Stefansson
H.
Rujescu
D.
Cichon
S.
Pietiläinen
O.P.
Ingason
A.
Steinberg
S.
Fossdal
R.
Sigurdsson
E.
Sigmundsson
T.
Buizer-Voskamp
J.E.
, et al. 
Large recurrent microdeletions associated with schizophrenia
Nature
2008
, vol. 
455
 (pg. 
232
-
236
)
36
Manolio
T.A.
Rodriguez
L.L.
Brooks
L.
Abecasis
G.
Ballinger
D.
Daly
M.
Donnelly
P.
Faraone
S.V.
Frazer
K.
Gabriel
S.
, et al. 
New models of collaboration in genome-wide association studies: the Genetic Association Information Network
Nat. Genet.
2007
, vol. 
39
 (pg. 
1045
-
1051
)
37
Kay
S.R.
Positive-negative symptom assessment in schizophrenia: psychometric issues and scale comparison
Psychiatr. Q.
1990
, vol. 
61
 (pg. 
163
-
178
)