Molecular Psychiatry (2003) 8, 706–709

& 2003 Nature Publishing Group All rights reserved 1359-4184/03 $25.00
www.nature.com/mp

ORIGINAL RESEARCH ARTICLE

Association study of neuregulin 1 gene with schizophrenia

JZ Yang*,1, TM Si*,1, Y Ruan1, YS Ling1, YH Han1, XL Wang1, M Zhou1, HY Zhang1, QM Kong1, C Liu1,
DR Zhang1, YQ Yu2, SZ Liu2, GZ Ju2, L Shu1, DL Ma3 and D Zhang1
1
Institute of Mental Health, Peking University, Beijing 100083, China; 2The Center for Genomic Medicine, University of Jilin,
Changchun 130021, China; 3Peking University Center for Human Disease Genomics, Beijing 100083, China

A number of studies have indicated that 8p22–p12 is likely to harbor schizophrenia

susceptibility loci. In this region, the candidate gene of interest, neuregulin 1 (NRG1), may
play a role in the pathogenesis of schizophrenia. Then in the present study, we performed the
linkage disequilibrium to determine the association between three genetic variants (SNPs:
rs3924999, rs2954041, SNP8NRG221533) on NRG1 gene and schizophrenia in 246 Chinese Han
schizophrenic family trios using PCR-based restriction fragment length polymorphism method
and denaturing high-performance liquid chromatography. The transmission disequilibrium
test analysis for each variant showed a significant difference between two transmitted alleles
even after Bonferroni correction (rs3924999, P ¼ 0.007752; rs2954041, P ¼ 0.0009309;
SNP8NRG221533, P ¼ 0.012606). The global v2 test for haplotype transmission also revealed
a strong association (v2 ¼ 46.068, df ¼ 7, Po0.000001). Our results suggest that the NRG1 gene
may play a role in conferring susceptibility to the disease.
Molecular Psychiatry (2003) 8, 706–709. doi:10.1038/sj.mp.4001377
Keywords: schizophrenia; single nucleotide polymorphism (SNP); neuregulin 1(NRG1); the
transmission disequilibrium test (TDT); haplotype

Schizophrenia is a complex disease, a common G38A), a G to A base change, occurs in position 12

mental disorder with a prevalence of 0.5–1% in the
general population. It is clinically characterized by
disturbed thought processes, delusions, hallucina-
tions, and/or reduced social skills.1 A number of
studies indicate a genetic component contributing to
schizophrenia, but the mode of inheritance does not
follow a simple Mendelian pattern. In spite of the
failure to identify a schizophrenia gene of major
import in multiply affected families, more than one
data set from genome-wide scans provides convincing
evidence that 8p22–p21 may harbor candidate genes
which may confer susceptibility to schizophrenia to
an individual.2–5 Interestingly, two studies carried out
by Stefansson et al6,7 reported that the gene coding for
neuregulin 1 (NRG1) was located in 8p21; its position
as well as function strongly supported NRG1 gene as a
susceptibility gene for schizophrenia. In their study,
however, they had not found a clear pathogenic
mutation. Several dozens of single nucleotide poly-
morphisms (SNPs) have been identified within the
locus (http://www.ncbi.nlm.nih.gov/SNP/), of which
we randomly selected two that can be detected by the
restriction fragment length polymorphism (RFLP)
analysis. Of the two SNPs, one SNP (rs3924999,
Correspondence: Dr Zhang, MD, PhD, Institute of Mental Health,
Peking University, Beijing 100083, China.
E-mail: daizhang@bjmu.edu.cn
*Contributed equally to this work.
Received 22 January 2003; revised 19 April 2003; accepted 22
April 2003
within the second exon of the NRG1 gene, and the
database (http://www.ncbi.nlm.nih.gov/SNP/) shows
that the SNP changes the amino acid from arginine
(Arg) to glutamine (Gln) (Arg 38 Gln). Another SNP
(rs2954041), located in fifth intron, may still be
informative, for example, by being in linkage dis-
equilibrium with a causal locus for schizophrenia.
Therefore, it is valuable to investigate the possibility
that these polymorphism sites on the NRG1 locus may
be associated with schizophrenia. We also genotyped
the single most significant SNP (SNP8NRG221533)
described in Stefansson’s findings6,7 by denaturing
high-performance liquid chromatography (dHPLC) in
order to clarify the similarities or differences of
SNP8NRG221533 polymorphism in the different
populations.
In the present study, we conducted association
analysis to determine the relationship between NRG1
gene and schizophrenia. As the transmission dis-
equilibrium test (TDT), which evaluates allele trans-
missions from heterozygous parents to affected
individuals, provides an index of association which
is unbiased by population structure. Therefore, we
performed a family-based test in the present study.
We analyzed the three polymorphisms of NRG1 gene
in 246 Chinese Han schizophrenic family trios by the
PCR-based RFLP and dHPLC. The genotype distribu-
tions and allelic frequencies are presented in Table 1.
The TDT analysis for each locus showed a significant
difference between two transmitted alleles (Table 2)
even after Bonferroni correction (rs3924999, P ¼
0.007752; rs2954041, P ¼ 0.0009309; SNP8NRG221533,
 Neuregulin 1 gene and schizophrenia
JZ Yang et al

707
Table 1 Genotype distributions and allelic frequencies of the SNPs

Genotype distributions Allele frequencies

Rs3924999 GG GA AA G A
Patients 26 126 94 178(0.36) 314(0.64)
Parents 33 243 216 309(0.31) 675(0.69)
Total 59 369 310 487(0.33) 989 (0.67)
Rs2954041 TT TG GG T G
Patients 51 131 64 233(0.47) 259(0.53)
Parents 64 262 166 390(0.40) 594(0.60)
Total 115 393 230 623(0.42) 853(0.58)
SNP8NRG221533 CC CT TT C T
Patients 113 102 31 328(0.67) 164(0.33)
Parents 205 205 82 615(0.625) 369(0.375)
Total 318 307 113 943(0.64) 533(0.36)

Table 2 The TDT for each SNP

SNPs Allele Transmitted Nontransmitted w2 P-value

rs3924999 G 145 98 9.0905 0.002584

A 98 145
rs2954041 T 169 93 22.0458 0.0003103
G 93 169
SNP8NRG221533 C 123 82 8.2 0.004202
T 82 123

P ¼ 0.012606). As haplotype has more accuracy and
statistical power than individual SNPs in linkage-
disequilibrium-based association study, we estimated
haplotype frequencies of the three SNPs with
TRANSMIT software.8 The global w2 test for haplotype
transmission also revealed a strong association be-
tween the NRG1 locus and schizophrenia
(w2 ¼ 46.068, df ¼ 7, Po0.000001). As shown in Table
3, the 1-df test for individual haplotypes demon-
strated that an excess of the TAG haplotype was
transmitted by the parent to the affected offspring
(w2 ¼ 17.134, df ¼ 1, Po0.000035).
NRG1 is one family member of structurally related
glycoproteins that include NRG2, NRG3, and NRG4.9
NRG1 plays a role as a multifunctional factor in the
neurodevelopment and many aspects such as cell
differentiation. In mouse embryos of 14.5 days, Orr-
Urtreger et al10 found that NRG1 expression was
confined predominantly to the central and peripheral
nervous systems, including the neuroepithelium

Table 3 Estimated haplotype frequencies and w2 test of three SNPs of NRG1gene

Estimated haplotype frequencies Observed Expected w2 P-value

Global w2 test 46.068 0.000000

CGT 45.093 32.663 12.11 0.000505
TGT 29.315 19.852 13.601 0.000228
CAT 105.84 87.046 10.882 0.000976
TAT 52.749 55.439 0.36266 0.547054
CGG 71.081 63.94 1.998 0.157492
TGG 32.511 38.045 2.0686 0.150345
CAG 105.98 123.85 7.7595 0.005358
TAG 49.425 71.165 17.134 0.000035

Note: CGT represents SNP8NRG221533 C+ rs3924999 G+ rs2954041 T; TGT represents SNP8NRG221533 T+ rs3924999 G+
rs2954041 T; CAT represents SNP8NRG221533 C+ rs3924999 A+ rs2954041 T; TAT represents SNP8NRG221533 T+rs3924999
A+ rs2954041 T; CGG represents SNP8NRG221533 C+ rs3924999 G+rs2954041 G; TGG represents SNP8NRG221533 T+
rs3924999 G+rs2954041 G; CAG represents SNP8NRG221533 C+ rs3924999 A+ rs2954041 G; TAG represents
SNP8NRG221533 T+ rs3924999 A+ rs2954041 G.

Molecular Psychiatry
 Neuregulin 1 gene and schizophrenia
JZ Yang et al
708
lining the lateral ventricles of the brain, the ventral
horn of the spinal cord, and the intestinal as well as
dorsal root ganglia. Through a ribozyme-based strat-
egy for the perturbation of NRG1 function in devel-
oping chick embryos, Zhao and Lemke11 discovered
that NRG1 could promote both muscle cell differ-
entiation in the heart and neuronal differentiation in
the central nervous system. Recently, Huang et al12
reported that neuregulins and their receptors, the
ERBB protein tyrosine kinases, are essential for
neuronal development and involved in long-term
potentiation in the hippocampal CA1 region without
affecting basal synaptic transmission, the brain
mechanisms considered important for memory for-
mation, and they also considered that the signaling
role of NRG in the central nervous system of adults
may be involved in the modulation of synaptic
plasticity.
In Stefansson et al’s study,6,7 they found that a core
at-risk haplotype represented a block of linkage
disequilibrium containing the 50 exon of NRG1
and upstream; however, the only SNP
(SNP8NRG433E1006) in the exon of NRG1 was not
found in significant excess in patients. In the present
study, we conducted an association analysis to
examine the relationship between the NRG1 locus
and schizophrenia in a Chinese population. Both the
TDT and the haplotype analysis showed a strong
association between the NRG1 locus and schizophre-
nia. In our sample, the SNP8NRG221533, which gave
the best single marker association in the study of
Stefansson et al showed significant evidence,
although its significance was less statistically than
the other two polymorphisms. These results not only
confirmed the finding reported by Stefansson et al,
but also extended the linked haplotype block from the
50 exon of NRG1 to the fifth intron in Chinese Han
people. In the haplotype block, the SNP rs3924999 is
located in the second exon of the NRG1 gene
(nucleotide sequence site 1200844), which is con-
servative in all isoforms of NRG1, and the database
(http://www.ncbi.nlm.nih.gov/SNP/) shows that this
SNP is a functional polymorphism (Arg 38 Gln).
However, to examine whether the SNP is a clear
Table 4 Details of PCR primers, length of PCR products, Tm, restriction enzyme used, and restriction fragments created for each
allele
Primer name Primer sequence 50 -30
SNP1 ACTGGTTTCACACCGAAGGAC
Forward
SNP1 CCAAGATGAGATCCATTTTCGC
Reversed
SNP2 TGACATTATTCATTGTTTGTTGCT
Forward GA
SNP2 GGATGCCATGGATATACTATGCA
Reversed GA
Note: SNP1: rs3924999; SNP2: rs2954041.
Molecular Psychiatry
pathogenic mutation in the protein level would be
required. The other SNP rs2954041 also showed the
linkage disequilibrium. Located in the fifth intron, the
result suggested that this site might be linked with
functional loci, either SNP rs3924999 or other loci.
Thus, we could not exclude the other linkage
polymorphic sites in NRG1 gene such as enhancers
or other controlling sites.
In summary, our results, along with Stefansson’s
findings, suggested that the NRG1 gene might either
play a role in predisposing an individual to schizo-
phrenia or be in linkage disequilibrium with a causal
locus for schizophrenia. Therefore, it is necessary to
carry out further investigation to confirm these
findings by replicating studies as well as genotyping
haplotypes within the NRG1 and adjacent loci.
Materials and methods
Subjects
Schizophrenic patients and their biological parents
were recruited for the present study at the Institute of
Mental Health, Peking University, China. All the
subjects were Chinese of Han descent. All patients
met the ICD-10 criteria. Of the patients, 138 (56%)
were male and 108 (44%) female, mean age 29 years.
The mean duration of illness was 5 years. All subjects
gave informed consent for genetic analysis. The study
was approved by the Ethics Committee of Peking
University.
Peripheral blood samples were obtained from
subjects and genomic DNA was extracted using the
phenol–chloroform method.
Genotyping
PCR-RFLP Two SNPs (rs3924999, rs2954041) were
genotyped by the PCR-based RFLP analysis for all
subjects. Primers and restriction enzymes used are
described in Table 4. PCR amplification was
performed in 25-ml reaction volume containing
10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2,
0.001% (w/v) gelatin, 200 mM of each dNTP, 0.4 mM of
each primer, 1.0 U of Taq DNA polymerase, and 40 ng
PCR (bp) Annealing Restriction Common Rare allele
temperature enzyme allele RFLP RFLP products (bp)
(1C) products (bp)
246 55 MunI A G
65,181 246
187 62 Tru1I G T
60,127 30,31,126

genomic DNA. The conditions used for PCR
amplification included an initial denaturation at
941C for 5 min, followed by 35 cycles at 941C for 30
s, 55–621C for 40 s, 721C for 1 min, and a final
elongation at 721C for 10 min. A 15-ml aliquot of the
PCR product mixtures was completely digested with
5 U of restriction enzyme and then separated on 4%
(SNP1) and 5% (SNP2) agarose gels.
dHPLC The DNA sequence containing the SNP8
NRG221533 was amplified with a pair of primers:
50 GCA TTA GAA CTA GAA CTT GCG TGA 30 and 50
TGG GAA CTC TCC ATC TCT TTC A 30 . Before
dHPLC analysis, PCR products were denatured by
heating to 941C for 1 min, and were then allowed to
reanneal by cooling to 501C over 25 min.
dHPLC. was performed using a WAVE DNA frag-
ment analyzer (Transgenomic). Column temperature
for dHPLC analysis was determined using software
that is available free of charge over the Internet
(http://insertion.standford.edu/melt.html).13 To en-
sure maximum sensitivity, in addition to the tem-
perature suggested by the software (n1C), each
fragment was also run at n721C. Where dHPLC
analysis suggested that a subject was heterozygous,
the PCR products from that individual were se-
quenced to determine the nature of the polymorphism
using the Big Dye Terminator Cycle Sequencing Kit
(Perkin-Elmer Applied Biosystems) as described by
the manufacturer.
Statistical analysis
The TDT14 was applied to analyze allelic association
in family trios consisting of father, mother, and
affected offspring with schizophrenia, where prefer-
ential allelic transmission from heterozygous parents
to affected offspring is tested by applying the (bc)2/
(b þ c) statistics and the w2 (McNemar’s test). The null
hypothesis of no association indicates that each allele
carried by a heterozygous parent has a 50% chance for
transmission to an affected child. If the allele plays a
causal role in the development of the disorder, then
its transmission should exceed 50%. Haplotype
frequency was estimated by TRANSMIT program,
version 2.5.2.8 We applied Bonferroni corrections for
all multiple statistical tests.
Acknowledgements
We express our gratitude to all patients and their
family members, without whose cooperation, this
work would not have been possible. We also thank
Professor Lian Zhang, Professor Kaifeng Pang in
Beijing Institute for Cancer Research in helping us
Neuregulin 1 gene and schizophrenia
JZ Yang et al
709
to carry out dHPLC. This work has been supported by
Grant H010210180112 from Beijing Biomedical R & D
Innovation Program, Grant 30000059 from the Na-
tional Natural Scientific Foundation of China, Grant
2000-A-A32 from Peking University Center for Hu-
man Disease Genomics, and Grant 2002BA711A07-06
from National Key Project.
References
1 Andreasen NC. Symptoms, signs, and diagnosis of schizophrenia.
Lancet 1995; 346: 477–481.
2 Kendler KS, MacLean CJ, O’Neill FA, Burke J, Murphy B, Duke F
et al. Evidence for a schizophrenia vulnerability locus on
chromosome 8p in the Irish Study of High-Density Schizophrenia
Families. Am J Psychiatry 1996; 153: 1534–1540.
3 Pulver AE, Lasseter VK, Kasch L, Wolyniec P, Nestadt G, Blouin JL
et al. Schizophrenia: a genome scan targets chromosomes 3p and
8p as potential sites of susceptibility genes. Am J Med Genet 1995;
60: 252–260.
4 Schizophrenia Linkage Group. Additional support for schizophre-
nia linkage on chromosomes 6 and 8: a multicenter study.
Schizophrenia Linkage Collaborative Group for chromosomes 3,
6 and 8. Am J Med Genet 1996; 67: 580–594.
5 Pulver AE, Mulle J, Swartz KL, Blouin J-L, Dombroski B, Liang K-Y
et al. Genetic heterogeneity in schizophrenia: stratification of
genome scan data using co-segregating related phenotypes. Mol
Psychiatry 2000; 5: 650–653.
6 Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S,
Sigmundsson T, Ghosh S et al. Neuregulin 1 and susceptibility to
schizophrenia. Am J Hum Genet 2002; 23,71: 877–892.
7 Stefansson H, Sarginson J, Kong A, Yates P, Steinthorsdottir V,
Gudfinnsson E et al. Association of neuregulin 1 with schizo-
phrenia confirmed in a Scottish population. Am J Hum Genet
2003; 72: 83–87.
8 Chiano MN, Clayton DG. Fine genetic mapping using haplotype
analysis and the missing data problem. Ann Hum Genet 1998; 62:
55––60.
9 Wolpowitz D, Mason TBA, Dietrich P, Mendelsohn M, Talmage
DA, Role LW. Cysteine-rich domain isoforms of the neure-gulin-1
gene are required for maintenance of peripheral synapses. Neuron
2000; 25: 79–91.
10 Orr-Urtreger A, Trakhtenbrot L, Ben-Levy R, Wen D, Rechavi G,
Lonai P et al. Neural expression and chromosomal mapping of
NEU differentiation factor to 8p12-p21. Proc Natl Acad Sci USA
1993; 90: 1867–1871.
11 Zhao JJ, Lemke G. Selective disruption of neuregulin-1 function in
vertebrate embryos using ribozyme–tRNA transgenes. Develop-
ment 1998; 125: 1899–1907.
12 Huang YZ, Won S, Ali DW, Wang Q, Tanowitz M, Du QS et al.
Regulation of neuregulin signaling by PSD-95 interacting with
ErbB4 at CNS synapses. Neuron 2000; 26: 443–455.
13 Jones A, Austin J, Hansen N, Hoogendoorn B, Oefner PJ, Cheadle
JP et al. Optimal temperature selection for mutation detection by
denaturing HPLC and comparison to single-stranded conformation
polymorphism and heteroduplex analysis. Clin Chem 1999; 45:
1133––1140.
14 Spielman RS, McGinnis RE, Ewens WJ. Transmission tests for
linkage disequilibrium: the insulin gene region and insulin-
dependent diabetes mellitus (IDDM). Am J Hum Genet 1993; 52:
506–516.
Molecular Psychiatry