Association of β-Secretase Functional Polymorphism with Risk of Schizophrenia

Abstract

Aim:

The NRG1-ERBB4 neurotransmitter signaling pathway plays a key role in the pathogenesis of schizophrenia (SZ). The intronic single-nucleotide polymorphism rs707284 in ERBB4 links to PI3K-AKT suppression in SZ. Another protein indirectly affecting NRG1-ERBB4 signaling is β-secretase, which is encoded by the BACE1 gene, and activates NRG1 by proteolytic cleavage. In this study, we aimed to investigate the association of ERBB4 rs707284 and BACE1 rs490460 with the risk of SZ in an Iranian population.

Subjects and Methods:

A total of 973 subjects, including 480 SZ patients and 493 healthy controls, matched by ethnicity, age, and gender, were recruited in a case-control study. Genomic DNA was extracted from peripheral blood of all subjects and genotyping of rs707284 and rs490460 was performed using polymerase chain reaction (PCR)-restriction fragment length polymorphism and tetra-primer amplification refractory mutation system (tetra-ARMS) PCR genotyping assays, respectively.

Results:

A significant association was observed between the rs490460 T allele and SZ (p = 0.0002, odds ratio 0.69, 95% confidence interval 0.57-0.84). There was no association between the risk of SZ and rs707284.

Conclusion:

Our data indicate that rs490460 is associated with the risk of SZ. In silico analysis indicates that rs490460 may be a potential splicing site, which affects protein structure. Replication studies are needed to confirm our data.

Introduction

Schizophrenia (SZ) is a chronic psychotic disorder affecting ∼1-4% of the population worldwide (Perala et al., 2007). SZ is characterized by hallucinations, delusions, social withdrawal, and cognitive symptoms (Park et al., 2011). Extensive genome-wide association studies leading to the discovery of numerous susceptibility loci support the substantial polygenic basis of SZ. Meta-analysis studies have demonstrated that the rate of mortality in SZ is higher than the general population. In addition, a linear increase in the disease incidence has been shown in the recent decades (Bushe et al., 2010). The high rate of mortality and the heritability of over 80%, necessitate investigating the underlying genetic basis of the disease. However, a comprehensive network of disease pathogenesis has not been elucidated at the molecular level.

Postmortem genetic studies have stated the importance of the NRG1-ERBB4 signaling pathway in the pathogenesis of SZ (Marenco et al., 2011). This pathway is linked to neural migration, synaptic plasticity, and the regulation of neurotransmitter expression (Falls, 2003). Stimulation of the cytoplasmic isoform (CYT-1) of Erb-B2 receptor tyrosine kinase 4 (ERBB4) by neuregulin1 (NRG1) leads to the triggering of several cascades, including the PI3K-AKT pathway. The CYT-1 isoform is overexpressed in the prefrontal cortex of patients with SZ (Law et al., 2012). The common single-nucleotide polymorphism (SNP), rs707284, in the intronic region of ERBB4 links to increases in ERBB4 CYT-1 isoform expression and class 1 PI3K enzymes (i.e., the catalytic subunit delta) (PIK3CD) (Silberberg et al., 2006; Law et al., 2007). Recent studies have demonstrated that PI3K catalytic subunit delta could be a promising therapeutic target for SZ (Law et al., 2012). Association of rs707284 with SZ has been studied previously in Irish and Han Chinese populations. However, these data should be further confirmed in other ethnic groups by independent replication studies.

Another gene that indirectly impacts the NRG1-ERBB4 signaling pathway is beta-site APP cleaving enzyme 1 (BACE1), encoding β-secretase enzyme (Wang et al., 2013). This enzyme activates NRG1 by proteolytic processing. Animal studies have shown that BACE1-null mice exhibit SZ-like behavioral traits (Savonenko et al., 2008). Moreover, studies on Brodmann's six area of human brain have reported an impaired proteolytic processing of NRG1 in SZ patients (Barakat et al., 2010). We hypothesized that the expression of BACE1 could be affected by certain critical polymorphisms in this gene. Herein, we designed a case-control study to investigate the association of ERBB4 rs707284 and BACE1 rs490460 with SZ in an Iranian population.

Subjects and Methods

Study population

A total of 973 unrelated Iranian individuals, including 480 SZ patients (350 males) and 493 healthy controls (343 males), were recruited in a case-control study. The two groups had no significant differences in the distributions of age and sex (Table 1). All cases met the diagnostic criteria by at least two experienced psychiatrists based on Diagnostic and Statistical Manual of mental disorders (DSM-IV) criteria for SZ (American Psychiatric Association, 2013). Patients with a history of head trauma, thyroid disease, and any other serious past medical illnesses other than SZ were excluded from the study. Moreover, first-episode patients and the subjects with alcohol dependence or hallucinogenic drug abuse at the time of the sampling were excluded. Normal controls were enrolled among healthy blood donors from different medical centers. All control subjects and their first-degree families were free of any history of psychiatric disorders. All participants provided written informed consent for participation in the present study. This study was approved by the Ethics Committee at Shahid Beheshti University of Medical Sciences, Tehran, Iran.

Table 1.

Distribution of Age and Sex in Schizophrenia and Control Groups

Variable	Schizophrenia	Control	p
Age (mean ± SD)	43.34 ± 7.21	44.11 ± 8.84	0.13
Gender
Male	350	343
Female	130	150	0.25
Total	480	493

SNP selection and replication study

No association study has been conducted between BACE1 and SZ to date. We searched the Hapmap and 1000 genome databases and excluded SNPs with minor allele frequency less than 0.05 or unknown frequency. We also excluded SNPs with unknown potential effect on the transcript or polypeptide product. Using the online software, Human Splicing Finder version 3.0 (HSF), we sought intronic nucleotides surrounding the exons, which could be considered as a potential splice site, and selected rs490460 located at the conserved donor site of intron 5. This region has obtained a consensus value of 95.9 in a 0-100 scoring range, which could possibly affect splicing.

The association of rs707284 in ERBB4 has been previously studied in the populations of Caucasian and Ashkenazi Jewish ancestry. We performed a replication study of this SNP in an independent population.

SNP genotyping

Genomic DNA was extracted from peripheral blood from all subjects using the salting out method (Miller et al., 1988). The rs707284 SNP was genotyped using the polymerase chain reaction-restriction fragment length polymorphism (PCR) method. PCRs were performed in a total volume of 19 μL PCR mixture. PCR products were digested by TaaI enzyme (Thermo Fisher Scientific) at 65°C for 16 h. The digested products were separated on 2% agarose gel stained by Gel Red stain (Takapoozist, Iran) and observed under UV light. The rs490460 SNP was genotyped through a tetra-primer amplification refractory mutation system (tetra-ARMS) PCR method. PCR was performed in a total volume of 21 μL PCR. PCR products were separated on 1% agarose gel and observed under UV light (Table 2).

Table 2.

The Primer Sequences and Polymerase Chain Reaction Cycles for rs490460 and rs707284

		PCR conditions (°C)
Polymorphism	Primer sequences (5→3)	Denature	Annealing	Extension	Alleles	PCR products
rs490460	F-O-TATAAGACCTTGCCACCCCAAGAAATCA	96/1 min	58/1 min	72/40 s	T	404 + 274
	R-O-TTTGTGACCATTAAGCCCCTGACTGTTC				G	404 + 185
	F-I-GATTTGAAAATGGACTGCAAGGAGGTGAG
	R-I-ATATTCCCAGGCTCTCCCTTGGTTCTA
rs707284	F-AGCTCAGGGTTATTCCACCA	72/30 s	66/30 s	96/30 s	A	237
	R-CAACCTGCACTTGTACTCCTGA				G	137 + 100

PCR, polymerase chain reaction.

The accuracy of the genotyping methods was confirmed by sequencing five samples of each genotype for all three SNPs.

Statistical analysis

Pearson's χ²-test was applied to test for significance in genotype and allele frequency differences between the two groups. Odds ratio with 95% confidence interval was estimated and a p-value of <0.05 (two-tailed) was considered to be significant. The Hardy-Weinberg equilibrium test was performed using Fisher's exact test. The distribution of genotype frequencies was analyzed under three different genetic models (additive, recessive, and dominant) using SNPassoc package of R, version 3.2.3 (Gonzalez et al., 2007). All other data were analyzed using R, version 3.2.3. Bonferroni correction was used to adjust the p-value for multiple testing, with a significance threshold of p < 0.016.

Results

A significant association was observed between rs490460 and SZ (Table 3). The T allele of this variant was significantly more prevalent in SZ patients compared with normal subjects (p = 0.0002). The association of rs490460 with SZ was significant under all three additive, dominant, and recessive models, with the strongest association under the additive model (Table 3). The association was still significant after correction for multiple testing. There was no association between rs707284 and risk of SZ in our study under any genetic models.

Table 3.

Analyses of Genotype and Allele Frequencies in Schizophrenia and Control Group in Studied Single-Nucleotide Polymorphisms

			Genotype frequencies (%)				Allele frequencies (%)
SNP	Gene	Subjects	AA	AG	GG	p	A	G	p	OR	95% CI
rs707284	ERBB4	Case	104 (22)	204 (42)	172 (36)	0.10	421 (43)	548 (57)	0.23	1.11	0.93-1.33
		Control	83 (17)	236 (48)	174 (35)		402 (41)	584 (59)

			GG	GT	TT		G	T
rs490460	BACE1	Case	26 (5)	201 (43)	253 (52)	0.0006^a	253 (26)	707 (74)	0.0002^a	0.69	0.57-0.84
		Control	48 (9)	240 (49)	205 (42)		336 (34)	650 (66)

SNP	Gene	Genetic models	p	OR	95% CI
rs707284	ERBB4	log-Additive (G/G = 0, A/G = 1 and A/A = 2)	0.35	1.09	0.91-1.29
		Dominant (A/A and A/G vs. G/G)	0.86	0.98	0.75-1.27
		Recessive (G/G and A/G vs. A/A)	0.056	1.37	0.99-1.88
rs490460	BACE1	log-Additive (T/T = 0, G/T = 1 and G/G = 2)	0.0001^a	0.67	0.55-0.82
		Dominant (G/G and G/T vs. T/T)	0.0005^a	0.64	0.50-0.82
		Recessive (T/T and G/T vs. G/G)	0.010^a	0.53	0.32-0.87

Considered as significant.

CI, confidence interval; OR, odds ratio; SNP, single-nucleotide polymorphism.

Discussion

In the present study, we report novel association between rs490460 and SZ. The minor G-allele of this SNP is located at the fifth position of the GTRAGT consensus splicing donor site, at the beginning of intron 5, and is conserved in vertebrates. This indicates that despite being the minor allele, the G nucleotide is necessary for normal splicing and substitution of G by T may affect this process (Human Splicing Finder, version 3.0) (Desmet et al., 2009).

BACE1 affects NRG1-ERBB4 signaling by proteolytic processing of NRG1 and generating active ligand for ERBB4 tyrosine kinase receptor. BACE1 has its highest expression in the spinal cord and brain, including cerebral cortex and frontal lobe. BACE1-null mice develop SZ-like behavioral cognitive deficits responding to atypical antipsychotic drugs (Savonenko et al., 2008). NRG1 C-terminal fragment is decreased in the same area of the brain, implicating a disruption in normal cleavage of NRG1 (Barakat et al., 2010). However, BACE1 protein levels were not altered in the Brodmann's area 6 and 9 of the cortex of SZ patients in another study (Dean et al., 2008). The latter inconsistency may be due to the fact that the protein product level may not correlate with transcript splice variant abnormalities. Functional studies are required to determine whether the NRG1-ERBB4 signaling pathway is abnormal in patients with rs490460 as a downstream cascade of BACE1.

The ERBB4 rs707284 was selected to replicate the prior association studies (Table 4). Similar to the findings of Nicodemus et al. (2006) in the Caucasian and African American races, we did not detect a positive association between rs707284 and SZ in the Iranian population. Previously, a Caucasian study reported a modest association between rs707284 and SZ (P = 0.01) (Silberberg et al., 2006), indicating that different ethnic backgrounds are subject to allelic heterogeneity. ERBB4 signaling mediated by NRG1 triggers multiple cascades, including the PI3K signaling pathway. The risk polymorphisms, rs707284, rs7598440, and rs839523 (GGA), are linked to increased ERBB4 expression, and subsequently increased levels of PI3K catalytic subunit delta, leading to suppression of PI3K signaling with an unknown mechanism. A recent study by Law et al. (2012) identified PIK3CD, the catalytic subunit of PI3K, as a promising therapeutic target for IC87114 in SZ. IC87114 blocks PIK3CD and triggers the signaling in a normal way (Rico, 2012). The presence of these polymorphisms within ERBB4 may affect PIK3CD response to the drug.

Table 4.

Comparison of Previous Results of Association of the rs707284 and Schizophrenia with the Results of the Present Study

	Frequency of G allele
Study	Cases	Controls	p
Silberberg et al.	0.66	0.52	0.01
Nicodemus et al.	0.66	0.66	<0.05
Chen et al.	0.37	0.39	<0.05
Present study	0.57	0.59	<0.05

In summary, we report novel association between BACE1 and SZ and suggest that this SNP is studied in other populations of larger sample size. Furthermore, we recommend functional analysis of the splice variant transcripts of BACE1 in SZ patients.

Limitations

This work had some limitations, including the relatively small sample size, lack of replication in an independent population, and lack of functional studies. Investigating the expression of the studied genes may elucidate their role in SZ pathogenesis.

Footnotes

Acknowledgments

This article has been written based on Simin Rahimi-Aliabadis' MSc thesis findings, which were entirely supported by Shahid Behshti University of Medical Sciences, Tehran, Iran. The authors would like to appreciate all subjects who participated in this study. The authors are sincerely grateful to Saray-e Ehsan Institute staff for cooperation in patient sampling and Dr. Morteza Karimi for his efforts in clinical diagnosis and giving scientific advice.

Author Disclosure Statement

No competing financial interests exist.

References

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders (DSM-5^®). Washington, DC: American Psychiatric Association.

Barakat

, Dean

, Scarr

, Evin

(2010) Decreased Neuregulin 1 C-terminal fragment in Brodmann's area 6 of patients with schizophrenia. Schizophr Res, 124:200-207.

Bushe

, Taylor

, Haukka

(2010) Mortality in schizophrenia: a measurable clinical endpoint. J Psychopharmacol, 24(4 Suppl):17-25.

Dean

, Soulby

, Evin

, Scarr

(2008) Levels of [(3)H]pirenzepine binding in Brodmann's area 6 from subjects with schizophrenia is not associated with changes in the transcription factor SP1 or BACE1. Schizophr Res, 106:229-236.

Desmet

, Hamroun

, Lalande

, et al. (2009) Human Splicing Finder: an online bioinformatics tool to predict splicing signals. Nucleic Acids Res, 37:e67.

Falls

(2003) Neuregulins: functions, forms, and signaling strategies. Exp Cell Res, 284:14-30.

Gonzalez

, Armengol

, Sole

, et al. (2007) SNPassoc: an R package to perform whole genome association studies. Bioinformatics, 23:644-645.

Law

, Kleinman

, Weinberger

, Weickert

(2007) Disease-associated intronic variants in the ErbB4 gene are related to altered ErbB4 splice-variant expression in the brain in schizophrenia. Hum Mol Genet, 16:129-141.

Law

, Wang

, Sei

, et al. (2012) Neuregulin 1-ErbB4-PI3K signaling in schizophrenia and phosphoinositide 3-kinase-p110delta inhibition as a potential therapeutic strategy. Proc Natl Acad Sci U S A, 109:12165-12170.

10.

Marenco

, Geramita

, van der Veen

, et al. (2011) Genetic association of ErbB4 and human cortical GABA levels in vivo. J Neurosci, 31:11628-11632.

11.

Miller

, Dykes

, Polesky

(1988) A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 16:1215.

12.

Nicodemus

, Luna

, Vakkalanka

, et al. (2006) Further evidence for association between ErbB4 and schizophrenia and influence on cognitive intermediate phenotypes in healthy controls. Mol Psychiatry, 11:1062-1065.

13.

Park

, Park

, Kim

, et al. (2011) Association of polymorphism in the promoter of the melatonin receptor 1A gene with schizophrenia and with insomnia symptoms in schizophrenia patients. J Mol Neurosci, 45:304-308.

14.

Perala

, Suvisaari

, Saarni

, et al. (2007) Lifetime prevalence of psychotic and bipolar I disorders in a general population. Arch Gen Psychiatry, 64:19-28.

15.

Rico

(2012) Finding a druggable target for schizophrenia. Proc Natl Acad Sci U S A, 109:11902-11903.

16.

Savonenko

, Melnikova

, Laird

, et al. (2008) Alteration of BACE1-dependent NRG1/ErbB4 signaling and schizophrenia-like phenotypes in BACE1-null mice. Proc Natl Acad Sci U S A, 105:5585-5590.

17.

Silberberg

, Darvasi

, Pinkas-Kramarski

, Navon

(2006) The involvement of ErbB4 with schizophrenia: association and expression studies. Am J Med Genet B Neuropsychiatr Genet, 141B:142-148.

18.

Wang

, Li

, Shen

(2013) Beta-Secretase: its biology as a therapeutic target in diseases. Trends Pharmacol Sci, 34:215-225.