Next Article in Journal
Pre-Treatment Neutrophil-to-Lymphocyte and Platelet-to-Lymphocyte Ratios as Predictors of Occult Cervical Metastasis in Clinically Negative Neck Supraglottic and Glottic Cancer
Previous Article in Journal
The SGLT-2 Inhibitors in Personalized Therapy of Diabetes Mellitus Patients
 
 
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evidence for the Association between the Intronic Haplotypes of Ionotropic Glutamate Receptors and First-Episode Schizophrenia

1
Institute of Biology and Medical Genetics, First Faculty of Medicine, Charles University, 12800 Prague, Czech Republic
2
Institute of Physiology, Czech Academy of Sciences, 14220 Prague, Czech Republic
3
Institute of Biotechnology, Czech Academy of Sciences, BIOCEV, 25250 Vestec, Czech Republic
4
Faculty of Science, Charles University, 12800 Prague, Czech Republic
5
Genomics Core Facility, EMBL, 69117 Heidelberg, Germany
6
The National Institute of Mental Health, 25067 Klecany, Czech Republic
7
Institute of Physiology, Czech Academy of Sciences, BIOCEV, 25250 Vestec, Czech Republic
*
Author to whom correspondence should be addressed.
These authors contributed equally to the work.
J. Pers. Med. 2021, 11(12), 1250; https://doi.org/10.3390/jpm11121250
Received: 31 August 2021 / Revised: 14 November 2021 / Accepted: 16 November 2021 / Published: 25 November 2021

Abstract

:
The heritable component of schizophrenia (SCH) as a polygenic trait is represented by numerous variants from a heterogeneous group of genes each contributing a relatively small effect. Various SNPs have already been found and analyzed in genes encoding the NMDAR subunits. However, less is known about genetic variations of genes encoding the AMPA and kainate receptor subunits. We analyzed sixteen iGluR genes in full length to determine the sequence variability of iGluR genes. Our aim was to describe the rate of genetic variability, its distribution, and the co-occurrence of variants and to identify new candidate risk variants or haplotypes. The cumulative effect of genetic risk was then estimated using a simple scoring model. GRIN2A-B, GRIN3A-B, and GRIK4 genes showed significantly increased genetic variation in SCH patients. The fixation index statistic revealed eight intronic haplotypes and an additional four intronic SNPs within the sequences of iGluR genes associated with SCH (p < 0.05). The haplotypes were used in the proposed simple scoring model and moreover as a test for genetic predisposition to schizophrenia. The positive likelihood ratio for the scoring model test reached 7.11. We also observed 41 protein-altering variants (38 missense variants, four frameshifts, and one nonsense variant) that were not significantly associated with SCH. Our data suggest that some intronic regulatory regions of iGluR genes and their common variability are among the components from which the genetic predisposition to SCH is composed.

1. Introduction

Schizophrenia is a severe neuropsychiatric disorder estimated to affect up to 0.5% of the population worldwide [1,2]. The strong genetic component of SCH has been demonstrated in studies of twins or affected families [3]. The disorder is characterized by positive symptoms, negative symptoms, and cognitive deficits [4,5]. Several lines of evidence, such as altered mRNA and protein expression or de novo mutations in protein-coding regions of iGluR genes [6,7,8], have supported the hypothesis of dysfunction of glutamate-dependent excitatory signaling in the development of SCH.
Excitatory signaling in the central nervous system is mediated by ionotropic glutamate receptors (iGluR), the ion channels which are divided pharmacologically into three subclasses: AMPA, kainate (KA), and NMDA receptors [9,10,11]. The iGluR family consists of sixteen genes encoding four AMPAR (GRIA1-4), five KAR (GRIK1-5), and seven NMDAR subunits (GRIN1, 2A-D, 3A-B) [12]. AMPA and KA receptors form homo- or heterotetramers [13,14]. In contrast, NMDAR is strictly a heterotetramer composed of two obligatory GluN1 subunits and two GluN2 or GluN3 subunits in different combinations [15,16]. The subunit composition determines the specific functional properties of each iGluR including its intracellular localization and trafficking to the synapse [9]. The AMPA and NMDA receptors located on the postsynaptic membrane are involved in excitatory signal transduction upon glutamate release into the synaptic cleft. The physiological function of AMPA and NMDA receptors is fundamental for learning, cognition, memory formation processes, and behavior [17,18]. The behavioral and cognitive defects are manifested in various neurodevelopmental disorders [19].
Previous pharmacological studies have shown that NMDAR antagonists can induce psychotic symptoms typical of SCH in healthy individuals and significantly exacerbate these symptoms in patients [20,21]. Therefore, NMDAR hypofunction is thought to play a role in the etiology of SCH. Furthermore, in genetically modified mice, altered expression of the GluN1 subunit to 5% of its physiological level resulted in hypofunctional glutamatergic signaling. GluN1-deficient mice exhibited behavioral deficits that mimicked pharmacologically induced SCH symptoms [22]. In addition, several genetic variations have been identified in the coding and noncoding regions of NMDAR genes. However, their association with SCH was low or limited to a specific group of patients. Most SNPs in NMDAR exons are synonymous and rare mutations [23,24,25,26,27,28,29,30,31]. Other variations have also been identified in the untranslated and promoter regions of NMDAR genes where they affect gene expression as determined by the in vitro luciferase assay [32,33]. In the last decade, genome-wide association studies underlined the polygenic nature of SCH with multiple risk variants involved [34,35,36]. Consistent with previous single gene studies, a number of identified genetic variants has been found in genes that form the signaling complex of NMDAR, including the postsynaptic complex PSD-95 or Arc gene [37,38,39,40]. A global study identified 108 gene loci associated with SCH, including loci harboring genes for ionotropic GluA1 and GluN2A. This work confirmed many previous studies indicating genetic alterations in genes involved in excitatory glutamatergic signaling [34].
The exonic mutations in iGluR genes previously found in SCH were rare or de novo and could contribute only to specific cases. However, an increased number of small-risk alleles in regulatory and intronic parts of iGluR genes, if accumulated, could be responsible for hypofunction of the glutamatergic system in patients.
To evaluate this possibility in this case–control study, we used the gene-targeted sequencing strategy and resequenced all sixteen iGluR genes. In addition to exonic mutations, we focused on searching for variations in noncoding parts of these genes to predict their possible effects on gene expression and the extent of their association with SCH and its brain morphological phenotype (gray matter volume). To test the cumulative effect of small-risk alleles in our cohort of SCH cases, we proposed a simple scoring model.

2. Materials and Methods

2.1. SCH Patients/Control Subjects

In this case–control study, sixty-three patients with first-episode schizophrenia-spectrum disorders (FES) (aged 24.6 ± 6.9 years) were diagnosed on the basis of the structured MINI International Neuropsychiatric Interview [41] and according to the ICD-10 criteria by two independent psychiatrists. FES subjects were diagnosed with schizophrenia and detected through their first hospitalization at the Bohnice psychiatric hospital with the catchment area of 1 million inhabitants living in Prague and the northern part of Central Bohemia. The diagnostic procedures were performed at the Prague psychiatric center, Prague, CR, or at the National Institute of Mental Health (NIMH), Klecany, CR, within the data collection period of 2014–2017. Patients were diagnosed with FES if they met the following criteria: (a) first hospitalization for schizophrenia-spectrum disorders (schizophrenia or acute and transient psychotic disorders) and (b) clinical interview identified the first psychotic and/or prodromal symptoms of psychosis not earlier than 24 months before (mean, 3.61 ± 5.05 months). The Positive and Negative Symptom Scale, PANSS [42], was rated by experienced psychiatrists (interclass correlation coefficient, 0.73; 95% confidence interval, 0.62–0.83). The majority of the patients were treated with atypical antipsychotics (olanzapine, 29; risperidone, 15; quetiapine, 6; amisulpride, 4; clozapine, 4; aripiprazole, 7; ziprasidone, 2; zotepine, 1; haloperidol, 1; levomepromazine, 1), 43 were on monotherapy, 4 were unmediated, and the rest were on a combination of antipsychotics. Thirty-two healthy control subjects had a similar sociodemographic background to the FES subjects, whom they were selected to match for age and sex. In the final sample, the controls had a slightly higher age and number of years of education than the FES subjects (Table 1). The higher number of SCH patients compared to the controls reflects the fact that we expected a more pronounced influence of risk loci on gray matter reduction (BrainAGE) in the SCH group, and the fixation index (Fst) allows us to estimate the portion of genetic variability in the unequally distributed samples.
The controls were screened using MINI and were excluded if they had a (a) lifetime history of any psychiatric disorders or (b) a family history of psychotic disorders up to the second degree (confirmed by diagnostic interviews and electronic medical records). Additional exclusion criteria for both groups were a history of seizures or significant head trauma, mental retardation, a history of substance dependence, and any MRI contraindications.

2.2. Genomic DNA Analysis

Genomic DNA was extracted from whole blood samples using a QIAamp DNA Blood Kit according to the manufacturer’s protocol. A NimbleGen Custom Sequence Capture Kit was used for ninety-six samples to capture a set of target regions for deep sequencing. The custom design included ~15 Mb of genomic sequence, approximately 0.5% of the human genome, which also targets the complex of all glutamate receptor genes. Briefly, genomic input DNA was sheared to a mean size of 178 bp. The captured fragments from each sample were sequenced as 75–80 bp paired-end reads using an Illumina HiSeq 2500 sequencing instrument at EMBL GeneCore. We used bwa mem [43] for sequence alignment against the genomic reference hg38. The average on-target rate (±100 bp of the target region) was estimated to be 85% using Alfred [44]. The percentage of duplicated reads varied in the range of 10–45% (mean: 24%). After deduplication, almost 90% of the captured targets had a mean coverage ≥ 30x for each sample. Based on the alignments, the sequencing error rate was estimated to be <1% for all the samples. Variant calling was performed using FreeBayes [45], the vcf file was filtered for read depth and mapping quality and finally annotated using VEP [46]. The complete alignment and variant calling workflow is available in the GitHub nRex repository (https://github.com/tobiasrausch/nRex (last accessed on 17 November 2021)). Further filtering based on our case–control design was performed using custom scripts.

2.3. Sequencing Data Analysis

Sixteen genes encoding ionotropic glutamate receptors span over 4.2 Mb and contain extended promoter regions (characterized as 1500 bp upstream of the canonical transcription start site, TSS), 5′UTRs, all coding parts of exons, intron regions, and a 3′UTRs region of variable length. The genomic coordinates of these regions were assigned according to the GENCODE annotation (Ensembl/Havana with consensus coding sequences selected according to CCDS) and saved as a BED file. Genomic variants within each region were then extracted and analyzed using the bcftools program from the merged regenotyped BCF-formatted file generated by nRex. All the identified variants were annotated based on the local copy of the Kaviar (Known VARiants) database and by Ensembl VEP. The distribution of variants was referenced to the canonical transcripts. For GRIA4 and GRIK1 where the TSL method for highlighting a well-supported model of transcript structure does not support a single transcript model, the canonical transcript was defined as the one with the longest coding length. Furthermore, annotation of variants identified within promoter regions was performed using the bedtools intersect with the wgEncode DNaseI Hypersensitivity Clusters and Transcription Factor ChIP-seq data. All necessary transformations of the data and processing of the intermediate results were performed using in-house scripts.

2.4. Homology Modeling

Models of the extracellular and transmembrane domains of human GluA2 (P42262), GluK2 (Q13002), and GluN1/GluN2B (Q05586/Q13224) were constructed with the MODELLER 9.14 suite of programs using the available crystal or cryoEM structures. We used structure 5ide as the GluA2 template, structure 5kuf for modeling GluK2, and structures 4pe5, 4tll, and 4tlm for the GluN1/GluN2B model.

2.5. Statistical Analysis

To identify candidate loci or chromosomal regions associated with SCH, we used VCFtools to estimate Fst for each autosomal diploid variant (calculated according to the Weir and Cockerham method). Only variants from the last percentile of the Fst distribution corresponding to Fst values above 5% inclusive were analyzed further. The Hardy–Weinberg equilibrium and statistically significant difference in the distribution of selected alleles and haplotypes between cases and controls (Fisher’s exact test) and all other statistical analyses were calculated using Statistica 12. Considering small sample sizes, statistical significance was set at p-value < 0.05.
Statistical significance of the case–control data for Manhattan plots was measured using Pearson’s chi-squared test as implemented in the R function chisq.test from the package stats. For a given contingency table (contg_table), we obtained the standardized Pearson residuals, which are residuals adjusted to have asymptotic standard normal distribution, from the R function chisq.test(contg_table)$stdres. The corresponding probability values were computed as chisq.test(contg_table)$p.value.
Pearson’s Chi-square test and the t-test were used for sociodemographic data.

2.6. Segmentation of the Brain and Estimation of the BrainAGE Index

Data were acquired on a 3T Siemens Trio MRI scanner (Siemens, Erlangen, Germany) equipped with a standard head coil. For post hoc analyses of the influence of gray matter volume, the MRI structural T1-weighted (T1W) 3D-MPRAGE sequence was used: repetition time (TR), 2300 ms; echo time (TE), 4.63 ms; bandwidth, 130 Hz/pixel; field of view (FOV), 256 × 256 mm2; matrix, 256 × 256; 160–224 contiguous sagittal slices; voxel size, 1 × 1 × 1 mm3; GRAPPA, acceleration factor 2.
T1W data were segmented using SPM12 (http://www.fil.ion.ucl.ac.uk/spm/soft-ware/spm (accessed on 13 January 2020)) implemented in MATLAB 9.1 (Math Works, Natick, MA, USA). After manual reorientation, T1W images were segmented using the standard unified segmentation model in SPM12. This generated the tissue probability maps reflecting the prior probability of a given voxel belonging to the gray and white matter (GM, WM) and cerebrospinal fluid (CSF). The Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra (DARTEL) algorithm [41] was used to create a study-specific template to generate GM, WM, and CSF images of each individual spatially normalized in the MNI space. Therefore, the volumes of GM, WM, and CSF were calculated using the MATLAB script get_totals (http://www.cs.ucl.ac.uk/staff/g.ridgway/vbm/get_totals.m (accessed on 14 November 2007)).

3. Results

3.1. Genetic Variations in iGluR Genes

In the canonical transcripts of all sixteen iGluR genes in SCH cases, 16,287 variants were detected, of which 98% were located in the intronic regions. Eighty-one variants were located within the adjacent promoter sequences (defined as 1500 bp upstream of a canonical TSS). A markedly increased number of variants (23%) were found in the SCH subjects in contrast to the controls (Table 2). In both groups, 91% of the variants were SNPs and 9% were insertion–deletion mutations (InDels), with the most common InDel being a single nucleotide deletion/duplication (61%). The vast majority of the variants (97%) found exclusively in SCH were located in intronic sequences, but there was also a substantial number of variants in 3′UTRs and promoters (Supplementary Materials: Supplementary Figure S1). However, most of these variants were rare (minor allele frequency (MAF) ≤ 0.016 in SCH) and randomly distributed. An exception was the GRIN2B 3′UTR variant rs1805503 which was present in six patients but not in the controls. It is a low-conserved nucleotide within a well-conserved segment of the 3′UTR region. The increased number of variants in SCH is due in part to the presence of extensive intronic haplotypes, such as the one we observed in GRIN3A, which spans a distance of 106,796 bp across introns 1 to 5 and consists of 63 SNPs and four InDels. Variants without an assigned reference number were assumed to be novel variants (2.4% in SCH). In 276 detected variants, the alternative allele was present on all chromosomes in both the SCH subjects and the controls and thus represents the ancestral allele in the Czech population.
The highest average variability across all the analyzed regions was found in GRIN2B (7.0 variants per 1000 bp), followed by GRIN2A (5.8), GRIN3A (5.6), GRIN3B (4.9), and GRIK4 (4.2). The lowest variability was found for GRIN1 (1.0) and GRIK5 (1.4).
One or more upstream ATGs (uATG) are annotated in the 5′UTRs of the canonical transcripts of twelve iGluR genes. Nevertheless, we found virtually no genetic variations near any of these uATGs in SCH, which are thus unaffected. In the 5′UTR of GRIA3, the alternative allele of the variant rs58044961 was present on all chromosomes in both the SCH subjects and the controls. The variant rs58044961 results in −2G duplication, but the consensus Kozak motif (RccATGG) is not disrupted or attenuated (GgCATGG instead of AgCATGG), so no effects on gene expression are expected.
Thirty-four percent of the exonic variants were missense mutations located predominantly in the N-terminal domain (NTD) and intracellular regulatory C-terminal domain (CTD) of glutamate receptors, suggesting a potentially less deleterious effect on the ion channel function (Figure 1). The observed missense variants largely represented benign population variability; even common premature stop gain-of-frameshift variants, which have a negative effect on protein expression of the allele, showed no significant difference in frequency between the SCH subjects and the controls (Table 3).
The GRIN3B frameshift variant p.Gly466AlafsTer18 (rs10666583) was detected in five SCH and three control homozygotes, consistent with the expected genotype frequencies according to the Hardy–Weinberg equilibrium. The second common GRIN3B frameshift variant p.Lys738GlnfsTer34 (rs545736648) is located in a GC-rich region, which complicates its robust variant calling. Based on the 50 bp rule [47], we hypothesize that both rs10666583 and rs545736648 frameshift variants are removed by nonsense-mediated decay (NMD). The influence of the p.Leu11ProfsTer13 (rs3841128) frameshift variant from the first exon of GRIA1, which was detected in seven SCH patients (six heterozygotes and one homozygote) and five control heterozygotes, remains to be elucidated. The GRIN2C frameshift variant p.Lys788ThrfsTer13 (rs754674133) from the twelfth exon was detected in one patient and is scored as “likely pathogenic” according to the ACMG classification. The novel nonsense variant (p.Gln56*) from the first exon of GRIN3A was scored as “pathogenic” according to the ACMG classification and was observed without co-occurrence with a common frameshift variant. The twenty exonic variants listed in Table 3 represent those with predicted deleterious effects. Missense variants were assessed using at least five prediction tools and must have a CADD_phred value above 20. Of the eighteen exonic variants unique to SCH, nine were classified as deleterious. The one novel missense variant we found, p.Pro1386Leu from the last exon of GRIN2A (hg38; chr16:9,763,387), was classified as benign. In addition, of the eight exonic variants present only in the controls, five were also classified as deleterious.
With few exceptions, the variants significantly associated with SCH were located in intronic regions (Supplementary Figure S1). These variants may represent deep intronic variants with a splicing effect that is difficult to assess or variants that are in linkage disequilibrium with the functional variants.

3.2. Genetic Variation Associated with FES

In the context of polygenic inheritance in SCH, we assumed a cumulative effect of the common susceptible variants. We used the fixation index (Fst) estimate to determine the degree of genetic divergence between the two subpopulations [48,49] that can be run even on small sample sizes. Nearly 20,000 variant loci were included for Fst detection. In case–control studies where both groups are from the same population, Fst detects differences at loci that could be associated with the development of the relevant complex phenotype. The weighted Fst estimate for a subpopulation of SCH cases and controls was −0.001 according to Weir and Cockerham [50]. The negative value indicated a slight excess of heterozygotes (usually caused by the presence of rare variants), but conclusively this analysis of variance did not detect any difference in genetic diversity between the two groups, consistent with their origin from the identical population. For the majority of the variants analyzed (94%), the Fst estimate was less than 2%. Only variants from the last percentile of the estimated variance difference distribution (divergence greater than 5%) became the object of our further interest. These variants were predominantly located in the intronic regions (Figure 2). The higher diversity in the controls was expected as they are derived from a larger population pool than the SCH subjects so only the variants with achieved divergence and higher frequency in SCH were further analyzed.
Loci-specific Fst estimate revealed 189 variants. After excluding 107 variants with higher frequency in the controls than in the SCH subjects, 82 variants remained to be further analyzed and might be associated with schizophrenia. Most of them were distributed among the nine different haplotypes. Certain variants seem to be driven by these haplotypes, and their increased frequency in SCH appears to be due to the observed genetic linkage. Data from the GRIA3 located on the X chromosome were excluded because the population genetics of sex chromosomes differ. However, we found a single haplotype TAAAAAGTT from intron twelve of GRIA3 (hg38; chrX:123,436,446-123,444,102) with a mean Fst = 0.061. The remaining eight haplotypes with detailed characterization are listed in Table 4. All haplotypes (labeled Hap1–8) associated with SCH are located in the intronic sequences. The Hap1 haplotype included regions with strong H3K27Ac histone modification, CpG island, curated transcription factor binding sites (OREG1520167; OREG1520175; OREG1684636), and candidate enhancer and promoter-like signatures. The rs59729868 from Hap1 is located in close proximity to the core-binding sequence for the transcription factor SPI1. The rs2485530 from Hap4 resides in the curated transcription factor binding site (OREG1369890; OREG1345162) within the recognized insulator element in close proximity to the CTCF binding sequence. The rs1939673 from Hap5 is assigned to the curated transcription factor binding site (OREG1938711), which was evaluated as an enhancer element. The rs2284412 from Hap7 is located in a region rich in H3K4me3 modifications (UCSF Brain DNA Methylation) that is co-occupied by multiple candidate transcription factor binding sites.
Besides the nine haplotypes, four additional SNPs were revealed based on the Fst estimate (Table 4). All of these SNPs were also intronic, had low conservation score, and were not located within regulatory sequences or DNaseI hypersensitivity sites. Therefore, they were not included in the proposed scoring model for estimating cumulative risk.

3.3. Scoring Model for the Estimation of Cumulative Risk

The polygenic trait is characterized by numerous variants, each contributing with a relatively small effect. These variants could be common and the predisposition to the phenotype is only an extreme combination of normally distributed genotypes from multiple risk loci. In our study, the value of the likelihood ratio (LR) for a specific haplotype suggests an additive risk model for most haplotypes. Therefore, to estimate a possible cumulative effect of discovered haplotypes on the emergence of SCH, a haplotype scoring model was proposed. The presence of a risk haplotype was penalized by one point for heterozygotes and two points for homozygotes. The Kolmogorov–Smirnov test of normality showed a normal score distribution in both the SCH subjects and the controls (p-values 0.0844 and 0.2616, respectively). The descriptive statistics of the scoring model is summarized in Figure 3. The medians of a given score in the SCH subjects and the controls were then compared. One-tailed Mann–Whitney U test detected a significant difference in the median of both groups (p < 0.00001). The scoring model was used as a simple test to detect SCH. The score of 6 was set as the cutoff point between the positive and the negative test. A positive test is interpreted as a genetic predisposition to schizophrenia. The sensitivity of the test was 0.67, but specificity was high, 0.93. The positive likelihood ratio for the scoring model test reached 7.11. The lower sensitivity of the test is consistent with the expected diversity of disease pathogenesis. According to our data, 9.4% (3 of 32 in our control group) of individuals from the Czech population would test positive. The portion of SCH with a score below six points was 33%.

3.4. Correlation of the BrainAGE Index with the Haplotype Scoring Model

We supplemented our study with post hoc analyses of gray matter loss using the BrainAGE index, a multivariate pattern recognition technique that quantifies age-related brain tissue loss by employing kernel regression methods in a large training database [51,52]. The SCH patients had an increased mean BrainAGE index of 2.44 ± 5.54 years while the control group had a mean BrainAGE index of 0.14 ± 3.94 years compared to their chronological age. There was a significant difference between the means of the two groups (p < 0.05). The mean score of haplotypes in the SCH patients (6.27 ± 1.58) was also higher than in the control group (3.69 ± 1.40), indicating a positive correlation between the overall higher biological brain age and the assumed cumulative effect of the discovered haplotypes in the SCH patients (Figure 4).

4. Discussion

In this study, genes of all the sixteen ionotropic glutamate receptors were sequenced and analyzed in a cohort of Czech SCH patients and matched healthy controls. GRIN2A, GRIN2B, GRIN3A, GRIN3B, and GRIK4 showed increased genetic variability compared to the other iGluR genes. In addition, eight intronic haplotypes were recognized that were significantly associated with SCH. A haplotype scoring model was proposed and used as a simple test to indicate genetic predisposition to schizophrenia. Several missense and frameshift variants were detected. None of these observed and previously annotated variants [53] were associated with SCH.
Recent GWAS studies have identified multiple risk loci associated with SCH [34,35,36]. Published data showed significantly increased genetic variation in genes of the excitatory glutamatergic signaling [37,38,39,40] and confirmed previous pharmacological and genetic studies linking SCH to dysfunction of iGluRs [6]. Previous studies focused on the coding parts of the corresponding genes, where a spectrum of genetic variants was discovered. However, the majority of variants identified were only common polymorphisms, and even those classified as rare missense mutations could only explain isolated SCH cases. The studies mapping the promoters and UTRs of GRIN2A and GRIN2B described a possible effect of noncoding variants on the emergence of SCH [23,24,32]. Much less is known about intronic variants or cumulative effects of variants.
In our study, overall increased genetic variability was in the SCH patients compared with the control subjects. However, most of these variants were rare (MAF ≤ 0.016 in the SCH cases) and randomly distributed. A significantly different frequency between samples from the SCH subjects and the controls was found for the variants located predominantly in intronic sequences, with a few in 3′UTRs or promoter regions. Our data are consistent with the previously published GWAS studies on 108 genomic loci associated with schizophrenia [34] as well as other studies that have repeatedly shown that intronic variants are significantly associated with SCH.
The identified missense variants that were estimated to be deleterious using the available prediction tools were mostly located in the NTD and CTD of the receptor, suggesting rather minor effects on the receptor function. However, a change in the receptor function, its biogenesis, or localization cannot be completely excluded [9]. Nevertheless, rare missense variants have been detected only in single SCH patients. The estimated pathogenic effect can be exerted on the individual level, but from the population perspective, these variants are not crucial. The polymorphic frameshift variant from exon 3 of GRIN3B (rs10666583) had a similar frequency in the SCH subjects and the controls, consistent with the gnomAD allele frequency (MAFCGTT in non-Finnish Europeans is 0.300). The expected frequency of homozygotes in the European population is approximately 9%, well above the lifetime prevalence of 0.5% in SCH [1,2]. The variant rs10666583 was of particular interest as previous studies indicated it as a causal variant and suggested that the truncated protein is produced and interferes with the activity of the wild-type variant, which would have a high impact on the phenotype [54,55]. However, these studies did not address the high frequency of the aberrant allele in the general population. By applying the 50 bp rule, we assumed that the premature stop codon determines the transcript to NMD machinery and acts as a null allele. Recent evidence shows that products of NMD can modulate the transcription of other genes and exert the so-called “genetic compensation response” enabling compensation for allele loss [56]. Thus, the frameshift variant rs10666583 might contribute to the population diversity of NMDAR expression and function but does not play a crucial role in the predisposition to SCH. The same appears to be true for the second common frameshift variant of GRIN3B (rs545736648). The effects of a common GRIA1 frameshift variant rs3841128 and a novel nonsense variant in GRIN3A p.Gln56* are rather unclear. The premature stop codon within the first exon can results in either NMD or activation of the new downstream start codon, with all the possible consequences this may have. The variant p.Lys788ThrfsTer13 (rs754674133) from the twelfth exon of GRIN2C observed in one SCH patient satisfies the 50 bp rule and should therefore escape the NMD machinery. Therefore, we assume its distinct pathogenicity. The variant was previously described as a rare loss-of-function mutation contributing to SCH [30]. The difference in the frequency of co-occurrence of common frameshift variants in the SCH subjects and the controls did not reach statistical significance (p-value = 0.1853), probably due to the small sample size. Nevertheless, the co-occurrence of more than one frameshift variant in two different iGluR genes could represent a risk factor for the development of SCH.
The 5′UTR sequences did not show remarkable variability, and the extent of genetic variation was similar to that in the exonic regions. This is consistent with the critical role of 5′UTRs in determining the mRNA stability and translational efficiency [57]. The higher rate of variability of 3′UTRs, despite their important role in the regulation of mRNA stability and localization, likely reflects the alternation of sequences under strong negative selection with those under mild selection [58]. Previous studies have shown that genetic variation of 3′UTR of NMDAR genes may have an impact on miRNA binding, which could repress the process of translation [59,60]. All the 3′UTR variants identified in our study were assigned to be potentially nonpathogenic (Figure 2). The C allele of rs890 from the 3′UTR of GRIN2B, previously associated with SCH (Chinese Han population), was more frequent in our controls (0.609) than in the SCH subjects (0.460), consistent with gnomAD allelic frequencies in Europeans (0.512). The C allele alters the sequence targeted by miR-4328 and creates a new miRNA site (miR-1468-5p) resulting in the repression of transcript activity [59]. However, it is quite difficult to assess functionality of the variants from 3′UTR sequences, as a number of proteins bind coordinately to 3´UTR elements, often in a tissue-specific manner.
For the given canonical transcripts, most of the observed genetic variation was located within introns. Fst calculation was used to identify those intronic variants associated with SCH. Strict constraints were applied, and only variants from the last percentile of the Fst distribution were further analyzed. The Fst estimate is used to detect population substructure due to inbreeding or genetic drift. To our knowledge, the SCH subjects were unrelated individuals recruited from the area of more than 1 million people with a generally high turnover of persons from all over the Czech Republic. Using Fst calculation, eight intronic haplotypes and four intronic SNPs with significantly higher frequency were detected in the SCH subjects compared to the controls. A highly conserved intronic nucleotide may highlight a splicing element, secondary structure-forming region, or enhancer element or affect protein abundance by some other mechanism [61]. Almost all of these haplotypes contain transcription factor binding sites and curated or putative regulatory elements. Haplotypes Hap1, Hap2, Hap4, and Hap8 are located at the 5′end (promoter, intron 1 or 2) of GRIK3, GRIA1, GRIN3A, and GRIK1 genes, respectively, and may have a direct impact on the efficiency of transcription initiation. The most interesting haplotype is Hap1 of GRIK3, as it combines a region with strong H3K27Ac histone modification, CpG island, and promoter-like signatures. GRIK3 encodes the GluK3 subunit, which plays a role in presynaptic facilitation of signal transduction in hippocampal mossy fibers [62]. In the event that Hap1 negatively affects the GRIK3 promoter, lower GluK3 expression could contribute to glutamatergic hypofunction in individual cases. Analysis of the GRIK3 promoter revealed another variant c.-1420G>T (rs588902) that was not in linkage with the Hap1 haplotype but had a high Fst estimate (0.049). Previously, common polymorphic variants from the promoter regions of GRIA1, GRIA3, GRIN1, or GRIN2A were associated with susceptibility to addiction, migraine, or schizophrenia [32,63,64,65]. Gene transcription requires the RNA polymerase complex II bound to the core of the promoter, which is decorated by regulatory factors that recognize specific DNA sequences [66]. Polymorphisms within promoter regions can modify binding motifs and alter promoter activity [67]. In the SCH subjects, the length of the microsatellite (GT)n repeat (rs3219790) within the GRIN2A promoter differed significantly from the controls and was associated with a change in transcriptional activity [32]. Similarly, the minor A allele of the c.-447-42G>A polymorphism (rs3764030) from the GRIN2B promoter was assigned using reporter gene assays as a gain-of-function variant associated with an increase in the GRIN2B mRNA levels [68]. The frequency of the A allele in the SCH subjects (0.294) was not significantly different (p-value = 0.1805) from that in the controls (0.203) and the gnomAD genomes frequency in the European population (0.190). The Fst estimate for rs3764030 was 0.012 (GERP = −1.74; CADD_phred = 17.16). Our data do not contradict the assigned influence of the A allele on GRIN2B expression but indicate its possibly small contribution to the emergence and development of SCH.
Since the likelihood ratios for individual haplotypes follow the simple additive model, we proposed a haplotype scoring model to estimate a possible cumulative effect in SCH. The scoring model was used as a simple test where a score of 6 indicates test positivity interpreted as a genetic predisposition to schizophrenia. With a lifetime prevalence of schizophrenia of 4.8/1000 [2], the population risk is approximately 0.5%. Test positivity could increase the lifetime risk of schizophrenia up to 3.3%. Such a test is not intended for population diagnosis, as only susceptibility to schizophrenia is inherited, but could be helpful in diagnosing the familial form of the disorder or pathogenesis. Diverse polygenic risk scores (PRSs) are beginning to be an effective tool for assessing familial risk of polygenic traits or response to treatment [69,70,71]. The estimated prevalence of score 6 in the Czech population is 9%, with 3.3% probability of schizophrenia in this group, this could explain about 60% of the lifetime prevalence of SCH in the affected Czech population. In 33% of FES patients with a score below six points, the pathogenesis of schizophrenia is not predominantly associated with the variation of iGluR genes or at least with the estimated risk haplotypes. However, such association of the defined haplotypes with SCH and the proposed test validity need to be verified in a larger group of SCH patients and controls.
A common sign of SCH is exaggerated brain maturation with age-altered loss of gray mater. The BrainAGE aggregates the complex multidimensional aging patterns in a single value, i.e., the deviation of brain atrophy from normal brain aging, expressed in years [72]. BrainAGE can be estimated either for the aggregated whole brain data or for individual tissues. A longitudinal neuroimaging study examining progressive brain tissue loss in schizophrenia patients revealed a brain age gap of 3.36 ± 5.87 years at baseline (the first MRI scans) compared to the chronological age of the patients [73]. In the study that focused on patients with early stages of SCH and bipolar disorder, a higher brain age (2.64 ± 4.15) than their chronological age was estimated [74]. Our group of FES patients had a brain age gap of 2.44 ± 5.54, similarly to previous studies, and this value correlated positively with a higher mean haplotype score in SCH. The largest difference in brain age between the SCH group and the control group was for two haplotype scores 3 and 5. In contrast, for a given haplotype score 4 or 6, we observed smaller differences in the brain age gap between the SCH group and the control group than should correspond to the estimated values of the mean in the SCH subjects and the controls. We suppose that this was caused by a small number of samples (n ≤ 6) within the specific score combined with genetic heterogeneity. This possibility supports studies showing that individual changes in the GM volume are influenced by the previous course of psychosis, medication, and other genetic or environmental factors [75,76,77].
Our data support the concept that genetic predisposition to SCH may also result from the cumulative effect of a number of polymorphisms, each contributing with a small impact. In our study of iGluRs, such polymorphic loci were located in conserved intronic sequences rather than in exonic ones. Variability in the regulatory elements of genes can affect the orchestration of gene activity and result in altered regulation of protein expression.
The study estimates genetic variability in iGluRs plays a meaningful role in two-thirds of FES patients. Genetic variability associated with SCH was found for variants in intronic sequences and variants in 3′UTRs or promoter regions, suggesting altered mRNA processing and lower gene expression that could possibly be compensated in certain cases by the administration of positive modulators of glutamatergic signaling.

5. Conclusions

In this study, we investigated the whole genes, inclusive introns, and extended promoters encoding all sixteen ionotropic glutamate receptors to capture the overall distribution of sequence variants in a cohort of SCH patients and matched controls. The Fst estimate highlighted eight intronic haplotypes significantly associated with FES patients in our study. These haplotypes were used in the proposed scoring model and moreover as a simple test where a score of 6 indicated positivity and was interpreted as a genetic predisposition to SCH. The positive likelihood ratio for the scoring model test reached 7.11. Genetic variability significantly associated with SCH was found for variants predominantly in intronic sequences, with few variants in 3′UTRs or promoter regions and none in 5′UTRs or exonic parts of the genes studied. Our data suggest that some intronic regulatory regions of iGluR genes and their common variability could be one of the components from which the genetic predisposition to SCH is composed.

Supplementary Materials

The following are available online at https://www.mdpi.com/article/10.3390/jpm11121250/s1, Figure S1: The position of the detected variants in the iGluR genes.

Author Contributions

Conceptualization, A.B., L.V. and J.H.; methodology, J.C., V.B., K.H. and J.H.; software, T.R.; investigation and analysis, K.H., J.C., P.B. and V.K.; validation, K.H., J.C. and A.B.; clinical data, J.H., F.S. and D.G.; writing, K.H., J.C., L.V. and A.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by MEYS of the Czech Republic: Program LT—INTER-EXCELLENCE, LTAUSA19122 (A.B.); the Czech Science Foundation: 20-179458 (L.V.); the Technology Agency of the Czech Republic: TN01000013 (L.V.); ERDF/ESF project: PharmaBrain (No. CZ.02.1.01/0.0/0.0/16_025/0007444) (LV); Charles University Grant Agency: 1206218 (V.K.); the Czech Health Res. Council: NU21-04-00405 (J.H.); Research Project of AS CR RVO: 67985823 and 86652036; MEYS of the Czech Republic: BIOCEV-FAR (LQ1604) ELIXIR CZ (LM2018131), e-INFRA CZ (LM2018140); and by the BIOCEV project (CZ.1.05/1.1.00/02.0109).

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of The National Institute of Mental Health (Klecany, Czech Republic).

Informed Consent Statement

Informed consent was obtained from all the subjects involved in the study.

Data Availability Statement

The data supporting the reported results can be found at https://zenodo.org/record/5347495 (accessed on 31 August 2021).

Acknowledgments

We would like to thank Katja Franke, PhD, for providing us with the BrainAge methodology, and all the individuals who participated in this study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Saha, S.; Chant, D.; Welham, J.; McGrath, J. A Systematic Review of the Prevalence of Schizophrenia. PLoS Med. 2005, 2, e141. [Google Scholar] [CrossRef] [PubMed]
  2. Simeone, J.C.; Ward, A.J.; Rotella, P.; Collins, J.; Windisch, R. An evaluation of variation in published estimates of schizophrenia prevalence from 1990─2013: A systematic literature review. BMC Psychiatry 2015, 15, 1–14. [Google Scholar] [CrossRef][Green Version]
  3. Shih, R.A.; Belmonte, P.L.; Zandi, P.P. A review of the evidence from family, twin and adoption studies for a genetic contribution to adult psychiatric disorders. Int. Rev. Psychiatry 2004, 16, 260–283. [Google Scholar] [CrossRef] [PubMed]
  4. Goldberg, T.E.; David, A.; Gold, J.M. Neurocognitive Deficits in Schizophrenia. In Schizophrenia; Blackwell Science Ltd.: Oxford, UK, 2003; pp. 168–184. [Google Scholar]
  5. Tandon, R.; Nasrallah, H.A.; Keshavan, M.S. Schizophrenia, “just the facts” 4. Clinical features and conceptualization. Schizophr. Res. 2009, 110, 1–23. [Google Scholar] [CrossRef]
  6. Kristiansen, L.; Huerta, I.; Beneyto, M.; Meador-Woodruff, J.H. NMDA receptors and schizophrenia. Curr. Opin. Pharmacol. 2007, 7, 48–55. [Google Scholar] [CrossRef]
  7. E McCullumsmith, R.; Hammond, J.H.; Shan, D.; Meador-Woodruff, J.H. Postmortem Brain: An Underutilized Substrate for Studying Severe Mental Illness. Neuropsychopharmacology 2013, 39, 65–87. [Google Scholar] [CrossRef][Green Version]
  8. Weickert, C.S.; Fung, S.J.; Catts, V.S.; Schofield, P.; Allen, K.M.; Moore, L.; Newell, K.; Pellen, D.; Huang, X.-F.; Catts, S.V.; et al. Molecular evidence of N-methyl-D-aspartate receptor hypofunction in schizophrenia. Mol. Psychiatry 2012, 18, 1185–1192. [Google Scholar] [CrossRef][Green Version]
  9. Traynelis, S.F.; Wollmuth, L.P.; Mcbain, C.J.; Menniti, F.S.; Vance, K.M.; Ogden, K.K.; Hansen, K.B.; Yuan, H.; Myers, S.J.; Dingledine, R. Glutamate Receptor Ion Channels: Structure, Regulation, and Function. Pharmacol. Rev. 2010, 62, 405–496. [Google Scholar] [CrossRef] [PubMed][Green Version]
  10. Hansen, K.B.; Wollmuth, L.P.; Bowie, D.; Furukawa, H.; Menniti, F.S.; Sobolevsky, A.I.; Swanson, G.T.; Swanger, S.A.; Greger, I.H.; Nakagawa, T.; et al. Structure, Function, and Pharmacology of Glutamate Receptor Ion Channels. Pharmacol. Rev. 2021, 73, 298–487. [Google Scholar] [CrossRef]
  11. Greger, I.H.; Mayer, M.L. Structural biology of glutamate receptor ion channels: Towards an understanding of mechanism. Curr. Opin. Struct. Biol. 2019, 57, 185–195. [Google Scholar] [CrossRef]
  12. Myers, S.J.; Dingledine, R.; Borges, K. Genetic Regulation of Glutamate Receptor Ion Channels. Annu. Rev. Pharmacol. Toxicol. 1999, 39, 221–241. [Google Scholar] [CrossRef] [PubMed]
  13. Gan, Q.; Salussolia, C.; Wollmuth, L.P.; Scientist, M.; Program, T.; Brook, S. Assembly of AMPA receptors: Mechanisms and regulation. J. Physiol. 2014, 593, 39–48. [Google Scholar] [CrossRef]
  14. Greger, I.H.; Watson, J.; Cull-Candy, S.G. Structural and Functional Architecture of AMPA-Type Glutamate Receptors and Their Auxiliary Proteins. Neuron 2017, 94, 713–730. [Google Scholar] [CrossRef] [PubMed]
  15. Paoletti, P.; Bellone, C.; Zhou, Q. NMDA receptor subunit diversity: Impact on receptor properties, synaptic plasticity and disease. Nat. Rev. Neurosci. 2013, 14, 383–400. [Google Scholar] [CrossRef]
  16. Hansen, K.B.; Yi, F.; Perszyk, R.; Furukawa, H.; Wollmuth, L.P.; Gibb, A.; Traynelis, S.F. Structure, function, and allosteric modulation of NMDA receptors. J. Gen. Physiol. 2018, 150, 1081–1105. [Google Scholar] [CrossRef]
  17. Elgersma, Y.; Silva, A.J. Molecular mechanisms of synaptic plasticity and memory. Curr. Opin. Neurobiol. 1999, 9, 209–213. [Google Scholar] [CrossRef]
  18. Hunt, D.; E Castillo, P. Synaptic plasticity of NMDA receptors: Mechanisms and functional implications. Curr. Opin. Neurobiol. 2012, 22, 496–508. [Google Scholar] [CrossRef][Green Version]
  19. Thapar, A.; Cooper, M.; Rutter, M. Neurodevelopmental disorders. Lancet Psychiatry 2017, 4, 339–346. [Google Scholar] [CrossRef][Green Version]
  20. Luby, E.D. Study of a New Schizophrenomimetic Drug—Sernyl. Arch. Neurol. Psychiatry 1959, 81, 363–369. [Google Scholar] [CrossRef]
  21. Malhotra, A.K.; Pinals, D.A.; Adler, C.M.; Elman, I.; Clifton, A.; Pickar, D.; Breier, A. Ketamine-Induced Exacerbation of Psychotic Symptoms and Cognitive Impairment in Neuroleptic-Free Schizophrenics. Neuropsychopharmacology 1997, 17, 141–150. [Google Scholar] [CrossRef]
  22. Mohn, A.R.; Gainetdinov, R.R.; Caron, M.G.; Koller, B.H. Mice with Reduced NMDA Receptor Expression Display Behaviors Related to Schizophrenia. Cell 1999, 98, 427–436. [Google Scholar] [CrossRef][Green Version]
  23. Miyatake, R.; Furukawa, A.; Suwaki, H. Identification of a novel variant of the human NR2B gene promoter region and its possible association with schizophrenia. Mol. Psychiatry 2002, 7, 1101–1106. [Google Scholar] [CrossRef][Green Version]
  24. Qin, S.; Zhao, X.; Pan, Y.; Liu, J.; Feng, G.; Fu, J.; Bao, J.; Zhang, Z.; He, L. An association study of the N-methyl-D-aspartate receptor NR1 subunit gene (GRIN1) and NR2B subunit gene (GRIN2B) in schizophrenia with universal DNA microarray. Eur. J. Hum. Genet. 2005, 13, 807–814. [Google Scholar] [CrossRef][Green Version]
  25. Rice, S.R.; Niu, N.; Berman, D.B.; Heston, L.L.; Sobell, J.L. Identification of single nucleotide polymorphisms (SNPs) and other sequence changes and estimation of nucleotide diversity in coding and flanking regions of the NMDAR1 receptor gene in schizophrenic patients. Mol. Psychiatry 2001, 6, 274–284. [Google Scholar] [CrossRef][Green Version]
  26. Sakurai, K.; Toru, M.; Yamakawa-Kobayashi, K.; Arinami, T. Mutation analysis of the N-methyl-d-aspartate receptor NR1 subunit gene (GRIN1) in schizophrenia. Neurosci. Lett. 2000, 296, 168–170. [Google Scholar] [CrossRef]
  27. Shen, Y.-C.; Liao, D.-L.; Chen, J.-Y.; Wang, Y.-C.; Lai, I.-C.; Liou, Y.-J.; Chen, Y.-J.; Luu, S.-U.; Chen, C.-H. Exomic sequencing of the glutamate receptor, ionotropic, N-methyl-d-aspartate 3A gene (GRIN3A) reveals no association with schizophrenia. Schizophr. Res. 2009, 114, 25–32. [Google Scholar] [CrossRef] [PubMed]
  28. Takata, A.; Iwayama, Y.; Fukuo, Y.; Ikeda, M.; Okochi, T.; Maekawa, M.; Toyota, T.; Yamada, K.; Hattori, E.; Ohnishi, T.; et al. A Population-Specific Uncommon Variant in GRIN3A Associated with Schizophrenia. Biol. Psychiatry 2013, 73, 532–539. [Google Scholar] [CrossRef] [PubMed][Green Version]
  29. Williams, N.M.; Bowen, T.; Spurlock, G.; Norton, N.; Williams, H.; Hoogendoorn, B.; Owen, M.J.; O’Donovan, M.C. Determination of the genomic structure and mutation screening in schizophrenic individuals for five subunits of the N-methyl-D-aspartate glutamate receptor. Mol. Psychiatry 2002, 7, 508–514. [Google Scholar] [CrossRef][Green Version]
  30. Yu, Y.; Lin, Y.; Takasaki, Y.; Wang, C.; Kimura, H.; Xing, J.; Ishizuka, K.; Toyama, M.; Kushima, I.; Mori, D.; et al. Rare loss of function mutations in N-methyl-d-aspartate glutamate receptors and their contributions to schizophrenia susceptibility. Transl. Psychiatry 2018, 8, 1–9. [Google Scholar] [CrossRef] [PubMed]
  31. Tarabeux, J.; Kebir, O.; Gauthier, J.; Hamdan, F.F.; Xiong, L.; Piton, A.; Spiegelman, D.; Henrion, É.; Millet, B.; Fathalli, F.; et al. Rare mutations in N-methyl-D-aspartate glutamate receptors in autism spectrum disorders and schizophrenia. Transl. Psychiatry 2011, 1, e55. [Google Scholar] [CrossRef] [PubMed][Green Version]
  32. Itokawa, M.; Yamada, K.; Yoshitsugu, K.; Toyota, T.; Suga, T.; Ohba, H.; Watanabe, A.; Hattori, E.; Shimizu, H.; Kumakura, T.; et al. A microsatellite repeat in the promoter of the N-methyl-d-aspartate receptor 2A subunit (GRIN2A) gene suppresses transcriptional activity and correlates with chronic outcome in schizophrenia. Pharmacogenetics 2003, 13, 271–278. [Google Scholar] [CrossRef]
  33. Iwayama-Shigeno, Y.; Yamada, K.; Itokawa, M.; Toyota, T.; Meerabux, J.M.; Minabe, Y.; Mori, N.; Inada, T.; Yoshikawa, T. Extended analyses support the association of a functional (GT)n polymorphism in the GRIN2A promoter with Japanese schizophrenia. Neurosci. Lett. 2005, 378, 102–105. [Google Scholar] [CrossRef]
  34. Ripke, S.; Neale, B.M.; Corvin, A.; Walters, J.T.R.; Farh, K.-H.; Holmans, P.A.; Lee, P.; Bulik-Sullivan, B.; Collier, D.A.; Huang, H.; et al. Biological insights from 108 schizophrenia-associated genetic loci. Nature 2014, 511, 421–427. [Google Scholar] [CrossRef][Green Version]
  35. International, T.; Consortium, S. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nat. Cell Biol. 2009, 460, 748–752. [Google Scholar] [CrossRef]
  36. Ripke, S.; O’Dushlaine, C.; Chambert, K.; Moran, J.; Kähler, A.K.; Akterin, S.; Bergen, S.; Collins, A.L.; Crowley, J.; Fromer, M.; et al. Genome-wide association analysis identifies 13 new risk loci for schizophrenia. Nat. Genet. 2013, 45, 1150–1159. [Google Scholar] [CrossRef]
  37. Frank, R.; McRae, A.F.; Pocklington, A.J.; Van De Lagemaat, L.N.; Navarro, P.; Croning, M.D.R.; Komiyama, N.H.; Bradley, S.J.; Challiss, R.A.J.; Armstrong, J.D.; et al. Clustered Coding Variants in the Glutamate Receptor Complexes of Individuals with Schizophrenia and Bipolar Disorder. PLoS ONE 2011, 6, e19011. [Google Scholar] [CrossRef]
  38. Kirov, G.; Pocklington, A.; Holmans, P.; Ivanov, D.; Ikeda, M.; Ruderfer, D.; Moran, J.; Chambert, K.; Toncheva, D.; Georgieva, L.; et al. De novo CNV analysis implicates specific abnormalities of postsynaptic signalling complexes in the pathogenesis of schizophrenia. Mol. Psychiatry 2011, 17, 142–153. [Google Scholar] [CrossRef][Green Version]
  39. Pocklington, A.J.; Rees, E.; Donovan, M.C.O.; Owen, M.J.; Pocklington, A.J.; Rees, E.; Walters, J.T.R.; Han, J.; Kavanagh, D.H.; Chambert, K.D. Novel Findings from CNVs Implicate Inhibitory and Excitatory Signaling Complexes in Schizophrenia Article Novel Findings from CNVs Implicate Inhibitory and Excitatory Signaling Complexes in Schizophrenia. Neuron 2015, 86, 1203–1214. [Google Scholar] [CrossRef][Green Version]
  40. Purcell, S.M.; Moran, J.; Fromer, M.; Ruderfer, D.; Solovieff, N.; Roussos, P.; O’Dushlaine, C.; Chambert, K.; Bergen, S.; Kähler, A.; et al. A polygenic burden of rare disruptive mutations in schizophrenia. Nature. 2014, 506, 185–190. [Google Scholar] [CrossRef] [PubMed][Green Version]
  41. Sheehan, D.V.; Lecrubier, Y.; Sheehan, K.H.; Amorim, P.; Janavs, J.; Weiller, E.; Hergueta, T.; Baker, R.; Dunbar, G.C. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 1998, 59, 22–33; quiz 34–57. [Google Scholar] [PubMed]
  42. Kay, S.R.; Fiszbein, A.; Opler, L.A. The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophr. Bull. 1987, 13, 261–276. [Google Scholar] [CrossRef]
  43. Li, H.; Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 2009, 25, 1754–1760. [Google Scholar] [CrossRef] [PubMed][Green Version]
  44. Rausch, T.; Fritz, M.H.-Y.; O Korbel, J.; Benes, V. Alfred: Interactive multi-sample BAM alignment statistics, feature counting and feature annotation for long- and short-read sequencing. Bioinformatics 2019, 35, 2489–2491. [Google Scholar] [CrossRef][Green Version]
  45. Garrison, E.; Marth, G. Haplotype-based variant detection from short-read sequencing. arXiv 2012, arXiv:1207.3907. [Google Scholar]
  46. McLaren, W.; Gil, L.; Hunt, S.E.; Riat, H.S.; Ritchie, G.R.S.; Thormann, A.; Flicek, P.; Cunningham, F. The Ensembl Variant Effect Predictor. Genome Biol. 2016, 17, 122. [Google Scholar] [CrossRef][Green Version]
  47. Zhang, Z.; Xin, D.; Wang, P.; Zhou, L.; Hu, L.; Kong, X.; Hurst, L.D. Noisy splicing, more than expression regulation, explains why some exons are subject to nonsense-mediated mRNA decay. BMC Biol. 2009, 7, 23. [Google Scholar] [CrossRef] [PubMed][Green Version]
  48. Wright, S. The interpretation of population structure by f-statistics with special regard to systems of mating. Evolution 1965, 19, 395–420. [Google Scholar] [CrossRef]
  49. Meirmans, P.G.; Hedrick, P.W. Assessing population structure: FST and related measures. Mol. Ecol. Resour. 2011, 11, 5–18. [Google Scholar] [CrossRef]
  50. Weir, B.S.; Cockerham, C.C. Estimating F-Statistics for the Analysis of Population Structure. Evolution 1984, 38, 1358–1370. [Google Scholar] [CrossRef]
  51. Franke, K.; Ziegler, G.; Klöppel, S.; Gaser, C. Estimating the age of healthy subjects from T1-weighted MRI scans using kernel methods: Exploring the influence of various parameters. NeuroImage 2010, 50, 883–892. [Google Scholar] [CrossRef]
  52. Franke, K.; Luders, E.; May, A.; Wilke, M.; Gaser, C. Brain maturation: Predicting individual BrainAGE in children and adolescents using structural MRI. NeuroImage 2012, 63, 1305–1312. [Google Scholar] [CrossRef] [PubMed]
  53. García-Recio, A.; Santos-Gómez, A.; Soto, D.; Julia-Palacios, N.; García-Cazorla, À.; Altafaj, X.; Olivella, M. GRIN database: A unified and manually curated repertoire of GRIN variants. Hum. Mutat. 2021, 42, 8–18. [Google Scholar] [CrossRef] [PubMed]
  54. Hornig, T.; Grüning, B.; Kundu, K.; Houwaart, T.; Backofen, R.; Biber, K.; Normann, C. GRIN3B missense mutation as an inherited risk factor for schizophrenia: Whole-exome sequencing in a family with a familiar history of psychotic disorders. Genet. Res. 2017, 99, e1. [Google Scholar] [CrossRef] [PubMed][Green Version]
  55. Matsuno, H.; Ohi, K.; Hashimoto, R.; Yamamori, H.; Yasuda, Y.; Fujimoto, M.; Yano-Umeda, S.; Saneyoshi, T.; Takeda, M.; Hayashi, Y. A Naturally Occurring Null Variant of the NMDA Type Glutamate Receptor NR3B Subunit Is a Risk Factor of Schizophrenia. PLoS ONE 2015, 10, e0116319. [Google Scholar] [CrossRef]
  56. El-Brolosy, M.A.; Kontarakis, Z.; Rossi, A.; Kuenne, C.; Günther, S.; Fukuda, N.; Kikhi, K.; Boezio, G.L.M.; Takacs, C.M.; Lai, S.-L.; et al. Genetic compensation triggered by mutant mRNA degradation. Nature 2019, 568, 193–197. [Google Scholar] [CrossRef]
  57. Mignone, F.; Gissi, C.; Liuni, S.; Pesole, G. Untranslated regions of mRNAs. Genome Biol. 2002, 3, reviews0004.1. [Google Scholar] [CrossRef]
  58. Hu, Z.; Bruno, A.E. The Influence of 3′UTRs on MicroRNA Function Inferred from Human SNP Data. Comp. Funct. Genom. 2011, 2011, 1–9. [Google Scholar] [CrossRef][Green Version]
  59. Zhang, Y.; Fan, M.; Wang, Q.; He, G.; Fu, Y.; Li, H.; Yu, S. Polymorphisms in MicroRNA Genes And Genes Involving in NMDAR Signaling and Schizophrenia: A Case-Control Study in Chinese Han Population. Sci. Rep. 2015, 5, 12984. [Google Scholar] [CrossRef][Green Version]
  60. Shen, H.; Li, Z. miRNAs in NMDA receptor-dependent synaptic plasticity and psychiatric disorders. Clin. Sci. 2016, 130, 1137–1146. [Google Scholar] [CrossRef] [PubMed][Green Version]
  61. D’Orazio, K.N.; Wu, C.C.-C.; Sinha, N.; Loll-Krippleber, R.; Brown, G.W.; Green, R. The endonuclease Cue2 cleaves mRNAs at stalled ribosomes during No Go Decay. eLife 2019, 8, 8. [Google Scholar] [CrossRef]
  62. Pinheiro, P.; Perrais, D.; Coussen, F.; Barhanin, J.; Bettler, B.; Mann, J.R.; Malva, J.; Heinemann, S.F.; Mulle, C. GluR7 is an essential subunit of presynaptic kainate autoreceptors at hippocampal mossy fiber synapses. Proc. Natl. Acad. Sci. USA 2007, 104, 12181–12186. [Google Scholar] [CrossRef] [PubMed][Green Version]
  63. Liu, Y.-P.; Ding, M.; Zhang, X.-C.; Liu, Y.; Xuan, J.-F.; Xing, J.-X.; Xia, X.; Yao, J.; Wang, B.-J. Association between polymorphisms in the GRIN1 gene 5′ regulatory region and schizophrenia in a northern Han Chinese population and haplotype effects on protein expression in vitro. BMC Med. Genet. 2019, 20, 26. [Google Scholar] [CrossRef][Green Version]
  64. Formicola, D.; Aloia, A.; Sampaolo, S.; Farina, O.; Diodato, D.; Griffiths, L.R.; Gianfrancesco, F.; Di Iorio, G.; Esposito, T. Common variants in the regulative regions of GRIA1 and GRIA3 receptor genes are associated with migraine susceptibility. BMC Med. Genet. 2010, 11, 103. [Google Scholar] [CrossRef] [PubMed][Green Version]
  65. Domart, M.-C.; Benyamina, A.; Lemoine, A.; Bourgain, C.; Blecha, L.; Debuire, B.; Reynaud, M.; Saffroy, R. Association between a polymorphism in the promoter of a glutamate receptor subunit gene (GRIN2A) and alcoholism. Addict. Biol. 2011, 17, 783–785. [Google Scholar] [CrossRef] [PubMed]
  66. Butler, J.E.F.; Kadonaga, J.T. The RNA polymerase II core promoter: A key component in the regulation of gene expression. Genes Dev. 2002, 2583–2592. [Google Scholar] [CrossRef][Green Version]
  67. Bai, G.; Hoffman, P.W. Transcriptional Regulation of NMDA Receptor Expression. In Biology of the NMDA Receptor; CRC Press: Boca Raton, FL, USA, 2009; pp. 79–102. [Google Scholar]
  68. Jiang, Y.; Lin, M.; Jicha, G.A.; Ding, X.; McIlwrath, S.L.; Fardo, D.; Broster, L.S.; Schmitt, F.A.; Kryscio, R.; Lipsky, R.H. Functional human GRIN2B promoter polymorphism and variation of mental processing speed in older adults. Aging 2017, 9, 1293–1306. [Google Scholar] [CrossRef] [PubMed][Green Version]
  69. Toulopoulou, T.; Zhang, X.; Cherny, S.; Dickinson, D.; Berman, K.F.; E Straub, R.; Sham, P.; Weinberger, D.R. Polygenic risk score increases schizophrenia liability through cognition-relevant pathways. Brain 2019, 142, 471–485. [Google Scholar] [CrossRef] [PubMed]
  70. Wimberley, T.; Gasse, C.; Meier, S.M.; Agerbo, E.; MacCabe, J.H.; Horsdal, H.T. Polygenic Risk Score for Schizophrenia and Treatment-Resistant Schizophrenia. Schizophr. Bull. 2017, 43, 1064–1069. [Google Scholar] [CrossRef]
  71. Chalmer, M.A.; Esserlind, A.-L.; Olesen, J.; Hansen, T.F. Polygenic risk score: Use in migraine research. J. Headache Pain 2018, 19, 29. [Google Scholar] [CrossRef][Green Version]
  72. Gogtay, N.; Thompson, P.M. Mapping gray matter development: Implications for typical development and vulnerability to psychopathology. Brain Cogn. 2010, 72, 6–15. [Google Scholar] [CrossRef][Green Version]
  73. Schnack, H.G.; van Haren, N.E.M.; Nieuwenhuis, M.; Pol, H.E.H.; Cahn, W.; Kahn, R.S. Accelerated Brain Aging in Schizophrenia: A Longitudinal Pattern Recognition Study. Am. J. Psychiatry 2016, 173, 607–616. [Google Scholar] [CrossRef][Green Version]
  74. Hajek, T.; Franke, K.; Kolenic, M.; Capkova, J.; Matejka, M.; Propper, L.; Uher, R.; Stopkova, P.; Novák, T.; Paus, T.; et al. Brain Age in Early Stages of Bipolar Disorders or Schizophrenia. Schizophr. Bull. 2019, 45, 190–198. [Google Scholar] [CrossRef] [PubMed]
  75. Cahn, W.; Rais, M.; Stigter, F.; van Haren, N.; Caspers, E.; Pol, H.H.; Xu, Z.; Schnack, H.; Kahn, R. Psychosis and brain volume changes during the first five years of schizophrenia. Eur. Neuropsychopharmacol. 2009, 19, 147–151. [Google Scholar] [CrossRef]
  76. Fusar-Poli, P.; Smieskova, R.; Kempton, M.; Ho, B.-C.; Andreasen, N.; Borgwardt, S. Progressive brain changes in schizophrenia related to antipsychotic treatment? A meta-analysis of longitudinal MRI studies. Neurosci. Biobehav. Rev. 2013, 37, 1680–1691. [Google Scholar] [CrossRef] [PubMed][Green Version]
  77. Brans, R.G.H.; Van Haren, N.E.M.; Van Baal, G.C.M.; Schnack, H.G.; Kahn, R.S.; Pol, H.E.H. Heritability of Changes in Brain Volume Over Time in Twin Pairs Discordant for Schizophrenia. Arch. Gen. Psychiatry 2008, 65, 1259–1268. [Google Scholar] [CrossRef] [PubMed][Green Version]
Figure 1. Positions of missense variants found in iGluR subunits. The structures of GluA and GluK represent a prototypical homomeric receptor. Green and yellow residues in the GluA or GluK subunit (in blue) mark missense variants. The structure of GluN represents the heteromeric iGluR (obligatory GLuN1 subunits in gray, variable GluN2/N3 subunits in orange). Missense variants in the GluN1 subunit are depicted in green. Yellow and red residues in the GluN2/N3 subunit (in orange) mark missense variants found in some of the six subunits.
Figure 1. Positions of missense variants found in iGluR subunits. The structures of GluA and GluK represent a prototypical homomeric receptor. Green and yellow residues in the GluA or GluK subunit (in blue) mark missense variants. The structure of GluN represents the heteromeric iGluR (obligatory GLuN1 subunits in gray, variable GluN2/N3 subunits in orange). Missense variants in the GluN1 subunit are depicted in green. Yellow and red residues in the GluN2/N3 subunit (in orange) mark missense variants found in some of the six subunits.
Jpm 11 01250 g001
Figure 2. The Fst (%) estimate for variants detected in iGluR genes. Boxplots show variants with Fst above 5% (last percentile of the Fst distribution) that are located nearly exclusively in intronic sequences. Non-intronic variants are highlighted. The rs61736488 is a synonymous substitution p.Lys798Lys from exon 15 of GRIA4 with a medium CADD_phred score of 10.66. The variant rs594592 is located in the 3′UTR of GRIA4 with a low CADD_pherd score of 1.841. The GRIK3 promoter variant rs2134846 is part of the haplotype Hap1. Another GRIK3 promoter variant rs588902 with a high CADD_phred score of 22.3 is highly conserved and has the GERP score of 2.51. The exonic variant rs1337677 is a synonymous variant p.Pro229Pro from the first exon of GRIN3A that is in almost complete linkage with the haplotype Hap4 and has a medium CADD_phred score of 11.62. The GRIN2D 3′UTR variant rs8111684 has a CADD_phred score of 16.41.
Figure 2. The Fst (%) estimate for variants detected in iGluR genes. Boxplots show variants with Fst above 5% (last percentile of the Fst distribution) that are located nearly exclusively in intronic sequences. Non-intronic variants are highlighted. The rs61736488 is a synonymous substitution p.Lys798Lys from exon 15 of GRIA4 with a medium CADD_phred score of 10.66. The variant rs594592 is located in the 3′UTR of GRIA4 with a low CADD_pherd score of 1.841. The GRIK3 promoter variant rs2134846 is part of the haplotype Hap1. Another GRIK3 promoter variant rs588902 with a high CADD_phred score of 22.3 is highly conserved and has the GERP score of 2.51. The exonic variant rs1337677 is a synonymous variant p.Pro229Pro from the first exon of GRIN3A that is in almost complete linkage with the haplotype Hap4 and has a medium CADD_phred score of 11.62. The GRIN2D 3′UTR variant rs8111684 has a CADD_phred score of 16.41.
Jpm 11 01250 g002
Figure 3. Score histograms and descriptive statistics of the haplotype scoring model. The top graph shows a histogram (%) of the scores distribution in both groups; SCH subjects (dark gray) and controls (light gray). In both groups, the haplotype scoring model did not deviate from the normal distribution. The bottom graph shows a boxplot of the scores in the SCH patients (dark gray) and the controls (light gray). There was a significant difference between the medians of the two groups (p < 0.00001). The mean score was 6.27 ± 1.58 in the SCH subjects and 3.69 ± 1.40 in the controls.
Figure 3. Score histograms and descriptive statistics of the haplotype scoring model. The top graph shows a histogram (%) of the scores distribution in both groups; SCH subjects (dark gray) and controls (light gray). In both groups, the haplotype scoring model did not deviate from the normal distribution. The bottom graph shows a boxplot of the scores in the SCH patients (dark gray) and the controls (light gray). There was a significant difference between the medians of the two groups (p < 0.00001). The mean score was 6.27 ± 1.58 in the SCH subjects and 3.69 ± 1.40 in the controls.
Jpm 11 01250 g003
Figure 4. BrainAGE index and haplotype scoring model. The graph shows the distribution of haplotype scores and BrainAGE indexes in both groups; SCH (red) and controls (gray). The arrows mark the mean haplotype score 6.27 ± 1.58 and mean BrainAGE index 2.44 ± 5.54 in SCH, and score 3.69 ± 1.40 and BrainAGE 0.14 ± 3.94 in controls, respectively. The bottom table shows the mean values of BrainAGE index for each scores in both groups.
Figure 4. BrainAGE index and haplotype scoring model. The graph shows the distribution of haplotype scores and BrainAGE indexes in both groups; SCH (red) and controls (gray). The arrows mark the mean haplotype score 6.27 ± 1.58 and mean BrainAGE index 2.44 ± 5.54 in SCH, and score 3.69 ± 1.40 and BrainAGE 0.14 ± 3.94 in controls, respectively. The bottom table shows the mean values of BrainAGE index for each scores in both groups.
Jpm 11 01250 g004
Table 1. Sociodemographic and clinical Data for the first-episode schizophrenia-spectrum and healthy control groups.
Table 1. Sociodemographic and clinical Data for the first-episode schizophrenia-spectrum and healthy control groups.
FES (n = 63)HC (n = 32)p-Value
(Two-Tailed t-Test)
Age, years; mean (SD)24.62 (6.94)27.84 (6.96)0.0355
Female, No. (%)24 (38.10)17.00 (53.13)0.16
Male, No. (%)39 (61.90)15.00 (46.88)
Education, years; mean (SD)12.72 (2.84)16.74 (2.40)0.00 +
Schizophrenia, No. (%)38 (60.32)
Other schizophrenia-spectrum disorders *, No. (%)25.00 (39,68)
PANSS Positive Subscale; mean (SD)14.82 (4.91)
PANSS Negative Subscale; mean (SD)18.48 (6.46)
PANSS General Psychopathology Subscale; mean (SD)36.17 (8.87)
PANSS total; mean (SD)69.48 (17.35)
Duration of untreated psychosis (months); mean (SD)3.61 (5.05)
Age at disease onset; mean (SD)24.13 (6.97)
Chlorpromazine equivalents, mg/d; mean (SD)399.67 (192.03)
Duration of antipsychotic treatment (months); mean (SD)1.79 (2.96)
Note: FES, first-episode schizophrenia-spectrum patients; HC, healthy controls; PANSS, Positive and Negative Symptom Scale; SD, standard deviation; +, Pearson’s chi-squared test; *, the ICD-10 diagnosis of acute and transient psychotic disorders is congruent with the DSM-IV-defined brief psychotic disorder.
Table 2. Genetic variations of iGluR genes. The summary of detected genetic variations within the promoter regions (1500 bp upstream of a TSS), 5′UTRs, exons, introns, and 3′UTRs of sixteen iGluR genes. The rate of variation in specific parts of the genes is expressed as the frequency of variants per 1000 bp. The RefSeq transcript corresponds to the canonical Ensembl transcript. Note: the RefSeq transcript does not necessarily have to be a canonical transcript itself in the hg38 assembly database.
Table 2. Genetic variations of iGluR genes. The summary of detected genetic variations within the promoter regions (1500 bp upstream of a TSS), 5′UTRs, exons, introns, and 3′UTRs of sixteen iGluR genes. The rate of variation in specific parts of the genes is expressed as the frequency of variants per 1000 bp. The RefSeq transcript corresponds to the canonical Ensembl transcript. Note: the RefSeq transcript does not necessarily have to be a canonical transcript itself in the hg38 assembly database.
SampleChriGluR GeneRefSeqEnsembl CanonicalPromotor (Existing/Novel)Per 1000 bp5´UTR (Existing/Novel)Per 1000 bpEXONS (Existing/Novel)Missense/Synonymous (Novel)Frameshift VariantPer 1000 bpINTRONS (Existing/Novel)Per 1000 bp3´UTR (Existing/Novel)Per 1000 bp
SCH caseschr5GRIA1NM_001258022.1ENST00000518783.110/06.70/00.03/00/211.1(1563) 1522/414.94/12.0
Controlschr5GRIA1NM_001258022.1ENST00000518783.18/05.30/00.03/01/111.1(1423) 1396/274.53/01.2
SCH caseschr4GRIA2NM_000826.4ENST00000296526.71/00.70/00.03/01/201.1(349) 333/162.58/13.4
Controlschr4GRIA2NM_000826.4ENST00000296526.72/01.31/03.12/00/200.8(302) 297/52.26/02.3
SCH caseschrXGRIA3NM_000828.4ENST00000622768.44/02.70/00.02/01/100.7(540) 526/141.76/02.7
ControlschrXGRIA3NM_000828.4ENST00000622768.42/12.00/00.01/00/100.4(446) 440/61.51/00.5
SCH caseschr11GRIA4NM_000829.4ENST00000282499.15/14.01/02.34/00/401.5(1107) 1068/393.09/03.8
Controlschr11GRIA4NM_000829.4ENST00000282499.14/02.71/02.32/00/200.7(948) 932/162.68/03.4
SCH caseschr21GRIK1NM_001330994.2ENST00000327783.44/02.70/00.04/01/301.4(1620) 1589/314.11/03.4
Controlschr21GRIK1NM_001330994.2ENST00000327783.41/00.70/00.04/00/401.4(1339) 1317/223.40/13.4
SCH caseschr6GRIK2NM_021956.4ENST00000421544.10/00.00/00.04/11/4 (1)01.8(2816) 2756/604.25/03.2
Controlschr6GRIK2NM_021956.4ENST00000421544.10/00.00/00.04/01/301.5(2339) 2307/323.53/01.9
SCH caseschr1GRIK3NM_000831.4ENST00000373091.85/03.30/00.03/01/201.1(938) 912/264.113/02.1
Controlschr1GRIK3NM_000831.4ENST00000373091.86/04.00/00.02/01/100.7(602) 592/102.64/00.6
SCH caseschr11GRIK4NM_014619.4ENST00000527524.210/17.31/03.57/01/602.4(1976) 1938/384.210/03.8
Controlschr11GRIK4NM_014619.4ENST00000527524.26/04.01/03.56/01/502.1(1429) 1397/323.08/03.0
SCH caseschr19GRIK5NM_001301030.1ENST00000301218.43/02.00/00.02/01/100.7(65) 63/21.11/03.0
Controlschr19GRIK5NM_001301030.1ENST00000301218.43/12.70/00.02/12/1 (1)01.0(50) 49/10.81/03.0
SCH caseschr9GRIN1NM_001185090.2ENST00000371553.32/01.30/00.03/00/301.1(67) 67/02.60/00.0
Controlschr9GRIN1NM_001185090.2ENST00000371553.32/01.30/00.02/00/200.7(60) 56/42.40/00.0
SCH caseschr16GRIN2ANM_001134407.3ENST00000330684.328/018.70/00.07/1(1) 4/401.8(2162) 2112/505.231/03.2
Controlschr16GRIN2ANM_001134407.3ENST00000330684.317/011.30/00.06/04/201.4(1842) 1816/264.529/03.0
SCH caseschr12GRIN2BNM_000834.4ENST00000609686.14/02.72/219.112/10/13 (1)02.9(1779) 1730/494.3134/16.0
Controlschr12GRIN2BNM_000834.4ENST00000609686.14/02.72/09.510/00/1002.2(1449) 1430/193.5107/14.8
SCH caseschr17GRIN2CNM_000835.5ENST00000293190.51/00.71/06.86/03/211.6(31) 30/12.30/00.0
Controlschr17GRIN2CNM_000835.5ENST00000293190.51/11.30/00.07/03/401.9(27) 27/02.01/02.4
SCH caseschr19GRIN2DNM_000836.2ENST00000263269.35/03.31/011.43/00/300.8(109) 107/22.41/01.0
Controlschr19GRIN2DNM_000836.2ENST00000263269.32/01.31/011.43/01/200.8(87) 84/31.91/01.0
SCH caseschr9GRIN3ANM_133445.2ENST00000361820.36/04.06/010.018/08/1005.4(760) 752/84.714/13.9
Controlschr9GRIN3ANM_133445.2ENST00000361820.34/02.75/08.315/06/904.5(603) 597/63.713/23.9
SCH caseschr19GRIN3BNM_138690.3ENST00000234389.313/210.00/00.025/116/9 (1)18.3(38) 36/26.30/00.0
Controlschr19GRIN3BNM_138690.3ENST00000234389.311/07.30/00.019/010/906.1(29) 29/04.80/00.0
Table 3. Missense variants of iGluR genes with assigned deleterious effect. The list shows nonsense, frameshift, and missense variants of iGluR genes. Missense variants were classified as deleterious by at least five prediction tools and must have a CADD_phred value above 20. The corresponding minor allele frequencies (MAF) in the SCH subjects and the healthy controls with assessed p-values are shown (ND = not determined due to low frequency of the variant in both groups).
Table 3. Missense variants of iGluR genes with assigned deleterious effect. The list shows nonsense, frameshift, and missense variants of iGluR genes. Missense variants were classified as deleterious by at least five prediction tools and must have a CADD_phred value above 20. The corresponding minor allele frequencies (MAF) in the SCH subjects and the healthy controls with assessed p-values are shown (ND = not determined due to low frequency of the variant in both groups).
IDhg38Transcript (Ensembl)ConsequenceiGluR GeneExonRef. AlleleALT AlleleSCH Cases, MAFControls, MAFp-Value *
rs145573036chr4:157336626ENST00000296526.12p.Glu575GlnGRIA211/16GC0.00790.0000ND
rs3841128chr5:153492228ENST00000518783.1p.Leu11ProfsTer13GRIA11/16C0.06350.07810.7643
rs146865938chr5:153650491ENST00000518783.1p.Arg218CysGRIA14/16CT0.00000.01560.3368
rs71509734chr9:101573337ENST00000361820.6p.Asn1062SerGRIN3A9/9TC0.00790.0000ND
chr9:101737814ENST00000361820.6p.Gln56 *GRIN3A1/9GA0.00790.0000ND
rs34755188chr9:101670973ENST00000361820.6p.Arg480HisGRIN3A3/9CT0.00790.03130.2628
rs10989589chr9:101670953ENST00000361820.6p.Gly487ArgGRIN3A3/9CT0.40480.48440.3528
rs41297895chr11:120819909ENST00000527524.8p.Ala167GlyGRIK46/21CG0.00000.01560.3368
rs137906208chr11:120861999ENST00000527524.8p.Arg262HisGRIK49/21GA0.00790.0000ND
rs765016248chr17:74843525ENST00000293190.10p.Gln871ArgGRIN2C13/13TC0.01590.00000.5509
rs754674133chr17:74844496–74844498ENST00000293190.10p.Lys788ThrfsTer13GRIN2C12/13TC0.00790.0000ND
rs375174698chr17:74850716ENST00000293190.10p.Arg389CysGRIN2C5/13GA0.00000.01560.3368
rs545736648 #chr19:1007862ENST00000234389.3p.Lys738GlnfsTer34GRIN3B5/9C0.02460.01610.7099
rs201638380chr19:1004804ENST00000234389.3p.Glu435LysGRIN3B3/9GA0.00790.0000ND
rs10666583chr19:1004897ENST00000234389.3p.Gly466AlafsTer18GRIN3B3/9CGTT0.30160.31250.8694
rs566603277chr19:1005200ENST00000234389.3p.Gly567SerGRIN3B3/9GA0.00790.0000ND
rs765625485chr19:1005375ENST00000234389.3p.Ala625ValGRIN3B3/9CT0.00000.01560.3368
rs2285906chr19:1008684ENST00000234389.3p.Ala845ThrGRIN3B7/9GA0.20630.09380.0643
rs61744375chr19:1008705ENST00000234389.3p.Glu852LysGRIN3B7/9GA0.00790.0000ND
rs137867437chr19:42062790ENST00000301218.8p.Ser104ProGRIK53/19AG0.00000.01560.3368
# The frameshift variant is in the GC-rich region with low coverage and lower confidence of variant calling; the ALT allele frequency in the Czech popuation is estimated to be 0.02 (non-published data). * The p-value was calculated using Fisher’s exact test.
Table 4. Haplotypes and SNPs revealed using the Fst estimate. Each haplotype is characterized by its SNPs, location, frequency of risk allele in the SCH subjects and the controls, and the corresponding Fisher exact test’s p-value. When possible, the odds risk ratio (OR) was calculated for each haplotype, as well as the likelihood ratio (LR) for specific genotypes. The LR suggests an additive model of the risk allele effect for most haplotypes. Variants that exhibit high conservation or colocalize with some regulatory elements or transcription factor binding sites are shown with appropriate symbols and are considered as functional candidate variants. An additional four SNPs are listed and characterized based on their Fst estimate above 0.05.
Table 4. Haplotypes and SNPs revealed using the Fst estimate. Each haplotype is characterized by its SNPs, location, frequency of risk allele in the SCH subjects and the controls, and the corresponding Fisher exact test’s p-value. When possible, the odds risk ratio (OR) was calculated for each haplotype, as well as the likelihood ratio (LR) for specific genotypes. The LR suggests an additive model of the risk allele effect for most haplotypes. Variants that exhibit high conservation or colocalize with some regulatory elements or transcription factor binding sites are shown with appropriate symbols and are considered as functional candidate variants. An additional four SNPs are listed and characterized based on their Fst estimate above 0.05.
iGluR GeneHaplotypehg38rsLocationAlleleFstSCH Cases freqControls freqp-Value 1ORGenotypeLRGenotypeLRGenotypeLR
GRIK3Hap1chr1:37014543rs3753776 &,§intron 1T0.0712810.3810.1880.00802.67wt/wt0.554wt/Hap11.904Hap1/Hap12.270
chr1:37017453rs2359647intron 1T0.071281
chr1:37023779rs6686296 #intron 1G0.071281
chr1:37028109rs59729868 £intron 1A0.071281
chr1:37031920rs3845491intron 1T0.083134
chr1:37035122rs2134846promoterA0.083134
GRIA1Hap2chr5:153527505rs566577 ¥,£intron 2C0.0627820.1830.0470.01264.54wt/wt0.754wt/Hap22.872Hap2/Hap2ND
chr5:153532297rs1493395intron 2G0.062782
chr5:153534826rs1908100intron 2T0.062782
GRIK2 chr6:101460819rs2518230intron 4T0.0550460.5480.3440.00912.31wt/wt0.473wt/Hap31.227Hap3/Hap32.032
Hap3chr6:101462357rs2245037 &intron 4G0.055046
chr6:101467933rs2579937intron 4A0.056160
GRIN3AHap4chr9:101711604rs2485530 #,§,¥,£intron 2C0.0507610.0790.0000.0174NDwt/wt0.841wt/Hap4NDHap4/Hap4ND
chr9:101714448rs2506363intron 1C0.050761
chr9:101715378rs2210991intron 1C0.055466
chr9:101716352rs2506364intron 1A0.050761
chr9:101717714rs2065965intron 1G0.050761
chr9:101720595rs7849059intron 1A0.050761
GRIK4Hap5chr11:120746795rs1945010 ¥,£intron 3C0.0581870.6830.5000.01772.15wt/wt0.361wt/Hap50.845Hap5/Hap52.027
chr11:120747370rs1939673 #,§,¥,£intron 3T0.051846
GRIK4Hap6chr11:120927187rs5795249intron13T0.1015370.2220.0470.00165.81wt/wt0.683wt/Hap63.372Hap6/Hap6ND
chr11:120927920rs1893816intron13T0.101537
chr11:120930476rs2186620 ¥,£intron13C0.094634
GRIN2Bnonechr12:13634912rs2160519intron 4C0.0571780.087 (C)0.000 (C)0.0171NDT/T0.825T/CNDC/CND
GRIN2Bnonechr12:13722646rs2110984intron 3T0.0802490.825 (T)0.641 (T)0.00662.65C/C0.508C/T0.481T/T1.985
GRIN2B chr12:13728086rs2284412 #,¥,£intron 3T0.0541350.3170.1560.02282.51wt/wt0.669wt/Hap71.419Hap7/Hap7ND
Hap7chr12:13741902rs10772710intron 3A0.054135
chr12:13744486rs2268121 $,¥,£intron 3C0.054135
GRIN2Bnonechr12:13871835rs10845851intron 1G0.0615790.095 (G)0.000 (G)0.0072NDA/A0.825A/GNDG/GND
GRIN2Dnonechr19:48437650rs56125279intron 10G0.0511180.190 (G)0.063 (G)0.01813.53C/C0.762C/G2.288G/GND
GRIK1Hap8chr21:29925689rs2832486intron 1T0.0520220.7300.5630.02272.10wt/wt0.308wt/Hap80.789Hap8/Hap81.808
chr21:29926886rs2832488intron 1T0.052022
chr21:29927152rs2832489intron 1A0.052022
chr21:29929318rs2832490intron 1G0.052022
chr21:29930154rs2255821intron 1A0.052022
chr21:29930365rs767253 ¥,£intron 1C0.052022
chr21:29931426rs2300333intron 1G0.052022
chr21:29931851rs2832492intron 1C0.052022
chr21:29933351rs9982938 ¥,£intron 1T0.052022
chr21:29933357rs9982699 ¥,£intron 1C0.052022
chr21:29933999rs3831792intron 1AATATAAATG0.052022
chr21:29934461rs2243533intron 1T0.052022
chr21:29934499rs2243535intron 1T0.052022
chr21:29934740rs2243435intron 1T0.052022
chr21:29934751rs2243436intron 1C0.052022
chr21:29936109rs2832494intron 1C0.052022
chr21:29937434rs2832495intron 1C0.052022
chr21:29937912rs2409355intron 1C0.052022
chr21:29938441rs2832496intron 1T0.052022
chr21:29938848-29938850rs66799775intron 1G0.052022
chr21:29938853rs75584618 &intron 1T0.052022
chr21:29938854rs77191337intron 1T0.052022
1 Fisher exact test statistic value. & highly conservative (GERP > 1.8); # DNaseI hypersensitivity peak cluster ENCODE (UCSC); $ enhancer (Ensembl); § regulatory region ORegAnno (UCSC); ¥ regulatory region (Ensembl); £ transcription factor binding site (Ensembl); OR = Odds ratio; LR = Likelihood ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Hirschfeldova, K.; Cerny, J.; Bozikova, P.; Kuchtiak, V.; Rausch, T.; Benes, V.; Spaniel, F.; Gregus, D.; Horacek, J.; Vyklicky, L.; Balik, A. Evidence for the Association between the Intronic Haplotypes of Ionotropic Glutamate Receptors and First-Episode Schizophrenia. J. Pers. Med. 2021, 11, 1250. https://doi.org/10.3390/jpm11121250

AMA Style

Hirschfeldova K, Cerny J, Bozikova P, Kuchtiak V, Rausch T, Benes V, Spaniel F, Gregus D, Horacek J, Vyklicky L, Balik A. Evidence for the Association between the Intronic Haplotypes of Ionotropic Glutamate Receptors and First-Episode Schizophrenia. Journal of Personalized Medicine. 2021; 11(12):1250. https://doi.org/10.3390/jpm11121250

Chicago/Turabian Style

Hirschfeldova, Katerina, Jiri Cerny, Paulina Bozikova, Viktor Kuchtiak, Tobias Rausch, Vladimir Benes, Filip Spaniel, David Gregus, Jiri Horacek, Ladislav Vyklicky, and Ales Balik. 2021. "Evidence for the Association between the Intronic Haplotypes of Ionotropic Glutamate Receptors and First-Episode Schizophrenia" Journal of Personalized Medicine 11, no. 12: 1250. https://doi.org/10.3390/jpm11121250

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop