Genetic Analysis of the ZNF804A Gene in Mexican Patients With Schizophrenia, Schizoaffective Disorder and Bipolar Disorder

Background: Evidence suggests that Schizophrenia (SCZ), Schizoaffective Disorder (SAD) and Bipolar Disorder (BPD) share genetic risk variants. ZNF804A gene has been associated with these disorders in different populations. The aim was to analyze the rs1344706 SNP and identify common and rare variants in a targeted region of the ZNF804A gene in SCZ, BPD and SAD Mexican patients compared with a control group. Methods: We genotyped the rs1344706 in 228 Mexican patients diagnosed with SCZ, SAD and BPD, and 295 controls. An additional sample of 167 patients and 170 controls was analyzed to identify rare and common variants using the Sanger-sequence analysis of a targeted region of ZNF804A gene. Results: We replicated the association between the rs1344706 and SCZ, SAD and BPD. In the sequence analysis, we did not identify rare variants; however, we identied three common variants: rs3046266, rs1366842 and rs12477430. A comparison of the three identied variants between SCZ, SAD and BPD did not show statistical differences. Conclusions: Our ndings support the evidence suggesting that ZNF804A gene is involved in the etiology of SZC, SAD and BPD. Future studies are needed to identify the pleiotropic effect of this gene in psychiatric disorders.


Background
Schizophrenia (SCZ), Schizoaffective Disorder (SAD) and Bipolar Disorder (BPD) are chronic and serious mental disorders that have been recognized as an important public health matter. These disorders are important worldwide contributors to years lived with disability, largely due to the young age at onset, high disability burden, and the often-chronic course of illness. Patients with SCZ, SAD and BPD demand a disproportionate amount of health resources and have premature mortality because of comorbid cardiovascular diseases or suicide (Correll et al. 2017).
SCZ and BPD are complex disorders and different genetic studies have found that genes represent around the 80% of the variance of the population at risk. However, the genetic risk variants (either common or rare) of these disorders are still under scrutiny. Previous studies have suggested that there is an overlap between SCZ, SAD and BPD genetic vulnerability, suggesting that these disorders shared genetic risk variants. ZNF804A gene has been importantly associated with SCZ, SAD and BPD in different populations, speci cally the common variants of the gene (Sun et al. 2015). Nevertheless, rare variants of ZNF804A may also play an important role in the risk of developing these disorders. In 2010, a study in UK population did not nd a signi cant association between rare variants of the ZNF804A gene and SCZ (Dwyer et al. 2010).
ZNF804A gene is known to be involved in different neuronal processes that are distorted in patients with SZC and BPD, such as embryonic neurodevelopment, neuroprotection, neurogenesis, neuronal migration, synaptic formation and function (Chang et al. 2017;Deans et al. 2016;. Studies have found evidence that sustain that ZNF804A protein might be implicated in pre-messenger RNA processing, RNA translational control and other genes expression regulation, including some genes involved in SZC, synaptic formation, neurodevelopment and in ammation (Chapman et  ZNF804A gene is located on chromosome 2q32.1 and is organized in 4 exons that encode for a protein with a zinc nger domain in the N-terminus. This domain allows the protein to bind to DNA, RNA or other proteins so it can regulate gene expression, RNA processing and protein translation. The coding sequence for ZNF804A gene consists in 542 nt in exon 1, 114 nt in exon 2, 131 nt in exon 3 and 3710 nt in exon 4 (Ensembl Genome Browser 2021).
Several common variants have been identi ed in the ZNF804A gene. The rs1344706 SNP (Single Nucleotide Polymorphism) is the most importantly common variant that has been associated with SZC, SAD and BPD. This SNP is located in the second intron of the gene and consists in a substitution of adenine for cytosine. O'Donovan et al. (2008) found an association between SCZ and BPD and rs1344706 in a GWAS, proposing the A allele as a risk factor (O'Donovan et al. 2008). The association of rs1344706 with SZC, SAD and BDP has been replicated in different studies and populations (Sun et al. 2015). It has been suggested that rs1344706 SNP may alter the gene expression of the ZNF804A modifying transcription activity or splicing factors binding to mRNA, or by acting as a cis variant that interacts with other polymorphisms that are in linkage disequilibrium (Hess & Glatt, 2014). In addition, rs1344706 has been associated with alterations in white or gray matter, brain connectivity and cognitive functions in patients with SCZ or BPD and in controls (Chang et al. 2017;Wang et al. 2019). Also, this variant has been associated with symptoms severity and response to treatment in both disorders (Wickramasinghe et al. 2015). Furthermore, rs1344705 has also been associated with psychosisproneness by indirectly impacting on a tendency to misconstrue external or internal cues, which is hypothesized as central feature for presenting psychotic experiences (Stefanis et al. 2013).
Another ZNF804A SNP that has been associated with SCZ and BPD is the rs1366842, located in exon four and consists in a change of cytosine to adenine, resulting in a threonine to lysine change in the amino acid sequence (Baek et al. 2017). Therefore, the aim of the present study was to analyze the rs1344706 of the ZNF804A gene and identify common or rare variants by sequence analysis of a 352 bp targeted region around the rs1366842 SNP in Mexican patients with SCZ, BPD and SAD compared with a control group.

Sample
We included a total of 228 Mexican psychiatric outpatients who met the DSM-IV-TR and DSM5 criteria for SCZ, SAD and BPD type I, using the Mini-International Neuropsychiatric Interview (MINI) (Bobes, 1998) and corroborated by at least two certi ed psychiatrists of the Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz. Inclusion criteria were: 1) Diagnostic Stability of two years; 2) Age ≥ 18 years; 3) Age of onset of the disorder < 50 years; and 4); Mexican patients, with parents and grandparents born in Mexico.
Also, we included 295 controls with the following inclusion criteria: Mexican subjects with parents and grandparents born in Mexico; individual without personal or family history of mental disorders evaluated with the MINI scale.
For the sequence analysis, we include an additional sample of 167 patients composed of 73 (44%) patients with SCZ, 23 (14%) with SAD and 71 (42%) with BPD, with the following characteristics: BPD patients with psychotic symptoms at least once in their life, family history of SCZ, SAD or BPD type I (at least one relative with a maximum of third degree), and patients carriers of A allele of the rs1344706 SNP. Also, we included 170 controls. All participants in the additional samples met the previous inclusion criteria.

Genetic analysis
Genotyping Genomic DNA was extracted from blood lymphocytes using the Genomic DNA Puri cation Kit FlexiGene® (QIAGEN). Genotyping of rs1344706 was performed using a TaqMan SNP Genotyping Assay by-Design (C_2834835_10). Genotypes were analyzed by the TaqMan allele speci c assay method using the ABI Prism® 7500 Sequence Detection System according to the manufacturer´s protocols (Applied Biosystems Inc., Foster City, CA) with a reaction mixture consisting of 20 ng of genomic DNA, 2.5 µl of TaqMan Universal Master Mix, and 0.125 µl of TaqMan SNP Genotyping Assay (Applied Biosystems Inc.).
After denaturing at 95 ° C for 10 min, 50 cycles of PCR were performed under the following conditions: denaturing at 95°C for 15 s, annealing and extension at 60°C for 1min.

Sequence analysis
Sequencing of the targeted region of exon four of the ZNF804A was performed in a sequence of 352 bp region spanning from 184937315 bp to 184937667 bp. The fragment was ampli ed using the oligonucleotide forward: 5'-AATGCTGGGAAATATCTATTGGAAC-3 'and reverse: 5'-GTGAATGTTGTCTACGTTTTCGAT-3', designed with the Primer-BLAST software (Ye et al., 2012). A 2% agarose gel electrophoresis was performed using 1X TAE as a running buffer and SYBR® Safe DNA Gel Stain as a dye to visualize the PCR product under UV light. The ampli ed product was puri ed using the QIAquick® PCR Puri cation Kit (QIAGEN) and its concentration and purity were evaluated by spectrophotometric analysis (NanoDrop ™ 2000). The BigDye ™ Terminator v3.1 Cycle Sequencing Kit reagent was used to perform the Sanger Sequencing, and the Centri-Sep ™ Spin Columns were used to remove the reaction residues. The sequences were obtained in the 3500XL Genetic Analyzer sequencer and, to identify the genetic variants, the sequencing results were compared with the reference sequence using the BioEdit Sequence Alignment Editor v7.1.11 software, the BLAST alignment tool, and the website of the National Center for Biotechnology Information-NCBI. All genotyping and sequencing were performed blind to sample outcome.

Statistical Analysis
The categorical data were analyzed using Chi-square or Fisher's exact test. Differences in dimensional variables were evaluated by Student's t-test. Data are presented as means ± SD, frequencies and percentages. Analysis was performed using IBM SPSS Statistics 20 program. Analysis of linkage disequilibrium (LD) and haplotype analysis was performed with the SHEsis software (Shi & He, 2005).
Most of the patients were single (70.2%) and 28.1% were unemployed. Average age and years of education were 41.1 ± 11.8 years and 12.4 ± 3.9 years, respectively. 86.8% of the patients had family history of psychiatric disorders, the most frequent was mayor depressive disorder (37.7%), followed by alcohol use disorders (31.1%). 64.9% of the patients had comorbidity with other psychiatric disorder. In the case of SZC, mayor depressive disorder was the most frequent comorbidity (55.4%). When considering all the patients, the most frequent comorbid conditions were anxiety disorders (28.9%) and substance use disorders (26.3%). 61.4% and 24.1% of the patients reported previous suicidal ideation and attempt, respectively. The age of onset of the disorder, the years of course and the number of hospitalizations were 22.9 ± 8.6, 18.2 ± 10.5, and 2 ± 2.2, respectively.

Genetic Analysis
The genotype and allele frequencies of the rs1344706 in the patients with SCZ, SAD and BPD showed signi cant differences compared with the control subjects (Table 1). In particular, A allele was associated with an increased risk for presenting SCZ, SAD and BPD (OR = 1.35, CI 95% = 1.054-1.739, p = 0.0208). The SNP was in Hardy-Weinberg equilibrium (p ≥ 0.05). In the sequence analysis, we identi ed a total of 3 genetic variants that had been previously reported in other populations. Of these, 2 were SNPs and 1 was a duplication. We did not identify rare variants in the sample analyzed in the present study. In Table 2, we showed the common variants identi ed in the sequence analysis among the groups ( Table 2).

MAF = Minor Allele Frequency
The SNP based-analysis of the common variants identi ed in the sequence analysis did not show differences between cases and controls ( Table 3). The three variants were in Hardy-Weinberg equilibrium (p ≥ 0.05). Also, the analysis of the common variants by sex did not show differences between cases and controls (data not shown).  We observed a haplotype block composed of rs3046266, rs1366842 and rs12477430 (D′=1, r 2 = 1). The haplotype analysis did not show statistical differences between cases and controls (data not shown).

Discussion
Previous studies reported association between the ZNF804A gene and SZC, SAD and BPD, particularly with the common variant rs1344706. Our nding supports that in Mexican population the rs1344706 is also associated with SCZ, SAD and BPD. It has been suggested that this SNP can alter the expression of the ZNF804A gene at different stages of development, affecting neuronal growth, dendritic spine maintenance, and synaptic function; processes that have been related to the pathophysiology of these disorders It is important to analyze if rs1344706 in LD with other SNPs localized in the same gene and/or the interaction between ZNF804A and other genes may be related with the etiology of these disorders. A bioinformatic study found an interaction between rs1344706 and rs11685645 of Myt1L gene, and between rs1344706 and rs1836796 of NCAM1 gene, concluding that these SNPs may be implicated in the early development of central nervous system (Hess et al. 2015;Riley et al. 2010). Also, interaction study between rs1344706 and a SNP of CACNA1C reported association with increase activation in brain regions of SCZ and BPD patients (Tecelão et al. 2019). In addition, it was reported an interaction effect of rs1344706 and COMT variant on brain activity in healthy individuals (Cui et al. 2019). However, more studies that analyze the interaction between rs1344706 and other SNPs are required. On the other hand, we do not know how these gene variants interact with the environment, in order to understand the biological relationship between SZC, SAD and BPD and the ZNF804A gene.
Genetic studies carried out in Caucasian and Asian populations reported association with the rs1388842; however, we did not nd an association in Mexican patients with SCZ, BPD and SAD. Also, we did not nd an association between rs3046266 and rs12477430 with these disorders. The present study replicates the previous ndings of negative association between rs12477430 and SCZ in Irish population (Riley et al. 2010 Our results replicate that ZNF804A gene is involved in the genetic risk of SCZ, SAD and BPD. However, much research is still required, as obviously the mechanisms involved in the development of these highly heterogeneous disorders are more complex. The differences between these disorders and even between individuals with the same disorder are probably explained by the numerous combinations of the different genetic and environmental risk factors that interact with each other; coupled with the sum of different additional modifying factors, either genetic and/or environmental, that do not in uence the risk of developing a disorder, but in uence its clinical presentation; explaining the di culty to understand the biological role of the ZNF804A gene in this disorders.
In addition, we consider that in the future it would also be important to study the role of ZNF804A gene in other psychiatric disorders, principally neurodevelopmental disorders, or those with psychotic symptoms, such as autism, primary delusional disorder, major depressive disorder with psychotic symptoms and substance-induced psychotic disorders.
The present study has several limitations. First, the small sample size in our analysis resulted in limited power to identify rare variants, principally ultra-rare variants. Second, we only sequenced a targeted region of exon four of the gene, consequently the existence of rare variants in other regions of the ZNF804A gene in SZC, SAD and BPD is possible; therefore, our study did not exclude the involvement of rare variants in the risk of presenting these disorders in Mexican population. It will be important sequencing the complete gene and replicate the association between rs1344706 and SZC, SAD and BPD in a large sample size.

Conclusions
In conclusion, our ndings support the evidence suggesting that ZNF804A gene is involved in the etiology of SZC, SAD and BPD. Future studies are needed to identify the pleiotropic effect of this gene in psychiatric disorders.