Functional Analysis of DNMT1 SNPs (rs2228611 and rs2114724) Associated with Schizophrenia

Saxena, Sonal; Maroju, Pranay Amruth; Choudhury, Sumana; Voina, Vidhya Chitta; Naik, Poonam; Gowdhaman, Kavitha; Kkani, Poornima; Chennoju, Kiranmai; Ganesh Kumar, S.; Ramasubramanian, C.; Prasad Rao, G.; Jamma, Trinath; Narayan, Kumar Pranav; Mohan, K. Naga

doi:https://doi.org/10.1155/2021/6698979

Genetics Research

On this page

Abstract Introduction Materials and Methods Results and Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2021 | Article ID 6698979 | https://doi.org/10.1155/2021/6698979

Functional Analysis of DNMT1 SNPs (rs2228611 and rs2114724) Associated with Schizophrenia

Sonal Saxena,¹Pranay Amruth Maroju,¹Sumana Choudhury,¹Vidhya Chitta Voina,¹Poonam Naik,¹Kavitha Gowdhaman,¹Poornima Kkani,²Kiranmai Chennoju,¹S. Ganesh Kumar,³C. Ramasubramanian,³G. Prasad Rao,⁴and Trinath Jamma¹ et al.

Academic Editor: Josephina Meester

Received25 Nov 2020

Revised10 Feb 2021

Accepted17 Mar 2021

Published01 Apr 2021

Abstract

A recent study showed the association of minor alleles of rs2228611 (T allele) and rs2114724 (T allele) of DNMT1 with schizophrenia (SZ) and suggested their effects on splicing of the transcripts. We performed a replication study using 310 controls and 304 SZ patients and confirmed the association of the homozygous minor allele genotypes with SZ ( = 0.04 for rs2114724 and = 0.007 for rs2228611). This significant association persisted after Bonferroni correction when the previously published data of 301 controls and 325 patients were also considered ( ≤ 0.0002). In addition, we found that the proportion of male patients with homozygous minor alleles at rs2114724 was significantly higher than that of females ( = 0.002). When haplotype analysis of both loci was performed, we observed a significant association of T/T–T/T and T/T–C/T ( = 0.04) haplotypes with SZ. To gain insights into the functional effects of the two SNPs on the levels of DNMT1 transcripts, quantitative real-time PCR experiments were performed using peripheral blood monocytes from 10 individuals each with T/T–T/T (homozygous minor allele), C/T–C/T (heterozygous), and C/C–C/C (homozygous major allele) haplotypes. Independently, the levels of DNMT1 protein were also compared in three individuals each by immunofluorescence. These results suggest that neither DNMT1 transcript nor the protein levels were significantly different in the peripheral blood monocytes among the individuals studied for the three groups. Taken together, our results confirm that the two minor alleles in homozygosity are associated with SZ but with no discernible effects on transcript or protein levels of DNMT1 in the peripheral blood monocytes of the small number of samples tested.

1. Introduction

Schizophrenia (SZ) is a complex disorder with a worldwide incidence of ∼0.7% and multiple etiologies that include genetic, environmental, and epigenetic mechanisms [1–4]. A number of family and case-control studies identified many single-nucleotide polymorphisms (SNPs) and copy-number variants (submicroscopic deletions and duplications) as significantly associated with SZ [5, 6]. Evidence for the involvement of epigenetic mechanisms in SZ comes from twin studies wherein monozygotic (MZ) twins show 41–65% concordance [1]. Of the different epigenetic processes in mammals, DNA methylation which involves addition of methyl group at the 5^th carbon of cytosine is the most well studied [7]. DNA methylation is established by de novo methyltransferases DNMT3A and DNMT3B and maintained by DNMT1. DNMT3L has no catalytic activity but regulates de novo methylation [8].

Recently, SNPs of DNMTs were studied among SZ patients and controls from South India [9]. These results showed a significant association of minor alleles of DNMT1 (rs2114724, = 0.02, and rs2228611, = 0.002), DNMT3B (rs2424932 in SZ males, = 0.006; rs1569686 in early onset as well as family history, = 0.03), and DNMT3L (rs2070565 in an early onset of SZ as well as family history, = 0.003). Among these, the minor alleles of rs2114724 and rs2228611 (Figure 1(a)), which are the focus of the current study, have been suggested to affect splicing of the DNMT1 transcript. Here, we performed a replication study to validate the association of the two DNMT1 minor alleles using an independent collection of 310 controls and 304 patients. Because defectively spliced or unspliced mRNAs in eukaryotic cells are subjected to degradation and an overall reduction of the transcript levels [10], we compared the levels of DNMT1 transcripts and proteins among individuals with three haplotypes (homozygous for both minor alleles, homozygous for both major alleles, and heterozygous). The results are discussed here.

(a)

(b)

2. Materials and Methods

2.1. Patient and Control Samples

This work was approved by the Institutional Human Ethics Committees of the participating institutions. Normal controls and SZ patients were identified by qualified psychiatrists using DSMV criteria [11]. The controls were age- and sex-matched and did not have any neuropsychiatric conditions. After obtaining informed consent, 5–10 ml blood was drawn by venipuncture and the peripheral blood monocytes were isolated by using Histopaque 1077 (Merck, USA) solution. The purified monocytes were washed with 1X phosphate-buffered saline (1X PBS: 37 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, and 1.8 mM KH₂PO₄), used for isolation of genomic DNA using SDS-proteinase K method [12] and total RNA using RNeasy Mini kit (Qiagen, Germany). Total RNAs were reverse-transcribed using SuperScript IV cDNA synthesis kit (Thermo Fisher, USA), and the resultant cDNAs were amplified using primers for ACTIN and DNMT1 by real-time PCR (see below).

2.2. SNP Genotyping and Analysis

Genomic DNAs from 20 normal individuals were used to amplify regions containing the rs2114724 and rs2228611 loci. Primer sequences and annealing temperatures are given in Supplementary Table 1. For amplification, 10 ng each of the genomic DNAs was used with 5 µM each of the primers in a volume of 10 µl. Amplification conditions include one cycle of denaturation at 95^°C for 3 min, followed by 40 cycles of PCR (95⁰C: 30 sec; 55°C: 30 sec; 68°: 30 sec) and one cycle of final extension at 68°C for 3 min. The PCR products were gel-purified using Qiaquick Gel-purification kit (Qiagen) and sequenced by Sanger’s dideoxy sequencing method. Using the sequencing data, individuals with homozygous major allele, homozygous minor allele, and heterozygous genotypes at both loci were identified. The corresponding genomic DNAs were used to develop Amplification Refractory Mutation System (ARMS) by designing primers (Supplementary Table 1) that can distinguish the major and minor alleles by PCR [13]. Amplification conditions include one cycle of denaturation at 95°C for 3 min, followed by 40 cycles of PCR (95°C: 30 sec; 55°C: 15 sec; 68°C: 20 sec) and one cycle of final extension at 68°C for 3 min.

Validation of the ARMS-PCR primers was performed using the same genomic DNA samples that were sequenced. Following validation, genomic DNAs from 310 controls and 304 patients were amplified by ARMS-PCR, and the observed amplification patterns were used to deduce the genotype and allele frequencies. The number of controls and patients with a specified genotype or haplotype was used to perform a Fisher’s t-test (https://www.graphpad.com/quickcalcs/contingency1/) using the two-tailed method and calculate odds ratios (https://www.medcalc.org/calc/odds_ratio.php) [14]. values of ≤0.05 were taken as significant. For a given genotype that is significantly associated with SZ, penetrance estimates were made using CalPen, an online tool [15].

2.3. Real-Time PCR Analysis

cDNAs were prepared from PBMCs of 10 normal individuals each with the three haplotypes (homozygous for major alleles at both loci, heterozygous for both loci, and homozygous for minor alleles at both loci) and amplified with primers for DNMT1 and β-ACTIN transcripts (Supplementary Table 1) using iTaq Universal SYBR Green Supermix (Bio-Rad, USA) and QuantStudio3 Real-time PCR System (Thermo Fisher, USA). DNMT1 expression was normalized with β-ACTIN as an endogenous control to obtain ΔCt values. The average of ΔCt values from ten individuals with homozygous major haplotype was used as a reference to calculate ΔΔCt for all the 30 samples with the three haplotypes. Relative abundances of transcripts were then determined by calculating 2^−ΔΔCt values [16]. One-way analysis of variance (ANOVA) was performed using Microsoft Excel to identify any significant differences in DNMT1 transcript levels among the three groups [17].

2.4. Immunocytochemistry

Purified PBMCs were washed and resuspended in 1X PBS and diluted to ∼10⁶ cells/ml. One millilitre of each of the cell suspensions was transferred into 6-well dishes containing coverslips and left stationary for 30 min at room temperature to allow attachment of cells [18]. The nonadherent cells were removed by aspiration, and the attached cells were fixed for 10 minutes with 500 μl of 4% paraformaldehyde (Merck, USA) added gently along the side of the well. The fixed cells were washed with 1X PBS, permeabilized with 500 μl of 0.5% Triton X-100 (HiMedia, India) for 10 min, washed with 1X PBS for 5 min, and blocked with 500 μl of 1% BSA (HiMedia, India) for 30 minutes. Primary antibodies for DNMT1 (Thermo Fisher, 60B1220.1) and HISTONE H3 (Cell Signaling Technologies, 4499) diluted in 1% BSA were added to the dishes and were incubated overnight at 4°C. The cells were washed again with 1X PBS and incubated with fluorescently labelled secondary antibodies (Cell Signaling Technologies, 4409 and 4412) for one hour at room temperature. Washings were repeated twice with 1X PBS, counterstained with DAPI (Thermo Fisher, D1306) for 5 min, washed thrice with 1X PBS, and mounted onto glass slides for visualization using a confocal microscope (Leica: TCSSP8). The fluorescence intensities were quantified using the Leica Las-X software, using HISTONE H3 signals as reference. After normalization of data for signals obtained with HISTONE H3, the relative signals for DNMT1 in 20 random cells for each individual were used for comparisons. The DNMT1 protein levels among individuals with the three haplotypes were compared by Student’s t-test (two-tailed), and ≤ 0.05 was taken as statistically significant.

3. Results and Discussion

We established an ARMS-PCR to achieve allele-specific amplification so that samples containing only minor alleles (homozygous minor alleles), only major alleles (homozygous major alleles), and both the alleles (heterozygous) at the two loci were identified reliably (Figure 1(b)). We employed ARMS-PCRs for genotyping 310 controls and 304 patients (Materials and Methods).

The genotypes of all 614 individuals were classified as homozygous major allele or homozygous minor allele or heterozygous, and the number of individuals for each genotype was tabulated (Table 1). From these genotypes, the minor allele frequencies were estimated for rs2114724 as 0.43 for controls and 0.45 for patients. The number of controls with the minor allele was not significantly different from patients. On the other hand, the number of patients with the minor allele for rs2228611 (frequency = 0.47) was significantly higher ( = 0.04) than controls (frequency = 0.41). Fisher’s t-tests were performed to identify whether any genotype frequency is significantly different between controls and patients. For example, 158 out of 310 controls have C/T genotype for rs2114724 with the remaining controls (152) having either C/C or T/T genotypes. In the case of patients, 125 out of 304 patients have C/T genotype and the remaining 179 have either C/C or T/T genotype. Using this information, when a two-tailed Fisher’s t-test was performed, a value of 0.01 was obtained, indicating that the proportion of C/T individuals among controls (51%) was significantly higher than that in patients (41%). The odds ratio for this genotype was 0.7, meaning that the odds for C/T genotype to be a control is higher than being schizophrenic. Similar analysis suggested that homozygous minor allele genotypes for rs2114724 ( = 0.04; odds ratio = 1.5253) and rs2228611 ( = 0.007; odds ratio = 1.7) were significantly higher in patients than in controls. A combined analysis with the previously published data [9] amounting to a total of 611 controls and 629 patients revealed a more significant association of the minor alleles of both loci with SZ after Bonferroni correction ( ≤ 0.0002) with individual odds ratios of 1.73 (rs2114724) and 1.83 (rs2228611), respectively.

We next examined whether there was any gender-specific difference in the distribution of the minor allele-associated genotypes in patients (Table 2). For rs2114724, the proportion of minor allele homozygous males (52 out of 170) was significantly higher ( = 0.002) than that in females (22 out of 134). There was no such gender-specific difference in the case of homozygous minor allele genotypes for rs228611.

Haplotype analysis between controls and patients revealed a significant association of homozygous minor allele genotype of rs2228611 (T/T) with either homozygous minor allele genotype ( = 0.04; odds ratio = 1.52) or heterozygous genotype of rs2114724 ( = 0.04; odds ratio = 4.7) in patients (Table 3).

Minor alleles of both rs2228611 and rs2114724 have been suggested to affect the splicing process of the DNMT1 mRNA [9]. Since mRNAs that are aberrantly spliced or fail to undergo successful splicing are unstable and degraded [10], we measured DNMT1 mRNA levels in peripheral blood monocytes of 10 individuals each with C/C–C/C, C/T–C/T, and T/T–T/T haplotypes. Although the quantitative real-time PCR data revealed variation in the transcript levels within a haplotype, there was no difference in the average DNMT1 transcript levels between the three haplotypes (Figure 2(a)).

(a)

(b)

(c)

As an independent approach, the DNMT1 protein levels were compared among three individuals each with the three haplotypes by immunocytochemistry (Figure 2(b)). As in the case of DNMT1 transcripts, the distribution of the relative DNMT1 protein amounts varied within a haplotype and there was no difference in the protein levels among the three haplotypes (Figure 2(c)). These results taken together with the transcript levels compared by real-time PCR indicate that the minor allele haplotype (T/T–T/T) may not be associated with any identifiable decrease in the DNMT1 transcript or protein levels.

4. Conclusions

Minor alleles of rs2114724 and rs2228611 have been reported to be associated with SZ, gastric, ovarian, and breast cancers, hypertension, and hearing loss [19–24]. The results reported here support a significant association of these alleles with SZ. Our observation is based on a total of 611 controls and 629 patients, and given the small size of samples studied, more studies are needed to confirm these findings. Contrary to the prediction that the minor alleles affect the splicing process, there was no discernible change in the transcript or protein levels in the PBMCs of individuals with only minor alleles at both loci. Since the sample size tested here is small, future studies involving a larger set of samples may identify events where DNMT1 transcript or protein levels are affected. It is also possible that the effects of these minor alleles are specific to neurons. In this context, a similar functional analysis on postmortem brain samples is warranted. The effects of these minor alleles may still end up to be too small to be detected experimentally even in the brain. Whether such small effects, if at all present, along with any other variants elsewhere in the DNMT1 gene or other genes result in an identifiable molecular abnormality related to SZ requires further investigation.

Data Availability

The data present in the manuscript does not include any files that need to be deposited in the public repositories. All necessary information on the data collected is given in the manuscript itself.

Conflicts of Interest

The authors do not have any competing financial interests.

Acknowledgments

KNM lab was supported by grants from DST-CSRI, OPERA, and the Centre for Human Disease Research. Immunofluorescence imaging was carried out in the Central Analytical Laboratory of BITS-Pilani, Hyderabad Campus. Sumana Choudhury is supported by a fellowship from the Centre for Human Disease Research.

Supplementary Materials

Supplementary Table 1: oligonucleotides used for sequencing, ARMS-PCRs, and real-time PCR analysis . (Supplementary Materials)

References

A. G. Cardno and Gottesman II, “Twin studies of schizophrenia: from bow-and-arrow concordances to Star Wars Mx and functional genomics,” American Journal of Medical Genetics, vol. 97, pp. 12–17, 2000.
View at: Publisher Site | Google Scholar
P. F. Sullivan, K. S. Kendler, and M. C. Neale, “Schizophrenia as a complex trait,” Archives of General Psychiatry, vol. 60, no. 12, pp. 1187–1192, 2003.
View at: Publisher Site | Google Scholar
S Akbarian, “Epigenetic mechanisms in schizophrenia,” Dialogues in Clinical Neuroscience, vol. 16, no. 3, pp. 405–417, 2014.
View at: Google Scholar
C. McDonald and R. M. Murray, “Early and late environmental risk factors for schizophrenia,” Brain Research Reviews, vol. 31, no. 2-3, pp. 130–137, 2000.
View at: Publisher Site | Google Scholar
S. Ripke, B. M. Neale, A. Corvin et al., “Biological insights from 108 schizophrenia-associated genetic loci,” Nature, vol. 511, pp. 421–427, 2014.
View at: Publisher Site | Google Scholar
D. Malhotra and J. Sebat, “CNVs: harbingers of a rare variant revolution in psychiatric genetics,” Cell, vol. 148, no. 6, pp. 1223–1241, 2012.
View at: Publisher Site | Google Scholar
A. Bird, “DNA methylation patterns and epigenetic memory,” Genes and Development, vol. 16, pp. 6–21, 2002.
View at: Publisher Site | Google Scholar
H. Gowher, K. Liebert, A. Hermann, G. Xu, and A. Jeltsch, “Mechanism of stimulation of catalytic activity of Dnmt3A and Dnmt3B DNA-(cytosine-C5)-methyltransferases by Dnmt3L,” Journal of Biological Chemistry, vol. 280, no. 14, pp. 13341–13348, 2005.
View at: Publisher Site | Google Scholar
K. R. Saradalekshmi, N. V. Neetha, S. Sathyan, I. V. Nair, C. M. Nair, and M. Banerjee, “DNA methyl transferase (DNMT) gene polymorphisms could be a primary event in epigenetic susceptibility to schizophrenia,” PLoS One, vol. 9, no. 5, pp. e98182–e98188, 2014.
View at: Publisher Site | Google Scholar
O. Jaillon, K. Bouhouche, J.-F. Gout et al., “Translational control of intron splicing in eukaryotes,” Nature, vol. 451, no. 7176, pp. 359–362, 2008.
View at: Publisher Site | Google Scholar
S. Saxena, P. Kkani, C. Ramasubramanian et al., “Analysis of 15q11.2 CNVs in an Indian population with schizophrenia,” Annals of Human Genetics, vol. 83, no. 3, pp. 187–191, 2019.
View at: Publisher Site | Google Scholar
M. R. Green and J. Sambrook, “Isolation of high-molecular-weight DNA using organic solvents,” Cold Spring Harbor Protocols, vol. 2017, no. 4, Article ID pdb.prot093450, 2017.
View at: Publisher Site | Google Scholar
S. Little, “Amplification‐refractory mutation System (ARMS) analysis of point mutations,” Current Protocols in Human Genetics, vol. 7, no. 1, 1995.
View at: Publisher Site | Google Scholar
M. Pagano and K. Gauvreau, Principles of Biostatistics, Chapman and Hall/CRC, Boca Raton, FL, USA, Second edition, 2018.
View at: Publisher Site
A. Addepalli, S. Kalyani, M. Singh, D. Bandyopadhyay, and K. Naga Mohan, “CalPen (Calculator of Penetrance), a web-based tool to estimate penetrance in complex genetic disorders,” PLoS One, vol. 15, no. 1, p. e0228156, 2020.
View at: Publisher Site | Google Scholar
K. J. Livak and T. D. Schmittgen, “Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method,” Methods, vol. 25, no. 4, pp. 402–408, 2001.
View at: Publisher Site | Google Scholar
D. R. Cox, Principles of Statistical Inference, Cambridge University Press, Cambridge, UK, 2006.
View at: Publisher Site
M. Tsang, J. Gantchev, F. M. Ghazawi, and I. V. Litvinov, “Protocol for adhesion and immunostaining of lymphocytes and other non-adherent cells in culture,” Biotechniques, vol. 63, no. 5, pp. 230–233, 2017.
View at: Publisher Site | Google Scholar
M.-Y. Sun, X.-X. Yang, W.-W. Xu, G.-Y. Yao, H.-Z. Pan, and M. Li, “Association of DNMT1 and DNMT3B polymorphisms with breast cancer risk in Han Chinese women from South China,” Genetics and Molecular Research, vol. 11, no. 4, pp. 4330–4341, 2012.
View at: Publisher Site | Google Scholar
A. Mostowska, S. Sajdak, P. Pawlik, M. Lianeri, and P. P. Jagodzinski, “DNMT1, DNMT3A and DNMT3B gene variants in relation to ovarian cancer risk in the Polish population,” Molecular Biology Reports, vol. 40, no. 8, pp. 4893–4899, 2013.
View at: Publisher Site | Google Scholar
Z. Jia, X. Wu, D. Cao et al., “Polymorphisms of the DNA methyltransferase 1 gene predict survival of gastric cancer patients receiving tumorectomy,” Disease Markers, vol. 2016, Article ID 8578064, 9 pages, 2016.
View at: Publisher Site | Google Scholar
H. Li, J.-w. Liu, L.-p. Sun, and Y. Yuan, “A meta-analysis of the association between DNMT1 polymorphisms and cancer risk,” BioMed Research International, vol. 2017, Article ID 3971259, 10 pages, 2017.
View at: Publisher Site | Google Scholar
E. Ding, J. Liu, H. Guo et al., “DNMT1 and DNMT3A haplotypes associated with noise-induced hearing loss in Chinese workers,” Scientific Reports, vol. 8, no. 1, pp. 1–10, 2018.
View at: Publisher Site | Google Scholar
H.-L. Chen, Z.-M. Li, J.-F. Liu et al., “Polymorphism of the DNA methyltransferase 1 gene is associated with the susceptibility to essential hypertension in male,” Clinical and Experimental Hypertension, vol. 40, no. 7, pp. 695–701, 2018.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2021 Sonal Saxena et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

1148

Downloads

1101

Citations