All published articles of this journal are available on ScienceDirect.
PRKCA and TCF4 Genetic Variants in Anxiety Symptoms and Generalized Anxiety Disorder in a Sample of Colombian Subjects Selected on the Basis of High Anxiety Scores
Abstract
Objective:
There are several anxiety disorders leading to a high burden of disease around the world, including Generalized Anxiety Disorder (GAD). The heritability of GAD suggests that genetic factors play an important role in its development; however, further research in this area is needed in Latin America. This study aimed to analyze the possible association between two single nucleotide polymorphisms (SNPs), rs2244497 and rs1452789, located in the PRKCA and TCF4 genes with anxiety symptoms and GAD based on high anxiety scores in a sample of selected Colombian subjects.
Methods:
We evaluated 303 participants using the Hospital Anxiety and Depression Scale (HADS) and Zung’s Self-Rating Anxiety Scale (ZSAS). Subjects with high scores in both scales (according to established cut-off points) participated in a psychiatric evaluation for the diagnosis of GAD. TaqMan assays were employed to genotype the SNPs, and statistical analyses were performed using logistic and linear regression.
Results:
In a sample of Colombian subjects selected on the basis of high anxiety scores, we found a significant association between the rs2244497 SNP in the PRKCA gene and higher scores in anxiety symptoms, where people carrying the T/T genotype had the highest scores for HADS scale. However, we did not observe this association in people diagnosed with GAD. In addition, the SNP in TCF4 (rs1452789) did not have an association with anxiety symptoms or GAD diagnosis.
Conclusion:
This study contributes to the analysis of the molecular basis of anxiety disorders in selected Latin American samples. However, further studies are necessary to understand the role of rs2244497 SNP in the PRKCA gene and the risk for higher scores in anxiety symptoms.
1. INTRODUCTION
Constant worrying and physical symptoms of tension are some of the manifestations of anxiety disorders [1]. Feelings of anxiety are usually future-oriented, such that perceived future threats activate physiological responses [2]. According to the National Institute of Mental Health (NIMH), anxiety is a normal part of life; many people worry about different aspects of their future, such as money, health, and family, among others [3]. However, in some cases, it is hard for people to regulate their anxiety, such that it worsens over time and, under certain conditions, leads to specific anxiety disorders [4-7], which are among the most common psychiatric disorders [8, 9]. Before the COVID-19 pandemic, the global prevalence of anxiety disorders was approximately 4.05% [10]. According to a recent study, anxiety disorders increased by 25.6% due to COVID-19 [11]. In 2015, for individuals between 18 and 44 years old, the prevalence of one or two anxiety symptoms in Colombia was 31.9%, while the prevalence of 5 or more anxiety symptoms was 6.7%; this represents a considerable burden for the health system [12].
Anxiety disorders include generalized anxiety disorder (GAD), panic disorder (PD), social anxiety disorder (SAD), and specific phobias [3]. GAD is one of the most chronic and prevalent of these disorders, with an early age of onset and symptoms that can extend throughout a person's life span, being characterized by uncontrollable and excessive worry present in different areas of life, restlessness, fatigue, irritability, and inability to concentrate or sleep, among others [13]. GAD has a heritability of around 35%, according to the meta-analysis of twin studies carried out in 2021 [14], which suggests that the genetic components play an important role in the development of this disorder [15]. It should be noted that further studies are necessary to understand the etiology of this disorder since GAD leads to a high burden of disease and has the potential to complicate the presentation of other psychiatric disorders [16].
Different studies have analyzed genetic factors related to GAD through two approaches, the genome-wide association studies (GWAS) and the candidate gene studies, finding several polymorphisms associated with this disorder in several populations [17]. In 2017, a GWAS found that several SNPs were associated with GAD symptoms in Latin American subjects in the United States; nevertheless, none of these polymorphisms reached the necessary significance threshold for a genome-wide study (p < 5 x 10-8) [18]. In contrast, a candidate gene study carried out in Mexican individuals found that BDNF-rs6265 confers an increased risk for developing GAD in this population [19]. Furthermore, other SNPs in different genes have been analyzed in relation to GAD, for instance, in the serotonin transporter (SLC6A4) [20], CD300 molecule like family member f (CD300LF) [21], and norepinephrine transporter (SLC6A2) [22]. It is important to highlight that there is an urgent need to study the genetic aspects of anxiety disorders in Latin American countries, including the Colombian population, due to the existence of risk factors specific to certain ancestries and taking into account the severe underrepresentation of Latin American countries in genetic studies of anxiety disorders. As an example of this, Fatumo et al. [23] recently found that the number of samples from Latin America in the GWAS catalog is very low (less than 2 percent), and van der Walt et al. [24] also identified, in a systematic review of GWAS for anxiety disorders, very low participation of individuals from Latin America.
The protein kinase C alpha (PRKCA) gene encodes a protein that is part of a family of serine- and threonine-specific protein kinases that influence many cellular processes, such as cellular development, proliferation, and differentiation, among others [25]. Variants in this gene have been previously associated with the risk for psychiatric disorders and neurological diseases, such as bipolar disorder, Alzheimer's disease, and post-traumatic stress disorder [26-29]. Another gene associated with neuropsychiatric disorders is transcription factor 4 (TCF4), which encodes a protein that is a member of the basic helix-loop-helix family of transcription factors, which has an important role in different processes in nervous system development [30]. Genetic variants located in this gene have been associated with schizophrenia, bipolar disorder, major depression, and post-traumatic stress disorder [30-34].
The possible influence of the PRKCA and TCF4 on anxiety disorders, particularly on GAD, has not been explored previously. Notwithstanding, there are two polymorphisms located in both genes, the rs2244497 C/T and rs1452789 T/A, respectively, which have been associated with neuroticism, a personality trait related to anxiety disorders [35-36]. Our hypothesis was that these novel SNPs in two candidate genes might be associated with levels of anxiety symptoms. Therefore, we tested this hypothesis in order to analyze the possible association between these variants in PRKCA and TCF4 genes and anxiety symptoms and GAD in a Colombian sample.
2. MATERIALS AND METHODS
2.1. Participants
About 303 Colombian individuals living in the capital city of Bogota were included in this study. All subjects were over 18 years old, with a mean age of 25.2 ± 8.4 years. About 200 subjects were females (66.01%).
2.2. Phenotypic Characterization and Classification of Individuals
We assessed the levels of anxiety symptoms using the Hospital Anxiety and Depression Scale (HADS), specifically, the subscale HADS-A [37] and Zung’s Self-Rating Anxiety Scale (ZSAS) [38]. We used previously established cut-off points of 8 for the HADS-A and 36 for the ZSAS [39-40]. Scores above these cut-off points suggest clinical levels of anxiety. For this study, the Cronbach’s alpha for the HADS-A was 0.8647, and for ZSAS was 0.9920. According to the results of both scales, individuals were classified into two different groups as follows: participants with high scores in anxiety symptoms (HADS-A ≥ 8, and ZSAS ≥36) and participants with low scores in anxiety symptoms (HADS-A < 8 and ZSAS < 36). Then, individuals who showed scores above cutoff points in both scales were invited to a clinical evaluation by a psychiatrist to confirm a possible GAD diagnosis. The psychiatric evaluation included a structured diagnostic interview using the Spanish version of the Mini-International Neuropsychiatric Interview. These interviews took place via remote settings (due to the pandemic), following MINI guidelines [41].
2.3. SNP Genotyping
In this study, we employed a candidate gene approach. We selected the SNPs in PRKCA (rs2244497) and TCF4 (rs1452789) using data from the GWAS catalog and Genome Aggregation Database (gnomAD); moreover, we considered the following selection criteria: minor allele frequency in samples from the Latin American Population (MAF ≥ 0.1) and previous reports of association with other diseases or disorders [35-36]. In addition, the RegulomeDB (https://regulomedb .org/regulome-search) web server was used for exploring the classification of variants based on their functional location in a regulatory region [42].
Isolation of DNA from peripheral blood samples was performed through the salting-out method [43]. DNA concentration was determined in a Qubit fluorometer (Thermo Fisher Scientific, MA, USA), and the DNA samples were normalized at 10ng/μL. The genotyping was carried out using TaqMan SNP Genotyping Assays (Applied Biosystems, Foster City, CA, USA), and employed the following pre-designed probes: PRKCA (C_370967_10) and TCF4 (C_9617300_10). For the qPCR reaction, we used the following reagents: 2ul (20ng) of genomic DNA, 1X TaqMan® Genotyping Master Mix, 1X TaqMan assay and water in a total volume of 10 μL. The reaction was carried out in a CFX96 Touch Real-Time PCR system (BioRad, Hercules, CA, USA). qPCR protocol included a 10 min denaturation step at 95ºC (1 cycle), 95 °C for 15 s, and 60 °C for 90 s (50 cycles).
2.4. Statistical Analysis
We used the Plink v1.9 and SNPStats programs [43-45]. First, for the 303 participants, we determined the allelic and genotypic frequencies, and then, the Hardy-Weinberg equilibrium (HWE) was assessed, considering the 5% level of significance [46]. Second, we performed a linear regression test to evaluate the possible association between the polymorphisms and the scores in anxiety symptoms. On the other hand, to analyze the association of the SNPs with the psychiatric diagnosis, we compared the cases of GAD (n=50) and controls (participants with scores below the cut-off points, n=101) using a logistic regression test. The regression tests were adjusted by age and sex, and a p-value lower than 0.05 was considered statistically significant. In addition, effect sizes were determined to describe the strength of the association using the Hedges' g measure [48].
3. RESULTS
3.1. Psychological Evaluation
The phenotypic evaluation using HADS-A and ZSAS scores showed that 137 individuals exceeded the cut-off points in both anxiety scales (HADS-A ≥ 8 and ZSAS ≥36), suggesting the presence of clinical levels of anxiety (high scores), and 101 participants presented scores below the cut-off points (low scores). About 65 subjects exceeded the cut-off point for one scale only. The psychiatric evaluation of subjects with high scores confirmed the diagnosis of GAD in 50 participants, while 54 did not have clinical evidence for this disorder. Moreover, for 33 subjects, the psychiatric assessment suggested other diagnoses. Sex and age distribution among the groups are shown in Table 1.
For the analysis of the association between the SNPs and anxiety symptom levels, all participants were included. On the other hand, for the association analysis with GAD, only people with a confirmed diagnosis and subjects with low scores in both scales -controls-, were considered.
3.2. Genetic Characterization
In a sample of Colombian subjects selected on the basis of high anxiety scores for rs2244497 in PRKCA, the allele frequencies were 0.34 for allele C and 0.66 for allele T; the genotypic frequencies were C/C (0.10), C/T (0.48), and T/T (0.42). For TCF4 rs1452789, allelic frequencies were 0.33 for allele T and 0.67 for allele A. The genotypic frequencies were A/A (0.43), A/T (0.49), and T/T (0.09). Both SNPs, rs2244497 and rs1452789, were in Hardy–Weinberg equilibrium, with p values of 0.2 and 0.09, respectively.
3.3. Association of Polymorphisms and Scores from Anxiety Scales
We analyzed the association between polymorphisms and scores on each anxiety scale using different genotypic models. We observed a significant association between the rs2244497 in the PRKCA gene with HADS-A scores under codominant, dominant, and overdominant models (Table 2). These results showed that carriers of the T/T genotype have higher scores than C/T and C/C carriers (p = 0.012). In contrast, for the ZSAS scores, we did not observe any significant differences; the mean score for each genotype was T/T (40.54), C/T (39.11), and C/C (39.21). On the other hand, for the rs1452789 of TCF4, we did not find significant associations between scores of HADS-A and the different genotypes; the scores according to each genotype were A/A (7.98), A/T (8.02), and T/T (7.62). Similarly, significant differences in the ZSAS scores were not identified, where the mean for each genotype was A/A (39.84), A/T (39.99), and T/T (37.65). Finally, the sum of both scales showed only a significant association with rs2244497 in the PRKCA gene (p =0.003) under a dominant model, where carriers of the genotype T/T had higher scores than C/T-C/C, and the mean scores were 49.90 and 46.52, respectively.
- | All Participants | High Scores | Low Scores | GAD | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sex | All | Female | Male | All | Female | Male | All | Female | Male | All * | Female | Male |
N | n=303 | n=200 | n=103 | n=137 | n=105 | n=32 | n=101 | n=52 | n=49 | n=50 | n=42 | n=8 |
% | 100% | 66% | 44% | 45.2% | 76.6% | 23.4% | 33.3% | 51.5% | 48.5% | - | 84% | 16% |
Age means (SD) | 25.19 (8.40) | 25.14 (8.36) | 25.27 (8.51) | 24.58 (7.99) | 24.29 (8.15) | 25.53 (7.50) | 25.72 (8.43) | 26.62 (8.57) | 24.78 (8.27) | 23.62 (7.28) | 23.5 (7.2) | 21.5 (3.1) |
HADS-A mean (SD) | 7.97 (4.33) | 8.59 (4.39) | 6.77 (3.99) | 11.71 (2.88) | 11.86 (2.98) | 11.22 (2.50) | 3.99 (2.05) | 4.04 (2.05) | 3.76 (2.06) | 12.2 (3.13) | 12.07 (3.45) | 12.00 (3.21) |
ZSAS mean (SD) | 39.72 (9.81) | 42.25 (10.07) | 34.82 (7.10) | 47.49 (7.64) | 49.03 (7.53) | 42.44 (5.64) | 29.67 (3.35) | 30.02 (3.34) | 29.31 (3.36) | 49.46 (8.29) | 50.57 (8.01) | 44.13 (7.57) |
Genotypic Model | N | HADS-A Mean (SD) | p-value | Hedges g |
---|---|---|---|---|
Codominant | T/T=128 C/T=146 C/C=29 |
8.75 (4.41) 7.42 (4.10) 7.28 (4.25) |
0.012 | T/T vs C/T= 0.313 C/T vs C/C=0.034 C/C vs T/T=0.335 |
Dominant | T/T=128 C/T-C/C=175 |
8.75 (4.41) 7.39 (4.10) |
0.003 | 0.321 |
Overdominant | T/T-C/T=274 C/C=29 |
8.48 (5.79) 7.28 (4.25) |
0.035 | 0.212 |
Polymorphism | Alleles and Genotypes |
GAD Cases n=101 |
Controls n=50 |
p-value |
---|---|---|---|---|
rs2244497 (PRKCA) | C | 0.37 | 0.27 | 0.055 |
T | 0.63 | 0.73 | ||
C/C | 0.09 | 0.04 | 0.220 | |
C/T | 0.56 | 0.46 | ||
T/T | 0.35 | 0.50 | ||
rs1452789 (TCF4) | A | 0.70 | 0.64 | 0.299 |
T | 0.30 | 0.36 | ||
A/A | 0.49 | 0.38 | 0.440 | |
A/T | 0.43 | 0.52 | ||
T/T | 0.09 | 0.10 |
The second analysis compared 50 cases (GAD-diagnosed individuals) and 101 controls (subjects with scores below the cut-off points). By assessing the allele distribution using a logistic regression adjusted by sex and age, we did not find any significant association with the rs2244497 in the PRKCA gene (p = 0.055) or rs1452789 in the TCF4 gene (p = 0.299). Similar results were detected when we compared cases and controls under different genotypic models (Table 3).
According to the RegulomeDB, the variant rs2244497 in PRKCA has a rank of 3a, which shows evidence for being a transcription binding site, and a probability score of 0.71; this indicates that there is a high probability it will be a regulatory variant [42]. The rs1452789 in TCF4 has a rank of 4 and a probability score of 0.61.
4. DISCUSSION
In this study, we analyzed for the first time the association between two polymorphisms and anxiety symptoms and GAD in a sample of Colombian subjects selected based on high anxiety scores. We observed that carriers of the T/T genotype had higher anxiety levels than other genotypes of rs2244497 in the PRKCA gene; however, the individual analysis of each scale showed that this association is only with HADS-A scores and not with ZSAS. Moreover, when comparisons were done for GAD cases and controls, we did not find any significant results. On the other hand, TCF4-rs1452789 did not show evidence to be involved in the anxiety phenotypes in the individuals of the analyzed population.
Here, 137 (45.21%) individuals showed high levels of anxiety based on the results from the HADS-A and ZSAS analyses; in this group, 76,64% were females (n=105). Previously, it has been reported that anxiety symptoms prevalence in Colombia ranged from 25.8 to 52.9% [12,48-50], and international reports have shown an anxiety prevalence between 14 and 25% [8, 9]. However, a meta-analysis detected a prevalence of 46% in people from different countries during the coronavirus pandemic, which suggests an increase in anxiety prevalence around the world [51]. Otherwise, based on a psychiatric examination, 16.5% of individuals were diagnosed with GAD; for this group, 82% were women. In both cases, female individuals were more affected than males, as indicated in previous reports, in which anxiety disorders appear to be 2 to 3 times more prevalent in women than in men [52].
The association analysis only identified, based on high anxiety scores in a sample of Colombian subjects selected, a significant result with the rs2244497 in the PRKCA gene. It should be noted that this variant has not been well studied; it is located in intron 3, and until now, there is no evidence about its functional role. In our results, carriers of the T/T genotype had high levels of anxiety. Interestingly, the allele T of this variant has been associated with neuroticism [35]. This is an important personality trait associated with the development of common mental disorders, including anxiety disorders [53]. On the other hand, although the variant rs1452789 in TCF4 has also been associated with neuroticism [36], in our study, this was not associated with anxiety symptoms or GAD.
Previously, PRKCA has been implicated in anxiety disorders; for example, a murine model showed that differential expression of the protein kinase C alpha altered neurobiological processes that led to altered anxiety phenotypes [54]. Additionally, there is evidence that this protein has different targets involved in the regulation of emotional behavior, which might result in anxiety-related problems [55]. Another study analyzed the interactome of several anxiety disorders, identifying that PRKCA is a key gene expressed in the striatum, an important region for different disorders, including GAD, panic disorder and specific phobias [56]. Interestingly, in an association study, a polymorphism (rs4790904) in PRKCA was statistically associated with posttraumatic stress disorder in a population of U.S. veterans [57]. This association highlights the potential role of the genetic variants of PRKCA in neuropsychiatric disorders.
It should be noted that in this study, the rs2244497-PRKCA was only associated with HADS-A scores and not with ZSAS. Both scales are used to measure anxiety symptoms, and in our analysis, we evidenced a positive correlation between their scores (Spearman's Rho: 0.784; p= 0.000); this was similar to other studies [58]. However, there might be some differences that could explain the association with only HADS-A. For instance, the type of symptoms assessed by each scale; the HADS-A is focused on evaluating psychological symptoms [39], whereas the ZSAS covers both somatic and psychological symptoms [40]. This difference suggests that rs2244497 might influence the psychological aspects of anxiety but not the somatization. In contrast, in spite that the assessment time and the number of items are differents between both scales, these seem not to affect anxiety detection rates [58, 59].
Although we observed a significant association with levels of anxiety symptoms for GAD, we did not observe the same results. Several reasons can explain these findings: first, the low number of individuals with a GAD diagnosis decreases the statistical power of our study. Second, the etiology of GAD might involve other pathways different from those involved in anxiety symptoms, and finally, we only analyzed the association with two single polymorphisms, and we did not perform any assessment of gene-gene or gene-environment interaction, which might also provide evidence about the etiology of GAD [17].
As we mentioned above, in this study, the main limitation was the sample size, particularly, for the group of people with GAD. In addition, the sample was selected on specific criteria and was not randomly chosen. This work was performed during the COVID-19 pandemic, and this limited our capacity to include more participants. Future studies should include larger samples and other novel SNPs in candidate genes.
CONCLUSION
In summary, in a sample of Colombian subjects selected on the basis of high anxiety scores, our results suggest that the PRKCA gene has an association with levels of anxiety symptoms, but it is not associated with GAD. However, further studies are necessary to elucidate the complete relationship between PRKCA and GAD. Moreover, our work is the first genetic analysis of GAD in the Colombian population and one of the first genetic studies of anxiety symptoms in samples from Latin American populations.
AUTHORS' CONTRIBUTION
DF, MLTG and YGG contributed to the conceptualization, methodology, validation, and review. DALR and LUM contributed to formal analysis and original draft preparation. SB, LM and WV contributed to administering and interpreting the psychological and psychiatric tests of this study. LUM, LM, BJQ, SZ, WV, YGG and MLTG participated in the formal analysis, investigation, and review. MLTG and YGG contributed to editing the work. All authors (DALR, LUM, BJQ, SZ, WV, LM, YGG, SB, DF and MLTG) contributed to critically revising the work, approved the final version of the article to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
LIST OF ABBREVIATIONS
GAD | = Generalized Anxiety Disorder |
SNPs | = Single Nucleotide Polymorphisms |
HADS | = Hospital Anxiety and Depression Scale |
HWE | = Hardy-Weinberg Equilibrium |
ETHICS APPROVAL AND CONSENT TO PARTICIPATE
The Ethics Committee of the Universidad Antonio Nariño approved the study (protocol code 27052019-1).
HUMAN AND ANIMAL RIGHTS
No animals were used that are the basis of this study. All the human procedures were conducted in accordance with the Declaration of Helsinki principles.
CONSENT FOR PUBLICATION
Informed consent was obtained from all participants of this study.
AVAILABILITY OF DATA AND MATERIALS
The data that support the findings of this study are available from the corresponding author, [M.L.T-G] upon reasonable request.
FUNDING
This study was supported by a research grant from Minciencias (Ministry of Science, Technology, and Innovation of Colombia; Grant No. 915-2019). Daniel A Leon R. was supported by a post-doctoral fellowship from Minciencias at Universidad Antonio Nariño (Grant No. 672-2021).
CONFLICT OF INTEREST
The authors declare no conflict of interest, financial or otherwise.
ACKNOWLEDGEMENTS
We would like to thank everyone who has contributed to the data collection, as well as all the study participants. The authors acknowledge the financial support of Minciencias (Ministry of Science, Technology, and Innovation of Colombia; Grant No. 915-2019), Universidad Antonio Nariño and Universidad de Los Andes. Daniel A Leon was supported by a post-doctoral fellowship at Universidad Antonio Nariño (Minciencias; Grant No.672-2021).