Dysmorphic contribution of neurotransmitter and neuroendocrine system polymorphisms to subtherapeutic mood states

Abstract Objective From an evolutionary perspective, emotions emerged as rapid adaptive reactions that increase survival rates. Current psychobiology includes the consideration that genetic changes affecting neuroendocrine and neurotransmission pathways may also be affecting mood states. Following this hypothesis, abnormal levels of any of the aminergic neurotransmitters would be of considerable importance in the development of a pathophysiological state. Materials and Methods A total of 668 students from the School of Medicine of the University of Malaga (Average = 22.41 ± 3; 41% men) provided self‐report measures of mood states using POMS and GHQ‐28 questionnaires and buccal cells for genotyping 19 polymorphisms from 14 selected neurotransmitter pathways genes (HTR1A; HTR2A; HTR2C; HTR3B; TPH1; SLC18A1; SLC18A2; COMT; MAOA; MAOB) and neuroendocrine system (AVPR1B; OPRM1; BDNF; OXTR). Results MAOA rs3788862 genotype correlates with decreasing levels of Tension among females (beta = −0.168, p‐value = 0.003) but it is neutral among males in this subscale. On the contrary, it correlates with lower GHQ‐28 depression scores among males (beta = −0.196, p‐value = 0.008). Equivalently, SLC18A1 and HTR2A variants correlated with anger and vigor scores, only among males. From the neuroendocrine system, OPRM1 rs1799971 correlated increasing levels of female's Anxiety, depression and Social Dysfunction scores. Conclusion Our findings suggest that these polymorphisms contribute to define general population mood levels, although exhibiting a clear sexual dimorphism.


| INTRODUC TI ON
The study of human personality, behavior, and mood has been addressed from multiple disciplines. Understanding the intimate nature of our emotions, let us give a rational voice to the feelings that condition our behavior in society. Emotions have been explained from the evolutionary perspective as rapid adaptive reactions that increase survival rates among vertebrates (Nesse & Ellsworth, 2009). Anxiety and fear, for example, are triggered from the amygdala even before our frontal cortex processes the origin of the warning stimulus. This alarm system is tightly regulated but allows an overreaction. From a purely biological point of view, it is much more economical to be alarmed without reason than not to once a situation deserves it (Marks & Nesse, 1994, Sanjuán and Casés, 2005, Garakani, Mathew, & Charney, 2006. Certain types of depression would emerge as a strategy of energy savings once facing the impossibility of achieving an objective, therefore reducing the risk to new stressors, a situation that would be reversed when these objectives are achieved (Sanjuán and Casés, 2005;Kinney & Tanaka, 2009). In this sense, depressed patients with a poor therapeutic response exhibit a significant improvement upon facing a favorable environmental change. This is also compatible with the hypotheses related to social competition, according to which the levels of serotonin (5-hydroxytyroxine or 5-HT) in the central nervous system are elevated upon the achievement of dominance, which is associated with the decrease in stress levels and mood enhancement (Raleigh et al., 1991;Price, Sloman, Gardner, Gilbert, & y Rohde, 1994). Each element that contributes to the mood state is influenced by a wide spectrum of individual and collective factors; therefore, it might be considered one of the most complex human traits to study. Knowing the biochemical pathways that comprise mood states is especially relevant when associated with the daily clinical practice. This happens whenever a subject reaches a pathological level of the different components of the mood state and exhibit anxiety disorder, bipolar disorder, or major depression.
We must take into account that the different dimensions of mood states are quantitative variables that are differently affecting general population. Currently, we have a diverse scope of technical approaches to study mood states. The Goldberg General Health Questionnaire (GHQ-28) is an instrument originally designed to identify nonpsychotic mental disorders in contexts of general medical practice. It allows to differentiate in a simple way, psychiatric patients from those considered healthy (Goldberg, 1978;Retolaza et al, 2003). The GHQ-28 consists of four subscales: A-scale refers to somatic symptoms, B to anxiety and insomnia, C to social dysfunction, and D to depression. GHQ-28 can be applied to the general population and is suggested for the assessment of mental health.
The Profile of Mood States (POMS) test consists of 65 items rated using a Likert type format, with five alternatives response ranging from 0 to 4 (McNair, Lorr, & Droppleman, 1971). It allows to obtain a general index of the alteration of seven partial measurements: tension, depression, anger, vigor, fatigue, confusion, and friendship. At the beginning, this test was used to evaluate the effects of psychotherapy and medication in external psychiatric patients although it was also tested with a variety of nonpsychiatric samples and has become a very popular instrument (Andrade et al., 2002).
Current psychobiology includes the consideration that genetic changes affecting neurotransmission pathways may also be affecting mood states. Following this hypothesis, abnormal levels of any of the aminergic neurotransmitters, dopamine, norepinephrine, and serotonin, would be of considerable importance in the development of a pathophysiological state (Baldwin & Birtwistle, 2002). Serotonin offers remarkable action on sleep-wake cycle, behavior, cardiac function, endocrine secretions, pain perception, appetite, and sexual activity. Tryptophan is the known precursor of serotonin. Functional mutations affecting the coding region of the tryptophan-hydrolase 2 gene (TPH2) have been found among families with bipolar disorder Grigoroiu-Serbanescu et al., 2008). Other studies have analyzed the role of genes involving the neurotransmitter synthesis, transport, and degradation such as SLC6A3, HTR2A, MAOA, COMT, and SLC6A4 (O'Donovan et al., 2008;Williams et al., 2011). Coding variants within the COMT gene, related to dopamine degradation, have been shown to be associated with bipolar disorder risk (Zhang et al., 2009). A single-nucleotide polymorphism (SNP) in the promoter region of the serotonin receptor gene HTR1A was also significantly associated with bipolar disorder risk (Kishi et al., 2011) as well as different genomic variants of the mono amine oxidase genes (MAOA, MAOB) (Fan et al., 2010). Polymorphisms within SLC6A4 (5-HTTLPR) have been studied among major depressive disorder patients and been included in several meta-analyses that demonstrated a small but significant association to bipolar disorder (Lasky-Su, Faraone, Glatt, & Tsuang, 2005;Cho et al., 2005). Metaanalysis studying the different alleles of the TPH1 gene concluded that it is not associated with major depressive disorder but rather with bipolar disorder (Halmoy et al., 2010). Other genes have been also found to affect different neuropsychiatric disorders such as the brain-derived neurotrophic factor (BDNF gene), which is involved in both the pathogenesis of depression and the mechanism of action of antidepressant treatments (Duman & Monteggia, 2006;Verhagen et al., 2010). However, in spite of the role of aforementioned genes in the development of pathological status, literature is scarce about how the different genetic configurations affect mood states among healthy subjects. In order to evaluate in a quantitative manner the role of these genetic variants over the different dimensions of the mood state within the general population, we initiated a study in which 20 genetic variants affecting different neuroendocrine biochemical pathways were analyzed in a series of volunteers from the University of Malaga who phenotyped using POMS and GHQ-28 questionnaires.

| DNA donors
The study subjects of this research were 668 healthy students of the University of Malaga who voluntarily decided to participate in the project. Inclusion criteria were being adult and fell healthy without apparent psychiatric disease. The following demographic variables were taken: weight, height, age, sex, and whether they were currently taking any drug treatment. DNA was extracted from buccal swap according to standard procedures. This research was carried out with the approval of the Ethics Committee of the University of Malaga and all the students signed an informed consent. This work was carried out in accordance with the principles of the Declaration of Helsinki.

| Single-Nucleotide Polymorphisms
Genotyping was outsourced to Genologica SL. SNP analysis was performed using the TaqMan Open Array Genotyping System from Applied Biosystems. The results obtained were processed using TaqMan
Kolmogorov-Smirnov test was used to determine the normality of the quantitative data series. For bivariate correlations studies, both the Pearson's correlation coefficient and Spearman's Rho were calculated. For models that included both the genetic variants and other covariates, the linear regression models were used. The level of significance was 0.05.

| RE SULTS
The study comprised 668 students from the School of Medicine of the University of Malaga recruited between 2011 and 2015. The age of the study subjects was relatively homogeneous (22.41 ± 3 years) although ranged between 18 and 51 years. The series was composed by 41% men and all from Caucasian origin. All of them were sampled for buccal swap for subsequent determination of genetic polymorphisms. Call ratios had an average of 96%, although they ranged from 98% for SNPs such as rs3813929, rs3027452, or rs2254298, and the minimum of 89% obtained with rs6313. Hardy-Weinberg equilibrium (HWE) was determined for those SNPs mapping autosomal chromosomes and only those with p > 0.05 were used for further analyses (all but rs2254298, rs324981, and rs1800532, Supporting Information Table S2). When volunteers were then invited to fill the POMS and GHQ-28 questionnaires, from the initial 668 students, 601 (90%) completed both tests. Regarding the variables under study, a summary of the mood variables determined using POMS and GHQ-28 is shown in Supporting Information Table S3.
We first determined the correlation between both tests and evaluated the effects attributed to age, sex, or BMI (Supporting Information Table S4). Gender exhibited statistically significant differences in Vigor T-score (lower among females, Spearman's p-value = 0.004) and chronic Anxiety (higher among females, Spearman's p-value = 0.008). Age also correlated with different parameters such as vigor, friendship, and new onset Depression, evidencing the need to use them as covariates to determine the potential role of the genetic variants under analyses. Beyond this, we found a relevant intercorrelation between the different variables within the same questionnaire (GHQ-28 chronic and new onset) as well as a significant correlation between equivalent variables interrogated in POMS and GHQ-28. As an example, we found that the POMS T-score measuring fatigue positively correlated with GHQ-28 chronic anxiety and depression levels (Rho > 0.435, p-value < 0.001) (Supporting Information Table S4). Therefore, both test might be considered to a certain extend an internal replica.
Next, we performed a multiple correlation analysis between the three genotypes for each genetic variant and the POMS Tscores. Results are shown in Table 1. A particular haplotype captured by the two variants within the MAOA gene correlated with a lower degree of Tension. HTR2A rs6313 also correlated with Vigor (Rho = 0.134, p-value = 0.004) suggesting that those subjects harboring the mutant homozygous genotype reported an increased Vigor than those with the reference genotype. We might mention the associations found for SLC18A1 variants and BDNF rs6265; however, the p-values obtained do not support multiple correction and therefore should be treated with caution. Of mention,  when quantified with GHQ-28 while being neutral among females.

TA B L E 1 Association between the genetic variants and the POMS variables measured
Equivalently, SLC18A1 and HTR2A variants correlated with increasing levels of anger and vigor, respectively, but only among males. From the neuroendocrine system-associated genes, we might highlight the association among females between OPRM1 polymorphism and increasing levels of anxiety and somatization, concomitantly with lower Social Dysfunction scores. conditions that did not give rise to a psychopathology. This sensitivity on detection mood variability and genotype among euthymic subjects could be supported by the high homogeneity of the age a sociocultural features of the population studied. Perhaps because of this, some statistical association may not support multiple testing correction.

| D ISCUSS I ON
Similar studies (Takeuchi et al., 2015) relate the polymorphism of DRD2 with the POMS test and find differences between sexes in a similar population. Yarosh, Meda, Wit, Hart, & Pearlson, (2015), performed a multivariate analysis of polymorphisms of a whole genome association study with the POMS test in healthy subjects treated with amphetamine, finding association with SNPs related to genes of the glutamatergic signal pathways, which seem to mediate behavior and in cardiovascular responses to amphetamine.
On the other hand, in the present study a sexual dimorphism is shown both when we correlate general items such as age, sex, and BMI and when we observe them associated with genotypes. Among the general items such as BMI, women have lower BMI than men in our population, which is reversed in the adult population, perhaps due to the low average age of the sample (Wells, 2007). The male sex correlates positively with the vigor. Age correlates positively with BMI (Livshits et al., 2012), and negatively with anger. This points that age correlates with lesser anger, higher vigor, friendship, and somatization. Thus, the irritable attitude decreases with age, as the sensation of activity and energy increases, what taking into account the age range of the population this could be associated with a greater hormonal balance, increases the capacity of relationships that is translated into friendship, and somatic sensations increase maybe due to greater recognition of one's body.
The correlations between the test items show that tension, nervousness, agitation, etc., positively correlate with depression, anger, fatigue, and TMD of POMS questionnaire and anxiety, somatization, depression, and social dystonia in their chronic profile from GHQ-28 and negatively with the vigor and friendship (POMS questionnaire shall be more predisposed to depression and less tension, inversely than men, according to a higher MAO activity found among women (Jansson, 2005 According to the literature, mutant allele is related to increased pain, suggesting a compromised protein function (Slavich, 2014).
In our study, we found an association to increased anxiety and somatization symptoms, while decreases Social Dysfunction, but only among women. This, in terms of quality of life, could be substantiated in a greater sensitivity to pain and less pleasant rewards to intense stimuli. Variant G carriers are adapted to relaxation stimuli or lower endorphinic pleasure than carriers of A. In fact could explain why this polymorphism and in particular the G allele is associated with a greater tendency to addiction and variations in the pharmacological response to it (van den Wildenberg et al., 2007;Anton et al., 2008). We did not find in the literature any association study on emotional response, most of them refer to predisposition to addiction, and some to depression, but indirect processes as a consequence of pain. Our results show an association in the responses to the Goldberg test of the G allele with anxiety and insomnia, somatization and very strongly negatively associated with social dysfunction among women. This result indicates that subjects in whom the degree of neurological reward mediated by the opioid effect is diminished are the most socially adapted or those with less social dysfunction indicate always that not fall in addiction. It could be interpreted that their state of less neurological pleasure leads to a more correct social response, perhaps abounding on the hypothesis that the greater tendency to pleasure is associated with a greater rebellion to social restrictions (Slavich, Tartter, Brennan, & Hammen, 2014). Overall, it can be deduced that genetic variations within the neurotransmitter (HTR2A, SLC18A1, MAOA) and neuroendocrine (BDNF and OPRM1) systems, determined in a euthymic population, are associated with emotional traits quantitatively assays using POMS and GHQ-28 questionnaires, although the gender shall be considered as both determining and excluding criteria in the different associations.

| Study limitations
Whenever a genetic association is reported, the suspect of facing a false positive arises. Given the amount of variables assayed, we might consider that defining an alpha of 0.05 might be too permissive. Here, we analyzed 14 genes and conducted two mood tests, but the different variables under study are not purely independent.
MAOA polymorphisms for instance are in linkage disequilibrium. In Caucasian and that the Spanish population is largely homogeneous (Gayan et al., 2010). For these reasons, an independent replica of the current findings would be required to confirm our findings.

ACK N OWLED G M ENT
We thank all the volunteers for their participation in this study.

CO N FLI C T O F I NTE R E S T S
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. This work was financed using internal sources from the Department of Surgery, Biochemistry and Immunology, Universidad de Málaga.