Gene Polymorphisms and Susceptibility to Functional Dyspepsia: A Systematic Review and Meta-Analysis

Functional dyspepsia (FD) is a common chronic gastrointestinal disorder with a complex, undefined mechanism. Clustering of patients with FD in families highlights the role of genetic factors in the pathogenesis of FD. We performed a systematic review and meta-analysis to clarify the associations between specific gene polymorphisms and FD susceptibility. PubMed, EMBASE, the Cochrane Library, and HuGE database were searched. An additive model was adopted to determine whether previous studied genes are associated with FD susceptibility. Carriers of minor allele in GNB3 825C>T (OR = 1.15, 95% CI 0.99-1.34, P = 0.07), SCL6A4 5HTTLPR (OR = 0.92, 95% CI 0.75-1.12, P = 0.40), and CCK-1R 779T>C (OR = 0.86, 95% CI 0.72-1.03, P = 0.09) genes failed to demonstrate susceptibility to FD. In a subgroup analysis, only minor allele (T) in GNB3 825C>T was associated with an increased susceptibility to the epigastric pain syndrome subtype (OR = 1.34, 95% CI 1.10-1.63, P = 0.003). Our meta-analysis based on available studies using an additive model failed to show that GNB3, SCL6A4, and CCK-1R polymorphisms are associated with FD susceptibility.

Emerging studies demonstrate that susceptibility to FD is influenced by hereditary factors. Clustering of patients with FD in families highlights the role of genetic factors in the pathogenesis of FD [7,8]. Furthermore, the presence of family history of abdominal pain or family history of indigestion increases the likelihood of developing FD [9]. Gene association studies to evaluate gene polymorphisms encoding neuromodulatory and immunomodulatory proteins related to gastrointestinal motility and visceral hypersensitivity, important in the pathogenesis of FD, have been performed [10]. Although previous studies identified several genes associated with FD susceptibility, the results are inconsistent. Therefore, we conducted a systematic review and meta-analysis to critically evaluate existing literature to determine whether specific genetic polymorphisms are associated with FD susceptibility and also stratified by EPS and PDS subtypes.
information of search strategy can be found in the supplementary file (available here).

Eligibility Criteria.
The inclusion criteria for studies were as follows: (1) case-control studies or cohort studies assessing the association between any gene polymorphism and FD, (2) sufficient data available to obtain genotypic frequencies to calculate odds ratio (OR) and 95% confidence interval (CI), and (3) studies in adult population. Non-English manuscripts, review articles, conference abstracts, or studies with insufficient demographic data were excluded.

Data Extraction.
Two investigators (L.D., H.R.) independently extracted data. Data on the author, publication year, demographic characteristics, FD diagnostic criteria, genotyping method, and distribution of genotypes were collected. The quality of the studies was assessed by using the Newcastle-Ottawa Scale (NOS) based on three components: selection, comparability, and ascertainment of outcome [11]. From a range of 1 to 9 stars, studies with higher stars were considered to be higher quality. The values of the gene polymorphisms for those with at least three or more available studies were pooled to perform a meta-analysis designed a priori. An additive model that assumes the contribution of each allele to the relative risk was used to prevent multiple testing of differences between each pair of genotypes [12].

Literature Search and Study
Meta-analysis was performed for the following three genes (GNB3 825C>T, SCL6A4 5HTTLPR, and CCK-1R 779T>C) with three or more studies available meeting the study criteria for meta-analysis. Quality scores of the selected studies ranged from seven to nine indicating moderate to high quality. The study selection process is summarized in Figure 1. Among the eligible studies, all were casecontrol studies. Only two studies demonstrated selection bias in the control group when calculating HWE [32,33]. However, the association was not significantly changed when the two studies were excluded from the metaanalysis. Detailed characteristics of the studies included in the meta-analyses are shown (Table 1).
In the sensitivity analysis removing one study at a time, no single study substantially influenced the pooled ORs. Similarly, there was little change in the estimated pooled ORs after excluding studies by Holtmann et al. [32] and Chung et al. [33], whose genotype distribution of the control group deviated from HWE. Finally, no evidence of publication bias was observed according to the Begg's test (Table 3).
In terms of individual studies, nNOS, CD14, MIF, and TRPV1 gene polymorphisms demonstrated increased susceptibility to FD [22,42]. Ghrelin was associated with feeling of hunger in FD patients in one study [27]. Furthermore, a single study showed that p22PHOX, IL-1b-31CC genotype, and SCN10A were associated with decreased susceptibility to FD [16,24,28]. IL-10, IL-17, TNF-α, and 5HT3A were not associated with FD in a single study, respectively [16,25,36,40]. Two studies exploring 5HT2A did not show any association [34,40]. For studies evaluating FD subtypes, one study suggested that RANTES promoter -28G carrier was associated with a reduced susceptibility to PDS especially in patients with H. pylori infection [15]. COX-1 was associated with susceptibility to EPS in one study [31].

Discussion
In the present meta-analysis, we used additive genetic model to assess and measure the associations of the most extensively studied gene polymorphisms and susceptibility to FD. Carriers of the minor allele in genes GNB3 825C>T, SCL6A4 5HTTLPR, and CCK-1R 779T>C in   FD failed to demonstrate susceptibility to FD. In the subgroup analysis, only minor allele (T) of GNB3 825C>T was associated with increased susceptibility to the EPS subtype. Guanine nucleotide-binding proteins (G-proteins) play an integral role in the function of stimulus-response coupling of membrane receptors that are linked to intracellular effector system [47]. Hormones, neurotransmitters, and inflammatory stimuli involved in the pathophysiology of FD exert effect on cells probably by binding to G-protein-coupled receptors (GPCRs) [43]. GNB3 is the most widely studied G-protein in various disease processes including depression, cardiovascular disease, obesity, and irritable bowel syndrome [48,49]. In FD, 825C>T variation-induced signal transduction contributes to the abnormalities in gastroduodenal sensory and motor functions in the setting of immune activation [32,50]. Although the association of 825C>T and FD was initially reported in the Caucasian population [32], the results were not replicated in others. Furthermore, two meta-analyses were conducted to investigate the association between GNB3 polymorphism and susceptibility to FD but provided inconsistent results [51,52]. Therefore, we performed an updated meta-analysis with additional studies and found that the carrier of the minor allele (T) in gene GNB3 825C>T was not associated with FD susceptibility. However, our results suggested GNB3 variation is associated with susceptibility to FD in patients with EPS (OR = 1 34, 95% CI 1.10-1.63, P = 0 003) and a trend towards susceptibility in PDS (OR = 1 19, 95% CI 0.99-1.43, P = 0 07) subtypes. Therefore, our result suggests that GNB3-mediated signal transduction is more closely linked to pain sensory rather than motility abnormalities. However, the observation of an increased susceptibility to FD in studies with sample size >200 but not in studies with sample size <200 suggested that heterogeneity present smaller studies may have impacted the effect estimate. Furthermore, the effect size of minor allele (T) on FD susceptibility in EPS subgroup was modest and should be interpreted with caution. 5-HT is the primary neurotransmitter involved in the regulation of psychological processes [53]. 5-HT plays a key role in the pathogenesis of both mood disorders and functional gastrointestinal disorders including FD [54,55]. Psychiatric comorbidities including anxiety disorder and depression are more common in patients with FD compared to the control population [56]. Furthermore, a large body of neurobiological research have demonstrated that psychological factors impact gut physiology such as heightened pain sensitivity to gastric distention in patients with FD [57]. Clinical studies have also demonstrated that 5-HT3 receptor antagonism leads to relief of dyspeptic and anxiety symptoms [58,59]. Given that 5-HT transporter (SERT) is the principle regulator of 5-HT levels by facilitating the reuptake, SERT gene-linked polymorphic region (5HTTLPR) may be involved in the pathogenesis of FD [60,61]. More specifically, the short (S) allele has been associated with lower transcription and impaired reuptake of 5-HT compared to the long (L) allele [62]. However both overall and subgroup analyses in our study did not show that the carrier of S allele is associated with susceptibility to FD. However,  Impaired gastric emptying is one of the major pathophysiologic mechanisms in FD. CCK-1, secreted by the neuroendocrine cells of the duodenal mucosa, has a physiologic function of delaying gastric emptying and inducing satiety. Furthermore, CCK-1 receptor belongs to the family of  GPCRs and has been shown to be associated with symptoms of dyspepsia [63], and hyperresponsiveness to CCK has been demonstrated in patients with FD [64]. 779T carrier of CCK-1 is a predictor of PDS in Japanese male patients [46]. However, our meta-analysis failed to show an association between CCK-1 gene and susceptibility to FD.
Several other candidate genes have been explored for FD susceptibility. Genes that play important roles in the regulation of enteric primary afferents and brain-gut interaction (5-HT receptor genes, TRPV1, COMT, and SCN10), induction of inflammatory response (CD14, MIF, TNF-α, IL-17, IL-10, IL-1b, and RANTES), and mediation of gastric accommodation or relaxation (NOS) were hypothesized to be associated to FD susceptibility [24,30,[65][66][67][68][69][70]. However, a meaningful meta-analysis was not able to be performed given the sparse number of studies. Furthermore, H. pylori is a potential confounder of FD. Previous studies have demonstrated that homozygous GNB3 825C>T may be associated with dyspeptic symptom among H. pylori-negative subjects [41]. However, the lack of data on H. pylori status in majority of the studies precluded a subgroup analysis to examine the association of genetic polymorphisms on FD susceptibility independent of H. pylori infection.
The present meta-analysis has limitations. First, the language of the manuscripts was restricted to English which may have excluded eligible studies in other languages. Furthermore, variable definition of dyspepsia, controls, and methodologies for measuring SNPs among studies led to study heterogeneity that may have affected the validity of the meta-analysis. In addition, the lack of data on longterm follow-up and treatment response precluded the assessment of the impact of gene polymorphisms on clinical application in FD. Finally, the lack of data on H. pylori status did not allow evaluation of genetic polymorphisms on FD susceptibility independent of H. pylori infection.
In conclusion, carriers of the minor allele in genes GNB3 825C>T, SCL6A4 5HTTLPR, and CCK-1R 779T>C were not associated with susceptibility to FD. In a subgroup analysis, only minor allele (T) of GNB3 825C>T was associated with an increased susceptibility to EPS subtype. The potential role of utilizing gene polymorphisms to decide diagnostic strategy and therapeutic interventions in FD is appealing, but robust evidence is lacking. Additional studies with larger sample size and detailed characterization of the patients are needed to clarify the role of genetic polymorphisms in FD.

Conflicts of Interest
No competing interests to declare.