In the MR analysis encompassing up to 20,043 patients with gastric diseases and 2,983,613 controls from the UKBB, the FinnGen consortium and MRC-IEU, we comprehensively explored the causal associations between gastric diseases and serum trace minerals for the first time. Our findings revealed distinct effects of genetically predicted serum calcium on gastritis and gastric polyps. Genetically predicted serum magnesium and iron exhibited contrasting associations with the risk of gastric cancer. Similarly, hazardous or protective effect was indicated of serum copper or selenium on gastric ulcers, and serum selenium demonstrated a significant protective influence against gastric diseases overall.
While many epidemiological and experimental studies had suggested elevated serum calcium might exacerbate gastric diseases, such as gastritis, gastric ulcer and gastric cancer [5–8, 45–48], contrasting findings still emerged [11, 12, 22, 24]. These discrepancies could be attributed to variable research ethnicities, geographic regions, and research durations. In addition, the presence of various biases, including potential confounders, might contribute to the observed discrepancies. As such, we conducted this MR analysis, which revealed serum calcium was associated with an increased risk of chronic gastritis. Intriguingly, a protective effect of serum calcium against gastric polyp was discovered for the first time. In our MR study, glucokinase regulator gene (GCKR) polymorphism rs780094 was suggested to significantly influenced the relationship between serum calcium and risk of gastric diseases, which might be partly responsible for discrepancies between previous studies. On the basis of the experimental studies, the potential mechanisms underpinning an association between serum calcium and chronic gastritis might be increased levels of calcium in the blood, contributing to an increase in the acid-peptic factor, leading to the development of hypermotor dyskinesia in the stomach and disruption of regional microcirculation and repair processes, aggravation of chronic gastritis [8]. In addition, Masaki Iimuro et al. reported intake of calcium compounds might exacerbate gastritis through Helicobacter pylori (Hp)-infection [5]. The mechanism involved might be related to the inhibitory effect of calcium on the extraction of urease from Hp [49–51]. On the other hand, while no prior study had separately examined the causal association between serum calcium and gastric polyps, positive data from gastrointestinal cancer prevention trials using polyp as a biomarker supported the protective role of serum calcium [52].
Our study suggested that serum magnesium exerted a hazardous effect against gastric cancer, while serum iron was inversely associated with gastric cancer. Previous studies had reported elevated rates of gastric cancer among groups with high serum magnesium. A study by N Sakamoto et al. examining the concentration of magnesium in drinking water across 98 towns and cities in Japan, indicated magnesium to be correlated significantly to the gastric cancer [11]. Through determining blood magnesium concentrations in patients with gastric cancer, similarly, T Tazawa et al. showed high magnesium levels were found in serum and whole blood of patients in the early stages of gastric cancer [12]. Our MR study suggested that Mucin 1 (MUC1) rs4072037 polymorphism might be a significant factor in increasing the risk of magnesium for gastric cancer. Expressing near the surface of glandular lumens in epithelial tissues of the gastrointestinal tract, abnormal expression of MUC1 was related to the metastasis and proliferation of tumor cells [53–55]. Several studies had reported MUC1 rs4072037 polymorphism correlated with susceptibility to gastric cancer [56–58]. Moreover, transient receptor potential melastatin subtype 6 (TRPM6) rs11144134 polymorphism had also been mentioned to be related to the susceptibility of gastrointestinal cancer [59]. Furthermore, previous research indicated that high concentrations of magnesium might induce conformational changes from B-DNA to Z-DNA, contributing to gene dysregulation, which was associated with carcinogenesis [60, 61].
Epidemiological studies investigating the causal relationship between iron and gastric cancer risk showed inconsistency. Several previous studies had demonstrated a negative correlation between dietary iron intake and gastric cancer incidence [62–65]. Recently, a pooled analysis of case-control studies also revealed an inverse relationship between serum iron and risk of gastric cancer [23]. Conversely, a systematic review and a European prospective investigation both showed a positive association between higher iron intake and gastric cancer risk [13, 14]. These disparities might stem from differences in research populations, geographic regions, and inevitable biases inherent in observational studies. Our study provided evidence supporting an inverse causal role for serum iron in gastric diseases. The MR result suggested homeostatic iron regulator (HFE) gene polymorphism rs1088562 might significantly influenced the causal relationship between serum iron and gastric cancer, potentially explaining the discrepancies observed in previous studies [66]. In addition, transferrin receptor 2 (TfR2) gene polymorphism rs7385804 and transmembrane serine protease 6 (TMPRSS6) gene polymorphism rs855791 might also influence the causal association between serum iron and gastric diseases. Furthermore, potential mechanisms underlying this causal association might be that iron deficiency could lead to DNA damage and increased levels of oxidative stress, which increased cancer risk [67, 68].
The association between serum copper and an increased incidence of gastric ulcers had been well-documented and acknowledged. A randomized controlled trial by M Chen et al. revealed a significant elevation in serum copper content in experimental gastric ulcer model rats compared to the control group. The MR result further indicated selenium-binding protein 1 (SELENBP1) polymorphism (rs2769264) and small-membrane molecule 1 (SMIM1) polymorphism (rs1175550) might serve as important factors contributing to the heightened risk of gastric ulcers induced by high serum copper levels. A study by Jin Zhang et al. observed abundant expression of SELENBP1 in most precursor lesions of gastric cancer, particularly in gastric ulcer [69]. Additionally, the causal relationship between serum copper and the heightened risk of gastric ulcers might be attributed to oxidative stress and reduced activities of antioxidant enzymes resulting from elevated serum copper levels [70].
As a trace element primarily acquired through dietary sources, selenium played a crucial role in various biological functions within our organism, in particular via the antioxidant properties of selenoproteins. Consistent with previous studies [27–29], our study also suggested serum selenium was significantly and inversely associated with gastric disorder, gastric ulcer and chronic gastritis. A controlled trial by Amr M Abbas et al. demonstrated the protective effect of selenium against gastric ulcers [71]. In addition, it had been suggested that selenium might play a pivotal role in the prevention of cancer [72–75]. Charalabopoulos et al. concluded that decreasing levels of serum selenium might be involved in the development and progression of gastric cancer [27]. The significant preventive functions of serum selenium were supposed to stem from the protection of membrane lipids and macromolecules from oxidative damage by counteracting reactive oxygen species and activating antioxidant proteins [71]. Additionally, selenium regulated the DNA damage, G1-phase of the cell cycle, and controls B-cell function and cell mediated immunity to be effective in cancer prevention [76,77].
Previous studies on the causal relationship between serum phosphorus, zinc and gastric diseases have not reached a consensus. Chen et al. reported zinc deficiency promoted gastrointestinal cancer [78], however, Zhang et al. did not find this association [79]. By examining 150 patients with peptic ulcer recurrence, L A Fomina et al. reported recurrence of peptic ulcer was accompanied by decreasing phosphorus in blood [9]. In contrast, a retrospective study involving 75 dogs reported a significant increase in serum phosphorus was noted in dogs with gastric diseases [80]. Consistent with many studies [79, 81, 82], our results suggested no significant causal association between serum phosphorus, zinc and gastric diseases. These inconsistencies might be attributed to various confounders and gene polymorphism. The precise role of serum zinc and phosphorus in the initiation or inhibition of gastric diseases remained unclear and more GWAS and experimental studies were warranted to investigate.
Our study presented several notable advantages. Firstly, we identified a protective effect of serum calcium, iron, and selenium against gastric diseases, suggesting that supplementation of these trace minerals could potentially be a clinical option for high-risk populations. However, it was essential to note that further studies are necessary to validate this conclusion. Secondly, our study utilized publicly available GWAS data to establish causal inferences, leveraging the statistical robustness afforded by the large sample size. Additionally, the MR design employed in our study was inherently less prone to bias compared to classical observational studies, enhancing the credibility of our results.
This study had several limitations. Firstly, it was worth noting that our results were based primarily on European groups, which might limit their applicability to other ethnic cohorts. Secondly, our strict criteria for IVs resulted in a restricted number of variables for certain serum trace minerals, for instance zinc. Thirdly, our study lacked the ability to perform subtype analysis of certain gastric diseases, such as gastric cancer. Conducted using aggregated GWAS data, the MR analysis could not analyze at individual-level. Lastly, it was significant to recognize that our study represented a foundational theoretical investigation. Further validation through cohort studies and animal experiments was warranted to verify the robustness of our results before they could be translated into clinical practice.