We established mouse models of H. pylori infection under different dietary patterns to investigate the association between H. pylori infection and NAFLD. In the CD groups, TG content in liver tissue, serum insulin, and HOMA-IR scores demonstrated that H. pylori infection could cause hepatic TG deposition and insulin resistance in mice. However, serum biochemical parameters, liver enzymes, and liver pathology indicated that H. pylori infection does not significantly affect mice's physiological metabolism. We suspected the possible reasons may be: (1) H. pylori infection duration is not long enough, the systemic chronic inflammatory response is not apparent, and (2) H. pylori infection alone is insufficient to produce significant changes in liver pathology. Analysis of the liver transcriptome sequencing results showed that H. pylori infection induced 767 DEGs in mouse liver tissues, of which 371 genes were up-regulated, and 396 genes were down-regulated. Enrichment analysis showed that some DEGs were significantly involved in the "fatty acid metabolism" and "non-alcoholic fatty liver disease" pathway. These transcriptome results illustrate the link between H. pylori infection and NAFLD.
Based on HFD feeding, H. pylori infection had a more evident effect on the physiological metabolism of mice. Although there was no statistical difference in body weight and liver weight between the two groups, liver TG content, serum TC, LDL-C, and serum inflammatory cytokines IL-6 and TNF-α were significantly higher in the H. pylori-infected group than in the non-infected group. H. pylori infection combined with HFD feeding resulted in more significant hepatic lipid deposition and hepatocyte macrovesicular steatosis in mice, and there were significant differences in NAFLD scores in mice, which coincided with the findings of He et al. [22]. At the same time, mice infected with H. pylori showed decreased sensitivity to glucose and insulin. Analysis of liver transcriptome sequencing results showed that H. pylori infection under HFD feeding conditions induced differential expression of 578 genes in mouse liver tissues, of which 245 genes were up-regulated and 333 genes were down-regulated. Enrichment analysis found that some DEGs were significantly involved in the "long-chain fatty acid metabolic process " and "regulation of lipid metabolic process." Meanwhile, "Retinol metabolism" and "PPAR signaling pathway" were significantly enriched in KEGG analysis. The results of these analyses illustrate that HFD-based H. pylori infection impacts hepatic lipid metabolism.
By comparing the effect of H. pylori strain infection with different Cag A status on liver transcriptomics under the uniform dietary pattern, we explored the possible role of the virulence factor Cag A in the relationship between H. pylori infection and NAFLD. The comparison revealed that the "PPAR signaling pathway" and "Fatty acid degradation" pathways were significantly enriched in DEGs from livers of Cag A- H. pylori-infected mice regardless of dietary pattern. In addition, Fabp5 was upregulated in the transcriptome DEGs, a critical regulator of lipid metabolism.
Several experimental studies have explored the relationship between H. pylori and NAFLD directly. He et al. reported that H. pylori infection combined with 12 weeks of HFD feeding promoted central obesity and IR in mice to a comparable extent as HFD feeding alone for 24 weeks, and dynamic changes in the gut microbiota may cause these effects [22]. Subsequently, the authors measured hepatic lipid deposition in the liver, and NAFLD scores revealed that H. pylori infection significantly aggravated HFD-induced NAFLD and different H. pylori strains, most notably SS1, had different exacerbating effects on NAFLD [23]. Notably, the H. pylori strains used in the above studies (SS1 and NCTC 11637) did not include the Cag A- strain, and we established a Cag A- strain control in combination with clinical epidemiological studies and H. pylori virulence factor studies to explore the effects of different H. pylori strains further. In addition, H. pylori infection has been demonstrated to promote CCl4-induced liver fibrosis in animal models [24]. In this study, it was possible that HFD plus H. pylori infection only intervened for 16 weeks, and no significant hepatic fibrosis was observed via Masson staining of liver sections. Combined with the reported literature [25], we estimated that HFD feeding alone requires at least 24 weeks to visualize significant fibrosis in the livers of mice.
Previous studies have confirmed that Cag A is closely related to the occurrence of gastric cancer. Reports on Cag A combined with extragastric diseases are common in patients with atherosclerosis [26, 27], and only two studies have reported the association between Cag A and NAFLD. Kang et al. suggested that the H. pylori Cag A- strain may be associated with NAFLD [20]. In contrast, Barreyro et al. reported no significant association between H. pylori infection, Cag A status, and ultrasonographically diagnosed NAFLD in NAFLD patients with dyspeptic symptoms [28]. Moreover, the results suggested that Cag A + but not Cag A- was associated with higher AST and fibrosis 4 scores in patients. Our study is the first transcriptomical research to mechanistically explore the relationship between Cag A, H. pylori, and NAFLD. Sequencing analysis of liver transcriptomes infected with different H. pylori strains revealed that “Nonalcoholic fatty liver disease” and “PPAR signaling pathway” were enriched according to KEGG enrichment analysis, and Fabp5 expression was significantly different in the Cag A- groups.
Fabp5 is a member of the fatty acid binding protein family, which is mainly involved in the uptake, transport, and metabolism of fatty acids and related metabolites in the cytoplasm and regulating lipid metabolism and cell growth [29]. Fabp5 is essential for the pathogenesis of IR associated with obesity and lipid metabolism [30, 31]. Loss of Fabp5 gene expression leads to increased systemic insulin sensitivity in animal models of obesity and IR, and adipocytes isolated from Fabp5 -/- mice also exhibit increased insulin-stimulated glucose transport capacity [30]. In contrast, mice with high Fabp5 expression in adipose tissue exhibited significantly decreased systemic insulin sensitivity, and Fabp5 may regulate blood glucose and blood lipid metabolism by affecting leptin expression.
In this study, we detected the differential expression of Fabp5 in each group by qRT-PCR. Interestingly, Fabp5 was highly expressed in both the Cag A- groups but not in the Cag A + groups, regardless of the dietary ingredient. In addition, two bioinformatics studies predicted the crucial role of Fabp5 in NAFLD. High Fabp5 expression was significantly associated with poor prognosis in NAFLD-related HCC patients [32, 33]. These results fit our results to some extent. Therefore, we speculated that other virulence factors of H. pylori, such as vacuolating toxin A, neutrophils activating protein, upregulated Fabp5 expression through some mechanisms, while the presence of Cag A, the most potent virulence factor, masked the mechanisms. However, additional experimental studies are needed to explore the underlying mechanisms of H. pylori virulence factors and extragastric diseases such as NAFLD.
The enrichment analysis results suggested that the retinol metabolic and PPAR signaling pathways were significantly enriched in the HFD groups. Retinol and its primary metabolites, retinal and all-trans retinoic acid (atRA), are collectively referred to as naturally occurring retinoids, which control energy balance, obesity, and inflammatory processes. Total cellular reflectance retinoic acid binding protein (CRABP) is the primary receptor for intracellular retinoid transport, and Fabp5 also has a high affinity for atRA and long-chain fatty acids. Fabp5 competitively binds atRA with CRABP2, and when the Fabp5/CRABP2 ratio is high, atRA binds Fabp5 and activates the downstream PPAR pathway, a crucial pathway regulating glucose and lipid metabolism [34, 35]. We speculate that the overexpression of Fabp5 in mouse hepatocytes caused by H. pylori infection inhibits CRABP2, binds to atRA for transport, activates the downstream PPAR pathway, and, in turn, regulates fatty acid degradation pathways. Our experimental results concatenate H. pylori infection exacerbating NAFLD into a complete clue.