Helicobacter pylori infection altered gastric microbiota in patients with chronic gastritis

Abstract

Objective: The present study aims to investigate the effect of Helicobacter pylori (Hp) infection on gastric mucosal microbiota in patients with chronic gastritis.

Methods: Here recruited a population of 193 patients with both chronic gastritis and positive rapid urease, including 124 patients with chronic atrophic gastritis (CAG) and 69 patients with chronic non-atrophic gastritis (nCAG). Immunoblotting was used to detect four serum Hp antibodies (UreA, UreB, VacA and CagA) to determine the types of virulent Hp-I and avirulent Hp-II infections. Gastric microbiota was profiled by 16S rRNA gene V3-V4 region, and R software was used to present the relationship between the microbial characteristics and the type of Hp infection.

Results: In the stomach of patients with Hp-positive gastritis, the dominant gastric bacterial genera included Ralstonia (23.94%), Helicobacter (20.28%), Pseudonocardia (9.99%), Mesorhizobium (9.21%), Bradyrhizobium (5.05%), and Labrys (4.75%). The proportion of Hp-I infection was significantly higher in CAG patients (91.1%) than in nCAG patients (71.0%) (P < 0.001). The gastric microbiota richness index (observed OTUs, Chao) was significantly lower in CAG patients than in nCAG patients (P <0.05). Compared with avirulent Hp-II infection, virulent Hp-I infection significantly decreased the Shannon index in CAG patients (P <0.05). In nCAG patients, Hp-I infected patients had lower abundances of several dominant gastric bacteria (Aliidiomarina, Reyranella, Halomonas, Pseudomonas, Acidovorax) than Hp-II infected patients. Meanwhile, in CAG patients, Hp-I infected patients occupied lower abundances of several dominant oral bacteria (Neisseria, Staphylococcus and Haemophilus) than Hp-II infected patients. In addition, bile reflux significantly promoted the colonization of dominant oral microbiota (Veillonella, Prevotella 7 and Rothia) in the stomach of CAG patients. There was no significant symbiotic relationship between Helicobacter bacteria and non-Helicobacter bacteria in the stomach of nCAG patients, while Helicobacter bacteria distinctly linked with the non-Helicobacter bacteria (Pseudolabrys, Ralstonia, Bradyrhizobium, Mesorhizobium and Variovorax) in CAG patients.

Conclusions: Virulent Hp infection alters the gastric microbiota, reduces microbial diversity, and enhances the symbiotic relationship between the Helicobacter bacteria and non-Helicobacter bacteria in patients with chronic gastritis. The data provides new evidence for treating Hp infection by improving the gastric microbiota.

Keywords: Helicobacter pylori; chronic gastritis; gastric mucosa; microbiota; symbiotic relationship.

Figure 1

Effect of Hp-infection typing on the alpha diversity of gastric microbiota in patients with chronic gastritis. Non-parametric Mann-Whitney U test was used to conducted the different analysis, and the asterisk (*) indicated P < 0.05.

Figure 2

Effect of Hp-infection typing on gastric microbial community in patients with chronic gastritis. (A) The stacked bar chart presented the community of gastric microbiota in total population, Hp-I infected patients, and Hp-II infected patients at phylum level. (B) Volcano plot showed the different bacteria between nCAG and CAG patients at OTU level.

Figure 3

Different analysis of gastric microbiota in CAG patients with Hp infection. (A) Based on nCAG patients, Venn analysis of gastric different OTUs in CAG patients was performed among total population, Hp-I infected patients, and Hp-II infected patients. (B) The abundant OTUs was related to CAG risk. (C) The abundant OTUs was related to the risk of Hp-I induced CAG. (D) The abundant OTUs was related to the risk of Hp-II induced CAG.

Figure 4

Effect of Hp-infection typing on the symbiotic network of gastric microbiota in patients with chronic gastritis. (A) Spearman correlation analysis was conducted to build the symbiotic networks at OTU level in CAG and nCAG patients, respectively. The red nodes referred to Helicobacter OTUs and the green nodes referred to non-Helicobacter OTUs, the size of node mean the relative abundance, the length of line mean absolute value of correlation coefficient (r). (B) Correlated number of 66 differential non-Helicobacter OTUs was extracted from the symbiotic network of gastric microbiota in CAG and nCAG patients. (C) Correlated number of eight Helicobacter OTUs was extracted from symbiotic network of gastric microbiota in CAG and nCAG patients. The symbiotic network was displayed between eight Helicobacter OTUs and non-Helicobacter OTUs in CAG (D) and nCAG patients (E), respectively. The yellow solid line indicated positive correlation, while the blue dashed line indicated negative correlation.

Figure 5

Effect of Hp-infection typing on the COG predictive function of gastric microbiota in patients with chronic gastritis. (A) Based on the nCAG patients, Volcano plots presented the different COG predictive functions of gastric microbiota in total population, Hp-I infected patients, and Hp-II infected patients, respectively. (B) Venn analysis was performed to screen the sharing and unique differential KEGG predictive functions. (C) The abundances of 82 sharing differential COG predictive functions were demonstrated in the four groups.

Figure 6

Effect of living habits on gastric microbiota in patients with chronic gastritis. (A) Based on the nCAG patients, Volcano plots presented the different gastric OTUs in drinking patients, no-drinking patients, smoking patients, and no-smoking patients. (B) Venn analysis was performed to screen the sharing and unique different OTUs. (C) The 20 shared OTUs were presented in patients with certain living habit.

Figure 7

Hp-infection typing exerted deeply impact on the gastric microbiota in the chronic gastritis patients. This figure was drawn by Figdraw (https://www.figdraw.com/).