The study was designed as a single-center, randomized, double-blind, placebo-controlled, and prospective clinical trial, adhering to the principles outlined in the Declaration of Helsinki, which obtained approval from the Medical Ethics Committee of the Nanjing First Hospital (KY20220314-05) and registered with the Chinese clinical trial registry (CHICTR2200060023).

From December 27, 2021 to June 22, 2022, a cohort of 100 H. pylori positive patients, ranging in age from 18 to 65, diagnosed by 13C-UBT (DOB > 6‰), histology, or culture, were recruited from Nanjing First Hospital in China. The main exclusion criteria were as follows: (1) allergy to the study drug; (2) patients with chronic gastritis with a confirmed peptic ulcer; (3) patients who had previously received H. pylori eradication therapy within 6 months; (4) use of antibiotics, bismuth, and probiotics within 4 weeks or H2 receptor antagonists, proton pump inhibitor (PPI), and potassium competitive acid blockers (P-CAB) within 2 weeks; (5) use of adrenocortical hormones, non-steroidal anti-inflammatory drugs, or anticoagulants; (6) history of esophageal or gastric surgery; (7) pregnant or lactating women; (8) suffer from a serious illness such as liver disease, cardiovascular disease, lung disease, kidney disease, or tumor; (9) alcoholism. Twenty H. pylori negative healthy volunteers (DOB < 4‰ of 13C-UBT) were selected as normal controls who had no history of smoking or alcohol consumption and oral disease.

The biostatisticians programmed the generation of the random allocation sequence for this clinical trial using the SAS 9.4 statistical analysis system and achieved it through the block randomization method with a block length of 4. Subjects meeting the enrollment criteria were randomly assigned to receive the vonoprazan-amoxicillin regimen along with probiotics in the BtT group or the placebo in the PT group for 14 days in a 1:1 ratio. The study medicine was administered to enrolled subjects based on their given randomization number, ensuring that both the subjects and the administering researchers remained blinded.

The BtT group was administered a regimen consisting of amoxicillin 750 mg qid, vonoprazan 20 mg bid, and Bifidobacterium tetravaccine tablets (3 tablets tid) for 14 days. The probiotics contained 0.5 g per tablet. The PT group received the identical vonoprazan-amoxicillin dual therapy in conjunction with the placebo, which was made of starch and had the same weight, appearance, and taste as Bifidobacterium tetravaccine tablets, except that it lacked the active ingredient. Both the probiotics and placebo were provided by Hangzhou Grand Biologic-Pharmaceutical, Inc. And the probiotics or placebo must be taken more than 2 h apart from amoxicillin.

Tongue coating samples were collected from H. pylori positive patients at three time points (before H. pylori eradication (W0), after H. pylori eradication (W2), and at confirmation of H. pylori infection cure (W6)) and H. pylori negative patients. Before the collection process, the oral cavity of each individual was washed with a normal saline solution two to three times. Sterile cotton swabs were utilized to collect samples from a specific region measuring 2 cm × 2 cm located in the central section of the dorsal surface of the tongue. Following 30 s of wiping, the samples were subsequently transferred into sterile test tubes and stored in a refrigerator set at a temperature of −80°C until they were deemed suitable for use.

Total genomic DNA was isolated using the CTAB technique. The concentration and purity of the DNA were evaluated on a 1% agarose gel. Specific primers (341F, CCTAYGGGRBGCASCAG, 806R, GGACTACNGGGTATTAAT) and barcodes were applied to amplify 16S genes from separate locations (V3-V4). All PCR reactions were performed using 15 μL of Phusion® High-Fidelity PCR Master Mix (New England Biolabs), 0.2 μM of forward and reverse primers, and roughly 10 ng of template DNA. The thermal cycle underwent initial denaturation at 98°C for 1 min, followed by denaturation at 98°C for 10 s, annealing at 50°C for 30 s, elongation at 72°C for 30 s, and finally elongation at 72°C for 5 min. The same amount of IX loading buffer (containing SYB green) was mixed with PCR products and detected by electrophoresis on a 2% agarose gel. The PCR products were mixed at an equal density ratio, then purified with the Qiagen Gel Extraction Kit (Qiagen, Germany). Sequencing libraries were generated using the TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, United States) following the manufacturer’s recommendations, and index codes were added. Finally, 250 bp paired-end reads were acquired by sequencing on the Illumina NovaSeq platform.

First, the technician screened the raw data and discarded any sequences that were shorter than 230 bp, had a quality score of ≤20, contained ambiguous bases, and did not precisely match the primer and barcode tag sequences, and then separated the sequences using sample-specific barcode tags. Then, the high-quality sequences were de-noised using the unoise3 method with usearch11 to produce amplicon sequence variants (ASVs; Edgar, 2016). Finally, all sequences were classified into different taxonomic groups against the SILVA138 database by the BLAST tool.

Qiime (v1.8.0) was used to calculate the richness and diversity indices (Chao1 richness index and Shannon diversity index) based on the ASV information. To compare the membership and structure of microbial communities in different samples, heat maps of the top 20 ASVs were generated using Mothur (Jami et al., 2013). According to the results of taxonomic annotation and relative abundance, PCoA based on bray-curtis distance and PLS-DA were performed using R packages (v3.6.0) to test the similarity between different samples (Wang et al., 2012). LEfSe was used to measure the unique species between H. pylori positive and negative groups, as well as before and after the H. pylori eradication. LDA value >3.0 was considered statistically significant. Microbial function was predicted by using PICRUST2 in the KEGG database.¹ Co-occurrence network analysis was performed by the spearman method.

SPSS 25.0 statistical software was used to process the data, and the clinical data were compared by analysis of variance. Continuous data were presented as mean ± standard deviation. Disaggregated information is expressed as frequencies or percentages. Comparisons of measured information among groups were analyzed by one-way ANOVA, and categorical information was analyzed by Chi-squared test. When p < 0.05, the data exhibited statistical significance.

A total of 120 participants were included in the study, consisting of 100 H. pylori positive patients and 20 H. pylori negative individuals. Tongue coating samples were not collected from 40 H. pylori positive patients because of rejection or loss of follow-up. Finally, 60 H. pylori positive patients (referred to as the Hp group) who had undergone successful H. pylori eradication treatment, with 29 patients belonging to the BtT group and 31 patients belonging to the PT group, and 20 H. pylori negative individuals (referred to as the NT group) were selected for oral microbiota analysis (Figure 1).

There were no significant differences among the BtT group, PT group, and NT group in demographic characteristics such as age, sex, BMI index, smoking, and alcohol (Supplementary Table S1).

H. pylori was detected in the oral cavity of infected subjects (34/60, 56.7%) and uninfected subjects (7/20, 35%) with no statistical differences (p > 0.05). The detection rate among eradicated subjects was 1.7% (1/60) by high-throught sequencing, significantly lower than the other two groups (p < 0.01).

The KEGG analysis demonstrated statistically significant variations in the signaling pathways of 16, 12, and 10, respectively, between the time points W0 and W2 (Figure 6A), W2 and W6 (Figure 6B), and W0 and W6 (Figure 6C) in the BtT group. The PT group exhibited variations of 16, 18, and 23 routes at W0 and W2 (Figure 7A), W2 and W6 (Figure 7B), and W0 and W6 (Figure 7C), respectively. The dissimilarities in signaling pathways between W0 and W2 primarily included genetic information processing, environmental information processing, celluar process, human diseases, and metabolism. After H. pylori eradication, there was a decrease in the signaling pathways associated with genetic information processing (including folding and sorting and degradation), environmental information processing (including signal transduction and signaling molecules and interaction), cellular process (including cellular motility and cellular community), and human diseases (including cancer, viral infectious disease, and substance dependence). Conversely, the signaling pathways involved in metabolism (including metabolism of cofactors and vitamins) and human diseases (including drug resistance: antineoplastic, endocrine and metabolic disease, immune disease, and neurodegenerative disease) experienced a decrease. The aforementioned modifications were partially reverted to their original levels. Moreover, notable disparities were seen between 4 signaling pathways in the BtT group (Figure 6D) and 8 signaling pathways in the PT group (Figure 7D) at W0, W2, and W6.

H. pylori is a gram-negative bacteria that predominantly inhabits in the epithelial and mucus layers of the gastric mucosa and can be transmitted by both the oral-oral and fecal-oral routes, which establishes the oral cavity as a possible reservoir of H. pylori (Al et al., 2014). Previous studies have identified the presence of H. pylori in dental plaque (Mendoza-Cantu et al., 2017), saliva (Seligova et al., 2020), tongue coating (Zhao et al., 2019), and dental pulp (Nomura et al., 2018). Nevertheless, the association between oral H. pylori and gastric H. pylori is still not well understood. According to a study (Yee, 2016), there was a notable disparity in the detection rate of oral H. pylori between H. pylori positive patients and H. pylori negative people (80% vs. 23%), suggesting a potential association between gastric H. pylori and the occurrence of oral H. pylori infection. However, Ji et al. (2020) found no statistical differences in the detection of tongue coating H. pylori among infected, uninfected, and eradicated patients through high-throughput sequencing, which suggested that there is likely no substantial association between oral H. pylori and gastric H. pylori. In our study, we observed a greater incidence of tongue coating H. pylori among H. pylori positive patients compared to H. pylori negative subjects. Nevertheless, the observed disparity did not exhibit statistical significance (p > 0.05). It was worth mentioning that the presence of H. pylori on the tongue coating significantly exhibited a substantial decrease following H. pylori eradication (p < 0.01). The decrease in H. pylori presence may be explained by two factors. Firstly, the use of antibiotics to treat H. pylori in the gastric tract inadvertently leads to the elimination of H. pylori in the oral cavity as well (Bago et al., 2011). Secondly, it is possible that H. pylori in the oral cavity is either temporary or exists as nonviable bacterial fragments. The abundance or scarcity of H. pylori in the oral cavity is significantly impacted by its identified depth and specific location.

The consistency of the results on the impact of H. pylori infection on oral bacteria diversity is not entirely uniform, perhaps due to regional disparities, variances in sample sources, and discrepancies in sample sizes. Multiple studies have demonstrated that there are no significant disparities in the alpha and beta diversity of the oral microbial community between H. pylori positive and negative individuals (Li et al., 2021; Wu et al., 2021). Other research suggested that there are noticeable variations in the beta diversity of the salivary microbiota between H. pylori infected and uninfected subjects (Ji et al., 2020). In their study, Chua et al. (2019) conducted an investigation wherein oral cavity samples were obtained from a total of 24 subjects during daylight hours. The researchers observed that H. pylori positive patients exhibited a notably lower Shannon index compared to H. pylori negative population (p = 0.03). Furthermore, the study revealed the presence of distinct oral microbial community structures between the groups infected with H. pylori and those who were not infected. Otherwise, a study conducted by Hu et al. (2022), demonstrated that H. pylori positive patients had higher richness and diversity and better evenness of oral microbiota than normal controls. Furthermore, a significant distinction in clustering patterns was observed between the two groups (p < 0.05). The results of our study indicate that there was a higher richness and diversity of tongue coating microbiota in H. pylori infected patients compared to healthy controls (p < 0.05 and p > 0.05, respectively). Additionally, there were significant differences in the bacterial composition (p < 0.05). In a comprehensive study conducted by Yamashita and Takeshita (2017), sequencing techniques were employed to examine the composition of the salivary microbial community. The results of this investigation revealed a positive correlation between an elevated abundance of the salivary microbiota and compromised oral health, which suggests a potential impairment of oral health due to H. pylori infection, potentially mediated by an augmentation in the diversity of oral flora.

The study revealed that the two groups under investigation exhibited a constant presence of the five primary phyla, namely Bacteroidetes, Firmicutes, Proteobacteria, Fusobacteria, and Actinomycetes, in a specific order. This finding is in accordance with previous research (Ji et al., 2020; Hu et al., 2022). There were notable differences in the oral microbial makeup between the two groups. The results of the LEfSe analysis revealed that there were discernible taxonomic differences between the two groups, which is consistent with previous research findings (Ji et al., 2020; Hu et al., 2022). However, these findings contradict the results reported by Li et al. (2021). In our study, we observed significant enrichment of Fusobacterium and Haemophilus in H. pylori infected individuals. Fusobacterium is classified as a commensal oral bacterium; however, Fusobacterium nucleatum is hypothesized to be linked to the development of periodontal disease (Signat et al., 2011). Haemophilus is generally regarded as non-pathogenic to humans; nonetheless, there have been documented cases linking Haemophilus parainfluenzae to endocarditis (Revest et al., 2016) and pseudomembranous colitis (Liu Q. et al., 2022). Porphyromonas, together with Fusobacterium and Haemophilus, is recognized as a significant indicator of tongue coating microbiota in H. pylori infected individuals, as shown in our results. Ahn et al. (2012) have suggested that specific microorganisms found in the mouth cavity, such as Porphyromonas and Fusobacteriaceae, have the potential to disturb the microecology within the stomach and contribute to the development of chronic inflammation, which may serve as a potential precursor to gastric cancer. In a study conducted by Hu et al. (2015), notable reductions were observed in the relative prevalence of Porphyromonas, Fusobacterium, and Haemophilus among individuals diagnosed with stomach cancer. In contrast, Stasiewicz and Karpinski (2022) have established a notable correlation between Porphyromonas gingivalis, a widely recognized oral pathogen, and Fusobacterium nucleatum, with the occurrence of oral squamous cell carcinoma. There is a suggestion that infection with H. pylori could potentially influence the colonization of oral bacteria, hence potentially impacting the overall health of the oral cavity.

Through our analysis of the oral microbiota in H. pylori positive patients and healthy controls, we have identified variations in the expression of signaling pathways. H. pylori infection has the potential to influence the gene expression of tongue fur flora, which may establish a connection between H. pylori infection and the development of cardiovascular diseases and metabolic diseases such as lipid metabolism. Prior studies (Chen et al., 2020; Sun et al., 2023) have provided evidence of a correlation between H. pylori infection and metabolic conditions, including dyslipidemia, hypertension, diabetes, and cardiovascular diseases. Nevertheless, the precise mechanism remains uncertain and could potentially be associated with chronic inflammatory responses, endothelial damage, and hyperhomocysteinemia induced by H. pylori infection (Vijayvergiya, 2015). Potential areas of investigation in future research may involve examining the correlation between H. pylori infection and metabolic illnesses, focusing on a microecological perspective.

The precise mechanism underlying the impact of H. pylori infection on oral microbiota composition and metabolism has yet to be fully elucidated. Insufficient research has been conducted on this topic; however, it may be linked to modifications in intragastric acidity, virulence factors, cytokines, and the immunological response of the host. A previous study has shown that the use of PPI can have an impact on oral flora (Mishiro et al., 2018). It is hypothesized that H. pylori infection may lead to the production of significant quantities of urease, an enzyme responsible for the breakdown of urea into carbon dioxide, hence reducing intragastric acidity. On the other hand, the virulence factors associated with H. pylori, including CagA and VacA, have the ability to consistently trigger cellular inflammatory reactions, leading to a reduction in stomach acid output. The modification of gastric acid secretion has the potential to result in a reduction in oral PH, which can have implications for the composition of the oral microbial population. The impact of H. pylori infection on the intestinal flora is mediated by this particular mechanism, as observed in the study conducted by Iino and Shimoyama (2021). Furthermore, existing evidence suggests that the presence of H. pylori infection triggers the activation of cytokines inside the inflamed gastric mucosa. These cytokines, such as TNFα and IL-6, play a crucial role in the development of metabolic diseases (Buzas, 2014). Hence, we propose the presence of a link between H. pylori infection and alterations in the metabolic pathways of the oral microbiota. The association between H. pylori infection and changes in host immunity has been postulated (Cheok et al., 2022). Therefore, any modification in the immune response has the potential to destabilize the oral microbial ecology.

Following the successful H. pylori eradication, we observed a significant decrease in alpha diversity and a notable alteration in the composition of the oral microbiota. These differences did not fully revert to the baseline level at W6. Additionally, notable discrepancies were seen in the signaling pathways of the oral microbiota at W0, W2, and W6. In summary, H. pylori eradication with vonoprazan-amoxicillin regimen has been observed to lower the diversity of tongue coating flora and impact its structure and function. Despite undergoing therapy, the flora exhibited only partial recovery within a limited timeframe, failing to reach the levels observed before eradication. The use of bismuth-containing quadruple therapy, intended for H. pylori eradication, has been demonstrated to result in a reduction in alpha diversity and alterations in the composition of the salivary microbiota (Ji et al., 2020). Jakobsson et al. (2010) observed alterations in oral microbiota for a period of 4 years after H. pylori eradication with a 7-day triple therapy and found that the diversity decreased immediately after eradication, then gradually returned to the original level over time. H. pylori eradication regimen consists mainly of antibiotics and PPI. Previous studies have shown that the use of antibiotics (Abeles et al., 2015) and PPI (Mishiro et al., 2018) may have an impact on oral flora, aligning with the aforementioned discoveries. Alternatively, it seems that the magnitude of the impact is associated with the length of treatment time and amount of antibiotics and PPI administered. Hu et al. (2022) investigated the efficacy of low-dose amoxicillin (2.0 g daily) and high-dose amoxicillin (3.0 g daily) combined with vonoprazan for 7 or 10 days to eradicate H. pylori, then found the salivary flora diversity was not affected and the change in oral microbial community structure was minimal after eradication, which differs from our findings. The observed discrepancy could potentially be attributed to the prolonged length of therapy and the administration of a higher dosage of the medicine in our prescribed regimen.

There was a consistent drop in the relative abundance of Firmicutes and Lactobacillus following H. pylori treatment. The observed decline in the relative abundance of Firmicutes was found to be statistically significant, aligning with the findings reported by Hu et al. (2022). Periapical periodontitis is attributed to the presence of polymicrobial infections, characterized by the coexistence of many bacterial species. Among them, Firmicutes have been identified as the most prevalent and abundant (Korona-Glowniak et al., 2021). At the genus level, there was a notable drop in the relative abundance of Lactobacillus, a bacterium that has been associated with the progression of dental caries (Zhang and Xu, 2022). The eradication of H. pylori has the potential to impede the proliferation of oral infections.

A prior study (Pellicano et al., 2020) has found an association between H. pylori infection and the occurrence of type 2 diabetes mellitus, autoimmune diseases, neuropathy, and metabolism-related diseases. Moreover, some studies have reported the potential improvement of glycemic control in type 2 diabetes mellitus and symptom control in Parkinson’s disease with H. pylori eradication (Pierantozzi et al., 2006; Song et al., 2021). Nevertheless, there is still limited comprehension of the pathophysiology of these disorders, and the precise impact of H. pylori infection and its eradication remains uncertain. Further research is necessary to fully explore these relationships. The findings of our research suggest that the eradication of H. pylori can impact the expression of signaling pathways associated with the aforementioned diseases. This provides an opportunity to investigate the connection between H. pylori and metabolic disorders, immune disorders, and neurodegenerative conditions from a microecological standpoint.

The results of our study indicate that there were noticeable differences in the alpha diversity, compositional structure, and metabolic pathways of the tongue microbiome before and after the eradication of H. pylori. Furthermore, it can be observed that the level of variation was similar in both the groups receiving probiotics and receiving the placebo. This suggests that the simultaneous administration of probiotics did not provide any relief in terms of the disruption to the structure and function of the oral flora. Probiotics are bacteria that provide benefits to the host organism, and numerous studies have demonstrated the efficacy of co-administering probiotics in the eradication of H. pylori for the regulation of gastrointestinal flora (Kakiuchi et al., 2020; Yuan et al., 2021; He et al., 2022). Nevertheless, there remains a divergence of opinions regarding the precise function of probiotics in the context of oral microecology. The literature has demonstrated that probiotics possess the ability to disturb oral pathogenic biofilms and uphold a state of equilibrium within the oral microecology (Nguyen et al., 2021; Radaic et al., 2022). For instance, Bifidobacteria have the ability to impede the proliferation of Porphyromonas gingivalis (Jasberg et al., 2016). In a recent meta-analysis conducted by Liu J. et al. (2022), a total of 11 randomized controlled trials were examined to assess the effects of probiotics on oral flora. The findings of this analysis indicated that there was no significant difference observed in the levels of gingival inflammation and the presence of oral pathogens, such as Porphyromonas gingivalis, Prevotella intermedia, and Fusobacterium nucleatum, between the groups receiving probiotics and those receiving the placebo. Dassi et al. (2018) conducted a study to analyze the impact of probiotics on oral microbiota in healthy volunteers through 16S amplicon sequencing before, during, and after the administration of probiotics. The results showed an increase in the diversity of the salivary microbiota, but the overall structure remained unaltered. To far, there has been a scarcity of research investigating the impact of probiotics on oral flora during the H. pylori eradication. A randomized, placebo-controlled trial (He et al., 2022) investigated the impact of probiotics combined with bismuth quadruple therapy on the oral microbiota. The study revealed that there was no statistically significant disparity in the variety of salivary flora between the groups administered with probiotics and those given the placebo. However, it did notice a more restricted impact on the organization of the microbial community. The probiotics group showed a notable decrease in the prevalence of harmful microorganisms such as Porphyromonas, Actinobacillus, and Prevotella. Nevertheless, the group administered with probiotics in our study did not exhibit a reduced disruption in terms of diversity, structural composition, and function of the tongue flora, and we took into account the following factors: (1) The utilization of dual therapy in this study to eradicate H. pylori, which solely involved the use of one antibiotic, inherently resulted in a minor impact on the composition of oral flora (Hu et al., 2022). (2) The guidelines for the administration of Bifidobacterium bifidum tetrapartum tablets indicate that penicillin hampers the viability of live bacteria. Although, in our study, patients were instructed to take amoxicillin and the probiotics at least 2 h apart. (3) Probiotics gain entry into the gastrointestinal tract via the oral cavity, where the potential for colonization by probiotics may be limited. (4) The oral microbiota is significantly influenced by local oral hygiene practices, and this particular source of variability posed challenges in maintaining control during our clinical study.