Microbiota in Reflux Esophagitis and Peptic Ulcer Disease

The Microbiome in Reflux Esophagitis

Gastroesophageal reflux disease (GERD) pathology is related to the repeated backflow of acidic gastric fluid into the esophagus. ¹ Gastroesophageal reflux disease occurs as a results of several inter-related factors including defective esophageal sphincter function, impaired esophageal clearance, as well as delayed gastric emptying. Several environmental factors that contribute to the development of GERD include diet, alcohol use, tobacco use (e.g., smoking), and recently the microbiota. Gastroesophageal reflux disease can be further classified into erosive reflux disease and non-erosive reflux disease based on the presence of mucosal erosion. ¹ Furthermore, exposure to refluxate and bile acids is a known risk factor for the development of Barrett's esophagus and progression to esophageal adenocarcinoma. ² Thus, understanding what drives reflux and whether the microbiota is involved in this process is fundamental to understanding, preventing, or delaying disease progression.

Analysis of the mucosal-associated microbiota of healthy individuals (NORM), patients with reflux (GERD), and those with metaplasia (MET) highlighted compositional differences. ³ A multiomic approach was undertaken to simultaneously analyze transcriptomics as well as microbiota and immune cell infiltration of NORM-, GERD-, and MET-biopsied tissue. ³ No major differences were observed in immune cell infiltration across groups with one exception: lower levels of resting and higher levels of active mast cells were identified when comparing MET with either GERD or NORM. Similarly, when evaluating transcriptomics, the differences were primarily in the MET group. There were 7,884 genes that were differentially regulated between MET and NORM, but only 12 genes were differentially regulated between GERD and NORM, five of which modulated in GERD in the same direction as MET. ³ Modulated pathways were related to genes involved in mucosal functionality including mucin glycan biosynthesis, acid and bile secretion, and carcinogenesis. These data indicate that there are profound transcriptome changes that occur in patients with metaplasia but not in those with GERD indicating that another factor likely drives GERD genesis and progression.

When the microbiota of the oropharynx and the esophagus was analyzed, the two microbiota demonstrated key differences. The differences were more pronounced in the GERD and MET groups compared with NORM with Streptococcus as the taxon mostly differentially represented among groups. When controlling for subject age, gender, proton pump inhibitors (PPI) use, body mass index, reflux symptoms, and disease, Campylobacter arose as the taxon that principally characterized the esophageal microbiota in patients with GERD and MET due to enrichment. Interestingly Campylobacter also correlated with infiltration by activated mast cells. ³ Based on the assumption that the esophageal microbiota principally resembled that of saliva, ⁷ Campylobacter species were isolated from saliva samples in each of the three groups. These isolates were tested for their proinflammatory potential, but no major differences were observed among groups. However Campylobacter rectus and some Campylobacter concisus GS1 isolates from GERD or MET demonstrated an enhanced ability to resist destruction within macrophages, elicited a more intense epithelial cell inflammatory response, and drove the expression of cancer genes. ³ Hence, this study links two surgically relevant clinical conditions, GERD and MET, with the presence of virulence factor enabled Campylobacter species capable of surviving longer in macrophages that also activates gene sequences believed responsible for esophageal mucosal carcinogenesis.

Another possible explanation for mucosal damage induced by acidic reflux comes from a rodent experimental reflux esophagitis (RE) model. ⁴ Proteomic analysis discovered that acyl-CoA synthetase long-chain family 4 (ACSL4), a key enzyme for ferroptosis (death induced by the accumulation of lipid peroxides due to cellular free iron) was a possible therapeutic target of RE. ⁴ Free iron can lead to the overgrowth of pathogenic bacteria within the microbiota, a condition present during dysbiosis that has been termed pathobiosis. The authors found that changes in esophageal and gut microbiota composition, together with increased circulating endotoxin (lipopolysaccharide [LPS]) triggered ferroptosis, suggesting a series of linked events that lead to mucosal damage. ⁴ Intervention with a combination of the ACSL4 inhibitor rosiglitazone (ROSI) and ferrostatin-1, a ferroptosis inhibitor, reduced mucosal injury and offered a potential therapeutic intervention to mitigate against the consequences of RE.

Because the microbiota is involved in the pathogenesis of RE, it is unsurprising that a concomitant Helicobacter pylori infection may further worsen dysbiosis. In fact, Helicobacter pylori infection is tied to non-malignant upper gastrointestinal conditions such as GERD, Barrett's esophagus, and peptic ulceration. ⁵ Saliva samples of patients with RE versus control individuals were analyzed taking into consideration the presence or absence of Helicobacter pylori infection. ⁶ Bacteroidetes were enriched in the RE group, whereas the control group was characterized by higher abundance of firmicutes. In a subpopulation of patients with RE, Helicobacter pylori infection reduced beta-diversity but enriched Veillonella, Haemophilus, Selenomonas, Megasphaera, Oribacterium, Butyrivibrio, and Campylobacter. ⁶ This indicates that Helicobacter pylori infection further exacerbates RE patient dysbiosis, and increased the prevalence of maladaptive species such as Campylobacter, which is associated with the mucosal metaplasia phenotype. ³

In a pediatric investigation, devoid of the confounders of smoking and alcohol intake, the oral, esophageal and gastric microbiota were analyzed for bacterial (16S rRNA) and fungal (ITS2) amplicon sequencing. ⁷ Healthy individuals demonstrate different microbiota profiles at each site, with the predominance of taxa flowing from the oropharynx. Importantly, gastroesophageal reflux led to more homogeneous bacterial but not fungal profiles at each site. ⁷ In patients with metaplasia, Prevotella enrichment was observed that followed a gradient of increase from healthy controls to either GERD, eosinophilic esophagitis, or metaplasia patients. Moreover, this association was confirmed using publicly available datasets as well. Within the same population, the use of a PPI was associated with an increased density of oral bacteria in gastric fluid and a proinflammatory profile characterized by increase of interleukin (IL)-1 concentration. ⁷

An Israeli study evaluated the effect of asymptomatic Helicobacter pylori infection on the intestinal microbiota (16S rRNA gene sequencing) in 169 children (six to nine years old) of diverse socioeconomic backgrounds. ⁸ The presence of Helicobacter pylori infection was confirmed by fecal antigen detection. Lower socioeconomic status correlated with Helicobacter pylori infection and was noted in 57% of children. Although no association was observed between microbiota diversity and Helicobacter pylori infection, Helicobacter pylori infection did correlate with enriched Prevotella copri and Eubacterium biforme species. Microbiota composition was also associated with socioeconomic status and likely reflects similarities in dietary patterns within similar social environments. ⁸ The long-term implications and clinical consequences of these findings, as well as therapy for asymptomatic Helicobacter pylori infection, remain to be determined.

The Microbiome in Peptic Ulcer Disease

The incidence of peptic ulcer disease (PUD) and chronic gastritis is increasing in developing countries.^9,10 The role of the microbiota in these disorders is increasingly important as new insights into pathogenesis and therapy are discovered. Oxidative stress has been implicated in the pathogenesis of gastrointestinal mucosal disorders, including PUD, but how and whether this is related to a dysbiotic microbiome remains to be elucidated. ¹¹ A recent study on a limited number of patients (10 with PUD and 10 with non-ulcer dyspepsia [NUD]) analyzed the microbiota (16s rRNA gene sequencing) in endoscopic gastric mucosal biopsies and correlated it to the tissue oxidation-reduction potential (ORP) measured on adjacent biopsy material. ¹² The authors analyzed the microbiota taxonomically, but also evaluated the presence of genes encoding for antioxidant defense mechanisms, and for the use of oxygen during respiration. These data were compared with the ORP, which serves as a direct measure of oxidative stress. In particular, they assessed the abundance of microorganisms preferring high rather than low ORP. The proportion of organisms that prefer a high ORP (aerobes) compared with a low ORP (anaerobes) allows one to determine the Microbial Redox Index that may serve as an indicator of mucosal health. Unsurprisingly, the gastric mucosa was characterized by aerobic species, including Helicobacter pylori and Sphingobacterium mizutaii. There were no major species differences between those with and without PUD, but high redox potential, alkaline gastric pH, correlated with Helicobacter pylori abundance.

Certain species correlated with gastrointestinal hemorrhage including Acinetobacter gulliouiae, Sulfuricurvum kujiense, Lactococcus garvieae, Butyriccoccus pullicaecorum, and Bifidobacterium longum. These data provide a rational basis for therapies that alter the redox state of the gastric mucosa including bisulfite and glutathione. Oxidation reduction potential monitoring may be a useful tool in assessing the impact of specific dietary interventions to reduce PUD promotion or progression including pre-, pro-, or synbiotics.

Helicobacter pylori infection is often associated with PUD. Stools from 19 Helicobacter pylori-infected patients and 16 controls were analyzed for microbiota and metabolome composition; importantly infected individuals were from a socioeconomically challenged and underprivileged inner city urban community embracing both environmental and dietary influences on the microbiota. ¹³ Similarly to children, Helicobacter pylori infection affected microbiota alpha diversity. However, the major difference in adult Helicobacter pylori patients was the enrichment of Atopobium, Gemellaceae, Micrococcaceae, Gemellales, and Rothia (R. mucilaginosa) and a decrease in the relative abundance of the phylum Verrucomicrobia compared with control subjects. Verrucomicrobia includes Akkermansia muciniphila, one of the bacteria that principally associates with healthy microbiota and that when administered to patients mitigates against metabolic syndrome. ¹⁴ Therefore, dysbiosis that decreases Verrucomicrobia bears potential implications for more than one organ system. The metabolomic analysis highlighted an increase in polyunsaturated fatty acids (PUFA) 22:4n6, 22:5n3, 20:3n6, and 22:2n6 and a decrease in 18:3n6 with the imbalances correlating with a deranged Bacteroidetes:Firmicutes ratio; this ratio serves as a measure of microbiota balance. Much larger studies have also linked microbiota composition with the amount of fecal Helicobacter pylori antigen. ¹⁵

Early metagenomic analyses suggested the presence of three discrete enterotypes that defined well-balanced host microbiome states. ¹⁶ Prevotella enrichment in particular correlates with Helicobacter pylori infection as Prevotella degrades mucin and may work in concert with the desfufating bacteria, Desulfovibrio, to increase the rate of glycoprotein degradation and increase ulceration risk. Both Prevotella and Desulfovibrio populate the more inflammatory, enterotype 2 profile. ¹⁶ Moreover, the species composition of each enterotype does not necessarily guarantee the presence of specific molecular functions, suggesting both crosstalk and collaboration between microbiota communities. For example, although the most abundant microbiota-related functions are provided by the most abundant species, there are a number of high-frequency functions that are provided by low-abundance species, especially regarding events that increase bacterial survival such as pilus assembly and plasmid transfer. ¹⁶

A related study of 55 children with gastrointestinal symptoms and a clinical indication for gastroduodenoscopy collected biopsy specimens for microbiota analysis (16S rRNA sequencing). ¹⁷ Thirty-seven were Helicobacter pylori pylori-positive (23 nonpeptic ulcer and 14 peptic ulcer) and 18 were Helicobacter pylori pylori-negative; Helicobacter pylori presence did not correlate with ulcer presence. After four weeks, a second biopsy was collected from those children (n = 11) undergoing eradication therapy. Eradication was successful in restoring a diversified microbiota, suggesting that it is important to monitor for Helicobacter pylori infection even in patients without ulcer formation, and that eradication is required to assure a eubiotic microbiota. Such therapy relies on antibiotic therapy that may be only one-arm or microbiome-directed therapy that should be used to help restore balance and adaptive function.

Microbiota Interventions in RE

Current treatments such as PPIs counteract acid-driven mucosal injury, but do not address the root cause of reflux. Moreover, PPIs may worsen dysbiosis by reducing gut microbiota diversity while enriching specific oral microbial taxa, most notably Streptococcaceae.^18,19 Long-term PPI can also lead to small intestinal bacterial overgrowth (SIBO),^18,20,21 which can foster chronic inflammation, reduce the reactivity of esophageal smooth muscle cells, ²² and increase methane production (which reduces intestinal transit and gastric emptying), ²³ thereby favoring reflux of gastric fluid into the esophagus. These studies suggest that current treatments may substantially exacerbate dysbiosis instead of favorably shaping or restoring the microbiota. Microbiota balance and homeostasis appears to be essential for those with GERD to avoid progression to Barrett's esophagus and esophageal metaplasia.

Focused inquiry into the role of microbiome therapeutics, including planned withdrawal of PPI therapy, seems warranted to reduce GERD-driven complications including malignancy. Gastroesophageal reflux disease is not the sole clinical condition that leverages PPI as a routine therapeutic. A currently clinical trial is exploring the role of PPI compared with no PPI in patients with complicated hepatic cirrhosis; time to rehospitalization or death will be evaluated as a composite primary end point. ²⁴ If the no PPI (placebo) arm demonstrates equivalency (non-inferiority) then it may provide a rational basis for exploring PPI withdrawal in those with GERD or RE provided they do not have an active ulcer, and if there is a non-PPUI therapeutic such as a microbiota-reshaping diet.

Microbiota transplantation is an effective tool to modulate the whole microbiota population of the recipient with a eubiotic one. This strategy has been shown to be effective in controlling Clostridium difficile infection as well as other conditions. ²⁵ A recent study in which washed microbiota transplantation (WMT) was undertaken in patients with non-erosive reflux disease but PPI-dependency versus a control population without microbiota transplantation has demonstrated encouraging results. ²⁶ Total remission rate was evaluated at one month and it was higher in patients undergoing WMT (n = 15; 93.3%) versus PPI (n = 12; 41.7%) group. The authors also succeeded in reducing the PPI dose in 80% of patients in the WMT group as well as 33.3% in the PPI group. Only one patient of 15 in the WMT group experienced a side effect (loose stool). Hence, this small but important study provides evidence that it is possible to reduce RE by a microbiota-based intervention. Washed microbiota transplantation is a powerful tool but it is not yet available in many hospitals or to a large proportion of patients based on a wide variety of barriers including donor sourcing and WMT batch homogeneity and dose. Alternatives may be considered, including deliberate diet manipulation, provision of probiotics, and administration of postbiotics (i.e., bacterial metabolites); the latter have recently emerged as novel therapeutics.^27–30

A randomized clinical trial was performed comparing placebo (n = 64) to probiotic administration (n = 66; Bacillus subtilis and Enterococcus faecium) in patients with RE being treated with PPI for 8 weeks. ³¹ Although there was no effect on healing rate or Reflux Disease Questionnaire score (a self-administered symptom assessment tool), patients in the probiotic group experienced reduced diarrhea symptoms and a lower SIBO rate. Furthermore, the risk of relapse was lower in the probiotic than in the placebo group during 12 weeks of follow-up. ³¹ A similar double-blinded randomized clinical trial was performed comparing placebo in 15 control patients with probiotic therapy in 15 patients (8 different strains, including Streptococcus thermophilus BT01; Bifidobacterium breve BB02, Bifidobacterium animalis subsp. Lactis BL03 and B. animalis subsp. lactisBI04; Lactobacillus acidophilus BA05, Lactobacillus plantarum BP06, Lactobacillus paracasei BP07, and Lactobacillus helveticus BD08); all patients had RE and were receiving PPI therapy. ³² In this study, shotgun metagenomic sequencing and metabolomic analyses were performed on stool samples, before, during, and after treatment. Unsurprisingly, PPI intervention modulated the microbiota with an enrichment in Streptococcaceae, Leuconostacaceae, and Pasteurellaceae at both family and genus levels and was accompanied by an increase in the metabolites, glycine, arginin, valine, and phenylacetic acid. ³² Probiotic intervention counteracted the microbial changes induced by the PPI with a significant reduction in the Leuconostacaceae family and led to fecal accumulation of 1H-Indole-4-carbaldehyde. ³² This is a metabolic product of tryptophan catabolism and is likely to be involved in binding the aryl hydrocarbon receptor and mediating the release of IL-22 which is important for mucosal healing. ³³ These data importantly identify that probiotic supplementation—at least in this focused fashion—can impact microbiota composition as well as function (metabolism) in a way that addresses the consequences (mucosal injury) of a clinical condition (RE).

In another study of patients with duodenal ulceration (DU) the combination of PPI and antibiotic therapy for Helicobacter pylori infection with or without probiotics was assessed for its impact on the intestinal microbiota; a control group of healthy individuals was included as well. ³⁴ The clinical trial was randomized and patients received PPIs plus clarithromycin and amoxicillin (PCA) or PCA plus Bacillus subtilis and Enterococcus faecium (BSEF). Fecal microbiome composition was analyzed by 16S rRNA sequencing. Patients with DU demonstrated decreased microbiota diversity prior to therapy compared with healthy controls. The use of PPI plus antibiotic agents further decreased diversity in the PCA group. Bacillus subtilis and Enterococcus faecium patients demonstrated no change in microbiota composition or diversity. Over the 10 weeks of monitoring, those in the PCA group consistently maintained microbiota different from those of healthy controls. In contrast, BSEF patients, supplemented with probiotics, slowly normalized their microbiota composition and diversity and approached that of healthy controls. These results confirm that Helicobacter pylori reduces intestinal microbiota diversity, but also suggest that PPI and antibiotic therapy accompanied by with probiotics can normalize microbiota composition. Furthermore, despite concerns regarding probiotic use during critical illness (bacteremia and recovery from presumed sterile sites), probiotic therapy in outpatients with clinical conditions related to dysbiosis appears to demonstrate a substantially different safety and efficacy profile.

Conclusions

Both RE and PUD are clinical conditions that are clearly linked to abnormal microbiota composition and diversity. Current therapies address complications and sequelae but much less frequently address the underlying dysbiosis. Unfortunately, some current therapeutics including PPIs and antibiotics may further derange the microbiota. Understanding how the microbiota interface with each of these clinical conditions has uncovered maladaptive influences on carcinogenesis as well as host metabolism. Specific therapies that support, re-shape, or repair the microbiome should be investigated as complementary therapies to existing approaches, and in some circumstances, alternative therapies that replace current approaches.