The Composition And Diversity Of Gastric Microbiome In Patients With Gastric Carcinoma And Polyposis

Background: The stomach is a special area in the gastrointestinal microecosystem. Itsunique ecological environment and characteristic microbial community are due to gastricacid secretion. It was once believed that gastric acid could kill the bacteria entering thestomach and that the stomach environment was not suitable for bacterial colonization.Advances in high-throughput DNA sequencing and bioinformatics analysis technologypermit the genetic analysis of complex microbial populations without requiring priorcultivation. Gastrointestinal microbiology has entered the metagenomics era and become ahot topic worldwide. However, most studies in this area have focused on the compositionand function of the intestinal microbiota and the relationship between intestinal microbiotaand metabolic diseases/intestinal disorders. Recently, only a few studies were performed toinvestigate the composition of gastric microbiota under health and gastric disorders basedon the sequencing methods, and showed that huge numbers of bacteria other thanHelicobacter pylori exist in the stomach. But the real composition of gastric microbiota inChinese healthy populations and patients with gastric disorders is still unclarified. Gastriccarcinoma (GC) as one of the most common malignancies worldwide, the pathogenesis ofwhich is a process of multiple stages and steps affected by multiple factors that involve ahuge number of molecules and complex regulation networks. The cause of GC is still notclear despite deep decades-long studies. It is generally believed that environmental, dietary,H. pylori infection and genetic factors participate in the pathogenesis of GC. It is stillunclear whether the microbiota in stomach is altered among GC patients. To date, no studyon the relationship between stomach flora and GC using high-throughput sequencingtechnology based on the metagenomics strategy has been performed. In addition, gastricmultiple polyposis (GP) are focal elevated lesions within the gastric epithelium mucosa.The current limited systematic studies of GP focus mainly on the relationship between GPand GC as well as the role of H. pylori in the pathogenesis. The etiology and biologicalcharacteristics of GP and its long-term effects on the human body are not yet clear. Theintestinal flora are involved in the pathogenesis of colonic polyps, while no study focusingon the relationship between stomach flora constitution and GP pathogenesis. In total, it isnecessary to investigate that whether the gastric microbiota in patients with gastric disorders is different from that in health subjects and further to clarify that whether thegastric dysbiosis contributes to the pathogenesis of gastric disorders.Aims: To describe the16S rRNA gene profiling of the gastric mucosal-associated bacterialmicrobiota in patients with GC and GP, and in health controls (HC).Methods: This study was registered at WHO ICTRP (ID: ChiCTR-OCC-12002573). Allsubjects were included after providing informed consent. The inclusive and exclusivecriteria for subjects in GC, GP and HC group were formulated. From May2012toDecember2013, the consecutive patients scheduled for GC and GP were included in thisstudy. Healthy subjects with normal gastric mucosal morphology and without any knownformer or current diseases were assessed as HC. Gastric mucosal biopsies were collectedduring routine endoscopy at the endoscopic center of the Chinese PLA General Hospital.Total genomic DNA from each sample was extracted with the QIAamp DNA Mini kitcombined with the glass beat beater method. PCR amplifications were conducted in withthe515f/806r primer set that amplifies the V4region of the16S rDNA gene.. The reverseprimer contains a6-bp error-correcting barcode unique to each sample. Bacterial16SrRNA V4gene was sequenced on the Illumina Miseq high-throughput sequencing platform.Pairs of reads from the original DNA fragments are merged by using FLASH softwareSequencing reads was assigned to each sample according to the unique barcode of eachsample. Sequences were analyzed with the QIIME software package and UPARSE pipeline,in addition to custom Perl scripts to analyze alpha and beta diversity. Sequences wereassigned to the operational taxonomic units (OTUs) at97%similarity. In order to computealpha divesity, we rarified the OTUs table and calculated three metrics: Chao1metric, theobserved species metric and Shannon index. Rarefaction curves were generated based onthese three metrics. QIIME calculated both weighted and unweighted unifrac, which arephylogenetically aware measures of beta diversity. We used unweighted unifrac to doprincipal coordinate analysis (PCoA) and unweighted pair group method with arithmeticmean (UPGMA) clustering.Results: Ninety subjects of the Han population in the Northen areas of Chinese mainlandwere enrolled: GC group (n=30)(mean age58.2±14.6years),19males and11females,all adenocarcinoma (18non-cardia and12cardia); GP group (n=30)(mean age56.7±14.2years),13males and17females,20fundic gland polyps and10hyperplastic polyps; HCgroup (n=30)(mean age45.6±14.8years),15males and15females (P>0.05). A total of240gastric biopsy samples were obtained:90in GC group (1gastric antrum,1gastric body and1cancer tissue for each subject),90in GP group (1antrum,1body and1polyps),and60in HC group (1antrum and1body). H. pylorir DNA was detected in75%(9/12)GC patients,69.2%(18/26) GP patients and70%(21/30) healthy subjects respectively, whohave been tested negative for H. pylori by conventional laboratory methods. The relativeabundance of H. pylori is similar among these three groups (P>0.05). Alpha diversityanalysis based on the Chao1metrics, rank abundance curves, Shannon index andrarefaction curves suggested a good sequenced coverage, and a good balance between thedepth of sequencing and the species richness among GC, GP and HC group. The bacterialcommunities varied among GC, GP and HC group. In general, more than two hundredphylotypes from nearly twenty bacterial phyla were identified on the gastric mucosa of GC,GP or HC group. Moreover, in HC group, the dominant phyla Proteobacteria, BacteroidetesActinobacteria, Firmicutes, Crenarchaeota, Acidobacteria, Verrucomicrobia, Cyanobacteria andNitrospirae are present in a vast majority. In addition, Proteobacteria, Crenarchaeota,Bacteroidetes, Firmicutes, Actinobacteria, Verrucomicrobia and Acidobacteria were thedomiant phyla with the abundance up to1%in GC group. In GP group, the most commonphyla include Proteobacteria, Firmicutes, Bacteroidetes, Actinobacteria, Acidobacteria,Crenarchaeota and Verrucomicrobia. Furthermore, at the genera level, the dominantgenera are also distinct among GC, GP and HC group, which are needed to be furtherinvestigated. The unweighted unifrac PCoA showed the close clustering of gastric antrumand body samples in HC group (PC36.5%and PC29.6%), and was also evidenced by theUPGMA clustering. The bacterial microbiota composition of the normal mucosa of gastricantrum and body, and the cancer/polyps tissues in GC/GP group could not be distinguishedby PCoA analysis. Moreover, the gastric bacterial communities between non-cardia andcardia cancer of GC group (PC112.7%and PC28.1%), and which between fundic glandpolyps and hyperplastic polyps of GP group (PC111.06%and PC29.04%), were also notclustered by PCoA and UPGMA analysis. However, beta diversity analysis indicated aseparate clustering of the gastric microbial communities between GC and HC group (PC133.6%and PC26.3%), GP and HC group (PC135.1%and PC26.7%), and GC and GPgroup (PC118.1%and PC25.7%).Conclusions:1. The human stomach harbors an abundant and diverse mucosal associatedbacterial microbiota that varies by individuals.2. There are little differences in bacterialmicrobiota between the gastric antrum and body in healthy subjects; the bacterialcommunities in the normal mucosa could not be distinguished from those in the gastric lesions such as cancer and polyps tissues; which indicate that the bacterial communities instomach may be not influenced by the anatomical sites and pathological types.3. Thedistinct gastric bacterial composition and diversity can be observed in patients with gastricadenocarcinoma and those with gastric multiple polyposis, compared with healthyindividuals with normal gastric mucosal morphology.4. The gastric dysbiosis could bepresent in those patients with gastric neoplasia. The gastric neoplasia associated bacterialcommunities or genus/species should be further identified based on the large amount ofsequenceing data, and be verified by functional studies.