| BackgroundCystitis glandularis is a mucosal proliferative disease with follicular solid lesions.The diagnosis of cystitis glandularis depends on biopsy under cystoscopy.Brunn's nest in the epithelium of bladder mucosa can be diagnosed as cystitis glandularis.So far,little is known about the pathogenesis of the disease.In recent years,the incidence of cystitis glandularis has been increasing year by year,which has become a common and frequent disease in urology.Patients often have no specific clinical manifestations,most of which are urinary tract irritation symptoms,such as frequency of urination,urgency and lower abdominal pain.Chronic infection is an important cause of cystitis glandularis.Inflammation is the essential feature of the disease.The relationship between microorganisms and infection and inflammation is inseparable.It has been found that diseases of multiple systems and organs of the human body have an important relationship with the imbalance of bacterial flora in the body,and bacterial flora often lead to the occurrence of diseases by participating in the metabolic process of the human body.Microbiomics and metabonomics are the rapid development of recent years in the study of microbial structure and metabolite composition of the omics technology,which provides a good technical platform for in-depth study of the structural differences and functions of microbial communities.In view of the close relationship between cystitis glandularis and infection,with the help of microbiomics and metabolomics technology,we analyzed the composition of bacteria and metabolites in bladder,intestinal tract and blood circulation system of cystitis glandularis patients,and explored the pathogenesis and diagnosis of cystitis glandularis.ObjectiveBy means of microbial diversity sequencing and metabonomics,the structure and metabolic characteristics of microbial community in bladder and intestinal and blood circulation closely related to cystitis glandularis will be studied.We will study the "core microbial group of cystitis glandularis" leading to the occurrence of diseases,and the disease diagnosis model based on different microflora and different metabolites will be established.To analyze the correlation between differential microflora and differential metabolites,and to elucidate the pathogenesis of cystitis glandularis.Materials and Methods1.Biological sample collection.It mainly includes collection of urine,feces and blood samples from patients with cystitis glandularis and normal controls.The admission criteria for patients with cystitis glandularis were:?1?pathologically diagnosed cystitis glandularis;?2?patients included in this study needed to be male because of gender factors can influence the results of the experiment;?3?age between 20 and 70 years;?4?BMI requirements between 18.5 and 24;?5?informed consent was signed and certified by ethics committee of Shanghai Changhai Hospital.The admission criteria of the normal control group were as follows:?1?physical examination was conducted in the physical examination center of Shanghai Changhai Hospital from January 2017 to December 2018,and the results were normal;?2?male participants were included;?3?age was between 20 and 70 years old;?4?BMI requirement was between 18.5 and 24;?5?informed consent was signed and certified by the ethics committee of Shanghai Changhai Hospital.Patients or healthy individuals with cystitis glandularis who received antibiotic treatment or consumed probiotic food for a long time in the past month could not be included in this study.2.Microbial diversity sequencing?1?Genomic DNA extraction and quality controlGenomic DNA was extracted from urine,feces and blood samples by genomic DNA extraction kit,and the extraction method was operated strictly according to the kit instructions.After the genomic DNA was extracted,the concentration and purity of DNA were detected by ultraviolet spectrophotometer.The extraction quality of DNA was detected by 1% agarose gel electrophoresis.?2?Design and synthesis of primersThe selection region for 16 S r DNA amplification was V3-V4.The common primers used were 341F?5'-ACTCCTACGGGAGCAGCAGCAG-3'?and 806R?5'-GGACTACHVGGGTWTCTAAT-3'?.?3?PCR amplification and product purificationUsing dilute genomic DNA as template,the V3-V4 region of 16 S r RNA gene was amplified by PCR using high fidelity enzyme,and PCR amplified product was detected by 2% agarose gel electrophoresis.After the target fragment was cut,the target fragment was recovered by gel Recovery Kit.?4?Quantification and homogenization of PCR productsQuantitative analysis of PCR products was carried out,and the products were mixed according to the corresponding proportion according to the data requirement of a single sample.?5?SequencingMicrobial diversity was sequenced using Illumina's Miseq platform,followed by bioinformatics analysis.The original data were mosaic and quality control.Reads were arranged from large to small according to abundance,and then OTUs were randomly selected for each sample.Then,Alpha diversity index dilution curve was analyzed to classify each OTU species.After classification,OTU abundance tables were obtained according to the number of sequences in each OTU,and subsequent bioinformatics analysis was carried out according to OTU abundance tables.?6?Bioinformatics analysisBioinformatics analysis of microbial diversity sequencing mainly includes the analysis of the composition of each sample at different classification levels;the analysis of the differences in flora structure between different samples and different groups;the construction of disease diagnosis model and evaluation of the diagnostic efficiency of the model based on different flora;and the prediction of the metabolic function of the sample flora based on 16 S r RNA gene sequencing results.3.Metabolomic analysis?1?Collection of biological samples.It mainly includes collection of urine and feces samples from patients with cystitis glandularis and normal controls.The criteria for enrollment of patients with cystitis glandularis and healthy controls were the same as those in the microbiological diversity sequencing mentioned above.?2?Testing.Urine metabonomics was analyzed by LC-MS.Agilent 1290 Infinity Ultra High Performance Liquid Chromatography and Agilent 6538 UHD and Accurate-Mass Q-TOF mass spectrometry were used for LC-MS analysis.Gas chromatography-time-of-flight mass spectrometry?GC-TOFMS,Pegasus HT,Leco Corp.,St.Joseph,MO,USA?is used for the detection of metabolites between bacteria and host.?3?Data analysis.Agilent Masshunter Qualitative Analysis B.04.00 software was used to convert the original data into a general format and conduct multivariate statistical analysis.The original data of fecal samples were exported to Metabolomics Metabolism Software Xplore MET?Metabo-Profile,Shanghai,China?through Chroma TOF?v4.51.6.0,Leco.,CA,USA?software for preliminary data processing and multivariate statistical analysis.The multivariate statistical analysis used in this study is mainly principal component analysis,partial least squares discriminant analysis,orthogonal partial least squares discriminant analysis,etc.Through analysis,the differences of metabolites and metabolic pathways between different samples and different groups are discussed,and the disease diagnosis model based on differential metabolites is constructed.4.Correlation Analysis of Differential Microflora and Differential MetabolitesUsing the difference significant bacterial flora in microbial diversity sequencing and the difference significant metabolite data in metabolomics detection,the correlation analysis of the selected difference bacterial flora and the difference metabolite data was carried out by Spearman correlation analysis method.Including correlation analysis between urine differential flora and urine differential metabolites;correlation analysis between intestinal differential flora and intestinal differential metabolites;correlation analysis between intestinal differential flora and urine differential metabolites.Results:1.Urine microbial diversity analysis included 70 patients with cystitis glandularis and 74 controls.There were 5704 OTUs in cystitis glandularis patients group and 6301 OTUs in normal control group.The number of OTUs in disease group was significantly lower than that in control group.Pan/Core OTU analysis showed that the amount of samples included in this study was sufficient,which could effectively reflect the species richness and the number of core species in urine samples of two independent groups.According to the gradient of flora abundance,Proteobacteria,Firmicutes,Bacteroidetes and Actinobacteria ranked the top four in the two groups.The abundance of Proteobacteria and Actinobacteria in the patients with cystitis glandularis was significantly higher than that in the normal control group,while the abundances of Firmicutes and Bacteroidetes in the normal control group were significantly higher than those in the patients with cystitis glandularis.The first four dominant genera with the highest abundance were Cupriavidus,Burkholderia-Paraburkholderia,Brucellaceae and Pelomonas.Among the four dominant genera,the abundance of Cupriavidus bacteria in patients with cystitis glandularis was significantly higher than that in normal control group.The difference in genus level between the two groups was Acinetobacter,Escherichia-Shigella,Anoxybacillus,Geobacillus,Lactobacillus and Brevundimonas.Alpha diversity analysis showed that the level of bacterial diversity in patients with cystitis glandularis was lower than that in normal control group.There was a significant difference in the degree of bacterial diversity?evenness?between the two groups?Shannon index: 3.13 vs 3.59,P = 0.01;Simpson index: 0.15 vs 0.10,P = 0.008?.Beta diversity analysis showed that there were significant differences in flora structure between the two groups.Among the top 50 flora of genus level,there were 17 different flora in the two groups?Wilcoxon rank-sum test,P < 0.05?.A random forest diagnosis model was constructed to distinguish the flora structure of the two groups by using 17 different flora.ROC curve analysis showed that the area under the curve was 0.84,which indicated that the diagnostic model had a high diagnostic accuracy and could effectively distinguish the patients with cystitis glandularis from the normal control group.2.Intestinal microbial diversity analysis included 72 patients with cystitis glandularis and 84 controls.There were 1078 OTUs in cystitis glandularis patients group and 1093 OTUs in normal control group.Pan/Core OTU analysis showed that the sample size included in this study was sufficient and could effectively reflect the species richness and core species number in urine samples of two independent groups.At the level of phylum,Firmicutes,Bacteroidetes,Proteobacteria,Actinobacteria and Fusobacteria were the most abundant phylum in the two groups.The abundance of Proteobacteria and Fusobacteria in cystitis glandularis group was significantly higher than that in normal control group?Proteobacteria,9.40 vs 3.57%,P=1.068e-7;Fusobacteria,1.28 vs 0.73%,P=0.000004?.Actinobacteria abundance in the normal control group was significantly higher than that in the cystitis glandularis group?6.01 vs 2.80%,P=0.0078?.At genus level,the top four dominant bacteria in the two groups were Bacteroides,Faecalibacterium,Prevotella9 and Subdogranulum in turn.The difference in genus level between the two groups was Bifidobacterium,Escherichia-Shigella,Lachnoclostridium and Lachnospira in order of abundance.Alpha diversity analysis showed that the flora richness of cystitis glandularis group was lower than that of normal control group?Chao index: 280.45 vs 292.56;ACE index: 283.78 vs 298.0?.There was no significant difference in diversity between the two groups.Beta diversity analysis showed that there were significant differences in flora structure between the two groups.Among the top 50 flora of genus level,there were 22 different flora?Wilcoxon rank-sum test,P < 0.05?between the two groups.A random forest diagnosis model was constructed based on 22 different flora.The area under ROC curve is 0.88,which indicates that the diagnostic accuracy of the model is high.3.Blood microbial diversity analysis included 16 patients with cystitis glandularis and 16 controls.There were 1576 OTUs in cystitis glandularis patients group and 1714 OTUs in control group.The highest abundance of the first six phyla in the two groups was Proteobacteria,Firmicutes,Deinococcus-Thermus,Bacteroidetes,Actinobacteria and Chloroflexi.The abundances of Proteobacteria and Actinobacteria in patients with cystitis glandularis were significantly higher than those in normal controls?Proteobacteria,90.78 vs 86.61%,P=0.00049;Actinobacteria,1.51 vs 1.29%,P=0.006287?.Firmicutes,Bacteroidetes and Deinococcus-Thermus in the normal control group were significantly higher than those in the glandular cystitis group?Firmicutes,4.57 vs 2.80%,P=0.00049;Bacteroidetes,2.078 vs 1.09%;Deinococcus-Thermus,3.31 vs 2.17%,P=0.003091?.At the genus level,the order of the two groups was Burkholderia-Paraburkholderia,Cupriavidus and Acinetobacter.The abundance of Acinetobacter in the patients with cystitis glandularis was significantly higher than that in the control group?32.88 vs 5.59,P = 0.00002235?.Alpha diversity analysis showed that the flora richness of cystitis glandularis group was significantly lower than that of normal control group?Chao index: 361.95 vs 403.04,P = 0.004988;ACE index: 357.06 vs 395.28,P = 0.003991?.The diversity of bacterial flora in cystitis glandularis group was significantly lower than that in normal control group?Shannon index: 2.23 vs 2.65,P = 0.0001305;Simpson index: 0.26 vs 0.21,P = 0.03121?.Beta diversity analysis showed that there were significant differences in flora structure between the two groups.Among the top 50 flora of genus level,there were 26 different flora?Wilcoxon rank-sum test,P < 0.05?between the two groups.A random forest diagnosis model was constructed by using 26 different flora.The area under ROC curve is 0.88,which indicates that the diagnostic accuracy of the model is high.4.Urine metabonomics analysis included 30 patients with cystitis glandularis and 29 controls.PCA and PLS-DA analysis showed that the trend of separation between the two groups was obvious.Sixty-one differential metabolites were screened out.The contents of 4-Hydroxy-5-?dihydroxyphenyl?-valeric acid-O-methyl-O-sulphate,Uric acid and 2,6 Dimethylheptanoyl carnitine were higher in cystitis glandularis group than in normal group.The other differential metabolites were lower than those in normal control group.Differential metabolites contain a variety of amino acid metabolites,such as L-Tyrosine,L-Isoleucine,L-Phenylalanine,Lys-Trp-OH,L-Glutamine,etc.5.Totally 30 patients with cystitis glandularis and 29 controls were included in the analysis of co-metabolism between bacteria and host.PCA analysis,PLS-DA analysis and OPLS-DA analysis showed that the patients with cystitis glandularis and the normal control group showed a distinct trend of separation.A total of 24 different metabolites were screened,mainly amino acids,organic acids and indoles.Among the differential metabolites,Hydroxypropionic acid and Citramalic acid were higher in cystitis glandularis group than in normal group,while the other 22 differential metabolites were lower in cystitis glandularis group than in normal control group.Boruta algorithm based on random forest was used to construct a disease diagnosis model.The area under ROC curve was 0.86,which showed that the diagnosis model had high accuracy and could effectively distinguish two groups of metabolites.6.There are strong positive correlations between urine differential metabolites and urine differential bacteria: 2,6 Dimethylheptanoyl carnitine and Actinobacteria;4-Hydroxy-5-?dihydroxyphenyl?-valeric acid-O-methyl-O-sulphate and Tencuteries,Actinobacteria;Uric acid and Actinobacteria;N-stearoyl and Proteobacteria,alpha-CEHC and Proteobacteria,Pyrogacid and Proteobacteria.7.There is a strong correlation between intestinal differential flora and intestinal differential metabolites at the phylum level.Actinobacteria are strongly and positively correlated with L-Histidine,L-Phenylalanine,L-Serine,L-Tyrosine,L-Tryptophan.Fusobacteria are strongly and negatively correlated with L-Isoleucine,3-Indolepropionic acid,1H-Indole-3-acetamide,L-Valine,Hydrocinnamic acid,N-acetyltryptophan,L-Tptophan.Han,L-Homoserine.Proteobacteria was negatively correlated with L-Histidine?L-histidine?and Valeric acid.8.The correlation analysis between intestinal differential flora and urine differential metabolites showed that at gate level,Proteobacteria was negatively correlated with 2,5,7,8-Tetramethyl-2-?2'-carboxyethyl?-6-hydroxychroman?alpha-CEHC?,N-stearoyl valine,Pyroglutamic acid and other metabolites,while Fusobacteria was negatively correlated with alpha-CEHC.CONCLUSIONThe composition of bacterial flora in bladder,intestinal tract and blood circulation of patients with cystitis glandularis has changed significantly compared with healthy people.The increase of "harmful" bacterial abundance and the decrease of "beneficial" bacterial abundance in disease group accord with the common characteristics of the changes of bacterial flora-related diseases.The abnormal increase of the abundance of Proteobacteria is the common characteristics of the three groups,and is also the structural change of the whole bacterial flora of the disease.The prominent characteristics of these three factors indicate that there are important interrelated and regulatory mechanisms among them,and blood circulation may be the link between them.There were significant differences in urine and intestinal metabolites between the disease group and the control group.The correlation analysis between microflora and metabolites showed that Proteobacteria was closely related to short-chain fatty acid metabolism,vitamin E metabolism and amino acid metabolism,suggesting that Proteobacteria might play a role in the pathogenesis of cystitis glandularis through the above related metabolic pathways.The disease diagnosis model based on differential microflora and differential metabolites has high accuracy and can effectively distinguish the disease group from the control group.This study is helpful to further understand the important role of microbial imbalance and metabolism in the occurrence of human diseases,and to provide new strategies for the diagnosis and treatment of bladder diseases. |
PDF Full Text Request