| The oral cavity harbors one of the most diverse microbiomes within the human body,including millions of bacteria,fungi and viruses.Dysbiosis of the microbial ecosystem is closely linked to various oral diseases,including dental caries,periodontitis and oral squamous cell carcinoma.Furthermore,oral microbes are found disseminated from the oral cavity to other body sites through the digestive tract or other routes,and contribute to systemic diseases like gastric cancer,inflammatory bowel disease,colorectal cancer,cardiovascular diseases and Alzheimer’s disease.Despite the crucial role of oral microbes in human health,its fundamental dynamics still remains elusive.Longitudinal oral microbiome studies to date are limited and are usually sampled in days,months or years.However,the oral cavity is frequently perturbed by daily activities such as food intake,teeth brushing and sleep.Therefore,high-frequency sampling is required to reveal the precise dynamics of the oral microbiome.Saliva is a non-invasive and easily accessible fluid.Here,we used the salivary microbiome as a model and conducted a high-frequency time series sampled at minute intervals.The high-resolution dynamics of the salivary microbiome under daily perturbations is fully investigated using bioinformatics analysis.The main contents and results were summarized as follow:(1)The fine-scale dynamics of the salivary microbiome and the role of daily activities such as eating and sleeping were investigated.Time series dataset sampled at10-to 60-min intervals was established.3 healthy adults were continuously tracked for 6 days.A total of 614 saliva samples were collected in this data set.About 17 million reads were obtained using 16 S amplicon sequencing.Microbiome analyses showed that this high-frequency sampling strategy did not systematically change the salivary microbiome and can be used for its dynamic study.Our dense time series dataset revealed that,despite the relative stability within each individual over time,the composition of the salivary microbiome was affected by daily activities and exhibited regular and highly dynamic fluctuations.For example,the relative abundance of Bacteroidetes increased during sleep;the relative abundance of Fusobacteria shifted around eating activities.Permutational multivariate analysis of variance(PERMANOVA)showed that eating and sleeping activities significantly affected the salivary microbiome.The functional capability of the salivary microbiome was predicted using PICRUSt2 algorithm,which was found to be significantly affected by eating activities.Microbial source tracking analysis combined with oral samples from the Human Microbiome Project(HMP)showed that the contributions of distinct oral niches to the salivary microbiome were also dynamically affected by daily activities.Notably,the sources of supragingival and subgingival plaque profoundly rose after eating.Therefore,our high-frequency time series reveals highly dynamic daily changes in the salivary microbiome.(2)Microbial temporal and spatial data were integrated to identify the co-occurrence patterns of salivary microbes shared among individuals.Different ASVs within Granulicatella,Haemophilus,Neisseria,and Streptococcus genus were found to be negatively correlated using Spearman’s correlation,and tended to be enriched on different oral surfaces.Co-occurrence network analysis based on large-scale longitudinal data reveals novel co-occurring relationships between microbes,which were associated with the intricate biogeography of the oral microbiome and were observed in multiple individuals: Haemophilus,Bergeyella HMT 206,Streptococcus,and Granulicatella elegans had high co-occurrence;Porphyrobacter tepidarius,Acidovorax temperans,and Delfia acidovorans had high co-occurrence;tongue-specific Campylobacter consisus and Oribacterium sinus had high co-occurrence;another cluster of tongue-specific microbes,including multiple Prevotella/Alloprevotella species and SR1 HMT 875,had high co-occurrence;dental plaque-specific Lautropia mirabilis,Rothia aeria,and Neisseria oralis had high co-occurrence.These observations were reproducible when different correlation and normalization methods were employed,which suggests that these conclusions were not artifacts caused by bioinformatic algorithms or the compositionality of microbiome data.This study uncovers novel co-occurrence patterns associated with site specificity of the oral microbiome.(3)The periodicity of the oral microbiota was next evaluated.Diurnal bacteria and their oscillation patterns were analyzed.The Lomb-Scargle algorithm indicated that the overall microbiome composition exhibited a significant oscillation with the periodicity of 24 h in all participants.A total of 5 phyla,14 genera and 22 ASVs that oscillated diurnally were identified.Most of the diurnal genera were also reported in another independent study.Despite diurnal variations among different subjects,consistent oscillation patterns were also observed in multiple subjects.At the phylum level,the relative abundances of Fusobacteria and Actinobacteria were higher in the daytime,whereas the relative abundances of Bacteroidetes and SR1 were higher in the evening.At the genus level,the relative abundance of Bergeyella slowly increased in the daytime;the relative abundance of Prevotella was higher in the evening;the relative abundance of Veillonella was higher in the daytime.At the ASV level,two specific diurnal patterns were identified: the relative abundances of Bergeyella HMT206 and Haemophilus were high in the evening,decreased upon waking,and slowly increased in the daytime.Multiple ASVs within Prevotella/Alloprevotella genus increased at night.This increasing trend apparently lasted beyond waking,until breakfast when there was a drastic reduction.The results reveal diurnal oscillations of the salivary microbiome,providing insights into the homeostasis of the oral microbiota.(4)Considering the significant effect of eating on the salivary microbiome,eating-responsive microbes were next determined and were further verified in an independent larger cohort.Food-derived chloroplast and mitochondria,fermented dairy products-introduced Lactobacillus species and other rare non-oral ASVs were detected in eating samples and disappeared soon afterward.Eating-responsive bacteria were identified using a self-developed z-score method.Their abundances uniquely increased after eating but not after other activities.The rise of these eating-increased microbes displayed a specific temporal order.Tongue-specific Campylobacter concisus and Oribacterium sinus increased earlier than dental plaque-specific Lautropia mirabilis,Neisseria oralis,and Rothia aeria.PICRUSt2 analysis showed that the relative abundance of KEGG pathways related to bacterial mobility significantly increased after eating,including bacterial chemotaxis,flagellar assembly and two-component system.Additionally,Prevotella nanceiensis,SR1 HMT875,and Prevotella melaninogenica were found decreased during eating.Our nutritional intervention experiment(including 60 saliva samples)and a third-party public dataset(soda intervention dataset,including 7 individuals and 19 saliva samples)together suggested that carbohydrates may not give rise to these eating-associated microbial changes.These observations were further validated in another independent cohort comprising 19 individuals--the lunch dataset(including222 saliva samples)and the validation dataset(including 94 saliva samples),and were shown to be reproducible in multiple subjects and across a longer period,suggesting potential generalizability in the population.Collectively,this study reveals novel dynamics of the salivary microbiome closely related to eating and diurnal oscillations in multiple datasets and cohorts,laying a scientific foundation for future microbiome-targeted tools for nutritional intervention and disease diagnosis and treatment.We also highlight the need to consider the influence of daily activities and diurnal oscillations in cross-sectional studies. |
PDF Full Text Request