Hundreds of bacterial species inhibit the oral cavity of humans; most of them are commensal and important for keeping equilibrium in our mouth ecosystem. However, some of this oral microbiome have a key role in the development of oral and non-oral diseases. The makeup of the normal flora may be influenced by many factors, including age, sex, genetics, stress, nutrition and diet of the individual. Our purposes were to utilize culture-independent molecular techniques to extend our knowledge on the breadth of bacterial diversity in the healthy human oral cavity in Egyptian individuals. Samples were collected from oral cavity of seventeen healthy subjects. Genomic DNA was extracted and used to amplify bacterial 16S rRNA genes which were sequenced on Illumina MiSeq platform. Over 2 million reads from the metagenomes of these seventeen samples were taxonomically assigned through a combined procedure provided that common species belong to five main phyla. Firmicutes which was the most predominant phyla followed by Proteobacteria, Bacteroidetes, Fusobacteria and Actinobacteria. Streptococcus, Veillonella, Prevotella, Neisseria, Haemophilus, Porphyromonas, Fusobacterium, Gamella and Granulicatella were the most predominant genera. Our data suggest that environmental factors can affect the composition of the oral microbiome in certain geographical regions. This study presents a new entry for subsequent analyses of human microbiome in Egypt.
Egypt, Oral microbiome, Metagenomes, Illumina
Oral microbiology is the study of the microorganisms of the oral cavity and the interactions between the oral microorganisms with each other and with the host 1 & refers to the population of microorganisms that inhabit skin and mucous membranes of normal healthy oral cavity and GIT. The oral microbiome plays a relevant role in the health status of the host and is a key element in a variety of oral and non-oral diseases. Some of these bacteria involved in oral diseases such as caries and periodontitis. In addition, specific oral bacterial species have been implicated in several systemic diseases, such as bacterial endocarditis 2, aspiration pneumonia 3, osteomyelitis in children 4and preterm low birth weight5. So, it is important to fully define the human microflora of the healthy oral cavity before we can understand the role of bacteria in oral disease. Several isolates belonging to the dominant bacteria in healthy individuals were cultured and shown to inhibit the growth of cariogenic bacteria, suggesting the use of these commensal bacterial strains as probiotics to promote oral health and prevent dental caries6.
Unlike most infectious diseases which a single pathogen may be found responsible for the infection, oral diseases appear to be resulted from the outcome of multiple microorganisms. In periodontitis, for instance, at least three bacterial organisms have been found to be directly associated with the development of the disease7. Similarly, the complexity of the microbial community in the oral cavity makes the identification of the single etiological agent for dental caries extremely difficult. So, dental caries is probably better understood as a polymicrobial disease 8where the interactions and synergistic effect of many multiple species should be considered for the future strategy of diagnosis, treatment and prevention. It has been demonstrated that Streptococcus sobrinus and above all S. mutans are acidogenic and play an important role in caries initiation9. However, using of molecular techniques like PCR amplification and cloning of the 16S rRNA gene have uncovered that a high number of samples derived from oral cavities do not contain mutans streptococci, whereas other acid-producing bacteria are present10. These bacteria include Lactobacillus, Actinomyces or Bifidobacterium. Recent molecular techniques have confirmed all these results and expanded the list of potential cariogenic species to Veillonella, Propionibacterium and Atopobium, among others11, most of which are poorly characterized species.
Traditional culture based techniques are unable to meet the demands of scientists to explore new world of bacterial communities. These techniques are restricted to very limited range 0.01-1% of bacteria which only survive in standard laboratory conditions excluding uncultivable bacteria 12-15. Because of there is no way for a high portion of oral bacteria to be cultured by the current laboratory methods, the presence of molecular techniques has provided significant improvement in understanding of the oral microbiota. Because PCR amplification still have a few significant bias that prevent microbial diversity to be fully characterized and studied, that many species and DNA fragments cannot be identified and detected, a metagenomic approach by which that a total DNA from a microbial community is obtained obviating the need for culture or PCR amplification and introduced a promising strategy for studying the full genetic pool of the human microbiome in health and disease16. In addition, the extraordinary increase in sequencing and the reduction in costs of next generation sequencing that have been applied to the study of oral microbiota, providing a high and more complete picture of the human bacterial communities. The aim of our study is to characterize healthy oral associated bacterial communities in samples taken from Egyptian volunteers.
Materials and Methods:
This study was approved by scientific research ethics committee at the faculty of pharmacy, Suez Canal University, Egypt (reference number of 20156H1). Seventeen subjects representing both genders, ranging in age from 25 to 50 participated in our study. The oral health status of each one of the individuals was evaluated and examined by a dentist following the instructions of Human Microbiome Project. Before sampling, at screening time, subjects must fulfill the following characters; no antibiotics or corticosteroids use for previous three months, no periodontal or related systemic disease, no gingivitis or dental caries, subjects either will be excluded if they have; chronic dry mouth, as assessed through questioning of the subject by an experienced clinician, periodontal pockets > 4 mm, untreated cavitated carious lesions or oral abscesses, evidence of precancerous or cancerous lesions, evidence of candidiasis, clinically meaningful halitosis as determined by organoleptic assessment by an experienced clinician, More than 10% of sites with BOP(Bleeding on probing), as stated by human microbiome project manual of procedures17 . From seventeen healthy volunteers who passed screening for required criteria, biological samples were collected from their oral cavities. The volunteers were asked not to brush their teeth 24 h before the sampling. Nine sites from these clinically healthy subjects were brushed using sterile dry cotton swabs pre-moisten with a sterile solution of (0.15 M NaCl with 0.1%Tween 20)18. Sites included tongue dorsum, lateral sides of tongue, buccal epithelium, hard palate, soft palate, supragingival plaque of tooth surfaces, subgingival plaque, maxillary anterior vestibule, and tonsils. A negative control, sterile pre-moisten swab without brushing of mouth, was used in parallel to check the presence of contaminants in cotton swabs. After collection, the cotton tips of the swabs were broken off directly into a 1.5 ml micro- centrifuge tube containing 1ml of lysis buffer (Mo Bio) with vortex at low speed to disperse the trapped bacteria.
Genomic DNA was extracted from the clinical specimens and the negative control using Mo Bio power soil DNA isolation kit (MO BIO Laboratories, Carlsbad, CA, USA cat. No 12888-50) with some modifications. In brief, the steps were: all cotton tips were removed, and the solution transferred to bead tubes. After that, we added 60 ?L of solution C1. Tubes were incubated at 65 °C for 10 min and then shaken horizontally at maximum speed for 2 min using the MO Bio vortex adapter. Then we followed all instructions mentioned by the manufacturer except one something, that we eluted DNA with only 30 ?L instead of 100 ?L of PCR DNAase free water. At this stage we obtained DNA yield, and now we can use it for further applications immediately or we can place it in 100 mL of TE buffer and store it at -80C until processing.
PCR amplification and sequencing of 16S rRNA gene
After DNA extraction, PCR reactions were carried out to amplify hyper variable regions V3V4 of 16S rRNA gene by using the 16S metagenomic sequencing (Illumina MiSeq platform).We amplified the DNA under standardized conditions using the primer set, Forward Primer 5’TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG Reverse Primer 5′ GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC. PCR reactions were occurred in 25?l reactions with 0.8?L for each forward and reverse primer (10?M, Metabion, Germany), 3 ?L of template DNA for specimens or 3 ?L of an elution solution for negative control, and 12.5 ?L of 1×of Hot Master Mix (Promega PCR mastermix). According to the protocol, Samples were preheated at 94°C for 3 min, followed by 35 cycles of denaturation at 95 °C for 30 s, then annealing at 60 °C for 30 s, and elongation at 72 °C for 30 s, with a final elongation step of 5 min at 72 °C. The PCR products were visualized by electrophoresis in a 1% agarose gel. DNA molecules were stained with ethidium bromide and visualized under UV light.
Purification of PCR products were done by using Agencourt XP Ampure Beads (Beckam Coulter, USA). The quality of the final products were assessed using a Bioanalyzer 2100 (Agilent Technologies, USA) and after quantification with a Qubit (Invitrogen, USA), the samples were pooled in equal proportions and sequenced paired-end in an Illumina MiSeq (Illumina, USA) with 600 cycles (300 cycles for each paired read and 12 cycles for the barcode sequence) at IGA Technology Services (Udine, Italy). To prevent focusing and phasing problems due to the sequencing of “low diversity” libraries such as 16S amplicons, 30% PhiX genome were spiked in the pooled library.
16S rRNA gene sequencing and data analysis:
Purified PCR-amplicons of 16S rRNA genes were sequenced using the metagenomics workflow of MiSeq platform v2.3 (Illumina). In brief, all sequences were demultiplexed based on index sequences. This step lead to generation of FASTQ files with Quality Score Trim sample-sheet set to make trimming. Then classification step was conducted using Classify Reads, an Illumina algorithm that make species-level classification for paired-end reads. This process involves matching short subsequences of that reads to a set of 16S reference sequences (template).The classification step assigned taxonomic classification to each read based on mapping of reads to a reference database for 6 taxonomic levels (Domain, Phylum, Class, Order, Family, Genus).The taxonomy database for the metagenomic workflow was an Illumina-curated version of the Greengenes database (greengenes.secondgenome.com/downloads/ database/13_5). All previous preprocessing steps were essentially repeated using MOTHUR (v.1.35.0) software package19. Forward and reverse reads were come together for each sample to build a contig from their paired ends using (make.contigs command).Then we need to remove poor quality sequences from our dataset using (screen.seqs command). This filtering step reject reads <440 nt and >490nt, excluding homopolymer runs >8nt, and 0 ambiguous bases. After quality filtering , (Unique.seqs command) search through all sequences and de-replicate the files. This command found identical sequences and select only one representative sequence for each. All these steps paved the way for creation of operational taxonomic units (OTUs) which is the proxy for traditional species designation. In order to create OTUs, there are three stages ; alignment, distance matrix creation, clustering by similarity(97% of similarity against SILVA database). Alignment of the unique sequences was performed by mapping them against reference database, SILVA 119 (arb silva.de/documentation/release-119/) 20. ). Chimeras sequences were removed using the MOTHUR implementation of UCHIME algorithm 21. Eventually , we obtained the (OTUs) by using (classify.seqs command ) with naïve Bayesian method 22.
Analysis of Diversity: Diversity measures and statistical significance were executed using random repeatedly resampling of contigs data with an average of 38,097 sequences per sample. In this study, we used two ways to analyze sequences representing just two points of view; the first way deal with each sample as a separate unique community, and the second one combined samples from the subjects to form groups representing age and sex dealing with each group as a community. Species diversity can be measured by two ways; species richness which refers to the total number of different species present in a sample, species evenness which refers to the spread of species in a sample. For instance, we may have two samples equal in richness, but one of them may be much more even than the other.
Phylotypes based analysis: Operational taxonomic units (OTUs) creation depends on grouping the sequences together by their sequence similarities, but phylotype creation is based on grouping sequences together by their similarity to a reference database. After the alignment step, we built a distance matrix to compare the sequences dissimilarity using (dist.seqs command). Then, this distance matrix is used for calculation of OTUs using (cluster command) with average-neighbor algorithm which is middle method between nearest neighbor”each of the sequences within an OUT are at most x% distant from the most similar sequence in the OTU”, furthest neighbor” all of the sequences within an OUT are at most x% distant from all of the other sequences within the OTU”. Cluster command created OTUs at every possible percentage of dissimilarity, so we needed to create a file to know which one of the OTUs occurred in each sample, and how often they occurred (diversity estimation) using ( make.shared command) with similarity cutoff of 97% that we do not create a shared file for every percentage of dissimilarity.
Alpha diversity: Describes the inherent characteristics of a given sample. Calculation of different species diversity indices( e.g.; rarefaction curves, Chao1 richness estimators, Inverse Simpson’s and Shannon diversity index ) were carried out using (sub.sample command) .Subsampling cutoff of 38,097 sequence/sample, while bacterial community evenness had been calculated with Shannon index using Microsoft Excel sheet.
Beta diversity: Describes differences between the samples. Interpersonal and intrapersonal diversity were detected using OTUs based analysis by using two measures, Jaccard index for shared community membership only that did not take consider into proportional abundance of each OUT, and Theta index for shared community structure, were calculated with the parameter thetayc and jclass 19, 23. Beta diversity was measured using UniFrac based principle coordinates analysis (PCoA), which is a phylogenetic method and a useful visualization technique for OTUs data. Relationships between samples were determined using (PCoA) based on both weighted and unweighted UniFrac distances 24, 25.
We used statistical analysis for assessment of significant differences between all samples and between all groups. Calculation of P values is based on Mann-Whitney U test (MW ); this test was used for determination of the significant difference between two samples or groups, while Kruskal-Wallis (KW) sum rank test was used for determining species with significant difference between more than two samples using (lefse command)26. Spearman correlation coefficient (SpCC) was used for assessment of the correlation between OTUs with (otu.association command) by cutoff value 0.001. R programming language version 3.3.2 2016 27was used for assessment of all statistics. Bacterial composition at genus level was used to calculate herarical dendrogram based on the Bray-Curtis distances 28 using Vegan package (v.1.15–1) in the R statistical framework 29. All charts and plots were generated using ggplot2 package R 30.
Data access Sequence data taken from this study had been submitted to NCBI BioProject (http://www.ncbi.nlm.nih.gov/ bioproject) under the accession number; PRJNA384402. Patient and sample metadata had been submitted to db biosample (http://www.ncbi.nlm.nih.gov/biosample) under the accession numbers; SAMN06840033 up to SAMN06840049