Comparative Analyses of Subgingival Microbiome in Chronic Periodontitis Patients with and Without IgA Nephropathy by High Throughput 16S rRNA Sequencing

In China, 80-90 percent of adults have periodontal problems, such as tartar, plaque, bleeding gums, and periodontitis, and about 15-20 percent of patients have chronic periodontitis at severe levels (CP) [1, 2]. The periodontitis incidence in China is higher than that in other countries, especially CP [2]. Periodontitis is a prevalent chronic inflammatory disease caused by enhanced inflammation induced by dysbiotic microbes forming on subgingival tooth sites, which may disturb the balance of the microbial composition in the biofilm and finally result in the progressive destruction of the periodontal ligament and alveolar bone with periodontal pocket formation and/or gingival recession [3]. The mechanism underlying destruction of periodontal tissues includes tissue damages caused by plaque bacterial products and bacterial induction of the host immune responses [3]. Periodontitis is considered as a leading cause of tooth loss in adults and historically viewed separately from those of the rest of the body [4, 5]. However, recent studies indicate that periodontal infection is a constant potential source of infection and is associated with numerous systemic diseases, including atherosclerosis, diabetes, cancer, rheumatoid arthritis, aspiration pneumonia, and adverse pregnancy outcomes [5-19]_ENREF_2. The mechanisms or pathways linking oral infections to secondary systemic effects are infections from oral cavity via transient bacteremia, injury caused by circulating oral microbial toxins, and inflammation caused by immunological injury induced by oral microorganisms. Therefore, promotion of oral health has been suggested as a way to promote systemic health [20].

IgA nephropathy (IgAN), also known as Berger’s disease, is a chronic glomerular disease that occurs when IgA deposits in the glomerular mesangium [21, 22]. IgAN usually progresses slowly over many years, and patients with IgAN usually present with proteinuria or microscopic hematuria, alone or in combination [23]. This results in local inflammation that may hamper kidneys’ function. About 25 % of adults with IgAN develop total kidney failure. A recent study indicates that the microbiome in IgAN is changed, such as a decrease in Clostridium, Enterococcus, Lactobacillus, Leuconostoc, and Bifidobacterium, and an increase in Streptococcus sp. and Firmicutes [24]. Till now, the relationship between periodontal infection and IgAN is not well studied. Our previous study has reported that the prevalence of CP and aggressive periodontitis in IgAN patients is higher than that in non-IgAN patients (P< 0.05), indicating that periodontitis is correlated with the onset and development of IgAN [25]. To further investigate the correlation between plaque bacterial microbiomes and IgAN incidence in CP patients, subgingival plaque samples from CP patients with or without IgAN were collected and comparative analyses of plaque bacterial microbiomes were further performed by using high throughput 16S rRNA sequencing.

Written informed consent was obtained from all the patients in this study. The study design, protocol, and informed consent were approved by Ethics Committee of China-Japan Friendship Hospital (2013-KY-3). The methods were carried out in accordance with relevant guidelines.

A total of 20 patient samples that included 9 CP patients without IgAN (Control) and 11 CP patients with IgAN (Disease) were selected and examined at China-Japan Friendship Hospital. Briefly, patients at the dental clinic and nephrology clinic without any treatment were selected for further sample collection. Subgingival plaque samples were collected from the first molars by means of a sterile excavating-spoon hand-instrument and then placed immediately into an Eppendorf tube containing 1 ml of sodium thiosulfate solution. Meanwhile, the clinical diagnoses of periodontal disease and renal disease were performed. Samples collected from patients with CP were considered as Control and samples collected from patients with CP and IgAN were considered as Disease.

Genomic DNA was extracted from subgingival samples using the PowerSoil® DNA isolation kit (MO BIO Laboratories, Inc. Cat. 12888-100) according to the manufacturer’s instructions. The amount of total DNA was determined using Nanodrop ND-2000 (ThermoScientific, Wilmington, DE, USA). Integrity and size of DNA were checked by 1% (w/v) agarose gel electrophoresis. The V3-4 hypervariable regions of bacterial 16S rRNA gene were amplified with the primers 357F (5’-CCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACHVGGGTWTC TAAT-3’). For each sample, 10-digit barcode sequence (Table 1) was added to the 5’ end of the forward and reverse primers (provided by Auwigene Company, Beijing) for differing each sample. PCR was carried out on a Mastercycler Gradient (Eppendorf, Germany) using 50 µl reaction volumes, containing 5 µl 10×Ex Taq Buffer (Mg²⁺ plus), 4 µl 12.5 mM dNTP Mix (each), 1.25 U Ex Taq DNA polymerase, 2 µl template DNA, 200 nM barcoded primers 357F and 806R each, and 36.75 µl ddH₂O. Cycling parameters were 94 °C for 2 min, followed by 30 cycles of 94 °C for 30 s, 57 °C for 30 s, and 72 °C for 30 s with a final extension at 72 °C for 10 min. Three PCR products per sample were pooled to mitigate reaction-level PCR biases. The PCR products were purified using a QIAquick Gel Extraction Kit (QIAGEN, Germany), quantified using Real-Time PCR, and sequenced at Auwigene Company, Beijing.

Raw sequencing data were processed by Beijing Auwigene Tech, Ltd. (Beijing, China) using the pipeline tools QIIME and MOTHUR. The overlapping paired-end reads were merged using FLASH (v 1.2.10). The sequences were removed from consideration if they were shorter than 200 bp, had a low-quality score (≤ 20), contained ambiguous bases or did not exactly match to primer sequences and barcode tags using Trimmomatic. Filtered reads were sorted into different samples according to their barcodes with MOTHUR. The retained high-quality sequences were further analyzed using MOTHUR and Usearch (version 8.0.1623). Specifically, MOTHUR was used for barcode and primer sequence removal, ‘trim.seqs (maxhomop = 10, minlength = 200)’ was used for quality filtering, and Usearch (version 8.0.1623) was used for de novo removal of chimeric reads. All the clean tags of all samples were clustered into OTUs using QIIME (v1.9.1) at 97% sequence similarity. These OTUs were used as a basis for calculating alpha-diversity and beta-diversity metrics using QIIME (v1.9.1). The sufficiency of the sampling effort was evaluated by drawing rarefaction curves, the bacterial community diversity within each individual sample was estimated using the Shannon-Wiener index, the species richness was estimated with the CHAO1 index, and the percentage of coverage was calculated by Good’s coverage estimator. The Ribosomal Database Project (RDP) Classifier tool was used to classify all sequences into different taxonomic groups.

The beta-diversity was performed with QIIME (v1.9.1) to assess the differences of microbial communities between Control and Disease based on their composition. A principal coordinate analysis (PCoA) of weighted UniFrac was performed to compare the overall structure of subgingival microbiome of all samples, based on the relative abundance of OTUs (at a 97% similarity level). The abundance of bacterial phyla and genus for each group was expressed as the percentage of total sequences and the bacterial community structures of Control and Disease were further compared at phylum and genus level using Mann–Whitney U test. P-values were corrected using a false discovery rate (FDR) correction to account for correction of multiple testing [26]. The statistical differences between Control and Disease were analyzed using ANOVA according to the methods provided by Mothur.

Subgingival plaque samples were collected from 9 CP patients without IgAN (Control) and 11 CP patients with IgAN (Disease). For Control and Disease, the average age were 37.3 and 36.1 years old, respectively; the probing depths (PD) were 4.8 mm and 5.2 mm, respectively; the clinical attachment levels (CAL) were 5.1 mm and 4.9 mm, respectively; and the percentage of surfaces of the plaque were 85.2 and 82.9, respectively (Table 2). No statistical differences were found in these parameters between Control and Disease.

In total, 408, 758 raw reads were obtained from all 20 subgingival plaque samples. After filtering, 350773 filtered clean tags (17538.65 tags per sample) and 8795 OTUs (439.75 OTUs per sample) were obtained from all the samples (Table 3). The Shannon-Wiener curve of all samples already reached a plateau at this sequencing depth (Fig. 1) and the coverage was higher than 95%, suggesting that the sequencing was deep enough. Though there were no statistically significant differences in the community richness estimator (Chao) and the diversity estimator (Shannon index) between Control and Disease (p=0.34 and 0.59, respectively), the diversity indices in Disease were a little higher than that in Control (Table 4).

Overall bacterial compositions for each group at phylum level were shown in Fig. 2 and Table 5. In general, a total of 23 phyla in all samples were obtained. The dominant phyla of all groups were Firmicutes, Proteobacteria, Bacteroidetes, and Fusobacteria, accounting for 81.79% of the total sequences. The most dominant phyla were Firmicutes in Control and Proteobacteria in Disease. In addition, the microbial composition, belonging to Proteobacteria and Bacteroidetes, varied greatly between Control and Disease (p< 0.05). Though no significant differences were found, the abundance of Actinobacteria was higher while the abundances of Bacteroidetes, Fusobacteria, Spirochaetae, and Synergistetes were lower in the subgingival microbiome of Disease than that in Control (Fig. 1). Overall microbial compositions for each group at class level were shown in Fig. 3 and Table 6. There were 37 classes in all samples. The dominant classes of both groups were Bacteroidia, Fusobacteriia, Negativicutes, Clostridia, Betaproteobacteria, Spirochaetes, Synergistia, Actinobacteria, Bacilli, and Flavobacteriia, representing 93% of total sequences. The most dominant class were Bacteroidia in Control and Betaproteobacteria in Disease. However, there were no significant differences between these two groups. At order level, a total of 70 orders were detected. Bacteroidales, Fusobacteriales, Neisseriales, Selenomonadales, and Lactobacillales represented 58.61 % of the total sequences. The abundance of Bacteroidales in Control (22.26%) was much higher than that in Disease (9.46%) while the abundance of Burkholderiales in Disease (8.23%) was much higher than that in Control (0.86%). At genus level, the abundances of Treponema_2 and Prevotella were lower while the abundances of Bergeyella, Lautropia, Actinomyces, Comamonas, Corynebacterium, Capnocytophaga, and Streptococcus were higher in Disease than that in Control (Fig. 4).

PCoA based on the compositions of OTUs in each sample was performed to compare the overall structure of subgingival microbiome between these two groups. There was an obvious separation of Control and Disease, PC1and PC2 accounted for 22.14% and 13.74% of the total variations, respectively (Fig. 5). Statistical analyses indicated that microbial composition of Disease was significantly different from that of Control (p< 0.001).

16S rRNA Miseq platform was used for assessing the relationship between the bacterial community and IgAN. The results showed that bacterial diversity in Disease was higher than that in Control. Some bacterial groups in Disease were also significantly different from that in Control. The structure of bacteria between the two groups varied significantly.

Periodontal disease is a prevalent chronic inflammatory disease of the oral cavity which is the major cause of tooth loss in adults [27]. Almost all forms of periodontal disease occur due to mixed microbial infections and many bacterial species are recognized as putative periodontal pathogens, such as Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, Bacteroides forsythus, Prevotella intermedia, Peptostreptococcus micros, Fusobacterium nucleatum, Filifactor alocis, Desulfobulbus sp. oral taxon 041 HOT 041, and Synergistetes [28-31]. Recent studies have indicated that periodontal disease and systemic health are closely linked and treating periodontal disease may help with the prevention of several other chronic inflammatory conditions, including IgAN [6, 16-19, 25]. IgAN is the most common primary glomerulonephritis wherein immune complexes consisting of IgA1 with galactose-deficient hinge region and anti-glycan antibodies deposit in glomeruli and induce renal injury [32]. Till now, knowledge about the impact of subgingival microbiome in CP patients on IgAN incidence is still limited. In this study, we used 16S rRNA sequencing to identify and compare bacteria present within Control and Disease and further investigated the correlation between subgingival microbiome and IgAN incidence in CP patients. Study subjects in Control and Disease were carefully selected, such that no statistically significant differences were present in average ages, probing depths, and clinical attachment levels. Moreover, statistically significant differences were found in subgingival microbial communities but not found in diversity.

The subgingival microbiome of Disease was compositionally distinct from that of Control. Disease had a higher relative abundance of phyla Proteobacteria and Actinobacteria, and a lower relative abundance of phyla Bacteroidetes, Fusobacteria, Spirochaetae, Synergistetes, and Saccharibacteria. Compared to Control, the ratio between Firmicutes/ Proteobacteria markedly decreased in Disease, which is consistent with the previous study that has demonstrated decreasing ratio between Firmicutes/Proteobacteria in the salivary microbiome of IgAN patients [33]. The changes of phyla Proteobacteria, Actinobacteria, Bacteroidetes, Fusobacteria, and Spirochaetae in Disease were mainly due to significant changes in class Betaproteobacteria and Gammaproteobacteria, Actinobacteria, Bacteroidia, Fusobacteriia, and Spirochaetes, respectively. Further comparison analysis at genus level indicated that compared to Control, genera Treponema and prevotella were less but Bergeyella, Lautropia, Actinomyces, Comamonas, Corynebacterium, Capnocytophaga, and Streptococcus were more abundant in Disease. The genus Treponema belongs to the phylum Spirochaetes. As reported, treponemes are involved in the etiology of chronic periodontitis and other forms of periodontal disease [34]. Treponema denticola resides in the human oral cavity and is highly correlated with the incidence and severity of human periodontal diseases [35, 36]. The genus Prevotella belongs to the phylum Bacteroidetes. A variety of Prevotella spp., including P. melaninogenica, P. intermedia, and P. loescheii, reside in the human oral cavity [37, 38]. P. intermedia might be a periodontal pathogen, whereas P. nigrescens is a marker of relative periodontal health [37]. Besides, it is found that, in the salivary microbiome, the relative abundances of Prevotella spp. (P. nigrescens, P. intermedia, P. pallens, and P. salivae) were higher in health control compared to IgAN patients and the only exception was P. aurantiaca [33]. Our data also indicated that higher abundance of Prevotella spp. was found in Control compared to Disease. The genus Bergeyella is a hard-to-cultivate taxon belonging to the phylum Bacteroidetes. The information regarding the role of Bergeyella in periodontal disease and IgAN is very limited [39]. The genus Lautropia belongs to the phylum Proteobacteria. Only one species, L. mirabilis, has been identified and was found to be associated with success in periodontal therapy [40]. The genus Actinomyces belongs to the phylum Actinobacteria (except A. meyeri, an obligate anaerobe). A previous study found that Actinomyces odontolyticus/meyeri and Actinomyces israelii were associated with chronic periodontitis (p=0.003) [41]. The genus Comamonas belongs to the phylum Proteobacteria. No information is available regarding the role of Comamonas in periodontal disease and IgAN. The genus Corynebacterium belongs to the phylum Actinobacteria. A previous study found that Corynebacterium diphtheria was in higher prevalence and level in the subgingival biofilm samples collected from patient with periodontitis, which is consistent with our data [42]. Piccolo, M. et al. found that the abundance of Corynebacterium sp. was lower in the salivary samples of IgAN patients compared to that in healthy control, showing an opposite conclusion [33]. The genus Capnocytophaga belongs to the phylum Bacteroidetes. Capnocytophaga spp. are often isolated from periodontal pockets, apical, and periodontal abscesses. It is found that Capnocytophaga spp. are more prevalent in gingivitis compared to healthy periodontium and periodontitis. Capnocytophaga spp. have the potential to cause periodontal disease [43-46]. The genus Streptococcus belongs to the phylum Firmicutes. In 2011, Huang et al. studied the relationship between oral microbiome and gingivitis status and found that Streptococcus was associated with gingivitis [47]. It is reported that Streptococcus has a correlation with IgAN [48]. A link between tonsillar infection caused by Streptococcus sp. and IgAN was hypothesized. Besides, streptococcal proteins were recognized by sera of IgAN patients [48, 49]. In the present study, Streptococcus was also upregulated in Disease, implying that Streptococcus spp. in CP patients has a correlation with IgAN.

In summary, our data indicated statistically significant changes in the subgingival microbiome in Disease compared to that in Control, implying a correlation between the changes in subgingival microbial structure and IgAN incidence in CP patients. However, till now, the knowledge regarding the roles of the changed subgingival microbial structure of CP patients in IgAN incidence is still limited and further investigation is needed.

This study was supported by Beijing Nova Program (No. xx2014B074).

The authors declare to have no competing interests.