The occurrence of Treponema spp. in gingival plaque from dogs with varying degree of periodontal disease

Periodontal disease is common in dogs and is initiated by gingival plaque composed of several hundred bacterial species. Some of these species have specifically been pointed out as potential periodontal pathogens, such as Treponema spp. Treponema spp. are difficult to culture and therefore the majority have been detected by culture-independent methods, such as PCR (Polymerase Chain Reaction). This leaves many Treponema spp. uncharacterized and unnamed. In this study, treponemes were investigated in gingival plaque from dogs with varying degree of periodontal disease with the aim to describe their occurrence and diversity in dogs. The methods used were culture, phase-contrast microscopy, PCR targeting the 16SrRNA-tRNAIle intergenic spacer region (ISR2), sequencing of the ISR2 and phylogenetic analysis. Treponema spp. were detected in samples from 10 out of 11 dogs and isolates were obtained from six dogs. Both healthy and periodontitis affected dogs were Treponema positive. Phylogenetic analysis, based on ISR2 sequences, revealed a large diversity of treponemes in the study population that were found to be distributed mainly in two groups, corresponding to the human oral treponeme phylogroups II (Treponema denticola) and IV (Treponema maltophilum) genetic groups. They were generally more distantly related to other treponemes in these groups. Treponemes from dogs with periodontitis and dogs with mild gingivitis without periodontitis did not differ in any obvious way. The results indicate that several phylotypes of oral treponemes are common in dogs regardless of periodontal status.


Introduction
Periodontal disease, gingivitis and periodontitis, is a common affliction in dogs as well as in humans [1]. Accumulation of bacterial plaque on the tooth surface is believed to play a part in the initiation of periodontal disease, and as it develops, a shift occurs in the bacterial flora from being predominantly aerobic to predominately anaerobic [2]. Recent studies have revealed a large diversity of bacterial species in subgingival plaque from dogs, and also extensive differences between dogs and humans in the composition of species [3][4][5][6]. It  that of 353 bacterial species in dog subgingival plaque, only 16.4% were shared with humans [4]. In human patients, Treponema denticola, Tannerella forsythia and Porphyromonas gingivalis as a group, referred to as the "Red complex", show a strong association to periodontitis [7][8][9], and some studies have indicated their involvement also in dogs [10][11][12][13][14][15]. Because of their fastidious nature [16], treponemes are difficult to cultivate and have often been excluded from culture-dependent studies [10,17,18]. Thus, only 10 oral treponemal species have yet been isolated and characterized: T. denticola [19], T. socranskii [20], T.pectinovorum [21], T.vincentii [22], T. lecithinolyticum [23], T. amylovorum [24], T. maltophilum [25], T. medium [26], T. parvum [27] and T. putidum [28]. So far, few studies have been published where treponemes from dogs have been cultured and isolated, and there is a limited knowledge on the occurrence and importance of these bacteria in periodontal disease in this species. Species-specific PCRs or methods like fluorescence in situ hybridization (FISH) have been used to identify treponemes from dogs at species-level, however, often, the methods used were developed to identify oral treponemes from humans [13,29]. The aim of this study was to describe the occurrence and diversity of oral Treponema spp. in dogs with periodontitis by: 1. Culturing and isolation of Treponema spp. from plaque samples from dogs with varying degrees of periodontal disease.
2. Analysing pure spirochetal cultures and uncultured gingival plaque samples with 16S rRNA-tRNA Ile intergenic spacer region PCR, using primers specific for the genus Treponema.
If CAL was greater than 0 periodontitis was present. Two dogs, with no signs of periodontitis and only mild signs of gingivitis (defined by pink gingiva color, less than one mm marginal hyperemia on maximum three teeth, no swelling of the gingival margin, knife-edge adaptation of gingiva to teeth, if probing could be performed only slight bleeding on probing on maximum three teeth, and no gingiva retraction) were used as controls.

Sampling and culture
Two supragingival samples were taken from each dog and two samples were taken from periodontal pockets if present. All samples were taken on the buccal surface. One of the supragingival samples was taken with a cotton swab from the tooth surface and gingiva of several teeth, and put in a flask with liquid growth medium (see below) for isolation of Treponema bacteria. The other supragingival sample was taken for DNA preparation with a plastic loop and put in a tube containing 500 μl PBS (Phosphate Buffered Saline). The periodontal pocket samples were collected with a plastic loop, one was put in a 30 ml flask with liquid growth medium and the other in a tube with 500 μl PBS. Each PBS-sample was examined directly with phase-contrast microscopy and the presence of spirochetes was noted. The PBS-samples were then centrifuged in 4000 rcf for 5 minutes and the pellets were stored in -20 C until DNA extraction. These samples will further on, in this text, be referred to as "original samples". Culturing was performed according to Svartström et al 2013 [31]. Shortly, flasks were prepared with 10 ml FAB + A (fastidous anaerobe broth from Lab 71 (LAB 71, Lab M, Heywood, Lancashire, UK) prepared with 2 grams of d-glucose per liter) with 720 μg thiamine pyrophosphate per liter, 10 μl each of isobutyric acid, isovaleric acid, 2-methylbutyric acid, valeric acid, solubilized in 0.1 M potassium hydroxide, 25% fetal calf serum, S 0115, (Biochrom AG, Germany), 10 μg/ml rifampicin Sigma (Sigma-Aldrich Sweden AB), and 10 μg/ml enrofloxacin (Fluka, Sigma-Aldrich Sweden AB)) and were incubated anaerobically at 37˚C and 90 rpm. Cultures were monitored macroscopically and by phase contrast microscope about once a week for detection of spirochetal growth. From cultures with spirochete growth, 100 μl was added on a filter (pore size 0.22 μm) placed directly on a fastidious anaerobe agar plate; (FAA plate, National Veterinary Institute, Uppsala, Sweden). The filter was removed after 2-3 days. Once bacterial growth was detected a piece of agar was transferred to FAB + A with supplements but without antibiotics. Cultures with spirochetes in abundance (about 10 spirochetes/field of vision), and free from other bacteria were frozen with glycerol in -80˚C and a bacterial pellet was stored in -20˚C for later DNA extraction.

Molecular analysis
For DNA extraction, pellets from isolates were washed once in PBS whereas pellets from original samples were not. Pellets were mixed with 50 μl Super-Q water, boiled for 10 min and centrifuged in 16200 rcf for 2 min at 4˚C, the supernatants were stored in -20˚C. PCR amplification of the 16SrRNA-tRNA Ile intergenic spacer region (ISR2) was performed using primers selective for the genus Treponema and the protocol described by Stamm et al. [32]. Agarose gel electrophoresis was performed to verify successful amplification, and if no or very weak bands were obtained, samples were run again with 35 PCR cycles. The PCR-products were sent for sequencing at Macrogen Inc, Korea using PCR primers as sequencing primers. Sequencing results were confirmed by sequencing of multiple PCR-products from the same sample. PCRproducts from all original samples and from isolates with sequences showing signs of mixed signals, were first cloned using the TOPO TA Cloning Kit for Sequencing, pCR 4-TOPO Vector (Invitrogen by Life Technologies). Transformation into competent Escherichia coli (One Shot TOP10 chemically competent cells) was made using the Rapid One Shot chemical transformation protocol; 50 μl of the cells were then spread on prewarmed LB (Luria-Bertani)-agar plates containing 50 μg/ml ampicillin, and incubated for 24 hours at 37˚C. Four isolated colonies from each plate were cultured in tubes with 5 ml LB broth and 5 μl ampicillin (50 μg/ml final concentration) at 37˚C and 160 rpm for 24 hours. The cultures were pelleted by centrifugation at 2900 rcf for 15 minutes, and plasmids were purified using the GeneJET Plasmid Miniprep Kit (Fermentas Life sciences) and eluted with BPC grade water. The plasmids were sent for sequencing at Macrogen Inc, Korea, using general primer M13F. Sequences were trimmed in CLC Main Workbench 7.8 (CLC bio). Homology searches were performed using the Basic Local Alignment Search Tool (BLAST) algorithm at National Center for Biotechnology Information (NCBI) [33]. Progressive alignments of sequences obtained from the study and sequences with high identity from the homology searches were made with default gap cost parameters and the "very accurate (slow)" algorithm and a phylogenetic tree was constructed. The "Model testing" tool in CLC Main Workbench 7.8 (CLC bio) showed that the best fit model was General Time Reversible (GTR), that was used to construct a maximum likelihood phylogenetic tree. If identical sequences were derived from the same sample only one was included in the tree. For isolates considered "pure" based on ISR-sequences, the 16S rRNA gene was amplified and sequenced using the primer pair pA and pH [34]. A phylogenetic tree was constructed with parts of the 16S rRNA including representatives of the human oral treponeme phylogroups. Accessions numbers of all GenBank derived sequences can be found in S2 Table.

Study population
Seven male and four female dogs of different breeds were sampled. Their age varied between 3-12 years (mean 7.9 years, SD 2.9) and weight 7-46 kg (mean 20.6 kg, SD 12.6). The TMPS-G scores and CAL at the sampling sites are shown in Table 1. Supragingival samples were collected from all 11 dogs and additional samples from periodontal pockets were collected from seven of the dogs. Tooth pocket depth in dogs with periodontitis ranged from 4-10 mm.

Sampling and culture
Spirochetes were observed by phase contrast microscopy in original or culture samples from 10 of the dogs, and were more commonly seen in samples from tooth pockets (6/7, 86%) than from supragingival samples (7/11, 64%). Results are presented in Table 1. Spirochetes were seen in the supragingival samples of both control dogs. Spirochetes were easily seen in samples from periodontitis affected dogs, two to four in the first field of vision, while several fields of vision had to be examined before spotting even one spirochete in other samples. A total of nine isolates of spirochetes (pure from non-spirochaetal bacteria as determined by phase-contrast microscopy) were obtained from six different dogs. Isolates were obtained from both supragingival samples and from periodontal pocket samples as well as from dogs with periodontitis and no periodontitis, including one of the control dogs (Table 1). Incubation time to obtain a pure culture of spirochetes with a density of at least 10 spirochetes/field of vision was between one and a half month and five months. From some of the cultures that initially contained large amounts of spirochetes, it was not possible to obtain a pure culture whereas some cultures, with no initial visible spirochetal growth, were positive after several months.

Molecular analysis
All isolates and 12 of 18 original samples were positive by PCR. Of the supragingival samples, five of 11 were positive by PCR. Six of seven samples from periodontal pockets were positive by PCR. The negative PCR sample was also negative in phase contrast microscopy. ISR2 sequences. In the original samples, different Treponema spp. may be present, which could generate similar size PCR fragments. For this reason, original sample PCR products were cloned into the pCR 4-TOPO Vector and four clones from each sample were chosen for sequencing to enhance the possibility of detecting different treponemes. The same approach was used on PCR products from isolates for which a first round of sequencing showed double peaks, indicating a mixed population. Twenty ISR2 sequences were obtained from isolates, of which 15 were deposited in GenBank (accession numbers MH482781-95). From four isolates complete ISR2 sequences (GenBank accession numbers MH482746-49) were assembled which were considered "pure" after up to five amplicons from the same sample yielded identical sequences. From original samples 38 sequences were retrieved and 31 were deposited in Gen-Bank (accession numbers MH482750-80). Names of the sequences and accession numbers can be found in S1 Table. The most closely related sequences (highest identity score on NCBIblast) for isolates and original samples are listed in Tables 2 and 3, respectively. Only two of the closely related sequences originated from isolates: Treponema denticola ATCC 35405 and Treponema sp. isoM1187, the rest were from uncultured clones. Four of the sequences originated from bovine ulcerative mammary dermatitis, three from porcine gingiva, and remaining sequences came from porcine ear necrosis and porcine shoulder ulcer. Analysis of genetic diversity. Fig 1 shows the phylogenetic tree constructed by the Treponema spp. derived from dogs in this study, closest related sequences by BLAST, and representatives from the different phylogroups in the systematic classification of human oral treponemes proposed by Dewhirst et al. [35]. Accession numbers and origin for all the sequences are found in S1 and S2 Tables. The sequences from dogs appeared in phylogroups I, II and IV, but also clustered with T. pedis. The majority of ISR2 sequences appeared in two main groups, of which one consisted of the phylogroup II (T. denticola), Treponema sp. clone C1BF-3-related sequences and T. pedis-related sequences. The other group clustered with phylogroup IV in two main clusters, one that comprised T. maltophilum and the other one comprised clone E1163b, originating from porcine ear necrosis. The remaining sequences clustered with Treponema medium (phylogroup I) and T. parvum. All obtained isolates grouped with the phylogroup IV associated group except one that was in phylogroup II clustering with T. denticola. In all samples but from one dog, more than one treponemal sequence was derived. The sequences from the two control dogs were found in both main groups.
Sequencing of the 16S rRNA resulted in about 1115 basepair long sequences of good quality. Two of the isolates, THI4a and b, that had identical ISR2 sequences, both showed double peaks on the chromatogram and was dismissed for further phylogenetic analysis. The 16S rRNA gene phylogenetic tree (Fig 2) showed that the isolates clustered with T. parvum or Treponema sp. canine oral taxon 356 clone and T. maltophilum.

Data analysis
The distribution of supragingival and periodontal treponemes in the phylogenetic groups and clusters are shown in Table 4. As shown in Table 1, a larger proportion of periodontal pocket-derived treponemes clustered with phylogroup IV than with phylogroup II, whereas more supragingival-derived treponemes clustered with phylogroup IV.

Discussion
The role of Treponema spp. in the etiology of periodontal disease is unclear. Several studies have related their presence in gingival plaque to the development of periodontal disease in  [7,9,34] and also in dogs [13,14]. However little is known of what treponemal species are of importance in dogs, and isolation has only been performed in one previous study [29]. The PCR results and the results from the microscopy, we detected treponema in supragingival plaque and/or periodontal pockets from 10 out of 11 dogs, and isolates were obtained from six of the dogs. An indication of the relative amount of spirochetes in the samples was given by observation of spirochetes using phase-contrast microscopy directly after sampling or during culturing. Spirochetes were easier to detect by phase-contrast microscopy in the tooth-pocket samples. Notably, spirochetes were detected by phase-contrast microscopy directly after sampling in both dogs with no periodontitis or mild gingivitis, which suggests that spirochetes are common in the oral microflora. The PCR results and the results from the microscopy directly after sampling correlated well for the periodontal pocket samples, whereas for the supragingival samples, five were positive by microscopy of which two were negative by PCR. From two of these samples isolates were obtained after a longer time period of culture. Possible explanations are that the quantity of bacteria in some samples was below the detection limit for the PCR, or that inhibitory factors were present in the samples. The primer pair used may also not fit with all oral spirochetes.
Together, the results indicate that the periodontal pocket samples contained more Treponema than the supragingival samples, an expected result as pockets creates a more anaerobic environment and a hideaway place for bacteria, and it is also in agreement with several other studies [13,14]. Clearly a swipe with a plastic loop in a periodontal pocket generated more material than a swipe along the supragingival surface. Thus, for future studies, a method for supragingival sampling should be used which result in more material. from dogs in this study and sequences from GenBank representing the phylogroups for human oral treponemes based on 16S rRNA sequences proposed by Dewhirst et al (2010). Accession numbers and origin for all the sequences from this study are found in S1 Table. https://doi.org/10.1371/journal.pone.0201888.g001 Two of the three supragingival isolates came from dogs without periodontitis. One dog was used as a control (mild gingivitis), and in the sample from this dog three spirochetes were observed in the first field of vision directly after sampling, which was similar to what was seen in subgingival samples from periodontitis affected dogs. It was also interesting that the only dog negative for Treponema, by PCR and also by microscopy, had periodontitis. This shows that, by these methods, Treponema spp. can appear quite abundant in more or less periodontally healthy dogs and uncommon in periodontitis affected dogs, even if the opposite seems to be more common. Several previous studies have also found oral Treponema in both healthy and periodontitis affected dogs [3,6,13]. Nordhoff and co-workers [13], found Treponema spp. more often with more severe periodontal disease, whereas Davis [3] and Wallis [6] found a reduction in health associated bacterial species to be more correlated with periodontal disease than an increase of putative periodontal pathogens such as Treponema spp. Thus, the significance of Treponema spp. in the etiology of periodontitis is still unclear and needs further investigation.
Culturing Treponema spp. is very time consuming and the possibility of obtaining a pure isolate seem unrelated to the amount of spirochetes in the original culture. As a result, not many strains have been described, especially not from dogs, compared to more easily grown bacteria. The methods used in the present study were those described previously by Svartström et al., [31], but for some of the treponemes, the methods could be further optimized. One observation made was that cultures regarded as "hopeless", after being left for several weeks in an unopened anaerobic jar, demonstrated dense population of apparently healthy spirochetes. One isolate was obtained first after five months. This indicate that some strains of Treponema spp. require a substantially longer incubation time.
One aim of our study was to analyse the diversity of oral Treponema in dogs. We chose to do this by comparing sequences of the 16SrRNA-tRNA Ile intergenic spacer region because it is believed to differentiate better between closely related Treponema spp. than 16S rRNA analysis [32]. However a drawback with the ISR2 region is the relatively few deposited Treponema spp. sequences in Genbank, that makes comparison to for example human oral treponemes difficult. Another problem is that some Treponema spp. will not be detected by the ISR2 PCR. Thus, considering the limitations of our method, the Treponema spp. oral flora is likely to be  more diverse, as has been implicated in recent studies [36]. However, our result show that dogs harbour treponemes similar to phylogroups I, II, IV and also T. pedis, which is in accordance with the study by Nordhoff et al. (2008) where Treponema spp. from phylogenetic groups I, II, IV and VI was found in subgingival plaque samples from dogs using Treponema specific oligonucleotide probes [13]. Similarly, Valdez et al (2000) identified T. denticola, T. socranskii, T. vincentii, T. maltophilum, T. medium, T. pectinovorum by species-specific PCRs [29]. Our results also show that individual dogs are colonized by several different phylotypes of oral Treponema spp. In our study there was a tendency to more Treponema spp. from supragingival plaque clustering in group II and Treponema spp. from periodontal pockets in group IV. We also did not see that sequences clustered depending on the periodontal status of the dogs. However, the isolates were concentrated to one group in the phylogenetic tree, corresponding to the phylogroup IV, which indicates that either these Treponema spp. were easier to culture, or occurred in larger numbers in our sample material. Limitations of the present study to consider are the relatively small number of dogs, and the lack of completely healthy control dogs. However, the control dogs in this study had extremely mild gingivitis without any signs of attachment loss and that is generally considered normal within an adult dog population.

Conclusion
Dogs harbour several different Treponema spp. in their oral cavity and they can be common in both healthy and periodontitis affected dogs, indicating they are part of the normal bacterial flora. A comparison of the ISR2 gene distributed the majority of these treponemes in two main groups that could be further divided in several more or less distinct clusters, related to human oral treponeme phylogroups I, II, IV. All but one of the cultivable treponemes clustered with the group IV. The random distribution of the sequences from both periodontitis free and periodontitis affected dogs does not indicate that a specific phylotype is more involved in the etiology of periodontitis in dogs.
Supporting information S1