Subgingival periodontal pathogens associated with chronic periodontitis in Yemenis

Periodontitis represents a range of clinical entities that are characterized by immunological destruction of the tooth supporting structures in response to chronic challenge by specific bacteria in subgingival biofilm[1]. The last three decades or so witnessed an explosion in our understanding of the microbiology of periodontitis. Earlier studies employing cultivation-based techniques recovered about 250 bacterial species (at > 1% relative abundance) from plaque samples[2]. With the extensive application of molecular techniques over the last decade, double this number of novel species has been identified[3‐5] bringing the richness of the subgingival microbiota to more than 700 bacterial species, about 50% of which are uncultivable.

While the majority of subgingival microbiota is considered commensal, several species have been implicated as periodontal pathogens. Porphyromonas gingivalis, Tannerella forsythia, and Treponema denticola (the so called red complex) have so far shown the strongest association with chronic periodontitis[6]. Other putative pathogens include Fusobacterium spp., Prevotella spp., Campylobacter rectus, Eubacterium nodatom, and Parvimonas micra (previously Peptostreptococcus micros)[6]. Uncultivable phylotypes such as Synergistetes, TM7 and Treponema taxa are also believed to play a pathogenic role in chronic periodontitis[3, 4, 7]. In fact, it is believed that periodontal destruction is triggered by a bacterial consortium rather than a single pathogen[1].

A number of molecular techniques have been employed for detection and quantification of periodontal pathogens in plaque samples including DNA-DNA hybridization, conventional and real time PCR, and 16S rRNA clone sequencing[8]. Of these, real-time PCR is the most sensitive allowing detection of as low as 1.6 cells per reaction[9, 10]. It also makes it possible to normalize target DNA counts to total bacterial counts in the sample (relative quantification), thus adjusting for variations in sampling and making comparisons between samples more reliable[11, 12]. Surprisingly, real-time PCR has not been as widely used in the study of microbiology of periodontitis as may be expected.

Subgingival microbial profile associated with periodontitis have been reported to significantly differ by geographical location independent of other factors known to modify subgingival microbial composition[8, 13]. It becomes prudent, therefore, that obtaining more information about the global distribution of periodontal pathogens and patterns of their association with disease can improve our understanding of the differences in the role they play in periodontitis in different populations. In the absence of data on this from the Middle East, the objective of the current study was to assess the association of seven putative periodontal pathogens with chronic periodontitis in a Yemeni population using quantitative PCR assays.

Studies from the Middle East are limited to those that assessed the effect of traditional oral hygiene such as miswak, or certain habits, like qat chewing, on levels of periodontal pathogens[17, 20, 21]; no studies have so far assessed which periodontal pathogens are particularly associated with periodontitis in an Arab population. The study compared counts of 7 putative pathogens between healthy and diseased sites in patients with moderate-severe chronic periodontitis, using real-time PCR. Despite the advantage of this techniques (sensitivity and possibility of relative quantification), a limited number of studies used it in the study of microbiology of periodontitis and even fewer used relative quantification for making comparisons between health and disease. Healthy sites within the same subjects rather than healthy individuals were used as controls to avoid the effects of inter-individual variations in factors other than microbial composition. Nevertheless, including healthy subjects as additional controls would have allowed for more comparisons and a broader view of differences in microbial composition between periodontal health and disease. So this could be considered as one limitation of the current study. Another limitation is that bleeding on probing was not recorded although it has been previously shown to be an important clinical variable. In addition, while the most important pathogens were tested, the panel could have included more species particularly the newer pathogens such as Filifactor alocis, oral synergistetes and TM7.

Absolute counts were reported in DNA copies rather than cell numbers because the target gene copy number per genome, particularly 16S rRNA gene, varies from one species to another and is not known for some of them. This implies that actual bacterial counts for some of the tested species/genera are less than the counts reported in the study. It, however, does not influence on validity of comparisons between sites or patients. Theoretically, as little as one gene copy per reaction can be detected by real-time PCR; however, in practice this is usually not possible. Some of the previous studies on periodontal pathogens were in fact not clear with respect to whether the reported lower detection limit was per reaction or per sample, but it can be concluded that at least 100 DNA copies per reaction could be detected[11, 22]. However, a detection limit of as low as 1.6 cells per reaction has been reported[9, 10] which is comparable to the sensitivity of q-PCR assays in this study (5 copies/reaction). The high sensitivity of q-PCR probably explains the higher detection rates of pathogens observed in studies that employed this technique compared to those found with culture techniques or DNA-DNA hybridization.

The median log counts of total bacteria as well as some of the test species/genera in this study are 1–3 log higher than those reported by most previous studies that used real-time PCR[9, 10, 12, 22]. In contrast, Lyons et al.[11] reported counts as high as 10¹⁴ and 10¹² for total bacteria and P. gingivalis, respectively, which sounds implausible. While these differences can be attributable to variations in sampling and DNA extraction efficiency, inaccuracies in preparation of standard curves probably account for the great part of them. Curves prepared using serial dilutions of bacterial cells or genomic DNA extract, as done in most previous studies, may not be as accurate as those generated using plasmids. On the other hand, quantification bias because of plasmid DNA conformation has been reported recently, especially with circular plasmids[23]. However, since values from this study are also close to those previously obtained by checkerboard DNA-DNA hybridization[24], they can be assumed to be accurate enough.

The relative abundances of T. denticola and T. forsythia in the current study are very comparable to those reported for a Japanese population using a similar quantification technique[12]. In the latter study, however, P. gingivalis and Prevotella intermedia were presented at exceedingly higher proportions and showed significant association with the disease which in contrast with the current findings. Using checker board DNA-DNA hybridization, much higher proportions of the red complex members have been reported in chronic periodontitis patients from USA, Brazil, Chili and Sweden[13] compared to the relative counts described here. However, this can be simply justified by the fact that proportions in the checker board technique are calculated by normalizing absolute counts of each species to total counts of the 40 probe species used rather than total bacteria as done in the current study.

Relative quantification data show that although periodontal pathogens were present at significantly higher proportions in periodontitis sites compared to healthy sites, they still constituted a minority of subgingival microbiota. This is, however, not surprising since assessment of earlier cultivation-based studies and more recent studies employing molecular techniques clearly reveals that periodontal pathogens, particularly members of the red complex, have almost always been detected at low abundance[3, 24, 25]. What has not been addressed adequately, on the other hand, is how these pathogens can cause periodontitis at such low abundance. One interesting, currently evolving view is that low abundance periodontal pathogens orchestrate periodontitis by inducing a dysbiotic “pathogenic” microbial community that in turn mediates bone destruction[26]. This is thought to result from ability of these pathogens to subvert some components of the host response rather than to act directly as proinflammatory bacteria as has been very recently demonstrated for P. gingivalis in vitro[27]. Accordingly, low abundant periodontal pathogens are claimed to function as keystone pathogens, a hypothesis that challenges the role of red complex members as conventional pathogens[28].

The current study did not show an association between P. gingivalis and periodontal destruction, which is very hard to defend against the existing overwhelming evidence. However, this can simply be a failure to detect existing association due to lack of adequate power, especially that there was a significant difference in absolute count at the 0.05 level. On the other hand, given the polymicrobial nature of periodontitis, and in view of the new keystone pathogen hypothesis, it is also plausible to propose that other members of the pathogenic team can in certain circumstances take over the role of P. gingivalis as a keystone pathogen. In fact, P. gingivalis has not always showed the strongest association with periodontitis[3, 29, 30]. In the current study, T. denticola and T. forsythia, both members of the red complex, did show significant association with periodontitis, which is consistent with the literature. However, P. micra (previously known as Peptostreptococcus micros) showed the strongest association with the disease being present at significantly higher absolute and relative counts in periodontitis sites in all study subjects. This species is a member of the orange microbial complex[6], and there is an expanding evidence on its role, along with other peptostreptococci, as a periodontal pathogen[3, 29, 31, 32].

Despite its presence in very low relative counts, P. micra showed the strongest association with periodontal destruction, which is suggestive of a potential role as keystone pathogen in place of P. gingivalis. However, this needs to be validated in a larger-scale study before it can be claimed to represent ethnic variations in pathogens’ association with periodontitis.