Periodontitis is one of the most widespread infectious diseases among adults, with an estimated 46% of dentate adults aged 30 years and older in the US suffering from some forms of the disease (Eke et al., 2015). Based on the data from the National Health and Nutrition Examination Surveys I, II, and III, periodontitis disproportionately affects different racial and ethnic groups. African Americans (AAs) and Hispanic Americans (HAs) experience higher incidences of periodontitis than Caucasian Americans (CAs) even after adjustment for all co-factors (Borrell et al., 2002; Thornton-Evans et al., 2013; Eke et al., 2020). Longitudinal studies on the natural history of periodontitis suggest that modifiable and non-modifiable risk factors potentially influence the onset, clinical presentation, and progression of the disease (Loe et al., 1978; Okamoto et al., 1988; Michalowicz et al., 1991; Grossi et al., 1994; Kornman et al., 1997; Borrell et al., 2004; Borrell and Talih, 2012; Genco and Borgnakke, 2013; Thornton-Evans et al., 2013). Smoking, age, and gender, as well as genetic and socioeconomic factors are well-known risk factors for periodontitis (Stashenko et al., 2011; Borrell and Crawford, 2012; Ioannidou, 2017; Naorungroj et al., 2017; Sutton et al., 2017). Additionally, the oral microbiome is a necessary, although insufficient, etiologic factor of periodontitis (Socransky and Haffajee, 2000; Jenkinson and Lamont, 2005). Porphyromonas gingivalis, one of the putative periodontal pathogens, has been well studied. P. gingivalis in the oral cavity, even in low abundance, can disrupt host-microbial homeostasis and elevate the virulence potential of oral microbial communities (Hajishengallis et al., 2011; Hajishengallis et al., 2012).

We previously reported an antagonistic relationship between P. gingivalis and Streptococcus cristatus (Wang et al., 2009). We found that S. cristatus can repress FimA expression in several P. gingivalis strains, including strains of fimA types I, II, and III (Xie et al., 2000; Xie et al., 2007; Wang et al., 2009). S. cristatus surface protein, arginine deiminase (ArcA), was identified as the signaling molecule to which P. gingivalis responds by repressing the expression of the fimA gene and consequently reducing production of fimbrial adhesin, FimA (Xie et al., 2007; Wu and Xie, 2010). We also discovered the expression of the arcA gene to be significantly higher in S. cristatus than in Streptococcus gordonii, suggesting that these two streptococcal species play distinct roles in the highly orchestrated dental plaque (Lin et al., 2008). Recently, we reported that the ratio of S. cristatus to P. gingivalis was significantly higher in CAs than in HAs and AAs with periodontitis (Wang et al., 2021). Based on these observations, we speculate that the interaction of specific bacteria may play an important role in periodontal health disparities. In the present work, we focused on levels of keystone pathogens and accessory bacteria and investigated a potential link between microbial compositions, particularly the prevalence of P. gingivalis and S. cristatus, and risks of periodontitis in AAs, CAs, and HAs with intact periodontium. Our results suggest that differences in the compositions of oral microbial communities may influence susceptibility to periodontitis.

To assess the potential risk factors for periodontitis development, we performed a comprehensive study to analyze the demographics, oral microbiological profiles, and clinical features of study subjects from different racial/ethnic backgrounds, all with an intact periodontium. Several factors that may contribute to periodontal health disparities were identified. Current models for the pathogenesis of periodontitis recognize P. gingivalis as a keystone pathogen that elevates the virulence of the entire microbial community (Hajishengallis et al., 2011; Darveau et al., 2012; Hajishengallis et al., 2012; Hajishengallis and Lamont, 2014). One key finding in this study is the differential detection rates of P. gingivalis among the AA, CA, and HA groups. P. gingivalis was detected in 23.5% of all participants; this percentage is similar to a previously reported rate of 25% (Griffen et al., 1998). However, the detection rate of P. gingivalis to be the highest in the HA group, when participants in specific groups were examined. Moreover, the detection rate of P. gingivalis in younger participants (aged 21–35 years) was higher in the AA and HA groups than in the CA group. More participants in the AA and HA groups exhibited higher levels of P. gingivalis in their dental plaques compared to the CA group. Our results suggest that the relatively high levels of P. gingivalis detected at younger ages may contribute to a higher risk for periodontitis in AA and HA populations. Our observations are in accordance with a previous analysis of 16S rRNA genes using deep sequencing that showed a higher abundance of Bacteroidetes and higher prevalence of P. gingivalis in AA than in CA subjects (Yang et al., 2019).

In the light of detection of P. gingivalis in dental plaques from both periodontitis patients and periodontally healthy subjects, it was suggested that different subsets of P. gingivalis strains exist in diseased and healthy periodontal sites (Griffen et al., 1998). An increasing body of evidence reveals that P. gingivalis strains with types II and IV FimA are often associated with periodontitis (Amano et al., 2000; van der Ploeg et al., 2004; Miura et al., 2005; Wang et al., 2020), while strains with type I fimA are associated with healthy periodontal tissues (Amano et al., 2000). In this study, the distribution of P. gingivalis fimA genotypes in the different racial/ethnic groups was compared. We found that the majority of P. gingivalis-positive individuals in the AA group carry the strain with type II fimA, which was significantly higher than that in the CA and HA groups. In addition, the type II and IV strains appeared to be associated with higher levels of BOP. These findings could explain a higher prevalence of periodontitis in the AA population (Amano et al., 1999; van der Ploeg et al., 2004; Miura et al., 2005; Wang et al., 2020). It is possible that P. gingivalis strains with fimA types II and IV in periodontally healthy subjects more likely break up a balanced host immune and induce uncontrolled inflammation in the periodontium (Yang et al., 2004).

We previously identified a negative correlation between P. gingivalis and S. cristatus in dental plaque samples from periodontitis patients (Wang et al., 2009). In the present study, higher ratios of S. cristatus to P. gingivalis in P. gingivalis-positive subjects in the CA group than those in the AA and HA groups with intact periodontium were also observed. It is reasonable to speculate that higher ratios of S. cristatus to P. gingivalis can keep P. gingivalis in check and inhibit the ability of the latter to elevate the virulence of the oral microbiota. In other words, the lower ratios found in the AA and the HA groups may render these individuals more susceptible to periodontitis.

It is well known that diabetes is a major risk factor for periodontitis (Preshaw et al., 2012). According to the CDC (https://www.cdc.gov), the incidence of new diabetes cases is higher among non-Hispanic AAs and people of Hispanic origin than among non-Hispanic Asians and non-Hispanic CAs. We did not observe an uneven distribution of diabetes cases among the different racial/ethnic groups in our study cohort of 340 participants. Smoking is another significant risk factor for periodontal disease development and progression (Zhang et al., 2019). According to the American Lung Association (https://www.lung.org/quit-smoking/smoking-facts/impact-of-tobacco-use/tobacco-use-racial-and-ethnic), adult smoking rates are 16.8% in AAs, 16.6% in CAs, and 10.1% in HAs, which are higher than what we observed in our cohort (9.2% in the AAs, 10.8% in the CAs, and 6.7% in the HAs). Although the smoking rate in the HA group appeared to be lower than that in the AA group and the CA group, the differences in percentages were not statistically significant. One plausible reason is that our sample size was not large enough to detect significant differences in diabetes incidence and smoking rate. In addition, participants in this study were recruited at dental clinic in an academic setting; hence, the cohort may not be representative of the general population. Future studies on larger and more representative patient cohorts are warranted. Moreover, our observation of zip code incomes agrees with a previously reported likelihood of an association between neighborhood socioeconomic circumstances and periodontitis (Borrell et al., 2006). However, further studies are needed to examine how neighborhood socioeconomic conditions impact the development of periodontitis and periodontal health disparities.

In this study, we collected dental plaque samples only once from the participants diagnosed as periodontal health or biofilm-induced gingivitis on an intact periodontium (without periodontitis) during their initial periodontal examination. The participants were not followed after the sample collection. Therefore, this study only provides preliminary information about the possible factors associated with periodontal health and further studies are needed in the future to discern their precise role in periodontitis development. Additionally, the present study was limited on P. gingivalis and several key oral bacteria. Whole bacterial profiles of the samples are under investigated using metagenome shotgun sequencing in our lab, which should provide clearer pictures on core microbiota associated with periodontal health disparities.

In conclusion, this work highlights the importance of differential oral microbiota in periodontal health disparities. Several risk factors may be linked to periodontal health disparities, including the presence of P. gingivalis at higher levels and at younger ages, the prevalence of specific subsets of P. gingivalis strains, and lower ratios of S. cristatus/P. gingivalis.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by the Committee for the Protection of Human Subjects of the University of Texas Health Science Center at Houston. The patients/participants provided their written informed consent to participate in this study.

HX and B-YW conceived the study and supervised the project. AC and DW performed and verified the statistical analyses. B-YW, SS, and H-WM enrolled study participants. EG, JH, and B-YW collected medical and dental history data. M-HH and JH helped sample process. All authors contributed to the article and approved the submitted version.