The Role of Epigenetics in Periodontal and Systemic Diseases and Smoking: A Systematic Review

: The aims of this systematic review were to identify and synthesize the evidence for an association in DNA methylation/histone modiﬁcations between periodontal diseases and systemic diseases/smoking. Electronic database searches using relevant search terms in PubMed, Embase, MEDLINE, CINAHL, Web of Science, Scopus, and SciELO, and manual searches, were independently conducted to identify articles meeting the inclusion criteria. Nine studies of 1482 participants were included. Periodontitis was compared to metabolic disorders, rheumatoid arthritis (RA), cancer, and smokers, as well as healthy controls. Substantial variation regarding the reporting of sample sizes and patient characteristics, statistical analyses, and methodology was found. IL6 and TNF were modiﬁed similarly in RA and periodontitis. While TIMP-3 and GSTP-1 were signiﬁcantly lower in periodontitis patients and controls than in cancer, SOCS-1, RMI2, CDH1, and COX2 were modiﬁed similarly in both cancer and periodontitis. While TLR4 in and CXCL8 were affected in periodontitis independent of smoking habit, smoking might change the transcription and methylation states of ECM organization-related genes, which exacerbated the periodontal condition. There was some evidence, albeit inconsistent, for an association between DNA methylation and periodontal diseases and systemic diseases or smokers compared to healthy patients or non-smokers.


Introduction
Periodontitis is a biofilm-induced condition that affects tooth-supporting tissues [1], which consists mainly of Gram-negative, anaerobic, and micro-aerophilic bacteria that can colonize the sub-gingival tissues [2,3]. The bacterial biofilm induces an inflammatory host response, which is influenced by environmental, genetic, and epigenetic factors [4][5][6]. Epigenetic factors refer to alterations in the gene expression that are not encoded in the DNA sequence [7,8], which include chemical alterations of the DNA and histones, resulting in the remodeling of the chromatin and activation or inactivation of a gene [9][10][11].
To date, the most recognized epigenetic mechanism is DNA methylation, which is regulated by DNA methyltransferases (DNMTs), resulting in the covalent addition of methyl

PECO Question: Population, Exposure, Comparison, and Outcomes
The focus question for the present systematic review was developed using the population, exposure, comparisons, and outcomes (PECO) criteria.
In human adults of any race (population), what are the effects of DNA methylation and histone modification profiles (outcomes) on periodontal diseases and systemic diseases or smoking (exposures) compared to no periodontal and no systemic diseases and nonsmokers (comparison)? Human clinical studies, including both interventional and observational studies: randomized controlled trials, cohort studies, case-control studies, and cross-sectional studies.

2.
Studies describing either an association in epigenetic marks (global, site-specific or genome-wide methylation of DNA) or histone modifications (methylation, phosphorylation, acetylation, ubiquitylation, and sumoylation).

Study Quality
Two reviewers (RB, IG) assessed the study quality independently by following the National Heart, Lung, and Blood Institute's Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [33]. The tool contains 14 criteria to assess the study design quality. First, each of the criteria was rated as either yes, no, or not reported. Then, each of the included studies received overall scores of good, fair, or poor as outlined in the assessment tool. Any disagreements were discussed, and a third reviewer (IK) moderated any disagreement if needed. Corresponding authors of the included studies were contacted via email for detailed information on study methodology when key criteria were determined to be not reported by the two reviewers.
outcomes (Supplementary Material SII). The data tables were organized by systemic diseases and smoking status. The data extraction tables included information on the general characteristics, methodology, and results of the included studies. The extracted data tables were compared and consolidated by the two independent reviewers (RB and MM). The final data extraction tables were reviewed by two reviewers (IG and IK) to ensure accurate data extraction and interpretation of the included studies.

Study Quality
Two reviewers (RB, IG) assessed the study quality independently by following the National Heart, Lung, and Blood Institute's Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [33]. The tool contains 14 criteria to assess the study design quality. First, each of the criteria was rated as either yes, no, or not reported. Then, each of the included studies received overall scores of good, fair, or poor as outlined in the assessment tool. Any disagreements were discussed, and a third reviewer (IK) moderated any disagreement if needed. Corresponding authors of the included studies were contacted via email for detailed information on study methodology when key criteria were determined to be not reported by the two reviewers.

Description of Included Studies General Characteristics of Included Studies
Nine studies examining a total of 1482 participants were included in the systematic review ( Table 1). One of the studies examined CP compared to metabolic disorders [34], two of the studies examined CP compared to RA [35,36], three of the studies examined CP compared to cancer [37][38][39], and three of the studies examined smokers [40][41][42]; all nine included studies also included a healthy patient comparison as a third study group. All Appl. Sci. 2021, 11, 5269 5 of 23 nine studies examined DNA modifications (i.e., CpG methylation) as an outcome, and no studies on histone modifications were identified in the search for inclusion in the review.

Setting and Study Population
Eight of the included studies recruited patients from university-based dental clinics; one of the studies did not report the study setting [40].
Three of the studies were conducted in Brazil [37,41,42], two of the studies in China [38,39], two of the studies in Japan [35,36], one of the studies in Serbia [34], and one of the studies did not report the study location [40].
One of the studies included only Caucasian patients of Serbian nationality [34], two of the studies included only Japanese patients [35,36], two of the studies included only patients from the Southeastern region of Brazil [41,42], and four of the studies did not report ethnic background and/or race of included patients [37][38][39][40].

Methods for Detecting DNA Methylation Changes
Bisulfate genomic sequencing [36], bisulfite modification followed by methylationspecific polymerase chain reaction (MSP) [34,38,39,42], direct sequencing of genomic DNA [35], the llumina NextSeq500 sequencing system [40], Infinium-based DNA Methylation Analysis, PCR, MS-HRM [37], and PCR and Methylation analysis with specific restriction enzymes [41] were applied as methods for detecting epigenetic changes. Table 2 summarizes the study outcomes reported by each of the studies, including the main site-specific methylation level findings, as well as additional observations/outcomes assessed, for example, mRNA expression, clinical assessment correlations, etc. Key findings of individual studies are reported below.          [34]. The authors reported that the CXCL12 promoter was predominantly unmethylated in all groups. However, both the CP and D/P groups had increased frequency and percent methylation of the CXCL12 promoter compared to healthy controls, albeit no statistically significant difference [34].

Characteristics of the Outcomes Measured
(b) Methylation of candidate genes in periodontitis and rheumatoid arthritis. Ishida et al. (2012) evaluated the DNA methylation status of 19 IL6 CpG motifs in peripheral blood from patients with CP compared to RA and controls [36]. The authors found that the methylation levels at the CpG motif at −74 bp and +19 bp were significantly lower in patients with CP and RA compared to controls; the methylation levels at −74 bp, but not +19 bp, were also associated with serum IL6 and IL6 production by mono-nuclear cells, and the other 17 CpG motifs exhibited comparable methylation levels between groups [36]. Kojima et al. (2016) evaluated the DNA methylation status of 12 CpG motifs in the TNF gene promoter region in peripheral blood from participants with CP compared to RA and controls [35]. The authors reported that both the CP and RA groups showed significantly higher methylation rates and frequencies than the control groups at −72 bp, and the RA group additionally exhibited significantly higher methylation rates and/or frequencies at 6 additional CpG motifs compared to controls. The levels of TNF produced were significantly different between individuals with and without methylation at −163 bp (significantly higher hypermethylation rate and frequency in the RA group) [35].
(c) Methylation of candidate genes in periodontitis and cancer. Loo et al. (2010) observed the hypermethylation status of E-cadherin and COX2 genes in blood samples and gingival tissues in CP patients compared to neoplastic tissues from breast cancer patients and blood samples from healthy controls [39]. The authors reported hypermethylation of candidate genes in both CP and cancer groups, with higher methylation frequencies in the cancer patients compared to CP patients and E-cadherin methylation compared to COX2 gene methylation [39]. Wang et al. (2014) analyzed the methylation status of TIMP1, GSTP1, 14-3-3σ genes in peripheral blood and gingival tissues from periodontitis patients compared to neoplastic tissues from breast cancer patients and blood samples from healthy controls [38]. The authors reported that the hypermethylation frequencies of TIM-3 and GSTP1 were similar in the periodontitis and control groups, but both were significantly lower than those for malignant disease patients. The methylation frequency of 14-3-3σ periodontitis was significantly higher than in the cancer and control groups [38]. Planello et al. (2016) profiled the DNA methylome of gingival tissue from CP patients, compared to oral squamous cell carcinoma samples (OSCC) and healthy controls [37]. The authors observed significant overlap between the altered DNA methylation patterns in CP and OSCC; specifically, the authors reported that for SOCS1 and RMI2, hypermethylated CpG sites in CP were also hypermethylated in OSCC, and hypomethylated CpG sites in CP were also hypomethylated in OSCC [37].
(d) Methylation of candidate genes in periodontitis and smokers/non-smokers. Oliveira et al. (2009) analyzed the DNA methylation status of the CXCL8 gene promotor region in epithelial oral mucosa cells from CP smokers, CP non-smokers, and non-periodontitis non-smokers [42]. The authors reported that independent of smoking status, the CP group had higher percentages of hypomethylation and higher expression of CXCL8 mRNA than non-periodontitis non-smoker controls. Additionally, there was no significant difference in results from gingival cells compared to blood leukocytes obtained from a sub-sample [42]. De Oliveira et al. (2011) analyzed the DNA methylation status in the TLR2 and TLR4 gene promotor regions in gingival tissue from CP smokers, CP non-smokers, and non-periodontal non-smokers [41]. The authors reported that the CpG sites analyzed were unmethylated in the majority of DNA samples of the three groups, and there was no statistically significant difference in DNA methylation and mRNA transcript levels between the groups [41]. Cho et al. (2017) analyzed the methylome and transcriptome of CP smokers, CP non-smokers, and non-periodontitis smokers and non-smokers [40]. The authors reported that smoking is associated with DNA methylation status and transcription of extracellularmatrix organization-related genes [40].

Genome Wide Methylation
Two studies reported the global and genome-wide methylation results [37,40]. DNA sequencing data showed that over 100 cell lines and 79 cell types exhibited DNA hypermethylation in CP [37].
In the presence of a chief disease-modifying factor, such as smoking, high throughput gene and methylation analysis in reference to the former's expression was able to decipher a pattern of expression in ECM degradation, adaptive immunity and skin pattern related genes among smokers and non-smokers who either have periodontitis or do not [40].

DNA Methylation of Candidate Genes
An inter-study comparison of candidate gene methylation could not be conducted as each of the candidate genes included were examined in only one of the included studies.

Global and Genome-Wide Methylation
An inter-study comparison of global and genome-wide methylation could not be conducted as the methodology for assessing outcomes was inconsistent across included studies.

Study Quality
Following the quality assessment guidelines [33], four studies [34][35][36]42] were assessed as overall good quality (Supplementary Material Table S2), four studies [39][40][41] were assessed as fair quality, and two studies [37] were assessed as poor quality. While the studies assessed as poor quality [37,38] lacked multiple exposure and outcome assessments, they were included in the review on the basis of pre-determined eligibility criteria; these two studies also presented extensive genomic and epigenomic data, which was determined as an additional strength of the study. A total of three quality assessment criteria (items #3, #7, and #11) were reported by all included studies. Four quality assessment criteria were rarely reported (items #5, #8, #12, and #14), and one of the criteria (item #10) was not reported by any of the included studies.

Patients with Systemic Diseases
Rheumatoid arthritis (RA) and periodontitis share a common pathogenesis pathway [21]; therefore, the epigenetic regulation in RA has been assessed and compared to periodontitis. Kojima et al. (2016) evaluated the DNA methylation of 12 CpG motifs in the TNF-α gene promoter region from peripheral blood of the Japanese population with CP compared to RA and controls [35], taking into consideration that TNF-α is an inflammatory cytokine that is elevated in active and progressing periodontitis [46][47][48]. Results of this study revealed that both CP and RA groups had higher DNA methylation rates and fre-quencies than the control groups at −72 bp [35], suggesting a pattern of hypermethylated CpG motifs in the TNF-α gene promoter in blood cells in Japanese adults with CP and RA.
In an earlier study on periodontitis and RA, Ishida et al. (2012) assessed the DNA methylation status of 19 CpG motifs in the IL-6 gene promoter from the peripheral blood of patients with CP compared to RA and controls [36]. Notably, IL-6 is a multifunctional pro-inflammatory cytokine that is highly expressed in periodontitis [49]. Ishida's group reported significantly lower DNA methylation levels at −74 bp and +19 bp in CP and RA patients when compared to controls. Moreover, the methylation levels at −74 bp but not +19 bp were also associated with the production of IL-6 by mononuclear cells, suggesting that hypomethylation of a single CpG motif in the IL-6 promoter region might result in increased levels of IL-6 in the serum. This observation could indicate the role of DNA methylation status at this specific CpG motif in the pathogenesis of RA and periodontitis [36].
The relationship between diabetes mellitus and periodontitis is well-established by the fact that periodontitis is the sixth complication of diabetes mellitus [50] and diabetes being a major risk factor for periodontitis [51]; therefore, the epigenetic modifications were evaluated in periodontitis and diabetic patients. In this context, Grdovic et al. (2016) evaluated DNA methylation of CXCL12 promoter in epithelial cells from CP, diabetes/CP patients and healthy controls. CXCL12 is a pro-and anti-inflammatory chemokine [34], and its role in the development and progression of both periodontitis and diabetes remains elusive as its effects range from protective to destructive [34]. Although Grdovic et al. (2016) demonstrated an increased DNA methylation percentage in the CXCL12 promoter in healthy patients with periodontitis and in diabetic patients with periodontitis, compared to healthy controls, these results failed to show a statistically significant difference [34]. Nonetheless, it can be inferred that chronic inflammation contributes to the change of DNA methylation of CXCL12 promoter in buccal epithelial cells, which might play a role in the development and progression of periodontal disease.
Since epidemiological data have demonstrated that CP patients tend to have a significantly higher incidence of squamous cell carcinoma [52,53], Planello et al. (2016) analyzed DNA methylation in CP and healthy controls compared to oral squamous cell carcinoma samples [37]. Their results demonstrated hypermethylated CpG sites of SOCS1 and RMI2 in samples from periodontitis and in samples from squamous cell carcinoma, and hypomethylated sites consistent in both periodontitis and squamous cell carcinoma. These findings might suggest the emergence of the pre-neoplastic epigenome in CP.
In a recent systematic review, periodontitis was associated with an increased longitudinal risk of cancer mortality [54]. Due to the possible association between chronic periodontitis and cancer on the molecular level [55,56], epigenetic changes in periodontitis and breast cancer have been evaluated. Wang et al. (2014) evaluated the DNA methylation of TIMP3, GSTP1, and 14-3-3 in CP, healthy controls, and patients with breast cancer, since these three genes were previously investigated in the diagnosis and treatment of cancer. The results of their study demonstrated that hypermethylation levels of TIMP3 and GSTP1 were significantly lower in periodontitis patients and controls than those with breast cancer [38]. However, the epigenetic silencing of 14-3-3 occurred more frequently in the CP group compared to the healthy and breast cancer groups, suggesting that 14-3-3 plays a role in the development of CP.
In an earlier investigation, Loo et al. (2010) evaluated the DNA methylation of Ecadherin and COX2 promoter in CP, breast cancer patients, and healthy patients. Both genes are key indicators of cancer and also have links to periodontitis [39]. In fact, COX-2 is reportedly downregulated in chronic periodontitis, which might serve as a new setpoint to restrict further tissue destruction [57]. The results by Loo et al. (2016) revealed that the promoters of Cadherin and COX2 were hypermethylated the most frequently in breast cancer, followed by periodontitis and the least frequently in controls [39]. These findings suggest that epigenetic changes presented in CP patients might demonstrate irreversible destruction in the tissues or organs similar to the effects of cancer. DNA hypermethylation in CP might be associated with DNA hypermethylation which is related to cancer risk factors.

Smokers and Non-Smokers
Since smoking is considered a major risk factor for periodontitis [58] and also influences epigenetic changes [59], DNA methylation patterns were evaluated in smokers with periodontitis. Toll-like receptors (TLRs) were evaluated, given their important role in the inflammatory process in periodontitis [60]. Gingival tissue samples from smokers and non-smokers affected by periodontitis as well as healthy patients were found to have major unmethylation of the TLR4 gene promoter, whereas the results for the TLR2 gene promoter were inconclusive given the mosaic of methylated and unmethylated DNA in the majority of samples [41]. There was also no statistically significant correlation between the DNA methylation status and mRNA transcription levels of both genes between the groups [41]. In another study, individuals with periodontitis displayed a higher percentage of hypomethylation of the CXCL8 gene, compared to healthy controls, independent of smoking habit [42]. This is in concordance with previously published papers that reported an increased expression of CXCL8, indicating its role in tissue destruction in periodontal diseases [61,62].
In another investigation assessing the methylation status of the extracellular matrix (ECM) organization-related genes, findings indicated that smoking might change the transcription and methylation status of ECM organization-related genes, which aggravated the periodontal disease [40]. These findings suggested that smoking-related changes in DNA methylation and successive changes in the expression of the ECM component-related genes may result in increased susceptibility to periodontitis in smokers by influencing ECM organization, which in turn may affect disease characteristics.
In a more recent investigation, the DNA methylation pattern of the SOCS1 promoter, a potent inhibitor of cytokine signaling [63], was evaluated in epithelial cells from the saliva of smoking and non-smoking patients with periodontitis [64]. Results showed that cells from the saliva of periodontitis patients who smoked were 7.08 times more likely to show a methylated SOCS1 promoter in comparison to periodontitis patients who are non-smokers, suggesting that SOCS1 may be a consequence of tobacco exposure and not periodontitis. This finding that methylation of SOCS1 promoter is solely due to tobacco is consistent with previous studies on SOCS1 and periodontitis in which SOCS1 hypomethylation was reported in chronic and aggressive periodontitis [63,65].

Limitations and Confounding Variables
The results of the present systematic review need to be interpreted with caution due to certain limitations. First, the number of available studies on the DNA methylation in periodontitis patients who have a systemic disease or smoke is quite limited to generalize conclusions. Most of the studies lacked information on the ethnic background of the study participants, which is an important factor that affects epigenetics. Regarding studies on smoking, there was inconsistency between the studies in identifying what a (smoker) patient was, i.e., how many cigarettes they smoked a day. These studies did not stratify smoking patients into slight, moderate, and heavy, which could affect the conclusions of the included studies.

Future Research
There is a need for further properly designed and conducted clinical human studies to evaluate the DNA methylation and periodontal and systemic diseases as well as smoking. Future investigations should further aim to identify DNA methylation of specific genes as potential diagnostic biomarkers for periodontitis in patients with systemic diseases at the chairside in dental practice. Furthermore, future research should aim to further understand the relationship in DNA methylation of specific genes between patients with periodontitis, systemic diseases, and smokers, in order to utilize this epigenetic modification as a tool for patient stratification and thereby a more personalized treatment approach based on the disease susceptibility. Moreover, experimental and clinical trials should be performed to further study the epigenetic association of short-and long-term interactions of biomaterials with gingival tissue.

Clinical Relevance
Understanding the link between periodontal diseases and systemic diseases/smoking through epigenetics will aid in increasing the knowledge and understanding of the pathogenesis and progression of periodontitis in patients with systemic diseases and smokers. In addition, the fact that epigenetic mechanisms are reversible makes them attractive targets and offers new treatment strategies. By identifying the pathways that can be targeted to promote epigenetic changes, periodontal tissue destruction can be limited. Moreover, epigenetic drugs, also known as epidrugs, can be used as an adjunct to periodontal therapy in order to improve the outcomes of periodontal therapy in these systemically compromised patients.

Conclusions
There was some evidence, albeit inconsistent, for an association between DNA methylation and periodontal diseases and systemic diseases or smokers compared to non-diseased or non-smokers. However, due to the limited number of studies and the heterogeneity in study design, population and outcomes, no definite conclusion may be drawn between DNA methylation and periodontal diseases and systemic diseases or smokers. The DNA modifications in periodontal and systemic diseases in human adults may lead to the identification of potential disease therapies. Therefore, further research in DNA methylation and periodontal diseases and systemic diseases or smokers in humans is needed in developing alternative therapeutic approaches to treat systemic conditions and periodontal diseases by improving wound healing and periodontal tissue regeneration.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/app11115269/s1, Supplemental SI: Search protocol used in the systematic review; Supplemental SII: Extracted information on study characteristics; Table S1: Studies Excluded After Full-text Review; Table S2: Risk of Quality Assessment.