Effect of cigarette smoking on subgingival bacteria in healthy subjects and patients with chronic periodontitis

The study was independently reviewed and approved by the Institutional Research Board (IRB) of Jordan University of Science and Technology (JUST). The purpose and methodology of the study, including possible future publication of clinical datasets, were fully explained to all participants or their parents if they were under the age of 18 years and their written consent was obtained before interviews, examinations and sample collection, complying with the tenets of the revised Declaration of Helsinki.

In total, 150 participants were enrolled in the study at the initial screening unit in the Dental Teaching Centre (DTC) of Jordan University of Science & Technology (JUST). Demographic data and medical history were collected, and brief extraoral and intraoral clinical examinations were performed. Patients with clinically healthy gingiva were referred to the oral diagnosis clinic and those with suspected periodontal disease were referred to the periodontics clinic, both located in the DTC. Selection criteria were: systemically healthy; Jordanian; no periodontal treatment or antimicrobial therapy for at least 6 months before the examination; at least seven natural teeth/quadrant with no fixed or removable prosthesis. Any participants using long-term medications, receiving orthodontic treatment, under the age of 16, or pregnant were excluded. Once in the designated clinic, supragingival plaque was scored and, when present, it was carefully removed. Teeth were isolated using cotton rolls, and subgingival plaque was collected prior to any periodontal examination by inserting a paper point for 30 s into the mesial sulcus of each posterior tooth apart from the third molar if present (16 sites in total). The clinical diagnosis was based on the 1999 Consensus Classification of Periodontal Diseases [16]. Four clinical variables were recorded for all participants: the periodontal pocket depth (PPD), clinical attachment loss (CAL), plaque index (PI) and gingival index (GI) [17]. The absence of periodontal disease was defined as the absence of sites with PPD of ≥4 mm in any tooth and confirmed by radiographic examination. Paper points contaminated with blood were not included as this might affect DNA quality, only points from non-bleeding sites were pooled together, placed in Eppendorf tubes and stored at –20 °C for extraction of genomic DNA and analysis. Only participants who satisfied the inclusion criteria and had a clear-cut diagnosis were included in the study and DNA was extracted from plaque samples. In total a study population of 94 unrelated individuals: 30 non-smoker CP cases (18–50 years), 9 current smoker CP cases (28–48 years), 37 non-smoker periodontally healthy subjects (17–53 years), and 18 current smoker periodontally healthy subjects (17–43 years).

In the first amplification, the 16S rRNA genes were amplified by PCR with universal primers 27 F and 1492R [18]. The primer sequences were: 27 F, 5′-AGA GTT TGA TCC TGG CTC AG-3′; and 1492R, 5′-TAC GGG TAC CTT GTT ACG ACT T-3′. PCR was performed in 25 μl volumes using 12.5 μl of PCR master mix (Promega, USA), 6.5 μl of nuclease free water, 1 μl of forward and reverse primers and 5 μl of DNA. Amplifications were performed using a PCR Thermal Cycler (Bio-RAD, USA) programed for 10 min at 95 °C for initial heat activation, 35 cycles of 45 s at 95 °C for denaturation, 45 s at 60 °C for annealing, 45 s at 72 °C for extension, and 7 min at 72 °C for final extension. The size of the PCR product using the universal primers was 1505 bp.

The presence or absence of the bacteria was determined by the presence or absence of the PCR band resulting from using previously reported species-specific primers [19] on the first universal PCR product as a template after being checked on Primer-BLAST (www.​ncbi.​nlm.​nih.​gov/​tools/​primer-blast/​). Primer sequences can be found in additional word document file (see Additional file 1). PCR mixtures were prepared using 5 μl of the first PCR amplification mixture and 20 μl of the reaction mixture containing 0.5 μl of species-specific primer, 13 μl of PCR master mix (Promega, USA) and 6.5 μl of nuclease free water. PCR amplification was performed in a PCR Thermal Cycler (Bio-RAD, USA) programed for 10 min at 95 °C for initial heat activation, 35 cycles of 1 min at 95 °C for denaturation, 45 s at 55 °C for annealing, 45 s at 72 °C for extension, and 7 min at 72 °C for final extension.

The PCR products were separated on 2% agarose gels stained with ethidium bromide and visualized under a UV light transilluminator. With each gel run, negative and positive control samples were run simultaneously. In any case of uncertainty about the presence or absence of a bacteria-specific band, the PCR was repeated.

Continuous variables were presented using mean ± SD and compared using the independent sample t-test, categorical variables were presented using frequencies and percentages, and were compared using Chi squared analysis. Binary logistic regression analysis was used to calculate the odds of having each bacterium in smokers compared to non-smokers for both healthy and CP patients. Smoking was the dependent variable and independent variables (covariates) were: age, gender, education level, PPD and type of bacteria. Gender, education level and type of bacteria were set as categorical variables, while age and PPD were set as continuous variables (default). Both crude and OR corrected for age, gender, education level and PPD (OR_adj) were presented. All statistical analyses were conducted using IBM SPSS 24 (SPSS Inc., Chicago, USA). The adjusted OR (OR_adj) was adopted for interpreting the results and a p-value ≤ 0.05 was considered to be statistically significant. SPSS could not calculate the OR for the bacteria if the prevalence was 100% or 0% in any of the compared groups; in that case, only the p-value for χ ² analysis was shown. A power calculation was conducted to determine the statistical power of our samples to detect an association using the G*Power program [20]. The input parameters were: effect size = 0.3 (small size = 0.1, medium size = 0.3), statistical significance level α = 0.05 and two tailed.