Smoking and periodontal disease: discrimination of antibody responses to pathogenic and commensal oral bacteria

Abstract

Smoking is an independent risk factor for the initiation, extent and severity of periodontal disease. This study examined the ability of the host immune system to discriminate commensal oral bacteria from pathogens at mucosal surfaces, i.e. oral cavity. Serum immunoglobulin (Ig)G antibody reactive with three pathogenic and five commensal oral bacteria in 301 current smokers (age range 21-66 years) were examined by enzyme-linked immunosorbent assay. Clinical features of periodontal health were used as measures of periodontitis. Antibody to the pathogens and salivary cotinine levels were related positively to disease severity; however, the antibody levels were best described by the clinical disease unrelated to the amount of smoking. The data showed a greater immune response to pathogens than commensals that was related specifically to disease extent, and most noted in black males. Significant correlations in individual patient responses to the pathogens and commensals were lost with an increasing extent of periodontitis and serum antibody to the pathogens. Antibody to Porphyromonas gingivalis was particularly distinct with respect to the discriminatory nature of the immune responses in recognizing the pathogens. Antibody responses to selected pathogenic and commensal oral microorganisms differed among racial groups and genders. The antibody response to the pathogens was related to disease severity. The level of antibody to the pathogens, and in particular P. gingivalis, was correlated with disease severity in black and male subsets of patients. The amount of smoking did not appear to impact directly serum antibody levels to these oral bacteria.

Fig. 1

Measures of the extent of inflammation [% of sites with bleeding on probing (BOP)] and periodontitis [full-mouth mean pocket depth (PD) and % of sites with pocket depth (PD) > 4 mm or >5 mm)] in subsets of patients stratified on race and gender. Bars show the group mean values and brackets identify 1 standard deviation.

Fig. 2

Salivary cotinine levels in the population stratified into categories based upon periodontitis severity. Bars show the mean group values and with brackets identify 1 standard deviation.

Fig. 3

Serum antibody levels to the pathogens: Actinobacillus actinomycetemcomitans (Aa), Porphyromonas gingivalis (Pg) and Treponema denticola (Td) and to the commensals: Veillonella parvula (Vp), Streptococcus sanguis (Ss), Prevotella loescheii (Pl), Actinomyces naeslundii (An) and Capnocytophaga ochracea (Co) in each subset of the population, separated by race and gender. Bars show the mean group values and brackets depict 1 standard error of the mean.

Fig. 4

Serum antibody levels in patients grouped according to mean pocket depth. The comparisons are made using summed (Σ) antibody across all tested pathogenic bacteria (Path) and all oral commensal bacteria (Comm) or the average antibody to the pathogens or commensals. Bars show the mean group values and brackets depict 1 standard deviation.

Fig. 5

Summary of antibody comparisons across each individual pathogenic or commensal bacterial species (see Fig. 3 legend) with patients grouped based upon whole-mouth mean pocket depth. Bars show the mean group values and brackets depict 1 standard error of the mean.

Fig. 6

Average serum antibody levels to pathogens and commensals in patients stratified by race and gender. Bars show the mean group values and brackets depict 1 standard deviation.

Fig. 7

Correlation between the summation (Σ) of immunoglobulin G antibody levels to pathogens and commensals in the entire population. The linear regression and significance value is depicted.

Fig. 8

Relationship of serum antibody levels to pathogens and commensals with patients stratified into categories based upon frequency of sites with bleeding on probing (BOP). Bars show the mean group values and brackets depict 1 standard deviation. No statistical differences were observed in responses across the various bleeding subsets of patients.

Fig. 9

Correlations of the summation (Σ) of immunoglobulin G antibody levels to pathogens and commensals in subsets of patients stratified based upon mouth mean pocket depths. The linear regression is depicted for each subgroup.