Patients with CG and CP were recruited from the Unit of Periodontology at Guy’s and St Thomas’ Foundation Trust and the Coronary Care Unit at St Thomas’ Hospital (Table 1). All subjects were aged between 38 and 58 years of age and had at least 15 standing teeth. Subjects were excluded from the study if they had systemic disease including diabetes mellitus, recurrent aphthous stomatitis, autoimmune disease, a history of malignancies, previous treatment for periodontitis, or a history of antibiotic therapy within the past six months prior to recruitment. Subjects were allocated into 4 groups: gingivitis, CP, Ging/CHD or CP/CHD. In each subject, probing depths and recession of all teeth were determined using a William’s probe. Recession was measured as the distance from the cemento-enamel junction to the gingival margin. Measurements were taken from 6 sites; probing attachment loss was then calculated as the sum of probing pocket depth and recession. Periodontitis was defined as probing attachment loss ³4 mm in at least 4 teeth and CG group as probing attachment loss <2 mm in all teeth. Ethical committee approval was obtained (code no 98/12/04) and subject consent obtained. 40 ml of venous blood were withdrawn from each subject.

The gingivitis subjects had minimal gingival inflammation and bleeding on probing (percentage of bleeding sites, mean ± s.d., 1.8 ± 0.8), compared with the diseased groups, with bleeding sites in CP (48.9 ± 11.7), coronary heart disease with gingivitis (Ging/CHD; 1.9 ± 0.7) and CP/CHD (45.9 ± 8.7), All Ging/CHD patients had CHD as determined by greater than 60% diameter stenosis in at least two major epicardial coronary arteries, and conversely only eight of the healthy control subjects were confirmed to be free of CHD based on angiography, the remaining members of the group denying any chest pain symptoms.

Recombinant HSP65 derived from Mycobacterium bovis was prepared at the National Institute of Public Health and Environmental Protection, Bilthoven, the Netherlands and used at a predetermined optimal concentration of 10 µg/ml. Human HSP60 was purchased from Stressgen (Victoria, Canada). The two HSPs were detoxified using Detoxi-gel columns (Pierce, Oxford, UK) and the endotoxin level was determined by Limulus Amoebocyte Lysate assay (Sigma-Aldrich, Poole, Dorset, UK). The concentration of endotoxin was < 0.007 U/µg or 7 pg endotoxin/µg for both HSPs.

^#significant p value for between group analysis of variance by Kruskal-Wallis test. ^*p value for pairwise comparison by Bonferroni between gingivitis and chronic periodontitis group. ^**p value for pairwise comparison by Bonferroni between gingivitis and gingivitis/CHD group. ^***p value for pairwise comparison by Bonferroni between chronic periodontitis and chronic periodontis/CHD group. NS = never smokers ES = ex-smokers PS = present smokers IQ range = interquartile range.

Peripheral blood mononuclear cells (PBMC) were seperated from defibrinated blood samples by density gradient centrifugation (Nycomed Pharma As, Oslo, Norway). Blood was diluted 1:1 with tissue culture medium (RPMI 1640, Sigma-aldrich, UK) containing penicillin 100 mg/ml, streptomycin 100 U/ml (Gibco, Paisley, Scotland, UK), 2 mM L-glutamine (Sigma-aldrich, Irvine UK) at room temperature. This mixture was then layered onto an equal volume of lymphoprep (Ficoll-Paque TM PLUS, Amersham Biosciences, Uppsala Sweden) in 50 ml plastic tubes (Bibby Sterilin Staffordshire UK) and centrifuged at 600 g for 30 minutes at 20˚C. Mononuclear cells were removed from the interface, and washed twice by spinning at 200 g for 10 minutes. The serum was retained and diluted with medium containing penicillin 100 mg/ml, streptomycin 100 U/ml (Gibco, Paisley, Scotland, UK), 2 mM L-glutamine (Sigma-aldrich, UK) to 10% and used in cell cultures.

PBMC from 100 patients were cultured (5 × 10⁵ cells/ well) in 24 well flat bottomed plates (Corning 25,820). Cytokine secretion by PBMC was blocked by adding monensin 4 mg/ml (Sigma-aldrich, Irvine UK) and brefeldin A 10 mg/ml (Sigma-aldrich, Irvine UK). PBMC were then activated with phorbol myristate acetate (PMA) 10 ng/ml plus ionomycin 1 µg/ml (Sigma-aldrich, Irvine UK), human HSP60 10 µg/ml (NSP-540 Stressgen UK), microbial HSP 65 10 µg/ml (NSP581 stressgen UK) or medium alone and incubated at previously predetermined optimal time of 18 hours (Figure 1) in a humidified atmosphere (5% CO₂ at 37˚C). To investigate the role of LPS contamination some of the patients PBMC were further incubated with heat treated LPS or HSP by boiling 100˚C for 30 minutes. After 18 hours of incubation, cells were transferred into 5 ml round-bottom tubes (Falcon 352,058). Cells were stained using CD4 fluorescein isothiocyanate (FITC) 5 µl (BD Pharmingen, San Diego), CD14 and CD8 peridinin chlorophyll-protein (PerCP) 5 µl (BD Pharmingen, San Diego), then incubated in the dark at room temperature for 30 minutes. Cells were then centrifuged at 500 g for 5 minutes and washed once in 2 mls of PBS/azide (×1 PBS, 0.5% BSA, 0.1% sodium azide). Cells were fixed with 1 ml FACS lysing solution containing <50% Diethylene glycol and < 15% formaldehyde (BD Pharmingen 349,202) diluted 1:10 with PBS/azide and incubated at room temperature for 10 minutes. The cells were washed in PBS/azide cells and then permeabilised with 500 µl of FACS permeabilising solution (BD pharmingen 340,973) containing 0.02% saponin, diluted 1:10 with deionised water. Cells were incubated for 10 minutes at room temperature and then centrifuged at 500 g for 5 minutes, washed once and resuspended in PBS/azide.

Cells were then stained for intracellular cytokines, using 1 µg/ml of the following monoclonal antibodies conjugated to phycoerythrin (PE): anti-human IL-2 phycoerythrin (PE) conjugate (554,566 BD Pharmingen, San Diego), anti-human IL-4PE conjugate (554,516 BD Pharmingen, San Diego), anti-human IL-6PE conjugate (554,545 BD Pharmingen, San Diego), anti-human IFN-γ PE conjugate (554,701 BD Pharmingen, San Diego), anti-humanTNF-α PE conjugate (554,513 BD Pharmin-

gen, San Diego), anti-human IL-17PE conjugate (560,438 BD Pharmingen) or CD64PE conjugate (Serotec Oxford). After 30 minutes cells were washed twice and resuspended in PBS/Azide. Samples were analysed in a Coulter Epics FACS machine, at least 10,000 events were analysed. Results were expressed as the mean fluorescence intensity (MFLI).

Data recorded on dental charts and data obtained from the different experiments were transcribed onto computer records and analysed using SPSS-11.0 for windows (2001). All continuous variables were examined to establish whether the data conformed to a normal distribution with the Lilliefors (Kolmogorov-Smirnov) test for normality. Skewness and kurtosis were also examined. Analysis revealed non-normal distribution for all variables tested which were not amenable to log or square root transformation. Therefore the non-parametric KruskalWallis ANOVA test was used to assess differences between the groups established. The significance level was set at p < 0.05. Post ANOVA pairwise comparisons between the gingivitis group, CP group, ging/CHD group and CP with CP/CHD group were carried out with MannWhitney U test, using the Bonferroni correction. Interrelations between two variables were tested using the Spearman ranked correlation coefficients (r_s); when the p value was <0.05, correction for smoking pack years was carried out. Non-parametric data are displayed in tables as median (1st and 3rd quartile). Data were also displayed as mean ± standard error of mean (SEM) to facilitate comparison with previously reported findings in the literature.

Pre-diluted standard (50 mL) and blank were added to a 96-well plate pre-coated with anti-serum CRP IgG (Kalon Biological Ltd). Serum samples were diluted 1:1000 with assay diluent (Kalon Biological Ltd.) and dispensed in duplicates to designated wells in CRP precoated plates. The plate was then incubated at room temperature for 60 minutes. Plates were washed 4 times with wash buffer (Kalon Biological Ltd); 100 mL of CRP tracer (affinity purified sheep anti-CRP labelled with alkaline phosphatase, Kalon Biological Ltd UK) were then dispensed to each well and incubated uncovered for 30 minutes at room temperature. Plates were washed again 4 times with washing buffer (Kalon Biological Ltd, UK).

Substrate solution (100 ml of 4-nitrophenylphosphate in substrate buffer Kalon Biological Ltd.) was then dispensed to each well and incubated at room temperature for 30 minutes. The reaction was stopped with 100 ml of (120 g/L) sodium hydroxide. Optical densities were read at 405 nm with microplate reader (Anthos 2001, Anthos labtec instruments UK). A standard curve was constructed with standard points and curve fitted with four parameter logistic curve fitting software. Test serum values were then read off the standard curve.

Unstimulated cells incubated in medium showed relatively low mean fluorescence intensity (MFLI) (Figure 2(g)) in all intracellular cytokines studied; no statistically significant difference was found amongst the 4 groups (ANOVA, p = 0.803, Table 2).

There was a statistically significant difference amongst the 4 groups in MFLI for intracellular TNF-α by ANOVA (p = 0.001) (Table 2, Figures 2 and 3); also between the gingivitis group compared to the CP group (p = 0.017) and between the gingivitis group and the ging/CHD