Analysis of matrix metalloproteinase-8 levels in gingival crevicular fluid and whole mouth fluid among smokers and nonsmokers using enzyme-linked immune-sorbent assay and a novel chair-side test

Abstract

Aim:

To indigenously prepare a chair-side test kit for investigating and comparing the matrix metalloproteinase (MMP)-8 levels in gingival crevicular fluid (GCF) and saliva in patients with healthy periodontium, gingivitis and chronic periodontitis in smokers and nonsmokers. To validate the diagnostic accuracy of indigenously prepared chair-side test against enzyme-linked immune-sorbent assay (ELISA). Furthermore, to assess the effect of nonsurgical periodontal therapy (NSPT) on the levels of MMP-8 in GCF and saliva among the test groups.

Materials and Methods:

GCF and saliva were collected from 250 subjects. The study population were divided into five groups; health periodontium-nonsmokers (Group 1; n = 50), chronic gingivitis-nonsmokers (Group 2; n = 50), chronic periodontitis-nonsmokers (Group 3; n = 50), chronic gingivitis-smokers (Group 4; n = 50), chronic periodontitis-smokers (Group 5; n = 50). A chair-side test kit was indigenously prepared using polyclonal antibodies (principle of immunochromatography) to detect the MMP-8 levels, and it was validated against ELISA at baseline and 3 months after NSPT.

Results:

The chair-side test detected MMP-8 levels with a sensitivity and specificity in accordance with ELISA. MMP-8 levels at baseline were higher in Group 2 and Group 3 as compared to controls (P < 0.05), and decreased after therapy (P < 0.05). MMP-8 levels in GCF were greater than in saliva for all the groups, indicating GCF to be a better sample to detect the MMP levels.

Conclusion:

The chair-side test detected MMP-8 levels accurately making it a viable chair side diagnostic tool. It was effective for early diagnosis of the periodontal disease among high-risk population such as smokers.

INTRODUCTION

Periodontal diseases are multifactorial, and smoking is one of the risk factors with a profound influence on the pathogenesis of periodontitis.[1] The host immune products in periodontitis are synthesized locally and appear in gingival crevicular fluid (GCF) and saliva, making oral fluids ideal for obtaining diagnostic information of periodontium.[2] These markers include cytokines, prostaglandins, bacterial and host-derived enzymes, connective tissue-degradation products, alongside bone matrix components that are primarily isolated in GCF[3] and may be expressed in saliva.

Matrix metalloproteinases (MMPs) are one such group of enzymes which are pivotal in the mediation of tissue destruction. MMP-8, a collagenase synthesized by neutrophils, is the major MMP implicated in degradation and remodeling of the extracellular matrix. MMP-8 has a strong affinity toward type I collagen, which is present in abundance in the periodontal tissues.[4] Various periodontal treatment modalities ranging from scaling and root planing to extensive regenerative procedures are directed toward curbing the menace of MMP-8 generation and thus disease progression.[5,6]

Since, MMP-8 is a host biomarker and it's testing from saliva or oral rinse can be useful and cost effective in periodontitis patients’ maintenance care interval determination,[7] we chose to detect the enzyme levels in GCF and saliva. Current techniques to detect enzymes in oral fluids include enzyme-linked immune-sorbent assay (ELISA), immunofluorometric assays or alike, which are technique sensitive and time-consuming. We prepared a point-of-care test to overcome these shortcomings and make identification of periodontal disease markers like MMP-8 easier and rapid, aiding in early diagnosis and treatment.

The first point-of-care test was developed by Sorsa et al., in 1999[8] to detect MMP-8 along chair-side. However, it was not popularized due to the high cost incurred by the use of monoclonal antibodies. To make it cost effective and suit the Indian economy, we indigenously developed a chair-side test incorporating polyclonal antibodies, which were less expensive than the monoclonal antibodies to human MMP-8 and helps in early detection of diseases. Since, periodontal disease is a major cause of tooth loss; its detection in early stages provided a strong rationale for us to develop our chair-side test strips. Further, our study group constituted smokers as they represent high-risk population to develop periodontal disease. Furthermore, this was the first ever effort to develop and validate a chair-side test strip to detect MMP-8 levels in the Indian population.

Therefore, the aim of the present study was first, to indigenously prepare a chair-side test to investigate and compare the MMP-8 levels in GCF and saliva in patients with healthy periodontium, gingivitis, and chronic periodontium among smokers and nonsmokers using ELISA and chair-side test stick. Second, to validate the diagnostic accuracy of the indigenously prepared chair-side test. Furthermore, to inspect the effect of nonsurgical periodontal therapy on the levels of MMP-8, if any.

MATERIALS AND METHODS

The study population consisted of 261 subjects (181 males and 80 females), 30–39 years of age, screened from the outpatient section of Department of Periodontics and 250 subjects fulfilled the inclusion criteria for the study. The power of the study was set to be 90%.

The exclusion criteria were: Medically compromised patients requiring prophylactic antibiotics, patients on antibiotic therapy within the last 6 months, patients who had received any form of periodontal therapy, surgical or nonsurgical within 6 months of baseline examination, pregnancy, recent orthodontic treatment, pulpal or periapical involvement on the qualifying teeth. Subjects satisfying the selection criteria were selected consecutively, and ethical clearance was obtained from the institutional review board, Rajiv Gandhi University of Health Sciences, in accordance with the guidelines of Indian Council of Medical Research's ethical guidelines for biomedical research on human subjects (2000). Written informed consent was obtained from those who agreed to participate in the study.

Each subject underwent a full mouth periodontal probing and charting, the subjects were categorized into five groups based on clinical examination of gingival index (GI) (Löe and Silness, 1963), plaque index (PI) (Turesky, Gilmore, Glickman modification of the Quigley and Hein PI, 1970), oral hygiene index simplified (OHI-S) (Green and Vermilion, 1964), and clinical attachment level (CAL) using a University of North Carolina 15 probe (UNC 15 mm, Hu-Friedy). Six sites on each were measured, namely; mesio-bucal, mid-buccal, disto-buccal, mesio-lingual, mid-lingual, and disto-lingual. CAL ≥5 mm on at least two nonadjacent sites had to be present. Based on the inclusion and exclusion criteria, the study population was divided into five groups. Fifty nonsmoker subjects with clinically healthy periodontium, mean PI (PI ≤1), mean GI of (GI ≤1), mean OHI-S (OHI-S 0.00–3.0) no attachment loss (AL) were included in Group 1 (healthy). Group 2 (gingivitis-nonsmokers) consisted of 50 nonsmoker subjects with gingival inflammation as indicated by the mean PI ≥2, mean GI ≥2, mean OHI-S ≥3 and no AL. Group 3 (chronic periodontitis-nonsmokers) consisted of 50 nonsmoker subjects with a mean PI ≥2, GI ≥2, mean OHI-S ≥3 and CAL ≥5 mm. Group 4 (gingivitis-smokers) included 50 subjects with gingival inflammation as indicated by the mean PI ≥ 2, GI ≥2, mean OHI-S ≥3, no AL and smokers who had a history of cigarette smoking >10 cigarettes/day for a minimum of 5 years. Finally, Group 5 (chronic periodontitis group-smokers) included 50 subjects with a mean PI ≥2, GI ≥2, mean OHI-S ≥3, CAL ≥5 mm and smokers who had a history of cigarette smoking >10 cigarettes/day for a minimum of 5 years.

Gingival crevicular fluid samples from a single site and whole mouth unstimulated saliva samples were collected from each patient at baseline and 3 months after therapy. Following initial examination and sampling Groups 2, 3, 4, and 5 were treated by a standard phase of Newfoundland Standard Time comprising scaling and root planning under local anesthesia using area specific scaler (Hu-Friedy), ultrasonic scaler (E.M.S. electro medical systems SA). This treatment was rendered by one examiner (G.A) without any time limitation.

SITE SELECTION AND COLLECTION OF GINGIVAL CREVICULAR FLUID AND SALIVA

This was a double-blind prospective study in which clinical examination, group allocation and sample site selection were performed by one examiner (GA), the samples were collected on the subsequent day by the second examiner (MLVP) who also carried out the posttreatment clinical examination. This was done to ensure masking of the sampling examiner and to prevent contamination of GCF with blood associated with the probing of inflamed sites. In subjects with a healthy periodontium, GCF sample was collected from the mesio-buccal aspect of maxillary premolars. In gingivitis patients, site with most severe clinical inflammatory signs was selected. In chronic periodontitis cases, site with probing pocket depth (PPD) ≥5 mm was identified using UNC 15 probe and sampled.

On the subsequent day, after drying the area with blast of air, supragingival plaque was removed without touching the marginal gingiva and GCF was collected. A standardized volume of 3 µl was collected from each of the test sites using color coded 1–5 μl calibrated volumetric microcapillary pipettes (Sigma Aldrich, St. Louis, MO) using the extracrevicular (unstimulated) method. Microcapillary pipettes suspected of being contaminated with blood and saliva were discarded, and sample was obtained from the tooth that showed the next highest PPD in the same patient. GCF collected was transferred to eppendorf tubes containing 0.5 ml of phosphate buffer saline and stored at −70°C until the time of assay.

Unstimulated saliva was collected by a modification of the method described by Navazesh.[9] Saliva was collected over a period of 5 min with the patient seated upright. Before collection, the mouth was emptied by rinsing with distilled water and an initial swallow. The patient was asked to empty the saliva collecting in the mouth every 30 s for a period of 5 min into the sterile containers which were centrifuged and stored at −70°C.

MATRIX METALLOPROTEINASE-8 ASSAY

The assays were conducted using commercially available ELISA kit, according to the manufacturer's instructions. Highly sensitive ELISA kit (Boster Biological Technology Co., LTD, Shanghai, China) was used to detect the enzyme levels in the sample in duplicates. The kit made use of biotinylated anti-human MMP-8 antibody and Avidin-Biotin-Peroxidase Complex. Absorbance of the substrate color reaction was read on ELISA reader using 450 nm wavelengths. The total MMP-8 level was determined in nanograms (ng), and the calculation of the concentration in each sample was performed by dividing the amount of enzyme by the volume of sample (ng/ml).

FABRICATION OF POINT-OF-CARE TEST STICKS FOR CHAIR-SIDE MONITORING

The chair-side test was fabricated based on the sandwich ELISA principle. Nitrocellulose membrane (Boster Biological Technology Co., LTD, Shanghai, China) of pore size 0.45 μm was cut into small strips. One end of the strip was treated with methanol to make it hydrophilic, and methanol was washed off using 0.01M phosphate-buffered saline (PBS) buffer. 1 μl of primary polyclonal antibody to human MMP-8 (Booster Biological Technology Co., LTD, Shanghai) was added to the hydrophilic end of the nitrocellulose membrane and allowed to dry. GCF and saliva samples collected were transferred to eppendorf tubes containing 0.5 ml of 0.01 M PBS buffer. The test strip was immersed into this tube to allow the sample to bind the primary antibody. The strip was removed and washed thoroughly in the PBS buffer to remove the unbound sample. It was then dipped in an eppendorf tube containing secondary antibody to human MMP-8 conjugated with a peroxidase system for 10–15 min. The strips were washed in the buffer again and transferred to another eppendorf tube to which a color developing solution was added. The solution turns blue for samples positive for MMP-8. A color change within 5 min was recorded as +++ (strongly positive), a change between 5 and 10 min was recorded as ++ (moderately positive) and a change in color after 15 min was recorded as + (weakly positive). The GCF and saliva samples were used to detect the MMP-8 levels with fabricated test sticks. For clinical use, the results were detected in the four groups based on time taken for the color change to become evident. For simplicity, the samples which turned color within 5 min were more helpful when recorded as positive and projected to the patient for patient education.

Statistical analysis

The distribution of the data was checked for normality using Shapiro wilks test and the test results revealed that the data were parametric for data regarding MMP-8 levels in the GCF and saliva, PPD and relative attachment levels; whereas the data regarding PI, GI, OHI-S were nonparametric. One-way analysis of variance (ANOVA) and Kruskal–Wallis test was applied for inter group comparisons in all the five groups from baseline to 3 months recall visit. Further, post-hoc test was carried out using Mann–Whitney U-test. Pre- and post-parameter values were compared using paired t-test within each group, and a P ≤ 0.05 was considered to be statistically significant. To check for agreement between ELISA and chair-side test Kappa statistics [Table 1] was applied. The statistical analysis was performed using the SPSS Inc., Chicago, IL USA, version 16.0.

Table 1.

Quantitative assay and test stick result tabulated for Kappa statistics

RESULTS

The study groups included subjects of similar age and proportions of males and females per Group. The gingival and oral hygiene status of all the groups as recorded by the PI, GI, and OHI-S were compared using Kruskal–Wallis test and the results showed a statistically significant improvement in gingival health at the 3^rd month follow-up as compared to baseline [Table 2]. The inter group comparison of the PI, GI, OHI-S using Mann–Whitney U-test showed a statistically significant difference between Group 1 and all the other test groups, however, there was no statistically significant difference between the test groups as such. Posttreatment reduction in PPD in Group 3 was greater than for Group 5, with a mean reduction of 1.78 ± 0.12 mm in Group 3 and a mean reduction of 1.22 ± 0.01 mm in Group 5, thus indicating persistent and progressing disease sites in Group 5 [Table 3].

Table 2.

Intra group comparison of PI, GI and oral hygiene and chair side test kit scores of MMP-8 GCF levels index at baseline and 3 months time points (Kruskal–Wallis test)

Table 3.

Intra group comparison of PPD and RAL at various follow-ups (Students t-test)

Matrix metalloproteinase-8 concentration (ng/ml) in GCF and saliva using ELISA among the study groups at baseline and posttreatment was tested using one-way ANOVA and Tukeys multiple post-hoc test. The results also showed a significant difference in the GCF MMP-8 levels among all test groups at baseline and 3 months posttreatment [Table 4]. The reduction after treatment was also statistically significant. On comparing Groups 3 and 5, it was observed that at baseline, the mean GCF enzyme levels were higher in the Group 5 but this difference was not statistically significant. However, posttreatment levels showed a greater reduction in Group 3 than Group 5 which was statistically significant (P < 0.05) indicating that smokers may be resistant to treatment. Salivary MMP-8 was much lower than GCF and pretreatment to posttreatment changes were statistically significant within the groups. But on inter group comparison of the gingivitis groups (Groups 2 and 4) and the periodontitis group (Groups 3 and 5), there was no statistically significant difference at the 3^rd month follow-up. Although the group means indicate that MMP-8 levels were higher in Group 3 when compared to Group 5 at baseline, posttreatment levels showed a greater reduction for Group 3 [Table 5].

Table 4.

Intra group comparison of MMP-8 levels in the GCF and saliva at various follow-ups (one-way ANOVA)

Table 5.

Post-hoc analysis for pair wise comparisons of MMP-8 in the saliva and GCF at baseline and 3 months follow-ups (Tukeys multiple post-hoc procedures)

On observing [Table 4], GCF and salivary MMP-8 levels in the gingivitis groups (Groups 2 and 4) showed that the enzyme levels were higher in these groups compared to control group, lower than in the periodontitis groups. However, smokers showed higher enzyme levels in GCF at baseline and posttreatment compared to nonsmokers.

Table 2 also indicates the MMP-8 chair-side test results, the specificity and sensitivity of the test were calculated, kappa statistics were applied to assess the agreement between ELISA and the chair-side test results which indicate that the test stick results are in good agreement with the ELISA readings at baseline as well as the follow-up visit.

DISCUSSION

Involvement of MMPs in the destruction of periodontal tissues is strongly evident; of these MMP-8 is the most important associated with the collagen breakdown during periodontal destruction and is present in GCF and saliva of patients with chronic periodontitis.[10] Extracellular matrix degradation by MMPs plays a pivotal role in the pathogenesis of various inflammatory diseases such as rheumatoid arthritis, acute myocardial infarctions (AMI), and repair processes of these diseases. Recent data on salivary MMP-8 activation levels in AMI cases suggest an enhanced MMP-8 activation in saliva, concluding that AMI is reflected in serum and saliva.[11] Apart from investigating MMP-8 in oral fluids to ascertain high-risk individuals, there are studies on MMP-8 gene polymorphism to aid their identification for developing periodontal diseases. A recent study showed that none of the investigated single nucleotide polymorphisms in MMP-8 gene was individually associated with periodontitis, although specific haplotype showed an association with clinical manifestation of chronic periodontitis.[12]

This was a double-blind prospective study to gain an insight into the possible role of MMP-8 as a mediator of periodontitis using the indigenously prepared chair-side test and compare MMP-8 levels in GCF and saliva among five study groups and posttreatment in Group 2, Group 3, Group 4, and Group 5. Age-dependent changes in inflammatory mediators have been reported, so we selected subjects in age groups of 30–39 years to control the influence of age on MMP-8.[13] A single site was selected for GCF collection, which precluded pooling of samples from multiple sites.[14] Micropipettes were used to reduce the risk of loss of sample due to evaporation from the paper, also there would be no fear of the protein of interest binding to periopaper and becoming unavailable for analysis. Unstimulated samples were collected as an increase in vascular permeability of blood vessels following gingival stimulation has been reported, suggesting MMP-8 levels in GCF to be influenced by stimulation in sampling. Unstimulated whole mouth saliva samples were collected by a modification of the method described by Navazesh.[9,15]

To the best of our knowledge, this was the first study to investigate in Indians the concentrations of MMP-8 in GCF and saliva of chronic periodontitis patients, incorporating a high-risk category of smokers. The findings of our study showed a strong correlation between MMP-8 levels and degree of inflammation as indicated by changes in clinical parameters GI and enzyme levels which were in accordance with previously conducted studies.[13] MMP-8 was elevated in both GCF and saliva as evident by ELISA [Figures 1 and 2] and chair-side test [Figure 3]. On comparison of MMP-8 levels in both GCF and saliva between smokers and nonsmokers, smokers had an elevated MMP-8 level when compared to nonsmokers at baseline. On the contrary, the third month follow-up showed a greater reduction in MMP-8 levels of nonsmokers as compared to smokers, indicating a poorer response to therapy among smokers.[16,17] This also reflected the reduction in PPD which in the nonsmoker group was greater than for smokers with a mean reduction of 1.78 mm in nonsmokers and a mean reduction of 1.22 mm in smokers, probably indicating persistent and progressing disease sites in smokers.

Figure 1.

Comparison of five groups (1–5) with respect to matrix metalloproteinase-8 gingival crevicular fluid scores at baseline and 3 months time points

Figure 2.

Comparison of five groups (1–5) with respect to matrix metalloproteinase-8 saliva scores at baseline and 3 months time points

Figure 3.

Comparison of five groups (1–5) with respect to test strip scores of matrix metalloproteinase-8 gingival crevicular fluid at baseline and 3 months time points

Matrix metalloproteinase-8 levels were higher in GCF than saliva; proving GCF as a better diagnostic sample than saliva although the latter has the advantage of better and easier availability. The chair-side test also detected MMP-8 levels in the GCF samples better than in the saliva samples, this also showed good agreement in this accord as depicted by kappa values. The specificity and sensitivity of the test strip were good [Table 2] and on par with the previously designed test.[18] This confirmed that the specificity and sensitivity obtained with either polyclonal or monoclonal antibodies remained almost the same validating our results.[19]

Although ELISA is a highly sensitive, widely used test certain drawbacks like technique sensitivity, time consumption, and escalated cost provided a strong rationale for us to develop this rapid chair-side test. The indigenously developed chair-side test proved a good, economic diagnostic tool to detect MMP-8 levels in the periodontal patient along chair-side. Furthermore, it could prove effective in patient education and motivation; thereby improving maintenance by patients. The rapid chair-side test enables easy identification of subjects at risk of developing periodontal disease and thus helps render early treatment. Also, if the cost of such tests is minimized, it becomes affordable for the general public, then it can be applied on a large scale. This study due to its cross-sectional nature, no definitive conclusion can be drawn in this regard. To elucidate this, more longitudinal multicenter trial should be undertaken.