Herr et al. 10.1073/pnas.0607254104. |
Fig. 6. Saliva MMP-8 level as a putative biomarker of periodontal disease. (A) ELISA measurements of the average salivary MMP-8 concentration (error bars are ± SEM, n = 3) by patient identifier. Patient characteristics are color-coded by clinical classification regarding periodontal disease state ("Healthy" or "Periodontitis"). (B) Comparison of mean values for each classification. MMP-8 concentrations were significantly different between the two patient classifications (P < 0.05). Mean (average ± SEM) ELISA-reported MMP-8 concentration for healthy patients was 64.6 ng/ml ± 16.4 ng/ml (n = 9) and periodontally-diseased patients was 623.8 ng/ml ± 204.0 ng/ml (n = 14). Both pocket depth and clinical attachment loss were also observed to be significantly different between the healthy and periodontally-diseased patients.
SI Experimental Procedures
Human Subjects Inclusion Criteria.
Subject inclusion was based on: possession of at least 20 teeth, not having received periodontal treatment or antibiotic-related therapy for medical or dental reasons for 3 months before study inclusion, not having received long-term treatment with medications known to affect periodontal status, and having no history of metabolic bone diseases (i.e., rheumatoid arthritis, postmenopausal osteoporosis). Healthy control classification was based upon the criteria: less than 3 mm of clinical tooth supporting attachment loss, no pocket depths greater than 4 mm, no radiographic bone loss, and less than 20 sites with bleeding on probing. Periodontitis classification required that subjects exhibit at least 4 tooth sites with evidence of radiographic bone loss, mean clinical attachment loss of greater than 3 mm, pocket depths greater than 4 mm, and bleeding on probing.Saliva Collection and Off-chip Preparation Protocol.
In the protocol used in this study, »2 ml of whole saliva was collected through passive drooling into a graduated plastic tube. Aliquoted saliva samples were shipped frozen via next day air and stored at less than -20°C until assayed. Before mCEI analysis, samples were minicentrifuged for 3 min. The supernatant was collected, diluted 5´ in running buffer, and immediately analyzed by the mCEI method.Laser Induced Fluorescence Detection.
Excitation light (HeNe laser, 633 nm) was frequency modulated using a mechanical chopper (220 Hz modulation) and reflected off of a dichroic mirror (XF 2035) through a 5722-BH 40´ microscope objective (New Focus, Inc, San Jose, CA) that defined the detection point on the chip (see Fig. 1 of article). Focusing and two-axis alignment of the HeNe laser beam with respect to the separation channel relied upon a custom fixture mounted on a 3-axis translation stage. Fluorescence was filtered spectrally (695AF55 notch filter, XF3076) and spatially before detection by a Hamamatsu H5784 photomultiplier tube (PMT). The signal from the PMT was demodulated using a lock-in amplifier (Stanford Research Systems, Sunnyvale, CA) and signal was collected via a custom LabVIEW data acquisition interface (National Instruments, Austin, TX).