There is growing interest in the use of human whole saliva for diagnostics and disease monitoring as an alternative to blood samples. In contrast to blood, whole saliva is a non-sterile body fluid. Proper hand-ling and storage are required to preserve the integrity of potential biomarkers. We investigated salivary autoproteolytic degradation using a variety of approaches. We determined inhibition of protease activities by monitoring the endogenous proteome. In addition, the stability of highly protease-susceptible proteins—histatin 5, statherin, and PRP1—was assessed. Experimental variables included (a) protease inhibitors, (b) salivary pH, (c) incubation temperatures, and (d) sample heating. A cocktail containing AEBSF, aprotinin, pancreatic trypsin inhibitor, leupeptin, antipain, and EDTA could not prevent histatin 5, statherin, or PRP1 degradation in whole saliva. Among the other treatments evaluated, short-term storage of freshly collected samples on ice was effective without interfering with the chemistry of the proteome. In conclusion, whole saliva contains a unique mixture of enzymes as evidenced from their resilience to protease inhibition. Analytical evidence on protein stability is needed to ensure the validity of salivary biomarker study outcomes. Analysis of the data presented will provide help and guidance for the use of saliva samples for diagnostic purposes.

Periodontal disease is characterised by proteolytic processes involving enzymes that are released by host immune cells and periodontal bacteria. These enzymes, when detectable in whole saliva, may serve as valuable diagnostic markers for disease states and progression. Because the substrate specificities of salivary proteases in periodontal health and disease are poorly characterised we probed these activities using several relevant substrates: 1) gelatin and collagen type IV; 2) the Arg/Lys–rich human salivary substrate histatin-5; and 3) a histatin-derived synthetic analog benzyloxycarbonyl-Arg-Gly-Tyr-Arg-methyl cumaryl amide (Z-RGYR-MCA). Substrate degradation was assessed in gel (zymography) and in solution. Whole saliva supernatant enzyme activities directed at gelatin, quantitated from the 42 kDa, 92 kDa and 130 kDa bands in the zymograms, were 1.3, 1.4 and 2.0 fold higher, respectively, in the periodontal patient group (p<0.01), consistent with enhanced activities observed towards collagen type IV. On the other hand, histatin 5 degraded equally fast in healthy and periodontal patients' whole saliva supernatant samples (p>0.10). Likewise, the hydrolysis rates of the Z-RGYR-MCA substrate were the same in the healthy and periodontal patient groups (p>0.10). In conclusion, gelatinolytic/collagenolytic activities but not trypsin-like activities in human saliva differentiate health from periodontal disease, and may thus provide an adjuvant to diagnosis for monitoring of disease activity.