Development of an enzyme-linked immunosorbent assay for determination of lysozyme in human parotid and submandibular-sublingual salivas

The specificity of lysozyme determinations in human parotid and submandibular-sublingual salivas of two subjects was assessed by comparison of lysozyme concentrations in native acidified salivas with purified enzyme obtained by immunadosorbent fractionation of the salivas. Lysozyme concentrations were measured by the turbidimetric catalytic method and by a newly developed enzyme-linked immunosorbent assay (ELISA). The validity of the assays was established by comparing assay results with enzyme concentration values determined from optical density-extinction coefficient calculations of the purified lysozyme peak. Values for purified enzyme were found to be similar, irrespective of the assay used to determine lysozyme concentrations, and were in agreement with extinction coefficient calculations. Based on the ELISA technique, recoveries of lysozyme from both parotid and submandibular-sublingual salivas were greater than 75 and 90%, respectively. Similar recoveries were noted for parotid saliva when determinations were based on the turbidimetric assay. However, the ELISA and turbidimetric assays differed with respect to lysozyme levels in submandibular-sublingual saliva because of the apparent presence of an enhancement factor which gave rise to higher lysozyme values in the catalytic assay and therefore resulted in low recoveries of purified enzyme. This catalytic enhancement factor was present in the nonadsorbed fraction of both subjects, as higher lysozyme activities were noted when nonadsorbed fractions were added to affinity-purified lysozymes. Lysozyme levels were also determined in the parotid and submandibular-sublingual salivas of caries-resistant and -susceptible adults. In general, levels of lysozyme in parotid saliva were lower in comparison to submandibular-sublingual saliva; however, significant differences in enzyme concentration were not evident between the caries-resistant and caries-susceptible subjects. Standard errors were within ±10% by the ELISA method.