Factors influencing measurement of human salivary lysozyme in lysoplate and turbidimetric assays

The use of different assay conditions has complicated the evaluation of studies relating salivary lysozyme levels to oral or systemic disease. The purpose of this study was to compare values obtained for lysozyme activity in mixed saliva of 104 healthy subjects by using two assay techniques and four variations in sample preparation. Lysozyme activity was assayed by the turbidimetric and lysoplate methods with human colostrum lysozyme as the standard. Lysozyme activity in saliva samples made 0.5 M with respect to NaCl was compared with that in untreated samples with and without centrifugation. Mean values for lysozyme concentration in centrifuged saliva were 2.2 micrograms/ml with the turbidimetric assay and 5.9 micrograms/ml with the lysoplate assay. In samples which were salt treated before centrifugation, mean concentrations increased to 17.3 and 72.9 micrograms/ml, respectively. The results for uncentrifuged saliva were four to five times higher than the results for centrifuged saliva in each of the assay systems. Salt treatment without centrifugation produced values comparable to those obtained with centrifugation. The addition of salt to human colostrum or hen egg white lysozyme generally resulted in a 20 to 25% increase in expressed activity. These results indicate that the measurement of lysozyme in the supernatant of centrifuged saliva is of questionable value, most of the lysozyme in whole saliva is inactive and may be activated by markedly increasing the ionic strength, and values for lysozyme activity in whole saliva are much greater in the lysoplate assay than in the turbidimetric assay when the same standard is used.