Pitfalls in the quantitative estimation of the secretion of granule proteins by eosinophils

The secretion of preformed granule proteins by eosinophils is an important correlate of eosinophil activation. However, a review of the literature reveals large disparities in the amounts of these substances which were reportedly secreted when eosinophils were activated. In the present study we report that our attempts to quantitate the secretion of eosinophil peroxidase and eosinophil-derived neurotoxin from activated eosinophils by measuring these substances in the incubation supernatants were uniformly unsuccessful. We found that, once they were secreted, both eosinophil peroxidase and eosinophil-derived neurotoxin were promptly lost to assay and presumably destroyed. Thus the measurement of the difference in the concentration of these substances in eosinophils prior to and after activation, revealed that as much as 65% of the eosinophil-derived neurotoxin and 62% of the peroxidase in the eosinophils were lost to assay during activation of the cells whereas the largest amount of these substances which could be measured in the incubation supernatants never exceeded 2%. Evidence is presented that the destruction of eosinophil-derived neurotoxin must occur prior to the release of this substance into the medium. Attempts to inhibit the destruction of eosinophil peroxidase and of eosinophil-derived neurotoxin by incorporating various inhibitors into the incubations were unsuccessful. These results emphasize the need to monitor the overall recoveries of secreted products from activated eosinophils and suggest that meaningful estimates of the secretion of these granule proteins from activated eosinophils can only be obtained by measuring the residual content of these substances in eosinophils after they have been activated and comparing these values to the contents of eosinophils prior to activation.