A comparative study of different methods for the assessment of apoptosis and necrosis in human eosinophils.

Eosinophils, prominent cells in asthmatic inflammation, undergo apoptosis or programmed cell death following deprivation of contact with survival-promoting cytokines such as IL-5 and GM-CSF. The aim of this study was to assess a number of techniques for the quantification of apoptosis in human eosinophils cultured with or without IL-5 or GM-CSF and following staurosporine treatment. The relationship between apoptosis and necrosis in eosinophils was also determined. Eosinophils 'aged' in vitro for 48 h exhibited endonuclease DNA degradation, apoptotic morphology, increased red autofluorescence and externalisation of phosphatidylserine (PS) as assessed by binding of FITC-labelled annexin V. Annexin V-FITC binding was first detectable in eosinophils maintained at 37 degrees C for 5 h post-purification. This method proved to be the most sensitive marker of apoptosis. Morphological assessment of wet preparations of eosinophils by Kimura staining was found to be the next most-sensitive marker followed by increased red autofluorescence. The latter was a relatively insensitive method for the detection of apoptosis. At 5, 20 and 24 h of culture trypan blue exclusion indicated that eosinophil viability was high (85-90% viable cells). However, propidium iodide (PI) staining and flow cytometry revealed that, by 24 h, approximately 75% of cells had compromised membrane integrity. Eosinophils maintained in IL-5 or GM-CSF exhibited a non-apoptotic morphology and levels of annexin V-FITC binding and PI uptake similar to that of freshly isolated cells. Staurosporine (10(-5) M) treatment of eosinophils maintained in IL-5 or GM-CSF resulted in significant levels of apoptotic morphology at 2 h (23.8% +/- 6.9, p < 0.025) which was associated with negligible annexin binding. At 6 h post-staurosporine treatment significant annexin-FITC binding (38% +/- 1.5, p < 0.025) was observed compared with 93% +/- 1.2 of eosinophils displaying apoptotic morphology. Exclusion of PI demonstrated membrane integrity at all time points up to 6 h. Thus, eosinophils aged in vitro in the absence of viability-promoting cytokines exhibit evidence of both apoptosis and necrosis simultaneously. In contrast, staurosporine-treated eosinophils exhibited both membrane integrity and rapid apoptosis-associated morphological changes detected by single step Kimura staining which preceded externalisation of PS.