Fixation of mammalian cells for flow cytometric evaluation of DNA content and nuclear immunofluorescence.

Mammalian tissue culture cells were fixed with 3 different alcoholic fixatives--acetone:methanol, EtOH, and MeOH. The quality of the resulting DNA histograms was evaluated by comparison of CV, G1/G2 ratio, G1 mode, cell aggregation, and debris formation; 81-90% MeOH (final concentration) was determined to be the optimal fixative by these criteria. A procedure was then examined using a prefix with paraformaldehyde followed by MeOH (PF/MeOH). This procedure produced cell preparations with reduced debris and aggregation, equivalent mode and ratio, but increased CV when compared with MeOH fixation. Both MeOH and PF/MeOH fixation procedures were then compared for their utility in dual staining for DNA and intracellular immunofluorescence for a nuclear protein, SV40 T antigen (Tag). Since alcohols are known to affect immunofluorescence staining of some antigens, fixation with paraformaldehyde followed by Triton X-100 permeabilization (PF/TX) was also included in this comparison to generalize the study by providing an alternative to MeOH permeabilization. The three procedures were evaluated for the quality of the sample by measuring the same descriptors of the DNA parameter as in the alcohol study. PF/TX consistently produced samples with decreased DNA CV and less debris and aggregation compared to MeOH methods. Two criteria were used to evaluate immunofluorescence--the amount of Tag measured and reproducibility. All MeOH methods were equivalently reproducible with CV's less than 3%. PF/TX was slightly less so with a CV of less than 6%. In contrast, different levels of Tag were measured for each procedure. For mouse 3T3 cells infected with a recombinant retroviral vector encoding T antigen, the level of T antigen measured after PF/MeOH was 21% greater than in MeOH fixed cells, and the level in PF/TX fixed cells was 37% less. The fraction of fluorescence specific to T antigen for these cells was 79-83% for all procedures. The lower levels measured after fixation by PF/TX were shown to be due to epitope masking. Why higher levels are measured with PF/MeOH procedures is unknown at present but may be due to antigen retention. Therefore, each of these fixation methods may be used with confidence in reliability but they are not equivalent with respect to the molecular architecture of the nucleus. It is postulated that PF/TX permeabilizes cells but cells retain native supramolecular structure, whereas MeOH based fixatives disrupt this structure and randomize availability of epitope to antibody. If so, the two procedures could be used as complementary procedures to study gene expression and function.