Abstract
Human peripheral blood lymphocytes were examined for the presence of membrane-associated immunoglobulin (Ig) by using direct immunofluorescence (DIF) and the mixed antiglobulin-rosetting reaction (MARR); to compare the two tests identical fluorescein-conjugated antiglobulin reagents were used both for fluorescent staining and for the rosetting reaction. The MARR was unaffected by incubation of lymphocytes in serum-free medium at 37°C or in acetate buffer (pH 4.0) before the actual rosetting procedure, whereas these conditions of preincubation reduced the number of fluorescent staining lymphocytes. Further, the MARR (in contrast to DIF) gave essentially the same results whether the fluorescent polyvalent antiglobulin reagent was IgG or a derived F(ab′)2 preparation. Evidence is presented that indicates that unlike DIF, the MARR demonstrates only membrane-incorporated Ig. With fluorescein-conjugated anti-µ, anti-κ, anti-λ, and polyvalent antiglobulin reagents, the MARR was found to be more sensitive than DIF under the present test conditions, and approximately 20% of all peripheral lymphocytes were found to be Ig-bearing with the MARR. It should be stressed, however, that although the MARR appears to have been performed optimally, DIF was performed by using methodology that is widely available but may be suboptimal. Of potential importance is the finding that summation of the number of sheep erythrocyte- (E) rosetting and mixed antiglobulin-rosetting lymphocytes approximated 100%, yet only 3% of lymphocytes in T-enriched preparations formed mixed antiglobulin rosettes. These findings suggest that E rosette-negative, DIF-negative null cells are B cells that have membrane-incorporated Ig that is not easily detectable by DIF. This interpretation is discussed with regard to possible technical difficulties with the MARR and with respect to other data that indicate the presence of a third lymphocyte population distinct from T and B lymphocytes.