Single cell studies on the antibody-forming potential of fractionated, hapten-specific B lymphocytes.

This study addresses itself to the problem of antibody formation in vitro by mouse splenic B lymphocytes enriched for reactivity to the hapten NIP by the hapten-gelatine binding and melting technique of Haas and Layton (1975). Small numbers of NIP-gelatine-bound B cells were placed in micro- cultures either by bulk dispensing of dilute cell suspensions, or by micromanipulation under direct microscopic visualization. Antibody formation was induced by the T cell-independent hapten-protein conjugate NIP-polymerized flagellin, using 10⁴ thymus cells per microlitre as 'filler' cells. The frequency of precursors of NIP-specific antibody- forming cells among bound cells was about 2.2 × 10^-2 (one cell in forty-five) by both statistical and direct evaluation, after adjustment for a background frequency of 6-10 × 10^-8 precursors in the thymus filler cells. Single clones commenced antibody secretion asynchronously, as shown by the fact that the incidence of positive cultures continued to rise over the whole three days of culture, and that very small clones of one to four plaque-forming cells (PFC) were still found on day 3. The mean PFC number per positive culture rose from 1.2 at day 1 to 4.7 at day 2 and about 20 at day 3. Negative feedback, presumably mediated by antigen-antibody complexes, was shown to be a powerful effect in microcultures containing bound cells. It was evident with 250 cells per well (five clones) and the effect was progressive, such that with 10⊃ bound cells per well, plaque numbers were 40-50 per cent of their potential optimum; with 10⁴, 10 per cent, and 10⁵, 1 per cent. With routine techniques of harvesting of cultured cells, a failure to disperse PFC uniformly into single cells was noted. Groups of two or more large cells of very similar morphology were frequently observed at the centre of haemolytic plaques. Micromanipulation studies revealed that essentially all large cells at the centre of plaques were PFC. It was found that a routine plaque count underestimated the true PFC number per culture by a factor of 2. In the absence of negative feedback, the system is capable of generating one PFC per two input B lymphocytes. This is by far the most efficient plaque- generating system yet described, and contrasts with a yield of one PFC per 10⁴ input cells in most typical tissue culture systems. Reasons for believing that both the culture and the enrichment systems can be further improved are discussed, and the potential of the method for understanding the cell and molecular biology of lymphocyte triggering is described. [ABSTRACT FROM AUTHOR]