Relation of antigen-binding cells to immunological memory

Priming of mice with a conjugate of HSA with sheep red cells induced a high level of memory to HSA, with very little antibody production ("pure priming"). HSA specific antigen binding cells in the spleens of the primed mice were assayed by means of a rosette technique, using HSA conjugated to donkey red cells. Rosette formation was almost completely inhibited by soluble HSA, thus confirming that the RFC were specific for this antigen. Spleens of primed mice contained up to 0.6% RFC, as compared to 0.08% HSA specific RFC in the spleens of non immunized animals. Suspensions enriched in rosettes (containing up to 16% RFC) were prepared by centrifugation on BSA density gradients. Adoptive transfer experiments showed that the rosette rich fraction contained all the memory cells. A marginal level of memory could be transferred to irradiated recipients with 3000 rosettes. A comparable degree of responsiveness to HSA could also be transferred with 70,000 RFC enriched from spleens of non immunized mice, but only when injected together with primed, RFC depleted spleen cells. Kinetic studies showed that the level of memory correlated well with the number of RFC up to two months after priming. The number of RFC decreased at later time intervals (though remaining higher than in controls at all times), without a corresponding decrease in the level of memory. A change in the quality of the memory cell with time is postulated.