Tissue culture techniques were combined with cell separation procedures to investigate the cellular requirements for a response to antigen, leading to the production of antibody-forming cells. Mouse spleen was dissociated, and the cells were separated into various groups on the basis of density, size, and active adherence. The ability of fractions to initiate a response in vivo, on transfer to an irradiated recipient, was compared to the response in vitro; and this ability was correlated with the presence or absence of phagocytic cells. Two different antigens were studied, sheep erythrocytes (SRC) and polymerized bacterial flagellin (POL).
Density distribution analysis of spleen showed a wide density range of cells responding to both antigens in vivo. The same fractions responded to POL in vitro as in vivo. By contrast, only the light density regions responded in vitro to SRC. Response occurred in regions of overlap between lymphocytes and phagocytic macrophages.
Separation by active adherence on columns of large glass beads gave a preparation containing large, medium, and small lymphocytes but no detectable phagocytic macrophages and very low levels of phagocytic polymorphs. This lymphocyte preparation responded to both antigens in vivo. In vitro it gave a full response to POL, but no response to SRC. Addition of a small quantity of the adherent fraction, enriched for phagocytic cells, restored response to SRC. The use of strain-specific antisera in a mixed culture containing a C57 phagocytic fraction and CBA lymphocytes showed that the lymphocyte fraction contributed the precursors of the final antibody-forming cells. The accessory cells from C57 spleen banded in the light regions of the density gradient where phagocytic macrophages were found. Irradiated spleen cells also activated the lymphocyte preparation, suggesting that the irradiated host provided the accessory cells for the in vivo response to SRC.
Small lymphocytes were purified from spleen by the small glass bead size filtration technique. This sample of small lymphocytes responded less well to POL than the total lymphocyte population, but it responded as well in vitro as in vivo. The small lymphocyte preparation responded in vivo to SRC but not in vitro. Addition of a small quantity of the phagocyte-rich fraction from adherence columns restored the in vitro response to SRC.
The results indicated that phagocytic cells are not required in the initiation of an immune response to POL. By contrast some accessory cell, possibly a phagocytic macrophage, is required for a response to SRC. The basis for this marked difference is discussed.