Two maturation-associated mouse erythrocyte receptors of human B cells. I. Identification of four human B-cell subsets.

Using rosetting tests with untreated mouse erythrocytes (M) and pronase-treated M (pro M), four human B cell subsets can be identified. Three of these, possessing the phenotypes BM+ pro M+, BM- pro M+ or BM- pro M-, constitute 17%, 61% and 22% of normal blood B cells respectively. The fourth subset, BM+ pro M-, does not occur in normal tissues but was found in the pre-B-cell line of Raji cells, indicating that this phenotype may be a marker for early B cells. Some differences in the proportion of each subset were found in cord blood, lymph nodes and tonsils. Surface-immunoglobulin-positive (SIg+) and -negative (SIg-) non-T cells were present in each subset. M and pro-M rosetting tests were applied to cells from blood of 27 cases of chronic lymphocytic leukaemia (CLL) and to cells from involved nodes, spleen or marrow in five cases of non-Hodgkin's lymphoma (NHL). In 15 cases of CLL, there was considerable increase in the BM+ pro M+ subset (BM+ pro M+ type CLL); in seven cases, there was a predominance of BM- pro M+ cells and in another four cases, BM- pro M- cells predominated. All five cases of NHL were greatly enriched in BM- pro M- cells. There was no obvious correlation between rosetting and other surface markers but BM- pro M- clones in CLL or NHL always stained brightly with FITC-anti-Ig. This was not found in BM+ pro M+ or BM- pro M+ clones. Rosette formation of neuraminidase-treated B cells with M identifies the same subset as B-pro-M rosetting in normals and CLL. Evidence is presented that two types of receptors are involved in M and pro-M rosetting, designated R1 and R2, binding to corresponding M ligands L1 and L2. M rosetting is due to R1-L1 binding while R2-L2 binding mediates B-pro-M rosetting. Shifts between subsets within the same clone in some cases of CLL suggest that the subsets are distinct maturational stage of B-cell development rather than families of B cells of different lineage. The following B-cell maturation sequence is proposed: R1+ R2- lead to R1+ R2+ leads to R1- R2+ leads to R1- R2-.