Background: Autoimmune hemolytic anemia (AIHA) represents a serious and potentially life-threatening illness. However, the mechanisms by which red blood cell (RBC) autoreactive B cells escape tolerance and subsequently receive the signals required to differentiate into plasma cells are unknown. It has been reported that B1 B cells, a special B cell lineage that arises from gut lymphatics, appear to be resistant to self-tolerance to RBC antigens. However, the existing murine models do not allow experimental juxtaposition of animals with and without the autoantigen. Moreover, in existing models the mice develop AIHA with systemic activation of complement, crosslinking of Fc receptors, phagocytosis of RBCs, and generation of large quantities of hemoglobin breakdown products. Although these events are relevant to the pathology of AIHA, they obscure the underlying immunology of the system. To circumvent these problems, we have designed a novel model of autoimmunity to self RBC antigens that allows the analysis of B cell tolerance in the absence of ongoing hemolytic pathology.

Materials/Methods: We generated the HOD mouse, which has RBC specific expression of the model humoral antigen hen egg lysozyme (HEL), linked to the cell membrane by a human blood group antigen (Duffy). HOD mice were crossed with B cell receptor transgenic IgHEL mice that express IgM specific for the HEL antigen. Heterozygotes were used for breeding to generate progeny that include HOD positive × IgHEL positive (“double positives”) and HOD negative × IgHEL positive (“controls”). Anti-HEL expressing B cells were identified by flow cytometry in spleen, bone marrow, and lymph nodes by the anti-HEL transgenic allotype (IgMa) and CD19 expression. B cell maturation was monitored by B220 expression. B1 B cells in the peritoneal cavity were identified utilizing CD11b (anti-MAC1). Serum anti-HEL IgM was determined by flow cytometric cross matching and HEL-specific ELISA (IgG is not secreted by anti-HEL BCR transgenic mice as they do not class switch).

Results: Double positive mice had significantly fewer B cells expressing anti-HEL (CD19+, IgMa+) than did control mice (7 fold fewer in spleen, 4 fold fewer in bone marrow, and essentially none in lymph nodes, as measured by % of total lymphocytes). These differences were observed predominantly in mature (B220 high) but not immature (B220 low) B cells. In contrast, while there was a dramatic decrease of conventional B cells in the peritoneum of double positive mice, B1 B cells persisted. These observations were not an artifact of expression of the HOD transgene, as no differences were seen in control HOD positive × IgHEL negative mice. Analysis of serum demonstrated that anti-HEL IgM levels in double positive mice were approximately 50% greater than that of control mice.

Conclusions: Utilizing our novel model of B cell tolerance to RBC autoantigens, we report a decrease of self reactive B cells in the spleen, bone marrow, and lymph nodes. This difference is likely generated during or after B cell maturation, as the decrease is restricted to B220 high mature B cells. However, autoreactive B1 B cells escape deletion in the double positive mice and persist in the peritoneum. Although the absolute number of self reactive B cells is significantly lower in the double positive as compared to control mice, autoantibody levels (anti-HEL IgM) are higher. Thus, the small percentage of autoreactive B cells that escape tolerance likely produce large amounts of autoantibody. Unlike models previously described by other investigators, the mice that have lost anti-RBC tolerance described herein show no signs of illness and have a hematocrit similar to that of control mice. Thus, our model provides a unique opportunity to perform well controlled studies on the mechanisms of tolerance to self-RBC antigens in the absence of systemic immune activation that occurs during clinical AIHA. Ongoing studies are investigating the role of B1 B cells on autoimmunity in this model system, as well as potential therapies to eliminate such cells.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author