Elevated Binding of Chronic Lymphocytic Leukemia Antibody to a Subset of Apoptotic Cells with Exposed Non-Muscle Myosin Heavy Chain IIA Correlates with Poor Patient Outcome.

Abstract 799

Persistent research has begun to provide an understanding of the pathogenesis of B-cell chronic lymphocytic leukemia (CLL), a clonal expansion of a single B lymphocyte bearing a B-cell antigen receptor (BCR) with a monoclonal antibody (mAb) of defined amino acid sequence. Analyses of large collections of CLL mAb sequences have shown that many CLL clones express mAbs with very similar or stereotyped sequences. Indeed, almost 30% of CLL patients can be grouped on the basis of stereotyped sequences. This remarkable similarity is improbable by chance alone and therefore suggests that CLL BCRs are reacting to a common antigen(s), which may provide the stimulation necessary for the leukemic clone to survive and expand.

We have been studying a large subset (at least 53 worldwide) of CLL mAbs that contain a rearranged heavy (H) chain encoded by IGHV1-69, IGHD3-16, and IGHJ3 with a stereotyped sequence (subset 6) and have shown that mAbs from this subset recognize non-muscle myosin heavy chain IIA (MYHIIA). In viable cells, MYHIIA resides in the cytoplasm as part of molecular motors involved in cell morphogenesis and locomotion. However, during apoptosis, MYHIIA structurally rearranges and becomes exposed on the cell surface, allowing interaction with CLL subset 6 mAbs. As assessed by immunohistochemistry of apoptotic cells generated spontaneously in a human T cell line (Jurkat), MYHIIA becomes exposed in large structures that do not contain condensed nuclear DNA. Using flow cytometry and staining with 7-amino-actinomycin and Annexin V-Phycoerytherin to determine early and late stages of apoptosis, we found that MYHIIA exposure (and concomitant CLL subset 6 mAb reactivity) occurred in a subgroup of both early and late apoptotic cells.

Because some CLL mAbs can bind apoptotic cells, we wondered if these mAbs bound autoantigens revealed in the subgroup of apoptotic cells that exposed MYHIIA. The revealed autoantigen(s) could be MYHIIA, other proteins such as vimentin or filamin B, or chemical modifications such as oxidized epitopes, which are induced and exposed during apoptosis. Using flow cytometry, we measured the binding of MYHIIA exposed apoptotic cells (MEACs) to a panel of CLL mAbs that included those in stereotyped subsets (18/26) with only two from subset 6. The majority of CLL mAbs (16/26) bound MEACs well. Among these high binders, the majority (14/16) belonged to characterized stereotyped CLL mAb subsets. Interestingly, the level of MEAC binding inversely correlated with the degree of IGHV mutation in a statistically significant association (p < 0.0032).

In CLL, IGHV mutation status associates with patient outcome, where patients with mutated (>2%) IGHV tend to have longer survival. Because MEAC binding correlated with IGHV mutation, we investigated the association of MEAC binding with patient outcome. Of the 26 CLL mAbs, survival data were available for 24 patients. Remarkably, CLL mAbs with high binding to MEACs (n = 15) correlated with shorter patient survival (median survival = 99 months), whereas CLL mAbs with low binding to MEACs (n = 9) correlated with longer patient survival (median survival not reached). This association was statistically significant (p < 0.0087). In comparison, the association with patient survival and IGHV mutation status (18 unmutated and 6 mutated) did not reach statistical significance in this patient cohort (p < 0.0582).

Thus, many stereotyped and non-stereotyped CLL mAbs recognize autoantigenic targets on MEACs, the subset of apoptotic cells exposing MYHIIA. Therefore CLL cells with these (or other) binding characteristics could be expanded in vivo, ultimately leading to poor patient outcome. Consistent with this hypothesis, our initial findings suggest that autoantigen (e.g., MEAC) binding, frequently found in members of stereotyped subsets, provides prognostic information for CLL patients regardless of IGHV mutation status and may better indicate patient outcome than IGHV mutation status. This possibility is provocative and requires further investigation with more mAbs from more patients. However if born out, this would be consistent with the notion that BCR binding and signaling can alter CLL cell biology and may represent the fact that IGHV mutation status implies the ability of the CLL BCR to bind antigen, whereas reactivity with MEACs is a true indicator of CLL BCR binding to antigen(s).