Co-Culture of CLL Cells with MEACs (Myosin Heavy Chain IIA Exposed Apoptotic Cells) Promotes Viability of Leukemic Clones

Abstract 624

Clinical outcome in B-cell chronic lymphocytic leukemia (CLL) correlates with the amount of mutation in the leukemic clone's unique B-cell antigen receptor (BCR) sequence. Patients whose CLL clone has a mutated (>2% from germline) IGHV sequence have better outcome, with an overall median survival about 10 years greater than patients whose CLL clone has an unmutated IGHV sequence. Furthermore, ∼30% of CLL patients share BCRs with very similar or stereotyped amino acid sequences; it is unlikely that this occurs by chance alone and suggests a shared BCR reactivity to common epitope(s) exists among CLL clones. To determine these antigenic epitope(s), we expressed CLL BCRs as recombinant human IgG monoclonal antibodies (mAbs) and identified cytoplasmic non-muscle myosin heavy chain IIA (MYHIIA) as an antigen that binds to CLL stereotype subset 6 mAbs. To explain how intracellular MYHIIA could come in contact with CLL BCRs, we found that MYHIIA is exposed to the cell surface during apoptosis in a subgroup of early and late apoptotic cells, which we named MEACs (MYHIIA exposed apoptotic cells). Both intrinsic and extrinsic apoptosis induction pathways produce MEACs by a caspase-3 cleavage dependent mechanism, in which MYHIIA cleavage products traffic from the cytoplasm to the cell membrane. Because binding to apoptotic cells has been observed for many CLL mAbs, we tested the binding of MEACs to a panel of recombinant CLL mAbs, including only two from subset 6, and found that most CLL mAbs bound MEACs well, with the level of binding inversely correlating to the degree of IGHV mutation and directly correlating with significantly shorter overall survival. Because MEACs display a large number of antigens, of which MYHIIA is only one, binding to MEACs by non-subset 6 mAbs is not limited to MYHIIA and may likely involve other novel exposed antigens exposed during MEAC formation. Based on these results, we hypothesized that autoantigens binding to CLL BCRs may promote CLL clone vitality and growth, leading to worse patient outcome.

To test this theory, we visually examined CLL cell binding to MEACs by confocal microscopy. CLL cells were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) and dead cells were labeled with Alexa Fluor 680, with the MEAC subgroup labeled with rabbit anti-MYHIIA followed by anti-rabbit IgG-Rhodamine. Alternatively, CLL cells were labeled with goat anti-CD23 followed by anti-goat IgG-DyLight 649 and dead cells labeled with CFSE, with MEACs labeled as before. Both labeling methods reproducibly identified CLL cells immediately adjacent to MEACs with apparent membrane juxtaposition, consistent with binding. To test our hypothesis that the functional consequence of such binding would result in increased CLL cell survival, we measured the affect of MEAC binding on spontaneous CLL cell apoptosis. Peripheral blood mononuclear cells (PBMCs) from 36 CLL patients were cultured for 2–4 days. Using flow cytometry, apoptosis in CD3⁻CD19⁺ cells was measured after AnnexinV and 7-actinomycin D staining. A typical wide range of spontaneous apoptosis was observed (median = 22.88% (3.11%–74.50%) AnnexinV⁺). Despite this wide range, co-culture of CLL PBMCs with MEACs (3:1-5:1 ratio) resulted in a decrease in apoptosis in nearly all tested patient samples (34/36) as compared to cultures without MEACs (median = 10.68% (1.75%–57.13%) AnnexinV⁺). This decrease in apoptosis between cultures with or without MEACs was highly significant (P=0.0001, paired t-test). To correlate this result with BCR binding to MEACs, recombinant CLL mAbs will need to be generated from these patient samples. In the mean time, we have obtained the IGHV sequence from 34 of these patients and noticed that only one belongs to a stereotype subset that we previously found binds to MEACs; the others that have not been tested. Interestingly, the CLL cells known to bind MEACs have one of the highest decreases in apoptosis (91.0%) among our tested samples (30.4% AnnexinV⁺ without MEACs versus 2.7% AnnexinV⁺ with MEACs). Thus, these results are consistent with the theory that antigenic stimulation of CLL cells via the BCR promotes CLL cell viability and may relate to the level of disease activity.