BAR-Bodies: B-Cell Receptor Antigens As the Targeting Moiety of Antibodies in Substitution for the Variable Region of Heavy and Light Chains

Background

Chronic antigenic stimulation of the B-cell receptor (BCR) seems to play a critical role in the pathogenesis of B-cell lymphomas. We recently identified ARS2 and LRPAP1 as the autoantigenic targets of the B-cell receptors of approximately 25% of diffuse large B cell lymphomas (DLBCLs) of the ABC type and 45% of mantle cell lymphomas (MCLs), respectively. These BCR antigens can be used to target lymphoma cells in an approach we designated as BAR (B-cell receptor antigens for reverse targeting). The optimal therapeutic format BARs can be integrated in has yet to be found. Since the most established approach to deliver therapeutic payloads to specific targets are antibodies which have well-defined pharmacokinetics, we constructed and tested an antibody like construct (BAR-body) incorporating the DLBCL-BAR ARS2 in substitution for the variable domains of the heavy and light chains.

Material and methods

To create the ARS2 BAR-body, we exchanged the heavy and light chain variable region sequences of an IgG1 antibody with a sequence of similar length (approximately 120 amino acids) of the ARS2 protein (aa 343 - 466) containing the DLBCL reactive epitope (aa 343 - 375). The construct was assembled in a pCR2.1 vector, then transferred to a pSfi FLAG Tag vector for fusion with the FLAG tag and transfected into HEK293 cells for production. Purification of the BAR-body was performed via anti-FLAG antibody affinity chromatography. The BAR-body was detected by western blot analysis and binding capacity to the ARS2-reactive lymphoma cell lines U2932 and OCI-Ly3 and the not ARS2-reactive control DLBCL cell line TMD8 was assessed by flow cytometry. ARS2 BAR-body induced cytotoxicity of lymphoma cells with an ARS2 reactive BCR was measured by LDH release assays with human PBMCs as effector cells at an E:T ratio of 10:1.

Results

We cloned, expressed and characterized an ARS2 containing BAR-body incorporating 4 molecules of the lymphoma-reactive epitope of ARS2 resulting in an antibody like construct using a BAR (ARS2) as binding moiety instead of normal variable regions. The ARS2 BAR-body could successfully be cloned and expressed as confirmed by western blot analysis, which showed the construct at approximately 150 kD as was to be expected. The BAR-body bound specifically to the ARS2-reactive lymphoma cell lines U2932 and OCI-Ly3 and did not bind to the DLBCL cell line TMD8, which has a B-cell receptor of different specificity or to lymphoma cell lines of different entities. In LDH release assays with 5 x 10⁴ PBMCs and 5 x 10³ lymphoma cells (E:T ratio of 10:1) the ARS2 BAR-body induced PBMC mediated specific lysis of the ARS2 reactive lymphoma cell lines U2932 and OCI-Ly3 but not the control DLBCL cell line TMD8 starting at a concentration of 0,1µg/ml. Cytotoxic effects were dose dependent, reached a maximum of 50% specific lysis at a concentration of 1µg/ml and did not increase at concentrations of 10µg/ml.

Conclusion

Here, we show that BARs can substitute for the variable domains as binding moiety in antibody like constructs to target the BCR of B-cell lymphomas. Because approaches using their specific cognate antigen for targeting the malignant B cells have an exclusive specificity for the BCR of the malignant clone, they can be expected to be less toxic than the currently available antibody derived therapies targeting B-cells, because they leave normal B-lymphocytes unaffected. By incorporating BARs into the well-known format of an antibody we hope to capitalize on years of experience with this therapeutic format from conducting and interpreting in vivo experiments to the translation of the BAR approach into the clinic.