Conditional activation of an anti-IgM antibody-drug conjugate for precise B cell lymphoma targeting

Abstract

Cancerous B cells are almost indistinguishable from their non-malignant counterparts regarding their surface antigen expression. Accordingly, the challenge to be faced consists in elimination of the malignant B cell population while maintaining a functional adaptive immune system. Here, we present an IgM-specific antibody-drug conjugate masked by fusion of the epitope-bearing IgM constant domain. Antibody masking impaired interaction with soluble pentameric as well as cell surface-expressed IgM molecules rendering the antibody cytotoxically inactive. Binding capacity of the anti-IgM antibody drug conjugate was restored upon conditional protease-mediated demasking which consequently enabled target-dependent antibody internalization and subsequent induction of apoptosis in malignant B cells. This easily adaptable approach potentially provides a novel mechanism of clonal B cell lymphoma eradication to the arsenal available for non-Hodgkin's lymphoma treatment.

Keywords: B cell lymphoma; B cell receptor; MMP-9; antibody-drug conjugate; conditional activated antibody; masked antibody; matriptase.

Figure 1

Concept Overview. Chicken immunization with IgM from human serum was followed by splenic RNA isolation and cDNA synthesis. Variable antibody domains were amplified and assembled as scFvs for yeast surface display and FACS. IgM binding scFvs were reformatted and cloned into bacterial and mammalian vectors for scFv/scFv-Fc expression. After selection and characterization of a lead candidate, the IgM binder was reformatted as full-length antibody, fused with the masking IgM domain and ultimately conjugated with MMAE resulting in a masked anti-IgM (aIgM) ADC. Created with BioRender.com.

Figure 2

Design and mode of action of masked aIgM ADC. (A) Schematic representation of the masked aIgM ADC. The anti-IgM LC is (N-terminally) fused to an IgM domain via a linker with dual-protease site, the HC is (C-terminally) modified with DBCO-PEG₄-Val-Cit-PAB-MMAE. DBCO, Dibenzocyclooctyne; PEG, polyethylene glycol; Val, valine, Cit: citrulline; PAB, p-aminobenzyl alcohol; MMAE, monomethyl auristatin E. Surface representation rendered with UCSF ChimeraX (35) from PDB: 1IGT/7XQ8. (B) Conceptional mode of action of the masked anti-IgM ADC. In systemic circulation the masked aIgM ADC is not able to bind to either soluble pentameric IgM nor membrane-bound IgM. Once reaching the tumor microenvironment, tumor-specific proteases such as MMP-9 or matriptase hydrolyze the linker connecting the aIgM antibody and the masking IgM domain. The activated aIgM ADC regains binding ability leading to specific ADC uptake and killing of IgM⁺ lymphoma B cells. Created with BioRender.com.

Figure 3

Epitope binning of aIgMscFv-Fc and CH2/IgM competition. (A) BLI-assisted epitope binning. The four His-tagged constant IgM domains of the HC (CH1, CH2, CH3 and CH4) were loaded onto Ni-NTA biosensor tips and associated with 150 nM aIgMscFv-Fc, followed by dissociation. (B) BLI-assisted competition assay. Biotinylated aIgMscFv-Fc was loaded onto SAX tips and IgM CH2/IgM from human serum were associated in sequence. (C) Cellular competition assay. IgM⁺ SUP-B8 and Ramos B cells were incubated with 100 nM aIgMscFv-Fc and varying concentrations of IgM CH2 domain (39-10,000 nM). Detection was performed using anti-human IgG Fc-PE staining and flow cytometry.

Figure 4

Protease-activation of CH2-masked aIgM. (A) Reducing SDS-PAGE of depicted antibodies with schematic representations of heavy and (masked) light chains. (B) BLI measurement. The four antibody constructs (Rituximab, aIgM, CH2-aIgM, Protein A purified CH2-aIgM+MMP-9) were loaded onto AHC biosensor tips and associated with 100 nM IgM from human serum.

Figure 5

Cellular binding of unmasked and CH2-masked aIgM variants. Flow cytometry analysis of IgM⁺ (SUP-B8, Ramos) and IgM^- (IM-9) B cells incubated with aIgM, CH2-aIgM and Protein A purified CH2-aIgM+MMP-9 antibodies and stained via anti-human IgG Fc-PE secondary detection antibody. (A) B cells were incubated with 100 nM of respective antibodies. Negative control samples (0 nM, black) represent cells stained with secondary detection antibody only. Histograms were created using FlowJoTM v10 Software (BD Life Sciences). (B) Cell titration of respective antibodies (0.125-200 nM) on B cells. On-cell K_Ds were determined using variable slope four-parameter fit. Results are shown as mean RFU, error bars represent standard deviation derived from experimental duplicates. Data is representative of three independent experiments.

Figure 6

Internalization and cytotoxicity of unmasked and CH2-masked aIgM ADC variants towards B cells. For cytotoxicity studies IgM⁺ (SUP-B8, Ramos) and IgM^- (IM-9) B cells were exposed to varying concentrations (0.014-90 nM) of aIgM, CH2-aIgM and Protein A purified CH2-aIgM+MMP-9 MMAE-conjugated antibodies for 72 h. Cell proliferation was normalized to untreated control cells (0 nM). For internalization studies pHAb-conjugated aIgM, CH2-aIgM and Protein A purified CH2-aIgM+MMP-9 (0.014-90 nM) were applied to B cells and incubated overnight. Fold internalization was defined by the ratio of relative fluorescence units (RFU) of the respective antibody sample and the untreated sample without antibody (0 nM). EC₅₀s were determined using variable slope four-parameter fit. Results are shown as mean, error bars represent standard deviation derived from experimental duplicates.

Figure 7

Apoptosis induction of aIgM and CH2-masked aIgM ADC in B cells. IgM⁺ (SUP-B8, Ramos) and IgM^- (IM-9) B cells were exposed to 0 nM, 50 nM of aIgM-MMAE and 50 nM CH2-aIgM-MMAE for 72 h. Cells were stained with Annexin V-FITC and propidium iodide (PI) and analyzed by flow cytometry. Percentage of Annexin V-FITC+/PI+ and Annexin V-FITC+/PI- (apoptotic cells) is depicted in right bar chart. Data is representative of two independent experiments.