Enhancement of complement-dependent cytotoxicity by linking factor-H derived short consensus repeats 19-20 to CD20 antibodies

Antibody-mediated complement-dependent cytotoxicity (CDC) on malignant cells is regulated by several complement control proteins, including the inhibitory complement factor H (fH). fH consists of 20 short consensus repeat elements (SCRs) with specific functional domains. Previous research revealed that the fH-derived SCRs 19–20 (SCR1920) can displace full-length fH on the surface of chronic lymphocytic leukemia (CLL) cells, which sensitizes CLL cells for e.g. CD20-targeting therapeutic monoclonal antibody (mAb) induced CDC. Therefore, we constructed lentiviral vectors for the generation of cell lines that stably produce mAb-SCR-fusion variants starting from the clinically approved parental mAbs rituximab, obinutuzumab and ofatumumab, respectively. Flow-cytometry revealed that the modification of the mAbs by the SCRs does not impair the binding to CD20. Increased in vitro lysis potency compared to their parental mAbs was corroborated by showing specific and dose dependent target cell elimination by CDC when compared to their parental mAbs. Lysis of CLL cells was not affected by the depletion of NK cells, suggesting that antibody-dependent cellular cytotoxicity plays a minor role in this context. Overall, this study emphasizes the crucial role of CDC in the elimination of CLL cells by mAbs and introduces a novel approach for enhancing CDC by directly fusing fH SCR1920 with mAbs.

In-house produced and purified antibodies were assessed for their quality and stability by DLS and nanoDSF.RTX-SCR1920 (Suppl.Fig. 1A) and OBI-SCR1920 (Suppl.Fig. 1B) were tested in three different buffer conditions.The unformulated sample in Tris-Gly-HCl elution buffer (pH 6.5; blue lines in Suppl.Fig. 1) was compared against a sample in elution buffer with 1:1,200 Trehalose, 10 mM L-His, 0.05% (w/v) Kolliphor P188 (red lines) or as a third option elution buffer with 1:1,200 Trehalose, 20 mM L-His, 0.05% (w/v) Kolliphor P188 (green lines).DLS analyses showed that in-house produced antibodies were of acceptable and sufficient purity.This was indicated by the sharp peaks of DLS measurement curves.Mean hydrodynamic radii between 5.93 to 6.97 nm for RTX-SCR1920 samples (Suppl.Fig. 1A) and 6.00 to 6.36 nm for OBI-SCR1920 samples (Suppl.Fig. 1B) were measured.Additional species between 100 to 10,000 nm were observed for both antibodies in all three formulation buffers.These higher molecular weight species were of low abundance and therefore neglectable.
Further, the turbidity signal determined by nanoDSF back-scatter measurements at rising temperatures served as indicator for microaggregation of the sample.One aggregation transition was observed between around 70°C and 80°C for all tested samples (Suppl.Fig. 2A).NanoDSF results showed the typical biphasic thermal melting transition of antibodies with two major unfolding events that corresponded most likely to the unfolding of individual antibody domains.One unfolding event for the CH2 domain and the second for the CH3 and Fab domains (see subgraphs "Thermal melting, first derivative" in the Suppl.Fig. 2A).
The thermal inflection points (T-IP) represent the temperatures at which approximately 50% of the sample/domain are unfolded.For RTX-SCR1920 two unfolding transitions were observed between 73.48-73.78°Cand 78.57-80.73°C(Suppl.Fig. 2A).For OBI-SCR1920 one broad unfolding transition was observed between 76.40-76.51°Cfor samples 1 and 2 and two unfolding transitions were observed for sample 3 (74.45°Cand 79.41°C; (Suppl.Fig. 2B)).This difference in sample 3 was associated to the higher concentration of this sample.The signal-to-noise level of all unfolding transitions sufficiently low and the unfolding transition temperatures could be determined with high accuracy.The unfolding inflection point values were similar for the three tested buffer conditions.

Antigen binding capacity of SCR-fusion constructs is not impaired compared to the parental antibody
The CD20 binding capacity of RTX-SCR constructs was tested on 2x10 5 CLL PBMCs of three different patients.The samples were treated with 5 μg RTX clinic, RTX in-house, RTX-SCR1920 or RTX-SCR1112 and incubated for 30 min at 4°C.After washing, the surface bound RTX was detected with a polyclonal rabbit anti-human IgG-FITC (Agilent Technologies) conjugated secondary antibody (incubation 30 min, 4°C).The mean fluorescence intensity (MFI) of the FITC-positive population was determined by flow-cytometry.
As patients differ in their CD20 surface expression, the MFIs varied between the three tested PBMC samples.The mean and SD were calculated and the statistical significance was analyzed by ordinary oneway ANOVA and a subsequent Tukey's multiple comparisons test (Suppl.Fig. 3).The antigen binding capacity did not differ significantly between the clinically used RTX, the in-house produced RTX, RTX-SCR1920 and RTX-SCR1112.Therefore, different binding capacities can be excluded as a reason for different target cell lysis capacities.

Verification of in vitro NK cell depletion in CLL PBMCs by flow-cytometry
The depletion of NK cells from CLL PBMCs was confirmed by staining PBMCs pre and post depletion with BD multitest CD3 FITC / CD16 PE + CD56 PE / CD45 PerCP / CD19 APC according to manufacturer's recommendations (BD Biosciences).
CD16+CD56+ PE-stained cells were quantified and compared.Suppl.Fig. 4 shows a representative sample of a patient.The proportion of NK cells decreased from 11.3% of the overall lymphocyte count before treatment to 0.52% after NK depletion, corresponding to a decreased NK cell count of approximately 95%.

ADCC assay
To evaluate ADCC induced by mAb-constructs, 2x10 5 MEC-1 cells (Leibniz Institute, DSMZ, Braunschweig, Germany) were combined with NK cells isolated from healthy donors in effector to target ratios of 1:5 or 1:10.Ab-constructs were added at 1-100 µg/mL in combination with 20% NHS or hiNHS.The samples were incubated at 37°C, 5% CO2 for 4 h and the amount of viable cells was determined by PI staining and flow cytometry as described above (Suppl.Fig. 5).Significances of changes induced through different E:T ratios were determined by two-way ANOVA and Tukey's multiple comparisons test.Significances were not included in Suppl.Fig. 5 for visibility reasons.The differences between means of OBI and OBI-SCR1920 treated cells with E:T ratios of 1:5 or 1:10 was only significant at 100 µg/mL antibody (*p=0.0172;** p=0.0015 respectively).For the hiNHS setting, only samples containing 50 and 100 µg/mL antibody and an E:T ratio of 1:5 displayed a significant difference when comparing OBI and OBI-SCR1920 treated samples (**p=0.0089;*p=0.0298respectively).

Phagocytosis inhibition assay
Spleen tyrosine kinase (SYK) inhibitor BI-1002494 (Boehringer Ingelheim, Ingelheim am Rhein, Germany) was used to inhibit phagocytosis exerted by macrophages and neutrophils.As suggested by the manufacturer, 1 µM of the inhibitor was added to each sample of a standard CLL lysis assay (as described in the main part) including 5 randomly chosen CLL patients.The patients' PBMCs were depleted from NK cells using the REAlease CD56 MicroBead Kit (Miltenyi Biotec), as described earlier to exclude that measured effects rely on ADCC rather than phagocytosis and CDC.All samples were normalized to a control sample, without antibody treatment or inhibitor, containing only NHS.The parental antibodies and SCR1920-conjugated constructs of RTX, OBI and OFA were tested at 60 µg/mL with 1 µM SYK ("+SYK-NK"), without SYK ("-SYK-NK") or without SYK but with NK cells (= PBMC composition prior to NK depletion; "-SYK+NK").