FIGURES AND TABLES

Figure 1.

Affinity and binding kinetics of the anti-ErbB3 A5 scFv. k _on and k _off rates were determined by surface plasmon resonance and used to determine the binding affinity (K _D) of the A5 scFv. (A) Sensorgram fit to 1 : 1 Langmuir binding model. (B) Analysis of data.

Figure 2.

The anti-ErbB2/ErbB3 bs-scFv ALM. (A) Cartoon of ALM depicting scFv orientation, linker sequence and kinetic constants of ALM for each target antigen. (B) UV adsorption spectrum chromatograph of ALM over Superdex 75 size-exclusion column.

Figure 3.

The A5-linker-ML3.9 bs-scFv selectively binds BT-474 tumour cells in vitro. Non-labelled BT-474 (ErbB2'+'/ErbB3'+') breast tumour cells were mixed with either an equal (A and B) or 18-fold excess (C) of fluorescently labelled MCF10a (ErbB2''/ErbB3'') normal breast epithelial cells. Cell mixtures were then incubated with buffer (A) or 100 nM ALM (B and C) and binding of ALM to each cell population was determined by flow cytometry with an anti-6XHis tag secondary antibody. MCF10a cells were sorted to the upper quadrants and the non-labelled BT-474 cells were sorted to the lower quadrants. Cells bound by the secondary antibody sorted to the respective right hand quadrants. Images on the left depict the raw flow cytometry data. Values on the right represent the absolute number and overall percentage of each cell type in the respective quadrants.

Figure 4.

Bispecific binding is required for optimal tumour targeting of the ALM bs-scFv in vivo. The biodistributions of radioiodinated ALM, ALD and DLM bs-scFv were analysed 24 h post-injection into xenograft-bearing SCID mice (n=5 per cohort). (A) Co-expression of ErbB2 and ErbB3 by the targeted tumour is required for optimal targeting of ALM in vivo. ¹²⁵I-ALM targeted ErbB2+/ErbB3+ tumour xenografts to 3-fold higher levels than xenografts that express only one of the target antigens. (B) Radioiodinated ALM (¹²⁵I-ALM), which is capable of bivalent association with the surface of Sk-OV-3 tumour cells, exhibited increased targeting as compared with ALD and DLM that targeted the tumours monovalently. Error bars represent the standard error of the mean (s.e.m.).

Figure 5.

The A5-linker-ML3.9 bs-scFv has intrinsic anti-tumour cell activity. (A) Treatment of BT-474 and MDA-361/DYT2 cells with ALM inhibits colony formation in clonogenicity assays. Treatment of (B) BT-474 or (C) MDA-361/DYT2 cells with A5 scFv, ML3.9 scFv or the combination of both indicates that the majority of the intrinsic anti-tumour cell activity of ALM is due to the anti-ErbB3 A5 scFv arm. Colonies larger than 0.35 mM were counted using an automatic colony counter. Error bars represent the standard deviation.

Table 1 - Expression levels of ErbB2 and ErbB3 on cell lines.

Table 2 - ALM selectively targets ErbB2/ErbB3 positive cells in vitro .