Vascular endothelial growth factor-A (VEGF) is a major target of anti-angiogenic cancer therapies. Alternative VEGFA gene splicing generates isoforms with different biological activities regulated by matrix binding and signalling through VEGF receptors and co-receptor neuropilin-1 (NRP1). Main human isoforms are VEGF121, VEGF165 and VEGF189 (VEGF120, VEGF164 and VEGF188 in mouse). Anti-VEGF therapy has made a huge clinical impact, but only in some patients, with clinical data suggesting that high circulating VEGF121 and low NRP1 may predict response. Using mouse fibrosarcoma cells expressing individual VEGF isoforms (fs120, fs164 and fs188) and differential NRP1 levels, this study explored mechanistic links between NRP1 and VEGF isoforms and their potential as predictive anti-VEGF therapy biomarkers.
Transient and CRISPR/Cas9n NRP1 knockdown in fs188 cells (fs188-NRP1KD) reduced migration without altering proliferation compared to wild type cells (fs188wt). NRP1 overexpression in fs120 cells also retarded migration compared to fs120wt cells. However, kinases driving proliferation and migration including AKT, Src and ERK-1/2 were unaffected by NRP1 overexpression and knockdown while signalling of hepatocyte growth factor that utilises NRP1 as co-receptor was only inhibited in one fs188-NRP1KD clone.
Vascularisation of fs188-NRP1KD and fs120-NRP1OE tumours was comparable to corresponding wt tumours; however, NRP1 modification resulted in smaller tumours. Fs120-NRP1OE tumours appeared more desmoplastic than fs120wt tumours with fewer dilated vessels.
Results of this study are consistent with the emerging idea that the prognostic and predictive value of NRP1 is context dependent. It is unlikely that VEGF isoforms and NRP1 alone are predictive of anti-VEGF therapy response; further work is required to unravel the complexities surrounding biomarker discovery.
