A key challenge in tissue engineering is overcoming cell death in the scaffold interior due to the limited diffusion of oxygen and nutrients therein. We here hypothesize that immobilizing a gradient of a growth/survival factor from the periphery to the center of a porous scaffold would guide endothelial cells into the interior of the scaffold, thus overcoming a necrotic core. Proteins were immobilized by one of three methods on porous collagen scaffolds for cardiovascular tissue engineering. The proteins were first activated with 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide/sulfo N-hydroxysuccinimide and then applied to the scaffold by one of three methods to establish the gradient: perfusion (the flow method), use of a source and a sink (the source–sink method) or by injecting 5 μl of the solution at the center of the scaffold (point source method). Due to the high reproducibility and ease of application of the point source method it was further used for VEGF-165 gradient formation, where an ∼2 ng ml⁻¹ mm⁻¹ gradient was formed in a radial direction across a scaffold, 12 mm in diameter and 2.5 mm thick. More endothelial cells were guided by the VEGF-165 gradient deep into the center of the scaffold compared with both uniformly immobilized VEGF-165 (with the same total VEGF concentration) and VEGF-free controls. All scaffolds (including the controls) yielded the same number of cells, but notably the VEGF-165 gradient scaffolds demonstrated a higher cell density in the centre. Thus we concluded that the VEGF-165 gradient promoted the migration, but not proliferation, of cells into the scaffold. These gradient scaffolds provide the foundation for future in vivo tissue engineering studies.