Novel techniques to study fibronectin adsorption

Objectives: It is known that fibronectin (Fn) affects cell-substrate adhesion, consequently, in implantology the structure and composition of the initially adsorbed Fn layer to a large extent determines the biological response to a biomaterial. We have used the novel technique of neutron reflectometry (NR) to study fibronectin adsorption at the solid/liquid interface on hydrophilic silica surfaces. Analysing NR data provides information on the fibronectin layer density, thickness and protein conformation not available using other techniques. NR measurements have been complimented with real-time quartz-crystal microbalance dissipation (QCM-D) and optical ellipsometry.

Methods: The adsorption of Fn was studied using NR at the solid/liquid interface on silica substrates at three different protein concentrations: 10, 30 and 100µg/ml. The kinetics of Fn adsorption was measured using QCM-D, and optical ellipsometry was used to measure the layer thickness thus complementing these techniques. For each experiment, the bare substrate was first characterised, then Fn in 0.1M PBS (pH7.4) was added at various concentrations.

Results: The neutron reflectivity data was modelled using the optical matrix method. This revealed a low volume fraction Fn layer of thickness 3.8nm. This layer thickness was confirmed by ellipsometry where 3.5 to 4nm thick layers were observed. QCM-D measured the kinetics of adsorption and revealed that after 20 minutes the majority of protein adsorption was complete. Using the Sauerbrey equation, the amount of Fn adsorbed was calculated as 120ng/cm². This agreed well with the value modelled from the NR data – 110ng/cm².

Conclusion: Neutron reflectometry gives information about protein layer thickness and density at the solid/liquid interface not available from other techniques. The thickness observed from NR and ellipsometry is likely to correspond to one diffuse monolayer of Fn arranged side-on, while QCM-D measurements agree with the low protein amount.

Supported by the UK Engineering and Physical Sciences Research Council, Grant EP/D066298/1.