Optimization of Nanodisc Array Generation on Silicon Photonic Microring Resonators for Lipid-Protein and Membrane Protein-Protein Interaction Characterization

Abstract

Biological membranes are essential for all life. Membranes govern compartmentalization between cells and provide frameworks for processes such as cell growth and signaling. The interactions that occur at the membrane interface are driven by charge state interactions between lipid headgroups and/or membrane proteins at the surface and binding domains on soluble proteins. Sometimes these interactions are also regulated by second messengers such as calcium ions. Understanding the processes that occur at the membrane surface is pertinent for designing therapeutics that may be needed for intervention in the physiological process. One physiological process that depends on membrane surface interactions is the blood coagulation cascade.

The extrinsic pathway of blood coagulation relies on the membrane interface for initiation, propagation, and amplification. Upon tissue damage, phosphatidylserine (PS) and an integral membrane protein, tissue factor (TF), are exposed to the extracellular matrix. Together, PS and TF are essential for recruitment of blood coagulation factors for complex formation and proteolytic activation that initiates the cascade and continue downstream to form a blood clot. The principal interactions driving these membrane associations are from GLA domains which are regions rich in γ-carboxyglutamate that chelate calcium ions to the surface to bind with PS.

This doctoral dissertation presents the development and optimization of a high throughput technique for characterization of molecular interactions at the membrane interface using Nanodiscs which are high-density lipoprotein mimetics formed with membrane scaffold protein (MSP) and lipids. These Nanodiscs are arrayed on silicon photonic microring resonators for refractive index-based detection of molecular interactions at the membrane surface.

Nanodiscs were arrayed on silicon photonic microring resonators using physisorption between the lipid headgroups and silicon oxide coated surface. These arrays were generated using molecular printing or by pipette spotting under a microscope to produce 7 or 9 lipid environment arrays. These Nanodiscs were used to observe GLA domain enhanced binding affinity for environments with increasing PS or phosphatidic acid (PA). Of particular interest was the increased binding affinity displayed by activated factor VII and activated protein C for PA over PS. These physisorption arrays also quantitated quantitate PS synergy with phosphatidylethanolamine (PE) for all GLA domain-containing blood coagulation factors. The most prominent synergistic behaviors were observed with factor X (fX) and prothrombin while the least synergy was seen with factor IX (fIX).

This physisorption technique is label-free and easy to use; however, the reliance on charge interactions limits the lipid environment and makes incorporation of membrane proteins difficult. Additionally, there is little to no control over the Nanodisc orientation on the surface. To overcome these challenges, DNA-tagged Nanodiscs were developed and optimized. Using DNA-tags, Nanodiscs are tethered above the surface which allows for more control over Nanodisc array loading with different lipid environments. Observation of Nanodisc loading overtime with DNA-tags provides quantitation of loading that was used for surface loading correction to calculate protein binding per leaflet. Using this technique, TF was incorporated into 7 different lipid environments of PC, PS, PA, and PE to observe differences in lipid binding preference of fIX and fX with and without TF. The multiplexity was then pushed to 26 for characterization the binding of fX-GLA domain mutants in 10 different lipid environments with 2-3 biological replicates. Through these studies, DNA-tagged Nanodiscs have demonstrated potential as a high throughput technology on silicon photonic microring resonators.