The overarching hypothesis of the proposal was “the hydrophobic effect dominates protein adsorption from aqueous-buffer solution to surfaces spanning the full range of water wettability”.  This hypothesis was to be tested using a novel implementation of the solution-depletion method to accomplish three specific aims: (1) Measure partition coefficients as a function of adsorbent surface energy (water wettability) for a broad array of purified proteins;  (2) Correlate interphase volume occupied by purified proteins with molecular dimensions, deduce interphase thicknesses/number of layers occupied by adsorbed protein, and follow changes in interphase volume as a function of adsorbent surface energy; and (3) Systematically examine the role of protein size in adsorption competition and evaluate how specific interactions among proteins (e.g. antibody-antigen interactions) affect measured adsorption competition.  Three key publications substantially accomplish these aims and results comprehensively support the overarching hypothesis in a manner that achieves energy-and-mass balance for protein adsorption from stagnant fluids.  In brief summary, supported work shows how adsorbent surface energy controls adsorption capacity by controlling the energy required to displace interfacial water and illuminates how protein can adsorb selectively from multi-component mixtures.  Understanding how interphase volume varies with adsorbent surface energy and how surface energy affects adsorption selectivity remains to be accomplished.

The experimental program accomplished Aim (2) first.  Interpretive mass-balance equations were derived from a model premised on the idea that protein reversibly partitions from bulk solution into a three-dimensional (3D) interphase volume separating the physical adsorbent surface from bulk solution.  A key concept of theory was that adsorption capacity was a material property that depended on surface energy and not on properties of adsorbing protein.  This maximal concentration capacity  scaled as mass (not moles) per unit volume.  Theory was shown to both anticipate and accommodate adsorption of tested proteins to the two different hydrophobic test surfaces (octyl sepharose and silanized glass), suggesting that the underlying model was descriptive of the essential physical chemistry of protein adsorption to these surfaces.  Application of mass balance equations to experimental data quantified partition coefficients , interphase volumes , and the number of hypothetical layers occupied by protein adsorbed within .  Partition coefficients measured the equilibrium ratio of interphase and bulk-solution-phase w/v (mg/mL) concentrations  and , respectively, such that .  Proteins studied were found to be weak biosurfactants with  and commensurately low apparent free-energy-of-adsorption  .  These measurements corroborated independent estimates obtained from interfacial energetics of adsorption (tensiometry) and were in agreement with thermochemical measurements for related proteins by hydrophobic-interaction chromatography.  Proteins with molecular weight  kDa were found to occupy a single layer at surface saturation whereas the larger proteins IgG and Fibrinogen required two layers.

Subsequently, Aim 1 was accomplished using methodology developed above.  It was shown that adsorbent capacity for albumin measured in interfacial-concentration units (mg/mL) decreased monotonically with increasing surface energy (water wettability) to detection limits near an adsorbent-particle water adhesion tension dyne/cm (nominal water contact angle ) and that albumin did not adsorb to (concentrate within the surface region of) more hydrophilic adsorbents.  These adsorbed-mass measurements corroborated predictions based on interfacial energetics and were consistent with AFM measurement of protein-surface adhesion.  Interpretive mass-balance equations derived from concepts developed in Aim 1 were shown to both anticipate and accommodate experimental results for all test adsorbents, suggesting that the underlying model was descriptive of the essential physical chemistry of albumin adsorption to surfaces spanning the full range of observable water wetting.  In particular, application of theory to experimental data showed that the free-energy cost of dehydrating the surface region by protein displacement upon adsorption increased with increasing adsorbent hydrophilicity in a manner that controled ultimate capacity for protein.  It was concluded that a simple, three-component free-energy rule adequately described protein adsorption from aqueous solution, at least for materials bearing varying surface concentrations of anionic (not cationic) functional groups.  These key concepts required modification for adsorption to ion exchange resin particles bearing strong Lewis acid/base functional groups that adsorb protein by an ion-exchange mechanism (as detailed in a fourth paper in review), but the overarching hypothesis was shown to apply to this class of surfaces. 

Aim 3 showed that a Vroman-like exchange of different proteins adsorbing from a concentrated mixture to the same hydrophobic adsorbent surface arises naturally from the selective pressure imposed by a fixed interfacial-concentration capacity (w/v, mg/mL) for which protein molecules compete.  A size (molecular weight, ) discrimination results because fewer large proteins are required to accumulate an interfacial w/v concentration equal to smaller proteins.  Hence, the surface region becomes dominated by smaller proteins on a number-or-mole basis through a purely-physical process that is essentially unrelated to protein biochemistry.  Under certain conditions, this size discrimination was shown to be amplified by the natural variation in protein-adsorption avidity (quantified by partition coefficients ) because smaller proteins (kDa) have been found to exhibit characteristically-higher  than larger proteins (kDa).