Novel approaches to understand and predict multicomponent protein sorption and kinetics in chromatographic materials

Understanding and modeling transport mechanisms in chromatographic media remains a key scientific challenge and obstacle to rational process design. Simple transport models may not capture changes in uptake and elution rates as a function of solution conditions. Further, these models often fail to predict changes in transport rates due to the presence of other species during multicomponent displacement and desorption. To describe such variations in transport rates, models frequently assume multiple mechanistic contributions and resistances to protein mobility. A major obstacle to developing more accurate models has been the inability to independently measure the contributions of individual transport mechanisms such as pore diffusion, binding kinetics, surface diffusion, film resistance and others. Additionally, whether for traditional process development or for model development and validation, performing the batch kinetic studies necessary to fully characterize transport rates over a parameter space is often prohibitively time-consuming. This dissertation presents multiple approaches, at different length scales, for obtaining, analyzing and interpreting results that can allow more reliable prediction of multicomponent chromatographic behavior. Improvements in multicomponent column breakthrough modeling through use of an improved intraparticle transport model are considered first. We use the shrinking-core model, which provides a reasonable description of particle uptake for proteins under strong binding conditions. Analytical column solutions for single-component systems are extended here to predict binary breakthrough chromatographic behavior. Additionally, analytical results for the location and profile of displacement effects and expected breakthrough curves are derived for limiting cases. More generally, straightforward numerical results have also been obtained through simultaneous solution of a set of ordinary differential equations obtained by extension of a transformation method to the multicomponent case. Both analytical and numerical predictions compare favorably with experimental column breakthrough data. The ability to model displacement behavior using simple analytical and numerical techniques is a significant improvement over current methods. While this model provides improved predictions, it is still limited by accuracy of the model inputs, which include individual and binary batch kinetic and isotherm data. Modern high-throughput liquid handling robotic techniques are investigated as a means of quickly obtaining the kinetic and isotherm data needed for chromatographic modeling over a variety of solution conditions. As these methods have become prevalent within the past five to ten years, a significant amount of time is spent on method development and validation. Moving forward, single-component isotherm and uptake data are collected and verified against literature results and these methods are extended to study multicomponent isotherm and uptake. The batch model described previously is applied successfully to describe multicomponent uptake, using the single- and multicomponent measurements as model inputs. The success of this model indicates that column behavior could be similarly predicted using the column model derived in the first part of this work. Moving to the molecular scale, novel confocal microscopy methods such as fluorescence recovery after photobleaching (FRAP) and fluorescence anisotropy are explored as means of rapidly measuring kinetic and diffusive contributions to protein mobility in chromatographic media. This study is especially motivated by the need for a better understanding of and improved models describing transport in polymer-modified and traditional ion-exchange media, especially for multicomponent separation. FRAP methods allowed estimation of kinetic desorption rates and intrinsic intraparticle pore diffusion rates of proteins in fully- and partially-loaded chromatographic media. The parameters estimated from these techniques were applied in an attempt to predict challenging multicomponent displacement profiles. Reasonable agreement between experiment and prediction was observed. In conclusion, each of these methods contributes to the fundamental understanding of and predictive capabilities for protein uptake, displacement and elution behavior. Most importantly, progress is achieved toward filling gaps in current methods of measurement and theoretical models for multicomponent protein separation.