UC Berkeley

To understand the function and dysfunction of cells in biological organisms, it is important to characterize one of the key functional actors of the cell, proteins. With indications of poor correlation between mRNA expression and proteomic expression at the single cell level, in vitro assays directly quantifying protein expression, in both the spatial and temporal context, are needed. To span the extensive cellular heterogeneity in gene and protein expression and activity observed in cells and tissues, proteomic assays should interrogate single- and low-cell resolution with sufficiently high throughput to identify cellular sub-populations. These proteomic assays would also require sufficiently high selectivity and analytical sensitivity with which to interrogate protein isoforms and post-translational modifications. To address this measurement gap, we introduce and further develop proteomic assays towards these specifications.

We enhanced the analytical sensitivity of the ultrathin isoelectric focusing assay (IEF) with subsequent immunoblot, by developing a highly-porous hydrogel matrix as a new substrate for the assay. We characterized the effect of this 10-fold increase in gel porosity on the IEF separation performance, paired with a reagent modification that directly impacts separation performance. Additionally, we assessed the benefits of the increased porosity on the in-gel immunoblot.

Furthermore, we investigated the compartmentalization of protein lysate from single cells within the microwells embedded in the hydrogel substrate in our proteomic assays. We characterized the height of the fluid film between multi-material interfaces. We used numerical modeling and experimental validation to assess the contribution of the fluid film to the diffusive losses that reduce analytical sensitivity in our assays.

We then re-imagined the ultrathin IEF assay for a 100-fold increase in throughput by developing 3D projection electrophoresis. We interrogated the IEF separation performance of this proof-of-concept high-throughput IEF assay with several optimizations. We conclude this section of this dissertation with an in-depth discussion of the potential further developments for this platform, towards a high-sensitivity, high-throughput proteomic assay with multiplexing capabilities.

In a parallel line of inquiry in this dissertation, to further understand and characterize cellular functions at a larger scale, in vitro biological models mimicking human physiology are needed. Due to inter-species differences in ion channels, biological pathways, and pharmacokinetic properties, animal models do not faithfully predict human cardiotoxicity. Human in vitro tissue models, with similar three-dimensional microenvironments to those found in in vitro human organs, that are predictive of human drug responses would be a significant advancement for understanding, studying, and developing new drugs and strategies for treating diseases. To assess the measurement needs in this space, we surveyed the breadth of in vitro cardiac devices mimicking human cardiac physiology.

The lipid storage and processing within adipose tissue strongly affects drug concentrations in vivo, and adipose tissue interacts with other organs via paracrine signals and fatty acid release, affecting the safety profiles of a large number of drug-like molecules. To address this measurement gap, we developed a microfluidic device with adipose tissue. We used numerical modeling and an analytical model to characterize the convective and diffusive transport within the device. We confirmed the maintenance of adipose cell viability and growth, extracellular matrix deposition, and adipose tissue functionality over two weeks.

We anticipate that the developments of analytical proteomic assays and in vitro biological models discussed in this dissertation will support quantitative characterizations of human biology, leading towards future development of targeted clinical therapies for improved length and quality of life.