Cameron Yamanishi

BME Ph.D. Defense Presentation

Date: Tuesday, August 27, 2019

Time: 1:30 pm, Atlanta time

Location: EBB CHOA Seminar Room

Advisor: Shuichi Takayama, PhD (GaTech BME)

Committee Members:

Rabindra Tirouvanziam, PhD (Emory Pediatrics)

Hang Lu, PhD (GaTech ChBE)

Shu Jia, PhD (GaTech BME)

Louise Hecker, PhD (University of Arizona, Medicine)

Title: Aqueous two-phase system micro-assays augmented by automated image analysis

Abstract:

Aqueous two-phase systems (ATPS) are a quirk of thermodynamics, wherein two immiscible solutions separate into stable liquid phases, containing primarily water. Conveniently, some biomolecules partition favorably into one phase or the other. Recently, these properties have been used for many applications, including micropatterning, purification, target concentration, and reagent segregation.

In this work, we examined a previously under-appreciated property of ATPSs – the spontaneous generation of circulating fluid flow within ATPS micro-droplets. First, we developed and validated a novel imaging modality, stigmatic microscopy, to perform the first 3D measurements of ATPS micro-droplet self-driven circulation. This technique borrows concepts from super-resolution microscopy to identify the 3D positions of fluorescent microbeads from individual image frames. Specifically, the use of a toroidal lens shifts the x-focal plane apart from the y-focal plane, enabling determination of z-position from the point-spread function of the image. We designed the microscope using optical modeling to determine usable lens strengths and distances. After constructing the microscope, we developed software to determine 3D positions of fluorescent micro-beads and link the positions from time frame to time frame to track flow.

Second, we enhanced the performance of an ATPS multiplex immunoassay. In previous work from our lab, antibody pairs were spotted in ATPS micro-droplets to co-localize them, preventing crosstalk between incorrect antibody pairs. We advanced this assay from a manually spotted method to a pre-dried format, wherein one phase of the ATPS is rehydrated by the other. We demonstrated that antibody segregation is maintained throughout the rehydration process. With further consideration of self-driven flow, we reduced assay incubation time drastically, first in a two-wash format and subsequently in a one-wash format. Further examination of the competing ATPS parameters identified an optimal medium between viscosity, partition behavior, and convective flow.

Third, we extended another ATPS bioassay from our lab - collagen microgel contraction to facilitate research on pulmonary fibrosis, a deadly disease with no effective treatments. The collagen gel contraction assay is a helpful measurement of the wound healing activity of fibroblasts. However, current formats use large numbers of cells, restricting their use with primary lung fibroblast cells, which have limited growth capacity. To address the issue, our lab previously used ATPS to generate microscale bioprinted collagen gels and demonstrated proof-of-concept with cell lines. However, the behavior of primary cells proved more difficult to study, due to small effect sizes. To overcome this obstacle, we incorporated higher throughput and used continuous imaging (as opposed to end point assays) with automated image processing. This yielded a markedly more reliable assay, which we used to observe the effects of current and potential therapeutics. Notably, we identify differences between normal and diseased fibroblasts in their contraction kinetics at moderate doses of anti-fibrotic drugs.

Bluejeans link: https://bluejeans.com/5733202109/