Open Access Dissertation
Date of Award
Proteomic workflows rely on sensitive and precise methods for purifying and detecting proteins, and one of the most popular methods is liquid chromatography (LC) followed by an inline detector. However, the most commonly used method, UV-Vis absorbance, provides relatively low sensitivity and requires long path lengths in the flow cell. In this thesis, we present a highly sensitive, label-free method to detect proteins in continuous flow, using a pressure-driven microfluidic droplet generator. The system consists of a cross junction or flow focusing structure where the flowing stream of proteins is combined with two coflowing streams of oil which break up the former stream into a train of droplets. In such a system, the droplet size predictably scales with interfacial tension (IFT) of the oil-water interface. When globular proteins (eg BSA) enter the junction, they adsorb to the interface and reduce IFT, and thereby the droplet radius. Temporal variations in droplet size therefore correlate with changes in protein concentration. Using microfluidic drop generators as an inline detection method provides two benefits: 1) it can improve available limits of detection, because the high surface area to volume ratio enhances adsorption phenomena, and 2) it provides fast time resolution, due to the high rate of drop generation. Four different carrier phases (oleic acid, octanol, hexadecane, and FC-40) were tested with four proteins (BSA, thyroglobulin, uridine, placental growth factor, galectin and blood plasma).
Droplet sizes were found to decrease when a hydrophobic protein was combined with a hydrophobic carrier. The best limit of detection is ~1 μg/mL in a 1 nL droplet, which equates to 1 fg of total protein. The low detection limits are due to favorable scaling: the high surface area to volume ratio in droplet systems increases the probability of adsorption, and therefore changes drop size even at low concentration. When used as a detector for high performance liquid chromatography (HPLC), it demonstrates sensitivity up to 100X better than conventional UV-Vis detectors. This detection method can potentially serve as a label-free, universal detector for proteins.
Kebriaei, Razieh, "Sub-Picogram, Inline Detection Of Proteins Using Microfluidic Drop Generators And Shape-Based Detection" (2016). Wayne State University Dissertations. 1645.