Doctoral Dissertation Defense – Jasmine Shirazi

BME PhD Candidate, Jasmine Shirazi, will be defending her dissertation.


Engineering Strategies for Quantifying Biological Phenomena

When:   Friday October 29 @ 2 PM EST
VIRTUAL:     Password: 117441
Committee Chair:   Jason Gleghorn, PhD
Committee:  Christopher Price, PhD; Eric Wommack, PhD; Ryan Zurakowski, PhD


Rigorously quantifying biological phenomena remains a challenge across disciplines. Different methods offer varying levels of resolution, throughput, and user effort and expertise. Our goal was to develop platforms for quantification of a range of biological entities that are accessible to users with no engineering background. In this work, we demonstrated three methods for quantification of biological objects: a platform for 3D imaging-based enumeration and size determination of suspended nano-scale objects, a platform for 2D imaging-based enumeration of cultured and environmentally isolated viruses, and a platform for storing, manipulating, and analyzing microfluidic droplets containing biological samples.

In Aim 1, we developed a fluorescent imaging-based method for enumerating and determining the size of suspended nano-scale particles based on Brownian motion. Existing methods for nanoparticle enumeration and characterization often do not provide the resolution needed to discern heterogeneous samples, or they rely on specialized equipment or engineering expertise. Whereas other fluorescent imaging-based approaches often rely on confocal microscopy or specialized custom imaging equipment, our method is compatible with standard widefield epifluorescent microscopy and requires inexpensive materials, making it accessible to more users. A simple device called the NanoVis consisting of two coverslips and double-sided tape can be created on demand and loaded with as little as 5 µL of suspended sample. For particle enumeration, samples were imaged across multiple z-stacks, and a custom image analysis algorithm was used to segment the resulting images and determine the count by averaging across the imaging fields. For size information, time lapses of single visual fields were taken, and each object was tracked temporally to determine its path length over a given time. The resulting mean square displacement was used to calculate the diffusion coefficient and particle size based on the Stokes-Einstein equation. This platform was validated using fluorescent nano-scale polystyrene beads. Enumeration and particle size data were compared to an established nanoparticle tracking analysis (NTA) platform called the NanoSight. As a proof-of-concept, we analyzed extracellular vesicles (EVs) isolated from pancreatic cancer cells and compared data from the NanoVis to the NanoSight and scanning electron microscopy (SEM). We found that the NanoVis effectively captured the size heterogeneity of the sample, whereas the NanoSight only captured particles that fell into a narrow size range.

In Aim 2, we established a platform for enumeration of cultured virus and environmental viral isolates. Viruses that infect microbes play a vital ecological role with profound effects on microbial diversity and biogeochemical cycles. The presence and infective activity of viruses has cycles. Common methods for counting viruses include the plaque assays, in which cultured hosts are infected with viruses and the resulting infectious units or “plaques” are counted, and quantitative polymerase chain reaction (qPCR), in which the presence of genetic sequences specific to a particular virus are determined. However, these approaches cannot be used on heterogeneous virus samples that have been isolated from the environment. As such the gold standard for viral isolates involves imaging of fluorescently stained viruses that have been adhered onto a ceramic filter (Anodisc). Using the Anodisc is costly in terms of both time and money, and we have developed an alternative called the Virometer which consists of surface-treated coverslips held together with double-sided tape. Stained viruses are loaded in between two coverslips and incubated so that they can adhere to the charged glass for 2D fluorescent imaging. We used T7, a well-characterized cultured bacteriophage that infects Escherichia coli (E. coli), to validate the performance of the Virometer against established methods. Digital PCR (dPCR) was chosen as a benchmark for the cultured T7 virus, and the counts obtained using the Virometer closely matched those obtained via dPCR. We then evaluated aquatic environmental isolates using the Virometer and found the counts to be comparable to the Anodisc, establishing the Virometer as a viable, cost-effective alternative to the Anodisc.

In Aim 3, we demonstrate a droplet-based liquid handling system-on-a-chip that is accessible to users without microfluidics experience. By encapsulating cells in thousands of microfluidic droplets, we are able create thousands of experimental vessels or replicates that can fit in a single dish. Our platform then stores the resulting droplets in a thin membrane array to be imaged over time, allowing for repeated measures for the same droplet and greater information density per sample. Each array consists of hundreds of micro-scale holes that serve as wells that can each fit a single droplet. Droplets are loaded by pipetting the droplet solution across the array and allowing the droplets to populate the wells through capillary action. This open-faced platform also offers experimental flexibility as the user has the option to deposit a second array of droplets containing other cells, viruses, or reagents on top of the initial population. These stacked droplet populations can then be merged via electrocoalescence, allowing the two components to interact. Loading efficiency of the array was evaluated using droplets containing colored dyes. As a proof-of-concept, we encapsulated E. coli in droplets and loaded them in our array. GFP-expressing E. coli were used to demonstrate that individual cells could be resolved via epifluorescent imaging and counted in this array. To demonstrate merging functionality, we added droplets containing a lipophilic membrane dye (PKH26) and merged them with the bacteria containing droplets. This platform can be used to study virus-host interactions by introducing viruses into the system and evaluating infection dynamics based on counts of stained bacteria over time.