(4jc) Postdoc Candidate: Quantifying Diffusive Mass Transport in Aqueous Two-Phase Systems for Vaccine Manufacturing

In 2018, vaccine preventable diseases led to the death of 700,000 children under the age of five years, almost entirely in low- or middle-income nations. One contributor is the difficulty of maintaining the cold chain, which is required to maintain the activity of most vaccines, in remote or low-income areas of the world. Another contributor is economic strain. Vaccines are even more challenging to manufacture than more common protein therapeutics, putting them out of reach of the communities that most need them. My goal is to decrease that financial burden by rethinking the downstream process accounting for the majority of manufacturing costs.

Many viral vaccines today are made of whole live or inactivated viruses. Their size and complexity mean they are highly sensitive to shear stress, temperature, and pH, making manufacturing particularly challenging. Traditional unit operations, like chromatography, struggle to adapt to whole-particle vaccines due to their size. Others are simply not scalable, like density gradient ultracentrifugation. However, aqueous two-phase systems (ATPS) create a stabilizing environment for biomolecules and adapt easily to continuous processing. ATPS are formed when sufficient concentrations of two polymers or a polymer and a salt are mixed with water and spontaneously separate to form two phases differing in charge and hydrophobicity. Since viral particles tend to be more hydrophobic than proteins, they tend to partition to one phase while contaminating proteins and DNA are removed in the other. I have co-authored work demonstrating consistent recoveries greater than 80% of infectious porcine parvovirus. Since then, we have extended that work to show similar yields for a panel of other vaccine models. We have also demonstrated a continuous ATPS process for vaccine manufacturing and potential for 90% reduction in capital costs and similar significant improvements in operating expenses compared with a conventional chromatography-based process. We believe ATPS could be the first platformable method to produce vaccines using a fully integrated, continuous process.

Right now, the biggest challenge to this reality is a limited knowledge of mass transport available to scale up a mixer-settler process for industrial production. We have discovered in the lab that insufficient mixing dramatically reduces virus yields in continuous ATPS, and that creating a process to efficiently mix and separate the phases is challenging even at the benchtop scale. Clearly, a tool to quantify and compare the mass transport of viral products and their primary protein contaminants is needed. To do this, I designed a microfluidic experiment that uses fluorescence microscopy to characterize mass diffusion in ATPS in real time. As the first to use microfluidics in my lab, I purchased equipment, learned to construct microdevices, and developed Python codes to compare concentration profiles collected in my experiments with simulated results using computational fluid dynamics solved by COMSOL multiphysics. One of the most interesting results was the strong aggregation of bovine serum albumin, the primary contaminant protein in our crude virus stocks, in the more concentrated ATPS that provide the best virus purification and recovery. This indicates that aggregation may play a significant role in the removal of proteins by precipitation from the polymer-rich phase and even accumulation at the two-phase interface while not affecting more dilute viral particles. I have also shown how real mass diffusion coefficients are significantly different than predicted by the Stokes-Einstein and Polson models often used to approximate transport behavior. These findings underscore the need to quantify these parameters experimentally and how using inexpensive microfluidic techniques can quantify diffusion coefficients.

During this time, as a contributor to other projects I have also become proficient at liquid-phase imaging of virus particles using an atomic force microscope and I am a certified user of Michigan Technological University’s Titan Themis transmission electron microscope, both tools we use to investigate the size, morphology, and surface chemistry of viruses produced in our lab. In addition to pursuing my own projects in concert with my GRFP award, the trajectory of my PhD was also affected by COVID-19. Michigan Technological University exists in a very rural area of Michigan’s upper peninsula and was asked to organize a testing center during the early days of the pandemic. Because of my training and work with viruses, I was fortunate to help establish a clinical testing facility and train technicians on RNA extraction and biosafety protocols. I have also co-organized a journal club for Michigan Tech’s Health Research Institute and taught three courses as instructor of record during my graduate studies. These tremendous opportunities have developed my leadership skills and prepared me for an academic career in ways a typical graduate career could not.

My goal after a postdoc is an academic tenure-track position where I can develop the next generation of scientists both in the lab and in the classroom. I have interests in transport and interfacial phenomena and would be interested in incorporating computational work or machine learning into my postdoc experience. I’m a detail-oriented person who really likes to understand things and study mechanism and methodology.