PhD Thesis Defense: Yongliang Fang

Wednesday, March 20, 2019, 2:00–4:00pm

Rm. MB01 (Zaleski Auditorium), MacLean ESC

“Innovative high-throughput screening technologies enable the discovery and optimization of therapeutic proteins”


Therapeutic proteins have revolutionized the pharmaceutical industry and clinical treatment options in many disease areas. Computational guided directed evolution is a powerful strategy for engineering therapeutic proteins with desirable properties and activities, but this approach typically relies on large combinatorial gene libraries that may encode millions or billions of unique protein variants.  Because only a few library members likely exhibit the desired properties, compatible assay methods are required to functionally screen the protein libraries in a high throughput manner.

This thesis describes the development and implementation of flow cytometric screening technologies that enable the discovery and isolation of improved therapeutic proteins from large combinatorial libraries. First, libraries of human lysozyme (hLYZ) were computationally designed in an effort to isolate variants that evade a pathogen-derived inhibitory molecule: Escherichia coli inhibitor of vertebrate lysozyme (Ivyc). A novel gel microdroplet based fluorescence-activated cell sorting (GMD-FACS) methodology was developed to screen millions of library members and ultimately isolate hLYZ variants able to kill bacteria in the presence of Ivyc. Second, the GMD-FACS technology was modified to develop a new high throughput screening platform for antibody engineering, enabling the efficient analysis of secreted full-length Immunoglobulin G (IgG) antibodies against integral membrane protein targets in their native contexts. The screening of a mock antibodies library demonstrated that GMD-FACS enables analysis and sorting of antibody libraries at rates of several thousand antibodies per second, and these results suggest that the method could be useful for the discovery and engineering of therapeutic antibodies against challenging molecular targets. Third, a large combinatorial library of a therapeutic toxin was computationally designed and constructed in an effort to reduce the toxin’s inherent immunogenicity, thereby improving safety and efficacy. An innovative fluorescence resonance energy transfer (FRET) based FACS technology was developed and used to screen a 200-million member library, and active toxin variants with lower immunogenicity were identified, isolated, and characterized.

In conclusion, the methodologies enabled screening of large recombinant libraries for properties and functions that are not readily addressable by conventional screening platforms.  Such innovative screens have the potential to accelerate biotherapeutic discovery and provide solutions to unmet medical needs.

Thesis Committee

For more information, contact Daryl Laware at