Pulmonary infections cause a greater global burden of disease than any other category including HIV/AIDS, cancer, heart attacks, and malaria. Even in the United States mortality rates from lung infections have failed to decline appreciably since the 1950's. A critical factor in the lack of progress treating lung infections is the accelerating emergence of antibiotic resistance in pathogenic bacteria. This reality is driving a medical and scientific imperative to design robust next generation antimicrobial therapies acting by mechanisms orthogonal to conventional inhibitory drugs. A key component of innate immunity in higher organisms is the biocatalyst lysozyme, which exerts antimicrobial activity via enzymatic hydrolysis of cell wall peptidoglycan as well as exhibiting a lesser known non-enzymatic mechanism towards both Gram-positive and Gram-negative strains. Inhibitory antibiotics act in a stoichiometric fashion, i.e. each therapeutic molecule typically acts on only one target molecule. In contrast, a single lysozyme protein has the capacity to hydrolyze thousands of glycosidic bonds in a short time. The catalytic nature of lysozyme's antimicrobial activity defines an entirely different class of antibacterial agent with the capacity for superior efficacy. However, native human lysozyme is unlikely to represent an optimized biotherapeutic for severe pulmonary infections as evidenced by the endogenous enzyme's failure to effectively protect the lung in these cases. Lysozyme's dysfunction in both acute and chronic lung infections indicates an inherent limitation under conditions far from stasis. Evidence suggests that this functional limitation derives in large part from the hyperinflammatory environment of the infected lung, which leads to extraordinarily high local concentrations of anionic polyelectrolytes such as extracellular DNA, F-actin, mucins, and in some cases alginate, a matrix component of some bacterial biofilms. We have shown that human lysozyme's cationic character drives electrostatic assembly of large aggregates with biological polyanions at physiological concentrations, ultimately resulting in inactivation of the enzyme. These observations led us to hypothesize that careful remodeling of human lysozyme's electrostatic potential could produce charge engineered antimicrobial proteins capable of evading electrostatic aggregation while maintaining high levels of lytic activity. To implement a combinatorial protein engineering strategy, structural and bioinformatics analysis were used to guide design of two saturation libraries containing 65,536 unique proteins each. To isolate therapeutic candidates from the library populations, we developed a customized high throughput functional screen that mimics inhibitory aspects of the infected lung environment. Iterative screening and recombination of isolated clones yielded more than twenty charge-modified lysozymes with enhanced performance characteristics. The bactericidal activities of the wild type and variant enzymes were characterized, and the best variants were shown to efficiently kill bacteria under physiologically relevant conditions that completely abolished wild type hLYS activity. In contrast to the native enzyme, the engineered variants avoided electrostatic aggregation in solution, and exhibited enhanced diffusivity in hydrogels containing anionic biopolymers. The impressive in vitro results have prompted in vivo efficacy studies using an established model of lung infection. The outcome of these preclinical studies may motivate further development of clinically useful, enzymatic, antimicrobial therapies.

Extended Abstract Status: Not Uploaded