UC Riverside

Dysregulation of proteases has been implicated in a variety of diseases such as cancer, inflammation, osteoporosis, neuropathic pain, and neurodegenerative diseases. Although several compound inhibitors of proteases have been approved by the FDA, many small molecule inhibitors of proteases failed in clinical trials due to severe side effects caused by non-specificity. Monoclonal antibodies thus hold a great promise as therapeutics able to inhibit pathogenic proteases with desired selectivity. However, proteases present a unique challenge for inhibitory antibodies discovery largely because of lack of a function based selection method.

This study has developed four inhibitory antibody functional high-throughput selection/screening methods: fast discovery by deep sequencing, epitope specific affinity maturation, conversion of inhibition selectivity by competitive FACS, and in vivo inhibition based genetic selection. Using these novel approaches, large panels of antibodies were discovered / engineered inhibiting a variety of proteases including matrix-metalloprotease-14 (MMP-14), MMP-9, autophagic serine protease (Alp2), aspartic acid protease β-secretase 1 (BACE1), and cysteine protease cathepsin B (CTSB). Biochemical characterizations by biolayer interferometry and inhibition kinetics suggested that isolated antibodies exhibited high binding affinities and high inhibition potencies both in nM range with decent selectivity and proteolytic stability. The isolated antibodies blocked their protease targets from hydrolyzing the associated physiological substrates either competitively or non-competitively. Competitive ELISA with native inhibitors and epitope/paratope mutation studies indicated that most generated antibodies achieved their inhibitory functions by recognizing protease’s catalytic clefts using their complementarity-determining region (CDR) H3. In a mouse model of drug induced neuropathic pain, intravenous application of IgG L13, a potent anti-MMP9 inhibitor, exhibited significant pain attenuation effects. And anti-Alp2 Fab 1 specifically stained Aspergillus fumigatus infected mouse lung tissues, demonstrating its potential in diagnosis and as a therapeutic. Collectively, these functional selection/screening methods developed in this study greatly facilitate the discovery and engineering of inhibitory antibodies, and can be applied for many proteases of clinical significance.