Discovery of human-like L-asparaginases with potential clinical use by directed evolution.

L-asparaginase is a chemotherapy drug used to treat acute lymphoblastic leukemia (ALL). The main prerequisite for clinical efficacy of L-asparaginases is micromolar K(M) for asparagine to allow for complete depletion of this amino acid in the blood. Since currently approved L-asparaginases are of bacterial origin, immunogenicity is a challenge, which would be mitigated by a human enzyme. However, all human L-asparaginases have millimolar K(M) for asparagine. We recently identified the low K(M) guinea pig L-asparaginase (gpASNase1). Because gpASNase1 and human L-asparaginase 1 (hASNase1) share ~70% amino-acid identity, we decided to humanize gpASNase1 by generating chimeras with hASNase1 through DNA shuffling. To identify low K(M) chimeras we developed a suitable bacterial selection system (E. coli strain BW5Δ). Transforming BW5Δ with the shuffling libraries allowed for the identification of several low K(M) clones. To further humanize these clones, the C-terminal domain of gpASNase1 was replaced with that of hASNase1. Two of the identified clones, 63(N)-h(C) and 65(N)-h(C), share respectively 85.7% and 87.1% identity with the hASNase1 but have a K(M) similar to gpASNase1. These clones possess 100-140 fold enhanced catalytic efficiency compared to hASNase1. Notably, we also show that these highly human-like L-asparaginases maintain their in vitro ALL killing potential.