Proteases substrate engineering

Engineering Substrate Specificity. Although the serine proteases use a common catalytic mechanism, the enzymes have a wide variety of substrate specificities. For example, the natural variant subtiHsins of B. amyloliquefaciens (subtiHsin BPN J and B. licheniformis (subtiHsin Carlsberg) possess very similar stmctures and sequences where 86 of 275 amino acids are identical, but have different catalytic efficiencies, toward tetraamino acid -nitroanilide substrates (67). 

Teplyakov, A. V., van der Laan, J. M., Lammers, A. A., Kelders, H., Kalk, K. H., Misset, O., Mulleners, L.J. Dijkstra, B. W. (1992). Protein engineering of the high-alkaline serine protease PB92 from Bacillus alcalophilus functional and structural consequences of mutation at the S4 substrate binding pocket. Protein Engineering, 5, 413-20. 

The protein superfamily of proteases [78, 79], however, is an ideal framework for a directed privileged structure-based masterkey concept. It has already been reported that the 5,5-trans-fused lactam moiety was systematically optimized and explored as a serine protease-directed scaffold by GlaxoSmithKline and has delivered progressible lead compounds for various members of that target class [3], such as thrombin [80, 81], elastase [82, 83], HCMV protease [84, 85], and the hepatitis C virus-encoded NS3-4A protease [86, 87]. Here, the initially identified scaffold was engineered toward the serine protease-wide commonality in substrate binding and processing [3],... 

Engineering members of the serine protease family which structurally consist of two homologous, stacked /1-barrels, Hopfner and coworkers designed a hybrid protease by fusing the N-terminal /1-barrel from trypsin to the C-terminal /1-barrel of factor Xa [74], The location of the fusion point in the linker region between the two /1-barrels was chosen following close inspection of the X-ray structures of trypsin and factor Xa. The resulting hybrid was shown to be fully functional as it hydrolyzed a broader but distinct spectrum of peptide substrates in comparison to the parental enzymes. 

The putative structure of the BAR protease can be inferred from its sequence homology to chymosin (see Fig. 1.5 b Protein Data Bank entry ICMS). When the mutations that were found in the final engineered BAR variant are mapped onto this structure, approximately 50% of the mutations lie in or close to the substrate binding site of the protease, while the other 50% are found at distant sites in the protein. This distribution confirms that directed evolution processes efficiently generate mutations at relevant positions. 

---