Subtilisin disulfide bonds

Yang, J. and Kramer, J.M. (1994) In vitro mutagenesis of Caenorhabditis elegans cuticle collagens identifies a potential subtilisin-like protease cleavage site and demonstrates that carboxyl domain disulfide bonding is required for normal function but not assembly. Molecular and Cellular Biology 14, 2722-2730. 

Mitchinson, C. and Wells, J.A. (1989) Protein engineering of disulfide bonds in subtilisin BPN. Biochemistry, 28, 4807-4815. 

Since there are strict stereochemical requirements for the relative positions and orientations of the two participating cysteine residues,11 addition of new disulfides to existing proteins by site-directed mutagenesis has not always produced the desired increase in stability. Introduction of disulfide bonds has been attempted for phage T4 lysozyme,4-71 phage A repressor,81 dihydrofolate reductase,91 and subtilisins.10-131 Among them the most extensive study has been performed on T4 lysozyme, and enhancement of protein stability has been successful. 

As for subtilisin BPN, the first attempt at molecular modeling of disulfide mutants was performed with computer graphics using coordinates from the crystal structure of the enzyme,10-121 but increasing enzyme stability was unsuccessful. We have studied a thermostable subtilisin-type protease, aqualysin I, and the introduction site of a disulfide bond was chosen on structural homology between aqualysin I and subtilisin E. Here we describe a successful study to increase the stability of subtilisin E and others done for subtilisin enzymes. 

At present, 16 cysteine-containing subtilisin-type enzymes are known and the position of the cysteine residues is restricted to the nine corresponding sites described above.42 Of the 16 enzymes, six enzymes other than aqualysin I and proteinase K have cysteine residues at positions where the cysteine residues are able to form disulfide bond(s) like the two enzymes. Although these disulfide bonds seem to have been acquired to increase protein stability, only four kinds of disulfide bonds are found in the subtilisin-type enzymes, suggesting that the positions of the disulfide bonds have been selected strictly in the process of molecular evolution of the enzyme. 

Introduction of a Disulfide Bond into Subtilisin by Site-directed Mutagenesis... 

As for aqualysin I, the disulfide bond seems to be one of the causes of its thermostability, because the enzyme is unstable in the presence of 2-mercaptoethanol.15) Accordingly, increasing the stability of subtilisin E was attempted by engineering disulfide bonds at the sites corresponding to those of aqualysin I. 

Mutant Subtilisin E Having a Cys61-Cys98 Disulfide Bond Gains Thermostability... 

Thermostability of subtilisin E is increased by the introduction of the disulfide bond (Table 12.3). The half-life of the disulfide mutant is 2-3 times longer than that of the wild-type enzyme at 45°-60°C. On the other hand, those of the single-cysteine mutants are... 

Table 12.3 Effect of the disulfide bond introduced in subtilisin E on the...

Identity of amino acid sequences between subtilisins E and BPN is 86%, so three-dimensional structures of the two enzymes are considered to be very similar. In the case of subtilisin BPN, residues 61 and 98 are located on the loop and turn structure, respectively, both of which connect /3-strand and a-helix (Fig. 12.5). Solvent exposures of the residues are both 9,45) indicating their presence on the surface of the enzyme molecule. The distance between the a-carbons of the two residues is 5.8 A. Accordingly, the positions seem appropriate for cysteine residues to form a disulfide bond without any strain in the enzyme structure. The disulfide bond formed is located close to the active site so as to stabilize the wall of the active-site pocket (Fig. 12.5). 

Fig. 12.5 Position of Cys61-Cys98 disulfide bond introduced into subtilisin E. Ribbon diagram of subtilisin51 is shown with the positions of substituted cysteines and active-site residues, Asp32, His64, and Ser221. The disulfide bond is represented by a dark line. ...

On the basis of sequence alignment, subtilisin-type proteases can be subdivided into class I and class II.42 Subtilisins, thermitase and others, none of which has a disulfide bond, belong to class I, and ten proteases including aqualysin I and proteinase K to class II. An alkaline protease from Aspergillus oryzae, which has no cysteine residue, belongs to class II. The sequence identity between aqualysin I and the alkaline protease is 44%.49 ... 

Figure 17. The effect of disulfide bond modification of turkey ovomucoid by alkali on inhibitory activity against trypsin (T), a-chymotrypsin, (C), and subtilisin (S). Turkey ovomucoid (O.lOmM) was treated with alkali (lOOmM NaOH) at 23°C. Sulfhydryl content (moles per mole of protein)...

B Elimination has been used to show differences among the disulfide bonds in various proteins including the ovomucoids (54). The e-elimination reaction has also been used to replace the hydroxyl group of the essential serine residue of subtilisin with a sulfhydryl group (55). The thiolsubtilisin had a small fraction of the activity of subtilisin but it has been quite useful in mechanistic studies of the serine and sulfhydryl proteases. 

Knight, E., Jr. Fathey, D. Human fibroblast interferon an improved purification. J. Biol. Chem. 1981,256, 3609-3611. Wells, J.A. Powers, D.B. In vivo formation and stability of engineered disulfide bonds in subtilisin. J. Biol. Chem. 1986, 261, 6564-6570. 

The combination of molecular modeling with genetic engineering to enhance protein stability has been successful in certain cases. For instance, introducing carefully sited novel disulfide bonds increased protein stability in T4 lysozyme (11-13) and in X-repressor (14). However, the results in other proteins, for instance, in subtilisin (15,16) and in dihydrofolate reductase (17) have been less predictable. 

For ribonuclease, removal by pepsin proteolysis of only a tetrapeptide sequence at the C-terminus (Asp-Ala-Ser-Val) lead to an inactive enzyme and unstable structure (Anfinsen 1956 Taniuchi, 1970). When this shortened enzyme, the so-called des(121-124) ribonuclease or pepsin inactivated ribonuclease (PIR), is reduced, it cannot reoxidize to 3deld the native pairing of disulfide bonds (Taniuchi, 1970). Removal of six C-terminal residues to form RNase 1-118 also yields a structureless and inactive enzyme (Lin, 1970 Andria and Taniuchi, 1978). The importance of the C-terminal end to the folding of these two proteins was also emphasized by the data reported in Chapter 9. The information contained in the C-terminal sequence of the two nucleases appears to be crucial for their refolding. It was proposed that the polypeptide chain of nuclease and ribonuclease cannot achieve the native structure, during biosynthesis, until the termination of the polypeptide chain. For RNase even the N-terminal end seems to be important for folding. After removal by proteolytic action of subtilisin of the first 20 amino adds (Richards, 1958), the RNase S protein is unstable and cannot refold correctly when disulfide bridges are reduced. 

---