Stabilization of substilisin E in organic solvents by site-directed mutagenesis

Stabilization of Subtilisin E in Organic Solvents by Site-Directed Mutagenesis

Pascal Martinez, Mariana E. Van Dam, Amy C. Robinson, Keqin Chen, and Frances H. Arnold” Division of Chemistry and Chemical Engineering 210-41, California Institute of Technology, Pasadena, California 91 125

Received April 26, 1991lAccepted June 14, 1991

Subtilisin E was rationally engineered to improve its stability in polar organic solvents such as dimethylformamide (DMF). A charged surface residue, Asp248, was substituted by three amino acids of increasing hydrophobicity, Asn, Ala, and Leu; all three variants were stabilized with respect to wild type in 80% DMF. This stabilization was only observed in the presence of high concentrations of the organic solvent: no stability enhancements were observed in 40% DMF. In contrast, the mutation Asn218 -+ Ser alters internal hydrogen bonding interactions and stabilizes subtilisin E in both 40% and 80% DMF. This study provides additional evidence that substitution of surface-charged residues is a generally useful mechanism for stabilizing enzymes in organic media and that the stabilizing effects of such substitutions are unique to highly altered solvent environments. The effects of the single amino acid substitutions on free energies of stabilization are additive in the Asp248 4 Asn + Asn218 -+ Ser combination variant, yielding an enzyme that is 3.4 times more stable than wild type in 80% DMF. Key words: protein engineering - enzymes in organic solvents - protein stabilization - subtilisin E

INTRODUCTION

Although there have been numerous reports of enzyme activity in nonpolar organic solvents such as octane,’ practical applications of biocatalysis in organic solvents are still somewhat limited. One reason is that substrates of interest are often poorly soluble in nonpolar media. Polar solvents such as acetone or dimethylformamide (DMF) provide a better reaction medium for many synthetic applications; unfortunately, these solvents are often strong denaturants and quickly lead to catalyst deactivation. The deleterious effects of polar organic solvents on enzyme structure and catalytic activity re- flect the solvent’s impact on the noncovalent interactions that determine the stability of a folded protein and stabilization of reaction transition states. A large set of competing interactions attain a new equilibrium when a protein is transferred to organic solvent, and that new balance can favor structural changes that lead to deactivation and denaturation.

Because the free energy by which a folded protein is stabilized with respect to unfolded form(s) is small (5-20 kcal mol-I) and corresponds to a few individual

* To whom all correspondence should be addressed.

interactions, minimal alterations in amino acid sequence can have dramatic effects on protein stability. Thus it is possible to compensate for some of the unfavorable effects of a nonaqueous solvent by engineering the protein’s amino acid sequence. General features of proteins that are stable in nonaqueous solvents have been dis- cussed, and the types of amino acid substitutions likely to improve stability have been proposed in the form of “design rules” for engineering nonaqueous solvent-stable

This laboratory has embarked on a program to validate and extend these design rules through protein engineering st~dies.’,’~ It is hoped this work will eventually allow the rational engineering of natural or de novo enzymes for biocatalysis in organic media.

For these studies we have chosen to work with the serine protease subtilisin E from Bacillus subtilis. 27 Sub- tilisin E consists of a single polypeptide chain without disulfide bridges. It is expressed as a pre-pro enzyme, with a pre sequence of 29 residues responsible for secre- tion, a pro sequence of 77 residues needed for proper folding of subtilisin, and a mature enzyme of 275 amino acids. Once exported and correctly folded, the pre and pro portions are cleaved by an autocatalytic reaction. Without the pro sequence, the unfolding of subtilisin E is irreversible. The X-ray crystallographic structure is known to high resolution for subtilisin BPN’ from Bacil- lus arnyloliq~efaciens’~ and can be used to model subtilisin E because of the high structural homology between the two enzymes.

Numerous protein engineering studies have been carried out on subtilisin to alter a variety of properties, including catalytic a~tivity,’~,’’ substrate specifi~ity,~’.’ stability to therma16x2’,30 and ~xida t ive” .~’ inactivation, pH-activity and metal ion binding.” Demonstrations of subtilisin’s utility in organic solvents have included peptide s y n t h e ~ i s ~ ~ ’ ~ ~ and regioselective transe~terification.’~

Although subtilisin exhibits some activity in a wide variety of solvents, its stability is dramatically reduced in polar organic solvents such as DMF. In this article we report on the results of site-directed mutagenesis studies designed to test two proposed mechanisms for stabilizing enzymes in organic solvents: (1) substitution of selected charged surface residues by uncharged amino

Biotechnology and Bioengineering, Vol. 39, Pp. 141-147 (1992) 0 1992 John Wiley & Sons, Inc. CCC 0006-3592/92/020141-07$04.00

acids and (2) improvement of internal (protein-protein) hydrogen bonds. To test the first hypothesis, Asp248 on the surface of subtilisin E was replaced by three amino acids of increasing hydrophobicity: Asn, Ala, and Leu. To test the second mechanism, the mutation Asn218 +- Ser was introduced. This mutation was discovered by random mutagenesis and screening to improve the thermostability of subtilisin BPN’6 and has been shown to increase the stability of subtilisin E in 40% DMF.‘

Structural perturbations that result from amino acid substitutions are often accomodated with only localized changes in the protein and the solvent.’ Consequently, the stabilizing effects of conservative changes in the amino acid sequence in spatially distant parts of the protein can be, at least to some extent, cumulative. Several investigators have demonstrated that mutations which improve enzyme stability can be combined to obtain a variant more stable than any single mutant.13.20.21,23 To test whether single mutations can be combined to yield cumulative effects on enzyme stability in nonaqueous solvents, the double mutant Asp248 --f Asn + Asn218 -+ Ser was constructed and characterized.

A protein engineering strategy for stabilizing enzymes in organic solvents must be carried out so as to maintain, or even improve, catalytic activity. Therefore, activities for the hydrolysis of amide substrates and pH profiles of these activities were also investigated. Amide hydrolysis by serine proteases is extremely sensitive to changes in solvent environment and enzyme structure4 and thus is a sensitive measure of the effects of amino acid substitutions on catalysis.

MATERIALS AND METHODS

All restriction enzymes and T4 DNA ligase were purchased from Boehringer Mannheim and used according to the manufacturer’s specifications. The DMF (HPLC grade) and diethanolamine were obtained from Aldrich. The peptide substrate succinyl-Ala-Ala-Pro- Phe-p-nitroanilide (sAAPF-pna) and all other chemicals were purchased from Sigma.

Site-Directed Mutagenesis

The wild-type subtilisin E gene used for the creation of the variants was kindly provided by Professor M. Inouye (Rutgers University) in plasmid pHI212 that allowed expression in Escherichiu coli. l8 The unique HindIII- BurnHI DNA fragment of 789 basepairs, covering the amino acid sequence from Ser49 to the C-terminus of the mature subtilisin E gene, was subcloned into M13 mp19. The following four deoxyoligonucleotides, corresponding to the specific mutations Asp248 -+ Asn, Ala, Leu and Asn218 + Ser were used for site-directed mutagenesis according to the in vitro site-directed mutagenesis kit from Amersham: 5’-TTTCTAAACGGTT-

ACGGACTTGC3’, 5’-CTAAACGAGCACGGACT3’, 5’-CTAAACGAAGACGGACT3 ’ , 5 ’ -CGCTTATAGC- GGAACGT3’, respectively. Mutants in E. coli TG1 were screened by DNA sequencing according to the T7 Sequencing kit from Pharmacia LKB. HindIII-BurnHI fragments were subcloned into pHI212, which was used to transform E. coli JA221. The unique HindIII-NcoI fragment of the Asp248 -+ Asn subtilisin E gene was replaced by the corresponding fragment of the Asn218 +

Ser gene to obtain the double mutant Asp248 -+ Asn + Asn218 --f Ser. All variants were sequenced a second time to confirm the mutations.

Plasmid pKWC was used for expression of subtilisin E in B. subtilis. This plasmid was derived from the plasmid pKWZ (provided by Professor R. Doi, University of California, Davis) by eliminating a SulI-EcoRI fragment which contains a HindIII site from the upstream region of the subtilisin gene. The HindIII-BurnHI DNA fragments of wild-type and mutated subtilisin genes were subcloned into pKWC which had been digested with HindIII and BurnHI and purified. The resulting plas- mids were transformed into B. subtilis cells DB428, which do not produce additional extracellular proteases (R. Doi, personal communication). Expression of subtilisin E in B. subtilis was detected by using modified Schaeffer’s medium agar platesI6 with casein. The enzyme was purified from the culture supernatants as de- scribed previously.”

Molecular Modeling

To model subtilisin E, the appropriate amino acids from subtilisin E from B. subtilis 16827 were substituted into the X-ray crystal structure of subtilisin BPN’ .IS Coordi- nates from the Brookhaven database were displayed on an Evans and Sutherland PS300 graphics terminal using the program BIOGRAF.

Enzyme Assays and Kinetic Stability

The amidase activity of subtilisin E was measured on sAAPF-pna at 37°C in 0.1M tris-HCI, 10 mM CaClz, pH 8.0.’ The reaction was initiated by adding a small aliquot of the enzyme solution to 0.95 mL of buffered substrate (0.5 mM). The amount of p-nitroaniline re- leased was followed spectrophotometrically at 410 nm over several minutes.

To study the pH dependence of amidase activity, a combination buffer system (0.1M MOPS, 0.05M tris, 0.05M diethanolamine) was used in order to obtain a buffering capacity over the pH range 6-9. The concentrations of the three buffer components guarantee that the change in ionic strength over this pH range is minimal. The ionic strength was adjusted to 0.1M by adding appropriate amounts of NaCl.”

To determine the kinetic stabilities, lyophilized enzymes (from 2mM CaC12, 10 mM tris, pH 8) were redis-

142 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 39, NO. 2, JANUARY 20, 1992

solved in the same volume of solvent mixture [SO% (v/v) DMF, 20% H 2 0 or 40% DMF, 60% H20] and incu- bated at 30°C or 50°C. Because the stability of subtilisin strongly depends on the calcium ion concentration, it was necessary to maintain a constant CaC12 concentration in the DMF/water incubation mixture. Enzyme concentrations were micromolar or less. The samples were prepared so that all the variants showed similar initial activities, and residual hydrolytic activities were measured at different times of incubation in the organic solvent.

RESULTS

Selection of Charged Surface Residues

Although many of the surface residues of subtilisin may be amenable to replacement, we have been careful to avoid replacing residues critical to enzyme function. Sequence homology and structural analysis were em- ployed to identify charged residues suitable for replacement. Subtilisin E belongs to a family of closely related serine pro tease^.'^ Certain regions of the amino acid sequence are conserved among all subtilisins, suggesting that these sequences are essential for the protein function or for its processing. Conserved amino acids were not considered for replacement. Subtilisin E also contains two calcium binding sites”; mutations that would directly interfere with calcium binding were avoided, as were side chains that participate in known stabilizing interactions (hydrogen bonds or ion pairs). The position we chose to test in this work, Asp248, is an acidic surface residue that does not appear to interact with other side chains and is not conserved. (Position 248 is occu- pied by Ser in subtilisin BPN’ and by Asn in subtilisin Carlsberg.) To test whether the stability of enzymes in nonaqueous environments depends on the hydrophobicity of the protein surface, three variants were constructed in which the aspartyl side chain at position 248 was replaced by three side chains of increasing hydrophobicity: Asn, Ala, and Leu.

Activities of Subtilisin Variants

As shown in Table I, the specific activities of wild-type subtilisin E and each of the variants at position 248 are very similar. In contrast, Am218 -+ Ser and the double mutant Asp248 + Asn + Asn218 + Ser exhibit ap- proximately twofold higher specific activities. This in- creased activity reflects an increase in the catalytic rate constant k,,, caused by the mutation at position 218, which is located close to the active site.7 All the variants display a pH-activity profile that is identical to that of wild-type subtilisin E illustrated in Figure 1.

Stability in DMF

Figure 2 shows the deactivation profiles of wild- type and three variant subtilisins E (Asp248 + Asn,

.., 6 7 a 9 1 0

PH Figure 1. The pH dependence of catalytic activity of wild-type and variant subtilisins E at 37°C. Hydrolysis of suc-Ala-Ala-Pro- Phe-p-nitroanilide, normalized with respect to maximum activity at pH 8.5. The differences between the percentage of activity measured for the various mutant subtilisins are smaller than the size of the symbol used to represent the data point.

Table I. Specific activities and half-lives for inactivation of wild-type and variant substilisins E.

Variant

80% DMF, 30°C 40% DMF, 50°C

Specific activity l l i 2 AAG t112 AAG (nmoi min-’ mg-’) (h) (kcal mol-I) (h) (kcal mol-’)

Wild type 27,900 5.7 - 4.7 - Asp248 + Asn 27,800 10.2 -0.35 3.7 +0.15 Asp248 + Ala 23,500 10.3 -0.36 3.6 +0.17 Asp248 + Leu 25,300 11.1 -0.40 4.2 +0.07 Asn218 4 Ser 42,100 10.2 -0.35 10.0 -0.48 Asp248 - Asn + Asn218 + Ser 42,200 19.2 -0.73 8.0 -0.34

Specific activities were determined for hydrolysis of sAAPF-pna in 0.1M tris-HCI, 10 mM CaC12, pH 8, 37°C. Errors in the AAG values are less than 10%.

MARTINEZ ET AL.: PROTEIN ENGINEERING IN ORGANIC SOLVENTS 143

1 0 ' I I

0 1 0 20 30

Time (h) Figure 2. Deactivation of wild-type and variant subtilisins E in 80% (v/v) DMF, 30"C, as measured by residual activity for hydrolysis of suc-Ala-Ala-Pro-Phe-p-nitroanilide: (W) wild type; ( + ) Asn218 + Ser; (0) Asp248 + Asn; (A) Asn218 + Ser + Asp248 + Asn.

Asn218 + Ser, and Asp248 + Asn + Asn218 -+ Ser) upon exposure to 80% DMF at 30°C. Deactivation rates of subtilisin E and its variants were also measured in 40% DMF; data for the wild-type and three variants are shown in Figure 3. Half-lives for all five variants in 40% and 80% DMF are listed in Table I. The three mutations at position 248 enhance enzyme stability in the presence of 80% DMF; the half-lives of the three vari-

0 1 0 2 0

Time (h) Figure 3. Deactivation of wild-type and variant subtilisins E in 40% (v/v) DMF, 50"C, as measured by residual activity for hydrolysis of suc-Ala-Ala-Pro-Phe-p-nitroanilide: (M) wild type; (+) Asn218 + Ser; (0) Asp248 ---f Asn; (A) Asn218 + Ser + Asp248 + Asn.

ants increase by roughly a factor of 2 (Table I). How- ever, in 40% DMF all three variants are slightly less stable than wild type. The stability of Asn218 +. Ser subtilisin E is twice that of wild type in both 40% DMF and 80% DMF.

DISCUSSION

Stability in DMF

The activity of subtilisin E is drastically reduced in 80% DMF, which renders accurate determination of kinetic stability directly in the mixed solvent difficult. However, since the unfolding of subtilisin is irreversible, following incubation in the mixed solvent the enzyme can be diluted into aqueous buffer to assay residual activity. To prove that the kinetic stability measured by assaying residual enzyme activity after dilution into aqueous buffer is identical to the rate of deactivation in the DMF/water mixtures, the stability of wild-type subtilisin E was also assayed directly in 40% DMF at 50°C. The half-lives determined by the two techniques are within experimental error (data not shown).

Many water-soluble proteins have highly hydrophilic surfaces with numerous charged residues that are solvated by water. Transfer to a medium that is less ef- fective in solvating charges could lead to unfavorable interactions as these charges deionize or search for better solvation through aggregation or misfolding. We have p r o p o ~ e d ~ , ~ that the replacement of noncritical charged surface residues by uncharged amino acids would stabilize a protein for use in nonaqueous solvents by rendering the protein surface less hydrophilic and presumably less dependent on solvation by water for proper folding.

A systematic study of the replacement of four charged residues on the surface of a-lytic protease by 14-18 of the remaining 19 amino acids resulted in the identifica- tion of seven variants that were significantly more stable than wild-type enzyme in 84% DMF.17 When two surface substitutions were combined, the resulting double mutant was 27 times more stable than wild-type a-lytic protease. Similarly, replacement of a surface lysine in an engineered subtilisin BPN' by (wild-type) tyrosine improved stability in DMF?5 In this study, the replacement of the isolated charged residue Asp248 by the uncharged amino acids Asn, Ala, and Leu enhances the stability of subtilisin E in the presence of high concentrations of DMF.

Whatever the mechanism by which the Asp248 surface substitutions enhance stability of subtilisin E in 80% DMF at 30"C, it is absent in 40% DMF and 50°C. For the experiments in 40% DMF, it was necessary to raise the temperature to 50°C in order to reduce the enzyme half-lives to accurately measurable levels. (At 50°C in 80% DMF, the enzymes deactivate too rapidly for accurate measurement .) Thus the differences in rela-


tive stabilities in 40% and 80% DMF could result from either the difference in solvent environment or temperature. Although we are unable to distinguish between the two at this time, it is tempting to speculate that the removal of the charged residue becomes increasingly important at low water concentrations, where charge de- solvation would occur. This would explain why surface charge substitutions enhance enzyme stability only in high concentrations of organic solvent. It is clear, however, from studies on both subtilisin E (this work) and a-lytic p r~ tease ’~ that surface charge substitutions generally do not alter enzyme stability in primarily aqueous media. The stability enhancement achieved by removing surface charges is unique to nonaqueous solvents and reflects greater compatibility between the altered enzyme and the unusual solvent environment.

A second general approach to enhancing enzyme stability in nonaqueous solvents is to incorporate stabilizing interior crosslinks in the form of disulfide bridges, hydrogen bonds, or electrostatic interaction^.^ The Am218 +. Ser mutation was originally discovered by random mutagenesis and screening for thermostable variants of subtilisin BPN’6 and was one of six mutations incorporated in a subtilisin BPN‘ variant resistant to deactivation in anhydrous DMF.34 The stability of As11218 +. Ser is greater than wild-type subtilisin E in both 40% DMF and 80% DMF (Table I). As this substitution increases the thermostability of the homolo- gous subtilisin BPN’ in fully aqueous media as well, its stabilizing influence is essentially independent of the solvent. In discussing the design rules for engineering enzymes for stability in organic solvents, it was proposed that certain types of stabilizing mutations (i.e., those that improved internal interactions such as hydrogen bonds) would enhance thermostability as well as stability in organic solvents. Asn218 -+ Ser, which serves to improve hydrogen bonds in subtilisin BPN’, is a good example of such a mutation.

Additivity of Mutations

It has been reported that mutations whose effect is to increase subtilisin E activity in organic solvents are additive when ~ombined .~ For catalytic activity, additivity is defined in terms of incremental free energies of transition state ~tabilization.~~ Additivity for stabilizing mutations is defined in terms of incremental effects on the free energy of protein stability or the free energy by which the folded protein is more stable than the unfolded protein. For subtilisin, this free energy is difficult to measure directly (e.g., by calorimetric techniques) since the enzyme unfolds irreversibly. However, experimental data for subtilisin BPN’ and subtilisin E show that kinetic stability (half-life for deactivation) and thermodynamic stability (as measured by differential scanning calorimetry) are strongly ~ o r r e l a t e d . ~ ~ ’ ~ . ~ ~ This relation between thermodynamic and kinetic stabilities

implies that loss of activity is a direct consequence of (irreversible) unfolding events. In organic solvents and at low enzyme concentrations, where the catalytic activity is extremely low and deactivation follows first-order kinetics and does not involve aggregation, it is reason- able to assume that deactivation mirrors the enzyme’s thermodynamic stability.

One can estimate the incremental free energy of unfolding that results from a particular amino acid substitution from the first-order deactivation rate constants, according to the formula

AAG

This assumes that the incremental activation energy for unfolding (due to an amino acid substitution) is equal to the incremental free energy of protein stability under the experimental conditions. Implicit is the assumption that the wild-type and mutant enzymes pass through a transition state of equal free energy; in other words, the mutation does not significantly alter the free energy of the transition state for unfolding. Fersht and co-workers have shown that mutations in the hydrophobic core of barnase affect the energies of both the folded protein and the transition state.” However, surface residues and specific hydrogen bonds are less likely to be involved in unfolding transition state stabilization.

Free energies AAG calculated according to Eq. (1) are reported in Table I. Stabilization free energies of up to 0.8 kcal mol-’ were observed for substitution of selected charged surface residues in a-lytic protease.” The increases in stability observed for the subtilisin E variants are relatively small, corresponding to 0.3-0.4 kcal mol-’ for the individual substitutions. However, if multiple substitutions can be combined in an active variant, significant enhancements in enzyme stability can be achieved by this strategy. This strategy for protein stabilization in nonaqueous solvents is also relatively easy to implement, since proteins are particularly tolerant of amino acid substitutions at surface positions.’”’ The free energies reported in Table I show that the effects of combining the two stabilizing subtilisin E mutations are additive, as would be expected, since positions 218 and 248 are located far apart in the subtilisin tertiary structure (-27 A).

Specific Activities and pH Profiles

Altering the surface charge of a serine protease can shift the pH-activity profile by affecting the pK, of the active site h i ~ t i d i n e . ~ ~ * ~ ~ However, the distance between Asp248 and the active site histidine (residue 64) is quite large (-30 A), and any electrostatic interaction between these two residues would be greatly attenuated. The substitution of this surface residue in fact has no effect on either the catalytic activity of subtilisin E or the pH dependence of that activity. A shift in the pH de-


pendence of catalytic activity is not expected for the Am218 -+ Ser substitution because it does not involve a change in the surface charge.

The results of site-directed mutagenesis of subtilisin E reported in this article support the argument that enzymes can be rationally engineered to improve stability in polar organic solvents. Small alterations, even single amino acid substitutions, can enhance the stability of the enzyme in high concentrations of DMF. Further- more, single substitutions can be combined, and significant stabilization can be achieved without decreasing enzyme activity. The types of changes in amino acid sequence that improve stability in organic solvents are not necessarily the same as those that enhance thermostability; there are clearly certain classes of substitutions whose effects are unique to organic solvents. The major- ity of surface charge substitutions studied fall into this category. Replacement of charged surface residues ap- pears to be a generally applicable strategy for enhancing enzyme stability in organic solvents.

This research is supported by the Energy Conversion and Utilization Technologies program of the Department of En- ergy. One of us (F. H. A.) gratefully acknowledges a National Science Foundation Presidential Young Investigator Award and a David and Lucile Packard Foundation Fellowship. An- other of us (A. C. R.) acknowledges a National Institutes of Health postdoctoral fellowship, GM13953-02.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

Alber, T., Daopin, S . , Nye, J. A., Muchmore, D.C., Matthews, B.W. 1987. Temperature-sensitive mutations of bacteriophage T4 lysozyme occur at sites with low mobility and low solvent accessibility in the folded protein. Biochemistry 26: 3754. Arnold, F. H. 1988. Protein design for nonaqueous solvents. Prot. Eng. 2: 21. Arnold, F. H . 1990. Enzyme engineering for non-aqueous solvents. Trends Biotechnol. 8: 244. Barbas 111, C. F., Matos, J. R., West, J. B., Wong, C.-H. 1988. A search for peptide ligase: cosolvent-mediated conversion of proteases to esterases for irreversible synthesis of peptides. J. Am. Chem. SOC. 110: 5162. Bonneau, P. R., Graycar, T. P., Estell, D.A., Jones, J. B. 1991. Alteration of the specificity of subtilisin BPN’ by site-directed mutagenesis in its S, and S,’ binding sites. J. Am. Chem. SOC. 113: 1026. Bryan, P. N., Rollence, M. L., Pantoliano, P.W., Wood, J., Finzel, B. C., Gilliland, G . L . , Howard, A . J., Poulos, T. L. 1986. Proteases of enhanced stability: characterization of a thermostable variant of subtilisin. Proteins: Struct. Func. Genet. 1: 326. Chen, K., Robinson, A.C., Van Dam, M. E., Martinez, P., Economou, C., Arnold, F. H. 1991. Enzyme engineering for nonaqueous solvents 11. Additive effects of mutations on the stability and activity of subtilisin E in polar organic media. Biotechnol. Progr. 7 125. Del Mar, E. G., Largman, C., Brodrick, J. W., Geokas, M. C. 1979. A sensitive new substrate for chymotrypsin. Anal. Biochem. 99: 316. Dordick, J. S. 1989. Enzymatic catalysis in monophasic organic solvents. Enz. Microb. Technol. 11: 194.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Ellis, J . K., Morrison, J. F. 1982. Buffers of constant ionic strength for studying pH-dependent processes. Meth. Enzy- mol. 87: 405. Estell. D. A., Graycar, T. P., Wells, J. A. 1985. Engineering an enzyme by site-directed mutagenesis to be resistant to chemical oxidation. J. Biol. Chem. 260: 6518. Hecht, M. H., Sturtevant, J. M., Sauer, R.T. 1984. Effect of single amino acid replacements on the thermal stability of the NHz-terminal domain of phage A repressor. Proc. Natl. Acad. Sci. USA. 81: 5685. Hecht, M. H., Sturtevant, J. M., Sauer, R.T. 1986. Stabilization of A repressor against thermal denaturation by site-directed Gly 4 Ala changes in a-helix 3. Proteins 1: 43. Hwang, J., Arnold, F. H. 1991. Enzyme design for nonaqueous solvents. In: D. S. Clark and H.W. Blanch (eds.), Applied Bio- catalysis. Marcel Dekker Vol. I, pp. 53-86. Kraut, J. 1971. Subti1isin:x-ray structure. Enzymes 3: 547. Leighton, T. J., Doy, R. H. 1971. The stability of messenger ribonucleic acid during sporulation in Bacillus subtilis. J. Biol. Chem. 246: 3189. Martinez, P., Arnold, F. H. 1991. Surface charge substitutions increase the stability of a-lytic protease in organic solvents. J. Am. Chem. SOC. 113: 6336. Masai, Y., Coleman, J., Inouye, M. 1983. Multipurpose expression cloning vehicles in Escherichia Colt. pp. 15-32 In: Inouye, M. (ed.), Experimental manipulation of gene expression. Aca- demic, New York. Matouschek, A, , Kellis, J.T. Jr., Serrano, L., Fersht, A. R. 1989. Mapping the transition state and pathway of unfolding by protein engineering. Nature 340: 122. Matsumura, M., Yasumura, S., Aiba, S. 1986. Cumulative effect of intragenic amino-acid replacements on the thermostability of a protein. Nature 323: 356. Matsumura, M., Signor, G., Matthews, B. W. 1989. Substantial increase of protein stability by multiple disulfide bonds. Nature 342: 291. Pantoliano, M. W., Whitlow, M., Wood, J. F., Rollence, M. L., Finzel, B. C., Gilliland, G . L., Poulos, L. P., Bryan, P. N. 1988. The engineering of binding affinity at metal ion binding sites for the stabilization of proteins: subtilisin as a test case. Bio- chemistry 27: 8311. Pantoliano, M.W., Whitlow, M., Wood, J.F., Dodd, S.W., Hardman, K. D., Rollence, M. L., Bryan, P. N. 1989. Large increases in general stability for subtilisin BPN’ through incremental changes in the free energy of unfolding. Biochemistry 28: 7205. Riva, S., Chopineau, J., Kieboom, A. P.G., Klibanov, A. M. 1988. Protease-catalyzed regioselective esterification of sugars and related compounds in anhydrous dimethylformamide. J. Am. Chem. SOC. 110: 584. Russell, A. J., Fersht, A. R. 1987. Rational modification of enzyme catalysis by engineering surface charge. Nature 328: 496. Russell, A. J . , Thomas, P. G., Fersht, A. R. 1987. Electrostatic effects on modification of charged groups in the active site cleft of subtilisin by protein engineering J. Mol. Biol. 193: 803. Stahl, M.L. , Ferrari, E. 1984. Replacement of the Bacillus subtilis subtilisin structural gene with in vitro-derived deletion mutation. J. Bacteriol. 158: 411. Takagi, H., Morinaga, Y., Ikemura, H., Inouye, M. 1988. Mu- tant subtilisin E with enhanced protease activity obtained by site-directed mutagenesis. J. Biol. Chem. 263: 19592. Takagi, H., Morinaga, Y., Ikemura, H., Inouye, M. 1989. The role of Pro-239 in the catalysis and heat stability of subtilisin E. J . Biochem. 105: 953. Takagi, H., Takahashi, T., Momose, H., Inouye, M., Maeda, Y., Matsuzawa, H., Ohta, T. 1990. Enhancement of the thermostability of subtilisin E by introduction of a disulfide bond engineered on the basis of structural comparison with a thermo- philic serine protease. J. Biol. Chem. 265: 6874.


31. Wells, J .A., Cunningham, B. C., Graycar, T. P., Estell, D.A. 1987. Recruitment of substrate-specificity properties from one enzyme into a related one by protein engineering. Proc. Natl. Acad. Sci. USA. 84: 5167.

32. Wells, J. A,, Estell, D. A. 1988. Subtilisin-an enzyme designed to be engineered. TIBS. 13: 291.

33. Wells, J. A. 1990. Additivity of mutational effects in proteins. Biochemistry 29: 8509.

34. Wong, C.-H., Chen, S.-T., Hennen, W. J . , Bibbs, J .A. , Wang, Y. F., Liu, J. L.-C., Pantoliano, M.W., Whitlow, M., Bryan,

P. N. 1990. Enzymes in organic synthesis: use of suhtilisin and a highly stable mutant derived from multiple site-specific mutations. J. Am. Chem. Soc. 112: 945.

35. Zhong, Z . , Liu, J . L - C . , Dinterman, L . M . , Finkelman, M. A. J. , Mueller, W.T., Rollence, M. L., Whitlow, M., Wong, C.-H. 1991. Engineering suhtilisin for reaction in dimethylformamide. J. Am. Chem. SOC. 113: 683.


Documents

Stabilization of substilisin E in organic solvents by site-directed mutagenesis