Protein Engineering, Vol. 14, No. 5, 343-347,
May 2001
© 2001 Oxford University Press
Comparing the effect on protein stability of methionine oxidation versus mutagenesis: steps toward engineering oxidative resistance in proteins
Department of Chemistry and Biochemistry, University of Arkansas, Fayetteville, AR 72701–1021, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The biological activity of some proteins is known to be sensitive to oxidative damage caused by a variety of oxidants. The model protein staphylococcal nuclease was used to explore the effect on protein structural stability of oxidizing methionine to the sulfoxide form. These effects were compared with the effects of substituting methionines with isoleucine and leucine, a potential strategy for stabilizing proteins against oxidative damage. Wild-type nuclease and various mutants were oxidized with hydrogen peroxide. Stabilities of both oxidized and unoxidized proteins were determined by guanidine hydrochloride denaturation. Oxidation destabilized the wild-type protein by over 4 kcal/mol. This large loss of stability supports the idea that in some cases loss of biological activity is linked to disruption of the protein native state. Comparison of mutant protein's stability losses upon oxidation showed that methionines 65 and 98 had a much greater destabilizing effect when oxidized than methionines 26 or 32. While substitution of methionine 98 carried as great an energetic penalty as oxidation, substitution at position 65 was less disruptive than oxidation. Thus a simple substitution mutagenesis strategy to protect a protein against oxidative destabilization is practical for some methionine residues.
Keywords: methionine oxidation/methionine sulfoxide/mutagenesis/protein stability/staphylococcal nuclease
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Proteins are well known to be sensitive to oxidative damage, often with important biological effects. Protein oxidation has been suggested as a causative or contributory factor in many diseases (Dean et al., 1997
). Oxidized proteins have been found to increase in aged organisms, leading to the proposal that protein oxidation contributes to the aging process (Berlett and Stadtman, 1998; Gao et al., 1998a; Schoneich, 1999).
Methionine, cysteine, tryptophan, tyrosine and histidine residues are susceptible to oxidation. The oxidation of methionine to the sulfoxide form is of particular interest as it has been shown to occur in a wide variety of proteins and often reduces or eliminates biological activity (Brot and Weissbach, 1991; Vogt, 1995). Natural biological oxidants (Glaser et al., 1992; Fliss and Menard, 1994; Fujii et al., 1994), cigarette smoke (Gadek et al., 1979; Boudier et al., 1983) and other environmental oxidants (Banerjee and Mudd, 1992; Blaurock et al., 1992) have all been demonstrated to cause methionine sulfoxide formation in proteins and peptides. Methionine oxidation is of serious concern when proteins are used as pharmaceuticals (Fransson et al., 1996; Gitlin et al., 1996; Berti et al., 1997; Konz et al., 1998; Liu et al., 1998; Kornfelt et al., 1999) because of the possible effects on activity and the possibility of oxidation during processing or storage.
While it is not always understood how methionine oxidation changes the biological activity of a protein, oxidation is known in some cases to destabilize a protein's native structure (Volkin et al., 1997; Gao et al., 1998b; Liu et al., 1998; DalleDonne et al., 1999) with obvious consequences for activity. It is well known that side chain substitution can have profound effects on protein stability and structure. In effect oxidation can be regarded as a chemical `mutagenesis' which substitutes the methionine side chain with methionine sulfoxide, a larger and more polar side chain. It should not be surprising that it can greatly disrupt protein structure and stability.
A potential strategy for protecting a protein against the effects of methionine oxidation is to replace those residues with some side chain which is resistant to oxidation. This has been attempted in a few cases, usually directed at preventing methionine oxidation that directly affects activity by altering an active or binding site. 
1-Antitrypsin (Rosenberg et al., 1984), subtilisin (Estell et al., 1985; Bott et al., 1988) and D-amino acid oxidase (Ju et al., 2000) have all had the effect of methionine replacement on oxidative inactivation assessed. However, in only one case that we are aware of has the effect of methionine oxidation upon stability been measured combined with mutagenesis to desensitize the protein toward oxidation. Even that study (Lu et al., 1999), on human granulocyte colony-stimulating factor, examined only the stability effects of oxidation itself and not the stability effects of mutagenesis.
Such a comparison is important since it is plausible that a site that is sensitive to the effects of oxidation might be equally sensitive to the effects of side chain substitution. The oxidation of methionine presumably affects protein structure and stability by reducing side chain hydrophobicity, increasing the capacity for hydrogen bonding and altering the size and shape of the methionine. The relative importance of these factors will depend on the particular environment of a given methionine. For example, somewhat counterintuitively it has been shown (Chao et al., 1997) that methionine oxidation can actually increase the hydrophobicity of a protein surface, presumably through conformational changes of some sort, but the energetic consequences of such rearrangement are not clear and neither are the effects of substitution. It is equally plausible to imagine that if the cause of destabilization upon oxidation is the increase in side chain polarity then the effects of substitution with a non-polar side chain may be minor.
At our present state of knowledge it is difficult to predict the degree of stability change upon oxidation of a given methionine residue, much less whether or not side chain substitution is a viable approach to moderate those effects. To further understanding of these issues, in this study we examined the effects of methionine oxidation upon the stability of staphylococcal nuclease, a well-characterized model protein. We determined the relative effects of oxidizing each methionine and we compared oxidation stability effects with those of mutating each methionine to alanine, glycine, isoleucine or leucine.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
All mutants of staphylococcal nuclease were prepared by procedures described previously (Byrne et al., 1995
). Protein stabilities were determined by guanidine hydrochloride denaturation as described previously (Stites et al., 1995; Schwehm and Stites, 1998). ß-Mercaptoethanol (BME) was purchased from Millipore. Hydrogen peroxide (H2O2, nominal 30%) was purchased from Fisher Scientific. Procedure for methionine oxidation
Hydrogen peroxide, a well-known oxidizing agent fairly specific for methionine and cysteine (Shechter et al., 1975), was used. To 0.12 µmol (192.3 µl) of wild-type or mutant (M26I, M32I, M65L, M98L) staphylococcal nuclease at a concentration of 2.6 mg/ml (pH 7.0, 25 mM sodium phosphate, 100 mM NaCl) were added 11.2 µl of 31.6% (w/w) hydrogen peroxide (~950 equiv. for each of the four methionines in wild-type). This solution was stirred for 30 min at room temperature. The reaction mixture was then quenched with a large excess (25.2 µl, ~3000 equiv. based on four methionines) of ß-mercaptoethanol. The oxidized proteins were stored at –20°C until stabilities were determined.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
We have previously published the stabilities of the M32I and M32L mutants of staphylococcal nuclease (Spencer and Stites, 1996
). The remaining three methionines present in wild-type staphylococcal nuclease were each mutated to isoleucine and leucine and the stabilities of the resulting proteins were determined. Previously published work also described the stability of the alanine and glycine substitutions at these sites (Shortle et al., 1990). The stabilities of all these mutant proteins are given in Table I.
|
Wild-type nuclease and the mutants M26I, M32I, M65L, M98L were each subjected to strongly oxidizing conditions. The stabilities of these oxidized proteins were then assessed by guanidine hydrochloride denaturation. Figure 1a
shows the normalized fluorescence intensity versus guanidine hydrochloride concentration for oxidized and unoxidized wild-type and M65L. All the unoxidized proteins had the characteristic sigmoid of an apparent two-state denaturation as represented by wild-type and M65L in Figure 1a. When these data are fitted to a two-state model as previously described (Stites et al., 1995; Schwehm and Stites, 1998) the logarithm of the apparent equilibrium constant varies linearly with guanidine hydrochloride concentration (Figure 1b) as usual for staphylococcal nuclease and its mutants. After oxidation, the transition from native to denatured structure was shifted to lower guanidine hydrochloride concentrations for all proteins. Wild-type, M26I and M32I after oxidation had a comparatively ragged and broad transition. The denaturation curve for wild-type is shown in Figure 1a as a representative of this type of behavior. As shown in Figure 1b, the data do not fit the two-state model well. In contrast, the oxidized mutants M65L and M98L, although less stable than before oxidation, had fairly smooth sigmoidal transitions. This is readily seen in the representative unfolding curve of oxidized M65L in Figure 1a. In Figure 1b it is also clear that a two-state model fits M65L reasonably well.
|
), oxidized wild-type (
), unoxidized M65L (•) and oxidized M65L (
). (a) Fluorescence intensity of tryptophan 140 as a function of guanidine hydrochloride concentration. (b) Variation of the log(apparent equilibrium constant) with guanidine hydrochloride concentration. The apparent equilibrium constant is calculated by a two-state model where Kapp = (IN – I)/(I – ID), IN is the intensity of fluorescence in the native state, ID is the denatured intensity and I is the intensity at a given guanidine hydrochloride concentration. A least squares fit to the data in the range 1 > logKapp > –1 is plotted as a solid line (Schwehm and Stites, 1998The stabilities of the oxidized proteins as determined by a fit to two-state model are given in Table II
. These values should be regarded as approximate, especially for wild-type, M26I and M32I. Fortunately, of the three parameters, mGuHCl, Cm and 
GH2O, it is the last, the stability difference between the denatured and native states, that is both most important for our purposes and the most accurate. This accuracy is due to the fact that after the oxidation the mixture of proteins is sufficiently destabilized that a reasonable estimation of the average equilibrium constant between the denatured and native states can be made without a long extrapolation. Thus GH2O is a reliable weighted average value of all the different protein species present.
|
As alluded to above, the deviations from two-state behavior for oxidized wild-type, M26I and M32I imply that two or more species of differing stabilities are present. It has been shown (Keck, 1996
) that hydrogen peroxide generally does not distinguish between exposed and buried methionine residues. However, to cite one of several examples, Lu et al. (1999) have shown that there are different rates at which methionines in human granulocyte colony-stimulating factor oxidize, with complete oxidation occurring at some sites in minutes, with other sites incompletely oxidized after days of peroxide exposure. Therefore, the cause of these deviations from two-state behavior is most likely the presence of proteins with different numbers of methionine sulfoxides. However, tryptophan and cysteine are the two other residues in proteins which are particularly sensitive to oxidation. There are no cysteine residues and only one tryptophan in wild-type staphylococcal nuclease. In order to confirm that partial oxidation of this tryptophan was not responsible for additional protein species, an oxidation reaction was performed under similar conditions to the protein oxidation. Tryptophan when treated overnight with 20 equiv. of H2O2, either alone or in a mixture with methionine, did not oxidize to oxindolylalanine as assayed by NMR (data not shown). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Is simple substitution of methionine with another side chain a generally viable method for preventing the destabilization of oxidized proteins? A substitution will, of course, prevent oxidation, but this does little good if the new protein is severely destabilized by the mutation. The most attractive choices for a replacement side chain resistant to oxidation are alanine and leucine. Substitution with alanine removes the sulfur, but greatly reduces side chain hydrophobicity and the potential for favorable van der Waals interactions and possibly introduces a void in the protein. Leucine preserves side chain hydrophobicity and the potential for favorable van der Waals interactions, but alters side chain geometry causing potentially unfavorable van der Waals interactions. A third choice, isoleucine, is less attractive because it introduces beta branching.
There are four methionines in wild-type staphylococcal nuclease. Their positions are shown in Figure 2. The stabilities of the isoleucine, leucine and alanine substitutions for each of the four methionines are summarized in Table I
. Data for the effects of glycine substitution are available and are also included for completeness, although such a mutation would usually be expected to be far too destabilizing for use in engineering oxidative resistance. In general, the effects of mutagenesis are relatively minor at positions 26 and 32, with much less than 1 kcal/mol of stability lost upon leucine or isoleucine substitution. Methionine 98 is clearly the position most sensitive to substitution, as expected given its lack of solvent exposure (Shortle et al., 1990). Methionine 65 shows the most complicated behavior, with leucine substitution having relatively little effect, whereas isoleucine has a significantly greater effect. The alanine substitution at 65 is only slightly worse than the alanine substitution at 26 or 32, the least sensitive sites, while glycine substitution at 65 is severely destabilizing.
|
The stability effects of oxidation upon wild-type are as large as the effects of the most destabilizing methionine substitutions. As noted in the results, the stability values for wild-type, M26I and M32I given in Table II
represent a rough weighted average because of the presence of multiple species. The changes in stability upon oxidation of the mutants M65L and M98L make it clear that most of the effects observed in wild-type are due to the oxidation of these two residues. This is revealed in two ways. First, the removal of either methionine 65 or 98, but not 26 or 32, greatly simplifies the complex denaturation behavior of wild-type. Presumably, 65 and/or 98 are only partly oxidized in the wild-type, M26I and M32I proteins and both must have considerable effects on the denaturation behavior or there would be little difference when removed.
More telling is the difference in stability between the oxidized and unoxidized proteins. This G value is nearly the same for wild-type, M32I and M26I, indicating that the removal of these residues does little to prevent the oxidative destabilization of staphylococcal nuclease. On the other hand, the removal of either methionine 98 or 65 reduced the value of G by about 2.0 kcal/mol in each case, nearly half of the wild-type G. This is consistent with these two residues accounting for most of the effects observed in wild-type.
The comparison of the effects of mutation to oxidation of each residue is informative. Not surprisingly, methionines 26 and 32 were not particularly sensitive to either mutation or oxidation. Indeed, substitution seemed, if anything, more disruptive than oxidation at these two sites. There would be little need to alter a methionine showing similar behavior in any protein unless such a residue interfered with activity because it was in an active or binding site. However, there would also be relatively little harm from a structural stability viewpoint in doing so if necessary.
The two residues that proved to cause most of the oxidative destabilization differed in their reaction to substitution. Oxidation of methionine 98 destabilized the protein by 2.0 kcal/mol, but the effects of substitutions were as bad or even worse. Hence although substitution of the methionine prevented oxidative damage in the case of positions such as methionine 98, it will also be necessary to make substitutions at other sites in the protein to find a stable structural solution. This is in contrast to the behavior of methionine 65. Oxidation of this side chain also destabilized the protein by 2.0 kcal/mol. Whereas three substitutions had effects that were about the same or worse than oxidation, one substitution, M65L, was significantly less destabilizing. The M65L substitution is not perfect, destabilizing the protein by 0.8 kcal/mol, but it is a noteworthy improvement over the effects of oxidation.
It has been proposed that methionine residues might act as intrinsic scavengers of various reactive oxygen species and thus protect other components of the cell against oxidative damage (Levine et al., 1999). This might lead to the supposition that removal of methionine would actually increase oxidative sensitivity. Our results show that this hypothesis is untrue in the specific case of staphylococcal nuclease, at least as far as damage to the protein itself is concerned. Of course, this does not mean that other sensitive components of a cell might be protected by sacrificial oxidation of methionine in a particular protein. However, the strategy suggested here is more likely to find application in situations where purified protein is being used and protection of other cellular components is moot.
In the future, as our ability to predict the effects of single and multiple mutations improves, it should be relatively easy to engineer a protein to resist oxidative destabilization. Unfortunately, our current state of knowledge is such that predicting the other changes needed to compensate completely for the effects of methionine substitution is not easy or reliable. Our results here show that simple substitution of some methionines helps to render a protein more structurally resistant to oxidation. Since every 1.34 kcal/mol represents a 10-fold increase in the amount of folded protein at 20°C, even modest prevention of stability loss can do much to retain activity, making this simple approach a viable strategy.
|
Notes |
|---|
1 Present address: Department of Microbiology, University of Arkansas for Medical Sciences, 4301 W. Markham, Little Rock, AR 72205, USA
2 To whom correspondence should be addressed. E-mail: wstites{at}uark.edu
|
Acknowledgments |
|---|
This research was supported in part by the NIH.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Banerjee,S.K. and Mudd,J.B. (1992) Arch. Biochem. Biophys., 295, 84–89.[ISI][Medline]
Berlett,B.S. and Stadtman,E.R. (1998) Drug Metab. Rev., 30, 225–243.[ISI][Medline]
Berti,P.J., Ekiel,I., Lindahl,P., Abrahamson,M. and Storer,A.C. (1997) Protein Express. Purif., 11, 111–118.[ISI][Medline]
Blaurock,B., Hippeli,S., Metz,N. and Elstner,E.F. (1992) Arch. Toxicol., 66, 681–687.[ISI][Medline]
Bott,R., Ultsch,M., Kossiakoff,A., Graycar,T., Katz,B. and Power,S. (1988) J. Biol. Chem., 263, 7895–7906.
Boudier,C., Pelletier,A., Pauli,G. and Bieth,J.G. (1983) Clin. Chim. Acta, 132, 309–315.[ISI][Medline]
Brot,N. and Weissbach,H. (1991) Biofactors, 3, 91–96.[ISI][Medline]
Byrne,M.P., Manuel,R.L., Lowe,L.G. and Stites,W.E. (1995) Biochemistry, 34, 13949–13960.[Medline]
Chao,C.C., Ma,Y.S. and Stadtman,E.R. (1997) Proc. Natl Acad. Sci. USA, 94, 2969–2974.
DalleDonne,I., Milzani,A. and Colombo,R. (1999) Biochemistry, 38, 12471–12480.[Medline]
Dean,R.T., Shanlin,F., Stocker,R. and Davies,M.J. (1997) Biochem. J., 324, 1–18.
Estell,D.A., Graycar,T.P. and Wells,J.A. (1985) J. Biol. Chem., 260, 6518–6521.
Fliss,H. and Menard,M. (1994) J. Appl. Physiol., 77, 2727–2733.
Fransson,J., Florin-Robertsson,E., Axelsson,K. and Nyhlen,C. (1996) Pharm. Res., 13, 1252–1257.[ISI][Medline]
Fujii,N., Ishibashi,Y., Satoh,K., Fujino,M. and Harada,K. (1994) Biochim. Biophys. Acta, 1204, 157–163.[Medline]
Gadek,J.E., Fells,G.A. and Crystal,R.G. (1979) Science, 206, 1315–1316.
Gao,J., Yin,D., Yao,Y., Williams,T.D. and Squier,T.C. (1998a) Biochemistry, 37, 9536–9548.[Medline]
Gao,J., Yin,D., Yao,Y., Sun,H., Qin,Z., Schoneich,C., Williams,T.D. and Squier,T.C. (1998b) Biophys J., 74, 1115–1134.
Gitlin,G., Tsarbopoulos,A., Patel,S.T., Sydor,W., Pramanik,B.N., Jacobs,S., Westreich,L., Mittelman,S. and Bausch,J.N. (1996) Pharm. Res., 13, 762–769.[ISI][Medline]
Glaser,C.B., Morser,J., Clarke,J.H., Blasko,E., McLean,K., Kuhn,I., Chang,R.J., Lin,J.H., Vilander,L. and Andrews,W.H. (1992) J. Clin. Invest., 90, 2565–2573.
Ju,S., Lin,L. Chein,H.R. and Hsu,W. (2000) FEMS Microbiol. Lett., 186, 215–219.[ISI][Medline]
Keck,R.G. (1996) Anal. Biochem., 236, 56–62.[ISI][Medline]
Konz,J.O., King,J. and Cooney,C.L. (1998) Biotechnol. Prog., 14, 393–409.[Medline]
Kornfelt,T., Persson,E. and Palm,L. (1999) Arch. Biochem. Biophys., 363, 43–44.[ISI][Medline]
Kraulis,P.J. (1991) J. Appl. Crystallogr., 24, 946–950.
Levine,R.L., Berlett,B.S., Moskovitz,J., Mosoni,L. and Stadtman,E.R. (1999) Mech. Ageing Dev., 107, 323–332.[ISI][Medline]
Liu,J.L., Lu,K.V., Eris,T., Katta,V., Westcott,K.R., Narhi,L.O. and Lu,H.S. (1998) Pharm. Res., 15, 632–640.[ISI][Medline]
Lu,H.S., Fausset,P.R., Narhi,L.O., Horan,T., Shinagawa,K., Shimamoto,G. and Boone,T.C. (1999) Arch. Biochem. Biophys., 362, 1–11.[ISI][Medline]
Rosenberg,S., Barr,P.J., Najarian,R.C. and Hallewell,R.A. (1984) Nature, 312, 77–80.[Medline]
Schoneich,C. (1999) Exp. Gerontol., 34, 19–34.[ISI][Medline]
Schwehm,J.M. and Stites,W.E. (1998) Methods Enzymol., 295, 150–170.[ISI][Medline]
Shechter,Y., Burstein,Y. and Patchornik,A. (1975) Biochemistry, 20, 4497–4503.
Shortle,D., Stites. W.E. and Meeker,A.K. (1990) Biochemistry, 29, 8033–8041.[Medline]
Spencer,D.S. and Stites. W.E. (1996) J. Mol. Biol., 257, 497-.[ISI][Medline]
Stites,W.E., Byrne,M.P., Aviv,J., Kaplan,M. and Curtis P.M. (1995) Anal. Biochem., 227, 112–122.[ISI][Medline]
Vogt,W. (1995) Free Rad. Biol. Med., 18, 93–105.[ISI][Medline]
Volkin,D.B., Mach,H. and Middaugh,C.R. (1997) Mol. Biotechnol., 8, 105–122.[ISI][Medline]
Received August 31, 2000; revised December 23, 2000; accepted February 22, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Y. Sroussi, J. Berline, and J. M. Palefsky Oxidation of methionine 63 and 83 regulates the effect of S100A9 on the migration of neutrophils in vitro J. Leukoc. Biol., March 1, 2007; 81(3): 818 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Iwakura, K. Maki, H. Takahashi, T. Takenawa, A. Yokota, K. Katayanagi, T. Kamiyama, and K. Gekko Evolutional Design of a Hyperactive Cysteine- and Methionine-free Mutant of Escherichia coli Dihydrofolate Reductase J. Biol. Chem., May 12, 2006; 281(19): 13234 - 13246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Chugha, H. J. Sage, and T. G. Oas Methionine oxidation of monomeric {lambda} repressor: The denatured state ensemble under nondenaturing conditions Protein Sci., March 1, 2006; 15(3): 533 - 542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. R. Sumner, A. Shanmuganathan, T. C. Sideri, S. A. Willetts, J. E. Houghton, and S. V. Avery Oxidative protein damage causes chromium toxicity in yeast Microbiology, June 1, 2005; 151(6): 1939 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Hou, I. Kang, R. E. Marchant, and M. G. Zagorski Methionine 35 Oxidation Reduces Fibril Assembly of the Amyloid Abeta -(1-42) Peptide of Alzheimer's Disease J. Biol. Chem., October 18, 2002; 277(43): 40173 - 40176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-H. Oh, S.-H. Nam, and H.-S. Kim Improvement of Oxidative and Thermostability of N-Carbamyl-D-Amino Acid Amidohydrolase by Directed Evolution Protein Eng. Des. Sel., August 1, 2002; 15(8): 689 - 695. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||