Protein Engineering, Vol. 13, No. 11, 739-743,
November 2000
© 2000 Oxford University Press
Trifluoroethanol may form a solvent matrix for assisted hydrophobic interactions between peptide side chains
Centre for Protein Analysis and Design, University of Bath, Claverton Down, Bath BA2 7AY, UK
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionReferences |
|---|
Several models for interactions between trifluoroethanol (TFE) and peptides and proteins have recently been proposed, but none have been able to rationalize the puzzling observations that on the one hand TFE can stabilize some hydrophobic interactions in secondary structures, but on the other can also melt the hydrophobic cores of globular proteins. The former is illustrated in this paper by the effect of TFE on a short elastin peptide, GVG(VPGVG)3, which forms type II ß-turns stabilized by hydrophobic interactions between two intra-turn valine side chains. This folding, driven by increasing the entropy of bulk water, is stimulated in TFE–water mixtures and/or by raising the temperature. To explain these apparently contradictory observations, we propose a model in which TFE clusters locally assist the folding of secondary structures by first breaking down interfacial water molecules on the peptide and then providing a solvent matrix for further side chain–side chain interactions. This model also provides an explanation for TFE-induced transitions between secondary structures, in which the TFE clusters may redirect non-local to local interactions.
Keywords: alcohol/ß-turn/elastin/entropy/folding
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionReferences |
|---|
During the last decade, there have been a large number of observations on the effect of trifluoroethanol (TFE) on peptides and proteins. TFE has been shown to induce and stabilize
-helices in sequences with intrinsic helical propensity (Sönnichsen et al., 1992
) and to induce ß-turns, ß-hairpins (Sönnichsen et al., 1992; Blanco et al., 1994; Graf von Stosch et al., 1995; Searle et al., 1996) and also ß-strands (Lu et al., 1984; Martenson et al., 1985). It can also promote switching between different secondary structures, generally from a ß-sheet to an -helical structure (Yang et al., 1994; Zhang et al., 1995; Narhi et al., 1996; Kuwata et al., 1998). A transition to a 310-helix has also been reported for a ß-sheet peptide (Graf von Stosch et al., 1995). It is implicit that intra- and intermolecular side chain interactions must play an important role in solvent-induced secondary structures in proteins (Zhong and Johnson, 1992; Li and Deber, 1993). Understanding the underlying mechanisms of these induced transitions is important for the elucidation of folding pathways but particularly for the de novo design of proteins. Several models explaining the effects of TFE on secondary structures have been proposed. The two contrasting views are that either that TFE directly associates with the folded peptide chain (Jasanoff and Fersht, 1994; Bodkin and Goodfellow, 1996; Rajan and Balaram, 1996; Hirota et al., 1998) or it destabilizes the unstructured form of the peptide (Cammers-Goodwin et al., 1996, and references therein). Recent observations have focused on the interaction of TFE with water molecules on the peptide, with two differing interpretations. One view suggests that TFE behaves as an indirect chaotrope disrupting the solvent shell on -helices (Walgers et al., 1998; Luo and Baldwin, 1999), thereby stabilizing them. The second view casts TFE as a kosmotrope where its desolvating effect destabilizes the unfolded state (Kentsis and Sosnick, 1998). There are, however, several puzzling contradictions. If TFE stabilizes or strengthens local hydrophobic interactions in -helices (Albert and Hamilton, 1995; Hirota et al., 1998; Padmanabhan et al., 1998), ß-turns and ß-hairpins (Searle et al., 1996; Reiersen et al., 1998), why is there a lack of experimental evidence that TFE actually binds to hydrophobic groups (Storrs et al., 1992)? Again, TFE disrupts tertiary interactions in proteins [by weakening non-polar interactions (Gast et al., 1999; Luo and Baldwin, 1998)] while preserving secondary structures.
Here we propose a structural model that may help to account for these observations. We have studied a temperature-induced structural transition from random coil to type II ß-turn in an elastin peptide in TFE solutions at different temperatures. The mechanism of contraction of elastin peptides derives from temperature-induced folding, also seen in the folding of cold-denaturated -helices in hexafluoro-2-propanol (HFIP) (Andersen et al., 1996). For elastin this involves physical- (temperature/pressure) or chemical- (solvent composition) induced collapse of hydrophobically bound waters generating a positive entropy of bulk water and increased hydrophobic interactions, a phenomenon well known known in protein folding (Dill, 1990; Livingstone et al., 1991; Urry, 1993; Cheng and Rossky, 1998). Specifically, the elastin transition involves loss of hydrated water at the higher temperatures from exposed valine side chains inducing Val–Val interactions that fold the VPGVG sequence into a type II ß-turn (Urry, 1988a,b, 1993; Reiersen et al., 1998). In this study, increasing temperature induced a type II ß-turn at low TFE concentrations. At higher concentrations of TFE (>30%, v/v), however, the amount of ß-turn structure decreased with increasing temperature. We propose a model in which TFE associates with elastin side chains in its cluster form and thus forms a solvent matrix which supports hydrophobic interactions.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionReferences |
|---|
The peptide in this study, Ac-GVG(VPGVG)3-NH2, was synthesized, purified and analysed by CD spectroscopy as described previously (Reiersen et al., 1998
). The elastin-derived peptide Ac-GVG(VPGVG)3-NH2 incorporates a rather hydrophobic repeating (VPGVG)n sequence without any charged groups and is thus a good model for studying the effect of temperature and TFE on intramolecular hydrophobic interactions. This peptide covers more that one turn of the suggested ß-spiral in larger elastin polymers (Urry, 1988a,b) and its transition has previously been shown to be independent of concentration with reversible temperature scans (Reiersen et al., 1998). The effect of TFE at two different temperatures is shown in Figure 1a and b. On increasing the concentration of TFE at 6.8°C there was a shift in the CD spectra from a global minimum at 199 nm typical for unordered peptides to a global maximum at 206 nm with a near isodichroic point around 218 nm (Figure 1a). This latter spectrum has previously been shown to be diagnostic of type II ß-turns (Urry, 1988a,b; Perczel et al., 1993; Woody, 1995). There was a significant effect on the proportion of type II ß-turn at 6.8°C at elevated concentrations of TFE. The transition to type II ß-turn in elastin results in a closer association of the two Val groups on the cis-face of the ß-turn at higher temperatures (Urry, 1988a,b; Wasserman and Salemme, 1990; Reiersen et al., 1998). Similarly, by increasing the temperature with phosphate buffer alone, the proportion of type II ß-turn increased slightly. This was also seen for concentrations of TFE below 30% (Figure 2a and b). However, on raising the temperature the amount of ß-turn structure decreased for the two highest TFE concentrations (Figure 2c and d) and the pseudo-isodichroic point, observed as a function of TFE concentration at 58.6°C, was absent (Figure 1b). The mean residual ellipticity at 206 nm was lowered to within the same ellipticity range for 30 and 98% TFE as it was increased for the lower concentrations of TFE and phosphate buffer measured between 6.8 and 58.6°C (Figure 2a–d).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionReferences |
|---|
It is tantalizing to suggest that the similar conformational effect on the elastin peptide of increasing concentrations of TFE and increasing temperature, as illustrated in Figures 1 and 2
, derives from the same or a similar mechanism. No synergistic effect of low concentrations of TFE and increasing temperature was seen. Previously we have shown that the amount of ß-turn in elastic-like peptides increases for all TFE concentrations up to 98% (v/v) (measured at 15.7°C) (Reiersen et al., 1998), suggesting that the required Val–Val hydrophobic interactions are not inhibited but rather stimulated in TFE. However, at higher temperatures (Figure 1b), a transition to structures other than coil and type II ß-turn may still be possible since there was no clear isodichroic point.
Some type II elastin ß-turns also fold in TFE without forming the 4
1 hydrogen bond (Dyson et al., 1988, and references therein; Bisaccia et al., 1998). In the sequence VPG–Hiv–G (Hiv = S--hydroxyvaleric acid, in which no hydrogen bond is possible), the type II CD spectrum was seen in TFE and compared well with the wild-type sequence VPGVG (Arad and Goodman, 1990).
Rajan and Balaram (1996) have argued that TFE binds through its fluoro groups to hydrophobic side chains on -helices, although there is no experimental evidence from NMR for this (Storrs et al., 1992). There have been observations that TFE strengthens intrahelical hydrogen bonds (Luo and Baldwin, 1997) and weakens hydrogen bonds to solvent (Cammers-Goodwin et al., 1996). Contrariwise, D–H isotope partitioning in a GCN4 coil–coil showed that intramolecular peptide hydrogen bonds are weakened by TFE (Kentsis and Sosnick, 1998).
The foregoing suggests that the main driving force for chemical folding of VPGVG-peptides in TFE is probably not hydrogen-bond formation. Recently, Luo and Baldwin (1999) suggested a role for hydrophobic residues in shielding polar groups from solvent in the helical state, thus stabilizing the helix against thermal denaturation (Urry, 1988a,b, 1993; Reiersen et al., 1998).
Some alcohols, in particular long-chain and halogenols, form clusters that are not homogeneous in size but exist in different alcohol–water clathrate structures depending on their concentration (Kuprin et al., 1995; Hirota et al., 1998; Gast et al., 1999). This implies that the local concentration of TFE in these clusters may be many times larger than the bulk concentration of TFE. It is also a thermodynamic certainty that the effective TFE concentrations within the clusters will change as a function of temperature. This may affect thermodynamic calculations based on concentration and size of molecules.
TFE and HFIP have been shown to form clathrate structures starting from 10% HFIP or around 20% TFE (Kuprin et al., 1995; Gast et al., 1999) and at even lower concentrations (Hong et al., 1999). This implies that there are small clusters of TFE/HFIP in water with an intra-cluster alcohol concentration that is much higher than the average bulk concentration. The bell-shaped profiles of the light-scattering data show that the sizes of the clusters vary between 5 and 10 Å (Gast et al., 1999). More recently, Hong et al. (1999) have shown by X-ray scattering that clusters of alcohols such as TFE and HFIP are formed between 0 and 80% (v/v), with a maximum scattering intensity at about 30% (v/v) for HFIP (with 14 Å clusters), although the curve for TFE is weaker and much broader. This correlated well with the transitions seen for mellitin and ß-lactoglobulin in HFIP and TFE at concentrations down to ~5% (v/v) and ~10% (v/v), respectively, for ß-lactoglobulin and even down to ~2% HFIP and ~5% TFE for mellitin. This is surprising because cluster formation at such low alcohol concentrations, frequently used for studies of protein folding, has not previously been correlated with structural transitions. However, these authors did not investigate the effect of temperature on the scattering behaviour of these cosolvents.
These data, along with our own observations, may explain the effect of TFE in supporting local hydrophobic interactions in -helices, ß-hairpins and ß-turns (Padmanabhan and Baldwin, 1994; Albert and Hamilton, 1995; Searle et al., 1996; Hirota et al., 1998; Reiersen et al., 1998). The clusters may interact with local hydrophobic regions by disrupting water structures around these groups and provide a solvent matrix for further assisted side chain associations (Figure 3a and b). Note that in the elastin type II ß-turn (Reiersen et al., 1998) the valine side chains are between ~4 Å (contracted/ß-turn) and ~6 Å (extended/coil) of each other, whereas the TFE clusters are in the range 5–10 Å. The ability of TFE to interact with water molecules suggests that the TFE clusters can 'pull' water from the surface, at least as an initial step. This may especially be relevant at lower alcohol concentrations (0–10%) where the TFE clusters are not fully developed or stabilized. However, in subsequent steps the clusters may also directly associate with appropriate hydrophobic side chains. Such a matrix may be important to lower the side chain conformational entropy, which is thought to be a key factor in the folding of -helices (Aurora et al., 1997). The differences in cluster size and composition, dependent on TFE concentration, may explain why TFE switches secondary structure conformations at concentrations >50%. At elevated concentrations the TFE clusters are smaller, but have a higher local intra-cluster TFE concentration (Figure 3c). This provides a better matrix to redirect non-local side chain interactions in an existing structure to a more local folding nucleus. As the majority of TFE-induced transitions occurs from a non-local (ß-sheets; Otzen and Fersth, 1995) to a local folding contact (helices, turns and hairpins; Thomas and Dill, 1993), the TFE-induced transition can easily be explained by the cluster model (Figure 3a and b).
|
The destabilization of tertiary and quaternary structures in proteins originates from the same phenomenon: by favouring the stabilization of short-range hydrophobic contacts the long-range hydrophobic interactions are systematically disrupted, resulting in denaturation of the non-polar core (Hirota et al., 1998
).
The observation in this study that concentrations of TFE 
30% have the opposite effect to low concentrations of TFE at higher temperatures (Figure 2a and b versus Figure 2c and d) suggests that the cluster structures are altered in composition and size at higher temperatures. At least for tert-butyl alcohol (Iwasaki and Fujiyama, 1979) and hexafluoro-2-propanol (Kuprin et al., 1995), both light scattering and low-angle X-ray scattering, respectively, have been shown to increase with temperature, possibly owing to the increased segregation of the alcohol from water (Iwasaki and Fujiyama, 1979; Kuprin et al., 1995). Interestingly, the scattering of clusters containing low concentrations of tert-butyl alcohol is lowered by increasing temperature (Iwasaki and Fujiyama, 1979). At higher temperatures it would be expected that the local TFE concentration in the clusters would increase owing to a more positive mixing enthalpy between alcohol and water and that the cluster will become more hydrophobic, ultimately leading to a macroscopic phase separation (Iwasaki and Fujiyama, 1979). This provides an explanation for the thermo-stabilization of helices whereby the TFE cluster stabilizes the helix side chains by offering a solvent matrix which can interact with the helix surface.
What can be said about cold denaturation? Cold denaturation of -helices in 8–10% hexafluoro-2-propanol has previously been described as a destabilization of the coil state (Andersen et al., 1996). By reducing the temperature the density of HFIP clusters will decrease, as seen by Kuprin et al. (1995), and will therefore diminish the ability of the HFIP matrix to provide an effective local environment for the helix side chains.
The importance of having an external supporting interface in the folding and stability of proteins has been well addressed in the literature. An exposed hydrophobic surface is important for chaperone function and activity (Das and Surewicz, 1995) and -helices are generally stabilized by tertiary hydrophobic contacts with other parts of the protein. The helix-stabilizing effect by reducing the non-polar surface area has been stressed in several previous studies (Padmanabhan and Baldwin, 1994, and references therein; Zhang et al., 1995; Butcher and Moe, 1996).
On the basis of the observations that TFE forms clusters in water solutions, we have proposed here a structural model in which TFE forms a temperature-sensitive solvent matrix that locally assists hydrophobic side chains of secondary structures. This model accounts for the various effects TFE on hydrophobic groups in peptides and proteins and may provide an explanation for the phenomenon of cold denaturation.
|
Notes |
|---|
1 To whom correspondence should be addressed. Present address while on academic leave: Syntem, Parc Scientifique Georges Besse, 30000 Nimes, France. E-mail: arrees{at}compuserve.com
|
Acknowledgments |
|---|
We thank Dr A.R.Clarke and the Department of Biochemistry, School of Medical Sciences, University of Bristol for giving us access to a CD instrument and Dr G.Bloomberg of the School of Medical Sciences, University of Bristol for amino acid analysis. This work was supported by grants from the Research Council of Norway (No. 110859/410) and DYNAL, Oslo, Norway.
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionReferences |
|---|
Albert,J.S. and Hamilton,A.D. (1995) Biochemistry, 34, 984–990.[Medline]
Andersen,N.H., Cort,J.R., Liu,Z., Sjöberg,S.J. and Tong,H. (1996) J. Am. Chem. Soc., 118, 10309–10310.
Arad,O. and Goodman,M. (1990) Biopolymers, 29, 1651–1668.
Aurora,R., Creamer,T.P., Srinivasan,R. and Rose,G.D. (1997) J. Biol. Chem., 272, 1413–1416.
Bisaccia,F., Castiglione-Morelli,M.A., Spisani,S., Ostuni,A., Serafini-Fracassini,A., Bavoso,A. and Tamburro,A.M. (1998) Biochemistry, 37, 11128–11135.[Medline]
Blanco,F.J., Jimenez,M.A., Pineda,A., Rico,M., Santoro,J. and Nieto,J.L. (1994) Biochemistry, 33, 6004–6014.[Medline]
Bodkin,M.J. and Goodfellow,J.M. (1996) Biopolymers, 39, 43–50.[Web of Science][Medline]
Butcher,D.J. and Moe,G.R. (1996) Proc. Natl Acad. Sci. USA, 93, 1135–1140.
Cammers-Goodwin,A., Allen,T.J., Oslick,S.L., McClure,K.F., Lee,J.H. and Kemp,D.S. (1996) J. Am. Chem. Soc., 118, 3082–3090.
Cheng,Y.-K. and Rossky,P.J. (1998) Nature, 392, 696–699.[Medline]
Das,K.P. and Surewicz,W.K. (1995) FEBS Lett., 369, 321–325.[Web of Science][Medline]
Dill,K.A. (1990) Biochemistry, 29, 7133–7155.[Medline]
Dyson,H.J., Rance,M., Houghten,R.A., Lerner,R.A. and Wright,P.E. (1988) J. Mol. Biol., 201, 161–200.[Web of Science][Medline]
Gast,K., Zirwer,D., Müller-Frohne,M. and Damaschun,G. (1999) Protein Sci., 8, 625–634.[Web of Science][Medline]
Graf von Stosch,A., Jiménez,M.A., Kinzel,V. and Reed,J. (1995) Proteins: Struct. Funct. Genet., 23, 196–203.[Web of Science][Medline]
Hirota,N., Mizuno,K. and Goto,Y. (1998) J. Mol. Biol., 275, 365–378.[Web of Science][Medline]
Hong,D.-P., Hoshino,M., Kuboi,R. and Goto,Y. (1999) J. Am. Chem. Soc., 121, 8427–8433.
Iwasaki,K. and Fujiyama, T (1979) J. Phys. Chem., 83, 463–468.
Jasanoff,A. and Fersht,A.R. (1994) Biochemistry, 33, 2129–2135.[Medline]
Kentsis,A. and Sosnick,T.R. (1998) Biochemistry, 37, 14613–14622.[Medline]
Kuprin,S., Gräslund,A., Ehrenberg,A. and Koch,M.H.J. (1995) Biochem. Biophys. Res. Commun., 217, 1151–1156.[Web of Science][Medline]
Kuwata,K., Hoshino,M., Era,S., Batt,C.A. and Goto,Y. (1998) J. Mol. Biol., 283, 731–739.[Web of Science][Medline]
Li,S.-C. and Deber,C.M. (1993) J. Biol. Chem., 31, 22975–22978.
Livingstone,J.R., Spolar,R.S. and Record,M.T., Jr. (1991) Biochemistry, 30, 4237–4244.[Medline]
Lu,Z.X., Fok,K.F., Erickson,B.W. and Hugli,T.E. (1984) J. Biol. Chem., 259, 7367–7370.
Luo,P. and Baldwin,R.L. (1997) Biochemistry, 36, 8413–8421.[Medline]
Luo,P. and Baldwin,R.L. (1999) 1999) Proc. Natl Acad. Sci. USA 96, 4930–4935.
Luo,Y. and Baldwin,R.L. (1998) J. Mol. Biol., 279, 49–57.[Web of Science][Medline]
Martenson,R.E., Park,J.Y. and Stone,A.L. (1985) Biochemistry, 24, 7689–7695.[Medline]
Narhi,L.O., Philo,J.S., Li,T., Zhang,M., Samal,B. and Arakawa,T. (1996) Biochemistry, 35, 11447–11453.[Medline]
Otzen,D.E. and Fersht,A.R. (1995) Biochemistry, 34, 5718–5724.[Medline]
Padmanabhan,S. and Baldwin,R.L. (1994) J. Mol. Biol., 241, 706–713.[Web of Science][Medline]
Padmanabhan,S., Jiménez,M.A., Laurents,D.V. and Rico,M. (1998) Biochemistry, 37, 17318–17330.[Medline]
Perczel,A., Hollósi,M., Sándor,P. and Fasman,G.D. (1993) Int. J. Pept. Protein Res., 41, 223–236.[Web of Science][Medline]
Rajan,R. and Balaram,P. (1996) Int. J. Pept. Protein Res., 48, 328–336.[Web of Science][Medline]
Reiersen,H., Clarke,A.R. and Rees,A.R. (1998) J. Mol. Biol., 283, 255–264.[Web of Science][Medline]
Searle,M.S., Zerella,R., Dudley,D.H. and Packman,L.C. (1996) Protein Eng., 9, 559–565.
Sönnichsen,F.D., Van Eyk,J.E., Hodges,R.S. and Sykes,B.D. (1992) Biochemistry, 31, 8790–8798.[Medline]
Storrs,R.W., Truckses,D. and Wemmer,D.E. (1992) Biopolymers, 32, 1695–1702.[Web of Science][Medline]
Thomas,P.D. and Dill,K.A. (1993) Protein Sci., 2, 2050–2065.[Web of Science][Medline]
Urry,D.W. (1988a) J. Protein Chem., 7, 1–34.[Web of Science][Medline]
Urry,D.W. (1988b) J. Protein Chem., 7, 81–114.[Web of Science][Medline]
Urry,D.W. (1993) Angew. Chem., Int. Ed. Engl., 32, 819–841.
Walgers,R., Lee,T.C. and Cammers-Goodwin,A. (1998) J. Am. Chem. Soc., 120, 5073–5079.
Wasserman,Z.R. and Salemme,F.R. (1990) Biopolymers, 29, 1613–1631.[Web of Science][Medline]
Woody,R.W. (1995) Methods Enzymol., 246, 34–71.[Web of Science][Medline]
Yang,J.J., Pitkeathly,M. and Radford,S.E. (1994) Biochemistry, 33, 7345–7353.[Medline]
Zhang,H., Kaneko,K., Nguyen,J.T., Livshits,T.L., Baldwin,M.A., Cohen,F.E., James,T.L. and Prusiner,S.B. (1995) J. Mol. Biol., 250, 514–526.[Web of Science][Medline]
Zhong,L. and Johnson,W.C., Jr. (1992) Proc. Natl Acad. Sci. USA, 89, 4462–4465.
Received July 3, 2000; revised October 11, 2000; accepted October 11, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Tada, T. Aizawa, Y. Shinohara, K. Matsubara, K. Miura, M. Yoshida, K. Shitara, T. Kouno, M. Mizuguchi, K. Nitta, et al. Roles of Aromatic Residues in the Structure and Biological Activity of the Small Cytokine, Growth-blocking Peptide (GBP) J. Biol. Chem., March 14, 2003; 278(12): 10778 - 10783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Roccatano, G. Colombo, M. Fioroni, and A. E. Mark Mechanism by which 2,2,2-trifluoroethanol/water mixtures stabilize secondary-structure formation in peptides: A molecular dynamics study PNAS, September 17, 2002; 99(19): 12179 - 12184. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||