13C-Labeled Tyrosine Residues as Local IR Probes for Monitoring Conformational Changes in Peptides and Proteins.

July 17, 2018 | Author: Anonymous | Category: Каталог , Без категории
Share Embed


Short Description

Download 13C-Labeled Tyrosine Residues as Local IR Probes for Monitoring Conformational Changes ...

Description

Angewandte

Chemie

Protein Folding 13

C-Labeled Tyrosine Residues as Local IR Probes for Monitoring Conformational Changes in Peptides and Proteins** Sandra Tremmel,* Michael Beyermann, Hartmut Oschkinat, Michael Bienert, Dieter Naumann, and Heinz Fabian* Fourier transform infrared (FTIR) spectroscopy has become a valuable tool for studying the secondary structure of proteins and for monitoring conformational changes associated with protein unfolding, folding, and misfolding.[1–3] Most studies in the past have focused on components of the intense amide I band (primarily the C=O stretch of the secondary amide group) in the IR spectra between 1610 and 1700 cm1 as established indicators of protein secondary structure. Besides the amide I band of the peptide backbone, several amino acid side-chain absorption bands may serve as intrinsic local monitors of conformational changes. Very useful is the relatively strong absorption band of the tyrosine side chain (CC stretching vibration of the aromatic ring) at  1515 cm1, which can be distinguished particularly well in the infrared spectra of proteins.[4] The frequency of this band may change in a characteristic manner upon unfolding, folding, and aggregation of proteins. For proteins whose unfolding is reversible, a discontinuous high-frequency shift of the tyrosine vibration around the corresponding denaturation temperatures was observed.[5–7] In a structureless peptide, the tyrosine band displayed only a minor gradual shift to lower wavenumbers upon increasing temperature, reflecting a general effect of temperature.[5] A relatively large downshift of the tyrosine vibration was observed for proteins upon aggregation, which has been suggested as indicative of stronger hydrogen bonds of the tyrosine OH group with acceptors in the protein aggregates.[6, 8] However, if the corresponding peptide or protein contains more than one tyrosine residue, the site-specific information is limited. Removal of all but one of the tyrosine residues by sitedirected mutagenesis is one possible solution to this problem. However, if the corresponding residues are, for example, part [*] S. Tremmel, Dr. M. Beyermann, Prof. H. Oschkinat, Prof. M. Bienert Forschungsinstitut f r Molekulare Pharmakologie Robert-R$ssle-Strasse 10, 13125 Berlin (Germany) Fax: (+ 49) 30-9479-3159 E-mail: [email protected] Prof. D. Naumann, Dr. H. Fabian Robert-Koch-Institut Nordufer 20, 13353 Berlin (Germany) Fax: (+ 49) 30-4547-2606 E-mail: [email protected] [**] The work at the Forschungsinstitut f r Molekulare Pharmakologie was supported by the Deutsche Forschungsgemeinschaft (DFGFOR 299). We thank Annerose Klose and Bernhard Schmikale for their assistance with the peptide synthesis. Supporting information for this article is available on the WWW under http://www.angewandte.org or from the author. Angew. Chem. Int. Ed. 2005, 44, 4631 –4635

DOI: 10.1002/anie.200500547

 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

4631

Communications served as model systems for studies of b-sheet stability and folding.[11–19] Among them, the FBP28WW domain was reported to be monomeric and stable, although under certain conditions aggregation and even fibril formation have been observed.[18] The preparation of WW analogues, including the incorporation of labeled amino acids, should be feasible by solidphase peptide synthesis (SPPS).[20] However, initial attempts to synthesize wild-type FBP28WW (peptide 1) by SPPS failed, which is not unexpected for the assembly of b-sheetforming sequences.[21] Fortunately, the synthesis could be improved considerably by insertion of two X-Ser and X-Thr dipeptides (X = pseudo-proline) in the form of the corresponding pseudo-proline derivatives.[22] In addition, aspartate in position 15 was replaced by asparagine (peptide 2) in order to minimize the high propensity for aspartimide formation[23] of the Asp15-Gly16 unit (Scheme 1; see the Supporting Information). The low-temperature IR spectra (Figure 1 a, blue traces) of the [Asn15]-FBP28WW variant (peptide 2) recorded between 10 and 90 8C during the course of a thermal denaturation experiment (see the Supporting Information) are characterized by a major amide I band centered at 1636 cm1 and a weaker band component at 1679 cm1, indicating the presence of antiparallel b-sheet structures.[1–3] The infrared bands between 1500 and 1612 cm1 are due to absorptions of amino acid side chains: tyrosine (1515 and 1612 cm1), glutamate (1565 cm1), and arginine (1587 cm1).[4] The amide I region of the spectrum recorded at 90 8C is dominated by bands at 1616 and 1685 cm1 (Figure 1 a, black trace). These bands are characteristic of intermolecular b-sheet structures,[1–3] indicating that thermal denaturation of the [Asn15]-FBP28WW variant involves aggregation. During our search for FBP28WW analogues with a reduced tendency toward aggregation, we surprisingly found that insertion of Gln residues in certain positions was very Scheme 1. Structural features of the FBP28WW domain. The sequence 6–33 (gray) successful (peptide 3, Table 1). Peptide 3 displayed the typical forms a defined three-dimensional structure, including three strands and two [11] fold of the FBP28WW domains when it was characterized by loops. The locations for the application of pseudo-proline building blocks to NMR spectroscopy[27] and did not form intermolecular benable the successful synthesis and the positions for 13C1(4)-tyrosine labeling are sheet structures at high temperatures, which is indicated by indicated (Oxa = Ser/Thr-derived oxazolidine-4-carboxylic acid). the lack of the characteristic IR “aggregate” bands at 1616 and 1685 cm1. Instead, the high-temperature spectra of Table 1). WW domains are the smallest naturally occurring peptide 3 are dominated by a broad and featureless amide I three-stranded b sheets identified so far. They are named band centered at 1648 cm1 (Figure 1 b), indicative of strucafter the two conserved tryptophan residues and function as tureless proteins.[1–3] This is also valid for peptide 4 (spectra noncatalytic domains of signaling proteins by recognizing not shown), in which additionally two Tyr residues have been proline-containing ligands.[10] Recently, several WW domains replaced by Phe in order to permit the assignment of tyrosinemediated spectroscopic changes to a single Tyr residue. The temperature-induced Table 1: Sequences of FBP28WW peptides (*Y = 13C1(4)-Tyr). decrease in intensity of the low-frequency component of the amide band at 1636 cm1 Peptide Sequence (Figure 2 a) and the frequency shift of the FBP28WW 1 GATAVSEWTEYKTADGKTYYYNNRTLESTWEKPQELK 15 [a] stretching vibration of tyrosineDs aromatic [N ]-FBP28WW 2 GATAVSEWTEYKTANGKTYYYNNRTLESTWEKPQELK 9,21,23,26,28 15 [a] ring at  1515 cm1 (Figure 2 b) revealed [Q ,N ]-FBP28WW 3 GATAVSEWQEYKTANGKTYYQNQRTQEQTWEKPQELK 9,21,23,26,28 11,20 15 [Q ,F ,N ]-FBP28WW 4 GATAVSEWQEFKTANGKTYFQNQRTQEQTWEKPQELK sigmoidal unfolding transitions between 10 5 GATAVSEWQE *YKTANGKTYYQNQRTQEQTWEKPQELK [*Y11,Q9,21,23,26,28,N15]-FBP28WW[a] and 90 8C for the two peptides. [*Y19,Q9,21,23,26,28,N15]-FBP28WW[a] 6 GATAVSEWQEYKTANGKT *YYQNQRTQEQTWEKPQELK The Tyr!Phe substitution, however, decreased the denaturation temperature [a] Amidated at the C terminus. of the hydrophobic core of the protein, mutating them may destabilize the protein and/or affect its function. Herein we demonstrate that an ideal alternative to site-directed mutagenesis is site-directed isotope labeling, such as the introduction of individual tyrosine residues labeled with 13C at the C4 position of the phenol ring. Consequently site-specific structural specificity is added to the spectrum since the vibration of the labeled residue is shifted by  10 cm1,[9] thus providing a well separated unique spectroscopic signature. While isotopic labeling of tyrosines to probe proteins by IR spectroscopy is not new, the current application to obtain information on the microenvironment of distinct tyrosine side chains and to monitor simultaneously local conformational changes in different regions of proteins is novel. To explore the potential of site-directed isotope labeling, we studied changes in the infrared-spectroscopic behavior of distinct tyrosine residues caused by the thermal unfolding of the formin-binding protein 28 (FBP28) WW domain. This protein contains four tyrosine residues, one located in strand 1 (position 11), whereas the remaining three (at positions 19, 20, and 21) are located in strand 2 (Scheme 1, peptide 1 in

4632

 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.angewandte.org

Angew. Chem. Int. Ed. 2005, 44, 4631 –4635

Angewandte

Chemie

by  5 K,[24] which demonstrates a drawback of the mutation approach. Therefore, we labeled peptide 3 with 13C1(4)-Tyr residues site-specifically. In order to probe simultaneously temperature-induced changes in the local environment of Tyr11 located in strand 1 of the WW domain (see Scheme 1), and of Tyr19, a highly conserved residue of the second strand of WW proteins,[11] we synthesized the labeled peptides 5 and 6. As expected, selective 13C1(4)-labeling of Tyr11 (Figure 3, middle trace) or of Tyr19 (Figure 3, upper trace) led to a drop of intensity at 1515 cm1 and the appearance of a new band at 1505 cm1.

Figure 1. Deconvolved infrared spectra of a) peptide 2 and b) peptide 3 in D2O buffer (50 mm phosphate, 50 mm sodium chloride, pD 7.6) as a function of temperature. Blue traces, 10–40 8C; green traces, 45– 70 8C; red traces, 75–85 8C; black traces, 90 8C.

Figure 3. Deconvolved infrared absorbance spectra of unlabeled (peptide 3, lower trace), site-specifically 13C1(4)-Tyr11-labeled (peptide 5, middle trace), and site-specifically 13C1(4)-Tyr19-labeled (peptide 6, upper trace) FBP28WW peptides in D2O buffer (50 mm phosphate, 50 mm sodium chloride, pD 7.6) at 10 8C.

Figure 2. a) Temperature dependence of the peak intensity of the amide I band at 1636 cm1 and b) temperature dependence of the position of the tyrosine band for peptide 3 (&) and peptide 4 (?). The peak intensities at 1636 cm1 were normalized relative to the signal obtained at 90 8C. The solid lines represent the best fit used to estimate the denaturation temperatures (Tm). Angew. Chem. Int. Ed. 2005, 44, 4631 –4635

The temperatures of thermal denaturation, derived from the IR bands of Tyr11 in strand 1 and of Tyr19 in strand 2, were found to be practically identical,[25] suggesting no major differences in the thermal response of the conformational changes in both strands of the FBP28WW peptide 3. An inspection of the transition profiles, however, revealed striking differences between the bands of the labeled Tyr11 and Tyr19 residues (Figure 4 b). While the band due to Tyr11 displayed only minor frequency changes, a relatively large shift to higher frequencies by  0.6 cm1 was observed for Tyr19. At high temperatures, the peak positions were practically identical, as expected for an unfolded peptide. Moreover, the frequency changes of less than 0.2 cm1 due to labeled Tyr11 observed around the thermal denaturation temperature of peptide 5 (Figure 4 b) were comparable to the minor mean spectroscopic changes due to unlabeled Tyr11 and Tyr20 in peptide 6 (Figure 4 a), suggesting a very similar temperature response of Tyr11 and Tyr20. Altogether, this indicates that in the folded state of the peptides the local

www.angewandte.org

 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

4633

Communications improve our insights into the role of the tyrosine residues in the stabilization and unfolding of the b-sheet structure of FBP28WW and other WW domains. Received: February 14, 2005 Published online: June 24, 2005

.

Keywords: IR spectroscopy · isotope effects · peptide synthesis · protein folding · WW domains

Figure 4. Temperature dependence of the position of the tyrosine bands of site-specifically 13C1(4)-Tyr11- and 13C1(4)-Tyr19-labeled [Gln9,21,23,26,28,Asn15]-FBP28WW. a) Positions of the unlabeled tyrosines of peptide 5 (&) and of peptide 6 (?). b) Positions of the labeled tyrosine of peptide 5 (?) and of peptide 6 (&).

environments of Tyr19, on one side, and of Tyr11/Tyr20, on the other side, must differ considerably. A plausible explanation for a strong shift of the tyrosine band to higher frequencies upon protein unfolding would be a strengthening of tyrosine OH bonds, caused by loosening of hydrogen bonds between the tyrosine OH group and neighboring acceptors.[5, 6] The available information on the FBP28WW structure, however, provides no hints for tyrosine OH groups being involved in hydrogen bonding. This suggests that effects other than hydrogen bonding may also affect the spectroscopic features of the corresponding ring vibration. A possible explanation may be altered p–p interactions between tyrosine rings and conjugated systems of other residues.[6, 26] In the FBP28WW domain Trp30 is close to Tyr19.[11] Thus, the temperature-dependent shift of the Tyr19 side-chain band observed herein might reflect a disruption of p–p interactions between Tyr19 and the indole ring of Trp30. If so, than the infrared band of Tyr19 provides a local probe of specific interactions between residues located in strands 2 and 3 of the WW domain. In summary, we have demonstrated that FTIR spectroscopy along with site-directed 13C1(4)-labeling of tyrosine residues is a promising new approach to obtain information on the microenvironment of distinct tyrosine side chains and to monitor simultaneously temperature-induced local conformational changes in different regions of a given protein. In combination with time-resolved IR-spectroscopic techniques, this strategy offers novel possibilities for studying protein unfolding and folding events at the level of individual groups in a peptide or protein. We are currently conducting both peptide synthesis and spectroscopic studies in order to

4634

 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

[1] M. Jackson, H. H. Mantsch, Crit. Rev. Biochem. Mol. Biol. 1995, 30, 95 – 120. [2] H. Fabian, W. MGntele in Handbook of Vibrational Spectroscopy. Infrared Spectroscopy of Proteins (Eds.: J. M. Chalmers, P. R. Griffiths), Wiley, Chichester, 2002, pp. 3399 – 3425. [3] A. Barth, C. Zscherp, Q. Rev. Biophys. 2002, 35, 369 – 430. [4] Y. N. Chirgadze, O. V. Fedorov, N. P. Trushina, Biopolymers 1975, 14, 679 – 694. [5] H. Fabian, C. Schultz, J. Backmann, U. Hahn, W. Saenger, H. H. Mantsch, D. Naumann, Biochemistry 1994, 33, 10 725 – 10 730. [6] C. Zscherp, H. AygLn, J. W. Engels, W. MGntele, Biochim. Biophys. Acta 2003, 1651,139 – 145. [7] J. Torrent, P. Rubens, M. Ribo, K. Heremans, M. Vilanova, Protein Sci. 2001, 10, 725 – 734. [8] J. L. R. Arrondo, N. M. Young, H. H. Mantsch, Biochim. Biophys. Acta 1988, 952, 261 – 268. [9] R. Hienerwadel, A. Boussac, J. Breton, B. A. Diner, C. Berthomieu, Biochemistry 1997 36, 14 712 – 14 723. [10] a) M. Sudol, T. Hunter, Cell 2000, 103, 1001 – 1004; b) L. J. Ball, R. KLhne, J. Schneider-Mergener, H. Oschkinat, Angew. Chem. 2005, 117, 2912- – 2930; Angew. Chem. Int. Ed. 2005, 44, 2852 – 2869. [11] M. J. Macias, V. Gervais, C. Civera, H. Oschkinat, Nat. Struct. Biol. 2000, 7, 375 – 379. [12] R. Kaul, A. R. Angeles, M. JGger, E. T. Powers, J. W. Kelly, J. Am. Chem. Soc. 2001, 123, 5206 – 5212. [13] M. JGger, H. Nguyen, J. C. Crane, J. W. Kelly, M. Gruebele, J. Mol. Biol. 2001, 311, 373 – 393. [14] a) G. T. Ibragimova, R. C. Wade, Biophys. J. 1999, 77, 2191 – 2198; b) N. Ferguson, J. R. Pires, F. Toepert, C. M. Johnson, Y. P. Pan, R. Volkmer-Engert, J. Schneider-Mergener, V. Daggett, H. Oschkinat, Proc. Natl. Acad. Sci. USA 2001, 98, 13 008 – 13 013. [15] N. Ferguson, C. M. Johnson, M. Macias, H. Oschkinat, A. Fersht, Proc. Natl. Acad. Sci. USA 2001, 98, 13 002 – 13 007. [16] J. Karanicolas, C. L. Brooks III, Proc. Natl. Acad. Sci. USA 2003, 100, 3954 – 3959. [17] H. Nguyen, M. JGger, A. Moretto, M. Gruebele, J. W. Kelly, Proc. Natl. Acad. Sci. USA 2003, 100, 3948 – 3953. [18] N. Ferguson, J. Berriman, M. Petrovich, T. D. Sharpe, J. T. Finch, A. R. Fersht, Proc. Natl. Acad. Sci. USA 2003, 100, 9814 – 9819. [19] S. Deechongkit, H. Nguyen, E. T*. Powers, P. E. Dawson, M. Gruebele, J. W. Kelly, Nature 2004, 430, 101 – 105. [20] G. B. Fields, R. L. Noble, Int. J. Pept. Protein Res. 1990, 35, 161 – 214. [21] a) M. Beyermann, M. Bienert, Tetrahedron Lett. 1992, 33, 3745 – 3748; b) M. Dettin, S. Pegoraro, P. Rovero, S. Bicciato, A. Bango, C. Di Bello, J. Pept. Res. 1997, 49, 103 – 111; c) C. Hyde, T. Johnson, D. Owen, M. Quibell, R. C. Sheppard, Int. J. Pept. Protein Res. 1994, 43, 431 – 440. [22] a) T. Haack, M. Mutter, Tetrahedron Lett. 1992, 33, 1589 – 1592; b) T. WPhr, F. Wahl, A. Nefzi, B. Rohwedder, T. Sato, X. Sun, M. Mutter, J. Am. Chem. Soc. 1996, 118, 9218 – 9227; c) W. R. Sampson, H. Patsiouras, H. J. Ede, J. Pept. Sci. 1999, 5, 403 – 409. [23] a) E. Nicolas, E. Pedroso, E. Giralt, Tetrahedron Lett. 1989, 30, 497 – 500; b) R. DPlling, M. Beyermann, J. Haenel, F. Kernchen,

www.angewandte.org

Angew. Chem. Int. Ed. 2005, 44, 4631 –4635

Angewandte

Chemie

[24]

[25]

[26]

[27]

E. Krause, P. Franke, M. Brudel, M. Bienert, J. Chem. Soc. Chem. Commun. 1994, 853 – 854; c) Y. Yang, W. V. Sweeney, K. Schneider, S. Thornqvist, B. T. Chait, J. P. Tam, Tetrahedron Lett. 1994, 35, 9689 – 9692. The denaturation temperatures (Tm) of the FBP28WW variants were estimated either from the intensity/temperature plot of the b band at 1636 cm1 and from the frequency/temperature changes of the tyrosine-ring vibration at 1515 cm1 (in brackets); peptide 3: 50.9 (52.8) 8C, peptide 4: 46.2 (46.4) 8C. The denaturation temperatures of strand 1 and strand 2 of the [Gln9,21,23,26,28,Asn15] FBP28WW variant were estimated from the frequency/temperature changes of the corresponding bands of 13 C-labeled tyrosine at 1505 cm1. For comparison, the Tm (8C) values calculated from the frequency/temperature changes of the unlabeled tyrosine bands at 1515 cm1 are also shown: Tyr11 (strand 1): 51.5; Tyr19 (strand 2): 51.1; Tyr11/20: 52.3; Tyr19/20: 51.1. IR studies of the histone-like protein of Bacillus subtilis, HBsu, revealed that also the absorption band of the phenylalanine side chain at 1498 cm1 may show a pronounced discontinuous highfrequency shift around the denaturation temperature: H. Fabian, unpublished results. J. H. Holtmann, H. Oschkinat, unpublished results.

Angew. Chem. Int. Ed. 2005, 44, 4631 –4635

www.angewandte.org

 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

4635

View more...

Comments

Copyright © 2017 UPDOC Inc.