ISSN 1068-1620, Russian Journal of Bioorganic Chemistry, 2008, Vol. 34, No. 5, pp. 578–585. © Pleiades Publishing, Ltd., 2008. Original Russian Text © D.A. Prokhorov, M.A. Timchenko, Yu.A. Kudrevatykh, D.V. Fedyukina, L.V. Gushchina, V.S. Khristoforov, V.V. Filimonov, V.P. Kutyshenko, 2008, published in Bioorganicheskaya Khimiya, 2008, Vol. 34, No. 5, pp. 645–653.
Study of the Structure and Dynamics of a Chimeric Variant of the SH3 Domain (SHA-Bergerac) by NMR Spectroscopy D. A. Prokhorova, M. A. Timchenkoa, Yu. A. Kudrevatykha, D. V. Fedyukinaa, L. V. Gushchinab, V. S. Khristoforova, V. V. Filimonovb, and V. P. Kutyshenkoa,1 a
Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, pr. Nauki, Pushchino, Moscow oblast, 142290 Russia b Institute of Protein Research, Russian Academy of Sciences, Pushchino, Moscow oblast, 142290 Russia Received November 12, 2007; in final form, December 28, 2007
Abstract—A structural-dynamic study of one of the chimeric proteins (SHA) belonging to the SH3-Bergerac family and containing the KATANGKTYE sequence instead of the N47D48 β-turn in the spectrin SH3-domain was carried out by high resolution NMR spectroscopy. The spatial structure of the protein was determined and its dynamics in solution was investigated on the basis of the NMR data. The elongation of the SHA polypeptide chain in comparison with the WT-SH3 original protein (by ~17%) exerts practically no effect on the general topology of the molecule. The presence of a stable β-hairpin in the region of insertion was confirmed. This hairpin was shown to have a higher mobility in comparison with other regions of the protein. Key words: SH3-domain, SHA-Bergerac protein, spatial structure, dynamics, NMR DOI: 10.1134/S1068162008050075
INTRODUCTION Protein self-assembly is one of the central and topical problems of modern biology that has discussed in many articles devoted to the process of protein folding in a unique spatial structure called as native.2 This structure corresponds, as a rule, to an energy minimum, which is spontaneously achieved by a polypeptide chain, and, therefore, this problem has three basic aspects: structural, thermodynamic, and kinetic. Formation of transient states, secondary structure, and folding nucleus of a protein and quantitative determination of driving forces of this process are of special interest. “Minimal” globular proteins that can form stable spatial structure in the absence of additional stabilizing factors (such as disulfide bridges) are optimal for studying mechanisms of the self-assembly. Isolated SH3 domains and their mutant analogues are widely used as 1
Corresponding author; phone: +7 (4967) 73-9226; fax: +7 (4967) 33-0553; e-mail:
[email protected]. 2 Abbreviations: SH3, Src-homologous domain 3; Src, tyrosine kinase contained in the genome of virus of the Rous sarcoma; WT-SH3, recombinant protein corresponding to the sequence of the spectrin domain of a wild type; SHA, chimeric variant of WTSH3; NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser enhancement spectroscopy; COSY, correlation spectroscopy; TOCSY, total correlation spectroscopy; HSQC, heteronuclear single-quantum correlation spectroscopy; PDB, Protein Databank structures; RMSD, root-mean-square deviation; τm, time of correlation of a protein as a whole.
convenient subjects for physicochemical and structural studies of β-structural proteins by various methods [1−3]. Natural SH3 domains are incorporated in large multidomain proteins (tyrosine kinases), which participate in signal transduction and other cellular processes. Many of these processes are associated with the development of various diseases [4–8]. An SH3 domain as a structural mediator module recognizes and interacts with proline-rich sequences that belong to both out-ofdomain regions of self polypeptide chain and to protein partners [9–12]. PDB contains more than 30 structures of the SH3 domains determined on the basis of X-ray analysis and NMR spectroscopy. Nevertheless, functions of SH3 domains in binding of the proline-rich protein sites, structure-thermodynamic binding determinants, and specificity of these interactions are still not elucidated in detail in spite of significant amount of accumulated structural information. The transient state (folding nucleus) of SH3 domains has a conserved structure [13]. Its most ordered element is the so-called distal loop. More than ten chimeric variants of the SH3 domain (including the SHA domain) in which various decapeptides substituted for its 47–48 β-turn were created for detailed studies of thermodynamics and kinetics of formation of the protein structure [14]. It was believed that such insertions would increase stability of the β-hairpin, and it will protrude from the domain like a “nose”. This
578
STUDY OF THE STRUCTURE AND DYNAMICS WT-SH3
β1
β2
RT
n-src
β3
579 β4
310 β5
(‡) (b)
1 5 30 40 45 50 55 60 10 15 20 25 35 M D E T G K E L V L A L Y D Y Q E K S P R E V T M K K G D I L T L L N S T N K D WW K V E V ND R Q G F V P A A Y V K K L D 20 1 10 15 25 30 35 60 70 5 40 45 65 SHA K A T A NG K T Y E 55 50
Fig. 1. (a) Scheme of the secondary structure of the WT-SH3 protein according to X-ray analysis [16]. The β-strands are designated by arrows. According to the nomenclature [13], the 14–27, 31–46, and 41–54 sequences are involved in the RT-loop, the n-src-loop, and the distal loop, respectively. (b) Amino acid sequences of (the upper row of numbers) the WT-SH3 (62 amino acid residues) and (the bottom row of numbers) the chimeric SHA-Bergerac protein (70 amino acid residues). Insertion of ten amino acid residues is bolded. The β-turns are printed in italics.
family of chimeric proteins was called as Bergerac. The structure of SHA variation turned out not to be the most stable, but it had a high solubility [14]. This property of SHA domain facilitates the structural studies by NMR spectroscopy. Sufficient amount of thermodynamic and kinetic information has been accumulated for proteins of the Bergerac family at present [14, 15], but this information should be adequately interpreted. In this study, we determined spatial structure of chimeric recombinant SHA protein by heteronuclear NMR spectroscopy and studied its dynamics. RESULTS AND DISCUSSION Polypeptide chain of the spectrin SH3 domain consists of 62 amino acid residues [16]. In the case of SHA, the natural β-hairpin is elongated by four pairs of amino acid residues, and the β-turn is modified by substitution of glycine for aspartic acid (Fig. 1b). It is known that isolated β-hairpins in short peptides with free termini are incompletely folded in solution even at room temperature. The content of secondary structure is no higher than 30–40% at best. The insertion that is not incorporated into a protein globule might be expected to behave similarly. Earlier, the D48G substitution in the wild type protein was found to increase stability of its native structure by 4 kJ/mol, and melting point of the protein increased from 66 to 74°ë at pH 4.0 [14]. The NG sequence was incorporated into SHA as a β-turn on the basis of this fact (Fig. 1b). The remaining eight amino acid residues of the insertion were chosen so that: (a) the fragment that protrudes from the globule really would form a β-hairpin, (b) this hairpin should stabilize the structure of the whole globule and form common cooperative system with it, where possible, and (c) the superstructure would increase and reinforce the folding nucleus and accelerate the self-assembly. It was also interesting to find how point substitutions that changed stability of the “nose” would affect structure and kinetics of folding of the whole protein. The structure of SHA variation proved to be a cooperative system in kinetic and thermodynamic studies [14]. RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
and 13C-labeled proteins were prepared for sequential assignment of NMR resonances to the corresponding residues of the primary structure of SHA. Combination of three-dimensional and two-dimensional heteronuclear NMR experiments with the application of gradients of magnetic field provided practically complete assignment of resonances of 15N and 13C nuclei and protons (see the Experimental section), which is a necessary precondition for the calculation of a spatial structure. The 15N-HSQC spectrum with indications of cross-peaks corresponding to the numbers of amino acid residues of SHA is presented in Fig. 2 as a result of one of the attribution steps. Spatial structure of SHA has been calculated on the basis of the NMR data (see the Experimental section). Initial data and statistics obtained in the course of structural calculations are given in the table. Results of the RMSD calculation for each residue of the protein from the series of 20 structures with minimum value of the energy are given in Fig. 3. Values of local RMSD for seven N- terminal and two ë-terminal residues of SHA point out to a high mobility of these regions. RMSDs of approximately 1 Å are observed for the 48–55 sequence residues of the elongated β-hairpin (“nose”), suggesting its relative lability. Values of RMSD for the remaining residues of the main chain are 0.8 Å or less. Final model of spatial structure of SHA is illustrated in Fig. 4 where β-structural regions are conventionally indicated by arrows. Superposition of skeletal models of the SHA structure determined by NMR in this paper and structure of the wild type protein determined by X-ray analysis [16] is presented in Fig. 5 for comparison. According to the results of X-ray analysis, SH3-domain looks like a compact small barrel that consists of five antiparallel β-strands [16] and forms by the all-or-nothing mechanism [17]. As one can see from Fig. 5, arrangement of the chains in the overlapping regions of both proteins coincides rather well despite different methods of determinations of the structures. Long “nose” is notable in the NMR structure of SHA (Fig. 4). It is formed by antiparallel β-sheet that is typically turned as a propeller. Remarkable elongation of the “Bergerac” polypeptide chain in comparison with
Vol. 34
15N-
No. 5
2008
580
PROKHOROV et al. 15N,
ppm 52
105 59
110
61
66
5
9
13 23 64
49 4
115
21
40 37
28
35
19 65
32 24
14 45 67
60
58 56
15
120 51
18 27 46
36 42
125
26
55 68
53
7 29
41
43
22
17
8
2
34 50
11 31
38
10 57
6 30
39 3
44
25
54
70 47 69 16
12 48
33
63
130 10
9
8
7 H, ppm
1
Fig. 2. The 15N HSQC spectrum of the 1 mM solution of the SHA-Bergerac chimeric protein in 20 mM sodium acetate buffer containing 0.03% NaN3 (pH 4.0) at 298 K. Designations of the 1H-15N cross-peaks correspond to the numbers of amino acid residues. Resonances without assignment belong to NH protons of side chains of glutamine, asparagines, arginine, and lysine residues.
5 Atoms of the main chain Heavy atoms
RMSD, Å
4
3
2
1
0
5
10
15
20
25
30 35 40 Residue number
45
50
55
60
65
70
Fig. 3. The RMSD values for each residue of the SHA-Bergerac. RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
Vol. 34
No. 5
2008
STUDY OF THE STRUCTURE AND DYNAMICS
the starting protein (~17%) has practically no effect on the general topology of the molecule, and all the elements of secondary structure of the original domain are preserved in comparison with both X-ray [16, 18, 19] and NMR structures [20–22]. Comparison of the spatial structures of WT-SH3 (X-ray [16]) and SHA determined by NMR (Fig. 5) demonstrates that the first seven residues of both structures have no fixed positions. The ϕ and ψ angles, which are the most reliably determined on the basis of NMR data, indicate that the 8-11 sequence forms the first β-strand similarly to the wild type protein. The second β-strand in the structure of WT-SH3 involves the 29–34 region of the polypeptide chain, and the backbone conformation in the region of residues 33 and 34 is not ideal [16], suggesting some tensions in this segment. The corresponding β-strand in SHA begins from position 30. Eleven and nine of the twenty best structures, which have been obtained on the basis of the NMR analysis, involve the corresponding β-strand that consists of 30–33 and 30–32 residues, respectively. Thus, this β-strand proved to be twice shorter in solution than in the crystal. This fact indicates weakness of hydrogen bonds on its terminus. The third β-strand in SHA, similarly to the protein of the wild type, begins from the 40th residue, but increases from five residues in the crystal to ten residues in the “Bergerac” (41–50) due to the “nose”. Accordingly, the fourth strand is also enlarged from six residues (48–53) in WT-SH3 to ten residues in the “Bergerac”. The rest of the elements of secondary structure (helix of the 310 type and the fifth strand) are completely identical in the wild type protein and the SHA chimera. The spatial structure that consists of two separate β-sheets (one β-sheet combines the third and fourth strands, and another β-sheet involves the first, second, and fifth strands) exists in solution distinct from the crystalline structure [16] where all the five strands form one conjugated structure of a β-barrel type. Data on 15N-NMR relaxation are widely used for studies of both rotating diffusion of a protein molecule and local dynamics of a polypeptide chain. The relaxation data (see Fig. 6) is analyzed using the DASHA program [23]. High values of NOE (approximately 0.8) are characteristic of the most residues of the main chain of SHA except seven N-terminal residues, C-terminal residue in position 70, and the residues of the “nose” (the 47-57 region) (Fig. 6‡). Thus, data on RMSD (see Fig. 3) well correlates with the results of 1H-15N NOE. Such dependence is observed during determination of S2 parameter in the model-free approach [24, 25] (Fig. 6c). Selection criterion of the corresponding spectral density function for different protein residues is determined as a minimum of the penalty function [26]. Internal motions of the majority of amino acid residues of SHA (8, 9, 11-19, 22-32, 34-46, and 59-68) are described by S2 parameters closed to 1. This fact points RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
581
Analysis of the calculated spatial structures of SHAa Used limitationsb Spatial limitations of the NOE contacts Interresidual (i – j = 0) 222 Sequential (|i – j| = 1) 300 Contacts of the medium range (1 < |i – j| ≤ 4) 89 Contacts of long range (|i – j| ≥ 5) 343 Limitations of the dihedral angles Torsion angles (ϕ/ψ) 2 × 66 Total limitations 1086 Violation of the NOE limitations at a distance of >0.5 Å 0 to the dihedral angles >5° 0 RMSD (Å)c SHA (the 8–69 sequence) atoms of the main chain 0.34 ± 0.1 all the heavy atoms 0.88 ± 0.11 d Analysis of the Ramachandran card Residues in the most preferable regions, % 84.6 Residues in the additional permitted regions, % 15.4 Residues in the conditionally permitted regions, % 0.0 Residues in the forbidden regions, % 0.0 Notes: a Statistics was obtained for 20 calculated structures of SHA with the best objective function. b The spatial limitations were obtained on the basis of the contacts. c RMSD was calculated by the pair comparison of the structures in the ensemble with the average structure. d The Ramachandran map for all the residues, including the first seven residues with the high flexibility was obtained with the use of the PROCHECK-NMR program (see www.rcsb.org).
to their relative rigidity in the domain. However, N-terminal and C-terminal regions and the “nose” (the 47-57 sequence) are rather flexible according to the lower values of S2. In addition, some residues are well described by the function of spectral density with larger number of parameters. Thus, internal motion of residues 5, 6, 10, 21, 47, 48, 50-54, 56-58, and 69 is more adequately described by parameters S2 and τ (coordinated internal motions) [24] where S2 and τ are the order parameter and the effective correlation time of an internal motion, respectively. Mobility of residues 7, 49, 55, and 70 is 2 2 mostly exactly described by parameters S f , τf , S s , and τs (the case of uncoordinated internal motions [25]) where S s and τs are the order parameter and the correlation time of a slow internal motion, respectively and 2 S f and τf are the order parameter and the correlation time of a fast internal motion, respectively. 2
Overall tumbling time of a rotated correlation of the protein as a whole (τm = 4.0 ± 0.2 ns) is determined from
Vol. 34
No. 5
2008
582
PROKHOROV et al.
K53 A50
β4 RT
β3 β2 L8
L33
I30
β1
only for the first seven amino acid residues and in region of polypeptide chain corresponding to the “nose” that is somewhat more mobile than other fragments of the molecule, including the RT-loop, which is not involved in the formation of the secondary structure (the 14-27 sequence, Fig. 1‡). Increase in the flexibility of the “nose” could be associated with the fact that the β-hairpin is not completely incorporated into the domain cooperative network due to its isolated position. Although, the domain increases its stability by fixation of its ends, it is insufficient for achievement of the complete rigidity. Different rigidity of various regions can create an additional “vibration” that is reflected in the 15N dynamics of the protein. This explanation seems quite logical, because the “nose” that protrudes from the general contour of the protein molecule can quite effectively interact with water molecules, which are capable to compete for formation of hydrogen bonds in the βstructure, weaken it, and make it blinking. Other factor that affects the dynamic characteristics of this region of the polypeptide chain can be a flexibility of the long “nose” as a whole.
A11 V66 K68
P62
W41
β5
Fig. 4. Ribbon model of the spatial arrangement of the main chain of the SHA-Bergerac protein on the basis of the NMR data. Numbers of the first and the last residues of the β1–β5 strands are numerated. The 63–65 sequence forms helix of the 310 type.
the í1/í2 ratio [27] that corresponds to the expected value for monomeric proteins with the molecular mass of 8 kDa [28]. The í1/í2 value for Leu33 is higher than one standard deviation (Fig. 6b) and suggests that this residue could participate in a conformational exchange [26]. Possibility of existence of this chain region in at least two various conformations [26] most probably is a reason of disturbance of the hydrogen bond in the second β-strand (30-33) and decreases its length (30-32) in comparison with that in the X-ray structure in two times. Anomalous behavior of Leu33 was found in the course of studies of the fast kinetics of folding of the wild type protein. Only one residue (Leu33) proved to have a negative value of parameter î, which, as a rule, varies from 0 to 1 and reflects a degree of involvement of this residue in formation of the transient state during the folding [1, 9]. It is considered that the anomalous values of î indicate the participation of this residue in negative interactions during the transient state (the alternative ordered conformation). According to the 15N (í1, í2, NOE) dynamic studies, considerable differences in the flexibility are observed
EXPERIMENTAL Expression, purification, and preparation of the SHA protein. The gene of SHA protein was prepared by PCR and incorporated into the pBAT-4 vector [29]. Expression of the gene and purification of the protein were performed by the procedure described previously [30]. Purity and homogeneity of the protein preparation were evaluated by the standard denaturing electrophoresis in polyacryl amide gel (SDS-PAGE). Concentration of the SHA protein was measured on a spectrophotometer at 280 nm using the molar extinction coefficient (17300 å–1cm–1) determined previously [31]. The uniformly 15N/13C-labeled protein was prepared by growing E. coli (the BL21(DE3) strain) on an M9 minimum medium with a concentration of [13C]glucose and [15N]ç4ël (CIL, United States) of 1 g/l. NMR spectroscopy. The sample for the NMR experiments contained 1mM [15N/13C]SHA in 20 mM sodium acetate buffer with 0.03% NaN3 (pH 4.0) in the mixture of H2O and 2ç2O (9 : 1). All the experiments were carried out on an AVANCE 600 spectrometer (Bruker, Germany) with an operating proton frequency of 600 MHz equipped with a triple resonance pulsed field gradient probe with an impulse Z-gradient. The sample temperature was 298 K. The pH 4 value was chosen so that the possibility of oligomerization was decreased to minimum with the preservation of sufficient stability of the protein spatial structure. Programs XWINNMR and TOPSPIN (Bruker Biospin, Karlsruhe, Germany) were used for a primary data processing and automatic NOE collection. The 1H, 13C, and 15N resonances were attributed using the CARA program [32, 33]. An AutoLink integrated module in the CARA program was used for the semiautomatic attribution of the polypeptide chain on the basis of
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
Vol. 34
No. 5
2008
STUDY OF THE STRUCTURE AND DYNAMICS
583
Fig. 5. Superposition of the skeletal models of structures of the SHA-Bergerac chimerical protein (the NMR model is drawn in black) and the wild type protein (the X-ray model is drawn in gray).
2D-1H/15N HSQC and 3D-HNCACB spectra and the primary structure of the protein. Further, we superposed fragments of the semiautomatic attribution on the basis of the HNCACB and C(C)(CO)NH spectra and validated the accuracy of this superposition. In addition, the C(C)(CO)NH spectrum was also used for attribution of 13C resonances of side chains of aliphatic residues. The corresponding proton resonances were found on the basis of the 3D-15N-TOCSY and HCCHTOCSY experiments. The chemical shifts of carbons of the carbonyl groups were determined from the HNCO spectrum. The 1H- and 13C resonances of side chains of aromatic residues were determined from the 1H/13C HSQC and HCCH TOCSY spectra. The 2D-1H/15N HSQC, 3D-15N TOCSY, and 3D-15N NOESY spectra were used for attribution of the indole protons of Trp41 and Trp42. The chemical shifts of protons are given relatively to the resonance of sodium 2,2-dimethyl-2-silapentane-5-sulfonate according to the recommendations of IUPAC [34]. Chemical shifts of other nuclei were calculated from the ratio of gyromagnetic factors [35]. The attributed resonances 1H, 15N, and 13C were placed in the database of BioMagResBank with the accession code of 11026. RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
Parameters of 15N relaxation were measured for the sample with the use of impulse sequences described by Farrow et al. [26] at a field strength of 14.1 T and temperature of 298 K for studies of dynamics of SHA. Values of times of the spin-lattice (í1) and spinspin (í2) relaxations of the 15N nuclei were calculated from the series of 2D-1H-15N correlation spectra recorded at the following variations of the relaxation delays: 5, 40, 100, 160, 240, 340, 460, 600, 800, and 1200 ms for í1 and 8, 48, 64, 80, 100, 128, 160, 208, and 256 ms for í2. Transformation of decays of the free induction was performed with the use of the LorentianGaussian apodization function applied in both dimensions in NMRPipe [36]. The í1 and í2 values were determined by fitting the integral values to two-parametric exponential function with the accuracy of 96%. The values of heteronuclear NOE between 1H-15N were obtained form the ratio of intensities of peaks with and without saturation of amide protons within 2.7 s. The value of time of the rotating correlation for every 15N nucleus of the SHA main chain was calculated from the í1/í2 ratio in the DASHA program [23]. Calculation of the spatial structure. The distance limitations were obtained from the three-dimensional 15N- and 13C NOESY experiments (for aliphatic and 15N-labeled
Vol. 34
No. 5
2008
584
PROKHOROV et al. NOE 0.8 (‡) 0.4 0 –0.4 T1/T2 4.0 3.5 3.0 (b)
2.5 2.0 1.5 S2 1.0 0.8 0.6
(c) 0.4 0.2 0
10
20
30
40
50
60
70 Residue
Fig. 6. The parameters of 15N NMR relaxation of the SHA-Bergerac protein measured at 600 MHz and 298 K: ‡ is the values of standard 1H-15N NOE depending on the number of a residue of the SHA sequence, b is the í1/í2 dependence on the number of a residue of the SHA sequence, and c is the values of the order parameter (S2) depending on the number of a residue in the SHA sequence. Lines that are parallel to abscissa indicate the average í1/í2 and one standard deviation from the average. Signs parallel to the ordinate designate the measurement error.
aromatic areas of a spectrum) with mixing time of the magnetization components of 80 ms. The collected resonances were edited and integrated in the CARA program [32, 33]. The ϕ and ψ torsion angles of the polypeptide chain were predicted by the PREDITOR program from the chemical shifts of nuclei of the atoms of the main chain: HN, 15N, 1Hα, 13Cα, 13CO, and 13Cβ [37]. The structure was calculated with the use of the CYANA program [38]. The standard protocol that included seven cycles of the combined automatic attribution of NOE and calculation of 100 structural conformations on each step of the cycle was used. As a result,
20 structures with the minimum value of the objective function [38] were chosen, and used for the final analysis. The obtained ensemble of structures correlated with the experimental data and demonstrated good statistics according to Ramachandran (table). Moreover, the structure satisfied convergence criteria necessary for the significant calculation in the CYANA program. The convergence of decisions corresponded to the X-ray resolution of approximately ~2 Å owing to the good quality of the three-dimensional NOESY spectra and practically complete attribution (96.5%). The MOLM program was used for visualization of the structure [39].
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
Vol. 34
No. 5
2008
STUDY OF THE STRUCTURE AND DYNAMICS
The experimental NMR limitations and atomic coordinates of 20 structures of the chimeric SHA protein were deposited into PDB with access code of 2rmo. ACKNOWLEDGEMENTS We thank Professor Luis Serrano and Doctor Ana Rosa Viguera from the European Molecular and Biological Laboratory (Heidelberg, Germany) for the plasmid with the SHA gene. This study was supported by the INTAS grant, project no. 03-51-5569 and the Program of the Presidium of the Russian Academy of Sciences (Molecular and Cellular Biology). REFERENCES 1. Nolting, B., Methods in Modern Biophysics, Berlin: Springer, 2004. Translated under the title Noveishie metody issledovaniya biosistem,, Moscow: Tekhnosfera, 2005. 2. Mittermaier, A., Korzhnev, D.M., and Kay, L.E., Biochemistry, 2005, vol. 44, pp. 15 430–15 436. 3. Li, J., Shinjo, M., Matsumura, Y., Morita, M., Baker, D., Ikeguchi, M., and Kihara, H., Biochemistry, 2007, vol. 46, pp. 5072–5082. 4. Musacchio, A., Gibson, T., Lehto, V.P., and Saraste, M., FEBS Lett., 1992, vol. 307, pp. 55–61. 5. Gmeiner, W.H. and Horita, D.A., Cell Biochem. Biophys., 2001, vol. 35, pp. 127–140. 6. Kay, B.K., Williamson, M.P., and Sudol, M., Faseb J., 2000, vol. 14, pp. 231–241. 7. Mayer, B.J., J. Cell. Sci., 2001, vol. 114, pp. 1253–1263. 8. Cowan-Jacob, S.W., Cell. Mol. Life Sci., 2006, vol. 63, pp. 2608–2625. 9. Viguera, A.R., Arrondo, J.L., Musacchio, A., Saraste, M., and Serrano, L., Biochemistry, 1994, vol. 33, pp. 10925– 10933. 10. Pisabarro, M.T. and Serrano, L., Biochemistry, 1996, vol. 35, pp. 10 634–10 640. 11. Arold, S.T., Ulmer, T.S., Mulhern, T.D., Werner, J.M., Ladbury, J.E., Campbell, I.D., and Noble, M.E., J. Biol. Chem., 2001, vol. 276, pp. 17 199–17 205. 12. Musacchio, A., Adv. Protein Chem., 2002, vol. 61, pp. 211–268. 13. Musacchio, A., Wilmanns, M., and Saraste, M., Prog. Biophys. Mol. Biol., 1994, vol. 61, pp. 283–297. 14. Viguera, A.-R. and Serrano, L., J. Mol. Biol., 2001, vol. 311, pp. 357–371. 15. Viguera, A.-R. and Serrano, L., Proc. Natl. Acad. Sci. USA, 2003, vol. 100, pp. 5730–5735. 16. Musacchio, A., Noble, M., Pauptit, R., Wierenga, R., and Saraste, M., Nature, 1992, vol. 359, pp. 851–855.
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY
585
17. Viguera, A.R., Martinez, J.C., Filimonov, V.V., Mateo, P.L., and Serrano, L., Biochemistry, 1994, vol. 33, pp. 2142– 2150. 18. Noble, M.E., Musacchio, A., Saraste, M., Courtneidge, S.A., and Wierenga, R.K., EMBO J., 1993, vol. 12, pp. 2617–2624. 19. Morton, C.J., Pugh, D.J.R., Brown, E.L.J., Kahmann, J.D., Renzoni, D.A.C., and Campbell, I.D., Structure, 1996, vol. 4, pp. 705–714. 20. Blanco, F.J., Ortiz, A.R., and Serrano, L., J. Biomol. NMR, 1997, vol. 9, pp. 347–357. 21. Hiroaki, H., Klaus, W., and Senn, H., J. Biomol. NMR, 1996, vol. 8, pp. 105–122. 22. Ortega, RoldenJ.L., Romero, RomeroM.L., Ora, A., Ab, E., Lopez-Mayorga, O., Azuaga, A.I., and van Nuland, N.A., J. Biomol. NMR, 2007, vol. 39, pp. 331–336. 23. Orekhov, V.Yu., Nolde, D.E., Golovanov, A.P., Korzhnev, D.M., and Arseniev, A.S., Appl. Magn. Reson., 1995, vol. 9, pp. 581–588. 24. Lipari, G. and Szabo, A., J. Am. Chem. Soc., 1982, vol. 104, pp. 4559–4570. 25. Clore, G.M., Szabo, A., Bax, A., Kay, L.E., Driscoll, P.C., and Gronenborn, A.M., J. Am. Chem. Soc., 1990, vol. 112, pp. 4989–4991. 26. Farrow, N.A., Muhandiram, R., Singer, A.U., Pascal, S.M., Kay, C.M., Gish, G., Shoelson, S.E., Pawson, T., FormanKay, J.D., and Kay, L.E., Biochemistry, 1994, vol. 33, pp. 5984–6003. 27. Kay, L.E., Torchia, D.A., and Bax, A., Biochemistry, 1989, vol. 28, pp. 8972–8979. 28. Daragan, V.A. and Mayo, K.H., Prog. Nucl. Magn. Reson. Spectrosc., 1997, vol. 31, pp. 63–105. 29. Peranen, J., Rikkonen, M., Hyvonen, M., and Kaariainen, L., Anal. Biochem., 1996, vol. 236, pp. 371–373. 30. Viguera, A.R., Serrano, L., and Wilmanns, M., Nature Struct. Biol., 1996, vol. 3, pp. 874–880. 31. Gill, S.C. and von Hippel, P.H., Anal. Biochem., 1989, vol. 182, pp. 319–326. 32. Keller, R., The Computer Aided Resonance Assignment Tutorial. ISBN, 3–85600-112-3, First Edition, 2004. 33. Sattler, M., Schleucher, J., and Griesinger, C., Prog. Nucl. Magn. Reson. Spectrosc., 1999, vol. 34, pp. 93– 158. 34. Markely, J.L., Bax, A., Arata, Y., Hilbers, C.W., Kaptein, R., Sykes, B.D., Wright, P.E., and Wutrich, K., Pure Appl. Chem., 1998, vol. 70, pp. 117–142. 35. Wishart, D.S., Bigam, C.G., Yao, J., Abildgaard, F., Dyson, H.J., Oldfield, E., Markley, J.L., and Sykes, B.D., J. Biomol. NMR, 1995, vol. 6, pp. 135–140. 36. Delaglio, F., Grzesiek, S., Vuister, G.W., Zhu, G., Pfeifer, J., and Bax, A., J. Biomol. NMR, 1995, vol. 6, pp. 277–293. 37. Berjanskii, M.V., Neal, S., and Wishart, D.S., Nucleic Acids Res., 2006, vol. 34, pp. W63–W69. 38. Güntert, P., Methods Mol. Biol., 2004, vol. 278, pp. 353– 378. 39. Koradi, R., Billeter, M., and Wuthrich, K., J. Mol. Graph., 1996, vol. 14, pp. 51–55.
Vol. 34
No. 5
2008