From Structure and Dynamics of Protein L7/L12 to Molecular Switching in Ribosome
2004; Elsevier BV; Volume: 279; Issue: 17 Linguagem: Inglês
10.1074/jbc.m313384200
ISSN1083-351X
AutoresEduard V. Bocharov, Alexander G. Sobol, Konstantin V. Pavlov, Dmitry M. Korzhnev, Victor Jaravine, A.T. Gudkov, Alexander S. Arseniev,
Tópico(s)Toxin Mechanisms and Immunotoxins
ResumoBased on the 1H-15N NMR spectroscopy data, the three-dimensional structure and internal dynamic properties of ribosomal protein L7 from Escherichia coli were derived. The structure of L7 dimer in solution can be described as a set of three distinct domains, tumbling rather independently and linked via flexible hinge regions. The dimeric N-terminal domain (residues 1-32) consists of two antiparallel α-α-hairpins forming a symmetrical four-helical bundle, whereas the two identical C-terminal domains (residues 52-120) adopt a compact α/β-fold. There is an indirect evidence of the existence of transitory helical structures at least in the first part (residues 33-43) of the hinge region. Combining structural data for the ribosomal protein L7/L12 from NMR spectroscopy and x-ray crystallography, it was suggested that its hinge region acts as a molecular switch, initiating "ratchet-like" motions of the L7/L12 stalk with respect to the ribosomal surface in response to elongation factor binding and GTP hydrolysis. This hypothesis allows an explanation of events observed during the translation cycle and provides useful insights into the role of protein L7/L12 in the functioning of the ribosome. Based on the 1H-15N NMR spectroscopy data, the three-dimensional structure and internal dynamic properties of ribosomal protein L7 from Escherichia coli were derived. The structure of L7 dimer in solution can be described as a set of three distinct domains, tumbling rather independently and linked via flexible hinge regions. The dimeric N-terminal domain (residues 1-32) consists of two antiparallel α-α-hairpins forming a symmetrical four-helical bundle, whereas the two identical C-terminal domains (residues 52-120) adopt a compact α/β-fold. There is an indirect evidence of the existence of transitory helical structures at least in the first part (residues 33-43) of the hinge region. Combining structural data for the ribosomal protein L7/L12 from NMR spectroscopy and x-ray crystallography, it was suggested that its hinge region acts as a molecular switch, initiating "ratchet-like" motions of the L7/L12 stalk with respect to the ribosomal surface in response to elongation factor binding and GTP hydrolysis. This hypothesis allows an explanation of events observed during the translation cycle and provides useful insights into the role of protein L7/L12 in the functioning of the ribosome. Ribosomes are complex and dynamic ribonucleoprotein assemblies, which provide the framework for protein biosynthesis in all organisms (for review see Refs. 1Ramakrishnan V. Cell. 2002; 108: 557-572Abstract Full Text Full Text PDF PubMed Scopus (646) Google Scholar, 2Doudna J.A. Rath V.L. Cell. 2002; 109: 153-156Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar, 3Wilson D.N. Blaha G. Connell S.R. Ivanov P.V. Jenke H. Stelzl U. Teraoka Y. Nierhaus K.H. Curr. Protein Pept. Sci. 2002; 3: 1-53Crossref PubMed Scopus (53) Google Scholar, 4Spirin A.S. FEBS Lett. 2002; 514: 2-10Crossref PubMed Scopus (79) Google Scholar). One of the most remarkable features of the large subunit of the ribosome is the presence of a highly flexible protuberance called the stalk (for review see Refs. 5Liljas A. Gudkov A.T. Biochimie (Paris). 1987; 69: 1043-1047Crossref PubMed Scopus (39) Google Scholar, 6Gudkov A.T. FEBS Lett. 1997; 407: 253-256Crossref PubMed Scopus (50) Google Scholar, 7Wahl M.C. Moller W. Curr. Protein Pept. Sci. 2002; 3: 93-106Crossref PubMed Scopus (88) Google Scholar, 8Conzalo P. Reboud J.-P. Biol. Cell. 2003; 95: 179-193Crossref PubMed Scopus (109) Google Scholar). Various conformations of the stalk are thought to reflect different functional states of the ribosome, which are essential for tRNA binding and translocation of peptidyl-tRNA from A- to P-site. In prokaryotes, the stalk includes a highly conserved acidic 12-kDa protein L7/L12 (L7 is the N-terminal acetylated form of L12), which is present in four copies as two dimers associated with other ribosomal components via protein L10. Both L7/L12 dimers are necessary for optimal rates of protein synthesis and proper function of elongation factors, but a single dimer is sufficient for ribosomal activity (9Griaznova O. Traut R.R. Biochemistry. 2000; 39: 4075-4081Crossref PubMed Scopus (33) Google Scholar). Spatial contacts of the L7/L12 stalk with elongation factors G and Tu (EF-G and EF-Tu) 1The abbreviations used are: EF, elongation factor; NTD, N-terminal domain; CTD, C-terminal domain; dNTD, dimeric NTD; r.m.s.d., root mean square deviation; RDC, residual 1H-15N dipolar coupling; Dr, diffusion tensor; PDB, Protein Data Bank. were observed by electron cryomicroscopy (10Stark H. Rodnina M.V. Rinke-Appel J. Brimacombe R. Wintermeyer W. van Heel M. Nature. 1997; 389: 403-406Crossref PubMed Scopus (316) Google Scholar, 11Stark H. Rodnina M.V. Wieden H.J. van Heel M. Wintermeyer W. Cell. 2000; 100: 301-309Abstract Full Text Full Text PDF PubMed Scopus (263) Google Scholar) in keeping with the importance of L7/L12 for either factor binding to ribosome or stimulation of factor-depended GTPase activity (12Mohr D. Wintermeyer W. Rodnina M.V. Biochemistry. 2002; 41: 12520-12528Crossref PubMed Scopus (128) Google Scholar). Moreover, conformational changes in L7/L12 were revealed by limited proteolysis upon EF-G and EF-Tu binding with GTP or GDP to ribosome (13Gudkov A.T. Bubunenko V.G. Biochimie (Paris). 1989; 71: 779-785Crossref PubMed Scopus (32) Google Scholar). Current x-ray crystallographic analysis failed to resolve the fine structural state of L7/L12 within the ribosome, because electron density for L7/L12 stalk was not apparent even in the recent 2.4-Å resolution structure of the large ribosomal subunit (14Ban N. Nissen P. Hansen J. Moore P.B. Steitz T.A. Science. 2000; 289: 905-920Crossref PubMed Scopus (2836) Google Scholar). Thus, high resolution x-ray crystallographic and NMR studies of the isolated protein L7/L12 and its complex with the protein L10 may still provide valuable insights into the atomic structures of the L7/L12 stalk. Based on structural and biochemical investigations, the protein L7/L12 is composed of two distinct, organized domains connected by an extensive linker, the so-called "hinge" region (5Liljas A. Gudkov A.T. Biochimie (Paris). 1987; 69: 1043-1047Crossref PubMed Scopus (39) Google Scholar, 6Gudkov A.T. FEBS Lett. 1997; 407: 253-256Crossref PubMed Scopus (50) Google Scholar, 7Wahl M.C. Moller W. Curr. Protein Pept. Sci. 2002; 3: 93-106Crossref PubMed Scopus (88) Google Scholar, 8Conzalo P. Reboud J.-P. Biol. Cell. 2003; 95: 179-193Crossref PubMed Scopus (109) Google Scholar, 13Gudkov A.T. Bubunenko V.G. Biochimie (Paris). 1989; 71: 779-785Crossref PubMed Scopus (32) Google Scholar, 15Bushuev V.N. Gudkov A.T. Liljas A. Sepetov N.F. J. Biol. Chem. 1989; 264: 4498-4505Abstract Full Text PDF PubMed Google Scholar). The N-terminal domain (NTD) is responsible for L7/L12 dimerization and for anchoring the protein to the ribosome, whereas the C-terminal domain (CTD) is involved in translation factor interaction (6Gudkov A.T. FEBS Lett. 1997; 407: 253-256Crossref PubMed Scopus (50) Google Scholar). In solution, the hinge regions of L7/L12 dimer have an unordered flexible structure that enables independent movement of both CTDs relative to each other and to the dimeric N-terminal domain (dNTD) (15Bushuev V.N. Gudkov A.T. Liljas A. Sepetov N.F. J. Biol. Chem. 1989; 264: 4498-4505Abstract Full Text PDF PubMed Google Scholar, 16Bocharov E.V. Gudkov A.T. Arseniev A.S. FEBS Lett. 1996; 379: 291-294Crossref PubMed Scopus (34) Google Scholar, 17Hamman B.D. Oleinikov A.V. Jokhadze G.G. Traut R.R. Jameson D.M. Biochemistry. 1996; 35: 16680-16686Crossref PubMed Scopus (45) Google Scholar, 18Hamman B.D. Oleinikov A.V. Jokhadze G.G. Traut R.R. Jameson D.M. Biochemistry. 1996; 35: 16672-16679Crossref PubMed Scopus (44) Google Scholar, 19Bocharov E.V. Gudkov A.T. Budovskaya E.V. Arseniev A.S. FEBS Lett. 1998; 423: 347-350Crossref PubMed Scopus (32) Google Scholar). The removal of one but not two CTDs from L7/L12 dimer gives no marked effect on the ribosomal translation activity (20Oleinikov A.V. Jokhadze G.G. Traut R.R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95 (03432): 4215-4218Crossref PubMed Scopus (21) Google Scholar), whereas the deletions of the hinge residues significantly reduce the mobility of the CTDs and decrease activity of the modified ribosome (21Bubunenko M.G. Chuikov S.V. Gudkov A.T. FEBS Lett. 1992; 313: 232-234Crossref PubMed Scopus (21) Google Scholar, 22Oleinikov A.V. Perroud B. Wang B. Traut R.R. J. Biol. Chem. 1993; 268: 917-922Abstract Full Text PDF PubMed Google Scholar). Earlier, the crystal structure of Escherichia coli L7/L12 C-terminal fragment consisting of three β-strands and three α-helices had been solved at a 1.7-Å resolution (23Leijonmarck M. Liljas A. J. Mol. Biol. Chem. 1987; 247: 3622-3629Google Scholar). NMR study of the E. coli L7 dimer in solution has shown α-helical hairpin conformation of NTD (16Bocharov E.V. Gudkov A.T. Arseniev A.S. FEBS Lett. 1996; 379: 291-294Crossref PubMed Scopus (34) Google Scholar, 19Bocharov E.V. Gudkov A.T. Budovskaya E.V. Arseniev A.S. FEBS Lett. 1998; 423: 347-350Crossref PubMed Scopus (32) Google Scholar). Recently, the crystal structure of isolated L12 from the hyperthermophilic bacterium Thermotoga maritima with two full-length molecules and two proteolysed N-terminal fragments associated into tetramer in the asymmetric unit was reported (24Wahl M.C. Bourenkov G.P. Bartunik H.D. Huber R. EMBO J. 2000; 19: 174-186Crossref PubMed Scopus (79) Google Scholar). The T. maritima protein L7/L12 is highly homologous to the E. coli counterpart with 64.7% sequence identity. In contrast to the parallel arrangement of the subunits in the dNTD assumed earlier (16Bocharov E.V. Gudkov A.T. Arseniev A.S. FEBS Lett. 1996; 379: 291-294Crossref PubMed Scopus (34) Google Scholar), this crystal structure reveals antiparallel arrangement. Another important finding is that in crystal the hinge region of the full-length molecules is contracted into long α-helix, which folds back on the NTD forming a compact overall protein structure, whereas the restricted hinge regions of the L12 fragments adopted elongated, unstructured conformations (24Wahl M.C. Bourenkov G.P. Bartunik H.D. Huber R. EMBO J. 2000; 19: 174-186Crossref PubMed Scopus (79) Google Scholar). Moreover, the hinge regions from the full-length molecules bind to each other, which result in additional intermolecular contacts between the N- and C-terminal domains in the crystalline L12 tetramer. Thus, the protein L7/L12 in crystal has a very compact structure different from the solution model in which the N- and C-terminal domains are separated by a stretched and flexible hinge region (19Bocharov E.V. Gudkov A.T. Budovskaya E.V. Arseniev A.S. FEBS Lett. 1998; 423: 347-350Crossref PubMed Scopus (32) Google Scholar). However, the high resolution structure of protein L7/L12 in solution was not obtained to date and the molecular details of function of this protein during the translation process are still not fully understood. In this study, we have determined the spatial structure and characterized the internal dynamics of protein L7 from E. coli in solution by heteronuclear NMR spectroscopy. Based on this information along with the available structural and biochemical data, a model of L7/L12 molecular switching during the translation process is proposed. This provides useful insights into the role of protein L7/L12 in the functioning of the ribosome. NMR Spectroscopy—NMR experiments were performed on 600 MHz (1H) Varian Unity spectrometer equipped with pulsed-field gradient unit and triple resonance probe. NMR spectra were acquired at 30 °C using 1 mm samples of uniformly 15N-labeled L7 dissolved at pH 6.9 in 600 μl of buffer solution containing 0.05 m sodium phosphate, 0.1 m KCl and either 90% H2O, 10% D2O or 99.9% D2O. Details of the protein preparation and NMR experiments used to assign 1H-15N resonances (BioMagResBank accession number 4429) and determine the L7 secondary structure are documented elsewhere (16Bocharov E.V. Gudkov A.T. Arseniev A.S. FEBS Lett. 1996; 379: 291-294Crossref PubMed Scopus (34) Google Scholar). During the course of NMR data collection, there were no detectable changes in NMR spectra because of spontaneous proteolysis of L7 dimer (25Liljas A. Prog. Biophys. Mol. Biol. 1982; 40: 161-228Crossref PubMed Scopus (110) Google Scholar). The interproton distance restraints used in this work for L7 structure calculation were obtained from three-dimensional 1H-15N NOESY-HSQC (60-ms mixing time) experiment featuring pulsed-field gradient coherence selection, sensitivity enhancement, and flip-back pulse for minimizing saturation of water (26Zhang O. Kay L.E. Olivier J.P. Forman-Kay J.D. J. Biomol. NMR. 1994; 4: 845-858Crossref PubMed Scopus (613) Google Scholar). Two-dimensional NOESY (60-ms mixing time) spectrum acquired for the D2O sample was also used as an additional source to the structure information concerning aromatic rings of phenylalanine residues. Dihedral angle restraints were estimated from homonuclear 3JHNα and heteronuclear 3JNβ coupling constants obtained quantitatively from two-dimensional 1H-15N HMQCJ (27Kay L.E. Bax A. J. Magn. Reson. 1990; 86: 110-126Google Scholar) and qualitatively from three-dimensional 1H-15N HNHB (28Archer S.J. Ikura M. Torchi D.A. Bax A. J. Magn. Reson. 1991; 95: 636-641Google Scholar) experiments, respectively. The preliminary stereospecific assignments of methylene β-protons were deduced by inspection of the relative cross-peak intensities in three-dimensional 1H-15N HNHB, 1H-15N NOESY-HSQC (60-ms mixing time), and 1H-15N TOCSY-HSQC (26Zhang O. Kay L.E. Olivier J.P. Forman-Kay J.D. J. Biomol. NMR. 1994; 4: 845-858Crossref PubMed Scopus (613) Google Scholar) (50-ms mixing time) spectra. The slowly exchanging amide protons were identified by reconstituting lyophilized L7 sample in D2O and immediately recording a series of 1H-15N HSQC spectra (30-min duration each) over 4 h at 23 °C. Residual 1H-15N dipolar couplings (RDCs) were extracted from 1JNH-modulated HSQC experiments (29Tjandra N. Grzesiek S. Bax A. J. Am. Chem. Soc. 1996; 118: 6264-6272Crossref Scopus (315) Google Scholar) collected on isotropic L7 sample and on partially aligned L7 samples containing the liquid crystalline phase of Tobacco mosaic virus (30Clore G.M. Starich M.R. Gronenborn A.M. J. Am. Chem. Soc. 1998; 120: 10571-10572Crossref Scopus (324) Google Scholar) or dihexanoylphosphatidylcholine/dimyristoylphosphatidylcholine mixture (1:2.9 molar ratio, 6.5% lipid (w/v)) (31Bax A. Tjandra N. J. Biomol. NMR. 1997; 10: 289-292Crossref PubMed Scopus (206) Google Scholar). Alignments of both media were confirmed by observation of the quadrupole splitting of the D2O signal. The splitting of 16 or 8 Hz was achieved for a sample containing virus particles or lipid bicelles, respectively. Sample homogeneity and stability were controlled before and after RDC measurements by arrayed deuterium experiments with slice selection gradients. In both lipid and virus media, the legitimate dipolar couplings were observed only for the residues from the CTD (see Supplementary data). 15N relaxation measurements were performed using the pulse sequences by Farrow et al. (32Farrow N.A. Muhandiram R. Singer A.U. Pascal S.M. Kay C.M. Gish G. Shoelson S.E. Pawson T. Forman-Kay J.D. Kay L.E. Biochemistry. 1994; 33: 5984-6003Crossref PubMed Scopus (2030) Google Scholar) at 14.1 T and 30 °C. The values of 15N longitudinal (R1) and transverse (R2) relaxation rates were obtained by exponential fitting of the intensities of cross-peaks in sets of two-dimensional 1H-15N correlation spectra recorded with relaxation delays ranging from 10 to 1000 ms and from 0 to 200 ms in R1 and R2 experiments, respectively. The R2 data were numerically corrected to account for off-resonance effects associated with the CPMG (Carr-Pur-cell-Meiboom-Gill) refocusing pulses (33Korzhnev D.M. Tischenko E.V. Arseniev A.S. J. Biomol. NMR. 2000; 17: 231-237Crossref PubMed Scopus (49) Google Scholar). The values of heteronuclear 15N{1H} steady-state NOE were measured as a ratio of signal intensities in two-dimensional 1H-15N correlation spectra recorded with and without prior saturation of amide protons achieved by a sequence of 120° 1H pulses spaced by 1.0 ms. The delay between scans in 15N R1 and R2 and in 15N{1H} NOE experiments was 2.5 and 5.0 s, respectively. The R1, R2, and NOE measurements were repeated several times, and resulting values were averaged over all of the recorded data sets. The relaxation data analysis and hydrodynamic calculations were performed using DASHA program (see Supplementary data) (34Orekhov V.Y. Nolde D.E. Golovanov A.P. Korzhnev D.M. Arseniev A.S. Appl. Magn. Reson. 1995; 9: 581-588Crossref Scopus (81) Google Scholar). Structure Calculations—Spatial structure calculations were performed using the programs DYANA 1.5 (35Güntert P. Mumenthaler C. Wüthrich K. J. Mol. Biol. 1997; 273: 283-298Crossref PubMed Scopus (2569) Google Scholar) and CYANA 1.01 (www.guentert.com), which employ simulated annealing combined with molecular dynamics in torsion angle space. Meaningful upper distance restraints were derived using CALIBA (36Güntert P. Braun W. Wüthrich K. J. Mol. Biol. 1991; 217: 517-530Crossref PubMed Scopus (919) Google Scholar) function of DYANA from the volumes of NOE cross-peaks integrated in the NOESY spectra by the program XEASY (37Bartels C. Xia T. Billeter M. Güntert P. Wüthrich K. J. Biomol. NMR. 1995; 6: 1-10Crossref PubMed Scopus (1614) Google Scholar). The semiautomatic assignments of NOE cross-peaks were obtained with the program XEASY and NOAH (38Mumenthaler C. Güntert P. Braun W. Wüthrich K. J. Biomol. NMR. 1997; 10: 351-362Crossref PubMed Scopus (134) Google Scholar) subroutine of DYANA package. Stereospecific assignments and torsion angle restraints for ϕ, φ, and χ1 were obtained by the analysis of local conformation in GRIDSEARCH (39Güntert P. Billeter M. Ohlenschlager O. Brown L.R. Wüthrich K. J. Biomol. NMR. 1998; 12: 543-548Crossref PubMed Scopus (55) Google Scholar) and GLOMSA (36Güntert P. Braun W. Wüthrich K. J. Mol. Biol. 1991; 217: 517-530Crossref PubMed Scopus (919) Google Scholar) subroutines of DYANA using the available homonuclear 3JHNα and heteronuclear 3JNβ spin-spin-coupling constants and sequential NOE data. The slowly exchanging amide protons were assigned as hydrogen bond donors with related hydrogen-acceptor partners on the basis of preliminary structure calculations. Corresponding hydrogen bond restraints were employed in subsequent calculations for d(O,N), d(O,HN), and d(C,HN) distances in accordance with angle and distance criteria for different types of hydrogen bonds (40Baker E.N. Hubbard R.E. Prog. Biophys. Mol. Biol. 1984; 44: 97-179Crossref PubMed Scopus (1668) Google Scholar). During preliminary rounds of structure calculations (obtained without account for RDCs), typically 100-200 structures were generated using 5,000 simulated annealing steps followed by 1,200 steps of conjugate gradient minimization with 10-20 structures being retained according to their standard DYANA target function values for relaxation matrix back-calculations of the NOEs and for analysis of hydrogen bonds, stereospecific assignments, and ambiguous NOEs. In the final cycle of calculations, the standard CYANA-simulated annealing protocol (modified to incorporate refinement against RDC restraints after employing distance and dihedral angle restraints) was applied to 200 random structures and the resulting 20 structures with the lowest target function were selected. Constrained energy minimization of the side chains of the 20 best CYANA structures was performed in the program FANTOM (41Schaumann T. Braun W. Wüthrich K. Biopolymers. 1990; 29: 679-694Crossref Scopus (96) Google Scholar) using ECEPP/2 potential with available distance and dihedral angle restraints. The mean structure calculation of the CYANA family and the analyses of root mean square deviations (r.m.s.d.), secondary structure, and hydrogen bonds were performed with MOLMOL (42Koradi R. Billeter M. Wüthrich K. J. Mol. Graphics. 1996; 14: 51-55Crossref PubMed Scopus (6519) Google Scholar). Spatial distribution of electrostatic potential on the protein solvent-accessible surfaces and figures of the structures were generated by MOLMOL (42Koradi R. Billeter M. Wüthrich K. J. Mol. Graphics. 1996; 14: 51-55Crossref PubMed Scopus (6519) Google Scholar). The molecular hydrophobicity potential created by protein atoms on the protein solvent-accessible surface was calculated as described previously (43Efremov R.G. Vergoten G. Phys. Chem. 1995; 99: 10658-10666Crossref Scopus (47) Google Scholar). The calculation and visualization of molecular hydrophobicity potential was done using the homemade software. N- and C-terminal Domains of L7 Dimer Tumble at Different Rates—The 15N{1H} NOE, 15N R1, and R2 values measured for the backbone 15N nuclei of the L7 dimer (Fig. 1, a-c) exhibit significant variations along the protein sequence. Moreover, it is clearly seen from the relaxation data (Fig. 1, a-c) that L7 consists of two relatively stable N- and C-terminal domains separated by mobile hinge region spanning residues 33-51. A more rigid part of the NTD comprises the residues from 3 to 31 with mean 15N R1 and R2 values of 1.53 ± 0.05 s-1 and 14.1 ± 1.7 s-1, respectively. The CTD spreads over the residues 52-120 with mean 15N R1 and R2 values of 1.72 ± 0.14 s-1 and 10.7 ± 1.1 s-1, respectively. The 15N nuclei of both N- and C-terminal domains have high positive 15N{1H} NOEs (Fig. 1a), which suggest restricted internal mobility for the NH vectors in pico-nanosecond time scale. Additionally, residues Phe30 and Asn64 from these domains have pronounced enhanced R2, pointing to the micro-millisecond conformational exchange (see Supplementary data). In contrast, the residues 1-2 and 33-51 from N terminus and hinge region, respectively, have nearly unrestricted mobility, resulting in low and negative 15N{1H} NOE and decreased 15N R1 and R2 rates (Fig. 1, a-c). The amides of Ser1 and the hinge region residues, with the exception of Val38 and Val40, have intensive water-exchange cross-peaks in the 1H-15N NOESY-HSQC spectrum. Besides, the residues within the hinge region display only trivial sequential NOE connectivities. Hence, interdomain linker of L7 exhibits all of the hallmarks of a mobile and unstructured region that is exposed to the aqueous environment. Different mean values of R1 and R2 for 15N nuclei of the L7 N- and C-terminal domains strongly suggest that rotational diffusion of the domains occurs at different rates. This becomes evident from the local rotational correlation times calculated from R2/R1 ratios of the individual 15N nuclei of L7 (Fig. 1d). The overall rotation correlation times, τR calculated from R2/R1 ratio averaged over 15N nuclei with 1H{15N} NOE higher than 0.6 (47Korzhnev D.M. Billeter M. Arseniev A.S. Orekhov V.Y. Prog. Nucl. Magn. Reson. Spectrosc. 2001; 38: 197-266Abstract Full Text Full Text PDF Scopus (193) Google Scholar) are 9.45 ± 0.25 and 7.28 ± 0.76 ns for the N- and C-terminal domains, respectively. Using the empirical dependence of the overall correlation time on the number of residues in globular proteins (44Daragan V.A. Mayo K.H. Prog. Nucl. Magn. Reson. Spectrosc. 1997; 31: 63-105Abstract Full Text Full Text PDF Scopus (224) Google Scholar), one can estimate that the dNTD and each of the CTDs of the L7 dimer, having a total of 240 residues, tumble at the same rates as compact proteins of 152 and 115 residues, respectively. These values exceed the actual molecular sizes of the dNTD and the CTD by approximately 88 and 46 residues, respectively. Hence, the interaction of CTD-dNTD-CTD over long flexible hinge regions gives almost equal "overweight" by 45 residues to the size of the NTD (residues 1-32) and the CTD (residues 52-120). This proves our choice of the domain boundaries within a few amino acids. Thus, according to the overall structure of L7/L12 in solution (19Bocharov E.V. Gudkov A.T. Budovskaya E.V. Arseniev A.S. FEBS Lett. 1998; 423: 347-350Crossref PubMed Scopus (32) Google Scholar) where it forms a stable dimer with subunits interacting by the NTDs, one can suggest a model for its rotational diffusion assuming that L7/L12 dimer behaves as a set of three almost independent domains linked by flexible hinges. Namely, the dNTD of L7/L12 dimer behaves as a core unit, whereas two mutually independent CTDs diffuse as wings weakly interacting with the core. That is in agreement with the previous results of NMR and fluorescence polarization measurements, suggesting that a hierarchy of motions exists in the L7/L12 molecule including facile motions of the dNTD and the two CTDs, in addition to the overall tumbling of the protein (17Hamman B.D. Oleinikov A.V. Jokhadze G.G. Traut R.R. Jameson D.M. Biochemistry. 1996; 35: 16680-16686Crossref PubMed Scopus (45) Google Scholar, 19Bocharov E.V. Gudkov A.T. Budovskaya E.V. Arseniev A.S. FEBS Lett. 1998; 423: 347-350Crossref PubMed Scopus (32) Google Scholar). Structure Determination—Taking into account that the domains of the L7 dimer are almost independent in solution and remain in the separated states most of the time, we calculated their spatial structures individually to optimize the computational task. The dNTD was modeled as a dimer of two L7 N-terminal fragments, Ser1-Ala37 and Ser1′-Ala37′, linked by 20 pseudo-residues (l-Gly) to give the monomers enough mutual arrangement. Structure calculation protocol for the dNTD is described in Supplementary data. The full set of input data for structure calculation of the dimer included 686 NOE distance restraints, upper and lower distance restraints for 30 hydrogen bonds, 198 backbone, ϕ and φ, and side chain, χ1, dihedral angle restraints. A representative ensemble of the 20 structures of the dNTD with small violations and low molecular energies is shown in Fig. 2a. The backbone r.m.s.d. of the dimer is 0.56 Å for well defined residues 3-31 and is 0.41 Å when comparing the monomers only. The structure of the CTD was calculated as the L7 C-terminal fragment, Ala47-Lys120, based on NMR-derived constraints including 670 NOE distance restraints, upper and lower distance restraints for 46 hydrogen bonds, 196 backbone, ϕ and φ, and side chain, χ1, dihedral angle restraints. Two final sets of 20 best CTD structures were obtained using separately 66 or 58 1H-15N RDC measured in the alignment media containing the virus particles (Fig. 2b) or the lipid bicelles, respectively. In both virus and lipid media, the backbone conformations of the CTD comprising residues 52-120 are well defined and the overall mean global r.m.s.d. from the average structure are, respectively, 0.36 and 0.47 Å for backbone atoms. A survey of the structural statistics and residual violations of experimental restraints for the dNTD and CTD ensembles is provided in the Table I. The atomic coordinates and experimental restrains for the representative structures and the ensembles of 20 structures of the dNTD and CTD (in virus alignment medium) of L7 dimer in solution have been deposited in the Protein Data Bank (PDB) under accession codes 1RQT and 1RQS, respectively. Additionally, two representative structures of whole L7 dimer (PDB accession code 1RQU and 1RQV) were generated using the NMR structures of the L7 domains mutually remote and linked by two hinge regions, both of which were unstructured or one of them had been turned in a helix with a fashion similar to the crystal structure (24Wahl M.C. Bourenkov G.P. Bartunik H.D. Huber R. EMBO J. 2000; 19: 174-186Crossref PubMed Scopus (79) Google Scholar).Table IStructural statistics for the ensembles of 20 lowest CYANA target function structures of the L7 dimer domainsCTDParameterQuantitydNTDVirus alignment mediumLipid alignment mediumTarget function (Å2)0.11 ± 0.030.51 ± 0.080.25 ± 0.07No. of distance constraintsNOE (intra-/inter-/co-monomer)aThe co-monomer NOE is a NOE connectivity, which has both intermonomer and intramonomer contribution in a symmetric oligomer (63).578/68/40670670H-bond (upper/lower)88/88135/135135/135No. of torsion angle constraintsBackbone146146146Side chains525050No. of orientation constraintsBackbone 1H-15N RDC6658No. of upper constraint violations>0.2 Å000No. of lower constraint violations>0.2 Å000No. of van der Waals constraint violations>0.2 Å000No. of torsion angle constraint violations>5°000No. of 1H-15N RDC constraint violations>0.2 Hz00RDC alignment tensorMagnitude DaThe co-monomer NOE is a NOE connectivity, which has both intermonomer and intramonomer contribution in a symmetric oligomer (63). (Hz)−3.7−6.6Rhombicity0.050.47RDC Q-factor0.140.15R.m.s.d. (Å) of residues (3-31)2 of dNTD and 52-120 of CTDBackbone0.56 ± 0.190.36 ± 0.090.47 ± 0.15All heavy atoms1.05 ± 0.210.99 ± 0.131.08 ± 0.17Ramachandran analysisbRamachandran statistics were determined using PROCHECK_NMR (64). of residues (1-32)2 of dNTD and 52-120 of CTD% residues in most favored regions96.690.087.4% residues in additional allowed regions3.410.012.6% residues in generously allowed regions0.00.00.0% residues in disallowed regions0.00.00.0a The co-monomer NOE is a NOE connectivity, which has both intermonomer and intramonomer contribution in a symmetric oligomer (63Nilges M. O'Donoghue S.I. Prog. Nucl. Magn. Reson. Spectrosc. 1998; 32: 107-139Abstract Full Text PDF Scopus (224) Google Scholar).b Ramachandran statistics were determined using PROCHECK_NMR (64Laskowski R.A. Rullman J.A.C. MacArthur M.W. Kaptein R. Thornton J.M. J. Biomol. NMR. 1996; 8: 477-486Crossref PubMed Scopus (4566) Google Scholar). Open table in a new tab Tertiary Fold of L7/L12
Referência(s)