Biomol NMR Assign DOI 10.1007/s12104-014-9554-2

ARTICLE

Resonance assignment of the ribosome binding domain of E. coli ribosomal protein S1 Pierre Giraud • Jean-Bernard Cre´chet • Marc Uzan • Franc¸ois Bontems • Christina Sizun

Received: 1 February 2014 / Accepted: 15 March 2014 Ó Springer Science+Business Media Dordrecht 2014

Abstract Ribosomal protein S1 is an essential actor for protein synthesis in Escherichia coli. It is involved in mRNA recruitment by the 30S ribosomal subunit and recognition of the correct start codon during translation initiation. E. coli S1 is a modular protein that contains six repeats of an S1 motif, which have distinct functions despite structural homology. Whereas the three central repeats have been shown to be involved in mRNA recognition, the two first repeats that constitute the N-terminal domain of S1 are responsible for binding to the 30S subunit. Here we report the almost complete 1H, 13C and 15N resonance assignment of two fragments of the 30S binding region of S1. The first fragment comprises only the first repeat. The second corresponds to the entire ribosome binding domain. Since S1 is absent from all high resolution X-ray structures of prokaryotic ribosomes, these data provide a first step towards atomic level structural characterization of this domain by NMR. Chemical shift analysis of the first repeat provides evidence for structural divergence from the canonical OB-fold of an S1 motif. In contrast the second domain displays the expected topology for an S1

P. Giraud
F. Bontems
C. Sizun (&) CNRS UPR 2301, Institut de Chimie des Substances Naturelles, 91190 Gif-sur-Yvette, France e-mail: [email protected]; [email protected] J.-B. Cre´chet Ecole Polytechnique, 91128 Palaiseau, France M. Uzan CNRS UMR 7099, Institut de Biologie Physico-Chimique, 75005 Paris, France M. Uzan Universite´ Paris-Diderot, 75005 Paris, France

motif, which rationalizes the functional specialization of the two subdomains. Keywords S1
rpsA
Ribosome
30S
Translation initiation
RNA binding
OB fold

Biological context Atomic resolution structures of prokaryotic ribosomes by X-ray crystallography and cryo-electron microscopy represented a major breakthrough for the comprehension of the molecular bases of protein synthesis (Schuwirth et al. 2005). However S1, a protein encoded by the rpsA gene specific to prokaryotes and the largest protein of the small ribosomal 30S subunit in Escherichia coli with 61 kDa (Subramanian 1983), has eluded structural analysis at the atomic level in the context of the ribosome. Since S1 exhibits less avidity for the ribosome than its other ribosomal counterparts, ribosomal preparations containing S1 are heterogeneous. S1 is therefore removed for X-ray diffraction and high-resolution electron microscopy of ribosomes. S1 promotes recognition of the translation initiation region on mRNAs by the 30S ribosomal subunit in Gramnegative bacteria, in particular when the upstream Shine– Dalgarno recognition signal is missing (Farwell et al. 1992). It is strictly required in E. coli (Sorensen et al. 1998). S1 is also involved in other related functions like transcriptional cycling in vitro (Sukhodolets et al. 2006). Several bacteriophages use host S1 at different stages of their infection cycle: S1 is a subunit of the phage Qb replicase (Wahba et al. 1974; Vasilyev et al. 2013) and activates phage T4 ribonuclease RegB (Ruckman et al. 1994). S1 binds to RNA in an apparently sequence

123

P. Giraud et al.

unspecific manner, even if A/U rich RNA sequences have been shown to activate translation initiation by S1 (Boni et al. 1991). It has been proposed that RNA-binding proceeds via a common mechanism independent of sequence or secondary structure (Aliprandi et al. 2008) and that it acts as an RNA-unwinding protein (Rajkowitsch and Schroeder 2007; Qu et al. 2012). S1 is a modular protein that contains six repeats of the S1 motif, a structural motif commonly found in proteins involved in RNA metabolism. The two first repeats, R1 and R2, constitute the N-terminal 30S binding domain of S1, whereas the downstream repeats are involved in mRNA recognition. The multidomain aspect of S1 likely accounts for the difficulty to characterize the structure of S1: very different shapes were reported for S1 in solution and on the ribosome (Subramanian 1983; Selivanova et al. 2003). The structures of two isolated S1 motifs in S1, corresponding to the fourth and the sixth repeat, R4 and R6, were solved by NMR and display canonical OB-folds (Salah et al. 2009). Evidence was also given for the tight association of the fourth and fifth RNA binding repeats by using X-ray scattering combined to NMR (Aliprandi et al. 2008). Two recent reports were aimed at localizing S1 and more particularly its N-terminal domain on the ribosome with respect to the other ribosomal proteins (Byrgazov et al. 2012; Lauber et al. 2012). Still, in order to build a more precise model of the anchoring of S1 to the 30S ribosome, high resolution structural data are needed. We decided to close the gap by using NMR to solve the structure of the N-terminal domain of S1. Here we report the NMR assignment of two S1 fragments, containing the first and both the first and second S1 motifs respectively, as a first step to determine the structure of the N-terminal domain of S1, the relative positions of the two motifs as well as interaction regions with other ribosomal proteins or RNA.

Methods and experiments Cloning, expression and purification The S1 fragment comprising the two N-terminal S1 motifs, F12 (Met1–Ser179), was cloned from rpsA (E. coli K12 strain ED8689) by PCR in the pET15b vector (Novagen) and overexpressed in E. coli BL21(DE3) (Novagen) (Bisaglia et al. 2003). Expression conditions were adapted to the toxicity of F12 for the host strain. A 15 mL preculture in M9 medium, supplemented with 100 lg/mL ampicillin and inoculated with a 1 mL glycerol stock, was grown until OD600 = 0.6 at 37 °C. It was then transferred into 1 L of M9 medium supplemented with ampicillin. Induction proceeded for 3 h at 37 °C after addition of 0.5 mM IPTG at OD600 = 0.6. Uniform 15N or 13C15N-

123

labeling was achieved by using 1 g/L 15N-NH4Cl (Cortecnet) and 3 g/L unlabeled or 13C-labeled glucose (Cortecnet). For triple labeling 80 % 2H–13C15N, cells were grown in 1 L unlabeled M9 medium in H2O until OD600 = 0.6, then collected by centrifugation, resuspended in 100 mL 13C15N-M9 medium in 99 % D2O (Eurisotop), adapted to this medium for 1 h at 37 °C, collected again by centrifugation and resuspended in 900 mL 2H13C15N-M9 medium. Induction was started 30 min later for 3 h at 37 °C and cell collected by centrifugation. After lysis of the cell pellet in 50 mM sodium phosphate pH 8, 500 mM NaCl buffer with a Cell Disruptor (Constant Systems Limited) and ultracentrifugation in a 45Ti rotor (Beckman) at 35 krpm and for 1 h, purification was performed in four steps. First affinity chromatography was carried out on a 5 mL Nickel HiTrap FF column (GE Healthcare) by elution with lysis buffer plus 500 mM imidazole. A second step consisted of ion exchange on a 6 mL Resource Q (GE Healthcare) starting from 50 mM Tris pH 8, 10 mM NaCl buffer and eluting with 50 mM Tris pH 8, 1 M NaCl, followed by concentration up to 500 lM on an Amicon centrifugal filter unit with a 10 kDa cut-off (Millipore). The N-terminal histidine tag was cleaved by thrombin (20U, thrombin from human plasma, Sigma) overnight at 4 °C. The cleaved fragment was recovered by loading the sample on a 1 mL Nickel HisTrap HP column (GE Healthcare). Finally gel filtration was performed on a Superdex 75 (GE Healthcare) using either sodium or potassium NMR buffer at pH 6.8 (sodium or potassium phosphate 25 mM, NaCl or KCl 200 mM). The final concentration of F12 was adjusted to 300 lM. The F1 fragment (Glu11–Lys100) was obtained by PCR from the pET-F12 plasmid, cloned into pET151/D-TOPO (Invitrogen) and overexpressed in E. coli BL21(DE3) (Novagen). A 15 mL preculture inoculated with a single colony was grown overnight at 37 °C in Luria Broth, supplemented with 0.2 % glucose and 50 lg/mL carbenicillin. After transfer into 1 L of M9 medium supplemented with carbenicillin, the culture was grown until OD600 = 0.6 at 37 °C. Overexpression was achieved with 0.5 mM IPTG overnight at 28 °C. Uniform 15N or 13C15N-labeled F1 samples were obtained by using 1 g/L 15N-NH4Cl (Cortecnet) and 3 g/L unlabeled or 13C-labeled glucose (Cortecnet) in M9. After lysis in the same conditions as for F12, purification was operated by Nickel affinity chromatography. The protein was then concentrated up to 250 lM on an Amicon centrifugal filter unit with a 10 kDa cut-off (Millipore). The double N-terminal 6xHis/V5 epitope tag was cleaved by Histagged TEV protease (0.1 OD280 TEV for 10 OD280 F1). The cleaved protein that contains six additional N-terminal residues (Gly5–Thr10) was isolated by loading the mixture on a 1 mL Nickel HisTrap HP column (GE Healthcare). F1 was exchanged into potassium NMR buffer at pH 6.5 by dialysis

Resonance assignment of the ribosome binding domain

(8 kDa cut-off tubing, Spectra/Por 6) and the final concentration adjusted to 0.2 mM for the 13C15N-F1 sample and 0.5 mM for 15N-F1. NMR spectroscopy NMR experiments were carried out at 293 K on Bruker Avance 600, 800 and 950 MHz spectrometers equipped

with triple resonance cryoprobes. 3D HNCO, HN(CA)CO, HNCA, HN(CO)CA and HNCACB triple-resonance experiments were run at 14.1 T with 200–250 lM 15N13Clabeled F1 and 80 %2H–U15N13C-labeled F12 samples in H2O buffer with 7 % D2O to achieve backbone resonance assignment. In the case of F12, an additional CBCA (CO)NH spectrum was recorded. HBHA(CO)NH spectra were recorded for U15N13C-labeled F1 and U15N13C126G

105 41G

169S

41G *

58G 21G

110

21G

44S 80G

26G

133G

63Q 55N

26G

*

149G

114G

58G

78G *

110G

78G

* 119G 113N 144T

* 80G * *

44S

165N

164N

161Q

125N

57Q 7Q 63Q 55N 178N 62I 155K 176S 82T 22S 121T 2T18T 163R 57Q 109T 154F 51E 53F 104D 165N 52Q 146H 34D * 97T 172A * * 153E 4S 65G 111V 68V 173V 27V 175E 137D 168V 33K 93E 7Q 164N 102Y 43K 116V135L75V 42L 143D 148E 150K 107T 106E 178N 96I 67E 157I 63Q 59E 142R 54K 108V 55N 31I64V77D 14K 174I 99E 101A147L 70V 66D 130F 56A 30A 103E 5F 115K 0H 179S 125N 127I 171R 8L177E 45E 60L 151E 23I98L 158K 136V 38V 120F 3E 17E 13L 105A 24V 76E 6A 61E 122V 25R 145L 39D 36V 160D 1M 72L 19R 141V 159L 48I 166V 29V 161Q 28V 50A 128R 112I 123E 152L 37L 71A 129A 113N 32D 167V 47A 124L * 131L *

57Q

N (ppm)

115

15

118G

120

51E 34D 65G

10T 53F

62I 57Q 52Q

Arg* 27V 97T 75V 33K 43K 9F 90K 95W 42L 63Q 46S 59E 55N 6I 67E 79F35V 84L 88K 91R 31I 81E 56A 64V 16I 30A 77D15E 93E 14K 66D 96I 99E 11E 23I 45E 60L 69D 25R 74A 38V 24V 76E 89A 94A13L 61E 83L 72L 39D 17E 19R 98L 36V 48I 50A 7D 29V 28V 100K 71A 37L 47A

125

130

68V

82T 22S 18T 12S 85S

Arg*

54K

156V

95W

32D

95W

*

40A

40A

10.0

9.5

9.0

8.5

8.0 1

7.5

7.0

6.5

H (ppm)

10.0

9.5

9.0

8.5

8.0 1

7.5

7.0

6.5

H (ppm)

Fig. 1 1H-15N HSQC spectra of 0.5 mM 15N-labeled S1-F1 (left) and 0.2 mM 13C15N-labeled S1-F12 (right) recorded on Bruker Avance 800 and 950 MHz spectrometers respectively, at 293 K. Assignments are given in one-letter code. Numbering starts with Met1. Side-chain

resonances of Asn and Gln are denoted with d and e letters and connected by horizontal lines. The Trp95 He-Ne correlation is indicated with e. Degradation peaks and/or Lys and Arg side-chain resonances are marked with an asterisk

Fig. 2 Sequence alignment of the two first S1 motif repeats of E. coli S1, R1 and R2, with the two repeats of known structure, R4 and R6 (Salah et al. 2009). Secondary structure prediction obtained with TALOS? (Shen et al. 2009) on S1 fragments F1 and F12 indicates

that the repeat R2 contains the five b-strands awaited for an OB-fold in contrast to R1 that lacks the fifth strand. The alignment was edited with EsPript 3.0 (Gouet et al. 2003)

123

P. Giraud et al.

labeled F12. Aliphatic side chain resonances were assigned using 3D CCH-TOCSY experiments for F1 and F12 and an HCCH-TOCSY for F12, all with 20 ms spin-lock durations, recorded on 200 lM U15N13C-labeled samples in 100 % D2O buffer at 22.3 T. Aromatic side chains were assigned using aromatic 3D HCCH-TOCSY experiments as well as a 2D TOCSY experiment with a 60 ms spinlock. Assignment was further assisted by analysis of 3D 13 C-NOESY-HSQC and 2D NOESY spectra in 100 % D2O buffer and 15N-NOESY-HSQC in 93 %/7 % H2O/D2O buffer, with mixing times of 80 ms, recorded at 18.1 or 22.3 T. Data were processed with Bruker Topspin 2.1 and NMRPipe (Delaglio et al. 1995). Spectra were peak-picked and analyzed with CcpNmr Analysis (Vranken et al. 2005). Assignments and data deposition The 15N-HSQC spectra of the S1 protein fragments F1 and F12 are shown in Fig. 1. Assignment completeness for F1 was 95.7 % for amide resonances, 94.5 % for backbone resonances (13Ca, 13Cb, 13C and 1Ha) and 95.4 % for side chain protons. For F12 assignment completeness was 90.8 % for amide resonances, 87.4 % for backbone resonances and 81.7 % for side chain protons. All side-chain amide groups of Asn and Gln were assigned, but none of the exchangeable side-chain protons of Arg and Lys. Chemical shift outliers were found for several nuclei expected to be proximal to aromatic moieties. These are Leu134–13Ca, Gln164–1Hc2, Gln164–1Ha, Val171–1Hca* and Ser172–15N. Missing assignments for F1 are in the region between Leu84 and Lys88 due to line broadening, with no information available for Arg86. This region was predicted to form the fifth b-strand of a b-barrel characteristic of S1 motifs on the basis of sequence alignments with S1 motifs of known structure. The alignment with the fourth and sixth repeats of the S1 protein is shown in Fig. 2. However backbone chemical shift analysis performed by running TALOS? (Shen et al. 2009) does not corroborate this hypothesis. The same region of R1 is even more severely broadened in F12, where chemical shift information is completely missing for residues Leu83– Lys88. Several other assignments are incomplete for residues located in loops where line broadening is also observed, like in regions Leu131–Pro140 or Lys158– Arg163. Apart from the Leu84–Lys88 region, secondary structure predictions with TALOS? and structure calculations with CS-Rosetta (Shen et al. 2008) agree well with homology modelling from sequence alignment. As illustrated in Fig. 2, two b-barrel subdomains can be identified in F12, spanning residues Ile23–Ala71 and Thr107– Ser169. They are separated by a linker with a-helical structure between Ala89 and Asp104. The N-termini of F1 and F12 appear to be disordered, as well as the C-terminus

123

of F1. F12 displays a C-terminal a-helix spanning Arg170– Glu177. It can be seen in Fig. 1 that chemical shifts for the R1 repeat are nearly identical in F1 and in F12, which helped with the assignment of the first subdomain in F12 and strongly indicates that the two subdomains in F12 are independent. The data have been deposited in the BioMagResBank under accession numbers 19550 and 19554 for the F1 and F12 fragments of E. coli S1 protein respectively. Acknowledgments We gratefully acknowledge the Centre National de la Recherche Scientifique (TGIR-RMN-THC FR3050) for access to the 950 MHz NMR spectrometer and the Institut de Chimie des Substances Naturelles (CNRS UPR2301) for a doctoral fellowship to P.G. Conflict of interest of interest.

The authors declare that they have no conflict

References Aliprandi P, Sizun C, Perez J, Mareuil F, Caputo S, Leroy JL, Odaert B, Laalami S, Uzan M, Bontems F (2008) S1 ribosomal protein functions in translation initiation and ribonuclease RegB activation are mediated by similar RNA-protein interactions: an NMR and SAXS analysis. J Biol Chem 283(19):13289–13301. doi:10. 1074/jbc.M707111200 Bisaglia M, Laalami S, Uzan M, Bontems F (2003) Activation of the RegB endoribonuclease by the S1 ribosomal protein is due to the cooperation between the S1 four C-terminal modules in a substrate-dependent manner. J Biol Chem 278(17):15261–15271 Boni IV, Isaeva DM, Musychenko ML, Tzareva NV (1991) Ribosome-messenger recognition: mRNA target sites for ribosomal protein S1. Nucleic Acids Res 19(1):155–162 Byrgazov K, Manoharadas S, Kaberdina AC, Vesper O, Moll I (2012) Direct interaction of the N-terminal domain of ribosomal protein S1 with protein S2 in Escherichia coli. PLoS ONE 7(3):e32702. doi:10.1371/journal.pone.0032702 Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, Bax A (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6(3):277–293 Farwell MA, Roberts MW, Rabinowitz JC (1992) The effect of ribosomal protein S1 from Escherichia coli and Micrococcus luteus on protein synthesis in vitro by E. coli and Bacillus subtilis. Mol Microbiol 6(22):3375–3383 Gouet P, Robert X, Courcelle E (2003) ESPript/ENDscript: extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res 31(13):3320–3323 Lauber MA, Rappsilber J, Reilly JP (2012) Dynamics of ribosomal protein S1 on a bacterial ribosome with cross-linking and mass spectrometry. Mol Cell Proteomics 11(12):1965–1976. doi:10. 1074/mcp.M112.019562 Qu X, Lancaster L, Noller HF, Bustamante C, Tinoco I Jr (2012) Ribosomal protein S1 unwinds double-stranded RNA in multiple steps. Proc Natl Acad Sci USA 109(36):14458–14463. doi:10. 1073/pnas.1208950109 Rajkowitsch L, Schroeder R (2007) Dissecting RNA chaperone activity. RNA 13(12):2053–2060. doi:10.1261/rna.671807 Ruckman J, Ringquist S, Brody E, Gold L (1994) The bacteriophage T4 regB ribonuclease. Stimulation of the purified enzyme by ribosomal protein S1. J Biol Chem 269(43):26655–26662

Resonance assignment of the ribosome binding domain Salah P, Bisaglia M, Aliprandi P, Uzan M, Sizun C, Bontems F (2009) Probing the relationship between Gram-negative and Gram-positive S1 proteins by sequence analysis. Nucleic Acids Res 37(16):5578–5588. doi:10.1093/nar/gkp547 Schuwirth BS, Borovinskaya MA, Hau CW, Zhang W, Vila-Sanjurjo A, Holton JM, Cate JH (2005) Structures of the bacterial ribosome at 3.5 A resolution. Science 310(5749):827–834 Selivanova OM, Shiryaev VM, Tiktopulo EI, Potekhin SA, Spirin AS (2003) Compact globular structure of Thermus thermophilus ribosomal protein S1 in solution: sedimentation and calorimetry study. J Biol Chem 278:36311–36314 Shen Y, Lange O, Delaglio F, Rossi P, Aramini JM, Liu G, Eletsky A, Wu Y, Singarapu KK, Lemak A, Ignatchenko A, Arrowsmith CH, Szyperski T, Montelione GT, Baker D, Bax A (2008) Consistent blind protein structure generation from NMR chemical shift data. Proc Natl Acad Sci USA 105(12):4685–4690. doi:10.1073/pnas.0800256105 Shen Y, Delaglio F, Cornilescu G, Bax A (2009) TALOS?: a hybrid method for predicting protein backbone torsion angles from NMR chemical shifts. J Biomol NMR 44(4):213–223. doi:10. 1007/s10858-009-9333-z

Sorensen MA, Fricke J, Pedersen S (1998) Ribosomal protein S1 is required for translation of most, if not all, natural mRNA in Escherichia coli in vivo. J Mol Biol 280:561–569 Subramanian AR (1983) Structure and functions of ribosomal protein S1. Prog Nucleic Acids Res Mol Biol 28:101–142 Sukhodolets MV, Garges S, Adhya S (2006) Ribosomal protein S1 promotes transcriptional cycling. RNA 12(8):1505–1513 Vasilyev NN, Kutlubaeva ZS, Ugarov VI, Chetverina HV, Chetverin AB (2013) Ribosomal protein S1 functions as a termination factor in RNA synthesis by Qbeta phage replicase. Nat Commun 4:1781. doi:10.1038/ncomms2807 Vranken WF, Boucher W, Stevens TJ, Fogh RH, Pajon A, Llinas M, Ulrich EL, Markley JL, Ionides J, Laue ED (2005) The CCPN data model for NMR spectroscopy: development of a software pipeline. Proteins 59(4):687–696. doi:10.1002/prot.20449 Wahba AJ, Miller MJ, Niveleau A, Landers TA, Carmichael GG, Weber K, Hawley DA, Slobin LI (1974) Subunit I of Q beta replicase and 30 S ribosomal protein S1 of Escherichia coli. Evidence for the identity of the two proteins. J Biol Chem 249(10):3314–3316

123