ebook img

Dissecting the in vivo assembly of the 30S ribosomal subunit reveals the role of RimM and general features of the assembly process. PDF

10.5 MB·English
by  GuoQiang
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Dissecting the in vivo assembly of the 30S ribosomal subunit reveals the role of RimM and general features of the assembly process.

Published online 4 January 2013 Nucleic Acids Research, 2013, Vol. 41, No. 4 2609–2620 doi:10.1093/nar/gks1256 Dissecting the in vivo assembly of the 30S ribosomal subunit reveals the role of RimM and general features of the assembly process Qiang Guo1, Simon Goto2, Yuling Chen1, Boya Feng1, Yanji Xu1, Akira Muto2, Hyouta Himeno2, Haiteng Deng1, Jianlin Lei1,* and Ning Gao1,* 1Ministry of Education Key Laboratory of Protein Sciences, Center for Structural Biology, School of Life Sciences, Tsinghua University, Beijing 100084, China and 2Department of Biochemistry and Molecular Biology, Faculty of Agriculture and Life Science, RNA Research Center, Hirosaki University, Hirosaki 036-8561, Japan Received August 20, 2012; Revised November 2, 2012; Accepted November 3, 2012 INTRODUCTION ABSTRACT Ribosome biogenesis is a tightly regulated multi-stepped Ribosome biogenesis is a tightly regulated, multi- process,assistedbyawidevarietyofproteinfactors,such stepped process. The assembly of ribosomal as transcription factors, endoribonucleases, rRNA subunitsisacentralstepofthecomplexbiogenesis helicases and chaperones, rRNA and ribosomal protein process,involvingnearly30proteinfactorsinvivoin modification enzymes and assembly factors (1). As to bacteria. Although the assembly process has been the 30S subunit, early in vitro reconstitution experiments extensively studied in vitro for over 40 years, very (2–6)havedemonstratedthatactive30Ssubunitscouldbe limited information is known for the in vivo formed from purified ribosomal proteins and 16S rRNA process and specific roles of assembly factors. in the absence of other cellular components. The in vitro Such an example is ribosome maturation factor assembly occurs very slowly and requires non- M (RimM), a factor involved in the late-stage physiological conditions, such as high Mg2+ concentra- tion, high ion strength and heat shock. In contrast, the assembly of the 30S subunit. Here, we combined assembly of the 30S subunit in vivo starts with rRNA quantitative mass spectrometry and cryo-electron primary transcripts (7) and occurs co-transcriptionally microscopy to characterize the in vivo 30S (8) in a much more efficient way, underscoring the essen- assembly intermediates isolated from mutant tial contribution of assembly factors. In recent years, Escherichia coli strains with genes for assembly application of new techniques, such as pulse-chase moni- factors deleted. Our compositional and structural tored by quantitative mass spectrometry (PC/QMS) (9), data show that the assembly of the 30-domain of time-resolved X-ray footprinting (10) and time-resolved the 30S subunit is severely delayed in these inter- electron microscopy (11), has brought our understanding mediates, featured with highly underrepresented oftheinvitroassemblyprocesstoanewlevel,providinga 30-domain proteins and large conformational differ- large amount of valuable kinetic and structural informa- encecomparedwiththemature30Ssubunit.Further tion. Together with earlier work [reviewed in (12)], these data have established that the in vitro 30S subunit analysis indicates that RimM functions not only to assembly starts from multiple sites on the 16S rRNA promote the assembly of a few 30-domain proteins (10), following parallel pathways (9–11) and the free but also to stabilize the rRNA tertiary structure. energy of the assembly can be represented by a complex More importantly, this study reveals intriguing landscape (9). More importantly, kinetic data revealed similarities and dissimilarities between the in vitro thatforseveralsubsetsof30-domainproteins,thethermo- and the in vivo assembly pathways, suggesting dynamic interdependence does not align well with that they are in general similar but with subtle measured kinetic cooperativity (11,13), and at these loca- differences. tions, the in vitro assembly often encounters kinetic traps, *To whom correspondence should be addressed. Tel:+86 01062794277; Fax:+86 01062771145; Email: [email protected] Correspondence may also be addressed to Jianlin Lei. Tel:+86 01062797699; Fax:+86 01062771145; Email: [email protected] The authors wish it to be known that, in their opinion, the first three authors should be regarded as joint First Authors. (cid:2)TheAuthor(s)2013.PublishedbyOxfordUniversityPress. ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense(http://creativecommons.org/licenses/by-nc/3.0/),which permitsnon-commercialreuse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.Forcommercialre-use,pleasecontact [email protected]. 2610 NucleicAcidsResearch,2013,Vol.41,No.4 suggestingthatassemblyfactorsmightbeinvolvedinsub- strain. Then, the kanamycin-resistant cassette was verting kinetic traps in the assembly landscape (9,11,13). removedfromtheintermediatestrainusinganFLPexpres- Over the past two decades, accumulating experimental sionplasmidpCP20(24)toproducetheA19(cid:2)rimMstrain. data, mainly through genetic approaches, has implicated a A19(cid:2)rbfA(cid:2)rsgAisanA19derivativeinwhichbothofthe number of factors, including RbfA, RsgA, KsgA, Era, rbfA and rsgA genes are replaced by a short peptide gene ribosome maturation factor M (RimM), RimP, RimJ containing an FRT sequence, constructed by transducing [(reviewed in (12)] and YqeH (14,15), in the maturation of rbfA::FRT-kan-FRT into A19 and removing kan using the30Ssubunitinbacteria.However,thespecificmolecular pCP20 and then transducing rsgA::FRT-kan-FRT into rolesofmostofthesefactorsremainunclear.Amongthese the resulted strain and removing kan using pCP20. factors,RimMwasfirstidentifiedasafactorrequiredfora Sources of rbfA::FRT-kan-FRT and rsgA::FRT-kan-FRT fastgrowthinrichmedium(16).Thegene-encodingRimM are intermediate strains produced during the construction (yfjA)inEscherichiacoliisco-localizedtothetrmDoperon ofW3110(cid:2)rbfA(25)andthersgA-disruptedstrainofKeio (17) with genesfor ribosomal proteinsS16 and L19, and a collection (23), respectively. Both the A19(cid:2)rimM and tRNAmethyltransferase(TrmD),ahintthatRimMmight A19(cid:2)rbfA(cid:2)rsgA strains were confirmed with polymerase be directly involved in ribosome-related function. Indeed, chain reaction (PCR). deletion of RimM confers a slow growth phenotype (18), with accumulation of 16S rRNA precursors and free 30S Spot assay and ribosome profile subunits (19) as well as reduced level of polysomes (20). A19, A19(cid:2)rimM and A19(cid:2)rbfA(cid:2)rsgA strains were RimM associates with free 30S subunit in vivo (18,20) and grown in liquid LB at 37(cid:2)C to OD 0.8 and diluted to a alsobindstoS19invitro(20,21).Moreover,suppressormu- series of concentrations, 100, 10(cid:3)1, 10(cid:3)2, 10(cid:3)3, 10(cid:3)4 and tations to the (cid:2)rimM mutant were found on S13 (18) and 10(cid:3)5. Three microliters of each dilution was dropped to a suppressormutationstoarimM-Y106AY107Amutantwere LB plate and incubated at 37(cid:2)C overnight. The cell foundonS19,helices31and33bofthe16SrRNA(20). extracts from the A19, A19(cid:2)rimM and A19(cid:2)rbfA(cid:2)rsgA Inthisstudy,wecharacterizetheimmature30Ssubunits strains were loaded onto a 10–40% sucrose gradient con- purified from an E. coli (cid:2)rimM strain biochemically and taining 10mM Mg(OAc) and centrifuged for 3.5h at structurally. Our data indicate that the immature 30S 2 39000rpm in a SW41 rotor (Beckman Coulter). The gra- subunits are a collection of assembly intermediates, with dients were analyzed with A254 absorbance using a the 30-head domain proteins severely underrepresented, Teledyne ISCO fractionation system. such as S10, S14, S13 and S19. Moreover, protein com- position analysis of another category of immature 30S Immature and mature 30S subunit purification subunits from a (cid:2)rbfADrsgA strain shows a different spectrum, with much enhanced levels for these proteins, Escherichia coli cells (A19, A19(cid:2)rimM and suggesting that RimM promotes the assembly of these A19(cid:2)rbfA(cid:2)rsgA strains) grown in LB medium were har- slow binding proteins in vivo. Structural analysis shows vested, lysed and clarified in opening buffer [20 mM Tris– that these (cid:2)rimM intermediates also differ largely in HCl (pH=7.5), 150mM NH Cl, 10mM Mg(OAc) and 4 2 rRNA conformation, particularly the rotational position 0.5mM ethylenediaminetetraacetic acid (EDTA)]. The of the 30-head domain relative to the body domain. An lysate was loaded onto the top of 5ml sucrose cushion incubation of recombinant RimM with the immature 30S [20mM Tris–HCl (pH=7.5),150mM NH4Cl, 10mM subunits significantly reduces the flexibility of the head Mg(OAc)2, 0.5mM EDTA and 1.1M sucrose] and centri- domain. More importantly, our data also suggest that fuged for 18h at 28000rpm in a 70Ti rotor (Beckman the in vivo assembly process occurs along multiple Coulter). The resulting pellets were resuspended in binding pathways in a certain degree as well, and the rRNA mat- buffer and centrifuged through a 10–40% sucrose gradient urationis tightly coupled with ribosomal protein binding. with 10mM Mg(OAc)2 for 7h at 30000rpm in a SW32 The functional depiction of RimM thus illustrates that rotor (Beckman Coulter). Fractions containing the there are possible checkpoints along the in vivo assembly immature 30S and 70S peaks were pooled separately and pathways where maturation factors come into play to concentrated with buffer changed to binding buffer for the direct the process to more efficient branches. 30S fractions and to separation buffer [20 mM Tris–HCl (pH=7.5), 150mM NH Cl and 2mM Mg(OAc) ] for the 4 2 70S fractions. The 70S fractions were further centrifuged MATERIALS AND METHODS through a 10–40% sucrose gradient with 2mM Mg(OAc) 2 to get the mature 30S and 50S subunits. Escherichia coli strains We used E. coli A19 (Hfr, rna-19, gdhA2, his-95, relA1, RimM preparation, rRNA extraction and identification of spoT1, metB1) (22) as the source of the 30S subunit. the 30 and 50ends of the 17S rRNA A19(cid:2)rimM is an A19 derivative in which the rimM gene Full details are available in the Supplementary Data. is replaced by a short peptide gene containing an FRT sequence, constructed as follows. The kanamycin-resistant Pelleting assay marker of a rimM disruptant from Keio collection (23), in which the rimM gene has been substituted by an Mature or immature 30S subunits (2.5pmol) were FRT-flanked kanamycin-resistant cassette, was transduced incubated with 30-fold excess of RimM for 15min at into A19 using phage P1vir to produce an intermediate 37(cid:2)C in binding buffer. The mixture was then layered NucleicAcidsResearch,2013,Vol.41,No.4 2611 onto a 150ml sucrose cushion and centrifuged at outliers. Ratios of two or more tryptic peptides from the 96409rpm for 4h in a TLA-120.1 rotor (Beckman same protein were used to calculate the means and the Coulter). The pellets and the supernatants were separated standard deviations (Supplementary Table S1). and 1/2 of total pellets and 1/20 of supernatants were resolved by 12% sodium dodecyl sulfate–polyacrylamide Cryo sample preparation and cryo-electron microscopy gel electrophoresis (SDS–PAGE). Cryo-grids for the immature 30S subunits were prepared as previously described (26). The grids were examined in Quantitative mass spectrometry an FEI Tecnai F20 microscope operated at 200kV, and For quantitation of targeted protein, samples with same images were recorded at a nominal magnification of A260 absorption value were separated on 1D 80000(cid:4) on a Gatan UltraScan 4000 CCD camera, Tricine-SDS–PAGE. Among all ribosomal proteins, S1 under low-dose conditions ((cid:5)20e-/A˚ 2). The complex of was not included in the QMS analysis, because it dissoci- the immature 30S subunit bound with RimM was ates readily from the 30S subunits during centrifugation- formed by an incubation of a 40-fold excess of RimM based purification. The gel bands corresponding to the with the immature 30S subunits at 37(cid:2)C for 15min. The targeted protein were excised from the gel, reduced with grids of the 30S complex were examined in an FEI Titan 10mMofDithiothreitol(DTT)andalkylatedwith55mM Krios cryo-TEM operated at 300kV, and images were iodoacetamide.Then,in-geldigestionwasperformedwith collected at a nominal magnification of 59000(cid:4) on an the sequence grade modified trypsin (Promega) in 50mM FEI Eagle 4k(cid:4)4k CCD camera, under low-dose condi- ammonium bicarbonate at 37(cid:2)C overnight. The peptides tion. Data collection was done with AutoEMation were extracted twice with 1% trifluoroacetic acid in 50% software package (27). acetonitrile aqueous solution for 30min. The extractions Image processing werethencentrifugedinaspeedvactoreducethevolume. Peptidesfromdifferentsampleswerelabeledwithtandem All the micrographs were decimated by a factor of 2. mass tags (TMT) reagents (Thermo, Pierce Particle picking was performed using the SPIDER Biotechnology) according to the manufacturer’s instruc- package(28)withamethodbasedonalocallynormalized tion (TMT 127, 129 and 130 for the samples from the cross-correlation function (29). The resulting particles A19 mature 30S, A19(cid:2)rimM and A19(cid:2)rbfA(cid:2)rsgA (125(cid:4)125 in window size, 2.76 and 3.0A˚ in effective samples, respectively). Briefly, the TMT label reagents pixel size, for the 30S and the 30S complex samples, re- were dissolved by anhydrous acetonitrile and carefully spectively) were manually verified using a method based added to each digestion products. The reaction was per- on correspondence analysis (30). To ensure the perform- formed for 1h at room temperature, and hydroxylamine ance of the 2D and 3D analysis, particles were further was used to quench the reaction. The TMT-labeled subjected to another round of manual screen, which peptides were desalted using the stage tips. finally rendered 164368 and 94535 particles for the 30S For LC-MS/MS analysis, the TMT-labeled peptides and 30S complex, respectively. The parameters of the were separated by a 65-min gradient elution at a flow contrast transfer function (CTF) were estimated using rate of 0.250ml/min with an EASY-nLCIITM integrated SPIDER at the micrograph level. Particles were then nano-HPLCsystem(Proxeon),whichisdirectlyinterfaced CTF corrected using the phase-flipping method (31). with a Thermo LTQ-Orbitrap mass spectrometer. The 2D image classification was performed using a analytical column was a home-made fused silica capillary maximum-likelihood approach (32) with the XMIPP column (75mm ID, 150mm length; Upchurch) packed software package (33). Particles from both samples were with C-18 resin (300A, 5mm; Varian). Mobile phase A classified into 100 groups in 100 iterations, and the per- consisted of 0.1% formic acid and mobile phase B con- formance of the classification was monitored by sisted of 100% acetonitrile and 0.1% formic acid. The log-likelihood function. To facilitate further comparison, LTQ-Orbitrap mass spectrometer was operated in the class average images were subjected to a multi-reference data-dependent acquisition mode using the Xcalibur alignmentto832Dprojectionsgeneratedfromacryo-EM 2.0.7 software and there was a single full-scan mass mapofthemature30Ssubunit(26),atanangularinterval spectrum in the Orbitrap (400–1800m/z, 30000 reso- of 15(cid:2) (Supplementary Figure S2). lution)followedbythreeMS/MSscansinthequadrupole 3D classification was performed using a 3D maximum- collisioncellusingthehigherenergycollisiondissociation. likelihood approach with XMIPP (34). The initial model The MS/MS spectra from each LC-MS/MS run were was generated by low-pass filtering (60A˚ ) of a cryo-EM searched against the selected database using an in-house map of the mature 30S subunit (26). Particles from the Mascot or Proteome Discovery searching algorithm. both samples were classified into five groups in 50 iter- Peptides that have XCorr/Charge scores >2.75 for 2+ ations, at an angular sampling of 10(cid:2). Refinements of and 3.0 for 3+were used for protein identification and theclassstructureswereperformedwithSPIDER,follow- MS/MS spectra for all matched peptides were manually ingthestandardreferenceprojectionmatchingprocedures interpreted and confirmed. The QMS experiments were (31), witha gradualdecreaseoftheangular stepfrom 15(cid:2) repeated for three times and similar results were to 1(cid:2). Amplitude correction to the density maps was per- obtained. For TMT quantification of a specific protein, formed as previously described (26,35). The final reso- ratios of 129:127 and 130:127 for each of the ribosomal lutions of the refined density maps were estimated with a proteins were examined by Grubbs’ test to remove soft Gaussian mask approach (36,37) using 0.5 cutoff 2612 NucleicAcidsResearch,2013,Vol.41,No.4 criterion of the Fourier Shell Correlation (Supplementary Table S2). Atomic model and temperature map building The head and body domains of a 30S subunit crystal structure [PDB ID: 3OFA, (38)] were docked into the cryo-EM maps as rigid bodies first using Chimera (39), followed by a flexible fitting method based on molecular dynamicssimulation (40)invacuofor1000000stepswith a 0.5-kcal mol(cid:3)1 scaling factor using NAMD (41). To avoid overfitting, ribosomal protein S2 and S3 were removed from the initial model before flexible fitting due Figure 1. Phenotypesofthe(cid:2)rimMand(cid:2)rbfA(cid:2)rsgAstrains.(A)Spot assayshowingthatboththe(cid:2)rimM((cid:2))and(cid:2)rbfA(cid:2)rsgA((cid:2)(cid:2))strains totheirlowoccupancies.ForclassNo.5oftheimmature grow slowly, compared with the wild-type strain (WT). (B) Ribosome 30Ssubunits(SupplementaryTableS2),allproteinsinthe profile analysis of the A19, (cid:2)rimM and (cid:2)rbfA(cid:2)rsgA strains. The head domain were removed and only the rRNA structure profile curves of the WT, (cid:2) and (cid:2)(cid:2) strains are colored in black, wasrefined.Forbettercomparison,afterfitting,S2andS3 green and red, respectively. Deletion of RimM or a combination of proteinswereaddedbacktothefittedstructureusingtheir RbfA and RsgA causes an accumulation of immature 30S subunits. Both experiments indicate that deletion of RimM is more deleterious. contactingrRNAhelicesasreference.The10modelswere aligned using the 30S body domain as reference and 10 protein gel analysis shows that some ribosomal proteins, temperature maps were constructed in PyMOL (42) by e.g.S2andS3,areunderrepresentedinthe(cid:2)rimMsample calculating the deviation of the 16S rRNA in the fitted models from the mature 30S structure. The scripts used (Figure 2B). The compositional heterogeneity suggests for root-mean-square deviation (RMSD) calculation and that the immature 30S subunits from the (cid:2)rimM strain temperature map visualization were downloaded from are a collection of in vivo assembly intermediates that are http://pldserver1.biochem.queensu.ca/(cid:5)rlc/work/pymol/. different in protein composition. ChimeraandPyMOLwereusedforgraphicvisualization. To determine the protein levels, similar to a previously established quantification method (45), we used a QMS technique based on TMT labeling (46). The QMS data RESULTS reveal that the levels of S21, S10, S14, S13, S19, S3, S2 and S5 are dramatically reduced in the (cid:2)rimM sample, Construction of a series of E. coli A19 strains <50% of those in the mature 30S subunits (Figure 2C RNase I is the major non-specific endoribonuclease and Supplementary Table S1). Most of them are second- localized in periplasm and often found to be in 30S aryandtertiarybindingproteinsfromthe30-headdomain subunit fractions in cell extracts (43). To avoid undesired of the 30S subunit, except that S21 and S5 are tertiary degradation of the rRNA precursors in the immature 30S binder from the central domain and the 50-domain, re- subunits during the sample preparation, we chose the spectively. S21 is known to easily dissociate in solution RNase I defective A19 strain (22) as the source of the (47) and is therefore not included for further analysis. 30S subunits. In this genetic background, we further con- Thus, these data clearly demonstrate that the deletion structedstrainswiththerimMgenedeletedandwithboth of RimM causes a severe delay in the assembly of the rsgA and rbfA genes deleted. The two resulting strains 30-domain of the 30S subunit in vivo (Figure 2C and grow poorly on LB medium and show an accumulation Supplementary Figure S3). Among these 30-domain offree30Ssubunits(Figure1).Interestingly,boththecell proteins, S7 has the highest occupancy (81%) in the growth test (Figure 1A) and the ribosome profile analysis (cid:2)rimM sample, which is in accordance with the in vitro (Figure 1B) show that the deletion of rimM is more dele- assembly map that S7 is a primary binder and directs the terious.Asaresult,thereisanintermediatepeakbetween binding of all the rest 30-domain proteins (48). the 30S and 50S peaks, probably representing immature Incontrast,theproteincompositionoftheimmature30S 50S precursors caused by globally decreased protein pro- subunitsfromthe(cid:2)rbfA(cid:2)rsgAstrainshowsintriguingdif- duction in the (cid:2)rimM strain (Figure 1B). ference and similarity (Supplementary Figure S3). The most underrepresented protein in the (cid:2)rbfA(cid:2)rsgA Compositional characterization of the immature 30S sample is still S21 (30%), followed by S7, S2, S10, S11 subunits from the A19 "rimM and "rbfA"rsgA strains and S19 (49–63%) (Figure 2C and Supplementary Table RNA gel analysis shows that the rRNAs in the 30S frac- S1),clearlyshowingadifferentpattern.Althoughmany30- tions from the (cid:2)rimM strain and the (cid:2)rbfA(cid:2)rsgA strain domainproteins,suchasS10,S13,S14,S19andS3,arealso are 16S rRNA precursors (Figure 2A), indicating that underrepresented, theirlevels aresignificantly higher than thesefree30Ssubunitsareindeedimmature30Sparticles. thoseinthe(cid:2)rimMsample(Figure2DandSupplementary Identificationofthetwosetsof16SrRNAprecursorsbya Table S1). In fact, the (cid:2)rbfA(cid:2)rsgA sample has a higher previously established 5030-rapid amplification of com- level for almost all the proteins, compared with the plementary DNA ends (RACE) technique (44) reveals (cid:2)rimM sample (Figure 2C and Supplementary Figure that a majority of these precursors are unprocessed at S3). For example, S10, S13 and S14 have an over 2-fold both the 50- and 30-ends (Supplementary Figure S1). The increase and S21, S19 and S3 have a moderate increase, NucleicAcidsResearch,2013,Vol.41,No.4 2613 Figure 2. Compositional characterizationoftheimmature30Ssubunits.Compositionofmature andimmature 30SsubunitsfromtheA19(cid:2)rimM ((cid:2))and(cid:2)rbfA(cid:2)rsgA((cid:2)(cid:2))strainswasanalyzedinboththeRNAandproteinlevels.(A)RNAgelanalysisoftherRNAsinthe(cid:2)and(cid:2)(cid:2)samples. (B)Tricine-SDS–PAGEanalysisoftheproteincompositionoftheimmature30Ssubunits.(C)QMSanalysisoftheproteincompositioninthe(cid:2)and the(cid:2)(cid:2)samples.Errorbarsshowstandarddeviations.Thedifferenceinproteinratiobetweenthe(cid:2)and(cid:2)(cid:2)sampleswassubjectedtoaone-tailed t-test,whichreportsasignificantdifferenceforS10,S14,S13,S3,S12,S5,S4,S7(P<0.01),S19andS6(P<0.05).(D)Therelativeproteinratiosof the(cid:2)(cid:2)sample tothe(cid:2)sample ((cid:2)(cid:2)/(cid:2))areplotted againsttheratios ofthe (cid:2)sampletothe matureone((cid:2)/mature).(E)Atomicstructureofthe mature 30S subunit (38) viewed from the inter-subunit and solvent sides, with proteins in the top-left part of (D) colored in green. from1.5-to2-folds.Interestingly,twoprimaryproteins,S7 method to our sample. First, a reference-free 2D image and S4, display significantly lower levels in the classificationtechniquebasedonmaximum-likelihoodop- (cid:2)rbfA(cid:2)rsgA sample than in the (cid:2)rimM sample (Figure timization(32)wasemployedtoestimatethelevelofstruc- 2C and D). Taking together, an evident pattern is that tural variation in the cryo-EM particles. The 2D analysis the immature 30S subunits from the (cid:2)rbfA(cid:2)rsgA strain reveals that a large number of the class average images havesignificantlyhigheroccupanciesforallthesecondary show smeared densities on the head domain of the 30S andtertiarybindingproteinsinthe30-domain(Figure2E), subunit.Incontrast,densitiesintheseaverageimagescor- suggesting that their 30-head domains are indeed further responding to the body domain are nicely resolved, and maturatedwithmoreproteinsincorporated. the features of the body domain could be easily identified ThisimmediatelysuggeststhataroleofRimMinvivois (Supplementary Figure S2). This suggests that the to promote the binding of 30-domain proteins, since the immature 30S subunits are truly composed of multiple immature30Ssubunitsfromthe(cid:2)rbfA(cid:2)rsgAstrainlikely assembly intermediates, with a highly flexible head resemble a stage downstream the RimM action. In agree- domain and a rather rigid body domain. ment with this conclusion, the in vitro kinetic data show Next, a multi-structure refinement method (34) was that RimM accelerates the binding of some head domain used to investigate possible metastable structural inter- proteins, S19, S10 and S3 (49). mediates at the 3D level. As a result, the particles were grouped into five classes, and as expected, the five Structural characterization of the immature 30S subunits cryo-EM maps (at 12–14A˚ resolution) display dramatic from the "rimM strain conformational differences (Figure 3). However, similar To explore the structural heterogeneity of the immature to a previous cryo-EM study on the immature 30S 30S subunits, we applied the cryo-EM single-particle subunits from a (cid:2)rsgA strain (50), we did not find 2614 NucleicAcidsResearch,2013,Vol.41,No.4 Figure 3. Overview of the five cryo-EM structures of the immature 30S subunits from the (cid:2)rimM strain. The five density maps (A–E or F–J, respectively) are displayed in transparent surface representation, superimposed with flexibly fitted crystal structures in cartoon representation. For eachmap,boththeinter-subunitview(A–E)andthesolventview(F–J)aredisplayed.The16SrRNA,S2,S3andtherest30Ssubunitproteinsare paintedinblue,green,redandpurple,respectively.Deviationsofthe16SrRNAbackbonesinthefittedmodelfromthatofthemature30Ssubunit are colored as indicated by the scale to form the temperature maps (K–O). significantdensitiesthatcouldbeattributedtotheunpro- different among the five maps (Supplementary cessed ends of the 17S rRNA. Figure S4). Third, in agreement with the QMS data, To facilitate the quantitative comparison of the struc- these structures differ in protein composition, as tural data, we built pseudo-atomic models for the five exemplified by S2 and S3 (Figure 3F–J). In fact, none of cryo-EM maps using a flexible fitting technique (40). the structures has a full occupancy for both factors, and Based on these models, five temperature maps for the interestingly the occupancy of S2 has no correlation with 16S rRNA were then constructed according to their devi- the occupancy of S3 (Supplementary Table S2). This ob- ations from the structure of the mature 30S subunit servation appears to align well with the in vitro assembly (Figure 3K–O). Structural difference can be directly data showing that S2 and S3 could bind in independent identified from these maps. First, the conformational dif- order and the prior binding of S2 ahead of S3 leads to ference is dominated by a relatively rigid rotational kinetically trapped intermediates (11). movement of the head domain, which changes the inter- Insummary, as seenin thetemperature maps,thehead domain orientation between the head and the body domainofthe16SrRNAishighlymobileintheimmature domains. Especially, one class has a nearly 60(cid:2) rotated 30S subunits. It is known that the motion between the headdomain(Figure3E,JandO).Thisrotatedstructure, head domain and the body domain is intrinsic and is derived from nearly one-third of all the particles believed to be required for the dynamic interaction with (Supplementary Table S2), is in fact very similar to one translationalcomponents(51).However,theheaddomain oftheGroupIIinvitroassemblyintermediatesdiscovered rotation observed in our structures is in a much larger in a time-resolved electron microscopy study (11), which scale, suggesting that hypo-level of proteins in the was shown to miss nearly all the 30-domain proteins. In 30-domain increases its flexibility. This structural observa- tion demonstrates that the in vivo intermediates from the additiontotherotation,intwoclasses(Figure3KandM), (cid:2)rimM strain vary not only in protein composition but the channel between the head and the body domains is also in rRNA conformation. closed up, resulting in a narrow down of the mRNA entrance. Second, four of the five maps show very incom- Structural characterization of the "rimM immature 30S plete, fragmented densities for helix 44 of the 16S rRNA, subunits bound with RimM except for one group (Figure 3A), which is close to the conformation of a mature 30S subunit and also has a less Next,weexaminedthebindingpreferenceofRimMtothe rotated head domain. Along with the conformational dif- immature and mature 30S subunits by pelleting assay. ference at helix 44, the decoding center is also sharply While RimM shows almost no binding to the mature NucleicAcidsResearch,2013,Vol.41,No.4 2615 30Ssubunit,itindeedbindstotheimmaturesubunit,with RimM also has a role in stabilizing the rRNA tertiary a low affinity (Figure 4). In contrast, both RfbA and structure in the 30-domain. RsgA show a considerably higher affinity to the mature 30S subunit containing the 16S rRNA than RimM does, Binding position of RimM on the 30S subunit and especially, RsgA displays a strong preference to the The pelleting assay indicates that the affinity of RimM to mature 30S subunit (25). Therefore, similar to our QMS the immature 30S subunits is very low (Figure 4). This data, the binding preferences of these factors also suggest apparently sets an obstacle for us to analyze contact that RimM acts, prior to RbfA and RsgA, in the in vivo sites of RimM on the 30S subunit in detail. Fortunately, assembly pathway. RimMbindstoS19invitro(20,21)andcouldco-crystalize We then sought to explore possible structural changes withS19(PDBID:3A1P).Therefore,thebindingposition oftheimmature30SsubunitsuponRimMbinding.Using of RimM could be deduced using S19 as a reference the same 2D and 3D image classification techniques, we (Figure 6), given that RimM does not change its found that the addition of RimM to the immature 30S contacts in the context of the immature 30S subunit. In subunits seems to stabilize the 30S head domain. At the fact, the cryo-EM maps of the RimM-treated immature 2D level, class averages of particles from the RimM- 30S subunits, although prepared with a 40-fold excess of treated sample still show smeared densities in the head RimM, have limited densities at locations expected to region, but the total fraction of the particles with an have RimM bound when the maps are displayed at a 3s unstableheaddomainissignificantlysmaller(Supplemen- level (Figure 5). Densities corresponding to RimM begin tary Figure S2). About 18% of particles from the un- to appear in lower threshold (Supplementary Figure S6). treated sample display apparent instability in the head Nevertheless,wecouldcomparetheaveragedensitiesstat- domain, whereas the percentage in the treated sample is istically, within a 3D binary mask generated from the decreased to 11%. alignedRimMcrystalstructure.Consistently,thedensities At the 3D level, similarly, we classified the at RimM-bound region in the cryo-EM maps from the RimM-treated data into five groups, and these cryo-EM RimM-treated sample are significantly higher than those maps (at 15–19A˚ resolution) also differ in conformation from the untreated sample (Supplementary Figure S6). and protein composition (Figure 5). First, the head This analysis, albeit rather preliminary, proves that domain rotation is apparently in a much less scale, as RimM is present in these cryo-EM maps. seen in the temperature maps (Figure 5K–O and Supple- The structure of RimM is composed of two b-barrels mentary Figure S5). Second, regions, such as the long containing domains (21). While the C-terminal domain is helix 44 and the decoding center still display a large showntointeractwithS19,theN-terminaldomainclosely amount of variation (Supplementary Figure S4), resembles a tRNA-binding domain of EF-Tu (21), sug- implying the final accommodation of helix 44 is gesting the ability of RimM to bind to the 16S rRNA. probably a later event, not related to RimM binding. Consistently, alignment of the structure of the Third, as expected, the occupancies of S2 and S3 are RimM-S19 complex immediately places the N-terminal both very low, but surprisingly, the levels of S2 and S3 domain of RimM at the junction of several helices, such seemtobeevenlowerthantheuntreatedsample(Supple- as h29, h30 and h42 (Figure 6). Since, prior to RimM mentary Table S2). This finding suggests that RimM sta- binding, the 30S assembly is in a stage with very limited bilizes the immature 30S subunits in a conformation that 30-domain protein incorporated (Figure 2C), the binding disfavors S2 and S3 binding, implying that S2 and S3 ofRimM atthismulti-helicesinterfacemightstabilizethe binding might be later events in the assembly pathway. rRNAconformationgloballyandthereforeallowsafaster Therefore, our structural analysis of the cryo-EM and/or more stable binding of 30-domain proteins. images from the RimM-treated sample shows that in addition to the role in ribosomal protein assembly, DISCUSSION The role of RimM in the assembly of the 30S subunit It is known that disturbance to protein translation might affect the subunit assembly in an indirect way, due to a shortage in the ribosomal protein production. Consequently, the impaired subunit assembly in E. coli strains with genes for assembly factors deleted stems not only from the defective assembly process itself but also from a reduced supply of ribosomal proteins. Never- theless, in this study, the composition of the in vivo inter- Figure 4. RimM preferentially binds to the immature 30S subunits from the (cid:2)rimM strain. Immature 30S subunits from the (cid:2)rimM mediatesfromthe(cid:2)rimMand(cid:2)rbfA(cid:2)rsgAstrainsclearly strainandmature30Ssubunitsfromthe70Sribosomeswereincubated displays a non-uniform level of ribosome proteins, with withor withouta30-fold excess ofRimM. Themixtures werepelleted mostly the 30-domain proteins significantly underrepre- by centrifugation. The pellets (P) and the supernatants (S) were sented (Figure 2), suggesting that the secondary effect separated and resolved by SDS–PAGE. RimM alone was centrifuged causedbyimpairedtranslationinthesestrainsisnegligible as a control. The asterisk denotes the weak binding of RimM to the immature 30S subunits. and does not over-shadow the assembly defects. 2616 NucleicAcidsResearch,2013,Vol.41,No.4 Figure 5. Overview of the five cryo-EM structures of the immature 30Ssubunits treated with RimM. The five density maps (A–E or F–J, respect- ively)aredisplayedintransparentsurfacerepresentation,superimposedwithflexiblyfittedcrystalstructuresincartoonrepresentation.Foreachmap, boththeinter-subunitview(A–E)andthesolventview(F–J)aredisplayed.The16SrRNA,S2,S3andtherest30Ssubunitproteinsarepaintedin blue,green,redandpurple,respectively.Deviationsofthe16SrRNAbackbonesinthefittedmodelfromthatofthemature30Ssubunitarecolored as indicated by the scale to form the temperature maps (K–O). TheroleofRimMuncoveredinthepresentworkinfact from a (cid:2)rsgA strain, was also quantitatively analyzed serves as a perfect illustration to the proposed general (50). The comparison of these quantitative data from dif- function of assembly factors, i.e. to subvert possible ferent genetic background would enable us toidentify the kinetic traps caused by mis-folded rRNA or rate-limiting temporal relationship of assembly factors. binding of certain proteins, during the in vivo assembly First, assembly intermediates isolated from a (cid:2)rsgA process (11,13,49). Previous kinetic data showed that the strain only have a very small subset of tertiary binding binding of 30-domain proteins is not obligatory to the proteins (S21, S2 and S3) largely underrepresented (50), prebinding of S7 (13), while in contrast, prebinding of S7 suggesting that RsgA acts at a very late stage when most andS19togetherdramaticallyacceleratesthebindingofthe of the components are already in place. Second, inter- rest 30-domain proteins (13), indicating the presence of a mediates from the (cid:2)rimM strain show underrepresented rate-limiting S7-independent assembly pathway for S19 levels for all secondary and tertiary binding proteins (13,52). Consistently, our data show that in the (cid:2)rimM from the 30-domain. Interestingly, the (cid:2)rimM intermedi- sample, S19 is among the most underrepresented proteins, atesaretosomeextentsimilarinproteincompositiontoa whereasinthe(cid:2)rbfA(cid:2)rsgAsample,thelevelofS19isdra- previously identified invivo 21Sintermediates (44,53), but maticallyincreased(Figure2).Thus,itislikelythattherole differ from the in vitro RI intermediates (54). The close ofRimMinvivoistocounteractthekinetictrapcausedby resemblance of the (cid:2)rimM intermediates to the naturally slow binding of S19. In support of this view, both RimM populated in vivo 21S intermediates suggests that they and S19 bind to a multi-helices junction of the 16S represent an early stage during the assembly of the 30-domain (Figure 6), highlighting their potential effect on 30-domain, likely the entry stage. Last, in contrast, the the global stabilization of the 30-domain. intermediates from the (cid:2)rbfA(cid:2)rsgA strain, with only a subset of 30-domain proteins largely underrepresented, Functional interplay of assembly factors do not resemble any of the known intermediates, In addition to the two sets of intermediates described in indicating that they represent a novel set of intermediates this study, another set of in vivo intermediates, isolated roughly in-between. NucleicAcidsResearch,2013,Vol.41,No.4 2617 Figure 6. MechanisticmodeloftheRimMfunctionintheinvivoassemblyofthe30-domain.(A)Theheaddomainofthe30Ssubunitviewedfrom theinter-subunitside.(B)Sameas(A),witha70(cid:2) rotationaroundY-axis.Theh33b,h31andtherestofthe16SrRNAarepaintedincyan,wheat andblue,respectively.TheC-terminaldomainofRimM(CTD),N-terminaldomainofRimM(NTD),S7,S13,S19andS14arepaintedinmagenta, orange, red, yellow, green and purple, respectively. (C) RimM, RbfA and RsgA act at different checkpoints during the in vivo assembly. The deficiencyofassemblyfactorsdivertstheassemblyintolessefficientbranches(coloreddashlines)andcausesaccumulationofasetofcloselyrelated intermediates(coloredboxes).Theribosomalproteinlevelsinthethreesetsofinvivointermediatesaredisplayedinthegrayscale.Thedataofthe intermediates from a(cid:2)rsgA strain is froma previous study (50).The large conformational differences amongthe threesets of invivo intermediate werealsoshownincartoon:the(cid:2)rimMone(red)withadramaticallyrotatedheaddomainandadisorderedhelix44;the(cid:2)rbfA(cid:2)rsgAone(green) with a disordered helix 44 only; the (cid:2)rsgA one (blue) with a well-resolved helix 44. Therefore,theproteinspectraoftheabovethreesetsof ‘dead-end’ products that are otherwise elusive in the intermediates clearly suggest an order for the actions of normal condition. these factors (Figure 6C), which is consistent with The in vivo assembly of the 30S 30-domain also follows previous genetic and biochemical data (19,25,55). It parallel pathways must be noted that it is difficult to unambiguously timestamp other biogenesis factors in the assembly Early chemical probing of the 16S rRNA conformation pathway due to the lack of biochemical data, although (56), as well as recent kinetic measurement of the protein genetic data have suggested both functional redundancy binding (9,11), showed that the 30-domain assembly is the andhierarchyforsomeassemblyfactors[reviewedin(12)]. latesteventduringtheinvitroassemblyofthe30Ssubunit, Nevertheless, if we view the in vivo assembly as a coincident with the 50- to 30-transcription order. On the multi-branched process, the seemingly function redun- otherhand,accumulatingevidences(9–11,56)suggestthat dancy among assembly factors is merely a sign of altered amajorgeneralfeatureoftheinvitroassemblyofthe30S contribution of different, inter-connected assembly subunitisthattheprocessproceedsalongmultipleroutes. pathways. As shown in Figure 6C, the late-stage Inthisstudy,weisolatedtheinvivoassemblyintermedi- assembly in vivo starts with an in vivo 21S intermediate ates from two genetically modified E. coli strains. The (44,53) and proceeds along a highly efficient pathway in most remarkable feature in the protein spectra of these the presence of all assembly factors. Assembly factors two sets of intermediates is that they both severely lack come in play at different time points to assist certain kin- 30-domain proteins, suggesting that the maturation of the etically disfavored assembly events. The disruption of a 30-domain is also a rate-limiting process in vivo. These factor or a combination of factors would avert the quantitative data also indicate that the intermediates assemblytolessefficientbranchesandcauseaccumulation from both strains are very heterogeneous in ribosomal of a certain category of kinetically trapped intermediates. protein composition, which means they do not represent Consistent with this view, most of the E. coli genes for asingle populated assembly intermediate state, butrather assembly factors are not essential. The remaining a collection of multiple-related intermediates with more questioniswhetherthesekineticallytrappedintermediates than one metastable state enriched. These differently from various genetic background with different factors prepared intermediates, although with a recognizable disrupted truly represent genuine snapshots of the temporal relationship, cannot be easily reconciled by a assembly process in the normal condition, or different single continuous assembly pathway. 2618 NucleicAcidsResearch,2013,Vol.41,No.4 With respect to the occupancies of individual proteins, the integration of structural (26,50) and QMS data [the there are a number of exceptions to the well-accepted present work and (50)], we could draw a conclusion that Nomura assembly map (5). To name a few, first, an un- the30-domainofthe16SrRNAmaturatesinaprogressive expectedobservationofourQMSdataisthattwoprimary manner in vivo, paralleling with the ribosomal protein proteins (S4 and S7) in the (cid:2)rbfA(cid:2)rsgA sample show assembly. decreased levels compared with the (cid:2)rimM sample. S4 is First, assembly intermediates from (cid:2)rimM cells show thermodynamically required for subsequent binding of dramaticconformationaldifferencesinthepositionofthe S16, S12 and S5 to the 50-domain (5), and more import- 30-domain (Figure 3 and Supplementary Figure S2). In antly, S4 was shown to have a global stabilization effect contrast, cryo-EM structures of intermediates from onthe50-domain(57).S7istheonlyprimaryproteininthe (cid:2)rsgA cells (50) also vary in the 30-domain position, but 30-head domain and directs the following binding of S9, with a much smaller scale. Second, the long helix 44 of S13andS19(5,48).However,inthe(cid:2)rbfA(cid:2)rsgAsample, the 30-minor domain is highly flexible and is almost invis- the occupancy of S4 and S7 (Figure 2) is even lower than ible in the structures of intermediates from (cid:2)rimM cells its follower proteins. Similarly, the level of S7 was also (Figures 3 and 5). However, cryo-EM structures of 30S reported to be lower than its follower S9, S13 and S19 intermediates from (cid:2)rsgA cells (50), which also contain in the (cid:2)rsgA sample (50). Second, levels of S10 and S14 17S rRNA, display well-resolved densities for helix 44, arelowerthantheirfollowertertiaryproteinsS2andS3in except for the upper decoding center region. This the (cid:2)rimM sample, suggesting that S3 could bind inde- suggests that helix 44 adapts its mature conformation pendent of S10. Third, the level of S2 is lower than S3 in only at a very late stage. In fact, hydroxyl radical the (cid:2)rbfA(cid:2)rsgA sample, although S3 binding is thermo- probing data (10,59) already showed that the full accom- dynamically dependent on the prior binding of S2 in the modation of helix 44 is a late event even when the experi- Nomura map. mentswereperformed withthe16SrRNA.Third, further The Nomura map was derived by single protein downstream is the cryo-EM structure of the 30S–RsgA omission reconstitution experiments with fully processed complex,whichdisplaysanalmostidenticalconformation 16S rRNA under equilibrium conditions and therefore to the mature 30S subunit (26). does not necessarily reflect the true order of serials of Based on the above structural comparisons, the matur- binding events during assembly (52). In addition to our ation of the 30-domain of the 16S rRNA follows the tran- QMSdata,deviationsfromtheNomuramaphavealready scription order in a progressive manner, first the 30-head been observed from both in vivo and in vitro studies. domain, next the 30-minor domain (Figure 6C), and more Previous genetic data show that S15, a primary protein importantly, the conformational maturation is coupled in the central domain, is dispensable for the 30S with the gradually increased protein level (Figure 6C). assembly in vivo (58). Furthermore, in vitro kinetic data Therefore, our data have revealed another common char- from Williamson group based on PC-QMS (11,13) or acteristic shared by the in vivo and the in vitro processes, fluorescence triple correlation spectroscopy (52) indicate i.e. a high cooperativity between protein binding and that S9 and S19 could bind independent of S7 (13,52), rRNA folding (10,11,60,61). and S2 could bind independent of S3 (11). All these data suggest that there are hidden assembly pathways that ACCESSION NUMBERS could not be directly inferred from the Nomura map. Thus,theseeminglydiscrepancybetweenourQMSdata The cryo-EM maps have been deposited in the andtheNomuramapcouldbeeasilyreconciledifweview EMDataBank (EMDB codes 5500-5504 and 5506-5510 theinvivoassemblyprocessasahighlybranchednetwork. for the immature 30S subunits with or without the In the presence of not fully processed 17S rRNA and the RimM treatment, respectively). The atomic models of absence of certain assembly factors, the 30S assembly the 16S rRNA have been deposited in the Protein Data in vivo takes alternative, kinetically inefficient, pathways Bank (PDB codes 3J28, 3J29, 3J2A, 3J2B, 3J2C, 3J2D, thatarenotpredictedbytheNomuramap.Therefore,the 3J2E, 3J2F, 3J2G, 3J2H). assembly intermediates isolated from these different assembly factor-deficient mutants might represent inter- mediates kinetically trapped in various parallel branches SUPPLEMENTARY DATA in the assembly network. Supplementary Data are available at NAR Online: Insummary,althoughthenumberofpossibleassembly Supplementary Materials and Methods, Supplementary pathwaysinvivoislimitedbytheco-transcriptionalnature Tables 1 and 2, Supplementary Figures 1–7 and and the presence of assembly factors in the in vivo condi- Supplementary References [26,32,44]. tion, our data provide additional strong evidence to the emerging idea that the in vivo assembly also proceeds along parallel pathways in a certain degree (54,58). ACKNOWLEDGEMENTS We thank Nara Institute of Science and Technology, The maturation of the 16S rRNA 30-domain in vivo is National BioResource Project and National Institute of highly coupled with protein assembly Genetics for providing the (cid:2)rimM strain from the Keio Our structural data reveal that the in vivo assembly inter- collection and Dr Krisana Asano for construction of an mediates differ largely in rRNA conformation. Through overproducing system of RimM. We also thank the

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.