Eukaryotic rpL10 drives ribosomal rotationSergey O Sulima1 Suna P Gulay1 Margarida Anjos2 Stephanie Patchett2

Arturas Meskauskas13 Arlen W Johnson2 and Jonathan D Dinman1

1Department of Cell Biology and Molecular Genetics University of Maryland College Park MD 20742 USA2Section of Molecular Genetics and Microbiology Institute for Cellular and Molecular Biology University ofTexas at Austin Austin TX 78712 USA and 3Department of Biotechnology and Microbiology VilniusUniversity Vilnius LT-03101 Lithuania

Received April 16 2013 Revised October 18 2013 Accepted October 21 2013

ABSTRACT

Ribosomes transit between two conformationalstates non-rotated and rotated through the elong-ation cycle Here we present evidence that aninternal loop in the essential yeast ribosomalprotein rpL10 is a central controller of thisprocess Mutations in this loop promote opposingeffects on the natural equilibrium between thesetwo extreme conformational states rRNA chemicalmodification analyses reveals allosteric interactionsinvolved in coordinating intersubunit rotationoriginating from rpL10 in the core of the largesubunit (LSU) through both subunits linking all thefunctional centers of the ribosome Mutationspromoting rotational disequilibria showed catalyticbiochemical and translational fidelity defects AnrpL3 mutation promoting opposing structural andbiochemical effects suppressed an rpL10 mutantre-establishing rotational equilibrium The rpL10loop is also involved in Sdo1p recruitment suggest-ing that rotational status is important for ensuringlate-stage maturation of the LSU supporting amodel in which pre-60S subunits undergo a lsquotestdriversquo before final maturation

INTRODUCTION

The ribosome is an essential and complex nanomachinethat provides a model for understanding principles ofmacromolecular assembly and functional coordinationThe eukaryotic yeast ribosome is a 36-MDa RNAndashprotein complex consisting of 79 intrinsic ribosomalproteins and 4 ribosomal RNAs (rRNAs) (1) Differentbiochemical functions are spatially separated from oneanother in the two subunits of the ribosome The smallsubunit (SSU) contains the mRNA decoding centerwhereas the large subunit (LSU) harbors separate regionswith distinct functions the peptidyltransferase center (PTC)

(responsible for catalysis) the peptide exit tunnel threetransfer RNA (tRNA) binding pockets and a singlebinding site that must distinguish between the two elong-ation factors and release factor in response to specific cir-cumstances The subunits must also interact with oneanother as a holoenzyme to coordinate a complex seriesof events particularly allosteric movements that occurthroughout the course of the elongation cycle The twoextreme conformational states are termed lsquorotatedrsquo (alsoknown as ratcheted or hybrid) and lsquonon-rotatedrsquo (alsoknown as classical) How information is exchanged overlong distances between spatially distinct functionalcenters and how these centers then work in concert toensure timely rotation proper ligand binding and ultim-ately unidirectional and faithful translation remains largelyunclear In addition while the roles of the SSU (particu-larly the head) and tRNAs in ribosome rotation have beeninvestigated the question of whether the LSU is an activeor passive participant in this process has not been exploredRibosomal protein L10 [rpL10 aka L16 (1)] plays es-

sential roles in ribosome biogenesis and translationalfidelity Incorporation of rpL10 into the LSU in the cyto-plasm (2) constitutes a late step of LSU maturation rpL10works in conjunction with Shwachman-Diamond proteinSdo1p and the eukaryotic elongation factor 2 (eEF2)-likeGTPase Efl1p to promote the release of the anti-associ-ation factor Tif6p (3ndash5) and the nuclear export adapterNmd3p (6) Thus LSUs lacking rpL10 are unable tojoin with the SSU (5) rpL10 is located near the corridorthrough which aminoacyl-tRNAs (aa-tRNAs) moveduring the process of accommodation (Figure 1a) and isinvolved in tRNA movement through this structure (7) Itis also located near several other functional centers of theLSU including the PTC the A-site finger (H38) theelongation factor binding site and the GTPase associatedcenter (Figure 1b) The C-terminus of rpL10 contacts 5SrRNA which interacts with rpL5 and rpL11 at the head ofthe central protuberance (8) Thus rpL10 is well pos-itioned to act as a sensor of activity near the PTC andtransduce that information to other functional centers tocoordinate ribosome function

To whom correspondence should be addressed Tel +1 301 405 0918 Fax +1 301 314 9469 Email dinmanumdedu

Published online 8 November 2013 Nucleic Acids Research 2014 Vol 42 No 3 2049ndash2063doi101093nargkt1107

The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby30) whichpermits unrestricted reuse distribution and reproduction in any medium provided the original work is properly cited

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

An essential internal loop of rpL10 (aa 102ndash112 previ-ously called the lsquoP-site looprsquo) that makes the closestapproach to the PTC (13 A) of any ribosomal protein iscrucial for Tif6p and Nmd3p release (59) Mutagenesis ofthe loop revealed two classes of mutants based on theireffects on ribosome biogenesis (5) This study employs thestrongest representative mutants of these two classesS104D (improper subunit joining and 60S biogenesisdefect) and A106R (high 60S40S ratio) Here wepresent evidence that these mutations promote opposingeffects on the natural equilibrium between the two extremeconformational states of the ribosome Structural and bio-chemical analyses of vacant ribosomes demonstrate thatthe S104D mutation drives the ribosome toward therotated state favoring binding of eEF2 and disfavoringbinding of eEF1Aaa-tRNAGTP (elongation ternarycomplex) and inhibiting peptidyltransfer In contrast

A106R causes vacant ribosomes to distribute toward thenon-rotated state favoring elongation ternary complexbinding over eEF2 and leaving peptidyltransferaseactivity unaffected We also show that the non-rotatedconformation favors Sdo1p binding stimulating bindingof tRNAs to the A-site Furthermore Sdo1p competeswith acetylated-aa-tRNA for binding to the P-site andinhibits peptidyltransfer suggesting that Sdo1p stabilizesthe non-rotated state through binding the P-site Largescale rRNA chemical modification analyses revealdistinct information transmission pathways originatingfrom rpL10 in the heart of the LSU and emanatingthroughout the LSU and the SSU

These observations lead us to propose that the rpL10loop plays a central role in much of the ribosomal lifecycleby helping to set the conformational status of the LSUcoordinate intersubunit rotation and communicate this in-formation to the decoding center on the SSU During lateLSU biogenesis the loop senses Sdo1p recruitment to theP-site initiating a lsquotest driversquo to ensure the functionality ofpre-60S subunits After ensuring proper 80S assembly theloop monitors the tRNA occupancy status of the PTCA-site in the absence of A-site ligand (aa-tRNA) it cansample this space while the introduction of ligand dis-places it We suggest that the positioning of the rpL10loop determines which state the LSU assumes in aprocess involving cascades of allosteric interactions thatlink functional centers in the LSU with those in theSSU The downstream effects of rotational disequilibriumare wide-ranging impacting translational reading framemaintenance the ability to discriminate between cognateand near- and non-cognate codons and termination codonrecognition Mutants of ribosomal protein L3 that conferopposing effects on ribosome structure and function cansuppress the structural biochemical and functional defectsof the rpL10 loop mutants by re-establishing the normalrotational equilibrium In sum we propose that the rpL10loop is a master controller of ribosome structure andfunction influencing critical steps in both ribosomeassembly and biogenesis and the protein-synthetic phaseof elongation We suggest that the unidirectionality oftranslation is aided by this intrinsic feature of theribosome and that the LSU alone has the ability to inde-pendently influence intersubunit rotation

MATERIALS AND METHODS

Strains plasmids and genetic manipulation

Saccharomyces cerevisiae strain AJY1437 containing wild-type RPL10 on a centromeric URA3 vector (pAJ392) haspreviously been described (5) In AJY3222 the wild-typevector was replaced by wild-type RPL10 on a centromericLEU2 vector (pAJ2522) through standard 5-FOAshuffling techniques AJY3209 harbors pAJ2609 a centro-meric LEU2 vector expressing the rpl10-S104D alleleSimilarly AJY3212 contains pAJ2612 which expressesthe rpl10-A106R mutant from a centromeric LEU2vector Generation of the rpl10-F94I and rpl10-G81Dexpressing strains JD1308F94I and JD1308G81D waspreviously described (7) AJY2104 (9) was transformed

tRNAin

PTC

tRNAout

rpL10

rpL10 Loop

5S rRNA

H38

H39

PTC H89 side ofaccommodationcorridor

Factorbinding

site

104 106

(a)

(b)

Figure 1 rpL10 is strategically positioned in the core of the LSU(a) The big picture rpL10 in the context of the subunit interface ofthe LSU (b) A close-up view of rpL10 and the local environment Thehypothetical loop structure is circled and indicated by dashed red linesand the approximate positions of S104 and A106 are indicated Theprotein is situated between Helices 38 and 89 and appears to be anextension of Helix 39 It is located in close proximity to several func-tional centers of the LSU including the PTC the aa-tRNA AC and theelongation factor binding site It is also positioned to communicatewith the SSU through Helix 38 and the 5S rRNA Images weregenerated using PyMOL

2050 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

with pAJ2522 or pAJ2609 and RPL3 or rpl3-W255C ex-pressed from 2 mm TRP1 vectors Standard molecularbiology techniques were used to subclone RPL3 andrpl3-W255C into the HIS3 selectable pRS423 AJY3209was subsequently transformed with pRS423-RPL3 andpRS423-rpl3-W255C

Translational fidelity and polysome analyses

The dual luciferase reporter plasmids pYDL-controlpYDL-LA pYDL-Ty1 pYDL-UAA (10) pYDL-AGC218 and pYDL-TCT218 (11) were employed tomonitor programmed 1 ribosomal frameshiftingprogrammed+1 ribosomal frameshifting suppression ofUAA and suppression of an AGC near-cognate serinecodon and a TCT non-cognate serine codon in place ofthe cognate AGA codon in the firefly luciferase catalyticsite respectively The reporters were expressed from high-copy URA3-based plasmids (pJD375 pJD376 pJD376pJD431 pJD642 and pJD643) Assays were performedas previously described (12) Sample readings were col-lected using a GloMax Multi-Microplate luminometer(Promega) All assays were repeated four times Sucrosedensity gradient analysis was carried out as described (5)

Ribosome preparation

Purification of active 80S ribosomes using cysteine-charged sulfolink columns was performed as described(13) with the following modifications after elution fromthe column ribosomes were treated with 1mM finalconcentration GTP and 1mM final concentration pH-neutralized puromycin at 30C for 30min to removeendogenous tRNAs After a 100 000 g 16ndash20 h spinthrough a high salt glycerol cushion [20mM HEPES-KOH pH 76 60mM NH4Cl 500mM KCl 10mMMg(OAc)2 2mM DTT 25 glycerol] ribosomes wereresuspended in elution buffer and passed through a lowsalt cushion [20mM HEPES-KOH pH 76 50mMNH4Cl 5mM Mg(OAc)2 1mM DTT 25 glycerol] toincrease purity

RibosometRNA interactions

eEF1A preparation purification of aa-tRNA synthetasescharging of tRNAPhe with [14C]-phenylalanine and purifi-cation of aa-tRNA and acetylated aa-tRNA were carriedout as described (1415) To assay steady-state dissociationrates (KD) of aa-tRNA to the ribosomal A-site two sets ofreactions were set up in parallel A mix containing 100 mgof polyU 50 pmol of ribosomes a 4-fold molar excess oftRNAPhe all in binding buffer [80mM TrisndashHCl pH 74at 30C 160mM NH4Cl 15mM Mg(CH3COOH)2 2mMspermidine 05mM spermine 6mM b-mercaptoethanol]in 150 ml total volume was prepared and incubatedfor 30min at 30C to block the P-site To prepare theternary complex ([14C]Phe-tRNAPhe

eEF1AGTP)100 mg of soluble protein factors 1mM final concentrationof GTP 125 pmol [14C]-Phe-tRNAPhe were mixed in 50 mltotal volume binding buffer and incubated for 30min at30C After incubation serial 2-fold dilutions of theternary complex reaction mix were prepared resulting ineight fractions containing decreasing amounts of ternary

complex (625ndash1 pmol) in 105 ml each An equal amountof the ribosome mix (5 pmol of ribosomes 15ml) wasadded to each dilution followed by incubation for30min at 30C The mixtures were applied onto pre-wetted nitrocellulose Millipore HA (045 mm) filterswashed with binding buffer and radioactivity wasmeasured via scintillation counting Background controlreactions without ribosomes were performed at eachligand dilution and subtracted from experimental onesBinding data were analyzed via single binding site withligand depletion models using GraphPad Prism Non-enzymatic binding studies were performed likewise butwithout eEF1A or by using 50 mg of soluble proteinfactors (half maximum activity) To test binding oftRNA to the P-site Ac-[14C]-Phe-tRNAPhe was used asthe ligand there was no need to pre-block with tRNAPhe

and 11mM Mg(CH3COOH)2 was used in the bindingbuffer All reactions were repeated four times

Peptidyltransferase activity

Single turnover peptidylpuromycin reactions were per-formed to assay apparent rates of peptidyltransfer asdescribed (14) with the following modificationRibosomes pre-bound with Ac-[14C]Phe-tRNAPhe andpolyU were loaded onto pre-wetted Millipore HA(045 mm) filters and washed with binding buffer Thefilters were placed into 15ml scintillation vials and 24mlof binding buffer+005 Zwittergent (EMD BioScience)followed by a 30min incubation on a rocker at 4CAliquots (1ml) were taken from each vial incubated for5min at 30C to activate ribosomes reactions wereinitiated by adding pH-neutralized puromycin to 10mMfinal concentrations and the procedure completed asdescribed (14) Reactions were repeated four times

Sdo1p cloning purification and labeling

Sdo1p was cloned into a modified pET-21 a expressingSdo1p with a C-terminal 6 histidine tag and thephosphorylatable kemptide The protein was expressedin Codon Plus bacteria (Stratagene) and purified by Ni-NTA (Invitrogen) chromatography followed by gel filtra-tion on sephacryl S200 (GE Healthcare) Labelingreactions containing 10 mg of Sdo1p 5 molar excess of[32P]-g-ATP 1 ml PKA (NEB) in 100ml 1 kinase buffer(NEB) were incubated for 15min at 30C passed throughG25 columns (GE) to remove unincorporated ATP andflow-through was measured by scintillation countingFlow-through values from control mixtures lacking PKAwere subtracted from values of the labeling reactions todetermine the specific activity of 32P-labeled Sdo1pyielding 40ndash60 label incorporation

Ribosomeprotein interactions

6-His-tagged eEF2 was purified from TKY675 yeast cells(a generous gift from Dr T Kinzy) as described (16)Aliquots containing ribosomes (2 pmol) were first pre-incubated with increasing concentrations of eEF2 (025ndash32pmol) 100mg of polyU 01mM final concentrationGDPNP and 4 molar excess of [14C] NAD over ribo-somes in 50ml total volume binding buffer at 30C for

Nucleic Acids Research 2014 Vol 42 No 3 2051

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

20min Diphtheria toxin (02mg) was added and reactionswere incubated for 30min at 30C After precipitation withTCA (final concentration 15) and 15min incubation onice reaction mixtures were applied onto GFC filterswashed with 5 TCA and the amount of [14C]-ADPribosylated eEF2 was determined by scintillationcounting Readings reflect unbound eEF2 and were sub-tracted from total eEF2 to obtain values bound Sdo1pbinding assays were performed by incubating ribosomes(2 pmol) with increasing concentrations of Sdo1p (025ndash32pmol) and 100mg of polyU in binding buffer [50mMTris-(OAc)2 pH 75 RT 50mM NH4(OAc)2 10mMMg(OAc)2 2mM DTT] in 50ml total volume at 30C for20min The reaction mixtures were then applied onto pre-equilibrated 1ml polyethylene filter spin columns (Pierce)containing 03ml cysteine-charged sulfolink resin andincubated on ice for 5min The columns were spun andflow-through measured via scintillation countingReadings reflect unbound Sdo1p and were subtractedfrom total loaded to obtain Sdo1p bound Backgroundcontrol reactions without ribosomes were performed ateach ligand dilution eEF2 assays were repeated fourtimes Sdo1p assays were performed in triplicate

Ribosome binding competition

To monitor effects of Sdo1p on tRNA binding to theA-site wild-type ribosomes (270 pmol) primed withpolyU (035mg) were first mixed with a 4-fold molarexcess of uncharged tRNAPhe 28 nmol GTP and 5mgof soluble protein factors (including eEF1A) in 700 mltotal volume of ribosome binding buffer After 10min in-cubation at 30C ribosometRNA complexes (50 mlaliquots) were added to 10 ml aliquots of 2-fold dilutionsof purified Sdo1p (500ndash31 pmol plus a no-Sdo1p control)and incubated at 30C for 10min Subsequently 25 pmolof [14C]Phe-tRNA was added to each ribosomeSdo1pcomplex and incubated at 30C for 10min The reactionmixtures were applied onto pre-wetted nitrocelluloseMillipore HA (045 mm) filters washed with bindingbuffer and radioactivity was measured via scintillationcounting Background control reaction values fromsamples without ribosomes were subtracted from experi-mental ones To monitor the ability of Sdo1p to competefor binding at the P-site the same conditions wereemployed except that binding reactions were performedin 11mM magnesium and [14C]Ac-Phe-tRNAPhe wasused Assays of peptidyltransferase activity were per-formed as described above in the presence of Sdo1p Allassays were performed twice in triplicate

Preparation of complexes for hSHAPE probing

Fifty pmoles 80S ribosomes isolated from isogenic strainswere incubated with 100 mg polyuridylic acid (polyU) inbinding buffer [80mM HEPES pH 75 50mM NaCl11mM Mg(OAc)2 6mM b-mercaptoethanol] at 30Cfor 10min One set of ribosomes so prepared (lsquovacant80Srsquo wild type rpL10-S104D rpL10-A106R) wasemployed for chemical protection assays To preparecontrol lsquonon-rotatedrsquo ribosomes 200 pmol N-acetyl-phenylalanyl tRNAPhe were added and incubations

continued for 20min To prepare lsquorotatedrsquo wild-type ribo-somes vacant 80S ribosomes were first incubated with200 pmol of deacylated tRNAPhe eEF2 (400 pmol) andGDPNP (final concentration of 1mM Sigma) were thenadded to the deacylated tRNAPhemdashribosome mixture andincubated for an additional 20min These conditions werebased on results of binding assays

rRNA structure probing

hSHAPE (17) of rRNA with 1M7 was performed asdescribed (18) using the following substrates vacant ribo-somes isolated from isogenic wild-type cells cells express-ing rpl10-S104D rpl10-A106R wild-type ribosomescontaining acetylated-aa-PhetRNAPhe in the P-site (non-rotated controls) and deacylated-tRNAPhe + eEF2-GDPNP (rotated control) The following primers wereemployed 969 and 1780 in the SSU 25-2 1466 26322836 25-7 and 3225 in the LSU Data were analyzedusing SHAPEFinder (19) For kethoxal studies 25 pmolof ribosomes in a 50 -ml volume were treated with 1 ml ofa 4 kethoxal solution (in pure ethanol) or 1 ml ofethanol as control and incubated for 10min at 30CReactions were stopped by addition of one halfvolume of stop solution (150mM sodium acetate250mM potassium borate) followed by analysis asabove using primer 969

Statistical analyses

Studentrsquos t-test for P-value calculations was used through-out Data analysis of dual-luciferase assays and rRNAprobing was carried out as described (1218) After gener-ation of rRNA chemical modification data usingShapeFinder (19) the median reactivities of each primerregion were used to normalize the raw integrated peakvalues

RESULTS

The S104D and A106R mutants promote opposing effectson ligand binding to the ribosomal A- and P-sites

Saturation mutagenesis of the rpL10 loop revealed twoclasses of mutants based on their ribosome biogenesisdefects Class I mutants typified by rpL10-S104D weredefective for subunit joining and displayed halfmer poly-somes and Class II mutants exemplified by rpL10-A106Rexhibited higher 60S40S subunit ratios (5) While therpL10 loop was not resolved by X-ray crystallography(8) cryo-EM studies suggested that the tip of this loopis in close proximity to the P-site tRNA (2021) Thus itwas speculated that P-site ligand binding would beaffected by these mutants However steady state bindingof acetylated-[14C]Phe-tRNAPhe to 80S ribosomes purifiedfrom isogenic strains expressing wild type and mutantforms of rpL10 showed no significant differences inbinding of this ligand to the ribosomal P-site (Figure 2aSupplementary Figure S1a) In contrast the mutantsshowed significant changes in their ability to bind elong-ation ternary complex ([14C]Phe-tRNAPhe

eEF1AGTP)to the ribosomal A-site specifically the rpL10-S104D

2052 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

mutant promoted a 2-fold increase in KD for this ligandwhile the rpL10-A106R mutant promoted a 2-folddecrease (Figure 2b Supplementary Figure S1b) To as-certain whether these differences were due to changes inaffinity for the elongation factor or the aa-tRNA itself thesame experiment was performed using [14C]Phe-tRNAPhe

alone (ie non-enzymatic binding) Under these condi-tions the aa-tRNA binding defect of the rpL10-S104Dmutant was exacerbated (6-fold increased KD relativeto wild-type ribosomes) and this was partiallyameliorated by addition of eEF1A (Figure 2bSupplementary Figure S1d) In contrast the binding of[14C]Phe-tRNAPhe to wild type or rpL10-A106R ribo-somes was not influenced by eEF1A These data indicatethat the tRNA binding defects are intrinsic to the riboso-mal A-site and not to binding sites unique to eEF1A

eEF2 which drives translocation is a structural mimicof the elongation ternary complex and includes a tRNA-like domain (Domain IV) that also interacts with the ribo-somal A-site of the decoding center (22) Steady-statebinding assays of purified eEF2 to 80S ribosomes asmonitored by the extent of diphtheria toxin [14C]-ADPribosylation also revealed reciprocal changes in KD

values for the two mutants (Figure 2a SupplementaryFigure S1c) Importantly while the S104D mutant ribo-somes exhibited decreased affinity for aa-tRNA andelongation ternary complex they displayed increasedaffinity for eEF2 Conversely A106R mutant ribosomes

displayed increased affinity for aa-tRNA and elongationternary complex and decreased affinity for eEF2Sdo1p is required at a late step in 60S maturation

coupling the GTPase activity of the Efl1p to release ofTif6p (4) Steady state binding assays using purified[32P]-labeled Sdo1p revealed that mutant ribosomes dis-played defects in binding this ligand similar to thoseobserved with elongation ternary complex ie the rpL10-S104Dmutant promoted decreased affinity for Sdo1p whilethe rpL10-A106R mutant promoted increased affinity for it(Figure 2c Supplementary Figure S1e) Sdo1p did not bindmRNA or tRNA alone and the Sdo1pN mutant lackingthe N-terminal FYSH domain that is essential for proteinfunction (23) displayed negligible binding to wild-type ribo-somes (Supplementary Figure S1e) Scatchard plot analysesdemonstrated that Sdo1p binds to a single site on the ribo-somes (Supplementary Figure S1f) Competition assays inwhich wild-type ribosomes were pre-incubated withincreasing amounts of Sdo1p revealed that this proteinstimulated binding of aa-tRNA to ribosomes in a concen-tration dependent manner while inhibiting both binding ofAc-aa-tRNA to the P-site and peptidyltransferase activity(Figure 2d)

The A106R and S104D mutants have opposing effects onthe rotational equilibrium of the ribosome

During the elongation cycle the ribosome transitsthrough a large number of conformations characterized

0

10

20

30

40

50

WT

S10

4DA

106R

WT

S10

4DA

106R

WT

S10

4DA

106R

P-sitetRNA

A-sitetRNA

eEF2

K (

nM

)

0

40

80

120

WT S104D A106ReEF1A 0 1 0 1 0 1 frac12

K (

Sd

o1p

nM

)

K (

Ph

e-tR

NA

n

M)

Ph

e

0

40

80

120

160

WT S104D A106R0

50

100

150

200

0 5 10 15 20

o

f tR

NA

Bo

un

d o

r

Pu

rom

ycin

rea

cted

Sdo1p80S

AcPhe-tRNA

Puromycin

Phe-tRNA

DD

D

(a) (b)

(c) (d)

Figure 2 rpL10 loop mutants inversely affect ligand binding to the A- and P-sites (andashc) Steady state binding of indicated ligands for ribosomes (a)Dissociation constants obtained from assays of binding of Ac-aa-tRNA to the P-site aa-tRNA to the A-site and eEF2 to ribosomes isolated fromcells expressing wild-type rpL10 and the rpL10-S104D and rpL10-A106R mutants (b) Titration of eEF1A into aa-tRNA binding reactions similar to(a) (c) Dissociation constants obtained from assays of binding of Sdo1p to indicated ribosomes (d) Competition assays Binding of aa-Phe-tRNAPhe

or Ac-Phe-tRNAPhe at half maximal concentrations was monitored in the presence of increasing amounts of Sdo1p Peptidyltransferase activity wassimilarly monitored Bars indicate SEM (n=4 for a b n=3 for c) Plt 005 Plt 001 (compared to wild type unless noted)

Nucleic Acids Research 2014 Vol 42 No 3 2053

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

at the extremes by states described as non-rotated androtated (824ndash28) Recent single molecule experimentsusing Escherichia coli ribosomes have shown that theaa-tRNAEF-TuGTP elongation ternary complex hashigher affinity for non-rotated ribosomes than rotatedribosomes and that the converse is true for EF-G (29)Thus the ligand binding data described above suggestedthat the A106R and S104D mutants drive the structuralequlibria of ribosomes toward either the non-rotated(substrate for binding elongation ternary complex) orrotated (substrate for eEF2) states respectively Whilechemical modification profiles and atomic resolutionstructures are well-defined for non-rotated and rotatedE coli ribosomes (2426ndash2830ndash34) no equivalent infor-mation exists regarding yeast ribosomes Thus toexamine the rotational status of yeast ribosomes it wasfirst necessary to demonstrate that chemical protectionpatterns of well-defined E coli and yeast ribosomecomplexes are similar For E coli the standard fornon-rotated ribosomes are those which are primed withpolyU and contain N-Ac-PhetRNAPhe in the P-sitewhile rotated ribosomes are primed with polyU andcontain deacylated tRNA in the PE site and EF-G-GDPNP (30) Similar complexes were prepared usingyeast ribosomes (rotated yeast ribosomes containedeEF-2-GDPNP instead of EF-G-GDPNP) These twocomplexes plus vacant wild-type control ribosomeswere chemically probed with 1M7 (or DMSO onlycontrols) and base reactivities were assessed byhSHAPE and quantified using ShapeFinder (1935)Representative complete electropherograms for thesethree samples are shown in Supplementary Figure S2Supplementary Table S1 shows that the chemical modi-fication profiles of landmark E coli rRNA bases are veryclosely matched by their yeast counterparts in the non-rotated and rotated states (compare columns 3 and 4with columns 6 and 7) Columns 8ndash10 ofSupplementary Table S1 also depict conversion of thestatistically normalized reactivity data (19) to a scalefrom 0 to 4 with 4 being most reactive (1819)Occupation of the P-site by Ac-aa-tRNA (non-rotatedribosomes) resulted in nearly identical chemical modifi-cation patterns at equivalent rRNA bases in the P-sitesof the SSU and LSU and in the LSU E-sites (indicatedby color coded boxes in Supplementary Table S1)Similarly the chemical modification patterns in theLSU P- and E-sites of rotated yeast ribosomes closelymatched published data for rotated E coli ribosomesIn addition the non-rotated control was verified bio-chemically by comparing the ternary complex and eEF2steady state binding profiles (Supplementary Figure S3)Note that the rotated control could not be biochemicallyassessed as the factor binding site was already occupiedby eEF2-GDPNP Collectively these data demonstratethat these two complexes represent non-rotated androtated yeast ribosomesHaving established standards defining the chemical re-

activity profiles of non-rotated and rotated yeast ribo-somes these were used to determine the rotationalstatuses of mutant ribosomes To this end the chemicalprotection profiles of lsquolandmarkrsquo basepair interactions

located in several universally conserved intersubunitbridges (82737ndash39) were examined using vacant ribo-somes Although FRET experiments revealed thatvacant bacterial ribosomes are not as structurallydynamic as pre-translocation ribosomes (ie containingdeacylated tRNA in the P site) (36) we nonetheless choseto make comparisons using vacant ribosomes for tworeasons (i) to monitor the intrinsic influence of theL10 internal loop on ribosome conformational statesie in the absence of trans-acting factors (eg tRNAs)and (ii) because ribosomes harboring deacylated tRNAsin the P-sites alone do not occur in physiological condi-tions The B7a intersubunit bridge undergoes dramaticrearrangements during ribosome rotation (37)Specifically when the ribosome is in the non-rotatedstate E coli A702 (yeast G913) of the SSU rRNA inter-acts with E coli A1847 (yeast A2207) of the LSU rRNAprotecting both bases from chemical attack This inter-action is disrupted when the ribosome assumes therotated state rendering both bases susceptible tochemical modification (Figure 3a SupplementaryFigure S4) (37) Kethoxal was used to probe G913 and1M7 was used to probe A2207 and the extent to whichthese bases were modified in purified vacant isogenic wildtype and mutant ribosomes and with control rotated andnon-rotated ribosomes was quantitatively assessed andnormalized using hSHAPE (18) Figure 3b and c showthat both of these bases were reactive along a continuumbeginning with non-rotated wild-type control (leastreactive) to rpL10-A106R vacant wild type rpL10-S104D and finally rotated wild-type control (mostreactive) The intermediate peak heights observed withvacant wild-type ribosomes are consistent with the viewthat the 80S ribosome is free to transit between the twostates in the absence of ligands (40) Additional keyrRNA bases known to undergo structural changesduring intersubunit rotation were also probed Forexample the B2a intersubunit bridge formed betweenthe distal loop of H69 of the LSU and h44 of the SSUis important for substrate selection on the ribosome (40)and bases comprising this bridge undergo significant re-arrangement during subunit rotation albeit not asdramatic as the B7a bridge (27) Table 1 shows thatbases involved in the B2 bridge are less reactive innon-rotated control and rpl10-A106R mutant ribosomesthan their rotated control and rpl10-S104D counterpartsThe B3 bridge was utilized as an internal control becauseit is not disrupted during intersubunit rotation and isthought to be the pivot around which the subunitsrotate (72741) Consistent with this no significantchanges in B3 base reactivities were observed The aa-tRNA accommodation corridor (AC) closes uponribosome rotation rendering the lsquogate basesrsquo lessreactive (741ndash43) This is reflected as increasedchemical reactivities of these bases (U2860 U2924 andU2926) in non-rotated control and A106R mutant ribo-somes as compared to S104D and rotated controls Inaddition specific bases in the SarcinRicin loop arereactive in non-rotated E coli ribosomes (30) Thesame protection patterns are observed at the analogousyeast rRNA bases (U3023 and A3027) Collectively the

2054 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

analyses shown in Table 1 support the hypothesis thatthe S104D mutant drives the structural equilibrium of80S ribosomes toward the rotated state and that theA106R mutant drives it toward the non-rotated state

The S104D and A106R mutants define local and longdistance changes in rRNA structure corresponding toribosome rotational status

Having established the rotational status of the mutantribosomes 1M7 and hSHAPE were used to quantitativelyassess the chemical reactivities of approximately half(2000 nt) of the rRNA content of salt-washed vacant80S ribosomes purified from isogenic cells expressingwild type and the mutant forms of rpL10 The highly re-producible quantitative data enabled subtraction analysesfor each base probed Supplementary Figure S5 employs aheat map based approach in which the reactivity dataobtained from A106R mutant ribosomes were subtractedfrom data obtained from S104D mutant ribosomes Thisanalysis yields maps highlighting the differences in rRNAbase reactivities between the conformational states of thetwo mutants In these maps warmer colors signify higherreactivity and implicit higher flexibility of rRNA bases inS104D relative to A106R while cooler colors correspondto lower reactivity and implicit higher structure and con-straint Reactivity scale numbers denote the extent of thedifferences with each step in color as statisticallydetermined (each step denoting one standard deviation)(18) These are mapped onto two-dimensional flat rRNAmaps (Supplementary Figure S5andashc) onto three-dimen-sional (3D) atomic resolution structures (SupplementaryFigure S5d and e) and are summarized in Figure 6cSignificant differences between the two mutants were

observed in the core of the PTC the rRNA structure

Non-rotated A106R WT S104D Rotated control control

A2207 chemical probing (1M7)

G913 chemical probing (kethoxal)

Reactivity 2 2 3 4 4

Reactivity 1 2 3 4 4

(a) (b)

(c)

Figure 3 rpL10 loop mutants alter the rotational equilibrium of the ribosome (a) The B7a intersubunit bridge In the non-rotated state A2207 (25SrRNA) forms an interaction with G913 (18S rRNA) In the rotated state that interaction is disrupted and the marked 20 OH-group on A2207becomes accessible to modification by 1M7 Similarly marked atoms on G913 become accessible to modification by kethoxal upon rotation(b) Reactivity peaks obtained by hSHAPE after probing of the landmark base A2207 (arrows) at the LSU side of the B7a intersubunit bridgewith 1M7 (c) Reactivity peaks obtained by hSHAPE after chemical probing of the landmark base G913 at the SSU side of the B7a intersubunitbridge with kethoxal Data were assigned five levels of reactivity 0 (for less than a tracersquos median value) 1 (for a value between the median and themean) 2 (for a value between the mean and the first standard deviation) 3 (for a value between the first and the second standard deviation) and 4(for values above the second standard deviation) Mean median and standard deviation were calculated on a per trace basis

Table 1 Establishing the rotational status of mutant ribosomes

Region rRNA base Non-rotated A106R S104D Rotated

B7a A2207 2 2 4 4G913 (SSU) 1 2 4 4

B2a U2258 2 2 3 4A2262 0 0 1 2C1644 (SSU) 1 1 2 2G1645 (SSU) 0 0 2 1

B3 U2301 0 0 1 1G2302 0 0 0 0A1655 (SSU) 1 2 2 2U1656 (SSU) 2 2 2 2

AC U2860 1 2 0 0U2924 4 2 0 1A2926 4 2 1 1

SRL U3023 1 1 0 0A3027 2 3 0 0

Non-rotated control yeast ribosomes were primed with polyU and con-tained Ac-Phe-tRNAPhe in the P-site Rotated control ribosomes wereprimed with polyU loaded with deacylated Phe-tRNA and incubatedwith eEF-2-GDPNP These complexes along with salt-washed vacantRpl10-A106R and Rpl10-S104D ribosomes were chemically probed with1M7 and analyzed by hSHAPE B7 B2a and B3 denote the correspond-ing intersubunit bridges Numerical values reflect statistically normalizedreactivities on a scale of 0ndash4 with 4 being most reactive (1819)

Nucleic Acids Research 2014 Vol 42 No 3 2055

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

most proximal to the rpL10 loop (Supplementary FigureS5a) Three of the four bases proposed to play centralroles in the induced fit model of peptidyltransfer (44)ie G2922 (E coli G2553) U2924 (E coli U2555) andU2954 (E coli U2584) were less reactive in the rpL10-S104D mutant as compared to rpL10-A106R(Supplementary Figure S5a) suggesting a role for yeastrpL10 in this process The universally conserved A2819(E coli A2450) G2874 (E coli G2505) and U2875 (Ecoli U2506) that constitute the lsquoentrancersquo to the PTC(43) also showed dramatic differences in chemicalreactivities (Supplementary Figure S5a and d)Moving outward from the loop rpL10 is framed by

Helices 38 (the A-site finger) 39 43 and 89 Significantdifferences in chemical reactivities between the twomutants are seen in all these structures (SupplementaryFigure S5a and b) suggesting that structural rearrange-ments involving the flexible loop may be transducedthrough the body of rpL10 to neighboring rRNA struc-tural elements Focusing on these elements in more detailHelices 89 and 90ndash92 (Supplementary Figure S5a) formthe lsquoACrsquo through which 30 ends of aa-tRNAs transit asthey enter the PTC (43) In this structure U2860 in H89(E coli U2491) and U2924 in H92 (E coli U2555) interactto lsquoclosersquo the AC in the rotated state Conversely they donot interact in the non-rotated state and should be moreextensively chemically modified when the AC is open Insupport of our model the difference map shown inSupplementary Figure S5a and summarized in Table 1demonstrates that these bases are more protected fromchemical attack in S104D mutant ribosomes ascompared to A106R mutantsrpL10 also interacts directly with Helix 39 Examination

of Supplementary Figure S5b reveals large scale structuraldifferences between the two mutant ribosomes extendingfrom this helix to Helix 44 (PTC-distal) Importantlythese include the GTPase activating center and the H43ndash44 structure upon which the P0P1P2 (E coli L7L12)stalk is assembled (45) This region of the LSU interactswith both the elongation ternary complex and eEF2 atdifferent stages during the elongation cycle As discussedbelow we suggest that the structural differences identifiedhere trace an information transmission pathway that helpsthe ribosome distinguish between different trans-actingfactors at different points in its lifecycle One such trans-mission pathway is shown in Supplementary Figure S5ehere the reactivity data from Supplementary Figure S5bare plotted onto the yeast ribosome high-resolution 3Dstructure to illustrate a network of rRNA helices connect-ing the PTC-proximal and PTC-distal portions of rpL10The three site allosteric model posits that structural

changes in the ribosomal A and E sites are linked (46)and studies from our laboratory have identified an exten-sive network of rRNA structural changes extending alongthe entire path taken by tRNAs as they transit theribosome (7424748) Examination of the chemical modi-fication patterns in Helices 82ndash88 (Supplementary FigureS5a) which trace the path of tRNAs from the A- to the E-sites in the LSU reveals extensive differences in rRNAbase reactivities between the two mutants consistent withlinkage between the A- and E-sites

The distal tip of H38 forms the LSU partner of the B1aintersubunit bridge Although no differences in basereactivities were observed in this structure(Supplementary Figure S5b) presumably because it is in-trinsically highly mobile (8334950) changes wereobserved in its SSU partner in the distal loop of h33part of the 30 major or lsquohead domainrsquo of the SSU(Supplementary Figure S5c) A block of changes inrRNA base reactivities were also observed along the uni-versally conserved helices 32ndash35 in the 30 major domain ofthe 18S rRNA These map to the mRNA entrance tunnelon the SSU opposite the decoding center (see Figure 6c)In the 30 minor domain A1755 (E coli A1492) whichplays a central role in stabilizing cognate codonanticodoninteractions in the decoding center (51) is much moreprotected from chemical modification in rpL10-S104Dcompared to rpL10-A106R ribosomes (SupplementaryFigure S5c) The head of the SSU undergoes a dramaticseries of rearrangements during intersubunit rotation(2852) Multiple differences in rRNA base reactivitieswere also noted throughout the head domain (h38ndashh43)providing evidence that the rpL10 loop may influenceintersubunit rotation through the B1bc bridge in apathway involving 5S rRNA and rpL11 (48) This is sup-ported by pronounced differences in rRNA chemical pro-tection patterns at the tip of H88 (Supplementary FigureS5a) which interacts with rpL11 as a monitor of P-sitetRNA occupancy (47) This pathway is shown inFigure 6c

Identification of important points of contact through whichstructural changes in rpL10 are communicated to thesurrounding rRNA

For structural changes in the loop of rpL10 to be broadlycommunicated to both subunits information needs to betransduced through points of contact between rpL10 andlocal rRNA structural elements For example structuralchanges at amino acid R7 located in the N-terminallsquohookrsquo of rpL10 are involved in opening and closing theAC (7) Examination of mutants identified in a previousgenetic screen of rpl10 mutants suggested two additionalcandidates F94 and G81 (Supplementary Figure S9) F94interacts with H38 and the F94I mutant was chosen fordeeper analyses because of its strong resistance toanisomycin (not shown) a competitive inhibitor of aa-tRNA binding to the A-site Replacement of thearomatic phenylalanine with isoleucine (F94I) producedribosomes with higher affinity for elongation ternarycomplex and decreased affinity for eEF2 ie distributedtoward the non-rotated state similar to the A106R mutant(Supplementary Figure S6a) hSHAPE analysis revealedthat rRNA bases proximal to F94 (SupplementaryFigure S6d circled) were deprotected in the F94Imutant presumably due to loss of interactions involvingthe aromatic ring of phenylalanine Interestingly basesfurther down the H38 structure (Supplementary FigureS6d boxed) that interact with the aa-tRNA D-loopshowed decreased reactivities similar to an rRNAmutant of these bases that also displayed increasedaffinity for elongation ternary complex (53) G81 is

2056 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

located at the opposite end of the body of rpL10 closer tothe solvent exposed (PTC-distal) side of the protein and isone of the amino acid residues closest to the elongationfactor binding site (H43) The G81D mutant was chosenbecause of its anisomycin hypersensitivity (not shown)and it promoted the opposite effects ie decreasedaffinity for elongation ternary complex and increasedaffinity for eEF2 (Supplementary Figure S6andashc)hSHAPE analysis also revealed that rRNA basescomprising the eEF2 binding site and the GTPase-associated center in G81D underwent changes in reactivitysimilar to those observed with S104D (SupplementaryFigure S5b) These findings suggest that like S104DG81D distributes ribosomes toward the rotated state

Effects of rpL10 loop mutants on peptidyltransferaseactivity and translational fidelity

A single round puromycin-based assay ofpeptidyltransferase activity (54) revealed that S104Dmutant ribosomes promoted 60 of peptidyltransferaseactivity relative to wild type and A106R mutant ribosomes(Figure 4a Supplementary Figure S7) As peptidyltransferoccurs in the context of the non-rotated state this furthersupports the hypothesis that the rpL10-S104D mutantdrives the equilibrium of ribosomes toward the rotatedstate Programmed 1 ribosomal frameshifting (1PRF) directed by L-A virus-derived sequence requiresslippage of both A- and P-site tRNAs (55) while Ty1directed +1 PRF only requires slippage of P-site tRNA(56) Both mutants promoted enhanced 1 PRF but hadno effects on +1 PRF (Figure 4b) To more completelyunderstand how the structural changes induced by theS104D and A106R mutations impact the ability of theribosome to accurately decode mRNAs we employeddual-luciferase translational fidelity assays designed tomeasure rates of misreading of near- and non-cognatetRNAs (11) and misreading of a UAA terminationcodon (10) The S104D mutant was generally more in-accurate than wild-type the magnitude of which becamegreater as the extent of codonanticodon mismatchesincreased (Figure 4c) In contrast decoding was largelyunaffected in cells expressing the A106R mutantalthough decoding of near-cognate tRNAs was slightlymore accurate

An intra-ribosomal suppressor of the S104D mutantreveals the importance of LSU allostery in functionof the mature ribosome

If the rpL10 mutants affect the distribution of ribosomesbetween the classical and rotated states by altering thedistribution of allosteric structures intrinsic to theribosome then mutants of other ribosomal componentsthat promote opposing effects may be able to suppressthe defects exhibited by the rpL10 mutants The L3-W255C mutant which lies in a flexible loop (theW-finger) on the opposite side of the aa-tRNA AC fromrpL10 (Figure 6c) promotes increased affinity of aa-tRNAto the A-site and decreased affinity for eEF2 (54) A highcopy plasmid expressing L3-W255C was able to almostcompletely suppress the biochemical (binding of

aa-tRNA and Sdo1p to the A-site) and translationalfidelity (1 PRF) defects of the rpL10-S104D mutant(Figure 5andashc Supplementary Figure S8c) The ability ofthis mutant to correct the changes in rRNA structure atthe B7a bridge promoted by the rpL10-S104D mutant(Figure 5d) demonstrates that the proper rotational equi-librium was re-established The control experimentco-expression of L3-W255C with the wild-type RPL3did not affect any of these parameters Interestingly thismutant did not suppress the rpL10-S104D slow growthdefect (data not shown) or the ribosome biogenesisdefect of rpl10-S104D as assessed by polysome profiling(Supplementary Figure S8) No mutants of L3 able tosuppress the rpL10-A106R mutant were identified

DISCUSSION

Eukaryotic rpL10 and its prokaryotic homologs L16 inbacteria and L10e in archaea contain a conservedinternal loop that approaches the catalytic center of theLSU This loop is not resolved in atomic resolution struc-tures of vacant yeast (8) or prokaryotic ribosomes (57)suggesting that this structure is dynamic However theinternal loop is resolved in high-resolution crystal struc-tures of bacterial ribosomes containing P-site tRNA(2058) and in cryo-EM imaging of translating yeast ribo-somes (59) suggesting that it is stabilized by tRNAbinding While the loop is shorter in the bacterial andarchaeal homologs the N-terminus of the L27 proteinsfrom these kingdoms appear to provide the structuralmimic for the tip of this loop (60) Here we have shownthat mutations of this loop affect the rotational status ofthe ribosome altering the ribosomersquos affinity for Sdo1p atthe P-site for aa-tRNAs and eEF2 at the A-site andglobally affecting ribosome function from biogenesisthrough translational fidelityCurrent models posit that ribosomal rotational status is

solely determined by binding of different tRNA speciesand trans-acting GTPases (26ndash2861) However thefindings presented here suggest that control of rotationis an intrinsic property of the ribosome We proposethat when the A-site is unoccupied by ligand the flexiblerpL10 loop can sample this space we call this the lsquoflippedinrsquo conformation In contrast occupation of the A-site bytRNAs displaces or lsquoflips outrsquo this loop (Figure 6a and b)The effects of the two rpL10 mutants on A-site associatedfunctions and on ribosome structure in the absence ofligands suggest that this loop has an intrinsic centralrole in establishing the rotational status of the ribosomeThe finding that non-enzymatic binding of aa-tRNA tothe A-site is exacerbated and that peptidyltransferaseactivity is diminished in rpL10-S104D provide additionalsupport for this idea Addition of the positively chargedarginine residue may create new chargendashcharge inter-actions with the A-site to stabilize the lsquoflipped inrsquo statemaking it more difficult to displace by aa-tRNA (Figure6a) Conversely addition of a negatively charged residuein the S104D mutation may interfere with the ability of theloop to stably occupy the A-site due to charge repulsion

Nucleic Acids Research 2014 Vol 42 No 3 2057

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

effects with this rRNA-rich environment thus stabilizingthe lsquoflipped outrsquo conformation (Figure 6b)How can a change in the position of a small peptide

loop affect the conformation of a large and complexmacromolecule like the ribosome We suggest thatligand occupancy information is first transduced locallyby lateral movement of the body of rpL10 as a result ofdynamic changes in loop positioning This information isallosterically transduced to more distant regions of theribosome through rRNA networks thus setting theribosome rotational mechanism in motion In thismodel the body of rpL10 functions akin to a piston

that is physically linked to multiple functional centers ofthe machine (visualized in Figure 6 primary hSHAPEdata summarized in Supplementary Figure S9) The allo-steric information pathways identified by hSHAPEanalysis frame the path through which tRNA movesthrough the ribosome during translation including theAC and decoding center the PTC and the E-site Theyalso encompass the critical intersubunit bridges involvedin rotation and the elongation factor binding site Previousstudies identified other flexible peptide elements that areused by the ribosome as allosteric switches These includethe N-terminal hook of rpL10 (7) the W-finger

0

10

20

30

40

50

WT WT +L3-W255C

S104D S104D + L3-W255C

0

40

80

120

160

0

10

20

30

40

-1 PRF

+1 PRF

K

(Ph

e-tR

NA

n

M)

Ph

e

K (

Sd

o1p

nM

)

Fra

mes

hif

tin

g (

)

WT S104D S104D + L3-W255C

WT WT +L3-W255C

S104D S104D + L3-W255C

WT WT +L3-W255C

S104D S104D + L3-W255C

D

D

(a) (b)

(c) (d)

Figure 5 A mutant of rpL10 can be intrinsically suppressed with a mutant of rpL3 (a and b) Binding aa-tRNA and Sdo1p to indicated ribosomes(c) Frameshifting analyses (d) Structural probing of the landmark base A2207 (arrows) at the LSU side of the B7a intersubunit bridge with 1M7Bars indicate SEM (n=4 for a and c n=3 for b) Plt 005 Plt 001

0

002

004

006

WT S104D A106R0

10

20

30

40

WT S104D A106R

-1 PRF+1 PRF

0

1

2

3

4

5

S104D A106R

Near-cognate

Non-cognate

Stop codon

PTa

se K

(m

in

)ap

p-1

Fra

mes

hif

tin

g (

)

Tran

slat

ion

al m

isre

adin

g(f

old

-WT

)

(b) (c)(a)

Figure 4 rpL10 loop mutants affect peptidyltransferase activity and translational fidelity (a) Apparent rates of peptidyltransfer from single turnoverpeptidylpuromycin reactions for indicated ribosomes (b) Programmed 1 and +1 ribosomal frameshifting values obtained using dual-luciferasereporters (c) Misincorporation of a stop codon and near- and non-cognate amino acids in mutants compared to wild-type levels as monitored usingdual-luciferase reporters Bars indicate SEM (n=4) Plt 005 Plt 001

2058 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

N-terminal extension and Basic Thumb elements of rpL3(145462) an internal loop located in rpL5 (63) and the P-site loop of rpL11 (47) Importantly there is a consider-able degree of overlap between the allosteric networksidentified in these and other studies (414248536465)indicating that they are all components a single networkof lsquoswitchesrsquo and lsquowiresrsquo that coordinate ribosome struc-ture with function This is further supported by the obser-vation in the current study that a mutant in one switchL3-W255C can suppress the biochemical and transla-tional fidelity defects of a second ie the rpL10-S104Dloop mutant What distinguishes the rpL10 loop fromother elements is that it is a primary sensor of thearrival of ligands at the A-site As a result defects inthis initial step are amplified through a series of outwardlyradiating allosteric networks broadly impacting ribosomebiogenesis elongation and termination It should be notedthat while we interpret our results as evidence for the in-fluence of rpL10 on ribosome equilibria alternative ex-planations are also possible For example these resultscould be interpreted as evidence for changes in rRNAfolding and the structure of the LSU However incorpor-ation of L10 occurs very late in the process of eukaryoticribosome biogenesis after the pre-60S subunit has beenexported to the cytoplasm well after the core of theLSU has been folded Thus the effects of these single

amino acid mutants of rpL10 on rRNA folding arelikely to be minimalThe rotational status of the 80S ribosome is an import-

ant determinant for binding of trans-acting factorsoptimizing PTC activity and translational fidelity Bothmutants promoted increased rates of 1 PRF(Figure 4b) which requires slippage of both A- andP-site tRNAs (66) but did not affect +1 PRF whichonly requires slippage of P-site tRNA (56) This is consist-ent with the observation that the rpL10 loop mutants onlyaffected A-site tRNA binding Increased utilization ofnear- and non-cognate tRNAs and termination codonreadthrough by the rpL10-S104D mutant (Figure 4c)may be explained by the observation that A1755 (E coliA1492) in h44 was hyperprotected from chemical modifi-cation (Supplementary Figure S5c) Flexibility of thisbase is critical for accurate decoding by stabilizing amini-helix between cognate codonanticodon interactions(67) Concurrently LSU A2256 (E coli A1913) inH69 was hyper-reactive in rpL10-S104D ribosomes(Supplementary Figure S5a) This base also plays acentral role in mRNA decoding where it is paired withA1492 in the non-rotated state but is unpaired andflexible in the rotated state (68) We suggest that the pro-pensity of rpL10-S104D ribosomes to assume the rotatedstated limits the ability of these bases to participate in

L10CP

Head

L11

S18S15

B1bc B1a

B7aB2ac

SRL

L3

H38

5S PTC

AC

DC

mRNA path

FactorBindingSite

H39

H38

H89

t

5S

eEFBinding

sitePDrpL10

t

H89

t t

H39

5SH38

eEFBinding

sitePD

(a)

(b)

(c)

Figure 6 Models of rpL10 function rpL10 is at the center of a cascade of allosteric communication pathways throughout the ribosome (a) TherpL10 loop lsquoflipped inrsquo conformation is the substrate for aa-tRNA and Sdo1p P PTC-proximal D PTC-distal (b) The lsquoflipped outrsquo loop con-formation substrate for eEF2 Binding of an aa-tRNA (indicated by the red lsquotrsquo) causes displacement of the loop from the A-site precipitatingstructural rearrangements in rpL10 These include lateral displacement of the main body of the protein (dashed black arrow) and H38 toward theelongation factor binding site creating the binding platform for eEF2 Release of the N-terminal hook of rpL10 from H89 enables closing of the aa-tRNA AC These movements also initiate allosteric transmission of information through the communication pathways shown in (c) to distantlylocated functional centers of the ribosome to set the stage for the next phase of elongation These include rearrangements in the E-site in preparationfor release of deacylated tRNA and interactions with the decoding center and SSU to initiate subunit rotation (c) Summary of chemical probingexperiments mapping the allosteric information exchange pathways emanating from rpL10 to all the functional centers of the ribosome to influenceintersubunit rotation Intersubunit bridges B1a B1bc B2ac and B7a and ribosomal proteins L3 L10 L11 S15 and L18 are labeled CP centralprotuberance of the LSU SRL SarcinRicin loop DC decoding center

Nucleic Acids Research 2014 Vol 42 No 3 2059

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

mini-helix formation leading to increased utilization ofnon-cognate ligandsConsidering the mechanisms in place to ensure fidelity

of gene expression we expect that safeguards have evolvedto ensure that only correctly functioning ribosomes areutilized in translation Ribosome assembly culminates incytoplasmic maturation where essential ribosomalproteins are added and transacting factors are released(69) A critical step in this pathway is the release of theanti-association factor Tif6p by the concerted action ofSdo1p and the eEF2 paralog Efl1p (7071) The observa-tion that suppressing mutations in Efl1p appear to facili-tate a conformational change in the protein akin to theconformational change observed in eEF2 engendered theproposal that the LSU undergoes a lsquotest driversquo in whichthe integrity of the P-site is assessed through activationof the GTPase of Efl1p (5) Notably similar functionalcheckpoints governing pre-40S maturation have recentlybeen proposed (7273) The findings presented inthe current study add further support and detail to thismodel Specifically the ability of Sdo1p to competefor Ac-aa-tRNA binding in the P-site inhibitpeptidyltransferase activity and stimulate aa-tRNAbinding to the A-site [similar to the stimulatory effect ofP-site bound Ac-aa-tRNA on aa-tRNA binding to theA-site (74)] all suggest that Sdo1p interacts with theP-site stabilizing a pseudo-non-rotated state of the LSUThe structure of Sdo1p has been likened to tRNA (75)and we propose that Sdo1p is a mimic for a P-site ligandthat couples the GTPase activity of the elongation factormimic Efl1p (4) to drive a pseudo-translocation event onthe LSU (Figure 7) Because Efl1p and Sdo1p appear toact on 60S independent of 40S we suggest that the

rpL10-S104D and -A106R mutants promote conform-ational changes within the 60S subunit alone that areanalogous to those in the 60S subunit in the context ofrotated and non-rotated 80S ribosomes The 60S subunitlsquotest driversquo appears to provide the primary quality controlcheck on assembly and function of the LSU before it isreleased into the pool of actively translating ribosomesSdo1p binding defects in the lsquopseudo-rotatedrsquo S104Dmutant can be intrinsically suppressed with the rpL3-W255C mutant likely due to its ability to re-establishthe correct pre-LSU conformation However the failureof this mutant to rescue the ribosome biogenesis defectsuggests that quality control involves more than meremonitoring of ligand binding

The biomedical importance of the rpL10 loop wasrecently highlighted by the discovery that a significantfraction of T-cell acute lymphoblastic leukemia patientsharbor mutations of R98 and in one patient at Q123both of which lie at the base of the loop (76)Remarkably these mutations prevent the release ofTif6p and Nmd3p but can be suppressed by mutationsin Nmd3p indicating that like the S104D mutation R98and Q123 mutations in human Rpl10 cause a failureduring the lsquotest driversquo of the 60S subunit These findingssuggest that defects in ribosome biogenesis andor intranslational fidelity may be drivers of this neoplasticdisease Investigations are currently underway in ourlaboratories to address these questions

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online

Loop in

Loop out

Non-Rotated Rotated

Loop in

PPEE APPE

eEF1A+ GTP

eEF2+ GTP

Sdo1p Efl1p+ GTP

Loop out

Tif6p

Pre-60S 60S

Loop in

Pseudo Rotated

P A

Pseudotranslocation

Non-Rotated

Pseudo Non-Rotated

(a)

(b)

Figure 7 Involvement of the rpL10 loop in ribosome rotation throughout the ribosomal life cycle (a) The elongation cycle of translation rpL10loop positioning and ligand binding The elongation cycle begins at left with the ribosome in the non-rotated state where the E-site contains adeacylated tRNA the P-site is occupied by peptidyl tRNA and the un-occupied A-site can be sampled by the rpL10 loop ie the lsquoflipped inrsquoconformation Following elongation ternary complex binding aa-tRNA accommodation and peptidyltransfer tRNAs assume the hybrid states andthe loop assumes the lsquoflipped outrsquo conformation signaling the ribosome to assume the rotated state eEF2 binds to rotated ribosomes resulting intranslocation and the elongation cycle begins anew (b) The lsquotest driversquo Sdo1p and Efl1p interact with the pre-60S subunit in the pseudo-rotatedstate Efl1p-mediated pseudo-translocation drives Sdo1p into the P-site stabilizing the pseudo-non-rotated state with the rpL10 loop in the lsquoflippedinrsquo conformation This is followed by release of the anti-association factor Tif6p and Sdo1p promoting the final steps of 60S maturation

2060 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

ACKNOWLEDGEMENTS

We wish to thank Ashton Belew SharmishthaMusalgaonkar Alicia Bowen Vivek Advani and RyanKobylarz for technical and intellectual assistance

FUNDING

Public Health Service [2 R01 GM058859-11 and 3 R01GM053655-15S1 to JDD 2 RO1 GM53655 and 3 R01GM053655-15S1 to AWJ] Partially supported by NIHtraining [T32GM080201 to SOS] Funding for openaccess charge The Public Health Service [R01GM053655-15S1 R01 GM058859 RO1 GM53655 andT32GM080201]

Conflict of interest statement None declared

REFERENCES

1 JennerL MelnikovS de LoubresseNG Ben ShemAIskakovaM UrzhumtsevA MeskauskasA DinmanJYusupovaG and YusupovM (2012) Crystal structure of the 80Syeast ribosome Curr Opin Struct Biol 22 759ndash767

2 WestM HedgesJB ChenA and JohnsonAW (2005) Definingthe order in which Nmd3p and Rpl10p load onto nascent 60Sribosomal subunits Mol Cell Biol 25 3802ndash3813

3 MenneTF GoyenecheaB Sanchez-PuigN WongCCTonkinLM AncliffPJ BrostRL CostanzoM BooneC andWarrenAJ (2007) The Shwachman-Bodian-Diamond syndromeprotein mediates translational activation of ribosomes in yeastNat Genet 39 486ndash495

4 FinchAJ HilcenkoC BasseN DrynanLF GoyenecheaBMenneTF GonzalezFA SimpsonP DrsquoSantosCSArendsMJ et al (2011) Uncoupling of GTP hydrolysis fromeIF6 release on the ribosome causes Shwachman-Diamondsyndrome Genes Dev 25 917ndash929

5 BussiereC HashemY AroraS FrankJ and JohnsonAW(2012) Integrity of the P-site is probed during maturation ofthe 60S ribosomal subunit J Cell Biol 197 747ndash759

6 HedgesJ WestM and JohnsonAW (2005) Release of theexport adapter Nmd3p from the 60S ribosomal subunit requiresRpl10p and the cytoplasmic GTPase Lsg1p EMBO J 24567ndash579

7 PetrovAN MeskauskasA RoshwalbSC and DinmanJD(2008) Yeast ribosomal protein L10 helps coordinate tRNAmovement through the large subunit Nucleic Acids Res 366187ndash6198

8 Ben ShemA GarreaudL MelnikovS JennerL YusupovaGand YusupovM (2011) The structure of the eukaryotic ribosomeat 30 A resolution Science 334 1524ndash1529

9 HoferA BussiereC and JohnsonAW (2007) Mutationalanalysis of the ribosomal protein RPL10 from yeast J BiolChem 282 32630ndash32639

10 HargerJW and DinmanJD (2003) An in vivo dual-luciferaseassay system for studying translational recoding in the yeastSaccharomyces cerevisiae RNA 9 1019ndash1024

11 PlantEP NguyenP RussJR PittmanYR NguyenTQuesinberryJT KinzyTG and DinmanJD (2007)Differentiating between near- and non-cognate codons inSaccharomyces cerevisiae PLoS One 2 e517

12 JacobsJL and DinmanJD (2004) Systematic analysis ofbicistronic reporter assay data Nucleic Acids Res 32 e160ndashe170

13 LeshinJA RakauskaiteR DinmanJD and MeskauskasA(2010) Enhanced purity activity and structural integrity of yeastribosomes purified using a general chromatographic methodRNA Biol 7 354ndash360

14 MeskauskasA and DinmanJD (2010) A molecular clampensures allosteric coordination of peptidyltransfer and ligand

binding to the ribosomal A-site Nucleic Acids Res 387800ndash7813

15 MeskauskasA PetrovAN and DinmanJD (2005)Identification of functionally important amino acids of ribosomalprotein L3 by saturation mutagenesis Mol Cell Biol 2510863ndash10874

16 OrtizPA UlloqueR KiharaGK ZhengH and KinzyTG(2006) Translation elongation factor 2 anticodon mimicry domainmutants affect fidelity and diphtheria toxin resistance J BiolChem 281 32639ndash32648

17 WilkinsonKA MerinoEJ and WeeksKM (2006) Selective20-hydroxyl acylation analyzed by primer extension (SHAPE)quantitative RNA structure analysis at single nucleotideresolution Nat Protoc 1 1610ndash1616

18 LeshinJA HeselpothR BelewAT and DinmanJD (2011)High throughput structural analysis of yeast ribosomes usinghSHAPE RNA Biol 8 478ndash487

19 VasaSM GuexN WilkinsonKA WeeksKM andGiddingsMC (2008) ShapeFinder a software system for high-throughput quantitative analysis of nucleic acid reactivityinformation resolved by capillary electrophoresis RNA 141979ndash1990

20 VoorheesRM WeixlbaumerA LoakesD KelleyAC andRamakrishnanV (2009) Insights into substrate stabilizationfrom snapshots of the peptidyl transferase center of the intact70S ribosome Nat Struct Mol Biol 16 528ndash533

21 ArmacheJP JaraschA AngerAM VillaE BeckerTBhushanS JossinetF HabeckM DindarG FranckenbergSet al (2010) Localization of eukaryote-specific ribosomal proteinsin a 55-A cryo-EM map of the 80S eukaryotic ribosome ProcNatl Acad Sci USA 107 19754ndash19759

22 SpahnCM Gomez-LorenzoMG GrassucciRA JorgensenRAndersenGR BeckmannR PenczekPA BallestaJP andFrankJ (2004) Domain movements of elongation factor eEF2and the eukaryotic 80S ribosome facilitate tRNA translocationEMBO J 23 1008ndash1019

23 ShammasC MenneTF HilcenkoC MichellSRGoyenecheaB BoocockGR DuriePR RommensJM andWarrenAJ (2005) Structural and mutational analysis of theSBDS protein family Insight into the leukemia-associatedShwachman-Diamond Syndrome J Biol Chem 28019221ndash19229

24 FrankJ and AgrawalRK (2000) A ratchet-like inter-subunitreorganization of the ribosome during translocation Nature 406318ndash322

25 ValleM ZavialovA SenguptaJ RawatU EhrenbergM andFrankJ (2003) Locking and unlocking of ribosomal motionsCell 114 123ndash134

26 FischerN KonevegaAL WintermeyerW RodninaMV andStarkH (2010) Ribosome dynamics and tRNA movement bytime-resolved electron cryomicroscopy Nature 466 329ndash333

27 ZhangW DunkleJA and CateJH (2009) Structures of theribosome in intermediate states of ratcheting Science 3251014ndash1017

28 DunkleJA WangL FeldmanMB PulkA ChenVBKapralGJ NoeskeJ RichardsonJS BlanchardSC andCateJH (2011) Structures of the bacterial ribosome in classicaland hybrid states of tRNA binding Science 332 981ndash984

29 ChenJ PetrovA TsaiA OrsquoLearySE and PuglisiJD (2013)Coordinated conformational and compositional dynamics driveribosome translocation Nat Struct Mol Biol 20 718ndash727

30 MoazedD and NollerHF (1989) Intermediate states in themovement of transfer RNA in the ribosome Nature 342142ndash148

31 SternS MoazedD and NollerHF (1988) Structural analysis ofRNA using chemical and enzymatic probing monitored by primerextension Methods Enzymol 164 481ndash489

32 AgrawalRK SpahnCM PenczekP GrassucciRANierhausKH and FrankJ (2000) Visualization of tRNAmovements on the Escherichia coli 70S ribosome during theelongation cycle J Cell Biol 150 447ndash460

33 CateJH YusupovMM YusupovaGZ EarnestTN andNollerHF (1999) X-ray crystal structures of 70S ribosomefunctional complexes Science 285 2095ndash2104

Nucleic Acids Research 2014 Vol 42 No 3 2061

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

34 GaoH SenguptaJ ValleM KorostelevA EswarNStaggSM Van RoeyP AgrawalRK HarveySC SaliAet al (2003) Study of the structural dynamics of the E coli 70Sribosome using real-space refinement Cell 113 789ndash801

35 MerinoEJ WilkinsonKA CoughlanJL and WeeksKM(2005) RNA structure analysis at single nucleotide resolution byselective 20-hydroxyl acylation and primer extension (SHAPE)J Am Chem Soc 127 4223ndash4231

36 CornishPV ErmolenkoDN NollerHF and HaT (2008)Spontaneous intersubunit rotation in single ribosomes Mol Cell30 578ndash588

37 TaylorDJ DevkotaB HuangAD TopfM NarayananESaliA HarveySC and FrankJ (2009) Comprehensivemolecular structure of the eukaryotic ribosome Structure 171591ndash1604

38 Ben ShemA JennerL YusupovaG and YusupovM (2010)Crystal structure of the eukaryotic ribosome Science 3301203ndash1209

39 SchmeingTM and RamakrishnanV (2009) What recentribosome structures have revealed about the mechanism oftranslation Nature 461 1234ndash1242

40 FischerN KonevegaAL WintermeyerW RodninaMV andStarkH (2010) Ribosome dynamics and tRNA movement bytime-resolved electron cryomicroscopy Nature 466 329ndash333

41 RakauskaiteR and DinmanJD (2011) Mutations of highlyconserved bases in the peptidyltransferase center inducecompensatory rearrangements in yeast ribosomes RNA 17855ndash864

42 RakauskaiteR and DinmanJD (2008) rRNA mutants in theyeast peptidyltransferase center reveal allosteric informationnetworks and mechanisms of drug resistance Nucleic Acids Res36 1497ndash1507

43 SanbonmatsuKY JosephS and TungCS (2005) Simulatingmovement of tRNA into the ribosome during decoding ProcNatl Acad Sci USA 102 15854ndash15859

44 SchmeingTM HuangKS StrobelSA and SteitzTA (2005)An induced-fit mechanism to promote peptide bondformation and exclude hydrolysis of peptidyl-tRNA Nature 438520ndash524

45 RemachaM Jimenez-DiazA SantosC BrionesEZambranoR Rodriguez GabrielMA GuarinosE andBallestaJP (1995) Proteins P1 P2 and P0 components of theeukaryotic ribosome stalk New structural and functional aspectsBiochem Cell Biol 73 959ndash968

46 NierhausKH (1990) The allosteric three-site model for theribosomal elongation cycle features and future Biochemistry 294997ndash5008

47 RhodinMH and DinmanJD (2010) A flexible loop in yeastribosomal protein L11 coordinates P-site tRNA binding NucleicAcids Res 38 8377ndash8389

48 RhodinMH and DinmanJD (2011) An extensive network ofinformation flow through the B1bc intersubunit bridge of theyeast ribosome PLoS ONE 6 e20048

49 NollerHF YusupovMM YusupovaGZ BaucomALiebermanK LancasterL DallasA FredrickK EarnestTNand CateJH (2001) Structure of the ribosome at 55 Aresolution and its interactions with functional ligands ColdSpring Harb Symp Quant Biol 66 57ndash66

50 SelmerM DunhamCM MurphyFV WeixlbaumerAPetryS KelleyAC WeirJR and RamakrishnanV (2006)Structure of the 70S ribosome complexed with mRNA andtRNA Science 313 1935ndash1942

51 OgleJM BrodersenDE ClemonsWM Jr TarryMJCarterAP and RamakrishnanV (2001) Recognition of cognatetransfer RNA by the 30S ribosomal subunit Science 292897ndash902

52 RatjeAH LoerkeJ MikolajkaA BrunnerMHildebrandPW StarostaAL DonhoferA ConnellSRFuciniP MielkeT et al (2010) Head swivel on the ribosomefacilitates translocation by means of intra-subunit tRNA hybridsites Nature 468 713ndash716

53 RakauskaiteR and DinmanJD (2006) An arc of unpairedlsquolsquohinge basesrsquorsquo facilitates information exchange among functionalcenters of the ribosome Mol Cell Biol 26 8992ndash9002

54 MeskauskasA and DinmanJD (2007) Ribosomal protein L3Gatekeeper to the A-site Mol Cell 25 877ndash888

55 DinmanJD (2012) Mechanisms and implications of programmedtranslational frameshifting Wiley Interdiscip Rev RNA 3661ndash673

56 BelcourtMF and FarabaughPJ (1990) Ribosomal frameshiftingin the yeast retrotransposon Ty tRNAs induce slippage on a 7nucleotide minimal site Cell 62 339ndash352

57 KavranJM and SteitzTA (2007) Structure of the base of theL7L12 stalk of the Haloarcula marismortui large ribosomalsubunit analysis of L11 movements J Mol Biol 3711047ndash1059

58 KorostelevA TrakhanovS LaurbergM and NollerHF (2006)Crystal structure of a 70S ribosome-tRNA complex revealsfunctional interactions and rearrangements Cell 126 1065ndash1077

59 ArmacheJP JaraschA AngerAM VillaE BeckerTBhushanS JossinetF HabeckM DindarG FranckenbergSet al (2010) Cryo-EM structure and rRNA model of atranslating eukaryotic 80S ribosome at 55-A resolution ProcNatl Acad Sci USA 107 19748ndash19753

60 GaoYG SelmerM DunhamCM WeixlbaumerAKelleyAC and RamakrishnanV (2009) The structure ofthe ribosome with elongation factor G trapped in theposttranslocational state Science 326 694ndash699

61 DunkleJA and CateJH (2010) Ribosome structure anddynamics during translocation and termination Annu RevBiophys 39 227ndash244

62 MeskauskasA and DinmanJD (2008) Ribosomal protein L3functions as a lsquorocker switchrsquo to aid in coordinating of largesubunit-associated functions in eukaryotes and Archaea NucleicAcids Res 36 6175ndash6186

63 MeskauskasA and DinmanJD (2001) Ribosomal protein L5helps anchor peptidyl-tRNA to the P-site in Saccharomycescerevisiae RNA 7 1084ndash1096

64 KiparisovS PetrovA MeskauskasA SergievPVDontsovaOA and DinmanJD (2005) Structural andfunctional analysis of 5S rRNA Mol Genet Genomics 27235ndash247

65 MeskauskasA RussJR and DinmanJD (2008) Structurefunction analysis of yeast ribosomal protein L2 Nucleic AcidsRes 36 1826ndash1835

66 JacksT MadhaniHD MasirazFR and VarmusHE (1988)Signals for ribosomal frameshifting in the Rous Sarcoma Virusgag-pol region Cell 55 447ndash458

67 CarterAP ClemonsWM BrodersenDE Morgan-WarrenRJ WimberlyBT and RamakrishnanV (2000)Functional insights from the structure of the 30Sribosomal subunit and its interactions with antibiotics Nature407 340ndash348

68 DemeshkinaN JennerL WesthofE YusupovM andYusupovaG (2012) A new understanding of the decodingprinciple on the ribosome Nature 484 256ndash259

69 LoKY LiZ BussiereC BressonS MarcotteEM andJohnsonAW (2010) Defining the pathway of cytoplasmicmaturation of the 60S ribosomal subunit Mol Cell 39 196ndash208

70 ValenzuelaDM ChaudhuriA and MaitraU (1982) Eukaryoticribosomal subunit anti-association activity of calf liver iscontained in a single polypeptide chain protein of Mr=25500(eukaryotic initiation factor 6) J Biol Chem 257 7712ndash7719

71 RussellDW and SpremulliLL (1979) Purification andcharacterization of a ribosome dissociation factor (eukaryoticinitiation factor 6) from wheat germ J Biol Chem 2548796ndash8800

72 StrunkBS NovakMN YoungCL and KarbsteinK (2012)A translation-like cycle is a quality control checkpoint formaturing 40S ribosome subunits Cell 150 111ndash121

73 LebaronS SchneiderC van NuesRW SwiatkowskaAWalshD BottcherB GrannemanS WatkinsNJ andTollerveyD (2012) Proofreading of pre-40S ribosome maturationby a translation initiation factor and 60S subunits Nat StructMol Biol 19 744ndash753

74 RheinbergerHJ SternbachH and NierhausKH (1981) ThreetRNA binding sites on Escherichia coli ribosomes Proc NatlAcad Sci USA 78 5310ndash5314

2062 Nucleic Acids Research 2014 Vol 42 No 3

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from

75 NgCL WatermanDG KooninEV WaltersADChongJP IsupovMN LebedevAA BunkaDHStockleyPG Ortiz-LombardiaM et al (2009) Conformationalflexibility and molecular interactions of an archaeal homologue ofthe Shwachman-Bodian-Diamond syndrome protein BMC StructBiol 9 32

76 De KeersmaeckerK AtakZK LiN VicenteC PatchettSGirardiT GianfeliciV GeerdensE ClappierE PorcuM et al(2013) Exome sequencing identifies mutation in CNOT3 andribosomal genes RPL5 and RPL10 in T-cell acute lymphoblasticleukemia Nat Genet 45 186ndash190

Nucleic Acids Research 2014 Vol 42 No 3 2063

by guest on August 15 2016

httpnaroxfordjournalsorgD

ownloaded from