10
Review The mechanics of ribosomal translocation John Achenbach a , Knud H. Nierhaus b, c, * a NOXXON Pharma AG, Max-Dohrn-Str. 8-10, 10589 Berlin, Germany b Institut für Medizinische Physik und Biophysik, Charit e, Charit eplatz 1, 10117 Berlin, Germany c Max-Planck-Institut für Molekulare Genetik, Ihnestr. 73, D-14195 Berlin, Germany article info Article history: Received 5 September 2014 Accepted 5 December 2014 Available online xxx Keywords: Protein synthesis Elongation cycle Translocation Control of tRNA 2 $mRNA movement on the ribosome EF-G dependent GTPase activity abstract The ribosome translates the sequence of codons of an mRNA into the corresponding sequence of amino acids as it moves along the mRNA with a codon-step width of about 10 Å. The movement of the million- dalton complex ribosome is triggered by the universal elongation factor G (EF2 in archaea and eukary- otes) and is termed translocation. Unraveling the molecular details of translocation is one of the most challenging tasks of current ribosome research. In the last two years, enormous progress has been ob- tained by highly-resolved X-ray and cryo-electron microscopic structures as well as by sophisticated biochemical approaches concerning the trigger and control of the movement of the tRNA 2 $mRNA complex inside the ribosome during translocation. This review inspects and surveys these achievements. © 2014 Elsevier B.V. and Soci et e française de biochimie et biologie Mol eculaire (SFBBM). All rights reserved. 1. Introduction Ribosomal protein-synthesis is divided in three functional phases: initiation, the elongation cycle and termination. Each phase is governed by specialized factors. Accordingly we distinguish fac- tors for initiation, elongation and termination. The elongation cycle is controlled by two universal factors, EF-Tu and EF-G (EF1 and EF2 in eukaryotes), present in all three domains and therefore devel- oped before domain separation around 3 billion years ago. The other two phases, viz. initiation and termination, contain domain- specic factors and thus have been tuned after domain separa- tion. It follows that the elongation phase developed rst, and that in early stages of life on this planet a ribosome just bound to the 5- end of an mRNA and started translation, which all kinds of ribo- somes can still do today, e.g. translating the articial mRNA poly(U) during poly(Phe) synthesis (for review see Ref. [1]). A ribosome separates into two subunits, the large subunit (50S in bacteria, 60S in eukaryotes) and the small subunit (30S/40S, respectively). The large 50S subunit contains two ribosomal RNAs (rRNAs) and 33 L-proteins (L-for large), the 30S subunit one rRNA and 21 S-proteins (S-for small). Fig. 1A shows both subunits from the interface with some structural landmarks. A ribosome decodes an mRNA with the help of adapters, the transfer RNAs (tRNAs). Up to 20 different aminoacyl-tRNA synthe- tases (aaRS) specically recognize their cognate tRNA and ligate the corresponding proteinogenic amino acid to the 3 0 -end of the tRNA, forming aminoacyl-tRNA. The anticodon of the tRNA is located at the other distal end of the L-shaped tRNA and can form three consecutive base pairs with a complementary codon of the mRNA within the ribosome, thereby decoding the genetic code. The elongation factor EF-Tu$GTP binds to aminoacyl-tRNA in a ternary complex (aa-tRNA$EF-Tu$GTP), and carries the aminoacyl- tRNA to the rst ribosomal tRNA binding site, the A site (A for aminoacyl-tRNA) (Fig. 1B). The ribosome harbors two further tRNA binding sites, the P-site (P for peptidyl-tRNA) and the E site (E for exit), through which each tRNA will be stepped in this order before it is released from the E site. At the beginning of an elongation cycle, the P site is occupied by a peptidyl-tRNA and the E site by a deacylated tRNA. The ternary complex aa-tRNA$EF-Tu$GTP interacts with the decoding center on the small subunit, which is part of the A site, and triggers the decoding process. If successful, the GTPase center of EF-Tu is acti- vated. Following GTP cleavage, EF-Tu changes it conformation and leaves the ribosome as EF-Tu$GDP, whereas the aa-tRNA accom- modates into the A site (Fig. 1B). Next, the peptidyl residue is transferred from the peptidyl-tRNA at the P site to the aminoacyl- tRNA at the A site via formation of a peptide bond. This reaction is catalyzed by the peptidyltransferase center located on the large ribosomal subunit. As a result, the nascent peptide is elongated by * Corresponding author. Institut für Medizinische Physik und Biophysik, Charit e, Charit eplatz 1, 10117 Berlin, Germany. Tel.: þ49 30 321 3590. E-mail address: [email protected] (K.H. Nierhaus). Contents lists available at ScienceDirect Biochimie journal homepage: www.elsevier.com/locate/biochi http://dx.doi.org/10.1016/j.biochi.2014.12.003 0300-9084/© 2014 Elsevier B.V. and Soci et e française de biochimie et biologie Mol eculaire (SFBBM). All rights reserved. Biochimie xxx (2014) 1e10 Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mechanics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/ 10.1016/j.biochi.2014.12.003

Mechanism of Ribosomal Translocation

Embed Size (px)

DESCRIPTION

RNA role in evolution

Citation preview

  • nerm

    Translocation

    e semRis trocatt riX-ronce d

    dividele and tccordininatio

    , EF-Tudomainund 3

    specic factors and thus have been tuned after domain separa-

    (rRNAs) and 33 L-proteins (L-for large), the 30S subunit one rRNAand 21 S-proteins (S-for small). Fig. 1A shows both subunits fromthe interface with some structural landmarks.

    tRNA to the rst ribosomal tRNA binding site, the A site (A fortwo further tRNAd the E site (E forn this order before

    site is occupied bytRNA. The ternaryecoding center on

    decoding process. If successful, the GTPase center of EF-Tu is acti-vated. Following GTP cleavage, EF-Tu changes it conformation andleaves the ribosome as EF-Tu$GDP, whereas the aa-tRNA accom-modates into the A site (Fig. 1B). Next, the peptidyl residue istransferred from the peptidyl-tRNA at the P site to the aminoacyl-tRNA at the A site via formation of a peptide bond. This reactionis catalyzed by the peptidyltransferase center located on the largeribosomal subunit. As a result, the nascent peptide is elongated by

    * Corresponding author. Institut fr Medizinische Physik und Biophysik, Charite,Chariteplatz 1, 10117 Berlin, Germany. Tel.: 49 30 321 3590.

    Contents lists availab

    Biochi

    .e l

    Biochimie xxx (2014) 1e10E-mail address: [email protected] (K.H. Nierhaus).respectively). The large 50S subunit contains two ribosomal RNAs the small subunit, which is part of the A site, and triggers thetion. It follows that the elongation phase developed rst, and that inearly stages of life on this planet a ribosome just bound to the 5-end of an mRNA and started translation, which all kinds of ribo-somes can still do today, e.g. translating the articial mRNA poly(U)during poly(Phe) synthesis (for review see Ref. [1]).

    A ribosome separates into two subunits, the large subunit (50Sin bacteria, 60S in eukaryotes) and the small subunit (30S/40S,

    aminoacyl-tRNA) (Fig. 1B). The ribosome harborsbinding sites, the P-site (P for peptidyl-tRNA) anexit), through which each tRNAwill be stepped iit is released from the E site.

    At the beginning of an elongation cycle, the Pa peptidyl-tRNA and the E site by a deacylatedcomplex aa-tRNA$EF-Tu$GTP interacts with the dother two phases, viz. initiation and termination, contain domain- ternary complex (aa-tRNA$EF-Tu$GTP), and carries the aminoacyl-1. Introduction

    Ribosomal protein-synthesis isphases: initiation, the elongation cycis governed by specialized factors. Ators for initiation, elongation and termis controlled by two universal factorsin eukaryotes), present in all threeoped before domain separation arohttp://dx.doi.org/10.1016/j.biochi.2014.12.0030300-9084/ 2014 Elsevier B.V. and Societe franaise

    Please cite this article in press as: J. Achenb10.1016/j.biochi.2014.12.003d in three functionalermination. Each phasegly we distinguish fac-n. The elongation cycleand EF-G (EF1 and EF2s and therefore devel-billion years ago. The

    A ribosome decodes an mRNA with the help of adapters, thetransfer RNAs (tRNAs). Up to 20 different aminoacyl-tRNA synthe-tases (aaRS) specically recognize their cognate tRNA and ligate thecorresponding proteinogenic amino acid to the 30-end of the tRNA,forming aminoacyl-tRNA. The anticodon of the tRNA is located atthe other distal end of the L-shaped tRNA and can form threeconsecutive base pairs with a complementary codon of the mRNAwithin the ribosome, thereby decoding the genetic code.

    The elongation factor EF-Tu$GTP binds to aminoacyl-tRNA in aribosomeEF-G dependent GTPase activityControl of tRNA2$mRNA movement on the 2014 Elsevier B.V. and Societe franaise de biochimie et biologie Moleculaire (SFBBM). All rightsreserved.Review

    The mechanics of ribosomal translocatio

    John Achenbach a, Knud H. Nierhaus b, c, *

    a NOXXON Pharma AG, Max-Dohrn-Str. 8-10, 10589 Berlin, Germanyb Institut fr Medizinische Physik und Biophysik, Charite, Chariteplatz 1, 10117 Berlin, Gc Max-Planck-Institut fr Molekulare Genetik, Ihnestr. 73, D-14195 Berlin, Germany

    a r t i c l e i n f o

    Article history:Received 5 September 2014Accepted 5 December 2014Available online xxx

    Keywords:Protein synthesisElongation cycle

    a b s t r a c t

    The ribosome translates thacids as it moves along thedalton complex ribosomeotes) and is termed translchallenging tasks of currentained by highly-resolvedbiochemical approaches ccomplex inside the ribosom

    journal homepage: wwwde biochimie et biologie Molecula

    ach, K.H. Nierhaus, The mechany

    quence of codons of an mRNA into the corresponding sequence of aminoNA with a codon-step width of about 10 . The movement of the million-iggered by the universal elongation factor G (EF2 in archaea and eukary-ion. Unraveling the molecular details of translocation is one of the mostbosome research. In the last two years, enormous progress has been ob-ay and cryo-electron microscopic structures as well as by sophisticatederning the trigger and control of the movement of the tRNA2$mRNAuring translocation. This review inspects and surveys these achievements.

    le at ScienceDirect

    mie

    sevier .com/locate/biochiire (SFBBM). All rights reserved.

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • s / BJ. Achenbach, K.H. Nierhau2one aminoacyl-residue and now attached to the A-site tRNA,whereas the P-site tRNA is deacylated. The next step is the EF-G$GTP catalyzed translocation, during which the tRNA2$mRNAcomplex is moved by one codon length inside the ribosome,shifting the tRNAs from A and P to P and E sites, respectively.Translocation nishes an elongation cycle, through which a ribo-some runs for each amino-acid incorporation into the nascentpeptide chain. During an elongation cycle a ribosome oscillatesbetween two main states, viz. the pre-translocational (PRE) and thepost-translocational (POST) state, which are separated by highactivation-energy barriers of about 80e100 kJ/mol [2]. The PREstate is characterized by tRNAs at the A and P sites and the POSTstate by tRNAs at the P and E sites.

    Several lines of evidence indicate that the tRNA translocated tothe E site is released during the PRE-state of the next elongationcycle. Accordant with biochemical and genetic data [3,4], cryo-electron microscopic (cryo-EM) structures of bacterial andmammalian POST-state ribosomes [5,6] and an X-ray crystalstructure of bacterial POST-state ribosomes [7] showed that theribosome maintains the E site tRNA after completion of an

    Fig. 1. Ribosomal landmarks and the elongation cycle. A, both subunits are shown from the(light gray and dark gray, respectively), and the small 30S subunit the 16S rRNA (light gray). Rin translocation, as well as the sarcinericin loop (SRL) of the 23S rRNA and the acceptor endssurface representation. The A-site, P-site and E-site tRNAs are also shown. For clarity, onprotuberance of the 50S subunit; PTC, peptidyltransferase center; SRL, a-sarcin-ricin loop. Tadelivered to the ribosomal A site by the elongation factor EF-TueGTP. Next, the nascent peptia deacylated tRNA at the P site. EF-G then catalyzes the translocation of peptidyl-tRNA from tin blue, 30S in yellow. Taken from [28].

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003iochimie xxx (2014) 1e10elongation cycle. In bacteria, tRNA is released from the E site duringan early contact of the ternary complex aa-tRNA$EF-Tu$GTP withthe ribosomal A site after decoding and before aa-tRNA accom-modation [8] (for review, see Ref. [9]), whereas in higher eukary-otes the release happens probably after the accommodationstep [6].

    2. An important ribosomal gross movement beforetranslocation: tRNAs in hybrid sites

    In the PRE state ribosomal subunits can rotate against each otherby 7 in the interface plane [10e13]. If we x the large subunit andlook onto the solvent side of the small one, the small subunit ro-tates counterclockwise (Fig. 2A). This rotation e sometimes calledratcheting e is accompanied by a movement of the tRNAs only onthe large subunit from A and P sites into P and E sites, respectively.On the small subunit, the tRNAs stay at A and P sites, respectively.The consequence is that the tRNAs are now in hybrid sites anno-tated A/P and P/E hybrid sites, where the rst letter indicates thelocation on the small subunit and the second that on the large

    interface. The large 50S subunit contains the 23S ribosomal RNA (rRNA) and 5S rRNAibosomal proteins are represented as colored ribbons, and those that have specic rolesof A- and P-site tRNAs within the peptidyl-transferase center (PTC), are highlighted by

    ly the anticodon stem-loops of the tRNAs are shown on the 30S subunit. CP, centralken from Ref. [29], modied. B, elongation cycle. At top, an aminoacyl-tRNA (aa-tRNA) isde chain is transferred from the peptidyl-tRNA to the A-tRNA via a peptide bond, leavinghe A site and deacylated tRNA from the P site to P and E sites, respectively. 50S subunits

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • s / BJ. Achenbach, K.H. Nierhausubunit (rotated PRE state); classical sites are accordingly termed A/A, P/P and E/E sites (classical PRE state). This rotation is easilyreversible, independent of factors and energy in the form of GTPand coupled to a strong 30 inward movement of the L1 stalk[14e16] (closed conguration of the L1 stalk; Fig. 2C, rst twopanels). The open conguration of the L1 stalk observed only inthe non-rotated classical PRE-state allows an easy release of the E-tRNA (tRNA at the E site). In the POST state, the L1 stalk adopts anintermediate conguration.

    Now we can rene the denition of translocation: it is amovement of the codoneanticodon duplices on the small subunitfrom A and P sites to the to the P and E sites, respectively.

    3. The EF-G:GTP catalyzed translocation

    When EF-G$GTP binds to a PRE state of a ribosome, it induces astrong conformational change of the 30S subunit within the 70S

    Fig. 2. Conformational changes before and during translocation. A, after peptidyl-transfer, tthe hybrid states H1 and H2 (H2 not shown) owing to intersubunit rotation. When elongatiinduced, leading to the formation of the translocation intermediate TIPOST, which later resolve30S back-rotation. 30S and 50S subunits are sketches roughly mimicking the subunit and tRstructures. Top row, view of the 70S ribosome from the 30S solvent side showing the interspositions. B, positions of the 16S rRNA base A790, which forms an important component of tgate is wide enough (23.6 ) to allow passage of the anticodon stem of the tRNA from the P-positions of the L1 stalk in the open conformation (corresponding to the classical state ofintermediate conformation (TIPOST and POST); the pivot point for rotation of the L1 stalk isCoordinates from PDBs 3J0T & 3J0U [31] (classical PRE); 3J10, 3J14 [31] and 3J0L [12] (PRE

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003iochimie xxx (2014) 1e10 3ribosome: the head of the 30S subunit rotates relative to the 30Sbody by ~18 towards the E site (Fig. 2A). This head rotation esometimes also called head swiveling e probably causes the localstructural events described in the following sections of this chapter,which allow or control the translocational movement of thetRNA2$mRNA complex from the PRE to the POST state. In inter-mediate states of translocation, ribosome$EF-G complexes con-taining mRNA and one or two tRNAs show the tRNAs usually inhybrid states, where the codoneanticodon duplices have leftalready the P and A sites of the small subunit, but did not yet arriveat the E and P sites, respectively. Accordingly, these hybrid statesobserved on translocational intermediates are intersubunit bindingstates and are termed pe/E and ap/P sites [13,17,18].

    A recent comparative analysis of 55 structures has revealed thatthe rotation is not a simple turn around the neck, but rather aexing at two hinge points [19]. Both occur at the two connectorsbetween head and body in the neck region. One connector is helix

    he tRNAs are in the classical state (A/A and P/P), which establishes an equilibrium withon factor G (EF-G) binds to one of these three PRE-states, swiveling of the 30S head iss into the post-translocational state (POST-state) after a reversal of the head swivel andNA arrangements of various functional states that we learned from cryo-EM and X-rayubunit movements. Bottom row, view from above the 70S ribosome showing the tRNAhe A790 gate, corresponding to the ribosomal states that are shown in part A. The A790to the E-site on the 30S subunit during translocation only in the TIPOST intermediate. C,the tRNAs), closed conformation (corresponding to the hybrid states H1 and H2) andindicated by the red dot. For further details see text; taken from Ref. [29], modied.hybrid); 2XUX & 2XUY [13] (TIPOST); 2WRI & 2WRJ [32] (POST).

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • h28, the only covalent connection between head and body. Thesecond non-covalent connector is the co-axially stacked helicesh35/36 of the head, whereby h36 interacts with the body-helix h2via conserved A minor motifs (Fig. 3A and B). Both connectors arethe basis of an arch of head helices h28-h29-h30-h32-h34-h35/36,which represents the core structure of the head and moves as arigid body during head rotation (Fig. 3C and D). Hinge 1 is located inthe middle of h28, around the universally conserved bulged G926,and hinge 2 between the co-axially stacked h35/h36 and h34,where h34 moves around the hinge point in the loop connectingh34 and h35 between the nucleotides 1064 and 1067. Motions atboth hinge points are strictly correlated. We do not know how EF-GGTP triggers head rotation; this is one of the most importantunresolved questions concerning translocation.

    3.1. An early event in translocation

    To understand EF-G's mode at the beginning of translocation,we have to consider briey some features of the decoding process.The decoding center in the A site monitors the correct matching ofcodon and anticodon and xes the codoneanticodon duplex duringsuccessful decoding with a number of hydrogen bonds (H-bonds).

    The rst base pair is monitored and locked by 16S rRNA nucleotideA1493, and the second base pair by nucleotides A1492 and G530;all of them are universally conserved. Additionally, Ser46 (Escher-ichia coli nomenclature) of ribosomal protein S12 is involved withthe second base-pair (Fig. 4A and B, respectively; [20]). All theseelements constitute the decoding center. Control of the third basepair is less stringent, e.g. allowing wobble base pairs. In contrast tothe rst and second base pair, a WatsoneCrick geometry is notstrictly required. Consequently, the interactions between decodingcenter and the third base pair are less tight.

    Note that all the hydrogen bonds between the elements of thedecoding center and the rst two base pairs of the codoneanticodonduplex are sequence independent. This is clear for the 20-OH groupsinvolved (Fig. 4A and B), but also true for contacts involving bases ofthe mRNA: Fig. 4A shows a hydrogen bond between the 20-OH ofA1493 and the O2 of themRNA pyrimidine base U1. Importantly, if apurine base would be in place of U1, the purine N3 would occupy aposition equivalent to the pyrimidine O2 in the minor groove of adouble helix [21] and like the O2 of pyrimidine bases, a purine N3can act as a hydrogen-bond acceptor.

    Although the precise role of the described hydrogen bonds forthe decoding process is still controversial [20,22,23], recent

    rtiaria cosesnnec

    J. Achenbach, K.H. Nierhaus / Biochimie xxx (2014) 1e104Fig. 3. Mechanics of the head rotation according to ref. [19]. A and B, secondary and tewith helices of the body (blue). The dotted lines indicate the connection of h36 and h2 vof the 30S body. C, the two connectors h28 and the coaxially stacked h35/h36 are the bathe universally conserved and bulged G926 in the middle of h28, hinge 2 by the loop co

    rotation is caused by straightening hinge 1 and up-movement of the core helices around thethe head rotate as a rigid body. This Figure was compiled from gures in Ref. [19] and are

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003y structure of the helices forming the core of the 30S head (red) and their connectionsnserved A minor motifs. Orange, structures of the 30S head; black and gray, structuresof the core-helices of the head. The two hinges are indicated. Hinge 1 is represented byting h35 and h34. Blue, non-rotated state of the 30S head; violet, rotated state. D, head

    loop connecting h34 and h35 (between nucleotides 1064 and 1067). The core helices ofreproduced with permission.

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • s / BJ. Achenbach, K.H. Nierhauevidence indicates that the complementary geometry and shape ofthe WatsoneCrick base pairs rather than the hydrogen bonding isresponsible for the decoding process [24]. Be that as it may, it isclear that the hydrogen bonds at the decoding center xcodoneanticodon duplex at the A site, thus stabilizing the PREstate. They also contribute in an important way to the energeticbarrier between the PRE and the POST states. All core residues ofthe decoding center belong to the 30S body; during head rotation,the tRNA2$mRNA complex moves with respect to the body. It fol-lows that this network of hydrogen-bonds has to be weakened ordisrupted, before the duplex can move from the A to the P siteduring translocation. In support of this expectation, it has beendemonstrated that preventing hydrogen-bond formation to the 20-OH group of the rst or second nucleotide of the A-site codon ac-celerates translocation [25].

    The tip of domain IV of EF-G contains two highly conservedloops (loop I and II) that are highly conserved and interact with thedecoding center through hydrogen bonds [17], substituting thenetwork of hydrogen bonds of the codon-anticodon duplex withthis center. This allows head rotation and the concomitant shift ofthe codoneanticodon duplex fromA to P site [26]. The conservationof the two loops covers all kingdoms of the bacterial domain as wellas the eukaryotic mitochondrial translocase mtEF-G1 (Fig. 4C).Interestingly, mitochondria contain a second EF-G factor, mtEF-G2,

    Fig. 4. WatsoneCrick base pairs of the A-site codon, interaction with the decoding center. Aand second nucleotide of the A-site codon, A36 and A35 nucleotides of the anticodon of thedisplayed in brown belong to the 16S rRNA, Ser46 (E. coli nomenclature) belongs to protein Sand its homolog mtEF-G1 and paralog mtEF-G2. The EF-G consensus sequence is also givesymbols were used for the amino acid residues: capitals (red), universally conserved in EF-Gamino acid; @, aromatic (H, F, W, Y); $, aliphatic (I, L, M, V); o, alcoholic (S, T); , negative

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003iochimie xxx (2014) 1e10 5which is highly related to mtEG-G1, but it cannot catalyze ribo-somal translocation; rather it is - in concert with mtRRF - specif-ically involved in the ribosomal recycling phase after termination[27]. Importantly, the sequences of loops I and II of mtEF-G2 arenot related to those of the translocase mtEF-G1 (Fig. 4C; [26])providing compelling evidence that the conserved loops I and II inEF-G andmtEF-G1 play an important role for translocation. Anotherinteresting aspect concerning evolution is revealed by a compari-son of the loop sequences of bacterial EF-G with those of eukaryoticcytoplasmic translocases eEF2: the strong loop-conservation seenin bacteria and mitochondria does not extend to the eukaryoticeEF2s. For example, the sequences of loops I and II in E. coli areQSGGRGQ and HDVDSSE, respectively, whereas the correspondingsequences of e.g. homo sapiens are ARQELKQ and HADAIHR. It fol-lows that the described mechanism of mobilizing the codon-anticodon duplex at the beginning of translocation has beenevolved after separation of the bacterial and eukaryotic domains.

    3.2. Control of the tRNA movement

    In addition to the movement-control of the codon-anticodonduplex of the A-tRNA from the A to the P site, the 30S subunit isinvolved in a second control, i.e. the movement of the anticodonstem-loop (ASL) of the P-tRNA from this site to the E site. The ASL

    and B, the rst two base-pairs of the A-site codon, respectively; U1 and U2 are the rsttRNA at the A site. All other residues are elements of the decoding center; nucleotides12. Coordinates from PDB 1IBM [20]. C, sequence alignments of loops I and II from EF-Gn under mtEF-G1 and mtEF-G2 as reference. From Ref. [26], modied. The followingand also observed in some sequences of mtEF-G1 and 2; lower case (blue), prevailing

    ly charged (D, E); , positively charged (K, R).

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • of the P-tRNA has to pass a gate on the way to the E site, and thisgate consists of the residue A790 of the 30S platform (part of the30S body) and the residues G1338-A-N-U1341 of the 30S headtermed the A790 gate [28,29], all residues except N are univer-sally conserved. This gate has a width of about 14 (closed gate),too narrow for an RNA double helix with a width of 20 (Fig. 2B).The Cate group, who published an X-ray structure of 70S E. coliribosomes 2005, suggested this hindrance [30]; remarkably theirribosome preparation did not contain mRNA or tRNAs. The gate isclosed in the PRE and POST states (panels 1, 2 [31] and 4 [7,32] inFig. 2B) and open to a width of up to 24 exclusively in trans-location intermediate states (panel 3 in Fig. 2B), which is docu-mented in a number of cryo-EM and X-ray structures oftranslocational intermediates [13,17,18,33]. The last two refer-ences [18,33] are particularly important, because they illustrate

    site codon are counted from 1 to 3 (counting is continueddownstream), whereas the upstream E-site codon from 1 to 3(continued upstream, see Fig. 5). Both nucleotides C1397 and A1503are again universally conserved and sit on top of massive secondarystructures interconnected by a Watson-Crick base pair [17]. A1503is retracted from the mRNA outside translocation via H-bondingwith ribose 927 and G925 of h28, the only covalent connector be-tween head and body mentioned above [19]. Head rotation in-terrupts these interactions with the result that A1503 ips into theintercalated state. Interestingly, the structures of the two hingesand their involvement in head rotation seem to be valid also ineukaryotic 80S ribosomes [19].

    3.4. Closing and opening of the L1 stalk

    J. Achenbach, K.H. Nierhaus / Biochimie xxx (2014) 1e106for the rst time authentic translocation-intermediate statescontaining EF-G and two tRNAs, and show both a strong headrotation of 18 and an open A790 gate. Opening of the A790 gate iscoupled to and caused by the EF-G induced head rotation of the30S subunit of at least 18e20 (the extent of head rotations invarious structures is summarized in Table S1 of [19]).

    Upon back-rotation of the 30S head and reversal of the inter-subunit rotation, the tRNAs are accommodated in the classical P/Pand E/E sites thus reaching the POST state, where the A790 gate isfully closed again re-erecting the activation-energy barrier beforedissociation of EF-G from the ribosome (Fig. 2B, right panel; [32]).This has the important consequence that back-sliding of thetRNA2$mRNA complex during back-rotation of the head is pre-vented by the presence of domain IV of EF-G at the A site, and whenthe POST state is reached by the closed A790 gate. EF-G and theA790 gate thus take turns in preventing a back-sliding of the tRNAsto a PRE state during and after translocation.

    3.3. Control of the mRNA movement

    After the network of hydrogen bonds between the decodingcenter at the A site and the codoneanticodon duplex has beensubstituted by a couple of hydrogen bonds between the decodingcenter and loops I and II of the EF-G domain IV, the tRNAmovementis controlled by the A790 gate. Interestingly, also the mRNAmovement is controlled during the transition fromPRE to POST. Thebases C1397 and A1503 of the 16S rRNA head domain intercalatebetween the nucleotides 9 and 10 and 1 and 2 exclusively inthe intermediate states of translocational and are thought to pre-vent a back-sliding of the mRNA during back-rotation of the 30Shead, thus exerting a pawl function [17] (Fig. 5, left panel). In thePOST state the two intercalating nucleotides do not touch themRNA (Fig. 5, right panel; [34]). Note that the nucleotides of the P-Fig. 5. Interaction of the bases C1397 and A1503 of the 16S rRNAwith the mRNA in the transbases are not in contact with the mRNA. Coordinates from PDBs 4KDG [17] and 3I8G [34].

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003The structures of the various elongation states, viz. PRE [14,31],translocational intermediates [13,17,18,35,36] and POST states[7,32], have revealed that a conspicuous movement of the L1 stalkaccompanies the transitions from one state to another (Fig. 2C). Thestalk swings by ~30, while the stalk-tip moves by ~50 . In theclassical PRE state during or after occupation the A site the L1 stalkswings out, away from the E-tRNA (open conformation), which cannow easily be released [8]. In the following rotated PRE state withthe tRNAs in the hybrid sites, the stalk shows a closed conformationcontacting the elbow of the tRNA in the P/E site. An intermediateconformation is observed in the translocation intermediates andthe POST state. Whereas structural data unequivocally show a tightcoupling of the functional state of the ribosome and the confor-mation of the L1-stalk, FRET measurements indicated only a loosecoupling between elongation state and stalk conformation [16,37].It is not yet clear, whether this loose coupling depends on thechosen in vitro conditions or reects physiological states of theribosome.

    Because the L1 stalk contacts the P/E-tRNA, it has been sug-gested that the L1 stalk pulls the tRNA from the P/E hybrid site tothe E site [14]. However, deletion of the L1 gene is not lethal andreduces the growth rate to about 50% corresponding to a 50%retardation of the protein-synthesis rate [38]. Furthermore, the L1-lacking mutant shows an unaltered performance of the EF-Gdependent translocation reaction, leaving it unlikely that L1 playsan active role in tRNA translocation [39]. The importance of the L1stalk might rather be its function as a gate for the E-site controllingthe release of the E-tRNA.

    4. Role and trigger of the EF-G dependent GTPase

    The role of the EF-G dependent GTPase activity for the trans-location reactions was controversially discussed in the last decades.location intermediate state according to ref. [17]. In the POST state (right panel), the two

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • 5. Various translocation inhibitors freeze the ribosomes atdistinct stages of translocation

    We saw in the preceding section that one group observed aHis92 conformation in a translocation intermediate different tocorresponding structures of three groups. Several reasons for thesediscrepancies can be discussed, e.g. standard arguments usually putforward are differences in ribosome preparations and/or bufferconditions. Here we see another conspicuous difference, namelythe groupwith the deviating His92 position used the non-cleavableGTP analog GDPNP for the complex formation, whereas the otherthree groups used GDPCP. Comparing the main accumulated statein the presence of EF-G in conjunction with the diverse trans-location inhibitors and GTP analogs used by various groups, a

    Fig. 6. Mechanism of the GTP hydrolysis on EF-G. A, the active GTPase center of EF-Gin complex with a translocation intermediate in the presence of the non-cleavableGTP analog GDPCP (green). The functional motifs of EF-G are shown, namely theP-loop, switch I (SW I) and switch II (SW II, in the middle), together with a part ofthe ribosomal sarcin-ricin loop (SRL). Interactions of His18 and the catalytic His92(E. coli nomenclature) with nucleotides of the SRL are indicated. In the active GTPasestate the catalytic His92 is oriented toward the g-phosphate (g-Ph) of the GDPCP(distance 3 ). Note that His18 and His92 interact with the backbone of the SRL(phosphate-OH groups of G2661 and A2662, respectively). B, left panel, His92 ofthree crystal structures of the translocation intermediate [35,36,52] have beenaligned according to the bound GDPCP; the His92 is in virtually identical positionsindicating an active GTPase center. Right panel, in one translocation intermediate(PDB 3SFS; [17]), the His92 (white) points away from the g-phosphate, similar to theHis92 (orange) in the inactive GTPase center of the POST state (PDB 2WRI; [32]). Theyellow histidine represents the corresponding catalytic residue of RF3 from anRF3$ribosome complex (PDB 3FSF; [53]). For reference, one histidine is shown(brown) in an active conformation.

    s / BOne group reported that the GTPase activity precedes the trans-location reaction coupling the energy of GTP hydrolysis to trans-location [40]. Later, the same group demonstrated that EF-Gdependent GTPase activity including Pi release is not strictlycoupled to the translocation reaction [41]. The following observa-tion shed light on these discrepancies: The binding of the antibioticsparsomycin to the ribosomal peptidyltransferase center on the 50Ssubunit can trigger an accurate translocation reaction in theabsence of EF-G$GTP [42]; this observation indicated that the en-ergy exploited for a translocation reaction is not derived from GTPhydrolysis but rather from other sources; candidates are thestructure of the tRNA2$mRNA after peptide-bond formation [42]and/or EF-G binding to the PRE state. Therefore, EF-G is not a mo-tor protein, but rather a classical G-protein: Upon binding to itstarget (here the PRE ribosome) the latter undergoes a conforma-tional change, which in turn activates the GTPase center of the G-protein. The G-protein falls into its GDP conformation, looses theafnity to the target and dissociates [43]. This view is supported bythe fact that the EF-G dependent GTP hydrolysis increases thetranslocation rate by 4-fold or a little more [40,44,45], which is amoderate effect if one considers that EF-G catalyzed translocation-rate with or without GTP hydrolysis is at least four orders ofmagnitude larger than the rate of spontaneous translocation in theabsence of EF-G [46].

    Our current understanding of the mechanism that triggers theG-factors GTPase activity was originally derived from the crystalstructure of isolated EF-Tu$GTP [47]. Translated into structures ofEF-G$ribosome complexes the mechanism goes as follows: Twohydrophobic residues, Ile19 and Ile61, form a hydrophobic gate.When the ribosome induces a conformational change on EF-G, thehydrophobic gate opens and allows His92 to coordinate a H2Omolecule, which attacks the g-phosphate of the GTP (Fig. 6A). Theresidues Ile19, Ile61 and His92 are located on highly conservedstructures of EF-G called P-loop, switch I (SW I) and switch II (SWII), respectively, which represent the GTPase center of EF-G andcontain the binding site for GTP. The structure shown in Fig. 6 alsoindicates how the factor and the ribosome cooperate to open thehydrophobic gate: It is the connection with the a-sarcin-ricin loop(SRL) of the ribosome, which is a unique functional hot-spot of theribosome with the longest universally conserved stretch of 12 nu-cleotides in a row [48,49]. We have comprehensive evidence thatthe SRL is the ribosomal tool to regulate the EF-G dependent GTPase[11,35,36,49,50]. The mechanism outlined above and shown inFig. 6A is thought to be universally valid for both elongation factorsEF-Tu [51] and EF-G [36].

    It is very satisfying that three independent structure de-terminations of translocation intermediates showed the catalyticHis92 in an identical orientation pointing to the g-phosphatewith adistance of 3 [35,36,52], providing compelling evidence that thisHis92 orientation represents the active state of the EF-G dependentGTPase (Fig. 6B, left panel). Interestingly, a fourth structure of atranslocation intermediate shows the His92 in a strikingly differentorientation, namely in a position practically identical to that seen ina POST-state ribosome, which has an inactive GTPase center(Fig. 6B, right panel, white and orange His92, respectively [17] and[32]). This translocation intermediate structure still contains anopen A790 gate, and the combination of an inactive GTPase centerand an open A790 gate was interpreted as being a late state of atranslocation intermediate just before reaching the POST state witha closed A790 gate [28,29]. Another ribosome structure withanother G-protein bound, viz. RF3 [53], showed the homolog his-tidine residue in an intermediate position (Fig. 6B, right panel,yellow His92), suggesting that His92 ips between two confor-mations, an activated and an inactivated state of the EF-G GTPase

    J. Achenbach, K.H. Nierhaucenter.

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003iochimie xxx (2014) 1e10 7possible correlation becomes apparent:

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • 1. Adding both antibiotics fusidic acid and viomycin to a PRE statein the presence of EF-G and GTP traps the factor on the ribo-some, which remains in the PRE state with the tRNAs in hybridstates [54]. Fusidic acid binds to EF-G on the ribosome, but not toisolated EF-G [55], mutations conferring resistance to fusidicacid are located in EF-G [56]. Fusidic acid does not preventtranslocation or GTP cleavage, but seems to block the EF-Gtransition into its GDP-conformation with the effect, that EF-Gis trapped on ribosomes (for review see Ref. [57]). In the ex-periments reported by Ref. [54], fusidic acid was used toimprove and stabilize EF-G binding to the ribosome. Viomycin isa cyclic peptide antibiotic that blocks translocation and stabi-lizes tRNAs in their hybrid state before translocation [58].Viomycin or fusidic acid together with EF-G$GTP allows GTPcleavage and blocks the ribosome in the PRE state with a headrotation 6 [19].

    2. GDPCP accumulates early translocation intermediates charac-terized by a moderate head rotation of ~6 with a deacylatedtRNA at the P site [35,36] (or up to 13 with vacant ribosomes[52]), a closed A790 gate, but an active GTPase center (His92).

    3. GDPNP gives rise to intermediates representing a late step oftranslocation with a strong head rotation of about 18, an openA790 gate and an inactive GTPase center (His92 in a positionsimilar to that of a POST state [17]).

    4. Fusidic acid can block ribosomes in complex with EF-G$GTP in alate translocation intermediate state after GTP hydrolysis withproperties as described above [18] or even in the POSTstate [7,32].

    6. A novel structural feature of elongating ribosomes:Subunit rolling

    We have seen that ribosomes from all three domains undergotwo gross-conformational changes during the elongation cycle: Therst is the subunit rotation during the PRE state without involve-ment of EF-G$GTP and leading to the two conformers of the PREstate, (i) viz. the classical PRE state with the two tRNAs in theclassical A/A and P/P states, respectively, and (ii) after subunitrotation the rotated PRE state with the tRNAs in the hybrid states A/P and P/E, respectively (Fig. 2A): The second gross conformationalchange is induced by EF-G$GTP and is the head rotation of the 30Ssubunit by about 18, which is responsible for widening the A790gate, the translocation of the codoneanticodon duplices at A and Psites to the P and E sites, and the intercalation of the residues C1397and A1503 between nucleotides of the mRNA (Fig. 5). Transloca-tional intermediates trapped in the presence of EF-G show thetRNAs in hybrid states called pe/E and ap/P sites, where thecodoneanticodon duplices have left the P and A sites but did notyet arrive at the P and E sites, respectively [13,17,18].

    A recent analysis of the conformational landscape of mamma-lian 80S ribosomes (rabbit liver) revealed another conformationalchange of the small subunit [6], which is a ~6 rolling of the smallsubunit on the interface of the large one toward the L1 stalk (Fig. 7),precisely a rotation around an axis in the upper part of h44 of 18SrRNA roughly orthogonal to the intersubunit rotation shown inFig. 2. Subunit rolling is a distinguishing feature between POST and

    in

    J. Achenbach, K.H. Nierhaus / Biochimie xxx (2014) 1e108Fig. 7. Subunit rolling of the 40S subunit during the transition from POST to PRE statesPhewith an mRNA coding for Met, Phe and Lys and contained two tRNAs, tRNA and N-acet

    translocation (POST), respectively. A comparison of both structures revealed a rotation (roll

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003mammalian 80S ribosomes (rabbit liver). The functional complexes were programmedLysylated Lys-tRNA at A and P sites (PRE) or at E and P sites after an eEF2 dependent

    ing) of the 40S subunit by ~6 toward the L1 stalk [6]. See text for more explanations.

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • s / BPRE states, and it occurs during the transition from POST to theclassical PRE state, precisely after the decoding step and duringaccommodation of the aminoacyl-tRNA into the A site. Back-rollinghappens during translocation from PRE to POST state. The impor-tant consequence is that rolling during the transition from POST toPRE decreases the distance between the subunits on the side of theA site by 13e15 , while the distance at the E site between thesubunits increases by 6e7 , facilitating the tRNA release from the Esite. On the other hand, back-rolling during translocation increasesthe subunit distance at the side of the A site, facilitating the accessof a ternary complex aa-tRNA$eEF1A$GTP to the decoding center.This new gross-conformational change has not been observed withbacterial ribosomes during the elongation cycle and thereforemight be a distinguishing feature of ribosomes from bacteria and(higher) eukaryotes.

    We have seen that in the last two years our knowledge about thetranslocation mechanism has made a breathtaking progress. Thelatest developments in cryo-EM participated in this remarkablesuccess, viz. the fact that since two years the introduction of a newgeneration of detectors, the direct electron detectors, led to atomicresolution below 4 . In this respect cryo-EM competes alreadywith classical X-ray, and ribosomal samples evading an X-rayanalysis are still amenable to cryo-EM. Nevertheless, importantproblems of the translocation remain still unresolved, probably themost important one is our ignorance concerning the essential rstcontacts of EF-G with the ribosome, which trigger the grossconformational change of the 30S head rotation, the bases of alltranslocation events discussed here. The only point we know in thisrespect is that GTP cleavage is not involved, because 30S headrotation can be triggered by EF-G complexed with non-cleavableGTP. Concerning translocation there are still exciting times ahead!

    Conict of interest

    There is no conict of interest.

    Acknowledgment

    We thank Drs. Tatyana Budkevich and Christian M.T. Spahn,Charite e Universitatsmedizin Berlin, Germany, for help, discus-sions and support.

    References

    [1] K.H. Nierhaus, Question 6: early steps of evolution and some ideas about asimplied translational machinery, Orig. Life Evol. Biosphere 37 (2007)391e398.

    [2] S. Schilling-Bartetzko, A. Bartetzko, K.H. Nierhaus, Kinetic and thermodynamicparameters for tRNA binding to the ribosome and for the translocation reac-tion, J. Biol. Chem. 267 (1992) 4703e4712.

    [3] H.J. Rheinberger, K.H. Nierhaus, Testing an alternative model for the ribosomalpeptide elongation cycle, Proc. Natl. Acad. Sci. U. S. A. 80 (1983) 4213e4217.

    [4] F.J. Triana-Alonso, K. Chakraburtty, K.H. Nierhaus, The elongation factor 3unique in higher fungi and essential for protein biosynthesis is an E site factor,J. Biol. Chem. 270 (1995) 20473e20478.

    [5] R.K. Agrawal, et al., Visualization of tRNA movements on the Escherichia coli70S ribosome during the elongation cycle, J. Cell. Biol. 150 (2000) 447e460.

    [6] T.V. Budkevich, et al., Regulation of the mammalian elongation cycle by sub-unit rolling: a eukaryotic-specic ribosome rearrangement, Cell 158 (2014)121e131.

    [7] S. Feng, Y. Chen, Y.G. Gao, Crystal structure of 70S ribosome with both cognatetRNAs in the E and P sites representing an authentic elongation complex, PLoSOne 8 (2013) e58829.

    [8] G. Dinos, D.L. Kalpaxis, D.N. Wilson, K.H. Nierhaus, Deacylated tRNA isreleased from the E site upon A site occupation but before GTP is hydrolyzedby EF-Tu, Nucleic Acids Res. 33 (2005) 5291e5296.

    [9] M. Pech, O. Vesper, H. Yamamoto, D.N. Wilson, K.H. Nierhaus, in: H.J. Gross,J.M. Bujnicki (Eds.), Recoding: Expansion of Decoding Rules Enriches GeneExpression, Sronger-Verlag, New York, 2010, pp. 345e362.

    J. Achenbach, K.H. Nierhau[10] J. Frank, R.K. Agrawal, A ratchet-like inter-subunit reorganization of theribosome during translocation, Nature 406 (2000) 318e322.

    Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003[11] S.R. Connell, et al., Structural basis for interaction of the ribosome with theswitch regions of GTP-bound elongation factors, Mol. Cell 25 (2007) 751e764.

    [12] T. Budkevich, et al., Structure and dynamics of the mammalian ribosomalpretranslocation complex, Mol. Cell 44 (2011) 214e224.

    [13] A.H. Ratje, et al., Head swivel on the ribosome facilitates translocation bymeans of intra-subunit tRNA hybrid sites, Nature 468 (2010) 713e716.

    [14] M. Valle, et al., Locking and unlocking of ribosomal motions, Cell 114 (2003)123e134.

    [15] J. Fei, P. Kosuri, D.D. MacDougall, R.L. Gonzalez Jr., Coupling of ribosomal L1stalk and tRNA dynamics during translation elongation, Mol. Cell 30 (2008)348e359.

    [16] J.B. Munro, et al., Spontaneous formation of the unlocked state of theribosome is a multistep process, Proc. Natl. Acad. Sci. U. S. A. 107 (2010)709e714.

    [17] J. Zhou, L. Lancaster, J.P. Donohue, H.F. Noller, Crystal structures of EF-G-ribosome complexes trapped in intermediate states of translocation, Science340 (2013) 1236086.

    [18] D.J. Ramrath, et al., Visualization of two transfer RNAs trapped in transitduring elongation factor G-mediated translocation, Proc. Natl. Acad. Sci. U. S.A. 110 (2013) 20964e20969.

    [19] S. Mohan, J.P. Donohue, H.F. Noller, Molecular mechanics of 30S subunit headrotation, Proc. Natl. Acad. Sci. U. S. A. 111 (2014) 13325e13330.

    [20] J.M. Ogle, et al., Recognition of cognate transfer RNA by the 30S ribosomalsubunit, Science 292 (2001) 897e902.

    [21] N.C. Seeman, J.M. Rosenberg, A. Rich, Sequence-specic recognition of doublehelical nucleic acids by proteins, Proc. Natl. Acad. Sci. U. S. A. 73 (1976)804e808.

    [22] N. Demeshkina, L. Jenner, E. Westhof, M. Yusupov, G. Yusupova, A new un-derstanding of the decoding principle on the ribosome, Nature 484 (2012)256e259.

    [23] X. Zeng, J. Chugh, A. Casiano-Negroni, H.M. Al-Hashimi, C.L. Brooks 3rd,Flipping of the ribosomal a-site adenines provides a basis for tRNA selection,J. Mol. Biol. 426 (2014) 3201e3213.

    [24] P.K. Khade, X. Shi, S. Joseph, Steric complementarity in the decoding center isimportant for tRNA selection by the ribosome, J. Mol. Biol. 425 (2013)3778e3789.

    [25] P.K. Khade, S. Joseph, Messenger RNA interactions in the decoding centercontrol the rate of translocation, Nat. Struct. Mol. Biol. 18 (2011)1300e1302.

    [26] G. Liu, et al., EF-G catalyzes tRNA translocation by disrupting interactionsbetween decoding center and codoneanticodon duplex, Nat. Struct. Mol. Biol.21 (2014) 817e824.

    [27] M. Tsuboi, et al., EF-G2mt is an exclusive recycling factor in mammalianmitochondrial protein synthesis, Mol. Cell 35 (2009) 502e510.

    [28] J. Achenbach, K.H. Nierhaus, Translocation at work, Nat. Struct. Mol. Biol. 20(2013) 1019e1022.

    [29] H. Yamamoto, et al., 70S-Scanning Initiation, a Novel and Frequent InitiationMode of Ribosomal Translation in Bacteria, 2014 submitted for publication.

    [30] B.S. Schuwirth, et al., Structures of the bacterial ribosome at 3.5 A resolution,Science 310 (2005) 827e834.

    [31] X. Agirrezabala, et al., Structural characterization of mRNA-tRNA translocationintermediates, Proc. Natl. Acad. Sci. U. S. A. (2012).

    [32] Y.G. Gao, et al., The structure of the ribosome with elongation factor G trappedin the posttranslocational state, Science 326 (2009) 694e699.

    [33] J. Zhou, L. Lancaster, J.P. Donohue, H.F. Noller, How the ribosome hands the A-site tRNA to the P site during EF-G-catalyzed translocation, Science 345 (2014)1188e1191.

    [34] L.B. Jenner, N. Demeshkina, G. Yusupova, M. Yusupov, Structural aspects ofmessenger RNA reading frame maintenance by the ribosome, Nat. Struct. Mol.Biol. 17 (2010) 555e560.

    [35] Y. Chen, S. Feng, V. Kumar, R. Ero, Y.G. Gao, Structure of EF-G-ribosomecomplex in a pretranslocation state, Nat. Struct. Mol. Biol. 20 (2013)1077e1084.

    [36] D.S. Tourigny, I.S. Fernandez, A.C. Kelley, V. Ramakrishnan, Elongation factor Gbound to the ribosome in an intermediate state of translocation, Science 340(2013) 1235490.

    [37] P.V. Cornish, et al., Following movement of the L1 stalk between three func-tional states in single ribosomes, Proc. Natl. Acad. Sci. U. S. A. 106 (2009)2571e2576.

    [38] A.R. Subramanian, E.R. Dabbs, Functional studies on ribosomes lacking proteinL1 from mutant Escherichia coli, Eur. J. Biochem. 112 (1980) 425e430.

    [39] G. Sander, Ribosomal protein L1 from Escherichia coli. Its role in the binding oftRNA to the ribosome and in elongation factor G-dependent GTP hydrolysis,J. Biol. Chem. 258 (1983) 10098e10103.

    [40] M.V. Rodnina, A. Savelsbergh, V.I. Katunin, W. Wintermeyer, Hydrolysis of GTPby elongation factor G drives tRNA movement on the ribosome, Nature 385(1997) 37e41.

    [41] F. Peske, A. Savelsbergh, V.I. Katunin, M.V. Rodnina, W. Wintermeyer,Conformational changes of the small ribosomal subunit during elongationfactor G-dependent tRNA-mRNA translocation, J. Mol. Biol. 343 (2004)1183e1194.

    [42] K. Fredrick, H.F. Noller, Catalysis of ribosomal translocation by sparsomycin,Science 300 (2003) 1159e1162.

    iochimie xxx (2014) 1e10 9[43] H.R. Bourne, D.A. Sanders, F. McCormick, The GTPase superfamily: conservedstructure and molecular mechanism, Nature 349 (1991) 117e127.

    anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

  • [44] A.V. Zavialov, M. Ehrenberg, Peptidyl-tRNA regulates the GTPase activity oftranslational factors, Cell 114 (2003) 113e122.

    [45] D. Pan, S.V. Kirillov, B.S. Cooperman, Kinetically competent intermediates inthe translocation step of protein synthesis, Mol. Cell 25 (2007) 519e529.

    [46] S. Shoji, S.E. Walker, K. Fredrick, Ribosomal translocation: one step closer tothe molecular mechanism, ACS Chem. Biol. 4 (2009) 93e107.

    [47] H. Berchtold, et al., Crystal structure of active elongation factor Tu revealsmajor domain rearrangements, Nature 365 (1993) 126e132.

    [48] Y. Endo, I.G. Wool, The site of action of alpha-sarcin on eukaryotic ribosomes.The sequence at the alpha-sarcin cleavage site in 28 S ribosomal ribonucleicacid, J. Biol. Chem. 257 (1982) 9054e9060.

    [49] T.P. Hausner, J. Atmadja, K.H. Nierhaus, Evidence that the G2661 region of 23SrRNA is located at the ribosomal binding sites of both elongation factors,Biochimie 69 (1987) 911e923.

    [50] N. Clementi, N. Polacek, Ribosome-associated GTPases: the role of RNA forGTPase activation, RNA Biol. 7 (2010) 521e527.

    [51] R.M. Voorhees, T.M. Schmeing, A.C. Kelley, V. Ramakrishnan, The mechanism foractivation of GTP hydrolysis on the ribosome, Science 330 (2010) 835e838.

    [52] A. Pulk, J.H. Cate, Control of ribosomal subunit rotation by elongation factor G,Science 340 (2013) 1235970.

    [53] J. Zhou, L. Lancaster, S. Trakhanov, H.F. Noller, Crystal structure of releasefactor RF3 trapped in the GTP state on a rotated conformation of the ribosome,Rna 18 (2012) 230e240.

    [54] A.F. Brilot, A.A. Korostelev, D.N. Ermolenko, N. Grigorieff, Structure of theribosome with elongation factor G trapped in the pretranslocation state, Proc.Natl. Acad. Sci. U. S. A. 110 (2013) 20994e20999.

    [55] J.W. Bodley, F.J. Zieve, L. Lin, S.T. Zieve, Formation of the ribosome-G factor-GDP complex in the presence of fusidic acid, Biochem. Biophys. Res. Commun.37 (1969) 437e443.

    [56] A. varsson, et al., Three-dimensional structure of the ribosomal translocase:elongation factor G from Thermus thermophilus, EMBO J. 13 (1994)3669e3677.

    [57] D.N. Wilson, The AeZ of bacterial translation inhibitors, Crit. Rev. Biochem.Mol. Biol. 44 (2009) 393e433.

    [58] D.N. Ermolenko, et al., The antibiotic viomycin traps the ribosome in an in-termediate state of translocation, Nat. Struct. Mol. Biol. 14 (2007) 493e497.

    J. Achenbach, K.H. Nierhaus / Biochimie xxx (2014) 1e1010Please cite this article in press as: J. Achenbach, K.H. Nierhaus, The mech10.1016/j.biochi.2014.12.003anics of ribosomal translocation, Biochimie (2014), http://dx.doi.org/

    The mechanics of ribosomal translocation1. Introduction2. An important ribosomal gross movement before translocation: tRNAs in hybrid sites3. The EF-G:GTP catalyzed translocation3.1. An early event in translocation3.2. Control of the tRNA movement3.3. Control of the mRNA movement3.4. Closing and opening of the L1 stalk

    4. Role and trigger of the EF-G dependent GTPase5. Various translocation inhibitors freeze the ribosomes at distinct stages of translocation6. A novel structural feature of elongating ribosomes: Subunit rollingConflict of interestAcknowledgmentReferences