To my Teachers

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals

I Helgstrand, M., Mandava, C. S., Mulder, F. A., Liljas, A., Sanyal, S., and Akke, M. (2007) The ribosomal stalk binds to translation factors IF2, EF-Tu, EF-G and RF3 via a conserved region of the L12 C-terminal domain, J Mol Biol 365, 468-479

II Huang, C., Mandava, C. S., and Sanyal, S. (2010) The ribosomal stalk plays a key role in IF2-mediated association of the ribosomal subunits, J Mol Biol 399, 145-153

III Mandava, C. S., Ederth, J., Peisker, K., Kumar, R., Szaflarski, W., and Sanyal, S. (2011) Bacterial ribosome requires more than one L12 dimer for efficient initiation and elongation of protein synthesis involving IF2 and EF-G. (Submitted Manuscript)

IV Ederth, J*., Mandava, C. S*., Dasgupta, S., and Sanyal, S. (2009) A single-step method for purification of active His-tagged ribosomes from a genetically engineered Escherichia coli, Nucleic Acids Res 37, e15. (* equal contribution)

Reprints were made with permission from the respective publishers.

Paper not included in this thesis

I Tobin, C., Mandava, CS., Ehrenberg, M., Andersson, DI., Sanyal,S. (2010) Ribosomes lacking protein S20 are defective in mRNA bind-ing and subunit association. Journal of Molecular Biology, 397(3):767–76

Contents

Background ................................................................................................... 11 Ribosome – a macromolecular complex .................................................. 11

Basic components of the ribosome ...................................................... 11 Structural landmarks on the bacterial ribosome .................................. 12 Functional sites on the ribosome ......................................................... 13

Ribosomal stalk ........................................................................................ 15 L7/L12 ................................................................................................. 15 Domains in the L12 dimer ................................................................... 16 The mode of dimerization .................................................................... 16 Structure of the L12 dimers off and on the ribosome .......................... 17 Stoichiometry of the L12 proteins on ribosome .................................. 18 The binding site of the L12 dimers ...................................................... 19 The factor interaction site on L12 ........................................................ 20 Role of L12 in translation .................................................................... 20

Translation ................................................................................................ 21 Translation components ....................................................................... 21 The steps in protein synthesis .............................................................. 23

Aim ............................................................................................................... 27 Interaction of L12 CTD with translational G-factors ............................... 27 Role of L12 in translation initiation ......................................................... 29 Why do bacteria need multiple copies of L12 protein on ribosome? ....... 31 Technological application with stalk protein L12 .................................... 34

Current work and future plan ........................................................................ 36 Kinetic studies on translation initiation .................................................... 36 Future plans .............................................................................................. 38

Identification of residues on L12 involved in the interaction of IF2 during subunit association ................................................................... 38 Comparison of E. coli ribosomes with single, double and triple L12 dimers .................................................................................................. 39

Conclusions ................................................................................................... 40

Summary in Swedish .................................................................................... 41 Ribosomala stjälkprotein L12: struktur, funktion och tillämpningar. ...... 41

Bakgrund ............................................................................................. 41

Samverkan mellan L12 och translationella G-faktorer ........................ 41 L12s roll i initieringen av proteinsyntesen .......................................... 42 Varför behöver bakterierna flera kopior av L12 protein på ribosomen? ........................................................................................... 42 Teknisk tillämpning av stjälkproteinet L12 ......................................... 42

Acknowledgement ........................................................................................ 43

References ..................................................................................................... 44

Abbreviations

DNA Deoxyribonucleic acid RNA Ribonucleic acid rRNA Ribosomal RNA mRNA Messenger RNA tRNA Transfer RNA aa-tRNA Amino-acyl tRNA r-protein Ribosomal protein Mda Mega Dalton kDa Kilo Dalton Å Ångström bps base pairs nt nucleotide CP Central Protuberance PTC Peptidyl-Transferase Centre SD Shine-Dalgarno sequence aSD anti-Shine-Dalgarno sequence TIR Translation Initiation Region RBS Ribosome Binding Site PIC Preinitiation Complex IC Initiation Complex A-site Amino-acyl site P-site Peptidyl site E-site Exit site IF Initiation Factor EF Elongation Factor RF Release Factor RRF Ribosome recycling Factor CTD C terminal domain NTD N terminal domain GTP Guanosine triphosphate GDP Guanosine diphosphate ATP Adenosine triphosphate FRET Fluorescence Resonance Energy Transfer fMet Formyl Methionine Met Methionine Leu Leucine

cryo-EM cryo-Electron Microscopy GEF Guanine nucleotide Exchange Factor GAC GTPase associated center E. coli Escherichia coli

11

Background

Ribosome – a macromolecular complex The basic mechanism of protein synthesis is similar throughout all forms of life, where ribosome plays a crucial role together with its accesso-ries. Ribosomes are large ribonucleoprotein complexes with 200- 250 Å in diameter and with an approximate mass of 2.5 MDa (for review see (Wilson and Nierhaus 2007; Schmeing and Ramakrishnan 2009)). The macromolecu-lar nature of the ribosome is evident from the diversity of its RNA and pro-tein components as discussed in the section below. The ribosomes are lo-cated in the cytoplasm; in eukaryotes they are also found on endoplasmic reticulum. Ribosomes translate the genetic information encoded in a mes-senger RNA (mRNA) into the corresponding sequence of amino acids, the-reby synthesizing a polypeptide chain or a protein. In Escherichia coli (E. coli) cells, ribosomes constitute about 50% of the total dry cell mass during exponential growth, their number varies from 40000 – 70000 per cell.

Basic components of the ribosome The ribosome is composed of ribosomal proteins (r-proteins) and ri-

bosomal RNAs (rRNAs) arranged in two subunits, the large subunit is about twice the size of the small subunit. The ribosomes, ribosomal subunits and rRNAs are called by their sedimentation coefficient (S). The intact ribosome is called 70S in prokaryotes and 80S in eukaryotes. In prokaryotic and euka-ryotic ribosomes the large subunits are called 50S and 60S, and the small subunits are called 30S and 40S, respectively. These ribosomes, although very similar in structural organization, differ in their protein and RNA com-ponents. In the prokaryotic 70S ribosome, the large subunit (50S) contains two RNAs, 23S rRNA (~2900 nucleotides) and 5S rRNA (~120 nucleotides) along with 33 proteins. The small subunit (30S) contains only one RNA component called the 16S rRNA (~1500 nucleotides) and about 21 proteins. Eukaryotic ribosomes are larger and much more complex than the prokaryo-tic ribosomes. The large subunit in eukaryotes is composed of 28S rRNA (~4700 nucleotides), 5S rRNA (~120 nucleotides) and an additional 5.8S rRNA (~160 nucleotides) together with 49 r-proteins. In contrast, the small subunit (40S) is composed of 18S rRNA (~1900 nucleotides) and 33 proteins (for review see (Wilson and Nierhaus 2003; Marintchev and Wagner 2004)).

12

The ribosomes contain around two- thirds of RNA and one third of protein. The high RNA content is attributed to the fact that the main functions of the ribosome i.e. mRNA decoding and peptide bond formation are driven by rRNA. In each ribosomal subunit rRNA makes the back bone or the core to which the r-proteins attach. The r-proteins often have distinct functions, yet in general, they play an important role in maintenance of the structural inte-grity of the rRNA core.

Ribosomal subunits are recycled and reused through the cycles of translation. During translation initiation, the subunits associate via interac-tions along their interface to form the 70S ribosome. However, after comple-tion of synthesis of a peptide the ribosomes split into the individual subunits, thus completing the cycle. The two subunits perform different roles in pro-tein synthesis. The 30S subunit participates in decoding the information on mRNA and the 50S subunit catalyzes the peptide bond formation. As men-tioned above, both functions are driven by RNA.

Structural landmarks on the bacterial ribosome Each ribosomal subunit has a unique structure with some ‘landmark’

features on them. The 30S subunit has a relatively simple structure. The up-per part of 30S is called the ‘head’, which is connected through a ‘neck’ to the lower part or ‘body’ carrying a shoulder (Figure 1). The 50S subunit presents a compact structure consisting of a hemisphere base with three pro-tuberances on it, named L1 stalk, central protuberance (CP) and the L7/L12 stalk (Figure 1). The L1 stalk is composed of L1 protein and its 23S rRNA binding region. The L7/L12 stalk, also commonly called the ‘ribosomal stalk’ contains multiple copies of L7/L12 protein (referred hereafter as only L12) and one L10 protein bound to the rRNA next to the binding site of another r-protein L11. Protein L11 constitutes the ‘stalk base’, which is a part of an important region of the ribosome called GTPase associated center (GAC). The 50S subunit also contains a tunnel through which the nascent peptide is believed to exit to the cytosol. The tunnel is about 100 Å long and up to 25 Å wide in diameter, it can accommodate a peptide of approximately 40 amino acids in length (Yonath, Leonard et al. 1987; Nissen, Hansen et al. 2000).

13

Figure 1: Structural landmarks on ribosomal subunits.

Functional sites on the ribosome

tRNA binding sites Both the ribosomal subunits have three distinct tRNA binding sites

named as A-site, P-site and E-site (Rheinberger, Sternbach et al. 1981). The A-site accommodates the aminoacyl tRNAs, the P-site holds the peptidyl tRNAs after translocation and the E-site binds nonspecifically the deacylated tRNAs (Figure 2).

A-site is the binding site for the incoming aminoacyl-tRNAs (aa-tRNA). The codon anticodon interactions are monitored in the A-site at the decoding center (DC) on the 30S subunit, this allows mostly cognate and some near cognate tRNAs to bind to the mRNA (for review see (Wilson and Nierhaus 2003)). The A-site is also the binding site for many translation factors during various steps of translation. The P-site is located next to the A-site on the ribosome. During initiation of protein synthesis, initiator tRNA (fMet-tRNAfMet) binds directly to the P-site on the 30S subunit. Other than that, peptidyl tRNAs occupy the P-site after translocation until the next pep-tidyl transfer. The deacylated tRNAs when translocated from the P-site move to the E-site. It has been suggested that binding of aminoacyl tRNA in the A-site decreases the affinity of deacylated tRNA in the E-site thereby trig-gering tRNA release from the E-site (Rheinberger and Nierhaus 1986). Thus, tRNA binding to A-site and E-site are possibly regulated in a coordinated manner.

14

Figure 2: Thermus thermophilus 70S ribosome with E, P and A site tRNAs, adopted from (Selmer, Dunham et al. 2006). DC is the decoding center, PTC is the peptidyl transferase center and GAC is the GTPase associated center.

Decoding center (DC) The decoding center (Figure 2) is located at the A-site of the 30S sub-

unit which includes the helix 44 and 34 and loop 530 of 16S rRNA (Schluenzen, Tocilj et al. 2000). Universally conserved bases A1492, A1493 and G530 of 16S rRNA are the important components at the decoding center (Ogle, Brodersen et al. 2001). The process of decoding is explained in the later part of the thesis.

Peptidyl transfer center (PTC) The peptidyl transfer center or PTC is the catalytic heart of the 50S

subunit which is located near the central protuberance and it is almost exclu-sively composed of RNA (Figure 2). N-terminal tail of L27 is the only pro-tein part that is present close to the PTC. The central loop of domain V of 23S rRNA is involved in the catalytic act of PTC, this loop is also known as the “PTC ring”. The major catalytic activity of PTC is the peptide bond for-mation during the elongation of protein synthesis. PTC is also involved in the release of the polypeptide chain during the termination of protein synthe-sis.

GTPase associated center (GAC) The GTPase associated center or GAC (Figure 2) is known to stimu-

late the GTPase activity of several G-protein factors during translation. GAC is composed of helices 42-44 located at the domain II of 23S rRNA and the protein L11 (Li, Sengupta et al. 2006). Protein L10 and L11 although bind to the same rRNA region but differ in their binding site, L10 interacts with helices 42 and 43, whereas L11 binds to helices 43 and 44. The GAC togeth-er with helix 95 or sarcin resin loop (SRL) of 23S rRNA are responsible for the binding of all translational GTPases and their GTPase activation.

15

Ribosomal stalk Ribosomal stalk is a finger like protuberance on the large subunit of

the ribosome (Figure 3) (Strycharz, Nomura et al. 1978; Agrawal, Heagle et al. 1999) (for review see (Chandra Sanyal and Liljas 2000)), composed mainly of two to three dimers of L12 protein associated with one L10 pro-tein. In E. coli, two dimers of L12 bind to one L10 protein (Pettersson and Liljas 1979) and this pentameric complex is known as the L8 complex. L12 is the only multicopy protein on the ribosome (Hardy 1975). The other unique feature it possesses is that it has no direct contact with rRNA. L12 proteins bind to rRNA via C terminal domain of the L10 protein. During translation, four translational G-factors (IF2, EF-Tu, EF-G, and RF3) interact with the ribosomal stalk, which is closely associated with a special region on the ribosome, known as the GTPase associated center (GAC). This region also includes the r-proteins L10 and L11 bound on a stretch of rRNA (nt 1028-1124) located at the base of the stalk (Schmidt, Thompson et al. 1981; Beauclerk, Cundliffe et al. 1984).

Figure 3: (a) A model of 50S and 30S ribosomal subunit showing relative organiza-tion of L12 dimers, L10 and L11 proteins. I, II, and III indicate alternative locations for the C-terminal domain of L12 on the 50S subunit. The figure is adopted from (Dey, Bochkariov et al. 1998) (b) High-resolution structure of the ribosome showing the components of ribosomal stalk and its base. The image was created in PyMol from the coordinates of T. thermophilus ribosome (PDB id 2WRI and 2WRJ) (Gao, Selmer et al. 2009).

L7/L12 In E. coli, L12 is composed of 120 amino acids with a molecular mass

of 12 kDa. Protein L12 is also known as L7/L12 as it exists in an N-terminal acetylated form named L7, where the Serine residue at the N-terminus is

(a) (b)I

16

modified by acetylation (Terhorst, Moller et al. 1972). The L7:L12 ratio in the cell changes depending on the growth conditions. L12 acetylation in-creases during stress such as when cells are grown in minimal media in nu-trient depleted condition. It is believed that the acetylation of L12 increases the stability of the pentameric complex (L8 complex) (Gordiyenko, Deroo et al. 2008) meaning both dimers of L12 remain stably bound to L10.

Domains in the L12 dimer L12 is one of the best studied r-protein. In general, L12 proteins form

a strong dimer in solution. The L12 dimer consists of two organized domains connected by one flexible hinge region. The N-terminal domain spanning residues 1-37 is responsible for the strong dimer interaction and for binding to the protein L10 (Bocharov, Gudkov et al. 1998). This domain in the L12 monomer contains two α-helices arranged in a hair pin conformation form-ing a four helix bundle in the L12 dimer. The C-terminal domain of L12 is globular, highly conserved and provides sites for interaction with the GTPase factors (Helgstrand, Mandava et al. 2007). This compact domain consists of 3 α-helices and 3-β sheets spanning the residues 50-120. The N and C-terminal domains are connected by the hinge region which includes the residues from 38 to 49 (Pettersson, Hardy et al. 1976; Gudkov 1997; Mulder, Bouakaz et al. 2004). The hinge region is highly flexible and can change conformation from a compact helix to an extended structure. The flexible nature of the hinge is very important for the function of L12. Due to the flexible hinge the CTDs of L12 can be seen at different locations on the ribosome (Figure 3a)(Olson, Sommer et al. 1986; Zecherle, Oleinikov et al. 1992) and also can be extended out of the ribosome for the recruitment of the translation factors (Liljas and Gudkov 1987; Diaconu, Kothe et al. 2005).

The mode of dimerization There have been several attempts to describe the dimerization mode of

L12 on and off the ribosome. Following the crystal structure of the incom-plete tetramer of L12 from Thermotoga maritima Wahl et al. discussed two possible dimerization modes for L12 (Wahl, Bourenkov et al. 2000); (i) a parallel dimer involving N- terminal helices α2 and α3 (hinge in an α-helical conformation) of two full length molecules arranged side by side (ii) an anti-parallel dimer where α1 and α2 at the N-terminus of a full length L12 mo-nomer is linked to the same helices on the N-terminus of the other (incom-plete) L12 monomer. In the parallel dimer model, the main interaction be-tween the L12 monomers involved the flexible hinge region (α3), compacted as helix (Figure 4a). Thus, this model although favoured by the authors was somewhat unrealistic as the flexible hinge is not always helical in shape. Instead, Sanyal and Liljas proposed the antiparallel dimer of L12 as the func-

17

tional dimer, where one of the two hinge regions form a compact α-helix whereas the other is unstructured and extended (Figure 4b) (Chandra Sanyal and Liljas 2000). Later, the antiparallel mode of dimerization of L12 was confirmed by Moens et al. with FRET assays where the distance between the L12 CTD and NTDs were measured by incorporation of specific fluores-cence probes (Moens, Wahl et al. 2005). The four helix bundle N-terminal dimers, as proposed in the antiparallel dimers, were also seen in the high resolution crystal structure of T. maritima L10-(L12-NTD)6 complex (Diaconu, Kothe et al. 2005).

Figure 4: (a) Arrangement of two full length L12 molecules in parallel dimerization mode involving α2 and α3 helices of the monomers. The image was created in Py-Mol from the coordinates of T. maritima L12 (PDB id 1DD4) (Wahl, Bourenkov et al. 2000) (b) Model of antiparallel L12 dimer involving α1 and α2 helices where one hinge is a helix and the other is an extended structure, adopted from (Chandra Sanyal and Liljas 2000) (c) NMR structure of L12 dimer in solution showing three distinct regions, adopted from (Bocharov, Sobol et al. 2004).

Structure of the L12 dimers off and on the ribosome In 2004, three groups have simultaneously reported the structure and dynamics of L12 dimers in solution based on heteronuclear NMR spectros-copy (Bocharov, Sobol et al. 2004; Christodoulou, Larsson et al. 2004; Mulder, Bouakaz et al. 2004). In all cases, the CTDs were seen in their usual globular form matching the crystal structures (Leijonmarck and Liljas 1987; Wahl, Bourenkov et al. 2000). The NTDs were seen entangled in a four helix bundle, in good agreement with the earlier NMR based observation (Bocharov, Gudkov et al. 1998) Most interestingly, both the hinges were flexible and unstructured, entirely different from the crystal structure of the incomplete tetramer of L12 (Wahl, Bourenkov et al. 2000).

In contrast to L12 dimers in solution, clear signals only from the C-terminal domain of L12 were seen when NMR was done with intact 70S ribosome and 50S subunit from E. coli. Somewhat weaker signals from the hinge residues were also seen while no signal from the N-terminus could be obtained. Thus it was reconfirmed that L12 CTDs are dynamic even on the ribosome and the NTDs are tightly bound to the ribosome. Comparing the

(a) (c)(b)

α1α1

α2 α2α1α2 α1

α2

Flexiblehinge

Helicalhinge

18

peak intensities of the residues from the CTD and the hinge region, the num-ber of the extended L12 molecules per ribosome was quantified (Mulder, Bouakaz et al. 2004). Two out of four L12 monomers were counted as flexi-ble on the ribosome. Based on that, Mulder et al. proposed two models of L12 dimers on the ribosomes (Figure 5). In one model, both L12 dimers were arranged as the typical antiparallel dimer (Chandra Sanyal and Liljas 2000) with one L12 molecule extended and the other in a compact shape. In the other model, one dimer had both the L12 molecules compact as in the crystal structure (Wahl, Bourenkov et al. 2000) while the second dimer had both the molecules in extended shape (Mulder, Bouakaz et al. 2004). Further experiments will be needed to conclude which model is more correct and if both the models can exist under special conditions.

Figure 5: Two models of L12 dimers bound on the ribosome. (a) Both L12 dimers have one extended and one retracted CTD. (b) One L12 dimer has both CTDs ex-tended, while the other dimer has both CTDs retracted, adopted from (Mulder, Bouakaz et al. 2004).

Stoichiometry of the L12 proteins on ribosome The numbers of L12 molecules on the ribosome varies in different

bacteria. In E. coli they are four in number, which exist as two dimers (Subramanian 1975) forming the pentameric L8 complex together with L10. In contrast, in thermophiles and archaebacteria, they are six in number form-ing three dimers on the ribosome (Diaconu, Kothe et al. 2005; Ilag, Videler et al. 2005; Maki, Hashimoto et al. 2007). The length and sequence of L10 α8 helix determines the number of L12 molecules per ribosome. In recent studies with mass-spectroscopy, Gordiyenko et al. have observed that the stoichiometry of the L12 proteins in the stalk complexes (pentametic or hep-tameric L8 complexes) changes in different growth stages of archaebacteria (Gordiyenko, Videler et al. 2010). Putting the evidence together one can conclude that there exists a strong correlation between the number of L12 dimers on the ribosome and the bacterial growth rate.

(a) (b)

19

The binding site of the L12 dimers It has been known for a long time that the L12 dimers are associated

with the ribosome by binding to the C-terminal end of the L10 protein. By serial deletion from the C-terminal end of L10, Griaznova et al. suggested that approximately the last 10 amino acids constitute the binding site of one L12 dimer and the next 10 amino acids, although different in sequence con-stitute the binding site of the second L12 dimer on the E. coli ribosome (Griaznova and Traut 2000). It was unclear how two identical L12 dimers could bind to the unidentical sequences on the two binding sites. Much later, when high-resolution crystal structure of L10-(L12-NTD)6 complex from T. maritima was solved the earlier finding about the L12 dimer binding site was confirmed. It turned out that all L12 dimers bind to the long α8 helix at the C-terminal end of L10 (Figure 6). These sites although in a different bacte-ria, spanned approximately 10 amino acids each. It was proposed that hydro-phobic interaction plays main role in the association of the L12 dimers on L10. We have utilized this knowledge about the L12 dimer binding sites in construction of homogeneous ribosomes containing only one L12 dimer.

Figure 6: Model of L10 and L12 complex, shows the binding of L12 NTDs on α8 helix of L10, and also the binding of two full length L12 molecules with one ex-tended hinge and other with compact (helical) hinge. The image was created in Py-Mol from the coordinates T.maritima L10 and L12 NTD complex and E. coli L12 dimer (PDB id 1ZAV and 1RQV) (Bocharov, Sobol et al. 2004; Diaconu, Kothe et al. 2005).

20

The factor interaction site on L12 The sites involved in translation factor interaction are located on the globular CTD of L12 molecules. Earlier mutational analysis identified some key residues on L12 CTD (K65, V66, K70, R73 and K84), which affected the functions of elongation factors EF-Tu and EF-G (Kothe, Wieden et al. 2004; Savelsbergh, Mohr et al. 2005). Many of these sites coincided with those identified later by means of heteronuclear NMR (Helgstrand, Mandava et al. 2007). Most interestingly, these residues were concentrated on two adjoined helices at the L12 CTD, and were common for all four major trans-lation factors IF2, EF-Tu, EF-G and RF3. From this observation, it was sug-gested that L12 CTDs interact with the translation factors in a general man-ner.

Other than these direct studies, the L12 CTDs were seen attached to EF-G and IF2 in the cryo-EM images of the complexes (Agrawal, Penczek et al. 1998; Allen, Zavialov et al. 2005). The recent high resolution crystal structure of 70S with EF-G bound on it showed one L12 CTD interacting directly with EF-G on the ribosome (Gao, Selmer et al. 2009). These obser-vations point towards the functional significance of L12 CTD and translation factor interaction.

Role of L12 in translation Protein L12 plays a key role in protein synthesis, L12 protein on ribo-some is needed for translation initiation (Kay, Sander et al. 1973; Huang, Mandava et al. 2010), elongation (Brot, Yamasaki et al. 1972; Agrawal, Penczek et al. 1998) and termination (Brot, Tate et al. 1974) of protein syn-thesis. During these steps L12 interacts with different GTPase factors (IF2, EF-Tu, EF-G and RF3) via its conserved C-terminal domain (Helgstrand, Mandava et al. 2007). It has been reported that both L12 dimers are required for the maximal rate of protein synthesis and minimal error frequency of protein synthesis (Pettersson and Kurland 1980) and in their absence the binding of elongation factors to the ribosome as well as GTP hydrolysis by the translation factors are severely impaired (Leijonmarck and Liljas 1987; Mohr, Wintermeyer et al. 2002). Two mutations in E. coli L12 CTD (G74D and E82K) led to decreased growth rate and increased error frequency in vivo (Kirsebom, Amons et al. 1986). Ribosomes from two missense suppres-sor mutant (mutations in rplLgene) strains (UY134 and UY143) which have altered L12 protein showed reduced translation efficiency in vitro (Kirsebom and Isaksson 1985; Bilgin, Kirsebom et al. 1988). Ribosomes with a deletion in the hinge region of L12 have shown reduced activity in the Poly U trans-lation(Gudkov, Bubunenko et al. 1991), implying that the hinge region in L12 protein is very important for the mobility of the conserved globular C-terminal domain.

21

Two best studied functions of L12 are GTPase activation and factor recruitment. It has been shown that these proteins are necessary on the ribo-somes for the stimulation of the GTPase factors, particularly for EF-Tu and EF-G (Hamel, Koka et al. 1972; Savelsbergh, Mohr et al. 2000; Mohr, Wintermeyer et al. 2002) and such activation is hindered by the mutations on the factor interaction site of L12 CTD (Kothe, Wieden et al. 2004; Savelsbergh, Mohr et al. 2005). Also, free L12 protein at high concentrations could stimulate GTPase activity of EF-G but not of EF-Tu. (Savelsbergh, Mohr et al. 2000). Recently, using fast kinetics we have shown that unlike EF-G, the L12 protein cannot stimulate the GTPase activity of IF2, irrespec-tive of whether it is in free state or in a functional ribosomal complex (Huang, Mandava et al. 2010).

It has also been suggested that highly mobile CTDs of L12 promote the recruitment of the G-factors, EF-Tu in ternary complex and EF-G, onto the ribosome (Diaconu, Kothe et al. 2005). Our experiments with IF2 show that L12 is needed for the recognition of IF2 on 30S preinitiation complex which plays a key role for rapid subunit association (Huang, Mandava et al. 2010). Thus, although different translation factors interact on a common site on the L12 CTD, it seems that L12 has specialized functions with different G-factors. The role of L12 proteins in RF3 function remains to be tested.

Translation Translation is the process in which ribosomes translate the genetic in-

formation from mRNAs into proteins. In bacteria, translation occurs in four steps, initiation, elongation, termination and recycling. All these steps are guided by different protein factors in coordination with the ribosome. The ribosome has been discussed at the beginning of this thesis. Below I describe other components involved in translation.

Translation components

mRNA (Messenger RNA) The messenger RNA (mRNA) carries the genetic information from DNA to the ribosome in three base long genetic codes, also called codons. In bacteria, mRNAs are polycistronic and contain multiple signal sequences for translational initiation and termination (Laursen, Sorensen et al. 2005). During translation, mRNAs interact with both tRNA and 30S ribosomal subunit.

Bacterial mRNA exists mainly in two forms, leadered mRNA and lea-derless mRNA. Canonical mRNAs start with a 5’ untranslated region (5’ UTR), which contains a ribosome binding site called Shine-Dalgarno se-

22

quence (SD sequence). The SD sequence is normally located six to nine nucleotides upstream of the translational initiation codon AUG. It interacts with the anti-SD sequence (aSD) located at the 3’ end of 16S rRNA of the 30S subunit (Shine and Dalgarno 1974). Leaderless mRNAs start either di-rectly from the initiation codon or a few nucleotides upstream from the initi-ation codon. These mRNAs are recognized by the 30S subunit, initiation factor IF2 and initiator tRNA (fMet-tRNAfMet) complex. It was shown that initiation factors IF2 and IF3 regulate the translation efficiency of the leader-less mRNA (Grill, Moll et al. 2001; Moll, Grill et al. 2002). In E. coli, trans-lation starts mainly with AUG initiation codon (90%) although GUG (8%) and UUG (1%) codons can also act as the initiation codons in some cases (Schneider, Stormo et al. 1986). The efficiency of translation initiation is highly influenced by the strength of the SD sequence (determined by the number of paired bases between SD and aSD sequence) as well as the spac-ing between the SD sequence and the initiation codon. The mRNA with a strong SD sequence (7 - 8 base complementarity) initiates translation four times more efficiently than that with a weak SD sequence (3 – 4 bases). (Ringquist, Shinedling et al. 1992). The XR7 mRNAs used in our experi-ments contain a strong SD sequence (7 base complementarity) and six nuc-leotide long spacing between the SD-sequence and the initiation codon.

tRNA (transfer RNA) The tRNAs are the adapter molecules that deliver the amino acids to the ribosome during the elongation of the polypeptide chain. There are 41 different types of tRNAs, which occur in the cell in different frequencies and carry 20 amino acids in a highly specific manner. The tRNAs carrying the same amino acid are called tRNA isoacceptors, which are recognized by an aminoacyl tRNA synthetase (aaRS) that links the corresponding amino acid to the universally conserved 3’ CCA end in an ATP dependent manner. This reaction is called aminoacylation or charging of the tRNA (Arnez and Moras 1997; Ibba and Soll 2000). The accuracy of the translation is highly depen-dent on the correct matching of the amino acid to the cognate tRNA. The tRNAs carry anticodons on a loop called the anticodon stem loop (ASL). The anticodons, usually complementary to the codons let the tRNAs bind specifically to the mRNA by Watson-Crick base pairing. In a cognate codon – anticodon interaction, the first two positions always complement each other. Compared to that near-cognate (one of first two bases do not match) and non-cognate (more than one base mismatch) are much weaker. The correct codon anticodon interaction is important for adding the correct amino acid in the chain.

Among all tRNAs, the initiator tRNAs have some special features, which are important for maintenance of the fidelity in translation initiation. These tRNAs are charged with methionine in all the organisms, in proka-ryotes the methionine is formylated. Initiator tRNA is the only tRNA which

23

can directly bind to the P site on the ribosome, three continuous G-C base pairs at the bottom of the anticodon stem facilitates this binding (Varshney, Lee et al. 1993; RajBhandary 1994).

Translation factors Each step of the translation in bacteria is carried out with the help of several protein factors. Initiation factors IF1, IF2 and IF3 are involved in the initiation of protein synthesis to form the 70S initiation complex. Elongation factors EF-Tu, EF-Ts and EF-G are needed for the elongation of protein synthesis. Release factors RF1, RF2 and RF3 regulate the steps in termina-tion of the translation. Finally, ribosome release factor (RRF) along with EF-G catalyzes the recycling of the ribosomal subunits. The function of the translation factors has been discussed in detail in a later section. Besides these commonly known translation factors, there are some other factors too such as EF-P, which is thought to be important for the formation of the first peptide bond (Aoki, Xu et al. 2008) and lepA, which is involved in back-translocation (Qin, Polacek et al. 2006).

IF2, EF-Tu, EF-G and RF-3 are the four GTPase proteins which are essential for the rapid protein synthesis in bacteria (Zavialov and Ehrenberg 2003). These GTPases belong to a large class of G proteins which switch between the active GTP and inactive GDP forms and participate in different cellular functions (Bourne, Sanders et al. 1991). The low intrinsic GTPase activity of these G proteins gets strongly stimulated upon binding to the GAC on the ribosome ((Rodnina, Stark et al. 2000), for review see (Ramakrishnan 2002)).

The steps in protein synthesis

Initiation After splitting of the post termination complex, IF3 rapidly binds to

the 30S subunit and promotes the dissociation of deacylated tRNAs and mRNA from it (Karimi, Pavlov et al. 1999; Peske, Rodnina et al. 2005). It also blocks premature subunit association, this anti association property of IF3 increases in the presence of IF1 (Grunberg-Manago, Dessen et al. 1975). The initiation step starts by the binding of a new mRNA to the 30S subunit (Antoun, Pavlov et al. 2006a), where SD - aSD interaction play a major role (Shine and Dalgarno 1974; Yusupova, Yusupov et al. 2001). Next, with the help of the initiation factors, mRNA places the initiation codon AUG into the P-site (La Teana, Gualerzi et al. 1995). Then the initiator tRNA (fMet-tRNAfMet) binds to the AUG codon with the help of IF2. Whether or not the initiator tRNA binds to the ribosome in complex with IF2·GTP has been debated for a long time (Gualerzi and Pon 1990; Wu and RajBhandary 1997). Recent kinetic studies have resolved the issue confirming that

24

IF2·GTP helps in recruitment of the initiator tRNA, but doesn’t bring it to the ribosome in a complex (Milon, Carotti et al. 2010). The speed of associa-tion of the initiator tRNA to the 30S subunit as well as the accuracy in initia-tion increases significantly in the presence of all three initiation factors (Antoun, Pavlov et al. 2006b). The whole complex with mRNA, initiator tRNA and three initiation factors is called the 30S preinitiation complex. Association of 50S subunit to 30S preinitiation complex leads to the formation of 70S initiation complex. IF2 in GTP form is needed for fast as-sociation of the subunits (Grigoriadou, Marzi et al. 2007a). We have recently shown that L12 proteins on the 50S subunit recognize IF2·GTP on the 30S preinitiation complex, a reaction that leads to rapid subunit association (Huang, Mandava et al. 2010).

There are two different opinions about the timing and role GTP hydro-lysis during translation initiation. According to Antoun et al., IF2 in GTP form is needed for fast subunit association, and GTP hydrolysis is needed for the dissociation of IF2·GDP from the 70S initiation complex (Antoun, Pavlov et al. 2003). In contrary, Grigoriadou et al. claimed that GTP hydro-lysis is required for stable subunit association (Grigoriadou, Marzi et al. 2007a). Our recent experiments with two IF2 G-domain mutants (V400G and H448E), which are impaired in GTP hydrolysis (Luchin, Putzer et al. 1999), did not show any defect in the rate of subunit association when com-pared with wild type IF2 (Figure 16a). This result is in accordance with An-toun et al. and confirmed that GTP hydrolysis is not needed for subunit asso-ciation. However, dipeptide formation was really slow, almost insignificant, when 70S initiation complex was formed with these two mutant IF2s. Thus, GTP hydrolysis is needed for the release of IF2 from 70S initiation complex. During the formation of 70S initiation complex all three initiation factors dissociate in various sequences getting it ready for the elongation step.

Elongation Elongation of protein synthesis starts with an empty A site and fMet-tRNAfMet in the P site of the 70S initiation complex. The two crucial steps in elongation are mRNA decoding and peptide bond formation. Elongation factors EF-Tu, EF-Ts and EF-G are the three main elongation factors which facilitate this step. EF-Tu brings an aminoacyl-tRNA (aa-tRNA) to the ribosomal A site in a ternary complex with GTP (Stark, Rodnina et al. 1997). An initial selec-tion governed by correct codon anticodon pairing (decoding) stabilizes the aa-tRNA on the ribosome and triggers GTP hydrolysis by EF-Tu (for review see (Ramakrishnan 2002)). After EF-Tu·GDP leaves the complex, the aa-tRNA repositions itself - a process called tRNA accommodation. This step is followed by rapid peptide bond formation between the amino acids on the tRNAs at the P site and A site. The peptidyl tranferase reaction is catalyzed

25

mainly by rRNA, the domain V of 23S rRNA being identified as the active site for this reaction. After peptidyl transfer, the elongation step continues with tRNA trans-location, a step driven by EF-G, when both deacylated and peptidyl tRNAs move on the mRNA by one codon. In the beginning of translocation the aminoacyl end of the tRNAs move with 50S subunit, thereby creating E/P and P/A hybrid states of tRNAs. The ribosome also goes in the ratcheted stage, with two subunits rotated with respect to each other. In the next step, EF-G·GTP catalyzes the movement of the ASL end of the tRNAs attached to the mRNA with respect to 30S (Moazed and Noller 1989). This leads to the post translocation state of the ribosome with an empty A site and places the peptidyl and deacylated tRNAs in the classical P/P and E/E state respectively for the entry of the next aa-tRNA. GTP plays a vital role in EF-G function. However, whether GTP hydrolysis is essential for tRNA translocation or not remains an open question (Rodnina, Savelsbergh et al. 1997; Zavialov, Hauryliuk et al. 2005).

Elongation factor EF-Ts, a guanine nucleotide exchange factor (GEF) exchanges the GDP on EF-Tu to GTP, which in turn forms a ternary com-plex with aa-tRNA. The EF-Tu ternary complex reinitiates another round of elongation. The elongation step repeats until a stop codon on the mRNA reaches the A site of the ribosome.

Termination In bacteria, there are three stop codons UAA, UAG and UGA, which

are recognized by two class-I release factors RF1 and RF2 in a semispecific fashion. Primarily RF1 recognizes UAG and RF2 recognizes UGA codon, while both factors recognize UAA codon (Kisselev and Buckingham 2000). These two release factors mimic the aa-tRNA in structure extending from the DC on the 30S subunit to the PTC on the 50S subunit. RF1 or RF2 binds to the A site and promotes the hydrolysis and release of nascent peptide from the peptidyl tRNA in the P site. The GGQ (Gly-Gly-Gln) motif of the class I release factors interacts with the PTC on the 50S subunit (Zavialov, Mora et al. 2002). Similarly SPF motifs have been identified on these for recognition of the stop codons (Ito, Uno et al. 2000). Once the peptide is released, the class II release factor RF3 prompts the release of RF1 and RF2 from the A-site. RF3 enters the ribosome with GDP and rapidly exchanges it to GTP (Zavialov, Buckingham et al. 2001). GTP binding on RF3 induces conforma-tional changes on the ribosome which break the interaction of class I release factors with the ribosome (Gao, Zhou et al. 2007). GTP hydrolysis on RF3 triggers its release from the ribosome (Zavialov, Buckingham et al. 2001).

Recycling Recycling begins with the post termination complex with a deacylated

tRNA in the P-site. Ribosome recycling factor (RRF) and EF-G modulates

26

the recycling step through many cycles of GTP hydrolysis (Pavlov, Antoun et al. 2008). The main event in recycling is the splitting of the ribosomal subunits, which is followed by the binding of the anti-association factor IF3 on the 30S subunit (Karimi, Pavlov et al. 1999; Hirokawa, Nijman et al. 2005).

The main steps of translation are summarized in the following scheme.

Figure 7: Overview of translation cycle, showing initiation, elongation, termination, and recycling.

27

Aim

The aim of this thesis is to determine the role of L12 in translation, both from functional and structural angles. In this thesis the following topics are discussed

1. Ribosomal stalk protein L12 has been implicated as a key player in the interaction of the translation factors; in Paper-I, with the help of NMR, we have mapped the interaction sites of IF2, EF-Tu, EF-G and RF-3 on the CTD of L12.

2. The role of L12 with respect to the function of EF-Tu and EF-G has been explored by other groups. In Paper-II, with fast kinetic experi-ments, we have discussed the role of L12 in translation initiation.

3. The numbers of L12 dimers on the ribosome vary in different bacteria (two to three); in Paper-III we have shown why more than one L12 di-mer is required on the bacterial ribosome.

4. Technological application with the stalk protein L12; in Paper-IV we have described a single step method for purification of active ribosomes carrying (His)6-tag on the CTD of L12.

Interaction of L12 CTD with translational G-factors The ribosomal stalk protein L12 is thought to be involved in transla-tion initiation (Kay, Sander et al. 1973), elongation (Brot, Yamasaki et al. 1972) and termination (Brot, Tate et al. 1974) of protein synthesis by inte-racting with different GTPase factors. Although not fully-evidenced, L12 proteins are thought to play important role in activation of translational GTPases as well as in their recruitment (Savelsbergh, Mohr et al. 2000; Mohr, Wintermeyer et al. 2002; Diaconu, Kothe et al. 2005). Using hetero-nuclear NMR spectroscopy we have shown that L12 directly binds to the factors IF2, EF-Tu, EF-G and RF-3 and we have mapped the sites on L12 which are involved in these interactions. Proteins with high purity and activity were used to investigate the interaction site between L12 and translation factors IF2, EF-Tu, EF-G and RF3. Each unlabelled translation factor was titrated separately into a solution of 15N- labeled L12, and the 1H-15N heteronuclear single-quantum coherence (HSQC) spectrum of the L12 residues were noted at each titration point. The

28

back bone amide group of each residue in 15N- labeled L12 protein corres-ponds to a single cross-peak in the 1H-15N HSQC spectrum, so residue spe-cific information can be obtained from the chemical shift and line width of the cross-peaks. By monitoring changes in the chemical shifts and line widths of the cross peaks in the HSQC spectrum of L12 by addition of the translation factors, the interaction site of individual translation factor on L12 was identified.

Upon factor binding, NMR signals from the CTD residues of L12 broadened significantly while the signals from the NTD of L12 remained unchanged confirming the fact that the factors interact with the CTD and not with the NTD of the L12 proteins. In addition, peak shift was seen in some specific residues on the CTD, surprisingly common for all four factors. The following figure (Figure 8) shows the chemical shift of the residues from the CTD of L12 upon IF2 (K70 and L80) and EF-Tu (V66 and E82) binding. Right side panel shows cross peaks from two residues (M13 and K108) which did not change upon factor binding.

Figure 8: Close-up views of selected cross-peaks in superimposed 1H-15N HSQC spectra of the L12 dimer, obtained at different equivalents (molar ratios), x, of added factor with respect to L12 dimer. Two representative cross-peaks exhibiting signifi-cant chemical shift changes are shown for each titration series (left-hand and middle panels), together with two representative cross-peaks, M17 and K108, that are unaf-fected by factor addition (right-hand panel). (a) Addition of IF2, showing residues K70 (left) and L80, (middle); x = 0, 0.4, 0.8, 1.2, 1.6. (b) Addition of EF-Tu, show-ing residues V66 (left) and E82 (middle); x = 0, 0.1, 0.2, 0.3, 0.4. The cross-peak contours are colored according to the increasing molar ratio of the added factor in the order yellow, red, brown, black.

Thus our data indicated that all the four major GTPase factors (IF2, EF-Tu, EF-G and RF3) interact essentially with the same site on the con-served CTD of the L12 protein. This region includes three strictly conserved

(a)

(b)

29

residues K70, L80 and E82 and some highly conserved residues, including V66, A67, V68 and G79 placed on two adjacent α-helices on the L12-CTD (Figure 9). Based on this result, we have concluded that CTD of L12 directly interacts with translational G-factors (IF2, EF-Tu, EF-G and RF3) following a general mechanism.

Figure 9: Binding surfaces of (a) IF2, (b) EF-Tu, (c) EF-G, and (d) RF3 on the CTD of L12. The residues forming the factor-binding surface are colored in red and la-beled with residue number. The figure is adopted from (Helgstrand, Mandava et al. 2007).

Role of L12 in translation initiation L12 interaction with the elongation factors EF-Tu and EF-G have been studied in the last decade in some detail (Savelsbergh, Mohr et al. 2000; Mohr, Wintermeyer et al. 2002). However, how the L12 proteins in-fluence the function of the other two translational GTPases IF2 and RF3 is quite unknown. In Paper-II we have investigated the role of L12 in IF2 me-diated initiation of bacterial protein synthesis. During protein synthesis IF2 in GTP form plays important role in subunit association (Antoun, Pavlov et al. 2003). To study the role of L12 protein in IF2 mediated initiation steps, we have compared L12 depleted 50S subunits (Hamel, Koka et al. 1972; Mohr, Wintermeyer et al. 2002) with native 50S subunits in different steps of initiation using an in vitro transla-tion system with all translation components purified from E. coli. As a con-trol, we have used L12 reconstituted 50S in the similar reactions.

Subunit association reaction, monitored by Rayleigh light scattering in a stopped flow instrument (Antoun, Pavlov et al. 2004), was performed si-multaneously with Pi release or IF2 release, using parallel detectors equipped with suitable cut-off filters. The depletion of L12 from 50S did not affect the association of naked subunits, indicating that L12 proteins do not constitute a structural component of 50S subunit that is needed for its associ-ation to 30S subunit. The addition of mRNA, initiator tRNA, and initiation factors IF1 and IF3 to 30S other than IF2 one at a time or in different combi-nation influenced the rate of subunit association, but did not show any dif-ference between the native and L12 depleted 50S. From these observations we concluded that L12 proteins do not have any functional or structural inte-ractions with mRNA, initiator tRNA, IF1 and IF3. The presence of IF2-GTP

30

on 30S preinitiation complex resulted in a fast rate of subunit association, ka

= 130±10 µM-1s-1 for native 50S. Upon L12 depletion from the 50S subunit, this has decreased ~40 -fold giving rise to a rate constant of about 3.5 µM-1s-

1, this defect can be restored just with the reconstitution of L12 on to 50S subunit (Figure 10).

Figure 10: Subunit association with different 50S subunits. The time course of asso-ciation of normal (black trace), L12-depleted (red trace) and L12 reconstituted (green trace) 50S subunit (0.5 μM), with (a) naked 30S subunit (0.5 μM), (b) 30S preIC with IF2 but without IF3, was followed by the increase in light scattering at 430 nm. The blue traces in (b) reflect association of the subunits in the absence of IF2 in otherwise identical conditions.

L12 protein by itself could not activate GTP hydrolysis of IF2 even at very high concentrations. However, as in the case of EF-Tu, the presence of L12 on naked 70S was essential for ribosome mediated GTPase activation of IF2. In this paper we show that L12 depleted 70S is primarily defective in IF2 binding, which in turn affects the subsequent steps i.e. ribosome stimu-lated GTP hydrolysis and Pi release even in translation uncoupled reactions. The association of 30S preinitiation complex with 50S is dependent on the presence of GTP on IF2. When GTP is removed or replaced with GDP, the rate of association with native 50S subunits decreases significant-ly. However, L12 depleted 50S core could not differentiate between GTP or GDP or no nucleotides present with IF2. Moreover, the slow rate of subunit association with L12 depleted 50S subunit matched closely with that ob-tained without IF2. These results demonstrate that rapid subunit association depends upon the specific interaction between the L12 protein on the 50S subunit and IF2·GTP on the 30S subunit.

We have also measured the individual rates of GTP hydrolysis, Pi re-lease and IF2 release with native and L12 depleted 50S subunits. In all these measurements the apparent (or observed) rates for L12 depleted samples were about 40 times slower than the native ones. However when actual rates of GTP hydrolysis, Pi release and IF2 release were estimated taking into account the average time spent in the steps before, there was almost no dif-ference between native and L12 depleted 50S reactions (Figure 11). Thus,

(a) (b)

31

we conclude that L12 protein is not a GTPase activating protein (GAP) for IF2 (Huang, Mandava et al. 2010), but is necessary for the recognition and interaction of IF2-GTP on 30S, that promotes rapid subunit association. In this respect L12-IF2 interaction is different from those between L12- EF-G and EF-Tu as suggested in the earlier reports.

Figure 11: Kinetic scheme of the steps in translation initiation for native and L12-depleted 50S subunits. kobs is the observed rate. taverage (average time spent in a step)=1/kobs (this step)−1/kobs (previous step) and estimated rate kest=1/taverage. Sym-bols used for L12-depleted 50S subunit are k′obs, k′est and t′average, respectively; and shown in red in the cartoon.

Why do bacteria need multiple copies of L12 protein on ribosome?

The numbers of the L12 molecules present on the ribosome are differ-ent in different organisms. For example in E. coli they are four in number and exist as two dimers (Subramanian 1975) but in thermophiles such as Thermatoga maritima they are six and exist as three dimers on the ribosome (Diaconu, Kothe et al. 2005). In all cases the L12 dimers bind on a C termin-al α8 helix on the protein L10. It has been reported that the binding sites of different L12 dimers are different in their amino acid sequence, the signific-ance of which is unknown. In this report (Paper III) we describe the con-struction and characterization of an E. coli strain JE105, which produces

32

ribosomes with single L12 dimer. The strain was created by deletion of 30 nucleotides from the 3’ end of the rplJ gene from the chromosome of the wild type strain MG1655 and substituted with an ampicillin resistance cas-sette (Amp) by means of lambda red recombineering (Figure 12) (Yu, Ellis et al. 2000; Costantino and Court 2003).

Figure 12: The chromosomal replacement of the last 30 nucleotides of rplJ gene with Amp gene in wild type MG1655 and the resulting JE105 strain (left side), car-toon of L8 complex (right side) from both MG1655 and JE105 strains.

This strain named JE105 holds only one L12 dimer on its ribosomes as checked by 2D gel and quantitative Western blot. It has been reported that ribosomes with single L12 dimer are active in poly U dependent poly Phe synthesis and EF-G dependent GTP hydrolysis (Griaznova and Traut 2000). In that study a plasmid based construct of L12 and truncated L10 were used, which were reconstituted on L8 depleted ribosomal cores. The in vivo situa-tion for the single L12 dimer ribosomes was missing in that study, and the depletion and reconstitution may lead to heterogeneity in the ribosome popu-lation. JE105 strain showed significant growth defect in LB media at 37oC. Further, the single L12 dimer ribosomes were also found defective in syn-thesis of a full-length protein in a cell-free translation system. These ribo-somes were also characterized with fast kinetics assays.

In our previous work we have reported that L12 protein on 50S sub-unit is not essential for naked subunit association (Huang, Mandava et al. 2010). Similar to that, the single L12 dimer 50S subunits produced the same rate of association of the naked subunits as the native ones. However, during proper subunit association in presence of mRNA, fMet-tRNAfMet and initia-tion factors IF1 and IF2, single L12 dimer 50S subunits showed two to three fold lower activity than double L12 dimer native 50S (Figure 13a). The dif-ference between the two ribosomes remained unchanged even at high con-centrations of 50S subunit. We have also observed an apparent slower rate of Pi release after subunit association. However, when the actual rate of Pi re-lease was estimated by ‘average time’ analysis (Huang, Mandava et al. 2010) it became clear that the defect in Pi release was a consequence of the defec-tive subunit association (Table I).

33

Table I: Ratio of various translation steps t (average time spent in a step) between the native and single L12 dimer ribosomes (last column of the table). kobs is the observed rate.

In addition we have also checked the dipeptide (ML) and tripeptide (MLL) formation starting from initiation complex to characterize the single L12 dimer ribosomes during translation elongation. The rate of dipeptide formation is similar for both single L12 dimer and normal ribosomes which is 50 ± 10 s-1. In tripeptide formation starting from initiation complex, where sequential steps occur (peptide bond formation between the first two amino acids, translocation of the A site tRNA to P site, and second peptide bond formation between second and third amino acid), the rate of tripeptide for-mation is slightly slower for single L12 dimer ribosomes (3.3 s-1) when compared to normal ribosome’s (5 s-1) (Figure 13 b). As we know that the single L12 dimer ribosomes are not defective in peptide bond formation, the defect that we see in tripeptide formation can be from EF-G binding to post translocation complex or from translocation of the tRNAs.

From the above experiments we can conclude that single L12 dimer ribosomes are defective in IF2 mediated translation initiation and EF-G me-diated elongation.

34

Figure 13: (a) Subunit association with different 50S subunits. The time course of association of normal MRE600 (black trace), single L12 dimer (red trace) and wild type MG1655 (blue trace) 50S subunit (0.5 μM), with 30S preIC was followed by the increase in light scattering at 430 nm. Inset shows the same experiment with long time scale (b) Tripeptide formation experiment starting from initiation complex. Color codes are same as in the subunit association figure.

Technological application with stalk protein L12 Purification of active ribosomes is the bottle neck step in all in vitro translation experiments. The conventional method of ribosome purification is quite expensive in terms of effort, time and cost for chemicals, this method involves several steps of ultracentrifugation (Antoun, Pavlov et al. 2004), so a simple and high-throughput method for purification of ribosomes is in demand. There have been several attempts to purify the ribosomes with affinity tags (Gan, Kitakawa et al. 2002; Leonov, Sergiev et al. 2003; Youngman and Green 2005). RNA fused affinity tag purification was mainly designed for the purification of E. coli ribosomes (Leonov, Sergiev et al. 2003; Youngman and Green 2005). These methods involved the plasmid based over expression of ribosomal components fused with affinity tags and the success of the method depends upon the over expression and the assembly of those components. To circumvent these hurdles we have created a novel E. coli strain JE28, in which a (His)6-tag has been inserted in frame at the C-terminus of the r-protein L12 directly on the chromosome using genetic recombineering method (Ellis, Yu et al. 2001; Costantino and Court 2003) (Figure 14). Since L12 is present in four copies on the large subunit of E. coli ribosome, the ribosomes of JE28 are homogeneously tetra-(His)6-tagged. The insertion of (His)6-tag did not affect the growth of the bacteria. Further we have developed a simple, fast, single-step affinity purifica-tion method to purify the tetra-(His)6-tagged ribosomes from JE28 strain.

(a) (b)

35

Affinity purified ribosomes contained only 70S either due to a much shorter preparation time (1–2 hrs) or due to high Mg2+ concentration during purifica-tion. The (His)6-tagged ribosomes were found to be equally active as the native ones in functional assays such as dipeptide formation and full-length protein synthesis in a cell free coupled transcription-translation system.

Figure 14: Insertion of linear DNA encoding (His)6, stop codon and Kan gene at the 3’ end of rplL gene in E. coli chromosome by λ Red recombineering.

Figure 15: (a) 20-50% sucrose density gradient separation profiles for conventional-ly purified ribosomes (black line), and affinity column purified ribosomes (blue line). (b) In vitro dipeptide formation experiment with 70S from MG1655 (black trace), JE28 conventional purification (red trace) and JE28 column purification (green trace).

We have also purified 30S and 50S subunits through this method. These tetra (His)6-tagged ribosomes can be used for the purification of vari-ous functional ribosomal complexes for biochemical and structural studies. It is also useful for the purification of ribosomes carrying mutations on rRNA or on r-proteins.

(a) (b)

36

Current work and future plan

Kinetic studies on translation initiation We have recently reported the role of ribosomal stalk protein L12 in IF2 mediated initiation of bacterial translation (Huang, Mandava et al. 2010). With fast kinetic measurements we have shown that L12 proteins are needed for the recognition and recruitment of IF2·GTP bound on 30S preinitiation complex, but not for the activation of its GTPase function. So, unlike in the cases of EF-Tu and EF-G, L12 is not a GAP (GTPase activating protein) for IF2. However, this study opened a general question about the sequence of events in translation initiation. The kinetic steps of translation initiation have been reported in recent articles (Antoun, Pavlov et al. 2006a; Fabbretti, Pon et al. 2007; Grigoriadou, Marzi et al. 2007a); there are some contradictory views for some of the steps in translation initiation. One among them is the IF2 mediated GTP hydrolysis during the subunit association whether it is required or not required for the rapid subunit association, if not then what is the role of GTP hydrolysis at the initiation step? There are two different opinions about the timing and role of GTP hydrolysis during the translation initiation. According to Antoun et al., IF2 in GTP form is needed for fast subunit association, and GTP hydrolysis is needed for the dissociation of IF2·GDP from the 70S initiation complex (Antoun, Pavlov et al. 2003). In contrary, Grigoriadou et al. from Barry Cooperman’s group reported that GTP hydrolysis is needed for stable sub-unit association (Grigoriadou, Marzi et al. 2007a). We thought to check both models by doing parallel kinetic experiments for different steps in initiation. To see whether GTP hydrolysis is necessary for subunit association or not, we have used GDPNP (non hydrolysable analogue of GTP) in subunit asso-ciation reactions followed by light scattering in stopped flow instrument. When added in excess, more or less same rate of subunit association was obtained with GDPNP as with GTP. We have also checked the subunit asso-ciation by titrating GDP, the rate of subunit association is very low and re-mained unchanged even with 1mM GDP (Figure 16b). From these results and the results from Paper-II suggests that GTP hydrolysis is not needed for subunit association but IF2 in GTP form is needed for rapid subunit associa-tion.

To confirm this finding we have used two IF2 G-domain mutants (V400G & H448E) which are impaired in GTPase activity (Luchin, Putzer et

37

al. 1999) in subunit association and in phosphate release experiments. When compared with wild type IF2, these two IF2 mutants did not show any defect in subunit association (Figure 16a), but because of their disability in GTPase activity there is no Pi released after 70S formation. These mutants again confirmed that subunit association can take place without any hydrolysis of GTP.

Figure 16: (a). Association of 50S subunits with 30S pre initiation complex contain-ing either WT (black trace) or two GTPase impaired mutant IF2s (V440G and H448E) (red and blue traces respectively) or without IF2 (green trace) followed by light scattering in stopped flow instrument. All variants of IF2 produced comparable rates of subunit association. (b). The rate constants for the subunit association reac-tion as in (a) with different guanine nucleotides as noted below the bars. The inset shows the rates against increasing concentration of GTP (black trace), GDP (blue trace) and GDPNP (green trace).

To investigate the role of GTP hydrolysis we have also checked the release of IF2 after subunit association in presence of GTP and GDPNP. In this experiment we have used a fluorescent labeled (rhodamine) IF2 and followed the decrease of fluorescence after the subunit association when IF2 is released. In the presence of GTP we can see the decrease of fluorescence indicating IF2 release, whereas with GDPNP we did not see any decrease in fluorescence. Subunits can associate in presence of GDPNP and GTP, but the peptide bond formation is very slow with GDPNP when compared to GTP, indicating that IF2 was able to promote the subunit association in pres-ence of GDPNP but it remained bound to 70S ribosome, thereby blocking the subsequent binding of EF-Tu ternary complex and peptide bond forma-tion (Antoun, Pavlov et al. 2003). When we have checked the dipeptide for-mation, both the IF2 mutants showed a slower rate of peptide bond forma-tion when compared to wild type IF2. This indicates the presence of mutant IF2s on 70S ribosome after subunit association, because of their inactivity in GTP hydrolysis. Thus the presence of IF2 blocks association of the EF-Tu ternary complex and peptide bond formation. Based on the above experi-ments we can conclude that GTP hydrolysis is needed for IF2 release, but

0.0 0.2 0.4 0.6 0.8 1.0

0.00

0.05

0.10

0.15

0.20

0.25

WT V400G H448E without IF2

Lig

ht S

catte

ring

(a

.u.)

Time (sec)

(a) (b)

38

not for subunit association. To further confirm this argument, we would like to check the release of the GTPase impaired IF2 mutants using the rhoda-mine labeled mutant IF2s. The prediction is that the mutant IF2s will be re-tained on the ribosome.

We have also checked the maximal rate for each step in translation in-itiation by titrating up the 50S concentration, further from these rates we have estimated the time needed for each individual step (Figure 17).

Figure 17: Kinetic scheme of the steps in translation initiation, maximal rate for individual steps are shown below the arrow of each step and the estimated times are shown above the arrow.

Future plans

Identification of residues on L12 involved in the interaction of IF2 during subunit association

We have mapped earlier the IF2 interaction sites on L12 CTD by us-ing heteronuclear NMR spectroscopy (Helgstrand, Mandava et al. 2007). However, for technical reasons the mapping was done with purified L12 protein and not with L12 bound on the ribosomes. Now that we have found that L12 plays an important role in IF2 recognition, it is important that we can trace the identification sites on L12 as well as on IF2 using mutagenesis approach. Based on the IF2 interaction sites mapped with NMR, we have now created some mutations on L12 CTD. The next step will be to reconsti-tute these mutant L12s on ribosomal cores (L12 depleted 50S and 70S) and the ribosomes with mutant L12 proteins will be tested primarily in subunit

39

association assay and then in GTP hydrolysis and Pi release, covering essen-tially all the steps in translation initiation.

Comparison of E. coli ribosomes with single, double and triple L12 dimers

As we mentioned earlier in this report we have created an E. coli strain (JE105) with single L12 dimer on its ribosomes. This strain has showed slower growth rate. We have also created another E. coli strain (CS111) where an additional L12 binding site has been added on the ribo-some, which contains three L12 dimer binding sites and therefore expected to bind three L12 dimers on its ribosomes. With the help of the above two strains we want to know why the number of L12 proteins per ribosome va-ries from two to three in different bacterial species and what is the need of multiple copies of L12 on the ribosome. We have already characterized the single L12 dimer ribosomes at different steps of translation and found out that they are defective in translation initiation and elongation when com-pared to the wild type ribosomes from E. coli strain MG1655. We will purify the triple L12 dimer ribosomes from CS111 and characterize those for their content as well as in different steps of translation.

40

Conclusions

From the above discussed works we can conclude that

1. L12 interacts with the translational GTPase factors (IF2, EF-Tu, EF-G and RF3) involving a common region on its CTD. The residues on L12 CTD involved in the interactions are K70, L80, E82 (strictly conserved), V66, A67, V68 and G79 (highly conserved). (Paper I).

2. Unlike for EF-G and EF-Tu, L12 is not a GAP for IF2. Instead, during initiation L12 is needed for the recognition and recruitment of IF2 in GTP form. (Paper II)

3. Ribosomes with single L12 dimer are defective in IF2 mediated subunit association and EF-G mediated elongation. Multiple L12 dimers are needed on the ribosome for efficient translation and fast growth of the bacteria. (Paper III)

4. His-tagging on multi copy L12 protein provides a good system for single step purification of active ribosomes. (Paper IV)

41

Summary in Swedish

Ribosomala stjälkprotein L12: struktur, funktion och tillämpningar.

Bakgrund Ribosomerna är stora ribonucleoprotein komplex med en approxima-tiv massa på 2,5 MDA och en diameter på 200 - 250 Å. Ribosomerna finns i cytoplasman, i eukaryoter finns de också på det endoplasmatiska nätverket. Ribosomer översätter den genetiska information kodad i budbärar-RNA (mRNA) till motsvarande sekvens av aminosyror och tillverkar därigenom ett protein. Ribosomen består av proteiner och ribosomala RNA (rRNA), delat i två subenheter, som brukar kallas den stora och den lilla subenheten. Proka-ryota (70S) och eukaryota (80S) ribosomer, som är mycket lika i organisa-tion och struktur, skiljer sig dock i både protein och RNA-komponenter. Där finns tre olika tRNA bindningsställen på ribosomen som heter A- (amino acyl), P- (peptidyl) och E- (exit) stäl.

I bakterier sker proteinsyntesen i fyra steg, initiering, elongering, ter-minering och återvinning. Flera GTPase protein faktorer (IF2, EF-Tu, EF-G och RF3) samverkar med den ribosomala stjälken under dessa steg. Stjälken består av 2-3 dimerer av L12 och ett L10 protein vilket utgör nedre delen av stjälken. I E. coli finns fyra kopior av L12 vilka existerar som en dimer av dimerer på ribosomernas 50S-subenhet. I denna avhandling har vi rapporte-rat följande saker i detalj.

Samverkan mellan L12 och translationella G-faktorer L12 proteiner är kända för att interagera med flera GTPase faktorer

under translationen. Vi har kartlagt interaktionsplatser för de fyra viktigaste G-faktorerna (IF2, EF-Tu, EF-G & RF3) på L12 molekylen med heteronuk-leär NMR-spektroskopi. Våra data visar att alla dessa fyra GTPase faktorer I huvudsak samverkar med samma plats på den konserverade C-terminala domänen (CTD) i L12 proteinet. Denna region innehåller tre helt konserve-rade aminosyrarester K70, L80 och E82 samt några starkt konserverade res-

42

ter på två intilliggande α-helixar. Detta tyder på en generell mekanism för samspelet mellan L12 och G-faktorer.

L12s roll i initieringen av proteinsyntesen L12 är känd för att stimulera GTPase aktivering av EF-Tu och EF-G

under elongeringsfasen av proteinsyntesen. I denna rapport har vi klargjort betydelsen av L12 i IF2 medierad initiering av proteinsyntesen genom att jämföra 50S utan L12 med vanliga 50S-subenheter i association av subenhe-terna till 70S ribosomer, GTP hydrolys på IF2, frisläppning av oorganiskt fosfat efter GTP hydroly samt frisläppning av IF2 från ribosomen. 50S sub-enheter utan L12 associerar mycket långsammare (~ 40 gånger) med 30S endast när IF2-GTP finns på 30S innan associeringen. Detta tyder på att en interaktion mellan L12 protein på 50S och IF2-GTP på 30S behövs för snabb associering av subenheterna. Avlägsnande av L12 från 50S påverkade inte de efterföljande stegen. Således har vi kommit fram till att L12 inte är ett GAP (GTPase aktiverande protein) för IF2.

Varför behöver bakterierna flera kopior av L12 protein på ribosomen?

Antalet L12 molekyler på ribosomen är olika i olika organismer. L12 binder till ribosomen via C-terminal α8 helix på L10 proteinet. För att under-söka behovet av fler kopior av L12 på ribosomen har vi skapat en genetiskt modifierad E. coli-stam (JE105), som producerar ribosomer med endast en L12 dimer. Denna stam uppvisade en betydande tillväxt defekt. In vitro var dessa ribosomer 2-3 gånger långsammare i initiering, men visade inga fel på EF-Tu och EF-G medierade steg i peptidelongering. Dessa resultat indikerar en viktig roll för L12 i initieringen av proteinsyntesen.

Teknisk tillämpning av stjälkproteinet L12 I denna rapport har vi också diskuterat tekniska tillämpningar av flera

kopior av L12 protein. Vi har skapat en genetiskt modifierad E. coli-stam (JE28) som har en (His)6-tag I slutet av L12 proteinet. Eftersom L12 protein finns i fyra kopior på ribosomen, innehåller ribosomer som produceras i denna stam fyra (His)6-tags. Vi har utvecklat en metod för att rena dessa ribosomer från JE28. Dessa ribosomker var lika aktiva som konventionellt renade ribosomer från vildtyp E. coli stammar, i in vitro dipeptid bildning och i cell fri proteinsyntes

43

Acknowledgement

I would like to acknowledge and extend my heart felt gratitude to my su-pervisor Suparna Sanyal for accepting me as her first PhD student. Suparna you have been very supportive throughout this period. You are very inspira-tional and I have always admired your passion for science. Thank you very much for being there whenever I needed your help in and out of the lab. My co-supervisor Måns Ehrenberg, thank you Måns for your suggestions and comments during the group meetings.

My co-authors specially Huang Chenhui, thank you Huang for teaching me all complicated preps. You’re a nice friend and colleague. Josefine thanks for generating very stable strains (JE28 and JE105). Chris you are a good friend and thanks for correcting my thesis and I wish you good luck.

I would like to thank my collaborators Santanu Dasgupta and Mikael Akke. Santanu thank you for the wonderful discussions during the coffee time about the Science, Politics, Food… what not

I would like to thank my dearest former and present office members. Su-zana you’re a wonderful friend, thanks for the green office room, some of the plants are still good and Bruce Lee is still hanging on the wall with all the birthday marks. Venkat garu thanks helping me in the lab during the initial days. Christiane thanks for your friendship and for the Milk mouse choco-lates. Kristin, Mikael and Ravi, thank you guys for the cake and chocolate supplies. Puli, good luck for your licentiate. Mikael thanks for the Swedish translations and especially for the Swedish summary in the thesis. Kristin thanks for the western blots. Chian, thanks for helping me with the red re-combineering. Ranjeet, thanks for the comments and corrections for my thesis and for the spiritual discussions. Gürkan thanks for helping me with the PyMol. Yanhong, good luck for your licentiate. Liang thanks for the L12 mutations and for the L10 plasmid. Antje good luck. Ray thanks for the supply of components in the general box and for proof reading my thesis. Deba da, thanks for the wonderful 'tonfisk’ curry; you’re a good researcher and I had good time working with you, good luck with the science and life.

I would like to thank present and former members of Måns Ehrenberg group, especially Lamine for teaching me the ribosome preparation. Misha, thanks for your suggestions and for the help with stopped flow. Anneli, Do-reen, Galyna, Harriet, Ikue, Jingji, Kaweng, and Magnus, I wish you all good luck for your science. David thanks for showing me the execution of the curve smoothing program. Thanks to all my friends whom I did not men-tion here.

Finally to family,, you are the best …!!!

44

References:

Agrawal, R. K., A. B. Heagle, et al. (1999). "EF-G-dependent GTP hydrolysis induces translocation accompanied by large conformational changes in the 70S ribosome." Nat Struct Biol 6(7): 643-647.

Agrawal, R. K., P. Penczek, et al. (1998). "Visualization of elongation factor G on the Escherichia coli 70S ribosome: the mechanism of translocation." Proc Natl Acad Sci U S A 95(11): 6134-6138.

Allen, G. S., A. Zavialov, et al. (2005). "The cryo-EM structure of a translation initiation complex from Escherichia coli." Cell 121(5): 703-712.

Antoun, A., M. Y. Pavlov, et al. (2003). "The roles of initiation factor 2 and guanosine triphosphate in initiation of protein synthesis." EMBO J 22(20): 5593-5601.

Antoun, A., M. Y. Pavlov, et al. (2006a). "How initiation factors tune the rate of initiation of protein synthesis in bacteria." EMBO J 25(11): 2539-2550.

Antoun, A., M. Y. Pavlov, et al. (2006b). "How initiation factors maximize the accuracy of tRNA selection in initiation of bacterial protein synthesis." Mol Cell 23(2): 183-193.

Antoun, A., M. Y. Pavlov, et al. (2004). "Ribosome formation from subunits studied by stopped-flow and Rayleigh light scattering." Biol Proced Online 6: 35-54.

Aoki, H., J. Xu, et al. (2008). "Interactions of elongation factor EF-P with the Escherichia coli ribosome." The FEBS journal 275(4): 671-681.

Arnez, J. G. and D. Moras (1997). "Structural and functional considerations of the aminoacylation reaction." Trends in biochemical sciences 22(6): 211-216.

Beauclerk, A. A., E. Cundliffe, et al. (1984). "The binding site for ribosomal protein complex L8 within 23 s ribosomal RNA of Escherichia coli." The Journal of biological chemistry 259(10): 6559-6563.

Bilgin, N., L. A. Kirsebom, et al. (1988). "Mutations in ribosomal proteins L7/L12 perturb EF-G and EF-Tu functions." Biochimie 70(5): 611-618.

Bocharov, E. V., A. T. Gudkov, et al. (1998). "Conformational independence of N- and C-domains in ribosomal protein L7/L12 and in the complex with protein L10." FEBS Lett 423(3): 347-350.

Bocharov, E. V., A. G. Sobol, et al. (2004). "From structure and dynamics of protein L7/L12 to molecular switching in ribosome." J Biol Chem 279(17): 17697-17706.

Bourne, H. R., D. A. Sanders, et al. (1991). "The GTPase superfamily: conserved structure and molecular mechanism." Nature 349(6305): 117-127.

Brot, N., W. P. Tate, et al. (1974). "The requirement for ribosomal proteins L7 and L12 in peptide-chain termination." Proc Natl Acad Sci U S A 71(1): 89-92.

Brot, N., E. Yamasaki, et al. (1972). "The properties of an E. coli ribosomal protein required for the function of factor G." Arch Biochem Biophys 148(1): 148-155.

Chandra Sanyal, S. and A. Liljas (2000). "The end of the beginning: structural studies of ribosomal proteins." Curr Opin Struct Biol 10(6): 633-636.

45

Christodoulou, J., G. Larsson, et al. (2004). "Heteronuclear NMR investigations of dynamic regions of intact Escherichia coli ribosomes." Proc Natl Acad Sci U S A 101(30): 10949-10954.

Costantino, N. and D. L. Court (2003). "Enhanced levels of lambda Red-mediated recombinants in mismatch repair mutants." Proc Natl Acad Sci U S A 100(26): 15748-15753.

Dey, D., D. E. Bochkariov, et al. (1998). "Cross-linking of selected residues in the N- and C-terminal domains of Escherichia coli protein L7/L12 to other ribosomal proteins and the effect of elongation factor Tu." J Biol Chem 273(3): 1670-1676.

Diaconu, M., U. Kothe, et al. (2005). "Structural basis for the function of the ribosomal L7/12 stalk in factor binding and GTPase activation." Cell 121(7): 991-1004.

Ellis, H. M., D. Yu, et al. (2001). "High efficiency mutagenesis, repair, and engineering of chromosomal DNA using single-stranded oligonucleotides." Proc Natl Acad Sci U S A 98(12): 6742-6746.

Fabbretti, A., C. L. Pon, et al. (2007). "The real-time path of translation factor IF3 onto and off the ribosome." Molecular cell 25(2): 285-296.

Gan, X., M. Kitakawa, et al. (2002). "Tag-mediated isolation of yeast mitochondrial ribosome and mass spectrometric identification of its new components." European journal of biochemistry / FEBS 269(21): 5203-5214.

Gao, H., Z. Zhou, et al. (2007). "RF3 induces ribosomal conformational changes responsible for dissociation of class I release factors." Cell 129(5): 929-941.

Gao, Y. G., M. Selmer, et al. (2009). "The structure of the ribosome with elongation factor G trapped in the posttranslocational state." Science 326(5953): 694-699.

Gordiyenko, Y., S. Deroo, et al. (2008). "Acetylation of L12 increases interactions in the Escherichia coli ribosomal stalk complex." J Mol Biol 380(2): 404-414.

Gordiyenko, Y., H. Videler, et al. (2010). "Mass spectrometry defines the stoichiometry of ribosomal stalk complexes across the phylogenetic tree." Mol Cell Proteomics 9(8): 1774-1783.

Griaznova, O. and R. R. Traut (2000). "Deletion of C-terminal residues of Escherichia coli ribosomal protein L10 causes the loss of binding of one L7/L12 dimer: ribosomes with one L7/L12 dimer are active." Biochemistry 39(14): 4075-4081.

Grigoriadou, C., S. Marzi, et al. (2007a). "A quantitative kinetic scheme for 70 S translation initiation complex formation." J Mol Biol 373(3): 562-572.

Grill, S., I. Moll, et al. (2001). "Modulation of ribosomal recruitment to 5'-terminal start codons by translation initiation factors IF2 and IF3." FEBS letters 495(3): 167-171.

Grunberg-Manago, M., P. Dessen, et al. (1975). "Light-scattering studies showing the effect of initiation factors on the reversible dissociation of Escherichia coli ribosomes." J Mol Biol 94(3): 461-478.

Gualerzi, C. O. and C. L. Pon (1990). "Initiation of mRNA translation in prokaryotes." Biochemistry 29(25): 5881-5889.

Gudkov, A. T. (1997). "The L7/L12 ribosomal domain of the ribosome: structural and functional studies." FEBS Lett 407(3): 253-256.

Gudkov, A. T., M. G. Bubunenko, et al. (1991). "Overexpression of L7/L12 protein with mutations in its flexible region." Biochimie 73(11): 1387-1389.

Hamel, E., M. Koka, et al. (1972). "Requirement of an Escherichia coli 50 S ribosomal protein component for effective interaction of the ribosome with T and G factors and with guanosine triphosphate." J Biol Chem 247(3): 805-814.

46

Hardy, S. J. (1975). "The stoichiometry of the ribosomal proteins of Escherichia coli." Mol Gen Genet 140(3): 253-274.

Helgstrand, M., C. S. Mandava, et al. (2007). "The ribosomal stalk binds to translation factors IF2, EF-Tu, EF-G and RF3 via a conserved region of the L12 C-terminal domain." J Mol Biol 365(2): 468-479.

Hirokawa, G., R. M. Nijman, et al. (2005). "The role of ribosome recycling factor in dissociation of 70S ribosomes into subunits." RNA 11(8): 1317-1328.

Huang, C., C. S. Mandava, et al. (2010). "The ribosomal stalk plays a key role in IF2-mediated association of the ribosomal subunits." J Mol Biol 399(1): 145-153.

Ibba, M. and D. Soll (2000). "Aminoacyl-tRNA synthesis." Annual review of biochemistry 69: 617-650.

Ilag, L. L., H. Videler, et al. (2005). "Heptameric (L12)6/L10 rather than canonical pentameric complexes are found by tandem MS of intact ribosomes from thermophilic bacteria." Proc Natl Acad Sci U S A 102(23): 8192-8197.

Ito, K., M. Uno, et al. (2000). "A tripeptide 'anticodon' deciphers stop codons in messenger RNA." Nature 403(6770): 680-684.

Karimi, R., M. Y. Pavlov, et al. (1999). "Novel roles for classical factors at the interface between translation termination and initiation." Molecular cell 3(5): 601-609.

Kay, A., G. Sander, et al. (1973). "Effect of ribosomal protein L12 upon initiation factor IF-2 activities." Biochem Biophys Res Commun 51(4): 979-986.

Kirsebom, L. A., R. Amons, et al. (1986). "Primary structures of mutationally altered ribosomal protein L7/L12 and their effects on cellular growth and translational accuracy." Eur J Biochem 156(3): 669-675.

Kirsebom, L. A. and L. A. Isaksson (1985). "Involvement of ribosomal protein L7/L12 in control of translational accuracy." Proc Natl Acad Sci U S A 82(3): 717-721.

Kisselev, L. L. and R. H. Buckingham (2000). "Translational termination comes of age." Trends in biochemical sciences 25(11): 561-566.

Kothe, U., H. J. Wieden, et al. (2004). "Interaction of helix D of elongation factor Tu with helices 4 and 5 of protein L7/12 on the ribosome." Journal of molecular biology 336(5): 1011-1021.

La Teana, A., C. O. Gualerzi, et al. (1995). "From stand-by to decoding site. Adjustment of the mRNA on the 30S ribosomal subunit under the influence of the initiation factors." RNA 1(8): 772-782.

Laursen, B. S., H. P. Sorensen, et al. (2005). "Initiation of protein synthesis in bacteria." Microbiology and molecular biology reviews : MMBR 69(1): 101-123.

Leijonmarck, M. and A. Liljas (1987). "Structure of the C-terminal domain of the ribosomal protein L7/L12 from Escherichia coli at 1.7 A." J Mol Biol 195(3): 555-579.

Leonov, A. A., P. V. Sergiev, et al. (2003). "Affinity purification of ribosomes with a lethal G2655C mutation in 23 S rRNA that affects the translocation." The Journal of biological chemistry 278(28): 25664-25670.

Li, W., J. Sengupta, et al. (2006). "Functional conformations of the L11-ribosomal RNA complex revealed by correlative analysis of cryo-EM and molecular dynamics simulations." RNA 12(7): 1240-1253.

Liljas, A. and A. T. Gudkov (1987). "The structure and dynamics of ribosomal protein L12." Biochimie 69(10): 1043-1047.

Luchin, S., H. Putzer, et al. (1999). "In vitro study of two dominant inhibitory GTPase mutants of Escherichia coli translation initiation factor IF2. Direct

47

evidence that GTP hydrolysis is necessary for factor recycling." J Biol Chem 274(10): 6074-6079.

Maki, Y., T. Hashimoto, et al. (2007). "Three binding sites for stalk protein dimers are generally present in ribosomes from archaeal organism." The Journal of biological chemistry 282(45): 32827-32833.

Marintchev, A. and G. Wagner (2004). "Translation initiation: structures, mechanisms and evolution." Quarterly reviews of biophysics 37(3-4): 197-284.

Milon, P., M. Carotti, et al. (2010). "The ribosome-bound initiation factor 2 recruits initiator tRNA to the 30S initiation complex." EMBO Rep 11(4): 312-316.

Moazed, D. and H. F. Noller (1989). "Interaction of tRNA with 23S rRNA in the ribosomal A, P, and E sites." Cell 57(4): 585-597.

Moens, P. D., M. C. Wahl, et al. (2005). "Oligomeric state and mode of self-association of Thermotoga maritima ribosomal stalk protein L12 in solution." Biochemistry 44(9): 3298-3305.

Mohr, D., W. Wintermeyer, et al. (2002). "GTPase activation of elongation factors Tu and G on the ribosome." Biochemistry 41(41): 12520-12528.

Moll, I., S. Grill, et al. (2002). "Leaderless mRNAs in bacteria: surprises in ribosomal recruitment and translational control." Molecular microbiology 43(1): 239-246.

Mulder, F. A., L. Bouakaz, et al. (2004). "Conformation and dynamics of ribosomal stalk protein L12 in solution and on the ribosome." Biochemistry 43(20): 5930-5936.

Nissen, P., J. Hansen, et al. (2000). "The structural basis of ribosome activity in peptide bond synthesis." Science 289(5481): 920-930.

Ogle, J. M., D. E. Brodersen, et al. (2001). "Recognition of cognate transfer RNA by the 30S ribosomal subunit." Science 292(5518): 897-902.

Olson, H. M., A. Sommer, et al. (1986). "Localization of two epitopes of protein L7/L12 to both the body and stalk of the large ribosomal subunit. Immune electron microscopy using monoclonal antibodies." J Biol Chem 261(15): 6924-6932.

Pavlov, M. Y., A. Antoun, et al. (2008). "Complementary roles of initiation factor 1 and ribosome recycling factor in 70S ribosome splitting." EMBO J 27(12): 1706-1717.

Peske, F., M. V. Rodnina, et al. (2005). "Sequence of steps in ribosome recycling as defined by kinetic analysis." Molecular cell 18(4): 403-412.

Pettersson, I., S. J. Hardy, et al. (1976). "The ribosomal protein L8 is a complex L7/L12 and L10." FEBS Lett 64(1): 135-138.

Pettersson, I. and C. G. Kurland (1980). "Ribosomal protein L7/L12 is required for optimal translation." Proc Natl Acad Sci U S A 77(7): 4007-4010.

Pettersson, I. and A. Liljas (1979). "The stoichiometry and reconstitution of a stable protein complex from Escherichia coli ribosomes." FEBS Lett 98(1): 139-144.

Qin, Y., N. Polacek, et al. (2006). "The highly conserved LepA is a ribosomal elongation factor that back-translocates the ribosome." Cell 127(4): 721-733.

RajBhandary, U. L. (1994). "Initiator transfer RNAs." Journal of bacteriology 176(3): 547-552.

Ramakrishnan, V. (2002). "Ribosome structure and the mechanism of translation." Cell 108(4): 557-572.

Rheinberger, H. J. and K. H. Nierhaus (1986). "Allosteric interactions between the ribosomal transfer RNA-binding sites A and E." The Journal of biological chemistry 261(20): 9133-9139.

48

Rheinberger, H. J., H. Sternbach, et al. (1981). "Three tRNA binding sites on Escherichia coli ribosomes." Proceedings of the National Academy of Sciences of the United States of America 78(9): 5310-5314.

Ringquist, S., S. Shinedling, et al. (1992). "Translation initiation in Escherichia coli: sequences within the ribosome-binding site." Molecular microbiology 6(9): 1219-1229.

Rodnina, M. V., A. Savelsbergh, et al. (1997). "Hydrolysis of GTP by elongation factor G drives tRNA movement on the ribosome." Nature 385(6611): 37-41.

Rodnina, M. V., H. Stark, et al. (2000). "GTPases mechanisms and functions of translation factors on the ribosome." Biological chemistry 381(5-6): 377-387.

Savelsbergh, A., D. Mohr, et al. (2005). "Control of phosphate release from elongation factor G by ribosomal protein L7/12." EMBO J 24(24): 4316-4323.

Savelsbergh, A., D. Mohr, et al. (2000). "Stimulation of the GTPase activity of translation elongation factor G by ribosomal protein L7/12." The Journal of biological chemistry 275(2): 890-894.

Schluenzen, F., A. Tocilj, et al. (2000). "Structure of functionally activated small ribosomal subunit at 3.3 angstroms resolution." Cell 102(5): 615-623.

Schmeing, T. M. and V. Ramakrishnan (2009). "What recent ribosome structures have revealed about the mechanism of translation." Nature 461(7268): 1234-1242.

Schmidt, F. J., J. Thompson, et al. (1981). "The binding site for ribosomal protein L11 within 23 S ribosomal RNA of Escherichia coli." The Journal of biological chemistry 256(23): 12301-12305.

Schneider, T. D., G. D. Stormo, et al. (1986). "Information content of binding sites on nucleotide sequences." Journal of molecular biology 188(3): 415-431.

Selmer, M., C. M. Dunham, et al. (2006). "Structure of the 70S ribosome complexed with mRNA and tRNA." Science 313(5795): 1935-1942.

Shine, J. and L. Dalgarno (1974). "The 3'-terminal sequence of Escherichia coli 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites." Proceedings of the National Academy of Sciences of the United States of America 71(4): 1342-1346.

Stark, H., M. V. Rodnina, et al. (1997). "Visualization of elongation factor Tu on the Escherichia coli ribosome." Nature 389(6649): 403-406.

Strycharz, W. A., M. Nomura, et al. (1978). "Ribosomal proteins L7/L12 localized at a single region of the large subunit by immune electron microscopy." J Mol Biol 126(2): 123-140.

Subramanian, A. R. (1975). "Copies of proteins L7 and L12 and heterogeneity of the large subunit of Escherichia coli ribosome." J Mol Biol 95(1): 1-8.

Terhorst, C., W. Moller, et al. (1972). "Amino acid sequence of a 50 S ribosomal protein involved in both EFG and EFT dependent GTP-hydrolysis." FEBS Lett 28(3): 325-328.

Varshney, U., C. P. Lee, et al. (1993). "From elongator tRNA to initiator tRNA." Proceedings of the National Academy of Sciences of the United States of America 90(6): 2305-2309.

Wahl, M. C., G. P. Bourenkov, et al. (2000). "Flexibility, conformational diversity and two dimerization modes in complexes of ribosomal protein L12." EMBO J 19(2): 174-186.

Wilson, D. N. and K. H. Nierhaus (2003). "The ribosome through the looking glass." Angewandte Chemie 42(30): 3464-3486.

Wilson, D. N. and K. H. Nierhaus (2007). "The weird and wonderful world of bacterial ribosome regulation." Critical reviews in biochemistry and molecular biology 42(3): 187-219.

49

Wu, X. Q. and U. L. RajBhandary (1997). "Effect of the amino acid attached to Escherichia coli initiator tRNA on its affinity for the initiation factor IF2 and on the IF2 dependence of its binding to the ribosome." The Journal of biological chemistry 272(3): 1891-1895.

Yonath, A., K. R. Leonard, et al. (1987). "A tunnel in the large ribosomal subunit revealed by three-dimensional image reconstruction." Science 236(4803): 813-816.

Youngman, E. M. and R. Green (2005). "Affinity purification of in vivo-assembled ribosomes for in vitro biochemical analysis." Methods 36(3): 305-312.

Yu, D., H. M. Ellis, et al. (2000). "An efficient recombination system for chromosome engineering in Escherichia coli." Proceedings of the National Academy of Sciences of the United States of America 97(11): 5978-5983.

Yusupova, G. Z., M. M. Yusupov, et al. (2001). "The path of messenger RNA through the ribosome." Cell 106(2): 233-241.

Zavialov, A. V., R. H. Buckingham, et al. (2001). "A posttermination ribosomal complex is the guanine nucleotide exchange factor for peptide release factor RF3." Cell 107(1): 115-124.

Zavialov, A. V. and M. Ehrenberg (2003). "Peptidyl-tRNA regulates the GTPase activity of translation factors." Cell 114(1): 113-122.

Zavialov, A. V., V. V. Hauryliuk, et al. (2005). "Guanine-nucleotide exchange on ribosome-bound elongation factor G initiates the translocation of tRNAs." Journal of biology 4(2): 9.

Zavialov, A. V., L. Mora, et al. (2002). "Release of peptide promoted by the GGQ motif of class 1 release factors regulates the GTPase activity of RF3." Molecular cell 10(4): 789-798.

Zecherle, G. N., A. Oleinikov, et al. (1992). "The proximity of the C-terminal domain of Escherichia coli ribosomal protein L7/L12 to L10 determined by cysteine site-directed mutagenesis and protein-protein cross-linking." J Biol Chem 267(9): 5889-5896.