Francois Franceschi

Structure of Functionally Activated Small Ribosomal Subunit at 3.3 Å Resolution

The small ribosomal subunit performs the decoding of genetic information during translation. The structure of that from Thermus thermophilus shows that the decoding center, which positions mRNA and three tRNAs, is constructed entirely of RNA. The entrance to the mRNA channel will encircle the message when a latch-like contact closes and contributes to processivity and fidelity. Extended RNA helical elements that run longitudinally through the body transmit structural changes, correlating events at the particles far end with the cycle of mRNA translocation at the decoding region. 96% of the nucleotides were traced and the main fold of all proteins was determined. The latter are either peripheral or appear to serve as linkers. Some may assist the directionality of translocation.

High Resolution Structure of the Large Ribosomal Subunit from a Mesophilic Eubacterium

We describe the high resolution structure of the large ribosomal subunit from Deinococcus radiodurans (D50S), a gram-positive mesophile suitable for binding of antibiotics and functionally relevant ligands. The over-all structure of D50S is similar to that from the archae bacterium Haloarcula marismortui (H50S); however, a detailed comparison revealed significant differences, for example, in the orientation of nucleotides in peptidyl transferase center and in the structures of many ribosomal proteins. Analysis of ribosomal features involved in dynamic aspects of protein biosynthesis that are partially or fully disordered in H50S revealed the conformations of intersubunit bridges in unbound subunits, suggesting how they may change upon subunit association and how movements of the L1-stalk may facilitate the exit of tRNA.

Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria

Ribosomes, the site of protein synthesis, are a major target for natural and synthetic antibiotics. Detailed knowledge of antibiotic binding sites is central to understanding the mechanisms of drug action. Conversely, drugs are excellent tools for studying the ribosome function. To elucidate the structural basis of ribosome–antibiotic interactions, we determined the high-resolution X-ray structures of the 50S ribosomal subunit of the eubacterium Deinococcus radiodurans, complexed with the clinically relevant antibiotics chloramphenicol, clindamycin and the three macrolides erythromycin, clarithromycin and roxithromycin. We found that antibiotic binding sites are composed exclusively of segments of 23S ribosomal RNA at the peptidyl transferase cavity and do not involve any interaction of the drugs with ribosomal proteins. Here we report the details of antibiotic interactions with the components of their binding sites. Our results also show the importance of putative Mg+2 ions for the binding of some drugs. This structural analysis should facilitate rational drug design.

Crystal structures of complexes of the small ribosomal subunit with tetracycline, edeine and IF3.

The small ribosomal subunit is responsible for the decoding of genetic information and plays a key role in the initiation of protein synthesis. We analyzed by X‐ray crystallography the structures of three different complexes of the small ribosomal subunit of Thermus thermophilus with the A‐site inhibitor tetracycline, the universal initiation inhibitor edeine and the C‐terminal domain of the translation initiation factor IF3. The crystal structure analysis of the complex with tetracycline revealed the functionally important site responsible for the blockage of the A‐site. Five additional tetracycline sites resolve most of the controversial biochemical data on the location of tetracycline. The interaction of edeine with the small subunit indicates its role in inhibiting initiation and shows its involvement with P‐site tRNA. The location of the C‐terminal domain of IF3, at the solvent side of the platform, sheds light on the formation of the initiation complex, and implies that the anti‐association activity of IF3 is due to its influence on the conformational dynamics of the small ribosomal subunit.

Structural basis for the antibiotic activity of ketolides and azalides.

The azalide azithromycin and the ketolide ABT-773, which were derived by chemical modifications of erythromycin, exhibit elevated activity against a number of penicillin- and macrolide-resistant pathogenic bacteria. Analysis of the crystal structures of the large ribosomal subunit from Deinococcus radiodurans complexed with azithromycin or ABT-773 indicates that, despite differences in the number and nature of their contacts with the ribosome, both compounds exert their antimicrobial activity by blocking the protein exit tunnel. In contrast to all macrolides studied so far, two molecules of azithromycin bind simultaneously to the tunnel. The additional molecule also interacts with two proteins, L4 and L22, implicated in macrolide resistance. These studies illuminated and rationalized the enhanced activity of the drugs against specific macrolide-resistant bacteria.

Characterization and preliminary attempts for derivatization of crystals of large ribosomal subunits from Haloarcula marismortui diffracting to 3 A resolution.

An improved form of crystals of large (50 S) ribosomal subunits from Haloarcula marismortui, formally named Halobacterium marismortui, diffracting to 3 A resolution, has been obtained by the addition of 1 mM-Cd2+ to the crystallization medium, which contained more than 1.9 M of other salts. The improved crystals, grown from functionally active particles to an average size of 0.3 mm x 0.3 mm x 0.08 mm, are isomorphous with the previously reported ones, which diffracted to 4.5 A. They are of space group C222(1), cell dimensions a = 210 A, b = 300 A, c = 581 A, and contain one particle in the asymmetric unit. Their superior internal order is reflected not only in their high resolution, but also in their reasonable mosaicity (less than 0.3 degrees). In contrast to the previously grown crystals, the new ones are of adequate mechanical strength and survive well the shock-cooling treatment. Due to their weak diffracting power, all crystallographic studies have been performed with synchrotron radiation. At cryotemperature, these crystals showed no measurable decay for a few days of irradiation and a complete diffraction data set could be collected from a single crystal. Efforts for initial phasing by specific and quantitative derivatization with super-dense heavy-atom clusters are in progress.

The Translational Apparatus

Robin Ray Gutell MCDBiology Campus Box 347 University of Colorado Boulder, Colorado 80309-0347 USA email: rgutell@boulder.colorado.edu The elucidation of 16S and 23S rRNA Higher-Order Structure has been addressed by Comparative Sequence Methods for more than a decade. During these years our comparative methods have evolved as the number of complete 16S and 23S rRNA sequences have increased significantly, resulting in the maturation of the higher-order structure models for 16S and 23S rRNA. With over 1000 16S (and 16S-like) and 200 23S (and 23S-like) sequences at this time, we have strong comparative evidence for the vast majority of all secondary structure base pairings, and are thus quite confident of the majority of the proposed Escherichia coli 16S and 23S rRNA secondary structure. Within the past few years additional rRNA Higher-Order structure constraints have been elucidated; constraints that reveal various RNA structural forms, including lone canonical pairings, pseudoknots, non-canonical pairings, tetra loops, canonical and non-canonical pairings that together forms a parallel (vs. the usual antiparallel) stranded structural element, and suggestive evidence for coaxial stacking of adjacent helices. At this time we question what additional RNA structural constraints can be deciphered with comparative structure methods. To answer such questions, the rRNA sequence collection will need to continue to grow in both number and diversity, and our comparative structure algorithms need to evolve to a more sophisticated level. In an effort to establish the limits for structural similarity, we need to address how different two higher-order structures can be and still be considered analogous. Introductory Statements Since the flrstcomplete 16S (Brosius et al. 1978) and 23S (Brosius et al. 1980) rRNA sequences were determined, comparative analyds of these molecules has progressed in a variety of ways. Maybe foremost for the majority (especially for this audience) is the resulting higher-order structures, which ribosome-ologists utilize to map andlor design their experiments onto. While there is a wealth of information that can and should be elucidated from the sequences that make up the 16S and 23S rRNA datasets, this article will focus on the most obvious and probably experimentally meaningful structural features, namely secondary structure helices, tertiary interactions, and a few interesting examples of other comparatively derived structural constraints. And since much has already been written on comparatively derived rRNA structure [and most recently for an upcoming book on ribosomal RNA (Gutell et al. 1993), this article will only briefly touch on some of the emerging RNA structural features and new The Translational Apparatus, Edited by K.H. Nierhaus et al., Plenum Press, New York, 1993 477 structural possibilities that have been uncovered within the past year or so, leaving the interested reader to investigate elsewhere for a more encompassing perspective of the details of the comparatively derived rRNA structures. Comparative Structure Analysis What is the basis of this method? What might we expect to decipher? This method is rooted in the simple concept that similar or analogous three-dimensional structure can be composed of different primary structures, or in other words many different primary structures can fold into the same isomorphic 3-dimensional structure. Thus natural selection can maintain and act on the higher-order structures of RNA while the primary structure is free to change, although constrained in its divergence. The ribosomal RNA is an ideal molecule to apply such methodology to due to its structural and functional role in the ribosome, and the ribosomes position in protein synthesis and the evolution of the cell (Woese 1980). Underlying this method are a number of key questions that cannot be answered a priori. How much and what types of variance can be tolerated in higher-order structure before these structures are not considered isomorphic? How much overall similarity should we expect to fmd for any RNA molecule? How much overall variance should we expect to find for any RNA molecule (i.e., tRNA: type I vs. type II)? To what extent can these methods identify general folding patterns and to what extent can these methods identify and distinguish subtle and detailed RNA structure (i.e., elucidate the generalized three-dimensional structure for tRNAs; elucidate the detailed features recognized by each of the aminoacyl synthetases for their cognate tRNA)? Will all RNA structural motifs be identified with such methods, or will only a subset of these structural elements be amenable to such methods, i.e., should we expect secondary and all tertiary interactions to be equally decipherable? And lastly, should we anticipate the same overall and/or detailed structural (and even biological) constraints within phylogenetic ally related vs. distant structures? A quick glimpse at the progression of our rRNA structure models Although these questions are not (yet) answerable, the comparative analysis of the rRNAs (and all RNAs for that matter) has advanced in stages, in part so the results from each stage with their underlying assumptions can be evalulated before moving on to the next stage, and in part due to the significant increase in the number of sequences, development of the underlying correlation analysis algorithms, and the fact that we believe there is more structural detail to be found at the completion of each stage. This analysis started with basic assumptions that were congruent with principles elucidated with experimental methods. Initially the comparative structure searched for the helices that compose the overall secondary structure [For 16S rRNA: (Woese et al. 1980, Stiegler et al. 1980, Zwieb et al. 1981), and 23S rRNA: (Nolier et al. 1981, Glotz et al. 1981, and Branlant et al. 1981) rRNAs]. These methods specifically searched for canonical base pairings (ie. A-U and G-C) arranged contiguously and in an antiparallel orientation. These structures were tested and evaluated with each new rRNA sequence, resulting in numerous refmements in the secondary structures [Many references not noted here]. The specific search for helical elements gave way to a more generalized, non-structure based method. This method (Gutell et al. 1985) transformed the pattern of nucleotides at each column in the sequence alignment to a number pattern, which was based on the pattern of conservation and variance at every position in the molecule. Similar number patterns were subsequently grouped and analyzed, resulting in refmements in the secondary structure, and several proposed tertiary interactions (Gutell et al. 1985, 1986). Equally significant this simple algorithm uncovered a few basic principles of RNA structure, namely canonical pairings, and contiguous and antiparallel arrangement of such pairings [It should be noted that this method searched for columns (nucleotide positions) with similar patterns of variation or covariance, regardless of the nucleotide and pairing types. It so happened that the underlying pairs

Crystallographic Studies on the Ribosome, a Large Macromolecular Assembly Exhibiting Severe Nonisomorphism, Extreme Beam Sensitivity and No Internal Symmetry

Crystals, diffracting best to around 3 A, have been grown from intact large and small ribosomal subunits. The bright synchrotron radiation necessary for the collection of the higher-resolution X-ray diffraction data introduces significant decay even at cryo temperatures. Nevertheless, owing to the reasonable isomorphism of the recently improved crystals of the small ribosomal subunits, reliable phases have been extracted at medium resolution (5-6 A) and an interpretable five-derivative MIR map has been constructed. For the crystals of the large subunits, however, the situation is more complicated because at higher resolution (2.7-7 A) they suffer from substantial radiation sensitivity, a low level of isomorphism, instability of the longest unit-cell axis and nonisotropic mosaicity. The 8 A MIR map, constructed to gain insight into this unusual system, may provide feasible reasoning for the odd combination of the properties of these crystals as well as hints for future improvement. Parallel efforts, in which electron-microscopy-reconstructed images are being exploited for molecular-replacement studies, are also discussed.

The suitability of multi-metal clusters for phasing in crystallography of large macromolecular assemblies

We owe exceptional gratitude to the late Professor HG Wittmann with whom we initiated these studies, to Drs M Pope and W Jahn who gave us generous gifts of heavy atom clusters and to the team working with us on this project. Data were collected at the EMBL and MPG beam lines at DESY; F1/CHESS, Cornell University; D2AM/ESRF, Grenoble; and PF/KEK, Japan. Support was provided by the Max-Planck Society, the US National Institute of Health (NIH GM 34360), the German Ministry for Science and Technology (BMFT 05-641EAC) and the Kimmelman Center for Macromolecular Assembly at the Weizmann Institute. AY holds the Martin S Kimmel Professorial Chair.

Elucidating the medium-resolution structure of ribosomal particles: an interplay between electron cryo-microscopy and x-ray crystallography

BACKGROUND Ribosomes are the universal cellular organelles that accomplish the translation of the genetic code into proteins. Electron cryo-microscopy (cryo-EM) has yielded fairly detailed three-dimensional reconstructions of ribosomes. These were used to assist in the determination of higher resolution structures by X-ray crystallography. RESULTS Molecular replacement studies using cryo-EM reconstructions provided feasible packing schemes for crystals of ribosomes and their two subunits from Thermus thermophilus, and of the large subunits from Haloarcula marismortui. For the large subunits, these studies also confirmed the major heavy-atom sites obtained by single isomorphous replacement combined with anomalous diffraction (SIRAS) and by multiple isomorphous replacement combined with anomalous diffraction (MIRAS) at approximately 10 A. Although adequate starting phases could not be obtained for the small subunits, the crystals of which diffract to 3.0 A, cryo-EM reconstructions were indispensable for analyzing their 7.2 A multiple isomorphous replacement (MIR) map. This work indicated that the conformation of the crystallized small subunits resembles that seen within the 70S ribosomes. Subsequently, crystals of particles trapped in their functionally active state were grown. CONCLUSIONS Single-particle cryo-EM can contribute to the progress of crystallography of non-symmetrical, large and flexible macromolecular assemblies. Besides confirming heavy-atom sites, obtained from flat or overcrowded difference Patterson maps, the cryo-EM reconstructions assisted in elucidating packing arrangements. They also provided tools for the identification of the conformation within the crystals and for the estimation of the level of inherent non-isomorphism.