Abstract

Antibody Equivalent RNA

Yoshikazu Nakamura

Recent crystal structure and cryo-electron microscopy studies have revealed that proteins called translation factors mimic the shape of tRNA. One of them, a polypeptide release factor, encodes a tripeptide that serves as an 'anticodon' to decipher stop codons in mRNA. These findings established the novel concept of macromolecular mimicry between protein and RNA. We aimed to prove this concept by creating novel RNA molecules that mimic proteins of interest. The systematic evolution of ligands by exponential enrichment (SELEX) method is based on in vitro selection of oligo-nucleotide ligands from large random-sequence libraries by repeated reactions of DNA transcription, RNA selection and RT-PCR amplification. The selected oligonucleotide ligands, having both high affinity and specificity to target molecules, are called 'aptamers'. We have initiated SELEX experiments using mammalian translation initiation factors including eIF4E, eIF4G, eIF1A and eIF4A. Selected RNA aptamers against these target proteins acquired several properties equivalent to, or more importantly, superior to antibodies. One of these aptamers had a Kd on the picomolar scale, an affinity which is a thousand-times stronger than normal antibody. Structural and biochemical analysis revealed that these aptamers need to be > 50 nucleotides long for specific and high affinity binding to their target proteins. Therefore, it might be argued that RNA aptamers to proteins without RNA recognition motifs or strong affinity to RNA can achieve specific high affinity to the target protein by capturing its global conformation. This is completely different from the pinpoint (i.e., epitope < 10 amino acids) recognition of target protein by antibodies. Previous, and ongoing studies (in this laboratory) of RNA aptamers to mammalian initiation factors and cell surface receptors contribute to strengthening the concept of conformational recognition of target proteins by RNA aptamer. For this reason, the RNA aptamer has promising potential to substitute for or complement the antibody as a new diagnostic or therapeutic tool that we refer to as "RNA super-antibody". We speculate that such a "natural aptamer" might be largely encoded in noncoding RNAs of current interest.

Autoantibodies in Connective Tissue Diseases as Useful Probes Analyzing Ribonucleoproteins

Tsuneyo Mimori

Autoantibodies directed against various cellular components are found in sera from connective tissue diseases. These autoantibodies have been demonstrated to be closely associated with certain diseases and clinical manifestations, and give us useful information for clinical practice such as to diagnose diseases and to predict clinical subsets, disease activity and prognosis. The development of new technologies for detecting autoantibodies has been able to identify more than 100 of autoantibodies and their target autoantigens. In recent years, the nature of many autoantigens has been clarified. Most of them are complexes with RNA and proteins (ribonucleoprotein), and act as intracellular enzymes or regulatory factors necessary for important biological function involved in gene replication, transcription, RNA processing and protein translation. Actually, almost all known ribonucleoproteins have been demonstrated to be the targets of autoantibodies, and most of such autoantibodies to ribonucleoproteins are especially important in clinical medicine. Patient sera containing these autoantibodies have been useful probes to analyze and study the structure and function of unknown ribonucleoprotein molecules. For example, anti-U1RNP and Sm antibodies found in SLE patients were used to analyze the precise mechanism of mRNA splicing. Thus, autoantibodies are useful not only in clinical medicine but also in basic cellular and molecular biology.

Connection between RNAi and Fragile X Syndrome

Haruhiko Siomi

Fragile X syndrome is the most common form of heritable mental retardation caused by loss-of-function mutations in the FMR1 gene. The FMR1 gene encodes an RNA-binding protein that associates with translating ribosomes and acts as a negative translational regulator. Recent work in Drosophila melanogaster has shown that the fly homolog of FMR1 (dFMR1) plays an important role in regulating neuronal morphology and function, which may underlie the observed deficits in behaviors of dFMR1 mutant flies. Biochemical analysis has revealed that dFMR1 forms a complex with AGO2, an essential component of RNA interference (RNAi) pathway in Drosophila. AGO2 associates tightly with exogenous and probably endogenous siRNAs and forms the RNA-induced silencing complex (RISC) that mediates cleavage of target RNAs. These findings suggest that dFMR1 may function in an RNAi-related apparatus to regulate the expression of its target genes. To test this, we are currently engaged in identifying small RNAs present in dFMR1/AGO2-associated complexes. The sequence information of these small RNAs should help determine target genes of dFMR1/AGO2-associated complexes.

Biological Aspects of the Frameshift Protein, Antizyme

Senya Matsufuji

Polyamines (putrescine, spermidine, and spermine) are ubiquitous and essential biogenic polycation. The concentrations of cellular polyamines significantly fluctuate according to growth states of the cells. At the same time, cellular polyamine contents are under negative feedback control, which is mediated by unique regulatory proteins, antizymes. Antizymes are induced by polyamines and negatively regulate polyamines by both repressing a key enzyme of polyamine synthesis, ornithine decarboxylase (ODC), and inhibiting cellular polyamine uptake. Antizyme is conserved among a wide range of eukaryotes, from yeast to human. In mammals, the antizyme family consists of three members. AZ1 and AZ2 are expressed systemically, while AZ3 only in male germ cells. We have demonstrated that polyamine-dependent translational frameshifting is required for the synthesis and regulation of antizyme. The mRNA of rat AZ1 has an in-frame UGA terminator at codon 69 and the remaining two thirds of the protein is encoded on the +1 reading frame. The frameshifting occurs at the last codon of the initiating frame, translating the mRNA sequence UCC UGA U as serine-aspartate. This event is signaled by the sequence of the frameshift site and enhanced by cis-acting elements on the mRNA, a downstream pseudoknot structure and an upstream GC-rich segment. The frameshifting is stimulated by polyamines, a maximal efficiency being up to 40%. Such stimulation by polyamines are not seen for -1 frameshifting, which is widely utilized by viral gene expressions.
The phylogenetic conservation and the intrinsic energy waste of antizyme-mediated polyamine regulation suggest its important role. To clarify the biological significance of the antizyme system, we prepared knockout mice of antizymes. Knockout of AZ1, the most abundant form of antizyme, caused systemic increases in polyamine synthesis and contents, resulting in partial embryonic mortality at a late embryonic stage (E15.5-18.5). The AZ1-deficient embryos were accompanied by anemia and liver hypoplasia. The erythroid differentiation were affected at multiple stages. The findings indicate that the polyamine regulation by AZ1 is crucial for the normal development, but unexpectedly, involved in rather specific differentiation processes.
Knockout of the minor isoform, AZ2, brought about no apparent phenotype, while AZ1-AZ2 double homozygous knockout mice were completely embryonic lethal around E13.5. This observation may suggest that AZ2 is a backup of AZ1. However, our recent study revealed some unique features of AZ2; it is dominantly localized in the nucleus while AZ1 is cytoplasmic, and AZ2, but not AZ1, is phosphorylated under physiological conditions. Yeast two-hybrid screening identified an onconeural antigen, cdr2, as an interacting protein with AZ2. Binding of cdr2 to AZ2 is strong and specific (it does not bind to AZ1 nor AZ3), and require its helix-leucine zipper domain. Since cdr2 has been implicated in negative regulation of c-Myc, its interaction with AZ2 might be a novel connection between the polyamines and cell proliferation.

RNA Silencing in Drosophila

Mikiko C. Siomi

Gene silencing pathways triggered by small RNAs are generically called RNA silencing. RNA interference (RNAi) triggered by short-interfering RNA (siRNA) of 21- to 23- nucleotide (nt) in length is a representative of it. Extensive studies on RNA silencing mechanisms revealed that members of the Argonate family play important roles in the pathways. In Drosophila, the Argoanute family, defined by the presence of PAZ and PIWI domains, consists of five members, which includes AGO1, AGO2, AGO3, Piwi, and Aubergine. One of our goals is to understand the functional differences among the fly Argonaute members in RNA silencing.
Previously we have shown that AGO1 and AGO2 in Drosophila function in gene silencing through specific binding with miRNA and siRNA, respectively. siRNA-loaded AGO2 functions in RNAi as Silcer, directly responsible for cleaving a target completely complementary to siRNA. miRNA-associated AGO1 is thought to repress translation of target mRNAs without cleaving them, but AGO1 also possesses Slicer activity as does AGO2. Recently we reported that Piwi, an essential factor in germline stem cell (GSC) self-renewal, is associated with repeat-associated siRNAs (rasiRNAs) derived from a variety of repetitive elements in the genome, and that recombinant Piwi protein produced in E.coli is able to exhibit Slicer activity in vitro. Nuclear localization of Piwi endogenously expressed in ovaries was confirmed by immunostaining with a specific antibody against it. Taken together, the results imply that Piwi functions in nuclear RNA silencing as Slicer by associating specifically with rasiRNAs that originate from the repetitive targets. We are currently engaged in identifying small RNAs specifically associated with Aubergine and AGO3.

Musashi and Hu: Translational Regulators of Cell Fates.

Hideyuki Okano

Musashi is an evolutionarily conserved family of RNA-binding proteins that is preferentially expressed in the nervous system (Reviewed by Okano et al., J. Cell Sci. 2002; Exp. Cell Res. 2005). The first member of the Musashi family was identified in Drosophila. This protein plays an essential role in regulating the asymmetric cell division of ectodermal precursor cells known as sensory organ precursor cells through the translational regulation of target mRNA (Nakamura et al., Neuron, 1994; Okabe, Imai et al. Nature, 2001). Its mammalian homologue, Musashi-1, is a neural RNA-binding protein that is strongly expressed in fetal and adult neural stem cells (NSCs). Mammalian Musashi-1 (Sakakibara et al., Dev. Biol., 1996; Proc.Natl.Acad.Sci.USA, 2002) augments Notch signaling through the translational repression of its target mRNA, m-Numb, thereby contributing to the self-renewal of NSCs (Imai et al. Mol. Cell. Biol., 2001). Musashi1 protein was also shown to be present in the translational machinery using sucrose gradient centrifugation analysis. As regards the mechanism of Musashi-1-mediated translational repression, we identified the poly(A)-binding protein (PABP) which is one of the proteins co-precipitated with Musashi-1. The PABP/eukaryotic initiation factor (eIF4G) interaction plays in key steps of the translation initiation pathway, which brings about the circularization of the mRNA. We showed the C-terminal region of Musashi-1 interacts with PABP, which possesses common binding domains for eIF4G. It is suggested that Musashi-1 acts in the translation initiation machinery. Further analysis has shown that Musashi-1 repressed the cap structure-dependent translation to the mRNA including the cap structure, poly(A)-tail and Musashi1-binding sequence.
In addition to its functions in NSCs, the role of mammalian Musashi-1 protein in epithelial stem cells, including intestinal and mammary gland stem cells, is attracting increasing interest. In this symposium, we also refer to our recent studies on the function of Hu family, a group of neuronal RNA-binding proteins required for neuronal differentiation in the developing nervous system (Akamatsu et al., Proc.Natl.Acad.Sci.USA, 1999; Akamatsu et al., Proc.Natl.Acad.Sci.USA, 2005) . Previously, Hu proteins have been shown to enhance the stabilization and/or translation of target mRNAs, such as p21 (CIP1), by binding to AU-rich elements in untranslated regions (UTRs) (Yano et al., J. Biol. Chem., 2005). Our results suggest a model in which the mutually antagonistic action of two RNA-binding proteins, Hu and hnRNP K, control the timing of the switch from proliferation to neuronal differentiation through the post-transcriptional regulation of p21 mRNA.

Molecular Evolution of Transfer RNAs in Archaea

Akio Kanai

It has been proposed that the last universal common ancestor (LUCA) was a thermophile or a hyperthermophile. Thus, genome-wide analysis of gene function in very ancient organisms such as the hyperthermophilic archaeon Pyrococcus species that thrive at temperatures above 95□C could lead to new insights into the fundamental knowledge of life. Moreover, recent findings of huge amount of non-coding RNAs and accumulating research articles for gene regulations at the RNA level promoted the importance of the concept of "RNA world" in the beginning of life. We have conducted expression cloning to screen a recombinant protein library for identifying RNA-related enzymes of Pyrococcus furiosus. An activity that changes the oligoribonucleotide migration on polyacrylamide gel containing 8 M urea was found in the library; and the corresponding gene PF0027 (termed Pf-ligR) that possessed heat-stable cyclic nucleotide phosphodiesterase (CPDase) activity was identified. The Pf-ligR gene encoded a 184-amino-acid protein with a calculated molecular mass of 21 kDa and an amino acid sequence that showed about 27% identity to that of the 2'-5' tRNA ligase protein, ligT (20 kDa), from Escherichia coli. We detected that Pf-ligR protein also possessed GTP-dependent tRNA ligase activity and found that synthetic tRNA halves bearing 2',3'-cyclic phosphate and 5'-OH termini were substrates for the ligation reaction. We determined the solution structure of Pf-ligR protein and the possible mechanism of tRNA ligation will be discussed.
In archaea, some tRNA precursors contain intron(s) not only in the anticodon loop region but also in diverse sites of the gene (intron-containing tRNA or cis-spliced tRNA). The parasite Nanoarchaeum equitans, which is a member of the Nanoarchaeota kingdom, creates a functional tRNA from separate genes, one encoding the 5'-half and the other encoding the 3'-half (so-called split tRNA or trans-spliced tRNA). We have developed SPLITS program, which is aimed at searching for any type of tRNA gene and is especially focused on intron-containing tRNAs or split tRNAs at the genome level. Using this software, we have collected more than 1,000 tRNA sequences from the complete genome sequences of 30 archaeal species and suggested that 5' and 3' halves of tRNA genes could have emerged independently. Moreover, we would like to propose that combinations of 5' and 3' halves of certain tRNA genes may generate mature tRNA genes through evolution.

Structural Biology of Protein Synthesis Systems

Shigeyuki Yokoyama

Protein synthesis systems consist of a large number of proteins and RNAs. The ribosome synthesizes polypeptide chains using aminoacyl-tRNAs, mRNA, and translation factors. Each of the twenty kinds of amino acids is ligated to its specific tRNA(s) by the cognate aminoacyl-tRNA synthetase (aaRS). aaRS first synthesizes aminoacyl-AMP from amino acid and ATP as the intermediate ("amino-acid activation"). Most aaRSs can catalyze amino-acid activation in the absence of tRNA, but three aaRSs specific for Glu, Gln, and Arg require the cognate tRNA for amino-acid activation. We determined the crystal structures of Thermus thermophilus glutamyl-tRNA synthetase (GluRS) complexed with tRNA, L-glutamate, ATP, and/or glutamyl-AMP analogues. Upon tRNA binding, the catalytic-site structure is changed so that the ATP-binding mode is switched from non-productive to productive. Furthermore, GluRS alone can bind not only L-glutamate but also D-glutamate and glutamine, whereas tRNA changes the amino-acid binding mode to be specific to L-glutamate. Therefore, the GluRS・tRNA complex is the functional enzyme for L-glutamate activation.
Isoleucyl-tRNA synthetase (IleRS), valyl-tRNA synthetase (ValRS), and leucyl-tRNA synthetase (LeuRS) have the "editing" activity to hydrolyze aminoacyl-tRNA formed with incorrect amino acid (e.g. Val for IleRS, Thr for ValRS, and Ile for LeuRS). We determined crystal structures of tRNA complexes of ValRS and LeuRS and the IleRS and ValRS editing domains, which revealed how these aaRSs recognize and hydrolyze incorrectly-formed aminoacyl-tRNAs.
We engineered bacterial and archaeal tyrosyl-tRNA synthetases in order to incorporate non-natural tyrosine analogues site-specifically into proteins with nonsense suppressor tRNA. 3-Iodotyrosine is thus incorporated into proteins, which is useful for phase determination in crystallographic analyses. A photo-crosslinker is also incorporated into a cell signaling protein in mammalian cells, for analysis of its complex formation with other proteins.
The crystal structures of elongation factor G-2 (EF-G-2, a homologue of EF-G) and the Era protein from T. thermophilus in the GTP-bound and GTP analogue-bound forms, respectively. The roles of GTP hydrolysis on the ribosome will be discussed for these two proteins.

Quinolone antimicrobials:
Mode of action and resistance mechanisms in bacteria

Yoshikuni Onodera

Quinolones have a broad spectrum of antibacterial activity, and now more than 10 congeners that are approved for clinical use around the world. On the other hand, resistance to quinolones has also emerged among some pathogens, as these antimicrobial agents are being used extensively in the clinical situation.
The target enzymes of quinolones are DNA gyrase and topoisomerase IV, both of which are essential for bacterial DNA replication. As the relative potency of quinolones against two target enzymes within bacterial cells are different, the quinolone resistance occurs stepwise by mutations in the two target enzymes genes, with the first mutation generally occurring in the more sensitive enzyme. From the first-step mutants second-step double mutants can then be selected with resistance mutations in the second target enzyme, thereby conferring a high-level resistance phenotype.
Efflux pumps also play a significant role in acquired clinical resistance. They appear to be ubiquitous among bacteria, and activation of them are affected either by substrate induction, or mutations in regulatory genes.
To overcome these problems, some approaches for the development of antibiotics have been advanced. One approach is developing a novel quinolone, which has dual inhibitory activity against two target enzymes as well as increasing activity against mutated enzymes. If the original sensitivities of both DNA gyrase and topoisomerase IV were the same, no single mutational alteration in either enzyme would result in an increase in the MIC. Resistance would require, instead, concurrent alteration in both enzymes.
Development of inhibitor of efflux pumps, which is used in conjunction with quinolones, is another approach to overcome quinolone resistance. Some compounds were shown to inhibit four clinically relevant Pseudomonas aeruginosa efflux pumps as well as other pumps from gram-negative bacteria. It increased quinolones susceptibility up to 64-fold in efflux pump overexpression strains of P. aeruginosa.
Other structural classes of topoisomerases inhibitors have also been explored to conquer quinolone resistant strains. To identify novel bacterial topoisomerase inhibitors, high-throughput screening and structure-based drug design have been conducted.