Takanori Oyoshi

Structure of noncoding RNA is a determinant of function of RNA binding proteins in transcriptional regulation.

The majority of the noncoding regions of mammalian genomes have been found to be transcribed to generate noncoding RNAs (ncRNAs), resulting in intense interest in their biological roles. During the past decade, numerous ncRNAs and aptamers have been identified as regulators of transcription. 6S RNA, first described as a ncRNA in E. coli, mimics an open promoter structure, which has a large bulge with two hairpin/stalk structures that regulate transcription through interactions with RNA polymerase. B2 RNA, which has stem-loops and unstructured single-stranded regions, represses transcription of mRNA in response to various stresses, including heat shock in mouse cells. The interaction of TLS (translocated in liposarcoma) with CBP/p300 was induced by ncRNAs that bind to TLS, and this in turn results in inhibition of CBP/p300 histone acetyltransferase (HAT) activity in human cells. Transcription regulator EWS (Ewings sarcoma), which is highly related to TLS, and TLS specifically bind to G-quadruplex structures in vitro. The carboxy terminus containing the Arg-Gly-Gly (RGG) repeat domains in these proteins are necessary for cis-repression of transcription activation and HAT activity by the N-terminal glutamine-rich domain. Especially, the RGG domain in the carboxy terminus of EWS is important for the G-quadruplex specific binding. Together, these data suggest that functions of EWS and TLS are modulated by specific structures of ncRNAs.

Identification of Ewing’s sarcoma protein as a G-quadruplex DNA- and RNA-binding protein

The Ewing’s sarcoma (EWS) oncogene contains an N‐terminal transcription activation domain and a C‐terminal RNA‐binding domain. Although the EWS activation domain is a potent transactivation domain that is required for the oncogenic activity of several EWS fusion proteins, the normal role of intact EWS is poorly characterized because little is known about its nucleic acid recognition specificity. Here we show that the Arg‐Gly‐Gly (RGG) domain of the C‐terminal in EWS binds to the G‐rich single‐stranded DNA and RNA fold in the G‐quadruplex structure. Furthermore, inhibition of DNA polymerase on a template containing a human telomere sequence in the presence of RGG occurs in an RGG concentration‐dependent manner by the formation of a stabilized G‐quadruplex DNA–RGG complex. In addition, mutated RGG containing Lys residues replacing Arg residues at specific Arg‐Gly‐Gly sites and RGG containing Arg methylated by protein arginine N‐methyltransferase 3 decrease the binding ability of EWS to G‐quadruplex DNA and RNA. These findings suggest that the RGG of EWS binds to G‐quadruplex DNA and RNA via the Arg residues in it.

G-Quadruplex DNA- and RNA-Specific-Binding Proteins Engineered from the RGG Domain of TLS/FUS

Human telomere DNA (Htelo) and telomeric repeat-containing RNA (TERRA) are integral telomere components containing the short DNA repeats d(TTAGGG) and RNA repeats r(UUAGGG), respectively. Htelo and TERRA form G-quadruplexes, but the biological significance of their G-quadruplex formation in telomeres is unknown. Compounds that selectively bind G-quadruplex DNA and RNA are useful for understanding the functions of each G-quadruplex. Here we report that engineered Arg-Gly-Gly repeat (RGG) domains of translocated in liposarcoma containing only Phe (RGGF) and Tyr (RGGY) specifically bind and stabilize the G-quadruplexes of Htelo and TERRA, respectively. Moreover, RGGF inhibits trimethylation of both histone H4 at lysine 20 and histone H3 at lysine 9 at telomeres, while RGGY inhibits only H3 trimethylation in living cells. These findings indicate that G-quadruplexes of Htelo and TERRA have distinct functions in telomere histone methylation.

Supramolecular gel electrophoresis of acidic native proteins.

Amphiphilic tris-urea molecules self-assemble into a supramolecular hydrogel in tris(hydroxymethyl)aminomethane-glycine buffer. The supramolecular hydrogel is used as a matrix for the electrophoresis of acidic native proteins, in which proteins are separated based on their isoelectric points rather than their molecular weights. The proteins remain in their native forms during migration, and their activities are retained after electrophoresis. Glucoside substituents on the amphiphilic tris-urea molecule allow for the affinity electrophoresis of a carbohydrate-binding protein to be performed. The proteins can be efficiently recovered from the supramolecular hydrogel using a simple procedure. This is a major advantage of using this noncovalent, self-assembled material.

Analysis of Guanine Oxidation Products in Double-Stranded DNA and Proposed Guanine Oxidation Pathways in Single-Stranded, Double-Stranded or Quadruplex DNA

Guanine is the most easily oxidized among the four DNA bases, and some guanine-rich sequences can form quadruplex structures. In a previous study using 6-mer DNA d(TGGGGT), which is the shortest oligomer capable of forming quadruplex structures, we demonstrated that guanine oxidation products of quadruplex DNA differ from those of single-stranded DNA. Therefore, the hotooxidation products of double-stranded DNA (dsDNA) may also differ from that of quadruplex or single-stranded DNA, with the difference likely explaining the influence of DNA structures on guanine oxidation pathways. In this study, the guanine oxidation products of the dsDNA d(TGGGGT)/d(ACCCCA) were analyzed using HPLC and electrospray ionization-mass spectrometry (ESI-MS). As a result, the oxidation products in this dsDNA were identified as 2,5-diamino-4H-imidazol-4-one (Iz), 8-oxo-7,8-dihydroguanine (8oxoG), dehydroguanidinohydantoin (Ghox), and guanidinohydantoin (Gh). The major oxidation products in dsDNA were consistent with a combination of each major oxidation product observed in single-stranded and quadruplex DNA. We previously reported that the kinds of the oxidation products in single-stranded or quadruplex DNA depend on the ease of deprotonation of the guanine radical cation (G•+) at the N1 proton. Similarly, this mechanism was also involved in dsDNA. Deprotonation in dsDNA is easier than in quadruplex DNA and more difficult in single-stranded DNA, which can explain the formation of the four oxidation products in dsDNA.

Identification of RNA Binding Specificity for the TET-family Proteins

The TET-family proteins (TAF15, EWS and TLS) are the RNA binding proteins involved in multiple levels of cellular functions. The RNA binding domain of those proteins is known as the important region for cellular functions. But little is known about the RNA binding specificity of TET-family proteins. In order to investigate the RNA binding properties of the TET-family proteins, we performed electrophoretic mobility shift assay using recombinant Flag-tagged TLS and guanine-rich and RNAs. It was found that TLS binds to human telomeric RNA in the presence of KCl, but not in the presence of LiCl.

Product analysis of photooxidation in isolated quadruplex DNA; 8-oxo-7,8-dihydroguanine and its oxidation product at 3′-G are formed instead of 2,5-diamino-4H-imidazol-4-one

The formation of quadruplex structure changed the site reactivity and the kinds of guanine photooxidation products of d(TGGGGT). In quadruplex DNA, 8-oxo-7,8-dihydroguanine (8oxoG) and dehydroguanidinohydantoin (Ghox) were mainly formed, although 2,5-diamino-4H-imidazol-4-one (Iz) was mainly formed in single-stranded DNA. In addition, 3′-guanine was specifically oxidized in quadruplex DNA compared with single-stranded DNA, which depended on the localization of the HOMO.

Plastic roles of phenylalanine and tyrosine residues of TLS/FUS in complex formation with the G-quadruplexes of telomeric DNA and TERRA

The length of a telomere is regulated via elongation and shortening processes. Telomeric DNA and telomeric repeat-containing RNA (TERRA), which both contain G-rich repeated sequences, form G-quadruplex structures. Previously, translocated in liposarcoma (TLS) protein, also known as fused in sarcoma (FUS) protein, was found to form a ternary complex with the G-quadruplex structures of telomeric DNA and TERRA. We then showed that the third RGG motif of TLS, the RGG3 domain, is responsible for the complex formation. However, the structural basis for their binding remains obscure. Here, NMR-based binding assaying revealed the interactions in the binary and ternary complexes of RGG3 with telomeric DNA or/and TERRA. In the ternary complex, tyrosine bound exclusively to TERRA, while phenylalanine bound exclusively to telomeric DNA. Thus, tyrosine and phenylalanine each play a central role in the recognition of TERRA and telomeric DNA, respectively. Surprisingly in the binary complexes, RGG3 used both tyrosine and phenylalanine residues to bind to either TERRA or telomeric DNA. We propose that the plastic roles of tyrosine and phenylalanine are important for RGG3 to efficiently form the ternary complex, and thereby regulate the telomere shortening.

G-quadruplex binding ability of TLS/FUS depends on the β-spiral structure of the RGG domain

Abstract The RGG domain, defined as closely spaced Arg-Gly-Gly repeats, is a DNA and RNA-binding domain in various nucleic acid-binding proteins. Translocated in liposarcoma (TLS), which is also called FUS, is a protein with three RGG domains, RGG1, RGG2 and RGG3. TLS/FUS binding to G-quadruplex telomere DNA and telomeric repeat-containing RNA depends especially on RGG3, comprising Arg-Gly-Gly repeats with proline- and arginine-rich regions. So far, however, only non-specific DNA and RNA binding of TLS/FUS purified with buffers containing urea and KCl have been reported. Here, we demonstrate that protein purification using a buffer with high concentrations of urea and KCl decreases the G-quadruplex binding abilities of TLS/FUS and RGG3, and disrupts the β-spiral structure of RGG3. Moreover, the Arg-Gly-Gly repeat region in RGG3 by itself cannot form a stable β-spiral structure that binds to the G-quadruplex, because the proline- and arginine-rich regions induce the β-spiral structure and the G-quadruplex-binding ability of RGG3. Our findings suggest that the G-quadruplex-specific binding abilities of TLS/FUS require RGG3 with a β-spiral structure stabilized by adjacent proline- and arginine-regions.

Supramolecular gel electrophoresis of large DNA fragments

Pulsed‐field gel electrophoresis is a frequent technique used to separate exceptionally large DNA fragments. In a typical continuous field electrophoresis, it is challenging to separate DNA fragments larger than 20 kbp because they migrate at a comparable rate. To overcome this challenge, it is necessary to develop a novel matrix for the electrophoresis. Here, we describe the electrophoresis of large DNA fragments up to 166 kbp using a supramolecular gel matrix and a typical continuous field electrophoresis system. C3‐symmetric tris‐urea self‐assembled into a supramolecular hydrogel in tris‐boric acid‐EDTA buffer, a typical buffer for DNA electrophoresis, and the supramolecular hydrogel was used as a matrix for electrophoresis to separate large DNA fragments. Three types of DNA marker, the λ‐Hind III digest (2 to 23 kbp), Lambda DNA‐Mono Cut Mix (10 to 49 kbp), and Marker 7 GT (10 to 165 kbp), were analyzed in this study. Large DNA fragments of greater than 100 kbp showed distinct mobility using a typical continuous field electrophoresis system.