Research Overview

Exploring the structural language of DNA and RNA to reveal the hidden codes of cellular function.

The Oyoshi Laboratory explores how the three-dimensional structures of DNA and RNA encode regulatory information that governs cellular function. Beyond the genetic sequence, nucleic acids can fold into higher-order architectures such as G-quadruplexes, which act as structural signals that are selectively recognized by proteins and translated into functional outcomes.

Our research is organized around four interconnected themes. We first focus on identifying (1) protein readers of G-quadruplex structures, uncovering how specific proteins selectively recognize non-canonical DNA and RNA conformations. We then investigate how G-quadruplexes function as (2) structural platforms for chromatin regulation, particularly at telomeres, where they coordinate protein and RNA interactions to control epigenetic states and genome stability.

At the molecular level, we dissect the (3) structural basis of G-quadruplex recognition by RGG domains, revealing how intrinsically disordered protein regions achieve selective binding through defined sequence and conformational features. Building on these mechanistic insights, we further pursue the (4) design of structure-selective peptides for G-quadruplex DNA and RNA, developing molecular probes that enable functional dissection of G-quadruplex roles in living cells.

By integrating protein identification, structural analysis, functional studies, and molecular tool development, our laboratory aims to establish a unified framework for understanding how the architecture of nucleic acids serves as a regulatory language in biology.

(1) Protein Readers of G-quadruplex Structures

G-quadruplexes (G4s) are unusual DNA and RNA structures formed by guanine-rich sequences. Unlike the familiar double helix, these structures fold into compact, square-shaped assemblies, and their formation can dramatically influence how genes and RNAs behave in cells. G-quadruplex structures are found not only in genomic DNA but also in mRNAs and non-coding RNAs, suggesting that they play broad roles in gene regulation.Accumulating evidence shows that G4 DNA is involved in transcriptional regulation, telomere maintenance, epigenetic control, and DNA replication. In parallel, G4 RNA has been implicated in diverse RNA-related processes such as transcription, RNA processing, subcellular localization, and translational regulation. Importantly, G-quadruplex structures are not static. In living cells, they are dynamically folded and unfolded through interactions with specific binding proteins, and this dynamic regulation is thought to be essential for their biological functions.

To understand how G-quadruplexes function in cells, it is therefore crucial to identify and characterize proteins that specifically recognize these structures. We have identified several G-quadruplex binding proteins, including Translocated in Liposarcoma/Fused in Sarcoma (TLS/FUS) and Ewing’s Sarcoma (EWS). These proteins share common structural features, such as RNA recognition motifs (RRMs) and Arg–Gly–Gly (RGG) domains, which are frequently found among G-quadruplex binding proteins.

Our studies have shown that specific amino acid residues within the RGG domain—particularly tyrosine, phenylalanine, and arginine—are critical for G-quadruplex recognition. Furthermore, recent findings suggest that the RGG domain does not bind G-quadruplexes simply as a flexible chain. Instead, it can adopt a β-turn–like structure, a small folded motif in which the peptide backbone bends sharply. For non-specialists, this β-turn can be thought of as a “corner” or “hinge” that allows the protein to fit the unique square-shaped surface of a G-quadruplex more precisely, thereby enabling selective recognition of these structures.

In addition to canonical RNA-binding proteins, we have discovered that Fibrillarin (FBL)—a core component of small nucleolar ribonucleoproteins and a key factor in ribosomal RNA processing—also binds G-quadruplex RNA. Using a G-quadruplex–based proximity labeling strategy, we selectively labeled and identified proteins in close proximity to G4 RNA, revealing fibrillarin as a previously unrecognized G-quadruplex-binding protein. This finding suggests that G-quadruplex RNA may function in unexpected cellular contexts, such as the nucleolus, and that G4-mediated interactions extend beyond classical RNA-binding proteins.

Through these studies, we aim to elucidate how diverse proteins recognize the structural features of DNA and RNA, rather than just their sequences, and how this structure-based recognition contributes to gene regulation and RNA function in cells.

DNA- and RNA-binding protein. Takahama, K., Kino, K., Arai, S., Kurokawa, R., Oyoshi, T. FEBS. J. 2011, 278, 988-998.

Loop lengths of G-quadruplex structures affect the G-quadruplex DNA binding selectivity of the RGG motif in Ewing’s sarcoma. Takahama, K., Sugimoto, C., Arai, S., Kurokawa, R., Oyoshi, T. Biochemistry 2011, 50, 5369-5378.

G-quadruplex-proximity protein labeling based on peroxidase activity. Masuzawa, T., Sato, S., Niwa, T., Taguchi, H., Nakamura, H., Oyoshi, T. Chem. Commun. 2020, 56, 11641-11644.

 

(2) G-quadruplexes as Structural Platforms for Chromatin Regulation

G-quadruplex structures can act as molecular platforms that organize proteins and RNAs to control chromatin structure and genome stability. Rather than being passive structural features, G-quadruplexes (G4s) actively participate in the regulation of gene expression by recruiting specific binding proteins and chromatin-modifying factors.

Our studies have demonstrated that TLS/FUS and Ewing’s sarcoma protein (EWS) specifically recognize G-quadruplex structures formed by telomeric DNA and telomeric repeat-containing RNA (TERRA). Through structure-specific binding, these proteins form stable ternary complexes with telomeric DNA and TERRA, thereby acting as molecular hubs that connect nucleic acids to chromatin regulatory machinery.

In particular, TLS/FUS directly interacts with the histone methyltransferase Suv4-20h2 as well as with TERRA, promoting the deposition of repressive histone marks such as trimethylation of histone H4 at lysine 20 (H4K20me3) and histone H3 at lysine 9 (H3K9me3). These modifications are hallmarks of heterochromatin and are essential for maintaining proper telomeric chromatin structure and telomere length control.

More recently, we showed that EWS regulates telomeric chromatin in a distinct but related manner. EWS binds both G4 DNA and G4 RNA at telomeres and modulates TERRA transcription in a chromosome-dependent manner. By interacting with histone acetyltransferases such as CBP/p300, EWS promotes histone H3 lysine 27 acetylation (H3K27Ac) at telomeres, a modification associated with transcriptionally active chromatin. These findings reveal that G-quadruplex-binding proteins can fine-tune telomeric chromatin states by balancing repressive and active histone modifications through G4-dependent interactions.

Together, these studies demonstrate that G-quadruplex structures function as active regulatory elements that integrate DNA, RNA, and protein interactions to control chromatin architecture and gene expression. By elucidating how G-quadruplexes organize epigenetic regulators at telomeres, our work highlights a fundamental mechanism by which nucleic acid structure contributes to genome stability and transcriptional regulation.

Regulation of Telomere Length by G-Quadruplex Telomere DNA- and TERRA-Binding Protein TLS/FUS. Takahama, K., Takada, A., Tada, S., Shimizu, M., Sayama, K., Kurokawa, R., Oyoshi, T. Chem. Biol. 2013, 20, 341-350. 

 Transcriptional regulation of telomeric repeat-containing RNA by the G-quadruplex-binding Ewing sarcoma protein. Ulum, L. L., Matsudaira, W., Yamanashi, M., Shibata, N., Saha, S., Ishihara, A., Oyoshi, T. Sci. Rep. 2026, 16, 715.

 

(3) Structural Basis of G-quadruplex Recognition by RGG Domains

The Arg–Gly–Gly (RGG) domain is an evolutionarily conserved sequence motif frequently found in DNA- and RNA-binding proteins. Proteins containing RGG domains are involved in diverse nucleic acid–related processes, including transcription, RNA processing, and chromatin regulation. However, the presence of an RGG domain alone does not necessarily confer G-quadruplex (G4) binding activity.

Our studies have shown that the C-terminal RGG domains of TLS/FUS and EWS specifically bind to G-quadruplex DNA and RNA, indicating that G4 recognition by RGG domains requires additional structural features beyond the simple presence of Arg–Gly–Gly repeats. In other words, not all RGG domains are functionally equivalent with respect to G-quadruplex binding.

Through systematic analyses, we have identified several key features that are critical for G-quadruplex recognition by RGG domains. First, tandem Arg–Gly–Gly repeats provide a multivalent array of arginine residues that can interact with the negatively charged phosphate backbone and guanine bases of G-quadruplex structures. Second, the presence of aromatic residues such as phenylalanine and tyrosine near the RGG repeats is essential, likely stabilizing G-quadruplex binding through π–π stacking interactions with guanine tetrads. Third, and importantly, the RGG domain can adopt a β-turn–like structure, which introduces a defined bend in the peptide backbone. For non-specialists, this β-turn can be viewed as a small “corner” that allows the RGG domain to fit the unique square-shaped surface of a G-quadruplex more effectively.

Together, these features enable specific RGG domains to recognize G-quadruplex structures with high selectivity. These findings reveal how intrinsically disordered RGG domains achieve structural selectivity in recognizing higher-order nucleic acid structures, providing a molecular basis for understanding how flexible protein regions can mediate highly specific nucleic acid recognition.

G-quadruplex Binding Ability of TLS/FUS Depends on the β-Spiral Structure of the RGG Domain. Yagi, R., Miyazaki, T., Oyoshi, T. Nucleic Acids Res. 2018, 46, 5894-5901.

Roles of the RGG domain and RNA recognition motif of Nucleolin in G‑quadruplex stabilization. Masuzawa, T. and Oyoshi, T. ACS Omega 2020, 5, 5202-5208.

 

(4) Design of Structure-Selective Peptides for G-quadruplex DNA and RNA

Based on our findings on the G-quadruplex binding properties of the RGG domain, we have developed engineered peptides derived from the RGG region of TLS/FUS that selectively recognize G-quadruplex structures. By rationally designing polypeptides containing tyrosine or phenylalanine residues, we succeeded in generating peptides that preferentially bind G-quadruplex DNA or G-quadruplex RNA, respectively.

These structure-selective peptides enabled us to dissect the distinct roles of DNA and RNA G-quadruplexes at telomeres. Using these molecules as molecular probes, we revealed that G-quadruplex DNA and G-quadruplex RNA contribute differently to histone modification pathways, highlighting functional differences between DNA- and RNA-based G-quadruplex structures in chromatin regulation. Importantly, we also demonstrated that a G4 DNA–binding peptide can suppress transcription of the bcl-2 gene, whose promoter region is known to form a G4 structure. This result provides a clear example that targeting G4 structures with designed molecules can directly modulate the expression of specific genes.

Because these peptides are derived from naturally occurring G-quadruplex-binding domains, they provide a versatile platform for selectively targeting G-quadruplex structures in living cells. We expect that these molecules will serve as powerful tools for elucidating the biological functions of G-quadruplex DNA and RNA. In addition, this peptide-based strategy offers a foundation for the future development of G-quadruplex–targeting molecular probes, with potential relevance to biomedical research. This peptide-based approach opens new possibilities for selectively probing DNA and RNA G-quadruplex functions in living cells.

Specific binding of modified RGG domain in TLS/FUS to G-quadruplex RNA: Tyrosines in RGG domain recognize 2’-OH of the riboses of loops in G-quadruplex. Takahama, K., Oyoshi, T. J. Am. Chem. Soc. 2013, 135, 18016-18019.

G-quadruplex DNA- and RNA-specific-binding proteins engineered from the RGG domain of TLS/FUS. Takahama, K., Miyawaki, A., Shitara, T., Mitsuya, K., Morikawa, M., Hagihara, M., Kino, K., Yamamoto, A., Oyoshi, T. ACS Chem. Biol. 2015, 10, 2564-2569.

DNA G‐Quadruplex-Binding Protein Developed Using the RGG Domain of Translocated in Liposarcoma/Fused in Sarcoma Inhibits Transcription of bcl‐2. Ulum, L. L., Karikome, Y., Yagi, R., Kawashima, T., Ishihara, A., Oyoshi, T. ACS Omega 2023, 8, 10459-10465. 

Identification of Guanine-Quadruplex-Binding Peptides from the RGG3 Domain of TLS/FUS. Takeo, S., Tabata, M., Okita, H., Shibata, N., Sato, K., Mase, N., Oyoshi, T., Narumi, T. Chem. Pharm. Bull. 2025, 73, 938-943.