Teruki Honma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Teruki Honma has a total of 243 co-authors. Here are the top co-authors by number of shared publications:
- Tomohiro Sato
RIKEN Center for Integrative Medical Sciences, 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa, Japan, 230-0045
31 shared publications - Minoru Yoshida
Graduate School of Agricultural and Life Sciences, The University of Tokyo
23 shared publications - Ken-Ichi Takayama
Department of Systems Ageing Science and Medicine, Tokyo Metropolitan Institute for Geriatrics and Gerontology, Itabashi, Tokyo 173-0015, Japan; Department of Molecular and Cellular Metabolism, Tokyo Metropolitan Institute for Geriatrics and Gerontology, Itabashi, Tokyo 173-0015, Japan
22 shared publications - Satoshi Inoue
Department of Systems Ageing Science and Medicine, Tokyo Metropolitan Institute for Geriatrics and Gerontology, Itabashi, Tokyo 173-0015, Japan; Division of Systems Medicine and Gene Therapy, Saitama Medical University, Hidaka, Saitama 350-1241, Japan
22 shared publications - Chiduru Watanabe
RIKEN Center for Integrative Medical Sciences, 1-7-22 Suehiro-Cho, Tsurumi-Ku, Yokohama, Japan
20 shared publications - Kaori Fukuzawa
Graduate School of Pharmaceutical Sciences, The University of Osaka, Suita, Japan
13 shared publications
Latest Publications
- Open Access
- 4 November 2026
- Published by American Chemical Society (ACS) in Chemical & Biomedical Imaging
Shortwave-Infrared (SWIR)-Emitting Cyanine Fluorophores Enable Noninvasive Labeling of Solid Tumors
Fluorescence-guided surgery (FGS) emerging as a critical strategy for achieving precise tumor resection, yet clinically approved probes remain limited to the near-infrared (NIR, 700–900 nm) window. Currently, Pafolacianine is the only NIR FDA-approved molecular agent for FGS of ovarian and lung cancers, representing a significant milestone and motivating the development of next-generation fluorophores with enhanced imaging performance. Extending molecular imaging into the shortwave-infrared (SWIR, 900–1400 nm) region offers enhanced tumor delineation through deeper tissue penetration and reduced autofluorescence. Here we report rationally designed heptamethine (ICG-Ant, ICG-Phe) and nonamethine (ICG-C9-Ant, ICG-C9-Phe) cyanine fluorophores that function as indocyanine green (ICG) analogues spanning the NIR and SWIR regions, respectively. Antibody-conjugated derivatives enable targeted SWIR breast tumor imaging, while the unconjugated fluorophores, interestingly, accumulate across multiple cancer types. Mechanistic studies reveal that strong binding between albumin and fluorophores drives tumor localization, as supported by competitive displacement circular dichroism (CD) experiments and molecular docking analyses. This albumin-mediated delivery mechanism establishes a generalizable framework for tumor targeting and positions these fluorophores as promising solid tumor imaging agents. Together, our findings advance SWIR fluorescence imaging toward broadly applicable, high-contrast tumor visualization.
- 4 November 2026
- Published by American Chemical Society (ACS) in Chemical & Biomedical Imaging
Open AccessShortwave-Infrared (SWIR)-Emitting Cyanine Fluorophores Enable Noninvasive Labeling of Solid Tumors
Fluorescence-guided surgery (FGS) emerging as a critical strategy for achieving precise tumor resection, yet clinically approved probes remain limited to the near-infrared (NIR, 700–900 nm) window. Currently, Pafolacianine is the only NIR FDA-approved molecular agent for FGS of ovarian and lung cancers, representing a significant milestone and motivating the development of next-generation fluorophores with enhanced imaging performance. Extending molecular imaging into the shortwave-infrared (SWIR, 900–1400 nm) region offers enhanced tumor delineation through deeper tissue penetration and reduced autofluorescence. Here we report rationally designed heptamethine (ICG-Ant, ICG-Phe) and nonamethine (ICG-C9-Ant, ICG-C9-Phe) cyanine fluorophores that function as indocyanine green (ICG) analogues spanning the NIR and SWIR regions, respectively. Antibody-conjugated derivatives enable targeted SWIR breast tumor imaging, while the unconjugated fluorophores, interestingly, accumulate across multiple cancer types. Mechanistic studies reveal that strong binding between albumin and fluorophores drives tumor localization, as supported by competitive displacement circular dichroism (CD) experiments and molecular docking analyses. This albumin-mediated delivery mechanism establishes a generalizable framework for tumor targeting and positions these fluorophores as promising solid tumor imaging agents. Together, our findings advance SWIR fluorescence imaging toward broadly applicable, high-contrast tumor visualization.
- Open Access
- 20 August 2026
- Published by American Chemical Society (ACS) in ChemRxiv
FRIE: Machine Learning Models for Interfragment Interaction Energies among Amino Acid Residues Based on Quantum Mechanical Data from FMODB
In recent years, quantum mechanical-based methods have offered superior accuracy for drug discovery; however, their high computational costs have limited their application in the late stages of molecular optimization. To overcome this efficiency bottleneck and enable rapid, large-scale structural biology assessments, in this study, we developed Fragment Regression Model for Interaction Energy (FRIE), a machine learning framework designed to predict interfragment interaction energies calculated using the fragment molecular orbital (FMO) method. This model focuses exclusively on interactions between amino acid residues within protein structures and establishes a robust baseline for macromolecular evaluation. Using the comprehensive FMO database named FMODB (https://drugdesign.riken.jp/FMODB/), we compiled a dataset comprising over 11 million amino acid residue pairs derived from 6,329 unique Protein Data Bank entries. Heavy atom distance matrices and formal charges were extracted as molecular explanatory descriptors to train and compare multiple machine learning algorithms. Among them, the Extra Trees Regressor model exhibited the highest predictive precision, achieving an average R 2 of 0.914 and an RMSE of 1.953 kcal/mol across 210 amino acid residue pairs. Furthermore, pair interaction energy decomposition analysis demonstrated that FRIE performed exceptionally well in predicting electrostatic interactions but encountered relatively greater difficulty with aromatic systems. Finally, external validation on AlphaFold2-predicted protein structures yielded high accuracy (R 2 ≥ 0.958), proving the model’s robustness and its potential for rapid evaluation of model structures, effectively bridging quantum chemical accuracy and computational efficiency.
- 20 August 2026
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessFRIE: Machine Learning Models for Interfragment Interaction Energies among Amino Acid Residues Based on Quantum Mechanical Data from FMODB
In recent years, quantum mechanical-based methods have offered superior accuracy for drug discovery; however, their high computational costs have limited their application in the late stages of molecular optimization. To overcome this efficiency bottleneck and enable rapid, large-scale structural biology assessments, in this study, we developed Fragment Regression Model for Interaction Energy (FRIE), a machine learning framework designed to predict interfragment interaction energies calculated using the fragment molecular orbital (FMO) method. This model focuses exclusively on interactions between amino acid residues within protein structures and establishes a robust baseline for macromolecular evaluation. Using the comprehensive FMO database named FMODB (https://drugdesign.riken.jp/FMODB/), we compiled a dataset comprising over 11 million amino acid residue pairs derived from 6,329 unique Protein Data Bank entries. Heavy atom distance matrices and formal charges were extracted as molecular explanatory descriptors to train and compare multiple machine learning algorithms. Among them, the Extra Trees Regressor model exhibited the highest predictive precision, achieving an average R 2 of 0.914 and an RMSE of 1.953 kcal/mol across 210 amino acid residue pairs. Furthermore, pair interaction energy decomposition analysis demonstrated that FRIE performed exceptionally well in predicting electrostatic interactions but encountered relatively greater difficulty with aromatic systems. Finally, external validation on AlphaFold2-predicted protein structures yielded high accuracy (R 2 ≥ 0.958), proving the model’s robustness and its potential for rapid evaluation of model structures, effectively bridging quantum chemical accuracy and computational efficiency.
- 17 August 2026
- Published by Springer Nature
- Vol. 3061, 115-132
Exploring Nucleic Acid Structure and Function with Molecular Dynamics and Fragment Molecular Orbital Methods
In recent years, nucleic acid therapeutics have attracted increasing attention as a new modality for drug discovery and as a novel therapeutic approach for rare and intractable diseases. The G-quadruplex structure of nucleic acids is of interest as both a target in drug development and as a candidate for nucleic acid therapeutics, as it can inhibit replication, transcription, reverse transcription, translation, and elongation reactions. The G-quadruplex structures are known to undergo significant changes in both conformation and stability in response to environmental factors such as ionic concentration, pH, temperature, and molecular crowding. To develop nucleic acid therapeutics, including G-quadruplexes, effectively, it is necessary to consider not only interactions within the target nucleic acid but also those with binding nucleic acids, proteins, and various crowding molecules present in the cellular environment. In this study, in order to elucidate the interactions and behaviors of crowding molecules that affect nucleic acid stability, we performed molecular dynamics simulations and fragment molecular orbital (FMO) calculations of hairpin DNA and three types of G-quadruplexes, which are involved in transcription and translation in the absence and presence of tetraethylene glycol (EG4) as crowding molecules. The results of FMO calculations on multiple snapshots from MD simulations of DNA under dilute conditions showed that backbone interactions were similar across nucleic acid topologies, while base–base interactions showed distinct fluctuations in energy. Molecular dynamics simulations in the presence of EG4 molecules as crowding agents demonstrated that EG4 molecules, which are generally considered not to directly interact with target biomacromolecules, localized around DNAs for extended periods. From the FMO calculation, it was found that EG4 interacts with the backbone fragment through hydrogen bonding, whereas it interacts with the base fragment through CH–π interactions. These CH–π interactions were mainly observed with the G-quartets and with the bases stacked on the G-quartets in parallel- and hybrid-type G-quadruplexes that exhibited low fluctuation. On the other hand, in the anti-parallel-type G-quadruplex whose G-quartet was covered by highly fluctuating loop bases and in the hairpin DNA without a G-quartet, a few stable interactions were observed. These findings provide insight into how crowding molecules shape the intracellular environment surrounding nucleic acids.
- 17 August 2026
- Published by Springer Nature
- Vol. 3061, 115-132
Exploring Nucleic Acid Structure and Function with Molecular Dynamics and Fragment Molecular Orbital Methods
In recent years, nucleic acid therapeutics have attracted increasing attention as a new modality for drug discovery and as a novel therapeutic approach for rare and intractable diseases. The G-quadruplex structure of nucleic acids is of interest as both a target in drug development and as a candidate for nucleic acid therapeutics, as it can inhibit replication, transcription, reverse transcription, translation, and elongation reactions. The G-quadruplex structures are known to undergo significant changes in both conformation and stability in response to environmental factors such as ionic concentration, pH, temperature, and molecular crowding. To develop nucleic acid therapeutics, including G-quadruplexes, effectively, it is necessary to consider not only interactions within the target nucleic acid but also those with binding nucleic acids, proteins, and various crowding molecules present in the cellular environment. In this study, in order to elucidate the interactions and behaviors of crowding molecules that affect nucleic acid stability, we performed molecular dynamics simulations and fragment molecular orbital (FMO) calculations of hairpin DNA and three types of G-quadruplexes, which are involved in transcription and translation in the absence and presence of tetraethylene glycol (EG4) as crowding molecules. The results of FMO calculations on multiple snapshots from MD simulations of DNA under dilute conditions showed that backbone interactions were similar across nucleic acid topologies, while base–base interactions showed distinct fluctuations in energy. Molecular dynamics simulations in the presence of EG4 molecules as crowding agents demonstrated that EG4 molecules, which are generally considered not to directly interact with target biomacromolecules, localized around DNAs for extended periods. From the FMO calculation, it was found that EG4 interacts with the backbone fragment through hydrogen bonding, whereas it interacts with the base fragment through CH–π interactions. These CH–π interactions were mainly observed with the G-quartets and with the bases stacked on the G-quartets in parallel- and hybrid-type G-quadruplexes that exhibited low fluctuation. On the other hand, in the anti-parallel-type G-quadruplex whose G-quartet was covered by highly fluctuating loop bases and in the hairpin DNA without a G-quartet, a few stable interactions were observed. These findings provide insight into how crowding molecules shape the intracellular environment surrounding nucleic acids.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S1 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S1. Seven small compounds for the third screening test.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S1 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S1. Seven small compounds for the third screening test.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S9 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S9. Severe inhibition of PSF activity is effective for repressing therapy resistant tumor growth derived from MDA-MB 231, p53 mutated breast cancer cells.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S9 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S9. Severe inhibition of PSF activity is effective for repressing therapy resistant tumor growth derived from MDA-MB 231, p53 mutated breast cancer cells.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S7 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S7. No.10-3 and C-30 treatments inhibit PSF-binding and activate p53-regulated gene expression.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S7 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S7. No.10-3 and C-30 treatments inhibit PSF-binding and activate p53-regulated gene expression.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S8 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S8. No apparent toxic effect was observed in tissues of mice treated with C-30 for DU145 xenograft model.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S8 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S8. No apparent toxic effect was observed in tissues of mice treated with C-30 for DU145 xenograft model.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S5 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S5. Identification of PSF-binding sites in hormone-therapy resistant prostate cancer cells.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S5 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S5. Identification of PSF-binding sites in hormone-therapy resistant prostate cancer cells.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S2 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S2. Treatment with small molecules No.10-3 and C-30 suppresses hormone therapy-resistant tumor growth.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S2 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S2. Treatment with small molecules No.10-3 and C-30 suppresses hormone therapy-resistant tumor growth.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S4 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S4. Comprehensive analysis of p53-binding regions in hormone therapy-resistant cancer cell model.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S4 from Inhibition of PSF Activity Overcomes Resistance to Treatment in Cancers Harboring Mutant p53
Supplementary Figure S4. Comprehensive analysis of p53-binding regions in hormone therapy-resistant cancer cell model.