Stepwise self-organization of hydrogen-bonded fibers in a minimalist glucose-pyrene system via CH–π-stabilized “iotamers”

Abstract

A kinetically controlled self-organization system was realized using the synthetically simple molecule 4,6-O-pyrenylidene glucose (Py-Glc). Unlike conventional supramolecular saccharide systems that converge into a thermodynamically stable state via strong hydrogen bonding, Py-Glc temporarily suppresses primary nucleation by forming kinetically favored CH–π-stabilized aggregates as a dormant product. These aggregates then directly transform into microfibers through hydrogen bonding. Notably, this transformation is not initiated from large aggregates but “iotamers” such as dimers, trimers, or tetramers. This stepwise transition is significantly accelerated by the addition of fibrous seeds. To the best of our knowledge, this is the first demonstration of the seeded self-assembly in a monosaccharide-based system. This study introduces a novel minimalist molecular-design strategy for the precise control over the self-assembly and provides a generalizable platform for carbohydrate–aromatic conjugate systems.

---

Spontaneous organization of molecules into self-assembled structures and ordered patterns is a common natural process, often critical to biological functions.¹ Among biomolecules, proteins with higher-order structures exhibit prominent self-organization, primarily driven by hydrogen bonding (H-bonding) among peptide groups and hydrophobic interactions among aliphatic moieties.² Interestingly, approximately 60% of proteins exist as glycoproteins, which are covalently modified with glycans that act as key regulators of cell–cell communication, viral infection, and protein folding.³ The self-assembly of glycoproteins often results in the formation of glyco-clusters, where the spatial density of saccharides is significantly increased. This elevated density enhances the binding affinity between saccharides and proteins, with biological effects that strongly depend on saccharide distribution.⁴ Inspired by these biological architectures, the use of saccharides in supramolecular chemistry has been attracting increasing attention. Due to the abundance of hydroxyl groups, saccharides can form H-bonds in organic solvents, typically leading to convergence into a thermodynamical supramolecular state.⁵ An alternative strategy to exploit interactions beyond H-bonding is the formation of saccharide assemblies in aqueous environments. There, the hydroxyl groups preferentially interact with water, allowing for the formation of saccharide-based hydrogels driven by π–π, CH–π, and/or hydrophobic interactions with aromatic substituents.⁵ In other words, establishing precise control over the self-assembly via interactions other than H-bonding remains challenging, particularly under non-aqueous conditions. To address the issue, we focused on the formation of kinetically metastable dormant states as a route to controlling the supramolecular assembly.⁶ In this context, it should be noted that previously reported strategies typically require complex molecular designs and interactions such as intramolecular H-bonding or fluorine–π interactions to afford dormant states.^6c,d

Here, we report on the design of a simple yet effective synthetic monosaccharide-based molecule, i.e., 4,6-O-pyrenylidene glucose (Py-Glc), in which a rigid π-conjugated pyrene unit bridges the C4- and C6-hydroxyl groups of glucose. This specific molecular design enables the formation of a dormant state via CH–π interactions between the saccharide and the π-conjugated moiety, which subsequently transforms into bundled microfibers through an on-pathway mechanism (Fig. 1). This process is significantly accelerated by the addition of seeds, thus showcasing the first seeded supramolecular growth from a monosaccharide-derived system.

Fig. 1 Schematic illustration of the self-organization of Py-Glc into fibers via amorphous aggregates.

For the molecular design, we selected methyl α-D-glucose (1α) and pyrene (2). Glucose derivatives are overall the most abundant monosaccharides, and C1-methylation can control isomerization of the anomeric position. Pyrene, an aromatic hydrocarbon system, is often used for CH–π interactions with saccharides in molecular cages.⁷ Two-factor combined Py-Glc was prepared in one step from 1α in 76% yield using synthetic dimethoxy pyrene 2 under acidic conditions (Scheme S1, ESI†). Py-Glc was characterized by nuclear magnetic resonance (NMR) spectroscopy (Fig. S1 and S2, ESI†) and mass spectrometry.

The formation of Py-Glc aggregates in methylcyclohexane (MCH) solution was investigated using ultraviolet-visible (UV-Vis) spectroscopy. The kinetically metastable dormant state was obtained from cooling an MCH solution comprising Py-Glc (300 μL; c_T = 100 μM) from 100 °C to 40 °C at a rate of 10.0 °C min⁻¹. Subsequently, the solution was aged at 40 °C for 480 min. Absorption spectra obtained during aging at 40 °C in the range of 250–400 nm indicated that the intensity of the absorption peaks merely decreased (hypochromic shift) without any wavelength shift (hypso-/bathochromic shift) (Fig. S3, ESI†) and the formation of a precipitate in the solution was observed. The morphology of the resulting aggregates composed of Py-Glc was examined by transmission electron microscopy (TEM). TEM images of Py-Glc just after cooling to 40 °C revealed the formation of amorphous agglomerates (Fig. 2A and B). Upon aging at 40 °C, the aggregates evolved into μm-scale fibrous structures (Fig. 2C). High-magnification TEM images revealed that these larger fibers were assembled through the bundling of thinner fibers (Fig. 2D–F and Fig. S4, ESI†). The TEM analysis demonstrated that the amorphous agglomerates spontaneously transition into the fibrous structures. To gain insights into the molecular orientation in the self-assembly, a clarification of the dominant interactions affecting self-assembly is essential.

Fig. 2 TEM images of Py-Glc self-assemblies (A and B) just after cooling and (C–F) after aging.

To establish the formation of aggregates in solution, solutions of Py-Glc in CDCl₃ of varying concentrations (50 μM–10 mM) were analyzed using ¹H NMR spectroscopy. In the ¹H NMR spectra of Py-Glc (100 μM), two proton signals corresponding to 2-OH and 3-OH of Py-Glc were detected at chemical shifts (δ) of 2.64 and 2.25 ppm, respectively (Fig. S6, ESI†). The two signals gradually shifted downfield with increasing Py-Glc concentration. At a concentration of 10 mM, the 2-OH and 3-OH signals were observed at 2.73 and 2.28 ppm, respectively (|Δδ|_2-OH = 0.09 ppm; |Δδ|_3-OH = 0.03 ppm). This finding indicates that H-bonding at the 2-OH and 3-OH groups is formed upon the self-assembly. Notably, the ¹H NMR signals corresponding to H_a and H_i in the pyrene backbone simultaneously exhibited slight shifts with increasing Py-Glc concentration (Fig. S6 and Table S1, ESI†). This result suggests that the interactions through the pyrene moiety are also active during self-assembly. Previous studies have shown that the intermolecular CH–π interactions between the saccharide and aromatic skeleton in another molecule act as an effective force on the self-assembly.⁸ The origin of the observed chemical shift of the proton signal in pyrene can be reasonably interpreted in terms of a CH–π interaction between pyrene and the saccharide, which was further corroborated by the slight shifts of the proton signals at the C2-, C4-, and C6-positions in the saccharide (Fig. S6 and Table S1, ESI†). To determine which interactions (H-bonding and/or CH–π interactions) are dominant to form fibrous aggregates, we synthesized reference molecule Py-Glc-Me, in which both the 2-OH and 3-OH of Py-Glc are substituted by methoxy groups (Fig. S7 and S8, ESI†). Then, the same self-assembling protocol was applied to Py-Glc-Me (200 μM). TEM images of Py-Glc-Me clearly revealed that only amorphous agglomerates were generated, even after aging (Fig. S9, ESI†), demonstrating that H-bonding is crucial for the production of fibrous aggregates, while the CH–π interactions are predominantly responsible for the formation of the observed temporal amorphous agglomerates.

The kinetics of the transformation from amorphous agglomerates to fibers were monitored by UV-Vis spectroscopy. Fig. 3A shows the time course of the degree of aggregation (α_agg) estimated from the absorption change at 340 nm during aging at 40 °C with Py-Glc solutions of varying monomer concentration (c_T = 100, 125, 150, 175, and 200 μM); the resulting structures are shown in Fig. S10–S13 (ESI†). The time-dependent α_agg values exhibited a sigmoidal response, and the half-time at α_agg = 0.5 (t₅₀) decreased with increasing monomer concentration. Such an aggregation response to the total monomer concentration indicates that the production of fibrous aggregates proceeds via an “on-pathway” process in which the agglomerates directly transform to thermodynamically stable aggregates without dissociation into the monomers.^6g,9 To gain additional evidence regarding the transformation from the amorphous agglomerates to fibers, the as-cooled MCH solution including amorphous agglomerates were further diluted with MCH, and the time profiles of α_agg were tracked (Fig. S14, ESI†). The dilution of amorphous agglomerates afforded a decrease in t₅₀ with 100 μL of additional MCH (c_T = 75 μM). However, further dilution of the amorphous agglomerates with MCH delayed the transformation (Fig. 3B). Notably, cooling followed by aging of the MCH solution containing 50 μM of Py-Glc caused no absorption change (Fig. S15, ESI†) albeit that a sigmoidal response for α_agg was observed upon diluting from c_T = 100 μM to c_T = 50 μM with 300 μL of MCH (red data points in Fig. S14, ESI†). This finding indicates that the dissociation of amorphous aggregates, once formed, is subject to kinetic suppression. Thus, the amorphous agglomerates with a relatively small number of monomers (e.g., dimers or trimers) formed via CH–π interactions become the nuclei for the self-assemblies, followed by an elongation through further interactions of the small number of the organized units. Here, we collectively assign the term “iotamers” to aggregates with a small number of monomeric units such as dimers, trimers, and tetramers, and the transformation from CH–π agglomerates into H-bonded fibers in this system is initiated from such iotamers.

Fig. 3 (A) Time course of the degree of aggregation (α_agg) during the aging of Py-Glc (c_T = 100, 125, 150, 175, and 200 μM) at 40 °C for 8 h. (B) Time course of α_agg during aging at 40 °C after diluting the amorphous agglomerates with MCH.

To support our experimental findings, we also carried out DFT calculations on Py-Glc dimers and trimers via H-bonding or CH–π interactions (for the details on the DFT calculations, please see the ESI†). The formation of H-bonded dimers is characterized by a positive ΔG value (+13.69 kJ mol⁻¹), which arises from the large entropic loss due to the V-shaped conformation (Table 1 and Fig. S16A, ESI†). Interestingly, an H-bonded trimer is significantly stabilized (ΔG = −50.40 kJ mol⁻¹) by a “zigzag-type” H-bonding motif (Fig. S16B, ESI†).^5c However, the transformation from dimer to trimer can be expected to be unfavorable given that the necessary H-bonding exchange contains a high activation barrier. In contrast, the Gibbs free energies for the aggregation via CH–π interactions are almost identical regardless of the aggregation number (Table 1 and Fig. S16C, D, S17, ESI†). This result indicates that the self-assembling via CH–π interactions is similar to an “isodesmic” process.^6a,f,10 The expected interactions in the self-assemblies based on these theoretical results are consistent with the results of the ¹H NMR analysis (Fig. S6 and Table S1, ESI†). Furthermore, the identification of CH–π aggregates as isodesmic products strongly supports the dilution-dependent kinetic study (Fig. 3B), suggesting that CH–π-stabilized iotamers act as the self-assembling units. Overall, the transformation from amorphous agglomerates into fibrous micro-structures proceeds via an “on-pathway” process that is the result of the assembly of iotamers through H-bonding.

Table 1 DFT-derived thermodynamic parameters for the formation of aggregates from monomers of Py-Glcvia H-bonding or CH–π interactions

Reaction	ΔH/kJ mol^−1	ΔS/J mol^−1 K^−1	ΔG (40 °C)/kJ mol^−1
2Mon → Dim (H-bonding)	−39.43	−169.70	+13.69
Dim + Mon → Trim (H-bonding)	−128.31	−248.91	−50.40
2Mon → Dim (CH–π)	−76.39	−192.36	−16.18
Dim + Mon → Trim (CH–π)	−76.97	−197.31	−15.21

---

Finally, we performed seeded-growth experiments of fibers by adding 1 μL of a seed solution to the as-cooled MCH solution containing amorphous agglomerates (c_T = 100 μM). The seed solution was prepared by sonication of an aged MCH solution of Py-Glc fibers with c_T = 100 μM (Fig. 4A). As shown in Fig. 4B, the addition of a small amount of seeds notably accelerated the growth of the fibrous aggregates. Hence, our simple but rational molecular design considering molecular interactions using saccharides such as Py-Glc in this study can be used to establish precise control over complicated systems such as supramolecular polymerizations.

Fig. 4 (A) Schematic illustration of the seeded-aggregation process. (B) Time course of α_agg during the aging of Py-Glc (c_T = 100 μM) in the non-seeded and seeded processes.

In summary, we have demonstrated the self-assembly of microfibers using 4,6-O-pyrenylidene glucose. The main driving force for the formation of fibers is H-bonding, while the CH–π interactions between the saccharide and pyrene moieties of another Py-Glc monomer effectively leads to the formation of amorphous agglomerates as a dormant metastable product. Here, the CH–π-stabilized aggregates that are composed of a relatively small number of monomers (“iotamers”) serve as the self-assembly intermediates that further organize into fibrous structures through H-bonding. Although the length distribution of the resulting fibers could not be determined owing to the very large size of the fibers whose termini were not observed in the TEM images, which prevented an estimation of the polydispersity index, the CH–π interactions between the saccharide and aromatic moieties seems to be a promising approach for establishing precise control over supramolecular polymerizations.

Mitsuaki Hirose, Keigo Tashiro: conceptualization, funding acquisition, project administration, writing – original draft, investigation, writing – review & editing. Naoya Tajima, Futa Sugiura, Shuhei Shimoda: investigation, writing – review & editing. Yoshiumi Kohno, Yasumasa Tomita, Kiichiro Totani: resources, writing – original draft, writing – review & editing.

This work was partially supported by the Japan Society for the Promotion of Science (JSPS) via KAKENHI grants JP24K17790 (to M. H.) and JP23K13828 (to K. Tashiro). The authors would like to thank the Joint Usage/Research Center for Catalysis (Hokkaido University) for providing assistance with TEM measurements, as well as the supercomputer facility of ACCMS (Kyoto University), where the DFT calculations were carried out.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data that support the findings of this study are available from the corresponding authors (Mitsuaki Hirose, Keigo Tashiro, and Kiichiro Totani) upon reasonable request. The data supporting this article have also been included in the ESI.†

Notes and references

1. (a) A. C. Steven, W. Baumeister, L. N. Johnson and R. N. Perham, Molecular Biology of Assemblies and Machines, Garland Science, New York, 1st edn, 2016; (b) D. Lingwood and K. Simons, Science, 2010, 327, 46–50; (c) H. Y. Kueh and T. J. Mitchison, Science, 2009, 325, 960–963.
2. (a) K. A. Dill, Biochemistry, 1990, 29, 7133–7155; (b) C. M. Dobson, Nature, 2003, 426, 884–890; (c) D. Guin and M. Gruebele, Chem. Rev., 2019, 119, 10691–10717; (d) J. Dutta, A. Kumar, A. S. Bhadauria and H. S. Biswal, J. Chem. Inf. Model., 2022, 62, 1998–2008.
3. A. Varki, Glycobiology, 2017, 27, 3–49.
4. (a) Y. C. Lee and R. T. Lee, Acc. Chem. Res., 1995, 28, 321–327; (b) J. J. Lundquist and E. J. Toone, Chem. Rev., 2002, 102, 555–578; (c) R. T. Lee and Y. C. Lee, Glycoconjugate J., 2000, 17, 543–551; (d) M. Mammen, S.-K. Choi and G. M. Whitesides, Angew. Chem., Int. Ed., 1998, 37, 2754–2794; (e) M. González-Cuesta, C. Ortiz Mellet and J. M. Garía-Fernández, Chem. Commun., 2020, 56, 5207–5222.
5. (a) O. Gronwald and S. Shinkai, Chem. – Eur. J., 2001, 7, 4328–4334; (b) K. Yoza, N. Amanokura, Y. Ono, T. Akao, H. Shinmori, M. Takeuchi, S. Shinkai and D. N. Reinhoudt, Chem. – Eur. J., 1999, 5, 2722–2729; (c) R. Luboradzki, O. Gronwald, M. Ikeda, S. Shinkai and D. N. Reinhoudt, Tetrahedron, 2000, 56, 9595–9599; (d) O. Gronwald, K. Sakurai, R. Luboradzki, T. Kimura and S. Shinkai, Carbohydr. Res., 2001, 331, 307–318; (e) L. S. Birchall, S. Roy, V. Jayawarna, M. Huhes, E. Irvine, G. T. Okorogheye, N. Saudi, E. D. Santis, T. Tuttle, A. A. Edwards and R. V. Ulijn, Chem. Sci., 2011, 2, 1349–1355; (f) S. L. Higashi and M. Ikeda, JACS Au, 2021, 1, 1639–1646; (g) J. H. Jung, G. John, M. Masuda, K. Yoshida, S. Shinkai and T. Shimizu, Langmuir, 2001, 17, 7229–7232; (h) G. John, J. H. Jung, M. Masuda and T. Shimizu, Langmuir, 2004, 20, 2060–2065; (i) N. Yan, G. He, H. L. Zhang, L. P. Ding and Y. Fang, Langmuir, 2010, 26, 5909–5917; (j) M. L. Ma, Y. Kuang, Y. Gao, Y. Zhang, P. Gao and B. Xu, J. Am. Chem. Soc., 2010, 132, 2719–2728.
6. (a) T. F. A. De Greef, M. M. J. Smulders, M. Wolffs, A. P. H. J. Schenning, R. P. Sijbesma and E. W. Meijer, Chem. Rev., 2009, 109, 5687–5754; (b) Z. Chen, A. Lohr, C. R. Saha-Möller and F. Würthner, Chem. Soc. Rev., 2009, 38, 564–584; (c) H. Choi, S. Ogi, N. Ando and S. Yamaguchi, J. Am. Chem. Soc., 2021, 143, 2953–2961; (d) O. Shyshov, S. V. Haridas, L. Pesce, H. Qi, A. Gardin, D. Bochicchio, U. Kaiser and G. M. Pavan, Nat. Commun., 2021, 12, 3134; (e) Q. Huang, N. Cissé, N. C. A. Stuart, Y. Lopatina and T. Kudernac, J. Am. Chem. Soc., 2023, 145, 5053–5060; (f) M. Wehner and F. Würthner, Nat. Rev. Chem., 2020, 4, 38–53; (g) H. Itabashi, K. Tashiro, S. Koshikawa, S. Datta and S. Yagai, Chem. Commun., 2023, 59, 7375–7378; (h) C. Wang, L. Xu, L. Zhou, N. Liu and Z.-Q. Wu, Angew. Chem., Int. Ed., 2022, 16, e202207028; (i) T. Fukui, N. Sasaki, M. Takeuchi and K. Sugiyasu, Chem. Sci., 2019, 10, 6770–6776; (j) Z. Jin, N. Sasaki, N. Kishida, M. Takeuchi, Y. Wakayama and K. Sugiyasu, Chem. – Eur. J., 2023, 29, e202302181; (k) S. Patra, S. Dhiman and S. J. George, Chem. Mater., 2024, 36, 9460–9468; Y. Chen, Z. Gao, L. Wang, J. Li, Y. Tang and C. Liu, ACS Appl. Polym. Mater., 2022, 4, 1055–1064; (l) G. Ghosh, P. Dey and S. Ghosh, Chem. Commun., 2020, 56, 6757–6769; (m) A. Das, S. Ghosh and S. J. George, Angew. Chem., Int. Ed., 2025, 64, e202413747; (n) A. Isobe, D. D. Prabhu, S. Datta, T. Aizawa and S. Yagai, Chem. – Eur. J., 2020, 26, 8997–9004.
7. (a) T. J. Mooibroek, J. M. Casas-Solvas, R. L. Harniman, C. M. Renney, T. S. Carter, M. P. Crump and A. P. Davis, Nat. Chem., 2016, 8, 69–74; (b) L. Montalvillo-Jiménez, A. G. Santana, F. Corzana, G. Jiménez-OsésJesús, J. Jiménez-Barbero, A. M. Gómez and J. L. Asensio, J. Am. Chem. Soc., 2019, 141, 13372–13384.
8. (a) M. Kumari, R. B. Sonoj and P. V. Balaji, Carbohydr. Res., 2012, 361, 133–140; (b) C. He, S. Wu, D. Liu, C. Chi, W. Zhang, M. Ma, L. Lai and S. Dong, J. Am. Chem. Soc., 2020, 142, 17015–17023; (c) M. del Carmen, F. J. Fernández-Alonso, J. Cañada, G. Jiménez-Barbero and G. Cuevas, J. Am. Chem. Soc., 2005, 127, 7379–7386; (d) A. Brito, S. Kassem, R. L. Reis, R. V. Ulijn, R. A. Pires and I. Pashkuleva, Chemistry, 2021, 7, 2943–2964; (e) L. Díaz-Casado, A. Villacampa, F. Corzana, J. Jiménez-Barbero, A. M. Gómez, A. G. Santana and J. L. Asensio, JACS Au, 2024, 4, 476–490; (f) A. M. Keys, D. W. Kastner, L. L. Kiessling and H. J. Kulik, Chem. Sci., 2025, 16, 1746–1761.
9. (a) T. Fukui, M. Takeuchi and K. Sugiyasu, Sci. Rep., 2017, 7, 2425; (b) M. Wehner, M. I. Silja Röhr, V. Stepanenko and F. Würthner, Nat. Commun., 2020, 11, 5460; (c) M. Sahu, I. Seth, S. J. George and U. Maitra, Chem. Mater., 2024, 36, 7792–7800; (d) K. Fu, Y. Zhao and G. Liu, Nat. Commun., 2024, 15, 9571.
10. (a) A. Sharko, D. Livitz, S. De Piccoli, K. J. M. Bishop and T. M. Hermans, Chem. Rev., 2022, 122, 11759–11777; (b) S. Kim, M. Kim, S. Woo, J. Kim, S. H. Jung and J. H. Jung, Nanoscale Adv., 2024, 6, 1917–1925; (c) H. M. M. ten Eikelder, B. Adelizzi, A. R. A. Palmans and A. J. Markvoort, J. Phys. Chem. C, 2019, 123, 6627–6642.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5cc02945g

‡ M. Hirose and K. Tashiro contributed equally to this work.

---