English

• The specific hydrogen-bonded base pairing between nucleobases constitutes the molecular foundation of genetic information storage and transfer in DNA and RNA [1–4]. This highly selective recognition mechanism enables the precise assembly of duplex structures that are essential for replication, transcription, and translation [5–8]. These nucleobase pairings could also support high-order supramolecular architecture or self-adaptative materials [9–13]. However, stabilizing base pairs of free mononucleotides in highly polar solutions (especially in water) remains a substantial challenge: DMSO and water are strongly competitive hydrogen-bonding media that can screen hydrogen bonds of host and anion, respectively. In addition, the charged phosphate groups of nucleotides further complicate the energetics of association by favoring strong anion hydration [14–16]. Consequently, complementary mononucleotides or short oligonucleotides typically do not form stable duplexes in bulk solutions under physiological conditions, unless their pairing is sheltered within either enzymatic pockets or artificial microenvironments, which can provide spatial confinement, desolvation, and extra complementary interactions [17,18].

  Inspired by the confined binding sites of ribosomes and polymerases, supramolecular chemists have built artificial hosts to mimic enzymatic microenvironments and thereby promote nucleobase pairing in solutions [19–24]. Successful strategies reported to date employ spatial confinement together with hydrophobic desolvation and electrostatic attraction to stabilize base-pairing motifs: examples include coordination cages and organic cationic cages (Fig. 1a) that can encapsulate two or more nucleotides and stabilize A·U/A·T (A: adenosine; U: uridine, T: thymidine) pairs through a combination of size-matching, hydrophobic packing, and Coulombic compensation of the phosphate charges. Recent work even demonstrates that elegantly designed cationic cages can promote H-bonded base pairing of nucleoside triphosphates in aqueous solution, highlighting the importance of cavity desolvation and multi-point host-guest contacts for overcoming water competition [25]. Despite these advances, the role of custom-designed hydrogen bonding between the host and the phosphate group of nucleotides, without Coulombic attraction, has been underexplored. In contrast to electrostatic attraction, hydrogen-bonding hosts would offer exceptional programmability in size, shape, and cavity length. This versatility makes them a promising alternative for not only inducing nucleobase pairing but also regulating its binding mode.

  Figure 1

  

  Figure 1.  Nucleobase pairing in (a) cationic hosts (metal-coordination cage and covalent cage, previous work) driven by hydrophobic effect and (b) oligourea macrocycle driven by phosphate coordination (hydrogen bonding, this work).

  DownLoad: Full-Size Img PowerPoint

  Recently, our group and others have reported a series of oligourea receptors, including linear [26], macrocyclic [27–33], and cage hosts [34–36], exhibiting remarkable efficiency in selective recognition, transport, and separation of phosphate anions in aqueous and highly polar media. Such behaviors originate from the formation of multiple hydrogen bonds between urea motifs and phosphate. To address the above-mentioned gap and build on our previous findings, herein we designed and synthesized a charge-neutral, expanded hexyl-linked oligourea macrocycle, H[14]U (H indicates a hexyl linkage, 14 indicates the total number of urea motifs, Fig. 1b) that can encapsulate two monophosphate nucleotides through oligourea–phosphate hydrogen bonds within a single cavity as supported by single-crystal X-ray diffraction structure and solution experiments. Importantly, distinctly diverging from cationic cage host, we find that primary oligourea-phosphate hydrogen bonding could also facilitate nucleobase pairing in highly polar solvents.

  Akin to the chelating effect of bipyridine (compared to monopyridine) in metal coordination chemistry [37–42], the ortho-phenylene-bridged bis(urea) motif exhibits strong anion coordination affinity and forms characteristic hydrogen bonds with oxyanions, capable of engaging in up to 12 hydrogen bonds with phosphate [43–47]. In the absence of an anion, the free bis(urea) motif forms intramolecular C═O⋯H–N hydrogen bonds, which have been shown to pre-organize the oligourea backbone and facilitate efficient macrocyclization [48], as demonstrated in the synthesis of tetraurea and octaurea macrocycles. This strategy of intramolecular hydrogen bond-directed macrocyclization is expected to extend to longer oligourea chains, such as the pentaurea-based precursor 1 (Fig. 2a) [49]. Given that the phosphate headgroup of a nucleotide carries two negative charges and typically requires more than ten hydrogen bonds for effective coordination, a pentaurea monomer was selected. As anticipated and further supported by theoretical calculations [50–57], the pentaurea diamine adopts a folded conformation stabilized by seven intramolecular hydrogen bonds (Fig. 2a), with the terminal amine groups positioned in a satisfying proximity (~3 Å) for macrocyclization. Through macrocyclization, two additional urea units can be incorporated, resulting in a total of seven urea units in the [1 + 1] macrocycle, thereby providing a flexible and adaptive oligourea chain suitable for phosphate coordination. A hexyl linker was chosen as the central spacer, as the commercially available hexamethylene diisocyanate (HDI) monomer has a length of approximately 6 Å, sufficient to accommodate nucleotide base pairing (typically 6–9 Å; Fig. S14 in Supporting information). Accordingly, diamine 1 and HDI were mixed in anhydrous N, N-dimethylformamide (DMF) at room temperature without any catalyst, readily undergoing [1 + 1] and [2 + 2] macrocyclization to afford the H[7]U and H[14]U macrocycles, respectively.

  Figure 2

  

  Figure 2.  (a) Concentration-dependent macrocyclization from pentaurea-based diamine (1) and hexamethylene diisocyanate to afford H[7]U and H[14]U macrocycles. The DFT-optimized geometry of the diamine precursor is shown as an inset. (b) Stacked ¹H NMR spectra (DMSO–d₆, 400 MHz, 298 K) for the reaction of macrocyclization at variable concentrations, and (c) NMR data derived distribution of target macrocycles. (d) Single crystal X-ray diffraction structures of H[7]U macrocycle indicating multiple intramolecular hydrogen bonds. Nonpolar hydrogens and solvent molecules are omitted for clarity.

  DownLoad: Full-Size Img PowerPoint

  The macrocyclization process exhibits concentration-dependent behavior. At dilute reaction concentrations (below 50 mmol/L), the [1 + 1] macrocyclization pathway is favored, while the [2 + 2] pathway becomes predominant at higher concentrations. The resulting macrocycles, H[7]U and H[14]U, were initially identified by mass spectrometry (Figs. S6 and S9 in Supporting information) and isolated straightforwardly by precipitation in tetrahydrofuran (THF). The smaller H[7]U macrocycle is readily soluble in THF, whereas the larger H[14]U analogue shows limited solubility. The distribution of the two macrocycles can be conveniently monitored by ¹H NMR spectroscopy (Fig. 2b). The newly formed urea proton (Hg) resonates at 6.52 ppm for H[14]U and 6.38 ppm for H[7]U. In highly diluted reaction solutions (1 mmol/L), H[7]U is obtained in 73% yield. As the concentration increases to 250 mmol/L, the formation of H[14]U rises significantly, reaching 83% yield (Fig. 2c). This concentration-dependent product distribution aligns with previous reports. It can be rationalized by enhanced intramolecular hydrogen bonding at relatively high concentrations, in which the folded oligourea precursor 1 introduces increased steric hindrance around the terminal amine groups and disfavors the [1 + 1] macrocyclization, thereby promoting the [2 + 2] pathway [58].

  The structures of obtained macrocycles were unambiguously established by single-crystal X-ray diffraction. Crystallographic analysis reveals that the free H[7]U macrocycle crystallizes as a racemic mixture of right-handed (P) and left-handed (M) folded conformers. The folded structure is clearly stabilized by an extensive network of six intramolecular hydrogen bonds (Fig. 2d), aligning well with the DFT-optimized geometry of the pentaurea precursor (Fig. 2a). This structural insight corroborates the role of efficient intramolecular hydrogen bonding in directing the macrocyclization process, highlighting the potential of this strategy for constructing functional oligourea-based macrocyclic hosts with tailored cavities [59–64].

  Fortunately, single crystals suitable for X-ray diffraction were obtained for the phosphate-bound macrocycles, providing direct structural evidence of strong phosphate coordination (Fig. 3). In the H[7]U complex, a 1:1 host–guest stoichiometry is observed, with the macrocycle adopting a "U"-shaped conformation that encapsulates one phosphate anion (PO₄^3–) within its central cavity. The anion is secured by up to fourteen N–H···O hydrogen bonds with a mean N–H···O distance of 2.04 Å, indicating very strong phosphate coordination affinity [65–67]. Notably, phosphate coordination induces a notable conformational change: unlike the folded, helical structure of free H[7]U, the complex exhibits approximate planar symmetry. In the case of H[14]U, a 1:2 host–guest complex is formed, with two phosphate anions bound at a P···P distance of 12.6 Å (Fig. 3b). Each anion is clamped by the oligourea chains via fourteen hydrogen bonds, with an average N–H···O distance of 2.06 Å, comparable to that in the H[7]U complex. A key structural feature is the full extension of the two central hexyl linkers, which adopt an almost parallel alignment with an N···N distance of approximately 8.4 Å (Fig. 3c). This spatial separation matches well with the typical length required to accommodate a nucleotide base pair (ca. 6–9 Å), suggesting the cavity's suitability for such recognition.

  Figure 3

  

  Figure 3.  Single-crystal X-ray diffraction structures of phosphate coordinated macrocycles (a) H[7]U and (b, c) H[14]U in front and top views. Nonpolar hydrogens and solvent molecules are omitted for clarity.

  DownLoad: Full-Size Img PowerPoint

  The phosphate coordination properties of the macrocycles were further investigated in solution using ¹H NMR spectroscopy. Upon addition of 2 equiv. of phosphate anions, the NH protons of H[14]U exhibited significant downfield shifts of approximately 2.6 ppm (Fig. 4a, bottom two spectra), consistent with the formation of strong hydrogen bonds between the oligourea chain and phosphate. Similar downfield shift changes were observed for H[7]U upon phosphate coordination (Fig. S3 in Supporting information). These spectral changes align with previous reports on oligourea-based hosts, where phosphate binding affinities on the order of 10⁶ L/mol (in DMSO) have been measured for 1:1 complexes. Encouraged by the strong phosphate coordination demonstrated in both solid and solution, and by the suitable spatial separation provided by the extended hexyl linkers in the H[14]U structure (Fig. 3c), we proceeded to evaluate the complexation profiles of these macrocycles with monophosphate nucleotides.

  Figure 4

  

  Figure 4.  (a) Stacked ¹H NMR spectra of free UMP guest, free H[14]U macrocycle and its complexes with PO₄^3– and UMP (DMSO–d₆, 400 MHz, 298 K, 1 mmol/L). (b) CD spectra of Cl₄^–⊂H[14]U complex by adding UMP guest (as TBA⁺ salts, 20 µmol/L, CH₃CN). (c) ITC plots for the titration of H[14]U macrocycle with UMP guest (DMSO, 298 K, 0.1 mmol/L). (d) Optimized structures of the UMP₂⊂H[14]U complex indicating self-complementary hydrogen bonding of uridine units (GFN2-xTB).

  DownLoad: Full-Size Img PowerPoint

  The host-guest complexation behavior of the macrocycles was evaluated with four monophosphate nucleotides, adenosine monophosphate (AMP), uridine monophosphate (UMP), guanosine monophosphate (GMP), and cytidine monophosphate (CMP), used as their tetrabutylammonium (TBA) salts. As expected, free nucleotides in the absence of a host do not undergo spontaneous base pairing. This is corroborated by the single-crystal X-ray structure of AMP isolated from DMSO (Fig. S10 in Supporting information), in which no direct adenosine–adenosine dimerization is observed. Instead, the AMP molecules self-assemble into a one-dimensional supramolecular polymer through hydrogen bonding between the phosphate group and the nucleobase of adjacent molecules. This observation underscores the inherent hydrogen-bonding basicity of the phosphate anion and highlights the significant challenge of achieving specific nucleobase pairing in the absence of a shielding microenvironment.

  The monophosphate nucleotides binding with H[14]U were comprehensively investigated using a combination of ¹H NMR spectroscopy, circular dichroism (CD), mass spectrometry (MS), and isothermal titration calorimetry (ITC). These techniques collectively confirmed that H[14]U encapsulates two nucleotide guests within its cavity. Using UMP as a representative example, MS analysis first verified the 1:2 binding stoichiometry, displaying a peak at 926.71 (m/z) corresponding to the complex [H[14]U + 2UMP + TBA]^3–. In the ¹H NMR spectra, upon addition of two equivalents of UMP, the urea NH protons of H[14]U exhibited significant downfield shifts, resonating between 9 ppm and 12 ppm (Fig. 4a, middle two spectra). This change is similar to that observed upon phosphate anion binding, confirming that the oligourea chain maintains strong hydrogen-bonding interactions with the phosphate group of the nucleotide. In contrast, these NH protons are observed to be relatively broad, indicating low fidelity and dynamic nature of the complexation. Such behavior suggests the macrocycle could undergo conformational adjustments for UMP complexation in solution, consistent with adaptable hydrogen bonding with the oligourea backbone. Notably, the U2 and U3 protons of the uridine nucleobase experienced clear upfield shifts compared to free UMP. This behavior aligns with the characteristic chemical shift changes reported for nucleobase pairing within the confined cavities of cationic cages in previous studies [19], suggesting potential base-pairing interactions inside H[14]U macrocycle. Furthermore, the methylene protons (NCH₂) adjacent to the hexyl linker also showed an upfield shift relative to the free H[14]U. This observation is tentatively attributed to C–H···π interactions between the hexyl spacer and the uridine units, indicating close contact and a constrained aromatic environment within the complexation with UMP.

  CD spectroscopy provided further evidence for the formation of a 1:2 host-guest complex between H[14]U and UMP. The chloride-complexed form of H[14]U (prepared as the complex with TBACl to enhance its solubility in acetonitrile) exhibited a newly emerging Cotton effect centered around 275 nm upon UMP addition. This signal is attributed to the induced supramolecular chirality originating from the encapsulated UMP guests. The CD intensity reached a saturation plateau after the addition of 2.5 equiv. of UMP (Fig. 4b), confirming the encapsulation of two UMP molecules per macrocycle. In contrast, the CD titration for H[7]U reached saturation at 1.0 equiv. of UMP, consistent with its 1:1 binding stoichiometry. Furthermore, distinct CD spectral responses were observed for H[14]U upon complexation with other nucleotides (AMP, GMP, CMP, Fig. S24).

  The specific chiroptical signatures induced in each case correlate with the identity of the nucleobase unit, suggesting that the supramolecular chirality is highly sensitive to the nature of the encapsulated nucleobase. These differential CD responses not only confirm the close interaction between the nucleobase units and the oligourea macrocycle but also align with the observed contacts between the nucleobases and the central hexyl linkages, as inferred from NMR analysis.

  ITC measurements quantitatively determined the binding affinity of H[14]U for UMP in DMSO (Fig. 4c). The titration data were well-fitted to a 1:2 binding model, yielding binding constants of K₁ = 3.1×10⁴ L/mol and K₂ = 6.5×10⁵ L/mol. The system exhibits strong positive cooperativity, with a cooperative factor (α = 4K₂/K₁) of 83. Cooperative factor (α) quantifies the cooperativity of stepwise guest binding and is defined as α = 4K₂/K₁, where K₁ and K₂ denote the first and second binding constants corresponding to the sequential binding of two guest molecules to the host, respectively [68]. A notable biphasic binding profile was observed: the first UMP binding is highly endothermic (ΔH₁ = 199 kJ/mol), suggesting a conformationally demanding encapsulation process driven by hydrogen bonding. In contrast, the second binding event is exothermic (ΔH₂ = –56 kJ/mol) and enthalpy-driven. This distinct thermodynamic shift implies that the first binding induces significant structural preorganization of H[14]U, such as realigning the oligourea chain, displacing ordered solvent, and flattening the macrocyclic framework, to facilitate the subsequent nucleotide inclusion. The two-step mechanism, transitioning from entropy-driven to enthalpy-driven binding, underscores the adaptive nature of the synthetic macrocyclic host.

  Theoretical calculations were carried out to gain deeper insight into UMP encapsulation within the H[14]U macrocycle. The optimized structure confirms stable inclusion of a self-paired UMP dimer inside the macrocyclic cavity, stabilized by a complementary hydrogen-bonding network (Fig. 4d). Between the uridine bases, two intramolecular N–H···O hydrogen bonds (2.8 Å and 3.2 Å) are formed, while twelve N–H···O interactions (d = 1.97 Å) are seen between the phosphate groups and the oligourea chain. We also observe C–H···π interactions between the hexyl linkers and the nucleobase pairs (Fig. S16 in Supporting information), correlating with the upfield shift of hexyl protons seen in ¹H NMR spectrum (Fig. 4a).

  For the AMP nucleotide, ITC measurements revealed binding behavior comparable to that of UMP with the H[14]U macrocycle. The determined binding constants are K₁ = 7.0 × 10⁴ L/mol and K₂ = 5.0 × 10⁵ L/mol, yielding a cooperative factor of 29. ¹H NMR spectra showed chemical shift changes similar to those observed in the UMP case. Furthermore, the optimized structure indicates the formation of two hydrogen bonds within the adenosine dimer (Fig. S17 in Supporting information), supporting a stable base-paired arrangement inside the macrocyclic cavity.

  Finally, we investigated whether a heterodimeric A·U base pair, representative of canonical RNA/DNA interactions, could form within the H[14]U macrocycle. A 1:1 mixture of AMP and UMP was prepared and used for binding studies. The ¹H NMR spectrum of the resulting host-guest complex showed increased complexity (Fig. 5a), yet key features support nucleotide binding: the urea NH protons resonate between 9.0–11.0 ppm, confirming phosphate coordination to the oligourea chain, while nucleobase protons (5.0–6.0 ppm) exhibit marked upfield shifts, analogous to those reported for A·U pairing in a Pt(Ⅱ)-coordinated cage. ESI-MS further substantiated the 1:1:1 complex of (AMP+UMP)⊂H[14]U assembly, with a peak observed at m/z 1522.72 (Fig. 5a). ITC titration of the AMP-UMP mixture into H[14]U yielded apparent binding constants of K₁ = 3.9 × 10⁴ L/mol and K₂ = 1.5 × 10⁶ L/mol. These values reflect enhanced affinity relative to titrations with individual nucleotides (Fig. 4c), and the significantly increased cooperative factor (α = 154) strongly suggests stabilization of the complementary A·U dimer. Enhanced cooperativity of A·U heterodimerization (α = 154) than UMP homodimer (α = 83) is presumably due to its relatively expanded macrocycle backbone (vide infra), where nucleobase pairing occurs outside the cavity thus reducing extra entropy compensation for the binding with the second guest. A relatively large K₂ binding constant is demonstrated (Fig. 5b).

  Figure 5

  

  Figure 5.  (a) Stacked ¹H NMR spectra for the 1:1 mixture of AMP and UMP, and their complex with H[14]U macrocycle (DMSO–d₆, 400 MHz, 298 K, 1 mmol/L), MS spectra for the nucleotide complex is inserted. (b) ITC plots for the titration of H[14]U with the 1:1 mixture of AMP and UMP (DMSO, 298 K, 0.1 mmol/L). (c) CD spectra of (AMP+UMP)⊂H[14]U complex (TBA⁺ salts, 20 µmol/L, CH₃CN). (d) Optimized structures (GFN2-xTB) of the (AMP+UMP)⊂H[14]U complex indicating Watson-Crick base pairing.

  DownLoad: Full-Size Img PowerPoint

  Additional support comes from CD spectroscopy studies, where a distinct Cotton effect emerges near 270 nm (Fig. 5c), indicative of supramolecular chirality for an organized heterodimer in the macrocyclic cavity. Collectively, these results confirm the co-encapsulation and stabilization of an A·U base pair inside the H[14]U macrocycle. The high binding affinity not only validates our design premise that phosphate-directed oligourea coordination can promote selective nucleobase pairing, but also provides a structural foundation for developing synthetic hosts toward biomimetic recognition.

  To gain structural insight into the (AMP+UMP)⊂H[14]U assembly, theoretical calculations were performed to evaluate possible base-pairing modes within the macrocyclic cavity. Geometries were optimized using the semi-empirical GFN2-xTB method [51,52], and single point energies were calculated at the ωB97M-V/def2-TZVPP level of theory [54–57]. Gibbs free energies (ΔG, [298 K]) were then obtained with Shermo using thermochemical corrections from the xTB vibrational analysis [69]. Among the A·U pairing patterns examined, including Watson-Crick, anti-Watson-Crick, Hoogsteen, and anti-Hoogsteen, the Watson-Crick type was the most stable hydrogen-bonded dimer, featuring two characteristic N–H···O hydrogen bonds between the adenine and uracil units (Fig. 5d). Notably, a non-base-paired 1:1:1 co-encapsulation structure was computed to be energetically more favorable than any hydrogen-bonded dimer. While this suggests that full Watson–Crick pairing is not the dominant driving force under the modeled conditions, the results do not preclude the occurrence of base pairing under experimental settings, where solvation and dynamic effects play significant roles. Importantly, the fact that hydrogen-bonded A·U pair can be accommodated and stabilized inside the macrocycle confirms that hydrogen bonding alone, without auxiliary electrostatic or strong hydrophobic contributions, can indeed promote nucleobase pairing within an extended oligourea cavity.

  In conclusion, we have successfully designed a neutral oligourea macrocycle, H[14]U, that functions as an effective synthetic host for nucleotide base pairing in solution. The key to its function lies in the strong, multiple hydrogen-bonding interactions between oligourea backbone and the phosphate anionic head, which anchors the nucleotides within a sufficiently large and adaptive cavity. This setup effectively creates a local microenvironment that shields the nucleobases from the bulk solvent, enabling hydrogen-bonding interactions. Comprehensive experimental data and theoretical calculations consistently support the encapsulation and stabilization of both homodimers and the complementary A·U Watson-Crick base pair. The observed high binding affinity and significant positive cooperativity underscore the efficiency of this design. Our findings validate the fundamental premise that hydrogen bonding alone can induce and stabilize nucleobase pairing within a well-designed neutral host, providing a new paradigm for the development of supramolecular systems for biomolecular recognition and assembly. Capitalizing on its phosphate-binding specificity and cavity adaptability, this neutral oligourea macrocycle holds promise for recognition and transport of anions and biological molecules, thus providing alternative candidates for designing new supramolecular biomaterials.

  CRediT authorship contribution statement

  Xiao-Wen Sun: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Wei Zhao: Writing – review & editing, Writing – original draft, Supervision, Project administration, Funding acquisition, Formal analysis, Conceptualization. Yougang Yao: Investigation, Data curation. Ji Wang: Methodology, Investigation. Yifei Liu: Investigation, Data curation. Kanglei Liu: Writing – review & editing, Supervision, Methodology, Funding acquisition. Biao Wu: Writing – review & editing, Project administration, Funding acquisition, Formal analysis, Conceptualization.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 22101024, 22171023, 22101025) and the Beijing Municipal Natural Science Foundation (No. 2222025). We also thank support from Analysis & Testing Center of Beijing Institute of Technology for data collection.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2026.112372.

  
  
• 1. [1]

     J.D. Watson, F.H.C. Crick, Nature 171 (1953) 737–738. doi: 10.1038/171737a0

  2. [2]

     W. Saenger, Principles of Nucleic Acid Structure, Springer, Berlin, 1984.

  3. [3]

     G.A. Jeffrey, W. Saenger, Hydrogen Bonding in Biological Structures, Springer, Berlin, 1991.

  4. [4]

     J. Li, L. Shi, K. Zhang, et al., Nucl. Acids Res. 46 (2018) D1039–D1048. doi: 10.1093/nar/gkx1039

  5. [5]

     K.D. Westover, D.A. Bushnell, R.D. Kornberg, Cell 119 (2004) 481–489. doi: 10.1016/j.cell.2004.10.016

  6. [6]

     T.C. Diller, Madhusudan, N.H. Xuong, S.S. Taylor, Structure 9 (2001) 73–82. doi: 10.1016/S0969-2126(00)00556-6

  7. [7]

     S. Takahashi, N. Sugimoto, Acc. Chem. Res. 54 (2021) 2110–2120. doi: 10.1021/acs.accounts.0c00734

  8. [8]

     M. Hu, T. Xie, Y. Hu, et al., Chin. Chem. Lett. 35 (2024) 109232. doi: 10.1016/j.cclet.2023.109232

  9. [9]

     J.J. Funke, P. Ketterer, C. Lieleg, et al., Sci. Adv. 2 (2016) e1600974. doi: 10.1126/sciadv.1600974

  10. [10]

      T.E. Ouldridge, A.A. Louis, J.P. Doye, Phys. Rev. Lett. 104 (2010) 178101. doi: 10.1103/PhysRevLett.104.178101

  11. [11]

      A. Rajendran, M. Endo, H. Sugiyama, Angew. Chem. Int. Ed. 51 (2012) 874–890. doi: 10.1002/anie.201102113

  12. [12]

      G.M. Peters, J.T. Davis, Chem. Soc. Rev. 45 (2016) 3188–3206. doi: 10.1039/C6CS00183A

  13. [13]

      P. Chen, C. Yu, J. Chen, L. Xu, H. Liu, Chin. Chem. Lett. 34 (2023) 108627. doi: 10.1016/j.cclet.2023.108627

  14. [14]

      J.L. Sessler, C.M. Lawrence, J. Jayawickramarajah, Chem. Soc. Rev. 36 (2007) 314–325. doi: 10.1039/B604119C

  15. [15]

      A.M. Agafontsev, A. Ravi, T.A. Shumilova, A.S. Oshchepkov, E.A. Kataev, Chem. Eur. J. 25 (2019) 2684–2694. doi: 10.1002/chem.201802978

  16. [16]

      S.J. Butler, K.A. Jolliffe, ChemPlusChem 86 (2021) 59–70. doi: 10.1002/cplu.202000567

  17. [17]

      D. Philp, J.F. Stoddart, Angew. Chem. Int. Ed. 35 (1996) 1154–1196. doi: 10.1002/anie.199611541

  18. [18]

      A. Korostelev, S. Trakhanov, M. Laurberg, H.F. Noller, Cell 126 (2006) 1065–1077. doi: 10.1016/j.cell.2006.08.032

  19. [19]

      T. Sawada, M. Yoshizawa, S. Sato, M. Fujita, Nat. Chem. 1 (2009) 53–56. doi: 10.1038/nchem.100

  20. [20]

      T. Sawada, M. Fujita, J. Am. Chem. Soc. 132 (2010) 7194–7201. doi: 10.1021/ja101718c

  21. [21]

      H. Cai, M. Li, X.R. Lin, et al., Angew. Chem. Int. Ed. 54 (2015) 10454–10459. doi: 10.1002/anie.201502045

  22. [22]

      L. Cheng, K. Liu, Y. Duan, et al., CCS Chem. 3 (2021) 2749–2763. doi: 10.31635/ccschem.020.202000509

  23. [23]

      L. Cheng, P. Tian, Q. Li, A. Li, L. Cao, CCS Chem. 4 (2022) 2914–2920. doi: 10.31635/ccschem.021.202101584

  24. [24]

      L. Zhao, L. Cheng, Y. Yang, et al., Angew. Chem. Int. Ed. 63 (2024) e202405150. doi: 10.1002/anie.202405150

  25. [25]

      Y. Li, L. Cheng, P. Wang, et al., Angew. Chem. Int. Ed. 64 (2025) e202505732. doi: 10.1002/anie.202505732

  26. [26]

      H. Li, L. Kou, L. Liang, et al., Chem. Sci. 13 (2022) 4915–4921. doi: 10.1039/d2sc00876a

  27. [27]

      X. Wu, P. Wang, P. Turner, et al., Chem 5 (2019) 1210–1222. doi: 10.1016/j.chempr.2019.02.023

  28. [28]

      X. Wu, J.R. Small, A. Cataldo, et al., Angew. Chem. Int. Ed. 58 (2019) 15142–15147. doi: 10.1002/anie.201907466

  29. [29]

      Y. Yang, H. Ying, Z. Li, et al., Nat. Commun. 12 (2021) 1572–1581. doi: 10.1038/s41467-021-21678-3

  30. [30]

      X. Wu, P. Wang, W. Lewis, Y.B. Jiang, P.A. Gale, Nat. Commun. 13 (2022) 4623–4629. doi: 10.1038/s41467-022-32403-z

  31. [31]

      J. Li, O. Catal, I. Marques, et al., J. Am. Chem. Soc. 147 (2025) 3392–3401. doi: 10.1021/jacs.4c14194

  32. [32]

      W. Zhao, H. Li, X. Lian, et al., Inorg. Chem. Front. 12 (2025) 4794–4803. doi: 10.1039/d5qi00611b

  33. [33]

      K. Diao, W. Xie, M. Wen, et al., Org. Lett. 27 (2025) 9738–9743. doi: 10.1021/acs.orglett.5c03042

  34. [34]

      L. Jing, E. Deplazes, J.K. Clegg, X. Wu, Nat. Chem. 16 (2024) 335–342. doi: 10.1038/s41557-024-01457-5

  35. [35]

      H. Shan, W. Zhao, J. Wang, et al., J. Am. Chem. Soc. 147 (2025) 14960–14965. doi: 10.1021/jacs.5c04399

  36. [36]

      Q. Nie, J. Tang, Y. Li, et al., Angew. Chem. Int. Ed. 64 (2025) e202509002. doi: 10.1002/anie.202509002

  37. [37]

      B.R. Serr, K.A. Andersen, C.M. Elliott, O.P. Anderson, Inorg. Chem. 27 (1988) 4499–4504. doi: 10.1021/ic00297a032

  38. [38]

      R. Krämer, J.M. Lehn, A.D. Cian, J. Fischer, Angew. Chem. Int. Ed. 32 (1993) 703–706. doi: 10.1002/anie.199307031

  39. [39]

      B. Hasenknopf, J.M. Lehn, B.O. Kneisel, G. Baum, D. Fenske, Angew. Chem. Int. Ed. 39 (1996) 1838–1840.

  40. [40]

      B. Hasenknopf, J.M. Lehn, N. Boumediene, et al., J. Am. Chem. Soc. 119 (1997) 10956–10962. doi: 10.1021/ja971204r

  41. [41]

      X. Feng, Y. Ren, H. Jiang, Coord. Chem. Rev. 438 (2021) 213907. doi: 10.1016/j.ccr.2021.213907

  42. [42]

      C. Verma, I.Y. Yaagoob, L.K.M.O. Goni, et al., Coord. Chem. Rev. 529 (2025) 216433. doi: 10.1016/j.ccr.2025.216433

  43. [43]

      S. Li, C. Jia, B. Wu, et al., Angew. Chem. Int. Ed. 50 (2011) 5721–5724. doi: 10.1002/anie.201180593

  44. [44]

      B. Wu, F. Cui, Y. Lei, et al., Angew. Chem. Int. Ed. 52 (2013) 5096–5100. doi: 10.1002/anie.201209930

  45. [45]

      W. Zhang, D. Yang, J. Zhao, et al., J. Am. Chem. Soc. 140 (2018) 5248–5256. doi: 10.1021/jacs.8b01488

  46. [46]

      L. Liang, W. Zhao, X.J. Yang, B. Wu, Acc. Chem. Res. 55 (2022) 3218–3229. doi: 10.1021/acs.accounts.2c00435

  47. [47]

      L. Liang, P. Su, Y. Wang, et al., J. Am. Chem. Soc. 146 (2024) 10908–10916. doi: 10.1021/jacs.4c01747

  48. [48]

      B. Wu, C. Jia, X. Wang, et al., Org. Lett. 14 (2012) 684–687. doi: 10.1021/ol2031153

  49. [49]

      G. Zhao, S.Q. Chen, W. Zhao, et al., CCS Chem. 4 (2022) 2498–2507. doi: 10.31635/ccschem.021.202101131

  50. [50]

      T. Lu, Molclus program, version 1.12, http://www.keinsci.com/research/molclus.html (accessed September 13, 2025).

  51. [51]

      S. Grimme, C. Bannwarth, P. Shushkov, J. Chem. Theory Comput. 13 (2017) 1989–2009. doi: 10.1021/acs.jctc.7b00118

  52. [52]

      C. Bannwarth, S. Ehlert, S. Grimme, J. Chem. Theory Comput. 15 (2019) 1652–1671. doi: 10.1021/acs.jctc.8b01176

  53. [53]

      M.J. Frisch, et al., Gaussian 09, Revision D. 01, Gaussian Inc., Wallingford, CT, 2009.

  54. [54]

      F. Neese, WIREs Comput. Mol. Sci. 2 (2012) 73–78. doi: 10.1002/wcms.81

  55. [55]

      F. Neese, WIREs Comput. Mol. Sci. 8 (2018) e1327. doi: 10.1002/wcms.1327

  56. [56]

      T. Lu, F. Chen, J. Comput. Chem. 33 (2012) 580–592. doi: 10.1002/jcc.22885

  57. [57]

      T. Lu, J. Chem. Phys. 161 (2024) 082503. doi: 10.1063/5.0216272

  58. [58]

      F. Wang, X. Shi, Y. Zhang, et al., J. Am. Chem. Soc. 145 (2023) 10943–10947. doi: 10.1021/jacs.3c01066

  59. [59]

      L. Yang, D. Zhang, H. Yan, et al., Inorg. Chem. 58 (2019) 4067–4070. doi: 10.1021/acs.inorgchem.9b00148

  60. [60]

      P. Wang, K. Liu, H. Ma, et al., Chem. Commun. 57 (2021) 13377–13380. doi: 10.1039/d1cc05603d

  61. [61]

      H. Nian, L. Cheng, L. Wang, et al., Angew. Chem. Int. Ed. 60 (2021) 15354–15358. doi: 10.1002/anie.202105593

  62. [62]

      H. Duan, T. Yang, Q. Li, et al., Chin. Chem. Lett. 35 (2024) 108878. doi: 10.1016/j.cclet.2023.108878

  63. [63]

      G. Yu, J. Zhou, J. Shen, G. Tang, F. Huang, Chem. Sci. 16 (2025) 17001-17001. doi: 10.1039/D5SC90190A

  64. [64]

      X. Zhang, Z. Lu, T. Ren, et al., Chin. Chem. Lett. 36 (2025) 110946. doi: 10.1016/j.cclet.2025.110946

  65. [65]

      W. Zuo, C. Jia, H. Zhang, et al., Chem. Sci. 10 (2019) 2483–2488. doi: 10.1039/c8sc04338h

  66. [66]

      Z.Y. Sun, S.Q. Chen, L. Liang, et al., Chem. Commun. 59 (2023) 12923–12926. doi: 10.1039/d3cc04122k

  67. [67]

      M. He, Y. Yao, Z. Yang, et al., Angew. Chem. Int. Ed. 63 (2024) e202406946. doi: 10.1002/anie.202406946

  68. [68]

      P. Thordarson, Chem. Soc. Rev. 40 (2011) 1305–1323. doi: 10.1039/C0CS00062K

  69. [69]

      T. Lu, Q. Chen, Comput. Theor. Chem. 1200 (2021) 113249. doi: 10.1016/j.comptc.2021.113249

• 

  Figure 1  Nucleobase pairing in (a) cationic hosts (metal-coordination cage and covalent cage, previous work) driven by hydrophobic effect and (b) oligourea macrocycle driven by phosphate coordination (hydrogen bonding, this work).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Concentration-dependent macrocyclization from pentaurea-based diamine (1) and hexamethylene diisocyanate to afford H[7]U and H[14]U macrocycles. The DFT-optimized geometry of the diamine precursor is shown as an inset. (b) Stacked ¹H NMR spectra (DMSO–d₆, 400 MHz, 298 K) for the reaction of macrocyclization at variable concentrations, and (c) NMR data derived distribution of target macrocycles. (d) Single crystal X-ray diffraction structures of H[7]U macrocycle indicating multiple intramolecular hydrogen bonds. Nonpolar hydrogens and solvent molecules are omitted for clarity.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Single-crystal X-ray diffraction structures of phosphate coordinated macrocycles (a) H[7]U and (b, c) H[14]U in front and top views. Nonpolar hydrogens and solvent molecules are omitted for clarity.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Stacked ¹H NMR spectra of free UMP guest, free H[14]U macrocycle and its complexes with PO₄^3– and UMP (DMSO–d₆, 400 MHz, 298 K, 1 mmol/L). (b) CD spectra of Cl₄^–⊂H[14]U complex by adding UMP guest (as TBA⁺ salts, 20 µmol/L, CH₃CN). (c) ITC plots for the titration of H[14]U macrocycle with UMP guest (DMSO, 298 K, 0.1 mmol/L). (d) Optimized structures of the UMP₂⊂H[14]U complex indicating self-complementary hydrogen bonding of uridine units (GFN2-xTB).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Stacked ¹H NMR spectra for the 1:1 mixture of AMP and UMP, and their complex with H[14]U macrocycle (DMSO–d₆, 400 MHz, 298 K, 1 mmol/L), MS spectra for the nucleotide complex is inserted. (b) ITC plots for the titration of H[14]U with the 1:1 mixture of AMP and UMP (DMSO, 298 K, 0.1 mmol/L). (c) CD spectra of (AMP+UMP)⊂H[14]U complex (TBA⁺ salts, 20 µmol/L, CH₃CN). (d) Optimized structures (GFN2-xTB) of the (AMP+UMP)⊂H[14]U complex indicating Watson-Crick base pairing.

  下载: 全尺寸图片 幻灯片
•