Peptide‐Based Pseudo[2]‐ and [3]rotaxanes Assembled via Square‐Planar Palladium Coordination

Abstract

Interlocked peptide molecules exhibit peptidase resistance as well as enzyme inhibitory activity, making them promising candidates for peptide‐based drugs. Lasso peptides possess a [1]rotaxane structure, in which a peptide chain threads through the cavity of the cyclic peptide moiety. However, the construction of such peptide‐based interlocked structures remains highly challenging, and the total chemical synthesis of natural lasso peptides has not yet been achieved. Here, we aim to develop a synthetic method for a peptide‐based rotaxane using a metal‐template strategy. A macrocyclic complex bearing a Pd(II) ion fixed by a pincer ligand is employed to examine the threading of Val‐8mer in DMSO. Pseudorotaxane formation is assessed by characteristic ¹H NMR shifts and DOSY measurements, and is confirmed by electrospray ionization mass spectrometry (ESI‐MS). Titration and variable‐temperature (VT) NMR experiments are also performed to elucidate the dynamic behavior of the rotaxanation. These results demonstrate the feasibility of creating lasso‐peptide‐like architectures via metal‐templated rotaxanation, providing a new strategy for the synthesis of natural and artificial lasso peptides.

Lasso peptides possess a [1]rotaxane structure in which a peptide chain threads through the inner cavity of the cyclic peptide moiety. However, constructing such peptide‐based interlocked structures remains highly challenging. Here, we aimed to develop synthetic methods for a peptide‐based rotaxane using a metal‐template strategy. A macrocyclic complex bearing a Pd(II) ion fixed by a pincer ligand was employed to examine the threading of Val‐8mer in DMSO. Pseudorotaxane formation was assessed by ¹H NMR and DOSY spectroscopy and confirmed by ESI‐MS. These results demonstrate the feasibility of constructing lasso‐peptide‐like architectures via metal‐templated rotaxane formation, providing an unprecedented approach for the synthesis of natural lasso peptides and their analogs.

1. Introduction

In the life sciences, extensive research has been devoted to reveal the relationship between the spatially entangled topological structures of peptide chains in natural proteins and their biological functions [1, 3]. Lasso peptides are naturally occurring middle‐size molecule peptides that possess a knotted [1]rotaxane‐like topology [4, 6]. Figure 1 shows the structure of Microcin J25 (MJ25), a representative lasso peptide [7, 9]. The lasso peptide consists of a peptide chain branching from the cyclic peptide and penetrating into the cyclic inner cavity. Bulky substituents at the C‐terminus suppress slippage of the peptide axle component, thereby kinetically stabilizing the [1]rotaxane structure [10, 12].

FIGURE 1.

The structure of Microcin J25.

Lasso peptides are known to exhibit a wide range of biological activities, including antibacterial [7, 9, 13], antifungal activities [14], atrial natriuretic peptide antagonism [15], anti‐HIV activity [16], endothelin B receptor antagonism [17], inhibition of HIV fusion [18, 19], prolyl endopeptidase inhibition [20], glucagon receptor antagonism [21, 23], antitubercular activity [24, 25], and antitumor activity [26, 27]. Notably, their interlocked conformation confers resistance to peptidase degradation [28, 29], which has attracted considerable attention to lasso peptides as a novel modality for peptide‐based medicines.

Despite such unique structures and interesting biological activities, the chemical synthesis of lasso peptides has not yet been achieved because threading a peptide chain into the inner cavity of cyclic peptide moiety is extremely difficult [30, 37]. This is because there are no specific interactions that induce threading of a peptide chain into the inner cavity, making the formation of an interlocked structure remarkably difficult. Recently, Numata et al. reported a pioneering work for the synthesis of peptide rotaxanes. Their method exploits a threading reaction driven by intermolecular hydrogen bonding between the carbonyl oxygens of a cyclic peptide and the NH group of an axle peptide. Although the efficiency of threading reactions strongly depends on the conformation of cyclic peptide and the size complementarity of the peptide‐based rotaxane, the obtained yields were extremely low, with a maximum of 1.2% [38]. Considering this background, we have focused on the development of a new rotaxanation method that is applicable to the synthesis of natural lasso peptides, with an eye toward future drug discovery research. In this study, we have focused on rotaxanation based on the metal‐template method. The synthesis of interlocked molecules by the metal‐template method has been systematically reviewed by Sauvage or Leigh [39, 41]. Because organometallic complexes can strictly control the spatial geometry of ligands through the central metal ion, the appropriate choice of a metal ion facilitates to rationally construct the rotaxane complexes [41]. In addition, metal‐coordination bonds are often stronger than hydrogen bonds. Even when the sizes of the components are not so complemental, the coordination bond tends to form efficiently the corresponding interlocked molecule. We envisioned that if the metal‐template method were applied to peptide molecules, the design and synthesis of peptide topology could be tuned freely.

On the other hand, our research group has also focused on processive catalytic reactions exhibited by naturally occurring cyclic enzymes [42, 49]. Recently, we reported the transformation of polymeric substrates using a macrocycle‐Pd(II) complex that mimics natural processive catalytic systems. In this Pd macrocyclic complex, the Pd(II) ion is immobilized by a pincer ligand on the macrocycle and adopts a square‐planar coordination geometry. The remaining exchangeable coordination site is oriented toward the internal cavity of the macrocycle [50, 53]. It has been revealed that, even in a polar solvent such as dimethyl sulfoxide (DMSO), coordination of the Pd(II) ion to a polymeric axle induces threading of the polymer chain through the macrocyclic cavity.

Furthermore, it has been reported that the Pd(II) ion can be removed from the resulting rotaxane structure under specific conditions [51, 52]. These findings suggest that a Pd‐template strategy could serve as an effective approach for constructing peptide‐based interlocked architectures such as lasso peptides. DMSO is a good solvent for peptides, and peptide amide bonds are known to coordinate to transition metals [49].

Recently, we reported that mixing Pd macrocyclic complex MC3 with PolyVal as a poly(α‐amino acid) in DMSO‐d ₆ afforded a polypseudorotaxane (Scheme 1a) [50]. The same study also involved complexation experiments between MC3 and Val‐8mer, which suggested the possible formation of pseudo [2]‐ and pseudo [3]rotaxanes, as indicated by characteristic changes in the ¹H NMR spectra. However, the assignments of the observed signals and the existence of the proposed pseudorotaxane species were still considered preliminary. Therefore, this study aimed to provide direct evidence for pseudorotaxane formation and to gain deeper insight into the Pd‐coordination‐mediated threading process by performing a detailed investigation of the MC3/Val‐8mer system (Scheme 1b). The formation of the pseudorotaxane structures was evaluated using a combination of ¹H NMR, diffusion‐ordered NMR spectroscopy (DOSY), and electrospray ionization mass spectrometry (ESI‐MS) measurements. Additionally, the rotaxanation behavior was examined through titration experiments and VT‐NMR measurements, which enabled the thermodynamic properties of the assemblies to be evaluated.

SCHEME 1.

(a) Previous work: synthesis of poly(pseudorotaxane) with PolyVal as the axle and (b) this work: synthesis of pseudorotaxanes with Val‐8mer as the axle.

2. Results and Discussion

Val‐8mer was synthesized by solid‐phase peptide synthesis using an automated peptide synthesizer according to the Fmoc strategy (Scheme 2). The Fmoc‐Val‐immobilized resin was first swollen in DMF, followed by removal of the Fmoc protecting group with piperidine. Fmoc‐Val was then coupled to the resin‐bound amino acid using N,N′‐diisopropylcarbodiimide (DIC) and Oxyma as coupling reagents, thereby elongating the peptide chain on the resin. This deprotection–coupling cycle was repeated seven times to sequentially extend the peptide chain from the C‐terminus, affording the formation of Val‐8mer as an octapeptide on the resin.

SCHEME 2.

Solid‐phase synthesis of Val‐8mer.

The peptide chain was cleaved from the resin by treatment with a mixture of trifluoroacetic acid (TFA), triisopropylsilane, and H₂O (95:2.5:2.5, v/v/v). The resin was removed by filtration, and the filtrate was concentrated. Cold diethyl ether (Et₂O) was added to the crude to precipitate the peptide. The products were collected by centrifugation, washed several times with Et₂O, and dried under reduced pressure to afford Val‐8mer.

The obtained product was characterized by ¹H NMR, IR, and matrix‐assisted laser desorption/ionization time of flight mass spectrometry (MALDI‐TOF MS). In the ¹H NMR spectrum, signals corresponding to the amide protons and the side‐chain isopropyl protons of the valine residues were observed. The IR spectrum exhibited characteristic absorption bands of amide linkages at 1628 (amide I) and 1539 cm⁻¹ (amide II), confirming formation of the peptide backbone. Furthermore, the molecular weight observed in the MALDI‐TOF MS was in good agreement with the theoretical value, indicating that Val‐8mer was obtained in high purity and high yield (Figures S1–S4). Next, the obtained Val‐8mer was employed as the axle component to attempt the synthesis of pseudorotaxanes (pseudo [2]rotaxane and pseudo [3]rotaxane) via a metal‐template approach. The macrocycle MC3 was prepared according to previously reported procedures [47, 50]. MC3 (500 mol%) was added to Val‐8mer (100 mol%) in DMSO‐d ₆, and the mixture was stirred at room temperature (25 °C) for 1 h (Scheme 3).

SCHEME 3.

Synthesis of the pseudorotaxanes.

The ¹H NMR spectrum of the reaction mixture revealed the appearance of new signals in the aromatic proton signal region (Figure S5). The new signals indicate that protons on the phenylene units would be upfield‐shifted as a result of shielding by the amide carbonyl groups of the axle component. Such upfield‐shifts are in good agreement with those typically observed in reported rotaxanation processes [49, 54]. Similar chemical shift changes were also observed in our previous studies on the threading of PolyVal systems [50].

To evaluate the chemical species generated in an equilibrium in DMSO‐d ₆ solution, we performed DOSY measurement of the reaction mixture. As a result of Contin analysis, four distinct sets of signals with different diffusion coefficients (D) were clearly observed in the DOSY spectrum (Figures 2 and S6–S10).

FIGURE 2.

Distributions of the D values of free Val‐8mer (blue dotted line), free MC3 (red dotted line), and mixture of MC3 (500 mol%) + Val‐8mer (black solid line) in DMSO‐d ₆.

The appearance of four sets of signals with different D values upon simple mixing of the components suggests that the host–guest interactions are in a dynamic equilibrium, supporting the reversible nature of the pseudorotaxane formation process. In particular, we observed two signals with significantly reduced D values (5.58 × 10⁻¹⁶ and 9.13 × 10⁻¹⁵ m²/s) that were assigned to the complexes compared to those of the original components. According to the Stokes–Einstein equation (R _h = k _B T/6πηD, R _h: hydrodynamic radius, k _B: Boltzmann constant, T: absolute temperature, and η: viscosity), smaller D corresponds to larger R _h. In the diffusion coefficient distribution of the mixture, signals exhibiting distinctly smaller D values indicate the formation of larger species through intermolecular association. These signals could be tentatively attributed to pseudo [2]rotaxane and pseudo [3]rotaxane formed by threading of Val‐8mer into MC3. The chemical shifts of each species were determined on the basis of the DOSY correlations. The fourth set of signals exhibited a D value of 1.64 × 10⁻¹¹ m²/s. This value is of the same order of magnitude as that of free MC3 (1.81 × 10⁻¹¹ m²/s), and the signal is therefore assigned to nonassociated MC3. The remaining signal, with a D value of 6.10 × 10⁻¹³ m²/s, could be attributed to Val‐8mer. Notably, this value is somewhat larger than that of free Val‐8mer (1.18 × 10⁻¹³ m²/s). This increase might suggest that Val‐8mer is aggregated in solution due to intermolecular interactions. The dynamic formation of pseudorotaxanes could promote its dissociation into monomeric species.

Furthermore, the reaction solution was subjected to mass spectrometric analysis. MALDI‐TOF MS measurements were unsuccessful and no clear signals were obtained, which may be attributed to the extremely low ionization efficiency of the products and/or possible decomposition of the pseudorotaxanes under MALDI conditions. Therefore, ESI‐MS, which is generally considered a softer ionization technique, was employed for the analysis. ESI‐MS measurements were performed under highly dilute conditions using a mixed solvent of CH₃CN/CH₂Cl₂/CHCl₃ (79:20:1, v/v/v). As a result, molecular ion peaks corresponding to pseudo [2]rotaxane and pseudo [3]rotaxane were clearly observed (Figure 3a,b).

FIGURE 3.

ESI‐MS spectra of (a) pseudo [2]rotaxane: calcd. for C₆₉H₁₀₆N₁₁O₁₆Pd⁺ [MC3+Val‐8mer‐CH₃CN‐Fmoc+H]⁺ 1449.7, found 1449.5 and (b) pseudo [3]rotaxane: calcd. for C₉₈H₁₃₈N₁₄O₂₃Pd₂ ²⁺ [2MC3+Val‐8mer‐2CH₃CN‐Fmoc + 2H]²⁺ 1044.9, found 1044.7.

Because the ESI‐MS measurements were performed under highly diluted conditions, the possibility of weakly associated species with small association constants, such as side‐on coordination complexes, should be ruled out. It is noted that these complexes exist in solution under dynamic equilibrium, making their isolation extremely difficult; consequently, single crystals suitable for X‐ray crystallographic analysis could not be obtained, and direct structural determination by X‐ray diffraction was not possible.

Nevertheless, on the basis of the combined results from ¹H NMR, DOSY, and ESI‐MS analyses, it is clear that Val‐8mer undergoes stepwise threading into MC3, leading to the successive formation of pseudo [2]rotaxane and pseudo [3]rotaxane species. Furthermore, the absence of signals attributable to pseudo [4]rotaxane indicates that a single MC3 molecule would sterically cover four valine residues. These findings strongly suggest that rotaxane formation proceeds through well‐defined host–guest self‐assembly driven by the square‐planar palladium coordination (Figure S11).

2.1. Titration Experiments of Pseudorotaxanes

Since pseudorotaxane formations between MC3 and Val‐8mer were suggested, titration experiments were conducted to clarify the effect of the host/guest ratio on the system. MC3 was added stepwise to Val‐8mer in amounts of 100, 200, and 500 mol%. The mixtures were mixed in DMSO‐d ₆ at 25 °C for 1 h prior to ¹H NMR measurements. As a result, chemical shift changes were observed in the aromatic region under all conditions, with signal intensities varying depending on the amount of MC3 added (Figures S12–S16). Figure 4a compares the spectral changes in the phenylene proton region.

FIGURE 4.

(a) ¹H NMR spectra of the mixtures of MC3 and Val‐8mer at the molar ratios of 1:1, 2:1, and 5:1, respectively (400 MHz, DMSO‐d ₆, 298 K) and (b) plots of the peak intensity as a function of the molar ratio of MC3. ^aThe percentages were calculated from the integral ratios of the proton signals corresponding to each component in the ¹H NMR spectra, according to the following equation: Integral value [%] = {[pseudo [2]‐ or pseudo [3]rotaxane]/([pseudo [2]rotaxane] + [pseudo [3]rotaxane] + [Val‐8mer])} ×100. Lines are guides for eyes.

The signal observed at 7.26–7.17 ppm upon addition of 100 mol% MC3 exhibited the highest intensity and gradually decreased with further increases in the amount of MC3. In contrast, the signals at 7.17–7.06 ppm increased monotonically as the MC3 concentration increased (Figure 4b). Such behaviors can be well explained by a sequential association model, in which one MC3 molecule initially coordinates to Val‐8mer to form a pseudo [2]rotaxane (7.26–7.17 ppm), followed by coordination of an additional MC3 molecule to the axle to afford a pseudo [3]rotaxane (7.17–7.06 ppm).

The combined yield of pseudo [2]rotaxane and pseudo [3]rotaxane was approximately 22% under all conditions examined, which is markedly higher than the yield of 1.2% reported in previous study on peptide rotaxanes [38].

2.2. VT‐NMR of Pseudorotaxanes

To evaluate in detail the stability and thermodynamic behavior of pseudorotaxane formation, VT‐NMR measurements were performed. This technique is effective for quantitatively evaluating the temperature dependence of the interactions between the macrocycle and the axle component. Val‐8mer was mixed with MC3 (200 mol%) in DMSO‐d ₆ and ¹H NMR spectra of the reaction solution were recorded at 25, 40, 60, and 80 °C (Figures S17–S21). Figure 5a shows the temperature‐dependent ¹H NMR spectra in the phenylene proton region. As the temperature increased, the characteristic aromatic proton signals attributable to the pseudorotaxane structure gradually decreased, whereas the signal intensities assigned to free MC3 increased correspondingly (Figure 5b).

FIGURE 5.

(a) Partial ¹H NMR of the mixtures of MC3 and Val‐8mer at the molar ratio of 2:1 at 25, 40, 60, and 80 °C, respectively (400 MHz, DMSO‐d ₆) and (b) plots of the peak intensity as a function of the molar ratio of pseudo [2]rotaxane, pseudo [3]rotaxane, and MC3. Lines are guides for eyes.

These results indicate that the pseudorotaxane structure would be preferentially stabilized at lower temperature region, whereas dissociation would become dominant under higher temperature region, reflecting a thermally equilibrated system. Because the threading complex formation between MC3 and Val‐8mer is reversible, the distribution of the resulting chemical species would dynamically change with temperature.

The molar fractions of pseudo [2]rotaxane, pseudo [3]rotaxane, free MC3, and Val‐8mer were calculated from the integrated peak intensities. On the basis of the equilibrium equations shown in Scheme 4, the corresponding complexation constants K ₁ and K ₂ were determined (Table 1).

SCHEME 4.

Equilibrium formation of pseudorotaxanes.

TABLE 1.

Equilibrium constants K ₁ and K ₂ at each temperature.

Temperature, °C	K 1, L/mol	K 2, L/mol
25	37.1	33.6
40	25.3	32.1
60	20.1	27.8
80	15.3	18.1

$$K_1=\frac{\mathbf{[}\mathbf{\text{pseudo}}\mathbf{[}\mathbf{2}\mathbf{\left]}\mathbf{\text{rotaxane}}\right]}{\left[\mathbf{MC}\mathbf{3}\right]\left[\mathbf{\text{Val‐8mer}}\right]}$$

$${\mathit{K}}_2=\frac{\left[\mathbf{\text{pseudo}}\mathbf{[}\mathbf{3}\mathbf{]}\mathbf{\text{rotaxane}}\right]}{[\mathbf{\text{pseudo}}\mathbf{[}\mathbf{2}\mathbf{]}\mathbf{\text{rotaxane}}][\mathbf{MC}\mathbf{3}]}$$

The logarithms of the obtained equilibrium constants (lnK ₁ and lnK ₂) were plotted against the inverse of the absolute temperature (1/T) (Figure 6a,b). Both plots exhibited good linearity over a wide temperature range, suggesting that each process can be treated as a single equilibrium. From the slopes and intercepts of the linear fits, the thermodynamic parameters were evaluated according to the van’t Hoff equation

FIGURE 6.

(a) van’t Hoff plots of the equilibrium reaction for K ₁ and (b) for K ₂.

$$\ln K=-\frac{\mathrm{\Delta }H}{R}\frac{1}{T}+\frac{\mathrm{\Delta }S}{R}$$

On the basis of the van’t Hoff equation (K: equilibrium constant, ΔH: enthalpy change, ΔS: entropy change, T: absolute temperature, and R: gas constant), ΔH and ΔS values along with the complexation processes were calculated. These thermodynamic parameters reflect the nature of the threading behaviors during pseudorotaxane formation and the obtained results are summarized in Table 2.

TABLE 2.

ΔH and ΔS values for K ₁ and K ₂.

	ΔH, kJ/mol	ΔS, J/mol
K 1	–13.4	–15.3
K 2	–9.5	–1.9

Since the ΔH values exhibit negative values, pseudorotaxane formation is an exothermic process. According to the free energy relationship, ΔG = ΔH – TΔS, this reaction would be more spontaneous under lower‐temperature conditions, whereas at elevated temperatures the macrocycle tends to dissociate more readily from the axle molecule. This behavior is consistent with previous reports that thermal stimuli promote shuttling in metal‐coordinated rotaxane systems [55, 57]. These results demonstrate that the peptide‐threading process mediated by the metal‐template strategy is thermally reversible and that host–guest equilibria can be precisely regulated by temperature control. The present findings highlight both the utility and the limitations of the metal‐template approach for constructing peptide topologies.

3. Conclusion

In this study, we propose a new method for constructing a threaded structure, which is the most critical challenge in the chemical synthesis of peptide‐based rotaxanes, by using the coordination space of Pd(II) ions as a template. The reaction between a Pd(II)‐macrocycle complex and a short peptide, Val‐8mer, was confirmed to form clear pseudorotaxane structures by ¹H NMR, DOSY, and ESI‐MS analyses. In addition, VT‐NMR analysis revealed that the threading behavior between the peptide chain and the macrocycle changes in a temperature‐dependent manner, suggesting that the threading behavior is controlled by reversible coordination bonds. These findings indicate that the metal‐template method is effective for constructing spatially oriented interlocked structures represented by lasso peptides and also suggest the potential for future development toward the total synthesis of natural lasso peptides or the creation of novel functional peptide topology materials. In future studies, we aim to synthesize natural lasso peptides and their analog molecules by developing rotaxanation reactions applicable to cyclic peptides using a Pd(II)‐ion‐based metal‐template method.

4. Experimental Section

4.1. General Methods

4.1.1. Materials

Fmoc‐Val‐Wang resin (Watanabe Chem. Ind., Ltd.), 20% piperidine in DMF (Watanabe Chem. Ind., Ltd.), Fmoc‐Val‐OH (Sigma‐Aldrich Co. LLC., St. Louis, USA), N,N′‐diisopropylcarbodiimide (DIC, FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), Oxyma pure (Sigma‐Aldrich Co. LLC., St. Louis, USA), i‐Pr₃SiH (Tokyo Chemical Industry Co., Inc., Tokyo, Japan), and CF₃COOH (TFA, Tokyo Chemical Industry Co., Inc., Tokyo, Japan) were used as received. MC3 was synthesized following the reported procedures [49].

4.1.2. Measurements

¹H NMR (400 MHz) and ¹³C NMR (100 MHz) spectra were recorded using JEOL ECZ 400R spectrometer using DMSO‐d ₆ and CF₃COOD as a solvent and calibrated using internal standards and/or signals from a residual nondeuterated solvent. DOSY measurements were also performed using the sample solution in DMSO‐d ₆ to estimate the D value. Experiments were performed without spinning to avoid diffusion. Bipolar pulse longitudinal relaxation measurements were employed using the standard JEOL pulse program bppleddosy. The D values were determined by the multivariate curve analysis (MCR)‐based Contin method (MCR‐Contin). FT‐IR spectra based on an attenuated total reflection (ATR) method were measured using an IR Spirit spectrometer. MALDI‐TOF mass spectrum was recorded on a JMS‐S3000 SpiralTOF‐plus mass spectrometer using 2,5‐dihydroxybenzoic acid (DHB) as a matrix. The analyzer of the JMS‐S3000 SpiralTOF‐plus mass spectrometer was used in positive spiral mode. ESI mass spectra were recorded on a scimaX Fourier transform ion cyclotron resonance mass spectrometry (FT‐ICR‐MS) in positive ion mode using an ESI source. The capillary voltage was set to 4500 V with an ion accumulation time of 0.2 s. The mass range was set from m/z 200–2500. The sample solution was introduced at a flow rate of 120 μL/h, with a nebulizer pressure of 1.0 bar and a dry gas flow rate of 4.0 L/min. CF₃COONa (TFANa) was used as an external calibration. Each spectrum was obtained by accumulating 20 scans to improve the signal‐to‐noise ratio.

4.2. Chemical Synthesis

4.2.1. Synthesis of Val‐8mer

Val‐8mer was synthesized by Biotage Initiator+ Alstra peptide synthesizer using a standard Fmoc chemistry on Fmoc‐Val‐Wang resin. The synthetic protocol of Val‐8mer included the following cycle: Deprotection of Fmoc was carried out by 20% piperidine in DMF; Fmoc‐Val‐OH, activated with DIC and Oxyma pure, was coupled with Val‐terminated resin. The cleavage of Val‐8mer from the resin was carried out by treating with a mixture containing i‐Pr₃SiH, TFA, and H₂O at room temperature. The mixture was filtered and the filtrate was concentrated in vacuo. Val‐8mer was purified by precipitation using the TFA solution of the crude material in Et₂O; ¹H NMR (400 MHz, CF₃COOD, 298 K): δ 7.84–7.33 (m, 8 H), 5.04–3.82 (m, 10 H), 2.42–1.97 (m, 8 H), 1.35–0.80 (m, 48 H) ppm; ¹³C NMR (100 MHz, CF₃COOD, 298 K) δ 175.2, 171.7, 139.5, 122.5, 66.7, 58.4, 56.4, 28.9, 28.7, 15.8, 15.2, 14.0 ppm; IR (ATR) υ 3272 (NH), 1628 (amide I), 1540 (amide II) cm⁻¹, MALDI‐TOF MS (matrix: DHB) calc’d for C₅₅H₈₄N₈NaO₁₁ ⁺ [M+Na]⁺ 1055.6, found 1055.5.

4.3. General Procedure for Titration Experiments to Form Pseudorotaxanes

The complexation experiments were performed by varying the molar ratio of macrocycle MC3 to axle Val‐8mer. The sample solutions were prepared directly in NMR tubes by dissolving Val‐8mer and MC3 in DMSO‐d ₆ (600 μL). The mixtures were allowed to stand at room temperature for 1 h. The formation of pseudo [2]‐ and pseudo [3]rotaxane complexes was confirmed by ¹H NMR spectra and ESI‐MS; ¹H NMR (400 MHz, DMSO‐d ₆, 298 K) δ 8.27–8.11 (m), 7.73–7.38 (m), 7.35–6.96 (m), 6.90–6.20 (m), 4.55–4.09 (m), 4.01 (br), 3.70 (br), 2.06 (br), 0.98–0.81 (m) ppm; ESI‐MS: calc’d for C₆₉H₁₀₆N₁₁O₁₆Pd⁺ [MC3 + Val‐8mer‐CH₃CN‐Fmoc+H]⁺ 1449.7, found 1449.5, calc’d for C₉₈H₁₃₈N₁₄O₂₃Pd₂ ⁺ [2 MC3+Val‐8mer‐2CH₃CN‐Fmoc + 2H]²⁺ 1044.9, found 1044.7.

4.4. VT‐NMR Experiments for van’t Hoff Plots

Val‐8mer (4.0 mg, 3.9 μmol) was dissolved in DMSO‐d ₆ and sonicated at 60 °C for 2 h, after which MC3 (5.3 mg, 7.8 μmol) was mixed and VT‐NMR measurements were carried out at 25, 40, 60, and 80 °C. Based on the integration values of the aromatic proton peaks corresponding to MC3, pseudo [2]rotaxane, and pseudo [3]rotaxane, the concentrations of each species were determined.

Funding

This study was supported by Daiichi Sankyo Foundation of Life Science, the Naito Foundation, and Tamura Science and Technology Foundation.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

¹H NMR, ¹³C NMR, DOSY, and MALDI‐TOF mass spectra of the synthesized compounds; IR spectra of Val‐8mer and mixtures of MC3 and Val‐8mer at mixing ratios of 1:1, 2:1, and 5:1 (Figures S22–S24); and a proposed structure of the square‐planar palladium coordination in the pseudo[2]rotaxane are provided in the Supporting Information.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.