Sandwich-Like Complexation of Carbohydrates by Hydrogen Bonding and CH-π Interactions

Abstract

Introduction: The design of carbohydrate-binding agents (artificial carbohydrate receptors) that enable selective and effective biomimetic recognition via noncovalent interactions is aimed either at a better understanding of natural recognition phenomena or at various potential applications in medicine and other fields. Although very interesting artificial receptors have been developed, the exact prediction of the receptor selectivity remains a challenge. Results and Methods: A molecular architecture based on a 1,3,5-substituted 2,4,6-triethylbenzene backbone bearing two aminopyridine- or aminopyrimidine-based recognition units and a purine moiety, which acts as both a hydrogen bonding site and a bridging component for the incorporation of additional substituents, has proved to be very useful for the development of effective carbohydrate-binding agents. This type of compounds has the ability to bind suitable carbohydrates through combined noncovalent interactions, where CH···π interactions can be formed on both faces of the carbohydrate substrate. The successful syntheses of the target compounds can be realized by the use of microwave irradiation and sealed tubes. The performed binding studies included ¹H NMR spectroscopic titrations and measurements by isothermal titration calorimetry. Conclusion: The new compounds were developed as artificial receptors especially for carbohydrates with an all-equatorial substitution pattern and have the ability to predictably form strong 1:1 complexes with a suitable substrate. The use of the purine moiety in the construction of carbohydrate receptors with a 1,3,5-substituted 2,4,6-triethylbenzene backbone has proved to be a very promising approach. The possibilities for structural variation of this molecular architecture are manifold. As a result, a wide range of compounds can be synthesized to perform extensive studies on the relationships between structure and binding efficiency.

Graphical Abstract

Short Text: New compounds were developed as artificial receptors especially for carbohydrates with an all-equatorial substitution pattern and have the ability to predictably form 1:1 complexes with a suitable substrate.

Key Topics: Artificial carbohydrate receptors; Combination of hydrogen bonding and CH۔۔۔π interactions; Molecular recognition

Introduction

Studies on the molecular recognition of carbohydrates, which act as information carriers in natural systems and thus play a key role in many vital processes, are of great importance and the subject of current research. The design of carbohydrate-binding agents (artificial carbohydrate receptors; for examples of reviews, see references^1‐11) which enable selective and effective biomimetic recognition via noncovalent interactions is aimed either at a deeper understanding of natural recognition phenomena or at various potential applications. In addition to their function as model systems for a better understanding of the molecular details of carbohydrate-mediated natural recognition processes, such compounds have the potential to serve as a the basis for the development of new therapeutic agents (anti-infective and anticancer agents)^12‐16 or diagnostics.

Although very interesting binding preferences of the artificial receptors have already been determined, the exact prediction of receptor selectivity remains a challenge.

X-ray diffraction studies of protein-carbohydrate complexes revealed that the selectivity and efficiency of carbohydrate-binding proteins is achieved by a combination of different noncovalent interactions^17‐25 such as neutral and charge-enhanced hydrogen bonding to the sugar hydroxy groups, CH···π interactions between the CH units of the sugar and the aromatic amino acid residues (see Figure 1a) as well as van der Waals contacts. The involvement of the sugar substrate in different types of noncovalent interactions, including CH···π interactions, has also been shown by crystallographic studies with artificial receptors,^26‐29 as illustrated exemplarily in Figure 1b-d. In addition to the crystallographic findings, various studies with artificial carbohydrate receptors^30‐56 (see notes in references^30,39,45 ) and other model systems^57‐62 in solution, especially NMR investigations, clearly showed the tendency of sugar molecules to participate in CH···π interactions with aromatic group(s) of the binding partners.

Figure 1.

(a) Examples of hydrogen bonds and CH···π interactions observed in the crystalline complex formed between D-galactose-binding protein (GBP) and D-glucose. ²² (b) and (c) Schematic representation of hydrogen bonds and CH···π interactions in two crystalline 2:1 receptor-carbohydrate complexes formed between a trimethyl- ²⁶ or triethylbenzene-based receptor ²⁷ and β-D-glucoside (octyl and methyl β-D-glucoside, respectively); the pyrimidine- (blue) and pyridine-based (green) recognition units of the receptor molecules are highlighted in color. (d) Hydrogen bonding and CH···π interactions observed in the crystalline 1:1 complex of a triethylbenzene-based receptor with methyl β-D-glucoside. ²⁷

In the natural complex shown in Figure 1a, the protein uses different aromatic groups (indole and benzene ring of tryptophan and phenylalanine, respectively) for the formation of CH···π interactions with the carbohydrate substrate. In the complexes formed by artificial receptors shown in Figure 1b and c, both sides of the glucoside (α- and β-face) are involved in CH···π interactions with the same type of aromatic group (here benzene ring). The sandwich-like complexation of the glucoside is realized in this case by the formation of 2:1 receptor-carbohydrate complexes, and the first crystal structure of such a complex ²⁶ (Figure 1b) acted as a source of ideas for the design of a new macrocyclic receptor architecture with flexible side-arms created by bridging the two receptor molecules.^39,41 By combining a macrocyclic backbone with two flexible side-arms (Figure 2), receptor molecules have been developed that predictably form 1:1 complexes with suitable carbohydrates, such as glucosides.^39,41‐44

Figure 2.

(Left) A representative ⁴¹ of the class of macrocyclic compounds bearing two flexible side-arms [here side-arms with pyrimidinyl groups (blue)]. (Right) Schematic representation of the hydrogen bonds and CH···π interactions stabilizing the complex of this macrocyclic compound with octyl β-D-glucopyranoside (the macrocyclic backbone is not completely depicted). Shown are interactions indicated by molecular modeling and confirmed by NMR spectroscopy).

In addition to macrocyclic compounds, well-designed acyclic receptor molecules also enable such sandwich-like complexation within a 1:1 complex. This was shown, for example, by initial studies with purine-containing receptor molecules, ³² whose binding mode is schematically shown in Figure 3. This type of compounds has the ability to bind suitable carbohydrates through combined noncovalent interactions, where the CH···π interactions can be formed with both faces of the carbohydrate substrate. Compared to the macrocyclic system, the structural variation of the acyclic architecture is easier to perform. As a result, compounds with two identical or different types of aromatic groups, which can be involved in the formation of the abovementioned CH···π interactions, are synthetically more easily accessible (e.g., a combination phenyl/indolyl groups, like in the case of the natural complexes shown in Figure 1a).

Figure 3.

Schematic representation of the structures and mode of action of purine-containing receptors. The pyridine- (green) and pyrimidine-based (blue) recognition units and the exchangeable aromatic unit (gray) on the purine ring (magenta) are highlighted in color.

The aim of the current studies was to synthesize a series of compounds (potential carbohydrate receptors), the structure of which is shown schematically in Figure 3, and to analyze in detail the influence of the nature of the structural subunits on the binding properties. In particular, the influence of the character of the aromatic group and the nature of its linkage to the purine ring was analyzed. In addition, it was investigated whether the replacement of the aminopyridine-based by aminopyrimidine-based recognition units would lead to more powerful (more efficient) carbohydrate receptors, as has been repeatedly observed by us for previously examined receptor systems. Such systematic studies of the structure-binding activity relationships provide useful information for the development of effective carbohydrate receptors with predictable binding properties.

Structures of the Target Compounds

Recently we have reported the synthesis of new 1,3,5-substituted 2,4,6-triethylbenzenes consisting of both purine unit(s) and pyridine-/pyrimidine-based recognition groups ³² (for a review on 1,3,5-substituted 2,4,6-triethylbenzene derivatives, see reference ⁶³  ). Compounds bearing 2-chloro-substituted purine unit(s) [2-chloro-9H(7H)-purin-6-yl unit(s)] were shown to be able to act as carbohydrate receptors and also as valuable precursors for further functionalizations, enabling systematic structural modifications. The nucleophilic displacement of the chlorine atom in the C2-position of the purine ring allows the incorporation of various substituents, including those bearing an aromatic group. A well-positioned aromatic group is capable of participating in the interactions with the sugar CH units on the opposite side to that involved in interactions with the central benzene ring. Thus, the purine moiety has two functions, it serves as a hydrogen bonding site and as a bridging unit for the introduction of substituents.

Initial binding studies with compounds prepared in this way indicated that their binding properties can be fine-tuned by the variation of the substituent attached to C2-position of the purine ring. Depending on the nature of the aromatic group and on the type and length of the linker unit, which connects this aromatic group with the purine ring, carbohydrate receptors with different binding properties can be developed.

In this study, we focused on the compounds shown in Figure 4, which contain phenyl, 1-naphthyl, 3-indolyl or 5-methoxy-3-indolyl group as an additional aromatic moiety. The unit –NH(CH₂)_n– with n = 1-3 was chosen as the linker, as illustrated in Figure 3. The binding properties of compounds containing aminopyridine-based recognition units (1a-7a) were compared with those of their analogs bearing pyrimidinyl groups (2b, 3b and 5b-7b).

Figure 4.

Structures of target compounds considered in this study.

Results and Discussion

Synthesis of the Target Compounds 1a-7a and 2b, 3b, 5b-7b

The syntheses of the target compounds 1a-7a, incorporating two pyridinyl groups, were carried out on the basis of the reaction of the educt 8a with selected amines, such as benzylamine, 2-phenylethylamine, 3-phenyl-1-propylamine, 1-naphthylmethylamine, 2-(1-naphthyl)ethylamine, tryptamine and 5-methoxytryptamine (see Scheme 1). To prepare the analogs 2b, 3b, and 5b-7b, bearing pyrimidinyl groups, compound 8b was treated with 2-phenylethylamine, 3-phenyl-1-propylamine, 2-(1-naphthyl)ethylamine, tryptamine or 5-methoxytryptamine, as shown in Scheme 1. 1-Naphthylmethylamine required for some of the abovementioned syntheses was prepared via a reduction of the commercially available 1-naphthylnitrile according to the procedure reported by Nystrom. ⁶⁴

Scheme 1.

Synthesis of compounds 1a-7a and 2b, 3b and 5b-7b (reaction conditions and yields are given in Table S1 in the Supporting Information).

The 1,3,5-substituted 2,4,6-triethylbenzene derivatives 8a and 8b were prepared by reacting compounds 9a and 9b, bearing an aminomethyl group, with 2,6-dichloropurine (Scheme 1), taking advantage of the reactivity differences in the 2- and 6-position of the purine derivative. These differences allow the controlled introduction of different substituents in the corresponding positions via nucleophilic substitution reaction [for literature examples, see references^65‐67].

The linkage of the amino group of 9a and 9b to the 6-position of 2,6-dichloropurine was carried out in n-butanol under microwave irradiation. By using this technique, the reaction time of 24 h, required with conventional heating, ³² was reduced to 1 h and allowed the preparation of 8a and 8b with 81% and 85% yield, respectively.

As shown in our previous study, ³² the nucleophilic displacement of the chlorine atom in the 2-position of the purine unit of 8a and 8b requires relatively drastic reaction conditions, which reflects the unfavorable influence of the bulky C6-substituent on the nucleophilic substitution at purine C2. The successful syntheses of the target compounds can be realized by the use of microwave irradiation and sealed tubes. The reaction conditions for the preparation of compounds 1a-7a, 2b, 3b, and 5b-7b as well as the yields are summarized in Table S1 in Supporting Information.

Binding Properties of Compounds 1a-7a, 2b, 3b, and 5b-7b: ¹H NMR Spectroscopic and Microcalorimetric Titrations

The compounds described in this work were developed as artificial receptors especially for carbohydrates with equatorial substitution pattern, such as β-glucosides. As a substrate for the binding studies was selected octyl β-D-glucoside (βGlc), the use of which enables the comparison of the binding properties of the new compounds with those of the previously studied receptor molecules. The performed binding studies included ¹H NMR spectroscopic titrations and measurements by isothermal titration calorimetry (ITC). In the course of these investigations, the influence of the different structural subunits on the binding properties was analyzed.

In addition to the ¹H NMR spectroscopic titrations in which the concentration of the tested compound was kept constant and that of the carbohydrate was varied, inverse titrations were also carried out in which the concentration of the sugar remained unchanged and that of the binding partner was varied. The extensive titration experiments thus enabled a detailed analysis of the complexation-induced shifts of the receptor signals and those of the carbohydrate substrate. The titration experiments were performed in CDCl₃ at 20 °C. The ¹H NMR titration data were analyzed using the WinEQNMR ⁶⁸ and SupraFit ⁶⁹ programs; the binding constants are summarized in Table 1. The complex stoichiometry was analyzed on the basis of the mole ratio method ⁷⁰ (for an example, see Figure S1 in Supporting Information).

Table 1.

Association Constants^a-c for Compounds 1a-7a, 2b, 3b, and 5b-7b and Octyl β-D-Glucoside (¹H NMR Titrations).

Aromatic group connected to the purine ring via linker -NH(CH2)n-	Compounds with pyridinyl groups		Compounds with pyrimidinyl groups
	Compound number	K11 / K21 [M-1]	Compound number	K11 / K21 [M-1]
NHCH2-phenyl	1a 32	42700 / 290c	—	—
NHCH2CH2-phenyl	2a	67800	2b	77500
NHCH2CH2CH2-phenyl	3a	81300	3b	101500
NHCH2-naphthyl	4a	64600 / 150c	—	—
NHCH2CH2-naphthyl	5a	95000	5b	113000
NHCH2CH2-indolyl	6a	66600	6b	74600
NHCH2CH2-(OMe)indolyl	7a	54200 / 590c	7b	66700 / 260c

^aAverage K _a values from multiple ¹H NMR titrations in CDCl₃.

¹H NMR titrations in CDCl.

^bErrors were estimated at ≤ 5%.

^cK₂₁ corresponds to the 2:1 receptor-sugar association constant.

It should be noted that although the molecular recognition of carbohydrates in water is of great importance, the binding studies in organic media should not be neglected. Analysis of the binding mode of carbohydrate-binding proteins has shown that the hydrogen bonds formed between these proteins and the sugar substrate are shielded from the bulk solvent, meaning that they exist in a lower dielectric environment.^21,23 Thus, binding studies in organic media such as chloroform provide valuable information about the factors that contribute to the affinity between receptors and sugars and can make an important contribution to our understanding of the complex carbohydrate-mediated processes in nature. Such studies are also an important tool in the search for new structural elements that are useful for the construction of artificial receptors with both a selective and effective mode of action.

The addition of βGlc to the solution of the corresponding compound led to significant changes in the ¹H NMR spectra. As an example, Figure 5 shows excerpts of the ¹H NMR spectra of compounds 5a and 5b, illustrating the chemical shifts of the NH and CH₂ signals as a function of increasing concentration of octyl β-D-glucoside. These NMR experiments indicated that among the tested compounds, the derivatives with aminopyrimidine-based recognition units (2b, 3b, and 5b-7b) possess better binding properties than their analogs with pyridinyl groups (2a, 3a, and 5a-7a).

Figure 5.

Excerpts from ¹H NMR spectra of compound 5a bearing pyridinyl groups (right, green) and of the analog 5b with pyrimidinyl groups (left, blue) during titration with octyl β-D-glucoside (βGlc). Shown are the shifts of the NH (marked by squares and triangles) and CH₂ (marked by circles) signals. [5a] = 1.00 mM, 0.00-4.00 equiv of βGlc; [5b] = 1.04 mM, 0.00-3.00 equiv of βGlc.

This was also confirmed by the analysis of the data obtained on the basis of the inverse titrations. It should be noted that in the case of the last-mentioned titrations, the CH signals of βGlc shifted upfield indicating its involvement in CH···π interactions with the corresponding receptor. Figure S1 in Supporting Information illustrates exemplary the changes in the ¹H NMR spectra of βGlc after the addition of compound 2b (the signals of the CH-2 and CH-4 protons of βGlc show the largest chemical shift).

The detailed analysis of the titration data revealed that in the case of the compounds bearing pyrimidinyl groups (2b, 3b, and 5b-7b), the association constants are increased by a factor in the range of 1.1 to 1.3 compared to the analogs with pyridinyl moieties (2a, 3a, and 5a-7a; see Table 1). It should be noted that in the case of the previously studied 1,3,5-substituted 2,4,6-triethylbenzenes with three identically substituted side-arms, the replacement of the aminopyridine- by aminopyrimidine-based units led to a greater increase in binding affinity. ⁷¹

For the purine-containing molecules described here, the aminopyridine- or aminopyrimidine-based recognition units are not the only hydrogen bonding sites, but an important role is also played by the purine residue. The latter also acts as a bridging unit, allowing the introduction of a well-positioned second aromatic group (in addition to the central benzene ring), which can participate in additional CH···π interactions contributing to the stabilization of the receptor-sugar complex. Thus, the importance of the combination of the different receptor subunits involved in the formation of various (combined) noncovalent interactions should be emphasized. The significant role of combined noncovalent interactions has also been identified in many other processes of molecular recognition (see, for example, reference ⁷² ; for examples of further discussions on supramolecular complexations, see references^73‐79).

The complexation properties of all tested compounds (1a-7a, 2b, 3b, and 5b-7b) depend on the substituent at C2 of the purine moiety. As expected, the type of the aromatic group and the length of the unit connecting this group with the purine ring have a strong influence on the binding properties.

In accordance with the aim pursued with the design of this type of carbohydrate receptors, the presence of a suitable and well-positioned aromatic moiety at C2 of the purine ring is favorable for the targeted formation of 1:1 receptor-carbohydrate complexes. At this point it should be mentioned that the tendency to form complexes with higher stoichiometry represents a disadvantage of some acyclic compounds.

When considering the influence of the type of aromatic unit (phenyl/naphthyl vs indolyl), it can be seen that the compounds equipped with a naphthyl or phenyl group are stronger receptors than those with an indolyl group. Furthermore, the complexation behavior is strongly influenced by the length of the –(CH₂)_n– unit. The correct positioning of the aromatic group when forming the complex with the carbohydrate substrate is only possible with a suitable linker. Among all tested compounds, the β-glucoside is complexed most strongly by the naphthyl-containing compound 5b under the conditions used (within the compounds bearing pyridinyl groups, compound 5a was identified as the most effective).

In the case of the phenyl group-bearing compounds 1a-3a, 2b, and 3b, a –(CH₂)₃– unit seems to be more favorable than a –(CH₂)₂– linker, and as expected, the –CH₂– unit (n = 1) is the least suitable. This behavior is also suggested by the results of molecular modeling calculations of the respective receptor-sugar complexes. For example, in the case of compound 1a, the linking unit is shorter, so that the phenyl ring is arranged at a less favorable inclination angle to the plane of the pyranose ring. In contrast, the longer –(CH₂)₃– unit in 3a causes the phenyl ring to align at an inclination angle more favorable for the formation of CH···π interactions and the associated sandwich-like complexation (see Figure 6).

Figure 6.

Energy-minimized structures of the 1:1 complexes formed between methyl β-glucopyranoside and compounds 1a (a), 3a (b), 4a (c) and 5a (d) (MacroModel, OPLS_2005 force field, MCMM, 50000 steps); examples of N-H···O and O-H···N hydrogen bonds and CH···π interactions in the complex. Color code: receptor N, blue; receptor C, gray; sugar O, red; sugar C, orange.

The presence of an indolyl group, which is also used by carbohydrate-binding proteins for the formation of CH···π contacts, should have a positive effect on the binding strength, but compounds 6a, 7a, 6b, and 7b show poorer binding properties than the corresponding derivative bearing a naphthyl group (compounds 5a and 5b).

As indicated by molecular modeling calculations, intramolecular hydrogen bonds may be responsible for this. For example, the energy-minimized structure of 7b (see Figure S2 in Supporting Information) shows an intramolecular hydrogen bond between the indole NH and one of the pyrimidine N atoms. The involvement of the receptor subunits in intramolecular interactions negatively affects their ability to participate in the formation of intermolecular interactions with the substrate. Negative effects of intramolecular interaction on binding properties are well known ⁸⁰ and have also been observed by us in the case of some acyclic^81,82 and macrocyclic^41‐43 compounds tested as carbohydrate-binding agents. However, it should also be noted that intramolecular interactions can play a positive role as so-called “structure-promoting elements”, ⁸⁰ for example, by stabilizing a molecular conformation that has a more favorable degree of preorganization for the recognition process.^80,83‐85

Among the compounds bearing a 3-indolyl or 5-methoxy-3-indolyl group (compounds 6a, 7a, 6b, and 7b), compound 6b was identified as the most effective. The presence of the methoxy group in the 5-position of the indole ring was found to be unfavorable. Compared to the structural analog with a naphthyl group (compound 5b), the affinity of 6b is noticeably lower (compound 5b has about 1.5 times higher affinity) (Figure 7).

Figure 7.

Binding efficiency of compounds 1a-7a and 2b, 3b, 5b-7b toward octyl β-D-glucoside (comparative consideration).

Compounds with an unsuitable linker or aromatic group represent weaker receptor molecules. In these cases, in addition to the 1:1 complexes, the formation of weaker 2:1 receptor-substrate complexes was also observed (for example in the case of compound 1a).

The binding properties of compound 5b, which proved to be the most effective among the tested compounds, were further analyzed by isothermal titration calorimetry (ITC). The microcalorimetric titrations were carried out in CHCl₃ at 20°C by adding increasing amounts of the sugar to a solution of compound 5b. The measurement data obtained from three independent microcalorimetric titrations were evaluated on the basis of NanoAnalyze and SupraFit programs (Figure 8). For all measurements the best fit of the titration data was obtained with the 1:1 receptor-sugar binding model. As given in Table 2, the results obtained with isothermal titration calorimetry confirmed those determined on the basis of the ¹H NMR method.

Figure 8.

Exemplary ITC thermogram (left) and titration curve-fitting (right) for the titration of 5b with octyl β-D-glucoside in CHCl₃ (the heat of dilution has been subtracted). Titration mode: addition of the monosaccharide (c_syringe= 6.11 mM) into 5b (c_cell= 0.51 mM) at 293 K in 46 steps (each 6 µL), calorimeter: MicroCal^TM VP-ITC.

Table 2.

Results of Microcalorimetric Titrations of Compound 5b with Octyl β-D-Glucoside.^a,b

Compound	K11 [m−1]	ΔG [kJ/mol]	ΔH [kJ/mol]	−TΔS [kJ/mol]
5b	126000	−28.6	−53.7	25.1

^aIn dry CHCl₃ at 293 K; calorimeter: MicroCal^TM VP-ITC.

^bStandard deviations for a minimum of three replicated ITC titrations: about 0.1 kJ/mol.

Abbreviation: ITC, isothermal titration calorimetry.

The results of these investigations are now being used to further improve the design of receptor molecules and to develop systems with accurately predictable binding properties in different solvents, including aqueous media (for a discussion of solvent effects, see reference ⁸⁶ ). It should be noted that the recognition of carbohydrates in water is particularly challenging because a receptor must select its target molecule from a large excess of competing water molecules.

Conclusion

1,3,5-Substituted 2,4,6-triethylbenzene derivatives equipped with two aminopyridine- or aminopyrimidine-based recognition units and a purine moiety, which serves both as a hydrogen bonding site and as a bridging component for the incorporation of additional substituents, have the potential to act as effective carbohydrate receptors and to form predictable 1:1 complexes with a suitable substrate. The use of the purine moiety allows the construction of compounds with two aromatic groups (central benzene ring and the additional aromatic group connected to the C2 position of the purine ring via a linker) positioned to interact with the sugar CHs in a sandwich-like manner. The effective complexation depends on the nature of this aromatic group and the length of the –(CH₂)_n– unit linking this group to the C2 position of the purine ring. Compared to the macrocyclic system shown in Figure 2, the acyclic architecture enables a simpler construction of compounds containing either two identical or different aromatic groups that can participate in the interactions with the sugar CH units. For example, it was found that the incorporation of a well-positioned phenyl or naphthyl group predictably leads to the formation of strong 1:1 complexes in which the substrate is sandwiched between two benzene rings or between a benzene and a naphthalene moiety.

The use of the purine moiety in the construction of carbohydrate receptors with a 1,3,5-substituted 2,4,6-triethylbenzene backbone has proved to be a very promising approach, and it should be emphasized that the possibilities for structural variations of this molecular architecture are manifold. Thus, a whole range of compounds can be synthesized to perform extensive studies on the relationships between structure and binding efficiency.

Materials and Methods

Experimental Section

Analytical TLC was carried out on silica gel 60 F₂₅₄ plates, column chromatography was carried out on silica gel. Melting points are uncorrected. Benzylamine, 2-phenylethylamine, 3-phenyl-1-propylamine, 2-(1-naphthyl)ethylamine, tryptamine and 5-methoxytryptamine are commercially available; the synthesis of 1-naphthylmethylamine was performed according to the procedure described in reference ⁶⁴ . The preparation of 8a and 8b is described in reference ³² .

General Procedure for the Preparation of Compounds 1a-7a, 2b, 3b, and 5b-7b

Route A (Sealed Tube)

1-[(2-Chloro-9H-purine-6-yl)-aminomethyl]-3,5-bis-[(4,6-dimethyl-pyridin-2-yl)-aminomethyl]-2,4,6-triethylbenzene (8a) or 1-[(2-Chloro-9H-purine-6-yl)-aminomethyl]-3,5-bis-[(4,6-dimethyl-pyrimidin-2-yl)-aminomethyl]-2,4,6-triethylbenzene (8b) and the corresponding amine in n-butanol (5 mL) were placed in a sealed tube. After purging with argon and closing of the tube, the reaction mixture was stirred at 125-140 °C for 6-7 d. Then, the solvent was removed by rotary evaporation, the residue washed with water and the crude product was purified via column chromatography, by means of preparative HPLC or recrystallization (for details, see below).

Route B (Microwave Conditions)

Compound 8a or 8b and the corresponding amine were dissolved in 3 mL n-butanol in a microwave vessel. After purging with argon and closing of the vessel, the microwave irradiation was applied for several hours at 160-170 °C (for details, see below). Afterwards, the solvent was removed under reduced pressure, the residue washed with water and the crude product was purified via column chromatography or recrystallization (for details, see below).

1-[(2-Benzylamino-9H-purin-6-yl)aminomethyl]-3,5-bis-[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (1a). ³²

Route A

Compound 1a was prepared from 8a (0.090 g, 0.146 mmol) and benzylamine (0.155 g, 1.46 mmol). Reaction time: 7 d, reaction temperature: 140 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, Isocratic CHCl₃/i-PrOH 91:9, t_R= 10.5 min). Yield 45% (0.045 g, 0.066 mmol).

Route B

Compound 1a was prepared from 8a (0.107 g, 0.175 mmol) and benzylamine (0.373 g, 3.49 mmol). Microwave irradiation (180 W) was applied for 8 h at 170 °C. Purification by column chromatography [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 79% (0.094 g, 0.138 mmol). R _f = 0.44 [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution]. Mp 117 °C. ¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.22 (t, J = 7.6 Hz, 6 H), 1.25 (t, J = 7.3 Hz, 3 H), 2.25 (s, 6 H), 2.34 (s, 6 H), 2.72-2.84 (m, 6 H), 4.38 (s, 4 H), 4.68 (s, 2 H), 4.73 (brs, 2 H), 6.19 (s, 2 H), 6.36 (s, 2 H), 7.24 (t, J = 7.3 Hz, 1 H), 7.32 (t, J = 7.6 Hz, 2 H), 7.40 (d, J = 7.4 Hz, 2 H), 7.54 (s, 1H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.12 (t, J = 7.1 Hz, 6 H), 1.21 (t, J = 7.5 Hz, 3 H), 2.22 (s, 6 H), 2.33 (s, 6 H), 2.45-2.60 (m, 4 H), 2.68-2.75 (m, 2 H), 4.29 (brs, 4 H), 4.61 (s, 4 H), 5.79 (brs, NH), 6.14 (s, 2 H), 6.32 (s, 2 H), 7.15 (s, 1 H), 7.22-7.32 (m, 5 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.5, 16.8, 21.2, 22.7, 22.9, 24.1, 38.4, 40.6, 45.8, 103.6, 113.7, 127.0, 127.4, 128.5, 132.2, 132.8, 134.8, 139.8, 143.6, 148.9, 151.5, 154.3, 156.6, 158.5, 159.8 ppm. HRMS (ESI): calcd for C₄₁H₅₁N₁₀: 683.429268 [M + H]⁺; found: 683.429271.

1-[(2-Phenylethylamino-9H-purin-6-yl)aminomethyl]-3,5-bis-[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (2a)

Route A

Compound 2a was prepared from 8a (0.088 g, 0.144 mmol) and phenylethylamine (0.174 g, 1.44 mmol). Reaction time: 7 d, Reaction temperature: 140 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, gradient CHCl₃/i-PrOH from 88:12 to 80:20 in 12 min, t_R= 6.8 min). Yield 18% (0.018 g, 0.026 mmol).

Route B1 (Preparative HPLC)

Compound 2a was prepared from 8a (0.080 g, 0.131 mmol) and phenylethylamine (0.157 g, 1.30 mmol). Microwave irradiation (200 W) was applied for 2 h at 160 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, gradient CHCl₃/i-PrOH from 88:12 to 80:20 in 12 min, t_R= 6.8 min). Yield 41% (0.037 g, 0.053 mmol).

Route B2 (Column Chromatography)

Compound 2a was prepared from 8a (0.108 g, 0.176 mmol) and phenylethylamine (0.43 g, 3.57 mmol). Microwave irradiation (200 W) was applied for 8 h at 160 °C. Purification by column chromatography [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 64% (0.078 g, 0.112 mmol). R _f = 0.51 [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution]. Mp 158 °C.¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.26 (t, J = 7.4 Hz, 9 H), 2.26 (s, 6 H), 2.34 (s, 6 H), 2.75-2.83 (m, 8 H), 2.95 (t, J = 6.9 Hz, 2 H), 4.37 (s, 4 H), 4.80 (s, 2 H), 6.20 (s, 2 H), 6.37 (s, 2 H), 7.18-7.26 (m, 3 H), 7.32 (dd, J = 8.2, 6.8 Hz, 2 H), 7.87 (s, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.06-1.14 (m, 6 H), 1.16-1.24 (m, 3 H), 2.22 (s, 6 H), 2.36 (s, 6 H), 2.48-2.62 (m, 4 H), 2.66-2.76 (m, 2 H), 2.86-2.97 (m, 2 H), 3.55-3.75 (m, 2 H), 4.31 (s, 4 H), 4.64 (s, 2 H), 5.32 (brs, NH), 5.54 (brs, NH), 6.15 (s, 2 H), 6.33 (s, 2 H), 7.20-7.30 (m, 6 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.5, 16.8, 21.1, 22.8, 23.0, 24.1, 39.3, 40.0, 40.6, 43.4, 103.9, 113.9, 117.9, 126.2, 128.5, 128.8, 131.2, 133.1, 138.8, 139.7, 143.8, 144.0, 149.0, 151.2, 154.3, 154.5, 156.5, 158.3 ppm. HRMS (ESI): calcd for C₄₂H₅₃N₁₀: 697.444918 [M + H]⁺; found: 697.444930.

1-{[2-(Phenylpropylamino)-9H-purin-6-yl]aminomethyl}-3,5-bis[4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (3a)

Route B

Compound 3a was prepared from 8a (0.071 g, 0.116 mmol) and phenylpropylamine (0.314 mg, 2.32 mmol). Microwave irradiation (250 W) was applied for 11 h at 170 °C. Purification by column chromatography [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 72% (0.059 g, 0.083 mmol). R _f = 0.42 [CHCl₃/i-PrOH incl. (2% 7 N NH₃ in CH₃OH) 7:1 v/v]. Mp 124 °C.¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.25 (t, J = 7.5 Hz, 6 H), 1.26 (t, J = 7.5 Hz, 3 H), 2.01 (p, J = 7.3 Hz, 2 H), 2.25 (s, 6 H), 2.34 (s, 6 H), 2.73-2.85 (m, 8 H), 3.48 (t, J = 7.0 Hz, 2 H), 4.38 (s, 4 H), 4.75 (s, 2 H), 6.20 (s, 2 H), 6.36 (s, 2 H), 7.14-7.19 (m, 1 H), 7.20-7.29 (m, 4 H), 7.55 (s, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.13 (t, J = 6.8 Hz, 6 H), 1.21 (t, J = 6.9 Hz, 3 H), 1.90 (s, 2 H), 2.22 (s, 6 H), 2.36 (s, 6 H), 2.45-2.60 (m, 4 H), 2.61-2.76 (m, 4 H), 3.41 (s, 2 H), 4.27 (s, 4 H), 4.61 (s, 2 H), 5.36 (brs, NH), 6.16 (s, 2 H), 6.33 (s, 2 H), 7.08-7.18 (m, 3 H), 7.19-7.25 (m, 2 H), 7.34 (s, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.6, 16.9, 21.3, 22.8, 23.0, 24.2, 31.3, 33.4, 38.5, 40.8, 41.6, 103.7, 113.8, 125.9, 128.4 (2C), 132.3, 132.7, 134.7, 141.9, 143.6 (2C), 149.1, 151.5, 154.3, 156.6, 158.6, 160.1 ppm. HRMS (ESI): calcd for C₄₃H₅₅N₁₀: 711.460568 [M + H]⁺; found: 711.460561.

1-{2-[(Naphth-1-yl)methylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis[4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (4a)

Route B (Preparative HPLC)

Compound 4a was prepared from 8a (0.080 g, 0.131 mmol) and 1-naphthylmethylamine (0.205 g, 1.28 mmol). Microwave irradiation (150 W) was applied for 4 h at 160 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, gradient CHCl₃/i-PrOH from 96:4 to 94:6 in 20 min, t_R= 6.9 min). Yield 34% (0.032 g, 0.044 mmol). R _f = 0.25 [CHCl₃/i-PrOH 15:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Mp 138 °C. ¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.20 (t, J = 7.2 Hz, 6 H), 1.25 (t, J = 7.4 Hz, 3 H), 2.25 (s, 6 H), 2.34 (s, 6 H), 2.73-2.83 (m, 6 H), 3.98 (s, 4 H), 4.37 (s, 4 H), 4.74 (s, 2 H), 5.14 (s, 2 H), 6.18 (s, 2 H), 6.36 (s, 2 H), 7.44 (dd, J = 8.2, 7.0 Hz, 1 H), 7.49-7.59 (m, 4 H), 7.79 (d, J = 8.3 Hz, 1 H), 7.86-7.90 (m, 1 H), 8.17 (d, J = 8.1 Hz, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.11 (t, J = 6.9 Hz, 6 H), 1.19 (t, J = 6.8 Hz, 3 H), 2.20 (s, 6 H), 2.32 (s, 6 H), 2.50-2.63 (m, 4 H), 2.66-2.77 (m, 2 H), 4.30 (s, 4 H), 4.63 (s, 2 H), 5.06 (s, 2 H), 5.60 (brs, NH), 6.11 (s, 2 H), 6.30 (s, 2 H), 7.34-7.41 (m, 1 H), 7.41-7.50 (m, 2 H), 7.51-7.56 (m, 1 H), 7.77 (d, J = 8.0 Hz, 1 H), 7.85 (d, J = 7.4 Hz, 1 H), 8.02-8.13 (m, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ 16.5, 16.8, 21.1, 22.7, 22.9, 24.0, 38.4, 40.6, 44.0, 103.6, 113.7, 113.8, 123.6, 125.4, 125.7, 125.9, 126.1, 127.9, 128.6, 131.6, 132.7, 133.7, 134.7, 134.9, 143.5, 143.6, 148.9, 151.5, 154.4, 156.6, 158.6, 159.8 ppm. HRMS (ESI): calcd for C₄₅H₅₃N₁₀: 733.444918 [M + H]⁺; found: 733.444920.

1-{{2-[(Naphth-1-yl)ethylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis[4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (5a)

Route B. Compound 5a was prepared from 8a (0.093 g, 0.15 mmol) and 2-(1-naphthyl)ethylamine (0.531 g, 3.10 mmol). Microwave irradiation (170 W) was applied for 14 h at 170 °C. Purification by column chromatography [CHCl₃/i-PrOH 10:1 (v/v) incl. 3% ammonia solution (7N in CH₃OH)]. Yield 80% (0.078 g, 0.104 mmol) R _f = 0.40 [CHCl₃/i-PrOH 10:1 (v/v) incl. 3% ammonia solution]. Mp 124 °C. ¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.23 (t, J = 7.6 Hz, 6 H), 1.24 (t, J = 7.8 Hz, 3 H), 2.24 (s, 6 H), 2.34 (s, 6 H), 2.72-2.82 (m, 6 H), 3.38-3.53 (m, 2 H), 3.80-3.87 (m, 2 H), 4.35 (s, 4 H), 4.78 (s, 2 H), 6.16 (s, 2 H), 6.34 (s, 2 H), 7.29-7.33 (m, 2 H), 7.35-7.50 (m, 3 H), 7.72-7.77 (m, 1 H), 7.82-7.88 (m, 1 H), 8.19 (d, J = 8.0 Hz, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ =  1.03-1.20 (m, 9 H), 2.21 (s, 6 H), 2.36 (s, 6 H), 2.43-2.58 (m, 4 H), 2.61-2.73 (m, 2 H), 3.31-3.44 (m, 2 H), 3.66-3.80 (m, 2 H), 4.26 (s, 4 H), 4.63 (brs, 2 H), 5.65 (brs, NH), 6.16 (s, 2 H), 6.33 (s, 2 H), 7.30-7.44 (m, 5 H), 7.64-7.74 (m, 1 H), 7.81 (d, J = 8.0 Hz, 1 H), 8.04-8.16 (m, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.5, 16.8, 21.2, 22.7, 24.1, 33.2, 38.4, 40.7, 42.8, 103.5, 113.7, 123.8, 125.5, 125.8, 126.7, 127.0, 128.7, 128.8, 132.0, 132.6, 132.7, 133.8, 134.8, 135.8, 143.5, 143.6, 148.9, 154.3, 156.6, 158.6, 159.8 ppm. HRMS (ESI): calcd for C₄₆H₅₅N₁₀: 747.460568 [M + H]⁺; found: 747.460569.

1-{{2-[(Indol-3-yl)ethylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis-[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (6a)

Route A

Compound 6a was prepared from 8a (0.080 g, 0.131 mmol) and tryptamine (0.26 g, 1.63 mmol). Reaction time: 7.5 d, Reaction temperature: 140 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, Isocratic CHCl₃/i-PrOH 94:6, t_R= 24.5 min). Yield 30% (0.030 g, 0.041 mmol).

Route B

Compound 6a was prepared from 8a (0.101 g, 0.165 mmol) and tryptamine (0.538 g, 3.36 mmol). Microwave irradiation (150 W) was applied for 10 h at 170 °C. Purification by column chromatography [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 60% (0.073 g, 0.10 mmol). R _f = 0.46 [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution]. Mp 142 °C. ¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.25 (t, J = 7.6 Hz, 6 H), 1.26 (t, J = 7.6 Hz, 3 H), 2.25 (s, 6 H), 2.34 (s, 6 H), 2.78 (q, J = 7.6 Hz, 2 H), 2.81 (q, J = 7.6 Hz, 4 H), 3.14 (t, J = 6.9 Hz, 2 H), 3.79 (t, J = 6.9 Hz, 2 H), 3.93 (s, 1H), 4.38 (s, 4 H), 4.78 (s, 2 H), 6.19 (s, 2 H), 6.35 (s, 2 H), 7.03-7.08 (m, 1 H), 7.12 (s, 1 H), 7.13-7.17 (m, 1 H), 7.39 (d, J = 8.1 Hz, 1 H), 7.53 (s, 1 H), 7.66 (d, J = 7.8 Hz, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.07-1.21 (m, 9 H), 2.21 (s, 6 H), 2.35 (s, 6 H), 2.47-2.63 (m, 4 H), 2.63-2.73 (m, 2 H), 3.02-3.12 (m, 2 H), 3.67-3.78 (m, 2 H), 4.29 (s, 4 H), 4.63 (s, 2 H), 5.32 (brs, NH), 6.15 (s, 2 H), 6.33 (s, 2 H), 6.92 (s, 1 H), 7.03 (t, J = 6.1 Hz, 1 H), 7.14 (t, J = 7.6 Hz, 1 H), 7.29 (d, J = 8.2 Hz, 1 H), 7.59 (d, J = 8.0 Hz, 1 H), 8.39 (brs, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.6, 16.8, 21.2, 22.8, 22.9, 24.1, 25.3, 38.5, 40.6, 42.3, 103.7, 111.2, 113.3, 113.5, 113.7, 118.8, 119.2, 121.9, 122.1, 127.4, 132.2, 132.8, 134.7, 136.4, 143.4, 143.6, 148.9, 151.5, 154.2, 156.6, 158.5, 159.8 ppm. HRMS (ESI): calcd for C₄₄H₅₄N₁₁: 736.455817 [M + H]⁺; found: 736.455817.

1-2-[(5-Methoxyindol-3-yl)ethylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis[4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (7a)

Route A

Compound 7a was prepared from 8a (0.080 g, 0.13 mmol) and 5-methoxytryptamine (0.25 g, 1.30 mmol). Reaction time: 7.5 d. Reaction temperature: 130 °C. Purification by preparative HPLC (Nucleodur 100-5 NH₂ Macherey-Nagel, Isocratic CHCl₃/i-PrOH 93:7, t_R= 12.6 min). Yield 65% (0.065 g, 0.02 mmol).

Route B

Compound 7a was prepared from 8a (0.10 g, 0.16 mmol) and 5-methoxytryptamine (0.63 mg, 3.31 mmol). Microwave irradiation (150 W) was applied for 12 h at 170 °C. Purification by column chromatography [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 58% (0.073 g, 0.10 mmol). R _f = 0.47 [CHCl₃/i-PrOH 7:1 (v/v) incl. 2% ammonia solution]. Mp 130 °C. ¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.25 (t, J  =  7.5 Hz, 6 H), 1.25 (t, J = 7.4 Hz, 3 H), 2.25 (s, 6 H), 2.34 (s, 6 H), 2.77 (q, J  =  7.5 Hz, 2 H), 2.81 (q, J  =  7.4 Hz, 4 H), 3.11 (t, J  =  6.9 Hz, 2 H), 3.77 (t, J  =  6.9 Hz, 2 H), 3.82 (s, 3 H), 3.89 (brs, 1 H), 4.37 (s, 4 H), 4.77 (s, 2 H), 6.18 (s, 2 H), 6.35 (s, 2 H), 6.82 (dd, J = 8.8, 2.4 Hz, 1 H), 7.09 (d, J = 2.5 Hz, 1 H), 7.10 (s, 1 H), 7.28 (d, J = 8.8 Hz, 1 H), 7.53 (s, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.06-1.22 (m, 9 H), 2.21 (s, 6 H), 2.35 (s, 6 H), 2.50-2.64 (m, 4 H), 2.64-2.75 (m, 2 H), 2.90-3.12 (m, 2 H), 3.67-3.78 (m, 2 H), 3.78 (s, 3 H), 4.31 (s, 4 H), 4.64 (s, 2 H), 5.29 (brs, NH), 6.14 (s, 2 H), 6.33 (s, 2 H), 6.81 (dd, J = 8.7, 2.0 Hz, 1 H), 6.90 (s, 1 H), 7.01 (s, 1 H), 7.18 (d, J = 8.7 Hz, 1 H), 7.22 (s, 1 H), 8.33 (br s, 1H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.6, 16.8, 21.1, 22.8, 22.9, 24.1, 25.3, 38.6, 40.6, 42.2, 55.9, 100.6, 103.7, 111.9, 112.2, 113.0, 113.1, 113.7, 122.9, 127.8, 131.6, 132.3, 132.9, 134.7, 143.6, 143.7, 148.8, 151.5, 153.9, 154.4, 156.6, 158.5, 159.8 ppm. HRMS (ESI): calcd for C₄₅H₅₆N₁₁O: 766.466382 [M + H]⁺; found: 766.466381.

1-[(2-Phenylethylamino-9H-purin-6-yl)aminomethyl]-3,5-bis-[(4,6-dimethylpyrimidin-2-yl)aminomethyl]-2,4,6-triethylbenzene (2b)

Route B

Compound 2b was prepared from 8b (0.100 g, 0.160 mmol) and phenylethylamine (0.45 mL, 3.56 mmol). Microwave irradiation (200 W) was applied for 12 h at 160 °C. Purification by filtration and recrystallization from acetonitrile. Yield 79% (0.089 g, 0.13 mmol). R _f = 0.65 [CHCl₃/i-PrOH 8:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH). Mp 140 °C.¹H NMR (500 MHz, CDCl₃): δ = 1.12 (t, J = 7.6 Hz, 6 H), 1.19 (t, J = 7.4 Hz, 3 H), 2.28 (s, 12 H), 2.42-2.56 (m, 4 H), 2.63-2.74 (m, 2 H), 2.85-2.93 (s, 2 H), 3.63 (s, 2 H), 4.49 (s, 4 H), 4.63 (s, 2 H), 5.60 (brs), 5.90 (brs), 6.33 (s, 2 H), 7.14-7.20 (m, 3 H), 7.20-7.25 (m, 2 H), 7.31 (s, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.4, 16.7, 22.7, 22.9, 23.9, 35.9, 38.3, 39.8, 43.5, 109.7, 126.2, 128.4, 128.8, 132.2, 132.5, 134.5, 139.7, 143.5, 143.6, 162.0, 167.6 ppm. HRMS (ESI): calcd for C₄₀H₅₁N₁₂: 699.43542 [M + H]⁺; found: 699.43619.

1-[(2-Phenylpropylamino-9H-purin-6-yl)aminomethyl]-3,5-bis-[(4,6-dimethylpyrimidin-2-yl)aminomethyl]-2,4,6-triethylbenzene (3b)

Route B

Compound 3b was prepared from 8b (0.102 g, 0.170 mmol) and 3-phenylpropylamine (0.24 mL, 1.69 mmol). Microwave irradiation (250 W) was applied for 16 h at 170 °C. Purification by recrystallization from ethyl acetate and acetonitrile. Yield 69% (0.082 g, 0.12 mmol). R _f = 0.63 [CHCl₃/CH₃OH 10:1 v/v]. Mp 129 °C.¹H NMR (500 MHz, CDCl₃): δ = 1.11 (t, J = 7.4 Hz, 6 H), 1.19 (t, J = 7.3 Hz, 3 H), 1.82-1.93 (m, 2 H), 2.30 (s, 12 H), 2.39-2.51 (m, 4 H), 2.61-2.67 (m, 2 H), 2.67-2.74 (m, 2 H), 3.33-3.49 (m, 2 H), 4.49 (s, 4 H), 4.57 (d, J = 4.4 Hz, 2 H), 5.58 (brs), 5.77 (brs), 6.34 (s, 2 H), 7.09-7.17 (m, 3 H), 7.20-7.25 (m, 2 H), 7.31 (s, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.4, 16.7, 22.7, 23.0, 24.0, 31.3, 33.4, 38.2, 39.8, 41.4, 109.6, 113.3, 125.8, 128.3, 132.3, 132.5, 134.5, 141.9, 143.4, 143.7, 151.7, 154.2, 160.0, 162.0, 167.6 ppm. HRMS (ESI): calcd for C₄₁H₅₃N₁₂: 713.45107 [M + H]⁺; found: 713.45328.

1-[(2-Naphthylethylamino-9H-purin-6-yl)aminomethyl]-3,5-bis-[(4,6-dimethylpyrimidin-2-yl)aminomethyl]-2,4,6-triethylbenzene (5b)

Route B

Compound 5b was prepared from 8b (0.080 g, 0.131 mmol) and 2-(1-naphthyl)ethylamine (0.098 mg, 0.57 mmol). Microwave irradiation (200 W) was applied for 26 h at 170 °C. Purification by recrystallization from acetonitrile. Yield 64% (0.062 g, 0.083 mmol). R _f = 0.73 [CHCl₃/CH₃OH 7:1 v/v]. Mp 140 °C.¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.24 (t, J = 7.6 Hz, 6 H), 1.25 (t, J = 7.4 Hz, 3 H), 2.32 (s, 12 H), 2.74-2.86 (m, 6 H), 3.40-3.49 (m, 2 H), 3.77-3.88 (m, 2 H), 4.60 (s, 4 H), 4.81 (s, 2 H), 6.38 (s, 2 H), 7.40-7.44 (m, 3 H), 7.46-7.60 (m, 3 H), 7.75 (dd, J = 6.0, 2.8 Hz, 1 H), 7.87 (d, J = 8.0 Hz, 1 H), 8.22 (d, J = 8.0 Hz, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃/CD₃OD 5:1): δ = 16.4 (2C), 23.0, 23.4, 30.1, 33.3, 39.8, 42.5, 109.9, 123.7, 125.5, 125.8, 126.7, 127.0, 128.7, 132.0, 132.3, 132.6, 133.9, 135.6, 143.9, 144.0, 161.4, 167.7 ppm. HRMS (ESI): calcd for C₄₄H₅₃N₁₂: 749.45107 [M + H]⁺; found: 749.45121.

1-{{2-[(Indol-3-yl)ethylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis-[(4,6-dimethyl-pyrimidin-2-yl)aminomethyl]-2,4,6-triethylbenzene (6b) ²⁷

Route B

Compound 6b was prepared from 8b (0.10 g, 0.16 mmol) and tryptamine (0.26 g, 1.36 mmol). Microwave irradiation (160 W) was applied for 6 h at 170 °C. Purification was performed by column chromatography [CHCl₃/i-PrOH 2:1 (v/v) incl. 2% ammonia solution (7N in CH₃OH)]. Yield 42% (0.05 g, 0.07 mmol). Mp 165 °C.¹H NMR (500 MHz, CDCl₃/CD₃OD 5:1): δ = 1.23 (t, J  =  7.4 Hz, 6 H), 1.26 (t, J = 8.4 Hz, 3 H), 2.32 (s, 12 H), 2.80 (q, J  =  7.4 Hz, 2 H), 2.82 (q, J  =  7.3 Hz, 4 H), 3.14 (t, J  =  6.8 Hz, 2 H), 3.79 (t, J  =  6.4 Hz, 2 H), 3.95 (brs, 1 H), 4.60 (s, 4 H), 4.78 (s, 2 H), 6.39 (s, 2 H), 7.04-7.08 (m, 1 H), 7.12 (s, 1 H), 7.13-7.17 (m, 1 H), 7.39 (d, J = 8.1 Hz, 1 H), 7.53 (brs, 1 H), 7.66 (d, J = 7.8 Hz, 1 H) ppm. ¹H NMR (500 MHz, CDCl₃): δ = 1.22-1.28 (m, 9 H), 2.32 (s, 12 H), 2.48-2.60 (m, 4 H), 2.64-2.74 (m, 2 H), 3.00-3.09 (m, 2 H), 3.67-3.77 (m, 2 H), 4.51 (s, 4 H), 4.63 (s, 2 H), 5.42 (brs, NH), 5.59 (brs, NH), 6.33 (s, 2 H), 6.94 (s, 1 H), 7.00-7.06 (m, 1 H), 7.11-7.14 (m, 1 H), 7.21-7.25 (m, 1 H), 7.29 (d, J = 8.1 Hz, 1 H), 7.60 (d, J = 7.9 Hz, 1 H), 8.39 (brs, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.5, 16.6, 22.8, 23.0, 23.9, 25.4, 39.9, 42.3, 43.3, 109.7, 111.2, 113.6, 115.1, 118.8, 119.2, 122.0 (2C), 127.5, 132.3, 132.7, 134.4, 136.4, 143.7 (2C), 151.9, 155.0, 160.0, 162.0, 167.6 ppm. HRMS (ESI): calcd for C₄₂H₅₂N₁₃: 738.446315 [M + H]⁺; found: 738.446317.

1-{{2-[(5-Methoxyindol-3-yl)ethylamino]-9H-purin-6-yl}aminomethyl}-3,5-bis-[(4,6-dimethyl-pyrimidin-2-yl)aminomethyl]-2,4,6-triethylbenzene (7b)

Route B

Compound 7b was prepared from 8b (0.10 g, 0.16 mmol) and 5-methoxytryptamine (0.35 g, 1.81 mmol). Microwave irradiation (200 W) was applied for 17 h at 170 °C. Purification was performed by column chromatography [CHCl₃/CH₃OH, 7:1] and recrystallization from acetonitrile. Yield 50% (0.06 g, 0.08 mmol). Mp 150 °C.¹H NMR (500 MHz, CDCl₃): δ  = 1.09 (t, J  =  7.5 Hz, 6 H), 1.17 (t, J = 7.4 Hz, 3 H), 2.24 (s, 12 H), 2.51-2.62 (m, 4 H), 2.63-2.74 (m, 2 H), 2.97-3.05 (m, 2 H), 3.69-3.75 (m, 2 H), 3.76 (s, 3 H), 4.49 (s, 4 H), 4.60 (s, 2 H), 6.31 (s, 2 H), 6.77 (dd, J = 8.7 Hz, 2.4 Hz, 1 H), 6.86 (s, 1 H), 6.96 (s, 1 H), 7.14 (d, J = 8.8 Hz, 1 H), 7.15 (s, 1 H), 8.57 (s, 1 H) ppm. ¹³C NMR (125 MHz, CDCl₃): δ = 16.5, 16.7, 22.9, 23.0, 24.0, 25.3, 38.4, 39.9, 42.3, 55.9, 100.6, 109.8, 112.0, 112.1, 113.1, 113.5, 123.0, 127.9, 131.6, 132.3, 132.6, 134.8, 143.7 (2C), 151.6, 153.9, 154.4, 159.9, 162.0, 167.7 ppm. HRMS (ESI): calcd for C₄₃H₅₄N₁₃O: 768.45688 [M + H]⁺; found: 768.45682.

Description of the ¹H NMR and Microcalorimetric Titrations

The ¹H NMR titrations were carried out in CDCl₃ at 20 °C [CDCl₃ was deacidified over basic aluminium oxide (Brockmann I) and stored over molecular sieve]. Stock solutions in CDCl₃ were prepared and homogenized for the respective receptor and sugar. These solutions and CDCl₃ were added together in a manner that the concentration of the receptor was kept constant and that of the sugar was varied. The receptor concentration was adjusted to about 1 mM to avoid self-aggregation. For the inverse titrations, the concentration of the sugar was kept constant and that of the receptor was varied accordingly. For each titration, 15-20 samples were prepared, thoroughly homogenized and the ¹H NMR spectra were recorded. For each system at least three titrations were carried out.

ITC was carried out in CHCl₃ [deacified over basic aluminium oxide (Brockmann I) and stored over molecular sieve] at 20 °C on MicroCal^TM VP-ITC. Three replicated ITC titrations were carried out; for details, see Figure 8 and Table 2.

Supplemental Material

sj-docx-1-npx-10.1177_1934578X241258352 - Supplemental material for Sandwich-Like Complexation of Carbohydrates by Hydrogen Bonding and CH-π Interactions

Supplemental material, sj-docx-1-npx-10.1177_1934578X241258352 for Sandwich-Like Complexation of Carbohydrates by Hydrogen Bonding and CH-π Interactions by Linda Köhler, Stefan Kaiser and Monika Mazik in Natural Product Communications

Supplemental Material

sj-pdf-2-npx-10.1177_1934578X241258352 - Supplemental material for Sandwich-Like Complexation of Carbohydrates by Hydrogen Bonding and CH-π Interactions

Supplemental material, sj-pdf-2-npx-10.1177_1934578X241258352 for Sandwich-Like Complexation of Carbohydrates by Hydrogen Bonding and CH-π Interactions by Linda Köhler, Stefan Kaiser and Monika Mazik in Natural Product Communications