Enantioselective Recognition of Neutral Molecules in Water by a Pair of Chiral Biomimetic Macrocyclic Receptors

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Dec 2020Enantioselective Recognition of Neutral Molecules in Water by a Pair of Chiral Biomimetic Macrocyclic Receptors Hongxin Chai†, Zhao Chen†, Sheng-Hua Wang, Mao Quan, Liu-Pan Yang, Hua Ke and Wei Jiang Hongxin Chai† Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, , Zhao Chen† Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543, , Sheng-Hua Wang Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, , Mao Quan Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, , Liu-Pan Yang Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, , Hua Ke Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, and Wei Jiang *Corresponding author: E-mail Address: [email protected] Shenzhen Grubbs Institute, Guangdong Provincial Key Laboratory of Catalysis, and Department of Chemistry, Southern University of Science and Technology, Xueyuan Blvd 1088, Shenzhen, 518055, https://doi.org/10.31635/ccschem.020.202000160 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Enantioselective recognition in water remains an ongoing challenge in supramolecular chemistry but is routine in nature. Herein, we report the first enantiopure pair of biomimetic macrocyclic receptors with hydrogen bonding donors in their deep hydrophobic cavities. The chiral naphthotubes can be efficiently synthesized through a chirality-directed macrocyclization strategy and are able to discriminate the enantiomers of neutral chiral molecules in water. Density functional theory calculations reveal that the "three-point contact" model effectively explains their enantioselectivity. The differential noncovalent interactions inside the hydrophobic cavity are responsible for the enantioselective recognition. Moreover, these chiral naphthotubes are both fluorescent and circular dichroism (CD)-active. In CD spectroscopy, they have been demonstrated to have the ability to detect nonchromophoric, achiral molecules in water. And, the use of fluorescence spectroscopy has aided in the determination of the enantiomeric excess (ee) values of chiral molecules. The results and conclusions obtained with these chiral biomimetic receptors can be used to better understand enantioselective recognition in biological systems. Download figure Download PowerPoint Introduction Enantioselective recognition in water is routine in nature and is the basis of life. Natural receptors are composed of chiral building blocks, resulting in chiral binding pockets that are very selective towards chiral substrates. Therefore, to be effective, most drug molecules must be enantiopure. Research on enantioselective recognition in water by artificial systems not only allows for a better understanding of enantioselective recognition in complex biological systems, but also provides a new means for enantioselective sensing and separation. However, enantioselective recognition of neutral molecules in water is still very challenging to achieve with synthetic receptors.1,2 Although many water-insoluble chiral hosts3–6 and achiral water-soluble hosts7–20 have been reported in the literature, water-soluble and chiral hosts are very rare.21–26 Cyclodextrins,21,22 which have intrinsically chiral cavities, are the most common receptors used for chiral recognition in water. Water-soluble peptidocalix[4]arene,23 cyclotriveratrylene-based cryptophane,24 molecular baskets,25 and other molecular receptors27,28 have also been used for chiral recognition in water. In addition to these molecular receptors, chiral metallosupramolecular cages29–32 and other noncovalent assemblies33,34 have shown to be capable of chiral recognition in water. Most of these receptors/assemblies, however, give rise to relatively low enantioselectivity (KS/KR or KR/KS < 1.3),23,28,35–38 and rarely achieve high enantioselectivity (>1.5).27,33,34,39–41 In the latter cases, molecular recognition involves dynamic covalent bonds27 and either charged chiral molecules33 or rigid chiral molecules39,40,41 used as guests. The binding constants of these molecules are usually smaller than 103 M-1, and their enantioselectivities are often determined by using only one of the enantiomers of the receptor. The possibility of the other enantiomer of the receptor showing the opposite enantioselectivity towards the same pair of guests has rarely been investigated. The contributions of the hydrophobic effect and other noncovalent interactions have not been well characterized. In fact, the key factors required to achieve enantioselective recognition in water are still poorly understood. More importantly, these synthetic receptors do not have a biomimetic cavity and are inadequate models by which enantioselective recognition in nature can be understood. According to the "three-point interaction" model,42–44 specific and directional noncovalent interactions, such as hydrogen bonding, should play an important role in enantioselective recognition. However, these noncovalent interactions are known to be significantly attenuated in the polar environment of water.45–47 Questions thus remain concerning how these interactions can be used effectively and the role of the hydrophobic effect in chiral recognition. A pair of enantiomers has identical shape and functional groups and should in principle have the same solvation and occupy similar spaces in the cavity of a chiral receptor. The hydrophobic effect may thus be the same for the binding of two enantiomers. Answering these questions is important for understanding the enantioselective recognition of bioreceptors. Ideally, an enantiomeric pair of chiral molecular receptors with a biomimetic cavity exhibiting enantioselective recognition would serve as a good model system. We recently reported a pair of water-soluble biomimetic macrocycles with inward-directed hydrogen bonding sites in their hydrophobic cavity, amide naphthotubes 1 and 2 (Scheme 1a).48–56 This cavity is analogous to those found in bioreceptors. The hydrogen bonding sites are shielded from bulk water in the hydrophobic cavity and experience low polarity. Therefore, hydrogen bonding can be efficiently used, even in water. These naphthotubes are capable of binding highly hydrophilic molecules, such as 1,4-dioxane, urethane, epoxides, and polyethylene glycols, in water by combining the hydrophobic effect with hydrogen bonding.48 Naphthotubes 1 and 2 are achiral, and although they can be used as sensors for the determination of the enantiomeric excess (ee) values of chiral epoxides through induced circular dichroism (CD) signals,49 they are unable to perform enantioselective recognition. In order to achieve enantioselective recognition, a chiral environment must be introduced into the cavity. Herein, we report the synthesis of the first enantio-resolved pair of biomimetic macrocyclic receptors with hydrogen bonding sites in their chiral hydrophobic cavities. These chiral naphthotubes (R2,S2- 3 and S2,R2- 3, Scheme 1) share similar structures to that of 1, but with a different arrangement of their carbonyl groups. This slight change in structure renders them chiral. The chiral naphthotubes show decent enantioselectivity toward neutral chiral molecules in water. Density functional theory (DFT) calculations reveal that their enantioselectivity is determined by the differential noncovalent interactions inside the hydrophobic cavity of the two diastereomeric complexes. In addition, these chiral naphthotubes are CD-active so they can be used to detect nonchromophoric, achiral molecules by CD spectroscopy. The naphthotubes also fluoresce and can be used to determine the ee values of chiral molecules by fluorescence spectroscopy. Scheme 1 | Chemical structures of amide naphthotubes (a) 1 and 2, (b) chiral naphthotubes R2,S2-3 and S2,R2-3, and (c) the chiral guests involved in this research. Download figure Download PowerPoint Experimental Methods Experimental Methods are available in the Supporting Information. Results and Discussion Synthesis and characterization of chiral naphthotubes Chiral amide naphthotubes can be constructed simply by rearranging one of the carbonyl groups in the syn configurational isomer 1. As shown in Scheme 1b, a pair of enantiomers, R2,S2- 3 and S2,R2- 3, can be obtained by shifting one of the two carbonyl groups from one bisnaphthalene to the other. Then, each bisnaphthalene cleft contains one carbonyl group and one aminomethyl group. This situation gives rise to two chiral centers on the acetal bridges that connect the two naphthalenes in the bisnaphthalene cleft. Thus, the cavities of R2,S2- 3 and S2,R2- 3 are chiral and are decorated with inwardly directed hydrogen bonding sites. This structure is biomimetic and is likely to be a good receptor for enantioselective recognition of neutral molecules in water. To synthesize chiral naphthotubes 3, it is necessary to introduce dissymmetry into the bisnaphthalene building block. The racemic mixture, rac- 4 (synthesized in four linear steps from 2,6-dihydroxynaphthalene, see the Supporting Information), which has one carboxylic acid and one aldehyde group (Figure 1a), was first synthesized from the corresponding dialdehyde through controlled oxidation (see the Supporting Information). Using a standard procedure reported previously, rac- 4 was further converted into rac- 5, with one protonated aminomethyl group and one carboxylic group (Figure 1a).48 A pair of rac- 5 can then macrocyclize under high-dilution conditions, generating two configurational isomers of the macrocycles, rac- 6 (yield: 15%) and 7 (yield: 29%). No other macrocycles were detected. The syn- (rac- 6) and anti- ( 7) configurational isomers were assigned according to their nuclear Overhauser effect spectroscopy (NOESY) NMR spectra ( Supporting Information Figures S1–S4). The ester side chain has a nuclear Overhauser effect (NOE) cross-peak with aromatic proton 8 (Figure 1a) for the anti-configurational isomer 7 but not for the syn-configurational isomer rac- 6. This assignment was further confirmed from the X-ray single-crystal structure of rac- 6 (Figure 1b). Figure 1 | (a) Synthetic procedures of chiral naphthotubes R2,S2-3 and S2,R2-3 through the chiral resolution of rac-6. (b) X-ray single-crystal structure of rac-6. A pair of enantiomers coexists in the crystal, and S2,R2-6 and R2,S2-6 are colored gray and yellow, respectively. Download figure Download PowerPoint The syn-configurational isomer rac- 6 is chiral with a pair of enantiomers embedded in the crystal structure (Figure 1b). Chiral resolution of rac- 6 could be achieved by passing it through a chiral High Performance Liquid Chromatography (HPLC) column (CHIRALPAK IG (ID00CD-UF004); eluent: MeOH: CH2Cl2 = 20: 80 (V/V)). The enantiopure R2,S2- 6 and S2,R2- 6 were then obtained; however, the achiral anti-configurational isomer 7 was obtained in a much higher yield (29%) than that of rac- 6 (15%) during the macrocyclization step. This is not considered a satisfactory yield for the synthesis of chiral naphthotubes. Careful analysis indicates that the two components of the bisnaphthalene building blocks in the R2,S2- 6 or S2,R2- 6 chiral naphthotubes have the same chirality, but are an enantiomeric pair in the mesomeric anti-configurational isomer 7. Therefore, only enantiopure R2,S2- 6 or S2,R2- 6 will be obtained if enantiopure R,S- 5 or S,R- 5 is used for macrocyclization, and the unwanted achiral anti-configurational isomer 7 can be fully eliminated. This process can be thought of as a chirality-directed macrocyclization strategy.57,58 Rac- 5, which has one protonated aminomethyl group and one carboxylic group, is very reactive and cannot be easily chirally-resolved by chiral HPLC. Compound rac- 4 is more stable and was thus selected and subjected to chiral HPLC for enantioseparation. Using a CHIRALPAK IG (ID00CD-UF004) chiral column and MeOH:CH2Cl2:Et2NH:AcOH (10∶90∶0.1∶0.3) as the eluent, the two enantiomers were successfully separated. Enantiopure R,S- 4 and S,R- 4 have symmetric CD spectra ( Supporting Information Figure S5), and their absolute configurations were confirmed by single-crystal X-ray diffraction (Figure 2) with Cu Kα radiation (λ = 1.54178 Å; Flack parameters = 0.04(2) and 0.01(3) for S,R- 4 and R,S- 4, respectively).59 Enantiopure R,S- 4 or S,R- 4 was further converted into the corresponding R,S- 5 or S,R- 5 in similar yields by following the same procedure used for the racemate. Enantiopure R,S- 5 or S,R- 5 was subjected to macrocyclization under high-dilution conditions, affording only the enantiopure syn-configurational isomers R2,S2- 6 or S2,R2- 6. R2,S2- 6 and S2,R2- 6 have symmetric CD signals (Figure 3). Most importantly, the yields of macrocyclization with the enantiopure building blocks were rather high (>50%), and no anti-configurational isomer 7 was detected. The resulting chiral naphthotubes have nearly identical CD spectra at the same concentrations as those obtained from the chiral resolution of rac- 6. These results show that the strategy of chirality-directed macrocyclization is quite successful in the synthesis of these chiral macrocycles . Figure 2 | Chirality-directed synthesis of R2,S2-3 and S2,R2-3 from the corresponding enantiopure R,S-4 and S,R-4, respectively, and X-ray crystal structures of R,S-4 and S,R-4. Download figure Download PowerPoint The absolute configurations of R2,S2- 6 and S2,R2- 6 were assigned according to the crystal structures of R,S- 4 and S,R- 4, respectively. This assignment is reliable because no racemization is possible during the reactions. Hydrolysis of R2,S2- 6 and S2,R2- 6 results in the formation of water-soluble sodium salts of naphthotubes R2,S2- 3 and S2,R2- 3, respectively. The sodium salts show the same binding ability as the ammonium salts but have much higher solubility in water.60 Enantiopure R2,S2- 3 and S2,R2- 3 also have symmetric CD spectra in water (Figure 3), but their spectra are slightly different from those of R2,S2- 6 and S2,R2- 6 in CH2Cl2. Figure 3 | CD spectra of R2,S2-6, S2,R2-6 in CH2Cl2 (0.10 mM) and R2,S2-3 and S2,R2-3 in H2O (0.10 mM). Download figure Download PowerPoint Comparison between 1 and Rac -3 In order to understand the difference in binding behaviors of 1 and rac 3, the hydrophilic molecule 1,4-dioxane, which is a good guest for 1,48 was selected as a test case. Both 1 (with the new assignment,55,61Ka = 2960 M−1, ΔH° = −17.2 kJ/mol, −TΔS° = −2.6 kJ/mol, T = 298 K) and rac- 3 (Ka = 2820 M−1, ΔH° = −26.1 kJ/mol, −TΔS° = 6.4 kJ/mol, T = 298 K; Supporting Information Figure S6) share rather similar binding constants but have different thermodynamic signatures. They have very similar cavity sizes and slightly different arrangements of hydrogen bonding sites in their cavities. NMR experiments ( Supporting Information Figures S7–S8) performed by titrating 1,4-dioxane into a solution of 1 or rac- 3 in H2O/D2O (9∶1) show that the hydrogen bonding of 1,4-dioxane in 1 and rac- 3 is different. These differences in cavity size and hydrogen bonding behavior obviously do not affect their binding affinity but change their thermodynamic signature, causing different entropy–enthalpy compensations.62,63A similar phenomenon has been observed in bioreceptors,64,65 but has rarely been studied in synthetic systems.66 Enantioselective recognition of neutral molecules Seven enantiomeric pairs of neutral molecules 8 – 14 (Scheme 1c) were selected to further study the enantioselective recognition ability of R2,S2- 3 and S2,R2- 3. The binding constants and thermodynamic parameters were determined by isothermal titration calorimetry (ITC) analyses ( Supporting Information Figures S9–S22). All the titration experiments were repeated three times, and the binding data are listed in Tables 1 and S1. In general, the binding constants are good , having values higher than 103 M−1, except for that of 8. Epoxide 10 is the best guest, with a Ka of up to 105 M−1. The molecular recognition of epoxides 8, 9, and 10 has been studied with naphthotube 1 (with the new assignment;55,61 for 8, 9, and 10, Ka = 621 M−1 (NMR), 1.28×104 M−1 (ITC), and 1.00×105 M−1 (ITC), respectively).49 The binding constants of R2,S2- 3 and S2,R2- 3 to epoxides 8, 9, and 10 are on the same order of magnitude as those of naphthotube 1. Guests 11– 14 contain more hydrophobic groups and one more heteroatom, which can act as a hydrogen bonding acceptor, than epoxides 8 and 9. It is expected that guests 11– 14 are more complementary to the cavity of R2,S2- 3 and S2,R2- 3 in terms of hydrogen bonding and hydrophobic effects. Surprisingly, guests 11– 14 generally show weaker binding affinities than those of epoxides 9 and 10. This finding may be attributed to the steric hindrance caused by the incongruent fit of the structures and the higher desolvation penalty resulting from the presence of additional heteroatoms. Table 1 | Thermodynamic Parameters of the Enantioselective Recognition of R2,S2-3 and S2,R2-3, as Determined by ITC Titrations in H2O at 25 °C.[a] Guest S2,R2- 3 R2,S2- 3 ΔG° ΔH° −TΔS° Ka KS/KR[b] ΔG° ΔH° −TΔS° Ka KR/KS[b] (kJ mol−1) (×103 M−1) (kJ mol−1) (×103 M−1) R- 8 −15.0 −18.8 3.8 0.42 ± 0.01 1.02 −15.0 −17.5 2.5 0.42 ± 0.01 1.05 S- 8 −15.0 −16.6 1.6 0.43 ± 0.01 −14.8 −20.6 5.8 0.40 ± 0 R- 9 −22.8 −18.1 −4.7 9.8 ± 0.4 1.48 −23.8 −24.5 0.7 14.7 ± 0.2 1.35 S- 9 −23.7 −24.9 1.1 14.5 ± 0.9 −22.9 −18.2 −4.7 10.9 ± 0.2 R,R- 10 −26.9 −23.1 −3.8 51.9 ± 2.4 1.97 −28.3 −26.0 −2.4 93.2 ± 1.6 1.95 S,S- 10 −28.6 −27.7 −0.9 102.4 ± 6.3 −26.7 −22.4 −4.3 47.7 ± 2.0 R,R- 11 −18.8 −16.2 −2.6 2.0 ± 0.1 1.80 −20.3 −14.7 −5.3 3.6 ± 0.1 1.80 S,S- 11 −20.3 −18.3 −2.0 3.6 ± 0.1 −18.9 −14.4 −4.5 2.0 ± 0.1 R- 12 −20.1 −17.9 −2.3 3.3 ± 0.1 1.48 −20.8 −20.8 -0.2 4.5 ± 0.1 1.50 S- 12 −20.9 −24.3 3.4 4.9 ± 0.5 −19.9 −22.5 2.6 3.0 ± 0.1 R- 13 −20.0 −24.9 2.9 7.2 ± 0.1 1.08 −22.9 −13.5 −9.4 10.2 ± 1.1 1.21 S- 13 −20.2 −23.9 1.7 7.8 ± 0.6 −22.4 −14.1 −8.3 8.4 ± 0.2 R- 14 −19.3 −17.5 −1.9 2.5 ± 0.3 2.04 −20.9 −23.1 2.1 4.7 ± 0.1 2.04 S- 14 −21.2 −17.2 −4.0 5.1 ± 0.4 −19.2 −15.8 −3.4 2.3 ± 0.1 [a]All of the experiments were repeated three times; [b]enantioselectivity, KS/KR or KR/KS, was calculated by comparing the binding affinities of the same host to the different enantiomers of the guests. Enantioselective recognition was indeed observed for all these guest pairs (Table 1). In general, R2,S2- 3 shows a preference towards R-configured guests, while S2,R2- 3 binds better to S-configured guests. The enantioselectivity (KS/KR) of S2,R2- 3 is almost identical to that (KR/KS) of R2,S2- 3, indicating that the binding data and enantioselectivity results are self-consistent and thus quite reliable. This enantioselective recognition is further supported by the results of the NMR experiments ( Supporting Information Figures S23–S26). The largest enantioselectivity (∼2.0) was achieved for guests 10 and 14, which is good enantioselectivity in water for such small and neutral organic molecules. Differential thermodynamic parameters show that the enantioselectivity is dominated by the favorable enthalpic contribution, but the entropic contribution is nevertheless unfavorable (Figure 4). A similar entropy–enthalpy compensation was observed in the enantioselective recognition of cyclodextrins.36 The dominant enthalpic contribution to the enantioselectivity suggests that the differential noncovalent interactions between two diastereomeric complexes with the same chiral host but different guests are responsible for the enantioselectivity, which will be discussed based on computational studies (see further). Figure 4 | Plot of the differential entropy changes against the differential enthalpy changes for the binding of the enantiomeric pairs of chiral guests with chiral naphthotubes R2,S2-3 or S2,R2-3. Download figure Download PowerPoint Influence of solvents on enantioselective recognition The binding parameters and enantioselectivity discussed thus far were all obtained in H2O. The enantioselective recognition may be further understood by performing experiments in different solvents.67 However, the binding of the systems discussed above are highly dependent on water and hydrophobic effects. Therefore, the solvent effect on enantioselective recognition was studied by adding 1% MeOH to the water solution or by substituting H2O with D2O. The addition of only 1% MeOH results in a four- to five fold decrease in the binding constants between 3 and 10 compared with those in pure H2O (Table 2 and Supporting Information Table S2, Figures S27–S28). This decrease is likely caused by the attenuated hydrophobic effect in the presence of MeOH. Although the polarity of H2O is reduced after adding MeOH (ɛMeOH = 33 and ɛwater = 78), the differential change in the binding free energy from H2O to 1% should be the same for the two diastereomeric complexes R,R- S2,R2- 3 and S,S- S2,R2- 3. the enantioselectivity may not be by the systems from H2O to 1% Surprisingly, the enantioselectivity (KR/KS) from 2.0 in H2O to 2.4 in 1% (Table 2 and Supporting Information Figures which could be caused by a slightly different of the solvents the and by MeOH introduced into the solvation This is supported by the enthalpic contribution to enantioselectivity after adding MeOH S2,R2- 3, changes from to for S2,R2- 3, changes from 3.6 to Table 2 | Thermodynamic Parameters of the Enantioselective Recognition of R2,S2-3 or S2,R2-3 to Guests or as Determined by ITC Titrations in 1% or at 25 °C.[a] Guest S2,R2- 3 R2,S2- 3 ΔG° ΔH° −TΔS° M−1) KS/KR[b] ΔG° ΔH° −TΔS° M−1) KR/KS[b] (kJ mol−1) (kJ mol−1) H2O R,R- 10 −26.9 −23.1 −3.8 ± 1.97 −28.3 −26.0 −2.4 ± 1.95 S,S- 10 −28.6 −27.7 −0.9 10.2 ± 0.6 −26.7 −22.4 −4.3 ± 1% MeOH R,R- 10 −23.8 −26.9 3.0 ± 0.1 0.6 ± 0.3 S,S- 10 −26.0 ± 0.1 −23.9 0.3 ± 0.1 R,R- 10 0.1 ± 0.2 ± 0.6 S,S- 10 ± 0.1 ± 0.1 [a]All of the experiments were repeated three times; enantioselectivity was calculated by comparing the binding affinities of the same host to the different enantiomers of the guests. substituting H2O with the binding constants as and a solvent effect = was between 3 and 10 ( Supporting Information Figures and Table The binding is more in than in H2O = −4.7 to bonding in is known to be than that in This solvent effect may be caused by the hydrogen between the host and guest because the amide are to in This effect to a slightly enantioselectivity in compared to that in H2O the solvent on the enantioselectivity are not very that the hydrophobic effect may not be the in enantioselective recognition. of enantioselective recognition In order to understand the the enantioselective recognition, we first must have a good of the complexes to No binding can be in for the naphthotube 6 ( Supporting Information Figures indicating that water and thus hydrophobic are very important for the recognition of these chiral NMR of the hosts by guests were performed in (9∶1) to show the contribution of hydrogen the amide are not fully with in this they remain in the NMR addition of guest 8, the of rac- 3 are and ( Supporting Information Figure that hydrogen bonding in the hydrophobic cavity between the host and the from the that the relatively more hydrophobic guest (Ka = M−1, Supporting Information Figure is a weaker to rac- 3 than 9, because it contains no hydrogen bonding Therefore, hydrophobic achieved through the cavity should be the the binding of these chiral guests. In addition, other noncovalent interactions between the host and guest, for hydrogen bonding and also likely even in because they are shielded inside the hydrophobic cavity. This situation is analogous to that in naphthotubes 1 and and is supported by the dominant enthalpic contribution to the binding free However, enantiomeric pairs have the same size and shape and should occupy the same in the cavity of the chiral differential hydrophobic may not to the enantioselectivity to a This analysis is supported by the solvent as discussed differential specific noncovalent interactions, such as hydrogen bonding and interactions, may be the of the enantioselectivity. According to the "three-point interaction" model for chiral should be three or interactions between chiral guests and chiral hosts to achieve enantioselective recognition. calculations for the complexes of R,R- 3 and S,S- 3 at the of theory were performed to into the interactions responsible for the high enantioselectivity. As shown in Figure the guest enantiomeric pair R,R- 10 and S,S- 10 different arrangements in the cavity of R2,S2- 3. model analysis based on the the noncovalent and their between the host and the guest, the "three-point Figure 5 | (a) structures and models of and obtained by calculations with the solution model in H2O. analysis shows the noncovalent interactions between the guests and the which can be further into the interactions with three of the (b) the (c) the group, and the group. the hydrogen Download figure Download PowerPoint The noncovalent interactions between the host and the guest can be into interactions with three of the the the group, and the group. These interactions to be compared between the two diastereomeric complexes to reveal the of the enantioselectivity. the interactions between the and the naphthotubes (Figure the hydrogen in R,R- 3 has a slightly but a than those in S,S- 3. interactions between the on the chiral centers of the guests and the naphthalenes of the hosts for both complexes. The parameters of these interactions are similar for the two complexes. to the group (Figure interactions only in the more stable complex R,R- 3. the group, the of R2,S2- 3 has