Selective Carboxylate Recognition Using Urea-Functionalized Unclosed Cryptands: Mild Synthesis and Complexation Studies

Herein we present the synthesis and evaluation of anion-binding properties of 12 new receptors from the unclosed cryptand family. Their core is built on the stable 26-membered tetraamidic macrocyclic scaffold, whereas various alkyl and aryl urea substituents were introduced after a yield-limiting macrocyclization step (65–98%). The receptors strongly bind anions, in particular carboxylates, even in a highly competitive solvent mixture (DMSO-d6 + H2O 95:5 v/v).


Diffractometer and data collection
Single crystal X-ray diffraction measurements were carried out on a Agilent Supernova diffractometer, at 100K with graphite monochromated Cu Ka radiation (1.54184 Å). Colorless crystal in a form of long needle was used.

Structure refirements
The data reduction was made by using CrysAlisPRO 1 software. The structures were solved by direct methods and refined on F2 by full-matrix least-squares by using SHELXS97 and SHELXL97. 2 All non-hydrogen atoms were refined as anisotropic while hydrogen atoms were placed in calculated positions, and refined in riding mode.

Crystallographic data
CCDC contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif

Crystal data and structure refinement details
The crystal suitable for the x-ray measurement was obtained by a slow diffusion of water to the solution of receptor 4l in DMSO-d6. Absorption correction multi-scan Refinement method Full-matrix least-squares on F 2 Restraints / parameters 30 / 544 Goodness-of-fit on F 2 1.068 S3 1.5 Thermal ellipsoid plots (ORTEP) for reported crystal structure with the thermal ellipsoids shown at a 50% probability level Figure S1. Thermal ellipsoid plot (ORTEP) diagram for structure 4l.

Computational studies 2.1 Conformational analysis of 4c-benzoate complex
The 1 H NMR titration experiments performed in DMSO/water solvent mixtures reveal the unusual selectivity of hosts 4c and 4d for benzoate over more basic acetate. In order to rationalize these results and to elucidate the potential role of aryl ring of lariat arm on this process, we decided to perform extensive computational calculations. As a representative, we carried out the conformational search analysis for the 4c-PhCO2complex as implemented in program Spartan'18 Parallel Suite (method conformer distribution). 3 Briefly, 400 starting structures were obtained from the conformer distribution method implemented in Spartan program and using Merck Molecular Force Field (MMFF) for the given torsion angles ( Figure S2).

Torsion Variance
. Torsions and variances of the torsion angles used to generate the conformers.
All these conformers were then geometrically-optimized without any constrains using fast semiempirical PM6 method. After that, energies of conformers were calculated at relatively fast DFT level of theory (B3LYP-D3/6-31(d)/gas-phase). From this set dozens of conformers with energy cutoff of 40 kJ/mol were selected, prior to removing the enantiomers and structurally similar conformers. These remaining ones were then fully geometrically-optimized without any constrains at DFT/B3LYP-D3/6-31(d)/gas-phase level of theory and 17 conformers with the lowest energies were further optimized at DFT/M06-2X/6-31G(d) 4,5 using a C-PCM approximation of the DMSO solvent which was used for the 1 H NMR titrations experiments. Three lowest energy conformers are shown in Figure S3. Detailed analysis of this set suggest that in conformation with lowest energy the phenyl ring of benzoate is not involved in any particularly short - interactions with aromatic residues of the host (the distance between C-H proton of lariat arm and PhC-O moiety is 3.66 Å). However, analysis of the two next conformers (higher in energy by 10 and 16 kJmol -1 , respectively) reveal presence of such interactions, in particular - stacking between benzoate and pyridine moiety of the host (dCC = 3.20 -3.63 Å, Figure S3b), edge-to-face interaction with pyridine moiety and aryl C-H proton from the lariat arm (dC-HN = 3.49 Å, Figure S3c), and C-HO hydrogen bond between carboxylate and CH residue of lariat arm (dOH-C = 3.35 Å, Figure S3b). In addition, the latter type of interaction might explain the unexpected higher affinity of host 4d vs 4c for carboxylates. The host 4d has a para-substituted OMe group whereas host 4c has an unsubstituted phenyl group. For this reason, one could expect that binding affinities should be lower for host 4d possessing an electron-donating group. Analysis of the electrostatic surface potential for the simple analogs of hosts 4c, 4d, and 4l reveals that C-H protons of the lariat arm are in fact more polarized for host 4d than for hosts 4c and 4l ( Figure S4). In addition 4l could not be engaged in edge-to-face interactions with benzoate due to steric factors.

Titration experiments
As the source of anions, commercially available tetrabutylammonium salts were used. HPLC grade water was added to the commercially available DMSO-d6 of 99.9% isotopic purity to obtain the appropriate water concentration. The host solution was titrated in a NMR tube with the solution of the respective TBA salt in receptor aliquots (details are given in Table S3 and S4). The binding constants were calculated from the changes in chemical shifts of ligand protons which were shifted during titration. Nonlinear curve fitting was carried out with HypNMR 2008 7 (Version 4.0.71) program with fitting to the appropriate global binding model (see Table S3 and S4).