Slovenian NMR Centre, National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia EN→FIST Centre of Excellence, Dunajska 156, SI-1000 Ljubljana, Slovenia Chemistry, University of Southampton, Southampton SO17 1BJ, United Kingdom Faculty of Chemistry and Chemical Technology, University of Ljubljana, SI-1000 Ljubljana, Slovenia This article is part of the Thematic Series "Supramolecular chemistry II". anion recognition conformation analysis host–guest systems NMR spectroscopy The conformational properties of 1,3-diindolylureas and thioureas were studied by a combination of heteronuclear NMR spectroscopy and quantum mechanics calculations. NOE experiments showed that the anti–anti conformer along the C7–N7α bonds was predominant in DMSO- d 6 solution in the absence of anions. Anion-induced changes in the 1 H and 15 N chemical shifts confirm the weak binding of chloride anions with negligible conformational changes. Strong deshielding of ureido protons and moderate deshielding of indole NH was observed upon the addition of acetate, benzoate, bicarbonate and dihydrogen phosphate, which indicated that the predominant hydrogen bond interactions occurred at the urea donor groups. Binding of oxo-anions caused conformational changes along the C7–N7α bonds and the syn – syn conformer was preferred for anion–receptor complexes. The conformational changes upon anion binding are in good agreement with energetic preferences established by ab initio calculations.

In the last two decades, remarkable efforts have been made in the field of the development of synthetic anion receptors, motivated by prospective applications in recognition, separation, guest inclusion and catalysis . The fundamental role of anions in biological and chemical processes drives much of this research. Biomolecules such as the sulfate binding protein and phosphate binding protein employ hydrogen bonds as the key driving force to bind or transport anions through cell membranes. Hydrogen bonding interactions are extensively employed in synthetic anion receptors comprising groups such as amides, pyrroles, indoles, ureas and triazoles, as well as in ammonium, guanidinium and imidazolium moieties used as hydrogen bond donors . Amongst neutral anion receptor systems, indole and related heterocycles, such as carbazole, 2,2'-biindole and indolo[1,2- b ]carbazoles, have recently attracted significant attention . Indole contains a single hydrogen bond donor group, which is employed in biological systems to bind anions such as chloride and sulfate . We have recently analyzed the conformational preferences of several 2,7-disubstituted indoles with amide substituents at C2 and urea substituents at C7, which showed preference for distinct conformations in the presence and in the absence of anions . In addition, indole and urea groups were strongly involved in hydrogen-bonding interactions with the bound anionic guest, whilst the amide group interacted only weakly with the bound anion. These observations led to the design of diindolylureas and diindolylthioureas . These compounds have remarkably high affinities for oxo-anions such as phosphate and sulfate for neutral receptors in DMSO- d 6 /0.5% water and have been shown to perturb the p K a of bound guests ( ) . X-ray crystal structures of a variety of complexes with anions revealed the adoption of the syn–syn conformation in the solid state upon anion complexation. With the urea analogues, such as 1 , this is accompanied by higher order complex formation with oxo-anions in the solid state. For example, with dihydrogen phosphate, three equivalents of receptor complex to a single anion, which has doubly deprotonated, resulting in the formation of a complex in which twelve hydrogen bonds stabilize the PO 4 3− anion. In solution, the thiourea analogues such as compound 2 show significantly lower affinities for oxo-anions than do the urea analogues. We had previously proposed that this may be due to the larger size of the sulfur atom resulting in a lower propensity of these systems to adopt a planar conformation. Whilst the conformational properties of these compounds have been explored by single crystal X-ray diffraction in the solid-state, a detailed analysis of the conformational properties of the these compounds in solution, in the absence and presence of oxo-anions, has yet to be performed and may help shed light on the high affinity of these systems for oxo-anions. Therefore, in the current work the conformational preorganization of bis-indole receptors 1 – 4 ( ), as well as the conformational changes of these systems upon binding of chloride and several oxo-anions, were studied by NMR spectroscopy and supported by energetic preferences established from ab initio calculations.

Compounds 1 – 3 were synthesized following a previously reported methodology . Compound 4 was prepared by reaction of 7-amino- N -phenyl-1 H -indole-2-carboxamide (0.27 g, 1.07 mM) with 7-isothiocyanato- N -phenyl-1 H -indole-2-carboxamide (0.31 g, 1.07 mM) in pyridine in 27% yield (see for details).

The conformational properties of diindolylureas and diindolylthioureas 1 – 4 , shown in , were evaluated by means of NMR spectroscopy. Proton and 13 C NMR resonances were assigned through 1D and 2D spectra, while 15 N chemical shifts were determined by 1 H– 15 N correlations in HSQC and HMBC spectra. Notable 1 H and 15 N NMR chemical shifts of 1 – 4 are shown in . It should be noted that only one set of signals was observed for both indole rings in all four receptors, due to the symmetry of the compounds. The full NMR data set together with 13 C NMR chemical shifts is reported in . Indole NH proton resonances were found between 10.8 and 11.7 ppm. Thioureido containing compounds 2 and 4 exhibited slight downfield shifts of H1 and H7α with respect to ureido receptors 1 and 3 ( ). N1 chemical shifts showed only minor variations as a result of structural differences in 1 – 4 . The most significant differences in chemical shifts between the ureido and thioureido functionalized receptors were observed for H7α and N7α atoms (Δδ H = 0.8 and Δδ N = 22 ppm, ). Compounds 3 and 4 contain phenylamide substituents at C2 and hence two more NH groups ( ). Considerable deshielding of the H3 and H6 resonances was observed in 3 and 4 with respect to the nonsubstituted indole moieties in 1 and 2 . The downfield shift of δ H3 was attributed directly to the presence of the phenylamide group at C2. Deshielding of H6 (Δδ 0.4–0.5 ppm) in 3 and 4 with respect to 1 and 2 , respectively, was much larger than the deshielding of H4 (Δδ 0.1 ppm), possibly due to a more efficient conjugation.

shows 1 H chemical shift changes of 1 in the presence of one equivalent of chloride, acetate, benzoate, bicarbonate and dihydrogen phosphate anions. The protons that are most affected by anion–receptor interaction were found to be H1, H6 and H7α. Only minor Δδ H7α and negligible differences of δ H1 and δ H6 were observed in the presence of chloride anions ( and ). The very weak interactions between chloride and 1 could be due to competing interactions of the polar DMSO molecules for the hydrogen bond donor groups of the receptor, and the weak basicity of the chloride. This is supported by the stability constant determinations previously reported, and presented in . Considerable downfield shifts of δ H7α were observed upon addition of acetate, benzoate and bicarbonate anions (Δδ ≈ 2 ppm, Figures 2c–2e), which suggested strong interaction of ureido NH protons with these anions. In addition, the strong deshielding of indolyl H1 protons corroborates its participation in the hydrogen bond formation with carboxylate and bicarbonate moieties (Δδ ≈ 1 ppm). Planar oxo-anions interact both with H1 and H7α due to their Y-shaped geometry. The tetrahedral geometry of the dihydrogen phosphate anion allows strong interaction with all four hydrogen bond donor groups, which is reflected in the substantial deshielding of the H1 and H7α protons ( ). Interestingly, one set of signals was observed for each type of anion on the NMR timescale, which suggested that the symmetry of the receptor 1 is preserved upon interactions with anions. The stability constant determinations presented in also support the finding that this compound interacts selectively with oxo-anions. Anion–receptor interactions assessed by 1 H chemical shift changes were corroborated by 15 N NMR data. Weak shielding of N1 in 1 was observed upon addition of acetate, benzoate and bicarbonate anions, whereas addition of dihydrogen phosphate anions led to deshielding of N1 ( ). In contrast, N7α was deshielded upon addition of anions ( ). The strongest deshielding of 11.2 ppm was observed for the 1 ·BzO − complex.

The rotational flexibility of the ureido moiety allows numerous conformations of receptor 1 . Among them three major, energetically preferred, conformers are likely to be observed ( ). The syn–syn conformer, where NH protons form a convergent hydrogen-bonding array, is expected to be adopted in the presence of bound anionic guests, based on the previous solid-state studies. On the other hand, this conformer is unlikely to be abundant in the absence of anions, due to repulsion between the hydrogen bond donor groups. The other two rotamers, namely anti–anti and syn–anti , can be stabilized by intramolecular NH–CO hydrogen bonds, which represent competition to anion–receptor interactions and therefore make conformational studies even more appealing. The conformational characteristics of diindolylurea 1 were assessed by the use of 1D difference NOE experiments in the absence and in the presence of anions. The orientation along the C7–N7α bonds was established by comparative evaluation of NOE enhancements of H7α with H1 and H6 protons. The saturation of H7α in 1 gave strong NOE at H6 (10.4%) and moderate NOE at H1 (4.3%), which suggested that the anti orientation prevails along the linkage between ureido moiety and indole ring ( ). As the observed NOE enhancements are primarily a function of the H6–H7α and H1–H7α distances, we compared their values in the optimized anti – anti (d(H6–H7α) = 2.28 Å) and syn–syn (d(H1–H7α) = 2.32 Å) structures and established a minor difference of 0.04 Å which would be reflected in a 1% change in the NOE enhancements. The observed difference between NOE enhancements in the uncomplexed form of receptor 1 was over 6%, which led us to conclude that the anti – anti conformer is predominant in the DMSO- d 6 solution. In addition, the anti–anti conformer of 1 with its plane of symmetry along the carbonyl bond is in agreement with the single set of resonances in the NMR spectra. On the other hand, the syn – anti rotamer shows a twofold rotational symmetry and is expected to exhibit distinct shielding of nuclei, imposed by the orientations of H6 and H1 protons in the two indole rings with respect to the carbonyl group. However, the populations of the two distinct conformational families are averaged on the NMR time scale. Only minor changes in the NOE enhancements were observed upon addition of chloride anions to 1 . The strongest NOE was observed between H7α and H6 (9.9%), which was of comparable magnitude to the NOE between the same protons in the absence of anions ( ). On the other hand, stronger NOE enhancement between H7α and H1 (7.0%) suggested predominance of the syn – anti rotamer of 1 in the presence of chloride anions in DMSO- d 6 solution. Interestingly, the syn – syn rotamer was observed in the crystal structure, where conformational preferences are dictated by other forces, such as crystal packing. These observations are in agreement with minute 1 H chemical shift changes and the weak stability constant of 1 for the binding of chloride anions. Considerable changes in the NOE enhancements were observed upon addition of acetate anions to 1 . The saturation of H7α resulted in a much stronger NOE at H1 (12.0%) with respect to H6 (4.8%), which suggested that addition of acetate anions led to conformational changes along the C7–N7α bond ( ). The syn – syn conformer is preferred for the 1 ·AcO − complex in solution ( ). In a similar manner, significant changes in the NOEs were observed upon addition of benzoate anions. The saturation of H7α gave much stronger NOE at H1 (10.4%) with respect to H6 (4.8%, ). Broad NH proton signals prevented the study of the conformation of the 1 ·HCO 3 − complex through NOE experiments. NOE enhancements between H1 and H7α (4.2–5.3%) were considerably stronger with respect to NOE between H7α and H6 (0–2.8%) upon addition of dihydrogen phosphate to 1 . The observed NOE enhancements for 1 ·H 2 PO 4 − complex suggest a preference for the syn–syn conformer in DMSO- d 6 .

The observations on the conformational equilibria in the absence and in the presence of anions were corroborated by quantum mechanical calculations at the B3LYP/6-311+G(d,p) level of theory by means of the Gaussian 03 and Gaussian 09 programs. Indole rings represent the rigid part of the anion receptors, while the substituents on the ureido moiety in 1 exhibit conformational freedom. [C6–C7–N7α–C7β] torsion angles were defined to follow energetic changes induced by reorientation of the indolyl moieties along the C7–N7α bonds. The energy surface for the [C6–C7–N7α–C7β] torsion angles, with 30° resolution, shows that the conformer with the lowest energy is in the anti – anti region ( ). Energy minimization of the anti–anti conformer of 1 was performed without any constraints, whereas syn orientations in the syn–anti as well as the syn–syn cases were restrained along the [C6–C7–N7α–C7β] torsion angle, while other degrees of freedom were freely optimized. The relative energies for the three representative conformers are reported in . The anti–anti conformer of 1 was found to be the lowest in energy, while the syn–syn conformer showed considerably higher energy (11.6 kcal·mol −1 ). The angle between the two indolyl rings in the freely optimized anti – anti conformer was 53.9° ( ). The relative energies of the three representative conformers were also computed for complexes of 1 with chloride, acetate and bicarbonate anions. The syn–syn conformer exhibited the lowest relative energy for all three anion–receptor complexes ( ). The anti–anti conformers of anion–receptor complexes exhibited considerably higher energies between 6.2 and 7.3 kcal·mol −1 . The angle between the two indolyl rings in the freely optimized syn – syn conformer of the 1 ·AcO − complex was found to be 21.6° ( ). The optimized structure, shown in , shows a single acetate anion bound to the four NH groups in the receptor 1 with N···O distances in the range from 2.75 to 2.95 Å and N–H···O angles in the range 170–176°. In order to evaluate the role of DMSO on the energetic preferences of rotamers, relative energies were calculated with the use of Tomasi's polarized continuum model . Preferences amongst the three rotamers were retained ( ). Only small differences below 1.2 kcal·mol −1 were found between the three distinct conformers in the case of the 1 ·Cl − complex. In particular, the negligible energy differences between syn–syn and syn–anti rotamers are in agreement with the NOE data that suggested predominance of the syn–anti conformer upon addition of chloride. The energetic preference of the syn–syn over the anti–anti conformer of 3.8 and 3.0 kcal·mol −1 was observed for the 1 ·AcO − and the 1 ·HCO 3 − complexes, respectively ( ). The energetic preferences of the 1 ·AcO – complex are in excellent agreement with the NOE experiments, which showed conformational reorganizations of 1 upon addition of acetate anions.

The choice of thio ( 2 and 4 ) versus oxo ( 1 and 3 ) ureido functionalities, as well as the C2 functionalization of the indole scaffolds with pendant phenyl amides in 3 and 4 , allows tuning of the binding affinities of the receptors. Negligible Δδ values were observed for 2 upon interaction with chloride anions ( ), which suggests a very weak interaction between chloride and 2 , similar to the weak interactions observed between chloride and 1 . Chemical shift changes showed that the main interaction between receptor 2 and trigonal planar anions (acetate, benzoate and bicarbonate) occurred at the H7α protons ( ). Addition of dihydrogen phosphate anions caused considerable deshielding of the H1 and H7α protons. Comparison of the Δδ values for 1 and 2 upon interaction with the anions showed that the urea derivative 1 exhibited a higher preference for anion binding relative to thiourea 2 (the data were supported by the stability constant determinations performed previously and shown in ). The larger sulfur atom can prevent the receptor 2 from adopting a planar conformation, which may reduce the affinity of this receptor for anionic guests. Conformational studies of 2 with the use of NOE enhancements showed that the anti – anti conformer is the preferred conformation in the absence of anions. No apparent conformational changes were observed upon addition of chloride anions to 2 . The overlap of the proton signals as well as the broad line-width of the H1 and H7α NMR resonances prevented conformational analysis of 2 upon addition of other anions used in the study. The conformational preferences of 2 were evaluated by quantum mechanical calculations at the B3LYP/6-311+G(d,p) level of theory. The freely optimized anti–anti conformer of 2 exhibited the lowest energy and the syn–syn conformer was 8.0 kcal·mol −1 higher in energy, in vacuo. Interestingly, the two indolyl rings were not coplanar, with the angle between the two indolyl planes found to be 98.9° ( ). In the case of 2 ·AcO − complex the syn–syn conformer exhibited the lowest energy, while the anti–anti conformer was 7.4 kcal·mol −1 higher in energy, in vacuo. The optimized structure of 2 ·AcO − complex is shown in , where the two acetate oxygen atoms are hydrogen bonded to the four NH groups, with N···O distances in the range from 2.76 to 2.94 Å and N–H···O angles in the range 168–177°. The angle between the indolyl rings in the freely optimized syn – syn conformer of the 2 ·AcO − complex was 68.0°. Bis-amide functionalized diindolylurea 3 exhibits two extra NH groups, which introduces additional possibilities for interactions with anions. The addition of chloride anions to 3 induced negligible chemical shifts, suggesting only weak interactions with this anion ( ). The strong deshielding of H7α and moderate deshielding of H1 protons in the 3 ·AcO − and 3 ·BzO − complexes suggests a significant interaction between the anions and the ureido functionality. Interestingly, negligible deshielding of H2β in 3 was observed and this suggests that the amide protons do not participate in the interaction with acetate and benzoate ( ). The observed Δδ values support the idea that carboxylates were strongly bound to the urea H7α protons which prevented interaction between the anions and the amide H2β protons. Analogously, large chemical shift changes of up to 1 ppm were observed for the H7α and H1 protons upon the addition of bicarbonate anions to 3 . Strong deshielding of H1, H2β and H7α protons in the 3 ·H 2 PO 4 − complex suggests that all the NH donor groups are involved in interactions with the dihydrogen phosphate anions ( ). The conformational properties of 3 and of its complexes with different anions were studied by NOE measurements. The saturation of the H1 protons resulted in moderately negative NOEs at the H7α and H2β protons. The cross peaks in the 2D NOESY spectra between the NH protons and bulk water suggest chemical exchange that complicated the conformational study along the C2–C2α and C7–N7α bonds. Nevertheless, strong NOE enhancements between the H2β and H3 protons suggest an orientation along the C2–C2α bond where the H2β and H3 protons are spatially close and the C2α carbonyl group is oriented towards the indole H1 proton. NOE enhancements between H2β and H3 protons were observed also in the 3 ·AcO − and 3 ·BzO − complexes, which suggests that the orientation of the carboxamide group along the C2–C2α bond is retained in 3 upon addition of carboxylate anions. This observation was supported by negligible Δδ values for the H2β protons in the 3 ·AcO − and 3 ·BzO − complexes with respect to 3 . The conformational preferences and the proposed binding mode in the 3 ·AcO − complex are shown in . A conformational study of 3 in the presence of bicarbonate and dihydrogen phosphate anions was hindered by the broadened and overlapped 1 H signals. In the solid state compound 3 crystallized with tetrabutylammonium dihydrogen phosphate as the monohydrogen phosphate complex . Only negligible chemical shifts were observed for 4 upon addition of chloride anions ( ). Considerable deshielding of H7α protons in 4 by up to 2.3 ppm in the 4 ·AcO − and 4 ·BzO − complexes suggested that the major interactions between carboxylates and receptor 4 occurred at the ureido functionality ( ). The conformational properties of the 4 ·AcO – and 4 ·BzO – complexes could not be determined due to the broad and overlapped proton signals. Unfortunately, excessively broad and overlapped 1 H signals for the 4 ·HCO 3 − and 4 ·H 2 PO 4 − complexes prevented unambiguous assignment of the NMR resonances and hence the conformational studies of these complexes. In the solid state compound 4 crystallized with tetrabutylammonium dihydrogen phosphate as the monohydrogen phosphate complex (see for more details).

The bis-indole receptors 1 – 4 were characterized by heteronuclear NMR spectroscopy. NOE based conformational analysis was supported by quantum mechanics calculations and revealed that diindolylurea 1 exhibited conformational preorganization in DMSO- d 6 solution. The anti – anti conformer, which could be stabilized by intramolecular hydrogen bonds between the C7β carbonyl group and indole NH proton, was predominant for 1 in the absence of anions. The energetically minimized structure of anti – anti conformer showed a plane of symmetry along the ureido carbonyl group and deviation from coplanarity amongst the indolyl rings. Anion-induced chemical shift changes suggested weak binding of chloride anions and negligible conformational changes for 1 . Addition of acetate, benzoate, bicarbonate and dihydrogen phosphate resulted in strong deshielding of the ureido protons and moderate deshielding of the indole H1, which indicated that the main hydrogen bond interaction occurred at the urea donor groups of 1 . Furthermore, binding of anions caused conformational changes along the C7–N7α bonds, and the syn–syn conformer was predominant in the anion–receptor complexes according to both NOE enhancements and ab initio calculations in solution. The freely optimized syn – syn conformer of the 1 ·AcO − complex retained a plane of symmetry along the carbonyl bond and showed a smaller deviation from indole ring coplanarity than did the anti – anti conformer of 1 . The conformational preferences for 2 were analogous to those observed for receptor 1 . Unfortunately, excessively broad and overlapped 1 H signals prevented a detailed conformational analysis of the anion–receptor complexes for 3 and 4 .

1 H, 13 C and 15 N NMR spectra were acquired on a Varian Unity Inova 300 MHz NMR spectrometer. All data were recorded in DMSO- d 6 at 298 K. Chemical shifts were referenced to the residual solvent signal of DMSO- d 6 at δ 2.50 ppm for 1 H (297.801 MHz) and δ 39.50 ppm for 13 C (76.190 MHz), while 15 N (30.188 MHz) chemical shifts were referenced relative to external benzamide (δ 103.55 ppm). Individual resonances were assigned on the basis of their chemical shifts, signal intensities, multiplicity of resonances, H–H coupling constants as well as by means of a series of 2D NMR experiments (COSY, gHSQC and gHMBC). The saturation delay in the 1D difference NOE experiment was 5.0 s. All anions were added as tetrabutylammonium salts except bicarbonate which was added as a tetraethylammonium salt. NOESY spectra were acquired with mixing time of 100 and 300 ms.

Initial structures were generated by Chem3D Pro 10.0 software and energy minimization at the B3LYP/6-311+G(d,p) level was performed for 1 and 2 without any constraints for the anti – anti conformers, by means of Gaussian 03 and Gaussian 09 . Syn orientations in the syn–anti as well as the syn–syn conformers of 1 and 2 were restrained along the [C6–C7–N7α–C7β] torsion angle while other degrees of freedom were freely optimized. Ab initio calculations of anion–receptor complexes were carried out without any constraints for the syn – syn conformers, where anions were placed initially at the expected equilibrium distance to the H1 and H7α protons. The positions of the anions were freely optimized. Energy minimizations of the syn–anti and anti–anti conformers of the anion–receptor complexes were restrained along the [C6–C7–N7α–C7β] torsion angle while other degrees of freedom were freely optimized. The tetrabutylammonium countercation was omitted in the geometry optimization of the anion–receptor complexes. Frequency calculations verified that the optimized geometries were stable points on the potential energy surface. Relative energies in solution were calculated by means of Tomasi's polarized continuum model, where the dielectric constant of DMSO was used (ε = 46.7).

Anion receptors 1 – 4 together with their atomic numbering scheme.

1 H NMR spectra of 1 in the absence of anions (a) and upon addition of one equivalent of the following anions: Chloride (b), acetate (c), benzoate (d), bicarbonate (e) and dihydrogen phosphate (f). All spectra were recorded in DMSO- d 6 at 298 K. (g) 1 H and 15 N NMR chemical shift changes, Δδ = δ (in the presence of anion) – δ (in the absence of anions), induced by addition of one equivalent of different anions to receptor 1 .

Three representative conformational families of rotamers of 1 . Notations refer to the orientations along [C6–C7–N7α–C7β] fragments.

NOE enhancements of 1 in the absence of anions (a) and upon addition of one equivalent of acetate anions (b).

Surface plot of the relative potential energy of 1 as a function of the two constitutive [C6–C7–N7α–C7β] torsion angles. Individual geometries were optimized at the B3LYP/6-311+G(d,p) level of theory at 30° resolution.

Freely optimized structure at the B3LYP/6-311+G(d,p) level of theory and side view showing deviation from coplanarity defined by the angle between the indolyl rings of receptor 1 in the absence of anions (a) and for the 1 ·AcO − complex (b).

1 H NMR chemical shift changes, Δδ = δ (in the presence of anions) – δ (in the absence of anions), induced by addition of one equivalent of different anions to receptors 2 (a), 3 (b) and 4 (c). Note, there is no H2β proton in 2 .

Freely optimized structures at the B3LYP/6-311+G(d,p) level of theory and side view showing deviation from coplanarity defined by the angle between the indolyl rings of 2 (a) and 2 ·AcO − complex (b).

Conformational preferences and proposed binding mode for the 3 ·AcO − 1:1 complex.

Anion a Compound 1 in DMSO- d 6 /0.5% water Compound 1 in DMSO- d 6 /10% water Compound 1 in DMSO- d 6 /25% water Compound 2 in DMSO- d 6 /0.5% water Cl − 128 17 – 74 CH 3 CO 2 − >10 4 774 20 1620 C 6 H 5 CO 2 − >10 4 521 precipitate 477 H 2 PO 4 − >10 4 5170 160 1630 a Anions added as tetrabutylammonium salts.

Stability constants of compound 1 measured in DMSO- d 6 /0.5% water, DMSO- d 6 /10% water and DMSO- d 6 /25% water and compound 2 in DMSO- d 6 /0.5% water at 298 K by 1 H NMR titration techniques .

H1 H2β H7α H2 H3 H6 N1 N2β N7α 1 10.78 – 8.64 7.35 6.44 7.08 136.5 – 102.7 2 11.03 – 9.48 7.36 6.46 7.03 136.3 – 124.9 3 11.62 10.29 8.97 – 7.49 7.59 134.5 128.6 104.6 4 11.68 10.26 9.72 – 7.48 7.39 134.9 129.0 126.6 a In DMSO- d 6 at 298 K.

Selected 1 H and 15 N NMR chemical shifts for 1 – 4 (in ppm). a

Saturated: H1 H6 H7α Enhanced: H7α H7α H1 H6 no anion 3.2 3.8 4.3 10.4 Cl – 4.2 2.2 7.0 9.9 AcO – 7.2 1.0 12.0 4.8 BzO – 7.2 0.9 10.4 4.8 HCO 3 – – a 0.0 – a 4.7 H 2 PO 4 – 4.2 0.0 5.3 2.8 a Broad NH signals in the baseline.

Selected NOE enhancements for 1 in the absence and in the presence of different anions (in %).

anion conformer in vacuo in DMSO no anion anti–anti 0.00 0.00 syn–anti 5.09 2.74 syn–syn 11.61 6.60 Cl − anti–anti 6.50 1.20 syn–anti 1.84 0.12 syn–syn 0.00 0.00 AcO − anti–anti 7.26 3.75 syn–anti 2.82 1.74 syn–syn 0.00 0.00 HCO 3 − anti–anti 6.21 2.97 syn–anti 2.02 1.31 syn–syn 0.00 0.00 a Relative energies are reported with respect to the lowest energy (arbitrarily set to 0.00 kcal/mol) in the absence and in the presence of anions. Geometry optimizations were carried out at B3LYP/6-311+G(d,p).

Relative energies a (in kcal·mol −1 ) of receptor 1 in vacuo and in DMSO, in the absence and in the presence of anions.

ZRJYVMPHGZJHQR-UHFFFAOYSA-N InChI=1S/C17H14N4O/c22-17(20-13-5-1-3-11-7-9-18-15(11)13)21-14-6-2-4-12-8-10-19-16(12)14/h1-10,18-19H,(H2,20,21,22) C1=CC2=C(C(=C1)NC(=O)NC3=CC=CC4=C3NC=C4)NC=C2 O=C(NC1=C2C(=CC=C1)C=CN2)NC3=CC=CC4=C3NC=C4 |(250.54,-33.23,;250.54,-44.73,;240.58,-50.48,;230.62,-44.73,;220.66,-50.48,;210.7,-44.73,;210.7,-33.23,;220.66,-27.48,;230.62,-33.23,;202.16,-52.43,;206.83,-62.93,;218.27,-61.73,;260.5,-50.48,;270.46,-44.73,;270.46,-33.23,;280.42,-27.48,;290.38,-33.23,;290.38,-44.73,;280.42,-50.48,;282.81,-61.73,;294.25,-62.93,;298.92,-52.43,)| AuxInfo=1/1/N:10,13,9,14,11,12,19,20,18,21,8,15,6,5,7,16,2,17,22,3,4,1/E:(1,2)(3,4)(5,6)(7,8)(9,10)(11,12)(13,14)(15,16)(18,19)(20,21)/rA:22nOCNNCCCCCCCCCCCCNCCCCN/rB:d1;s2;s2;s4;s3;d6;s7;d8;s9;s6d10;d5;s12;d13;s14;s5d15;s7;s17;s8d18;s15;d20;s16s21;/rC:250,5403,-33,2319,0;250,5403,-44,7319,0;240,5810,-50,4819,0;260,4995,-50,4819,0;270,4588,-44,7319,0;230,6217,-44,7319,0;220,6624,-50,4819,0;210,7031,-44,7319,0;210,7031,-33,2319,0;220,6624,-27,4819,0;230,6217,-33,2319,0;270,4589,-33,2319,0;280,4182,-27,4819,0;290,3774,-33,2319,0;290,3774,-44,7319,0;280,4182,-50,4819,0;218,2715,-61,7306,0;206,8345,-62,9328,0;202,1570,-52,4271,0;298,9236,-52,4269,0;294,2461,-62,9327,0;282,8091,-61,7306,0; C17H14N4O CDK 04282614012D 0 0 0 0 0 999 V3000 M V30 BEGIN CTAB M V30 COUNTS 22 25 0 0 0 M V30 BEGIN ATOM M V30 1 O 250.54027 -33.2319 0 0 M V30 2 C 250.54027 -44.7319 0 0 M V30 3 N 240.58098 -50.4819 0 0 M V30 4 N 260.49954 -50.4819 0 0 M V30 5 C 270.45883 -44.7319 0 0 M V30 6 C 230.62169 -44.7319 0 0 M V30 7 C 220.6624 -50.4819 0 0 M V30 8 C 210.70309 -44.7319 0 0 M V30 9 C 210.70309 -33.2319 0 0 M V30 10 C 220.6624 -27.4819 0 0 M V30 11 C 230.62169 -33.2319 0 0 M V30 12 C 270.45886 -33.23189 0 0 M V30 13 C 280.41815 -27.48189 0 0 M V30 14 C 290.37744 -33.2319 0 0 M V30 15 C 290.37744 -44.7319 0 0 M V30 16 C 280.41815 -50.4819 0 0 M V30 17 N 218.27151 -61.73062 0 0 M V30 18 C 206.8345 -62.93279 0 0 M V30 19 C 202.15695 -52.42706 0 0 M V30 20 C 298.92358 -52.4269 0 0 M V30 21 C 294.24609 -62.93266 0 0 M V30 22 N 282.80908 -61.73059 0 0 M V30 END ATOM M V30 BEGIN BOND M V30 1 2 1 2 M V30 2 1 2 3 M V30 3 1 2 4 M V30 4 1 4 5 M V30 5 1 3 6 M V30 6 2 6 7 M V30 7 1 7 8 M V30 8 2 8 9 M V30 9 1 9 10 M V30 10 2 10 11 M V30 11 1 11 6 M V30 12 2 5 12 M V30 13 1 12 13 M V30 14 2 13 14 M V30 15 1 14 15 M V30 16 2 15 16 M V30 17 1 16 5 M V30 18 1 7 17 M V30 19 1 17 18 M V30 20 2 18 19 M V30 21 1 19 8 M V30 22 1 15 20 M V30 23 2 20 21 M V30 24 1 21 22 M V30 25 1 22 16 M V30 END BOND M V30 END CTAB M END Experimental for the synthesis of compound 4 and details of the crystal structure of the HPO 4 2− complex of 4 , 1 H and 13 C NMR data for 1 – 4, 1D difference NOE spectra for 1 in the absence and upon addition of one equivalent of acetate anions. Crystallographic data of the complex of compound 4 with tetrabutylammonium dihydrogen phosphate ( 4 ·TBA 2 ·HPO 4 ). We gratefully acknowledge the financial support of the Slovenian Research Agency (ARRS, program no. P1-0242), COST D-31, EAST-NMR (grant no. 228461) and Bio-NMR (grant no. 261863) FP7 projects. PAG thanks the EPSRC for support and the access to the crystallographic facilities at the University of Southampton. Supramolecular Chemistry of Anions Chem. Rev. Acc. Chem. Res. J. Org. Chem. Angew. Chem., Int. Ed. Acc. Chem. Res. Angew. Chem., Int. Ed. Acc. Chem. Res. Acc. Chem. Res. Anion Receptor Chemistry Naturwissenschaften Acc. Chem. Res. Eur. J. Inorg. Chem. Nature Nature Coord. Chem. Rev. Org. Biomol. Chem. Angew. Chem., Int. Ed. Chem. Soc. Rev. Chem. Soc. Rev. Chem. Commun. Chem. Soc. Rev. Chem. Commun. Org. Lett. J. Am. Chem. Soc. Angew. Chem., Int. Ed. Chem. Commun. Org. Biomol. Chem. Chem. Commun. Pure Appl. Chem. New J. Chem. Nature Org. Biomol. Chem. Eur. J. Org. Chem. Supramol. Chem. Chem. Commun. Chem.–Eur. J. Chem.–Asian J. Chem. Sci. Gaussian 03 Gaussian 09 Chem. Phys. Chem. Phys. This is an Open Access article under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The license is subject to the Beilstein Journal of Organic Chemistry terms and conditions: (http://www.beilstein-journals.org/bjoc)