Encapsulation complexes are assemblies in which small molecular guests are completely by large molecular hosts. [1±4] The hosts are made up of subunits held together by intermolecular forces: hydrogen bonds, van der Waals forces, and/or metal ± ligand interactions. The assemblies are formed reversibly and are dynamic; they come together and dissipate on time scales ranging from milliseconds to days, long enough for their study by NMR methods. When multiple hosts can assemble from a given set of subunits, template effects can be expected and these have recently been reported. [5±9] Here we report an encapsulation complex with the additional characteristic that hydrogen bonds maintain an im-printÐthe ghostÐof a long-departed guest. Specifically, an asymmetric microenvironment is imprinted in a reversibly formed capsular host by a chiral guest template. Removal of the template leaves a chiral, nonracemic (g)host capsule that can be characterized by NMR spectroscopy. Recognition of the guest rather than its mirror image persists for hours in organic solvents.We recently reported a capsule known as the chiral ªsoftballº 1 a´1 a (Scheme 1 a). [5] It is formed when two selfcomplementary subunits 1 a dimerize in organic solvents through a seam of eight hydrogen bonds. The subunits feature a plane of symmetry and are achiral, but the dimer has only C 2 axes and exists as a pair of enantiomers. The cavity of the capsule is a distorted sphere and asymmetric guests generally prefer one enantiomer of the capsule to its mirror image. The enantiomeric capsules can interconvert (racemize) only by complete dissociation and recombination of their subunits (monomer exchange). Evidence reported elsewhere [10±12] indicates that guests get in and out of these capsules through flaps that are opened by the breaking of hydrogen bonds as conformational changes occur.A related structure 1 b was prepared for the imprinting studies. [13] It dimerizes into capsules that are more robust and assemble in a number of solvents. The phenolic groups introduce four additional hydrogen bonds in the corresponding capsule 1 b´1 b and slow its rate of racemization. Figure 1 shows how the NMR spectrum of 1 b´1 b with added chiral Scheme 1. a) Structure of the monomer 1 b and model of its dimeric assemblies. [22] Some protons and n-heptylphenyl groups in the dimers are omitted for clarity. Atoms are colored as follows: carbon and hydrogen: orange, oxygen: red,s nitrogen: blue. b) Guests used in the study. Figure 1. Portions of the 1 H NMR spectra showing the selected NH peaks of the dimer (8.65 ± 8.45 ppm) and H a peaks for the guest inside the capsules (2.90 ± 2.60 ppm). The labels A and A' denote the thermodynamically more-stable complexes of ()-2 and (À)-2, respectively, while B and B' denote the respective less-stable complexes. a) 1 b´1 b in [D 10 ]p-xylene (8.25 Â 10 À4 m); b) 4 min and c) 120 h after addition of 3 equiv of ()-2; d) 17 min and e) 191 h after addition of 30 equiv of (À)-2. guest, ()-pinanediol (()-2; Scheme 1 b), varies with tim...

Hypersecretion and alterations in the biological activity of the incretin hormone, glucose-dependent insulinotropic polypeptide (GIP), have been postulated as contributing factors in the development of obesity-related diabetes. However, recent studies also point to weight-reducing effects of GIP receptor activation. Therefore, generating precise experimental tools, such as specific and effective GIP receptor (GIPR) antagonists, is of key significance to better understand GIP physiology. Thus, the primary aim of the current study was to uncover improved GIPR antagonists for use in rodent studies, using human and mouse GIP sequences with N- and C-terminal deletions. Initial in vitro studies revealed that the GIPR agonists, human (h) GIP(1-42), hGIP(1-30) and mouse (m) GIP(1-30), stimulated ( P < 0.01 to P < 0.001) insulin secretion from rat BRIN-BD11 cells. Analysis of insulin secretory effects of the N- and C-terminally cleaved GIP peptides, including hGIP(3-30), mGIP(3-30), h(Pro3)GIP(3-30), hGIP(5-30), hGIP(3-42) and hGIP(5-42), revealed that these peptides did not modulate insulin secretion. More pertinently, only hGIP(3-30), mGIP(3-30) and h(Pro3)GIP(3-30) were able to significantly ( P < 0.01 to P < 0.001) inhibit hGIP(1-42)-stimulated insulin secretion. The human-derived GIPR agonist sequences, hGIP(1-42) and hGIP(1-30), reduced ( P < 0.05) glucose levels in mice following conjoint injection with glucose, but mGIP(1-30) was ineffective. None of the N- and C-terminally cleaved GIP peptides affected glucose homeostasis when injected alone with glucose. However, hGIP(5-30) and mGIP(3-30) significantly ( P < 0.05 to P < 0.01) impaired the glucose-lowering action of hGIP(1-42). Further evaluation of these most effective sequences demonstrated that mGIP(3-30), but not hGIP(5-30), effectively prevented GIP-induced elevations of plasma insulin concentrations. These data highlight, for the first time, that mGIP(3-30) represents an effective molecule to inhibit GIPR activity in mice.