The study of the interaction of ghrelin (1), the endogenous ligand for the GH secretagogues receptor (GHS-R1a), and des-acyl ghrelin (2) with the GHS-R1a by NMR using living cells is presented, using GHS-R1a stably transfected cell lines (CHO and HEK 293) and wild type cells. Therefore, the interaction of 1 and 2 with the GHS-R1a receptor has been performed using quasi-physiological conditions. Ghrelin (1), showed a higher number of residues affected by chemical shift perturbation (CSP) or chemical shift exchange (CSE) effects: Ser3, Phe4, Leu5, Val12, Gln13/Gln14, Lys16/Lys19, Glu17 and Lys24 were much more affected in 1 than in des-acyl ghrelin (2). The chemical shift index CSI values indicated the presence of a possible α-helical region between Glu8 and Lys20 for ghrelin (1). After analysing the NMR data, two possible structures have arisen, which present different proline rotamers: the EEZE and the EZEZ conformers, at positions Pro7, Pro21, Pro22 and Pro27, respectively, keeping a left-handed α-helix from Glu8 to Lys20. These experimental evidences might imply that the GHS-R1a receptor is acting as a prolyl-cis/trans isomerase.
Ghrelin (1), the endogenous ligand for the growth hormone secretagogue receptor,1 is a 28-amino residue peptide with a post-translational octanoyl modification on Ser3 (Fig. 1), which was first discovered in rat and human stomach tissues.2 This hormone is mainly synthesized in the stomach, but substantially lower amounts have been detected in other tissues.3 Functionally, ghrelin (1) stimulates growth hormone (GH) secretion from pituitary somatotropes2, 4 and increases food intake and body weight.5 Indeed, it has been proposed that 1 acts directly on the hypothalamic regulatory nuclei that control energy homeostasis acting as an orexigenic peptide.6 On the other hand, des-acyl ghrelin (2), which presents the same structure with the exception of the n-octanoyl modification on Ser-3, does not show the same functionalities.7 The receptor GHS-R1a transduces the information provided by ghrelin (1) and the group of growth hormone secretagogues (GHS), not structurally related to it. These striking properties have been explained on the basis of the existence of a common binding domain, as demonstrated, using GHS peptide and non-peptide agonists, by site-directed mutagenesis studies assisted by molecular modelling procedures.8
Since the discovery of ghrelin (1), there have been several studies aimed at determining which are the minimal structural requirements that permit to detect ghrelin receptor biological activity. Bednarek et al. reported the first ghrelin-based minimally active structure–activity studies, demonstrating that the minimum sequence necessary for GHS-R1a activation encompassed the first five residues, with the octanoyl modification on Ser-3. The employed protocol involved binding assays and activation of GHS-R1a by measuring intracellular calcium mobilisation, using HEK 293 cells transfected with GHS-R1a.9 However, it was later demonstrated that this truncated analogue was not capable of stimulating GH secretion from somatotrope cells.10 In principle, this discrepancy might be attributed to the fact that the rise in calcium obtained for the truncated analogues does not reflect the complete activation of the signal transduction systems as required, for example, to activate GH secretion. At the present moment, no data are available regarding the bioactive conformation of ghrelin (1) and its mode of interaction of GHS-R1a at the key binding site. In principle, given the presence of four Pro residues within the 28-amino acid primary sequence, as well as the n-octanoylation at Ser3, one could expect the presence of conformational heterogeneity and a fair amount of flexibility. To the best of our knowledge, three publications have reported approaches to determine the 3D structure of ghrelin (1) in solution. The 1 H NMR studies performed by Silva Elipe et al.11 showed that ghrelin behaves as an unstructured and/or fast interconverting peptide at acidic pH. Later, Beevers and Kukol 12 reported a molecular dynamics (MD) simulation study at neutral pH in water and in the presence of a lipid bilayer, proposing the existence of stable secondary structural features for 1 in the latter case. In particular, the presence of a short α-helix from Pro7 to Glu13 and a hairpin structure with Glu17 to Lys20 in the bending region. Very recently, Dehlin et al. have reported on the CD study of ghrelin (1) and des-acyl ghrelin (2) in the presence of Tris pH 7.4 and of the α-helix stabilizing solvent, TFE. Although rather qualitative, it was described that the helical content in 1 and 2 was enhanced from 12% to 23% and 49%, respectively.13
In any case, the proper structure–activity relationship study should be based on the knowledge of the bioactive conformation of 1 and 2 when bound to the receptor. NMR techniques are suitable to this end when working with isolated protein receptors,14, 15, 16 and even recently, new NMR experiments using receptor-rich living cells have been reported.17 Provided that the system holds the right kinetic features, these experiments, with living cells, could avoid the requirement for isolating the protein receptor. On this basis, the approach presented herein aims at the determination of the bioactive conformation of ghrelin (1) when interacting with its receptor, at neutral pH. Indeed, since the NMR study of ghrelin (1) is performed with living cells highly decorated with the GHS-R1a receptor, physiological conditions similar to those taking place in Nature are kept. Moreover, as the n-octanoyl modification of 1 has been shown to be essential for its physiological role, the des-acyl ghrelin analogue (2) has also been analysed using the same approach. Thus, the NMR analysis of the interaction of 1 and 2 with the ghrelin receptor (GHS-R1a) was performed in CHO and HEK 293 cell lines.
To assure time stability of the samples, and therefore reliability and reproducibility of the NMR experiments, a series of 1D 1 H NMR spectra of the samples with living cells were collected at regular intervals during the course of one day. These spectra were recorded for ghrelin (1) and des-acyl ghrelin (2) in the presence of wild type and GHS-R1a-containing cells. The 1 H NMR spectra of the 1:CHO-GHSR1a sample and of the 1:CHO sample showed drastic changes for most of the exchangeable amide
Ghrelin (1) was purchased from Global Peptides (Fort Collins, Co, USA). Des-acyl ghrelin (2) was obtained from Bachem AG (Bubendorf, CH). F-12 Ham was purchased from Sigma Chemical Co (St. Louis, MO, USA). DMEM was purchased from Cambrex Bio Science (Walkersville, MD, USA). D 2 O was purchased from Spectra Stable Isotopes (Columbia, MD, USA).
HEK 293 and CHO cell lines were cultured as described by the supplier (ECACC, Wiltshire, UK). Briefly, cells were seeded in 100-mm dishes and cultured in
We thank Dr. R. G. Smith (Huffington Center on Aging and Department of Molecular and Cellular Biology, Houston, Texas) for providing HEK-GHS-R1a cell line and Dr. C. Llorens-Cortes (Institut National de la Santé et de la Recherche Médicale, Paris, France) for providing CHO-GHS-R1a cell line. This work was supported by grants from the FIS and the Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación ( PI050382 , PI060239 , and PI060705 ) and Xunta de Galicia ( PGIDIT05BTF20802PR , and
The role of C-terminal part of ghrelin in pharmacokinetic profile and biological activity in rats 2011, Peptides These results support the importance of N-terminal part of ghrelin for the activation of GHS-R1a. In contrast, little is so far known about the biological role of the C-terminal (8–28) region in ghrelin, except that ghrelin binds to human plasma lipoproteins via both its N- and C-terminal parts [4], and that some of amino acids at position 12–24 of ghrelin biochemically interact with GHS-R1a by NMR analysis [17]. To explore the function of the C-terminal (8–28) region in ghrelin, we prepared human ghrelin, C-terminal truncated ghrelin derivatives and a small molecular GHS compound, anamorelin [7,14], and evaluated GHS-R1a agonist activity in vitro, and pharmacokinetic (PK) profile in rats after a single intravenous (iv) injection in the present study.
