Development of new potent agonists able to interact with two postulated subsites of the cholecystokinin CCK-B receptor

Since the biochemical and pharmacological profile of BC 197 and BC 264,two CCK₈-derived agonists with high specificity for CCK-Breceptors, suggests their potential interaction with two CCK-B receptorsubsites, it appeared essential to design new series of compounds that wouldbe able to discriminate between these two subsites. As CCK₄ isthe shortest fragment of CCK which interacts selectively with CCK-Breceptors, compounds derived from the C-terminal tetrapeptide domain of BC264, Boc-Trp-(NMe)Nle-Asp-Phe-NH₂, and of the cyclic compoundBC 197, were prepared. While RB 360(N-(cycloamido)--Me(R)Trp-[(2S)-2-amino-9-((cycloamido)carbonyl)nonanoyl]-Asp-Phe-NH₂), like BC 197, has a CCK-B₁ profilewith anxiogenic-like effects in the elevated plus-maze test, RB 400(HOOC-CH₂-CO-Trp-(NMe)Nle-Asp-Phe-NH₂), like BC264, seems to be a specific CCK-B₂ agonist, able to increaseattention and/or memory processes in the Y-maze test.     

Dual Modulation of Dopamine Release from Anterior Nucleus Accumbens Through Cholecystokinin-B Receptor Subsites

Abstract: Previous binding studies have suggested the existence of two affinity states for cholecystokinin-B (CCK-B) receptor. One study, using BC 197 and BC 264, two highly selective CCK-B agonists, has shown that BC 197 is selective for one subsite, B₁, and that BC 264 has the same affinity for the two subsites, B₁ and B₂. Therefore, the possible involvement of CCK-B subsites in the modulation of endogenous dopamine (DA) release from slices of the anterior part of the nucleus accumbens was investigated with these two agonists in order to associate a functional response with activation of each subsite. The selective B₁ agonist BC 197 produced a dose-dependent increase of 35 m M K⁺-stimulated DA release. In contrast, at a low concentration (20 n M ), BC 264 inhibited the K⁺-evoked DA release, whereas at a higher concentration (1 µ M ), it stimulated the DA release. These two opposing effects were suppressed by the CCK-B antagonist PD-134,308, but not by the CCK-A antagonist L-364,718 and were not prevented by tetrodotoxin, a Na⁺-channel blocker. Moreover, BC 264 at 20 n M , in the presence of PD-134,308 at a concentration that would block the B₂ subsites (0.1 n M ), increased the evoked DA release. All together, these results support further the existence of distinct CCK-B subsites and suggest that, in the anterior nucleus accumbens, their stimulation mediates opposite effects on K⁺-stimulated DA release via a presynaptic mechanism.     

Peptidomimetics as receptors agonists or peptidase inhibitors: a structural approach in the field of enkephalins, ANP and CCK.

Stabilization of biologically active conformations of native peptides by cyclization or introduction of hindering residues led to peptidominetics endowed with high affinity and selectivity for one class of receptors and able to cross the blood brain barrier. This is the case of BUBU, Tyr-D-Ser(OtBu)-Gly-Phe-Leu-Thr(OtBu) and BUBUC, Tyr-D-Cys-(OtBu)-Gly-Phe-Leu-Thr(OtBu) for the opioid delta receptors and of BC 254, Boc-gamma-D-Glu-Tyr(SO3H)-Nle-D-Lys-Trp-Nle-Asp-PheNH2 and of BC 264, Boc-Tyr(SO3H)gNle-mGly-Trp-MeNle-Asp-PheNH2 for central CCK-B receptors. Inhibition of metabolizing peptidases such as aminopeptidase N and endopeptidase 24.11 (NEP) for enkephalins and of NEP and ACE for atrial natriuretic peptide and angiotensin I by mixed inhibitors such as kelatorphan and RB 101 or ES14, rationally designed by taking into account the structural differences in the active site of these zinc-metallopeptidases, led to potent analgesics devoid of the major morphine side effects or to new antihypertensives.     

Conformational analysis of CCK-B agonists using 1H-nmr and restrained molecular dynamics: Comparison of biologically active Boc-Trp-(N-Me)Nle-Asp-Phe-NH2 and inactive Boc-Trp-(N-Me)Phe-Asp-Phe-NH2

The tetrapeptide Boc-Trp-(N-Me)Nle-Asp-Phe-NH₂ is a potent CCK-B agonist. Replacement in this analogue of the norleucine residue by a phenylalanine, to yield Boc-Trp-(N-Me)Phe-Asp-Phe-NH₂, led to a 740-fold decrease in affinity whereas the same decrease in affinity was not observed in their nonmethylated counterparts. In order to ascertain the conformational preferences of these two N-methylated tetrapeptides, a study by two-dimensional (2D) nmr spectroscopy and molecular modeling was undertaken. The solution conformation of the two peptides was examined by ¹H-nmr in a d₆-DMSO/H₂O (80 : 20) mixture. A cis-trans equilibrium, induced by N-methylation, was observed for both analogues, and the proton spectra of the two retamers were fully characterized in each case. ¹H-¹H distance constraints, derived from 2D nuclear Overhauser effect spectroscopy and rotating frame nuclear Overhauser effect spectroscopy experiments, were used as inputs for subsequent restrained molecular dynamics simulations. Comparisons of the nmr and molecular modeling data point toward distinct conformational preferences for these two peptides with an opposite spatial orientation of the Trp residue, and could explain the large difference in their biological activities. Furthermore, the tridimensional structure of Boc-Trp-(N-Me)Nle-Asp-Phe-NH₂ could serve as a model for the design of nonpeptide CCK-B agonists. © 1994 John Wiley & Sons, Inc.     

Association of enkephalin catabolism inhibitors and CCKB antagonists: A potential use in the management of pain and opioid addiction

The overlapping distribution of opioid and cholecystokinin (CCK) peptides and their receptors (μ and δ opioid receptors; CCK-A and CCK-B receptors) in the central nervous system have led to a large number of studies aimed at clarifying the functional relationships between these two neuropeptides. Most of the pharmacological studies devoted to the role of CCK and enkephalins have been focused on the control of pain. Recently the existence of regulatory mechanisms between both systems have been proposed, and the physiological antagonism between CCK and endogenous opioid systems has been definitely demonstrated by coadministration of CCK-B selective antagonists with RB 101, a systemically active inhibitor, which fully protects enkephalins from their degradation. Several studies have also been done to investigate the functional relationships between both systems in development of opioid side-effects and in behavioral responses. This article will review the experimental pharmacology of association of enkephalin-degrading enzyme inhibitors and CCK-B antagonists to demonstrate the interest of these molecules in the management of both pain and opioid addiction. Special issue dedicated to Dr. Eric J. Simon.     

Effects of Peripheral Axotomy on Cholecystokinin Neurotransmission in the Rat Spinal Cord

Abstract : Because cholecystokinin (CCK) acts as a "functional" endogenous opioid antagonist, it has been proposed that changes in central CCKergic neurotransmission might account for the relative resistance of neuropathic pain to the analgesic action of morphine. This hypothesis was addressed by measuring CCK-related parameters 2 weeks after unilateral sciatic nerve section in rats. As expected, significant decreases (-25-38%) in the tissue concentrations and in vitro release of both substance P and calcitonin gene-related peptide were noted in the dorsal quadrant of the lumbar spinal cord on the lesioned side. In contrast, the tissue levels and in vitro release of CCK were unchanged in the same area in lesioned rats. Measurements in dorsal root ganglia at L4-L6 levels revealed no significant changes in proCCK mRNA after the lesion. However, sciatic nerve section was associated with a marked ipsilateral increase in both CCK-B receptor mRNA levels in these ganglia (+70%) and the autoradiographic labeling of CCK-B receptors by [³H]pBC 264 (+160%) in the superficial layers of the lumbar dorsal horn. Up-regulation of CCK-B receptors rather than CCK synthesis and release probably contributes to increased spinal CCKergic neurotransmission in neuropathic pain.     

Conformational analysis of cholecystokinin fragments CCK4, CCK5, and CCK6 by 1H-NMR spectroscopy and fluorescence-transfer measurements

The confortmational behavior of the cholecystokinin-related fragments CCK₄, CCK₅, and CCK₆ as determined by ¹H-nmr spectroscopy in DMSO-d₆ and water and fluorescence-transfer measurements in aqueous medium are greatly dependent on the ionization states of these peptides. Under netral conditions, the backbones of CCK₅ and CCK₆ preferentially adopted folded forms with a β-turn including the four residues Gly-Trp-Met-Asp, probably stabilized by a hydrogen bond between the CO of Gly and the NH of Phe. In these structures, possible induced by an ionic interaction between the carboxylic group of Asp³² and the NH group of the N-terminal amino acid, the lateral chains of the various residues are quite distant from each other (15–16 Å). Under acidic conditions, extended structures without interactions between side chains predominate for CCK₅ and CCK₆, while for CCK₄, a conformational change drawing the Trp and Phe side chains in close proximity was shown by fluorescence. The conformations observed in aqueous medium at physiological pH are discussed in relation to the biological activity of these peptides.     

CCK4 contains the full hormonal information of cholecystokinin in isolated pancreatic acini

In mouse and rat isolated pancreatic acini, the C-terminal tetrapeptide amide of CCK (CCK₄) fully mimicked the actions of the physiological octapeptide hormone (CCK₈) although CCK₄ was 10–100 thousand fold less potent than CCK₈. Parallelism was observed for stimulation of both amylase secretion (including the submaximal secretion observed at supramaximal concentrations of agonist), and stimulation of glucose transport. Furthermore, CCK₄ and CCK₈ were able to comletely inhibit the binding of radioiodinated CCK₃₃ to CCK receptors on acini. Therefore, the CCK₄ sequence appears to be the minimal functional unit which possesses all of the information required to elicit the actions of CCK on the pancreas. The additional 4 amino acids present in CCK₈ increase the affinity of the CCK molecule for pancreatic CCK receptors and thus enhance target organ specificity and sensitivity.     

Synthesis of tantalum and niobium complexes that contain the diamidoamine ligand, [(3,4,5-F3C6H2NCH2CH2)2NMe], and the triamidoamine ligand, [(3,5-Cl2C6H3NCH2CH2)3N]

Several niobium and tantalum compounds were prepared that contain either the diamidoamine ligand, [(3,4,5-F₃C₆H₂NCH₂CH₂)₂NMe]²⁻ ([F₃N₂NMe]²⁻), or the triamidoamine ligand, [(3,5-Cl₂C₆H₃NCH₂CH₂)₃N]³⁻ ([Cl₂N₂NMe]³⁻). The former include [F₃N₂NMe]TaCl₃, [F₃N₂NMe]NbCl₃, [F₃N₂NMe]TaMe₃, [F₃N₂NMe]NbMe₃, [(F₃N₂NMe)TaMe₂][MeB(C₆F₅)₃], [F₃N₂NMe]Ta(CHSiMe₃)(CH₂SiMe₃), [F₃N₂NMe]Ta(CH₂-t-Bu)Cl₂, [F₃N₂NMe]Ta(CH-t-Bu)(CH₃), and [F₃N₂NMe]Ta(η²-C₂H₄)(CH₂CH₃). The latter include [Cl₂N₂NMe]TaCl₂, [Cl₂N₂NMe]TaMe₂, [Cl₂N₂NMe]Ta(η²-C₂H₄), and [Cl₂N₂NMe]Ta(η²-C₂H₂).X-ray diffraction studies were carried out on [F₃N₂NMe]Ta(CHSiMe₃)(CH₂SiMe₃), [F₃N₂NMe]Ta(η²-C₂H₄)(CH₂CH₃), and [Cl₂N₂NMe]TaMe_2..     

Localization of the murine cholecystokinin A and B receptor genes

We have determined the chromosomal locations of the two cholecystokinin (CCK) receptor genes in the mouse. Genetic localization utilized an interspecific backcross panel formed from the cross (C57BL/6J x Mus spretus) F₁ x Mus spretus. Genomic DNAs from 94 individuals in the backcross were analyzed by Southern hybridization with rat CCK_A and CCK_B receptor cDNA probes. Unique map positions were determined by haplotype analysis with 650 previously mapped loci in the mouse backcross. The CCK_A receptor gene (Cckar) mapped to mouse Chromosome (Chr) 5, in tight linkage with the DNA marker D5Bir8. The CCK_B receptor gene (Cckbr) mapped to mouse Chr 7, tightly linked to the -hemoglobin locus (Hbb). This localization places Cckbr in the same region as the mouse obesity mutation tubby (tub), which also maps near Hbb (2.4±1.4 cM). Since CCK can function as a satiety factor when administered to rodents, localization of Cckbr near the tub mutation identifies this receptor as a possible candidate gene for this obesity mutation.     

Cholecystokinin and insulin regulate insulin-like growth factor II binding to pancreatic receptors; Evidence of a role for intracellular calcium

The binding of ¹²⁵I-labeled insulin-like growth factor II (¹²⁵I-IGF II) to mouse pancreatic acini was stimulated (45%) by insulin and inhibited (30%) by cholecystokinin octapeptide (CCK₈). When CCK₈ and insulin were added together, the effect on IGF II binding was similar to that seen when CCK₈ was added alone. Two lines of evidence suggest that this effect of cholecystokinin on basal and insulin-stimulated ¹²⁵I-IGF II binding was mediated via a change in intracellular calcium: (1) the cholinergic agent carbachol inhibited IGF II binding to its receptors; (2) addition of the Ca²⁺ ionophore A23187 mimicked the effects of CCK₈ and carbachol. In contrast to its effects on IGF II binding to acini, CCK₈ had only small effects on IGF I binding and no effects on insulin binding.     

1H NMR conformational study of sulfated and non-sulfated cholecystokinin fragment CCK27–33: Influence of the sulfate group on the peptide folding

¹H NMR study of cholecystokinin fragment (CCK_27–33) in (C²H₃)₂SO and in ²H₂O at different pH shows that sulfated (CCK₇) and non sulfated (NS-CCK₇) peptides are under preferentially folded conformations characterized by a β-turn including the sequence Gly-Trp-Met-Asp with a H-bond between the CO of Gly and the NH of Asp. This structure is probably stabilized by an ionic interaction between Tyr and Asp. Moreover, the N-terminal part of CCK₇ forms a C₇ structure with a weak H-bond between the CO of Gly and the NH of Trp. In this model all CCK₇ hydrophobic side chains are in close vicinity, far from the hydrophilic sulfate group. Full interaction with brain CCK₈ receptors could require both the sulfate group and the maintening of conformational constraints.     

Subsite Structure of α-Amylase II from Thermoactinomyces vulgaris R-47

We estimated the subsite structure of α-amylase II (TVA II) from Thermoactinomyces vulgaris R-47 expressed in Escherichia coli. TVA II has eight subsites, and the catalytic site is between the 5th and 6th subsite from the non-reducing end side. The subsite affinities, A_-5, A_-4, A_-3, A_-2, (A_-1+A₊₁), A₊₂, and A₊₃, were calculated to be -0.35, 0.93, 0.55, 2.56, 1.18, 1.71, and 0.01 kcal mol^-1, respectively.     

Stimulation of δ-Opioid Receptors Reduces the In Vivo Binding of the Cholecystokinin (CCK)-B-Selective Agonist [3H]pBC 264: Evidence for a Physiological Regulation of CCKergic Systems by Endogenous Enkephalins

Cholecystokinin (CCK) and enkephalins appear to be colocalized in several brain structures, and a physiological interaction between these peptides has been suggested by a large number of pharmacological studies. In this work we have shown, by in vivo binding experiments, that the endogenous enkephalins, protected from degrading enzymes by mixed inhibitors such as kelatorphan and N-[(R,S)-2-benzyl-3-[(S)-2-amino-4-methylthiobutyldithio]-1-oxo pro pyl]- L-phenylalanine benzyl ester (RB 101), a systemically active prodrug, modulate CCK release in mouse brain, leading to an overall increase in the extracellular levels of CCK. This was quantified by measuring the effects of both inhibitors on the in vivo binding of [3H]propionyl-Tyr(SO3H)-gNle-mGly-Trp-(N-Me)Nle-Asp-Phe-NH2 ([3H]pBC 264), a selective and highly potent CCK-B agonist. Thus, intracerebroventricular injection of kelatorphan produced a dose-dependent inhibition of the in vivo binding of [3H]pBC 264 with a maximal effect (40%) at 50 nmol. A similar response was observed after intravenous injection of RB 101 (40 mg/kg). The specific binding of [3H]pBC 264 was also inhibited (25%) by intravenous injection of the selective delta-opioid agonist H-Tyr-D-Cys(StBu)-Gly-Phe-Leu-Thr(OtBu)-OH (BUBUC; 2 mg/kg) but not by the mu-agonist H-Tyr-D-Ala-Gly-(N-Me)Phe-Gly-ol (5 mg/kg), suggesting a preferential involvement of delta-opioid receptors in the modulation of CCK release. This was confirmed by using the selective delta-opioid antagonist naltrindole, which prevented the inhibitory effects of BUBUC and of enkephalin-degrading enzyme inhibitors on [3H]pBC 264 binding.(ABSTRACT TRUNCATED AT 250 WORDS)     

The Synergistic Roles of Cholecystokinin B and Dopamine D5 Receptors on the Regulation of Renal Sodium Excretion

Renal dopamine D₁-like receptors (D₁R and D₅R) and the gastrin receptor (CCK_BR) are involved in the maintenance of sodium homeostasis. The D₁R has been found to interact synergistically with CCK_BR in renal proximal tubule (RPT) cells to promote natriuresis and diuresis. D₅R, which has a higher affinity for dopamine than D₁R, has some constitutive activity. Hence, we sought to investigate the interaction between D₅R and CCK_BR in the regulation of renal sodium excretion. In present study, we found D₅R and CCK_BR increase each other’s expression in a concentration- and time-dependent manner in the HK-2 cell, the specificity of which was verified in HEK293 cells heterologously expressing both human D₅R and CCK_BR and in RPT cells from a male normotensive human. The specificity of D₅R in the D₅R and CCK_BR interaction was verified further using a selective D₅R antagonist, LE-PM436. Also, D₅R and CCK_BR colocalize and co-immunoprecipitate in BALB/c mouse RPTs and human RPT cells. CCK_BR protein expression in plasma membrane-enriched fractions of renal cortex (PMFs) is greater in D₅R^-/- mice than D₅R^+/+ littermates and D₅R protein expression in PMFs is also greater in CCK_BR^-/- mice than CCK_BR^+/+ littermates. High salt diet, relative to normal salt diet, increased the expression of CCK_BR and D₅R proteins in PMFs. Disruption of CCK_BR in mice caused hypertension and decreased sodium excretion. The natriuresis in salt-loaded BALB/c mice was decreased by YF476, a CCK_BR antagonist and {"type":"entrez-protein","attrs":{"text":"Sch23390","term_id":"1052733334"}}Sch23390, a D₁R/D₅R antagonist. Furthermore, the natriuresis caused by gastrin was blocked by {"type":"entrez-protein","attrs":{"text":"Sch23390","term_id":"1052733334"}}Sch23390 while the natriuresis caused by fenoldopam, a D₁R/D₅R agonist, was blocked by YF476. Taken together, our findings indicate that CCK_BR and D₅R synergistically interact in the kidney, which may contribute to the maintenance of normal sodium balance following an increase in sodium intake.     

Basic Amino Acids at the C-Terminus of the Third Intracellular Loop Are Required for the Activation of Phospholipase C by Cholecystokinin-B Receptors

Abstract: In common with other G_q protein-coupled receptors, the third intracellular loop of the cholecystokinin-B (CCK-B) receptor contains three basic amino acids (K333/K334/R335) at the C-terminal segment. To determine the importance of these conserved basic residues in G_q-protein activation and stimulation of phospholipase C, these basic amino acids were mutated. Subsequently, the ability of resulting mutant receptors to activate phospholipase C was investigated by measuring inositol phosphate formation in COS-7 cells and recording Ca²⁺-activated Cl^? currents from Xenopus oocytes. Site-directed mutagenesis was performed to mutate the three basic amino acids, K333/K334/R335, to neutral amino acids, M333/T334/L335. When the resulting mutant CCK-B receptors were expressed in COS-7 cells and Xenopus oocytes, sulfated cholecystokinin octapeptide (CCK-8) failed to induce inositol phosphate formation in COS-7 cells and evoke Ca²⁺-activated Cl^? currents from oocytes. Each basic amino acid was also mutated (K333M, K334T, and R335L). All three single-point mutations resulted in a significant reduction in CCK-8-induced inositol phosphate formation and CCK-8-activated Ca²⁺-dependent Cl^? currents. It is interesting that substituting the basic amino acids, K333/K334/R335, with three other basic residues, R333/R334/K335, did not change the maximal CCK-8-simulated inositol phosphate formation and the amplitude of CCK-8-evoked Ca²⁺-dependent Cl^? currents. Radioligand-binding studies showed that the above-mentioned mutations did not affect the affinity for CCK-8 and receptor expression level in COS-7 cells. These findings suggest that basic amino acids at the C-terminus of the third cytoplasmic loop are required for the signal transduction by CCK-B receptors.     

Anthranilic acid based CCK1 receptor antagonists: Blocking the receptor with the same ‘words’ of the endogenous ligand

The anthranilic acid diamides represent the more recent class of nonpeptide CCK₁ receptor antagonists. This class is characterized by the presence of anthranilic acid, used as a molecular scaffold, and two pharmacophores selected from the C-terminal tetrapeptide of CCK. The lead compound coded VL-0395, endowed with sub-micromolar affinity towards CCK₁ receptors, was characterized by the presence of Phe and 2-indole moiety at the C- and N-termini of anthranilic acid, respectively. Herein we describe the first step of the anthranilic acid C-terminal optimization using, instead of Phe, aminoacids belonging to the primary structure of CCK-8 and other not coded residues. Thus we demonstrate that the CCK₁ receptor affinity depends on the nature of the aminoacidic side chain as well as that the free carboxy group of the alpha-aminoacids is crucial for the binding. The R enantiomers of the most active compounds represent the eutomers of this class of antagonists confirming thus the stereo preference of the receptor. Moreover this SAR study demonstrates that the receptor binding pocket, that host the aminoacidic side chain, results much more tolerant respect to that accommodating the indole ring. As a result, an appropriate variation of the aminoacidic side chain could provide a better CCK₁ receptor affinity diorthosis.     

Rapakinin, Arg-Ile-Tyr, derived from rapeseed napin, shows anti-opioid activity via the prostaglandin IP receptor followed by the cholecystokinin CCK(2) receptor in mice

Rapakinin, Arg-Ile-Tyr, is a vasorelaxing, anti-hypertensive and anorexigenic peptide derived from rapeseed napin. In this study, we found that rapakinin intracerebroventricularly administered to mice inhibited the analgesic effect of morphine, evaluated by the tail-pinch test. The anti-opioid activity of rapakinin was blocked by LY225910, an antagonist of the cholecystokinin (CCK) CCK₂ receptor, but not by lorglumide, an antagonist of the CCK₁ receptor. The anti-opioid activity of rapakinin was also blocked by CAY10441, an antagonist of the prostaglandin (PG) IP receptor. These results suggest that the anti-opioid activity of rapakinin is mediated by the CCK₂ and IP receptors. The anti-opioid activity induced by ciprostene, an IP receptor agonist, was blocked by LY225910, while that of CCK-8 was not blocked by CAY10441. Thus, it is demonstrated that the CCK-CCK₂ system was activated downstream of the PGI₂-IP receptor system. Taken together, rapakinin shows anti-opioid activity via the activation of the PGI₂-IP receptor system followed by the CCK-CCK₂ receptor system.     

[3H] BOC [NLE28,31]CCK27−33, a new highly labelle ligand for CCK receptors: Binding on brain and on pancreas

Preliminary results on the binding of [³H]Boc[Nle^28,31]CCK_27^?33, designated [³H]Boc[diNle]CCK₇, on mouse brain and rat pancreas membranes are presented. This new ligand for CCK receptors possesses a high specific activity (144 Ci/mmole), and binds in a saturable manner to mouse brain (Kd = 0.49 nM, Bmax = 49 fmoles/mg protein) and rat pancreas (Kd = 4.4 nM, Bmax = 696 fmoles/mg protein). Unlabelled Boc[diNle]CCK₇ displaces [¹²⁵I]CCK₈ from its binding sites on mouse brain membranes with a high affinity, slightly superior to that of CCK₈. The order of potencies to displace [³H]Boc[diNle]CCK₇ from its binding sites was the same on mouse brain and rat pancreas: $\text{[}{}^3\text{HBoc[diNle]CCK}{}_7\text{>}\text{CCK}{}_8\text{,}\text{Boc-CCK}{}_7\text{>}\text{non-sulfated CCK}{}_8$, the pancreas binding sites being more discriminative than the brain binding sites. Thus, [³H]Boc[diNle]CCK₇ is a very promising new probe for the characterization of CCK receptors and their interaction with different CCK fragments.