Discovery of Novel Potent and Selective Agonists at the Melanocortin-3 Receptor

Abstract

The melanocortin receptors 3 and 4 control energy homeostasis, food-intake behavior, and correlated pathophysiological conditions. The melanocortin-4 receptor (MC4R) has been broadly investigated. In contrast, the knowledge related to physiological roles of the melanocortin-3 receptor (MC3R) is lacking because of the limited number of known MC3R selective ligands. Here, we report the design, synthesis, biological activity, conformational analysis, and docking with receptors of two potent and selective agonists at the human MC3 receptor.

Graphical Abstract

INTRODUCTION

The family of melanocortin receptors (MCRs) together with their endogenous ligands is engaged in the modulation of numerous pathophysiological pathways including feeding behavior and energy homeostasis, skin pigmentation, sexual function, as well as several other critical biological activities.¹ The endogenous melanocortin agonists, also known as the melanocortin peptides, consist of α-, β-, and γ-melanocyte-stimulating hormones (MSH) and adrenocorticotropic hormone (ACTH).² In humans, five different receptor subtypes, referred to as human melanocortin receptors 1–5 (hMC1–5R), have been discovered, to date. All of them belong to the rhodopsin family of seven-transmembrane-helix G protein-coupled receptors (GPCRs) and convey their effects in the cytosol by activating cAMP-dependent pathways, generating a cascade of specific intracellular events.³ All melanocortin peptides are agonists toward hMCRs, except for the hMC2R, which is exclusively recognized by ACTH that acts as a full agonist as well.⁴ In contrast, the endogenous hMCRs antagonists comprise the agouti signaling protein (hASIP, also known as the “agouti protein”), which mainly binds to the hMC1R and hMC4R but displaying lower affinity toward the hMC3R, and the agouti-related protein (hAGRP) that acts as selective ligand at the hMC3,4R subtypes.⁵ Intense efforts are addressed to the development of selective molecules active at the melanocortin receptors. In particular, shedding light on the role of melanocortin receptor subtypes expressed in the brain, i.e., hMC3R, hMC4R, and hMC5R, is intriguing considering the discovery that MC3R and MC4R are engaged in the control of food-intake behavior.^5,6 Extensive structure–activity relationship studies (SARs) on melanocortins, particularly upon α-MSH, have turned out to be useful for the development of small cyclic peptides such as MT-II, Ac-Nle⁴-c[Asp⁵-His⁶-DPhe⁷-Arg⁸-Trp⁹-Lys¹⁰]-NH₂,⁷ a potent but not selective agonist at the hMC1,3,4,5R, and SHU-9119, Ac-Nle⁴-c[Asp⁵-His⁶-D(2′)Nal⁷-Arg⁸-Trp⁹-Lys¹⁰]-NH₂, a nonselective antagonist with high affinity at the hMC3,4R and an agonist at the hMC1,5R.⁸ Further structure–activity studies have revealed that exchanging the His residue in position 6 in both sequences of MT-II⁹ and SHU-9119¹⁰ with Pro residue is well tolerated, generating cyclic peptides with similar/improved activity profiles compared to the corresponding parent peptide.

Herein, we report the synthesis, the biological activity at hMC1 and hMC3–5R, and the conformational analysis of a few analogues in which a Xaa⁶-Pro⁷ dipeptide replaced the His⁶ residue of MT-II or SHU9119 (Table 1), some of which resulted in potent and selective agonists at the human MC3 receptor.

Table 1.

Binding Affinities and cAMP Activities of the Designed Peptides.^a

		hMC1R			hMC3R			hMC4R			hMC5R
peptide (code)	structure	IC50 (nM)a	EC50 (nM)a	Act%c	IC50 (nM)a	EC50 (nM)b	Act%c	IC50 (nM)a	EC50 (nM)b	Act%c	IC50 (nM)a	EC50 (nM)b	Act%c
1 (PG-989)	Ac-Nle-c[D-P-P-DPhe-R-W-K]-NH2	>5000	860 ± 100	100 ± 7	590 ± 12	NAd		1300 ± 11	NAd		NBe	NAd
2 (PG-990)	Ac-Nle-c[D-P-P-DNal(2′)-R-W-K]-NH2	>5000	940 ± 100	100 ± 7	2.4 ± 0.1	1.9 ± 0.1	100 ± 7	190 ± 13	>1000		300 ± 15	10.1 ± 0.1	100 ± 9
3 (PG-991)	Ac-Nle-c[D-W-P-DPhe-R-W-K]-NH2	>5000	26 ± 3	100 ± 8	580 ± 18	>1000	0	>1000	220 ± 11	84 ± 9	>1000	73 ±9	89± 11
4 (PG-992)	Ac-Nle-c[D-W-P-DNal(2′)-R-W-K]-NH2	NBe	>1000	100 ± 11	11.1 ± 0.1	42 ± 12	94 ± 8	950 ± 22	>1000		580 ± 10	20 ± 4	100 ± 13
MT-II	Ac-Nle-c[D-H-DPhe-R-W-K]-NH2	1.2 ± 0.2	1.8 ± 0.2	100 ± 8	1.2 ± 0.2	1.8 ± 0.2	100 ± 8	1.1 ± 0.3	2.9 ± 0.5	100 ± 12	7.5 ± 0.2	3.3 ± 0.7	100 ± 17

^aIC₅₀ = concentration of peptide at 50% specific binding (N = 4).

^bEC₅₀ = effective concentration of peptide that was able to generate 50% maximal intracellular cAMP accumulation (N = 4).

^cAct% max is ratio of the highest cAMP level triggered by peptides over the highest cAMP level triggered by MT-II. The peptides were tested at a range of concentrations from 10⁻¹⁰ to 10⁻⁵ M.

^dNA: 0% cAMP accumulation at 10⁻⁵ M.

^eNB: no binding at 10⁻⁵ M.

RESULTS AND DISCUSSION

Design of Novel Ligands

Starting from Pro⁶ analogue of both MT-II and SHU9119, novel derivatives were designed by the insertion of an additional residue before Pro⁶; in fact, the increase of ring size from 23- to 26-mer did not significantly affect the ligand binding at human melanocortin receptors in His⁶ as well as in Pro⁶ derivatives.^11,12 It is well-known that l-proline residue in proteins and peptides has remarkable effects on their secondary structures. The presence of prolines induces unique properties to proteins due to their conformational restriction, lack of a hydrogen bond donor on the amide, and their propensity to form a cis amide bond, due mainly to the differential sterics of a secondary vs tertiary amide bond. In this context, residues preceding proline in Xaa-Pro influence the percentage of the cis Xaa-Pro bonds in peptides and proteins.¹³ In particular, Pro and Trp residues in the Xaa position gave the lowest (6%) and highest (37.7%) fraction of cis isomer, respectively. Hence we chose those two amino acids obtaining the analogues 1–4 (Table 1).

Peptide Synthesis

Peptides 1–4 were manually synthesized by adopting a solid-phase strategy and a Fmoc/tBu chemistry for orthogonal protection of functional groups. The physicochemical properties and purity of these peptides are reported in Table S1 and Figures S1–S4 (Supporting Information (SI)).

Biological Data

Table 1 summarizes the binding affinities of the novel analogues along with intracellular cAMP accumulation in cloned human MC1R, MC3R, MC4R, and MC5R. For comparison, MT-II data are also reported. The Pro⁶-Pro⁷-DPhe⁸ analogue (1) shows no appreciable binding at the hMC1R and hMC5R and low binding affinity at the hMC3R and hMC4R. Accordingly, low potency (hMC1R) or undetectable activity (hMC3–5Rs) is observed in functional assays. In contrast, the Pro⁶-Pro⁷-DNal(2′)⁸ analogue (2) shows strong binding affinity and selectivity for the hMC3R (K_i = 2.4 nM) vs the hMC1R and hMC4R–hMC5Rs by 2–3 orders of magnitude. It also displays potent activity at the hMC3R (EC₅₀ = 1.9 nM) behaving as a full agonist and is about 10-fold more selective at this receptor compared with the hMC5R, whereas it does not increase cAMP through the hMC4R. The Trp⁶-Pro⁷-DPhe⁸ analogue (3) shows weak binding at the hMC3R and no affinity for the hMC1R, hMC4R–hMC5Rs (Table 1). It also exhibits no stimulation of cAMP accumulation at the hMC3R and weak and moderate activation of the hMC1R, hMC4R–hMC5Rs. To note, the Trp⁶-Pro⁷-DNal(2′)⁸ analogue (4) shows strong binding affinity and selectivity for the hMC3R (K_i = 11.0 nM) vs the hMC1R, hMC4R–hMC5Rs by at least 2 orders of magnitude. It also shows moderate activity at the hMC3R (EC₅₀ = 42 nM) with a partial agonist behavior and is about 500-fold more selective at this receptor compared with the hMC4R while it also stimulates cAMP accumulation at hMC5R (EC₅₀ = 20 nM). Hence, peptides 2 and 4 are potent and selective (at least considering hMC4R) agonists at the hMC3R. It is well-known that MC3R and MC4R agonist ligands reduce food intake.¹⁴ By the way, the use of selective agonists toward the hMC4R promotes hypertension because of a MC4R-mediated process that has not been completely figured out yet.¹⁵ Therefore, acting selectively at the hMC3R may be a more interesting target suitable for energy homeostasis therapies. Although the compounds 2 and 4 show an agonist activity at hMC5R, the low profile in binding affinity at this receptor limits their capability to be used as useful ligands at this receptor. The development of peptides 2 and 4 can thus prove to be very useful in the perspective of existence of very few selective and “clean” MC3R ligands that can be used to fully elucidate the receptor function.

NMR Analysis

NMR spectra of active compounds 2 and 4 were collected in a 200 mM aqueous solution of DPC. For comparison purpose also inactive peptide 1 was studied by NMR. DPC micelle solutions were employed because they mimic membrane environments and are diffusely used for structural studies of peptide neurotransmitters and hormones.¹⁶ As a consequence of the conformational restrictions imposed by the cis/trans interconversion of the Pro amide bond, peptides 1–2 showed two slowly interchangeable conformational states (indicated as I and II) with populations of about 50 and 40% for state I in peptides 1 and 2, respectively. For both states, an almost complete assignment could be achieved for all proton resonances according to the Wüthrich¹⁷ strategy via the usual systematic application of DQF-COSY,¹⁸ TOCSY,¹⁹ and NOESY²⁰ experiments with the support of the XEASY software package (Supporting Information, Tables S2–S5, SI). Sequential NOE connectivities of the α-protons of Asp⁵ and Pro⁶ with the δ-protons of Pro⁶ and Pro⁷, respectively, provide evidence that in state I both Xaa-Pro amide bonds have the trans configuration (I: trans–trans). Conversely, NOE connectivity between α-proton of residues Asp⁵ and the δ-protons of Pro⁶ and between α-protons of residues Pro⁶ and Pro⁷, indicates the trans–cis configurations of the consecutive Asp⁵-Pro⁶ and Pro⁶-Pro⁷ amide bonds in state II (II: trans–cis). Considering peptide 4, only one state could be observed which could be determined as the trans–cis state (SI, Table S6). Thus, according to the design strategy, a Trp residue preceding the Pro⁷ residue stabilized a cis amide bond. A single configuration state of 4 is observed also in water solution (data not shown) and supports the idea that the trans–cis state is the one interacting with the receptor. Chemical shift values of protons belonging to state I of peptides 1–2 (SI, Tables S2 and S4) proved to be very similar to those observed in random coil peptides (Δδ < 0.1 ppm),²¹ while proton resonances of state II of peptides 1–2 and of the unique signal system of peptide 4 (SI, Tables S3, S5, and S6) were significantly different (|Δδ| > 0.1 ppm for all the H_α-resonances), indicating a marked conformational stability of this state. The NOESY spectra of peptides 1, 2, and 4 are shown in SI, Figures S5–S7, respectively. Considering the state I of peptides 1–2, only intra-residue or sequential NOEs could be unambiguously assigned indicating random conformations associated with this isomer as indicated also by the chemical shift analysis. In contrast, several NMR parameters indicate that the cis Xaa⁶-Pro⁷ state of all the peptides is highly structured. In fact, ³J_HN-Hα coupling constants (SI, Tables S3, S5 and S6), H_α CSI values²¹ (SI, Figure S8), and some NOE signals (SI, Table S7) point to a β-hairpin structure centered on the Xaa⁶-Pro⁷ β-turn. Diagnostic medium range NOE interactions d_αN(i, i + 2) 6–8 observed in the NOESY spectra are typical of a β-turn structure.

This result was enforced by the registration of low value of the temperature coefficient for NH resonance of residue 8 (−Δδ/ΔT < 3.0 ppb/K). Large values of ³J_HN-Hα coupling constants (³J_HN-Hα > 8 Hz) of residues 5 and 8–9 and downfield shift of α-proton resonances of residues 4–5 and 8–10 flanking the β-turn together with long-range NOE’s between H_α of Nle⁴ and H_α of Trp¹⁰, H_α of Nle⁴, and H_N of Lys¹¹ point to a short antiparallel β-sheet structure along those residues. Different NOEs connected the Nle⁴ side chain with both DPhe⁸/DNal(2′)⁸ and Trp¹⁰ aromatic moieties pointing to spatial closeness of these side chains. NMR-derived constraints registered for the analyzed peptides (SI, Table S7) were used as the input data for a simulated annealing structure calculation. Only state II was considered for structure calculation of peptides 1 and 2. For each peptide, 10 calculated structures satisfying the NMR-derived constraints (violations smaller than 0.20 Å) were chosen (Figure 1a–c). All peptides show a distorted type VIa1 β-turn structure flanked by a short β-sheet encompassing residue 4–5 and 9–10 (backbone rmsd values are 0.41, 0.34, and 0.35 Å, for peptides 1, 2, and 4, respectively). Considering the side chains orientation, DNal(2′)⁸ showed a large preference for g⁺ rotamer of χ1 angle, thus closely interacting with Pro⁶ in peptide 2 (Figure 1b) or Trp⁶ in peptide 4 (Figure 1c). Moreover, the indole moiety of Trp⁶ of peptide 4 strongly interacts with the Pro⁷ side chain. These interactions give rise to the observed stability of Xaa-Pro⁷ cis amide bond. A consequence of such close proximity is the ring current effect on α-proton of residue 6 (Δδ H_α = −0.61, and −0.71 ppm for 2, and 4, respectively, compared to random coil peptides).²¹ Furthermore, in peptide 4, a dramatic upfield shift is observed for α- and β-protons of Pro⁷, shielded by Trp⁶ indole moiety (Δδ ~ −2 ppm), and for some protons of Trp⁶ indole itself (Δδ H_ε3 = −1.39 ppm, H_ζ3 = −0.67 ppm), shielded by DNal(2′)⁸ naphthyl moiety. In contrast, the DPhe⁸ side chain orientation of peptide 1 is less defined. In most cases (9/10), χ1 angle has a trans orientation placing the DPhe⁸ phenyl far from Pro⁶ (Figure 1a).

Figure 1.

Superposition of the 10 lowest energy conformers of peptide 1 (a), 2 (b), and 4 (c). Structure models were superimposed using the backbone heavy atoms. Heavy atoms have different colors (carbon, green; nitrogen, blue; oxygen, red; sulfur, yellow). Many hydrogen atoms are hidden for a better view.

The peptides surface is amphipathic. As shown in Figure 1, considering the pseudoplane determined by the backbone atoms, the hydrophobic residues Nle⁴, Pro⁶/Trp⁶, DPhe⁸/DNal(2′)⁸, and Trp¹⁰ lie on one side (right) and the positively charged residue Arg⁹ lies on the other side. Such an amphipathic arrangement was already observed for the NMR structure of parent peptides MT-II and SHU-9119.¹⁶ For each peptide, ensembles of 10 NMR structures were submitted to the PDB (PDB ID: 2N7N, 1; 2N7O, 2; 2N7T, 4).

hMC3R and hMC4R Models and Docking

Three-dimensional models of hMC3R and hMC4R were generated based on the structure of other GPCR’s, using the I-TASSER server.²² Five models of both receptors were generated. I-TASSER output also contained top ranks of templates used for the structure prediction. The top template used is the high-resolution crystal structure of a human A2A adenosine receptor (PDB ID: 3EML).²³ Statistical analyses on the obtained models are reported in SI, Table S8. Interestingly, the selected models maintain the molecular signatures which feature class A GPCRs,²⁴ in fact, in both models are preserved 24 out of the 24 inter-TM contacts of the consensus network found in the GPCR structures. Peptide 2 NMR lowest energy structure was docked with both hMC3R and hMC4R models. Docking procedures using the program AUTODOCK²⁵ clustered 100 poses in three clusters (80/100 poses in the first cluster) for the complexes 2/hMC3R and 100 poses in 11 clusters for 2/hMC4R (12/100 poses in the first cluster). Statistics and energy terms are reported in SI, Table S9.

Clearly, the high number of clusters and their high energy (SI, Table S9) indicate that docking did not provide a suitable 2/hMC4R complex in accordance with the binding data. In contrast, a high populated low energy cluster was obtained for 2/hMC3R complex, and the best scored pose is shown in Figure 2. The predicted binding site is placed among TM3/TM7, EL1, and EL3 (Figure 2a). Main interactions between the peptide and hMC3R are shown in Figure 2b. As shown, most of the peptide interactions are established with side chains of residues belonging to EL1 and EL3, whose sequences are less conserved between hMC3R and hMC4R compared to the TM’s regions, again in accordance with the observed selectivity. Moreover, many interactions are observed between the naphthyl group of DNal(2′)⁸ and the receptor. Hence, its replacement with a smaller phenyl group in peptide 1 as well as the different orientation of the same phenyl compared to the naphthyl can tentatively explain the observed inactivity of the DPhe⁸ containing peptide 1 (and 3).

Figure 2.

(a) hMC3R model complexed with peptide 2. Peptide 2 heavy atoms (carbon, green; nitrogen, blue; oxygen, red; sulfur, yellow). Receptor backbone is represented in gray ribbon. (b) Peptide 2 within the binding pocket of hMC3R. Hydrogen bonds are represented with dashed lines.

CONCLUSION

In conclusion, we have successfully developed two novel selective hMC3R agonists. These compounds will represent suitable pharmacological tools for elucidating the receptor functions and can be considered as lead compounds for the design of novel hMC3R ligands.

EXPERIMENTAL SECTION

Materials

N_α-Fmoc-protected amino acids, HBTU, and HOBt were purchased from Inbios (Naples, Italy). Rink amide resin was purchased from GL Biochem (Shanghai, China).

99.9% ²H₂O were obtained from Aldrich (Milwaukee, USA), 98% DPC-d₃₈ was obtained from Cambridge Isotope Laboratories, Inc. (Andover, USA), and [(2,2,3,3-tetradeuterio-3-(trimethylsilanyl)]-propionic acid (TSP) was obtained from MSD Isotopes (Montreal, Canada).

Synthesis

Synthesis of peptides 1–4 was performed by standard solid-phase peptide synthesis (SPPS) by using a Fmoc/tBu orthogonal strategy on a Rink amide resin as solid support.²⁶ Further couplings were carried out with standard in situ activating reagents, such as uronium salts (HBTU), in the presence of a tertiary base (DIPEA), to generate HOBt esters. Lactam cyclization was performed on the resin after removal of the allyl-derived protecting groups employing the procedure described by Thieriet et al.^27,28 All compounds were purified by RP-HPLC and verified by analytical UHPLC (SI). The physicochemical properties and purity (>95%) of these peptides were assessed by LC-MS and HRMS (SI, Table S1, Figures S1–S4).

Binding Assays

Competition binding experiments were performed on whole cells. Stably transfected HEK293 cell lines having the individual hMCRs^29,30 were seeded on 96-well plate 48 h before the assay and grown to 100000 cells/well. For the assay, the medium was removed and cells were washed twice with a freshly prepared binding buffer containing 100% minimum essential medium with Earle’s salt (MEM, GIBCO), 25 mM HEPES (pH 7.4), 0.2% bovine serum albumin, 1 mM 1,10-phenanthrolone, 0.5 mg/L leupeptin, and 200 mg/L bacitracin. Cells were then incubated with different concentrations of unlabeled peptides and ¹²⁵I-labeled [Nle⁴,DPhe⁷]-α-MSH (PerkinElmer Life Science, 100000 cpm/well, 0.1386 nM) for 40 min at 37 °C. The medium was subsequently removed, and each well was washed twice with the binding buffer. The cells were lysed by the addition of 250 µL of 0.1 mM NaOH and 250 µL of 1% Triton X-100. The lysed cells were transferred to the 12 mm × 75 mm glass tubes, and the radioactivity was measured using a Wallac 1470 WIZARD γ counter. Data were analyzed using Graphpad Prism 3.1 (Graphpad Software, San Diego, CA).

Adenylate Cyclase Assays

HEK 293 cells stably transfected with individual human melanocortin receptors^29,30 were grown to confluence in MEM medium (GIBCO) containing 10% fetal bovine serum, 100 units/mL penicillin and streptomycin, and 1 mM sodium pyruvate. The cells were seeded on 96-well plates 48 h before assay and grown to 100000 cells/well. For the assay, the medium was removed and cells were rinsed with 1 mL of MEM buffer (GIBCO) or with Earle’s Balanced Salt Solution (EBSS, GIBCO). An aliquot (0.4 mL) of the Earle’s Balanced Salt Solution was placed in each well along with isobutylmethylxanthine (IBMX, 5 µL, 0.5 mM) for 1 min at 37 °C. Next, various concentrations of melanotropins (0.1 mL) were added and the cells were incubated for 3 min at 37 °C. The reaction was stopped by aspirating the buffer and adding ice-cold Tris/EDTA buffer to each well (0.15 mL). After the cells were dislodged with the help of trypsin, the cells were transferred to polypropylene micro-centrifuge tubes, capped, and placed in a boiling water bath for 15 min. The cell lysate was then centrifuged for 2 min (6500 rpm), and 50 µL of the supernatant was aliquoted into a clean Eppendorf tube. The total cAMP content was measured by competitive binding assay according to the assay kit instructions (TRK 432, Amersham Corp., Arlington Heights, IL

Data Analysis

IC₅₀ and EC₅₀ values represent the mean of duplicate experiments performed in triplicate. IC₅₀ and EC₅₀ estimates and their associated standard errors were determined by fitting the data using a nonlinear least-squares analysis, with the help of Graphpad Prism 3.1 (Graphpad Software, San Diego, CA).

NMR Spectroscopy

The samples for NMR spectroscopy were prepared by dissolving the appropriate amount of peptides in 0.27 mL of ¹H₂O (pH 5.5) AND 0.03 mL of ²H₂O to obtain a concentration 1–2 mM of peptide. For the sample in micelle solution, DPC-d₃₈ was also added to a concentration of 200 mM. NMR spectra were recorded on a Varian INOVA 700 MHz spectrometer equipped with a z-gradient 5 mm triple-resonance probe head. 1D and 2D NMR spectra were recorded and processed as described in the SI.

Structural Determinations

The NOE-based distance restraints, obtained in DPC solution, were obtained from NOESY spectra collected with a mixing time of 100 ms. The NOE cross peaks were integrated with the XEASY program and were converted into upper distance bounds using the CALIBA program incorporated into the program package DYANA.³¹ Only NOE derived constraints (SI, Table S7) were considered in the annealing procedures. An ensemble of 100 structures was generated with the simulated annealing calculations followed by successive steps of restrained and unrestrained energy minimization using the Discover algorithm (Accelrys, San Diego, CA) as previously described.¹⁶ From the produced 100 conformation, 10 structures were chosen whose interproton distances best fitted NOE derived distances.

Receptor Models and Docking

Three-dimensional structure predictions of hMC3R and hMC4R were generated by I-TASSER server for protein structure and function prediction, which is based on a threading alignment algorithm.²² The best scored model for each receptor was used for docking studies (SI, Table S8). The initial poses for the hMCR–peptide 2 complex are generated by docking the lowest energy conformers of peptide 2 obtained by NMR to the hMC3R or hMC4R model using the program AUTODOCK 4.0.²⁵ Only the side chain of Arg⁹ of peptide 2 was considered flexible in the docking procedure. Statistics are reported in SI, Table S9. Refinement of lowest energy pose of hMC3R–2 complex was achieved by in vacuo energy minimization with the Discover algorithm using the steepest descent and conjugate gradient methods until a RMSD of 0.05 kcal/mol per Å was reached. The backbone atoms of the TM and IL domains of the hMC3R were held in their position; the ligand and EL’s were free to relax.

ABBREVIATIONS USED

hMCR

human melanocortin receptor

cAMP

cyclic adenosine monophosphate

GPCR

G-protein-coupled receptor

MSH

melanocyte stimulating hormones

ACTH

adrenocorticotropic hormone

POMC

proopiomelanocortin

AGRP

agouti-related protein

ASIP

agouti signaling protein

DPC

dodecyl phosphocholine

NMR

nuclear magnetic resonance

DQF-COSY

double quantum filtered correlated spectroscopy

TOCSY

total correlated spectroscopy

NOESY

nuclear Overhauser enhancement spectroscopy

NOE

nuclear Overhauser effect

TSP

3-(trimethylsilanyl)propionic acid

IL

intracellular loop

EL

extracellular loop

TM

trans-membrane domain

Nal(2′)

(2′)-naphthylalanine

Fmoc

9-fluorenylmethoxycarbonyl

HBTU

N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uroniumhexa-fluorophosphate

HOBt

1-hydroxybenzotriazole

DIPEA

N,N-diisopropylethylamine

All

allyl

Aloc

allyloxycarbonyl

MALDI-TOF

matrix-assisted laser desorption ionization/time-of-flight mass spectrometry

RP-HPLC

reversed-phase high performance liquid chromatography

SPPS

solid-phase peptide synthesis