Scholarly article on topic 'A hydrophobic site on the GLP-1 receptor extracellular domain orients the peptide ligand for signal transduction'

A hydrophobic site on the GLP-1 receptor extracellular domain orients the peptide ligand for signal transduction Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Molecular Metabolism
OECD Field of science
Keywords
{GLP-1 / Exendin-4 / Glucagon / GPCR / "Receptor activation"}

Abstract of research paper on Biological sciences, author of scientific article — James T. Patterson, Pengyun Li, Jonathan W. Day, Vasily M. Gelfanov, Richard D. DiMarchi

Abstract Structure–function studies have analyzed substitutions within the glucagon-like peptide-1 (GLP-1) sequence that increase resistance to proteolysis, however, the investigation into how such substitutions alter interactions at the GLP-1 receptor (GLP-1R) has captured less attention. This work describes our efforts at identifying relevant interactions between peptide ligands and the GLP-1R extracellular domain that contribute to the positioning of the peptide N-terminus for receptor activation. Alanine substitutions at hydrophilic (Glu127⁎ and Glu128⁎) and hydrophobic (Leu32⁎) GLP-1R residues were previously shown to differentially interact with GLP-1 and exendin-4. We examined if these receptor residues influence the activity of GLP-1- and exendin-4-based peptides containing either alanine or glycine at position 2. Additionally, a series of glucagon-based peptides were studied to determine how the central to C-terminal region affects activity. Our results suggest that peptide binding to the GLP-1R is largely driven by hydrophobic interactions with the extracellular domain that orient the N-terminus for activation.

Academic research paper on topic "A hydrophobic site on the GLP-1 receptor extracellular domain orients the peptide ligand for signal transduction"

A hydrophobic site on the GLP-1 receptor (g)™,,

extracellular domain orients the peptide ligand for signal transduction

James T. Patterson*'1, Pengyun Li, Jonathan W. Day2, Vasily M. Gelfanov, Richard D. DiMarchi

ABSTRACT

Structure-function studies have analyzed substitutions within the glucagon-like peptide-1 (GLP-1) sequence that increase resistance to proteolysis, however, the investigation into how such substitutions alter interactions at the GLP-1 receptor (GLP-1R) has captured less attention. This work describes our efforts at identifying relevant interactions between peptide ligands and the GLP-1R extracellular domain that contribute to the positioning of the peptide N-terminus for receptor activation. Alanine substitutions at hydrophilic (Glu127* and Glu128*) and hydrophobic (Leu32*) GLP-1R residues were previously shown to differentially interact with GLP-1 and exendin-4. We examined if these receptor residues influence the activity of GLP-1- and exendin-4-based peptides containing either alanine or glycine at position 2. Additionally, a series of glucagon-based peptides were studied to determine how the central to C-terminal region affects activity. Our results suggest that peptide binding to the GLP-1R is largely driven by hydrophobic interactions with the extracellular domain that orient the N-terminus for activation.

& 2013 Elsevier GmbH. All rights reserved.

Keywords GLP-1; Exendin-4; Glucagon; GPCR; Receptor activation

1. INTRODUCTION

Glucagon-like peptide-1 (GLP-1) is a hormone produced by intestinal L-cells and released upon nutrient ingestion to act on its G proteincoupled receptor (GPCR) to stimulate insulin secretion from pancreatic b-cells. This glucose-dependent process is known as the incretin effect [1,2]. GLP-1 binds the GLP-1 receptor (GLP-1R) causing a conforma-tional change in the GPCR that stimulates cyclic adenosine monophosphate (cAMP) synthesis by adenylyl cyclase to enable subsequent downstream signaling [3]. The hormone further promotes glucose homeostasis by suppressing glucagon release, decreasing gastric motility, and increasing b-cell mass [4]. Treatment with GLP-1 has also been shown to reduce body weight and enhance fat metabolism thus providing several clinical benefits for treatment of patients with type 2 diabetes mellitus [5]. Nonetheless, GLP-1 is proteolytically inactivated by dipeptidyl peptidase-IV (DPP-IV) within minutes which has led to the therapeutic application of peptides, such as exendin-4 (Ex-4), that have prolonged in vivo action [6]. The extended pharmacokinetic profile of Ex-4 relative to GLP-1 is often associated with the Ala (GLP-1) to Gly (Ex-4) substitution at the second amino acid position as this change decreases susceptibility to DPP-IV cleavage (Figure 1) [7,8]. However, GLP-1 has proven to be much more sensitive to glycine substitution at this site impeding the effective use of this substitution in GLP-1-based analogs [9]. Binding and activation

Department of Chemistry, Indiana University, Bloomington, IN, USA

1 Current address: Departments of Molecular Biology and Chemistry, The Scripps Research Institute, La Jolla, CA, USA.

2Current address: Eli Lilly and Company, Indianapolis, IN, USA.

"Corresponding author. Tel.: +1 812 320 0135. Email: jtpatter@scripps.edu (J.T. Patterson)

at the GLP-1R is highly dependent on the orientation of His1 with small perturbations causing significant losses in potency [9-12]. Given the structural and biophysical data available, the source of the differential activities was hypothesized to be derived from the Gly16 (GLP-1) for Glu16 (Ex-4) substitution that stabilizes the a-helical peptide structure [13,14]. Indeed, incorporation of Glu16 to GLP-1 as well as a,a-dialkyl substitution and backbone lactamization has confirmed the importance of this helix-promoting site [15-18]. Nevertheless, additional residues within the central to C-terminal Ex-4 sequence, specifically Leu21 and Glu24, also provided considerable functional contributions to Gly2 acceptance as was determined by substitution of these residues within the GLP-1 Glu16 backbone [18].

We have explored the structural elements required to provide glucagon-related peptides tolerance to N-terminal substitution by examining a series of GLP-1R extracellular domain (ECD) mutants. Ex-4's a-helical character has appeared to contribute to ECD recognition as this hormone can bind the isolated ECD with much high affinity than GLP-1 [19,20]. Several hydrophilic (Glu127* and Glu128*) and hydrophobic (Leu32*) contacts were also observed between the peptide and ECD in the Ex-4 (9-39)a ligand-bound structure that were absent in the ligand-bound GLP-1 structure [21,22]. We reasoned that such interactions with the ECD could help align the peptide's N-terminus within the juxtamembrane region of the receptor for activation. To examine if contacts between the peptide and these ECD residues participate in the

Received November 7, 2012 • Revsion received December 26, 2012 • Accepted January 3, 2013 • Available online 16 January 2013 http://dx.doi.org/10.10167j.molmet.2013.01.003

GLP-1 Exendin-4 GLP-1 E16L21E24

Glucagon E16 Glucagon E16K20 Chimera 2 Chimera 2X

Compound 2

HAEGTFTSDVSSYLEGQAAKEFIAWLVKGR-NH2

hgegtftsdlskqmeeeavrlfiewlknggpssgappps-nh2 haegtftsdvssyleeqaaklfiewlvkgr-nh2

hsqgtftsdyskylderraqdfvqwlmnt-nh2 HSQGTFTSDYSKYLDERRAKDFVQWLMNT-NH2 HSQGTFTSDYSKYLDEEAAKEFIAWLMNT-NH2

hsqgtftsdyskyldeeavrlfiewlmnt-nh2

Figure 1: Ligands examined at mutant receptors. Primary sequences of C-terminally amidated GLP-1 (red),Ex-4 (blue), glucagon (purple), and hybri d pepti des are shown.The C-termi nal extension of Ex-4 (Cex) i s underli ned, but peptides examined herein truncated this extension.Top, conserved residues between GLP-1 and Ex-4 are black. Middle, residues conserved between GLP-1,Ex-4,and glucagon are shown in black.Bottom, structure of the allosteric small molecule Compound 2.(For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

ability to tolerate Gly2 substitution, we have compared a series of GLP-1, Ex-4, and related hybrid peptides at the hydrophilic GLP-1R mutants E127A* and E128A* as well as the hydrophobic GLP-1R mutant L32A*.

2. MATERIALS AND METHODS

2.1. Peptide synthesis

Peptides were generated by solid-phase peptide synthesis using the in situ neutralization method for ferf-butyloxycarbonyl (Boc) chemistry. Peptide synthesis was performed using 0.2 mmol 4-methylbenzhydrylamine resin (Midwest Biotech) on a modified Applied Biosystems 430A synthesizer. Amino acids (Midwest Biotech) were side-chain protected with the following groups: Arg(Tos), Asn(Xan), Asp(OcHex), Glu(OcHex), His(BOM), Lys(2-Cl-Z), Ser(Bzl), Thr(Bzl), Trp(CHO), and Tyr(Br-Z). Activation of amino acids (2 mmol) was performed with 0.5 M 3-(diethoxy-phosphoryloxy)-3H-benzo[d| [1,2,3] triazin-4-one in dimethylformamide and N,N-diisopropylethylamine (4:1 v/v). Cleavage of the peptides from the solid support used HF/p-cresol, 95:5 v/v, for 1 h at 0 °C followed by in vacuo evacuation of the HF. Peptides were precipitated in diethyl ether and collected using a 50 mm Teflon filter funnel. Acetic acid (10%) was used to solubilize the cleaved peptides which were then lyophilized and stored at 4 °C until purification.

2.2. Peptide purification

Reversed-phase HPLC (RP-HPLC) was used for peptide purification. A C18 stationary phase (Vydac 218TP, 22 x 250 mm2, 10 mm) was employed with a linear acetonitrile gradient in 0.1% trifluoroacetic acid during the preparative RP-HPLC purification. Analytical analysis was performed on peak fractions by employing RP-HPLC with a C8 column (Zorbax 300SB, 4.6 x 50 mm2, 3.5 mm). Peptide identity and purity was assessed by analytical RP-HPLC and electrospray ionization- or matrix-assisted laser desorption/ionization-mass spectrometry. All peptides were found to have the correct molecular weight and were approximately 95% pure. Purified peptides were lyophilized and stored at 4 °C.

2.3. GLP-1 receptor mutant cloning

The human GLP-1R cDNA (Open Biosystems) was subcloned into the mammalian expression vector pcDNA3.1(+) (Invitrogen). Mutant receptors (L32A*, E127A*, and E128A*) were produced by a modified megaprimer mutagenesis method using a single flanking reverse primer and mutagenic forward primers. Plasmid DNA was then isolated using the Plasmid Midi Kit (QIAGEN). Mutations were confirmed using the ABI Big Dye version 3.1 kit and sequenced using an ABI3730 DNA analyzer (Applied Biosystems).

2.4. Homogenous time-resolved fluorescence assays Human embryonic kidney (HEK) 293 cells were maintained at 37 °C and 5% CO2 in Dulbecco-modified Eagle's medium (Invitrogen) supplemented with 10% bovine growth serum (HyClone), 10 mg/mL gentamicin (BioWhit-taker), and 10 mM HEPES (HyClone). Cells were plated in 10-cm tissue culture dishes (2.5 x 106 cells/dish) in the absence of antibiotics and transfected after 24 h with 6 mg DNA using the FuGENETM 6 transfection reagent (Roche). Surface expression of the mutant receptors has been reported to be similar to wild-type [22], and no significant variations in mutant receptor expression were noted using our assay system. Cells were trypsinized 24 h after transfection and resuspended at 0.5 x 106 cells/ml, then 40 ml/well was transferred to 96 half-well plates (BD Biosciences). Serial dilutions of peptides (10 ml) were added to plates containing cell suspensions and shaken for 30 min at room temperature. Following incubation, 25 ml of diluted cAMP-XL665 (CIS bio international) was added to all the wells followed by 25 ml anti-cAMP cryptate conjugate (CIS bio international). Assay plates were then gently shaken (500 rpm) for an additional hour and time-resolved fluorescence output was read at 665 nm and 590 nm on an Envision plate reader (Perkin-Elmer). The half-maximal inhibitory concentration (IC50) values corresponding to the inhibition of labeled cAMP binding to antibody by cAMP synthesized in response to ligand activation were calculated using Origin software (OriginLab). Assays were performed in triplicate and each concentration point was tested in duplicate. IC50 values from individual experiments were normalized to the averaged IC50 value of GLP-1 at wild type receptor.

3. RESULTS

3.1. Peptide potency at the GLP-1R and influence of the second amino acid

We and others have previously reported that GLP-1 potency was decreased by approximately five-fold upon substitution of Ala2 for the Ex-4 residue Gly2 whereas Ex-4 was equally tolerant to either amino acid [9,18,23]. The sensitivity of the fluorescence resonance energy transfer (FRET)-based assay used for this study was increased compared to our previously described gene reporter assay as we observed a fifteen-fold reduction in GLP-1 (1-30)a (1) potency upon Gly2 (2) substitution (1 IC50=0.017 nM, 2 IC50=0.25 nM) (Figure 2A and Table 1). Ex-4 (1-30)a (3) still tolerated its native Gly2 residue relative to Ala2 (4) with only a very slight 1.4-fold preference for alanine at this site (3 IC50=0.007 nM, 4 IC50=0.005 nM). Nonetheless, the potency of Ex-4 (1-30)a (3) and GLP-1 (1-30)a Glu16Leu21Glu24 (5) maintained their enhanced activity relative to GLP-1 (1-30)a (1) (5 IC50=0.007 nM). Substitution of peptide 5 with Gly2, GLP-1 (1-30)a Gly2Glu16Leu24Glu24 (6), was also well tolerated with only a 1.6fold reduction in potency relative to the alanine-containing peptide (6 IC50=0.011 nM).

3.2. E127A* and E128A* are not critical determinants for GLP-1R activation

Neither GLP-1 (1) or Ex-4 (3) (1-30)a activity was significantly altered at the GLP-1R hydrophilic ECD receptor mutants E127A* and E128A* relative to the native receptor (Figure 2A-C). Receptor activation was reduced no more than 1.5-fold for peptide 1 at either receptor and 1.7-fold for peptide 3 (Table 1). A slight two-fold reduction in potency was observed for GLP-1 (1-30)a Gly2 (2) at these receptors whereas Ex-4 (1-30)a Ala2 (4) was marginally less discriminating. GLP-1 (1-30)a Glu16Leu21Glu24 (5) displayed a comparable response with the hydrophilic E127A* and E128A* mutations as did the GLP-1 and Ex-4 (1-30)a peptides (Figure 2A-C). Regardless, receptor activation was not affected by more than two-fold for

50.000 ■ 45.000 ■ 40.000 ■ 35.000 ■ 30.000 ■ 25.000 ■ 20.000 ■ 15.000 ■ 10.000 ■ 5.000 0

■ GLP-1

• GLP-1 G2

A Ex-4 (1-30)a

T Ex-4 (1-30)a A2

A GLP-1 E16L21E24

4 GLP-1 G2E16L21E24

10"2 10"1 [Peptide] (nM)

50.000 -, 45.000 -40.000 -E 35.000 -¿Ü 30.000 -

CD ■

25.000 -c -

m 20.000 -

15.000 -10.000 -5.000 -0 -10"4

E127A*

■ GLP-1

• GLP-1 G2 A Ex-4 (1-30)a T Ex-4 (1-30)a A2 A GLP-1 E16L21E24

-4 GLP-1 G2E16L21E24

10"2 10"' [Peptide] (nM)

50.000 -, 45.000 -40.000 -35.000 -30.000 -25.000 -20.000 -15.000 -10.000 -5.000 -0 -

GLP-1 GLP-1 G2 Ex-4 (1-30)a Ex-4 (1-30)a A2 GLP-1 E16L21E24 GLP-1 G2E16L21E24

000 000 000 000 000 000 000 000 000 000 0

GLP-1 GLP-1 G2 Ex-4 (1-30)a Ex-4 (1-30)a A2 GLP-1 E16L21E24 GLP-1 G2E16L21E24

[Peptide] (nM) [Peptide] (nM)

Figure 2: GLP-1R-mediated cAMP induction. GLP-1, Ex-4, and GLP-1 Glu16Leu21Glu24 (1-30)a peptides were examined at the wild-type (A), E127A* (B), E128A* (C), and L32A* (D) receptors.

# Peptide GLP-1R (nM) E127A* (nM) E128A* (nM) L32A* (nM)

1 GLP-1 (1-30)a 0.017±0.002 0.021 ± 0.005 (1.2) 0.026 ± 0.004 (1.5) 0.037 ±0.006 (2.2)

2 GLP-1 (1-30)a G2 0.25 7 0.03 0.46 ± 0.09 (1.8) 0.49 ±0.07 (2.0) 2.4 ±0.3 (9.6)

3 Ex-4 (1-30)a 0 . 007 ±0 . 001 0.011 ± 0.002 (1.6) 0.012 ± 0.002 (1.7) 0.044 ±0.009 (6.3)

4 Ex-4 (1-30)a A2 0 . 005 ±0 . 001 0.006 ± 0.001 (1.2) 0.007 ±0.001 (1.4) 0.010 ±0.002 (2.0)

5 GLP-1 (1-30)a E16L21E24 0.007 ±0.001 0.008 ± 0.001 (1.1) 0.008 ± 0.001 (1.1) 0.014 ±0.004 (2.0)

6 GLP-1 (1-30)a G2E16L21E24 0 . 011 ±0 . 003 0.014 ± 0.003 (1.3) 0.017 ± 0.003 (1.5) 0.095 ±0.007 (8.6)

7 Glucagon (1-29)a E16 0.060 ±0.009 0.072 ± 0.005 (1.2) 0.059 ± 0.003 (1.0) 0.43 ±0.09 (7.2)

8 Glucagon (1-29)a E16K20 0.083 ±0.01 0.10 ± 0.02 (1.2) 0.14± 0.02 (1.7) 0.45 ±0.08 (5.4)

9 Chimera 2 (1-29)a 0 . 038 ±0 . 007 0.051 ± 0.003 (1.3) 0.040 ± 0.007 (1.1) 0.076 ±0.01 (2.0)

10 Chimera 2X (1-29)a 0.076 ±0.01 0.084 ± 0.004 (1.1) 0.078 ± 0.008 (1.0) 0.17 ±0.004 (2.2)

11 Compound 2 300 ±10 420 ± 20 (1.4) 570 ±50 (1.9) 450 ±30 (1.5)

Table 1: Summary of IC50 values at the wild-type and mutant GLP-1 receptors (see methods).The fold reduction relative to wild-type Is Indicated In parentheses for the hydrophilic/hydrophobic mutant receptors. Numbering for GLP-1 (7-36)a corresponds to positons (1-30)a for consistency with Ex-4.

any of the peptides examined (Table 1). These data are fairly consistent with the published mutagenesis data, but our results indicate less substantial losses in activity for the E128A* receptor [22].

3.3. Differential activity at the GLP-1R mutant L32A* is dependent on position 2

GLP-1R ECD hydrophobic mutant L32A* conferred differential responses between the GLP-1 (1) and Ex-4 (3) (1-30)a peptides (Figure 2A and D). The potency losses for peptides 1 and 3 were approximately two- and

six-fold respectively at L32A* relative to the wild-type receptor (Table 1). Again, our data are consistent with the previous report with only small differences in the observed potency reductions, likely attributable to differences in our assay systems [22]. The data validate that Ex-4 receptor activation was more substantially reduced than that of GLP-1 at the hydrophobic L32A* receptor mutant (Figure 2A and D). However, GLP-1 (1-30)a Gly2 (2) exhibited an even more significant potency loss (~ 10-fold) than either peptides 1 or 3 at this receptor, and the activity of Ex-4 (1 -30)a (3) was rescued by Ala2 substitution (4) (Table 1). A similar trend was seen

for peptides 5 and 6 with the Ala2-containing peptide being reduced only approximately two-fold relative to wild-type and Gly2 being reduced approximately nine-fold. Hence, sensitivity to the L32A* receptor mutant was dependent on the identity of the second amino acid of the peptide, and not the specific hormone backbone.

3.4. GLP-1R activation by GLP-1R/GCGR co-agonists and Compound 2 To determine if glucagon-based peptides interact with the GLP-1R in a similar manner as the GLP-1 and Ex-4 based peptides, we further explored a set of GLP-1R and glucagon receptor (GCGR) co-agonists at the wild-type and mutant receptors. These co-agonists have already been characterized [15,18], and the performance of these peptides at the native GLP-1R is consistent with our previous analysis (Figure 3A and Table 1). Glucagon (1-29)a Glu16 (7) was slightly enhanced in potency relative to glucagon (1-29)a Glu16Lys20 (8) (7 IC50 = 0.060 nM, 8 IC50 = 0.083 nM). Similarly, the potency of Chimera 2 (1-29)a (9) was two-fold greater than to Chimera 2X (1-29)a (10) (9 IC50 = 0.038 nM, 10 IC50 = 0.076 nM). We also included the allosteric GLP-1R small molecule agonist Compound 2 (6,7-dichloro-2-methyl-sulfonyl-3-ferf-butylaminoquinoxaline) in our analysis to confirm that the ECD mutations were directly participating in orthosteric binding and not altering the conformation of the receptor [24]. The response of Compound 2 was greatly reduced relative to the peptide agonists (11 IC50=300 nM), but the small molecule did maintain full efficacy at the GLP-1R (Figure 3A and Table 1).

MOLECULAR METABOLISM

3.5. GLP-1R/GCGR co-agonists respond similarly as GLP-1R agonists at the E127A* and E128A* ECD mutants

Examination of the glucagon-based co-agonist peptides at the hydrophilic E127A* and E128A* GLP-1R mutants revealed that potency for these peptides was not substantially changed relative to the wild-type receptor (Figure 3A-C). Much like the GLP-1- and Ex-4-based peptides, potency reductions were less than two-fold for all the co-agonists at these receptor mutants (Table 1). This was also the case for the allosteric small molecule Compound 2 (11), albeit the decreased potency of the small molecule relative to the peptide agonists could possibly make its response less sensitive to small perturbations in receptor contacts. The activity of glucagon (1-29)a Glu16Lys20 (8) relative to glucagon (1-29)a Glu16 (7) at E128A* suggests a putative interaction between Lys20 and Glu128* (7=no change, 8 =~2-fold reduction) (Table 1). Also, the response of the other glucagon-based peptides besides peptide 7 and small molecule 11 at E128A* was not reduced relative to the wild-type receptor. The magnitude of the responses for the co-agonists (7-10) and Compound 2 (11) at E127A* and E128A* verify that these hydrophilic residues are not major determinants of receptor activation.

3.6. Differential activity of GLP-1R/GCGR co-agonists at the L32A* ECD mutant is dependent on the peptide backbone

GLP-1R hydrophobic ECD mutant L32A* resulted in more substantial potency losses for the GLP-1R/GCGR co-agonist peptides than did the hydrophilic alanine substitutions (Figure 3 and Table 1). Moreover, glucagon (1-29)a Glu16 (7) and glucagon (1-29)a Glu16Lys20 (8) activation at the L32A*

50.000 45.000 40.000 35.000

0 30.000

1 25.000

^ 20.000 <D

10 15.000 10.000 5.000 0

Glucagon E16 Glucagon E16K20 Chimera 2 Chimera 2X Compound 2

45.000 40.000

.000 .000 .000 .000 .000 0

■ GLP-1

• Glucagon E16 A Glucagon E16K20 T Chimera 2 A Chimera 2X

< Compound 2

[Peptide] (nM)

[Peptide] (nM)

50.000 45.000 -40.000 -35.000 -30.000 -25.000 -20.000 -15.000 -10.000 -5.000 -0 -

■ GLP-1

• Glucagon E16 A Glucagon E16K20 ▼ Chimera 2 A Chimera 2X

4 Compound 2

50.000 45.000 40.000 35.000 30.000 25.000 20.000 15.000 10.000 5.000 0

Glucagon E16 Glucagon E16K20 Chimera 2 Chimera 2X Compound 2

[Peptide] (nM)

10-1 10° 101 [Peptide] (nM)

Figure 3: cAMP induction at GLP-1 receptors. GLP-1R/GCGR co-agonist peptides and the allosteric small molecule Compound 2 were studied at the wild-type (A), E127A* (B), E128A* (C), and L32A* (D) GLP-1 receptors.

Leu32*

Figure 4: Superposition of the GLP-1- and Ex-4 (9-39)a-bound GLP-1R ECD crystal structures. Shown is the structures of GLP-1 (red, 3IOL) and Ex-4 (9-39)a (blue, 3C5T) bound to the GLP-1R ECD. Ex-4 Cex residues were omitted for clarity. Receptor residues that were substituted to alani ne are i ndicated.Leu32* stabi lizes the peptide ligand through hydrophobic interactions to promote the optimal orientation of the N-terminal peptide within the receptor core to stimulate signal transduction.(For i nterpretation of the references to color i n thi s figure legend, the reader is referred to the web version of thi s article.)

receptor relative to the native GLP-1R was reduced more than the Chimera 2 (9) and Chimera 2X (10) peptides (Figure 3A and D). Responses of co-agonists 7 and 8 resembled those of the Gly2-containing GLP-1R agonists at L32A*, but co-agonists 9 and 10 as well as Compound 2 (11) were more tolerant of this mutation much like the Ala2-containing GLP-1R agonists (Table 1). The Ser2 substitution was, however, previously demonstrated to be less favorable than Ala2 at the GLP-1R for the glucagon-based peptides [18]. Nevertheless, the response of the GLP-1R/GCGR peptides at L32A* was not dependent on the identity of the second amino acid as all these analogs contain serine at this position (Figure 1).

4. DISCUSSION

Our receptor mutagenesis study has demonstrated that GLP-1- and Ex-4-based peptides containing Gly2 were similarly dependent on the hydro-phobic GLP-1R residue Leu32* for receptor activation, and the hydrophilic residues Glu127* and Glu128* did not make significant contributions (Figure 4). Alanine substitution at position 2 of these peptides was capable of rescuing activity at the L32A* mutant receptor. A glycine residue at position 2 is expected to provide more conformational flexibility making the stable positioning of N-terminal residues for activation more reliant on contacts between the central peptide and the ECD. Alanine substitution analysis for GLP-1 has shown the significance of hydrophobic residues Phe22 and Ile23, which bind near Leu32* pocket, for receptor binding and activation [10,21]. Collectively, these results advocate for a reliance on hydrophobic contacts with the GLP-1R ECD to orient the peptide N-terminus for receptor activation.

Related family B members such as secretin [25] and pituitary adenylate cyclase-activating peptide [26] also require the proper positioning of their N-terminal peptides to efficiently facilitate signal transduction. Secretin activity was greatly reduced when Leu19 was substituted with alanine resulting in destabilization of the peptide backbone and disruption of a hydrophobic interaction with Val13* of the secretin receptor ECD [27]. Molecular dynamics simulations revealed that this alanine-substituted secretin analog was much less restricted when bound to the receptor. Our own data draws striking correlations with this work in that interactions between the central to C-terminal GLP-1 sequence with GLP-1R residue Leu32* was found to be important for the acceptance of glycine at position 2. Interestingly, we have demonstrated that the conversion of N-truncated GLP-1 to an antagonist (Jant-4) required the introduction of a valine at position 19 leading us to postulate the prospect of a contact with GLP-1R ECD

residue Leu32* [28]. Hence, hydrophobic interactions between central to C-terminal peptide residues within the ECD appear to be strongly correlated with proper receptor binding.

Substitutions that alter contacts within the juxtamembrane domain could also impart selectivity between homologous receptors much like we observed for the preference of GLP-1 or Ex-4 sequence within positions 17-24 of the glucagon-based co-agonists at the L32A* mutant receptor. Given that each of these co-agonists shares a relatively common backbone as well as contains Ser2, this result demonstrates that the central to C-terminal peptide sequence is capable of conferring differential responses as well. Previous analysis of the co-agonists revealed the importance of these residues in glucagon-based sequences for GLP-1R activation [15,29], so the presence of glucagon sequence in peptides 7 and 8 may further destabilize receptor binding relative to peptides 9 and 10 as fewer direct compensatory interactions are possible. The residues that confer these respective activities are also located within the same region that confers the different responses to Gly2 substitution at the native receptor for GLP-1R agonists. Furthermore, glucagon was recently demonstrated to be reliant on residues Phe22, Val23, Leu26, and Met27 for binding to the glucagon ECD [30] perhaps indicating the conservation of such hydrophobic interactions between receptors. It may be surmised that family B GPCRs share a common mechanistic basis for receptor activation as GLP-1 but depend on differential structural properties and receptor contacts for selectivity [31]. There appears to be a consensus amongst family B members that the central to C-terminal region of the hormone interacts with the ECD, and the N-terminal peptide extends into the receptor core to trigger a conforma-tional response. Chimeric ligand/receptor pairs for calcitonin/parathyr-oid hormone and glucagon/GLP-1 have confirmed that specific transmembrane segments in the juxtamembrane region substantially contribute to signal transduction [32,33]. Our observations suggest that GLP-1-related peptides communicate structural information throughout the a-helix in response to specific receptor contacts that functionally connect the central peptide region and the N-terminus to regulate receptor activation.

ACKNOWLEDGMENTS

We thank David Smiley and Jay Levy for their technical assistance. Cassandra Koole and Patrick Sexton of Monash University graciously provided Compound 2. Partial funding was provided by Indiana University and Marcadia Biotech.

CONFLICT OF INTEREST

None declared.

REFERENCES

[1] Elrick, H., Stimmler, L., Hlad, C.J., Jr., and Arai, Y., 1964. Plasma insulin response to oral and intravenous glucose administration. Journal of Clinical Endocrinology and Metabolism 24:1076-1082.

[2] Perley, M.J., and Kipnis, D.M., 1967. Plasma insulin responses to oral and intravenous glucose: studies in normal and diabetic sujbjects. Journal of Clinical Investigation 46:1954-1962.

[3] Takhar, S., Gyomorey, S., Su, R.C., Mathi, S.K., Li, X., and Wheeler, M.B., 1996. The third cytoplasmic domain of the GLP-1[7-36 amide] receptor is required for coupling to the adenylyl cyclase system. Endocrinology 137:2175-2178.

[4] Drucker, D.J., 2002. Biological actions and therapeutic potential of the glucagon-like peptides. Gastroenterology 122:531-544.

[5] Nogueiras, R., Perez-Tilve, D., Veyrat-Durebex, C., Morgan, D.A., Varela, L., Haynes, W.G., et al., 2009. Direct control of peripheral lipid deposition by CNS GLP-1 receptor signaling is mediated by the sympathetic nervous system and blunted in diet-induced obesity. Journal of Neuroscience 29:5916-5925.

[6] Drucker, D.J., and Nauck, M.A., 2006. The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet 368:1696-1705.

[7] Young, A.A., Gedulin, B.R., Bhavsar, S., Bodkin, N., Jodka, C., Hansen, B., et al., 1999. Glucose-lowering and insulin-sensitizing actions of exendin-4: studies in obese diabetic (ob/ob, db/db) mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys (Macaca mulatta). Diabetes 48:1026-1034.

[8] Kieffer, T.J., McIntosh, C.H., and Pederson, R.A., 1995. Degradation of glucose-dependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and in vivo by dipeptidyl peptidase IV. Endocrinology 136:3585-3596.

[9] Deacon, C.F., Knudsen, L.B., Madsen, K., Wiberg, F.C., Jacobsen, O., and Holst, J.J., 1998. Dipeptidyl peptidase IV resistant analogues of glucagon-like peptide-1 which have extended metabolic stability and improved biological activity. Diabetologia 41:271-278.

[10] Adelhorst, K., Hedegaard, B.B., Knudsen, L.B., and Kirk, O., 1994. Structure-activity studies of glucagon-like peptide-1. Journal of Biological Chemistry 269:6275-6278.

[11] Hareter, A., Hoffmann, E., Bode, H.P., Goke, B., and Goke, R., 1997. The positive charge of the imidazole side chain of histidine7 is crucial for GLP-1 action. Endocrine Journal 44:701-705.

[12] Lopez de Maturana, R., and Donnelly, D., 2002. The glucagon-like peptide-1 receptor binding site for the N-terminus of GLP-1 requires polarity at Asp198 rather than negative charge. FEBS Letters 530:244-248.

[13] Andersen, N.H., Brodsky, Y., Neidigh, J.W., and Prickett, K.S., 2002. Medium-dependence of the secondary structure of exendin-4 and glucagon-like-peptide-1. Bioorganic and Medicinal Chemistry 10:79-85.

[14] Neidigh, J.W., Fesinmeyer, R.M., Prickett, K.S., and Andersen, N.H., 2001. Exendin-4 and glucagon-like-peptide-1: NMR structural comparisons in the solution and micelle-associated states. Biochemistry 40:13188-13200.

[15] Day, J.W., Ottaway, N., Patterson, J.T., Gelfanov, V., Smiley, D., Gidda, J., et al., 2009. A new glucagon and GLP-1 co-agonist eliminates obesity in rodents. Nature Chemical Biology 5:749-757.

[16] Miranda, L.P., Winters, K.A., Gegg, C.V., Patel, A., Aral, J., Long, J., et al., 2008. Design and synthesis of conformationally constrained glucagon-like peptide-1 derivatives with increased plasma stability and prolonged in vivo activity. Journal of Medicinal Chemistry 51:2758-2765.

[17] Murage, E.N., Schroeder, J.C., Beinborn, M., and Ahn, J.M., 2008. Search for alpha-helical propensity in the receptor-bound conformation of glucagon-like peptide-1. Bioorganic and Medicinal Chemistry 16:10106-10112.

[18] Patterson, J.T., Day, J.W., Gelfanov, V.M., and DiMarchi, R.D., 2011. Functional association of the N-terminal residues with the central region in glucagon-related peptides. Journal of Peptide Science 17:659-666.

[19] Runge, S., Schimmer, S., Oschmann, J., Schiodt, C.B., Knudsen, S.M., Jeppesen, C.B., et al., 2007. Differential structural properties of GLP-1 and

exendin-4 determine their relative affinity for the GLP-1 receptor N-terminal extracellular domain. Biochemistry 46:5830-5840.

[20] Lopez de Maturana, R., Willshaw, A., Kuntzsch, A., Rudolph, R., and Donnelly, D., 2003. The isolated N-terminal domain of the glucagon-like peptide-1 (GLP-1) receptor binds exendin peptides with much higher affinity than GLP-1. Journal of Biological Chemistry 278:10195-10200.

[21] Runge, S., Thogersen, H., Madsen, K., Lau, J., and Rudolph, R., 2008. Crystal structure of the ligand-bound glucagon-like peptide-1 receptor extracellular domain. Journal of Biological Chemistry 283:11340-11347.

[22] Underwood, C.R., Garibay, P., Knudsen, L.B., Hastrup, S., Peters, G.H., Rudolph, R., et al., 2010. Crystal structure of glucagon-like peptide-1 in complex with the extracellular domain of the glucagon-like peptide-1 receptor. Journal of Biological Chemistry 285:723-730.

[23] Al-Sabah, S., and Donnelly, D., 2003. A model for receptor-peptide binding at the glucagon-like peptide-1 (GLP-1) receptor through the analysis of truncated ligands and receptors. British Journal of Pharmacology 140:339-346.

[24] Knudsen, L.B., Kiel, D., Teng, M., Behrens, C., Bhumralkar, D., Kodra, J.T., et al., 2007. Small-molecule agonists for the glucagon-like peptide 1 receptor. Proceedings of the National Academy of Sciences of the United States of America 104:937-942.

[25] Gespach, C., Bataille, D., Vauclin, N., Moroder, L., Wunsch, E., and Rosselin, G., 1986. Secretin receptor activity in rat gastric glands. Binding studies, cAMP generation and pharmacology. Peptides 7 (1), 155-163.

[26] Bourgault, S., Vaudry, D., Segalas-Milazzo, I., Guilhaudis, L., Couvineau, A., Laburthe, M., et al., 2009. Molecular and conformational determinants of pituitary adenylate cyclase-activating polypeptide (PACAP) for activation of the PAC1 receptor. Journal of Medicinal Chemistry 52:3308-3316.

[27] Dong, M., Le, A., Te, J.A., Pinon, D.I., Bordner, A.J., and Miller, L.J., 2011. Importance of each residue within secretin for receptor binding and biological activity. Biochemistry 50:2983-2993.

[28] Patterson, J.T., Ottaway, N., Gelfanov, V.M., Smiley, D.L., Perez-Tilve, D., Pfluger, P.T., et al., 2011. A novel human-based receptor antagonist of sustained action reveals body weight control by endogenous GLP-1. ACS Chemical Biology 6:135-145.

[29] Day, J.W., Li, P., Patterson, J.T., Chabenne, J., Chabenne, M.D., Gelfanov, V.M., et al., 2011. Charge inversion at position 68 of the glucagon and glucagon-like peptide-1 receptors supports selectivity in hormone action. Journal of Peptide Science 17:218-225.

[30] Koth, C.M., Murray, J.M., Mukund, S., Madjidi, A., Minn, A., Clarke, H.J., et al., 2012. Molecular basis for negative regulation of the glucagon receptor. Proceedings of the National Academy of Sciences of the United States of America 109:14393-14398.

[31] Karnik, S.S., Gogonea, C., Patil, S., Saad, Y., and Takezako, T., 2003. Activation of G-protein-coupled receptors: a common molecular mechanism. Trends in Endocrinology and Metabolism 14:431-437.

[32] Bergwitz, C., Gardella, T.J., Flannery, M.R., Potts, J.T., Jr., Kronenberg, H.M., Goldring, S.R., et al., 1996. Full activation of chimeric receptors by hybrids between parathyroid hormone and calcitonin. Evidence for a common pattern of ligand-receptor interaction. Journal of Biological Chemistry 271: 26469-26472.

[33] Runge, S., Wulff, B.S., Madsen, K., Brauner-Osborne, H., and Knudsen, L.B., 2003. Different domains of the glucagon and glucagon-like peptide-1 receptors provide the critical determinants of ligand selectivity. British Journal of Pharmacology 138:787-794.