Eleven amino acid glucagon-like peptide-1 receptor agonists with antidiabetic activity

Article abstract of DOI:10.1021/jm900752a

Glucagon-like peptide 1 (GLP-1) is a 30 or 31 amino acid peptide hormone that contributes to the physiological regulation of glucose homeostasis and food intake. Herein, we report the discovery of a novel class of 11 amino acidGLP-1 receptor agonists. These peptides consist of a structurally optimized 9- mer, which is closely related to the N-terminal 9 amino acids ofGLP-1, linked to a substituted C-terminal biphenylalanine (BIP) dipeptide. SAR studies resulted in 11-mer GLP-1R agonists with similar in vitro potency to the native 30-mer. Peptides 21 and 22 acutely reduced plasma glucose excursions and increased plasma insulin concentrations in a mouse model of diabetes. These peptides also showed sustained exposures over several hours in mouse and dog models. The described 11-mer GLP-1 receptor agonists represent a new tool in further understanding GLP-1 receptor pharmacology that may lead to novel antidiabetic agents.

Full text of DOI:10.1021/jm900752a

7788 J. Med. Chem. 2009, 52, 7788–7799
DOI: 10.1021/jm900752a
Eleven Amino Acid Glucagon-like Peptide-1 Receptor Agonists with Antidiabetic Activity
Claudio Mapelli,† Sesha I. Natarajan,† Jean-Philippe Meyer,† Margarita M. Bastos,† Michael S. Bernatowicz,† Ving G. Lee,†
Jelka Pluscec,† Douglas J. Riexinger,† Ellen S. Sieber-McMaster,† Keith L. Constantine,† Constance A. Smith-Monroy,#
Rajasree Golla,# Zhengping Ma,# Daniel A. Longhi,# Dan Shi,§ Li Xin,# Joseph R. Taylor,# Barry Koplowitz,z Cecilia L. Chi,z
Ashish Khanna,z Gordon W. Robinson,# Ramakrishna Seethala,# Ildiko A. Antal-Zimanyi,# Robert H. Stoffel,§
Songping Han,# Jean M. Whaley,# Christine S. Huang,z John Krupinski,*,# and William R. Ewing*,‡
§Department of Applied Biotechnologies and ‡Discovery Chemistry and zPreclinical Candidate Optimization and #Diabetes Drug Discovery,
Bristol-Myers Squibb Company Research & Development, Pennington, New Jersey 08534, and †Department of Applied Biotechnologies,
Bristol-Myers Squibb Company Research & Development, Princeton, New Jersey 08543
Received May 29, 2009
Glucagon-like peptide 1 (GLP-1) is a 30 or 31 amino acid peptide hormone that contributes to the
physiological regulation of glucose homeostasis and food intake. Herein, we report the discovery of a
novel class of 11 amino acid GLP-1 receptor agonists. These peptides consist of a structurally optimized 9-
mer, which is closely related to the N-terminal 9 amino acids of GLP-1, linked to a substituted C-terminal
biphenylalanine (BIP) dipeptide. SAR studies resulted in 11-mer GLP-1R agonists with similar in vitro
potency to the native 30-mer. Peptides 21 and 22 acutely reduced plasma glucose excursions and increased
plasma insulin concentrations in a mouse model of diabetes. These peptides also showed sustained
exposures over several hours in mouse and dog models. The described 11-mer GLP-1 receptor agonists
represent a new tool in further understanding GLP-1 receptor pharmacology that may lead to novel
antidiabetic agents.
Introduction
from intestinal L-cells in response to oral food intake potentiates
insulin secretion from pancreatic β-cells as part of the “incretin
effect”.2-4 However, unlike direct administration of insulin,
GLP-1 receptor agonism has a low potential to induce hypogly-
cemia because both GLP-1-dependent stimulation of insulin
release and inhibition of glucagon secretion from the pancreas
are glucose-dependent and do not occur when plasma glucose
decreases below normal concentrations.5 Importantly, GLP-1
can also inhibit both gastric emptying6 and food intake,7 result-
ing in weight loss following chronic administration in animal
models8 and obese, type 2 diabetic patients.9 In animal models,
GLP-1 has also been shown to preserve or increase β-cell mass
through inhibition of apoptosis, stimulation of proliferation, and
possibly, stimulation of β-cell differentiation.10
Continuous coverage with a GLP-1 receptor agonist is
the optimal pharmacokinetic profile, as type 2 diabetic pati-
ents infused with GLP-1 for 24 h showed more beneficial
effects on fasting plasma glucose when compared to patients
infusedfor16h.11 Endogenous GLP-1, however, is short-lived
in vivo, with a half-life of ∼2 min.12 Therefore, in a broad
patient population requiring chronic treatment, the unmodi-
fied GLP-1 peptide is not suitable as a therapeutic agent. The
major degradation pathway of GLP-1 is cleavage of the His-
Ala dipeptide from the N-terminus abolishing agonist activ-
ity, a process mediated by dipeptidyl peptidase IV (DPP4).13
However, DPP4 has other natural substrates in addition to
GLP-1, and the plasma concentrations of GLP-1 achieved
by DPP4 inhibition are limited, resulting in a somewhat
distinct clinical profile. Importantly, DPP4 inhibitors are
weight neutral, while GLP-1 receptor agonists promote
weight loss.14
Type 2 diabetes has been increasing worldwide at an
alarming rate. It has been projected that the number of people
suffering from diabetes will rise to 366 million by 2030, more
than doubling the prevalence of 171 million patients estimated
for the year 2000.1 The prevalence and severity of diabetes has
resulted in an increased need for new therapeutic agents that
have the potential to counter the progressive hyperglycemia,
weight gain, and loss of pancreatic β cell function that typify
the natural history of the disease and ultimately result in
significant morbidity and mortality.
Glucagon-like peptide 1 (GLP-1a) is a gastrointestinal hor-
mone that exists predominantly as a 30 amino acid, C-terminally
amidated peptide (GLP-1(7-36)NH2) or as the equally active,
glycine-extended form, GLP-1(7-37), either of which is a func-
tional agonist of the GLP-1 receptor, a class II GPCR. The
importance of GLP-1 and its receptor in regulating glucose
homeostasis and food intake supports the view that GLP-1
receptor agonists have the potential to improve on the current
standard of care in treating type 2 diabetes. Release of GLP-1
*To whom correspondence should be addressed. For J.K.: phone,
609-818-4977; fax, 609-818-3239; e-mail, john.krupinski@bms.com.
For W.R.E.: phone, 609-818-3483; fax, 609-818-3331; e-mail, william.
ewing@bms.com.
a Abbreviations: GLP-1, glucagon-like peptide-1; GLP-1R, gluca-
gon-like peptide-1 receptor; BIP, biphenylalanine; DPP4, dipeptidyl
peptidase IV; GPCR, G-protein-coupled receptor; cAMP, cyclic ade-
nosine monophosphate or adenosine 30,50-cyclic monophosphate; rmsd,
root-mean-square deviation; IPGTT, intraperitoneal glucose tolerance
test; ANOVA, analysis of variance; Cmax, maximum plasma concentra-
tion; LC-MS/MS, liquid chromatography-tandem mass spectrome-
try; LLQ, lower limit of quantitation.
r
pubs.acs.org/jmc
Published on Web 08/24/2009
2009 American Chemical Society

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7789
Scheme 1. Solid Phase Suzuki Synthesis of a Biphenylalanine Residue in Positions 10 and 11a
a (a) 20% piperidine/DMF; (b) Boc-4-I-Phe-OH, BOP, DIEA in DMF; (c) Fmoc-amino acid/DIC/HOAt, NMP; (d) Pd(0) cat., 25% DIEA/NMP,
85-90 °C, arylboronic acid; (e) TMS-OTf/2,6-lutidine/CH2Cl2 (1:1:3); (f) 20% MeOH/DMF.
Scheme 2. Preparation of Fmoc-(S)-(20-methyl)biphenylal-
anine and Fmoc-(S)-(20-ethyl-40-methoxy)biphenylalaninea
All of the GLP-1 receptor agonists approved or reported to
be in clinical development are based on the amino acid
sequence of GLP-1 or the 39-amino acid sequence of exen-
din-4, a GLP-1R agonist isolated from the saliva of the
gila monster, Heloderma suspectum.15 Synthetic exendin-4
(exenatide, Byetta) is a very potent, DPP4-resistant GLP-1
receptor agonist and is the first agent in this class approved for
the treatment of type 2 diabetes. However, the subcutaneous
formulation of exenatide is administered twice daily16 and
considerable effort has been devoted to enhancing the phar-
macokinetic properties of exenatide and other peptidic or
protein-based GLP-1R agonists in development.17 Although
a significant discovery, the nonpeptidic GLP-1R agonists
disclosed to date are not sufficiently potent to be suitable for
clinical development.18,19
a (a) K2CO3, Pd(PPh3)4, toluene, 80 °C, 2-Me-phenylboronic acid or
2-Et-phenylboronic acid; (b) THF/MeOH (3:1), NaOH; (c) HCl/
CH2Cl2; (d) THF/H2O (4:1), Fmoc-OSu.
Scheme 3. Preparation of 2-Ethyl-4-methoxyphenylboronic
Acid, Routes A and Ba
Chemistry
Amino Acid Synthesis. The described biphenylalanines (2-
amino-3-(biphenyl-4-yl)propanoic acids, BIP) were synthe-
sized on a solid support using Rink resin or by using solution
phase chemistry. In Scheme 1, the process to synthesize
position 10 or 11 biphenylalanines is described. 4-Iodophe-
nylalanine is coupled to Rink resin and is subjected to Suzuki
coupling conditions using either tetrakis(triphenylphos-
pine)palladium(0) or palladium(II) acetate/2-(dicyclohexyl-
phosphino)biphenyl catalyst. In this process, a 10-fold excess
of boronic acid and heating to 80 °C is needed for efficient
coupling. In a similar fashion (Scheme 1), position 10 bi-
phenylalanines could be synthesized via coupling iodophe-
nylalanine to the desired position 11-amino acid on the Rink
resin followed again by Suzuki coupling conditions.
Once identified, optimal position 10 and position 11
biphenylalanines were synthesized individually using solu-
tion phase chemistry following the procedures shown in
Scheme 2. Both procedures start with the conversion of the
phenol of Boc-L-Tyr methyl ester to the corresponding
triflate. In the synthesis of the optimal position 11 amino
acid, the resulting product was coupled under Suzuki con-
ditions with 2-methylphenylboronic acid. The ester of the
resulting product was hydrolyzed, and then the Boc group
a (a) Ph3PCH2Br, n-BuLi, THF; (b) 10%Pd/C, H2, EtOAc; (c) n-
BuLi, B(OBu)3; (d) 1.5 N HCl; (e) CH3I, K2CO3; (f) NBS.
was converted to the Fmoc group in a two step sequence to
yield (S)-2-(Fmoc)-3-(20-methylbiphenyl-4-yl)propanoic acid
(BIP-(20-Me)). Also shown in Scheme 2, an analogous proce-
dure was used to synthesize (S)-2-(Fmoc)-3-(20-ethyl-40-meth-
oxybiphenyl-4-yl)propanoic acid (BIP-(20-Et-40-OMe)), which
was found to be the optimal amino acid at position 10.
The synthesis of 2-ethyl-4-methoxyphenylboronic acid needed
for the preparation of BIP-(20-Et-40-OMe) is shown in
Scheme 3. The desired boronic acid could be conveniently
prepared from either 2-bromo-5-methoxybenzaldehyde or 3-
ethylphenol.
Fmoc-(S)-2-Fluoro-R-methylphenylalanine, used in posi-
tion 6 of the 11-mers, was prepared using an oxazolone
alkylation route, followed by enzymatic resolution of the
resulting racemic amino acid using carboxypeptidase A
(Scheme 4). This amino acid is also commercially available
in chiral form. Fmoc-(S)-2,6-difluoro-R-methylphenylal-
anine was prepared in a similar manner.

7790 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Mapelli et al.
Scheme 4. Preparation of Fmoc-(S)-2-fluoro-R-methylphenyl-
in functional activity.27,28 Among these, the GLP-1 analogue
with Aib was found to be the most stable to DPP4 cleavage
and to have potency similar to native GLP-1. Other modifica-
tionswere found that increased biological half-life, withone of
the most significant identified by the Novo Nordisk group,
showing that γ-glutamyl-mediated acylation with a fatty acid
of either a native or installed Lys residue across the GLP-1
sequence gave 31-mers with significant improvements in half-
life.29,30 Their research ultimately resulted in the discovery of
liraglutide which has completed phase III trials. Through all
of the structural advances, a model had developed in which
three regions of the GLP-1 peptide were determined to be
necessary for agonist activity at the GLP-1 receptor.31 Herein
we report the discovery of modified GLP-1 peptides that
represent a new minimal sequence for highly potent full
agonists of the GLP-1 receptor, a class II GPCR.
Our approach started with the iterative addition of one
amino acid to the N-terminal 9-mer sequence of GLP-1. This
9-mer sequence was chosen on the basis of research showing
that five of the amino acids critical for binding or functional
activity were located in this region and were thought to
represent one of the domains important for interactions with
the GLP-1R. Starting with this sequence, mono- and dipep-
tide sequences were added and assessed in vitro for functional
activity by measuring cAMP accumulation in Chinese ham-
ster ovary cells overexpressing the human GLP-1R. Through
our initial scan, it was found that an 11-mer peptide contain-
ing two biphenylalanine residues (BIP) at the C-terminus, 1, is
a submicromolar (545 ( 117 nM) full agonist for the GLP-1
receptor in the cAMP accumulation assay (Table 1), and the
functional activity can be inhibited by the known GLP-1R
antagonist exendin-4(9-39) (Figure 2).32
The concentration of exendin-4(9-39) that causes a 2-fold
increase in the EC50 of peptide 1 was estimated to be 37 nM
(from the curve fitting, pA2 = -1.57 ( 0.16 nM), and the
Schild slope was not statistically significantly different from 1,
consistent with simple competitive inhibition by exendin-
4(9-39) in the functional assay. Peptide 1 also displaced
125I-GLP-1 in a competitive radioligand binding assay (data
not shown), consistent with literature reports that five of the
first nine amino acids of GLP-1 are important for recep-
tor binding.24,25 In the following SAR studies, primary em-
phasis was placed on optimizing potency in the functional
assay because the 11-mers are structurally distinct GLP-1R
agonists.
Both BIP residues were determined to be critical for
activity, as the 10-mer 2 was completely inactive (Table 1).
As well, the C-terminal carboxamide was important for
potency, as the corresponding acid (peptide 3) was found to
be inactive. After the identification of 1, peptides were synthe-
sized in which BIP was kept constant at either position 10 or
11 (equivalent to position 16 or 17 in GLP-1(7-36) amide)
and the corresponding open position was scanned with
selected amino acids (Figure 3). From this study, no dipeptide
was identified that was as potent as 1. From the collection,
only one BIP replacement at position 10 (2-naphthyl-Ala) was
found that resulted in weak agonist activity (9000 nM).
However, replacements for the BIP in position 11 did identify
other modestly active sequences (BIP with 2-naphthyl-Ala, 4-
MePhe, 4-nitroPhe or Tyr) with agonist activity of approx
10000 nM.
alaninea
a (a) DCC/CH2Cl2; (b) 2-fluorobenzyl bromide, DIEA, DMF; (c) 6 N
HCl, room temp; (d) conc HCl, dioxane, reflux, 48 h; (e) (CF3CO)2O,
TFA; (f) (1) carboxypeptidase A, phosphate buffer, pH 7.2; (2) 6 N HCl,
pH 1; (h) NaCO3, pH 7-8, Fmoc-OSu, THF.
Figure 1. Amino acid sequence of GLP-1(7-36) amide.
Peptide Synthesis. Peptides were prepared by solid-
phase synthesis using a NR-Fmoc/tert-butyl protection strat-
egy on 4-(20,40-dimethoxyphenyl-Fmoc-aminomethyl)phe-
noxy (Rink) resin.20 After assembling on solid phase, the
resulting protected 11-mers were cleaved from the resin,
deprotected using either TFA/TIS/water or the TFA-based
reagent K.21 Coupling to R-methylphenylalanine and fluori-
nated analogues in position 6 required use of a large excess
(10-fold) of Fmoc-Thr(OtBu)-OH and HOAt/DIC in NMP
for the reaction to go to completion. Methods for purifica-
tions by preparative RP-HPLC, biology methods, and char-
acterization data are in the Experimental Section and
Supporting Information.
Results and Discussion
Since the discovery of GLP-1 (Figure 1), significant ad-
vances have been made in the understanding of the structural
requirements for the functional activity of this endogenous
peptide. As well, modifications to improve the biological half-
life of GLP-1 have been discovered through key amino acid
substitutions, modifications, and conformational constraints
linking amino acid side chains.
SAR studies focused on the determination of the key
residues important for the binding and function of GLP-1
demonstrated that the minimal sequence that retained func-
tional agonist activity, i.e., GLP-1R-mediated stimulation
of cAMP accumulation, is the 7-34 sequence of GLP-
1(7-37).22,23 From alanine scans, the amino acids of greatest
importance were determined to be those at positions 7, 10, 12,
13, 15, 19, 21, 28, and 29.24,25 Residues His-7 and Asp-15, in
particular, appear to be necessary and critical for functional
activity.26 With most of the sequence and with many amino
acids needed for agonist activity, approaches to improve on
the GLP-1 structure have involved specific amino acid sub-
stitutions andmodifications tolesscriticalaminoacidresidues
of the 30 amino acid sequence. Stabilization of GLP-1 to the
primary cleavage event by DPP4 was accomplished through
the recognition that Ala-8 could be replaced with other amino
acids including Aib, Gly, Ser, and Thr without significant loss
With peptide 1 remaining as the most active 11-mer,
attempts to increase potency were envisioned through sub-
stitutions on the phenyl rings of the BIPs (Table 1). A methyl

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7791
Table 1. 11-mer GLP-1R Agonists: Positions 10 and 11 SAR
peptide no.
Xaa10
Xaa11
hGLP-1R cAMP EC50 (nM)
GLP-1(7-36)-NH2
1
0.034 ( 0.01
545 ( 117
>10000
>10000
787 ( 522
480 (n = 2)
215 (n = 2)
390
BIP
BIP-NH2
2
BIP-NH2
BIP
BIP
3
BIP-OH
4
BIP-BIP-NH2
5
BIP
BIP
(20-Me)-BIP-NH2
(30-Me)-BIP-NH2
(40-OMe)-BIP-NH2
(40-Me)-BIP-NH2
(40-F)-BIP-NH2
(20-Me)-BIP-NH2
(30-Me)-BIP-NH2
BIP(40-Me)-NH2
BIP-NH2
6
7
BIP
BIP
8
270
430
9
BIP
10
11
12
13
14
15
16
17
18
(20-Me)-BIP
(20-Me)-BIP
(20-Me)-BIP
(20-Et)-BIP
(20-Et)-BIP
(20-n-Pr)-BIP
(20-Me, 40-OMe)-BIP
(30-OMe)-BIP
(20-Et, 40-OMe)-BIP
148 ( 13
87 ( 59
95 ( 73
27
(20-Me)-BIP-NH2
(20-Me)-BIP-NH2
(20-Me)-BIP-NH2
(20-Me)-BIP-NH2
(20-Me)-BIP-NH2
22 (n = 2)
93
35 ( 15
965 (n = 2)
7.0 ( 5.4
Figure 2. Concentration-response of peptide 1-mediated stimula-
tion of cAMP accumulation was measured in CHO cells over-
expressing the human GLP-1 receptor in the presence of the
concentrations of the antagonist, exendin-4(9-39), indicated in
the legend. All of the data were fit simultaneously to a modified
Gaddum-Schild model (GraphPad Prism software, version 4.03)
to estimate the concentration of the antagonist that caused a 2-fold
increase in the EC50 of peptide 1.
Figure 3. 11-mer SAR: scan of amino acids with BIP in position 10
or 11. Analogues with weak GLP-1R agonist activity are shown
with EC50 values in parentheses.
20-ethyl-BIP in 10 and the 20-methyl-BIP in 11 was found to be
equally potent(22nM, peptide 14). Peptide 15, with20-propyl-
BIP in position 10, is less active with EC50=93 nM. Other
substitutions in the distal ring of the position 10 BIP that
increased functional activity were also found. Peptide 16 with
the 20-methyl-40-methoxy substitution showed an increase in
functional activity relative to peptide 5, whereas the 30-meth-
oxy, 17 (1000 nM), is somewhat less active relative to peptide 5
in comparison. From the SAR studies on the position 10 BIP,
an analogue combining the optimized substitutions was
synthesized. The resulting analogue containing 20-ethyl-40-
methoxy-BIP in position 10, peptide 18, is a 7 nM full agonist
for the GLP-1 receptor.
walk around the distal ring of the position 11 BIP (peptides 5,
6, 8) resulted in analogues that were essentially equipotent to
1. Likewise, the addition of an electron withdrawing fluoro
(peptide 7) or an electron donating methoxy (peptide 9) did
not affect activity. While no improvements in functional
activity were achieved through these modifications, many
groups were identified that retained activity. In contrast,
substitutions to the distal phenyl ring of the BIP in position
10 had a larger effect on activity. Notably, addition of a
methyl to the 20-position increased activity and could be
paired with any of the position 11 substituted BIPs to improve
functional activity to the 100 nM range (peptides 10, 11, 12).
Other 20 substitutions were explored. Peptide 13, with an ethyl
in the 20 position, increased activity to 27 nM, a 20-
fold improvement over 1. Likewise, the analogue with the
With the optimization of the BIPs in positions 10 and 11
yielding analogues with nearly 100-fold improvements in
GLP-1 receptor functional activity over the lead peptide 1,

7792 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Mapelli et al.
further gains in potency were attempted through modifica-
tions of the amino acids in positions 1-9. Studies on the
conformation of GLP-1 have identified several regions that
have R-helical character. These are found between residues
Thr13-Glu20 and Ala24-Gly35, with the latter being more
stable as determined by solution NMR studies. However, by
contrast the N-terminal His7-Thr 13 was found to be a
random coil structure in solution. The described 11-mers
can then be characterized as overlaying with both the domains
containing the random coil and R helix.33 Recently, studies on
the receptor bound conformation of GLP-1 have shown that
the N-terminal R helix extends over residues 11-2134 (which
would start at position 5 in the described 11-mers).
In the design of truncated GLP-1 analogues, it was re-
cognized that conformational order would be needed to
provide 11-mers with GLP-1 like potency. In an attempt to
further induce helical structure in the above-described 11-mer
peptides, R-methylamino acids were tried in several positions.
Peptide 19 (Table 2), with an R-methyl-Phe in position 6,
showed modest gains in potency relative to peptide 18. When
this finding was paired with the structurally enforcing and
DPP4 stabilizing Aib in position 2, a subnanomolar peptide,
20 (EC50=0.28 nM), was obtained. The 11-mer, peptide 20 is
within 10-fold of the activity of the 30 amino acid native
peptide, GLP-1. Further improvements in potency were
realized through SAR efforts focusing on substituting the
phenyl of R-methyl Phe. Of the substitutions explored, fluoro
groups at the 2 position or the 2 and 6 positions were found
to substantially improve activity. Peptide 21, containing an
R-methyl-2-fluoro-Phe, and peptide 22, containing R-methyl-
2,6-difluoro-Phe, show significant improvements in func-
tional activity and were found to have similar potency to
GLP-1.
1
Using 2D H-1H NMR methods, the solution conforma-
tion of peptide 21 was determined and the NMR structure
ensemble is shown in Figure 4. This ensemble reflects a
population of ordered conformations for peptide 21, with
residues 6-11 forming a 310 helix showing some distortion
from the canonical conformation for residues 9-11. There is a
kink at position 2, and residues 3 and 4 could be characterized
as a distorted type I turn or a type VIIa turn. The region around
the R-methyl-2-fluoro-Phe in position 6 shows the formation of
an incipient 310 helix, most likely in synergy with the Aib in
position 2. Analogues lacking R-methylation at position 6
exhibited signatures of nascent turns or helices in their NMR
spectra but did not yield sufficient numbers of NOE distance
restraints to produce well-defined conformations.
Peptide 21 was found to be highly selective (EC50 or IC50
values of >10 μM) against several related class II GPCRs:
glucagon, glucose-dependent insulinotropic (GIP), the pitui-
tary adenylate cyclase-activating polypeptide (PACAP), va-
soactive intestinal peptide (VIP), and parathyroid hormone
(PTH) receptors. In addition, peptide 21 was also profiled
against 31 different receptors, enzymes, channels, and trans-
porters (CEREP screen) and was determined to be >10000-
fold selective against all of these targets.
In preparation for in vivo testing, peptides 21 and 22 were
profiled in cells expressing the mouse GLP-1 receptor. The
potency of native GLP-1 was not significantly different at the
two receptors. However, peptide 21 showed a ∼30-fold
decrease in functional potency with an EC50 value of 2.4 (
0.8 nM (mean ( SEM; n=11) versus 0.087 ( 0.01 nM (n=15)
in the mouse versus human GLP-1R cAMP accumulation
assays, respectively. For peptide 22, the EC50 values for
activation of the mouse versus human GLP-1R were 0.8 (
0.17 nM (n=9) and 0.12 ( 0.02 nM (n=12), respectively. The
potencies of both peptides 21 and 22 were statistically sig-
nificantly different (p < 0.05) when compared at receptors
across species but not when compared to each other at a single
Table 2. 11-mer GLP-1R Agonists: Positions 2 and 6 SAR
His-Xaa2-Glu-Gly-Thr-Xaa6-Thr-Ser-Asp- BIP(20-Et, 40-OMe)-
BIP-(20-Me)-NH2
hGLP-1R cAMP
peptide no.
Xaa2
Xaa6
EC50 (nM)
19
20
21
22
Ala
Aib
Aib
Aib
(L)-R-Me-Phe
(L)-R-Me-Phe
(L)-R-Me-(2-F)-Phe
(L)-R-Me-(2,6-di-F)-Phe
2.4 ( 2.8
0.28
0.087 ( 0.04
0.093 ( 0.07
Figure 4. Stereoview of an ensemble of 20 NMR structures of peptide 21 computed using 62 sequential, 67 medium-range, and 8 long-range
˚
NOE restraints. The average backbone atom rmsd to the mean structure is 0.29 A, and the average all-heavy-atom rmsd to the mean structure is
1
1.00 A. Atoms are color-coded according to type; hydrogens are not shown. His is near the bottom on the right side of the image.
˚

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7793
Figure 5. Effects of a single subcutaneous dose of peptides 21 and 22 in an IPGTT in male ob/ob mice. Peptide (dose in nmol/kg indicated in
parentheses in the legends) or vehicle (1:1 propylene glycol/20 mM sodium phosphate, pH 7.4) was administered to groups of six fasted male ob/
ob mice 30 min prior to the intraperitoneal administration of 2 g/kg glucose at time 0. The time course for plasma glucose changes during the
IPGTT from two separate studies are shown in panels A and C, and the corresponding insulin time courses are shown in panels B and D.
receptor (p > 0.05). The species dependence of the activity of
the 11-mers may reflect binding of the unnatural amino acids
in positions 10 and 11 to a region of the receptor which can
vary across species because it is not essential for binding of the
native ligand.
Peptide 21 was profiled in vivo in the insulin resistant ob/
ob mouse following a single subcutaneous dose of 30 or
300 nmol/kg with 1 nmol/kg exendin-4 as a positive control
(Figure 5A,B). The acute efficacy of the peptides was assessed
using a glucose tolerance test (GTT) initiated by intraperito-
neal (ip) injection of 2 g/kg glucose 30 min after subcutaneous
administration of the peptides or vehicle. Figure 5A shows the
time course for plasma glucose changes during the IPGTT.
Plasma glucose values in all three peptide-treated groups were
statistically significantly lower than those in the vehicle group
at each time point from 30 to 180 min after the glucose
challenge (ANOVA with Dunnett’s test, p < 0.001 vs vehicle
at t = 30-180 min). The plasma glucose values were not
statistically significantly different at any time point following
administration of either 300 nmol/kg peptide 21 or 1 nmol/kg
exendin-4 (Figure 5A). The decreases in plasma glucose were
accompanied by increases in plasma insulin concentrations,
consistent with a GLP-1-dependent mechanism (Figure 5B).
The time courses for plasma insulin concentrations were again
very similar following treatment with 300 nmol/kg peptide 21
or 1 nmol/kg exendin-4. With either of these treatments, the
plasma insulin concentrations were statistically significantly
greater than those of the vehicle group at both the 30 min (p <
0.01) and 60 min (p < 0.05) time points based on an ANOVA
followed by Dunnett’s post hoc test. Peptide 21 exhibits a
similar efficacy profile but is considerably less potent than
exendin-4 in the ob/ob mouse IPGTT model. Peptides 21 and
22 also produced similar glucose lowering to each other and
stimulated insulin secretion when compared in the IPGTT
model at a dose of 300 nmol/kg (Figure 5C and Figure 5D).
Consistent with their similar in vitro potencies, it was not
possible to differentiate peptides 21 and 22 in the ob/ob
IPGTT model.
Figure 6. Pharmacokinetics of peptides 21 and 22 in the ob/ob
mouse. The vehicle was 1:1 propylene glycol/20 mM sodium
phosphate, pH 7.4. Samples were collected at 1, 2, 4, 6, 8, and 24
h postdosing and were analyzed by LC-MS/MS.
The pharmacokinetics of peptides 21 and 22 were measured
in the ob/ob mouse, the pharmacodynamic model (Figure 6).
The peptides were administered via subcutaneous injection of
300 nmol/kg in 1:1 propylene glycol/20 mM sodium phos-
phate buffer, pH 7.4. The Cmax values were greater than 2000
nM for peptides 21 and 22, and both peptides achieved similar
measurable exposures over 24 h in ob/ob mice. Plasma
concentrations of exendin-4 were below the lower limit of
quantitation (∼5 nM) of the LC-MS method following
administration of the 1 nmol/kg dose used in the efficacy
study, underscoring the marked difference in potency between
these peptides (Figure 5). Even accounting for decreased
potency at the mouse GLP-1 receptor, these relatively high
efficacious plasma concentrations suggest that improvements
in the in vivo potency will be required to produce an 11-mer
that could be administered as a ∼1 mg dose in human diabetic
patients, similar to liraglutide.35
The pharmacokinetic properties of peptides 21 and 22 were
also investigated in a normal dog model. Subcutaneous
administration of 67 μg/kg (∼45 nmol/kg) in 1:1 propylene
glycol/20 mM sodium phosphate buffer, pH 7.4, resulted in
measurable exposure over a 24 h period for peptide 22, but
peptide 21 could not be detected beyond the 8 h time point
(Figure 7). This was the first significant difference in the

7794 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Mapelli et al.
methyluronim hexafluorophosphate (HATU) (Applied Bio-
sytems, Foster City, CA); HPLC-grade acetonitrile
(Mallinckrodt Baker, Inc.).
Solution Synthesis of Substituted Biphenylalanines (2-Amino-
3-(biphenyl-4-yl)propanoic Acid or BIPs). The biphenylalanine
derivatives described in this work were synthesized using the
procedures exemplified below, starting from the commercially
available Boc-tyrosine methyl ester to prepare the triflate and
using the appropriate boronic acid to produce the desired
biphenylalanine analogues. When a required boronic acid was
not available from commercial sources, its synthesis was per-
formed as exemplified below.
Synthesis of (S)-2-(9H-fluoren-9-yl)methoxy)carbonylamino)-
3-(20-methylbiphenyl-4-yl)propanoic Acid or Fmoc-(20-Me)-BIP-
OH. a. Preparation of Boc-L-tyrosine O-Triflate Methyl Ester.
To a solution of Boc-tyrosine methyl ester (126 mmol, 37 g) and
pyridine (25.4 mL, 314 mmol) in CH2Cl2 (100 mL), at -15 °C
under N2, was added slowly triflic anhydride (25.4 mL, 151
mmol). The solution was stirred at -15 °C for 15 min, and after
this time HPLC analysis indicated that the reaction was com-
plete. To the reaction was added water (150 mL). The cooling
bath was removed, and the layers were separated. The organic
layer washed with 0.5 M aqueous NaOH (2 ꢀ 150 mL) and 15%
aqueous citric acid (2 ꢀ 150 mL). The organic layer was dried
over MgSO4, filtered, and concentrated to give the crude
product as a red oil. The crude product was used “as is” in the
next step.
b. Preparation of (S)-2-(tert-Butoxycarbonylamino)-3-(20-
methylbiphenyl-4-yl)propanoic Acid Methyl Ester. A degassed
suspension of o-tolylboronic acid (140 mmol, 19.0 g), K2CO3
(175 mmol, 24.1 g), and tetrakis(triphenylphosphine)palladium
(0) (4.0 mol, 4.6 g, 0.034 equiv) in toluene (580 mL) is heated to
80 °C. To the resulting suspension was added the crude Boc-L-
tyrosine O-triflate methyl ester from the previous reaction step
(above) dissolved in toluene (70 mL). The resulting reaction
mixture was maintained at 80 °C under N2 for 3 h. After this
time, the reaction mixture was cooled to room temperature and
then filtered through a plug of Celite. The filtrate was washed
with 0.5% aqueous NaOH (2 ꢀ 150 mL) and 15% aqueous citric
acid (2 ꢀ 150 mL). The organic layer was then dried over
MgSO4, filtered, and concentrated. The obtained crude mixture
was purified by silica gel column chromatography, eluting with
EtOAc/heptane (1:9) [the crude mixture was preabsorbed on
silica gel (2 g silica gel/g crude mixture), 1:35 mixture/silica gel
used for the column], and yields varied from 50% to 80%.
c. Preparation of (S)-2-(tert-Butoxycarbonylamino)-3-(20-methyl-
biphenyl-4-yl)propanoic Acid. (S)-2-(tert-Butoxycarbonylamino)-
3-(20-methylbiphenyl-4-yl)propanoic acid methyl ester (120
mmol, 44.5 g) is dissolved in a solution of MeOH (147 mL)
and THF (442 mL). To the resulting solution at room tempera-
ture was added 1 N NaOH (147 mmol, 147.4 mL). HPLC
analysis indicated that the reaction was complete after 1 h. After
this time, the reaction mixture was concentrated at room
temperature to remove the MeOH and THF. The resulting
solution was partitioned between water (500 mL) and Et2O
(300 mL). The layers were separated, and the aqueous layer was
acidified with 1 N HCl (160 mL) and then extracted with Et2O
(2 ꢀ 250 mL). The combined Et2O extracts were dried over
MgSO4, filtrated, and concentrated, resulting in 41.5 g of title
compound.
Figure 7. Pharmacokinetics of peptide 21 and peptide 22 in the dog.
The figure shows the plasma concentration-time profiles following
subcutaneous (sc) administration in male beagle dogs (67 μg/kg,
n = 3). The dosing vehicle was 1:1 propylene glycol/20 mM sodium
phosphate buffer, pH 7.4. Samples were collected at 0.25, 0.5, 0.75,
1, 2, 4, 6, 8, 24, and 30 h postdosing and were analyzed by LC-MS/
MS. The plasma concentration of peptide 21 was below the LLQ at
24 and 30 h.
profiles of thesetwo peptides. Peptide 22 had a mean residence
time of 20 ( 5 h, suggesting that this peptide may potentially
meet an important PK criterion for a once a day agent.
Moreover, the peptide shows a relatively low peak to trough
ratio. This could be significant if this PK profile were to
translate to humans because rapid increases in plasma con-
centration and a high Cmax appear to contribute to the dose-
limiting side effects of nausea and vomiting reported for
peptide GLP-1 receptor agonists in the clinic.36-38
In conclusion, we have discovered a novel class of small
peptide agonists for the GLP-1 receptor. Through replace-
ment of the C-terminal 21 amino acids with a biphenylalanine
dipeptide followed by sequential optimization of several key
N-terminal amino acid residues, 11-mer peptides were dis-
covered that have in vitro functional activity similar to that of
GLP-1 while maintaining selectivity versus other related class
II GPCRs. Peptides 21 and 22 were shown to stimulate insulin
secretion and reduce glucose excursions in vivo in a mouse
model of diabetes, consistent with the expected mechanism
of action for a GLP-1R agonist. In addition, these 11-mers
show an enhanced pharmacokinetic half-life relative to
GLP-1. In particular, peptide 22 displays favorable pharma-
cokinetic properties in mouse and dog PK models, demon-
strating a low peak to trough with 24 h of exposure. The
11-mers represent a unique tool for further investigating the
pharmacology of the GLP-1 receptor and, with addi-
tional improvements of in vivo potency, could represent a
new series of clinical agents for the treatment of type 2
diabetes.
Experimental Section
Materials. NR-Protected amino acids and the 4-(20,40-di-
methoxyphenyl-Fmoc-aminomethyl)phenoxyacetyl-p-methyl-
benzhydrylamine resin (Rink amide MBHA resin) were
purchased from Novabiochem (San Diego, CA). The following
reagents were purchased from the commercial sources indicated:
trifluoroacetic acid (TFA, EMD Chemicals); N,N-diisopropyl-
carbodiimide (DIC), N,N-diisopropylethyl amine (DIEA), N-
bromosuccinimide (NBS), piperidine, acetic anhydride, and
triisopropylsilane (TIS) (Aldrich, Milwaukee, WI); dichloro-
methane (CH2Cl2, Mallinckrodt Baker, Inc.); N-methylpyrro-
lidone (NMP) and N,N-dimethylformamide (DMF) (Burdick
and Jackson); 1-hydroxy-7-azabenzotriazole (HOAt), 2-(1H-
benzotriazol1-yl)-1,1,3,3-tetramethyluronium hexafluoropho-
sphate (HBTU), and O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetra-
d. Preparation of (S)-2-((9H-Fluoren-9-yloxy)carbonylamino)-
3-(20-methylbiphenyl-4-yl)propanoic Acid. To a solution of (S)-
2-(tert-butoxycarbonylamino)-3-(20-methylbiphenyl-4-yl)pro-
panoic acid (117 mmol, 41.5 g) in CH2Cl2 (1 L) kept at room
temperature was slowly bubbled in gaseous HCl. After 5 min, a
white solid started to precipitate. After 2 h, the reaction was
shown to be complete by HPLC. The reaction mixture was then
concentrated, and the resulting residue was redissolved in THF
(600 mL). To the resulting solution was added water (150 mL)
followed by the slow portionwise addition of solid NaHCO3

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7795
until the pH of the mixture was basic. A white solid precipitated
out, and Fmoc-OSu (115 mmol, 38.9 g) was then added. The
resulting reaction mixture was stirred at room temperature, and
after 1 h, a homogeneous biphasic solution was obtained.
Stirring was continued at room temperature for an additional
16 h. After this time, the layers were separated. The THF layer
was acidified with 2 N aqueous HCl (58 mL) and then diluted
with EtOAc (400 mL). The layers were separated, and the
organic layer was washed with water (2 ꢀ 100 mL), dried over
MgSO4, filtered, and concentrated. The crude product was
purified using silica gel column chromatography using CH2Cl2
as eluant until most of the fast running impurities had been
removed. The eluant was then changed to 25% EtOAc in
heptane containing 1% acetic acid. The yield was >90% for
the three steps.
(S)-2-((9H-Fluoren-9-yloxy)carbonylamino)-3-(20-ethyl-40-metho-
xybiphenyl-4-yl)propanoic Acid or Fmoc-BIP-(20-Et-40-OMe).
Method A. Synthesis of (2-Ethyl-4-methoxy)phenylboronic Acid. a.
Preparation of (2-Bromo-5-methoxy)styrene. A suspension of
methyltriphenylphosphonium bromide (199.5 g, 0.465 mol) in dry
THF (800 mL) was purged for 10 min and cooled to 10 °C. Then
2.75 M n-BuLi in hexanes (169 mL, 0.465 mol) was added slowly
over 30 min. The resulting solution was stirred for 1 h. To the
reaction mixture was slowly added over a period of 30 min a solution
of (2-bromo-5-methoxy)benzaldehyde (100 g, 0.465 mol) in dry
THF (300 mL). After the addition was complete, the reaction
mixture was stirred for 1 h. Petroleum ether (2 L) was added, and
the reaction mixture was stirred for an additional 30 min and then
filtered through a silica gel pad. The pad was washed with Et2O. The
combined organic washes were concentrated with the temperature
kept below 30 °C. The resulting the crude product was purified by
silica gel column chromatography, using 100% petroleum ether as
eluent. Yield: 92 g, 90%, as pale-yellow liquid.
b. Preparation of (4-Bromo-3-ethyl)anisole. 2,20-Bipyridyl
(24.3 g, 0.15 mol) and (2-bromo-5-methoxy)styrene (0.31 mol,
65 g) were dissolved in EtOAc (650 mL), and the resulting
solution was cooled to 0 °C. The solution was purged with
nitrogen, and 10% palladium on carbon (16.25 g, 25%) was
added under a stream of nitrogen. The reaction mixture was
stirred under 2 kg pressure in a Parr shaker for 3 days under
hydrogen. The reaction progress was monitored by HPLC.
Upon completion, the reaction mixture was filtered through
Celite, and the filtrate was washed with 5% aqueous KHSO4,
dried over Na2SO4, filtered, and concentrated, keeping the
temperature below 30 °C. Yield: 60 g, 91%, as pale-yellow
liquid. LC-MS: (M þ H)þ at m/z 214, 216. 1H NMR (300
MHz, CDCl3): δ 7.37 (dd, 1H, J=3, 10 Hz); 6.77 (d, 1H, J=3
Hz); 6.65 (dd, 1H, J=3, 10 Hz); 3.80 (s, 3H); 2.77 (2H, q, J=7.6
Hz); 1.24 (3H, t, J=7.6 Hz).
c. Preparation of (2-Ethyl-4-methoxy)phenylboronic Acid. A solu-
tion of (4-bromo-3-ethyl)anisole (94 g, 0.437 mol) in THF (900
mL) was cooled to -78 °C, and n-BuLi (249 mL, 0.55 mol) was
added dropwise while maintaining the same temperature. Stir-
ring was continued for 1 h at -78 °C, and tri-n-butyl borate
(177 mL, 0.655 mol) was then added slowly at -78 °C. The
cooling bath was removed, and the reaction mixture was
allowed to warm to 0 °C and was then quenched with 1.5 N
hydrochloric acid at 0 °C. The organic layer was separated, and
the aqueous layer was extracted with EtOAc. The combined
organic layers were washed with saturated aqueous NaCl, dried
over MgSO4, filtered, and concentrated. The resulting residue
was stirred in petroleum ether for 30 min, and the solid obtained
was collected by filtration and then dried under vacuum. Yield:
65 g, 82%, as a white solid. 1H NMR (300 MHz, DMSO-d6): δ
7.80 (br s, 2H); 7.41 (d, 1H, J=9); 6.68 (d, 2H, 7.5); 3.70 (s, 3H);
2.75 (q, 2H, J=7.5); 1.14, 3H, J=7.5). 13C NMR (75 MHz,
DMSO-d6): δ 16.4, 28.0, 54.2, 109.3, 113.3, 125.9, 134.9, 150.01,
159.6.
phenol (50 g, 0.4 mol) and K2CO3 (283 g, 2.05 mol) in dry acetone
(500 mL) was added methyl iodide (290 g, 2.05 mol). The reaction
mixture was transferred to an autoclave and refluxed at 70 °C
overnight. The reaction mixture was filtered through a Celite pad,
the pad was washed with acetone, and the combined filtrate and
washes were concentrated. The product was dissolved in CH2Cl2,
filtered, and evaporated to dryness. Yield: 50 g, 90%, as a brown
liquid. 1H NMR (400 MHz, CDCl3): δ 7.20-7.28 (m, 1H);
6.74-6.94 (m, 3H); 3.83 (s, 3H); 2.66 (q, 2H, J=7.2); 1.28 (t,
3H, J=7.7).
b. Preparation of (4-Bromo-3-ethyl)anisole. 3-Ethylanisole (50
g, 0.3676 mol) and N-bromosuccinimide (72 g, 0.4 mol) in
acetonitrile (1 L) were stirred for 8 h in the dark at room
temperature. The reaction mixture was concentrated, keeping
the temperature below 40 °C, and the residue obtained was
redissolved in CCl4 and filtered. The filtrate was concentrated,
and the product was purified by fractional distillation. Yield: 35
1
g, 43%, as pale-yellow liquid. H NMR (300 MHz, CDCl3): δ
7.35-7.45 (m, 1H); 6.80 (d, 1H, J=2.9 Hz); 6.63 (dd, 1H, J=3.0,
8.7 Hz;); 3.80 (s, 3H); 2.74 (q, 2H, J=7.5); 1.23 (t, 3H, J=7.6).
c. Preparation of (2-Ethyl-4-methoxy)phenylboronic Acid. The
(4-bromo-3-ethyl)anisole was converted to the corresponding
boronic acid as described in method A, part c, above. The
1
boronic acid product had H and 13C NMR spectra identical
with those of the product obtained in method A, part c, above.
Synthesis of (S)-2-((9H-Fluoren-9-yloxy)carbonylamino)-
3-(20-ethyl-40-methoxybiphenyl-4-yl)propanoic Acid. a. Prepara-
tion of Boc-L-tyrosine O-Triflate Methyl Ester. To a solution of
Boc-L-tyrosine methyl ester (85 mmol, 25 g) and 2,6-lutidine
(339 mmol, 36.25 g) in CH2Cl2 (200 mL) at -40 °C under N2 was
slowly added a solution of triflic anhydride ((169.5 mmol, 47.74
g) in CH2Cl2 (100 mL) over 30 min. The solution was stirred
at -40 °C for an additional 2 h, and after this time HPLC
analysis indicated that the reaction was complete. The reaction
was quenched by addition of water (20 mL). The layers were
separated, and the organic layer was washed with 1 N aqueous
HCl (3 ꢀ 200 mL), saturated aqueous Na2CO3 (200 mL), water
(200 mL), and saturated aqueous NaCl (200 mL). The organic
layer was dried over MgSO4, filtered, and concentrated to give
the crude product as a red oil. The crude product was purified by
silica gel column chromatography (300 g silica gel, 0-50%
EtOAc in hexanes as a gradient) to give the desired compound
(27 g, 75% yield) as a white solid. 1H NMR (300 MHz, CDCl3):
δ 7.10-7.30 (m, 4H); 5.01-5.06 (br m, 1H); 4.55-4.65 (br
m, 1H); 3.70 (s, 3H); 3.15-3.23 (m, 1H); 3.01-3.11 (m, 1H);
1.42 (s, 9H).
b. Preparation of (S)-2-(tert-Butoxycarbonylamino)-3-(20-eth-
yl-40-methoxybiphenyl-4-yl)propanoic Acid Methyl Ester. Boc-L-
tyrosine O-triflate methyl ester (81 g, 0.19 mol) in dry toluene
(600 mL) was purged for 10 min with nitrogen. K2CO3 (36 g,
0.26 mol) in water (200 mL) was added, followed by 2-ethyl-4-
methoxyphenylboronic acid (36 g, 0.2 mol). To the solution
were added Pd(PPh3)4 (16.18 g, 0.014 mol), ethanol (200 mL),
and THF (400 mL), and the reaction mixture was then heated to
100 °C for 4 h. After this time, the reaction mixture was
concentrated under vacuum and the residue was dissolved in
CH2Cl2 (1.0 L). The organic layer was washed with 10%
aqueous sodium hydroxide, 15% aqueous citric acid, dried over
Na2SO4, filtered, and concentrated. The crude product was
purified by silica gel column chromatography, eluting with
10% EtOAc in petroleum ether. Yield: 50 g, 65%, as a yellow
liquid. 1H NMR (300 MHz, CDCl3): δ 7.10-7.30 (m, 5H); 6.86
(d, 1H, J=2.4 Hz); 6.78 (dd, 1H, J=2.5, 8.4 Hz); 4.95-5.05 (br
m, 1H); 4.60-4.70 (br m, 1H); 3.85 (s, 3H); 3.71 (s, 3H);
3.05-3.20 (br m, 2H); 2.56 (q, 2H, J=7.5 Hz); 1.43 (s, 9H);
1.11 (t, 3H, J=7.5 Hz).
c. Preparation of (S)-2-(tert-Butoxycarbonylamino)-3-(20-eth-
yl-40-methoxybiphenyl-4-yl)propanoic Acid. To a mixture of (S)-
2-(tert-butoxycarbonylamino)-3-(20-ethyl-40-methoxybiphenyl-
4-yl)propanoic acid methyl ester (60 g, 0.146 mol) in THF
Method B. Synthesis of (2-Ethyl-4-methoxy)phenylboronic
Acid. a. Preparation of 3-Ethylanisole. To a mixture of 3-ethyl-

7796 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Mapelli et al.
(450 mL) and MeOH (85 mL) was added a solution of sodium
hydroxide (24 g, 0.58 mol) dissolved in water (85 mL). The
reaction mixture was stirred at room temperature overnight.
After this time the solution was concentrated and the residue
dissolved in water (100 mL) The aqueous layer was washed with
Et2O and then was acidified to pH 1 using 20% aqueous citric
acid. The aqueous solution was extracted with EtOAc. The
organic extracts were combined and washed with saturated
aqueous NaCl, dried over Na2SO4, filtered, and concentrated.
Yield: 55 g, 94%, as colorless liquid. 1H NMR (300 MHz,
DMSO): δ 7.26-7.29 (m, 2H); 7.15-7.20 (m, 2H); 7.04 (d,
1H, J=8.3 Hz); 6.75-6.87 (m, 3H); 4.10-4.18 (br m, 1H); 3.77
(s, 3H); 3.02-3.05 (m, 1H); 2.70-2.85 (m, 1H); 2.50 (q, 2H, J=
7.5 Hz); 1.32 (s, 9H); 1.02 (t, 3H, J=7.5 Hz).
Procedures for Cleavage of the Boc Group. Procedure A. The
Boc-protected resins prepared as described in method A or B
were treated with 0.5 mL of a solution consisting of trimethylsi-
lyl trifluoromethanesulfonate, 2,6-lutidine, and CH2Cl2 (1:1:3
by volume). After three treatments for 1 h each with stirring, the
resins were washed with 4 ꢀ 1 mL of CH2Cl2, 3 ꢀ 1 mL of DMP,
3 ꢀ 1 mL of 20% MeOH in DMF, and 4 ꢀ 1 mL CH2Cl2 prior to
the next coupling step.
Procedure B. The Boc-protected resins prepared as described
in method A or B were treated with 1.0 mL of 1 N HCl in
anhydrous 1,4-dioxane for 1 h at room temperature with shak-
ing. The resins were washed with 4 ꢀ 1 mL of CH2Cl2, 3 ꢀ 1 mL
of 5% DIEA in CH2Cl2 (vol/vol), 3 ꢀ 1 mL of CH2Cl2, and 5 ꢀ
1 mL of N-methylpyrolidinone to provide the free amino resins
ready for the next coupling step.
d. Preparation of (S)-2-((9H-Fluoren-9-yloxy)carbonylamino)-
3-(20-ethyl-40-methoxybiphenyl-4-yl)propanoic Acid. (S)-2-(tert-
Butoxycarbonylamino)-3-(20-ethyl-40-methoxybiphenyl-4-yl)-
propanoic acid (55 g, 0.138 mol) was dissolved in dry CH2Cl2 (1
L), and dry HCl gas was passed through the reaction mixture at
room temperature for 6 h. A precipitate formed which was
collected by filtration and then dried under vacuum to give the
desired product (46 g, 100%). The resulting free amino acid
hydrochloride salt (30 g, 0.089 mol) was dissolved in THF
(700 mL), and a solution of NaHCO3 (29 g, 0.358 mol) predis-
solved in water (240 mL) was added. To the resuling solution
was added Fmoc-OSu (30 g, 0.089 mol) portionwise over a
period of 30 min. The reaction mixture was stirred overnight at
room temperature. The THF was removed under vacuum, and
water (2.0 L) was added. The clear solution was washed with
Et2O. The aqueous solution was acidified to pH 1 and extracted
with EtOAc. The organic layer was washed with water and
saturated aqueous NaCl. The EtOAc layer was dried over
Na2SO4, filtered, and concentrated to yield 37 g (80%) of
product as a pale-yellow solid. LC-MS: (M þ H)þ at m/z
522.0; [R]2D0 -14.5° (c 1% in DMF). 1H NMR (400 MHz,
DMSO-d6): δ 7.88 (2H, d, J=8.0 Hz); 7.82 (d, 1H); 7.67 (2H,
t, J=7.0 Hz); 7.32 (4H, t, J=8.0 Hz); 7.14 (2H, d, J=7.9); 6.99
(1H, d, J=8.4); 6.84 (1H, d, J=2.4); 6.75, (1H, d); 4.10-4.30
(4H, m); 3.77 (3H, s); 3.10-3.20 (1H, m); 2.88-2.98 (1H, m);
2.44 (2H, q, J=7.6); 0.94 (3H, t, J=7.5). 13C NMR (100 MHz,
DMSO-d6): δ 14.0, 15.4, 20.7, 25.7, 36.3, 46.6, 55.0, 55.4, 59.7,
65.7, 11.0, 114.0, 120.1, 125.2, 125.3, 127.0,127.6, 128.9, 130.7,
133.4, 136.2, 139.2, 140.7,142.4, 143.7, 143.8, 156.0, 158.6,
170.3, 173.4.
Solid Phase Synthesis of Substituted Biphenylalanines,. Meth-
od A. 4-(20,40-Dimethoxyphenylaminomethyl)phenoxy (Rink) re-
sin (50 mg, 0.025 mmol) was derivatized with a N-Boc-4-
iodophenylalanine residue either attached directly to the solid
support to synthesize a variety of position 11 biphenylalanine
analogues or via coupling to Rink resin already derivatized with an
amino acid to synthesize variety of position 10 biphenylalanine
analogues. The resulting derivatized Rink resins were weighed into
13 mm ꢀ 100 mm glass culture tubes with screw caps. Arylboronic
acids (0.5 mmol) were dissolved in 0.75 mL of 25% by volume
DIEA in N-methylpyrolidinone and added to the resins followed
by 0.05 mL of an N-methylpyrolidinone solution containing
1.0 mg of tetrakis(triphenylphospine)palladium(0) catalyst
(∼3.5 mol %). The resulting mixtures were blanketed with a
stream of nitrogen and the reaction vessels tightly capped and
maintained at 85-90 °C for 17-20 h with periodic shaking. The
resins were washed with 5 ꢀ 1 mL of N-methylpyrolidinone and
5 ꢀ 1 mL of CH2Cl2 prior to Boc group cleavage as described
below.
Synthesis of Fmoc-(S)-2-fluoro-r-methylphenylalanine. a.
Preparation of (R,S)-2-Fluoro-r-methylphenylalanine Hydro-
chloride. A suspension of N-benzoylalanine (50 g, 0.259 mol)
in CH2Cl2 (350 mL) under nitrogen was cooled to 0 °C, and a
solution of DCC (58.7 g, 0.28 mol) dissolved in CH2Cl2 (150 mL)
was added. The mixture was stirred at 0 °C for 30 min followed
by stirring at room temperature for 30 min. The reaction
progress was monitored by TLC. Upon disappearance of start-
ing material the reaction mixture was cooled to 0 °C and filtered.
The organic layer was concentrated, keeping the temperature
below 30 °C. Yield: 44 g as a pale-greenish yellow liquid, which
was used in the next step immediately. The oxazolidone from the
previous step (44 g, 0.25 mol) was dissolved in DMF (220 mL)
under nitrogen. To this solution was added 2-fluorobenzyl
bromide (38 g, 0.2 mol). The reaction mixture was cooled to
0 °C, and DIEA (48 g, 0.38 mol) was added dropwise over a
period of 30 min. The reaction mixture was stirred at 0 °C for
30 min and at room temperature for 30 min with TLC monitor-
ing. The reaction mixture was quenched with water (500 mL)
and extracted with EtOAc (3 ꢀ 200 mL). The combined organic
layers were washed with water, saturated aqueous NaCl, dried
over Na2SO4, filtered, and concentrated. Yield: 50 g, as a pale-
yellow liquid. The crude product was taken into the next step
without purification. To the benzylated oxazolidone from the
previous step (50 g, 0.176 mol) was added 6 N HCl (150 mL) and
Et2O (300 mL). The reaction mixture was stirred at room
temperature for 1 h. A precipitate formed, and the solid product
was collected by filtration and dried under vacuum to yield 50 g
of solid product, 60% overall yield for the last three steps. The
benzoylated amino acid from the previous step (98 g, 0.42 mol)
was dissolved in dioxane (300 mL), and to this solution was
added concentrated HCl (700 mL). The reaction mixture was
refluxed for 48 h. After removal of the dioxane in vacuo, the
aqueous layer was washed with CH2Cl2 and the aqueous layer
was concentrated. The residue was triturated with petroleum
1
ether. Yield: 69 g (70%) as an off-white solid. H NMR (300
MHz, DMSO-d6): δ 7.30-7.40 (2H, m); 7.12-7.22 (2H, m);
3.38-3.42 (2H, m); 1.46 (3H, s).
b. Preparation of N-Trifluoroacetyl-(R,S)-2-fluoro-r-methyl-
phenylalanine. To (R,S)-2-fluoro-R-methylphenylalanine hy-
drochloride from the previous step (220 g, 0.95 mol) was
added TFA (2.2 L) under nitrogen. The solution was cooled to
0 °C. Trifluoroacetic anhydride (500 g, 2.4 mol) was added
through an addition funnel dropwise over a period of 30 min,
and the reaction mixture was stirred at room temperature over-
night. After removal of TFA in vacuo, water (1 L) was added
and the mixture was extracted with EtOAc. The combined
EtOAc layers were washed with water and saturated aqueous
NaCl, dried over Na2SO4, filtered, and concentrated. The
product was precipitated with Et2O and collected by filtration.
The collected solid was washed with Et2O and then dried under
vacuum. Yield: 220 g, 75% as a white solid. 1H NMR (300 MHz,
DMSO-d6): δ 7.19-7.21 (1H, m); 7.18-7.13 (3H, m); 3.43-3.48
(1H, m); 3.03-3.08 (1H, m); 1.29 (3H, s).
Method B. The reactions were performed as in method A
except that a different catalyst was employed. The catalyst
solution was prepared by dissolving 9.0 mg of palladium(II)
acetate and 56 mg of 2-(dicyclohexylphosphino)biphenyl in
2.0 mL of N-methylpyrolidinone. For 0.025 mmol scale reac-
tions, 0.038 mL (∼3 mol %) of catalyst solution was used.

Article
c. Preparation of N-Fmoc-(S)-2-fluoro-r-methylphenylal-
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7797
mL of DMSO or 50% aqueous acetic acid for purification by
preparative HPLC as follows.
anine. To a suspension of N-trifluoroacetyl-(R,S)-2-fluoro-R-
methylphenylalanine (35 g, 0.12 mol) in water (300 mL) was
added 1 N aqueous sodium hydroxide until pH 8 was attained.
The suspension then became a clear solution. Phosphate buffer
(600 mL, pH 7.2) was added to maintain pH 7.2. The mixture
was heated to 37 °C. Carboxypeptidase A (60 000 units) was
added, and the reaction mixture was stirred overnight. HPLC
analysis showed 39% conversion. Then 6 N HCl was added to
bring the pH between 3 and 4. Activated charcoal (35 g) was
added to the solution, and stirring was continued at room
temperature for 15 min. The reaction mixture was filtered
through a Celite pad, and the pad was washed with water. The
filtrate was acidified to pH 1 using 6 N HCl and extracted with
EtOAc and CH2Cl2 to remove unreacted trifluoroacetylamino
acid. The aqueous layer was concentrated, and the pH was
brought up to 7 using a NaHCO3 solution. THF (300 mL) and
Fmoc-OSu (23.6 g, 0.07 mol) were added. The pH was adjusted
to 8 by adding additional NaHCO3. The reaction mixture was
stirred for 72 h. The solution was acidified with 6 N HCl to pH 2
and extracted with EtOAc. The organic layer was washed with
water, dried over Na2SO4, filtered, and concentrated. The
product was purified by silica gel column chromatography,
eluting with a gradient of using CH2Cl2/EtOAc starting from
100% CH2Cl2 to 10% EtOAc/CH2Cl2. Yield: 8 g, 16%.
LC-MS: (M þ H)þ at m/z 420.0;. [R]2D6 -40.8° (c 0.634% in
Preparative HPLC Purification of the Crude Peptides. Pre-
parative HPLC was carried out on either a Waters model 4000
or a Shimadzu model LC-8A liquid chromatograph. Each
solution of crude peptide was injected into a YMC S5 ODS
(20 mm ꢀ 100 mm) column and eluted using a linear gradient of
MeCN in water, both buffered with 0.1% TFA. A typical
gradient used was from 20% to 70% of 0.1% TFA/MeCN in
0.1% TFA/water over 15 min at a flow rate of 14 mL/min with
effluent UV detection at 220 nm. The desired product eluted was
well separated from impurities, typically after 10-11 min, and
was usually collected in a single 10-15 mL fraction on a fraction
collector. The desired peptides were obtained as amorphous
white powders by lyophilization.
HPLC Analysis of the Purified Peptides. Peptide purity was
determined by RP-HPLC using the conditions and methods
described below. In most cases, the purity was g95%. Excep-
tions are noted in Table S1 in the Supporting Information. After
purification by preparative HPLC, each peptide was analyzed
by analytical RP-HPLC on a Shimadzu LC-10AD or LC-10AT
analytical HPLC system consisting of a SCL-10A system con-
troller, a SIL-10A autoinjector, and a SPD10AV or SPD-M6A
UV/vis detector or a SPD-M10A diode array detector. The
column used was a YMC ODS S3 (4.6 mm ꢀ 50 mm, 3 μm)
column or a Phenomenex Luna C18 (4.6 mm ꢀ 50 mm, 5 μm)
column. Elution was performed using one of the following
gradients: 0-90% B in A over 8 min, 2.5 mL/min (method A);
20-60% B in A over 8 min, 2.5 mL/min (method B); 20-80% B
in A over 8 min, 2.5 mL/min (method C); 10-100% B in A over
8 min, 2.5 mL/min (method D); 10-80% B in A over 8 min, 2.5
mL/min (method E); 10-65% B in A over 8 min, 2.5 mL/min
(method F); 5-70% B in A over 8 min, 2.5 mL/min (method G);
30-70% B in A over 8 min, 2.5 mL/min (method H). Mobile
phase A was 0.1% TFA/water. Mobile phase B was 0.1% TFA/
acetonitrile.
1
DMF). H NMR (400 MHz, DMSO-d6): δ 7.90-7.93 (2H, d,
m), 7.73-7.76 (2H, m); 7.40-7.44 (3H, m), 7.33-7.36 (2H, m),
7.27-7.29 (1H, m), 7.10-7.14 (1H, m); 7.04-7.12 (1H, m);
6.90-6.95 (1H, m); 4.49-4.53 (1H, m); 4.20-4.30 (2H, m);
3.26-3.37 (1H, m), 3.05-3.10 (1H, m), 1.18 (3H, s). 13C NMR
(100 MHz, DMSO-d6): δ 22.3, 32.9, 46.9, 58.5.0, 65.1, 114.8,
115.0, 120.1, 123.5, 123.6, 123.9, 125.1, 125.3, 127.0, 127.6,
128.5, 128.6, 133.0, 140.8, 142.4, 143.8, 143.9, 154.9, 160.0,
162.4, 175.1.
Peptide Synthesis. The 11-mer GLP-1analogues were prepared
by solid-phase methods employing N-Fmoc/tert-butyl protection
strategy and chemistry. The following side chain protection
strategy was employed for standard amino acid residues: Asp-
(OtBu), Glu(OtBu), His(Trt), Ser(OtBu), Thr(OtBu), and Tyr-
(OtBu). Solid phase assembly was carried out in a stepwise manner
on an Advanced ChemTech MPS 396 multiple peptide synthesizer
or on an Applied Biosystems 433A peptide synthesizer using
N-[(1H-benzotriazol-1-yl)(dimethylamino)methylene]-N-methyl-
methanaminium hexafluorophosphate N-oxide (HBTU)/DIEA
or 1-hydroxy-7-azabenzotriazole (HOAt)/diisopropylcarbodii-
mide (DIC) coupling chemistry at 0.01-0.25 mmol resin scale
(Fmoc Rink or Sieber amide resin). For each coupling cycle, 5-10
equiv of N-Fmoc-amino acid, 20-40 equiv of DIEA, and 5-10
equiv of coupling agent were used. The coupling time was,
on average, 1-2 h but was extended to 16 h when coupling a
quaternary amino acid in manual mode. Fmoc deprotections were
carried out with two treatments of a 15% piperidine in DMF
solution, the first for 3 min and the second for an additional
18 min.
Peptide Cleavage and Global Deprotection. The target pep-
tides were cleaved/deprotected from their respective peptidyl
resins by treatment with a TFA cleavage mixture as follows. For
peptide arrays synthesized on the MPS 396 multiple peptide
synthesizer at 10-30 μmol/well, a solution of TFA/water/
triisopropylsilane (94:3:3) (1.0 mL) was added to each well in
the reactor block, which was then vortexed for 2 h. The TFA
solutions from the wells were collected by positive pressure into
pretared vials located in a matching 96-vial block on the bottom
of the reactor. The resins in the wells were rinsed twice with an
additional 0.5 mL of the TFA mixture, and the rinses were
combined with the solutions in the vials. These were dried in a
SpeedVac (Savant) to yield the crude peptides, typically in
>100% yields (20-60 mg). The crude peptides were either
washed with ether or more frequently redissolved directly in 2
Characterization by Mass Spectrometry. Each peptide was
characterized by electrospray mass spectrometry (ES/MS) either
in flow injection or in LC-MS mode. Finnigan SSQ7000 single
quadrupole mass spectrometers (ThermoFinnigan, San Jose,
CA) were used in all analyses in positive and negative ion
electrospray modes. Full scan data were acquired over the mass
range of 300-2200 amu for a scan time of 1.0 s. The quadrupole
was operated at unit resolution. For flow injection analyses, the
mass spectrometer was interfaced to a Waters 616 HPLC pump
(Waters Corp., Milford, MA) and equipped with an HTS PAL
autosampler (CTC Analytics, Zwingen, Switzerland). Samples
were injected into a mobile phase containing 50:50 water/acet-
onitrile with 0.1% ammonium hydroxide. The flow rate for the
analyses was 0.42 mL/min, and the injection volume was 6 μL. A
ThermoSeparations Constametric 3500 liquid chromatograph
(ThermoSeparation Products, San Jose, CA) and HTS PAL
autosampler were used for LC-MS analyses. Chromatographic
separations were achieved employing a Luna C18, 5 μm column,
2 mm ꢀ 30 mm (Phenomenex, Torrance, CA). The flow rate for
the analyses was 1.0 mL/min and column effluent was split so that
the flow into the electrospray interface was 400 μL/min. A linear
gradient from 0% to 100% B in A over 4 min was run, where
mobile phase A was 98:2 water/acetonitrile with 10 mM ammo-
nium acetate and mobile phase B was 10:90 water/acetonitrile
with 10 mM ammonium acetate. The UV response was mon-
itored at 220 nm. The samples were dissolved in 200 μL 50:50
H2O/MeCN (0.05% TFA). The injection volume was 5 μL. In all
cases, the experimentally measured molecular weight was within
0.5 Da of the calculated monoisotopic molecular weight (see
Supporting Information).
cAMP Accumulation Assay. Functional activity was assessed
by measuring the ability of the peptides to stimulate intracellular
cAMP accumulation in clonal Chinese hamster ovary (CHO)
cell lines stably overexpressing either human or mouse GLP-1

7798 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Mapelli et al.
receptors. These cell lines were cultured in F-12 nutrient medium
(Ham, Gibco) supplemented with 10% fetal bovine serum and
0.4 mg/mL (human GLP-1R) or 1 mg/mL (mouse GLP-1R)
Geneticin at 37 °C in a humidified 5% CO2 atmosphere. Cells
were plated at 2 ꢀ 104 cells/well in sterile 96-well white clear
bottom Costar plates and cultured overnight before assays.
Peptides were serially diluted at half-log concentrations in
100% DMSO and diluted to the desired test concentration in
assay buffer (Dulbecco’s phosphate-buffered saline buffer with
0.1 mM 3-isobutyl-1-methylxanthine and 0.05% bovine serum
albumin) to give a final concentration of 1% DMSO. To initiate
the assay, the medium was aspirated and the cells were treated
for 30 min with the diluted agonists (or vehicle) in assay buffer.
The cAMP in the cells was measured using the scintillation
proximity assay (SPA) method according to the manufacturer’s
protocol (Amersham). The counts per minute of 3H-cAMP
measured were converted to picomoles of cAMP by interpolat-
ing from a cAMP standard curve. The data for the test peptides
were normalized and plotted as the percentage of the response
stimulated by saturating (10 nM) GLP-1. The data were ana-
lyzed by nonlinear regression curve fitting (four-parameter
sigmoidal dose-response curve) to determine the EC50, which
was defined as the concentration of peptide that produced a
half-maximal effect. Each concentration-response was run in
duplicate in a given assay, and each peptide was assayed on at
least three separate occasions unless otherwise noted. Global
fitting to the modified Gaddum/Schild model (GraphPad Prism
software, version 4.03) was used to determine the concentration
of an antagonist that caused a 2-fold increase in the EC50 value
for cAMP accumulation.
Intraperitoneal Glucose Tolerance Tests. The protocols for all
in vivo studies were approved by the Institutional Animal Care
and Use Committee. Male ob/ob mice were purchased at 16
weeks of age (Jackson Laboratories) and were maintained on
Teklad 2018 rodent chow (Harlan) until the study was initiated.
For the study, 19 week old ob/ob mice were randomized into
groups of 6 mice per group based on fed plasma glucose and
body weight. After an 18 h overnight fast, mice were weighed
and placed in the experimental lab. After 30 min in the environ-
ment, the mice were bled via tail tip at -30 min and immediately
injected subcutaneously with vehicle (1:1 propylene glycol,
20 mM sodium phosphate, pH 7.4) or the peptides dissolved
in vehicle (0.1 mL solution/100 g body weight at a concentration
in nmol/mL equivalent to the dose that was to be administered in
nmol/kg). At time 0 the mice were bled and then injected
intraperitoneally with 50% glucose (2 g/kg) to initiate the
glucose tolerance test (IPGTT). The mice were bled 30, 60,
120, and 180 min after the glucose injection. Plasma was
prepared from blood samples that had been drawn into potas-
sium EDTA. Plasma glucose concentrations were measured
with the Cobas Mira analyzer, and plasma insulin concentra-
tions were determined using the ultrasensitive mouse insulin
ELISA kit (Crystal Chem Inc.).
precipitating plasma proteins with two volumes of acetonitrile
containing an internal standard. The samples were vortex
mixed, and the precipitated proteins were removed by centrifu-
gation. The resulting supernatants were transferred to a 96-well
plate, and an amount of 10 μL was injected for analysis. Samples
were prepared with the Packard Multiprobe II and Quadra 96
liquid handling system.
The HPLC system consisted of two Shimadzu LC10AD
pumps (Columbia, MD) and a CTC PAL autosampler (Leap
Technologies, Switzerland). The column used was a YMC
Hydrosphere C18 (2.0 mmꢀ50 mm, 3 μm) (YMC, Inc., Milford,
MA). The column temperature was maintained at 50 °C, and the
flow rate was 0.3 mL/min. The mobile phase A consisted of
10 mM ammonium formate and 0.1% formic acid in water, and
mobile phase B consisted of 0.1% formic acid in acetonitrile.
The initial mobile phase composition was 5% B and remained at
5% B for 1 min to equilibrate the column. The composition was
ramped to 95% B over 2 min and held there for an additional
1 min. The mobile phase was then returned to initial conditions
in 1 min. Total analysis time was 5 min. A switching valve was
used. The eluents between 0 and 1 min were diverted to the
waste.
The HPLC was interfaced to a Sciex API 4000 mass spectro-
meter (Applied Biosystems, Foster City, CA) and was equipped
with a TurboIonspray ionization source. Ultrahigh purity
nitrogen was used as the nebulizing and turbo gas. The tem-
perature of turbo gas was set at 300 °C, and the interface heater
was set at 60 °C. Data acquisition utilized selected reaction
monitoring (SRM). Ions representing the (M þ 2H)2þ species
for peptides 21 and 22 and for the internal standard were
selected in Q1 and were collisionally dissociated with high purity
nitrogen at a pressure of 3.5 ꢀ 10-3 torr to form specific product
ions which were subsequently monitored by Q3.
NMR Spectroscopy. NMR experiments were performed at
600 MHz using 2.8 mM samples of peptide 21 dissolved in 75%
DMSO-d6 and 25% H2O (or 25% D2O) at 10 °C. Assignments
were obtained from analysis of 2D NOESY, TOCSY, and DQF-
COSY spectra. Chemical shifts are referenced to the residual
DMSO line at 2.62 ppm. Additional details are presented in
Supporting Information.
Acknowledgment. The authors acknowledge Drs. Chet
Meyers, John Wetterau, Simeon Taylor, Robert Zahler, and
Wesley Cosand for their support of this work.
Supporting Information Available: Analytical and spectral
characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Wild, S.; Roglic, G.; Green, A.; Sicree, R.; King, H. Global
prevalence of diabetes: estimates for the year 2000 and projections
for 2030. Diabetes Care 2004, 27, 1047–1053.
(2) Holst, J. J.; Orskov, C.; Nielsen, O. V.; Schwartz, T. W. Truncated
glucagon-like peptide I, an insulin-releasing hormone from the
distal gut. FEBS Lett. 1987, 211, 169–174.
(3) Kreymann, B.; Williams, G.; Ghatei, M. A.; Bloom, S. R. Gluca-
gon-like peptide-1 7-36: a physiological incretin in man. Lancet
1987, 2, 1300–1304.
(4) Mojsov, S.; Weir, G. C.; Habener, J. F. Insulinotropin: glucagon-
like peptide I (7-37) co-encoded in the glucagon gene is a potent
stimulator of insulin release in the perfused rat pancreas. J. Clin.
Invest. 1987, 79, 616–619.
(5) Nauck, M. A.; Heimesaat, M. M.; Behle, K.; Holst, J. J.; Nauck,
M. S.; Ritzel, R.; Hufner, M.; Schmiegel, W. H. Effects of
glucagon-like peptide 1 on counterregulatory hormone responses,
cognitive functions, and insulin secretion during hyperinsulinemic,
stepped hypoglycemic clamp experiments in healthy volunteers.
J. Clin. Endocrinol. Metab. 2002, 87, 1239–1246.
Pharmacokinetics Assessment. The pharmacokinetic para-
meters of peptides 21 and 22 were determined in male beagle
dogs (n=3, 14 ( 1 kg). Following an overnight fast, each animal
received peptide by subcutaneous injection given near the
shoulder blades (67 μg/kg). Each animal received a subcuta-
neous dose with a 1-week washout between doses. The dosing
vehicle for both routes of administration was propylene glycol/
20 mM sodium phosphate buffer, pH 7.4 (1:1). Serial blood
samples were collected in EDTA-containing microcentrifuge
tubes at predose, 0.083, 0.25, 0.5, 0.75, 1, 2, 4, 6, 8, 24, and 30 h
postdose after intravenous administration and at predose, 0.25,
0.5, 0.75, 1, 2, 4, 6, 8, 24, and 30 h postdose after subcutaneous
administration. Approximately 0.3 mL of blood was collected
at each time point. Blood samples were immediately centrifuged
at 4 °C. The obtained plasma was frozen with dry ice and stored
at -20 °C. Plasma drug levels were determined using a LC-MS/
MS assay. Plasma samples were prepared for analysis by
(6) Nauck, M. A.; Niedereichholz, U.; Ettler, R.; Holst, J. J.; Orskov,
C.; Ritzel, R.; Schmiegel, W. H. Glucagon-like peptide 1 inhibition
of gastric emptying outweighs its insulinotropic effects in healthy
humans. Am. J. Physiol. 1997, 273, E981–E988.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7799
(7) Turton, M. D.; O’Shea, D.; Gunn, I.; Beak, S. A.; Edwards, C. M.;
Meeran, K.; Choi, S. J.; Taylor, G. M.; Heath, M. M.; Lambert, P.
D.; Wilding, J. P.; Smith, D. M.; Ghatei, M. A.; Herbert, J.; Bloom,
S. R. A role for glucagon-like peptide-1 in the central regulation of
feeding. Nature 1996, 379, 69–72.
(8) Davis, H. R., Jr.; Mullins, D. E.; Pines, J. M.; Hoos, L. M.; France,
C. F.; Compton, D. S.; Graziano, M. P.; Sybertz, E. J.; Strader, C.
D.; Van Heek, M. Effect of chronic central administration of
glucagon-like peptide-1 (7-36) amide on food consumption and
body weight in normal and obese rats. Obes. Res. 1998, 6, 147–156.
(9) Zander, M.; Madsbad, S.; Madsen, J. L.; Holst, J. J. Effect of
6-week course of glucagon-like peptide 1 on glycaemic control,
insulin sensitivity, and beta-cell function in type 2 diabetes: a
parallel-group study. Lancet 2002, 359, 824–830.
(10) Drucker, D. J. Glucagon-like peptides: regulators of cell proliferation,
differentiation, and apoptosis. Mol. Endocrinol. 2003, 17, 161–171.
(11) Larsen, J.; Hylleberg, B.; Ng, K.; Damsbo, P. Glucagon-like
peptide-1 infusion must be maintained for 24 h/day to obtain
acceptable glycemia in type 2 diabetic patients who are poorly
controlled on sulphonylurea treatment. Diabetes Care 2001, 24,
1416–1421.
(12) Vilsboll, T.; Agerso, H.; Krarup, T.; Holst, J. J. Similar elimination
rates of glucagon-like peptide-1 in obese type 2 diabetic patients
and healthy subjects. J. Clin. Endocrinol. Metab. 2003, 88, 220–224.
(13) Mentlein, R. Dipeptidyl-peptidase IV (CD26);role in the inacti-
vation of regulatory peptides. Regul. Pept. 1999, 85, 9–24.
(14) Holst, J. J. Glucagon-like peptide-1: from extract to agent. The
Claude Bernard Lecture, 2005. Diabetologia 2006, 49, 253–260.
(15) Eng, J.; Kleinman, W. A.; Singh, L.; Singh, G.; Raufman, J. P.
Isolation and characterization of exendin-4, an exendin-3 analo-
gue, from Heloderma suspectum venom. Further evidence for an
exendin receptor on dispersed acini from guinea pig pancreas.
J. Biol. Chem. 1992, 267, 7402–7405.
(16) Buse, J. B.; Henry, R. R.; Han, J.; Kim, D. D.; Fineman, M. S.;
Baron, A. D. Effects of exenatide (exendin-4) on glycemic control
over 30 weeks in sulfonylurea-treated patients with type 2 diabetes.
Diabetes Care 2004, 27, 2628–2635.
(17) Green, B. D.; Flatt, P. R. Incretin hormone mimetics and analo-
gues in diabetes therapeutics. Best Pract. Res., Clin. Endocrinol.
Metab. 2007, 21, 497–516.
(18) Chen, D.; Liao, J.; Li, N.; Zhou, C.; Liu, Q.; Wang, G.; Zhang, R.;
Zhang, S.; Lin, L.; Chen, K.; Xie, X.; Nan, F.; Young, A. A.; Wang,
M. W. A nonpeptidic agonist of glucagon-like peptide 1 receptors
with efficacy in diabetic db/db mice. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 943–948.
(19) Knudsen, L. B.; Kiel, D.; Teng, M.; Behrens, C.; Bhumralkar, D.;
Kodra, J. T.; Holst, J. J.; Jeppesen, C. B.; Johnson, M. D.; de Jong,
J. C.; Jorgensen, A. S.; Kercher, T.; Kostrowicki, J.; Madsen, P.;
Olesen, P. H.; Petersen, J. S.; Poulsen, F.; Sidelmann, U. G.; Sturis,
J.; Truesdale, L.; May, J.; Lau, J. Small-molecule agonists for the
glucagon-like peptide 1 receptor. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 937–942.
glucagon release from the isolated perfused rat pancreas. Endocri-
nology 1989, 125, 3109–3114.
(24) Adelhorst, K.; Hedegaard, B. B.; Knudsen, L. B.; Kirk, O.
Structure-activity studies of glucagon-like peptide-1. J. Biol.
Chem. 1994, 269, 6275–6278.
(25) Gallwitz, B.; Witt, M.; Paetzold, G.; Morys-Wortmann, C.;
Zimmermann, B.; Eckart, K.; Folsch, U. R.; Schmidt, W. E.
Structure/activity characterization of glucagon-like peptide-1.
Eur. J. Biochem. 1994, 225, 1151–1156.
(26) Siegel, E. G.; Gallwitz, B.; Scharf, G.; Mentlein, R.; Morys-
Wortmann, C.; Folsch, U. R.; Schrezenmeir, J.; Drescher, K.;
Schmidt, W. E. Biological activity of GLP-1-analogues with
N-terminal modifications. Regul. Pept. 1999, 79, 93–102.
(27) Deacon, C. F.; Knudsen, L. B.; Madsen, K.; Wiberg, F. C.;
Jacobsen, O.; Holst, J. J. Dipeptidyl peptidase IV resistant analo-
gues of glucagon-like peptide-1 which have extended metabolic
stability and improved biological activity. Diabetologia 1998, 41,
271–278.
(28) Ritzel, U.; Leonhardt, U.; Ottleben, M.; Ruhmann, A.; Eckart, K.;
Spiess, J.; Ramadori, G. A synthetic glucagon-like peptide-1
analog with improved plasma stability. J. Endocrinol. 1998, 159,
93–102.
(29) Knudsen, L. B.; Nielsen, P. F.; Huusfeldt, P. O.; Johansen, N. L.;
Madsen, K.; Pedersen, F. Z.; Thogersen, H.; Wilken, M.; Agerso,
H. Potent derivatives of glucagon-like peptide-1 with pharmaco-
kinetic properties suitable for once daily administration. J. Med.
Chem. 2000, 43, 1664–1669.
(30) Madsen, K.; Knudsen, L. B.; Agersoe, H.; Nielsen, P. F.; Thogersen,
H.; Wilken, M.; Johansen, N. L. Structure-activity and protraction
relationship of long-acting glucagon-like peptide-1 derivatives:
importance of fatty acid length, polarity, and bulkiness. J. Med.
Chem. 2007, 50, 6126–6132.
(31) Parker, J. C.; Andrews, K. M.; Rescek, D. M.; Massefski, W., Jr.;
Andrews, G. C.; Contillo, L. G.; Stevenson, R. W.; Singleton, D.
H.; Suleske, R. T. Structure-function analysis of a series of
glucagon-like peptide-1 analogs. J. Pept. Res. 1998, 52, 398–409.
(32) Raufman, J. P.; Singh, L.; Singh, G.; Eng, J. Truncated glucagon-
like peptide-1 interacts with exendin receptors on dispersed acini
from guinea pig pancreas. Identification of a mammalian analogue
of the reptilian peptide exendin-4. J. Biol. Chem. 1992, 267, 21432–
21437.
(33) Thornton, K.; Gorenstein, D. G. Structure of glucagon-like pep-
tide (7-36) amide in a dodecylphosphocholine micelle as deter-
mined by 2D NMR. Biochemistry 1994, 33, 3532–3539.
(34) Murage, E. N.; Schroeder, J. C.; Beinborn, M.; Ahn, J. M. Search
for alpha-helical propensity in the receptor-bound conformation of
glucagon-like peptide-1. Bioorg. Med. Chem. 2008, 16, 10106–
10112.
(35) Agerso, H.; Jensen, L. B.; Elbrond, B.; Rolan, P.; Zdravkovic, M.
The pharmacokinetics, pharmacodynamics, safety and tolerability
of NN2211, a new long-acting GLP-1 derivative, in healthy men.
Diabetologia 2002, 45, 195–202.
(20) Rink, H. Solid-phase synthesis of protected peptide fragments
using a trialkoxy-diphenyl-methylester resin. Tetrahedron Lett.
1987, 28, 3787–3790.
(21) King, D. S.; Fields, C. G.; Fields, G. B. A cleavage method which
minimizes side reactions following Fmoc solid phase peptide
synthesis. Int. J. Pept. Protein Res. 1990, 36, 255–266.
(22) Gefel, D.; Hendrick, G. K.; Mojsov, S.; Habener, J.; Weir, G. C.
Glucagon-like peptide-I analogs: effects on insulin secretion and
adenosine 30,50-monophosphate formation. Endocrinology 1990,
126, 2164–2168.
(23) Suzuki, S.; Kawai, K.; Ohashi, S.; Mukai, H.; Yamashita, K.
Comparison of the effects of various C-terminal and N-terminal
fragment peptides of glucagon-like peptide-1 on insulin and
(36) Elbrond, B.; Jakobsen, G.; Larsen, S.; Agerso, H.; Jensen, L. B.;
Rolan, P.; Sturis, J.; Hatorp, V.; Zdravkovic, M. Pharmacoki-
netics, pharmacodynamics, safety, and tolerability of a single-dose
of NN2211, a long-acting glucagon-like peptide 1 derivative, in
healthy male subjects. Diabetes Care 2002, 25, 1398–1404.
(37) Fineman, M. S.; Shen, L. Z.; Taylor, K.; Kim, D. D.; Baron, A. D.
Effectiveness of progressive dose-escalation of exenatide (exendin-
4) in reducing dose-limiting side effects in subjects with type 2
diabetes. Diabetes Metab. Res. Rev. 2004, 20, 411–417.
(38) Ritzel, R.; Orskov, C.; Holst, J. J.; Nauck, M. A. Pharmacokinetic,
insulinotropic, and glucagonostatic properties of GLP-1 [7-36
amide] after subcutaneous injection in healthy volunteers. Dose-
response-relationships. Diabetologia 1995, 38, 720–725.