Discovery of potent, cyclic calcitonin gene-related peptide receptor antagonists

Article abstract of DOI:10.1002/psc.1358

Calcitonin gene-related peptide (CGRP), a potent dilator of cerebral and dural vasculature, is known to be elevated in plasma and cerebral spinal fluid during migraine attacks. Selective blockade of the CGRP receptor offers the promise of controlling migraine headache more effectively and without the side-effects associated with the use of triptans. Our efforts to develop a novel, peptide-based CGRP antagonist focused on the C-terminal portion of the peptide which is known to bind the receptor but lack agonist properties. Extensive SAR studies of the C-terminal CGRP (27-37) region identified a novel cyclic structure: Bz-Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH₂ (23) with a kb value of 0.126 nM against the cloned human CGRP receptor. Additional SAR studies directed at enhancement of potency and improvement of physicochemical properties yielded a series of analogs with kb values in the 0.05-0.10 nM range.

Full text of DOI:10.1002/psc.1358

Research Article
Received: 1 October 2010
Revised: 4 January 2011
Accepted: 11 January 2011
Published online in Wiley Online Library: 15 March 2011
(wileyonlinelibrary.com) DOI 10.1002/psc.1358
Discovery of potent, cyclic calcitonin
gene-related peptide receptor antagonists‡
Liang Zeng Yan, Kirk W. Johnson, Emily Rothstein, David Flora,
Patrick Edwards, Baolin Li, Junqing Li, Renee Lynch, Renee Vaughn,
Amy Clemens-Smith, Deborah McCarty, Charles Chow, Kevin L. McKnight,
Jirong Lu, Eric S Nisenbaum and John P. Mayer∗
Calcitonin gene-related peptide (CGRP), a potent dilator of cerebral and dural vasculature, is known to be elevated in plasma
and cerebral spinal fluid during migraine attacks. Selective blockade of the CGRP receptor offers the promise of controlling
migraine headache more effectively and without the side-effects associated with the use of triptans. Our efforts to develop a
novel, peptide-based CGRP antagonist focused on the C-terminal portion of the peptide which is known to bind the receptor
but lack agonist properties. Extensive SAR studies of the C-terminal CGRP (27–37) region identified a novel cyclic structure:
Bz-Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2 (23) with a kb value of 0.126 nM against the cloned human CGRP
receptor. Additional SAR studies directed at enhancement of potency and improvement of physicochemical properties yielded
c
a series of analogs with kb values in the 0.05–0.10 nM range. Copyright ꢀ 2011 European Peptide Society and John Wiley &
Sons, Ltd.
Keywords: CGRP; migraine; peptide antagonists; solubility
Introduction
N-terminal, disulfide bond-containing region is primarily involved
in receptor activation, while the central and C-terminal regions are
thought to be responsible for receptor binding. Deletion of the
N-terminal heptapeptide from the native CGRP sequence yields
the competitive antagonist CGRP 8–37 reported by Chiba [11,12]
as well as more potent analogs disclosed recently by Miranda [13].
Further truncations have produced fragments with substantially
diminished receptor affinity [14]. Our efforts to develop a novel
CGRP antagonist focused on the C-terminal region encompassing
residues 27–37, which had previously been shown to be the
minimal fragment necessary for receptor binding [15,16]. High
in vitro potency was a primary goal for this series, driven partly by
the need to achieve high selectivity in order to minimize off-target
activity and facilitate alternatives to administration by injection.
The relatively low bioavailability associated with delivery methods
such as intranasal inhalation elevates the need for high potency to
achieve an acceptable commercial cost in peptide manufacture.
Migraine is a common neurovascular disorder characterized by
episodes of severe headache, nausea, and increased sensitivity
to light and sound lasting from 4 to 72 h [1,2]. Conventional
therapeutic agents include disease nonspecific remedies such
as aspirin, acetaminophen, and nonsteroidal anti-inflammatory
drugs, which are sometimes used in combination with ergotamine
or caffeine. The 5HT1B/1D selective agonist class of ‘triptans’
such as sumatriptan and rizatriptan revolutionized migraine
therapy and continue to offer an important therapeutic option
for migraine sufferers [3]. The triptans, however, fail to provide
pain relief in more than 50% of patients and are in addition
contraindicated in individuals with cardiovascular disease [4]. This
deficiency has stimulated extensive investigation of alternative
mechanism-based approaches for both prevention and treatment
of migraine headache [5]. One of the most promising is based on
the selective blockade of calcitonin gene-related peptide (CGRP)
interaction with its receptor. CGRP is a neuropeptide which is
expressed in both the central and peripheral nervous systems and
is thought to be a likely causative factor in migraine pathogenesis
[6,7]. This was demonstrated in a double-blind crossover study of
migraine sufferers who experienced attacks following intravenous
administration of human CGRP (hCGRP) [8]. Validation of antago-
nismcamefromaclinicalstudyofBIBN4096, apotentnon-peptide
CGRP antagonist which delivered rapid symptomatic relief with a
high overall response rate following intravenous administration to
migraineurs [9]. Despite these favorable clinical results, however,
the therapeutic utility of BIBN 4096 is limited by intravenous
administration. Structurally, α-CGRP(1)isa37aminoacidpolypep-
tide with a disulfide bridge spanning residues Cys2 and Cys7 [10].
Structure–activity relationship studies have determined that the
Methods
Peptide Synthesis
All peptide sequences were assembled by automated α-
Fmoc chemistry starting from 4-(2ꢁ,4ꢁ-dimethoxyphenyl-Fmoc-
∗
Correspondence to: John P. Mayer, Lilly Research Laboratories, Indianapolis, IN
46285, USA. E-mail: j.mayer@lilly.com
Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, IN 46285, USA
‡
Special issue devoted to contributions presented at the 12th Naples Worshop
on Bioactive Peptides and 2nd Italy-Korea Symposium on Antimicrobial
Peptides, 4–7 June 2010, Naples, Italy.
c
J. Pept. Sci. 2011; 17: 383–386
Copyright ꢀ 2011 European Peptide Society and John Wiley & Sons, Ltd.

YAN ET AL.
aminomethyl)-phenoxy ‘Rink’ resin on an ABI 433 Peptide Syn-
thesizer (PE Applied Biosystems Inc., Foster City, CA, USA) using
the manufacturer’s standard FastMoc HBTU/DIEA protocols. The
N-α-Fmoc-protected amino acids utilized the following protecting
group scheme: Arg(Pbf), Asn(Trt), Asp(OtBu), Cys(Trt), Agp(Boc)2
(Agp: α-amino-3-guanidino-propionic acid), Dap(Boc) (Dap: 1,3
diamino-propionic acid), His(Trt), Hyp(tBu) (Hyp: hydroxy pro-
line), Lys(Boc), Orn(Boc), Ser(tBu), Thr(tBu), Tyr(tBu). Other building
blocks, includingFmoc-CitandFmoc-3Pal(3Pal:3-pyridyl-alanine),
were used without side chain protection. Following peptide
chain assembly, the N-terminal Fmoc group was removed us-
ing 20% piperidine in DMF. N-terminal modification was carried
out using five equivalents of either the appropriate anhydride or
activated carboxylic acid in DMF or N-methyl-2-pyrrolidinone for
1 h at room temperature. The linear peptides were simultaneously
deprotected and cleaved from the resin using a scavenger cock-
tail of TFA/H2O/triisopropylsilane (TIS)/1,2-ethanedithiol (EDT)
(95/2/1/2, v/v/v/v) for 2 h at room temperature. The solvents
were removed under reduced pressure and the crude peptides
precipitated and washed with cold diethyl ether three times to
remove scavengers, then dried under vacuum. The oxidation of
the free cysteine sulfhydryl groups was carried out in 0.2 M am-
monium acetate buffer containing 20% DMSO overnight followed
by purification using preparative HPLC techniques. Final charac-
terization was performed using analytical HPLC and mass spectral
analysis with the data summarized in Table 1.
In vitro activity was assessed by measurement of cyclic adeno-
sine monophosphate (cAMP) using SK-N-MC cells, a human
neuroepitheliomacelllineendogenouslyexpressingthehCGRPre-
ceptor1purchasedfromReceptorBiology(Beltsville,MD,USA).SK-
N-MC cells were grown in minimal essential medium(Invitrogen,
Carlsbad, CA, USA) supplemented with 10% fetal bovine serum,
2 mM L-glutamine, 0.1 mM nonessential amino acids, 1 mM sodium
pyruvate, 100 U/ml penicillin/streptomycin at 37 ◦C incubator
with 95% humidity and 5% CO2. For cAMP measurement, SK-N-
MC cells were dissociated from the flask with enzyme-free cell
dissociation solution (Specialty Media, S-014-B, Billerica, MA, USA),
then suspended in the cAMP assay stimulation buffer (100 ml
of Hank’s balanced salt solution supplemented with 5 mM 4-(2-
hydroxyethyl)-1-piperazineethanesulfonic acid, 0.1% BSA, 100 µM
ascorbic acid + 200 ml of Dulbecco’s phosphate buffered saline
+ 300 µl of 500 mM 3-isobutyl-1-methylxanthine). The reaction
was performed in low volume (0.5 ml) black polystyrene 96 well
plates (Costar, Lowell, MA, USA). Each well contained approx-
imately 15 000 SK-N-MC cells and 2 nM hCGRP for the hCGRP
receptor cAMP assay. The reaction was carried out in the presence
of various concentrations of CGRP receptor antagonists. One hour
after incubation at room temperature, cAMP levels were mea-
sured using an homogeneous time-resolved fluorescence cAMP
assay kit (Dynamic 2 Bulk, Cisbio, Bedford, MA, USA) according to
the manufacturer’s instructions. EC50 or IC50 values were calcu-
lated by fitting the competition curves with Graphpad Prism 4.01.
(GraphPad Software, Inc., La Jolla, CA 92037 USA) kb values were
calculated by using the Cheng–Prusoff equation:
Table 1. Characterization of new calcitonin gene-related peptide
antagonist peptides
Theoretical Observed HPLC purity HPLCa Saline solution
No. MW (Da)
MW (Da)
(%)
RT (min)
(mg/ml)b
2
1165.35
1195.38
1211.39
1261.45
1273.47
1259.48
1275.44
1317.52
1277.45
1227.39
1303.49
1277.39
1089.27
1089.27
1089.27
1089.27
1063.23
1013.17
1344.59
1394.65
1247.47
1351.58
1332.57
1380.62
1378.65
1406.66
1047.64
1364.62
1336.57
1337.55
1337.55
1337.55
1275.48
1355.54
1397.58
1165.40
1195.40
1211.76
1261.79
1273.79
1259.82
1275.78
1317.82
1277.78
1227.8
96.75
97.61
96.43
99.62
99.9
20.61
30.82
30.58
34.23
33.83
32.49
30.43
31.81
30.65
25.77
31.59
26.43
37.98
40.2
>50.00
<0.10
n/a
3
4
5
n/a
6
n/a
7
99.79
99.77
98.7
n/a
8
n/a
9
n/a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
98.32
99.16
99.66
97.58
91.61
99.038
99.02
96.93
98.71
99.81
97.90
95.34
94.93
97.36
94.76
98.91
94.85
90.01
96.72
91.99
95.73
98.55
97.6
n/a
n/a
1303.83
1227.79
1089.58
1089.58
1089.33
1089.69
1063.63
1013.64
1344.67
1394.76
1246.91
1351.71
1332.76
1380.77
1378.78
1405.96
1406.75
1364.10
1336.45
1337.94
1338.06
1337.19
1275.30
1355.19
1397.47
n/a
n/a
n/a
n/a
37.87
37.95
36.27
29.73
36.52
38.70
34.55
41.73
33.13
38.20
39.43
39.82
40.10
39.43
39.88
29.52
31.07
31.12
26.72
32.20
32.10
n/a
n/a
n/a
n/a
n/a
n/a
n/a
0.02
4.67
n/a
0.17
0.05
0.03
0.04
0.05
4.57
6.70
10.00
15.40
4.30
1.00
98.7
97.08
98.80
98.58
MW, molecular weight.
a HPLCconditions:mobilephase:bufferA,waterwith0.1%TFA;bufferB,
acetonitrilewith0.1%TFA. Flowrate:1.0 ml/min;gradient:10%B5 min,
10–55% B 45 min, 90% B 5 min, 10% B 5 min. Column: Phenomenex,
Jupiter C18 (0.46 × 25 cm), 5 µm, 300 Å.
b Solubility measurement: weighted out 1.0 mg of peptide material
and dissolved in 0.5 ml of saline (0.9% sodium chloride), then removed
any non-dissolved material by centrifugation. Concentration was
calculated based on the measurement of the UV absorption (extinction
coefficients: 2200 M−1 cm−1 for 3-pyridyl-alanine).
Results
Our lead optimization strategy focused on the C-terminal CGRP
27–37 fragment (2) previously explored by Beck-Sickinger and
coworkers who reported several potent antagonists including
Asp31, Pro34, Phe35-CGRP 27–37 (3), and Hyp29, Asp31, Pro34
,
Phe35 CGRP 27–37 (4) [15–17]. The discovery of Tyr0CGRP 28–37
by Chakder and Rattan [18,19] resulting from the substitution
of Tyr for Phe at position 27, prompted us to conduct a Tyr
scan of peptide analog 4 in order to optimize the position of
this residue (peptides 5–13). In vitro evaluation of this series
kb = IC50/(1 + ([agonist]/agonist EC50))
and summarized in Table 2.
c
wileyonlinelibrary.com/journal/jpepsci Copyright ꢀ 2011 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2011; 17: 383–386

POTENT CYCLIC CALCITONIN GENE-RELATED PEPTIDE RECEPTOR ANTAGONISTS
Table 2. Antagonistic data of CGRP peptides in cAMP assay
No.
Structure
hCGRP1 cAMP Antag kb(nM)
H-Ala-cyclo[Cys-Asp-Thr-Ala-Thr-Cys]-Val-Thr-His-Arg-Leu-Ala-Gly
Leu-Leu-Ser-Arg-Ser-Gly-Gly-Val-Val-Lys-Asn-Asn-Phe-Val-Pro-Thr-Asn-Val-Gly-Ser-Lys-Ala-Phe-NH2
Phe-Val-Pro-Thr-Asn-Val-Gly-Ser-Lys-Ala-Phe-NH2
1
n/a
>100
2
3
Phe-Val-Pro-Thr-Asp-Val-Gly-Pro-Phe-Ala-Phe-NH2
1.72 1.19
1.12 1.51
13.3 13.7
>100
4
Phe-Val-Hyp-Thr-Asp-Val-Gly-Pro-Phe-Ala-Phe-NH2
5
Phe-Val-Tyr-Thr-Asp-Val-Gly-Pro-Phe-Ala-Phe-NH2
6
Phe-Val-Hyp-Tyr-Asp-Val-Gly-Pro-Phe-Ala-Phe-NH2
7
Phe-Val-Hyp-Thr-Tyr-Val-Gly-Pro-Phe-Ala-Phe-NH2
1.73 0.49
>100
8
Phe-Val-Hyp-Thr-Asp-Tyr-Gly-Pro-Phe-Ala-Phe-NH2
9
Phe-Val-Hyp-Thr-Asp-Val-Tyr-Pro-Phe-Ala-Phe-NH2
>100
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Phe-Val-Hyp-Thr-Asp-Val-Gly-Tyr-Phe-Ala-Phe-NH2
>100
Phe-Val-Hyp-Thr-Asp-Val-Gly-Pro-Tyr-Ala-Phe-NH2
7.09 0.00
>100
Phe-Val-Hyp-Thr-Asp-Val-Gly-Pro-Phe-Tyr-Phe-NH2
Phe-Val-Hyp-Thr-Asp-Val-Gly-Pro-Phe-Ala-Tyr-NH2
3.20 0.35
>100
Bz-Thr-Asp-Val-cyclo[Cys-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Thr-Asp-cyclo[Cys-Val-Gly-Pro-Phe-Cys]-Phe-NH2
>100
Bz-Thr-cyclo[Cys-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
>100
Bz-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
9.14 6.68
>100
Bz-cyclo[Cys-Thr-Asp-Val-Gly-Cys]-Phe-Ala-Phe-NH2
Bz-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Cys]-Ala-Phe-NH2
>100
Phe-Val-Hyp-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Phe-Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Thr-Asp-Val-Gly-Pro-Lys-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Thr-Asp-Lys-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Lys-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Arg-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Cit-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Orn-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-Phe-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-3Pal-Cys]-Phe-NH2
Bz-Val-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-Phe-Cys]-3Pal-NH2
Bz-(D-Val)-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-Phe-Cys]-3Pal-NH2
Ac-(D-Val)-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-Phe-Cys]-3Pal-NH2
(4-F-Bz)-(D-Val)-Tyr-cyclo[Cys-Dap-Asp-Val-Gly-Pro-Phe-Cys]-3Pal-NH2
(4-F-Bz)-(D-Val)-Tyr-cyclo[Cys-Agp-Asp-Val-Gly-Pro-Phe-Cys]-3Pal-NH2
59.4 16.4
4.73 3.29
1.72 0.00
0.126 0.042
0.761 0.388
51.0 5.6
0.184 0.093
0.122 0.079
0.277 0.202
0.305 0.068
0.106 0.068
0.351 0.042
0.290 0.230
0.0436 0.0149
0.103 0.042
0.0491 0.0032
0.0286 0.0230
CGRP, calcitonin gene-related peptide; hCGRP, human CGRP; cAMP, cyclic adenosine monophosphate; 3Pal, 3-pyridyl-alanine.
revealed that Tyr could be tolerated at positions 27, 29, 35,
and 37 (peptides 5, 7, 11, 13) with only modest attenuation of
activity. We next applied disulfide constraint in order to stabilize
the β-turn in this region as noted by Carpenter [20], which
produced mainly inactive peptides (14–19), except analog 17
with a respectable potency of 9.14 nM. We attempted to boost
potency further by modifying the N-terminus of cyclic analog
17 with the N-terminal tripeptide from the linear series. The
addition of the Phe-Val-Hyp (Hyp: 4-hydroxy proline) tripeptide
resulted in disappointing potency (peptide 20, kb = 59.4 nM),
however the substitution of Tyr for Hyp (shown to be tolerated
in the linear series) resulted in an improvement to 4.73 nM
(21).
hereinafter. Deletion of Phe1 gave a still better kb value of 1.72
nM (22), while N-benzoylation of the free N-terminus resulted in
the highly potent analog 23 with a kb value of 0.126 nM. While
gratifying in terms of receptor affinity, we noted that this peptide
exhibited unacceptably low aqueous solubility (0.02 mg/ml in
saline) which complicated plans for in vivo studies.
Attempts to enhance solubility of this peptide included
substitution of Lys at several positions which revealed that while
this residue could not be tolerated at position 10 (analog 24), it
could be introduced at position 7 and 5 without significant loss
of potency (analog 25 and 26). Analog 26 exhibited a potency of
0.184 nM as well as a slightly improved solubility of 0.17 mg/ml.
At this point, we sought to optimize the side chain at position 5
in terms of steric parameters and charge. The Arg (27), Cit (28),
Orn (29), and Dap (30) analogs demonstrated excellent potency
but uniformly low solubility. Next, we turned to the replacement
of one or both Phe residues with a functionally equivalent but
Insertion of the disulfide bridge necessitated a renumbering
from the linear CGRP convention to the following: Phe1-Val2-Tyr3-
Cys4-Thr5-Asp6-Val7-Gly8-Pro9-Phe10-Cys11-Phe12-NH2 (21). This
numbering applies to all cyclic peptides in the series discussed
c
J. Pept. Sci. 2011; 17: 383–386 Copyright ꢀ 2011 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

YAN ET AL.
more polar side chain in order to enhance solubility. Substitution
of 3Pal for Phe10 (analog 31) resulted in a kb value of 0.351 nM,
but a much improved solubility of 4.57 mg/ml. Substitution of 3Pal
for the exocyclic Phe12 gave a similar potency of 0.290 nM with a
solubility of 6.70 mg/ml (analog 32). In an attempt to restore the
higher potency of the earlier, less soluble analogs, we carried out
a series of iterative modification to the N-terminus. Inversion of
configurationatVal2 producedanalog33withasurprisingboostin
potencyto0.0436nM, alongwithfurtherimprovementinsolubility
of 10.0 mg/ml. The choice of the N-terminal capping group was
noted to be an important factor in achieving a balance between
potency and solubility. Additional improvement in solubility to
15.4 mg/ml was achieved with replacement of the hydrophobic
N-benzoyl group with an N-acetyl group (analog 34, kb = 0.103
nM). Conversely, the 4-fluorobenzoyl group (analog 35, kb =
0.0491 nM) gave better potency at the cost of lower solubility of
4.30 mg/ml. The best absolute potency of the series was obtained
by incorporating the strongly basic Agp (α-Amino-3-guanidino-
propionic acid) group in place of Dap at position 5 which resulted
in remarkable potency of 0.0286 nM, but diminished solubility of
1.0 mg/ml (analog 36).
4 Dodick D, Lipton RB, Martin V, Papademetriou V, Rosamond W,
Maassen VanDenBrink A, Loutfi H, Welch KM, Goadsby PJ, Hahn S,
Hutchinson S, Matchar D, Silberstein S, Smith TR, Purdy A, Saiers J,
Consensus Statement: Cardiovascular Safety Profile of Triptans (5-
HT1B/1D Agonists) in the Acute Treatment of Migraine. Headache
2004; 44: 414–425.
5 Ramadan NM, Buchanan TM. New and future migraine therapy.
Pharmacol. Ther. 2006; 112: 199–212.
6 Edvinsson L. New therapeutic target in primary headaches-blocking
the CGRP receptor. Expert. Opin. Ther. Targets 2003; 7: 377–383.
7 Edvinsson L. CGRP blockers in migraine therapy; where do they act?
Br. J. Pharmacol. 2008; 155: 967–969.
8 Lassen LH, Haderslev PA, Jacobsen VB, Iversen HK, Sperling B,
Olesen J. CGRP may play a causative role in migraine. Cephalalgia
2002; 22: 54–61.
9 Olesen J, Diener H-C, Husstedt IW, Goadsby PJ, Hall D, Meier U,
Pollentier S, Lesko LM. Calcitonin gene-related peptide receptor
antagonist BIBN 4096 for the acute treatment of migraine. N. Engl. J.
Med. 2004; 350: 1104–1110.
10 Morris HR, Panico M, Etienne T, Tippins J, Girgis SI, MacIntyre I.
Isolation and characterization of calcitonin gene-related peptide.
Nature 1984; 308: 746–748.
11 Chiba T, Yamaguchi A, Yamatani T, Nakamura A, Morishita T, Inui T,
Fukase M, Noda T, Fujita T. Calcitonin gene-related peptide receptor
antagonist human CGRP-(8–37). Am. J. Physiol. 1989; 256:
E331–E335.
12 Maggi CA, Chiba T, Giuliani S. Human α-calcitonin gene-related
peptide (8–37) as an antagonist of exogenous and endogenous
calcitonin gene-related peptide. Eur. J. Pharmacol. 1991; 192: 85–88.
13 Miranda LP, Holder RJ, Shi L, Bennett B, Aral J, Cegg CV, Wright M,
Walker K, Doellgast G, Rogers R, Hongyan L, Valladares V, Salyers K,
Johnson E, Wild K. Identification of potent, selective, and
metabolically stable peptide antagonists to the calcitonin gene-
related peptide (CGRP) receptor. J. Med. Chem. 2008; 51: 7889–7897.
14 Mimeault M, Fournier A, Dumont Y, St-Pierre S, Quirion R.
Comparative affinities and antagonistic potencies of various human
CGRP fragments on CGRP receptors in brain and periphery.
J. Pharmacol. Exp. Ther. 1991; 258: 1084–1090.
15 Rist B, Entzeroth M, Beck-Sickinger AG. From micromolar to
nanomolaraffinity:asystematicapproachtoidentifythebindingsite
of CGRP at the human calcitonin gene-related peptide 1 receptor.
J. Med. Chem. 1998; 41: 117–123.
16 Rist B, Lacroix JS, Enzenroth M, Doods HN, Beck-Sickinger AG. CGRP
27–37 analogs with high affinity to the CGRP1 receptor show
antagonistic properties in the rat blood flow assay. Regul. Pept. 1999;
79: 153–158.
Discussion
Our ligand optimization strategy which combined positional
scanninganduseofdisulfide-inducedconstraintyieldedanumber
of potent cyclic peptides with subnanomolar kb values at the
hCGRP receptor. An unexpected challenge that surfaced in this
peptide series pertained to the generally low solubility observed
among linear and cyclic analogs. We attribute this phenomenon
to the substitution of Ser34Lys35 with Pro34Phe35, a modification
which significantly enhanced potency but caused a dramatic drop
insolubilitywhencomparinganalogs2and3.Literatureprecedent
suggests that the presence of proline in a β-turn conformation
can increase the propensity for aggregation. Studies by Morimoto
et al. noted extensive aggregation when proline was introduced
into the β-turn region of amyloid peptides [21]. Since Pro34 was
a prerequisite for high potency in our series, we investigated
alternative methods of restoring solubility. Modification of the
aromatic side chains at positions 10 and 12 with a 3-pyridyl group
offered a practical solution to this problem. These substitutions
improved the aqueous solubility from less than 0.1 mg/ml to
a range of 5–15 mg/ml while maintaining potencies in the
0.050–0.10 nM range.
17 Lang M, De Pol S, Baldauf C, Hofmann H-J, Reiser O, Beck-
Sickinger AG. Identification of the key residue of calcitonin gene-
related peptide (CGRP) 27–37 to obtain antagonists with picomolar
affinity at the CGRP receptor. J. Med. Chem. 2006; 49: 616–624.
18 Chakder S, Rattan S. [Tyr0]-calcitonin gene-related peptide 28–37
(rat) as a putative antagonist of calcitonin gene-related peptide
responses on opossum internal anal sphincter smooth muscle.
J. Pharm. Exp. Ther. 1990; 253: 200–206.
19 Maton PN, Pradhan T, Zhou Z-C, Gardner JD, Jensen RT. Activities of
calcitonin gene-related peptide (CGRP) and related peptides at the
CGRP receptor. Peptides 1990; 11: 485–489.
20 Carpenter KA, Schmidt R, von Mentzer B, Haglund U, Roberts E,
Walpole C. Turn structures in CGRP C-terminal analogues promote
stable arrangements of key residue side chains. Biochemistry 2001;
40: 8317–8325.
21 Morimoto A, Kazuhiro I, Murakami K, Ohigashi H, Shindo M,
Nagao M, Shimizu T, Shirasawa T. Aggregation and neurotoxicity
of mutant amyloid β (Aβ) peptides with proline replacement:
importance of turn formation at positions 22 and 23. Biochem.
Biophys. Res. Commun. 2002; 295: 306–311.
References
1 Scher AI, Stewart WF, Liberman J, Lipton RB. Prevalence of frequent
headache in a population sample. Headache 1998; 38: 497–506.
2 Lipton RB, Diamond S, Reed M, Diamond ML, Stewart WF. Migraine
diagnosis and treatment: results from the American migraine study
II. Headache 2001; 41: 638–645.
3 Ferrari MD, Roon KI, Lipton RB, Goadsby PJ. Oral triptans (serotonin
5-HT1B/1D agonists) in acute migraine treatment: a meta-analysis of
53 trials. Lancet 2001; 358: 1666–1675.
c
wileyonlinelibrary.com/journal/jpepsci Copyright ꢀ 2011 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2011; 17: 383–386