Synthesis of oxytocin analogues with replacement of sulfur by carbon gives potent antagonists with increased stability

Article abstract of DOI:10.1021/jo050539l

The neuropeptide oxytocin 1 controls mammary and uterine smooth muscle contraction. Atosiban 2, an oxytocin antagonist, is used for prevention of preterm labor and premature birth. However, the metabolic lifetimes of such peptide drugs are short because of in vivo degradation. Facile production of oxytocin analogues with varying ring sizes wherein sulfur is replaced by carbon (methylene or methine) could be achieved by standard solid-phase peptide synthesis using olefin-bearing amino acids followed by on-resin ring-closing metathesis (RCM). These were tested for agonistic and antagonistic uteronic activity using myometrial strips taken from nonpregnant female rats. Peptide 8 showed agonistic activity in vitro (EC₅₀ = 1.4 × 10³ ± 4.4 10² nM) as compared to 1 (EC₅₀ = 7.0 ± 2.1 nM). Atosiban analogues 17 (pA₂ = 7.8 ± 0.1) and 18 (pA₂ = 8.0 ± 0.1) showed substantial activity compared to the parent oxytocin antagonist 2 (pA₂ = 9.9 ± 0.3). Carba analogue 35 (pA₂ = 6.1 ± 0.1) had an agonistic activity over 2 orders of magnitude less than its parent 3 (8.8 ± 0.5). A comparison of biological stabilities of 1,6-carba analogues of both an agonist 8 and antagonist 18 versus parent peptides 1 and 2 was conducted. The half-lives of peptides 8 and 18 in rat placental tissue were shown (Table 2) to be greatly improved versus their parents oxytocin 1 and atosiban 2, respectively. These results suggest that peptides 8 and 18 and analogues thereof may be important leads into the development of a long-lasting, commercially available therapeutic for initiation of parturition and treatment of preterm labor.

Full text of DOI:10.1021/jo050539l

VOLUME 70, NUMBER 20
SEPTEMBER 30, 2005
© Copyright 2005 by the American Chemical Society
Synthesis of Oxytocin Analogues with Replacement of Sulfur by
Carbon Gives Potent Antagonists with Increased Stability
Jake L. Stymiest,† Bryan F. Mitchell,‡ Susan Wong,‡ and John C. Vederas*,†
Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2, and Perinatal
Research Centre, Department of Obstetrics and Gynecology, University of Alberta, Canada TH5 3V9
john.vederas@ualberta.ca
Received March 16, 2005
The neuropeptide oxytocin 1 controls mammary and uterine smooth muscle contraction. Atosiban
2, an oxytocin antagonist, is used for prevention of preterm labor and premature birth. However,
the metabolic lifetimes of such peptide drugs are short because of in vivo degradation. Facile
production of oxytocin analogues with varying ring sizes wherein sulfur is replaced by carbon
(methylene or methine) could be achieved by standard solid-phase peptide synthesis using olefin-
bearing amino acids followed by on-resin ring-closing metathesis (RCM). These were tested for
agonistic and antagonistic uteronic activity using myometrial strips taken from nonpregnant female
rats. Peptide 8 showed agonistic activity in vitro (EC50 ) 1.4 × 103 ( 4.4 × 102 nM) as compared
to 1 (EC50 ) 7.0 ( 2.1 nM). Atosiban analogues 17 (pA2 ) 7.8 ( 0.1) and 18 (pA2 ) 8.0 ( 0.1)
showed substantial activity compared to the parent oxytocin antagonist 2 (pA2 ) 9.9 ( 0.3). Carba
analogue 35 (pA2 ) 6.1 ( 0.1) had an agonistic activity over 2 orders of magnitude less than its
parent 3 (8.8 ( 0.5). A comparison of biological stabilities of 1,6-carba analogues of both an agonist
8 and antagonist 18 versus parent peptides 1 and 2 was conducted. The half-lives of peptides 8
and 18 in rat placental tissue were shown (Table 2) to be greatly improved versus their parents
oxytocin 1 and atosiban 2, respectively. These results suggest that peptides 8 and 18 and analogues
thereof may be important leads into the development of a long-lasting, commercially available
therapeutic for initiation of parturition and treatment of preterm labor.
Introduction
roles, including mammary and uterine smooth muscle
contraction,1 neurotransmission in the central nervous
system, and autocrine and/or paracrine functions in the
ovaries and testes.2,3 Oxytocin 1 is clinically used to
Oxytocin 1 (Figure 1, Table 1) is a mammalian nona-
peptide hormone synthesized by the magnocellular neu-
rons of the hypothalamus.1 It has many physiological
(1) Andersson, K. E.; Forman, A.; Ulmsten, U. Clin. Obstet. Gynecol.
1983, 26, 56-77.
(2) Nicholson, H. D. Rev. Reprod. 1996, 1, 69-72.
(3) Harris, J. C.; Nicholson, H. D. J. Endocrinol. 1998, 156, 35-42.
* To whom correspondence should be addressed.
† Department of Chemistry.
‡ Department of Obstetrics and Gynecology.
10.1021/jo050539l CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/30/2005
J. Org. Chem. 2005, 70, 7799-7809
7799

Stymiest et al.
FIGURE 1. General structures of oxytocin (A), atosiban (B), and antagonist 3 (C). See Table 1 below.
1, Table 1) 211-13 is a competitive antagonist of oxytocin
receptors (OTR) that has been approved in Europe for
the short-term treatment of preterm labor.14-18
TABLE 1. Structures of Carbocyclic Peptides 1-12,
15-18 and 35
structure compd
m
n
X-X
isomer
ring size
However, 2 has a short metabolic half-life (ca. 16-18
min), is usually adminstered intravenously, and can have
unwanted side effects due to lack of selectivity for the
oxytocin receptor over the vasopressin receptors (VPR).13,19
Arginine[8]vasopressin (AVP), a hormone synthesized in
the posterior pituitary that plays a key role in osmotic
regulation and vasoconstriction, is structurally very
similar to atosiban 2 and oxytocin 1.19 Considerable effort
over the past decade has been devoted to the synthesis
of nonpeptidic OTR antagonists in an effort to increase
oral availability and longevity. However, drug approval
and liability issues20 as well as side effects, lack of
potency, poor bioavailability, or insufficient OTR specific-
ity have kept such compounds21-27 from clinical use.9,28
To address the selectivity problem with peptide ana-
logues, Manning and co-workers synthesized a series of
compounds including 3 (Figure 1, Table 1), which differs
A
1
1
1
1
2
2
1
2
2
2
2
2
2
-
1
1
2
1
1
2
2
2
1
1
1
2
-
S-S
20
20
21
21
21
21
22
22
22
22
21
22
20
20
20
20
20
20
4
CHdCH
S-SCH2
CHdCH
CHdCH
CHdCH
CHdCH
CHdCH
cis
cis
5
6
7
trans
8
2:1 cis/trans
cis
trans
9
10
11
12
15
16
2
CH2dCHCH2O cis
CH2dCHCH2O trans
CH2-CH2
CH2-CH2
S-S
CHdCH
CH2-CH2
S-S
CHdCH
CH2-CH2
B
C
17
18
3
20
35
-
-
1:4 cis/trans
-
-
-
-
-
-
cis/trans
-
-
induce labor and control postpartum hemorrhage,1,4 but
it has a short half-life (2-5 min) in vivo and is admin-
istered to patients intravenously.5,6 Antagonists of oxy-
tocin are of interest as tocolytic agents that inhibit
preterm labor and delay premature birth. There are an
estimated 13 million premature births worldwide per
annum7 that account for 66% of all neonatal mortality
and contribute to serious complications and infant mor-
bidity.8,9 In the US, about 470 000 infants (ca. 12%) were
born prematurely in 2002, and their hospital stays cost
in excess of $13.6 billion.10 Atosiban (Tractocile) (Figure
(10) Nelson, R. Am. J. Nursing 2004, 104, 23-25.
(11) Melin, P.; Trojnar, J.; Johansson, B.; Vilhardt, H.; Akerlund,
M. J. Endocrinol. 1986, 111, 125-131.
(12) Wilson, L. J.; Parsons, M.; Ouano, L.; Flouret, G. Am. J. Obstet.
Gynecol. 1990, 163, 195-202.
(13) Tsatsaris, V.; Carbonne, B.; Cabrol, D. Drugs 2004, 64, 375-
382.
(14) Gyetvai, K.; Hannah, M. E.; Hodnett, E. D.; Ohlsoon, A. Obstet.
Gynecol. 1999, 94, 869-877.
(15) Moutquin, J. M.; Sherman, D.; Cohen, H.; Mohide, P. T.;
Hochner-Celnikier, D.; Fejgin, M.; Liston, R. M.; Dansereau, J.; Mazor,
M.; Shalev, E.; Boucher, M.; Glezerman, M.; Zimmer, E. Z.; Rabinovici,
J. Am. J. Obstet. Gynecol. 2000, 182, 1191-1199.
(16) Worldwide Atosiban versus Beta-agonists Study Group. Br. J.
Obstet. Gynaecol. 2001, 108, 133-142.
(4) Sweetman, S. C. Martindale: The Complete Drug Reference, 34th
ed.; Pharmaceutical Press: London, 2004.
(5) Engstrom, T.; Barth, T.; Melin, P.; Vilhardt, H. Eur. J. Phar-
macol. 1998, 355, 203-210.
(6) Gazis, D. Proc. Soc. Exp. Biol. Med. 1978, 158, 663-665.
(7) Havass, J.; Bakos, K.; Marki, A.; Gaspar, R.; Gera, L.; Stewart,
J. M.; Fulop, F.; Toth, G. K.; Zupko, I.; Falkay, G. Peptides 2002, 23,
1419-1425.
(8) Villar, J.; Ezcurra, E. J.; Gurtner de la Fuente, V.; Campodonico,
L. Res. Clin. Forums 1994, 16, 9-38.
(9) Goldenberg, R. L.; Rouse, D. J. New Engl. J. Med. 1998, 339,
313-320.
(17) Goodwin, T. M.; Paul, R.; Silver, H.; Spellacy, W.; Parsons, M.;
Chez, R.; Hayashi, R.; Valenzuela, G.; Creasy, G. W.; Merriman, R.
Am. J. Obstet. Gynecol. 1994, 170, 474-478.
(18) French/Australian Atosiban Investigators Group. Eur. J. Obstet.
Gynecol. Reprod. Biol. 2001, 98, 177-185.
(19) Manning, M.; Stoev, S.; Cheng, L. L.; Wo, N. C.; Chan, W. Y.
J. Pept. Sci. 2001, 7, 449-465.
(20) Goodwin, T. M. J. Soc. Gynecol. Investig. 2001, 11, 339-341.
(21) Evans, B. E.; Leighton, J. L.; Rittle, K. E.; Gilbert, K. F.;
Lundell, G. F.; Gould, N. P.; Hobbs, D. W.; DiPardo, R. M.; Veber, D.
F.; Pettibone, D. J.; Clineschmidt, B. V.; Anderson, P. S.; Freidinger,
R. M. J. Med. Chem. 1992, 35, 3919-3927.
7800 J. Org. Chem., Vol. 70, No. 20, 2005

Dicarba Antagonists of Oxytocin with Increased Stability
from atosiban 2 by the presence of a bulky cyclohexyl
moiety at position 1 and introduction of a tyrosine residue
at position 9 as well as a D-thienylalanine at position 2.19
Nonapeptide 3 has comparable potency to 2, but the
selectivity for the OTR over the VP receptor set is greatly
enhanced,19 probably due primarily to the increased
conformational rigidity as a result of the cyclohexyl
group.
The discovery of antagonistic analogues of 1 that have
a longer duration of in vivo activity and greater bioavail-
ability than 2 is highly desirable for development of
tocolytic drugs. As the essential disulfide moiety that
restricts the conformational mobility of oxytocin and its
analogues is inherently sensitive to biochemical reduc-
tion, its substitution by a more stable linkage could
generate more metabolically robust derivatives. Replace-
ment of disulfide functionality in a variety of biologically
active peptides with methylene (CH2) or amide linkages
has afforded analogues that in certain cases retain
activity at reduced levels or are antagonists of the natural
compound.29-32 It has been demonstrated that substitu-
tion of the disulfide in oxytocin with methylene,33
methine,33d or amide functionalities33a yields analogues
with potent bioactivity and in some cases increased
metabolic stability.33c We recently reported that replace-
ment of the disulfide in 1 with a cis-alkene to give 4
FIGURE 2. Overlaid energy minimized structures of oxytocin
1 (A) and analogue 4 (B). Side chains on Y, I, Q, N, L are
omitted for clarity.
for a small kink imposed by the preference of the sulfurs
for a close to 90° dihedral angle (Figure 2).
In 1976, Smith and Ferger reported that increasing the
ring size of 1 by one methylene unit (substitution of a
homocysteine for cysteine residue) resulted in an ana-
logue 5 (Figure 1, Table 1) with antagonistic activity
(pA2 ) 6.0) and greatly reduced agonistic activity.34 This
was attributed to an increase in steric bulk within the
cyclic portion of the peptide.
(Figure 1, Table 1) generates a potent agonist (EC50
)
38 nM) that is only 10-fold less active than oxytocin,
whereas the corresponding trans-alkene and hydroge-
nated (methylene) analogues are 100-fold less active.33d
Overlaying energy-minimized structures of 1 and 4 shows
that the two peptides are very similar in structure except
Realizing that a variety of carbocyclic rings are poten-
tially available by ring-closing metathesis (RCM) reac-
tions of resin-bound linear peptides having two alkenyl
side chains of varying length,33d,35-44 we investigated the
synthesis, activity, and stability of a series of analogues
of oxytocin 1, atosiban 2, and the OTR selective antago-
nist 3. In each case, the disulfide bridge was replaced
with methylene or methine (olefinic) units in an effort
to enhance the biological half-life of the analogues
without compromising inherent activity.
(22) Williams, P. D.; Bock, M. G.; Evans, B. E.; Freidinger, R. M.;
Gallicchio, S. N.; Guidotti, M. T.; Jacobson, M. A.; Kuo, M. S.; Levy,
M. R.; Lis, E. V.; Michelson, S. R.; Pawluczyk, J. M.; Perlow, D. S.;
Pettibone, D. J.; Quigley, A. G.; Reiss, D. R.; Salvatore, C.; Stauffer,
K. J.; Woyden, C. J. Bioorg. Med. Chem. Lett. 1999, 9, 1311-1316.
(23) Kuo, M. S.; Bock, M. G.; Freidinger, R. M.; Guidotti, M. T.; Lis,
E. V.; Pawluczyk, J. M.; Perlow, D. S.; Pettibone, D. J.; Quigley, A. G.;
Reiss, D. R.; Williams, P. D.; Woyden, C. J. Bioorg. Med. Chem. Lett.
1998, 8, 3081-3086.
(24) Wyatt, P. G.; Allen, M. J.; Chilcott, J.; Gardner, C. J.; Livermore,
D. G.; Mordaunt, J. E.; Nerozzi, F.; Patel, M.; Perren, M. J.; Weingar-
ten, G. G.; Shabbir, S.; Woollard, P. M.; Zhou, P. Bioorg. Med. Chem.
Lett. 2002, 12, 1405-1411.
(25) Wyatt, P. G.; Allen, M. J.; Chilcott, J.; Foster, A.; Livermore,
D. G.; Mordaunt, J. E.; Scicinski, J.; Woollard, P. M. Bioorg. Med.
Chem. Lett. 2002, 12, 1399-1404.
Results and Discussion
Synthesis of 1,6-Carba Analogues of Oxytocin
Antagonists. Ring-closing olefin metathesis (RCM) reac-
tions have been previously done on peptide substrates
linked to solid support.33d,37,42,43 However, the tendency
of the organometallic catalysts to bind to resin-held
(26) Cirillo, R.; Tos, E. G.; Schwarz, M. K.; Quattropani, A.; Scheer,
A.; Missotten, M.; Dorbais, J.; Nichols, A.; Borrelli, F.; Giachetti, C.;
Golzio, L.; Marinelli, P.; Thomas, R. J.; Chevillard, C.; Laurent, F.;
Portet, K.; Barberis, C.; Chollet, A. J. Pharmacol. Exp. Ther. 2003,
306, 253-261.
(27) Gal, C. S.-L.; Valette, G.; Foulon, L.; Germain, G.; Advenier,
C.; Naline, E.; Bardou, M.; Martinolle, J.-P.; Pouzet, B.; Raufaste, D.;
Garcia, C.; Double-Cazanave, E.; Pauly, M.; Pascal, M.; Barbier, A.;
Scatton, B.; Maffrand, J.-P.; Le Fur, G. J. Pharmacol. Exp. Ther. 2004,
309, 414-424.
(28) Higby, K.; Suiter, C. R. Drug Safety 1999, 21, 35-56.
(29) Spinella, M. J.; Malik, A. B.; Everitt, J.; Anderson, T. T. Proc.
Natl. Acad. Sci. U.S.A. 1991, 88, 7443-7446.
(30) Grieco, P.; Carotenuto, A.; Patacchini, R.; Maggie, C. A.;
Novellino, E.; Rovero, P. Bioorg. Med. Chem. 2002, 10, 3731-3739.
(31) Hargittai, B.; Sole, N. A.; Groebe, D. R.; Abramson, S. N.;
Barany, G. J. J. Med. Chem. 2000, 43, 4787-4792.
(32) Thruieau, C.; Janiak, P.; Krantic, S.; Guyard, C.; Pillon, A.;
Kucharczyk, N.; Vilaine, J. P.; Fauchere, J. L. Eur. J. Med. Chem. 1995,
30, 115-122.
(33) (a) Smith, C. W.; Walter, R. J. Med. Chem. 1978, 21, 117-120,
and references therein. (b) Keller, O.; Rudinger, J. Helv. Chim. Acta
1974, 57, 1253-1259. (c) Servitova, L.; Barth, T. Collect. Czech. Chem.
Commun. 1975, 40, 215-217. (d) Stymiest, J. L.; Mitchell, B. F.; Wong,
S.; Vederas, J. C. Org. Lett. 2003, 5, 47-49.
(34) Smith, C. W.; Ferger, M. F. J. Med. Chem. 1976, 19, 250-254.
(35) Brouwer, A. J.; Liskamp, R. M. J. J. Org. Chem. 2004, 69,
3662-3668.
(36) Gao, Y.; Wei, C.; Burke, T. R. Org. Lett. 2001, 3, 1617-1620.
(37) Jarvo, E. R.; Copeland, G. T.; Papaioannou, N.; Bonitatebus,
P. J. J.; Miller, S. J. J. Am. Chem. Soc. 1999, 121, 11638-11643.
(38) Kasmaier, U.; Maier, S. Org. Lett. 1999, 1, 1763-1766.
(39) Miller, S. J.; Blackwell, H. E.; Grubbs, R. H. J. Am. Chem. Soc.
1996, 118, 9606-8614.
(40) Reichwein, J. F.; Liskamp, R. M. J. Eur. J. Org. Chem 2000,
2335-2344.
(41) Reichwein, J. F.; Versluis, C.; Liskamp, R. M. J. J. Org. Chem.
2000, 65, 6187-6195.
(42) Schafmeister, C. E.; Po, J.; Verdine, G. L. J. Am. Chem. Soc.
2000, 122, 5891-5892.
(43) Schmiedeberg, N.; Kessler, H. Org. Lett. 2002, 4, 59-62.
(44) Williams, R. M.; Liu, J. J. Org. Chem. 1998, 63, 2130-2132.
J. Org. Chem, Vol. 70, No. 20, 2005 7801

Stymiest et al.
SCHEME 1. Structures of 13 and 14 and General Synthesis of 1,6-Dicarba Analogues Using RCM Reaction
products complicates purification and drastically lowers
or abolishes yields. Fortunately, we recently found that
treatment of the resin after RCM reaction with dimethyl
sulfoxide (DMSO) removes organometallic byproducts
and allows facile recovery of the desired cyclized com-
pounds.33d It has also been uncertain whether a variety
of larger rings would be available via the RCM process
because conformational preferences of the precursors
could hinder the cyclization. However, the syntheses of
analogues 6-12 (Figure 1, Table 1), containing rings
larger than oxytocin 1 (i.e., >20 members) proceed readily
as outlined in Scheme 1. For 21-membered rings 6-8,
the syntheses employed SPPS of linear precursors using
Fmoc methodology with incorporation of L-homoallyl-
glycine48,49 at either positions 1 or 6, with L-allylglycine
added to the opposing position. Use of L-homoallylglycine
at both residues 1 and 6 generates the linear precursor
to analogues 9 and 10 having 22-membered rings. To
examine the influence of a ring oxygen on biological
activity, L-O-allylserine50 was placed at position 1 with
L-allylglycine at position 6 (compounds 11 and 12). RCM
reactions on suspensions of resin-bound peptides using
10 mol % solutions of either Grubbs first-generation
(45) Silverstein, R. M.; Webster, F. X. Spectrometric Identification
of Organic Compounds, 6th ed.; John Wiley and Sons: New York, 1998.
(46) Mitchell, B. F.; Fang, X.; Wong, S. Biol. Reprod. 1997, 57, 807-
812.
(47) Schild, H. O. Br. J. Pharmacol. 1947, 2, 189-206.
(48) Baldwin, J. E.; Hulme, C.; Schofield, C. J. J. Chem. Res.,
Miniprint 1992, 6, 1517-1526.
(49) Biagini, S. C. G.; Gibson, S. E.; Keen, S. P. J. Chem. Soc., Perkin
Trans. 1 1998, 16, 2485-2500.
(50) Ramtohul, Y. K. Ph.D. Thesis, University of Alberta, 2002, pp
192-193.
7802 J. Org. Chem., Vol. 70, No. 20, 2005

Dicarba Antagonists of Oxytocin with Increased Stability
catalyst 1351,52 (Scheme 1) or second-generation catalyst
1453 (Scheme 1) in dry degassed CH2Cl2 required reflux
at 40 °C for 18 h. Treatment with DMSO (50 equiv
relative to catalyst) for another 12 h at 20 °C assists
removal of byproducts and facilitates purification.33d
Following removal of the N-terminal Fmoc with piperi-
dine, acidic cleavage from the resin with concomitant
removal of acid-labile side-chain protecting groups gives
cyclic peptides in overall yields of 8-28% (Scheme 1) after
purification by RP-HPLC. The lower yields can be at-
tributed to cyclizations with the less active Grubbs first-
generation catalyst 13. The cis/trans isomers of all but
one set of olefin-containing peptides, compound 8, are
readily separated by HPLC. In each case, the cyclic
olefinic peptides elute prior to their linear precursors
during RP-HPLC analysis.
The assignment of cis and trans configuration could
be accomplished by a novel simultaneous 1H NMR
double-decoupling experiment at 500 MHz. In these
experiments, two sets of allylic protons at different
chemical shifts are simultaneously irradiated, thereby
allowing for the observation of the decoupled AB quartet
due to the adjacent olefinic protons. The coupling con-
stants of the two olefinic protons are then readily
measurable, and are 8.5-10.7 Hz for the cis isomers and
14.9-15.9 Hz for the trans isomers. Saturated derivatives
of both 21- and 22-membered ring analogues are avail-
able by quantitative reduction of olefinic peptides 8 and
9 using standard hydrogenation to yield peptides 15 and
16 (Figure 1, Table 1), respectively. The peptides were
then examined for both agonistic and antagonistic activ-
ity in comparison to peptides 1 and 2 (see below).
Synthesis of 1,6-dicarba analogues of atosiban 2 relies
on analogous methodologies. The salient features are that
position 1 of 2 contains 3-mercaptopropionic acid, which
was replaced with commercially available 4-pentenoic
acid. L-Allylglycine was placed at position 6 for the
purposes of RCM reaction. Fmoc SPPS yields a linear
resin-bound precursor that could be cyclized and subse-
quently cleaved from the resin to give analogue 17
(Figure 1, Table 1) as a 1:4 mixture of cis/trans isomers
in 21% overall yield. Various attempts to separate these
isomers using RP-HPLC failed, and the two component
mixture 17 was tested directly. Reduction of 17 quanti-
tatively gives a fully saturated pure dicarba analogue 18
(Figure 1, Table 1), which was also tested for antagonistic
activity and compared to atosiban 2.
1), but instead gave a dimerized product. It appears that
this dimerization occurs beween allylglycine residues
between neighboring peptides due to the steric bulk of
the 1-(vinylcyclohexyl)-1-acetic acid residue. To counter-
act this, a linear precursor 21 having L-crotylglycine at
position 6 was made so as to provide additional steric
bulk to the olefin terminus and thereby prevent dimer-
ization. However, attempted RCM on 21 yielded no
metathesis products. Linear precursor 19 was also made
on Rink amide resin with a much lower loading (0.12
mmol/g) to achieve greater dilution and thereby retard
dimerization. Unfortunately, this also yielded no metath-
esis products under RCM conditions. Finally, linear
precursor was made on Sieber amide resin which permit-
ted cleavage of the peptide from the resin under mild
conditions (1% TFA) without side-chain deprotection to
give 22. Numerous RCM reaction attempts on dilute
solutions of this free peptide did not generate any
monomeric cyclized product. Clearly, the steric crowding
caused by the cyclohexyl group completely inhibits in-
tramolecular cyclization by the bulky organometallic
reagent.
To overcome the difficulty of making a carbocyclic
analogue of 3 via RCM reaction, the strategy was
changed to a more classical approach. This involved
incorporation an orthogonally protected cyclohexyl de-
rivative of R-aminosuberic acid 23 with intramolecular
amide bond formation as the ring-closing step.54 Com-
pound 23 could be readily made as shown in Scheme 3.
The key step, photolytic decomposition of a diacyl per-
oxide at low temperature in the absence of solvent, is
based on a novel methodology recently reported by us
that allows facile synthesis of amino acid derivatives from
two carboxylic acids without crossover products.55 The
known56 anhydride 24 is readily opened to acid 25 using
sodium methoxide in refluxing methanol. Acid 25 reacts
with 50% hydrogen peroxide in the presence of sulfuric
acid to give the peracid 26 in 78% yield. Coupling of
peracid 26 with N-carboxybenzylglutamic acid 1-benzyl
ester (N-Cbz-Glu-OBn) 27 using dicyclohexyldicarbox-
imide (DCC) produces the corresponding diacyl peroxide
28 in 80% yield. Photolysis of neat 28 at -78 °C for 48 h
affords 29 in 34% yield with recovery of 36% of 28, which
could be recycled. Quantitative deprotection of 29 by
hydrogenolysis gives amine 30, and subsequent Fmoc
protection generates 23.
Unfortunately, the yield of the last step is low, possibly
because of solubility problems, but recovered starting
material could be reacted again to afford additional
product. This could then be incorporated in the synthesis
of the linear precursor 31 on Sieber amide resin using
standard SPPS (Scheme 4). Peptide 31 is cleaved from
the resin using 1% TFA to yield the fully protected
peptide methyl ester 32, which could be hydrolyzed using
2 N lithium hydroxide to yield peptide acid 33. This
reaction proceeds to only 59% completion after 3 days at
room temperature, but attempts to increase the yield by
heating at 50 °C for 3 h lead to decomposition of the
Based on the successful syntheses of these carbocyclic
peptide analogues by RCM reactions, it appeared that a
similar approach should provide access to an analogue
of 3, which has superior selectivity for OT vs VP recep-
tors, with methine or methylene groups in place of sulfur.
Synthesis of a linear precursor with 1-(vinylcyclohexyl)-
1-acetic acid at position 1 and L-allylglycine at postion 6
using standard SPPS on Rink amide resin (0.6 mmol/g)
proceeds readily (Scheme 2). However, RCM reaction of
this resin-bound precursor 19 under a variety of condi-
tions generated no cyclized derivative 20 (Figure 1, Table
(51) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996,
118, 100-110.
(52) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2039-2041.
(53) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953-956.
(54) Taunton, J.; Collins, J. L.; Schreiber, S. L. J. Am. Chem. Soc.
1996, 118, 10412-10422.
(55) Spantulescu, M. D.; Jain, R. P.; Derksen, D. J.; Vederas, J. C.
Org. Lett. 2003, 5, 2963-2965.
(56) Callahan, J. F.; Newlander, K. A.; Bryan, H. G.; Huffman, W.
F.; Moore, M. L.; Yim, N. C. F. J. Org. Chem. 1988, 53, 1527-1530.
J. Org. Chem, Vol. 70, No. 20, 2005 7803

Stymiest et al.
SCHEME 2. Attempted Syntheses of 1,6-Dicarba Analogue 20 via RCM
TABLE 2. Oxytocin Agonism and Antagonisma of
SCHEME 3. Synthesis of r-Aminosuberic Acid
Derivative 23
Compounds 1, 2, 3, 8, 17, 18, and 35 and Placental Tissue
Half-Livesb for 1, 2, 8, and 18
compd
EC50 (nM)
[pA2] (nM)
pA2
half-life (min)
1
7 ( 2
4,a 16b
8
(1.4 ( 0.4) × 103
24,a 15b
2
3
17
18
35
0.2 ( 0.1 9.9 ( 0.3
156,c 108d
3 ( 2
17 ( 6
10 ( 1
8.8 ( 0.5
7.8 ( 0.1
8.0 ( 0.1
318,c 306d
750 ( 20 6.1 ( 0.1
a Experiments done in triplicate. b Four different animals used,
two each for agonists 1 and 8 (a and b) and atagonists 2 and 18
(c and d).
35 (Figure 1, Table 1) could be isolated in 76% yield after
HPLC purification and was tested for biological activity
and stability.
Testing of Analogues as Agonists and Antago-
nists of Oxytocin.33d,46 Both atosiban 2 and the selective
antagonist 3 were tested to compare their activities to
the 1,6-carba analogues of oxytocin (Table 2). As ex-
pected, none of the carba analogues of 2 and 3 show any
significant agonistic activity in vitro.12,19 During biological
testing, with the exception of 8, all of the larger ring
carbocyclic compounds also display no agonistic behavior.
peptide. The acid 33 undergoes intramolecular cyclization
upon reaction with 1:1 PyBOP/HOBt in DMF at room
temperature for 2.5 days.54 Treatment of the resulting
cyclic analogue 34 (29% yield) with 95% TFA removes
all side chain protecting groups. The dicarba analogue
7804 J. Org. Chem., Vol. 70, No. 20, 2005

Dicarba Antagonists of Oxytocin with Increased Stability
SCHEME 4. Incorporation of 23 into Peptide Chain via SPPS and Cyclization to 35
Surprisingly, 8 exhibits agonistic activity at higher doses
(EC50 ) (1.4 ( 0.4) × 103 nM) and shows no antagonistic
behavior when tested. This contrasts with the literature
report34 on the corresponding disulfide analogue 5, which
indicates that it has no oxytocin agonist properties but
is a potent antagonist (pA2 ) 6.0). The pA2 values are
defined as the negative logarithm of the concentration
of antagonist that diminishes the OT activity of a double
dose to that of a single dose.12,47 The exact interpretation
of this result is uncertain as the 2:1 mixture of cis/trans
isomers of 8 could not be easily separated, and it is
unclear which isomer is exhibiting the agonistic activity.
All carbocyclic analogues were screened for antagonis-
tic activity. Only the 1,6-dicarba analogues 17, 18, and
35 possess significant antagonistic activity, and these
values were compared to atosiban 2 and the selective
antagonist 3, respectively (Table 2). The pA2 values were
calculated for active analogues 17, 18, and 35 using the
methodology developed by Schild47 (see the Supporting
Information). Peptides 17 and 18 have potent OT an-
tagonistic activity (pA2 ) 7.8 ( 0.1 and 8.0 ( 0.1,
respectively) approaching that of 2 (pA2 ) 9.9 ( 0.3)
(Table 2). Interestingly, the presence of the olefin func-
tionality in 17, which imposes considerable conforma-
tional restraint on the ring system, appears to have only
a slight influence on the activity as the more flexible
saturated bis-methylene derivative 18 has nearly the
same antagonistic effect on the OT receptors. This
contrasts the agonistic effect of oxytocin analogue 4 and
its congeners, where activity is reduced by an order of
magnitude for the trans isomer and the saturated
counterpart. The dicarba analogue 35 (pA2 ) 6.1 ( 0.1)
is considerably less active than its disulfide parent 3
(pA2 ) 8.8 ( 0.5). A possible explanation for the nearly
3 orders of magnitude decrease in activity of 35 may be
deviation in structure from the preferred 90° dihedral
J. Org. Chem, Vol. 70, No. 20, 2005 7805

Stymiest et al.
angle between the sulfurs of the disulfide bridge in 3.
This may be promoted by the steric crowding of the
cyclohexyl group. Introduction of the conformationally
less constrained cyclohexyl-bearing methylene linker that
can readily rotate to alleviate nonbonded interactions
may alter the rest of the peptide geometry and thereby
hinder the ability of 35 to readily dock with the OTR.
Potentially the optimal 90° dihedral angle of the disulfide
in 3 could be restored through incorporation of a confor-
mational lock, for example, an additional ring structure
on the carbon bridge.
Stability of Analogues in Rat Placental Tissue.
A new methodology was developed for testing the stabil-
ity of 1,6-dicarba analogues of oxytocin using fresh rat
placental tissue. To determine if this methodology is
useful as a general method to examine stablility, we
tested both an agonist 8 and antagonist 18. Four separate
animals were used for the testing, two each for the
agonist and antagonist. The results for analogues 8 and
18 were then compared to their parent disulfide contain-
ing peptides, oxytocin 1 and atosiban 2, respectively.
Half-lives were calculated for both the agonists 1 and 8
and the antagonists 2 and 18 (Table 2). Values are
reported for each experiment as there is variation in the
levels of enzyme(s) responsible for breaking down oxy-
tocin in placental tissue in different animals.46 The
results show that the 1,6-carba analogue 8 has a consid-
erably greater half-life (8-11 min longer) than 1 when
incubated in placental tissue from two different rats.
Similar results are observed with the 1,6-dicarba ana-
logue 18 of atosiban. Analogue 18 has a half-life in
placental tissue more than twice that of atosiban 2. These
results demonstrate that replacement of disulfide bridges
in peptidic oxytocin agonists or antagonists with carbon
linkers (saturated or unsaturated) can significantly
stabilize the compounds and hinder degradation in
placental tissue while still retaining high levels of inher-
ent activity.
nists of the oxytocin receptor (OTR). Interestingly, such
replacement of sulfur by carbon markedly increases the
half-life of the peptide analogues in placental tissue. This
observation provides a basis for design of new oxytocin
antagonists with increased potency and metabolic stabil-
ity that may prove valuable for the development of an
improved therapeutic for treatment of pre-term labor and
premature birth. It will also be interesting to determine
whether similar carbon for sulfur substitution in other
peptides will give analogues with improved properties.
Finally, the methodology would also be applicable for
generation of derivatives that are functionalized on the
bridge carbons (e.g., via bis-hydroxylation of the olefin),
thereby offering additonal avenues for improvement of
drug properties by medicinal chemistry.
Experimental Section
Simultaneous Double-Decoupling for NMR Assign-
ment of Cis/Trans Stereochemistry. Selective multiple site
homonuclear decoupling experiments on olefinic peptides (in
D2O) were accomplished using shifted laminar pulses on a
Varian Inova 600 MHz spectrometer running at 599.933 MHz.
The instrument was equipped with a Sun Microsystems Ultra
5, running VNMR 6.1C software, and a waveform generator
to produce shaped pulses. The program PBOX, part of the
spectrometer software package VNMR 6.1C, was used to
calculate the WURST-2 pulse shape required for the decou-
pling experiments. To calculate the WURST-2 shape, peaks
belonging to protons that were to be decoupled from the olefin
protons were interactively input into the program PBOX, along
with a reference coupling constant between these protons.
Additionally, a transmitter power of 53 and a pulse width of
1
5.5 for a 90° rectangular H pulse, was required as input for
PBOX. The parameters calculated by PBOX were then input
as standard decoupling parameters, using a decoupling power
of 21, in a normal 1H pulse sequence. Finally, the transmitter
and decoupler offsets were set to be equal (tof ) dof). The
spectrum was acquired with a total of 16 scans, a spectral
width of 8000 Hz, and a total of 48 000 points. Coupling
constants between the adjacent olefin protons were measured
directly and accurately from the aquired spectrum and used
to assign cis/trans geometry about the double bond.
Conclusion
Assays for OTR Agonistic Activity. Testing of peptides
employed freshly excised uteri from mature nonpregnant,
female Sprague-Dawley rats (∼250 g), and the experiments
were done in triplicate using tissue samples from three
separate animals. Muscle bath preparations were done on the
basis of published methodology.33d,46 Rat uteri were cut into
strips (1 cm × 0.3 cm) and mounted vertically on a Biopac
Systems, Inc., myobath apparatus, using wire hooks in sepa-
rately jacketed organ baths at 30 °C containing 10 mL of Krebs
buffer with the following composition: 118 mM NaCl; 4.7 mM
KCl; 2.5 mM CaCl2; 1.2 mM KH2PO4; 0.59 mM MgSO4; 25 mM
NaHCO3; 11.7 mM D-glucose at pH 7.4 with constant CO2
aeration. One end of each strip was anchored in the bath, and
the other end was attached to a FT-03C force-displacement
transducer connected to a World Precision Instruments, Inc.,
detection system. Resting tension was set to 1 g to provide
maximum active tension. Oxytocin 1 and analogue injections
were made directly into the baths. Measurements were
recorded in 5 min blocks with injections of increasing concen-
tration being added at the end of each 5 min interval. Dose-
response curves for 1 and analogues were constructed at the
conclusion of each 60 min experiment. Microsoft Excel was
used to calculate regression outputs for these curves in order
to determine analogue EC50 values.
The ring-closing metathesis (RCM) reaction is cur-
rently one of the most popular methods for carbon ring
formation. Its utility for cyclization of protected peptides
bound to solid phase to produce large (20-22 members)
carbocyclic olefins is demonstrated by synthesis of a
series of oxytocin analogues. The resulting compounds
have an olefin, or after reduction, methylene groups, in
place of the essential disulfide moiety. A key feature is
the use of DMSO after RCM cyclization to remove
organometallic impurities that otherwise lead to product
degradation during cleavage from the resin. A limitation
of the RCM reaction is that the steric bulk of the catalysts
can prevent cyclization in cases where the substrate is
already sterically hindered (e.g., 22 f 20). Fortunately,
the use of solvent-free photolysis of diacyl peroxides
permits linking of sterically crowded moieties and affords
rapid access to functionalized and selectively protected
amino acids (e.g., 23). These can then be incorporated
via linear solid-phase synthesis to make carbocyclic
peptides (e.g., 35) via standard amide bond formation.
Analogues of atosiban 2 and antagonist 3, wherein sulfur
is replaced by carbon (e.g., 17, 18, and 35) display
reduced, but still very high, levels of activity as antago-
Assays for OTR Antagonistic Activity. The same basic
procedure as for agonist tests was used with the following
modifications. Muscle strips were incubated with antagonists
7806 J. Org. Chem., Vol. 70, No. 20, 2005

Dicarba Antagonists of Oxytocin with Increased Stability
at varying concentrations for 4 min. Oxytocin 1 was then added
directly into the baths containing antagonists every 3 min with
increasing concentration (0.78, 3.12, 12.50, 50.00, and 200 nM,
respectively). All antagonist results were compared to either
2 or 3, and oxytocin-inhibitory curves were constructed from
the data using Microsoft Excel. The pA2 values (defined as the
negative logarithm of the concentration of antagonist that
diminished the OT activity of a double dose to that of a single
dose) for each OT antagonist analogue were calculated using
the Schild plot analysis method.47
J ) 7.2 Hz), 0.53 (d, 3H, J ) 7.0 Hz); 13C NMR (D2O, 150 MHz)
δ 175.3, 175.1, 174.9, 174.8, 174.4, 173.4, 171.9, 171.2, 158.0,
131.4, 129.5, 116.0, 68.4, 65.2, 59.9, 59.5, 56.7, 54.3, 51.9, 48.9,
43.2, 39.6, 36.5, 31.0, 30.3, 28.5, 25.6, 25.0, 24.0, 19.5, 17.5,
15.1, 14.4, 11.9; MALDI-TOF (MS) calcd for C43H67N11O12S2
993.4, found 994.4 (M + H, 100), 1016.4 (M + Na, 73), 1032.3
(M + K, 48).
(2S)-N-(9H-Fluorenylmethoxycarbonyl)-2-amino-5-hex-
enoic Acid (Fmoc-L-homoallylglycine). (2S)-2-Amino-5-
hexenoic acid (L-homoallylglycine) was prepared according to
literature precident.48 To a solution of L-homoallylglycine (0.62
g, 4.8 mmol) in H2O at 0 °C was added NaHCO3 (1.61 g, 19.1
mmol). This reaction mixture was allowed to stir at 0 °C for
10 min. A solution of 9H-fluorenylmethyl chloroformate (Fmoc-
Cl) (1.41 g, 5.3 mmol) in acetone was then added dropwise over
15 min. The reaction mixture was stirred at 0 °C for 1 h and
then allowed to warm to room temperature where stirring
continued for another 4 h. The mixture was then concentrated
in vacuo, and the resulting residue was taken up in H2O (50
mL). The aqueous layer was extracted with EtOAc (2 × 75
mL) and acidified to pH 2 with 1 M HCl and extracted with
EtOAc (4 × 75 mL). The combined organic layers were dried
(Na2SO4) and concentrated in vacuo to yield a white solid,
which was recrystallized from CH2Cl2 and hexanes (1.21 g,
Biological Stability. A new method was developed to test
the duration of activity of oxytocin analogues in placental
tissue. Stability tests were conducted on compounds 1, 2, 8,
and 18. Homogenate was made from placental tissue taken
from pregnant female Sprague-Dawley rats at day 19 of
gestation to provide for maximum oxytocinase activity. Solu-
tions containing 1 µM of agonists 1 and 8 were incubated with
the homogenate for 0, 5, 10, 15, 20, 25, and 30 min at 37 °C.
Freshly excised uteri from mature, nonpregnant, female
Sprague-Dawley rats (∼250 g) were cut into strips as in the
agonistic assays. Muscle bath preparations and the general
procedure were as described for agonistic and antagonistic
assays. Samples of 1 (10 nM) and 8 (1 µM) were then taken
from placental homogenate solutions and added directly into
the muscle baths. Maximum activity was measured at time 0
min for (each compound), and the decrease in percent activity
of the analogues was measured at each 5 min incubation
interval. Percent activity versus incubation time curves were
constructed for compounds 1 and 8, and half-lives (in minutes)
were calculated for each trial and averaged. Antagonists 2 and
18 were tested in a manner similar to that of the above
agonists with both being incubated as 100 nM solutions in the
placental homogenate. Incubation times were set at 0, 1, 2, 3,
4, and 6 h, respectively. Muscle baths were incubated for 4
min with each antagonist-homogenate solution (2; 10 nM, 18;
100 nM) followed by addition of 1 every 3 min at increasing
concentrations of 0.98, 3.9, 15.6, 62.5, and 250 nM, respec-
tively. Percent activity of oxytocin response was measured at
4 nM of 1, and these trials were conducted on four separate
animals, two of each for the agonist and antagonist. The
maximum inhibitions for compounds 2 and 18 were measured
at time 0 h, and any decrease in percent inhibition was
measured at each of the incubation intervals indicated above.
Glycinamide, O-Ethyl-N-(3-mercapto-1-oxopropyl)-D-
tyrosyl-L-isoleucyl-L-threonyl-L-asparaginyl-L-cysteinyl-
L-propyl-L-ornithyl-, Cyclic (1f5)-Disulfide (Atosiban)
(2). This is the general procedure used for the synthesis of
disulfide peptides 2 and 3. The linear precursor to peptide 2
was synthesized in a manner analogous to that of the other
OT analogues using standard Fmoc SPPS. (S-Trt)-3-mercap-
topropionic acid was prepared as according to literature
precident.57 Disulfide formation of the linear precursor to 2
was done in 0.1 mM NH4HCO3 buffer at pH 8.0 with constant
O2 aeration and vigorous stirring for a period of 18 h. This
mixture was then concentrated to one-third of the original
volume, and the remainder of the buffer was removed by
lyopholization. Peptide 2 was isolated as a single peak using
C18, RP-HPLC, 10% MeCN, 90% H2O (0.1% TFA), 5 min, 10-
90% MeCN over 20 min, tR ) 14.94 min (11 mg, 7.3% overall
based on 0.10 mmol of resin cleaved): 1H NMR (D2O, 600 MHz)
δ 7.22 (apparent doublet, 2H, J ) 8.6 Hz), 6.96 (apparent
doublet, 2H, J ) 8.6 Hz), 4.85 (dd, 1H, J ) 3.2, 10.0 Hz), 4.65
(dd, 1H, J ) 6.3, 9.9 Hz), 4.61 (m, 1H), 4.44 (dd, 1H, J ) 6.0,
8.0 Hz), 4.35 (m, 2H), 4.17 (m, 1H), 4.14 (d, 1H, J ) 4.9 Hz),
4.10 (q, 2H, J ) 7.1 Hz), 3.92 (m, 2H), 3.84 (m, 1H), 3.73 (m,
1H), 3.13 (m, 1H), 3.20 (m, 3H), 2.96 (m, 2H), 2.92-2.82 (m,
3H), 2.81-2.68 (m, 3H), 2.58 (m, 1H), 2.32 (m, 1H), 2.03 (m,
2H), 1.92 (m, 2H), 1.80 (m, 4H), 1.37 (t, 3H, J ) 7.1 Hz), 1.20
(d, 3H, J ) 6.4 Hz), 1.05 (m, 1H), 0.83 (m, 1H), 0.74 (t, 3H,
72%): IR (CHCl3, cast) 3400-2400, 1717, 1521, 1417 cm-1
;
1H NMR (CDCl3, 500 MHz) δ 10.63 (br. s, 1H), 7.78 (apparent
doublet, 2H, J ) 7.5 Hz), 7.58 (m, 2H), 7.38 (apparent triplet,
2H, J ) 7.4 Hz), 7.28 (apparent triplet, 2H, J ) 7.4 Hz), 5.75
(m, 1H), 5.28 (d, 1H, J ) 8.3 Hz), 5.04 (m, 2H), 4.32 (m, 3H),
4.22 (t, 1H, J ) 6.9 Hz), 2.18 (m, 2H), 2.02 (m, 1H), 1.8 (m,
1H); 13C (CDCl3, 125 MHz) δ 176.9, 156.0, 143.6, 141.2, 136.5,
127.6, 127.0, 124.9, 119.9, 115.9, 67.1, 53.3, 47.2, 31.6, 29.4;
MS (ES) calcd for C21H21NO4 351.3957, found (M+); [R]D (c 1.0,
CHCl3) ) +13.32.
(1:4, cis/trans)-(2S)-N-(9H-Fluorenylmethoxycarbonyl)-
2-amino-4-hexenoic (Fmoc-L-crotylglycine). (2S)-2-Amino-
4-hexenoic acid (L-crotylglycine) was prepared according to
literature precident.58 A solution of L-crotylglycine (2.70 g, 20.9
mmol) in 10% aqueous NaHCO3 (60 mL) and dioxane (20 mL)
was cooled to 0 °C and stirred for 15 min. Fmoc-Cl (6.12 g,
23.0 mmol) in dioxane (30 mL) was then added dropwise over
15 min, and the mixture was stirred at 0 °C for 2.5 h. The
reaction was warmed to rt and stirred for a further 8 h. H2O
(600 mL) was added, and the aqueous layer was extracted with
EtOAc (2 × 150 mL). The aqueous layer was acidified to pH 1
with concd HCl and extracted with EtOAc (3 × 250 mL). The
combined organic layers were dried (Na2SO4) and concentrated
yielding a colored residue which was recrystallized from CH2-
Cl2 and hexanes to give Fmoc-L-crotylglycine as a white solid
in a 1:4 cis/trans mixture (3.00 g, 41%): IR (CHCl3, cast) 3500-
2650, 1719, 1518, 758 cm-1 1H NMR (CDCl3, 300 MHz) for
;
major isomer only δ 10.82 (br s, 1H, CO2H), 7.75 (ap d, 2H,
J ) 7.4 Hz, ArH), 7.57 (m, 2H, ArH), 7.38 (ap t, 2H, J ) 7.4
Hz, ArH), 7.30 (ap t, 2H, J ) 7.4 Hz, ArH), 5.70-5.50 (m, 1H,
CH3CHdCH), 5.34 (m, 2H, CONHCH, CH3CHdCH), 4.45 (m,
3H, NHCHCO2H, Ar2CHCH2), 4.22 (t, 1H, J ) 7.0 Hz,
ArCHAr), 2.78-2.32 (m, 2H, CHdCHCH2), 1.66 (d, 3H, J )
6.1 Hz, CH3CHdCH); 13C NMR (CDCl3, 125 MHz) δ 176.8,
156.0, 143.8, 143.7, 141.3, 130.6, 127.8, 127.1, 125.1, 124.1,
120.1, 67.2, 53.5, 47.2, 35.2, 18.0; HRMS (ES) calcd for C21H21-
NO4Na 374.1363, found 374.1367 (M + Na).
Synthesis of Peptides 6 and 7. This is a general method
used for cyclization of oxytocin analogues 6-12 and 17 using
ring-closing metathesis.33d All cyclizations involving resin-
bound oxytocin analogues were done on Rink amide NovaGel.
To a suspended solution of resin-bound peptide precursor to 6
and 7 (645 mg, 0.25 mmol) in degassed CH2Cl2 (5 mL) was
added 10 mol % of either bis(tricyclohexylphosphine)ben-
zylideneruthenium(IV) dichloride (Grubbs first-generation
(57) Gazal, S.; Gelerman, G.; Ziv, O.; Karpov, O.; Litman, P.; Bracha,
M.; Afargan, M.; Gilon, C. J. Med. Chem. 2002, 45, 1665-1671.
(58) Goering, H. L.; Cristol, S. J.; Dittmer, K. J. Am. Chem. Soc.
1948, 70, 3310-3313.
J. Org. Chem, Vol. 70, No. 20, 2005 7807

Stymiest et al.
catalyst) 1351,52 (21 mg, 0.03 mmol) or 1,3-bis(2,4,6-trimeth-
ylphenyl)-2-imidazolidinylidene)dichloro(phenylmethylene)-
(tricyclohexylphosphine)ruthenium (Grubbs second-generation
catalyst) 1453 (21 mg, 0.03 mmol) dissolved in degassed CH2-
Cl2 (2 mL). This reaction mixture was then allowed to gently
reflux for ∼18 h. The mixture was then cooled to room
temperature, DMSO (50 equiv relative to the catalyst) was
then injected into the sample, and the mixture was allowed
to stir at room temperature for an additional 12 h. The resin-
bound peptide was then filtered through a sintered glass frit
and washed successively with CH2Cl2 and MeOH. Fmoc
removal was carried out while the peptide was on resin using
20% (v/v) piperidine/DMF followed by filtration and washing
as mentioned above. Cleavage of the peptide from the resin
was done using a cocktail of 18:1:1 TFA/CH2Cl2/Et3SiH. Upon
purification, the cyclized olefinic peptides 6 and 7 elute from
the column just prior to their linear precursor. In all cases,
except for 8 and 17, the peptides were isolated as their cis/
trans isomers. It is important to note that Fmoc deprotection
prior to cleavage is essential not only for the separation of cis/
trans isomers but also for reduction via hydrogenation.
concentrated to give a white precipitate. The precipitate was
dissolved in H2O and subjected to the same RP-HPLC purifi-
cation. The saturated products eluted from the column after
their olefin precursors under the same conditions. In all cases,
yields were excellent (approximately quantitative) according
to HPLC and 1H NMR analyses. A sample of mixture 8 (3 mg,
31 µmol) was reduced as mentioned above with 10% Pd/C (3
mg) to give peptide 15 (∼3 mg, quant). Peptide 15 was isolated
as a single peak using C18, RP-HPLC, 10% MeCN, 90% H2O
(0.1% TFA), 5 min, 10-55% MeCN over 20 min, tR ) 13.25
min: 1H NMR (D2O, 600 MHz) δ 7.16 (apparent doublet, 2H,
J ) 8.4 Hz), 6.83 (apparent doublet, 2H, J ) 8.5 Hz), 4.70 (m,
2H), 4.40 (m, 2H), 4.28 (m, 1H), 4.14 (d, 1H, J ) 5.1 Hz), 4.08
(dd, 1H, J ) 5.6, 8.8 Hz), 3.98 (dd, 1H, J ) 5.1, 6.8 Hz), 3.86
(AB quartet, 2H, J ) 17.1 Hz), 3.74 (m, 1H), 3.60 (m, 1H),
3.15 (dd, 1H, J ) 6.1, 14.4 Hz), 2.95 (dd, 1H, J ) 8.6, 14.3
Hz), 2.85 (dd, 1H, J ) 5.4, 15.7 Hz), 2.75 (dd, 1H, J ) 8.8,
15.7 Hz), 2.37 (m, 2H), 2.25 (m, 1H), 2.00 (m, 3H), 1.88 (m,
5H), 1.66 (m, 3H), 1.58 (m, 2H), 1.42-1.21 (m, 7H), 1.14 (m,
1H), 1.05 (m, 1H), 0.91 (d, 3H, J ) 6.2 Hz), 0.89 (d, 3H, J )
6.9 Hz), 0.86 (d, 3H, J ) 6.5 Hz), 0.84 (t, 3H, J ) 7.4 Hz); 13
C
NMR (D2O, 150 MHz) δ 178.4, 176.1, 175.7, 175.2, 174.9, 174.3,
173.6, 173.2, 172.5, 155.2, 131.3, 128.5, 116.5, 72.3, 61.2, 59.5,
56.0, 55.4, 53.9, 53.5, 53.0, 50.9, 48.2, 43.1, 40.3, 39.2, 37.5,
36.5, 32.0, 31.2, 30.7, 30.0, 27.1, 26.0, 25.5, 24.1, 23.2, 21.5,
15.7, 11.7; ES (MS) calcd for C46H72N12O12 984.5, found 986
(M + H, 100), 1008 (M + Na, 69).
cis-[1,6-r,r′-L,L-Diaminonon-γ-enedioic acid]oxytocin
(6). Peptide 6 was isolated as a single peak using C18, RP-
HPLC, 10% MeCN, 90% H2O (0.1% TFA), 5 min, 10-55%
MeCN over 20 min, tR ) 12.99 min (24 mg, 10% from 0.25
mmol of resin bound peptide cleaved): 1H NMR (D2O, 600
MHz) δ 7.19 (apparent doublet, 2H, J ) 8.2 Hz), 6.87 (apparent
doublet, 2H, J ) 8.2 Hz), 5.68 (hidden AB quartet, 1H, J )
8.5 Hz), 5.51 (hidden AB quartet, 1H, J ) 8.5 Hz), 4.64 (m,
1H), 4.42 (m, 1H), 4.30 (m, 1H), 4.16 (m, 1H), 4.12 (m, 1H),
3.93 (m, 1H), 3.89 (AB quartet, 2H, J ) 17.3 Hz), 3.74 (m,
1H), 3.61 (m, 1H), 3.10 (dd, 1H, J ) 6.8, 14.3 Hz), 3.04 (dd,
1H, J ) 7.2, 14.3 Hz), 2.90 (m, 1H), 2.75 (m, 3H), 2.36 (m,
3H), 2.26 (m, 1H), 2.16 (m, 1H), 2.02 (m, 6H), 1.92 (m, 3H),
1.76-1.56 (m, 5H), 1.17 (m, 1H), 0.94 (d, 3H, J ) 5.9 Hz), 0.89
(1-Methoxycarbonylmethylcyclohexyl)acetic Acid (25).
To 1,1-cyclohexanediacetic acid (5.00 g, 25.0 mmol) was added
acetyl chloride (7 mL, 98.0 mmol). This mixture was then
heated at 60 °C for 3 h and was observed to go from a
suspension to a homogeneous solution. The volatiles were then
removed at 100 °C under reduced pressure leaving behind a
clear liquid which solidified upon cooling to room temperature
to give cyclohexylglutaric anhydride 24 as a semitransparent
solid (4.42 g, 97%). This was used immediately in the next
reaction without further purification. The anhydride 24 was
dissolved in dry MeOH (100 mL), NaOMe (1.56 g, 25.0 mmol)
was added, and the reaction mixture was stirred at reflux for
12 h. The mixture was then acidified to pH 4 with AcOH, and
the methanol was removed under reduced pressure. The
residue was dissolved in water (100 mL) and extracted with
EtOAc (2 × 125 mL). The combined organic layers were
washed with brine, dried (Na2SO4), and concentrated in vacuo
to yield a colorless, viscous oil, 25 (6.27 g, quant): IR (cast,
CHCl3) 3450-2350, 1740, 1705, 1440, 1409, 1336, 1286, 1254,
1166 cm-1; 1H NMR (CDCl3, 300 MHz) δ 11.61 (br s, 1H), 3.60
(s, 3H), 2.52 (s, 2H), 2.49 (s, 2H), 1.20 (m, 10H); 13C (CDCl3,
125 MHz) δ 177.9, 173.1, 51.3, 35.9, 35.3, 25.7, 21.6, 21.5;
HRMS (EI) calcd for C11H18O4Na 237.1097, found 237.1097
(M + Na).
(1-Hydroperoxycarbonylmethylcyclohexyl)acetic Acid
Methyl Ester (26). Compound 25 (5.04 g, 23.5 mmol) was
dissolved in concd H2SO4 (3 mL) at 0 °C. Aqueous H2O2 (50%)
was then slowly added, and the mixture was stirred at 0 °C
for 2 h. The mixture was warmed to rt and stirred for an
additional 1 h. It was then quenched with ice and diluted with
Et2O (50 mL). The organic layer was separated, washed with
H2O (50 mL) and saturated NaHCO3 (50 mL), dried (Na2SO4),
and concentrated in vacuo to yield the peracid 26 as a colorless,
viscous oil (4.22 g, 78%). The peracid was used without further
purification in the next step due to volatility: IR (cast, CHCl3)
3274, 2930, 2857, 1733, 1456, 1440 cm-1; 1H NMR (CDCl3, 300
MHz) δ 11.12 (br s., 1H), 3.60 (s, 3H), 2.59 (s, 1H), 2.55 (d,
2H, 36.3 Hz), 2.48 (d, 2H, 17.61 Hz); 13C NMR (CDCl3, 100
MHz) δ 172.4, 171.9, 51.3, 40.7, 37.4, 35.7, 35.4, 25.3, 21.2;
HRMS (ES) calcd for C11H19O5 231.1227, found 231.1228
(M + H).
(d, 3H, J ) 6.0 Hz), 0.86 (d, 3H, J ) 6.8 Hz), 0.81 (m, 3H); 13
C
NMR (D2O, 150 MHz) δ 176.1, 175.6, 175.2, 174.9, 174.1, 173.9,
172.7, 155.1, 135.1, 131.4, 128.5, 123.0, 116.5, 61.2, 60.8, 56.7,
55.6, 53.4, 52.8, 50.8, 49.0, 48.8, 43.3, 40.2, 36.0, 32.1, 29.6,
27.0, 26.3, 25.4, 23.2, 21.7, 16.0, 11.6; MALDI-TOF (MS) calcd
for C46H70N12O12 982.5, found 983.5 (M + H, 15%), 1005.5
(M + Na, 100), 1021.5 (M + K, 44).
trans-[1,6-r,r′-L,L-Diamino-δ-non-γ-enedioic acid]oxy-
tocin (7). Peptide 7 was isolated as a single peak using C18,
RP-HPLC, 10% MeCN, 90% H2O (0.1% TFA), 5 min, 10-55%
MeCN over 20 min, tR ) 13.32 min (16 mg, 6% from 0.25 mmol
resin bound peptide cleaved): 1H NMR (D2O, 600 MHz) δ 7.19
(apparent doublet, 2H, J ) 7.1 Hz), 6.87 (apparent doublet,
2H, J ) 8.3 Hz), 5.70 (hidden AB quartet, 1H, J ) 15.6 Hz),
5.45 (hidden AB quartet, 1H, J ) 15.6 Hz), 4.65 (m, 1H), 4.44
(m, 2H), 4.28 (m, 1H), 4.12 (m, 1H), 4.07 (m, 1H), 4.03 (d, 1H,
J ) 6.2 Hz), 3.94 (m, 1H), 3.90 (AB quartet, 2H, J ) 17.2 Hz),
3.76 (m, 1H), 3.62 (m, 1H), 3.16 (dd, 1H, J ) 7.0, 14.4 Hz),
2.98 (dd, 1H, J ) 8.42, 14.4 Hz), 2.78 (m, 3H), 2.58 (m, 1H),
2.38 (m, 2H), 2.29 (m, 1H), 2.12 (m, 1H), 2.04 (m, 5H), 1.86
(m, 2H), 1.70 (m, 2H), 1.66 (m, 2H), 1.60 (m, 1H), 1.30 (m,
1H), 1.04 (m, 1H), 0.94 (d, 3H, J ) 6.1 Hz), 0.91 (d, 3H, J )
7.1 Hz), 0.89 (d, 3H, J ) 6.1 Hz), 0.84 (t, 3H, J ) 7.3 Hz); 13
C
NMR (D2O, 150 MHz) 178.5, 176.0, 175.1, 174.9, 174.1, 173.9,
172.2, 170.0, 155.3, 137.0, 131.5, 128.8, 122.5, 116.5, 61.4, 60.4,
56.3, 55.5, 53.6, 53.2, 52.3, 51.2, 48.7, 42.9, 40.8, 37.0, 36.5,
36.1, 34.6, 32.0, 30.2, 29.9, 27.3, 25.8, 25.3, 23.2, 22.3, 16.2,
11.3; MALDI-TOF (MS) calcd for C46H70N12O12 982.5, found
1005.5 (M + Na, 100), 1021.5 (M + K, 32).
[1,6-r,r′-L,L-Diaminononanedioic acid]oxytocin (15).
Hydrogenation of cyclic olefinic peptides 8, 9, and 17 was done
using standard hydrogenation techniques. The general proce-
dure for the reduction is as follows. The olefinic peptides were
dissolved in anhydrous EtOH followed by the addition of 10%
Pd/C. The reaction mixture was then stirred under a hydrogen
atmosphere at atmospheric pressure for ∼36 h. The reaction
mixtures were then filtered through a pad of Celite and
(2S)-2-Benzyloxycarbonylamiono-5-((1-methoxycar-
bonylmethylcyclohexyl)acetylperoxy)-5-oxopentanoic
Acid Benzyl Ester (28). To a solution of 26 (1.03 g, 4.51
mmol) in CH2Cl2 (25 mL) at 0 °C were added (N-Cbz-Glu-OBn)
7808 J. Org. Chem., Vol. 70, No. 20, 2005

Dicarba Antagonists of Oxytocin with Increased Stability
27 (1.34 g, 3.61 mmol) and DCC (0.93 g, 4.51). The reaction
mixture was stirred at 0 °C for 2 h, slowly warmed to rt, and
filtered through Celite. The filtrate was concentrated in vacuo
and the residue dissolved in EtOAc and cooled to -20 °C for
several hours and filtered. The filtrate was concentrated in
vacuo to yield a gum which was then subjected to column
chromatography (SiO2, 1:1 hexanes/EtOAc) to give 28 as a
colorless gum (1.69 g, 80%): IR (cast, CHCl3) 3346, 3033, 2931,
1806, 1776, 1731, 1524, 1454 cm-1; 1H NMR (CDCl3, 400 MHz)
δ 7.31 (m, 10H), 5.49 (d, 1H, J ) 7.6 Hz), 5.18 (s, 2H), 5.08 (s,
2H), 4.42 (m, 1H), 3.60 (s, 3H), 2.63 (m, 2H), 2.54 (s, 2H), 2.43
(m, 1H), 2.30 (m, 1H), 2.06 (m, 1H), 1.60-1.38 (m, 10H); 13C
NMR (CDCl3, 125 MHz) δ 172.1, 171.2, 168.3, 167.4, 167.0,
156.0, 136.1, 135.0, 128.7, 128.6, 128.5, 128.4, 128.2, 128.1,
67.6, 67.2, 53.2, 51.3, 40.8, 37.0, 36.1, 35.6, 35.5, 35.4, 27.6,
26.2, 25.6, 25.4, 21.4, 21.3; HRMS (ES) calcd. for C31H37NO10
583.2490 (M + H), found 584.2488; [R]D (c 1.0, CHCl3) ) +2.5.
2H, J ) Hz), 7.58 (m, 2H), 7.39 (apparent triplet, 2H, J ) Hz),
7.23 (apparent triplet, 2H, J ) Hz), 5.43 (d, 1H, J ) Hz), 4.40
(m, 3H), 4.21 (dd, 1H, J ) Hz), 3.59 (s, 3H), 2.25 (s, 2H), 1.88
(m, 1H), 1.69 (m, 1H), 1.54-1.21 (m, 14H); 13C NMR (CDCl3,
125 MHz) δ 177.0, 173.1, 156.2, 143.8, 141.3, 127.8, 127.7,
127.1, 125.1, 120.0, 67.2, 53.8, 51.3, 47.2, 36.0, 35.8, 32.9, 26.1,
21.6, 21.6, 18.9; HRMS (ES) calcd for C29H35NO6Na 516.2362,
found 516.2362 (M + Na); [R]D (c 1.0, CHCl3) ) +8.1.
[(1,6-r-L-Amino-r′-deamino-â′,â′-cyclohexylsuberic acid)-
2-D-thienyl-9-L-tyrosyl]atosiban (35). Peptide sythesis was
done on a 0.15 mmol scale using Sieber amide resin (0.62
mmol/g) as solid support. The general method for Fmoc SPPS
was similar to that for the other peptides. The fully protected
linear precursor 31 was cleaved from the resin using 1% TFA
in CH2Cl2 Purification by C18, RP-HPLC, 10% MeCN, 90% H2O
(0.1% TFA), 5 min, 10-90% MeCN over 20 min (tR ) 24.52
min) gave pure peptide methyl ester 32 as a white solid (42
mg, 18%): MALDI-TOF (MS) calcd for C87H125N11O15SNa
1602.9, found 1603.0 (M + Na). The methyl ester 32 (42 mg,
0.03 mmol) was treated with 2 M LiOH (2 mL) for 3 d at rt
and purified with prep-HPLC as above (tR ) 23.88 min) to give
the peptide acid 33 (25 mg, 59%) as a white solid: MALDI-
TOF (MS) calcd for C85H119N11O15SNa 1589.9, found 1588.9
(M + Na). The peptide acid 33 (25 mg, 18 µmol) was dissolved
in dry DMF (15 mL), and PyBOP (56 mg, 0.11 mmol), HOBt
(15 mg, 0.11 mmol), and NMM (24 µL, 0.22 mmol) were added.
The reaction mixture was then stirred in the dark for 2.5 d,
concentrated, and purified by RP-HPLC (tR ) 26.92 min) as
above to give the protected cyclized product 34 (8 mg, 29%):
MALDI-TOF (MS) calcd for C85H117N11O14SNa 1570.8, found
1571.8 (M + Na). The protected, cyclized peptide 34 (8 mg, 50
µmol) was treated with 18:1:1, TFA/CH2Cl2/Et3SiH for 4 h at
rt. Purification with RP-HPLC as above (tR )16.39 min)
yielded the fully deprotected peptide 35 (4 mg, 76%) as a white
solid: 1H NMR (D2O, 600 MHz) δ 7.31 (d, 1H, J ) 4.4 Hz),
6.98 (m, 4H), 6.55 (apparent doublet, 2H, J ) 8.1 Hz), 4.75
(m, 1H), 4.62 (m, 1H), 4.55 (m, 1H), 4.51 (m, 1H), 4.46 (t, 1H,
J ) 7.8 Hz), 4.41-4.31 (m, 2H), 4.16 (m, 2H), 3.77 (m, 1H),
3.62 (m, 1H), 3.32 (m, 2H), 2.94 (m, 2H), 2.76 (m, 2H), 2.56
(m, 1H), 2.45 (m, 1H), 2.25 (m, 2H), 2.12 (m, 1H), 2.19 (m,
4H), 1.84-1.61 (m, 5H), 1.52 (m, 1H), 1.40 (m, 5H), 1.26 (m,
8H), 1.15 (m, 6H), 0.83 (m, 6H); 13C NMR (D2O, 150 MHz)
(HMQC diagnostic peaks only) δ 131.2 (D-Thi), 128.2 (Tyr),
125.9 (D-Thi), 119.4 (Tyr); MALDI-TOF (MS) calcd for
C53H79N11O12S 1093.6, found 1094.2 (M + H, 100), 1116.2
(M + Na, 35), 1132.1 (M + K, 30).
(2S)-2-Benzyloxycarbonylamino-5-((1-methoxycarbon-
ylmethyl)cyclohexyl)pentanoic Acid Benzyl Ester (29).
A solution of 28 (500 mg, 0.86 mmol) in CH2Cl2 (5 mL) was
spread out over the bottom of a 500 mL Pyrex reaction vessel.
The solvent was removed with a steady stream of argon to
afford a thin layer of neat diacyl peroxide. The vessel was then
covered with a quartz glass plate and sealed under an argon
atmosphere. The reaction vessel was then immersed in a cold
bath at -78 °C. A 0.9 Amp UV lamp was placed directly onto
the quartz plate, and the neat sample was irradiated at 254
nm for 36 h. The reaction mixture was warmed to rt, and the
residue was extracted extensively with CH2Cl2. The solvent
was removed in vacuo to yield a yellow oil which was subjected
to column chromatography (SiO2, 1:6 to 1:1 EtOAc/hexanes)
to afford 29 as a colorless gum (146 mg, 34%) as well as some
recovered starting material (∼30%): IR (cast, CHCl3) 3349,
3064, 3032, 2929, 2855, 1728, 1523, 1455 cm-1 1H NMR
;
(CDCl3, 400 MHz) δ 7.35 (m, 10H), 5.39 (d, 1H, J ) 8.2 Hz),
5.24-5.08 (m, 4H), 4.42 (m, 1H), 3.60 (s, 3H), 2.21 (s, 2H), 1.86
(m, 1H), 1.68 (m, 1H), 1.45-1.10 (m, 14H); 13C NMR (CDCl3,
125 MHz) δ 172.7, 172.4, 155.9, 136.3, 135.4, 128.6, 128.5,
128.4, 128.3, 128.2, 128.1, 128.0, 127.9, 67.1, 54.0, 51.0, 41.4,
35.9, 35.6, 35.2, 33.4, 26.0, 21.6, 18.7; HRMS (ES) calcd for
C29H38NO6 496.2694 (M + H), found 496.2690; [R]D (c 1.0,
CHCl3) ) -8.0.
(2S)-2-Amino-5-((methoxycarbonylmethyl)cyclohexyl)-
pentanoic Acid (30). Compound 29 (602 mg, 1.21 mmol) was
dissolved in anhydrous EtOH (10 mL), and 10% Pd/C (58 mg)
was added. The reaction mixture was then stirred under H2
at 1 atm for 12 h. The mixture was then filtered through a
pad of Celite, and the filtrate was concentrated in vacuo to
give the amine 30 as a white solid (328 mg, quant), which was
used in the next step without any further purification: IR
Acknowledgment. We thank Glen Bigam, Thomas
Nakashima, Albin Otter, and Randy Whittal (Chemistry
Department, University of Alberta) for assistance with
spectral analyses. We are grateful to Steven Schendel
and Xuan Chen for help with peptide synthesis. These
investigations were supported by the Natural Sciences
and Engineering Council of Canada (NSERC), the
Alberta Heritage Foundation for Medical Research
(AHFMR), and the Canada Research Chair in Bioor-
ganic and Medicinal Chemistry. The authors gratefully
acknowledge Anne Tykwinski (Evita McConnell Graph-
ics, Inc.) for cover art design and Klaus-Peter Kelber
(Institute of Mineralogy, University of Wurzburg, http://
www.uni-wuerzburg.de/mineralogie) for the photograph
of sulfur crystals.
(µscope) 3450-2450, 1733, 1586, 1455 cm-1 1H NMR (CD3-
;
OD, 300 MHz) δ 3.62 (s, 3H), 3.56 (dd, 1H, J ) Hz), 2.32 (s,
2H), 1.81 (m, 2H), 1.61-1.30 (m, 14H); 13C NMR (CD3OD, 100
MHz) δ 174.6, 173.4, 63.6, 51.7, 43.2, 42.3, 33.1, 27.2, 22.7,
19.9; HRMS (ES) calcd. for C14H26NO4 272.1856, found 272.1852
(M + H); [R]D (c 1.0, CH3OH) ) +7.4.
(2S)-2-(9H-Fluorenylmethoxycarbonylamino)-5-(1-meth-
oxycarbonylmethylcyclohexyl)pentanoic Acid (23). Com-
pound 30 (328 mg, 1.21 mmol) was dissolved in 10% aqueous
NaCO3 (20 mL) and dioxane (10 mL) and cooled to 0 °C. A
solution of 9H-fluorenylmethoxyl chloroformate (374 mg, 1.39
mmol) in dioxane (15 mL) was then added dropwise over 20
min, and the mixture was stirred at 0 °C for 2.5 h upon
completion. The reaction mixture was then warmed to rt and
stirred for 1.5 h. Concentration of the mixture in vacuo yielded
a residue that was dissolved in water (50 mL) and extracted
with Et2O (2 × 50 mL). The aqueous layer was acidified to pH
3 with 1 M HCl and extracted with EtOAc (4 × 50 mL). The
combined organic layers were dried over Na2SO4 and concen-
trated to give the product 23 as a white foam (224 mg, 38%):
Supporting Information Available: All general proce-
dures, full experimental details (peptides 3, 8-12, and 16-
18), 1H NMR assignments for peptides 2, 3, 6-12, 16-18, and
35, Schild plots for 2, 3, 17, 18, and 35, and a sample double
decoupled spectrum of peptide 6. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
IR (CHCl3 cast) 3325-2860, 1722, 1525, 1450 cm-1; H NMR
(CDCl3, 300 MHz) δ 9.40 (br s, 1H), 7.75 (apparent doublet,
JO050539L
J. Org. Chem, Vol. 70, No. 20, 2005 7809