Synthesis of a Novel Esterase-Sensitive Cyclic Prodrug System for Peptides That Utilizes a "Trimethyl Lock"-Facilitated Lactonization Reaction

Article abstract of DOI:10.1021/jo961778z

This paper describes a unique strategy for preparing cyclic prodrugs of peptides that have increased metabolic stability and increased cell membrane permeability when compared to the linear peptides. By taking advantage of a unique "trimethyl lock"-facilitated lactonization system, an esterase-sensitive cyclic prodrug of a model hexapeptide H-Trp-Ala-Gly-Gly-Asp-Ala-OH was synthesized by linking the N-terminal amino group to the C-terminal carboxyl group. The key intermediate for both approaches was compound 9 with Boc-Ala attached to the phenol hydroxyl group of the "trimethyl lock" linker through an ester bond, which can then be incorporated into the peptide using a normal coupling reagent for peptide synthesis. The synthesis of the linear peptides was accomplished using both solution-phase and solid-phase approaches with the solution-phase approach having the advantage of using the key intermediate 9 most efficiently. Cyclization using standard high-dilution techniques provided cyclic prodrug 13. In 90% human plasma, prodrug 13 released the original peptide, as designed, through an apparent esterase-catalyzed hydrolysis of the phenol ester bond.

Full text of DOI:10.1021/jo961778z

J . Org. Chem. 1997, 62, 1363-1367
1363
Syn th esis of a Novel Ester a se-Sen sitive Cyclic P r od r u g System for
P ep tid es Th a t Utilizes a “Tr im eth yl Lock ”-F a cilita ted
La cton iza tion Rea ction
Binghe Wang,† Sanjeev Gangwar,‡ Giovanni M. Pauletti, Teruna J . Siahaan, and
Ronald T. Borchardt*
Department of Pharmaceutical Chemistry, 2095 Constant Avenue, The University of Kansas,
Lawrence, Kansas 66047
Received September 17, 1996X
This paper describes a unique strategy for preparing cyclic prodrugs of peptides that have increased
metabolic stability and increased cell membrane permeability when compared to the linear peptides.
By taking advantage of a unique “trimethyl lock”-facilitated lactonization system, an esterase-
sensitive cyclic prodrug of a model hexapeptide H-Trp-Ala-Gly-Gly-Asp-Ala-OH was synthesized
by linking the N-terminal amino group to the C-terminal carboxyl group. The key intermediate
for both approaches was compound 9 with Boc-Ala attached to the phenol hydroxyl group of the
“trimethyl lock” linker through an ester bond, which can then be incorporated into the peptide
using a normal coupling reagent for peptide synthesis. The synthesis of the linear peptides was
accomplished using both solution-phase and solid-phase approaches with the solution-phase
approach having the advantage of using the key intermediate 9 most efficiently. Cyclization using
standard high-dilution techniques provided cyclic prodrug 13. In 90% human plasma, prodrug 13
released the original peptide, as designed, through an apparent esterase-catalyzed hydrolysis of
the phenol ester bond.
Sch em e 1. Design of a n Ester a se-Sen sitive
P r od r u g System for Cyclic Der iva tiza tion of
P ep tid es
In tr od u ction
Two of the major obstacles to the development of
biologically active peptides as clinically useful therapeutic
agents have been their low permeability through biologi-
cal barriers (e.g., intestinal mucosa, blood-brain barrier)
and their metabolic lability.1 Earlier, we reported that
masking the C-terminal and N-terminal polar functional
groups of a peptide through cyclization with an (acyloxy)-
alkoxy linker can greatly enhance the membrane perme-
ability and metabolic stability of the linear peptide.2,3 In
this paper, we wish to report a new and generally
applicable approach to synthesizing esterase-sensitive
prodrugs of peptides by taking advantage of a unique
“trimethyl lock”-facilitated lactonization system.
Substituted phenol propionic acid derivatives such as
2, upon unmasking of the hydroxyl group, undergo a
facile spontaneous intramolecular cyclization to release
the moieties attached to the carboxyl functional group
(Scheme 1).4-6 The facile cyclization reaction is the result
of the “trimethyl lock,” which was shown earlier to
increase the rate of the cyclization reaction on the order
of 105-107.4-7 The result of such facilitation is that
intermediate 2 has a half-life of only approximately 100
s at room temperature in aqueous solution.8,9 Such
systems have been used to develop prodrugs of amines
and alcohols8-10 and redox-sensitive protecting groups of
amines.11 However, the facile lactonization of this system
also makes it an attractive target for the development of
a unique linker in an esterase-sensitive cyclic prodrug
system for peptides and peptide mimetics (Scheme 1). In
this approach, the C- and N-terminal ends of a linear
peptide can be masked by forming an ester and an amide
bond with the phenol hydroxyl and side-chain carboxyl
* To whom correspondence should be addressed. Phone: (913) 864-
3427. Fax: (913) 864-5736. E-mail: Borchardt@smissman.hbc.ukans.edu.
† Current address: Department of Chemistry, North Carolina State
University, Raleigh, NC 27695-8204.
‡ Current address: Prolinx, Inc., Bothwell, WA 98021.
X Abstract published in Advance ACS Abstracts, February 15, 1997.
(1) Oliyai, R.; Stella, V. J . Annu. Rev. Pharmacol. Toxicol. 1993, 32,
521-544.
(2) Gangwar, S.; Pauletti, G. M.; Siahaan, T. J .; Stella, V. J .;
Borchardt, R. T. J . Org. Chem. 1997, 62, 1363-1367.
(3) Pauletti, G. M.; Gangwar, S.; Okumu, F. W.; Siahaan, T. J .;
Stella, V. J .; Borchardt, R. T. Pharm. Res. 1996, 13, 1615-1623.
(4) Borchardt, R. T.; Cohen, L. A. J . Am. Chem. Soc. 1972, 94, 9175-
9182.
(5) King, M. M.; Cohen, L. A. J . Am. Chem. Soc. 1983, 105, 2752-
2760.
(6) Milstein, S.; Cohen, L. A. J . Am. Chem. Soc. 1972, 94, 9158-
9165.
(7) Wang, B.; Nicolaou, M. G.; Liu, S.; Borchardt, R. T. Bioorg. Chem.
1996, 24, 39-49.
(8) Amsberry, K. L.; Borchardt, R. T. J . Org. Chem. 1990, 55, 5867-
5877.
(9) Amsberry, K. L.; Gerstenberger, A. L.; Borchardt, R. T. Pharm.
Res. 1991, 8, 455-461.
(10) Ueda, Y.; Mikkilineni, A. B.; Knip, J . O.; Rose, W. C.; Casazza,
A. M.; Vyas, D. M. Bioorg. Med. Chem. Lett. 1993, 3, 1761-1766.
(11) Wang, B.; Liu, S.; Borchardt, R. T. J . Org. Chem. 1995, 60, 539-
543.
S0022-3263(96)01778-1 CCC: $14.00 © 1997 American Chemical Society

1364 J . Org. Chem., Vol. 62, No. 5, 1997
Wang et al.
Sch em e 2. Syn th etic Ap p r oa ch to th e
aspartic acid as the first amino acid to Wang’s resin13
through activation of the carboxyl group directly with
diisopropylcarbodiimide (DIC). On the basis of the
substitution density of the resin (0.95 mmol/g) and the
weight increase after the coupling of protected aspartic
acid, about 50% loading was achieved. Synthesis of the
linear pentapeptide attached to the resin 10 (Scheme 3)
was accomplished using standard Fmoc (9-fluorenyl-
methoxycarbonyl) chemistry, and the ninhydrin test was
used to determine whether the coupling was completed
for each step.14,15 The “trimethyl lock” linker 9 was
attached as the last protected amino acid using BOP
((benzotriazoyloxy)tris(dimethylamino)phosphonium
hexafluorophosphate) as the activating reagent to give
11a . The cleavage of the peptide from the solid support
with 50% trifluoroacetic acid (TFA) in DCM gave the
linear peptide 12 in about 50% yield. This linear peptide
was used directly for the cyclization reaction without
purification. Cyclization of 12 with BOP-Cl [N,N-bis(2-
oxo-3-oxazolidinyl)phosphinic acid] as the activating
reagent16 gave the cyclized product in about 8% yield,
which was then converted to the final product 13 through
catalytic hydrogenation. It should be noted that although
using more than 1 equiv of an amino acid is a common
practice in solid-phase peptide synthesis, it results in a
significant waste of the key intermediate 9. Because the
preparation of 9 represents the most time-consuming
process, we were interested in finding a synthetic ap-
proach to the final product, which would allow for a more
efficient use of 9. Therefore, we also studied the solution-
phase approach.
In ter m ed ia te w ith Boc-Ala Atta ch ed to th e Lin k er a
a
Key: (a) BocAla-OpNP, DMAP/DCM, reflux; (b) HOAc/THF/
H2O; (c) PCC; (d) KMnO4.
groups, respectively, of the linker, which is also referred
to as the pro-moiety (Prom) (Scheme 1). To demonstrate
the feasibility of this approach, we synthesized the cyclic
prodrug of a model hexapeptide (H-Trp-Ala-Gly-Gly-Asp-
Ala-OH) by linking the N-terminal amino group to the
C-terminal carboxyl group via this “trimethyl lock” linker.
The cyclic prodrug was shown to be 70 times more able
to permeate an in vitro cell culture model of the intestinal
mucosa than was the parent linear peptide.12 As it was
designed, the parent peptide was shown to be released
by an apparent esterase-catalyzed cleavage of the phenol
ester linkage (Scheme 1) in 90% human plasma.12
In the solution-phase approach, linear pentapeptide 14
was prepared using standard Boc-amino acid chemis-
try.15,17 The key to this solution-phase approach was the
selective protection of the R- and â-carboxyl groups of the
Asp residue. We successfully used the trichloroethyl
(Tce) ester protecting group for the R-carboxyl group of
Asp residue. Compound 14 was treated with 50% TFA
in DCM to give 15 in quantitative yield. The pentapep-
tide 15 was reacted with 9 in the presence of 1-(3-
(dimethylamino)propyl)ethylcarbodiimide hydrochloride
(EDC), 1-hydroxybenzotriazole (HOBt), and N-methyl-
morpholine (NMM) to give the fully protected linear
hexapeptide 11b in 68% yield. The protecting groups
were removed using zinc/HOAc (acetic acid) and 50%
TFA/DCM to provide 12 in 50% overall yield. Compound
12 was purified by preparative reversed-phase HPLC.
Cyclization was then accomplished by a standard high-
dilution technique using BOP-Cl as an activating re-
agent16 in the presence of NMM and DMAP to afford the
cyclic peptide with Asp-â-benzyl protection in 15% yield.
Hydrogenolysis of the protected cyclic peptide provided
the desired cyclic prodrug 13 in quantitative yield.
Through comparison of the two methods, it can be seen
that the solution-phase approach allows for somewhat
more efficient use of the key intermediate 9. In the solid-
phase approach, the yield of the linear peptide 12
calculated from the key intermediate 9 was 25%, and in
the solution-phase approach the yield was 38%. The
Resu lts a n d Discu ssion
Due to the facile cyclization of the system described,
the incorporation of the “linker” into the cyclic system
requires first the protection of the phenol hydroxyl group.
Therefore, the key intermediate for successful synthesis
of cyclic prodrug 13 was compound 9 with Boc-Ala
attached to the phenol hydroxyl group of the “linker”
through an ester bond (Scheme 2). The synthesis of 9
started with commercially available 3,5-dimethylphenol.
The tert-butyldimethyl silyl (TBDMS)-protected diol 5
was prepared by a literature procedure9 in three steps
with over 90% yield for each step. Reaction of activated
p-nitrophenol (pNP) ester of the N-Boc-alanine with 5
in the presence of 4-(N,N-dimethylamino)pyridine (DMAP)
in refluxing dichloromethane (DCM) gave 6 in 80% yield.
Treatment of the TBDMS ether 6 with acetic acid in the
presence of water and THF yielded the primary alcohol
7, which was converted in a two-step oxidation with
pyridinium chlorochromate (PCC) and KMnO4 to the key
intermediate 9. It should be noted that the Boc-alanyl
group of 7 has a tendency to migrate to the primary
hydroxyl group, presumably due to the high steric
hindrance of the phenol hydroxyl position. This migra-
tion occurs more readily during column chromatography.
Therefore, compound 7 was used for the oxidation
without purification.
Because solid-phase peptide synthesis normally allows
for faster synthesis of a peptide than does the solution-
phase approach, we first employed the solid-phase ap-
proach to the synthesis of the linear peptide. This
synthesis involves the attachment of â-benzyl-protected
(13) Wang, S. S. J . Am. Chem. Soc. 1973, 95, 1328-1333.
(14) Atherton, E.; Sheppard, R. C. Solid Phase Peptide Synthesis:
A Practical Approach; IRL Press: New York, 1989.
(15) Stewart, J . M.; Young, J . D. Solid Phase Peptide Synthesis;
Pierce Chemical Company: Rockford, IL, 1984.
(16) Tung, R. D.; Rich, D. H. J . Am. Chem. Soc. 1985, 107, 4342-
4343.
(12) Pauletti, G. M.; Gangwar, S.; Wang, B.; Borchardt, R. T. Pharm.
Res. 1997, 14, 11-17.
(17) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis;
Springer-Velag: New York, 1984.

Synthesis of an Esterase-Sensitive Cyclic Prodrug System
J . Org. Chem., Vol. 62, No. 5, 1997 1365
Sch em e 3. Syn th etic Ap p r oa ch to th e Cyclic P r od r u g of th e Mod el Hexa p ep tid e
Sch em e 4. Solu tion -P h a se Syn th etic Ap p r oa ch to
Cyclic P r od r u g 13a
enzymatic degradation by peptidases present in Caco-2
cell monolayers (t1/2 ) 14 min). However, even when the
linear hexapeptide was stabilized to peptidase-catalyzed
metabolism, using a cocktail of potent peptidase inhibi-
tors, it was still substantially less able to permeate the
Caco-2 cellular barrier than was the cyclic prodrug 13.3
Con clu sion s
We have described a strategy for preparing a cyclic
prodrug of a model hexapeptide via the N- and C-
terminal ends by utilizing a unique “trimethyl lock”-
facilitated lactonization system. However, it should be
feasible to use this methodology to cyclize other biologi-
cally active peptides by linking the C-terminal carboxyl
group to a side-chain amino (e.g., Lys, Arg) or hydroxyl
(e.g., Ser, Thr, Tyr) group or linking a side-chain carboxyl
group (e.g., Asp, Glu) to a side-chain amino (e.g., Lys, Arg)
or hydroxyl (e.g., Ser, Thr, Tyr) group. The application
of this methodology to biologically active peptides (e.g.,
opioid peptides) and peptide mimetics [e.g., Arg-Gly-Asp
(RGD) analogs] is currently under investigation in our
laboratories.
a
(a) TFA/DCM, 50%; (b) 9, EDC, HOBt, NMM; (c) Zn/HOAc;
(d) TFA/DCM, 50%; (e) BOP-Cl, NMM, DMAP; (f) H2Pd-C.
overall low yield (8-15%) for the cyclization was probably
due to (1) possible oligomerization even under high-
dilution conditions and (2) low recovery from preparative
HPLC purification due to the small amount of sample
that was used. One could certainly use even higher
dilution conditions to minimize the possible oligomeriza-
tion problem.
Prodrug 13 was designed to release its original peptide
through an esterase-catalyzed hydrolysis of the phenol
ester bond (Scheme 1). The stability of cyclic prodrug
13 was evaluated in 90% human plasma and was
compared to its chemical degradation in a physiological
buffer system (Hanks’ balanced salt solution (HBSS), pH
7.4).12 In HBSS, cyclic prodrug 13 degraded slowly (t1/2
) 1795 ( 289 min) but quantitatively to the hexapeptide.
The rate of degradation of cyclic prodrug 13 was signifi-
cantly faster in 90% human plasma (t1/2 ) 508 ( 24 min),
and it could be slowed (t1/2 ) 1729 ( 245 min) substan-
tially by inclusion of paraoxon, a known esterase inhibi-
tor, in the incubation mixture. In 90% human plasma,
only trace amounts of the hexapeptide were observed
when cyclic prodrug 13 was incubated in the presence
or absence of paraoxon because of the instability of this
natural peptide (t1/2 ) 3.7 min) to peptidases in this
biological medium. The relative membrane permeabili-
ties of cyclic prodrug 13 and the linear hexapeptide were
evaluated using monolayers of Caco-2 cells, an in vitro
cell culture model of the intestinal mucosa.18 The ap-
parent permeability coefficient (Papp) of the linear hexapep-
tide was estimated to be less than 0.17 × 10-8 cm/s,
whereas cyclic prodrug 13 exhibited a Papp value of 12.1
× 10-8 cm/s. The low permeation of the linear model
hexapeptide was due in part to its high susceptibility to
Exp er im en ta l Section
Gen er a l Meth od s. 1H NMR spectra were recorded on
either a 500 MHz or a 300 MHz instrument. High-perfor-
mance liquid chromatography (HPLC) was conducted using a
dual pump system with a UV detector. All starting materials
were purchased from Aldrich Chemical Co., Sigma, Fluka
Chemicals, or Bachem Bioscience, Inc., and used as received.
Wang’s resin for solid-phase peptide synthesis was obtained
from Bachem Bioscience, Inc., with 0.95 mmol/g substitution.
1-O-(t er t -Bu t yld im e t h ylsilyl)-3-(2′-h yd r oxy-4′,6′-d i-
m eth ylp h en yl)-3,3-d im eth ylp r op a n ol (5). To a mixture of
11.63 g of 3-(2′-hydroxy-4′,6′-dimethylphenyl)-3,3-dimethyl-
propanol9 and 9.3 g of TBDMS chloride in 70 mL of methylene
chloride cooled in an ice bath was added dropwise a solution
of 30 mL of methylene chloride and 31 mL of triethylamine
during a period of 1 h with stirring. The reaction mixture was
then warmed to room temperature and stirred for 1 d. Then
solvent was evaporated, and the residue was dissolved in 250
mL of methylene chloride and washed with water (20 mL ×
4). After the mixture was dried over MgSO4 for 1 h, solvent
evaporation gave 18.09 g (100%) of a white solid product, the
NMR spectrum of which is identical with that reported in the
literature.9
1-O-(ter t-Bu t yld im et h ylsilyl)-3-(2′-Boc-a la n yl-4′,6′-d i-
m eth ylp h en yl)-3,3-d im eth ylp r op a n ol (6). To a mixture of
5.18 g of the TBDMS-diol 5, 5 g of Boc-Ala-OpNP, and 2 g of
DMAP was added 20 mL of anhydrous methylene chloride.
(18) Hidalgo, I. J .; Raub, T. J .; Borchardt, R. T. Gastroenterology
1989, 96, 736-749.

1366 J . Org. Chem., Vol. 62, No. 5, 1997
Wang et al.
After the mixture was refluxed for 7 h, another 0.4 g of DMAP
was added, and the reaction was refluxed for an additional 7
h. After solvent evaporation, the residual was dissolved into
200 mL of methylene chloride and washed with saturated
NaHCO3 (20 mL × 4), 1% HCl (20 mL × 2), and brine (20 mL
× 2). After the mixture was dried over Na2SO4 and the solvent
evaporated, the residue was purified by column chromatog-
raphy (silica gel) using a gradient of 5-10% ethyl acetate in
hexane to give a colorless oil, which later solidified to give a
white solid product (6.87 g, 84%). 1H-NMR (CDCl3, δ): 6.80
(1 H, s, aromatic H), 6.50 (1 H, s, aromatic H), 5.18 (1 H, m,
NH), 4.51 (1 H, m, NHCH), 3.47 (2 H, t, J ) 7.6 Hz, OCH2),
2.51 (3 H, s, PhCH3), 2.21 (3 H, s, PhCH3), 2.01 (2 H, m,
OCH2CH2), 1.54 (3 H, d, J ) 7.3 Hz, CHCH3), 1.45 (15 H, s,
OC(CH3)3, C(CH3)2), 0.83 (9 H, s, SiC(CH3)3, -0.04 (6 H, s,
Si(CH3)2). CIMS (NH3) (rel intensity): 494 (M + 1, 1), 438
(4), 394 (9). HRMS: calcd for C27H48NO5Si 494.3302, found
494.3316. Anal. Calcd for C27H47NO5Si: C, 65.68; H, 9.59;
N, 2.84. Found: C, 65.90; H, 9.91; N, 2.76.
3-(2′-Boc-a la n yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r o-
p a n ol (7). To 5.0 g of the Boc-Ala-TBDMS-diol 6 were added
19 mL of THF, 19 mL of H2O, and 58 mL of HOAc. The
reaction was stirred at rt for 1 h. Then solvent was evaporated
to give a colorless oily residue (4.8 g). This product was used
directly for the oxidation without purification because of the
potential for rearrangement during column chromatography.
1H-NMR (CDCl3, δ): 6.79 (1 H, s, aromatic), 6.51 (1 H, s,
aromatic), 5.35 (1 H, m, NH), 4.47 (1 H, m, NHCH), 3.49 (2 H,
m, HOCH2), 2.49 (3 H, s, PhCH3), 2.20 (3 H, s, PhCH3), 2.04
(2 H, m, HOCH2CH2), 1.54 (3 H, d, J ) 7.4 Hz, CHCH3), 1.46
(6 H, s, C(CH3)2), 1.42 (9 H, s, C(CH3)3).
3-(2′-Boc-a la n yl-4′,6′-d im eth ylp h en yl)-3,3-d im eth ylp r o-
p ion ic Acid (9). To a solution of 4.38 g of PCC in 200 mL of
DCM was added dropwise a solution of alcohol 7 (3.90 g) in
200 mL of DCM. The reaction color changed from orange to
black during the addition. The solution was then stirred at
rt for 1 h. After filtration through Celite, solvent evaporation,
and subsequent purification with a short silica gel column, 3.08
g of an oily product was obtained (79%). The oily product was
then dissolved in 40 mL of acetone and added dropwise to a
solution of 1.42 g of KMnO4 in 40 mL of acetone-H2O (1:1).
The reaction mixture was stirred at rt for 17 h. Acetone was
then evaporated from the reaction mixture, and the residue
was filtered through Celite. After filtration, the pH (about 8)
was adjusted to 3 with dilute HCl. The mixture was then
extracted with ethyl acetate (100 mL × 3), and the combined
organic extracts were dried over MgSO4 overnight. Solvent
evaporation gave a white solid product (2.9 g, 88%). 1H-NMR
(CDCl3, δ): 6.72 (1 H, s, aromatic), 6.44 (1H, s, aromatic), 5.09
(1 H, m, NH), 4.30 (1 H, m, NHCH), 2.74 (2 H, b, CH2COO),
2.44 (3 H, s, PhCH3), 2.13 (3 H, s, PhCH3), 1.46 (9 H, m,
CHCH3, C(CH3)2), 1.35 (9 H, s, C(CH3)3). FABMS: 394 (M +
1), 338, 294. HRMS: calcd (M + 1) 394.2230, found 394.2231.
Anal. Calcd for C21H31NO6: C, 64.10; H, 7.94; N, 3.56.
Found: C, 63.92; H, 8.10; N, 3.30.
Syn th esis of Lin ea r P ep tid e 12. A typical synthesis used
Wang’s alkoxy resin (Bachem Bioscience, 0.95 mmol/g, 100-
200 mesh). Fmoc-protected amino acids were used for solid-
phase peptide synthesis unless otherwise indicated. For a
scale of 0.5 g of resin, Asp (â-benzyl-aspartic acid) (4.5 equiv)
was preactivated with DIC for 10-20 min before being added
to the resin together with 1 equiv of DMAP. The coupling
reaction was carried out for 2.5 h and then washed with DMF
(20 mL × 4) and DCM (30 mL × 3). The coupling reaction
was repeated using 2.3 equiv of Asp using the same conditions.
For subsequent coupling reactions, about 4 equiv of the amino
acids were used with DIC/HOBt as the activating reagents,
and reactions were carried out for about 2 h. The coupling
reaction was monitored using a ninhydrin test as well as by
the weight of the resin and was repeated if the ninhydrin test
indicated incomplete reaction after 2 h. Deprotection was
accomplished using 20% piperidine in DMF in 10-15 min. The
coupling of compound 9 (1-1.5 equiv) was accomplished using
BOP as the activating reagent in the presence of triethylamine.
The final cleavage of the peptide from the resin was ac-
complished by treatment with 50% TFA/DCM at rt for 30 min.
Solvent evaporation under reduced pressure gave the final
linear peptide, which was directly used for cyclization without
purification. 1H-NMR (CD3OD, δ): 7.54 (1 H, d, aromatic),
7.31 (6 H, m, aromatic), 7.10 (1 H, m, aromatic), 7.02 (1 H, m,
aromatic), 6.98 (1 H, b, aromatic), 6.81 (1 H, b, aromatic), 6.58
(1 H, b, aromatic), 5.10 (2 H, s, PhCH2), 4.83 (1 H, t, R-H),
4.54 (1 H, t, R-H), 4.37 (1 H, m, R-H), 4.15 (1 H, m, R-H), 3.88
(2 H, b, Gly-R-H), 3.61 (2 H, b, Gly-R-H), 3.17-2.89 (4 H, m,
Asp-CH2, Trp-CH2), 2.63 (2 H, s, (CH3)2CCH2CO), 2.43 (3 H,
s, PhCH3), 2.19 (3 H, s, PhCH3), 1.71 (3 H, d, Ala-CH3), 1.46
(6 H, s, C(CH3)2), 1.21 (3 H, d, Ala-CH3). FABMS: calcd for
C45H55N7O11 869, found 870.5 (M + 1).
Cyclic P ep tid e 13. To 94 mg of linear peptide 12 was
added 10 mL of anhydrous DMF, 240 mL of anhydrous DCM,
100 µL of NMM, and 158 mg of BOP-Cl. The reaction solution
was stirred for 34 h. The reaction mixture was then washed
with water (20 mL × 2) and dried over MgSO4 for 2 h.
Purification with HPLC (reversed phase, 70% methanol in
water) afforded 7 mg of a pure cyclic peptide with benzyl
protection of the â-carboxyl group of Asp. 1H-NMR (CDCl3,
δ): 6.5-7.6 (19 H, NH, aromatic), 5.14 (2 H, m, PhCH2), 4.82
(1 H, m, R-H), 4.52 (1 H, m, R-H), 4.38 (1 H, m, R-H), 3.66-
4.03 (5 H, m, R-H), 2.88-3.11 (4 H, Asp-CH2, Trp-CH2), 2.71
(1 H, d, (CH3)2CCH2), 2.51 (3 H, s, CH3), 2.40 (1 H, d, (CH3)2-
CCH2), 2.17 (3 H, s, CH3), 1.5-1.7 (9 H, m, CH3), 1.13 (3 H, d,
CH3). FABMS: calcd for C45H53N7O10 851, found 852.7 (M +
1). HRMS: calcd (M + 1) 852.3932, found 852.3953.
To 7 mg of the aforementioned cyclic peptide with benzyl
protection of the â-carboxyl group of aspartic acid was added
20 mL of absolute ethanol and 3 mg of Pd-C (10%). The
reaction was stirred at rt under 1 atm of hydrogen for 22 h.
Filtration of the reaction mixture and washing of the Pd/C with
methanol (10 mL × 4) followed by solvent evaporation gave a
colorless residue, which was washed with chloroform. After
drying under vacuum, 3.5 mg of product was obtained. 1H-
NMR (CD3OD, δ): 6.59-7.51 (7 H, aromatic), 4.84 (1 H, m,
R-H), 4.40-4.62 (2 H, m, R-H), 3.60-4.23 (5 H, m, R-H), 2.78-
3.20 (4 H, m, Asp-CH2, Trp-CH2), 2.44 (1 H, d, (CH3)CCH2),
2.36 (3 H, s, CH3), 2.18 (3 H, s, CH3), 2.04 (1 H, d, (CH3)CCH2),
1.10-1.70 (12 H, m, CH3). FABMS: 784 (M + Na), 762 (M +
1). HRMS: calcd for C38H47N7O10 (M + 1) 762.3463, found
762.3481.
Boc-Asp (OBn )-OTce. Boc-Asp(OBn)-OH (1 g, 3 mmol),
2,2,2-trichloroethanol (0.4 mL, 3 mmol), and DMAP (0.18 g,
1.5 mmol) were dissolved in DCM (30 mL) and cooled to 0 °C.
To this cooled solution was added EDC (0.57 g, 3 mmol), and
the reaction mixture was stirred at 0 °C for 3 h and then at
ambient temperature for 21 h. The precipitate was filtered
out, and the filtrate was diluted with EtOAc (100 mL). The
EtOAc layer was successively washed with saturated NaHCO3
(2 × 20 mL), H2O (2 × 50 mL), and saturated aqueous NaCl
(20 mL). The EtOAc layer was dried over anhydrous Na2SO4
and concentrated under reduced pressure to yield Boc-Asp-
(OBn)-OTce (1.13 g, 83%) as a yellow oil. 1H-NMR (CDCl3,
δ): 1.45 (9H, s), 2.93 and 3.15 (2H, dd, J ) 17, 4.5 Hz), 4.67
and 4.75 (2H, dd, J ) 12.3 Hz), 4.7-4.75 (1H, m), 5.13 (2H,
s), 5.56 (1H, d, J ) 9 Hz), 7.34-7.38 (5H, m). MS (FAB) m/ z:
454 (M+ + 1). Anal. Calcd for C18H22NO6Cl3: C, 47.68; H,
4.86; N, 3.09. Found: C, 48.20; H, 4.70; N, 3.20. HMRS: calcd
for C18H22NO6Cl3 454.0591, found 454.0594.
H-Asp (OBn )-OTce. TFA (5 mL) was added to a stirred
solution of Boc-Asp(OBn)-OTce (1 g, 2.2 mmol) in DCM (5 mL).
The reaction mixture was stirred at room temperature for 45
min. Volatile compounds in the reaction mixture were re-
moved using a rotary evaporator under vacuum. The residue
was triturated and washed with anhydrous Et2O, and the solid
was isolated by decantation. The solid H-Asp(OBn)-OTce (0.78
g, 100%) was dried under vacuum to remove the residual Et2O
and was used in the next step without further purification.
FABMS: calcd for C13H14Cl3NO4 353, found 354 (M+ + 1).
Boc-Ala -Gly-Gly-Asp (OBn )-OTce. To a cooled (0 °C)
stirred solution of Boc-Ala-Gly-Gly-OH (1.6 g, 5.13 mmol), Asp-
(OBn)-OTce (2.4 g, 5.13 mmol), HOBt (0.69 g, 5.13 mmol), and
NMM (0.5 mL, 5.13 mmol) in DCM (100 mL) was added EDC
(0.99 g, 5.16 mmol) in one portion. The reaction mixture was
stirred at 0 °C for 4 h and at ambient temperature for 24 h.

Synthesis of an Esterase-Sensitive Cyclic Prodrug System
J . Org. Chem., Vol. 62, No. 5, 1997 1367
The reaction mixture was diluted with DCM (250 mL) and
washed successively with 10% aqueous citric acid (2 × 50 mL),
H2O (100 mL), saturated NaHCO3 (2 × 50 mL), H2O (100 mL),
and saturated aqueous NaCl (50 mL). The organic layer was
dried over anhydrous Na2SO4 and concentrated under reduced
pressure to furnish Boc-Ala-Gly-Gly-Asp(OBn)-OTce (2.8 g,
86%) as pale yellow oil. 1H-NMR (DMSO-d6, δ): 1.17 (3H, d,
J ) 7.2 Hz), 1.37 (9H, s), 2.85 and 2.97 (2H, dd, J ) 17, 6.2
Hz), 3.71-3.79 (3H, br), 3.98 (2H, br), 4.86 (2H, d, J ) 4.3
Hz), 5.12 (2H, s), 7.01 (1H, d, J ) 6.5 Hz), 7.33-7.37 (5H, br),
8.07 (2H, d, J ) 6 Hz), 8.54 (1H, d, J ) 7.8 Hz). FABMS:
under vacuum to remove the residual Et2O and was used in
the next step without further purification. FABMS: calcd for
C31H35N6O8Cl3 724, found 725 (M+ + 1).
Boc-Ala -(P r om )-Tr p -Ala -Gly-Gly-Asp (OBn )-OTce (11b).
EDC (0.09 g, 0.44 mmol) was added into a cooled (0 °C) and
stirred solution of Boc-Ala-(Prom)-OH (0.17 g, 0.44 mmol),
H-Trp-Ala-Gly-Gly-Asp(OBn)-OTce (0.32 g, 0.44 mmol), HOBt
(0.6 g, 0.44 mmol), and NMM (0.09 mL, 0.89 mmol) in DCM
(50 mL). The mixture was stirred for 2 h at 0 °C and for 30 h
at ambient temperature. The reaction mixture was diluted
with DCM (250 mL) and washed successively with 10%
aqueous citric acid (2 × 50 mL), H2O (100 mL), saturated
NaHCO3 (2 × 50 mL), H2O (100 mL), and saturated aqueous
NaCl (50 mL). The organic layer was dried over anhydrous
Na2SO4 and concentrated under reduced pressure to furnish
an oil of Boc-Ala-(Prom)-Trp-Ala-Gly-Gly-Asp(OBn)-OTce (0.33
g, 68%). The peptide was analyzed by analytical reversed-
phase HPLC using a C-18 column (5 µm, 300 Å, 25 cm × 4.6
mm, flow rate 1 mL/min) eluting with a gradient starting from
20% of solvent A (0.1% TFA/H2O:5%ACN v/v) to 100% of
solvent B (ACN) in 20 min. The retention time of the peptide
is 16.85 min. 1H-NMR (CDCl3, δ): 0.99 (3H, d, J ) 6.7 Hz),
1.38 (9H, s), 1.49-1.72 (9H, m), 2.11 (3H, s), 2.43-2.71 (2H,
m), 2.98-3.17 (4H, m), 3.71-3.82 (1H, m), 3.92-4.02 (4H, m),
4.11-4.18 (1H, m), 4.33 (1H, t, J ) 6.8 Hz), 4.53-4.71 (3H,
m), 5.13 (2H, s), 6.39 (2H, s), 6.71 (2H, s), 6.95-7.53 (9H, m),
8.32 (1H, s). FABMS: calcd for C52H64N7O13Cl3 1099, found
1100 (M + 1).
calcd for C25H33N4O9Cl3 638, found 645 (M+ + Li), 661 (M+
+
Na).
H -Ala -Gly-Gly-Asp (OBn )-OTce. Boc-Ala-Gly-Gly-Asp-
(OBn)-OTce (0.29 g, 0.44 mmol) was dissolved in DCM (5 mL)
and the solution was cooled to 0 °C. To this clear solution was
added TFA (5 mL), and the reaction mixture was stirred at
room temperature for 1 h. Volatile compounds in the reaction
mixture were removed using a rotary evaporator under
vacuum. The residue was triturated and washed with anhy-
drous Et2O, and the solid was isolated by decantation. H-Ala-
Gly-Gly-Asp(OBn)-OTce (0.27 g, 95%) was dried under vacuum
to remove the residual Et2O and was used in the next step
without further purification. 1H-NMR (DMSO-d6, δ): 1.36
(3H, d, J ) 7.2 Hz), 2.85 and 2.98 (2H, dd, J ) 17.0, 6.0 Hz),
3.70-4.01 (6H, m), 4.86 (2H, d, J ) 4.1 Hz), 5.12 (2H, s), 7.36-
7.37 (5H, br), 8.09 (2H, br), 8.24 (1H, br), 8.64 (1H, d, J ) 7.2
Hz). FABMS: calcd for C20H25N4O7Cl3 538, found 538 (M+).
Boc-Tr p -Ala -Gly-Gly-Asp (OBn )-OTce (14). To a cooled
(0 °C) stirred solution of H-Ala-Gly-Gly-Asp(OBn)-OTce (2.2
g, 5.13 mmol), HOBt (0.69 g, 5.13 mmol), Boc-Trp-OH (1.56g,
5.13 mmol), and NMM (0.5 mL, 5.13 mmol) in DCM (100 mL)
was added EDC (0.99 g, 5.16 mmol) in one portion. The
reaction mixture was stirred at 0 °C for 4 h and at ambient
temperature for 24 h. The reaction mixture was diluted with
DCM (250 mL) and washed successively with 10% aqueous
citric acid (2 × 50 mL), H2O (100 mL), saturated NaHCO3 (2
× 50 mL), H2O (100 mL), and saturated aqueous NaCl (50 mL).
The organic layer was dried over anhydrous Na2SO4 and
concentrated under reduced pressure to furnish Boc-Trp-Ala-
Gly-Gly-Asp(OBn)-OTce (3.6 g, 86%) as a pale yellow oil. 1H-
NMR (CDCl3, δ): 1.21 (3H, d, J ) 7.2 Hz), 1.43 (9H, s), 2.99
and 3.17 (2H, m), 3.23 (2H, d, J ) 6.8 Hz), 3.52-3.59 (1H, m),
3.73-3.89 (1H, m), 3.97 (2H, d, J ) 5.9 Hz), 4.23-4.27 (1H,
m), 4.41-4.48 (1H, m) 4.75 (1H, d, J ) 11.8 Hz), 4.41 (1H, d,
J ) 11.8 Hz), 5.02-5.09 (1H, m), 5.14 (2H, s), 6.25 (1H, m),
6.49 (1H, m), 7.04-7.40 (9H, m), 7.64 (1H, d, J ) 7.4 Hz). Anal.
Calcd for C36H43N6O10Cl3: C, 52.43; H, 5.22; N, 10.19.
Found: C, 52.34; H, 5.05; N, 9.94.
H-Ala -(P r om )-Tr p -Ala -Gly-Gly-Asp (OBn )-OH (12). To
a
stirred solution of Boc-Ala-(Prom)-Trp-Ala-Gly-Gly-Asp-
(OBn)-OTce (0.22 g, 0.2 mmol) in HOAc (50 mL) was added
Zn dust (1 g) over 1 h. After the reaction mixture was stirred
for 24 h at room temperature, the insoluble material was
filtered out, and the filtrate was concentrated under reduced
pressure to give an oily residue. The resulting residue was
dissolved in DCM (10 mL), and the solution was cooled to 0
°C. To this clear solution were added TFA (5 mL), phenol (2
g), and ethanedithiol (0.2 mL). After the mixture was stirred
at room temperature for 2 h, DCM was evaporated and the
residue was triturated with anhydrous Et2O. The solid was
washed with anhydrous Et2O, isolated by decantation, and
dried under reduced pressure to afford H-Ala-(Prom)-Trp-Ala-
Gly-Gly-Asp(OBn)-OH (0.09 g, 50%). The product was purified
further by preparative reversed-phase HPLC using a C-18
column (12 µm, 300 Å, 25 cm × 21.4 mm, flow rate 5 mL/min)
eluting with a gradient starting from 30% of solvent A (0.1%
TFA/H2O:5%ACN v/v) to 100% of solvent B (ACN) in 70 min.
The peptide was analyzed by analytical reversed-phase HPLC
using a C-18 column (5 µm, 300 Å, 25 cm × 4.6 mm, flow rate
1 mL/min) eluting with a gradient starting from 20% of solvent
A (0.1% TFA/H2O:5%ACN v/v) to 100% of solvent B (ACN) in
20 min. The retention time of the peptide is 14.97 min.
H-Tr p -Ala -Gly-Gly-Asp (OBn )-OTce (15). Boc-Trp-Ala-
Gly-Gly-Asp(OBn)-OTce (0.29 g, 0.44 mmol) was dissolved in
DCM (5 mL), and the solution was cooled to 0 °C. To this clear
solution was added TFA (5 mL), and the reaction mixture was
stirred at room temperature for 1 h. Volatile compounds in
the reaction mixture were removed using a rotary evaporator
under vacuum. The residue was triturated and washed with
anhydrous Et2O, and the solid was isolated by decantation.
H-Trp-Ala-Gly-Gly-Asp(OBn)-OTce (0.3 g, 95%) was dried
Ack n ow led gm en t. Financial support was provided
by Glaxo, Inc., The United States Public Health Service
(DA09315), and the Swiss National Science Foundation.
J O961778Z