Simplified pepstatins: Synthesis and evaluation of N-terminally modified analogues

Article abstract of DOI:10.1021/jm9807306

The promising strategy of gastric ulcer healing with perorally administered epidermal growth factor (EGF) is so far strongly limited by the pepsinic degradation of this therapeutic polypeptide in the stomach. The incorporation of EGF in a bioadhesive polymer-pepsin inhibitor conjugate used as drug carrier matrix, however, might provide sufficient protection toward pepsinic degradation. The synthesis of appropriate pepsin inhibitors represents a prerequisite for the development of such polymer-inhibitor conjugates. The presented study demonstrates that modifications at the N- terminus of simplified analogues of pepstatin which can be synthesized in a simple and straight way result only in slight variations of the inhibitory activity. These analogues display only 10-fold reduced inhibitory activity, compared to pepstatin A, when bearing a greater N-terminal group like isovaleryl, Boc, or Cbz. Compounds which are substituted at the N-terminus by a shorter N-acyl group like propionyl or cyclopropylcarbonyl show further reduced activity (0.01, compared to pepstatin A). The presence of an amide or a urethane moiety at the N-terminus has no considerable effect on enzyme inhibition. Therefore, the N-terminus of these analogues is able to be modified forming a covalent bond to various bioadhesive polymers via a suitable functionality.

Full text of DOI:10.1021/jm9807306

J . Med. Chem. 1999, 42, 2041-2045
2041
Sim p lified P ep sta tin s: Syn th esis a n d Eva lu a tion of N-Ter m in a lly Mod ified
An a logu es
Martin Kratzel,*,† Birgit Schlichtner,† Roland Kirchmayer,‡ and Andreas Bernkop-Schnu¨rch‡
Institute of Pharmaceutical Chemistry and Institute of Pharmaceutical Technology, Center of Pharmacy, Althanstrasse 14,
A-1090 Vienna, Austria
Received December 28, 1998
The promising strategy of gastric ulcer healing with perorally administered epidermal growth
factor (EGF) is so far strongly limited by the pepsinic degradation of this therapeutic polypeptide
in the stomach. The incorporation of EGF in a bioadhesive polymer-pepsin inhibitor conjugate
used as drug carrier matrix, however, might provide sufficient protection toward pepsinic
degradation. The synthesis of appropriate pepsin inhibitors represents a prerequisite for the
development of such polymer-inhibitor conjugates. The presented study demonstrates that
modifications at the N-terminus of simplified analogues of pepstatin which can be synthesized
in a simple and straight way result only in slight variations of the inhibitory activity. These
analogues display only 10-fold reduced inhibitory activity, compared to pepstatin A, when
bearing a greater N-terminal group like isovaleryl, Boc, or Cbz. Compounds which are
substituted at the N-terminus by a shorter N-acyl group like propionyl or cyclopropylcarbonyl
show further reduced activity (0.01, compared to pepstatin A). The presence of an amide or a
urethane moiety at the N-terminus has no considerable effect on enzyme inhibition. Therefore,
the N-terminus of these analogues is able to be modified forming a covalent bond to various
bioadhesive polymers via a suitable functionality.
In tr od u ction
drastically improved by an appropriate galenic. An
initial step in this direction can be seen in the develop-
ment of a bioadhesive drug delivery system containing
a polymer-enzyme inhibitor conjugate, which provides
a strong protective effect for the embedded therapeutic
agent.4 The efficient pepsin inhibitor pepstatin A (1) was
selected. However, although this system turned out to
be very successful in various studies,4 its practical use
is strongly limited by the extensive costs of pepstatin
A. Following studies have been focused on the synthesis
of simplified pepstatin A analogues which guarantee low
production costs in large-scale preparation and form the
basis for the development of practically usable polymer-
pepsin inhibitor conjugates as vehicles for the peroral
administration of EGF.5
The peroral administration of therapeutic peptides
and proteins is a great challenge to pharmaceutical
sciences. So far, however, this route is of limited value
because the gastrointestinal tract provides a hostile
environment for (poly)peptides. In particular, luminally
secreted and brush border membrane-bound enzymes
lead to a rapid degradation of perorally administered
peptide and protein drugs. A promising strategy to
overcome this so-called ‘enzymatic barrier’ represents
the use of auxiliary agents such as mucoadhesive
polymers and enzyme inhibitors.1 On the one hand,
these polymers can provide intimate contact with the
gastrointestinal mucosa, thereby excluding a presys-
temic metabolism of the therapeutic (poly)peptide on the
way between the delivery system and the absorption
membrane. Enzyme inhibitors, on the other hand, are
able to inactivate proteases which penetrate into the
polymeric carrier system. In addition, toxic side effects
of enzyme inhibitors can be excluded by the covalent
attachment of these auxiliary agents to the unabsorb-
able polymeric carrier matrix.
Among such polymer-inhibitor conjugates, the de-
velopment of mucoadhesive polymer-pepsin inhibitor
conjugates has gained considerable interest for the
peroral administration of epidermal growth factor (EGF).
Due to the oral administration of this polypeptide drug,
an enhanced healing of gastric ulcers could already by
verified in a double-blind controlled clinical study.2 As
EGF is rapidly degraded in the stomach by pepsin,3 its
ulcer-healing capability after oral dosing could be even
Ch em istr y
McConnell et al. have described tripeptides (2) acting
as pepsin inhibitors, which contain only the first three
N-terminal amino residues of pepstatin A (1).6 Recently,
we have reported on comparably simplified analogues
(3), which utilize a vic-amino alcohol in place of the
C-terminal statine unit.5 These compounds show com-
parable activities to pepstatin A itself. Due to synthetic
considerations our compounds 3 have a Boc group at
the N-terminus, while the analogues 2, described by
McConnell et al., are N-terminated by a Cbz unit.6 It
* To whom correspondence should be addressed.
† Institute of Pharmaceutical Chemistry.
‡ Institute of Pharmaceutical Technology.
10.1021/jm9807306 CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/06/1999

2042 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Brief Articles
Sch em e 1a
a
(a) RCOCl, MgO, Et2O/H2O; (b) H2N-Val-OBn, DCC, THF; (c)
Pd/C/H2, MeOH; (d) EEDQ, THF.
Ta ble 1. Inhibitory Effect of Pepstatin A and Analogues
toward the Pepsin Hydrolysis of Horseradish Peroxidase
seems to be that the protective group at the N-terminus
has no considerable effect on enzyme inhibition. In the
following, our aim was to produce analogues of our most
active compounds with different groups at the N-
terminus to elucidate structure-activity dependences.
Obviously, the shortest way to synthesize compounds
of the general structure 4 is the synthesis of 3 (R1
)
isobutyl, R2 ) hexyl), cleavage of the Boc protecting
graded peroxidase was quantified by the capability of
the enzyme to oxidize o-phenylenediamine leading to
an orange-colored stain. The concentration of added
inhibitors, at which peroxidase displayed half of its
activity after incubation with pepsin, represented there-
fore the IC50. Samples without pepstatin analogues and
with neither pepstatin analogues nor pepsin served as
references.
group, and finally N-acylation with various acid halides
or chloroformates, respectively, though there is the risk
of ester formation at the hydroxy group. To avoid
additional steps caused by protection and deprotection
of the hydroxy group, we decided to construct the target
compounds starting from the N-terminus. The selected
strategy is outlined in Scheme 1. L-Valine (5) was
N-acylated with different acid chlorides (and chlorofor-
mates, respectively) (f6) and reacted with L-valine
benzyl ester in the presence of N,N′-dicyclohexylcarbo-
diimide. Hydrogenolytic cleavage of the benzyl group
(f8) and coupling with amino alcohol 9 afforded the
target compounds 4. The C-terminus of the described
compounds, represented by 9, has been derived from
L-leucine, subsequent reaction (of the Weinreb amide)
with hexylmagnesium bromide, and reduction with
sodium borohydride.5 For the synthesis of 4b we directly
used the commercially available dipeptidic acid Cbz-Val-
Val-OH (8b).
Resu lts a n d Discu ssion
The tested compounds possess a uniform backbone of
two valine residues and a vic-aminol C-terminus; all
analogues exhibit a strong inhibitory effect toward
pepsin. Results of inhibition studies are listed in Table
1. Compared to the compound with the original N-
isovaleryl moiety (of pepstatin A), derivatives with a
shorter group at the nitrogen show a lower inhibiting
activity. These smaller and similar groups comprised
ethyl, isopropyl, and cyclopropyl. Elongation of the
N-substituent does not have any adverse effect on the
activity. Replacement of the N-terminating acyl group
by a urethane moiety does not generally lead to more
active compounds (see 4b vs 4e and 4c vs 4f) as might
have been assumed by comparing the N-Boc derivative
4a with the carba analogue 4d . In the urethane series,
the N-Cbz-terminated compound is surpassed by the
N-Boc-protected analogue.
Eva lu a tion of th e In h ibitor y Activity
To evaluate the enzyme inhibitory activity, the IC50
of all compounds has been determined by an enzyme
assay as previously described by our research group.5
Horseradish peroxidase was thereby used as protein
substrate for pepsin. The remaining activity of unde-

Brief Articles
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2043
(s, 5H, arom. H); 13C NMR (CDCl3) δ 17.6, 18.8, 19.1, 19.2,
21.9 [3 × CH(CH3)2], 30.9, 31.2 [2 × CH(CH3)2], 50.1, 57.0 (2
× R-C), 67.4 (benzyl-C), 68.3 [OCH(CH3)2], 128.2, 128.4 (arom.
CH), 135.2 (arom. C), 156.6, 171.4, 171.6 (3 × CdO); MS m/z
392.3 (M+).
In summary, compared to the simplified pepstatin A
analogues, recently described by Kratzel et al.,5 deriva-
tives with other groups at the N-terminus exhibit only
slightly varying activities. Based on the results obtained
within this study, the following structure-activity de-
pendences could be concluded: (i) A substituent smaller
than C3 leads to a significant reduction of inhibitory
activity. (ii) Compounds with N-substituents such as
isovaleryl or longer carbon chains show similar activi-
ties. (iii) Amides and urethanes at the N-terminus
display comparable activities.
Taking these results into consideration, a more gen-
erous variation of the N-substituent should be possible,
e.g., incorporation of a terminating carboxy or hydroxy
group. Hence, the prerequisites are given for the cova-
lent attachment to a bioadhesive polymer also at the
N-terminus.
N-(3,3-Dim eth ylbu tyr yl)-L-va lyl-L-va lin e Ben zyl Ester
(7d ). Starting from 1.00 g (5 mmol) of N-(3,3-dimethylbutyryl)-
L-valine (6d ), 1.27 g (63%) of 7d was obtained as colorless
crystals: mp 160 °C; IR (KBr, cm-1) 3290 (NH), 1735, 1690,
1
1655 (CdO); H NMR (CDCl3) δ 0.83-0.96 [m, 12H, 2 × CH-
(CH3)2], 1.05 [s, 9H, C(CH3)3], 1.85-2.10 [m, 2H, 2 × CH(CH3)2],
2.28 [s, 2H, COCH2C(CH3)3], 4.55 (dd, 1H, J ) 4.6 Hz, J ) 8.6
Hz, R-H), 4.63 (dd, 1H, J ) 5.0 Hz, J ) 8.8 Hz, R-H), 5.11,
5.21 (AB-system, 2H, J ) 12.4 Hz, benzyl-H), 5.90, 6.14 (each
br.d, each 1H, 2 × NH), 7.35 (s, 5H, arom. H); 13C NMR (CDCl3)
δ 17.5, 17.7, 18.9, 19.1 [2 × CH(CH3)2], 29.9 [C(CH3)3], 31.2 [2
× CH(CH3)2], 31.6 [C(CH3)3], 47.2 [COCH2C(CH3)3], 57.2, 58.3
(2 × R-C), 67.0 (benzyl-C), 128.3, 128.4, 128.5 (arom. CH), 135.2
(arom. C), 171.3, 171.4, 172.0 (3 × CdO); MS m/z 404.5 (M+).
N-(3-P h en ylp r op ion yl)-L-va lyl-L-va lin e Ben zyl Ester
(7e). Starting from 1.25 g (5 mmol) of N-(3-phenylpropionyl)-
L-valine (6e), 1.88 g (86%) of 7e was obtained as colorless
crystals: mp 117-120 °C; IR (KBr, cm-1) 3290 (NH), 1735,
Exp er im en ta l Section
Melting points were determined on a Kofler hot plate
apparatus and are uncorrected. Infrared spectra were recorded
1
1655, 1635 (CdO); H NMR (CDCl3) δ 0.76-0.96 [m, 12H, 2
1
on a Perkin-Elmer 1600 FTIR spectrophotometer. H and 13C
× CH(CH3)2], 1.90-2.08, 2.10-2.25 [each m, each 1H, 2 ×
CH(CH3)2)], 2.48-2.63, 2.92-3.03 (each m, each 2H, 2 × CH2),
4.40 (dd, 1H, J ) 4.8 Hz, J ) 8.6 Hz, R-H), 4.55 (dd, 1H, J )
7.2 Hz, J ) 8.8 Hz, R-H), 5.11, 5.22 (AB-system, 2H, J ) 12.0
Hz, benzyl-H), 6.46, 6.83 (each br.d, each 1H, 2 × NH), 7.14-
7.28 (m, 5H, arom. H), 7.35 (s, 5H, arom. H); 13C NMR (CDCl3)
δ 17.6, 18.1, 18.8, 18.9 [2 × CH(CH3)2], 30.8, 30.9 [2 × CH-
(CH3)2], 31.1, 38.1 (2 × CH2), 57.2, 58.3 (2 × R-C), 67.0 (benzyl-
C), 126.1, 126.3, 128.2, 128.3, 128.4, 128.5 (arom. CH), 135.2,
140.6 (arom. C), 171.4, 171.5, 172.1 (3 × CdO); MS m/z 438.4
(M+).
NMR spectra were measured on a Varian Unity Plus 300
1
spectrometer. H NMR spectra were referenced to tetrameth-
ylsilane (δ 0.0); in 13C NMR spectroscopy CDCl3 (or MeOD-d3,
respectively) served as the internal standard [δ 77.0 (or δ 49.0,
respectively)]. MS spectra were measured on a Shimadzu QP
5000 instrument. For inhibition studies a microtitration plate
reader (Anthos Reader 2001) was used.
Flash chromatography was performed on Merck silica gel
60, TLC on plastic sheets (Merck silica gel 60 F254). Tetrahy-
drofuran was dried over sodium/benzophenone, dichloromethane
over phosphorus pentoxide.
N-Isova ler yl-L-va lyl-L-va lin e Ben zyl Ester (7f). Starting
from 1.00 g (5 mmol) of N-isovaleryl-L-valine (6f), 1.50 g (77%)
of 7f was obtained as colorless crystals: mp 130-132 °C; IR
(KBr, cm-1) 3285 (NH), 1735, 1655, 1635 (CdO); 1H NMR
(CDCl3 + MeOD-d3) δ 0.76-0.98 [m, 18H, 3 × CH(CH3)2],
1.80-2.24 [m, 4H, 2 × CH(CH3)2, COCH2CH(CH3)2], 4.30, 4.51
(each m, each 1H, 2 × R-H), 5.03, 5.12 (AB-system, 2H, J )
12.2 Hz, benzyl-H), 6.28, 6.63 (each br.d, each 1H, 2 × NH),
7.29 (s, 5H, arom. H); 13C NMR (CDCl3 + MeOD-d3) δ 17.6,
18.0, 18.3, 18.9, 22.3, 25.5 [3 × CH(CH3)2], 27.0 [COCH2CH-
(CH3)2], 31.0, 31.2 [2 × CH(CH3)2], 57.2, 58.3 (2 × R-C), 66.9
(benzyl-C), 128.3, 128.5 (arom. CH), 135.4 (arom. C), 171.3,
171.6, 172.6 (3 × CdO); MS m/z 390.3 (M+).
N-Cyclop r op ylca r bon yl-L-va lin e (6i). To a well-stirred
mixture of L-valine (17 mmol, 2.0 g) and 1.8 g of magnesium
oxide in a solvent mixture of diethyl ether (20 mL) and water
(100 mL) was added cyclopropylcarbonyl chloride (18.7 mmol,
1.70 mL) at 0 °C. After stirring for 4 h at room temperature
the mixture was acidified with HCl (6 N) to pH 2-3. The
organic layer was separated, followed by extraction of the
aqueous layer with ethyl acetate (2 × 20 mL). The combined
organic extracts were washed with brine, dried (Na2SO4), and
evaporated to afford 2.40 g (76%) of 6i as a colorless oil; IR
(KBr/liquid film, cm-1) 3340 (OH), 1720, 1630 (CdO); 1H NMR
(CDCl3) δ 0.65-0.74, 0.75-1.10 [each m, 2H, 8H, 2 × cyclo-
propyl-CH2, CH(CH3)2], 1.35-1.62 (m, 1H, cyclopropyl-CH),
2.09 [m, 1H, CH(CH3)2], 4.51 (dd, 1H, J ) 4.8 Hz, J ) 8.6 Hz,
R-H), 6.61 (br.d, 1H, NH), 9.70 (br, 1H, COOH); 13C NMR
(CDCl3): δ 8.6 (2 × cyclopropyl-CH2), 12.8 (cyclopropyl-CH),
17.6, 18.9 [CH(CH3)2], 31.0 [CH(CH3)2], 57.2 (R-C), 175.1, 176.8
(2 × CdO): MS m/z 185.1 (M+).
N-P r op ion yl-L-va lyl-L-va lin e Ben zyl Ester (7g). Starting
from 870 mg (5 mmol) of N-propionyl-L-valine (6g), 1.43 g
(79%) of 7g was obtained as colorless crystals: mp 180 °C; IR
(KBr, cm-1) 3290 (NH), 1730, 1640 (CdO); 1H NMR (CDCl3) δ
0.80-1.05 [m, 12H, 2 × CH(CH3)2], 1.12 (t, 3H, J ) 7.2 Hz,
CH2CH3), 1.15-1.45, 1.55-1.90 [each m, each 1H, 2 ×
CH(CH3)2)], 2.23 (q, 2H, J ) 7.2 Hz, CH2CH3), 4.45, 4.55 (each
m, each 1H, 2 × R-H), 5.13, 5.23 (AB-system, 2H, J ) 12.0
Hz, benzyl-H), 6.57, 7.02 (each br.d, each 1H, 2 × NH), 7.34
(s, 5H, arom. H); 13C NMR (CDCl3) δ 9.9 (CH2CH3), 17.6, 18.0,
18.9, 19.3 [2 × CH(CH3)2], 29.0 (CH2CH3), 30.9, 31.2 [2 × CH-
(CH3)2], 55.6, 57.4 (2 × R-C), 66.9 (benzyl-C), 128.2, 128.3,
128.5 (arom. CH), 135.3 (arom. C), 171.3, 171.8, 173.9 (3 ×
CdO); MS m/z 362.2 (M+).
N-Isobu tyr yl-L-va lyl-L-va lin e Ben zyl Ester (7h ). Start-
ing from 935 mg (5 mmol) of N-isobutyryl-L-valine (6h ), 1.33
g (71%) of 7h was obtained as colorless crystals: mp 137 °C;
IR (KBr, cm-1) 3285 (NH), 1735, 1640 (CdO); 1H NMR (CDCl3)
δ 0.82-0.98, 1.08-1.20 [each m, 18H, 3 × CH(CH3)2], 1.40-
1.65, 1.70-1.95 [each m, each 1H, 2 × CH(CH3)2], 2.17 [hept,
1H, J ) 6.8 Hz, COCH(CH3)2], 3.96 (dd, 1H, J ) 7.0 Hz, J )
8.5 Hz, R-H), 4.21 (dd, 1H, J ) 5.0 Hz, J ) 8.5 Hz, R-H), 4.84,
4.93 (AB-system, 2H, J ) 12.4 Hz, benzyl-H), 6.84, 7.55 (each
br.d, each 1H, 2 × NH), 7.35 (s, 5H, arom. H); 13C NMR (CDCl3)
δ 17.4, 17.8, 18.0, 18.7, 19.0, 19.6 [3 × CH(CH3)2], 30.9, 31.2,
35.8 [3 × CH(CH3)2], 57.3, 58.0 (2 × R-C), 66.8 (benzyl-C),
Gen er a l P r oced u r e for th e P r ep a r a tion of Dip ep tid es
7. To a solution of N-acylated L-valine (5 mmol) and L-valine
benzyl ester (5 mmol, 1.05 g) in 30 mL of tetrahydrofuran was
added N,N′-dicyclohexylcarbodiimide (5.5 mmol, 1.14 g), dis-
solved in 10 mL of tetrahydrofuran, at 0 °C. The reaction
mixture was stirred overnight, evaporated in vacuo, and
redissolved in ethyl acetate. The solution was washed with
citric acid (10%, 2 × 10 mL), saturated aqueous NaHCO3
solution (2 × 10 mL), and brine (2 × 10 mL). The organic layer
was dried (Na2SO4) and evaporated giving colorless solids.
N-Isop r op oxyca r bon yl-L-va lyl-L-va lin e Ben zyl Ester
(7c). Starting from 1.02 g (5 mmol) of N-isopropoxycarbonyl-
L-valine (6c), 1.41 g (72%) of 7c was obtained as colorless
crystals: mp 107-110 °C; IR (KBr, cm-1) 3285, 3260 (NH),
1735, 1690, 1655 (CdO); 1H NMR (CDCl3) δ 0.80-0.96, 1.14-
1.22 [each m, 18H, 3 × CH(CH3)2], 1.70-2.05, 2.08-2.25 [each
m, each 1H, 2 × CH(CH3)2], 4.08 (dd, 1H, J ) 7.3 Hz, J ) 8.8
Hz, R-H), 4.52 (dd, 1H, J ) 5.0 Hz, J ) 8.8 Hz, R-H), 4.70-
4.92 [m, 1H, OCH(CH3)2], 5.07, 5.17 (AB-system, 2H, J ) 12.2
Hz, benzyl-H), 5.38, 6.83 (each br.d, each 1H, 2 × NH), 7.31

2044 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Brief Articles
128.2, 128.3, 128.4 (arom. CH), 135.2 (arom. C), 171.4, 172.0,
177.8 (3 × CdO); MS m/z 376.2 (M+).
(q, 2H, J ) 7.5 Hz, CH2CH3), 4.12 (dd, 1H, J ) 4.8 Hz, J ) 7.8
Hz, R-H), 4.22 (m, 1H, R-H), 7.83, 8.00 (each br.d, each 1H, 2
× NH); 13C NMR (MeOD-d3) δ 10.5 (CH2CH3), 18.3, 18.8, 19.5,
19.8 [2 × CH(CH3)2], 29.9 (CH2CH3), 31.6, 31.8 [2 × CH(CH3)2],
58.9, 59.0 (2 × R-C), 174.0, 174.6, 177.1 (3 × CdO); MS m/z
272.2 (M+).
N-Isobu tyr yl-L-va lyl-L-va lin e (8h ). Starting from 1.13 g
(3 mmol) of 7h , 805 mg (94%) of 8h was obtained as colorless
crystals: mp 148-151 °C (methanol); IR (KBr, cm-1) 3420
(NH, OH), 1715, 1650, 1635 (CdO); 1H NMR (MeOD-d3): δ
0.95-1.08 [m, 12H, 2 × CH(CH3)2], 1.08-1.23 [m, 6H, COCH-
(CH3)2], 2.05, 2.20 [each m, each 1H, 2 × CH(CH3)2], 2.61 [hept,
1H, J ) 7.4 Hz, COCH(CH3)2], 4.26 (dd, 1H, J ) 3.2 Hz, J )
8.0 Hz, R-H), 4.37 (m, 1H, R-H), 7.92, 8.14 (each br.d, each
1H, 2 × NH); 13C NMR (MeOD-d3) δ 18.3, 18.9, 19.5, 19.6, 19.7,
20.2 [3 × CH(CH3)2], 31.7, 31.8 [2 × CH(CH3)2], 36.0 [COCH-
(CH3)2], 58.8, 59.0 (2 × R-C), 174.0, 174.5, 180.1 (3 × CdO);
MS m/z 286.2 (M+).
N-Cyclop r op ylca r bon yl-L-va lyl-L-va lin e (8i). Starting
from 1.12 g (3 mmol) of 7i, 1.25 g (66%) of 8i was obtained as
colorless crystals: mp 200-203 °C; IR (KBr, cm-1) 3490 (NH,
OH), 1705, 1660, 1640 (CdO); 1H NMR (MeOD-d3) δ 0.76-
0.85 (m, 2H, cyclopropyl-CH2), 0.90-1.08 [m, 14H, 2 × CH-
(CH3)2, cyclopropyl-CH2], 1.68-1.83 (m, 1H, cyclopropyl-CH),
2.02-2.30 [m, 2H, 2 × CH(CH3)2], 4.28 (dd, 1H, J ) 5.1 Hz, J
) 7.8 Hz, R-H), 4.36 (m, 1H, R-H), 8.13, 8.18 (each br.d, each
1H, 2 × NH); 13C NMR (MeOD-d3): δ 7.4, 7.6 (2 × cyclopropyl-
CH2), 14.5 (cyclopropyl-CH), 18.3, 18.8, 19.5, 19.7 [2 × CH-
(CH3)2], 31.7, 32.0 [2 × CH(CH3)2], 58.9, 60.2 (2 × R-C), 174.0,
174.6, 176.5 (3 × CdO); MS m/z 284.2 (M+).
N-Cyclop r op ylca r bon yl-L-va lyl-L-va lin e Ben zyl Ester
(7i). Starting from 925 mg (5 mmol) of N-cyclopropylcarbonyl-
L-valine (6h ), 1.25 g (66%) of 7h was obtained as colorless
crystals: mp 158-160 °C; IR (KBr, cm-1) 3290 (NH), 1735,
1
1635 (CdO); H NMR (CDCl3) δ 0.70-0.82, 0.83-1.00 [each
m, 2H, 10H, 2 × CH(CH3)2, 2 × cyclopropyl-CH2], 1.50-1.65
(m, 1H, cyclopropyl-CH), 1.90-2.30 [m, 2H, 2 × CH(CH3)2],
4.33 (dd, 1H, J ) 4.8 Hz, J ) 8 4 Hz, R-H), 4.57 (t, 1H, J ) 8.8
Hz, R-H), 5.13, 5.22 (AB-system, 2H, J ) 12.2 Hz, benzyl-H),
6.45, 6.60 (each br.d, each 1H, 2 × NH), 7.36 (s, 5H, arom. H);
13C NMR (CDCl3) δ 7.3 (2 × cyclopropyl-CH2), 14.2 (cyclopro-
pyl-CH), 17.5, 18.2, 18.8, 19.1 [2 × CH(CH3)2], 30.9 [2 × CH-
(CH3)2], 56.1, 57.2 (2 × R-C), 67.0 (benzyl-C), 128.3, 128.4,
128.5 (arom. CH), 135.2 (arom. C), 171.4, 173.6 (3 × CdO);
MS m/z 374.2 (M+).
Gen er a l P r oced u r e for th e P r ep a r a tion of Dip ep tid ic
Acid s 8. Dipeptide 7 (3 mmol) was dissolved in 20 mL of
methanol and hydrogenated with Pd/C (80 mg, 10%) as
catalyst overnight. After filtration the solution was concen-
trated in vacuo, redissolved in ethyl acetate, and extracted
with aqueous NaOH solution (0.5 N, 2 × 10 mL). The combined
aqueous layers were washed with ethyl acetate (1 × 10 mL),
acidified with HCl (2 N) to pH 2, and extracted with ethyl
acetate (2 × 10 mL). The solution was dried (Na2SO4) and
evaporated under reduced pressure yielding dipeptidic acids
8 as colorless oils or solids, respectively.
N-Isop r op oxyca r bon yl-L-va lyl-L-va lin e (8c). Starting
from 1.18 g (3 mmol) of 7c, 830 mg (92%) of 8c was obtained
as colorless crystals: mp 150 °C (methanol); IR (KBr, cm-1):
Gen er a l P r oced u r e for Am id e F or m a tion Yield in g
Com p ou n d s 4. A mixture of amino alcohol 9 (1 mmol, 200
mg) (diastereoisomeric ratio: 4S,5S:4S,5R ) 2.5:1)5 and the
selected dipeptidic acid 8 (1 mmol) in a solvent mixture of
tetrahydrofuran and dichloromethane was treated with EEDQ
(2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline) (1.2 mmol,
300 mg) and stirred at room temperature for 24 h. Then the
solution was evaporated, redissolved in ethyl acetate, and
washed with HCl (1 N, 2 × 10 mL), saturated aqueous
NaHCO3 solution (2 × 20 mL), and brine (2 × 20 mL). The
organic phase was dried (Na2SO4) and evaporated. Flash
chromatography of the residue (ethyl acetate-hexane, 1:1)
afforded the pure products (Rf: 0.45-0.76).
1
3315 (NH, OH), 1720, 1690, 1640 (CdO); H NMR (CDCl3) δ
0.95-1.02, 1.23-1.28 [each m, 18H, 3 × CH(CH3)2], 2.05-2.20,
2.22-2.35 [each m, each 1H, 2 × CH(CH3)2], 4.06, 4.59 [each
dd, each 1H, J ) 4.6 Hz, J ) 8.6 Hz, 2 × R-H], 4.82-5.00 [m,
1H, OCH(CH3)2], 5.45, 6.80 (each br.d, each 1H, 2 × NH), 10.80
(br, 1H, COOH); 13C NMR (CDCl3) δ 18.0, 18.4, 19.3, 19.6, 21.8,
22.4 [3 × CH(CH3)2], 31.1, 31.4 [2 × CH(CH3)2], 50.5, 57.8 (2
× R-C), 69.4 [OCH(CH3)2], 172.2, 172.4 (3 × CdO): MS m/z
302.2 (M+).
N-(3,3-Dim eth ylbu tyr yl)-L-va lyl-L-va lin e (8d ). Starting
from 1.17 g (3 mmol) of 7d , 580 mg (64%) of 8d was obtained
as colorless crystals: mp 157-160 °C (methanol); IR (KBr,
cm-1) 3360 (NH, OH), 1710, 1660, 1620 (CdO); 1H NMR
(CDCl3) δ 0.92-1.08 (m, 21H, 7 × CH3), 2.08-2.35 [m, 4H,
COCH2C(CH3)3, 2 × CH(CH3)2], 4.40, 4.66 (each m, each 1H,
R-H), 6.19, 7.36 (each br.d, each 1H, 2 × NH), 10.87 (br, 1H,
COOH); 13C NMR (CDCl3) δ 17.5, 17.7, 19.1, 19.3 [2 × CH-
(CH3)2], 29.8 [C(CH3)3], 30.9 [C(CH3)3], 31.0 [2 × CH(CH3)2],
50.5 [COCH2C(CH3)3], 57.0, 59.0 (2 × R-C), 172.4, 174.1, 175.0
(3 × CdO); MS m/z 314.2 (M+).
4-(N-Ben zyloxyca r bon yla m in o-L-va lyl-L-va lyla m in o)-
2-m eth ylu n d eca n -5-ol (4b). Starting from 350 mg (1 mmol)
of Cbz-L-Val-L-Val-OH (8b), 395 mg (74%) of 4b was obtained
as a colorless oil: IR (KBr/liquid film, cm-1) 3310 (NH, OH),
1
1715, 1650 (CdO); H NMR (CDCl3) δ 0.70-0.95 (m, 21H, 7
× CH3), 1.05-1.60 [m, 13H, chain-CH2, CH2CH(CH3)2], 1.98,
2.10 [each m, each 1H, 2 × CH(CH3)2], 3.60, 3.95, 4.08, 4.15
(each m, each 1H, 2 × R-H, 4-H, 5-H), 5.05 (s, 2H, benzyl-H),
5.80, 7.03 (each br.d, each 1H, 2 × NH), 7.13 (s, 5H, arom. H);
MS m/z 533.4 (M+).
N-(3-P h en ylp r op ion yl)-L-va lyl-L-va lin e (8e). Starting
from 1.31 g (3 mmol) of 7e, 950 mg (91%) of 8e was obtained
as colorless crystals: mp 149-152 °C (methanol); IR (KBr,
cm-1) 3295 (NH, OH), 1710, 1665, 1630 (CdO); 1H NMR
(MeOD-d3): δ 0.72-0.90 [m, 12H, 2 × CH(CH3)2], 1.80-1.95,
1.97-2.15 [each m, each 1H, 2 × CH(CH3)2], 2.82 (t, 2H, J )
7.4 Hz, CH2), 2.40-2.55 (m, 2H, CH2), 4.09 (dd, 1H, J ) 5.5
Hz, J ) 8.6 Hz, R-H), 4.21 (dd, 1H, J ) 4.6 Hz, J ) 7.8 Hz,
R-H), 7.06-7.14 (m, 5H, arom. H), 7.87, 7.94 (each br.d, each
1H, J ) 8.4 Hz, 2 × NH); 13C NMR (MeOD-d3) δ 18.3, 18.8,
19.5, 19.6 [2 × CH(CH3)2], 31.5, 31.7 [2 × CH(CH3)2], 32.8,
38.5 (2 × CH2), 58.8, 60.1 (2 × R-C), 66.8 (benzyl-C), 127.1,
129.3 (arom. CH), 141.9 (arom. C), 173.8, 174.5, 175.0 (3 ×
CdO); MS m/z 348.3 (M+).
N-Isova ler yl-L-va lyl-L-va lin e (8f). Starting from 1.17 g
(3 mmol) of 7f, 810 mg (90%) of 8f was obtained as colorless
crystals. Physical and spectroscopic data are described in ref
7.
N-P r op ion yl-L-va lyl-L-va lin e (8g). Starting from 1.09 g
(3 mmol) of 7g, 750 mg (91%) of 8g was obtained as a colorless
oil: IR (KBr, cm-1) 3295 (NH, OH), 1735, 1635 (CdO); 1H NMR
(MeOD-d3) δ 0.80-0.95 [m, 12H, 2 × CH(CH3)2], 1.03 (t, 3H,
J ) 7.5 Hz, CH2CH3), 1.80-2.05 [m, 2H, 2 × CH(CH3)2], 2.18
4-(N-Isop r op oxyca r bon yla m in o-L-va lyl-L-va lyla m in o)-
2-m eth ylu n d eca n -5-ol (4c). Starting from 300 mg (1 mmol)
of 8c, 315 mg (65%) of 4c were obtained as a colorless oil: IR
(KBr/liquid film, cm-1) 3305 (NH, OH), 1700, 1650 (CdO); MS
m/z 485.4 (M+).
4-[N -(3,3-Dim e t h ylb u t yr yl)-L-va lyl-L-va lyla m in o]-2-
m eth ylu n d eca n -5-ol (4d ). Starting from 315 mg (1 mmol)
of 8d , 310 mg (62%) of 4d was obtained as a colorless oil: IR
(KBr/liquid film, cm-1) 3295 (NH, OH), 1725, 1645 (CdO); 1H
NMR (CDCl3) δ 0.70-0.90 (m, 21H, 7 × CH3), 1.05-1.70 [m,
15H, chain-CH2, CH2CH(CH3)2], 1.25 [s, 9H, C(CH3)3], 2.00-
2.20 [m, 2H, 2 × CH(CH3)2], 2.80 (br, 1H, OH), 3.58, 3.80, 3.95,
4.08 (each m, each 1H, 2 × R-H, 4-H, 5-H), 5.06, 6.28 (each
br.d, each 1H, 2 × NH); MS m/z 497.4 (M+).
4-[N -(3-P h e n ylp r op ion yl)-L -va lyl-L -va lyla m in o]-2-
m eth ylu n d eca n -5-ol (4e). Starting from 350 mg (1 mmol)
of 8e, 355 mg (67%) of 4e was obtained as a colorless oil: IR
(KBr/liquid film, cm-1) 3290 (NH, OH), 1715, 1640 (CdO); 1H
NMR (CDCl3) δ 0.70-1.00 (m, 21H, 7 × CH3), 1.05-1.65 [m,
13H, chain-CH2, CH2CH(CH3)2], 2.05, 2.20 [each m, each 1H,
2 × CH(CH3)2], 2.45-2.58, 2.80-2.95 [each m, each 2H,

Brief Articles
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2045
PhCH2CH2CO], 3.60, 3.90, 4.10, 4.28 (each m, each 1H, 2 ×
R-H, 4-H, 5-H), 6.25, 6.80 (each br.d, each 1H, 2 × NH), 7.05-
7.12 (m, 5H, arom. H); MS m/z 531.4 (M+).
µg of horseradish peroxidase (Sigma, St. Louis, MO) in 75 µL
of 0.05 N HCl was transferred to each well, and the reaction
mixtures were incubated for 2 h at 37 °C. Thereafter, samples
were diluted in 1:2 (v:v) steps with 0.05 N HCl in the following
11 horizontal wells. To each 100 µL of diluted reaction mixture,
100 µL of the substrate medium containing 24 mg of o-
phenylenediamine dihydrochloride, 12 mL of 0.2 M phosphate
buffer (pH 6.5), and 24 µL of 30% H2O2 was added, and the
enzymatic reaction was allowed to proceed at room tempera-
ture for 15 min. Optical density was read at 492 nm with a
microtitration plate reader. Reaction mixtures containing
increasing pepstatin A (Sigma, St. Louis, MO) concentrations
instead of test compounds were used as references. For
negative and positive controls, the enzyme assay was carried
out as described above, but omitting test compounds or test
compounds as well as pepsin, respectively.
4-(N-Isovaler yl-L-valyl-L-valylam in o)-2-m eth ylu n decan -
5-ol (4f). Starting from 300 mg (1 mmol) of 8f, 305 mg (63%)
of 4f was obtained as a colorless oil: IR (KBr/liquid film, cm-1
)
3290 (NH, OH), 1720, 1640 (CdO); 1H NMR (CDCl3 + MeOD-
d3): δ 0.78-1.05 (m, 21H, 7 × CH3), 1.10-1.75 [m, 13H, chain-
CH2, CH2CH(CH3)2], 2.00, 2.15 [each m, each 1H, 2 ×
CH(CH3)2], 3.60, 3.85, 4.05, 4.40 (each m, each 1H, 2 × R-H,
4-H, 5-H), 6.80, 7.00 (each br.d, each 1H, 2 × NH); MS m/z
483.4 (M+).
4-(N-P r opion yl-L-valyl-L-valylam in o)-2-m eth ylu n decan -
5-ol (4g). Starting from 270 mg (1 mmol) of 8g, 305 mg (61%)
of 4g was obtained as a colorless oil: IR (KBr/liquid film, cm-1
3290 (NH, OH), 1735, 1645 (CdO); MS m/z 455.4 (M+).
)
4-(N-Isobu tyr yl-L-valyl-L-valylam in o)-2-m eth ylu n decan -
5-ol (4h ). Starting from 300 mg (1 mmol) of 8h , 305 mg (63%)
Sta tistica l Da ta An a lysis. Statistical data analysis was
performed using the Student t-test with p < 0.05 as the
minimal level of significance.
of 4h was obtained as a colorless oil: IR (KBr/liquid film, cm-1
3290 (NH, OH), 1735, 1640 (CdO); MS m/z 469.4 (M+).
)
Refer en ces
4-(N -Cyclop r op ylca r b on yl-L-va lyl-L-va lyla m in o)-2-
m eth ylu n d eca n -5-ol (4i). Starting from 300 mg (1 mmol) of
8i, 305 mg (63%) of 4i was obtained as a colorless oil: IR (KBr/
liquid film, cm-1) 3290 (NH, OH), 1735, 1640 (CdO); 1H NMR
(CDCl3) δ 0.66-0.80 (m, 2H, cyclopropyl-CH2), 0.80-1.05 (m,
23H, 7 × CH3, cyclopropyl-CH2), 1.15-1.75 (m, 14H, CH2CH-
(CH3)2, CH2-chain, cyclopropyl-CH), 1.95-2.20 [m, 2H, 2 ×
CH(CH3)2], 3.60, 3.95, 4.20, 4.35 (each m, each 1H, 2 × R-H,
4-H, 5-H), 5.80, 6.25 (each br.d, each 1H, 2 × NH); MS m/z
467.4 (M+).
Eva lu a tion of th e In h ibitor y Activity. The enzyme
inhibitory activity of all compounds was determined in a
slightly modified way as described previously by our research
group.5 Each test compound was dissolved in dimethyl sul-
foxide (DMSO) in a final concentration of 1.00 mg/mL and
diluted in 1:10 (v:v) steps with the same solvent; 100 µL of an
appropriate dilution was transferred to the first well of a
microtitration plate (96-well, not binding) and diluted in 1:2
(v:v) steps with DMSO in the following five vertical wells. To
each 50 µL of diluted pepstatin analogue, 75 µL of a fresh
prepared pepsin solution [10 µg of pepsin equivalent to 35 units
(Sigma, St. Louis, MO) per mL of 0.05 N HCl] was added; 60
(1) Bernkop-Schnu¨rch, A. The use of inhibitory agents to overcome
the enzymatic barrier to perorally administered therapeutic
peptides and proteins. J . Controlled Release 1998, 52, 1-16.
(2) Itoh, M.; Matsuo, Y. Gastric ulcer treatment with intravenous
human epidermal growth factor: A double-blind controlled
clinical study. J . Gastroenterol. Hepatol. 1994, 9, 78-83.
(3) Slomiany, B. L.; Nishikawa, H.; Bilski, J .; Slomiany, A. Colloidal
Bismuth Subcitrate Inhibits Peptic Degradation of Gastric
Mucus and Epidermal Growth Factor in Vitro. Am. J . Gastro-
enterol. 1990, 85, 390-393.
(4) Bernkop-Schnu¨rch, A.; Dundalek, K. Novel bioadhesive drug
delivery system protecting (poly)peptides from gastric enzymatic
degradation. Int. J . Pharmacol. 1996, 138, 75-83.
(5) Kratzel, M.; Hiessbo¨ck, R.; Bernkop-Schnu¨rch, A. Auxiliary
Agents for the Peroral Administration of Peptide and Protein
Drugs: Synthesis and Evaluation of Novel Pepstatin Analogues.
J . Med. Chem. 1998, 41, 2339-2344.
(6) McConnell, R. M.; Frizzell, D.; Evans, A. C. A.; J ones, W.; Cagle,
C. New Pepstatin Analogues: Synthesis and Pepsin Inhibition.
J . Med. Chem. 1991, 34, 2298-2300.
(7) Bartlett, P. A.; Hanson, J . E.; Giannousis, P. P. Pepsin Inhibition
of Pepsin and Penicillopepsin by Phosphorus-Containing Peptide
Analogues. J . Org. Chem. 1990, 55, 6268-6274.
J M9807306