Reversed hydroxamate-bearing thermolysin inhibitors mimic a high-energy intermediate along the enzyme-catalyzed proteolytic reaction pathway

Article abstract of DOI:10.1016/S0960-894X(03)00720-0

A series of inhibitors that bear a reversed hydroxamate moiety have been evaluated as transition state analogue inhibitors for thermolysin. A linear correlation is observed between the K_i values of these inhibitors and the kinetic parameters (K

Full text of DOI:10.1016/S0960-894X(03)00720-0

Bioorganic & MedicinalChemistry Letters 13 (2003) 3161–3166
Reversed Hydroxamate-Bearing Thermolysin Inhibitors Mimic a
High-Energy Intermediate Along the Enzyme-Catalyzed
Proteolytic Reaction Pathway
Jung Dae Park and Dong H. Kim*
Center for Integrated Molecular Systems, Division of Molecular and Life Sciences,
Pohang University of Science and Technology, San 31 Hyojadong, Pohang 790-784, South Korea
Received 12 May 2003; accepted 4 July 2003
Abstract—A series of inhibitors that bear a reversed hydroxamate moiety have been evaluated as transition state analogue inhibi-
tors for thermolysin. A linear correlation is observed between the Ki values of these inhibitors and the kinetic parameters (KM/kcat
)
of the parallel series of related substrates, satisfying the criterion stipulated for transition state analogue inhibitors by Bartlett and
Marlowe. Furthermore, examination of the binding mode of a related reversed hydroxamate bearing thermolysin inhibitor, in
comparison with a transition state postulated for the enzyme-catalyzed proteolytic reaction revealed that the inhibitors under study
mimic the electronic as well as the geometric characteristics of the transition state. On the basis of these results it may be concluded
that the hydroxamate-bearing zinc protease inhibitors are a new type of transition state analogue inhibitors.
# 2003 Elsevier Ltd. All rights reserved.
Zinc proteases constitute one of largest families of pro-
teolytic enzymes.1 These enzymes carry a catalytically
essentialzinc ion at their active site. Thermoyl sin (TLN,
EC 3.4.24.4) isolated from Bacillus thermoproteolyticus
is one of most extensively studied zinc proteases. Ther-
molysin cleaves the amino side peptide bond of the
amino acid residue having a hydrophobic side chain.2
The enzyme has been used as a modeltarget enzyme for
developing inhibitor design strategies that can be trans-
ferred to zinc proteases of medicinalinterests, such as
angiotensin converting enzyme3 and matrix metallopro-
teases.4 Most of small molecule inhibitors for the zinc
proteases are designed by incorporating a zinc ligating
group into a molecular scaffold that resembles substrate
so that it can be recognized by the enzyme and forms a
complex with it.5 Carboxylate, thiol and hydroxamate
are commonly used for such zinc-coordinating group.
Of these zinc ligating groups, hydroxamate occupies a
prominent position because of its strong chelating
property toward zinc ion. Another widely employed
design strategy involves construction of a small mol-
ecule whose structure can mimic a transition state in the
enzyme catalyzed reaction, and this approach has its
theoreticalbasis on the transition state theory proposed
for enzymic catalytic reactions.6 According to the the-
ory, enzyme lowers the activation energy in the enzymic
reaction by stabilizing a transition state along the reac-
tion path through stronger interactions relative to the
ground state of substrate.6 It is, therefore, expected that
chemically stable analogues of the transition state
would bind the enzyme more strongly than substrate
and thus inhibit the enzymic activity.7
Although the theoretical background for transition state
analogue inhibition is well founded, the transition state
analogue inhibitors has not been rigorously defined and
the terminology is used as a conceptual descriptor.
Thus, inhibitors are ordinarily classified as transition
state analogue inhibitors if they are designed in such a
way that they bear, in place of the scissile peptide bond,
a moiety that resembles structurally a postulated tran-
sition state in the enzymic reaction and exhibit such
kinetic features as a very low Ki value and slow inhibi-
tion rate. Recently, Bartlett and Marlowe8 have pro-
posed a criterion for testing transition state analogue
inhibitors: According to them, there exists a linear
correlation between Ki values for a series of transition
state analogue inhibitors and Km/kcat values for a par-
allel series of the corresponding substrates, as repre-
sented by eq 1:
*Corresponding author: Tel.: +82-54-279-2101; fax: +82-54-279-
5877; e-mail: dhkim@postech.ac.kr
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00720-0

3162
J. D. Park, D. H. Kim / Bioorg. Med. Chem. Lett. 13 (2003) 3161–3166
log Ki ¼ log ðKm=kcatÞ þ log dkuncat
ð1Þ
Thus, for transition state analogue inhibitors, there is
obtained a straight line with a slope=1 when log Ki
values of a series of the transition state analogue inhib-
itors are plotted against log Km/kcat value for a series of
the corresponding substrates whose structures are
altered in such a way that the alteration does not affect
the rate of the noncatalyzed reaction.8,9 The criterion has
been successfully applied for assessing phosphorous-con-
taining substrate analogues as transition state analogue
inhibitors for TLN8,9 and carboxypeptidase A.10
Recently, we have reported that N-formyl-N-hydroxy-b-
phenylalanine N - methylamide (HCONOH - b - Phe -
NHMe) is a potent competitive TLN inhibitor having
the Ki value of 1.66ꢀ0.05 mM.11 It was thought to be of
interest to evaluate the reversed hydroxamate-bearing
TLN inhibitors as transition state analogue inhibitors
and we undertook the present study to find that inhibi-
tors of this type fulfill the criterion set forth by Bartlett
and Marlowe for transition state analogue inhibitors.
The synthetic route for the preparation of the inhibitors
is outlined in Scheme 1. (3R)-N-Benzyloxy-3-isobutyl-2-
azetidine (2) that was prepared from 4-methylvaleric
acid in eight steps according to the method reported by
Jin and Kim11 was treated with methanolin the pres-
ence of trimethylsilyl chloride to give 3. N-Formylation
of the latter with sodium formate and acetic anhydride
in anhydrous formic acid gave 4. Treatment of 4 with
concentrated ammonium hydroxide in the presence of
isobutylchloroformate and N-methylmorpholine gave
5a which was converted into 1a by hydrogenolysis.
Similarly, treatment of 4 with an appropriate amino
acid in the form of a benzylester afforded 5b–e, which
were then subjected to hydrogenolysis to obtain 1b–e.12
We chose to use the same series of tripeptide substrates
as those used by Bartlett and Marlowe8 for their study
for assessing phosphonamidates as transition state
mimics in the TLN-catalyzed proteolysis. These 0sub-
strates bear a l-Leu residue in common at the P1 site
0
and varied P2 residues. We have then synthesized a
series of hydroxamate-bearing0 inhibitors (1a–e), which
bear a l-Leu residue at the P1 site and the same series
of residues as those in the substrates at the P2 site.
0
Scheme 1. Reagents, conditions, and yield: (a) chlorotrimethylsilane (4 equiv), MeOH, rt, 2 h, 92%; (b) (i) 3 N HCl, reflux, 2 h; (ii) NaCO2H (1.2
equiv), Ac2O (3 equiv), HCO2H, rt, 12 h, 83%; (c) isobutylchloroformate, 4-methylmorpholine, NH4OH, rt, 6 h, 56%; (d) H2, Pd/C, MeOH, rt, 2 h,
88–95%; (e) isobutylchloroformate, 4-methylmorpholine, the desired amino acid benzyl ester, rt, 2 h, 65–78%.

J. D. Park, D. H. Kim / Bioorg. Med. Chem. Lett. 13 (2003) 3161–3166
3163
The compounds synthesized were assayed for TLN
inhibition. Inhibition constants (Ki values) were deter-
mined using N-[3-(2-furyl)acryloyl]-Gly-l-leucineamide
(FA-Gly-l-Leu-NH2) as substrate at pH 7.2 as descri-
bed in the literature13 and are collected in Table 1.
Inhibitor 1e having Phe at the C-terminalbinds TLN
most tightly with the Ki value of 1.84 mM. Inhibition
constants for 1a and 1d are decreased by 29- and 208-
fold, respectively, compared with that of 1e, suggesting
A variety of tools and methods such as X-ray crystal-
lographic studies,14 kinetic investigations,15 site directed
mutagenesis,16 interactive computer graphics,17 and
quantum chemicalcacl ulations 18 have been employed in
the investigations, and numerous reaction pathways,
although differ each other only in minor detail, have
been proposed. All these proposed mechanisms postu-
late in common a tetrahedral gem-diolate transition
state that is generated by the nucleophilic attack of the
zinc-bound activated water molecule on the carbonyl
carbon of the scissile peptide bond. The carbonyl oxy-
gen of the scissile bond is formally uncharged in the
ground state, but acquires a negative charge as the
reaction proceeds, and the transition state is stabilized
via coordination with the active site zinc ion and
through hydrogen bonding with the imidazole of His-
231 as depicted schematically in Figure 3. The binding
mode of HCONOH-b-Phe-NHMe to TLN has been
revealed by the X-ray diffraction analysis of the com-
plex of TLN with the inhibitor.19 Figure 4 is a schematic
representation of the binding mode of HCONOH-b-
Phe-NHMe to TLN. It can be seen in Figure 4 that the
two oxygen atoms of the hydroxamate moiety bind the
zinc ion in a bidentate fashion with distances of 2.00
and 2.36 A and the protonated carboxylate of Glu-143
is engaged in a hydrogen bonding with the carbonyl
oxygen of the inhibitor. The distance of 2.00 A for the
bond between the N-hydroxyloxygen and the zinc ion is
worthy of noting and suggests that the hydroxyloxygen
0
that the amino acid residue at the P2 plays an important
role in binding of the inhibitors to TLN. Comparison of
the Ki values of the series of inhibitors with the Km
values of substrates in a parallel series is made graphi-
cally (Fig. 1). As can be seen in Figure 1, no correlation
is found between Ki values for the inhibitors and the Km
values of the corresponding substrates, indicating that
the inhibitors are not ground state analogues. However,
plotting of log Ki versus log Km/kcat according to eq 1
gives a straight line (Fig. 2), whose linear regression
yields a slope of 0.99 (R=0.98), which suggests strongly
that the reversed hydroxamate moiety in the inhibitors
may mimic a transition state or a high-energy inter-
mediate generated by the attack of a water molecule to
the scissile peptide bond of substrate in the enzymatic
proteolysis reaction pathway.
An additionalevidence for support that hydroxamate
may mimic a transition state in the TLN-catalyzed pro-
teolytic reaction has been obtained by comparing the
binding mode of a hydroxamate-bearing inhibitor to
TLN with a postulated transition state in the enzymic
reaction. Mechanism of TLN-catalyzed proteolytic
reaction has been the subject of intensive investigations.
Table 1. Kinetic constants for inhibition of thermolysin and ther-
molysin-catalyzed proteolysis
Inhibitor
Ki (mM)
Corresponding substratea
Ki (mM) Km/Kcat (mM/s)
1d
1a
1b
1e
1c
383
16.6
20.6
10.87
2.4
3200
196
165
20
52.8
14.7
1.84
2.49
10.6
13.6
aFrom ref 21.
Figure 2.
Figure 3.
Figure 1.

3164
J. D. Park, D. H. Kim / Bioorg. Med. Chem. Lett. 13 (2003) 3161–3166
3. Zuman, M. A.; Oparil, S.; Calhoun, D. A. Nat. Rev. Drug
Discov. 2002, 1, 621.
4. Wittaker, M.; Floyd, D. D.; Brown, P.; Gearing, A. J. H.
Chem. Rev. 1999, 99, 2735.
5. Powers, J. C.; Harper, J. W. Inhibitors of Metalloproteases.
In Protease Inhibitors; Barrett, A. J., Salvesan, G., Eds.;
Elsevier Science: Amsterdam, 1986; p 219.
6. (a) Kraut, J. Science 1988, 242, 533. (b) Cannon, W. R.;
Sergleton, S. F.; Benkovic, S. J. Nat. Struct. Biol. 1996, 3, 821.
7. (a) Wolfenden, R. Acc. Chem. Res. 1972, 5, 10. (b) Lien-
hard, G. E. Science 1973, 180, 149. (c) Wolfenden, R. Transi-
tion State Affinity and the Design of Enzyme Inhibitors. In
Enzyme Mechansims; Page, M. I., Williams, A., Eds.; Royal
ChemicalSociety: London, 1987; p 97.
8. Bartlett, P. A.; Marlowe, C. K. Biochemistry 1983, 22,
4618.
9. Mader, M. M.; Bartlett, P. A. Chem. Rev. 1997, 97, 1281.
10. Hanson, J. E.; Kaplan, A. P.; Bartlett, P. A. Biochemistry
1989, 28, 6294.
11. Jin, Y.; Kim, D. H. Bioorg. Med. Chem. Lett. 1998, 8,
3515.
Figure 4.
binds the zinc ion in a deprotonated form in reminis-
cence of the postulated diolate transition state.20 The
deprotonated hydroxyloxygen of the hydroxamate is
hydrogen bonded to a nitrogen of the imidazole moiety
in His-231. The tetrahedrality of the zinc-bound hydro-
xamate moiety may be visualized by envisioning an
imaginary bond between the hydroxamate nitrogen and
the formyloxygen, the distance of which is estimated to be
2.23 A. This distance is considerably longer than a nor-
malC–O bond but can be reconcield with the proposi-
tion that the hydroxamate may represent an early stage
of the postulated transition state. It may be thus infer-
red from the foregoing comparison that the reversed
hydroxamate moiety mimics the electronic as well as the
geometric characteristics of the transition state in the
enzyme catalyzed proteolytic reaction.
12. (R)-2-(N-(O-Benzyloxy)aminomethyl)-4-methylpentanoic
acid methyl ester hydrochloride (3). To the solution of 2 (9.0 g,
38.6 mmol) in anhydrous MeOH (100 mL) was added slowly
chlorotrimethylsilane (19.6 mL, 0.154 mol) under nitrogen
atmosphere. The resulting solution was stirred for 2 h and
evaporated under reduced pressure to afford a white solid,
which was recrystallized from methanol and ether to give a
white crystal(10.7 g, 92%). Mp 116.5–117 ꢁC; [a]D20 +6.5ꢁ (c
0.65, DMSO); IR (KBr) 1738 cmꢂ1 1H NMR (DMSO-d6,
;
300 MHz) d 0.83 (q, 6H), 1.31–1.48 (m, 3H), 2.88 (m, 1H),
3.14–3.34 (m, 2H), 4.95 (s, 2H), 7.32–7.39 (m, 5H); 13C NMR
(DMSO-d6, 300 MHz) d 22.6, 23.5, 26.2, 40.0, 51.2, 52.5, 75.6,
129.4, 129.7, 129.8, 135.1, 174.9.
(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methylpen-
tanoic acid (4). A mixture of 3 (10.0 g, 33.1 mmol) and 3 N
HCl(100 mL) was refluxed for 2 h, then evaporated under
reduced
pressure
to
give
aminomethyl)-4-methylpentanoic acid hydrochloride as
(R)-2-(N-(O-benzyloxy)-
a
white powder (quantitative). To an ice-chilled solution of
sodium formate (2.7 g, 39.8 mmol) and (R)-2-(N-(O-benzyl-
oxy)aminomethyl)-4-methylpentanoic acid hydrochloride in
anhydrous formic acid (50 mL) was added slowly acetic
anhydride (9.4 mL, 99.4 mmol) and the resulting solution was
stirred for 12 h at room temperature. The reaction mixture
was evaporated under reduced pressure. The residue was dis-
solved in ethyl acetate (100 mL), washed successively with 1 N
HClsoul tion (50 mL ꢃ3) and brine (50 mLꢃ3), and dried over
anhydrous MgSO4. The crude product was purified by column
chromatography (EtOAc/n-hexane=1/2) to give the product
as a pale yellow oil (7.68 g, 83%): [a]2D5 +2.6ꢁ (c 0.5, MeOH);
Conclusion
The linear correlation observed between Ki values and
Km/kcat for inhibitors 1a–e and the corresponding sub-
strates coupled with the binding feature of the reversed
.
hydroxamate in forming the TLN HCONOH-b-Phe-
NHMe complex suggest strongly that compounds 1a–e
are a noveltype of transition state anaol gue inhibitors
for TLN, and a hydroxamate moiety would mimic a
diolate transition state postulated for the TLN-cata-
lyzed proteolytic reaction pathway. The results of the
present study imply that many of reversed hydroxamate
as well as hydroxamate type inhibitors developed for
zinc proteases including MMPs may function as transi-
tion state analogue inhibitors.
1
IR (CHCl3) 1669, 1731 cmꢂ1; H NMR (CDCl3, 300 MHz) d
0.89 (d, 6H), 1.24 (m, 1H), 1.59 (m, 2H), 2.04 (d, 1H), 2.82 (m,
1H), 3.85 (m, 1H), 4.81 (s, 2H), 7.35 (m, 5H), 7.97 (s, 3/10H),
8.12 (s, 7/10H); 13C NMR (CDCl3, 300 MHz) d 14.6, 20.7,
22.4, 23.3, 26.2, 39.2, 42.3, 46.0, 129.1, 129.4, 129.8, 134.5,
163.2, 180.4. HRMS (FAB+) (M+H)+: calcd for
C15H22NO4, 280.1549; found 280.1546.
Acknowledgements
(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methylpen-
tanoic acid amide (5a). To an ice-chilled solution of 4 (0.35 g,
1.25 mmol) in anhydrous THF was added isobutyl-
chloroformate (0.18 mL, 1.25 mmol) and NMO (0.14 mL, 1.25
mmol) and the resulting mixture was stirred for 15 min at 0 ꢁC.
To the reaction mixture was added concd NH4OH solution
(0.6 mL, 5 mmol) and then stirred for 6 h at room tempera-
ture. The reaction mixture was evaporated under reduced
pressure. The residue was dissolved in ethyl acetate (20 mL),
washed successively with 1 N HCl solution (20 mLꢃ3), 5%
NaHCO3 solution (20 mLꢃ3), and brine (20 mLꢃ3), and
The work was supported by Korea Research Founda-
tion Grant (KRF-2000-015-DS0025).
References and Notes
1. Lipscomb, N. W.; Strater, N. Chem. Rev. 1996, 96, 2375.
2. Beynon, R. J.; Beaumont, A. In Handbook of Proteolytic
Enzymes; Barrett, A. J., Rawlings, N. D., Woessner, J. F. Eds:
Academic: New York, 1998; p 1037.

J. D. Park, D. H. Kim / Bioorg. Med. Chem. Lett. 13 (2003) 3161–3166
3165
dried over anhydrous MgSO4. The crude product was purified
by column chromatography (EtOAc/n-hexane=1/1) to give
the product as an oil(0.20 g, 56%). [ a]2D5 +1.0ꢁ (c 1.0, CHCl3);
was filtered and the filtrate was evaporated under reduced
pressure to give 1a–1e. In cases of 1b–1e, the products were
isolated and characterized as salts of dicyclohexylamine.
(R)-2-((N-formyl-N-hydroxy)aminomethyl)-4-methylpentanoic
acid amide (1a). (Yield, 90%). [a]2D0 ꢂ13.6ꢁ (c 1.72, MeOH); IR
1
IR (CHCl3) 1666, 1685 cmꢂ1; H NMR (CDCl3, 300 MHz) d
0.88 (d, 6H), 1.26 (m, 1H), 1.59 (m, 2H), 2.67 (m, 1H), 3.69
(m, 2H), 4.84 (s, 2H), 5.55–5.74 (m, 2H), 7.24–7.38 (m, 5H),
7.82 (s, 3/10H), 8.21 (s, 7/10H); 13C NMR (CDCl3, 300 MHz)
d 22.5, 23.4, 26.2, 40.0, 43.0, 62.4, 129.1, 129.6, 129.9, 134.5,
163.9, 176.6, 205.3. HRMS (FAB+) (M+H)+: calcd for
C15H23N2O3, 279.1709; found 279.1708.
General procedure for the preparation of 5b–5e. To an ice-chil-
led solution of 4 in anhydrous CH2Cl2 was added isobutyl-
chloroformate (1 equiv) and NMO (2 equiv) and the resulting
mixture was stirred for 15 min at 0 ꢁC. To the reaction mixture
was added the solution of the corresponding amino acid ben-
zylester (1 equiv) and then stirred for 2–4 h at room tem-
perature. The reaction mixture was evaporated under reduced
pressure. The residue was dissolved in ethyl acetate (50 mL),
washed successively with 1 N HCl solution (50 mLꢃ3), 5%
NaHCO3 solution (50 mLꢃ3), and brine (50 mLꢃ3), and
dried over anhydrous MgSO4. The crude product was purified
by column chromatography (EtOAc/n-hexane=1/2) to give
the product as an oil.
N-[(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methyl-
pentanoyl]glycine benzyl ester (5b). (Yield, 65%). [a]2D0 ꢂ13.1ꢁ
(c 2.04, CHCl3); IR (CHCl3) 1653, 1684, 1750 cmꢂ1; 1H NMR
(CDCl3, 300 MHz) d 0.87 (d, 6H), 1.07–1.25 (m, 1H), 1.59 (m,
2H), 2.67 (m, 1H), 3.06–3.73 (m, 2H), 3.894.12 (m, 2H), 4.68–
4.92 (m, 2H), 5.14 (s, 2H), 6.30 (m, 1H), 7.26–7.35 (m, 10H),
7.86 (s, 3/10H), 8.15 (s, 7/10H); 13C NMR (CDCl3, 300 MHz)
d 22.5, 23.4, 26.1, 39.9, 41.8, 43.8, 47.1, 67.5, 128.8, 128.9,
129.0, 129.3, 129.5, 129.8, 134.5, 135.5, 163.5, 170.6, 204.4.
HRMS (EI): calcd for C24H30N2O3, 426.2155; found 426.2155.
N-[(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methyl-
pentanoyl]-L-alanine benzyl ester (5c).
(Yield, 71%). [a]2D0 ꢂ9.6ꢁ (c 0.63, CHCl3); IR (CHCl3) 1656,
1681, 1743 cmꢂ1; 1H NMR (CDCl3, 300 MHz) d 0.86 (d, 6H),
0.96–1.23 (m, 1H), 1.28 (d, 3H), 1.57 (m, 2H), 2.59 (m, 1H),
3.06–3.73 (m, 2H), 4.59 (t, 1H), 4.75–4.98 (m, 2H), 5.15 (q,
2H), 6.16 (m, 1H), 7.267.37 (m, 10H), 7.86 (s, 3/10H), 8.14 (s,
7/10H); 13C NMR (CDCl3, 300 MHz) d 14.6, 18.5, 21.4, 22.5,
23.4, 26.0, 48.4, 60.8, 67.5, 128.6, 128.8, 129.1, 129.5, 129.9,
130.0, 133.9, 135.7, 163.2, 173.6, 204.4. HRMS (EI): calcd for
C25H32N2O5, 440.2311; found 440.2312.
N-[(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methyl-
pentanoyl]-D-alanine benzyl ester (5d). (Yield, 72%). [a]2D0 ꢂ3.7ꢁ
(c 0.64, CHCl3); IR (CHCl3) 1655, 1682, 1743 cmꢂ1; 1H NMR
(CDCl3, 300 MHz) d 0.86 (d, 6H), 1.12 (m, 1H), 1.36 (d, 3H),
1.56 (m, 2H), 2.61 (m, 1H), 3.68 (m, 2H), 4.56 (m, 1H), 4.73–
5.01 (m, 2H), 5.09 (s, 2H), 6.25 (m, 1H), 7.24–7.32 (m, 10H),
7.82 (s, 3/10H), 8.15 (s, 7/10H); 13C NMR (CDCl3, 300 MHz)
d 18.8, 22.5, 23.4, 26.2, 39.8, 43.7, 47.3, 48.5, 67.5, 128.6,
128.8, 129.0, 129.1, 129.4, 129.7, 134.6, 135.7, 163.6, 173.5,
204.4. HRMS (EI): calcd for C25H32N2O5, 440.2311; found
440.2307.
N-[(R)-2-((N-(O-Benzyloxy)-N-formyl)aminomethyl)-4-methyl-
pentanoyl]-L-phenylalanine benzyl ester (5e). (Yield, 78%). [a]2D0
ꢂ6.8ꢁ (c 0.34, CHCl3); IR (CHCl3) 1656, 1682, 1742 cmꢂ1; 1H
NMR (CDCl3, 300 MHz) d 0.80 (d, 6H), 1.21 (m, 1H), 1.50
(m, 2H), 2.53 (m, 1H), 2.93 (m, 2H), 3.66 (m, 2H), 4.67 (t, 1H),
4.88 (s, 2H), 5.08 (q, 2H), 6.06 (m, 1H), 6.96–7.35 (m, 15H),
7.82 (s, 3/10H), 8.09 (s, 7/10H); 13C NMR (CDCl3, 300 MHz)
d 22.5, 23.4, 26.0, 38.1, 39.9, 43.9, 46.8, 53.4, 67.6, 127.4,
128.9, 129.0, 129.1, 129.7, 130.1, 135.5, 136.2, 163.3, 171.6,
204.3. HRMS (EI): calcd for C31H36N2O5, 516.2624; found
516.2625.
1
(KBr) 1668, 1689 cmꢂ1; H NMR (CD3OD, 300 MHz) d 0.93
(d, 6H), 1.20 (m, 1H), 1.57 (m, 2H), 2.95 (m, 1H), 3.33–3.75
(m, 2H), 7.85 (s, 7/10H), 8.23 (s, 3/10H); 13C NMR (CD3OD,
300 MHz) d 21.4, 22.6, 26.1, 39.1, 42.3, 52.8, 158.5, 163.0,
177.8; HRMS (FAB+) (M+H)+: calcd for C15H24N2O3,
189.1239; found 189.1234.
N-[(R)-2-((N-Formyl-N-hydroxy)aminomethyl)-4-methylpent-
anoyl]glycine (1b). (Yield, 92%). Mp 153.5–154.5 ꢁC; [a]D20
ꢂ13.3ꢁ (c 0.12, MeOH); IR (KBr) 1651, 1669, 3389 cmꢂ1; H
1
NMR (CD3OD, 300 MHz) d 0.89 (d, 6H), 1.10–1.36 (m, 11H),
1.55 (m, 2H), 1.67 (d, 2H), 1.81 (m, 4H), 2.03 (m, 4H), 2.61–
2.89 (m, 1H), 3.14 (m, 2H), 3.34–3.55 (m, 2H), 3.81–3.86 (m,
1H), 4.04–4.10 (m, 1H), 7.83 (s, 3/10H), 8.25 (s, 7/10H); 13C
NMR (CD3OD, 300 MHz) d 21.3, 22.6, 24.5, 25.2, 26.1, 29.5,
38.5, 43.7, 50.5, 53.4, 158.5, 163.3, 175.0. Anal. Calcd. for
C22H41N3O51/2H2O: C, 60.52; H, 9.70; N, 9.62. Found: C,
60.34; H, 9.73; N, 9.63.
N-[(R)-2-((N-Formyl-N-hydroxy)aminomethyl)-4-methylpent-
anoyl]-L-alanine (1c). (Yield, 88%). Mp 182–183 ꢁC; [a]2D0
ꢂ19.2ꢁ (c 0.59, MeOH); IR (KBr) 1645, 1669, 3341 cmꢂ1; H
1
NMR (CD3OD, 300 MHz) d 0.81–0.88 (m, 6H), 1.09–1.32 (m,
13H), 1.49 (m, 2H), 1.63 (d, 2H), 1.77 (m, 4H), 2.00 (m, 4H),
2.63–2.87 (m, 1H), 3.10 (m, 2H), 3.24–3.51 (m, 1H), 3.61–3.69
(m, 1H), 4.15 (t, 1H), 7.74 (s, 7/10H), 8.19 (s, 3/10H); 13C
NMR (CD3OD, 300 MHz) d 18.7, 21.7, 22.5, 24.5, 25.2, 26.0,
29.5, 38.7, 43.0, 50.9, 53.4, 158.6, 163.1, 173.5. Anal. calcd for
C23H43N3O51/4H2O: C, 61.92; H, 9.83; N, 9.42. Found: C,
61.95; H, 10.08; N, 9.44.
N-[(R)-2-((N-Formyl-N-hydroxy)aminomethyl)-4-methylpent-
anoyl]-D-alanine (1d). (Yield, 90%). Mp 87–89 ꢁC; [a]2D0 ꢂ10.5ꢁ
1
(c 0.65, MeOH); IR (KBr) 1651, 1670, 3388 cmꢂ1; H NMR
(CD3OD, 300 MHz) d 0.91 (d, 6H), 1.26 (m, 1H), 1.31–1.34
(m, 15H), 1.56 (m, 2H), 1.82 (m, 4H), 2.05 (m, 4H), 2.70 (m,
1H), 3.13 (m, 2H), 3.42–3.79 (m, 2H), 4.26 (t, 1H), 7.85 (s, 3/
10H), 8.26 (s, 7/10H); 13C NMR (CD3OD, 300 MHz) d 17.6,
21.4, 22.5, 24.6, 25.2, 26.1, 38.7, 43.5, 50.4, 53.6, 158.7, 163.3,
174.3. Anal. calcd for C23H43N3O51/3H2O: C, 61.72; H, 9.83;
N, 9.39. Found: C, 61.97; H, 10.09; N, 9.40.
N-[(R)-2-((N-Formyl-N-hydroxy)aminomethyl)-4-methylpent-
anoyl]-L-phenylalanine (1e) (Yield, 94%). Mp 170–171 ꢁC; [a]2D0
+17.4ꢁ (c 0.23, MeOH); IR (KBr) 1651, 1668, 3408 cmꢂ1; H
1
NMR (CD3OD, 300 MHz) d 0.74–0.89 (m, 6H), 1.12–1.43 (m,
13H), 1.65 (d, 2H), 1.80 (m, 4H), 1.99 (m, 4H), 2.63–2.85 (m,
1H), 2.96 (m, 1H), 3.07–3.11 (m, 3H), 3.34–3.53 (m, 2H), 4.41
(m, 1H), 7.75 (s, 3/10H), 8.14 (s, 7/10H); 13C NMR (CD3OD,
300 MHz) d 21.3, 22.7, 24.5, 25.2, 25.8, 29.5, 38.6, 43.5, 50.5,
53.3, 56.4, 126.3, 128.2, 129.6, 138.7, 158.5, 163.2, 173.9. Anal.
calcd for C29H47CN3O51/3H2O: C, 66.51; H, 9.17; N, 8.02.
Found: C, 66.67; H, 9.10; N, 8.06.
13. Feder, J.; Broughham, L. R.; Wildi, B. S. Biochemistry
1974, 13, 1186.
14. Matthews, B. W. Acc. Chem. Res. 1988, 21, 333.
15. Izquierdo, M. C.; Stein, R. L. J. Am. Chem. Soc. 1990,
112, 6054.
16. Beaumont, A.; O’Donohue, M. J.; Peredes, N.; Rousselet,
N.; Assicot, M.; Bohuon, C.; Fourie-Zaluske, M.-C.; Roques,
B. J. Biol. Chem. 1995, 270, 16803.
17. Hangauer, D. G.; Monzingo, A. F.; Matthews, B. W.
Biochemistry 1984, 23, 5730.
18. (a) Antonczak, S.; Monard, G.; Ruiz-Lopez, M. F.;
Revail, J.-L. J. Am. Chem. Soc. 1998, 120, 8825. (b) Anton-
czak, S.; Monard, G.; Ruiz-Lopez, M. F.; Revail, J.-L. J. Mol.
Model 2000, 6, 527. (c) Pelmenschikov, V.; Blomberg, M. R. A.;
Siegbahn, P. E. M. J. Biol. Inorg. Chem. 2002, 7, 284.
General Procedure for the preparation of 1a–1e. A solution of
5a-5e in MeOH (20 mL) was stirred for 2 h under hydrogen
atmosphere in the presence of 10% Pd-C. The resulting mixture

3166
J. D. Park, D. H. Kim / Bioorg. Med. Chem. Lett. 13 (2003) 3161–3166
19. Kim, S.-J.; Kim, D. H.; Park, J. D.; Woo, J.-R.; Jin, Y.;
Ryu, S. E. Bioorg. Med. Chem. 2003, 11, 2421.
20. (a) The deprotonation may most likely be resulted as a
consequence of the high Lewis acidity of the zinc ion:
Izquierdo-Martin, M.; Stein, R. L. J. Am. Chem. Soc. 1992,
114, 325. (b) Cross, J. B.; Duca, J. S.; Kaminski, J. J.; Madi-
son, V. S. J. Am. Chem. Soc. 2002, 124, 11004.
21. Morihara, K.; Tsuzuki, H. Eur. J. Biochem. 1970, 15, 374.