Evaluation of hadacidin analogues

Article abstract of DOI:10.1016/j.bmcl.2010.10.088

Several derivatives of hadacidin have been developed and evaluated for activity against adenylosuccinate synthetase.

Full text of DOI:10.1016/j.bmcl.2010.10.088

Bioorganic & Medicinal Chemistry Letters 21 (2011) 517–519
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Evaluation of hadacidin analogues
Nidhi Tibrewal a, Gregory I. Elliott a,b,
⇑
a Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, KY 40536, USA
b Department of Chemistry, University of Kentucky, Lexington, KY 40536, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Several derivatives of hadacidin have been developed and evaluated for activity against adenylosuccinate
synthetase.
Received 22 September 2010
Accepted 18 October 2010
Available online 30 October 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keyword:
Adenylosuccinate
Hadacidin (1, N-formyl-N-hydroxyglycine) was originally
isolated from a fermentation broth of Penicillium frequentans. It
was later identified as a potent anti tumor agent.1 In 1962, Gordon
and co-workers reported the inhibitory effect of hadacidin on the
de-novo biosynthesis of adenylic acid in Ehrlich ascites tumor
and rat liver cells.2,3 Hadacidin is a known inhibitor of adenylosuc-
cinate synthetase (AdSS). Structurally, hadacidin resembles aspar-
tic acid (Fig. 1). Our group is investigating a series of newly
developed hadacidin analogues and testing the inhibitory effects
on adenylosuccinate synthetase.
GDP + Pi
GTP + aspartate
IMP
+
Mg2
ASMP
AdSS
O
HO2C
NH2
OH
N
CO2H
CO2H
Aspartic acid
Hadacidin
1
Figure 1. The conversion of IMP into ASMP.
Cells in approximately 25% of many common human cancers
are deficient in the enzyme methylthioadenosine phosphorylase
(MTAP)4 and rely entirely on the de-novo biosynthesis of adenine
nucleotides. Adenylosuccinate synthetase catalyses the first com-
mitted step in the de-novo biosynthesis of adenosine monophos-
phate (AMP). The enzyme catalyses the reaction in two steps;
treated with hydroxylamine hydrochloride in aqueous NaOH
which provided the corresponding oximes 4a–f in 80–90% yield.
Reduction of the oximes 4a–f with NaBH3CN in acetic acid
provided the racemic branched hydoxylamino acids 5a–f in mod-
erate yield. Installation of the required formyl group by treatment
of 5a–f with a mixture of formic acid in acetic anhydride provides
first, transfer of c-phosphate group of GTP to O-6 of IMP forming
a highly reactive 6-phosphoryl IMP derivative along with GDP. This
step is facilitated by Mg2+ ion. Second, the amino group of aspartic
acid displaces the 6-phosphoryl group of 6-phosphoryl-IMP to
form adenylosuccinate monophosphate (ASMP) (Fig. 1).5 The en-
zyme adenylosuccinate lyase then converts ASMP into AMP.
In 1984, Jahngen and Rossomondo evaluated hadacidin ana-
logues that were functionalized at both terminal ends.6 In this Let-
ter, we decided to examine the compounds that would probe the
aspartic acid active site by increasing the chain length of carbon
atoms in between the terminal ends and by adding a branched
the
(Scheme 1).
A similar method, shown in Scheme 2, was used for the synthe-
sis of compound 9, which is a one carbon extension of 6a. Commer-
cially available 4-ketopentanoic acid (7) was treated with hydroxyl
amine in acetic acid; this was followed by NaCNBH3 reduction of
the oxime and treatment with concentrated hydrochloric acid to
provide compound 8. Formylation of 8 with formic acid in acetic
anhydride provided compound 9 in 56% yield.
Compound 12, a one carbon homolagation of hadacidin, was
synthesized in four steps. This was acquired through a Michael
type addition of N-benzyloxyamine onto ethyl acrylate which
made available compound 10 in 40% yield.9 The b-ben-
zylhydroxylamino ester was hydrolyzed with lithium hydroxide
providing the intermediate acid that was debenzylated with palla-
dium on carbon. This two-step sequence supplied compound 11 in
90% yield. Formylation of 11 with formic acid in acetic anhydride
a-substituted hadacidin analogues 6a–f in excellent yields
a-chain. Herein, we describe our syntheses and evaluation of these
hadacidin analogues.
The synthesis of branched hadacidin compounds began with
commercially available
a a-keto acids were
-keto acids (3).7,8 The
⇑
Corresponding author. Tel.: +1 859 257 2798.
E-mail addresses: g2elliott@ucsd.edu, gelli3@uky.edu (G.I. Elliott).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.10.088

518
N. Tibrewal, G. I. Elliott / Bioorg. Med. Chem. Lett. 21 (2011) 517–519
OH
O
O
Boc
N
O
a
b, c
N
a
b, c
Br
n
EtO
EtO
OBn
R
CO2H
3 (a-f)
R
CO2H
4 (a-f)
n
R
13(a-b)
14 (a-b)
a = n = 3
a = Me
b = Et
O
b = n = 4
OH
N
OH
c
= n-Pr
HN
O
O
Boc
N
d
d
H
= CH(CH3)CH2CH3
e
d
N
e = CH2CO2H
f = CH2CH2CO2H
R
CO2H
6 (a-f)
R
CO2H
5 (a-f)
HO
OH
16 (a-b)
HO
HO
OH
n
n
15 (a-b)
Scheme 1. Synthesis of hadacidin analogues with different R groups at
a-carbon.
Reagents and conditions: (a) hydroxylamine hydrochloride, NaOH; 80–90%; (b)
NaBH3CN, 50% acetic anhydride; 40–50%; (c) concd HCl; (d) acetic anhydride,
formic acid, 70–80%.
O
N
O
OH
n
17 (a-b)
OH
Scheme 4. Reagents and conditions: (a) BocNHOBn, NaH, NaI, dry DMF, 60–70%;
(b) 1 N LiOH, THF/MeOH/H2O, 90%; (c) H2/Pd-C, quant.; (d) TFA, DCM, 0 °C, quant.;
(e) acetic anhydride, formic acid, 50–80%.
O
HN
a, b, c
CO2H
CO2H
7
8
18 °C and 250 rpm for ꢀ9 h (or until OD600 = 0.5), and gene
expression was induced with 0.1 mM IPTG. Cells were harvested
by centrifugation after incubation at 18 °C and 250 rpm for 14 h
post-induction. Pelleted cells were resuspended in 100 mM Tris–
HCl pH 8.0 and 300 mM KCl and lysed using a French Press using
one pass at 15,000 psi. Following high-speed centrifugation, the
soluble extract was loaded onto a Profinia™ protein purification
system equipped with a 1 mL Bio-Scale™ Mini Profinity IMAC
cartridge and a 10 mL Bio-Scale™ Mini Bio-GelÒ P-6 desalting
cartridge (Bio-Rad). The purified protein was concentrated using
an Amicon Ultra 10,000 MWCO centrifugal filter (Millipore Corp.)
and stored as 50% glycerol stocks at À20 °C. Under these condi-
tions, no loss in activity of recombinant AdSS was observed for
minimally 6 months.
OH
O
N
d
9
CO2H
Scheme 2. Reagents and conditions: (a) hydroxylamine hydrochloride, NaOH; 52%;
(b) NaBH3CN, 50% acetic anhydride; 18 h; (c) concd HCl; 92%; (d) acetic anhydride,
formic acid, 56%.
proceeded smoothly and provided N-formyl-N-hydroxy-b-amino-
propionic acid (12; Scheme 3).
The extended derivatives, compounds 17a and 17b, of hadaci-
din were available through the reaction of N-benzyloxy-N-tert-but-
oxycarbonylamino10 with either 3- or 4-bromoesters (13a or 13b)
providing compound 14a or 14b. The yield for 14a was 70%. Com-
pound 14b was carried forward without purification when
0.9 equiv of sodium hydride was utilized. The hydrolysis of the es-
ter was accomplished with LiOH. For either substrate the yields
were excellent. The deprotection of the benzyl hydroxamate with
hydrogenolysis afforded compounds 15a and 15b. The overall yield
of compound 16 after hydrolysis, benzyl deprotection and treat-
ment with TFA for the removal of the Boc group was 90%. Introduc-
tion of a formyl group on nitrogen provided the analogues with
either 3 or 4 carbon linkers between the N- and C-terminus of
hadacidin (Scheme 4).
The ability of hadacidin derivatives to inhibit AdSS was demon-
strated by HPLC detection. Adenylosuccinate synthetase catalyses
the condensation of IMP with aspartate forming adenylosuccinate
monophosphate. The enzymatic reaction also requires the conver-
sion of GTP to GDP. The mixture (100
(pH 7.5), 50 M IMP, 25 M GTP, 500
pound and 2.0 g of purified AdSS enzyme. The assay was initiated
at room temperature with the addition of either 100 or 500 M of
ll) contained, 20 mM Hepes
l
l
lM MgCl2, 10 mM test com-
l
l
aspartate. After 5 min, the solution was stopped by freezing in a
methanol/dry ice bath. The solution was allowed to reach room
temperature before 20 ll of reaction mixture was injected onto
the HPLC. The mobile phase for the assay contained 65 mM potas-
sium phosphate, 1 mM PIC A, and 10% methanol. The buffer has a
pH of 4.4. The HPLC assay procedure allows separation of all UV
For AdSS production, pMOB4511 was introduced into Escherichia
coli BL21(DE3), and positive transformants were selected on LB
agar supplemented with 30
was transferred to 3 mL LB supplemented with 30
cin and grown at 37 °C and 250 rpm for 10 h. A 0.5 mL aliquot was
transferred to 50 mL LB supplemented with 30 g/mL kanamycin
l
g/mL kanamycin. A single colony
lg/mL kanamy-
l
5:1asp to cmpd
1:1asp to cmpd
and incubated at 37 °C and 250 rpm overnight. Subsequently,
5 mL of the culture was used to inoculate 500 mL LB supplemented
with 30
lg/mL kanamycin. The 500 mL culture was incubated at
O
O
O
a
b
c
EtO
EtO
d
NHOBn
10
O
OH
O
OH
HO
N
HO
N
a
2
H
11
12
Scheme 3. Reagents and conditions: (a) NH2OBnÁHCl, Et3N, EtOH, À80 °C, 3 days,
40%; (b) 1 N LiOH, THF/MeOH/H2O, 80%; (c) H2/Pd-C, quant.; (d) acetic anhydride,
formic acid, 80%.
Figure 2. Inhibition of adenylosuccinate synthetase activity by hadacidin analogues
with different R groups at a
-C. aN-hydroxy glycine.

N. Tibrewal, G. I. Elliott / Bioorg. Med. Chem. Lett. 21 (2011) 517–519
519
with two-carbon linker showed almost negligible inhibition. The
formylated derivative, compound 12, unexpectedly did not show
any improvement in activity. There was a similar observed trend
of weak inhibition for both the compounds 17a and 17b.
5:1 asp to cmpd
1:1 asp to cmpd
In conclusion, several structurally similar hadacidin analogues
were tested for possible inhibition of adenylosuccinate synthetase.
The introduction of methyl substituent at the
a-carbon reduced
the ability of the compound to inhibit the enzyme by 50% when
tested at aspartate to substrate of 5:1. Since this was a racemic
compound, the possibility exists that one isomer may be active.
Addition of more bulky groups on the
a-carbon further reduced
a
2
the activity. By increasing the carbon chain length between the ter-
minal ends of the N-hydroxy and carboxylic acid resulted in nearly
complete loss of activity. None of the analogues tested for activity
against adenylosuccinate synthetase showed activity better than
hadacidin. We are currently working on the synthesis of optically
pure 6a to further evaluate the active site and to provide com-
pounds with better activity.
Figure 3. Inhibition of AdSS by hadacidin analogues with varying carbon linker
between N-hydroxy and carboxylic acid. aN-hydroxy glycine.
active components (GDP, GTP, and ASMP) of the mixture at
266 nm. The amount of adenylosuccinate monophosphate formed
was quantified by the integration of peak areas over three runs.
At 10 mM hadacidin completely inhibited the enzyme (Fig. 2).
However, by adjusting the ratio of aspartate to hadacidin to 5:1,
the inhibition dropped to 80%. We tested all final compounds at
both concentrations. Compound 6a, having a methyl group at the
Acknowledgements
This work was supported by the American Cancer Society (IRG
85-001-22). The authors are grateful to Mike Rosenbach and
Dennis Carson of the University of California, San Diego Moores
Cancer Center for providing the enzyme and very helpful
discussions.
a
-carbon demonstrated approximately 50% reduced ability to inhi-
bit the enzyme when asp:6a was 5:1. The inhibition improved to
90% when equimolar concentrations of aspartate and compound
6a were tested.
Supplementary data
Addition of more bulky groups on the a-carbon (6b–d) resulted
in nearly complete loss of activity at 5:1 ratio of aspartate:test
compound. When the concentration of the test compound was in-
creased to 1:1 some inhibition was observed, albeit very low.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.10.088.
When the
a-chain went from ethyl to propyl, the expected loss
of activity was observed. But when the diastereomers of isobutyl
6d were tested, some activity against the enzyme was retained.
Furthermore, compounds 6e and 6f with polar carboxylic
References and notes
1. Kaczka, E. A.; Gitterman, C. O.; Dulaney, E. L.; Folkers, K. Biochemistry 1962, 1,
340.
2. Shigeura, H. T.; Gordon, C. N. J. Biol. Chem. 1962, 237, 1932.
3. Shigeura, H. T.; Gordon, C. N. J. Biol. Chem. 1962, 237, 1937.
4. Kamatani, N.; Nelson-Rees, W. A.; Carson, D. A. Proc. Natl. Acad. Sci. U.S.A. 1981,
1219.
groups off the
a-carbon showed 30% and 60% inhibition, respec-
tively, with equimolar concentration of aspartate and compounds.
Without the formyl group, N-hydroxyglycine showed minimal
effect on AdSS activity. With other analogues (5a–c, f, 8) inhibition
also fell significantly. This confirmed that the formyl group was
necessary for inhibition.
In our second series of hadacidin analogues (Fig. 3) the result of
increasing the chain length between the N-hydroxy-N-formyl and
the carboxylic acid of hadacidin were as expected. At a concentra-
tion of 5:1, all tested compounds showed very weak to nonexistent
inhibition. At equimolar concentration to aspartate, compound 11,
5. Lieberman, I. J. Biol. Chem. 1956, 223, 327.
6. Jahngen, E. G. E.; Rossomando, E. F. Arch. Biochem. Biophys. 1984, 229, 145.
7. Kato, Y.; Tsuda, T.; Asano, Y. Biochem. Biophys. Acta 2007, 1774, 856.
8. Ahmad, A. Bull. Chem. Soc. Jpn. 1974, 47, 1819.
9. Yakirevitch, P.; Rochel, N.; Albrecht-Gary, A. M.; Libman, L.; Shanzer, A. Inorg.
Chem. 1993, 32, 1779.
10. Kurtz, T.; Geffken, D.; Wackendorff, C. Z. Naturforsch. 2003, 58b, 457.
11. A gift from Dennis Carson of the Moores Cancer Center at UCSD.