Design, synthesis, and evaluation of inhibitors of pyruvate phosphate dikinase

Article abstract of DOI:10.1021/jo3018473

Pyruvate phosphate dikinase (PPDK) catalyzes the phosphorylation reaction of pyruvate that forms phosphoenolpyruvate (PEP) via two partial reactions: PPDK + ATP + P_i → PPDK-P + AMP + PP_i and PPDK-P + pyruvate → PEP + PPDK. Based on i

Full text of DOI:10.1021/jo3018473

Article
pubs.acs.org/joc
Design, Synthesis, and Evaluation of Inhibitors of Pyruvate
Phosphate Dikinase
Chun Wu, Debra Dunaway-Mariano, and Patrick S. Mariano*
Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, New Mexico 87131, United States
S
* Supporting Information
ABSTRACT: Pyruvate phosphate dikinase (PPDK) catalyzes
the phosphorylation reaction of pyruvate that forms
phosphoenolpyruvate (PEP) via two partial reactions: PPDK
+ ATP + Pi → PPDK-P + AMP + PPi and PPDK-P + pyruvate
→ PEP + PPDK. Based on its role in the metabolism of
microbial human pathogens, PPDK is a potential drug target.
A screen of substances that bind to the PPDK ATP-grasp domain active site revealed that flavone analogues are potent inhibitors
of the Clostridium symbiosum PPDK. In silico modeling studies suggested that placement of a 3−6 carbon-tethered ammonium
substituent at the 3′- or 4′-positions of 5,7-dihydroxyflavones would result in favorable electrostatic interactions with the PPDK
Mg-ATP binding site. As a result, polymethylene-tethered amine derivatives of 5,7-dihydroxyflavones were prepared. Steady-state
kinetic analysis of these substances demonstrates that the 4′-aminohexyl-5,7-dyhydroxyflavone 10 is a potent competitive PPDK
inhibitor (Ki = 1.6 0.1 μM). Single turnover experiments were conducted using 4′-aminopropyl-5,7-dihydroxyflavone 7 to
show that this flavone specifically targets the ATP binding site and inhibits catalysis of only the PPDK + ATP + Pi → PPDK-P +
AMP PPi partial reaction. Finally, the 4′-aminopbutyl-5,7-dihydroxyflavone 8 displays selectivity for inhibition of PPDK versus
other enzymes that utilize ATP and NAD.
phosphorylated histidine enzyme intermediate PPDK-P, which
then participates in phosphorylation of pyruvate to form PEP.
The PPDK monomer comprises three sequential functional
domains (see Figure 1).8−11 The N-terminal domain has an
ATP-Grasp fold12 that forms an active site crevice where ATP,
Pi ,and Mg2+ bind. The C-terminal domain is an α,β-barrel, the
N-terminal edge of which forms the subunit interface of the
functional homodimer, with the C-terminal edge supporting an
active site crevice where pyruvate and Mg2+ bind. The small
globular α,β-fold central domain is connected to the terminal
domains by helical peptide linkers. The catalytic His455
residue, which is located on the surface of this domain,
transfers the phosphoryl group from donor to acceptor as the
central domain alternates with a swivel-like motion between the
active site of the N-terminal domain (conformer 1) and the
active site of the C-terminal domain (conformer 2) (Figure
1).11,13
INTRODUCTION
■
Pyruvate phosphate dikinase (PPDK; EC 2.7.9.1) catalyzes the
reversible phosphorylation reaction of pyruvate, adenosine 5′-
triphosphate (ATP), and inorganic phosphate (Pi) that forms
phosphoenolpyruvate (PEP), pyrophosphate (PPi) and AMP.1
PPDK is activated by divalent and monovalent metal ions. In
vivo, Mg2+ and NH4+ serve as the cofactors.2,3 The results from
extensive studies4−7 demonstrated that the chemical pathway of
the Clostridium symbiosum PPDK catalyzed reaction, shown in
Scheme 1, is initiated by nucleophilic substitution of the His455
imidazole N-3 nitrogen at the β-phosphorus of ATP to yield
AMP and a pyrophosphorylated enzyme intermediate (PPDK-
PP). This step is followed by Pi attack at the terminal
phosphoryl group of the PPDK-PP to form PPi and the
The biological range of PPDK is limited to specialized plants,
protozoa, and bacteria.14−23 Because of the small ΔGo of its
catalyzed reaction, PPDK can function in either direction and
thus fill different functional niches.1 PPDKs of plants24 and
photosynthetic bacteria (e.g., Chlorobium tepidum)25 function in
PEP production, a requirement for carbon dioxide fixation. In
the typanosomatids, PPDK is located in the glycosomes where
it functions in the recycling of pyrophosphate.26 For the single
cell protists and anaerobic bacteria, PPDK functions in a novel
Scheme 1
Special Issue: Howard Zimmerman Memorial Issue
Received: September 2, 2012
Published: October 24, 2012
© 2012 American Chemical Society
1910
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
ma. PPDK has also been found in the Wolbachia endosymbiont
from the filarial parasite Brugia malayi.16 Because it is not
present in higher organisms, PPDK holds great potential as a
target for the development of therapeutic agents to treat the
infectious diseases caused by these pathogens. The goal of the
effort described below was to identify a small molecule inhibitor
that might serve as a lead in the development of such agents.
PPDK catalysis operates from two active sites, either of
which could be targeted for inhibition. The active site that
catalyzes the third partial reaction (PPDK-P + pyruvate ⇔
PPDK + PEP) is located at the top of the β-barrel and is largely
comprised of polar residues. Oxalate, a mimic of the pyruvate
enolate anion intermediate, is a powerful PPDK inhibitor;10
however, this substance is not a promising lead for inhibitor
design. Instead, our attention focused on the N-terminal
domain active site where catalysis of the first two partial
reactions (PPDK + ATP + Pi ⇔ PPDK-PP + AMP + Pi ⇔
PPDK-P + AMP + PPi) occurs. As shown in Figure 2A, this
domain contains a large cleft that is formed by the association
of two subdomains. The snapshot of conformer 2 shown in
Figure 1 indicates that the two subdomains dissociate, thereby
expanding the cleft to facilitate substrate and Mg2+ cofactor
binding, and then reassociate to close the active site for catalysis
(conformer 1; Figures 1 and 2A).
As has been demonstrated with other ATP-dependent
enzymes, the potential exists for inhibiting catalysis by using
a tight-binding, space-filling ligand that blocks ATP binding.30
The goal is to combine the binding energy derived from
desolvation of matched nonpolar regions of the ligand and
binding site with the binding energy and specificity derived
from favorable electrostatic interaction between matched ligand
and binding site polar substituents. In general, larger ligands
provide a higher interaction potential than do smaller ligands,
and thus, a “space-filling” ligand is desirable.
Inhibitor targeting of the ATP binding site of ATP-grasp
enzymes is a relatively new endeavor.31 Because the
architecture of the ATP site of the ATP-grasp domain is so
similar to that found in eukaryotic protein kinases, a logical
starting point for inhibitor discovery would be screening of
chemical libraries originally used for protein kinase inhibitor
discovery. In this manner, a group from Pfizer identified several
tight binding inhibitors belonging to the pyridopyrimidine
structural class, which specifically target the ATP-grasp domain
enzyme biotin carboxylase.32 X-ray structural analyses of the
enzyme−inhibitor complexes showed that the inhibitors bind
to the ATP binding site where their respective phenyl
substituents engage in hydrophobic interactions with the
hydrophobic ribose binding region and the corresponding
pyridopyrimidine unit engages in numerous hydrogen bond
interactions with polar residues located in the adenine binding
region.32
Figure 1. Pymol diagrams of the Clostridium symbiosum PPDK
monomer in conformations 1 and 2. Catalysis of the ATP + Pi partial
reaction takes place in the N-terminal domain (subdomains are
colored forest green and wheat) active site in conformation 1, and
catalysis of the PPDK-P + pyruvate partial reaction takes place in the
C-terminal domain (colored blue) active site in conformation 2.
Movement of the catalytic His455 residue (shown as black stick)
between the two active sites is perceived to occur through
interconversion of conformers 1 and 2 via swiveling of the central
domain (colored yellow) about its helical linkers (colored red and
magenta). The Mg2+ cofactors are shown as cyan spheres. The PEP is
shown in stick form with carbon atoms colored yellow, nitrogen blue,
oxygen red, and phosphorus orange. The Pi and ATP ligands are
shown in stick form with the same color-coding except the carbon
atoms, which are colored magenta.
pyrophosphate-dependent glycolytic pathway,21,27,28 which is
the key source of chemical energy because the proteins that
support the citric acid cycle and oxidative phosphorylation are
absent. The Entamoeba histolytica PPDK has been shown to
exert significant control of glycolytic flux.27 PPDK gene
knockdown experiments carried out in Giardia lamblia have
shown that the level of ATP is reduced to 3% of the normal
level.28 These organisms in particular are likely to be especially
sensitive to inhibitors that block the function of PPDK.
PPDK has long been recognized as a logical target for
herbicide development29 because C4 plants, in which it is
mainly found, are primary members of the “weed” family.
PPDK is also present in the economically important bacterial
plant pathogens of the genus Phytophthora, which are known to
damage agricultural crops and to destroy natural ecosystems.21
Moreover, the PPDK gene is found in numerous human
bacterial pathogens including Porphyromonas gingivalis, Trepo-
nema palladium, Rickettsia conorii, Clostridium tetani, and
Streptococus faecalis and the protozoan human pathogens
Giardia, Entamoeba, Trichomonas, Leishmania, and Trypanoso-
PPDK has been long recognized as a potential herbicide29
and drug target33 investigations aimed at inhibitor development
have been limited to screenings of marine fungi and marine
sponge derived extracts for inhibition of plant-derived
PPDK.34−36 Notably, the hydroxyquinones, ilimaquinone,
ethylsmenoquinone, and smenoquinone, were reported to
have inhibition constants in the high micromolar range.
However, because these compounds show mixed-type inhib-
ition vs ATP it is not clear at which site in the enzyme they are
binding. On the other hand, the results from an in silico based
screen of small molecule libraries against a C. symbiosum PPDK-
derived homology model of the Entamoeba histolytica PPDK
1911
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
the cleft. The rear of the binding site is lined with clusters of
nonpolar residues, which could be paired with nonpolar
segments of an inhibitor. Sprinkled throughout the cleft are
individual uncharged, polar residues that can be targeted using
hydrogen bond forming substituents of an inhibitor.
The PPDK encoding gene spread from bacteria to eukaryotes
via lateral gene transfer and, consequently, the pairwise
sequence identity between bacterial PPDK and eukaryotic
PPDK is unusually high.17 For instance, the sequence identity
between C. symbiosum PPDK and Entamoeba histolytica is 52%,
and notably the residues of the C. symbiosum PPDK N-terminal
domain binding site pictured in Figure 2B are conserved.
Furthermore, the X-ray structure of the PPDK from
Trypansoma brucei PPDK38 reveals that its ATP binding site
is very similar to that in the bacterial enzyme. The C. symbiosum
PPDK was utilized as a starting platform for the inhibitor
discovery investigation because, in contrast to eukaryotic
PPDKs, it is most amenable to biochemical and structural
analysis. Here, we describe the results of studies that have led to
the identification of flavone derivatives as selective, tight
binding inhibitors of PPDK catalysis that act by competing with
ATP for binding to the N-terminal domain active site.
RESULTS AND DISCUSSION
■
Discovery of a First-Generation PPDK Inhibitor. For
the initial inhibitor screening effort, a large panel of compounds
derived from several structural classes of protein kinase
inhibitors,39−41 including 3-substituted indolin-2-ones, purines,
pyridopyrimidines, and flavonoids, were tested. In addition, the
natural products balanol,42 wortmannin,43 and gossypol44 were
included in the screen. The previously described4 spectropho-
tometric assay, based on the lactate dehydrogenase-NADH
coupling reaction, was employed to monitor the initial
velocities of the C. symbiosum PPDK-catalyzed reaction AMP
+ PPi + PEP → ATP + Pi + pyruvate in the presence and
absence of the inhibitor. The AMP concentration was set at 20
μM, which is twice the AMP Km of 10 μM. Typically, owing to
the limited water solubility, the maximum concentration of
inhibitor that could be used in the experiment was restricted to
100 μM. In the event that no inhibition was observed, the
smallest possible Ki value is reported to be greater than the
inhibitor concentration. Given that 69 compounds were tested,
structures and Ki values are reported in the Supporting
Information (see Table SI1) rather than the text. Of the
compounds screened, only two proved to be strong inhibitors.
One of these, quercetin (1), is a member of the flavone natural
product family, and the other is gossypol, a polyphenolic,
binaphthyl disesquiterpene (2). Further kinetic analysis (Figure
3A,B) revealed that both compounds display competitive
inhibition vs AMP with respective Ki values of 11.7 0.5 and
15 1 μM. The fact that competitive inhibition is observed is
evidence that the inhibitors bind to the ATP binding site only.
Figure 2. Pymol diagrams of the Clostridium symbiosum PPDK. (A) N-
terminal domain with one subdomain colored wheat and the other
forest green. The ATP (carbon is yellow, oxygen red, nitrogen blue,
and phosphorus orange) atoms and Mg2+ ion (cyan) are shown as
spheres. (B) The amino acid residues (shown in stick) that form the
N-terminal domain active site. The carbon atoms of the nonpolar
amino acids are colored white, those of the polar uncharged green,
those of the negatively charged cyan and those of the positively
charged yellow. All oxygen atoms are red, all nitrogen atoms blue, and
all sulfur atoms yellow.
suggest that the ATP binding site is a good candidate for
inhibition by a large space-filling ligand.37
Indeed, inspection of C. symbiosum PPDK N-terminal
domain active site reveals that the pattern of amino acid side
chains that project into this expansive binding site is ideal for
inhibitor binding (Figure 2B). Specifically, a cluster of charged
residues, which are available for favorable electrostatic
interaction with an inhibitor, are located toward the front of
As a consequence of our expectation that the inhibitor
development process would be more productive employing
quercetin as a lead compound, a second screening that focused
1912
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
Figure 3. Double-reciprocal plot for the inhibition of PPDK by (A) 0(λ), 32.5(5), 50.0(τ) μM gossypol; (B) by 6.8 (λ), 14.8 (ν), 29.6(τ) μM)
quercetin; (C) by 0(λ), 9.4(ν), 18.9 (τ) μM apigenin; (D) by 0(λ), 3.0 (ν), 4.5 (τ) μM aminoalkylflavone 10. Assay solutions contained 5−80 μM
AMP (range of 0.5- to 8-fold Km), 0.5 mM PEP, 1 mM PPi, 5 mM MgCl2, 40 mM NH4Cl, 0.2 mM NADH, 20 units/mL lactate dehydrogenase in
20 mM imidazole buffer at pH 6.8 and 25 °C.
on flavones was carried out (see Table 1). Twenty-four
commercially available and two synthetic (flavone and
quercetin-3,5,7,3′,4′-pentamethyl ether) flavones were tested
as inhibitors using the single-point diagnostic test (at 57−100
μM, vide supra) to estimate the Ki values. In cases where no
inhibition was observed, the minimal value of Ki possible is
reported to be greater than the concentration of inhibitor used
in the assay. Flavones that do display significant inhibition were
subjected to full kinetic analysis to demonstrate competitive
inhibition vs AMP and to determine the inhibition constant.
The results are reported in Table 1.
An analysis of the inhibition data displayed in Table 1 reveals
factors that govern binding of flavones to the ATP-grasp site of
PPDK. First, it appears that substances containing isoflavone-
type structures, such genistein (4) and daidzin (5), have low
PPDK binding affinities. Second, ( )-taxifolin (6), a flavonol
which is otherwise identical to quercetin except for the absence
of a 2,3-double bond, binds 3-fold less tightly to the ATP-site of
the enzyme than does quercetin. Thus, the presence of double
bond in the flavone B-ring is moderately important for
inhibitory activity. Third, flavones that possess either no or
one phenolic hydroxyl group are also poor PPDK inhibitors
(e.g., flavone and quercetin-3,5,7,3′,4′-pentamethyl ether). In
contrast, when at least two hydroxyl groups are present on the
flavone skeleton, dramatic improvements in inhibitory activities
take place. Among the group of flavones containing three
hydroxyl groups, apigenin displays the highest PPDK inhibitory
activity (Ki = 6 μM) (Figure 3C). Importantly, removal of the
C-5 and C-7 hydroxyl groups (e.g., 4′-hydroxyflavone, Ki > 68
μM) results in a decrease in binding to PPDK. However,
removal of the C-4′ hydroxyl group (e.g., 5,7-dihydroxyflavone
Ki = 16 μM) causes only a small decrease in binding affinity.
Furthermore, the presence of bulky polar groups, such as an o-
rhamnoside, in place of a hydroxyl hydrogen leads to greatly
reduced inhibition (quercetin, Ki = 11.7 μM, vs quercetrin, Ki
>100 μM). Finally, flavones containing five hydroxyl groups
have similar inhibitory activities as those with four hydroxyl
groups as long as two are located at the C-5 and C-7 positions.
Second-Generation Inhibitor Design and Docking
Studies. Comparison of the structures of the Src family
tyrosine kinase Hck bound with AMPPNP vs bound with
quercetin reveals that quercetin binds in the same position as
does the AMPPNP adenine ring. The adenine binding site in
PPDK combines nonpolar side chains that contribute to
desolvation-driven binding with backbone amides that can
potentially be engaged in hydrogen bonding with one or more
of the flavone carbonyl/hydroxyl substituents. Notably, the
remainder of the expansive active site is yet to be filled. Three
residues Glu-323, Asp-321, and Gln-335, clustered on one side
of the unfilled site, are believed to surround the magnesium ion
1913
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
a
Table 1. Flavone Competitive Inhibition Constants (Ki) or IC50 Values for PPDK
substituents
flavone
4, genistein
R3
R5
R6
R7
R2′
R3′
R4′
solvent
b,cKi (μM)
>100c
>100c
2% DMSO
0.4% DMSO
H2O
5, daidzin
6, ( )-taxifolin
quercitrin
27 1b
>100c
ORib
OMe
OMe
H
OH
OMe
OH
OH
OH
H
H
OH
OMe
OMe
OH
OH
H
H
OH
OH
OMe
OMe
OMe
OMe
H
2% DMSO
2% DMF
8%DMSO
H2O
quercetin pentamethyl ether
quercetin tetramethyl ether
acacetin
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
OH
H
H
H
H
H
H
H
OMe
OMe
H
>100c
H
>70c
H
>100c
dismetin
H
H
OH
H
2% DMSO
5% DMSO
4% DMSO
4% DMSO
H2O
16.4 0.2b
>100c
flavone
H
H
3-hydroxyflavone
5-hydroxyflavone
7-hydroxyflavone
4′-hydroxyflavone
5,4′-dihydroxyflavone
5,7-dihydroxyflavone
7,2′-dihydroxyflavone
7,3′-dihydroxyflavone
apigenin
OH
H
H
H
H
H
H
>75c
OH
H
H
H
H
H
>76c
H
OH
H
H
H
H
>57a
H
H
H
H
OH
OH
H
H2O
>68c
H
OH
OH
H
H
H
H
H2O
18 1c
16 1b
15 1b
22 4b
5.9 0.1b
15 1b
21.9 0.2b
13 1b
14 1b
13.4 0.5b
8.8 0.7b
11.7 0.5b
9.1 0.4b
H
OH
OH
OH
OH
H
H
H
H2O
H
OH
H
H
H
H2O
H
H
OH
H
H
H2O
H
OH
H
H
OH
OH
H
H2O
3,3′,4′-trihydroxyflavone
baicalein
OH
H
H
OH
H
H2O
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
H
2% DMSO
H2O
5,7,2′-trihydroxyflavone
lueolin
H
OH
OH
H
H
H
H
OH
OH
H
H
H2O
fisetin
H
OH
OH
OH
OH
H2O
kaempferol
OH
OH
OH
H
H2O
quercetin
H
OH
H
H2O
morin
OH
H2O
a
The 1 mL assay solutions initially contained 5−80 μM AMP (Ki determination) or 20 μM AMP (IC50 determination), 0.5 mM PEP, 1 mM PPi, 5
b
mM MgCl2, 40 mM NH4Cl, 0.2 mM NADH, 20 units/mL lactate dehydrogenase in 20 mM imidazole buffer (pH 6.8). Determined using a full
kinetic analysis to demonstrate competitive inhibition vs AMP. Estimated using a single-point diagnostic test (at 57−100 mM, vide supra) and a
minimal value is reported to be greater than the concentration of inhibitor used in the assay.
c
of the bound magnesium−ATP complex. By tethering an alkyl
ammonium group to the phenyl ring of 5,7-dihydroxyflavone
we hoped to target the amino acid triad through favorable
electrostatic interactions. In order to test this proposal, several
dihydroxyflavones (7−12) with structures that contain amino-
terminated alkyl substituents attached to the C-ring were
designed, prepared, and evaluated as second-generation PPDK
inhibitors. A selected member of this series, 4′-aminopropyl-
5,7-dihydroxyflavone 7, was first subjected to docking studies to
determine if it might bind to the active site in an orientation
that places the flavone moiety in the adenine binding site and
the ammonium group juxtaposed to the side chains of the side
chains of Glu-323, Asp-321, and Gln-335. Inspection of the
lowest energy conformer, shown superimposed on ATP in the
PPDK complexes in Figure 4, reveals that the surface the of 5,7-
dihydroxyflavone ring in 7 overlaps nearly completely with that
of the ATP adenine ring. Furthermore, the terminal ammonium
ion is well positioned to interact with the PPDK Glu 323, Asp
321, and Gln 335 side chains.
linker is longer (8−10) for testing as PPDK inhibitors. The
aminoalkyldihydroxyflavone 11 was also prepared to evaluate
the affect of repositioning the linker, whereas 12 was prepared
to test the impact of the restriction of conformational motion in
the alkyl linker. The synthetic sequences employed for the
preparation of 7−12 are shown in Schemes 2 and 3.
The aminoalkylflavones 7−12 were evaluated as competitive
inhibitors vs AMP, and the inhibition constants obtained are
displayed in Table 2. The results of this kinetic analysis clearly
show that incorporation of 3−6 carbon-containing, ammonium
ion terminated chains at the 4′-position of the 5,7-
dihydroxyflavone ring system results in a decrease in Ki,
which in the best inhibitor has been reduced by an order of
magnitude. In addition, placement of a 6-carbon-tethered
ammonium substituent at the 3′-position of 5,7-dihydroxy-
flavone (as in 11) results in the same degree of improvement in
PPDK binding as is seen in the 4-linked analogs. Finally, the
observation that the amino-trans-butenyl-flavone 12 binds 10-
fold less tightly than its saturated four-carbon analogue 8
suggests that side chain flexibility is an important factor
governing binding.
Encouraged by the results from the docking exercise, we
undertook the synthesis of the aminopropyl-5,7-dihydroxy-
flavone 7 as well as analogues in which the length of the alkyl
1914
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
located in an ATP-Grasp domain, it does not closely resemble
ATP binding sites in other ATP-Grasp domains. This is a
consequence of the fact that the ATP-Grasp domain is used to
catalyze ligase reactions. Thus, as a kinase, PPDK is an outlier
to the ATP-Grasp enzyme superfamily.31 The ATP binding
sites of protein kinases in general, and cAMP-dependent
protein kinase in particular, most closely resemble that of
PPDK. Therefore, we tested the inhibition of cAMP-dependent
protein kinase by the representative aminoalkyldihydroxyfla-
vone 8 and observed that it is a competitive inhibitor of this
kinase vs ATP with a Ki = 15 μM (Table 4). Thus, this inhibitor
binds to PPDK 8-fold more tightly than it does to cAMP-
dependent protein kinase.
Next, we examined a variety of kinases that function in
cellular metabolism. Of these, 8 served as an inhibitor of only
pyruvate kinase, which like the other kinases possesses an ATP
binding site that has no resemblance to that of PPDK.
Nevertheless, the Ki = 34 μM value observed is still 20-fold
greater than that observed for inhibition of PPDK by 8. Finally,
hexokinase, myokinase, and acetate kinase were not inhibited at
129 μM 8 nor was the NADH-dependent enzyme lactate
dehydrogenase.
CONCLUSIONS
■
The results presented here demonstrate that 5,7-dihydroxy-
flavones possessing properly positioned aminoalkyl chains are
potent inhibitors of PPDK and that they act through
competitive binding at the enzyme’s ATP binding site. The
flavone unit is an effective replacement of the ATP adenine ring
as shown by tight binding of apigenine. Placement of 3−6
carbon-tethered ammonium substituents at the 3′- or 4′-
positions of the 5,7-dihydroxyflavone skeleton results in tighter
binding which we propose is the result of favorable electrostatic
interactions between the positively ammonium ion and one or
more of the residues that form the triad Glu323, Asp321, and
Gln-335, otherwise positioned to interact with the ATP bound
Mg2+.
Figure 4. Overlays of docked structures of the PPDK complexes with
aminopropylflavone 7 (carbon, gray; nitrogen, blue; oxygen, red;
hydrogen, pink) and ATP (cyan). PPDK residues are colored as
follows: nitrogen, blue; oxygen, red; sulfur, orange; carbon, yellow.
Some residues are removed for clarity.
Aminoflavone 7 Targets the PPDK ATP Site. To
demonstrate that PPDK inhibition by the aminoalkyldihydrox-
yflavones derives solely from binding to the N-terminal domain
active site, we employed transient kinetic techniques to
separately test the inhibition of a single turnover reaction
catalyzed at the N-terminal domain active and the inhibition of
a single turnover reaction catalyzed at the C-terminal domain
active. Accordingly, the time course for the single turnover
reaction, PPDK + [14C]ATP + Pi → PPDK-P + [14C]AMP +
PPi, catalyzed by the enzyme in the presence and absence of the
representative aminoalkyldihydroxyflavone 7 is shown in Figure
5. The Ki = 3.3 μM derived from these data closely match the
Ki = 2.8 μM (Table 2) determined using steady-state kinetic
techniques.
The results of steady-state kinetic analysis demonstrate that,
in the series of substances explored, 4′-aminohexylflavone 10
(Ki = 1.6
0.1 μM) is the most potent competitive PPDK
inhibitor vs AMP. To confirm that the second-generation
flavones specifically target the ATP binding site, single turnover
experiments were conducted using aminopropylflavone 7 to
demonstrate that this substance inhibits catalysis of the PPDK
+ ATP + Pi → PPDK-P + AMP partial reaction and not the
ensuing PPDK-P + pyruvate → PEP + PPDK partial reaction.
Finally, the second-generation 4′-aminobutyl-flavone 8 was
observed to display selectivity for inhibition of PPDK versus
other enzymes that utilize ATP as a substrate.
Next, single time point (600 ms) measurements (percent
conversions) were made for the single turnover reaction, PPDK
+ [14C]PEP → PPDK-P + [14C]pyruvate, in the presence and
absence of 7 (see Table 3). As the results show, even at a very
high concentration (800 μM, from Table 2 the Ki measured for
the steady-state reaction is 2.8 μM) of 7, no inhibition takes
place. Thus, studies of the two partial reactions clearly
demonstrate that aminopropylflavone specifically targets the
N-terminal domain active site.
Aminoalkylflavone (8) Enzyme Specificity. The final
question that we addressed concerns whether the amino-
alkyldihydroxyflavones discriminate between the ATP binding
site in PPDK and the ATP binding sites in other ATP-
dependent enzymes. Although the PPDK ATP binding site is
EXPERIMENTAL SECTION
General Methods. All reactions were run under a nitrogen
■
atmosphere. Unless otherwise noted, all reagents were obtained from
commercial sources and used without further purification. H and 13
C
1
NMR spectra were recorded on CDCl3 solutions unless otherwise
specified, and chemical shifts are reported in ppm relative to residual
CHCl3 at 7.24 ppm (for 1H NMR) and 77.0 ppm (for 13C NMR). 13
C
NMR resonance assignments were aided by the use of the DEPT
technique to determine numbers of attached hydrogens. Mass spectra
were recorded by using fast atomic bombardment (FAB) or
electrospray (ES) techniques and a TOF instrument.
PPDK Purification and Assay. PPDK was purified from
Escherichia coli JM 101 carrying the plasmid pACYC184-D12 as
described previously.45 The spectrophotometric assay, described
1915
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
Scheme 2
Scheme 3
previously,4 was used to monitor the initial velocity of PPDK-catalyzed
reaction of AMP, PPi, and PEP to ATP, Pi, and pyruvate. The overall
reaction was monitored in the reverse direction (pyruvate formation)
at 340 nm by using the lactate dehydrogenase (20 units/mL)−NADH
(0.2 mM) coupling reaction. The initial velocity (v0) data were
analyzed using eq 1 and the computer programs described by
Cleland.46 The kcat was calculated using eq 2
kcat = vmax/[E]
(2)
where v0 is the initial velocity, vmax is the maximum velocity, [E] is the
enzyme concentration, [S] is the substrate concentration, and Km is
the Michaelis constant. The inhibition constants (Ki) for the inhibitors
were determined from initial velocity data obtained at varying AMP
concentrations and fixed, saturating concentrations of cosubstrates and
cofactors. The initial velocity data were analyzed using eq 3, where Ki
is the inhibition constant and [I] is the inhibitor concentration.
v0 = vmax[E][S]/(Km + [S])
(1)
1916
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
a
v0 = vmax[E][S]/[Km(1 + [I]/Ki) + [S]]
Table 2. PPDK Inhibition Constants of Aminoalkylflavones
(3)
flavone
Ki (μM)
flavone
Ki (μM)
For initial inhibitor screening, single-point diagnostic testing was
carried out on 1 mL assay solution containing 20 μM substrate AMP
(considering both the Km of PPDK (ca. 10 μM) and the lowest
detection limits of the UV spectrometer) and saturating concen-
trations of cosubstrates (0.5 mM PEP, 1 mM PPi) and metal ion
cofactors (5 mM MgCl2 and 40 mM NH4Cl) in 20 mM imidazole (pH
6.8, 25 °C) as described previously.47 The concentrations of inhibitors
tested in the 1 mL assay solution are 50 or 100 μM depending on
compound availability. In each case, the reaction was initiated by the
addition of PPDK.
10
1.6 0.1
7
8
9
2.8 0.1
1.8 0.3
2.6 0.5
11
12
2.7 0.4
18.3 0.1
5,7-dihydroxyflavone
16
2
a
The 1 mL assay solutions initially contained 5−80 μM AMP, 0.5 mM
PEP, 1 mM PPi, 5 mM MgCl2, 40 mM NH4Cl, 0.2 mM NADH, 20
units/mL of lactate dehydrogenase in 20 mM imidazole buffer (pH
6.8).
When a compound showed obvious inhibitory activity (<40 μM), a
more accurate assay was carried out in the manner as described below
for quercetin. A quercetin stock solution was prepared (50 mg in 1 mL
of 1 N NaOH). The stock solution was diluted 100-fold by H2O
addition to reach a concentration of 1.5 mM with a pH of 8. Aliquots
(0, 10, and 20 μL aliquots of the 1.5 mM) of the stock inhibitor
solution were added to 1 mL assay solutions containing 5−80 μM
AMP, 0.5 mM PEP, 1 mM PPi, 5 mM MgCl2, 40 mM NH4Cl, 0.2 mM
NADH, and 20 units/mL of lactate dehydrogenase in 20 mM
imidazole buffer (pH 6.8). In each case, the reaction was initiated by
the addition of 0.012 μM PPDK. The initial velocities of reactions
were measured as a function of AMP concentration in the absence and
the presence of 5,7-dihydroxy-4′-[3-aminoprop-1-yl]flavone 7 (0, 6.4,
9.6, and 12.8 μM), 5,7-dihydroxy-4′-(3-aminobut-1-yl)flavone 8 (0,
4.3, 8.6 μM), 5,7-dihydroxy-4′-(3-aminobut-1-en-1-yl)flavone 12 (0,
39, 58 μM), 5,7-dihydroxy-4′-(3-aminopent-1-yl)flavone 9 (0, 4.2, 6.3
μM), 5,7-dihydroxy-4′-(3-aminohex-1-yl)flavone 10 (0, 3.0, 4.5 μM),
and 5,7-dihydroxy-3′-(3-aminohex-1-yl)flavone 11 (0, 5.8, 7.7 μM).
Evaluation of Kinase and Dehydrogenase Inhibition by 5,7-
Dihydroxy-4′-aminobutylflavone 8. Hexokinase. The initial
velocity of Saccharomyces cerevisiae hexokinase-catalyzed reaction of
glucose and ATP to glucose-6-phosphate and ADP was measured
using the NADP/lactate dehydrogenase coupling system. Reaction
solutions contained Km levels of glucose and ATP, 5 mM MgCl2, 0.4
mM NADP, 2 units/mL glucose-6-phosphate dehydrogenase, 0.03
units/mL hexokinase, and 0 or 130 μM 5,7-dihydroxy-4′-amino-
butylflavone in 50 mM K+HEPES (pH 7.0, 25 °C). The overall
reaction was monitored at 340 nm (ε = 6.2 mM−1 cm−1).
Pyruvate Kinase. The initial velocity of rabbit muscle pyruvate
kinase-catalyzed reaction of PEP and AMP to pyruvate and ATP was
measured using the NADH/lactate dehydrogenase coupling system.
Reaction solutions contained the Km level of PEP, 0.5−10-fold Km
levels of ADP, 5 mM MgCl2, 0.2 mM NADH, 20 unit/mL lactate
dehydrogenase, 40 units/mL pyruvate kinase, and 0, 43, or 86 μM 5,7-
dihydroxy-4′-aminobutylflavone in 50 mM K+HEPES (pH 7.0, 25 °C).
The overall reaction was monitored at 340 nm (ε = 6.2 mM−1 cm−1).
Myokinase. The initial velocity of rabbit muscle myokinase-
catalyzed reaction of ATP and AMP to 2 × ADP was measured using
the pyruvate kinase/PEP-NADH/lactate dehydrogenase coupling
system. Reaction solutions contained Km levels of AMP and ATP, 5
mM MgCl2, 0.2 mM NADH, 1 mM PEP, 20 units/mL lactate
dehydrogenase, 0.1 unit/mL myokinase, and 0 or 130 μM 5,7-
dihydroxy-4′-aminobutylflavone in 50 mM K+HEPES (pH 7.0, 25 °C).
The overall reaction was monitored at 340 nm (ε = 6.2 mM−1 cm−1).
Acetate Kinase. The initial velocity of E. coli myokinase-catalyzed
reaction of ATP and acetate to acetylphosphate and ADP was
measured using the pyruvate kinase/PEP-NADH/lactate dehydrogen-
ase coupling system. Reaction solutions contained Km levels of acetate
and ATP, 5 mM MgCl2, 0.2 mM NADH, 1 mM PEP, 20 units/mL of
lactate dehydrogenase, 0.1 unit/mL of acetate kinase, and 0 or 130 μM
5,7-dihydroxy-4′-aminobutylflavone in 50 mM K+HEPES (pH 7.0, 25
°C). The overall reaction was monitored at 340 nm (ε = 6.2
mM−1 cm−1).
Figure 5. Time courses for [14C]AMP formation in single turnover
PPDK (7.2 μM) catalyzed reactions of 1.8 μM [14C]ATP, 11 mM Pi, 5
mM MgCl2, and 40 mM NH4Cl in 50 mM K+HEPES (pH 7.0, 25 °C)
in the presence of 0 μM ( ), 75 μM ( ), and 150 μM (Δ) of the
aminoalkylflavone 7. The solid lines were generated by time course
simulation using the simulation program KinSimand, the known
kinetic model of which included the microscopic rate constants
governing the steps of ATP + Pi + pyruvate partial reaction7 and the
dissociation constant governing the binding of the inhibitor to the free
enzyme.
□
○
Table 3. Percent Conversions at 600 ms in the Single
Turnover Reaction of PPDK + [14C]PEP → PPDK-P +
[14C]Pyruvate Using 8 μM PPDK, 2 μM [14C]PEP, 5 mM
MgCl2, and 40 mM NH4Cl in 50 mM K+HEPES (pH 7.0, 25
°C) along with 0, 82, and 800 μM of 7
concentration of 7 (μM)
conversion (%)
0
82
63
67
64
800
Table 4. Inhibition Constants of Aminobutylflavone 8
toward Lactate Dehydrogenase and Members of the Protein
Kinase Family
enzyme
Ki (μM)
Ki(enzyme)/Ki(PPDK)
PPDK
1.8 0.3
>129
1
>72
>72
>72
>72
34
lactate dehydrogenase
hexokinase
>129
myokinase
>129
Lactate Dehydrogenase. Reaction solutions contained 100 μM
pyruvate, 200 μM NADH, 0.1 unit/mL lactate dehydrogenase, and 0
or 130 μM 5,7-dihydroxy-4′-aminobutylflavone in 50 mM K+HEPES
(pH 7.0, 25 °C). The overall reaction was monitored at 340 nm (ε =
6.2 mM−1 cm−1).
acetate kinase
>129
pyruvate kinase
61
15
7
3
cAMP dependent protein kinase
8
1917
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
Rapid Quench Analysis of the PPDK-Catalyzed Single
Turnover Reaction E + [14C]ATP + Pi →E−P +[14C]AMP + PPi.
Single-turnover reactions of 7.2 μM wild-type PPDK, 1.8 μM
[14C]ATP, 11 mM Pi, 5 mM MgCl2, and 40 mM NH4Cl in 50 mM
K+HEPES (pH 7.0, 25 °C) were carried out in the presence 0, 75, and
150 μM 5,7-dihydroxy-4′-aminopropylflavone (7) using a rapid
quench device from KinTek Instruments. Reactions were initiated
by mixing 32 μL of 14 μM wild-type PPDK, 0, 147, or 295 μM 5,7-
dihydroxy-4′-aminopropylflavone (7), 9.7 mM MgCl2, and 77.5 mM
NH4Cl in 50 mM K+HEPES (pH 7.0, 25 °C) with 30 μL of 3.7 μM
[14C]ATP and 22.7 mM Pi in 50 mM K+HEPES (pH 7.0, 25 °C) and
then terminated at specified times with 182 μL of 0.6 M HCl. The
protein was removed from the acid quenched samples by
centrifugation through a 500 μL filter (cutoff molecular mass of 30
kDa) purchased from Pall Gelman, Inc. The unconsumed [14C]ATP
and [14C]AMP product were separated by HPLC using a Beckman
Ultrasphere C18 reversed-phase analytical column and 25 mM
KH2PO4, 2.5% triethylamine, and 5% methanol as the mobile phase.
The fractions containing ATP and AMP were analyzed for 14C content
by liquid scintillation counting. The percent conversion was deduced
from the ratio of unconsumed [14C]ATP and [14C]AMP product. The
AMP concentration was calculated by multiplying the initial substrate
concentration by the percent conversion. The time course data of
AMP formation were fit to single exponential equation (eq 4) using
the KaleidaGraph program to yield the rate constants from the single
turnover reactions
Synthesis of Aminoalkylflavones 7−10 and 12 (Scheme 2).
5,7-Dimethoxy-4′-bromoflavone (15). To a solution of methyl 4-
bromobenzoate (13) (9.80 g, 45.5 mmol) and NaH 1.20 g (95%
dispersion, 50 mmol) in DMF (20 mL) at 0 °C was added to a
solution of 2,4-dimethoxy-6-hydroxyacetophenone (14) (3.00 g, 15.2
mmol) in DMF (50 mL). The mixture was stirred at 25 °C for 12 h
and poured onto ice. The pH of the resulting solution was adjusted to
10 by addition of 1 N HCl and extracted with CH2Cl2. The CH2Cl2
extracts were concentrated in vacuo giving a residue, which was
dissolved in CHCl3. The CHCl3 solution was saturated with gaseous
HCl, stirred for 4 h at 25 °C, made basic by the addition of aq
Na2CO3, and extracted with CH2Cl2. The CH2Cl2 extracts were dried
and concentrated in vacuo giving a solid which was crystallized from
DMF to afford 2.90 g (53%) of 3,5-dimethoxy-4′-bromoflavone (15)
1
as white needles: mp 198.6−199.4 °C; H NMR 7.71 (d, 2H, J = 8.6
Hz), 7.61 (d, 2H, J = 8.6 Hz), 6.63 (s, 1H), 6.54 (d, 1H, J = 2.2 Hz),
6.37 (d, 1H, J = 2.2 Hz), 3.93 (d, 2H, J = 19.7 Hz); 13C NMR 177.0,
164.0, 160.7, 159.2, 132.0, 130.2, 127.1, 125.6, 108.9, 96.1, 92.6, 56.2,
55.6; HRMS (FAB) m/z (M)+ calcd for C17H12O479Br 359.9997,
found 360.0011.
5,7-Dimethoxy-4′-[3-(N-phthalimido)prop-1-en-1-yl]flavone
(20). A mixture of 5,7-dimethoxyl-4′-bromoflavone (15) (2.0 g, 5.5
mmol), N-allylphthalimide 16 (1.2 g, 6.0 mmol), P(o-tol)3 (275 mg,
0.3 mmol), and Pd2(dba)3 (400 mg, 1.2 mmol) in 50 mL of
triethylamine was stirred at 110 °C for 12 h, cooled to 25 °C, and
concentrated in vacuo. A solution of the residue in CH2Cl2 and was
washed with H2O, dried, and concentrated in vacuo giving a residue
that was subjected to flash chromatography (silica gel, EtOAc then
EtOH) to give 1.3 g (50%) of 5,7-dimethoxy-4′-(3-N-phthalimido-1-
[B]t = [B]∞{1 − exp(kobst)}
(4)
where [B]t is the product AMP concentration at time t, [B]∞ is the
product AMP concentration at equilibrium, and kobs is the observed
rate constant for the reaction.
1
propen-1-yl)flavone (20) as white needles: mp 238 °C dec; H NMR
7.86 (m, 2H), 7.82 (m, 2H), 7.72 (m, 2H), 7.45 (m, 2H), 6.70 (m,
2H), 6.54 (s, 1H), 6.35 (m, 2H), 4.46 (d, 2H, J = 6.5 Hz), 3.91 (d, 6H,
J = 20.9 Hz); 13C NMR 177.4, 167.9, 164.0, 160.8, 160.1, 139.0, 134.0,
132.5, 130.6, 126.9, 126.0, 125.0, 123.3, 108.8, 95.1, 92.8, 55.4,55.7,
39.5; HRMS (FAB) m/z (M + H)+ calcd for C28H22O6N 468.1447,
found 468.1465.
5,7-Dimethoxy-4′-[3-(N-phthalimido)prop-1-yl]flavone (24).
A mixture of 5,7-dimethoxy-4′-[3-N-(phthalimido)prop-1-en-1-yl]-
flavone (20) (1.6 g, 34 mmol), 10% Pd/C (0.1 g) in 40 mL of
CH2Cl2 was stirred under an H2 atmosphere 25 °C for 12 h, filtered,
and concentrated in vacuo to give 1.5 g (94%) of 24 as a white solid:
1H NMR 7.82 (m, 2H), 7.74 (m, 2H), 7.68 (m, 2H), 7.32 (m, 2H),
6.58 (d, 2H, J = 14.5 Hz), 6.38 (s, 1H), 3.94 (d, 2H, J = 21.0 Hz), 3.77
(t, 2H, J = 7.0 Hz), 2.77 (m, 2H), 2.10 (m, 2H), 13C NMR 177.5,
168.2, 163.9, 160.5, 159.7, 144.6, 133.8, 131.9, 129.0, 128.7, 128.2,
127.8, 125.9, 123.0, 109.0, 108.3, 96.0, 92.7, 56.3, 55.6, 37.5, 32.9, 29.1;
HRMS (M)+ (FAB) m/z (M + H) calcd for C28H23O6N 469.1525,
found 469.1513.
Rapid Quench Analysis of Competitive Inhibition by
Flavones toward the Single Turnover Reaction E + [14C]PEP
→ E−P +[14C]Pyruvate Catalyzed by the Wild-Type PPDK.
Single-turnover reactions of 8 μM wild-type PPDK, 2 μM [14C]PEP, 5
mM MgCl2, and 40 mM NH4Cl in 50 mM K+HEPES (pH 7.0, 25 °C)
were carried in the presence 0, 82, and 800 μM 5,7-dihydroxy-4′-
aminopropylflavone (7) using a rapid quench device from KinTek
Instrument. Reactions were initiated by mixing 32 μL of 15.5 μM wild-
type PPDK, 0, 0.16, or 1.5 mM 5,7-dihydroxy-4′-aminopropylflavone
(7), 9.7 mM MgCl2, and 77.5 mM NH4Cl in 50 mM K+HEPES (pH
7.0, 25 °C) with 30 μL of 4.0 μM [14C]PEP and 22.7 mM Pi in 50 mM
K+HEPES (pH 7.0, 25 °C) and then terminated at 600 ms with 182
μL of 0.6 M HCl. The protein was removed from the acid-quenched
samples by centrifugation through a 500-μL filter (cutoff molecular
mass of 30 kDa) purchased from Pall Gelman, Inc. The unconsumed
[14C]PEP and product [14C]pyruvate were separated by HPLC using a
Beckman Ultrasil anon-exchange column and 0.3 M KH2PO4, 0.6 M
HCl pH = 5.0 as the mobile phase. The fractions containing PEP and
pyruvate were analyzed for 14C radioactivity by liquid scintillation
counting. The percent conversion was deduced from the ratio of
unconsumed [14C]PEP and [14C]pyruvate product.
5,7-Dimethoxy-4′-(3-aminoprop-1-yl)flavone (28). A mixture
of 5,7-dimethoxy-4′-[3-(N-phthalimido)prop-1-yl]flavone (24) (120
mg, 0.26 mmol) in 20 mL of CH2Cl2 containing 1 mL N2H4−H2O at
25 °C was stirred for 12 h and concentrated in vacuo. An aqueous
solution of the residue was extracted with CH2Cl2, acidified to pH 3.0
by the addition of 1 M HCl, and extracted with CH2Cl2. The aqueous
layer was made alkaline by addition of 1 M NaOH and extracted with
CH2Cl2. The CH2Cl2 extracts were concentrated in vacuo to give 65
mg (73%) of 5,7-dimethoxy-4′-[3-aminoprop-1-yl]flavone (28) as a
Synthetic Procedures. General Methods. 1H NMR spectra
were recorded using CDCl3 solutions (unless otherwise noted) at 500
and 250 MHz, and 13C NMR spectra were recorded at 62.9 MHz.
Chemical shifts are reported in ppm relative to internal standards.
Infrared spectral bands are recorded in cm−1. High-resolution mass
spectra were obtained by using electron-impact ionization or FAB. All
reagents and solvents were purchased from commercial sources and
used without further purification unless otherwise specified. 2,4-
Dimethoxy-6-hydroxyacephenone 1448 and phthalimido alkenes 17−
2049 were prepared by using known procedures. All reactions were
carried out under a nitrogen atmosphere. Analytical TLC was
performed on 250 μm silica gel plates. Flash chromatography was
performed by using 200−400 mesh silica gel 60. All synthetic
intermediates were isolated as oils unless otherwise noted and shown
to be >90% pure by 1H and/or 13C NMR analysis. The target
aminoalkyl-flavones used for biochemical analyses were purified by
using HPLC (C18 reversed-phase, H2O−MeOH gradient) and shown
to be comprised of single substances (>95% purity).
1
yellow solid: H NMR 7.78 (d, 2H, J = 7.9 Hz), 7.31 (d, 2H, J = 7.9
Hz), 6.63 (s, 1H), 6.56 (s, 1H), 6.48 (s, 1H), 6.38 (s, 1H), 3.91 (m
2H), 2.71 (m, 4H), 1.78 (m, 2H); 13C NMR 177.6, 163.9, 160.9,
160.8, 159.9, 145.9, 129.1, 129.0, 128.0, 126.0, 109.3, 108.6, 96.1, 92.8,
56.4, 55.7, 41.6, 35.1, 33.2; HRMS (FAB) m/z (M)+ calcd for
C20H22O4N 340.1760, found 340.1751.
5,7-Dihydroxy-4′-(3-aminoprop-1-yl)flavone (7). A slurry of
5,7-dimethoxy-4′-(3-aminoprop-1-yl)flavone (28) (65 mg, 0.19
mmol) and BBr3 (2.0 mL of 1.0 M in CH2Cl2, 2.00 mmol) in 10
mL of dry CHCl3 was stirred at reflux for 12 h, cooled to 25 °C,
quenched by addition of MeOH, and concentrated in vacuo. The
residue was partitioned between CH2Cl2 and 1 M NaOH. The
aqueous layer was washed with CH2Cl2, acidified to pH 7 with 1 M
1918
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
HCl, and concentrated in vacuo to give a residue, which was dissolved
in MeOH containing a large excess of n-butylamine. The resulting
mixture was stirred at 25 °C for 12 h and concentrated in vacuo giving
a residue, which was subjected to HPLC separation (C18 reversed-
phase, H2O−MeOH gradient). Concentration of combined fractions
in vacuo afforded 13 mg (21%) of 5,7-dihydroxy-4′-(3-aminoprop-1-
(15), 1.9 g (8.8 mmol) of 1-buten-4-ylphthalimide 18, 100 mg (0.3
mmol) of P(o-tol)3, and 500 mg (0.5 mmol) of Pd2(dba)3 in 50 mL of
triethylamine was stirred at 110 °C for 12 h, cooled to 25 °C, and
concentrated in vacuo. A solution of the residue in CH2Cl2 was
washed with H2O, dried, and concentrated in vacuo, giving a residue
which was subjected to flash chromatography (silica gel, EtOAc−
EtOH) to give 2.1 g (51%) of 22: 1H NMR 7.75 (m, 2H), 7.67 (d, 2H,
J = 8 Hz), 7.61 (m, 2H), 7.31 (d, 2H, J = 8 Hz), 6.60 (s, 1H), 6.47 (s,
1H), 6.37 (d, 1H, J = 16 Hz), 6.29 (s, 1H), 6.26 (m, 1H), 3.81 (d, 6H,
J = 26), 3.69 (t, 2H, J = 7 Hz), 2.25 (q, 2H, J = 9 Hz), 1.85 (m, 2H);
13C NMR 177.3, 168.2, 164.0, 160.5, 160.1, 159.5, 140.2, 133.7, 131.8,
131.5, 129.5, 129.4, 126.1, 125.8, 122.9, 108.2, 95.9, 92.6, 56.1, 55.5,
37.3, 30.2, 27.6; HRMS (FAB) m/z:(M + H)+ calcd for C30H26O6N
496.1760, found 496.1745.
1
yl)flavone (7) as a yellow solid: H NMR 7.91 (d, 2H, J = 8.3 Hz),
7.42 (d, 2H, J = 8.3 Hz), 6.67 (s, 1H), 6.45 (d, 1H, J= 7.0 Hz), 6.21 (d,
1H, J = 7 Hz), 2.97 (m, 2H), 2.81 (m, 2H), 2.02 (m, 2H); 13C NMR
183.8, 166.5, 165.5, 163.2, 159.5, 146.4, 130.7, 130.3, 127.7, 105.7,
105.5, 100.3, 95.2, 40.3, 33.3, 30.0; HRMS (FAB) m/z (M + H)+ calcd
for C18H18O4N 312.1236, found 312.1246.
5,7-Dimethoxy-4′-[3-(N-phthalimido)but-1-en-1-yl]flavone
(21). A mixture of 1.0 g (2.8 mmol) of 5,7-dimethoxy-4′-bromoflavone
(15), 0.6 g (3.0 mmol) of 1-buten-4-ylphthalimide 17, 100 mg (0.3
mmol) of P(o-tol)3, and 500 mg (0.5 mmol) of Pd2(dba)3 in 50 mL of
triethylamine was stirred at 110 °C for 12 h, cooled to 25 °C, and
concentrated in vacuo. A solution of the residue in CH2Cl2 was
washed with H2O, dried, and concentrated in vacuo, giving a residue
which was subjected to flash chromatography (silica gel, EtOAc−
5,7-Dimethoxy-4′-[3-(N-phthalimido)pent-1-yl]flavone (26).
A solution of 5,7-dimethoxy-4′-[3-(N-phthalimido)pent-1-en-1-yl]-
flavone (22) (1.6 g,3.2 mmol) and 10% Pd/C (0.1 g) in 40 mL of
CH2Cl2 was stirred under 1 atm of H2 at 25 °C for 12 h and filtered.
1
The filtrate was concentrated in vacuo to give 26 (1.2 g, 75%): H
NMR 7.80 (d, 2H, J = 2.85 Hz), 7.73 (m, 2H), 7.67 (d, 2H, J = 2.2
Hz), 7.25 (d, 2H, J = 9.5 Hz), 6.6 (s, 1H), 6.53 (s, 1H), 6.35 (s, 1H),
3.91 (d, 2H, J = 21.3 Hz), 3.66 (t, 2H, J = 2.1 Hz), 2.65 (t, 2H, J = 7.3
Hz), 1.66 (m, 6H); 13C NMR 176.7, 167.5, 163.2, 159.9, 158.9, 145.4,
133.2, 131.2, 128.2, 128.0.126.8, 125.1, 122.3, 108.1, 107.3, 95.3, 92.1,
55.5, 55.1, 37.0, 34.7, 29.7, 27.5, 25.6 ; HRMS (FAB) m/z (M + H)+
calcd for C30H28O6N 498.1917, found 498.1923.
1
EtOH) to give 0.8 g (61%) of 21: H NMR 7.81 (m, 2H), 7.79 (m,
2H), 7.69 (m, 2H), 7.38 (m, 2H), 6.63 (m, 1H), 6.54 (m, 1H), 6.45
(m, 1H), 6.36 (s, 1H), 6.31 (m, 1H), 3.92 (d, 6H, J = 21 Hz), 3.70 (t,
2H, J = 7 Hz), 2.63 (m, 2H); 13C NMR 177.5, 168.3, 164.0, 160.8,
160.3, 159.8, 140.0, 133.9, 132.0, 131.5, 130.0, 128.7, 126.5, 126.1,
123.2, 108.6, 96.1, 92.8, 56.4, 55.7, 37.3, 32.3; HRMS (FAB) m/z (M
+ H)+ calcd for C29H24O6N 482.1604, found 482.1628.
5,7-Dimethoxy-4′-(3-aminopent-1-yl)flavone (30). A solution
of 5,7-dimethoxy-4′-[3-(N-phthalimido)pent-1-yl]flavone (26) (900
mg, 1.45 mmol) and 1 mL of NH2NH2·H2O in CH2Cl2 (20 mL) was
stirred at 25 °C for 12 h and concentrated in vacuo. A CH2Cl2 solution
of the residue was extracted with aqueous acid (pH 3.0). The aqueous
solution was made basic and extracted with CH2Cl2. The CH2Cl2
extract was concentrated in vacuo to give 30 (526 mg, 79%): 1H NMR
7.77 (m, 2H), 7.28 (m, 2H), 6.63 (s, 1H), 6.54 (d, 1H, J = 1.85 Hz),
6.35 (s, 1H), 3.91 (d, 2H, J = 18.8 Hz), 2.68 (m, 2H), 2.47 (t, 2H, J =
7.6 Hzd), 1.64 (m, 4H), 1.45 (m, 2H), 1.37 (m, 2H); 13C NMR 177.4,
163.8, 160.6, 159.6, 146.2, 128.7, 127.4, 125.8, 125.7, 109.0, 108.1,
95.9, 92.6, 56.1, 55.5, 55.0, 41.8, 35.0, 32.8, 30.2, 26.3; HRMS (FAB)
m/z (M + H)+ calcd for C22H26O4N 368.1862, found 368.1847.
5,7-Dihydroxy-4′-(3-aminopent-1-yl)flavone (9). A mixture of
5,7-dimethoxy-4′-[3-aminopent-1-yl]flavone (30) (50 mg, 0.14 mmol)
and BBr3 (1 mL of 1.0 M in CH2Cl2, 1.00 mmol) in 10 mL of CHCl3
was stirred at reflux for 12 h, cooled, diluted with MeOH, and
concentrated in vacuo. The residue was partitioned between CH2Cl2
and 1 M NaOH. The aqueous layer was separated, washed with
CH2Cl2, and acidified to pH 7 by addition of 1 M HCl. The solution
was concentrated to give a residue, which was dissolved in MeOH and
diluted by addition of butylamine. The mixture was stirred at 25 °C for
12 h and concentrated in vacuo giving a residue which was subjected
to HPLC (ultrasphere C18 reversed-phase column and H2O−MeOH)
to afford 15 mg of 9 (32%): 1H NMR (CD3OD) 7.70 (d, 2H, J = 8.0
Hz), 7.40 (d, 2H, J = 8.0 Hz), 6.70 (s, 1H), 6.48 (s, 1H), 6.23 (s, 1H),
2.92 (t, 2H, J = 7.5 Hz), 2.75 (d, 2H, J = 7.5 Hz), 1.72 (m, 4H),
1.45(m, 2H); 13C NMR (CD3OD) 183.9, 166.2, 165.8, 163.3, 159.5,
148.3, 130.3, 127.6, 105.5, 100.3, 95.1, 40.7, 36.4, 31.7, 28.9, 27.0;
HRMS (FAB) m/z (M + H)+ calcd for C20H22O4N 340.1549, found
340.1562.
5,7-Dimethoxy-4′-[3-(N-phthalimido)hex-1-en-1-yl]flavone
(23). A mixture of 1.6 g (4.4 mmol) of 5,7-dimethoxy-4′-bromoflavone
(14), 1.2 g (5.2 mmol) of 1-buten-4-ylphthalimide 19, 100 mg (0.3
mmol) of P(o-tol)3, and 500 mg (0.5 mmol) of Pd2(dba)3 in 50 mL of
triethylamine was stirred at 110 °C for 12 h, cooled to 25 °C, and
concentrated in vacuo. A solution of the residue in CH2Cl2 was
washed with H2O, dried, and concentrated in vacuo, giving a residue
which was subjected to flash chromatography (silica gel, EtOAc−
EtOH) to give 1.2 g (51%) of 23: 1H NMR 7.81 (m, 2H), 7.76 (d, 2H,
J = 8.4 Hz), 7.69 (m, 2H), 7.40 (d, 2H, J = 8.4 Hz), 6.63 (s, 1H), 6.54
(d, 1H, J = 2.1 Hz), 6.40 (d, 1H, J = 15.9 Hz), 6.35 (s, 1H), 6.33 (m,
1H), 3.91 (d, 6H, J = 20.9 Hz), 3.70 (t, 2H, J = 7.2 Hz), 2.28 (q, 2H, J
= 7.2 Hz), 1.72 (m, 2H), 1.55(m, 2H); 13C NMR 177.5, 168.3, 163.9,
5,7-Dimethoxy-4′-[3-(N-phthalimido)but-1-yl]flavone (25). A
solution of 5,7-dimethoxy-4′-[3-(N-phthalimido)but-1-en-1-yl]flavone
(21) (0.8 g, 1.7 mmol) and 10% Pd/C (0.1 g) in 40 mL of CH2Cl2
was stirred under 1 atm of H2 at 25 °C and filtered. The filtrate was
concentrated in vacuo to give 25 (0.7 g, 87%): 1H NMR 7.81 (m, 2H),
7.75 (m, 2H), 7.68 (m, 2H), 7.27 (m, 2H), 6.3 (s, 1H), 6.53 (s, 1H),
6.35 (s,1H), 3.91 (d, 2H, J = 32 Hz), 3.70 (t, 2H, J = 6 Hz), 2.71 (t,
2H, J = 7.2 Hz), 1.71 (m, 4H); 13C NMR 177.6, 168.2, 163.9, 160.7,
159.7, 145.6, 133.8, 131.9, 128.9, 125.9, 123.0, 108.3, 96.1, 92.7, 56.5,
55.6, 37.5, 35.6, 28.1, 27.9; HRMS (FAB) m/z (M + H)+ calcd for
C29H26O6N 484.1760, found 484.1776.
5,7-Dimethoxy-4′-(3-aminobut-1-yl)flavone (29). A solution
of 5,7-dimethoxy-4′-[3-(N-phthalimido)but-1-yl]flavone (25) (700
mg, 1.45 mmol) and 1 mL of NH2NH2·H2O in CH2Cl2 (20 mL)
was stirred at 25 °C for 12 h and concentrated in vacuo. A CH2Cl2
solution of the residue was extracted with aqueous acid (pH 3.0). The
aqueous solution was made basic and extracted with CH2Cl2. The
CH2Cl2 extract was concentrated in vacuo to give 29 (450 mg, 88%):
1H NMR 7.54 (d, 2H, J = 7.8 Hz), 7.0 (d, 2H, J = 7.8 Hz), 6.40 (s,
1H), 6.32 (s, 1H), 6.12 (s, 1H), 3.70 (d, 2H, J = 17 Hz), 2.53 (t, 2H, J
= 6.7 Hz), 2.47 (t, 2H, J = 7.6 Hz), 1.49 (m, 2H), 1.31 (m, 2H); 13C
NMR 177.7, 163.9, 160.9, 159.9, 146.2, 129.0, 128.6, 127.5, 109.3,
108.5, 96.1, 92.8, 56.4, 55.7, 41.7, 35.6, 33.3, 28.4; HRMS (FAB) m/z
(M + H)+ calcd for C21H24O4N 354.1705, found 354.1696.
5,7-Dihydroxy-4′-(3-aminobut-1-yl)flavone (8). A mixture of
5,7-dimethoxy-4′-[3-aminobut-1-yl]flavone (29) (50 mg, 0.14 mmol)
and BBr3 (1 mL of 1.0 M in CH2Cl2, 1.00 mmol) in 10 mL of CHCl3
was stirred at reflux for 12 h, cooled, diluted with MeOH, and
concentrated in vacuo. The residue was partitioned between CH2Cl2
and 1 M NaOH. The aqueous layer was separated, washed with
CH2Cl2, and acidified to pH 7 by addition of 1 M HCl. The solution
was concentrated to give a residue, which was dissolved in MeOH and
diluted by addition of butylamine. The mixture was stirred at 25 °C for
12 h and concentrated in vacuo giving a residue which was subjected
to HPLC (ultrasphere C18 reversed-phase column and H2O−MeOH)
1
to afford 19 mg of 8 (31%): H NMR (CD3OD) 7.89 (d, 2H, J = 8
Hz), 7.40 (d, 2H, J = 8 Hz), 6.69 (s, 1H), 6.46 (s, 1H), 6.21 (s, 1H),
3.21 (m, 2H), 2.96 (m, 4H); 2.77 (m, 2H); 13C NMR (CD3OD)
183.9, 166.2, 165.7, 163.3, 159.5, 147.7, 130.4, 127.9, 105.5, 100.3,
95.1, 40.6, 36.0, 28.9, 28.1; HRMS (FAB) m/z (M + H)+ calcd for
C19H20O4N 326.1318, found 326.1379.
5,7-Dimethoxy-4′-[3-(N-phthalimido)pent-1-en-1-yl]flavone
(22). A mixture of 3.0 g (8.3 mmol) of 5,7-dimethoxy-4′-bromoflavone
1919
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
160.7, 160.4, 159.7, 140.5, 133.8, 132.6,132.0,129.5, 129.4, 126.3,
126.0, 125.8, 123.0, 109.1, 108.3, 96.0, 92.7, 56.3, 55.6, 37.6; 32.5, 28.0,
26.2; HRMS (FAB) m/z (M + Na)+ calcd for C33H31O6NNa
538.2230, found 538.2243.
which was subjected to HPLC (ultrasphere C18 reversed-phase column
and H2O−MeOH) to afford 28 mg of 12 (31%): 1H NMR (CD3OD)
7.89 (d, 2H, J = 37.8 Hz), 7.55 (d, 2H, J = 22.5 Hz), 6.61 (m, 2H),
6.42 (m, 2H), 6.18 (s, 1H), 2.91 (t, 2H, J = 7.3 Hz), 2.63 (m, 2H); 13C
NMR (CD3OD) 183.7, 166.2, 165.1, 163.2, 159.4, 142.0, 133.8, 131.2,
128.4, 128.0, 127.6, 105.7, 105.6, 100.3, 95.1, 40.2, 32.1; HRMS (FAB)
m/z (M + H)+ calcd for C19H18O4N 324.1236, found 324.1208.
Synthesis of Aminoalkylflavone 11 (Scheme 3). 5,7-
Dimethoxy-3′-bromoflavone (34). To a solution of methyl 3-
bromobenzoate (33) (9.80 g, 45.5 mmol) and NaH 1.20 g (95%
dispersion, 50 mmol) in DMF (20 mL) at 0 °C was added a solution
of 2,4-dimethoxy-6-hydroxyacetophenone 13 (3.00 g, 15.2 mmol) in
DMF (50 mL). The mixture was stirred at 25 °C for 12 h and poured
onto ice. The pH of the resulting solution was adjusted to 10 by
addition of 1 N HCl and extracted with CH2Cl2. The CH2Cl2 extracts
were concentrated in vacuo giving a residue, which was dissolved in
CHCl3. The CHCl3 solution was saturated with gaseous HCl, stirred
for 4 h at 25 °C, made basic by the addition of aq Na2CO3, and
extracted with CH2Cl2. The CH2Cl2 extracts were dried and
concentrated in vacuo giving a solid which was crystallized from
DMF to afford 2.28 g (42%) of 3,5-dimethoxy-3′-bromoflavone (34)
as white solid: 1H NMR 7.99 (s, 1H), 7.73 (d, 2H, J = 7.8 Hz), 7.59 (d,
1H, J = 7.95 Hz), 7.35 (m, 1H), 6.62 (s, 1H), 6.55 (d, 1H, J = 1.9 Hz),
6.35 (d, 1H, J = 1.7 Hz), 3.91 (d, 2H, J = 15.0 Hz); 13C NMR 176.0,
163.3, 163.0, 159.8, 158.7, 157.7,133.1, 132.4, 129.7128.1, 127.7, 124.9,
123.5, 122.2, 108.4,108.1, 95.5, 92.0, 55.5, 55.1; HRMS (FAB) m/z
(M)+ calcd for C17H13O479Br 361.0075, found 361.0073.
5,7-Dimethoxy-3′-[3-(N-phthalimido)hex-1-yl]flavone (35). A
mixture of 1.6 g (4.4 mmol) of 5,7-dimethoxy-4′-bromoflavone (34),
1.2 g (5.2 mmol) of 1-hexen-4-ylphthalimide (19), 100 mg (0.3 mmol)
of P(o-tol)3, and 500 mg (0.5 mmol) of Pd2(dba)3 in 50 mL of
triethylamine was stirred at 110 °C for 12 h, cooled to at 25 °C, and
concentrated in vacuo. A solution of the residue in CH2Cl2 was
washed with H2O, dried, and concentrated in vacuo. The residue
containing of crude 5,7-dimethoxy-3′-[3-(N-phthalimido)hex-1-en-1-
yl]flavone was used for the next step without purification. A 40 mL
volume of CH2Cl2 was added to the above residue followed by adding
10% Pd/C (0.1 g). The mixture was stirred under 1 atom of H2 at 25
°C for 12 h and filtered. The filtrate was concentrated in vacuo. The
residue was subjected to flash chromatography (silica gel, EtOAc−
EtOH) to give 35 (1.2 g, 53%): 1H NMR 7.80 (m, 2H), 7.67 (m, 2H),
7.48 (m, 1H), 7.36(m, 1H), 7.28 (m, 1H), 6.65 (m, 1H), 6.57 (s, 1H),
6.35 (s, 1H), 3.91 (d, 2H, J = 17.7 Hz), 3.67 (tm, 4H), 2.64 (m, 2H),
1.65 (m, 4H), 1.38 (m, 2H); 13C NMR 177.6, 168.4, 164.0, 160.9,
160.0, 143.5, 133.8, 132.1,131.5, 130.8, 128.9, 128.8, 125.9, 123.7,
109.0, 96.2, 92.9, 56.4, 55.8, 37.9, 35.8, 31.1, 28.7, 28.4, 26.6.
5,7-Dihydroxy-3′-(3-aminohex-1-yl)flavone (11). A solution of
5,7-dimethoxy-4′-[3-(N-phthalimido)hex-1-yl]flavone (35) (500 mg,
0.98 mmol) and 1 mL of NH2NH2·H2O in CH2Cl2 (20 mL) was
stirred at 25 °C for 12 h and concentrated in vacuo. A CH2Cl2 solution
of the residue was extracted with aqueous acid (pH 3.0). The aqueous
solution was made basic and extracted with CH2Cl2. The CH2Cl2
extract was concentrated in vacuo giving residue containing crude 5,7-
dimethoxy-3′-(3-aminohex-1-yl)flavone, which was used for the next
step without purification. A mixture of the above crude and BBr3 (1
mL of 1.0 M in CH2Cl2, 1.00 mmol) in 10 mL of CHCl3 was stirred at
reflux for 12 h, cooled, diluted with MeOH, and concentrated in
vacuo. The residue was partitioned between CH2Cl2 and 1 M NaOH.
The aqueous layer was separated, washed with CH2Cl2, and acidified
to pH 7 by addition of 1 M HCl. The solution was concentrated to
give a residue, which was dissolved in MeOH and diluted by addition
of n-butylamine. The mixture was stirred at 25 °C for 12 h and
concentrated in vacuo giving a residue which was subjected to HPLC
(ultrasphere C18 reversed-phase column and H2O−MeOH) to afford
58 mg of 11 (17%): 1H NMR (CD3OD) 7.95 (m, 1H), 7.86 (m,1H),
7.62 (m, 1H), 7.44 (m, 1H) ; 6.76 (d, 1H, J = 12.2 Hz), 6.21 (s, 1H),
6.26 (s, 1H), 2.99 (t, 2H, J = 7.5 Hz), 2.92 (t, 2H, J = 7.9 Hz), 2.77 (t,
2H, J = 7.4 Hz), 2.36 (m, 2H), 1.75 (m, 2H), 1.65 (m, 2H); 13C NMR
(CD3OD) 183.9, 166.3, 163.3, 159.6, 133.3, 130.3, 127.3, 125.0, 105.5,
5,7-Dimethoxy-4′-[3-(N-phthalimido)hex-1-yl]flavone (27). A
solution of 5,7-dimethoxy-4′-[3-(N-phthalimido)hex-1-en-1-yl]flavone
(23) (1.0 g, 2.0 mmol) and 10% Pd/C (0.1 g) in 40 mL of CH2Cl2
was stirred under 1 atom of H2 at 25 °C for 12 h and filtered. The
1
filtrate was concentrated in vacuo to give 27 (1.2 g, 82%): H NMR
7.70 (m, 4H), 7.73 (m, 2H), 7.25 (m, 2H), 6.65 (m, 1H), 6.59 (m,
1H), 6.36 (m, 1H), 3.92 (d, 2H, J = 20.3 Hz), 3.66 (t, 2H, J = 7.3 Hz),
2.64 (t, 2H, J = 7.7 Hz), 1.66 (m, 4H), 1.46 (m, 2H), 1.23(m,2H); 13C
NMR 177.1, 167.9, 163.6, 160.4, 160.2, 159.3, 146.0, 133.4, 131.5,
130.8, 128.5, 128.2, 127.8, 125.4, 122.6, 108.5, 108.2, 107.7, 95.7, 92.4,
55.9, 55.3, 37.4, 35.2, 30.4, 29.2, 28.0, 26.2; HRMS (FAB) m/z (M +
H)+ calcd for C31H31O6N 512.2073, found 512.2080.
5,7-Dimethoxy-4′-(3-aminohex-1-yl)flavone (31). A solution
of 5,7-dimethoxy-4′-[3-(N-phthalimido)hex-1-yl]flavone (27) (800
mg, 1.6 mmol) and 1 mL of NH2NH2·H2O in CH2Cl2 (20 mL)
was stirred at 25 °C for 12 h and concentrated in vacuo. A CH2Cl2
solution of the residue was extracted with aqueous acid (pH 3.0). The
aqueous solution was made basic and extracted with CH2Cl2. The
CH2Cl2 extract was concentrated in vacuo to give 31 (428 mg, 72%):
1H NMR 7.74 (d, 2H, J = 7.0 Hz), 7.25 (m, 2H), 6.62 (s, 1H), 6.54 (d,
1H, J = 1.0 Hz), 6.34 (s, 1H), 3.90 (d, 2H, J = 19.8 Hz), 2.63 (m, 4H),
1.67 (m, 2H), 1.52 (m, 2H), 1.33 (m, 4H); 13C NMR 177.5, 163.9,
160.7, 159.7, 146.4, 128.8, 125.8, 109.1, 108.2, 96.0, 92.7, 56.2, 55.6,
41.4, 35.6, 32.2, 31.0, 28.9, 26.5; HRMS (FAB) m/z (M + H)+ calcd
for C23H28O4N 382.2018, found 382.2010.
5,7-Dihydroxy-4′-(3-aminohex-1-yl)flavone (10). A mixture of
5,7-dimethoxy-4′-(3-aminohex-1-yl)flavone (31) (50 mg, 0.13 mmol)
and BBr3 (1 mL of 1.0 M in CH2Cl2, 1.00 mmol) in 10 mL of CHCl3
was stirred at reflux for 12 h, cooled, diluted with MeOH, and
concentrated in vacuo. The residue was partitioned between CH2Cl2
and 1 M NaOH. The aqueous layer was separated, washed with
CH2Cl2, and acidified to pH 7 by addition of 1 M HCl. The solution
was concentrated to give a residue, which was dissolved in MeOH and
diluted by addition of n-butylamine. The mixture was stirred at 25 °C
for 12 h and concentrated in vacuo giving a residue which was
subjected to HPLC (ultrasphere C18 reversed-phase column and
H2O−MeOH) to afford 15 mg of 10 (32%): 1H NMR (CD3OD) 7.70
(m, 2H), 7.39 (d, 2H, J = 7.9 Hz), 6.69 (s, 1H), 6.48 (s, 1H), 6.23 (s,
1H), 2.93 (t, 2H, J = 7.2 Hz), 2.73 (t, 2H, J = 7.5 Hz), 1.70 (m, 4H),
1.33(m, 2H), 1.00(m, 2H); 13C NMR (CD3OD) 183.8, 166.1, 165.7,
163.2, 159.4, 148.6, 130.3, 129.9, 127.4, 105.5, 105.4, 100.3, 95.1, 40.7,
36.6, 31.9, 29.7, 28.5, 27.3.
5,7-Dimethoxy-4′-(3-aminobut-1-en-1-yl)flavone (32). A sol-
ution of 5,7-dimethoxy-4′-[3-(N-phthalimido)but-1-en-1-yl]flavone
(21) (300 mg, 0.62 mmol) and 1 mL of NH2NH2·H2O in CH2Cl2
(20 mL) was stirred at 25 °C for 12 h and concentrated in vacuo. A
CH2Cl2 solution of the residue was extracted with aqueous acid (pH
3.0). The aqueous solution was made basic and extracted with CH2Cl2.
The CH2Cl2 extract was concentrated in vacuo to give 32 (162 mg,
1
74%): H NMR 7.78 (d, 2H, J = 7.8 Hz), 7.44 (d, 2H, J =7.8 Hz),
6.64(s, 1H), 6.65 (s, 1H), 6.48 (d, 1H, J = 15.8 Hz), 6.36 (s, 1H), 6.32
(m, 1H), 3.92 (d, 2H, J = 20.2 Hz), 2.86 (t, 2H, J= 6.0 Hz), 2.39 (m,
2H); 13C NMR 177.6, 164.0, 161.0,160.3, 159.8, 140.3, 129.8, 129.0,
126.4, 126.1, 109.3, 108.6, 96.1, 92.8, 56.4, 55.7, 41.5, 37.4; HRMS
(FAB) m/z (M + H)+ calcd for C21H22O4N 352.1549, found
352.1559.
5,7-Dihydroxy-4′-(3-aminobut-1-en-1-yl)flavone (12). A mix-
ture of 5,7-dimethoxy-4′-(3-aminobut-1-en-1-yl)flavone (32) (100 mg,
0.28 mmol) and BBr3 (1 mL of 1.0 M in CH2Cl2, 1.00 mmol) in 10
mL of CHCl3 was stirred at reflux for 12 h, cooled, diluted with
MeOH, and concentrated in vacuo. The residue was partitioned
between CH2Cl2 and 1 M NaOH. The aqueous layer was separated,
washed with CH2Cl2, and acidified to pH 7 by addition of 1 M HCl.
The solution was concentrated to give a residue, which was dissolved
in MeOH and diluted by addition of butylamine. The mixture was
stirred at 25 °C for 12 h and concentrated in vacuo giving a residue
1920
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
106.0, 100.3, 95.2, 40.7, 36.6, 32.3, 29.7, 28.5, 27.3; HRMS (FAB) m/z
(M + H)+ calcd for C21H24O4N 354.1705, found 354.1713.
(3) Moskovitz, B. R.; Wood, H. G. J. Biol. Chem. 1978, 253, 884−
888.
Synthesis of (E)-3[4-(Morpholin-4-yl)benzylidenyl]indoline.
To a solution of 7.5 mL of dimethylformamide in 25 mL of anhydrous
1,2-dichloroethane was added dropwise 5 mL (50 mmol) of
phosphorus oxychloride at 0 °C. The mixture was stirred for 30 min
at room temperature and cooled to 0 °C. 4-Phenylmorpholine (8.3 g,
51 mmol) was added portionwise over 15 min, and the mixture was
stirred at reflux for 48 h. Then 1.25 mL of triethylamine was added,
and the solution was stirred for another 24 h and poured into ice-cold
1 N sodium hydroxide solution. The pH was adjusted to 10, and the
resulting mixture was stirred at room temperature for 1 h. The organic
layer was separated and extracted with dichloromethane. The
combined organic layer was washed with NaCl solution until pH =
7, dried over anhydrous sodium sulfate, and concentrated in vacuo
giving a residue that was subjected to silica gel column
chromatography, eluting with a mixture of ethyl acetate and hexane,
to afford 5.6 g (58%) of 4-(morpholin-4-yl)benzaldehyde as a white
(4) Wang, H. C.; Ciskanik, L.; Dunaway-Mariano, D. Biochemistry
1988, 27, 625−633.
(5) Thrall, H.; Mehl, A. F.; Carroll, L. J.; Dunaway-Mariano, D.
Biochemistry 1993, 32, 1803−1809.
(6) Goss, N. H.; Evans, C. T.; Wood, H. G. Biochemistry 1980, 19
(25), 5805−5809.
(7) Dunaway-Mariano, D. Methods Enzymol. 1999, 308, 149−176.
(8) Carroll, L. J.; Xu, Y.; Thrall, S. H.; Martin, B. M.; Dunaway-
Mariano, D. Biochemistry 1994, 33, 1134−1142.
(9) Xu, Y.; McGuire, M.; Martin, B. M.; Dunaway-Mariano, D.
Biochemistry 1995, 34, 2195−2202.
(10) Thrall, S. A.; Dunaway-Mariano, D. Biochemistry 1994, 33,
1103−1107.
(11) Herzberg, O.; Chen, C. C. H.; Kapadia, G.; McGuire, M.;
Carroll, L. J.; Noh, S. J.; Dunaway-Mariano, D. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 2652−2657.
1
solid: H NMR (500 MHz, CDCl) 9.74 (s, 1H), 7.72 (m, 2H), 6.86
(12) Galperin, M. Y.; Koonin, E. V. Protein Sci. 1997, 6 (12), 2639−
2643.
(m, 2H), 3.80 (m, 4H), 3.29 (t, 4H, J = 4.0 Hz) ppm. The
spectroscopic data match those reported by Sun et al.50
(13) Lim, K.; Read, R. J.; Chen, C. C.; Tempczyk, A.; Wei, M.; Ye,
D.; Wu, C.; Dunaway-Mariano, D.; Herzberg, O. Biochemistry 2007,
46, 14845−14853.
A mixture of 2 g (15 mmol) of oxindole, 3 g (15 mmol) of 4-
(morpholin-4-yl)benzaldehyde, and 2 mL of piperidine in 10 mL of
ethanol was stirred at 90 °C for 24 h. The precipitate formed on
cooling was separated by filtration, washed with cold ethanol, and
dried to yield 3.1 g (68%) of (E)-3[4-(morpholin-4-yl)benzylidenyl]-
(14) Hatch, M. D.; Slack, C. R. Biochem. J. 1969, 112, 549−558.
(15) Hrdy, I.; Mertens, E.; Nohynkova, E. Exp. Parasitol. 1993, 76,
438−441.
1
indoline as a yellow solid: H NMR (250 MHz, DMSO) 10.50 (s,
(16) Raverdy, S.; Foster, J. M.; Roopenian, E.; Carlow, C. K. Mol.
Biochem. Parasitol. 2008, 160, 163−166.
1H), 7.73 (d,1H, J = 7.9 Hz), 7.63 (d, 2H, J = 8.6 Hz), 7.53 (s, 1H),
7.18 (t, 1H, J = 7.6 Hz), 7.02 (d, 2H, J = 8.8 Hz), 6.87 (t, 2H, J = 7.4
Hz), 3.72 (t, 4H, J = 5.0 Hz), 3.24 (t, 4H, J = 4.7) ppm. The
spectroscopic data match those reported in Sun et al.50
Flavone and Quercetin-3,5,7,3′,4′-pentamethyl Ether. Fla-
vone and quercetin-3,5,7,3′,4′-pentamethyl ether were prepared by
using the procedures described by Kim et al.51 and Picq et al.,52
respectively.
(17) Slamovits, C. H.; Keeling, P. J. Eukaryotic Cell 2006, 5, 148−
154.
(18) Bringaud, F.; Baltz, D.; Baltz, T. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 7963−7968.
(19) Benziman, M.; Eizen, N. J. Biol. Chem. 1971, 246, 57−61.
(20) Reeves, R. E.; Menzies, R. A.; Hsu, D. S. J. Biol. Chem. 1968,
243, 5486−5491.
(21) Marshall, J. S.; Ashton, A. R.; Govers, F.; Hardham, A. R.
Current Genetics 2001, 40, 73−81.
ASSOCIATED CONTENT
■
S
* Supporting Information
(22) Liapounova, N. A.; Hampl, V.; Gordon, P. M.; K., Sensen, C.
W.; Gedamu, L.; Dacks, J. B. Eukaryotic Cell 2006, 5, 2138−2146.
(23) Coustou, V.; Besterio, S.; Biran, M.; Diolez, P.; Bouchaud, V.;
Voisin, P.; Michels, P. A. M.; Canion, P; Baltz, T.; Bringaud, F. J. Biol.
Chem. 2003, 278, 49625−49635.
1
Inhibition constants for pyruvate phosphate dikinase and H
and 13C NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Crafts-Brandner, S. J.; Salvucci, M. E. Plant Physiol. 2002, 129
(4), 1773−1780.
AUTHOR INFORMATION
■
Corresponding Author
* Tel: 505-277-6390. Fax: 505-277-6202. E-mail: mariano@
unm.edu.
(25) Eisen, J. A.; Nelson, K. E.; Paulsen, I. T.; Heidelberg, J. F.; Wu,
M.; Dodson, R. J.; Deboy, R.; Gwinn, M. L.; Nelson, W. C.; Haft, D.
H.; Hickey, E. K.; Peterson, J. D.; Durkin, A. S.; Kolonay, J. L.; Yang,
F.; Holt, I.; Umayam, L. A.; Mason, T.; Brenner, M.; Shea, T. P.;
Parksey, D.; Nierman, W. C.; Feldblyum, T. V.; Hansen, C. L.; Craven,
M. B.; Radune, D.; Vamathevan, J.; Khouri, H.; White, O.; Gruber, T.
M.; Ketchum, K. A.; Venter, J. C.; Tettelin, H.; Bryant, D. A.; Fraser,
C. M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (14), 9509−9514.
(26) Acosta, H.; Dubourdieu, M.; Quinones, W. A.; Caceres, A.;
Bringaud, F.; Concepcion, J. L. Comp. Biochem. Physiol. B Bichem Mol.
Biol. 2004, 138, 347−356.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support for these studies was provided by a grant
from the NIH (GM36260). This contribution, as well as all
others made by P.S.M. in the past, have benefited from the
excellent and broad educational background that he received as
a student in the late Professor Howard E. Zimmerman’s
laboratory at the University of WisconsinMadison. As a
result, this contribution is dedicated to the memory of H.E.Z.
and his numerous ground-breaking studies that helped create
the foundation of the science of organic chemistry.
(27) Saavedra, E.; Encalada, R.; Pineda, E.; Jasso-Chavez, R.;
Moreno-Sanchez, R. FEBS J. 2005, 272, 1767−1783.
(28) Feng, X. M.; Cao, L. J.; Adam, R. D.; Zhang, X. C.; Lu, S. Q.
Biochem. Biophys. Res. Commun. 2008, 367, 394−398.
(29) Lawyer, A. L.; Kelley, S. R.; Allen, J. I. Z. Naturforsch., C: Biosci.
1987, 42 (6), 834−836.
(30) Norman, R. A.; Toader, D.; Feruson, D. Trends Pharmacol. Sci.
2012, 33, 273−278.
REFERENCES
(31) Fawaz, M. V.; Topper, M. E.; Firestine, S. M. Bioorg. Chem.
2011, 39, 185−191.
■
(1) Wood, H. G.; O’Brien, W, E.; Micheales, G. Adv. Enzymol. Relat.
Areas Mol. Biol. 1977, 45, 85−155.
(2) Michaels, G.; Milner, Y.; Moskovitz, B. R.; Wood, H. G. J. Biol.
Chem. 1978, 253, 7656−7661.
(32) Mochalkin, I.; Miller, J. R.; Narasimhan, L.; Thanabal, V.;
Erdman, P.; Cox, P. B.; Prasad, J. V. N. V.; Lightle, S.; Huband, M. D.;
Stover, C. K. ACS Chem. Biol. 2009, 6, 473−483.
1921
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922

The Journal of Organic Chemistry
Article
(33) SaavedraLira, E.; PerezMontfort, R. Arch. Med. Res. 1996, 27,
257−264.
(34) Haines, D. S.; Burnell, J. N.; Doyle, J. R.; Llewellyn, L. E.; Motti,
C. A.; Tapiolas, D. M. J. Agric. Food Chem. 2005, 53, 3856−3862.
(35) Motti, C. A.; Bourguet-Kondracki, M. L.; Longeon, A.; Doyle, J.
R.; Llewellyn, L, E.; Tapiolas, D. M.; Yin, P. Molecules 2007, 12, 1376−
1388.
(36) Motti, C. A.; Bourne, D. G.; Burnell, J. N.; Doyle, J. R.; Haines,
D. S.; Liptrot, C. H.; Llewellyn, L. E.; Ludke, S.; Muirhead, A.;
Tapiolas, D. M. Appl. Environ. Microbiol. 2007, 73, 1921−1927.
(37) Stephen, P.; Vijayan, R.; Bhat, A.; Subbarao, N.; Bamezai, R. N.
J. Comput. Aided Mol. Des. 2008, 22, 647−660.
(38) Cosenza, L. W.; Bringaud, F.; Baltz, T.; Vellieux, F. M. D. J. Mol.
Biol. 2001, 318, 1417−1432.
(39) Sielecki, T. M.; Boylan, J. F.; Benfield, P. A.; Trainor, G. L. J.
Med. Chem. 2000, 43 (1), 1−18.
(40) Bridges, A. J. Chem. Rev. 2001, 101 (8), 2541−2571.
(41) Dumas, J. Expert Opin. Ther. Pat. 2001, 11 (3), 405−429.
(42) Narayana, N.; Diller, T. C.; Koide, K.; Bunnage, M. E.;
Nicolaou, K. C.; Brunton, L. L.; Xuong, N. H.; TenEyck, L. F.; Taylor,
S. S. Biochemistry 1999, 38 (8), 2367−2376.
(43) Norman, B. H.; Shih, C.; Toth, J. E.; Ray, J. E.; Dodge, J. A.;
Johnson, D. W.; Rutherford, P. G.; Schultz, R. M.; Worzalla, J. F.;
Vlahos, C. J. Med. Chem. 1996, 39 (5), 1106−1111. Walker, E. H.;
Pacold, M. E.; Perisic, O.; Stephens, L.; Hawkins, P. T.; Wymann, M.
P.; Williams, R. L. Mol. Cell 2000, 6 (4), 909−919.
(44) Baumgrass, R.; Weiwad, M.; Erdmann, F.; Liu, J. O.;
Wunderlich, D.; Grabley, S.; Fischer, G. J. Biol. Chem. 2001, 276
(51), 47914−47921.
(45) Pocalyko, D. J.; Carroll, L. J.; Martin, B. M.; Babbitt, P. C.;
Dunaway-Mariano, D Biochemistry 1990, 29, 10757.
(46) Cleland, W. W. Methods Enzymol. 1979, 63, 103−138. Cleland,
W. W. Methods Enzymol. 1979, 63, 500−513.
(47) Wei, M.; Li, Z.; Ye, D.; Herzberg, O.; Dunaway-Mariano, D. J.
Biol. Chem. 2000, 275, 41156−41165.
(48) Zembower, D. E.; Zhang, H. J. Org. Chem. 1998, 63, 9300−
9305.
(49) Delugo, G.; Faedda, G.; Gladiali, S. J. Organomet. Chem. 1984,
268, 167−174.
(50) Sun, L.; Tran, N.; Tang, F.; App, H.; Hirth, P.; McMahon, G.;
Tang, C. J. Med. Chem. 1998, 41 (14), 2588−2603.
(51) Kim, K. S. PCT Int. Appl. WO 9813344 A1 19980402, 1998;
CAN 128:270537, AN 208527 5,908, 934.
(52) Picq, M.; Prigent, A. F.; Chabannes, B.; Pacheco, H. Tetrahedron
Lett. 1984, 25 (21), 2227−2230.
1922
dx.doi.org/10.1021/jo3018473 | J. Org. Chem. 2013, 78, 1910−1922