Synthesis and analysis of potential prodrugs of coenzyme A analogues for the inhibition of the histone acetyltransferase p300

Article abstract of DOI:10.1016/S0968-0896(03)00265-7

Lys-CoA (1) is a selective inhibitor of p300 histone acetyltransferase (HAT) but shows poor pharmacokinetic properties because of its multiply charged phosphates. In an effort to overcome this limitation, truncated derivatives of 1 were designed, synthesized and tested as p300HAT inhibitors as well as substrates for the CoA biosynthetic bifunctional enzyme phosphopantetheine adenylyltransferase-dephospho-CoA kinase (PPAT/DPCK). Lys-pantetheine (3) and Lys-phosphopantetheine (2) showed no detectable p300HAT inhibition whereas 3′-dephospho-Lys-CoA (5) was a modest p300 inhibitor with IC₅₀ of 1.6 μM (compared to IC₅₀ of ～50 nM for 1 blocking p300). Compound 2 was shown to be an efficient substrate for PPAT whereas 5 was a very poor DPCK substrate. Further analysis with 3′-dephospho-Me-SCoA (7) indicated that DPCK shows relatively narrow capacity to accept substrates with sulfur substitution. While these results suggest that truncated derivatives of 1 will be of limited value as lead agents for p300 blockade in vivo, they augur well for prodrug versions of CoA analogues that do not require 3′-phosphate substitution for efficient binding to their targets, such as the GCN-5 related N-acetyltransferases.

Full text of DOI:10.1016/S0968-0896(03)00265-7

Bioorganic & Medicinal Chemistry 11 (2003) 3307–3313
Synthesis and Analysis of Potential Prodrugs of Coenzyme
A Analogues for the Inhibition of the Histone
Acetyltransferase p300
Marek Cebrat,a,b Cheol M. Kim,a Paul R. Thompson,a
Matthew Daughertyc and Philip A. Colea,*
aDepartment of Pharmacology and Molecular Sciences, Johns Hopkins University,
School of Medicine, 725 N. Wolfe St., Baltimore, MD 21205, USA
bFaculty of Chemistry, University of Wroclaw, 50-383 Wroclaw, Poland
cIntegrated Genomics, Incorporated, Chicago, IL 60612, USA
Received 27 January 2003; accepted 21 March 2003
Abstract—Lys-CoA (1) is a selective inhibitor of p300 histone acetyltransferase (HAT) but shows poor pharmacokinetic properties
because of its multiply charged phosphates. In an effort to overcome this limitation, truncated derivatives of 1 were designed, syn-
thesized and tested as p300HAT inhibitors as well as substrates for the CoA biosynthetic bifunctional enzyme phosphopantetheine
adenylyltransferase-dephospho-CoA kinase (PPAT/DPCK). Lys-pantetheine (3) and Lys-phosphopantetheine (2) showed no
detectable p300HAT inhibition whereas 30-dephospho-Lys-CoA (5) was a modest p300 inhibitor with IC50 of 1.6 mM (compared to
IC50 of ꢀ50 nM for 1 blocking p300). Compound 2 was shown to be an efficient substrate for PPAT whereas 5 was a very poor
DPCK substrate. Further analysis with 30-dephospho-Me-SCoA (7) indicated that DPCK shows relatively narrow capacity to
accept substrates with sulfur substitution. While these results suggest that truncated derivatives of 1 will be of limited value as lead
agents for p300 blockade in vivo, they augur well for prodrug versions of CoA analogues that do not require 30-phosphate sub-
stitution for efficient binding to their targets, such as the GCN-5 related N-acetyltransferases.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
While it has been employed for in vitro transcription
studies7 and in cells via microinjection or with the use
of cell permeabilizing agents,8 compound 1 has gen-
erally been ineffective by simple addition to cell culture
media.
CoA derivatives can be powerful and selective enzyme
inhibitors.1ꢁ5 However, since the CoA moiety contains
three phosphate groups which are negatively charged at
neutral pH, CoA containing compounds tend to have
poor pharmacokinetic performance in vivo. Our lab
has recently developed several bisubstrate inhibitors of
the acetyl-CoA-dependent histone acetyltransferases
which contain CoA moieties.5 Histone acetyl-
transferases (HATs) catalyze the transfer of an acetyl
group from acetyl-CoA to the e-amino group of lysines
in histones and other proteins (Fig. 1) and are impor-
tant in the regulation of chromatin remodeling and
gene expression.6 One compound in particular, Lys-
CoA (1) (Fig. 2), has proved useful for blocking the
HAT activity of transcriptional coactivator p300.5
A theoretical approach9 to overcoming this problem is
to develop truncated versions of 1 (compounds 2–5)
which are either p300 inhibitors on their own or
could be capable of being bio-converted into Lys-
CoA (1) inside cells containing the target. The bio-
synthetic pathway of the conversion of pantetheine to
CoASH has long been known10 and recently the
protein responsible for the final two steps in CoASH
production in human cells was isolated and char-
acterized.11 This bifunctional protein, phospho-
pantetheine
adenylyltransferase-dephospho-CoA
kinase (PPAT/DPCK), converts phosphopantetheine
to CoASH (Fig. 3) but its ability to process phos-
phopantetheine analogues has not yet been addressed.
In this study, we describe the design and synthesis of
*Corresponding author. Tel.: +1-410-614-8849; fax: +1-410-614-
7717; e-mail: pcole@jhmi.edu
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00265-7

3308
M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
Results
Synthesis of Lys-pantetheine (3) and analogues
Fmoc-Lys containing Dde (dimethyldioxocyclohex-
ylidene) protection on the e-amino group was first
immobilized on Rink amide resin via standard amide
bond formation. Removal of the Fmoc group with
piperidine was followed by treatment with acetic anhy-
dride, affording the resin bound a-acetamide (Fig. 4).
The Dde group was then removed with hydrazine and
the free e-amino group was reacted with bromoacetic
acid. The derivatized Lys-bromacetyl compound (8) was
then cleaved from the resin with trifluoroacetic acid and
purified by HPLC. The purified compound 8 was trea-
ted with freshly prepared d-pantetheine and the product
Lys-pantetheine (3) purified by HPLC. Phosphorylation
of 3 was carried out in three steps:4b first regio-selective
modification of the less hindered hydroxy group with
dibenzyl isopropyl phospoamidate, oxidation to the
dibenzyl phosphate triester (9), and then acidolysis of
the benzyl groups with trifluoroacetic acid yielded the
target compound Lys-phosphopantetheine (2).
Figure 1. Histone acetyltransferase reaction catalyzed by p300.
several sub-structural analogues (compounds 2–5)
related to Lys-CoA (1), their action against p300HAT
activity, and their properties as substrates of the
human CoASH biosynthetic bifunctional enzyme
PPAT/DPCK.
Figure 2. Lys-CoA (1) and analogues.

M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
3309
Figure 3. PPAT/DPCK catalyzed conversion of phosphopantetheine to CoASH.
Figure 4. Synthetic approach to Lys-phosphopantetheine (2).
A variety of strategies have been used to protect organo-
phosphate compounds in a fashion that allows them to
be cell permeable and for efficient liberation of the
phosphate moiety inside cells by nonspecific esterases.12
We turned our attention to the synthesis of derivative 4
because bis[(pivaloyloxy)methyl] (POM) protection of
phosphate moieties has been effective in enhancing the
action of several published prodrugs.13 Of the several
approaches that we tested, two synthetic methods
proved feasible, although both methods proceeded in
low yields (5.0–7.5%). The first of them utilized the
reaction of Lys-phosphopantetheine (2) with tributyltin
methoxide and iodomethyl pivaloate. The other
involved the reaction of Lys-pantetheine (3) with bis[(-
pivaloyloxy)methyl] hydrogen phosphate and diisopro-
pylcarbodiimide (DIC). Poor yields of the POM
derivative 4 may have resulted from the fact that the
secondary hydroxy group of pantetheine could form an
intramolecular interaction with the phosphoester sys-
tem. Though low-yielding, these methods may provide a
general solution to preparing bio-cleavable CoA pre-
cursors for cellular studies.
5 was synthesized both to evaluate how much the
30-phosphate of 1 contributes to p300 inhibition as
well as to determine the effects of lysyl-substitution on
PPAT/DPCK biotransformation. Finally, other con-
trol compounds for the PPAT/DPCK biotransforma-
tion studies that were prepared in straightforward
fashion include Me-SCoA (6) and 30-dephospho-Me-
SCoA (7).
p300HAT inhibition studies
The Lys-pantetheine analogues 2, 3, and 5 were eval-
uated as p300HAT inhibitors using recombinant
p300HAT domain, radiolabelled acetyl-CoA, and
H4-derived 20 amino acid peptide. Under these condi-
tions, the compound Lys-CoA (1) displayed an IC50
ꢀ50 nM. Both Lys-pantetheine (3) and Lys-phospho-
pantetheine (2) showed IC50 values > >6.4 mM (<10%
inhibition at 6.4 mM) indicating a dramatic loss of inhi-
bition. Interestingly, 30-dephospho-Lys-CoA (5), while
still reasonably potent at blocking p300 with IC50 of 1.6
mM, was about 33-fold weaker at blocking p300 com-
pared to Lys-CoA (1) itself. This behavior stands in
contrast to a bisubstrate analogue inhibitor of serotonin
N-acetyltransferase where the 30-phosphate of the com-
pound was found to be dispensable.4b
An additional compound that we synthesized was
30-dephospho-Lys-CoA (5) which was prepared simply
by reacting 8 with 30-dephospho-CoASH. Compound

3310
M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
PPAT/DPCK biotransformation studies
30-dephospho-Me-SCoA (7) was examined as a DPCK
substrate. This compound was found to be about
15-fold reduced as a DPCK substrate compared to
30-dephospho-CoASH, again highlighting the sensitivity
of DPCK to substrates containing structural substitu-
tion on sulfur. In summary, the PPAT component
shows relatively large tolerance of pantetheinyl sulfur
substitution in substrates whereas DPCK has quite strict
substrate requirements even at this remote sulfur site.
Since it became clear that substructures related to Lys-
CoA (1) show significantly weakened potency in the
blockade of p300 compared to Lys-CoA (1) itself, it was
important to investigate the potential for Lys-phospho-
pantetheine (2) and 30-dephospho-Lys-CoA (5) to be
enzymatically processed by PPAT/DPCK. An HPLC
assay was developed that cleanly separated each of these
three compounds from each other as well as other
reagents in the buffer (Fig. 5). Studies of the PPAT
reaction with Lys-phosphopantheine (2) revealed a
relatively rapid conversion to 30-dephospho-Lys-CoA
Discussion
(5) with a pseudo first-order rate constant of 0.02 sꢁ1
.
Although we were successful in the synthesis of a variety
of Lys-CoA (1) derivatives, it was found that an intact
CoA moiety in 1 is required for its potent blockade of
p300HAT. That removal of the 30-phosphate from 1
results in an apparent 33-fold reduction in affinity sug-
gests that the p300 active site makes specific interactions
with the 30-phosphate within the natural substrate ace-
tyl-CoA. This is interesting because other known HATs
such as the GCN-5 related N-acetyltransferases do not
appear to show a requirement for 30-phosphate in
molecular recognition.4,15 Thus, in addition to known
mechanistic differences with the GNAT superfamily,5c
p300 appears to have a distinct strategy for binding to
the acetyl-CoA substrate. Since CoA binding represents
the most conserved feature of this diverse superfamily,16
this result provides further evidence against a prior
proposal suggesting close structural resemblance of
p300 to the GNATs.17
Compared to published values, this was calculated to be
within 20-fold of the rate measured for the under-
ivatized phosphopanthetheine itself being processed by
PPAT.11,14
In contrast, in the second step, conversion of 30-depho-
spho-Lys-CoA (5) to Lys-CoA (1) by the DPCK activity
was found to be very slow, with only 5% conversion
after 17.5 h in the presence of 2 mM enzyme. To rule out
that the DPCK activity was damaged in our prepara-
tion, we tested this enzyme activity with its natural
substrate, 30-dephospho-CoA. As expected, 30-depho-
spho-CoASH was transformed to CoASH very effi-
ciently, about 53,000-fold faster than the rate of
conversion of 30-dephospho-Lys-CoA (5) to Lys-CoA
(1) under the same conditions, and at a rate in approx-
imate agreement with the literature value.11,14 It was
formally possible that 30-dephospho-Lys-CoA (5) was
acting primarily as a competitive inhibitor rather than
an efficient DPCK substrate. To test this, compound 5
was added to the reaction containing 30-dephospho-
CoASH and DPCK. However, compound 5 did not
significantly block the conversion of 30-dephospho-
CoASH to CoASH in the concentration (500 mM) tested
(data not shown).
It was our design that Lys-phosphopantetheine POM
derivative (4) which was prepared here could be bio-
transformed inside cells, first to the free phosphate and
then ultimately to Lys-CoA (1). While Lys-phospho-
pantetheine (2) proved to be a good PPAT substrate,
the next step in the conversion catalyzed by DPCK was
extremely inefficient. Since 30-dephospho-Lys-CoA (5) is
not a very strong p300 inhibitor, these results suggest
that a Lys-CoA prodrug approach will probably not
lead to the most potent p300 inhibitors in human cells.
For the purposes of cell biology research it is possible
Thus DPCK appeared to be poorly able to accept sub-
strates with lysyl substitution of the thiol moiety of
30-dephospho-CoASH. To further address this,
Figure 5. HPLC trace showing separation of Lys-phosphopantetheine (2) and products 30-dephospho-Lys-CoA (5) and Lys-CoA (1). HPLC con-
ditions: C-18 analytical column, flow rate of 1 mL/min and gradient of potassium phosphate pH 4.5 (A) and methanol (B) as follows, 0–5 min 100%
A, 5–55 min linear gradient to 80% A, UV detection at 214 nm. Retention times are 42, 46, and 54 min for 2, 1, and 5, respectively. See Experimental
for further details.

M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
3311
that overexpression of PPAT/DPCK in cells of interest
could be used to permit an efficient conversion of the
desired reaction. On the other hand, if the lysyl moiety
of 5 could be further modified to enhance inhibitory
potency, it is conceivable that one could sacrifice the
added affinity associated with the DPCK-catalyzed
phosphorylation reaction.
compound was precipitated from the cleavage mixture
with ice-cold diethyl ether, dissolved in water, lyophi-
lized and purified by reversed phase HPLC using a
water/acetonitrile gradient (0.05% trifluoroacetic acid)
with a C-18 column. The yield was 300 mg of the pure
compound 8 obtained after lyophilization. ESI-MS
calcd for C10H18N3O3Br 308.1, found m/z 330.0, 332.0
(Na+ salts, bromine isotopes). 1H NMR (400 MHz,
CD3OD) d 1.41 (m, 2H), 1.54 (m, 2H), 1.65 (m, 1H),
1.80 (m 1H), 1.99 (s, 3H), 3.20 (m, 2H), 4.28 (m, 1H);
13C NMR (100 MHz, CD3OD) d 21.30, 22.96, 27.62,
28.59, 31.54, 39.38, 53.23, 168.32, 172.20, 175.98.
More generally, since many CoA enzymes such as the
GNAT superfamily members do not require the
30-phosphate in substrates or inhibitors to achieve high
affinity, this prodrug strategy may be best directed at
such enzymes. For example, serotonin N-acetyltransfer-
ase is a potential drug target for disorders of circadian
rhythm and is potently blocked by bisubstrate analo-
gues containing a CoA moiety.4 Removal of the
30-phosphate from such a bisubstrate analogue only
shows a 2-fold effect on diminishing inhibitory poten-
cy.4b Thus, appropriately designed prodrug analogues
of these bisubstrate serotonin N-acetyltransferase inhi-
bitors may well show promise at blocking melatonin
production in vivo.
Lys-pantetheine (3). d-Pantetheine (0.46 g) was dis-
solved in 5 mL dioxane:water (4:1) mixture and 0.24 g
of triphenylphosphine was added. The pH was adjusted
to 2 by adding 6 M HCl and the mixture was stirred for
3 h at room temperature Ac-Lys(AcBr)-NH2 (8) (285
mg) dissolved in 5 mL dioxane/water mixture (4:1) was
added, and the pH was adjusted to 9.8 by adding trie-
thylamine. The reaction mixture was stirred for 2.5 h at
room temperature and then lyophilized. The crude mix-
ture was dissolved in water, the insoluble material
removed by filtration, and the solution was subjected to
reversed phase HPLC purification affording 460 mg
pure 3 after lyophilization. FABHRMS (M+H+)
calcd for C21H40N5O7S 506.2648, found m/z 506.2649.
1H NMR (400 MHz, CD3OD) d 0.91 (s, 6H), 1.40 (m,
2H), 1.54 (m, 2H), 1.66 (m, 1H), 1.80 (m, 1H), 1.99 (s,
3H), 2.43 (m, 2H), 2.69 (m, 2H), 3.21 (m, 4H), 3.36–3.49
(bm, 6H), 3.89 (s, 1H), 4.28 (m, 1H); 13C NMR
(100 MHz, CD3OD) d 19.74, 20.19, 21.36, 23.06, 28.80,
31.58, 31.72, 34.74, 35.22, 35.30, 38.50, 39.20, 53.29,
69.15, 76.08, 171.28, 172.20, 172.70, 174.87, 176.00.
Experimental
General
The bifunctional human PPAT/DPCK enzyme and the
p300 catalytic domain (N-terminal His tag, residues
1284–1673) were prepared as recombinant proteins as
described.11,18 Bis[(pivaloyloxy)methyl] hydrogen phos-
phate and iodomethyl pivaloate were synthesized fol-
lowing procedures described elsewhere.13c All other
reagents were commercially available unless otherwise
described. NMR spectra were recorded on a Varian
Mercury 400 MHz instrument, signals are reported in
ppm from tetramethylsilane (s, d, m, b for singlet,
doublet, multiplet and broad, respectively). Electro-
spray ionization MS was carried out on a Psciex single
quadrupole machine. FABHRMS were obtained at the
UC Riverside MS facility.
Lys-phosphopantetheine (2). Dibenzyl diisopropyl phos-
phoamidate (105 mL, 1 equiv) was added slowly to ice-
cold, stirred solution of Lys-pantetheine (3) (150 mg)
and 1H-tetrazole (22 mg, 1.1 equiv) in 3 mL DMF.
After allowing gradual warming to room temperature,
the mixture was stirred for 1 h. The reaction mixture
was cooled down again to 0 ꢂC and 2 equivof 6 M
t-butyl hydroperoxide solution in isooctane was added.
After 2 h, the mixture was concentrated in vacuo and
then purified by reversed phase HPLC to give 120 mg of
pure Lys-phosphopantetheine dibenzyl ester (9).
FABHRMS (M+H+) calcd for C35H53N5O10P1S
766.3251, found m/z 766.3242. 1H NMR (400 MHz,
CD3OD) d 0.89 (s, 3H), 0.95 (s, 3H), 1.39 (m, 2H), 1.52
(m, 2H), 1.64 (m, 1H), 1.79 (m, 1H), 1.98 (s, 3H), 2.41
(m, 2H), 2.68 (m, 2H), 3.19 (m, 4H), 3.37 (m, 2H), 3.46
(m, 2H), 3.84 (m, 2H), 4.00 (m, 1H), 4.28 (m, 1H), 5.05
(d, 4H), 7.36 (s, 10H).
Ac-Lys(AcBr)-NH2 (8). 2.0 g of Rink Amide AM resin
(0.63 mmol/g) was deprotected by 20% piperidine solu-
tion in DMF and treated with 3 equivof Fmoc-
Lys(Dde)-OH in the presence of 3 equivBOP (benzo-
triazole - 1 - yl - oxy - tris - (dimethylamino) - phospho-
niumhexafluorophosphate) and
3 equivof HOBt
(N-hydroxybenzotriazole) in a minimal volume of 0.4 M
solution of N-methylmorpholine (NMM) in DMF for 3
h at room temperature. The Fmoc group was removed
by standard piperidine treatment and the N-terminal
amino group was acetylated by treatment with excess
50% acetic anhydride and 0.4 M NMM (DMF) over 1 h
at room temperature. The Dde group was removed by
reaction with 2% hydrazine (DMF) for 2 h and the e-
amino group was treated with 5 equivof bromoacetic
acid and 5 equivdiisopropylcarbodiimide (DIC) in a
minimal volume of DMF overnight at room tempera-
ture. The product was cleaved from the resin by treat-
ment with a mixture of 95% TFA/2.5% water/2.5%
triisopropylsilane for 2 h at room temperature Crude
Benzyl protection was removed by treatment with 95%
TFA over 3 h at room temperature. The mixture was
concentrated in vacuo and then purified by reversed
phase HPLC to afford 54 mg of the pure product 2.
FABHRMS (M+H+) calcd for C21H41N5O10
586.2312, found m/z 586.2334. 1H NMR (400 MHz,
D2O) d 0.86 (s, 3H), 0.93 (s, 3H), 1.35 (m, 2H), 1.50 (m,
2H), 1.68 (m, 1H), 1.77 (m, 1H), 1.99 (s, 3H), 2.46 (m,
2H), 2.66 (m, 2H), 3.18 (m, 2H), 3.23 (s, 2H), 3.35 (m,

3312
M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
2H), 3.47 (m, 2H), 3.60 (m, 1H), 3.79 (m, 1H), 3.99 (s,
1H), 4.16 (m, 1H).
TFA. Reaction yield of purified 4 was ca. 5%. HRMS
(M+H+) calcd for C33H61N5O14P1S1 814.3673, found
m/z 814.3651. H NMR (400 MHz, C2D5OD) d 0.95 (s,
1
Lys-CoA (1). Compound 8 was dissolved in a small
volume of 0.5 M triethylammonium bicarbonate buffer,
pH 8.2, and treated with 2 equivcoenzyme A dilithium
salt (Sigma), overnight at room temperature. The mix-
ture was diluted with water, lyophilized and purified by
reversed phase HPLC in water/acetonitrile solvent sys-
tem. Analytical data for the compound agreed with
those measured previously. ESI-MS (M+H+) calcd for
C31H53N10O19P3S 994.8, found m/z 996.0. 1H NMR
(400 MHz, D2O) d 0.65 (s, 3H), 0.75 (s, 3H), 1.18 (m,
2H), 1.32 (m, 2H), 1.49 (m, 1H), 1.57 (m, 1H), 1.82 (s,
3H), 2.27 (m, 2H), 2.47 (m, 2H), 3.05 (s, 2H), 3.16 (m,
2H), 3.28 (m, 2H), 3.47 (m, 1H), 3.70 (m, 1H), 3.82 (s,
1H), 3.98 (m, 1H), 4.10 (m, 2H), 4.41 (s, 1H), 4.72 (m,
2H), 5.99 (d, 1H), 8.23 (s, 1H), 8.31 (s, 1H); 13C NMR
(100 MHz, D2O) d 179.82, 177.37, 177.01, 176.58,
175.02, 152.44, 151.11, 147.44, 145.07, 121.11, 90.13,
85.88, 77.06, 76.84, 76.47, 74.85, 71.83, 67.82, 56.31,
49.29, 42.02, 40.92, 38.13, 37.99, 37.44, 33.83, 33.25,
30.51, 25.12, 24.32, 23.47, 21.01.
3H), 1.01 (s, 1H ), 1.22 (s, 18H), 1.39 (m, 2H), 1.54 (m,
2H), 1.66 (m, 1H), 1.79 (m 1H), 1.98 (s, 3H), 2.43 (m,
2H), 2.69 (m, 2H), 3.21 (m, 4H), 3.38 (m, 2H), 3.48 (m,
2H), 4.32 (m, 1H), 5.64 (s, 2H), 5.67 (s, 2H), 7.91 (m,
1H), 8.06 (m, 1H), 8.22 (m, 1H), 8.30 (m, 1H); 31P
NMR (160 MHz, C2D5OD) ꢁ3.68 (s).
Methyl-SCoA (6). Gasous methyl bromide was bubbled
through a solution of CoASH dilithium salt in 0.5 M
triethylammonium bicarbonate buffer, pH 8.1, for 30
min. The reaction mixture was stirred for another 1.5 h,
flash-frozen and lyophilized. Compound 6 and S-dime-
thyl-coenzyme A formed during the reaction were easily
separated by reversed-phase HPLC in water/acetonitrile
system containing 0.05% TFA and the known19 com-
pound 6 was obtained in high purity. ESI-MS (M+H+)
1
calcd for C22H38N7O16P3S 781.1, found m/z 782.0. H
NMR (400MHz, D2O) d 0.65 (s, 3H, NOE coupling with
s at 0.78 and s at 3.86), 0.78 (s, 3H), 1.91 (s, 3H, NOE
coupling with m at 2.45, m at 3.20, and m at 3.31), 2.31 (m,
2H, NOE coupling with m at 3.31), 2.45 (m, 2H, NOE
coupling with m at 3.20), 3.20 (m, 2H), 3.31 (m, 2H), 3.45
(d, 1H), 3.71 (d, 1H), 3.86 (s, 1H), 4.10 (m, H), 4.44 (s,
1H), 4.73 (m, 2H), 6.06 (d, 1H), 8.28 (s, 1H), 8.51 (s, 1H).
30-Dephospho-Lys-CoA (5). It was synthesized following
the same procedure as for 1 but employed 30-depho-
spho-CoASH in place of CoASH. HRMS (M+H+)
calcd for C31H53N10O16P2S1 915.2837, found m/z
915.2810. H NMR (400 MHz CD3OD) d 0.89 (s, 3H),
30-Dephospho-Me-SCoA (7). This compound was syn-
thesized and purified similarly to 6 except that 30-depho-
spho-CoASH was used in place of CoASH. FABHRMS
(M+H+) calcd for C22H38N7O13P2S1 702.1724, found
m/z 702.1744. 1H NMR (400 MHz, D2O) d 0.67 (s, 3H),
0.80 (s, 3H), 1.93 (s, 3H), 2.32 (m, 2H), 2.47 (m, 2H), 3.21
(m, 2H), 3.33 (m, 2H), 3.46 (d, 1H), 3.72 (d, 1H), 3.88 (s,
1H), 4.10 (m, 2H), 4.26 (m, 1H), 4.40 (m, 1H), 6.03 (d,
1H), 8.29 (s, 1H), 8.52 (s, 1H).
1
1.05 (s, 3H), 1.40 (m, 2H), 1.53 (m, 2H), 1.66 (m, 1H),
1.79 (m, 1H), 1.99 (s, 3H), 2.43 (m, 2H), 2.68 (m, 2H),
3.20 (m, 4H), 3.37 (m, 2H), 3.47 (m, 2H), 3.79 (m, 1H),
3.98 (s, 1H), 4.09 (d, 1H), 4.30 (m, 3H), 4.38 (m, 1H),
4.45 (m, 1H), 4.64 (m, 1H), 6.10 (d, 1H), 8.38 (s, 1H),
8.67 (s, 1H); 13C NMR (100 MHz, CD3OD) d 176.06,
174.26, 172.75, 172.27, 171.34, 158.48, 150.36, 144.48,
142.61, 118.76, 88.61, 84.38, 75.23, 74.38, 74.01, 73.14,
70.59, 65.96, 53.34, 39.20, 38.50, 35.33, 35.24, 34.72,
31.67, 31.55, 28.78, 23.06, 21.35, 20.63, 18.62.
PPAT/DPCK assay
Lys-phosphopantetheine bis[(pivaloyloxy)methyl)] ester
(4). Method 1. Two equivalents of tributyltin methox-
ide was added to 2 dissolved in a minimal volume of
methanol and the mixture was stirred for 30 min at
room temperature. The mixture was concentrated in
vacuo, dissolved in a similar volume of acetonitrile, and
2 equivof tetrabutylammonium bromide and 10 equiv
of iodomethyl pivaloate were added, and the mixture
was stirred for 3 h at 40 ꢂC. The mixture was con-
centrated in vacuo, dissolved in 50% methanol/water
and purified by reversed phase HPLC using a water-
acetonitrile gradient without the addition of TFA. The
yield of purified 4 was ca. 7.5%.
A modified procedure adopted from Daugherty et al.
was used.11 The reaction mixture contained 50 mM Tris
(pH 8.0), 10 mM MgCl2, 20 mM KCl, 1 mM dithio-
threitol, 50 mg/mL bovine serum albumine, 5 mM ATP,
0.5 mM of the tested compound and 2 nM to 2 mM pur-
ified recombinant PPAT/DPCK enzyme. Reaction mix-
tures were incubated at 37ꢂC and timed aliquots (75 mL)
samples were removed and quenched by the addition of 6
mL 0.5 M EDTA. The quenched solutions were passed
through Microcon filters (10 kDa-MW cutoff, Millipore
Corp.) and analyzed by reversed-phase HPLC using a
gradient of 50 mM KH2PO4 (pH 4.5) in water and
methanol with UV detection at 260 and/or 214 nm.
Assignment of the HPLC peaks were based on co-injected
standards as well as mass spec identification of each iso-
lated compound from peak collection. Conversion rates
were determined by comparison of relative peak areas.
Method 2. Compound 3 was reacted with 2 equivof
bis[(pivaloyloxy)methyl] hydrogen phosphate and 2
equivof DIC in a minimal volume of DMF at room
temperature overnight. Following the addition of 2
equivmore DIC, stirring was continued for another 2 h
and then the precipitate was removed by filtration and
the product purified by reversed-phase HPLC in a
water/acetonitrile gradient without the addition of
Histone acetyltransferase p300 inhibition assay
The procedure was adapted from previously descri-
bed methods.5b Briefly, the concentrations of
[14C]acetyl-coenzyme A (Amersham Pharmacia) and

M. Cebrat et al. / Bioorg. Med. Chem. 11 (2003) 3307–3313
3313
peptide substrate (N-terminal 20-amino acid fragment
of histone H4) were fixed at 20 and 50 mM, respectively;
buffer conditions included 50 mM Tris–HCl, pH 8, 1
mM dithiothreitol, 0.1 mM EDTA, and 50 mg/mL
acetylated bovine serum albumin. Reactions employed
purified recombinant human p300HAT domain enzyme
(residues 1284–1673)18 at a concentration of 5 nM.
Assays were carried out in 0.5-mL plastic tubes at 30 ꢂC,
and the reaction volumes were 30 mL. Reactions were
initiated with [14C]acetyl-coenzyme A after allowing the
other components to equilibrate at 30 ꢂC for 10 min,
and quenched after 5–10 min with 6 mL 6ꢃTris–Tricine
gel loading buffer as described previously.5b Mixtures
were run out on 16% SDS Tris–Tricine polyacrylamide
gels, dried, and radioactivity quantified by phosphor-
image analysis (Molecular Dynamics) by comparison to
known quantities of 14C-labeled bovine serum albumin
standard (Amersham Pharmacia). Readings for back-
ground reactions carried out in the absence of p300
were subtracted from the total signal to net that used for
calculation of enzyme rates. All assays were performed
at least twice and duplicates agreed within 20%. For
inhibition assays, the specified amount of inhibitors in
all cases were at least 5-fold greater than the enzyme
concentration employed.
Thompson, P. R.; Kurooka, H.; Nakatani, Y.; Cole, P. A.
J. Biol. Chem. 2001, 276, 33721. (d) Poux, A. N.; Cebrat, M.;
Kim, C. M.; Cole, P. A.; Marmorstein, R. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 14065.
6. Richards, E. J.; Elgin, S. C. Cell 2002, 108, 489.
7. (a) Kundu, T. K.; Palhan, V. B.; Wang, Z. Y.; An, W.;
Cole, P. A.; Roeder, R. G. Mol. Cell 2000, 6, 551. (b) Asahara,
H.; Santoso, B.; Du, K.; Cole, P. A.; Davidson, I.; Montminy,
M. Mol. Cell. Biol. 2001, 21, 7892. (c) Lu, H.; Pise-Masison,
C. A.; Fletcher, T. M.; Schiltz, R. L.; Nagaich, A.; Radono-
vich, M.; Hager, G.; Cole, P. A.; Brady, J. N. Mol. Cell. Biol.
2002, 22, 4450.
8. (a) Polesskaya, A.; Naguibneva, I.; Duquet, A.; Ait-Si-Ali,
S.; Robin, P.; Vervish, A.; Cole, P.; Harel-Bellan, A. EMBO J.
2001, 20, 6816. (b) Costanzo, A.; Merlo, P.; Pediconi, N.;
Fulco, M.; Sartorelli, V.; Cole, P. A.; Fontemaggi, G.; Fan-
ciulli, M.; Schiltz, L.; Blandino, G.; Balsano, C.; Levrero, M.
Mol. Cell 2002, 9, 175. (c) Subbaramaiah, K.; Cole, P. A.;
Dannenberg, A. J. Cancer Res. 2002, 62, 2522. (d) Bandyo-
padhyay, D.; Okan, N. A.; Bales, E.; Nascimento, L.; Cole,
P. A.; Medrano, E. E. Cancer Res. 2002, 62, 6231.
9. Nguyen, T.-G.; Gerbing, K.; Eggerer, H. Hoppe-Seyler’s Z.
Physiol. Chem. 1984, 1, 365.
10. (a) Begley, T. P.; Kinsland, C.; Strauss, E. Vitam. Horm.
2002, 61, 157. (b) Tahiliani, A. G.; Beinlich, C. J. Vitam.
Horm. 1991, 46, 165.
11. Daugherty, M.; Polanuyer, B.; Farrell, M.; Scholle, M.;
Lykdis, A.; de Crecy-Lagard, V.; Osterman, A. J. Biol. Chem.
2002, 277, 21431.
12. (a) Srivasta, D. N.; Farquhar, D. Bioorg. Chem. 1984, 12,
118. (b) Jiang, T.; Sweeney, G.; Rudolph, M. T.; Klip, A.;
Traynor-Kaplan, A.; Tsien, R. Y. J. Biol. Chem. 1998, 273,
11017.
Acknowledgements
We thank C. Gross and V. Sagar for help with NMR
and p300 purification, respectively. We thank other
members of the Cole lab for helpful discussions. We are
grateful for support from the NIH and Ellison Medical
Foundation. PRT was supported in part by a Canadian
Institutes for Health Research post-doctoral fellowship.
13. (a) Iyer, R. P.; Phillips, L. R.; Biddle, J. A.; Thakker,
D. R.; Egan, W.; Aoki, S.; Mitsuya, H. Tetrahedron Lett.
1989, 30, 7141. (b) Freed, J. J.; Fraquhar, D.; Hampton, A.
Biochem. Pharm. 1989, 38, 3193. (c) Farquhar, D.; Khan, S.;
Srivasta, D. N.; Sanders, P. P. J. Med. Chem. 1994, 37, 3902.
(d) Benzaria, S.; Pelicano, H.; Johnson, R.; Maury, G.;
Imbach, J.-L.; Aubertin, A.-M.; Obert, G.; Gosselin, G.
J. Med. Chem. 1996, 39, 4958. (e) Stankovic, C. J.; Surendran,
N.; Lunney, E. A.; Plummer, M. S.; Para, K. S.; Shahripour,
A.; Fergus, J. H.; Marks, J. S.; Herrera, R.; Hubbel, S. E.;
Humblet, C.; Saltiel, A. R.; Stewart, B.; Sawyer, T. K. Bioorg.
Med. Chem. Lett. 1997, 7, 1909. (f) Kang, S. H.; Sinhababu,
A. K.; Cory, J. G.; Mitchel, B. S.; Thakker, D. R.; Cho, M.
J. Pharm Res. 1997, 14, 706. (g) Kang, S. H.; Sinhababu,
A. K.; Cho, M. J. Nucleosides Nucleotides 1998, 17, 1089.
14. In the published assay, the rate constants for PPAT/
DPCK were based on a spectrophotometric assay where PPi
and ADP formation were coupled to NADH to NAD con-
version.
15. Wolf, E.; Vassilev, A; Makino, Y.; Sali, A.; Nakatani, Y.;
Burley, S. K. Cell 1998, 94, 439.
16. Marmostein, R. J. Mol. Biol. 2001, 311, 433.
17. Martinez-Balbas, M. A.; Bannister, A. J.; Martin, K.;
Haus-Seuffert, P.; Meisterernst, M.; Kouzarides, T. EMBO J.
1998, 17, 2886.
18. Thompson, P. R., Cole, P. A., In preparation.
19. Blaschkowski, H. P.; Knappe, J.; Wieland, T. FEBS Lett.
1979, 98, 81.
References and Notes
1. Chase, J. F.; Tubbs, P. K. Biochem. J. 1969, 111, 225.
2. Williams, J. W.; Northrup, D. B. J. Antibiot. 1979, 32,
1147.
3. Paige, L. A.; Zheng, G. Q.; DeFrees, S. A.; Cassady, J. M.;
Geahlen, R. L. J. Med. Chem. 1989, 32, 1665.
4. (a) Khalil, E.; Cole, P. A. J. Am. Chem. Soc. 1998, 120,
6195. (b) Khalil, E. M.; De Angelis, J.; Ishii, M.; Cole, P. A.
Proc. Nat. Acad. Sci. U.S.A. 1999, 96, 12418. (c) Hickman,
A. B.; Namboodiri, M. A.; Klein, D. C.; Dyda, F. Cell 1999,
97, 361. (d) Kim, C. M.; Cole, P. A. J. Med. Chem. 2001, 44,
2479. (e) Wolf, E.; De Angelis, J.; Khalil, E. M.; Cole, P. A.;
Burley, S. K. J. Mol. Biol. 2002, 317, 215. (f) Zheng, W.; Cole,
P. A. Curr. Med. Chem. 2002, 9, 1187.
5. (a) Lau, O. D.; Kundu, T. K.; Soccio, R. E.; Ait-Si-Ali, S.;
Khalil, E. M.; Vassilev, A.; Wolffe, A. P.; Nakatani, Y.; Roe-
der, R. G.; Cole, P. A. Mol. Cell 2000, 5, 589. (b) Lau, O. D.;
Courtney, A. D.; Vassilev, A.; Marzilli, L. A.; Cotter, R. J.;
Nakatani, Y.; Cole, P. A. J. Biol. Chem. 2000, 275, 21953. (c)