Design, synthesis and biological evaluation of carboxy analogues of arginine methyltransferase inhibitor 1 (AMI-1)

Article abstract of DOI:10.1002/cmdc.200900459

Here we report the synthesis of a number of compounds structurally related to arginine methyltransferase inhibitor 1 (AMI-1). The structural alterations that we made included: 1) the substitution of the sulfonic groups with the bioisosteric carboxylic gro

Full text of DOI:10.1002/cmdc.200900459

MED
DOI: 10.1002/cmdc.200900459
Design, Synthesis and Biological Evaluation of Carboxy
Analogues of Arginine Methyltransferase Inhibitor 1
(AMI-1)
Sabrina Castellano,[a] Ciro Milite,[a] Rino Ragno,[b] Silvia Simeoni,[b] Antonello Mai,[b]
Vittorio Limongelli,[c] Ettore Novellino,[c] Ingo Bauer,[d] Gerald Brosch,[d] Astrid Spannhoff,[e]
Donghang Cheng,[e] Mark T. Bedford,*[e] and Gianluca Sbardella*[a]
Dedicated to Professor Marino Artico on the occasion of his 75th birthday.
Here we report the synthesis of a number of compounds struc-
turally related to arginine methyltransferase inhibitor 1 (AMI-1).
The structural alterations that we made included: 1) the substi-
tution of the sulfonic groups with the bioisosteric carboxylic
groups; 2) the replacement of the ureidic function with a bis-
amidic moiety; 3) the introduction of a N-containing basic
moiety; and 4) the positional isomerization of the aminohy-
droxynaphthoic moiety. We have assessed the biological activi-
ty of these compounds against a panel of arginine methyl-
transferases (fungal RmtA, hPRMT1, hCARM1, hPRMT3,
hPRMT6) and a lysine methyltransferase (SET7/9) using histone
and nonhistone proteins as substrates. Molecular modeling
studies for a deep binding-mode analysis of test compounds
were also performed. The bis-carboxylic acid derivatives 1b
and 7b emerged as the most effective PRMT inhibitors, both in
vitro and in vivo, being comparable or even better than the
reference compound (AMI-1) and practically inactive against
the lysine methyltransferase SET7/9.
Introduction
Besides allowing the cell to expand its repertoire over the con-
straints imposed by the twenty encoded amino acids, post-
translational modifications of proteins play pivotal roles in
chromatin-templated nuclear events, such as transcription and
DNA damage repair.[1–3] The methylation of arginine residues is
a prevalent post-translational modification, found on both nu-
clear and cytoplasmic proteins, catalyzed by the protein argi-
nine N-methyltransferase (PRMT) family of enzymes. Arginine-
methylated proteins are involved in a number of different cel-
lular processes, including transcriptional regulation, RNA me-
tabolism and DNA damage repair.[4–7] Most PRMTs methylate
glycine- and arginine-rich patches (GAR motifs) within their
substrates, using S-adenosylmethionine (SAM) as the methyl
donor.[8,9] The complexity of the methylarginine marker is en-
hanced by the ability of this residue to be methylated in three
different ways on the guanidino group: monomethylated
(MMA), symmetrically dimethylated (sDMA) and asymmetrically
dimethylated (aDMA), each of which has potentially different
functional consequences.[4]
increasing amount of evidence shows the involvement of
PRMTs in a wide variety of cellular processes, including nuclear
hormone receptor-mediated signaling, protein–protein interac-
tions, protein trafficking, mRNA splicing and processing, and
transcriptional regulation.[4,6,9,10] In particular, PRMT1 plays a
key role in the shuttling of heterogeneous nuclear ribonucleo-
proteins (hnRNPs) between the cytoplasm and the nucleus[9,11]
and is a transcriptional coactivator for multiple nuclear recep-
tor family members (e.g., the androgen and estrogen recep-
tors),[12–14] being recruited to promoters by a number of differ-
[a] Dr. S. Castellano, Dr. C. Milite, Prof. G. Sbardella
Dipartimento di Scienze Farmaceutiche, Universitꢀ degli Studi di Salerno
Via Ponte Don Melillo, 84084 Fisciano SA (Italy)
Fax: (+39)089-96-9602
E-mail: gsbardella@unisa.it
[b] Dr. R. Ragno, Dr. S. Simeoni, Prof. A. Mai
Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza Universitꢀ di
Roma, P. le A. Moro 5, 00185 Roma (Italy)
[c] Dr. V. Limongelli, Prof. E. Novellino
Dipartimento di Chimica Farmaceutica e Tossicologica, Universitꢀ di Napoli
“Federico II”, Via D. Montesano 49, 80131 Napoli (Italy)
To date, ten mammalian PRMTs have been identified; they
are classified as type I, type II, type III or type IV enzymes.[9]
Types I, II and III methylate the terminal (or w) guanidino nitro-
gen atoms; type I PRMTs (PRMT1, 3, 4, 6 and 8) catalyze the
production of aDMA, whereas type II PRMTs (PRMT5, PRMT7
and FBXO11) catalyze the formation of sDMA. PRMT7, a type III
enzyme, catalyzes the formation of MMA on certain substrates,
but does catalyze the formation of sDMA catalysis. A type IV
enzyme that catalyzes the monomethylation of the internal (or
d) guanidino nitrogen atom has been described in yeast.[9] An
[d] Dr. I. Bauer, Prof. G. Brosch
Division of Molecular Biology, Biocenter-Innsbruck Medical University
Fritz-Preglstrasse 3, 6020 Innsbruck (Austria)
[e] Dr. A. Spannhoff, Dr. D. Cheng, Prof. M. T. Bedford
University of Texas M.D. Anderson Cancer Center, Science Park-Research
Division, Smithville, Texas 78957 (USA)
Fax: (+1)512-237-2475
E-mail: mtbedford@mdanderson.org
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.200900459.
398
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
ent transcription factors.[4,9] Dysregulation of nuclear receptor
signaling is a hallmark of hormone-dependent cancers, such
as, breast cancer.[15–18] In addition to this, PRMT1 has recently
emerged as a potential new target for the development of a
novel therapeutic for heart disease,[19–26] as it appears to be re-
sponsible for generating the majority of aDMA and it is over-
expressed in the hearts of patients with coronary heart dis-
ease.[27]
binding.[36] To date, AMI-1 is the most active nonpeptidic[47] in-
hibitor reported to be selective against PRMT1; other com-
pounds are from 1.5- to approximately 45-fold less active.[41,42]
Prompted by our interest in the discovery of small-molecule
modulators of epigenetic targets,[37,38,40,48–55] we focused our at-
tention on AMI compounds and noticed the occurrence of
dye-like scaffolds as a common feature.[40] Therefore, we carried
out molecular modeling studies and described the binding
mode analysis of a focused library of small molecules into the
catalytic domain of both hPRMT1 and its fungal homologue
RmtA, validated by us as a preliminary screening tool for argi-
nine methyltransferase inhibitors.[40] We also performed enzy-
matic assays on recombinant RmtA and PRMT1 proteins using
SAM and histones as substrates, and obtained a good agree-
ment between biological and computational results. The major
outcomes of the docking and binding mode analysis (the relia-
bility of which was then confirmed by structure-based three-di-
mensional QSAR models[56]) were the positioning of AMI-1 be-
tween the SAM and arginine binding sites without fully occu-
pying them, whereas the nonarginine-selective derivative AMI-5
(Figure 1) was positioned in the SAM binding site.[40] This is
consistent with previously reported kinetics experiments.[36]
Moreover, these analyses hinted that two regions in the RmtA
catalytic site, the pocket formed by Ile12, His13, Met16, and
Thr49 (dark gray area in Figure 2)[57] and the SAM methioninic
portion binding site delimited by Arg22, Asp44, Gly46, Cys47,
Ile51, Leu52 and Glu112 (light gray area in Figure 2),
should be taken into account when designing novel
inhibitors.[40]
Similarly, CARM1, sometimes referred to as PRMT4, has been
shown to methylate many coactivators, including p300/CBP
and AIB1 (amplified in breast cancer-1),[4] as well as proteins in-
volved in splicing[28] and RNA-binding proteins,[3,29,30] thus play-
ing a crucial role in modulating gene expression at multiple
critical levels.[9] Recently it was reported that CARM1 is up-
regulated during the progression of prostate cancer,[20] and
that CARM1 and PRMT1 synergistically coactivate NF-kB-de-
pendent gene expression.[31] Therefore, convincing evidence
supports the hypothesis that targeting PRMTs would be a
viable approach in anticancer therapy.
Despite extensive research aimed at better understand the
role of PRMTs in physiological and pathological path-
ways,[9,12–24,26] elucidating the structure[32–34] of these enzymes,
and gaining insights into the mechanism of methyl transfer,[35]
there have been only a few publications to date describing
small-molecule chemical modulators of the PRMTs
(Figure 1).[36–46]
However, before undertaking the exploration of
the two aforementioned additional pockets, we real-
ized that AMI-1 should be optimized as it is likely to
have low bioavailability and would probably not pen-
etrate the blood–brain barrier due to the bisanionic
structure. Moreover, it is related to suramin-type sul-
fonated ureas, reported to give pleiotropic interac-
tions with many proteins.[58,59]
Therefore, we designed a number of derivatives
characterized by the substitution of the sulfonic
groups with the bioisosteric carboxylic groups, the
replacement of the ureidic function with a bisamidic
moiety, the introduction of a N-containing basic
moiety or the positional isomerization of the amino-
hydroxynaphthoic moiety (Figure 3).
Here, we describe in detail the synthesis of com-
pounds 1–9 and their biological evaluation against a
panel of PRMTs (fungal RmtA, hPRMT1, hCARM1,
hPRMT3, hPRMT6), as well as against a lysine methyl-
transferase, SET7/9.
Figure 1. Small molecule inhibitors of PRMTs.
Results and Discussion
Among them, 7,7’-carbonylbis(azanediyl)bis(4-hydroxynaph-
thalene-2-sulfonic acid) (AMI-1, Figure 1) was identified in 2004
through a random screening effort by Bedford and co-workers
as a compound able to inhibit arginine, but not lysine methyla-
tion, being noncompetitive with S-adenosylmethionine (SAM)
Chemistry
The key step in the synthesis of derivatives 1–6 (Scheme 1)
was the preparation of 12a. Following a highly efficient proce-
dure recently described by us,[60] we used a Wittig reaction be-
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
399

G. Sbardella, M. T. Bedford et al.
MED
furan furnished the corresponding acid 1b (Scheme 1), where-
as the treatment of the ester 1a with aqueous ammonia gave
the bisamidic derivative 1c. Conversely, the unsymmetrical
ureidic derivatives were prepared by treating 12a with trichloro-
acetylchloride in dichloromethane and reacting the resulting
trichloroacetamide 14 with the appropriate amine to obtain
compounds 4a, 5a and 6a, the hydrolysis of which yielded
acids 4b and 6b, respectively (Scheme 1).[62]
With regard to bisamidic derivatives, the reaction of 12a
with succinyl chloride in the presence of triethylamine yielded
ester 3a, which was hydrolyzed to the acid 3b following the
aforementioned protocol. The reaction of 12a with malonyl
chloride under these conditions failed and the malonyl dia-
mide 2a was obtained via a different route in two steps: one
equivalent of compound 12a was reacted with an excess of
neat diethyl malonate by microwave irradiation, and then a
second equivalent of the compound was added as a solution
in N-methylpyrrolidone/toluene (1:10). Again, subsequent hy-
drolysis of 2a gave the corresponding acid 2b.
The absence of regioselectivity
Figure 2. The two additional binding pockets in the RmtA catalytic site that
emerged from three-dimensional QSAR studies. The area highlighted in light
gray[57] is delimited by Arg22, Asp44, Gly46, Cys47, Ile51, Leu52 and
Glu112, whereas the area depicted in dark gray is formed by Ile12, His13,
Met16, and Thr49. The binding mode of AMI-1 (stick representation, carbon
atoms in gray) is also shown.
in the preparation of the 7-
amino-4-hydroxy-2-naphthoic
ester 12a, which had to be sepa-
rated from the 5-amino-substi-
tuted isomer 12b,[60] hindered
the assembly of a focused library
based on this intermediate with
which we intended to explore
the two aforementioned pockets
that emerged from computa-
tional studies (Figure 2). There-
fore, we decided to synthesize
derivatives 7a,b, 8a,b and 9a,b,
positional isomers of 1a,b, 2a,b
and 4a,b, respectively, and to
evaluate their biological activi-
ties.
The Wittig reaction between
4-nitrobenzaldehyde and carboxy-
phosphorane 10, followed by
the microwave-assisted Friedel–
Figure 3. Novel AMI-1 analogues.
Crafts-type ring closure, yielded
only methyl 4-acetoxy-6-nitro-2-
tween 3-nitrobenzaldehyde and carboxyphosphorane 10[61] to
regioselectively prepare the (E)-nitrophenylitaconate 11 a,
which was selectively reduced with zinc dust in acetic acid to
give the amino derivative 11 b (Scheme 1). Ring closure via mi-
crowave-assisted Friedel–Crafts acylation[60] followed by the hy-
drolysis of the crude product and subsequent esterification fur-
nished a mixture of 5-amino- and 7-amino-substituted naph-
thoic esters from which 12a was conveniently obtained by
precipitating its 5-amino- isomer 12b as an insoluble cobalt(II)
complex salt.
naphthoate 15, which was promptly reduced to the key inter-
mediate 16 by heterogeneous catalytic hydrogenation
(Scheme 2).
The reaction of 16 with diphenyl carbonate in refluxing
chlorobenzene followed by the hydrolysis of the acetoxy
group with potassium carbonate gave the symmetrical ureidic
derivative 7a. The subsequent hydrolysis with aqueous sodium
hydroxide and pyridine in tetrahydrofuran furnished the corre-
sponding acid 7b (Scheme 2). Conversely, the unsymmetrical
ureidic derivatives were prepared by treating 16 with trichloro-
acetylchloride to give the corresponding trichloroacetamides
18[63] and then reacting these with tryptamine to obtain the
ester 9a. The hydrolysis of the latter yielded the corresponding
acid 9b (Scheme 2). Finally, the two-step reaction under micro-
The symmetrical ureidic derivative 1a was directly obtained
by reacting 12a with diphenyl carbonate in refluxing chloro-
benzene in the presence of DMAP. The following hydrolysis
with aqueous sodium hydroxide and pyridine in tetrahydro-
400
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
Scheme 1. Reagents and conditions: a) benzene, RT, 48 h; b) Zn, AcOH, RT, 24 h; c) NaOAc, Ac2O, MW (300 W, 5 min); d) aq 8n HCl, 5 h; e) EtOH, H2SO4, reflux,
24 h; f) Co(OAc)2·4H2O, AcOH/NaOAc pH 5 buffer, CH3OH; g) diphenylcarbonate, DMAP, chlorobenzene, reflux, 72 h; h) aq NaOH, pyridine, THF, RT; i) aq NH3,
RT, 12 h; j) succinyl chloride, Et3N, acetone, RT, 4 h; k) diethyl malonate, MW (300 W, 30 min), neat; l) 12a, toluene/NMP (10:1), MW (300 W, 3ꢁ30 min);
m) ClCOCCl3, CH2Cl2, RT, 4 h; n) R1NH2, K2CO3, DMF, 1508C, 1 h, sealed tube.
Scheme 2. Reagents and conditions: a) benzene, RT, 48 h; b) NaOAc, Ac2O, MW (300 W, 5 min); c)H2, Pd/C, EtOH, 2 h; d) 1) diphenylcarbonate, DMAP, chloro-
benzene, reflux, 72 h; 2) K2CO3, EtOH, 708C, 2 h; e) aq NaOH, pyridine, THF, RT; f) diethyl malonate, MW (300 W, 30 min), neat; g) 1) 16, toluene/NMP (10:1),
MW (300 W, 3ꢁ30 min); 2) K2CO3, EtOH, 708C, 2 h; h) ClCOCCl3, CH2Cl2, RT, 4 h; i) tryptamine, K2CO3, DMF, 1508C, 1 h, sealed tube.
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
401

G. Sbardella, M. T. Bedford et al.
MED
wave irradiation between 16 and diethyl malonate, followed
by the hydrolysis of the acetoxy group with potassium carbon-
ate, yielded the malonyl diamide 8a. Again, subsequent hy-
drolysis of 8a gave the corresponding acid 8b (Scheme 2).
The introduction of the tyramine nucleus in place of one of
the two naphthalenic moieties resulted in derivatives with ac-
tivities comparable to those of their counterparts (cf. activities
of 1a and 4a, or 1b and 4b). On the other hand, replacement
with the isosteric indole-2-carboxylic moiety gave less homo-
geneous results. In fact, indolic derivatives 6a and 6b showed
decreased RmtA inhibition (Table 1) in comparison with their
naphthalenic counterparts 1a and 1b, respectively, but the ac-
tivities against PRMT1 were similar (Table 1 and Table 2).
Strangely, in this case, carboxylic acid 6b was less active than
the corresponding ester 6a. This outcome could be justified
by the formation of an intramolecular H bond between the
indole NH and the COOH group, thus reducing the availability
of both groups for interaction with the binding pocket of the
enzyme.
Biology
In accordance with our previous studies,[37,40] we first per-
formed a preliminary screening of the activities of compounds
1–9 against Aspergillus nidulans RmtA, a fungal PRMT acting on
histone H4 substrate and validated by us as a useful, predictive
model for studying PRMT inhibition in mammals.[40] Then we
tested the derivatives against human recombinant PRMT1 in vi-
tro, using histone as well as nonhistone (the RNA-binding nu-
clear shuttling protein, Npl3) proteins as a substrate, to con-
firm their inhibitory activity and to observe the influence of
substrates different from histones on the inhibitory activity.
Subsequently, selected compounds were tested (50 mm)
against a panel of human PRMTs (PRMT1, PRMT3, CARM1, and
PRMT6), using histone H4 (for PRMT1), histone H3 (for CARM1
and PRMT6) or GAR (for PRMT3) motifs as substrates. Further-
more, to assess the selectivity of our compounds against lysine
methyltransferases, we also tested our compounds against the
HKMT SET7/9 using histone H3 as a substrate.
Regarding compounds resulting from the formal shift of the
ureidic function from the C-7 to the C-6 position of the naph-
thalene ring, it is noteworthy that their inhibitory activity was
greatly enhanced. In fact, compounds 7, 8, and 9 were more
potent than their positional isomers. Moreover, the biscarbox-
ylic acid derivative 7b, the isomer of 1b, showed the highest
inhibitory efficacy, comparable (Table 1) or even better
(Table 2) than AMI-1.
Finally, the introduction of a tertiary amine, like the dimethyl-
aminopropyl moiety in compound 5a, led to a substantial de-
crease of the inhibitory potency against hPRMT1 (Table 2).
To determine whether the compounds that showed arginine
methyltransferase inhibitory properties were able to inhibit
PRMT activity within a cellular context, we used a fusion be-
tween green fluorescence protein (GFP) and the yeast protein
Npl3. We previously established that mammalian PRMT1 can
methylate Npl3 in vitro,[36] thus we reasoned that this reaction
could also take place within a mammalian cell line. A destabi-
lized GFP variant was used that displays rapid turnover rates.
This shorter half-life makes destabilized variants suitable for
use in quantitative reporter assays. The GFP–Npl3 was transi-
ently transfected into HeLa cells; post-transfection the cells
were treated for 24 h with derivatives 1b, 7b, 8b and 9b (10,
50, and 100 mm), using AMI-1 and 2’,3’-acycloadenosine-2’,3’-di-
aldehyde (adenosine dialdehyde, AdOx), an indirect methyl-
transferase inhibitor,[66,67] as reference compounds. Because
GFP and Npl3 are fused, the aGFP antibody was used to estab-
lish equal loading and aNpl3 antibody (1E4)[68] acted as the
methylation sensor (Figure 4A). Thus, the relative degree of ar-
ginine methylation in the presence of the different inhibitors
can be established. Using this assay system we demonstrated
that all tested derivatives were able to inhibit methylation of
the GFP-Npl3 fusion, even if to varying extents (data not
shown). We thus focused our attention on 7b, the compound
that showed the highest inhibitory efficacy in enzymatic
assays. A concentration gradient of 7b (10, 50, 100 mm) was
used to treat GFP–Npl3 transiently transfected HeLa cells for
24 h, using AMI-1 (10 and 100 mm) and AdOx (10 and 20 mm)
as reference compounds. Total cell extracts were then subject-
ed to Western analysis with aGFP and 1E4 (methyl-sensitive
aNpl3) antibodies. Derivative 7b inhibited the methylation of
Npl3 within the cell in a dose-dependent manner and more
Inhibitory activities against RmtA and PRMT1
Compounds 1–9 were preliminarily tested against Aspergillus
nidulans RmtA, a fungal PRMT with significant sequence simi-
larity to human PRMT1 and specific for methylation at Arg3 of
histone H4,[64] and against hPRMT1, using core histones as sub-
strate as previously described.[40,64] The inhibition (%) at a fixed
dose (nearly 100 mm) were first determined (data not shown),
and then the IC50 values for the active compounds were estab-
lished (Table 1).
Moreover, the derivatives were also tested against hPRMT1,
using the heterogeneous nuclear ribonucleoprotein (hnRNP)
Npl3p, an in vivo substrate of HMT1 from Saccharomyces cere-
visiae,[65] as a substrate. The inhibition (%) at fixed doses (10
and 50 mm) were determined (Table 2). AMI-1 was used as ref-
erence compound in both assays.
The first result that emerged from both assays was that the
substitution of the AMI-1 sulfonic group with its carboxylic iso-
ster gave only a slight decrease in inhibiting activity (cf. AMI-1
and 1b, Tables 1 and 2). Conversely, the replacement of the
carboxylic group with an ester or an amide function dimin-
ished the activity against PRMT1. There was no difference in
the order of activity when histone or nonhistone proteins were
used as the substrate (1b>1c>1a), however, a slightly differ-
ent order resulted when compared to the results obtained
against RmtA (1b>1a>1c).
The substitution of the ureidic group with bisamidic moiet-
ies was detrimental to the inhibitory potency of the resulting
derivatives, with the decrease being proportional to the length
of the aliphatic spacer (cf. inhibition (%) values of compounds
1b, 2b and 3b).
402
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
Table 1. Inhibitory activities (IC50 values) of compounds 1–9 against hPRMT1 and RmtA using histone substrates.[a, b]
Compd
W
X
Y
Z
R1
R2
IC50 ꢀSD [mm]
RmtA
PRMT1
1a
1b
1c
2a
2b
3a
3b
OH
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
COOEt
COOH
CONH2
COOEt
COOH
COOEt
COOH
H
H
H
H
H
H
H
–
–
–
CH2
CH2
–
–
–
–
–
–
–
190.8ꢀ7.6
137.7ꢀ4.1
445.1ꢀ26.7
121.0ꢀ3.6
ND[c]
272.2ꢀ10.9
122.8ꢀ6.2
447.8ꢀ19.9
298.0ꢀ8.9
405.9ꢀ20.3
233.2ꢀ9.3
ND[c]
676.1ꢀ33.8
342.7ꢀ17.2
CH2CH2
CH2CH2
4a
4b
OH
OH
H
H
COOEt
COOH
H
H
–
–
195.7ꢀ7.8
345.6ꢀ6.9
170.1ꢀ8.5
262.3ꢀ13.1
5a
6a
OH
OH
H
H
COOEt
COOEt
H
H
–
–
180.2ꢀ3.6
875.0ꢀ35.0
510.0ꢀ30.6
620.0ꢀ24.8
6b
OH
H
COOH
H
–
329.8ꢀ16.5
0[d]
7a
7b
8a
8b
H
H
H
H
COOMe
COOH
COOMe
COOH
H
H
H
H
OH
OH
OH
OH
–
–
CH2
CH2
–
–
–
–
204.3ꢀ8.2
94.7ꢀ3.8
195.2ꢀ7.8
111.7ꢀ3.4
132.8ꢀ5.3
271.6ꢀ13.6
204.9ꢀ12.3
395.1ꢀ15.8
9a
H
COOMe
H
OH
–
290.0ꢀ14.5
320.2ꢀ12.8
9b
H
COOH
H
H
OH
H
–
–
226.7ꢀ9.1
88.2ꢀ2.6[e]
205.4ꢀ8.2
92.1ꢀ3.7[e]
AMI-1
OH
SO3H
–
[a] Chicken erythrocyte core histones and SAM were used as substrates at the concentration of 25 mm and 0.13 mm, respectively; [b] Values are means de-
termined for at least two separate experiments; [c] Not determined; [d] No inhibition at 230 mm; [e] Literature value (Reference [41]): 33.2ꢀ7.8 mm (RmtA)
and 1.2ꢀ0.5 mm (PRMT1), fluorescence assay.
effectively than the reference AMI-1 (Figure 4b). In addition,
the inhibitor of global methylation, AdOx, also reduced the
methylation status of this reporter.
substrates, respectively, and also against the lysine methyl-
transferase SET7/9, using histone H3 as a substrate. AMI-1 was
used as reference compound in all assays.
As seen in Table 3, all of the derivatives tested are generally
more selective for arginine methyltransferases than AMI-1. In
fact, they are practically inactive against the lysine methyltrans-
ferase SET7/9, whereas AMI-1 shows a minor inhibition of this
HKMT enzyme. This, together with its capability to inhibit all
tested PRMTs, support the pleiotropic nature of the interac-
tions established by the sulfonic groups.
Inhibition against a panel of arginine methyltransferases
The most active derivatives were selected and tested at 50 mm
against a panel of arginine methyltransferases, as well as
against a lysine methyltransferase, to assess their selectivity.
Compounds 1b, 1c, 2b, 4a, 4b, 7b, 8b, 9a, and 9b were
tested against the human recombinant arginine methyltrans-
ferases PRMT1, PRMT3, CARM1 and PRMT6, using histone H4,
GAR motifs and histone H3 (for both CARM1 and PRMT6) as
In contrast, compound 1b, the carboxylic analogue of AMI-1,
is inactive against SET7/9 but its activity is fairly comparable to
that of its sulfonic counterpart against PRMT3, and to a lesser
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
403

G. Sbardella, M. T. Bedford et al.
MED
Table 2. Inhibitory activities of compounds 1–9 against hPRMT1 using nonhistone substrates.[a,b]
Compd
W
X
Y
Z
R1
R2
inhibition [%] (hPRMT1/Npl3)
50 mm
10 mm
1a
1b
1c
2a
2b
3a
3b
OH
OH
OH
OH
OH
OH
OH
H
H
H
H
H
H
H
COOEt
COOH
CONH2
COOEt
COOH
COOEt
COOH
H
H
H
H
H
H
H
–
–
–
CH2
CH2
–
–
–
–
–
–
–
40.81
75.16
63.47
34.65
60.22
31.66
30.85
4.91
41.73
0.13
10.77
21.73
0.92
CH2CH2
CH2CH2
-0.41
4a
4b
OH
OH
H
H
COOEt
COOH
H
H
–
–
61.24
71.65
30.21
20.21
5a
6a
OH
OH
H
H
COOEt
COOEt
H
H
–
–
ꢁ7.20
ꢁ1.70
53.20
27.73
6b
OH
H
COOH
H
–
32.47
8.01
7a
7b
8a
8b
H
H
H
H
COOMe
COOH
COOMe
COOH
H
H
H
H
OH
OH
OH
OH
–
–
CH2
CH2
–
–
–
–
59.07
100.00
52.07
19.62
75.29
25.13
ꢁ8.59
90.99
9a
H
COOMe
H
OH
–
83.78
20.47
9b
H
COOH
H
H
OH
H
–
–
73.17
20.15
66.89
AMI-1
OH
SO3H
–
100.00
[a] Npl3p was used as a nonhistone substrate, SAM as a cofactor (see Experimental Section); [b] Values given are means determined for at least two sepa-
rate experiments.
degree against CARM1. Interestingly, the use of histone H4 in-
stead of core histones or the nonhistone protein Npl3p as a
substrate for the PRMT1 assay yielded an appreciably weaker
inhibition of PRMT1. The malonic bisamidic derivative 2b ex-
hibited a similar activity profile against the enzyme panel
(Table 3).
and CARM1, and was comparable or even better than that
exhibited by AMI-1. However, it was practically inactive against
both PRMT6 and SET7/9 (Table 3).
The bisamidic malonic analogue 8b was consistently less
active than 7b against both PRMT1 and PRMT3, yet displayed
a positive modulating effect on the enzymatic activity of
CARM1 (Table 3). Similarly, the tryptamine derivatives 9a,b
showed little or no activity against PRMT1, PRMT3 and PRMT6,
but strongly increased enzymatic activity of CARM1. In con-
trast, the isomeric derivatives 4a,b showed only weak inhibi-
tion against all enzymes (see Supporting Information for
fluorographs[69]).
The bisamide 1c was consistently less active than 1b and
2b against both PRMT3 and CARM1, but was the only com-
pound among those tested that was able to inhibit PRMT6,
with a potency comparable to that of AMI-1 or even higher.
Regarding compound 7b, the positional isomer of 1b, this
compound was confirmed as the most active in the series
showing very good inhibitory activities against PRMT1, PRMT3,
404
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
Figure 4. Effects of compounds on cellular arginine methyltransferase activi-
ty: a) a depiction of the GFP–Npl3 fusion protein with the position of me-
thylated region and the antibodies that recognize it; b) HeLa cells were
grown in 12-well plates and then transiently transfected with d2GFP–Npl3.
Three hours post-transfection, the cells were incubated with the indicated
compounds for 24 h. The cells were lysed in RIPA buffer, and Western analy-
sis was performed with either the 1E4 antibody (top panel) or aGFP anti-
body (bottom panel). The effects of the compounds on GFP-Npl3 methyla-
tion status were established with the methyl-specific antibody, 1E4. The
aGFP antibody showed the protein levels of GFP-Npl3. DMSO (0.25% v/v)
was used as a vehicle (lanes 1, 4, 7, 11); compounds concentrations: AdOx
(10 and 20 mm, lane 2,3), AMI-1 (10 and 100 mm, lanes 5,6), 7b (10, 50, and
100 mm, lanes 8–10).
Figure 5. Autodock/X-Score selected binding conformations of compounds
1b (light gray), 7b (dark gray) and AMI-1 (black) docked into the RmtA cata-
lytic site. The volumes occupied by Arg and SAM are represented in mesh
and filled transparent gray, respectively.
Binding mode studies
The binding modes of selected PRMT inhibitors were studied
in an attempt to rationalize the differences in activity. To this
aim, compounds 1b, 7b, and AMI-1 were docked (Auto-
dock 3)[70] into the homology
model of the PRMT1 orthologue
RmtA, previously reported by
us[40] and used to describe three
different binding modes of
PRMT inhibitors: a) molecules
docked in the arginine pocket
(DAP); b) molecules docked in
the SAM pocket (DSP); and
c) molecules partially overlap-
ping with both sites (docked in
both pockets, DBP).[40] The analy-
sis of the Autodock conforma-
tions selected by the X-Score[71]
external
scoring
function
showed that 1b belongs to the
DSP group, while 7b seems to
bind in both sites and so be-
longs to the DBP group
(Figure 5).
In particular, the binding con-
formation selected for 1b is sim-
ilar to the one observed for the
SAM co-factor (Figure 6). In fact,
one naphthalenic group lays in a
Figure 6. Comparison between binding conformations of SAM cofactor and compound 1b in the RmtA catalytic
site. a) SAM cofactor (light gray carbon atoms); b) 1b (dark gray carbon atoms). The RmtA residues within 4.0 ꢂ
from the docked compounds are reported in white. For the sake of clarity, hydrogen atoms are not displayed.
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
405

G. Sbardella, M. T. Bedford et al.
MED
Table 3. Inhibition activities of selected compounds against different methyltransferases.[a]
Compd
W
X
Y
Z
R1
R2
Inhibition [%] at 50 mm
PRMT1/H4[b]
PRMT3/GAR[c]
CARM1/H3[d]
PRMT6/H3[d]
SET7/9/H3[d]
1b
1c
2b
OH
OH
OH
H
H
H
COOH
CONH2
COOH
H
H
H
–
–
CH2
–
–
–
12.36
42.42
48.44
90.92
29.46
91.94
53.06
9.26
54.96
ꢁ7.23
43.77
6.40
ꢁ8.07
13.36
ꢁ1.40
4a
4b
OH
OH
H
H
COOEt
COOH
H
H
–
–
24.73
9.86
42.13
13.98
ꢁ46.56
ꢁ12.40
6.65
39.53
18.30
ꢁ5.13
7b
8b
H
H
COOH
COOH
H
H
OH
OH
–
CH2
–
–
61.65
16.76
72.81
32.56
83.07
ꢁ298.99
ꢁ21.72
ꢁ2.09
3.80
14.44
9a
H
COOMe
H
OH
–
20.50
ꢁ5.03
ꢁ328.49
9.70
ꢁ8.46
9b
H
COOH
H
H
OH
H
–
–
38.89
87.67
ꢁ8.17
ꢁ217.98
ꢁ11.13
ꢁ10.41
AMI-1
OH
SO3H
–
101.33
74.76
40.69
34.34
[a] Values given are means determined for at least two separate experiments; [b] Histone H4 (1.5 mm) and SAM (0.42 mm) were used as substrates;
[c] Glycine- and arginine-rich (GAR) motifs (0.41 mm) and SAM (0.42 mm) were used as substrates; [d] Histone H3 (1.1 mm) and SAM (0.42 mm) were used as
substrates.
sandwich-like mode between the Met69 and Met123 side
chains (SAM adenine binding site) making positive van der
Waals interactions, while the corresponding carboxylate group
makes a weak H bond with the hydroxy group of Thr126. The
other aromatic moiety binds in the SAM methionine
maintaining the same protein orientation (Figure 8), and by
comparison of the bonding interactions (Table 4) made by
each inhibitor with the residues in the RmtA binding pockets.
pocket, delimited by Arg22, Asp44, Gly46, Cys47,
Table 4. Summary report of the interactions between AMI-1, 7b, 1b and RmtA amino
Ile51, Leu52 and Glu112, and the carboxylate func-
tion establishes either an electrostatic or a H-bond
interaction with the Arg22 side chain. Moreover, the
two ureidic NH are within H bonding distance of the
Asp68 carboxylate group (Figure 6b), thus mimick-
ing the two hydroxy groups of the SAM ribosyl
moiety (cf. Figure 6a and 6b).
acid residues.
AMI-1
7b
1b
d [ꢂ] residuegsubstituent d [ꢂ] residuegsubstituent d [ꢂ] residuegsubstituent
Electrostatic/H-bond interactions
2.9
3.1
2.9
Glu15gOH
Arg22gSO3
His261gSO3
2ꢁ
ꢁ
2.7
Arg22gCO2
2ꢁ
ꢁ
3.2
3.2
His261gCO2ꢁ
Notably, derivative 7b displays a substantially dif-
ferent binding scenario from the one described
above for its positional isomer 1b, as the binding
conformation selected by X-Score shows it belongs
to the DBP group (Figure 7). Significant H bonds
may be observed between 7b and both arginine-an-
choring residues Glu112 and Glu121 and Asp68
(Figure 7b).
Tyr116gCO2
3.3
2.8
Met114gNH
Glu112gNH
3.6
2.8
Glu112gOH
Glu121 NH
2ꢁ
2.7
Thr49 NHgSO3
2.9
2.7
Asp68gOH
Met96-NHgCO2
2.8
2.8
2.3
3.3
2.9
Asp68gNH
Met96-NHgOH
Glu97gOH
Leu52-NHgCO2
Tyr126gCO2
ꢁ
ꢁ
ꢁ
The differences in affinity values among deriva-
tives that are highly structurally related (like 1b, 7b
and AMI-1) could be better highlighted by direct
comparison of their respective binding modes while
Hydrophobic interactions
Met16-Thr49-Gly46
Trp262-His261-Tyr116
–
–
Met16-Thr49-Gly46
Trp262-His261-Tyr116
Met123-Met69
–
Met123-Met69
406
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
Figure 7. Binding conformations of compound 7b in the RmtA catalytic site. a) SAM cofactor (dark gray carbon atoms) and arginine substrate (light gray
carbon atoms); b) 7b (dark gray carbon atoms). The RmtA residues within 4.0 ꢂ from the docked compounds are reported in white. For the sake of clarity,
hydrogen atoms are not displayed.
Like AMI-1, derivative 7b can be classified as a DBP binding
compound, even though it shows a stretched conformation,
while AMI-1 was found to bind in a bent shape. In fact, similar
to AMI-1, half of 7b structure is buried in a hydrophobic
pocket delimited by Trp262, His261, and Tyr116 side chains.
However, the second half of 7b is located in the SAM adenine
binding pocket, while the second half of AMI-1 is placed into
the SAM methionine site.
Ile12-His13-His16-Thr49 pocket, as well as to the lack of any
interaction with the arginine-anchoring residues Glu112 and
Glu121, which seem to play an important role in the enzyme
inhibition.
This scenario is also supported by the X-Score protein–
ligand complex binding affinity estimations calculated by the
HMScore (hydrophobic matching algorithm), which correlates
with the IC50 values (Supporting Information, table B). Consid-
ering the HMScore, when a hydrophobic ligand atom is placed
in a hydrophobic site of a protein, it is expected to contribute
favorably to the binding process. As a matter of fact, AMI-1
and 7b present higher hydrophobic matching values than 1b,
which contributes positively to the protein–ligand binding pro-
cess and consequently to the activity (Supporting Information,
table B).
On the other hand, 1b is a DSP binding derivative and thus
shares fewer interactions with AMI-1 than 7b does. Interesting-
ly, the binding profile of 1b seems to be intermediate between
those displayed by 7b and AMI-1. In fact, half of its structure is
docked in the SAM methionine site, similar to AMI-1, and the
other half occupies the same SAM adenine binding region that
is also filled by a 7b naphthyl group (Figures 5 and 8).
In all three derivatives, either sulfonic or carboxylic acid
groups act as anchoring points to the protein establishing rele-
vant interactions. In particular, one AMI-1 sulfonic group inter-
acts with Arg22 (guanidinic side chain) and Thr49 (amidic NH)
(Table 4), and the second with main chain His261 amidic NH.
On the other hand, 1b and 7b carboxylic groups, while shar-
ing some interactions with the AMI-1 sulfonic group (H bonds
with Arg22 and His261 for 1b and 7b, respectively, Table 4),
establish new interactions with either Tyr116 and Met96 (7b)
or Leu52 and Tyr126 (1b).
Conclusions
We started by stating the rationale by which 7,7’-carbonylbis-
(azanediyl)bis(4-hydroxynaphthalene-2-sulfonic acid) (AMI-1), a
selective PRMT inhibitor[36] with a bisanionic structure that is
related to compounds known to generate pleiotropic interac-
tions with many proteins, should be further optimized before
exploring additional binding pockets. On the basis of these ob-
servations, we have described the synthesis of compounds 1–
9, which are structurally related to AMI-1 and are characterized
by the substitution of the sulfonic groups with the bioisosteric
carboxylic groups, the replacement of the ureidic function
with a bisamidic moiety, the introduction of a N-containing
A deeper analysis of the binding modes described above
could help explain the observed activity trend. The lower activ-
ity of 1b with respect to those observed for 7b and AMI-1
could be due to its weaker interaction with the important
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
407

G. Sbardella, M. T. Bedford et al.
MED
basic moiety or the positional isomerization of the aminohy-
droxynaphthoic moiety. We assessed their biological activity
against a panel of arginine methyltransferases (fungal RmtA,
hPRMT1, hCARM1, hPRMT3, hPRMT6), as well as against SET7/9
lysine methyltransferase, using histone and nonhistone pro-
teins as substrates.
Preliminary structure–activity relationships were obtained
from the biological data. Substitution of the AMI-1 sulfonic
group with the carboxylic isoster gave compound 1b, which
maintained a fairly good activity. Moreover, derivatives result-
ing from the formal shift of the ureidic function from the C-7
to the C-6 position of the naphthalene ring (compounds 7, 8,
and 9) were more potent than their positional isomers. The
biscarboxylic acid 7b, an isomer of 1b, showed the highest in-
hibitory efficacy in vitro and was able to prevent arginine
methylation of cellular proteins in whole-cell assays, with activ-
ities comparable or even better than AMI-1.
All derivatives evaluated were found to be selective for argi-
nine methyltransferases, and practically inactive against the
lysine methyltransferase SET7/9, whereas AMI-1, due to the
pleiotropic nature of the interactions established by the sulfon-
ic groups, inhibits all the enzymes tested, albeit with different
potencies, including a minor inhibition of the HKMT SET7/9.
To rationalize the observed differences in terms of activity,
we also performed molecular modeling studies that yielded a
deep binding mode analysis of tested molecules. In both deriv-
atives 1b and 7b the carboxylic acid groups act as anchoring
points to the protein by establishing relevant interactions. In
particular, while sharing some common interactions with the
AMI-1 sulfonic groups (H bonds with Arg22 and His261 for 1b
and 7b, respectively), they establish new interactions with
either Tyr116 and Met96 (7b), or Leu52 and Tyr126 (1b).
Moreover, derivative 7b presents higher hydrophobic match-
ing values than 1b, which contributes positively to the pro-
tein–ligand binding process and to activity. Consequently it
emerged as a promising candidate for further derivatization,
and represents a step towards potent and selective arginine
methyltransferase inhibitors.
Experimental Section
Chemistry
All chemicals were purchased from Aldrich Chimica (Milan, Italy) or
from Alfa Aesar GmbH (Karlsruhe, Germany) and were of the high-
est purity. All solvents were reagent grade and, when necessary,
were purified and dried by standard methods. All reactions requir-
ing anhydrous conditions were conducted under a positive atmos-
phere of nitrogen in oven-dried glassware. Standard syringe tech-
niques were used for anhydrous addition of liquids. All microwave
reactions were conducted using a CEM Corporation (Cologno al
Serio, Italy) Discover LabMate system using the standard 10 mL re-
action vessel. Reactions were routinely monitored by TLC per-
formed on aluminum-backed silica gel plates (Merck DC, Alufolien
Figure 8. Comparison between binding conformations of a) AMI-1, b) 7b
and c) 1b in the RmtA catalytic site. The RmtA residues within 4.0 ꢂ from
the docked compounds are reported in white. For the sake of clarity, hydro-
gen atoms are not displayed.
Kieselgel 60 F254) with spots visualized by UV light (l=254,
365 nm) or using a KMnO4 alkaline solution. Solvents were re-
moved using a rotary evaporator operating at a reduced pressure
of ~10 Torr. Organic solutions were dried over anhydrous Na2SO4.
Chromatographic separations were performed on silica gel (silica
408
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
gel 60, 0.063–0.200 mm; Merck DC) or on alumina (aluminum
oxide 90, active, neutral, 0.063–0.200 mm; Merck DC) columns.
Melting points were determined on a Gallenkamp melting point
apparatus in open capillary tubes and are uncorrected. Infrared (IR)
spectra (KBr) were recorded on a Shimadzu FTIR-8000 instrument.
1H NMR spectra were recorded at 300 MHz on a Bruker Avance 300
spectrometer. Chemical shifts are reported in d (ppm) relative to
the internal reference tetramethylsilane (TMS). Mass spectra were
recorded on a Finnigan LCQ DECA TermoQuest (San Jose, USA)
mass spectrometer using an electrospray ion source (ESI-MS). Com-
bustion analyses were performed by our Analytical Laboratory at
the University of Salerno (Italy). All compounds showed ꢂ98%
purity. When the elemental analysis is not included, crude com-
pounds were used in the next step without further purification. As
a rule, samples prepared for physical and biological studies were
dried in high vacuum over P2O5 for 20 h at temperatures ranging
from 25 to 1108C, depending on the sample melting point.
7-(3-(7-Carboxy-5-hydroxynaphthalen-2-ylamino)-3-oxopropan-
amido)-4-hydroxy-2-naphthoic (2b): White solid (96%); mp:
>2908C; H NMR ([D6]DMSO): d=10.50 (s, 2H), 10.45 (s, 2H), 8.33
(s, 2H), 8.14 (d, J=9 Hz, 2H), 7.93 (s, 2H), 7.73 (d, J=9 Hz, 2H),
7.29 (s, 2H), 3.64 (s, 2H); MS (ESI): m/z: 475 [M+H]+; Anal. calcd
for C25H18N2O8: C 63.29, H 3.82, N 5.90, found: C 63.38, H 3.82, N
5.89.
1
7-(4-(7-Carboxy-5-hydroxynaphthalen-2-ylamino)-4-oxobutan-
amido)-4-hydroxy-2-naphthoic (3b): White solid (90%); mp:
>2908C; 1H NMR (D2O+NaOD): d=7.92 (d, J=9 Hz, 2H), 7.55 (d,
J=9 Hz, 2H), 7.27 (s, 2H), 7.12–7.07 (m, 4H), 2.55 (s, 4H); MS (ESI):
m/z: 489 [M+H]+; Anal. calcd for C26H20N2O8: C 63.93, H 4.13, N
5.74, found: C 64.03, H 3.75, N 5.73.
7-(3-(2-(1H-Indol-3-yl)ethyl)ureido)-4-hydroxy-2-naphthoic (4b):
1
White solid (98%); mp: >2908C; H NMR ([D6]DMSO): d=11.77 (s,
1H), 10.83 (s, 1H), 10.25 (s, 1H), 8.75 (s, 1H), 8.00 (d, J=7 Hz,1H),
7.98 (s, 1H), 7.81 (s, 1H), 7.57 (d, J=8 Hz, 1H), 7.53 (dd, J1 =9 Hz,
J2 =2 Hz, 1H), 7.33 (d, J=8 Hz, 1H), 7.17 (d, J=2 Hz, 1H), 7.15 (d,
J=1 Hz, 1H), 7.08–7.03 (m, 1H), 6.99–6.94 (m, 1H), 6.24 (t, J=
12 Hz, 1H), 3.45–3.49 (m, 2H), 2.87 (t, J=14 Hz, 2H); MS (ESI): m/z:
390 [M+H]+; Anal. calcd for C22H19N3O4: C 67.86, H 4.92, N 10.79,
found: C 67.98, H 4.92, N 10.78.
Ethyl 7-amino-4-hydroxy-2-naphthoate (12a): A solution of 12a
and 12b (3.00 g, 12.97 mmol), obtained as previously described,[60]
in CH3OH (150 mL) was treated dropwise with a solution of cobalt-
(II) acetate tetrahydrate (4.00 g, 16.06 mmol) in AcOH/AcONa pH 5
buffer (20 mL) over 30 min. The resulting slurry was heated at 508C
for 3 h and then left at RT for an additional 3 h. The black solid
was filtered off and the solution was concentrated in vacuo. Satu-
rated aq NaHCO3 (150 mL) was added to the resulting oil and the
mixture was extracted with EtOAc (3ꢁ75 mL). The organic phase
was dried, filtered and concentrated in vacuo to give 12a as a
white solid (0.96 g, 33%); mp: 245.5–246.58C (dec); 1H NMR
(CDCl3): d=10.57 (s, 1H), 8.03 (d, J=8.0 Hz, 1H), 7.92 (d, J=1.2 Hz,
1H), 7.18 (d, J=1.2 Hz, 1H), 7.02–6.96 (m, 2H), 5.50 (br s, 2H), 4.40
(q, J=7.3 Hz, 2H), 1.41 (t, J=7.3 Hz, 3H); MS (ESI): m/z: 232 [M+
H]+.
5-(3-(7-Carboxy-5-hydroxynaphthalen-2-yl)ureido)-1H-indole-2-
carboxylic (6b): White solid (70%); mp: >2908C; 1H NMR
([D6]DMSO): d=11.62 (s, 1H), 10.31 (s, 1H), 8.91 (s, 1H), 8.67 (s,
1H), 8.08–8.03 (m, 2H), 7.88–7.85 (m, 2H), 7.63 (d, J=8 Hz, 1H),
7.37 (d, J=8 Hz, 1H), 7.26 (s, 1H), 7.21 (s, 1H), 7.02 (s, 1H); MS
(ESI): m/z: 406 [M+H]+; Anal. calcd for C25H23N3O6: C 65.07, H 5.02,
N 9.11, found: C 65.19, H 5.01, N 9.10.
7,7’-Carbonylbis(azanediyl)bis(4-hydroxy-2-naphthamide) (1c): A
mixture of compound 1a (0.3 g, 0.61 mmol) and NH4OH (30% w/w,
20 mL) was stirred at RT for 12 h. The solution was then acidified
to pH 5 with 3n aq HCl and isolation of the light brown solid by
filtration gave 1c (0.157 g, 60%); mp: >2908C; 1H NMR
([D6]DMSO): d=10.23 (s, 2H), 9.13 (s, 2H), 8.15 (s, 2H), 8.07 (d, J=
9 Hz, 2H), 7.98 (s, 2H), 7.81 (s, 2H), 7.54 (d, J=9 Hz, 2H), 7.31 (s,
2H), 7.18 (s, 2H); MS (ESI): m/z: 431 [M+H]+; Anal. calcd for
C23H18N4O5: C 64.18, H 4.22, N 13.02, found: C 64.29, H 4.21, N
13.00.
Diethyl 7,7’-carbonylbis(azanediyl)bis(4-hydroxy-2-naphthoate)
(1a): Diphenylcarbonate (0.463 g, 2.16 mmol) and DMAP (0.053 g,
0.43 mmol) were added to a solution of 12a (1.0 g, 4.32 mmol) in
chlorobenzene (20 mL) and the mixture was refluxed for 72 h. The
solvent was removed in vacuo, and the residue was washed with
petroleum benzene (2ꢁ20 mL). The crude was dissolved in EtOAc
(100 mL) and washed with 3n aq HCl (3ꢁ70 mL). The organic
phase was washed with brine, dried, filtered and concentrated in
vacuo. Purification by column chromatography on silica gel
(CH2Cl2/CH3OH, 98:2!90:10) gave 1a as a white solid (0.633 g,
60%); mp: 266.2–266.88C (dec); 1H NMR ([D6]DMSO): d=10.43 (s,
2H), 9.11 (s, 2H), 8.16 (s, 2H), 8.10 (d, J=8 Hz, 2H), 7.94 (s, 2H),
8.33 (d, J=8 Hz, 2H), 7.25 (s, 2H), 6.45 (q, J=14 Hz, 4H), 1.36 (t,
J=14 Hz, 6H); MS (ESI): m/z: 489 [M+H]+; Anal. calcd for
C27H24N2O7: C 66.39, H 4.95, N 5.73, found: C 66.52, H 4.94, N 5.72.
Ethyl 7-(3-ethoxy-3-oxopropanamido)-4-hydroxy-2-naphthoate
(13): A suspension of 12a (1.0 g, 4.32 mmol) in diethyl malonate
(6.5 mL) was irradiated for 30 min at 300 W keeping the tempera-
ture below 1008C (air flow cooling). EtOAc (100 mL) was added to
the crude and the solution was washed with 6n aq HCl (3ꢁ
70 mL). The organic phase was dried, filtered and concentrated in
vacuo. Purification by column chromatography on silica gel
(CH2Cl2/CH3OH, 99:1!95:5) gave 13 as a light yellow solid (1.34 g,
7,7’-Carbonylbis(azanediyl)bis(4-hydroxy-2-naphthoic acid) (1b):
Pyridine (1.22 mmol, 98 mL) was added to a solution of ester 1a
(0.3 g, 0.61 mmol) in THF (4 mL). The reaction was treated drop-
wise with aq 1n NaOH (12 mL) and the mixture was stirred at RT
until starting material disappeared (silica TLC; AcOEt/AcOH, 99:1).
THF was removed under a flow of N2 and 2n aq HCl (10 mL) was
added. Filtration of the mixture gave 1b as a white solid (0.260 g,
1
90%); mp: 201–2028C; H NMR ([D6]DMSO): d=10.47 (s, 1H), 10.44
(s, 1H), 8.29 (s, 1H), 8.10 (d, J=9 Hz, 1H), 7.92 (s, 1H), 7.64 (d, J=
9 Hz, 1H), 7.27 (s, 1H), 4.32 (q, J=14 Hz, 2H), 4.12 (q, J=14 Hz,
2H), 3.51 (s, 2H), 1.34 (t, J=14 Hz, 3H), 1.20 (t, J=14 Hz, 3H); MS
(ESI): m/z: 346 [M+H]+.
Ethyl 7-(3-(7-(ethoxycarbonyl)-5-hydroxynaphthalen-2-ylamino)-
3-oxopropanamido)-4-hydroxy-2-naphthoate (2a): Compound
12a (0.740 g, 3.19 mmol) was added to a solution of 13 (1.00 g,
2.90 mmol) in toluene (10 mL) and 1-methyl-2-pyrrolidinone (1 mL).
The mixture was irradiated (3ꢁ30 min) at 300 W keeping the tem-
perature below 1208C (air flow cooling). The reaction was concen-
trated in vacuo and redissolved in EtOAc (150 mL). The solution
was washed with 1n aq HCl (3ꢁ60 mL), dried, filtered and concen-
1
98%); mp: >2908C; H NMR ([D6]DMSO): d=10.37 (s, 2H), 9.09 (s,
2H), 8.13–8.10 (m, 4H), 7.93 (s, 2H), 7.67 (dd, J1 =9 Hz, J2 =1 Hz,
2H), 7.25 (d, J=1 Hz, 2H); MS (ESI): m/z: 433 [M+H]+; Anal. calcd
for C23H16N2O7: C 63.89, H 3.73, N 6.48, found: C 64.04, H 3.73, N
6.47. Acids 2b, 3b, 4b and 6b were obtained from the corre-
sponding esters 2a, 3a, 4a, and 6a, respectively, following the
same procedure.[62]
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
409

G. Sbardella, M. T. Bedford et al.
MED
trated in vacuo. Purification by column chromatography on silica
gel (CH2Cl2/CH3OH, 99:1!90:10) gave 2a as a white solid (0.770 g,
50%); mp: 282.2–282.78C; 1H NMR ([D6]DMSO): d=10.50 (s, 2H),
10.47 (s, 2H), 8.34 (s, 2H), 8.11 (d, J=9 Hz, 2H), 7.92 (s, 2H), 7.70
(dd, J1 =9 Hz, J2 =1 Hz, 2H), 7.27 (s, 2H), 4.32 (q, J=14 Hz, 4H),
3.61 (s, 2H), 1.33 (t, J=14 Hz, 6H); MS (ESI): m/z: 531 [M+H]+;
Anal. calcd for C29H26N2O8: C 65.65, H 4.94, N 5.28, found: C 65.76,
H 4.93, N 5.27.
([D6]DMSO): d=10.36 (s, 1H), 8.76 (s, 1H), 8.03–8.00 (m, 2H), 7.85
(s, 1H), 7.55 (dd, J1 =9 Hz, J2 =1 Hz, 1H), 7.19 (s, 1H), 6.34 (t, J=
9 Hz, 1H), 4.33 (q, J=14 Hz, 2H), 3.17–3.11 (m, 2H), 2.36–2.31 (m,
2H), 2.20 (s, 6H), 1.63–1.58 (m, 2H), 1.34 (t, J=14 Hz, 3H); MS (ESI):
m/z: 360 [M+H]+; Anal. calcd for C19H25N3O4: C 63.49, H 7.01, N
11.69, found: C 63.57, H 7.00, N 11.67.
Ethyl
5-(3-(7-(ethoxycarbonyl)-5-hydroxynaphthalen-2-yl)urei-
do)-1H-indole-2-carboxylate (6a): White solid (50%); mp: 210.8–
1
Ethyl 7-(4-(7-(ethoxycarbonyl)-5-hydroxynaphthalen-2-ylamino)-
4-oxobutanamido)-4-hydroxy-2-naphthoate (3a): A solution of
12a (1.0 g, 4.32 mmol) in acetone (6 mL) was cooled to ꢁ158C and
treated dropwise with a solution of succinyl dichloride (0.401 g,
2.59 mmol) in acetone (12 mL) over 30 min. The resulting solution
was stirred for 12 h at RT, then aq NaHCO3 (6 mL) was added and
the mixture concentrated in vacuo. The crude oil was redissolved
in aq NaHCO3 (100 mL) and extracted with EtOAc (3ꢁ60 mL). The
combined organic phases were washed with 2n aq HCl (3ꢁ
60 mL), dried, filtered and concentrated in vacuo to give a crude
solid. Purification by flash chromatography on silica gel (CH2Cl2/
CH3OH, 99:1!90:10) gave 3a as a white solid (0.470 g, 40%); mp:
211.78C; H NMR ([D6]DMSO): d=11.74 (s, 1H), 10.39 (s, 1H), 8.98
(s, 1H), 8.75 (s, 1H), 8.17 (s, 1H), 8.08 (d, J=9 Hz, 1H), 7.94–7.88
(m, 2H), 7.64 (d, J=8 Hz, 1H), 7.39 (d, J=8 Hz, 1H), 7.27 (d, J=
9 Hz, 1H), 7.23 (s, 1H), 7.07 (s, 1H), 4.35 (q, J=13 Hz, 4H), 1.38–
1.32 (m, 6H); MS (ESI): m/z: 462 [M+H]+; Anal. calcd for
C25H23N3O6: C 65.07, H 5.02, N 9.11, found: C 65.17, H 5.01, N 9.10.
(E)-3-(Methoxycarbonyl)-4-(4-nitrophenyl)but-3-enoic acid: Car-
boxyphosphorane 10[61] (13.0 g, 35.2 mmol) was suspended in dry
benzene (150 mL) and 4-nitrobenzaldehyde (5.74 g, 38.0 mmol)
was then added. The resulting mixture was stirred at RT for 48 h
and then extracted with saturated aq NaHCO3 (3ꢁ70 mL). The
aqueous phase was washed with Et2O, acidified with concd HCl
and the white solid formed was collected by filtration (8.60 g,
92%) and directly used in the following step without further purifi-
cation; mp: 158–1608C; 1H NMR ([D6]DMSO): d=8.30 (d, J=9 Hz,
2H), 7.86 (s, 1H), 7.69 (d, J=9 Hz, 2H), 3.78 (s, 3H), 3.43 (s, 2H); MS
(ESI): m/z: 266 [M+H]+.
1
>2908C; H NMR ([D6]DMSO): d=10.50 (s, 2H), 10.37 (s, 2H), 8.33
(s, 2H), 8.07 (d, J=9 Hz, 2H), 7.87 (s, 2H), 7.68 (d, J=9 Hz, 2H),
7.31 (s, 2H), 4.31 (q, J=21 Hz, 4H), 2.83 (s, 2H), 1.33 (t, J=21 Hz,
6H); MS (ESI): m/z: 545 [M+H]+; Anal. calcd for C25H23N3O6: C
65.07, H 5.02, N 9.11, found: C 65.15, H 5.01, N 9.10.
Ethyl 4-hydroxy-7-(2,2,2-trichloroacetamido)-2-naphthoate (14):
Et3N (0.674 g, 6.67 mmol) was added to a solution of 12a (1.0 g,
4.32 mmol) in dry CH2Cl2 (100 mL) under a flow of N2. After 30 min,
trichloroacetylchloride (1.212 g, 6.67 mmol) was added dropwise
over 10 min under a flow of N2 and the reaction was stirred at RT
for 4 h. The solution was then washed with saturated aq NaHCO3
(3ꢁ40 mL) and 3n aq HCl (3ꢁ40 mL), then dried, filtered and con-
centrated in vacuo. Purification by column chromatography on
silica gel (CH2Cl2/CH3OH, 99:1!90:10) gave 14 as a light yellow
solid (1.34 g, 83%); mp: 232.0–232.78C; 1H NMR ([D6]DMSO): d=
11.06 (s, 1H), 10.56 (s, 1H), 8.29 (s, 1H), 8.16 (d, J=9 Hz, 1H), 7.99
(s, 1H), 7.80 (d, J=9 Hz, 1H), 7.33 (s, 1H), 4.34 (q, J=14 Hz, 2H),
1.34 (t, J=14 Hz, 3H); MS (ESI+): m/z (%): 376 (100) [M+H]+, 378
(96) [M+H+2]+, 380 (36) [M+H+4]+, 382 (4) [M+H+6]+.
Methyl 4-acetoxy-6-nitro-2-naphthoate (15): A round-bottomed
flask containing a magnetic stirring bar and fitted with a reflux
condenser was charged with a mixture of (E)-3-(methoxycarbonyl)-
4-(4-nitrophenyl)but-3-enoic acid (5.0 g, 18.85 mmol), Ac2O (35 mL)
and NaOAc (2.30 g, 28.27 mmol). The flask was subjected to MW ir-
radiation (power 300 W) for 5 min keeping temperature below
1208C (air flow cooling). The reaction mixture was filtered and the
filtrate was concentrated to give 15 as a yellow solid (5.17 g, 95%);
1
mp: 167–1688C; H NMR (CDCl3): d=8.89 (d, J=2 Hz, 1H), 8.62 (s,
1H), 8.38 (dd, J1 =9 Hz, J2 =2 Hz, 1H), 8.17 (d, J=9 Hz, 1H), 8.05 (s,
1H), 4.05 (s, 3H), 2.59 (s, 3H); MS (ESI): m/z: 290 [M+H]+.
Methyl 4-acetoxy-6-amino-2-naphthoate (16): Pd/C (5 wt% on ac-
tivated carbon, 0.1 equiv) was added to a solution of 15 (3.0 g,
10.37 mmol) in EtOH (100 mL) and the reaction was stirred under
H2 (1 atm, balloon) for 2 h. The reaction mixture was filtered and
concentrated to give 16 as a light yellow solid (2.66 g, 99%); mp:
Ethyl
7-(3-(2-(1H-indol-3-yl)ethyl)ureido)-4-hydroxy-2-naph-
thoate (4a): Na2CO3 (1.40 g, 13.25 mmol) and tryptamine (0.467 g,
2.91 mmol) were added to a solution of 14 (1.0 g, 2.65 mmol) in
dry DMF (10 mL) and the resulting mixture was heated at 1508C
for 1 h in a sealed tube. The reaction was concentrated in vacuo
and then redissolved in EtOAc (150 mL). The solution was then
washed with 2n aq HCl (3ꢁ50 mL) and saturated aq NaHCO3 (3ꢁ
50 mL), dried, filtered and concentrated in vacuo. Purification by
column chromatography on silica gel (CH2Cl2/CH3OH, 99:1!90:10)
gave 4a as a white solid (0.696 g, 63%); mp: 214.7–215.58C;
1H NMR ([D6]DMSO): d=10.82 (s, 1H), 10.39 (br s, 1H), 8.79 (d, J=
4 Hz, 1H), 8.03–7.99 (m, 2H), 7.84 (s, 1H), 7.59–7.51 (m, 2H), 7.34
(d, J=9 Hz, 1H), 7.19–7.18, (m, 2H), 7.06–7.03 (m, 1H), 6.97–6.95
(m, 1H), 6.29 (t, J=12 Hz, 1H), 4.31 (q, J=14 Hz, 2H), 4.43–4.41 (m,
2H), 2.91 (t, J=25 Hz, 2H), 1.33 (t, J=14 Hz, 3H); MS (ESI): m/z:
418 [M+H]+; Anal. calcd for C24H23N3O4: C 69.05, H 5.55, N 10.07,
found C 69.14, H 5.54, N 10.06. Derivatives 5a and 6a were ob-
tained following the same procedure starting from 14 and N1,N1-di-
methylpropane-1,3-diamine or ethyl 5-amino-1H-indole-2-carboxyl-
ate, respectively, as the nucleophile.
1
195–1968C; H NMR ([D6]DMSO): d=8.28 (s, 1H), 7.86 (d, J=9 Hz,
1H), 7.54 (s, 1H), 7.06 (d, J=9 Hz, 1H), 6.79 (s, 1H), 6.04 (s, 2H),
3.87 (s, 3H), 2.42 (s, 3H); MS (ESI): m/z: 260 [M+H]+.
Dimethyl
6,6’-carbonylbis(azanediyl)bis(4-hydroxy-2-naphth-
oate) (7a): Diphenylcarbonate (0.411 g, 1.92 mmol) and DMAP
(0.046 g, 0.38 mmol) were added to a solution of 16 ( 1.0 g,
3.85 mmol) in chlorobenzene (20 mL) and the mixture was refluxed
for 72 h. The solvent was removed in vacuo and the resultant dark
brown solid was washed with petroleum benzene (2ꢁ20 mL). The
crude was redissolved in EtOH (50 mL) and K2CO3 (0.585 g,
4.23 mmol) was added. The mixture was heated at 708C for 2 h,
then concentrated and redissolved in EtOAc (100 mL). The organic
phase was washed with 3n aq HCl (3ꢁ70 mL) and brine, dried, fil-
tered and concentrated. Purification of the crude by column chro-
matography on silica gel (CH2Cl2/CH3OH, 98:2!90:10) gave 7a as
a
white solid (0.443 g, 50%); mp: 276.6–277.68C; 1H NMR
([D6]DMSO): d=10.39 (s, 2H), 9.13 (s, 2H), 8.42 (d, J=2 Hz, 2H),
8.00 (s, 2H), 7.95 (d, J=9 Hz, 2H), 7.56 (dd, J1 =9 Hz, J2 =2 Hz, 2H),
7.33 (s, 2H), 3.85 (s, 6H); MS (ESI): m/z: 461 [M+H]+; Anal. calcd
Ethyl 7-(3-(2-(dimethylamino)propyl)ureido)-4-hydroxy-2-naph-
thoate (5a): White solid (63%); mp: 161–1638C; 1H NMR
410
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
for C25H20N2O7: C 65.21, H 4.38, N 6.08, found: C 65.29, H 4.37, N
6.07.
characterization purposes, analytical samples of 18a and 18b were
obtained by column chromatography on silica gel (CH2Cl2/CH3OH,
99:1!95:5); 18a: mp: 162–1638C; 1H NMR ([D6]DMSO): d=11.25
(s, 1H), 8.56 (s, 1H), 8.32 (s, 1H), 8.27 (d, J=9 Hz, 1H), 8.03 (d, J=
9 Hz, 1H), 7.80 (s, 1H), 3.93 (s, 3H), 2.51 (s, 3H); MS (ESI): m/z: 404
[M+H]+; 18b: mp: 225–2268C; 1H NMR ([D6]DMSO): d=11.13 (s,
1H), 10.63 (s, 1H), 8.55 (s, 1H), 8.07–8.05 (m, 2H), 7.85 (dd, J1 =
9 Hz, J2 =2 Hz, 1H), 7.39 (s, 1H), 3.89 (s, 3H); MS (ESI): m/z: 362
[M+H]+.
Acids 7b, 8b and 9b were obtained from the corresponding
esters 7a, 8a, and 9a, respectively, following the same general
procedure described above for compounds 1b, 2b, 3b, 4b and
6b.
6,6’-Carbonylbis(azanediyl)bis(4-hydroxy-2-naphthoic acid) (7b):
1
White solid (90%); mp: >2908C; H NMR ([D6]DMSO): d=10.33 (s,
2H), 9.13 (s, 2H), 8.44 (d, J=2 Hz, 2H), 8.00 (s, 2H), 7.96 (d, J=
9 Hz, 2H), 7.59 (dd, J1 =9 Hz, J2 =2 Hz, 2H), 7.35 (s, 2H); MS (ESI):
m/z: 433 [M+H]+; Anal. calcd for C23H16N2O7: C 63.89, H 3.73, N
6.48, found: C 63.95, H 3.73, N 6.47.
Methyl 6-(3-(2-(1H-indol-3-yl)ethyl)ureido)-4-hydroxy-2-naphtho-
ate (9a): Prepared from a crude mixture of 18a and 18b (1.0 g,
2.77 mmol) following the same procedure described above for the
preparation of 4a, to yield a white solid (0.704 g, 63%); mp:
1
218.8–219.88C; H NMR ([D6]DMSO): d=10.83 (s, 1H), 10.26 (s, 1H),
6-(3-(6-Carboxy-8-hydroxynaphthalen-2-ylamino)-3-oxopropan-
8.87 (s, 1H), 8.26 (d, J=2 Hz, 1H), 7.95 (s, 1H), 7.86 (d, J=9 Hz,
1H), 7.57 (d, J=8 Hz, 1H), 7.51 (dd, J1 =9 Hz, J2 =2 Hz, 1H), 7.33 (d,
J=8 Hz, 1H), 7.27 (s, 1H), 7.17 (d, J=2 Hz, 1H), 7.06–7.03 (m, 1H),
6.97–6.93 (m, 1H), 6.22 (t, J=12 Hz, 1H), 3.84 (s, 3H), 3.44–3.40 (m,
2H), 2.87 (t, J=14 Hz, 2H); MS (ESI): m/z: 404 [M+H]+; Anal. calcd
for C23H21N3O4: C 68.47, H 5.25, N 10.42, found: C 68.58, H 5.24, N
10.41.
amido)-4-hydroxy-2-naphthoic acid (8b): White solid (85%); mp:
>2908C; H NMR (CD3OD): d=8.54 (d, J=2 Hz, 2H), 8.09 (s, 2H),
7.91 (d, J=9 Hz, 2H), 7.76 (dd, J1 =9 Hz, J2 =2 Hz, 2H), 7.39 (s, 2H),
3.51 (s, 2H); MS (ESI): m/z: 475 [M+H]+; Anal. calcd for C25H18N2O8:
C 63.29, H 3.82, N 5.90, found: C 63.38, H 3.81, N 5.89.
1
6-(3-(2-(1H-Indol-3-yl)ethyl)ureido)-4-hydroxy-2-naphthoic acid
(9b): White solid (98%); mp: >2908C; 1H NMR ([D6]DMSO): d=
10.86 (s, 1H), 10.21 (s, 1H) 8.87 (s, 1H), 8.29 (d, J=2 Hz, 1H), 7.94
(s, 1H), 7.86 (d, J=9 Hz, 1H), 7.61 (d, J=8 Hz, 1H), 7.53 (dd, J1 =
9 Hz, J2 =2 Hz, 1H), 7.37 (d, J=8 Hz, 1H), 7.30 (d, J=1 Hz, 1H),
7.21 (d, J=1 Hz, 1H), 7.09–7.05 (m, 1H), 7.00–6.95 (m, 1H), 6.24 (t,
J=12 Hz, 1H), 3.47–3.43 (m, 2H), 2.91 (t, J=14 Hz, 2H); MS (ESI):
m/z: 390 [M+H]+; Anal. calcd for C22H19N3O4: C 67.86, H 4.92, N
10.79, found: C 67.96, H 4.91, N 10.78.
Biochemistry
Preparation of GST-RmtA and GST-PRMT1 fusion proteins
GST-PRMT1 fusion protein was expressed in E. coli BL21 cells. A cul-
ture of transformed E. coli BL21 cells was grown overnight in
10 mL of lysogeny broth (LB)[72] with ampicillin (100 mgmLꢁ1). Fresh
LB (100 mL) with antibiotic was added and the culture was agitat-
ed for 1 h at 378C. Then iso-propyl-b-d-1-thiogalactopyranoside
(IPTG) was added to give a final concentration of 0.1 mm, followed
by additional shaking at 308C for 4 h. The bacterial culture was
then centrifuged at 5000 rpm for 5 min at 48C, and the superna-
tant was discarded. The pellet was resuspended in 500 mL of cold
PBS (137 mm NaCl, 2.7 mm KCl, 4.3 mm Na2HPO4, 1.4 mm KH2PO4,
pH 7.4) and sonicated for 30 s. The suspension was spun down at
5000 rpm for 10 min at 48C. During that time, glutathione sephar-
ose beads (GE Healthcare) were washed once with ice-cold PBS,
then 100 mL of clean beads were placed into a 1.5 mL microcentri-
fuge tube and the supernatant added; the beads are rocked for 3–
5 h at 48C. The beads were washed three times with ice-cold PBS.
Fresh glutathione-reduced buffer was prepared by mixing elution
buffer (100 mm Tris-HCl, pH 8.0; 120 mm NaCl) with glutathione re-
duced (0.01 gmLꢁ1). Glutathione-reduced buffer (100 mL) was
added to the beads and rocked for 2 h at 48C. Finally, the beads
were spun down and the supernatant collected.
Methyl 4-acetoxy-6-(3-ethoxy-3-oxopropanamido)-2-naphthoate
(17): Obtained from 16 (1.0 g, 3.85 mmol) following the same pro-
cedure described above for the preparation of 13, to yield a white
solid (1.30 g, 90%); mp: 170.5–171.98C; 1H NMR ([D6]DMSO): d=
10.66 (s, 1H), 8.52 (s, 1H), 8.32 (s, 1H), 8.20 (d, J=9 Hz, 1H), 7.80–
7.76 (m, 2H), 4.16 (q, J=14 Hz, 2H), 3.93 (s, 3H), 3.56 (s, 2H), 2.47
(s, 3H), 1.22 (t, J=14 Hz, 3H); MS (ESI): m/z: 374 [M+H]+.
Methyl 4-hydroxy-6-(3-(6-(methoxycarbonyl)-8-hydroxynaphtha-
len-2-ylamino)-3-oxopropanamido)-2-naphthoate (8a): Com-
pound 16 (0.762 g, 2.94 mmol) was added to a solution of 17
(1,00 g, 2.68 mmol) in toluene (10 mL) and 1-methyl-2-pyrrolidi-
none (1 mL) and the mixture was irradiated (3ꢁ30 min) at 300 W,
keeping temperature below 1208C (air flow cooling). The reaction
was concentrated in vacuo and redissolved in EtOAc (150 mL). The
organic phase was washed with 1n aq HCl (3ꢁ60 mL), dried, fil-
tered and concentrated in vacuo. The crude residue was dissolved
in EtOH (50 mL) and K2CO3 (0.407 g, 2.94 mmol) was added. The
mixture was heated at 708C for 2 h, then concentrated and redis-
solved in EtOAc (100 mL). The organic phase was washed with 3n
aq HCl (3ꢁ70 mL) and brine, then dried, filtered and concentrated.
Purification by column chromatography on silica gel (CH2Cl2/
CH3OH, 99:1!90:10) gave 8a as a white solid (0.673 g, 50%); mp:
The RmtA coding sequence[64] was cloned into a pGEX-5X-1 expres-
sion vector (GE Healthcare). RmtA protein was expressed in BL21
cells in LB medium. 250 mL cultures (A600 =0.4) were induced with
a final concentration of 1 mm IPTG and grown for 4 h at 378C.
After centrifugation of cells at 4000 g, the pellet was resuspended
in 6 mL of GST-binding buffer (140 mm NaCl, 2.7 mm KCl, 10 mm
Na2HPO4, 1.8 mm KH2PO4, pH 7.3) containing one protease inhibitor
tablet (Complete, Roche, Mannheim, Germany) per 50 mL of buffer.
For cell lysis, lysozyme was added at a final concentration of
5 mgmLꢁ1 binding buffer and cells were passed through a French
press with pressure setting of 1000 psi. The resulting lysate was
centrifuged at 20000 g for 10 min at 48C. GST fusion protein was
purified from soluble extracts by binding to a GST-HiTrap column
(GE Healthcare). Proteins were eluted with 50 mm Tris-HCl, 10 mm
1
287.8–288.88C; H NMR ([D6]DMSO): d=10.51 (s, 2H), 10.44 (s, 2H),
8.55 (d, J=2 Hz, 2H), 8.01 (s, 2H), 7.98 (d, J=9 Hz, 2H), 7.73 (dd,
J1 =9 Hz, J2 =2 Hz, 2H), 7.32 (s, 2H), 3.86 (s, 6H), 3.60 (s, 2H); MS
(ESI): m/z: 503 [M+H]+; Anal. calcd for C27H22N2O8: C 64.54, H 4.41,
N 5.58, found: C 64.61, H 4.40, N 5.57.
Methyl
4-acetoxy-6-(2,2,2-trichloroacetamido)-2-naphthoate
(18a) and methyl 4-hydroxy-6-(2,2,2-trichloroacetamido)-2-
naphthoate (18b): Title compounds were obtained as a mixture
starting from 16 (1.0 g, 3.85 mmol) following the same procedure
described above for the preparation of 14, to yield a crude white
solid (1.14 g, 82%) that was directly used in the following step. For
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
411

G. Sbardella, M. T. Bedford et al.
MED
reduced glutathione, pH 8.0 and assayed for histone methyltrans-
ferase activity.
diluted HCl/EtOH 9:1, pH 2.0–2.5, GE Healthcare) and PBS (137 mm
NaCl, 2.7 mm KCl, 4.3 mm Na2HPO4, 1.4 mm KH2PO4, pH 7.4). To de-
termine the specificity of their inhibitory activities, test compounds
were incubated with GST-PRMT1 (10ꢁ10ꢁ8 m) and histone H4 (all
1.5ꢁ10ꢁ6 m), GST-PRMT3 (9.0ꢁ10ꢁ8 m) and GAR (4.1ꢁ10ꢁ7 m), GST-
PRMT4 (9.8ꢁ10ꢁ8 m) and histone H3 (all 1.1ꢁ10ꢁ6 m), GST-PRMT6
(9.7ꢁ10ꢁ8 m) and histone H3 and SET 7/9 (1.3ꢁ10ꢁ7 m) and histo-
ne H3. Histones were purchased from Roche. Substrates (0.5 mg,
concentration range from 1.1ꢁ10ꢁ6 m to 4.06ꢁ10ꢁ7 m) were incu-
bated with recombinant enzymes (0.2 mg, concentration range
from 1.3ꢁ10ꢁ7 m to 10ꢁ10ꢁ8 m) in the presence of 0.5 mg [3H]SAM
(0.42 mm) and 50 mm of compound for 90 min at 308C in a final
volume of 30 mL. Reactions were run on a 10% SDS-PAGE, trans-
ferred to a polyvinylidene fluoride (PVDF) membrane (Millipore),
sprayed with Enhanceꢃ (PerkinElmer) and exposed to film over-
night. Band intensities were calculated using a Kodak Image Sta-
tion 440 and 1D Image Analysis Software (Eastman Kodak Co.).
GST-PRMT3, GST-PRMT4, and GST-PRMT6 have been described pre-
viously.[73] Unless stated otherwise, all chemicals/reagents were pur-
chased from Sigma–Aldrich.
Protein methyltransferase assays
RmtA and PRMT1 (histone substrate) inhibitory assays and determi-
nation of IC50 values: For inhibition assays and determination of IC50
values, affinity-purified GST-RmtA and GST-PRMT1 fusion proteins
were used as the enzyme source. HMT activities were assayed as
described previously[64] using chicken erythrocyte core histones as
substrate. GST-RmtA and GST-PRMT1 fusion proteins (500 ng) were
incubated with five different concentrations of each compound for
15 min at RT and 2.5 mL of chicken core histones (8 mgmLꢁ1) and
1 mL of [3H]-S-adenosyl-l-methionine ([3H]SAM, 0.55 mCi) were
added, resulting in concentrations of 25 and 0.13 mm, respectively.
This mixture was incubated for 30 min at 308C. The reaction was
stopped by TCA precipitation (25% final concentration) and sam-
ples were kept on ice for 20 min. Whole sample volumes were col-
lected onto a glass fiber filter (Whatman GF/F) preincubated with
25% TCA. Filters were washed three times with 3 mL of 25% TCA
and then three times with 1 mL of EtOH. After drying the filters for
10 min at 708C, radioactivity was measured by liquid scintillation
spectrophotometry (3 mL scintillation cocktail). Assays were per-
formed in triplicate. IC50 values were determined by fitting activity
data to Equation (1) with nonlinear regression analysis using
SigmaPlot v. 10.0 (Systat Software, San Jose, USA).
In vivo methylation assay: HeLa cells were labeled using a previous-
ly described in vivo methylation assay.[36,74] Briefly, HeLa cells were
grown in 12-well plates and then transiently transfected with
d2GFP-Npl3. For this GFP-Npl3 fusion construct, full-length Npl3
from pGEX-Npl3 was amplified by PCR primer 1 (5’-GTGGGATCC-
CACCATGTCTGAAGCTCAAGA-3’) and primer 2 (5’-AGAGGATC-
CAACCTGGTTGGTGATCTTTCACG-3’). The 5’ primer introduced a
mammalian “Kozak” sequence (CACC) just upstream of the initiator
sequence, ATG. The fragments were subcloned in pd2EGFP-N1
(Clontech, Palo Alto, USA) to generate an N-terminal fusion of
Npl3. Three hours post-transfection, the cells were incubated with
the indicated compounds (as DMSO solutions) for 24 h. The cells
were lysed in a “mild” buffer (RIPA buffer; 150 mm NaCl, 5 mm
EDTA, 1% Triton X-100, 10 mm Tris-HCl, pH 7.5) and Western analy-
sis was performed with either the 1E4 antibody[68] or aGFP anti-
body (Clontech, Palo Alto, USA). The effects of the compounds on
GFP-Npl3 methylation status were established with the methyl-spe-
cific antibody, 1E4. The aGFP antibody showed the protein levels
of GFP-Npl3.
1
fractional enzyme activity ¼
ð1Þ
½lꢃ
1 þ IC
50
PRMT1 (Npl3p substrate) inhibitory assay: A colorimetric assay was
used as previously described[36] to determine inhibitory activities of
test compounds against hPRMT1 using Npl3p as a nonhistone sub-
strate. Briefly, 50 mL of a 10 mgmLꢁ1 solution of GST-Npl3 protein in
Tris-buffered saline (TBS) was added to each well of a clear 96-mi-
crotiter plate (Greiner Bio-one) with high binding affinity. After in-
cubation overnight at 48C, the plate was rinsed twice with TBST
(25 mm Tris (pH 7.5), 150 mm NaCl, 0.1% Tween 20), and once with
20 mm TBS (pH 8.0). The compounds were added (1 mL of 0.5 mm
or 2.5 mm DMSO solutions) to the GST-Npl3-coated plates. After a
15 min incubation, 100 ng of hPRMT1 (0.03 mm) and 1 mL of SAM
(0.1 mm stock, Sigma–Aldrich) were added to each well in a final
volume of 50 mL of 20 mm Tris buffer (pH 8.0) and incubated at
308C for 1 h. The plate was washed with TBST and then blocked
with 5% BSA in TBST buffer for 1.5 h at RT. Anti-Npl3 antibody 1E4
(1:1000)[68] was then added (50 mL) to each well and the plate was
shaken for 2 h at RT. The wells were washed three times with TBST.
Anti-mouse HRP IgG (GE Healthcare; 1:5000 dilution in TBST, 5%
BSA) was then added as a secondary antibody to each well, and
the mixture was incubated for 1 h at RT. The wells were again
rinsed three times with TBST. The peroxidase substrate 2,2’-(hydra-
zine-1,2-diylidene)bis(3-ethyl-2,3-dihydrobenzo[d]thiazole-6-sulfonic
acid) (ABTS, Roche) was added to each well (50 mL of 1 mgmLꢁ1 so-
lution), and the mixture was incubated for 30 min. The absorbance
was measured at 405 nm with a plate reader (Bio-Rad).
Molecular docking
All compounds were built, starting from ASCII text, using the
standalone version of PRODRG,[75,76] in conjunction with the GRO-
MACS suite.[77] Docking studies were performed on Autodock 3.0.5
using a grid spacing of 0.375 ꢂ and 39ꢁ50ꢁ56 number of points
that embraced both the SAM and Arg binding sites. The grid was
centered on the mass centre of the experimental bound SAM and
Arg substrates. The GA-LS method was adopted using the default
setting, except for the maximum number of energy evaluations
which was increased from 250000 to 2500000. Autodock generated
100 possible binding conformations for each molecule that were
clustered using a tolerance of 2.0 ꢂ. The AutoDockTool (ADT)
graphical interface[78] was used to prepare the enzyme PDBQS file.
The protein atom charges as calculated during the complex mini-
mization were retained for the docking calculations.
The SAM and N-acetyl,O-methyl-capped Arg PRODRG generated
conformations were docked into the RmtA structure to assess the
docking protocol. The SAM and Arg were docked back into their
binding sites and the Xscore[71] selected conformations showed
root mean square deviations (RMSD) of 0.60 and 1.47, respectively.
The Autodock scoring function did not select conformations with
lower RMSD values. The same trials were conducted using the
DOCK program, however, attempts to dock these substrates back
in to the RmtA binding site were unsuccessful, therefore we con-
PRMT1, PRMT3, PRMT4/CARM1, PRMT6 and SET7/9 in vitro methyla-
tion assays: In vitro methylation assays were described in detail
previously.[36] Briefly, all methylation reactions were carried out in
the presence of [3H]SAM (0.42 mm, 79 Cimmolꢁ1, stock solution in
412
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414

Carboxy Analogues of AMI-1
[12] S. S. Koh, D. Chen, Y.-H. Lee, M. R. Stallcup, J. Biol. Chem. 2001, 276,
1089–1098.
[13] B. D. Strahl, S. D. Briggs, C. J. Brame, J. A. Caldwell, S. S. Koh, H. Ma, R. G.
Cook, J. Shabanowitz, D. F. Hunt, M. R. Stallcup, C. D. Allis, Curr. Biol.
2001, 11, 996–1000.
[14] H. Wang, Z.-Q. Huang, L. Xia, Q. Feng, H. Erdjument-Bromage, B. D.
Strahl, S. D. Briggs, C. D. Allis, J. Wong, P. Tempst, Y. Zhang, Science
2001, 293, 853–857.
tinued to use Autodock. ADT was used to analyze the docking re-
sults and Chimera 1.3 (build 2577)[79] was used to produce the
images.
The outcomes of the docking experiments were supported by the
application of our previous three-dimentional QSAR models,[40]
yielding low errors of prediction (Supporting Information).
[15] S. L. Anzick, J. Kononen, R. L. Walker, D. O. Azorsa, M. M. Tanner, X.-Y.
Guan, G. Sauter, O.-P. Kallioniemi, J. M. Trent, P. S. Meltzer, Science 1997,
277, 965–968.
Glossary
[16] J. Torchia, D. W. Rose, J. Inostroza, Y. Kamei, S. Westin, C. K. Glass, M. G.
Rosenfeld, Nature 1997, 387, 677–684.
AIB1, amplified in breast cancer-1; BL21, Escherichia coli B cells lack-
ing the Lon protease; BSA, bovine serum albumin; CARM1, coacti-
vator-associated arginine methyltransferase 1; CBP, CREB binding
protein; CREB, c-AMP response-element binding; aDMA, asymmet-
rical dimethylarginine; sDMA, symmetrical dimethylarginine;
FBXO11; F-box protein 11; GAR, glycine- and arginine-rich; GST, glu-
tathione-S-transferase; HKMT, histone lysine methyltransferase;
hnRNP, heterogeneous nuclear ribonucleoprotein; HRP, horseradish
peroxidase; IPTG, isopropyl-b-d-1-thiogalactopyranoside; LB, lysog-
eny broth; MMA, monomethylarginine; NF-kB, nuclear factor kB;
Npl3, nuclear shuttling protein; p300, E1A binding protein
300 kDa; PBS, phosphate buffer saline; PRMT, protein arginine
methyltransferase; PVDF, polyvinylidene fluoride; RmtA, fungal ar-
ginine methyltransferase A; SAM, S-adenosyl methionine; SDS-
PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis;
SET, Su(var.) 3–9, enhancer-of-zeste and trithorax; TBS, Tris-buffered
saline; TBST, Tris-buffered saline Tween-20; TCA, trichloroacetic
acid.
[17] S.-K. Lee, S. L. Anzick, J.-E. Choi, L. Bubendorf, X.-Y. Guan, Y.-K. Jung, O. P.
Kallioniemi, J. Kononen, J. M. Trent, D. Azorsa, B.-H. Jhun, J. H. Cheong,
Y. C. Lee, P. S. Meltzer, J. W. Lee, J. Biol. Chem. 1999, 274, 34283–34293.
[18] Y. Zhu, C. Qi, S. Jain, M. M. Le Beau, R. Espinosa, G. B. Atkins, M. A. Lazar,
A. V. Yeldandi, M. S. Rao, J. K. Reddy, Proc. Natl. Acad. Sci. USA 1999, 96,
10848–10853.
[19] H. Hong, C. Kao, M.-H. Jeng, J. N. Eble, M. O. Koch, T. A. Gardner, S.
Zhang, L. Li, C.-X. Pan, Z. Hu, G. T. MacLennan, L. Cheng, Cancer 2004,
101, 83–89.
[20] S. Majumder, Y. Liu, O. H. Ford III, J. L. Mohler, Y. E. Whang, Prostate
2006, 66, 1292–1301.
[21] R. H. Bçger, K. Sydow, J. Borlak, T. Thum, H. Lenzen, B. Schubert, D.
Tsikas, S. M. Bode-Bçger, Circ. Res. 2000, 87, 99–105.
[22] C. T. L. Tran, J. M. Leiper, P. Vallance, Atheroscler. Suppl. 2003, 4, 33–40.
[23] P. Vallance, J. Leiper, Arterioscler. Thromb. Vasc. Biol. 2004, 24, 1023–
1030.
[24] J. Leiper, J. Murray-Rust, N. McDonald, P. Vallance, Proc. Natl. Acad. Sci.
USA 2002, 99, 13527–13532.
[25] A. Leone, S. Moncada, P. Vallance, A. Calver, J. Collier, Lancet 1992, 339,
572–575.
[26] J. Leiper, M. Nandi, B. Torondel, J. Murray-Rust, M. Malaki, B. O’Hara, S.
Rossiter, S. Anthony, M. Madhani, D. Selwood, C. Smith, B. Wojciak-Sto-
thard, A. Rudiger, R. Stidwill, N. Q. McDonald, P. Vallance, Nat. Med.
2007, 13, 198–203.
[27] X. Chen, F. Niroomand, Z. Liu, A. Zankl, H. A. Katus, L. Jahn, C. P. Tiefen-
bacher, Basic Res. Cardiol. 2006, 101, 346–353.
[28] D. Cheng, J. Cꢄtꢅ, S. Shaaban, M. T. Bedford, Mol. Cell 2007, 25, 71–83.
[29] J. Lee, M. T. Bedford, EMBO Rep. 2002, 3, 268–273.
[30] T. Fujiwara, Y. Mori, D. L. Chu, Y. Koyama, S. Miyata, H. Tanaka, K. Yachi,
T. Kubo, H. Yoshikawa, M. Tohyama, Mol. Cell. Biol. 2006, 26, 2273–2285.
[31] P. O. Hassa, M. Covic, M. T. Bedford, M. O. Hottiger, J. Mol. Biol. 2008,
377, 668–678.
[32] N. Troffer-Charlier, V. Cura, P. Hassenboehler, D. Moras, J. Cavarelli, EMBO
J. 2007, 26, 4391–4401.
[33] X. Zhang, X. Cheng, Structure 2003, 11, 509–520.
[34] X. Zhang, L. Zhou, X. Cheng, EMBO J. 2000, 19, 3509–3519.
[35] O. Obianyo, T. C. Osborne, P. R. Thompson, Biochemistry 2008, 47,
10420–10427.
Acknowledgements
Many thanks to Professors Gabriele Cruciani and Sergio Clementi
(Molecular Discovery and MIA srl) for the use of the GOLPE pro-
gram in their chemometric laboratory (University of Perugia,
Italy) and for providing the GRID program. This work was partial-
ly supported by grants from Regione Campania 2003, LR 5/02
(G.S.), Ministero dell’Universitꢀ e della Ricerca Scientifica e Tecno-
logica -PRIN 2007, (G.S.), Universitꢀ di Salerno (G.S.), Ministero
dell’Universitꢀ e della Ricerca Scientifica e Tecnologica -PRIN
(A. M.), RETI FIRB, (A. M.), Fondazione Roma (A. M.) and The
Welch Foundation Grant G-1495 (M.T.B.).
Keywords: enzymes · epigenetics · histone methylation ·
inhibitors · transferases
[36] D. Cheng, N. Yadav, R. W. King, M. S. Swanson, E. J. Weinstein, M. T. Bed-
ford, J. Biol. Chem. 2004, 279, 23892–23899.
[37] A. Mai, D. Cheng, M. T. Bedford, S. Valente, A. Nebbioso, A. Perrone, G.
Brosch, G. Sbardella, F. De Bellis, M. Miceli, L. Altucci, J. Med. Chem.
2008, 51, 2279–2290.
[38] A. Mai, S. Valente, D. Cheng, A. Perrone, R. Ragno, S. Simeoni, G. Sbar-
della, G. Brosch, A. Nebbioso, M. Conte, L. Altucci, M. T. Bedford, Chem-
MedChem 2007, 2, 987–991.
[39] A. V. Purandare, Z. Chen, T. Huynh, S. Pang, J. Geng, W. Vaccaro, M. A.
Poss, J. Oconnell, K. Nowak, L. Jayaraman, Bioorg. Med. Chem. Lett.
2008, 18, 4438–4441.
[40] R. Ragno, S. Simeoni, S. Castellano, C. Vicidomini, A. Mai, A. Caroli, A.
Tramontano, C. Bonaccini, P. Trojer, I. Bauer, G. Brosch, G. Sbardella, J.
Med. Chem. 2007, 50, 1241–1253.
[1] B. D. Strahl, C. D. Allis, Nature 2000, 403, 41–45.
[2] A. Shilatifard, Annu. Rev. Biochem. 2006, 75, 243–269.
[3] H. Li, S. Park, B. Kilburn, M. A. Jelinek, A. Henschen-Edman, D. W. Aswad,
M. R. Stallcup, I. A. Laird-Offringa, J. Biol. Chem. 2002, 277, 44623–
44630.
[4] M. T. Bedford, S. Richard, Mol. Cell 2005, 18, 263–272.
[5] S. Pahlich, R. P. Zakaryan, H. Gehring, Biochim. Biophys. Acta Proteins Pro-
teomics 2006, 1764, 1890–1903.
[6] C. D. Krause, Z.-H. Yang, Y.-S. Kim, J.-H. Lee, J. R. Cook, S. Pestka, Phar-
macol. Ther. 2007, 113, 50–87.
[7] J. M. Aletta, J. C. Hu, M. R. El-Gewely, Biotechnol. Annu. Rev. 2008, 14,
203–224.
[41] A. Spannhoff, R. Heinke, I. Bauer, P. Trojer, E. Metzger, R. Gust, R. Schule,
G. Brosch, W. Sippl, M. Jung, J. Med. Chem. 2007, 50, 2319–2325.
[42] A. Spannhoff, R. Machmur, R. Heinke, P. Trojer, I. Bauer, G. Brosch, R.
Schꢆle, W. Hanefeld, W. Sippl, M. Jung, Bioorg. Med. Chem. Lett. 2007,
17, 4150–4153.
[8] Y. Zhang, D. Reinberg, Genes Dev. 2001, 15, 2343–2360.
[9] M. T. Bedford, J. Cell Sci. 2007, 120, 4243–4246.
[10] F. Bachand, Eukaryotic Cell 2007, 6, 889–898.
[11] F. Herrmann, M. Bossert, A. Schwander, E. Akgun, F. O. Fackelmayer, J.
Biol. Chem. 2004, 279, 48774–48779.
ChemMedChem 2010, 5, 398 – 414
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
413

G. Sbardella, M. T. Bedford et al.
MED
[43] T. Osborne, R. L. Weller Roska, S. R. Rajski, P. R. Thompson, J. Am. Chem.
Soc. 2008, 130, 4574–4575.
[59] R. P. McGeary, A. J. Bennett, Q. B. Tran, K. L. Cosgrove, B. P. Ross, Mini-
Rev. Med. Chem. 2008, 8, 1384–1394.
[44] M. Allan, S. Manku, E. Therrien, N. Nguyen, S. Styhler, M.-F. Robert, A.-C.
Goulet, A. J. Petschner, G. Rahil, A. R. MacLeod, R. Dꢅziel, J. M. Bester-
man, H. Nguyen, A. Wahhab, Bioorg. Med. Chem. Lett. 2009, 19, 1218–
1223.
[45] R. Heinke, A. Spannhoff, R. Meier, P. Trojer, I. Bauer, M. Jung, W. Sippl,
ChemMedChem 2009, 4, 69–77.
[60] S. Castellano, C. Milite, P. Campiglia, G. Sbardella, Tetrahedron Lett. 2007,
48, 4653–4655.
[61] S. Doulut, I. Dubuc, M. Rodriguez, F. Vecchini, H. Fulcrand, H. Barelli, F.
Checler, E. Bourdel, A. Aumelas, J. Med. Chem. 1993, 36, 1369–1379.
[62] All attempts to obtain the pure acid 5b from the corresponding ester
5a failed.
[46] H. Wan, T. Huynh, S. Pang, J. Geng, W. Vaccaro, M. A. Poss, G. L. Trainor,
M. V. Lorenzi, M. Gottardis, L. Jayaraman, A. V. Purandare, Bioorg. Med.
Chem. Lett. 2009, 19, 5063–5066.
[63] These experimental conditions lead to the partial hydrolysis of the 4-
acetoxy to a 4-hydroxy group. As this did not influence the reaction,
the crude mixture of the trichloroacetamides 18 was directly used in
the following step.
[64] P. Trojer, M. Dangl, I. Bauer, S. Graessle, P. Loidl, G. Brosch, Biochemistry
2004, 43, 10834–10843.
[65] A. E. McBride, J. T. Cook, E. A. Stemmler, K. L. Rutledge, K. A. McGrath,
J. A. Rubens, J. Biol. Chem. 2005, 280, 30888–30898.
[66] R. L. Bartel, R. T. Borchardt, Mol. Pharmacol. 1984, 25, 418–424.
[67] C. Schwerk, K. Schulze-Osthoff, Oncogene 2005, 24, 7002–7011.
[68] C. W. Siebel, C. Guthrie, Proc. Natl. Acad. Sci. USA 1996, 93, 13641–
13646.
[47] Recently, Thompson and co-workers reported the preparation of a
potent and selective bisubstrate inhibitor by the PRMT1-catalyzed reac-
tion between peptide AcH4-21 and a SAM congener, 5’-(diaminobutyric
acid)-N-iodoethyl-5’-deoxyadenosine ammonium hydrochloride (AAI).[43]
)
Such a peptide-based bisubstrate analogue is unlikely to possess drug-
like properties, but it could be a very useful tool for “chemical genetics”
studies of PRMT function in vitro and in vivo.
[48] S. Massa, A. Mai, G. Sbardella, M. Esposito, R. Ragno, P. Loidl, G. Brosch,
J. Med. Chem. 2001, 44, 2069–2072.
[49] A. Mai, S. Massa, R. Ragno, M. Esposito, G. Sbardella, G. Nocca, R. Scate-
na, F. Jesacher, P. Loidl, G. Brosch, J. Med. Chem. 2002, 45, 1778–1784.
[50] P. Ornaghi, D. Rotili, G. Sbardella, A. Mai, P. Filetici, Biochem. Pharmacol.
2005, 70, 911–917.
[51] S. Bartolini, A. Mai, M. Artico, N. Paesano, D. Rotili, C. Spadafora, G. Sbar-
della, J. Med. Chem. 2005, 48, 6776–6778.
[69] In the case of CARM1/H3 incubations with tested compounds, we were
not able to get strong bands on the fluorograph film and therefore we
did not scan it.
[70] D. S. Goodsell, G. M. Morris, A. J. Olson, J. Mol. Recognit. 1996, 9, 1–5.
[71] R. Wang, L. Lai, S. Wang, J. Comput.-Aided Mol. Des. 2002, 16, 11–26.
[72] G. Bertani, J. Bacteriol. 2004, 186, 595–600.
[52] G. Sbardella, S. Bartolini, S. Castellano, M. Artico, N. Paesano, D. Rotili, C.
Spadafora, A. Mai, ChemMedChem 2006, 1, 1073–1080.
[53] A. Mai, D. Rotili, D. Tarantino, P. Ornaghi, F. Tosi, C. Vicidomini, G. Sbar-
della, A. Nebbioso, M. Miceli, L. Altucci, P. Filetici, J. Med. Chem. 2006,
49, 6897–6907.
[73] A. Frankel, N. Yadav, J. Lee, T. L. Branscombe, S. Clarke, M. T. Bedford, J.
Biol. Chem. 2002, 277, 3537–3543.
[74] Q. Liu, G. Dreyfuss, Mol. Cell. Biol. 1995, 15, 2800–2808.
[75] A. W. Schꢆttelkopf, D. M. F. van Aalten, Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2004, 60, 1355–1363.
[54] G. Sbardella, S. Castellano, C. Vicidomini, D. Rotili, A. Nebbioso, M.
Miceli, L. Altucci, A. Mai, Bioorg. Med. Chem. Lett. 2008, 18, 2788–2792.
[55] S. Castellano, D. Kuck, M. Sala, E. Novellino, F. Lyko, G. Sbardella, J. Med.
Chem. 2008, 51, 2321–2325.
[76] PRODRG2 is freely available for academic research purposes at: http://
davapc1.bioch.dundee.ac.uk/prodrg/index.html.
[77] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A. E. Mark, H. J. C. Be-
rendsen, J. Comput. Chem. 2005, 26, 1701–1718.
[56] R. Ragno, A. Mai, S. Simeoni, A. Caroli, S. Valente, A. Perrone, S. Castella-
no, G. Sbardella, Small-Molecule Inhibitors of Histone Arginine Methyl-
transferases: Updated Structure-Based 3-D QSAR Models with Improved
Robustness and Predictive Ability; Frontiers in CNS and Oncology Medici-
nal Chemistry conference, ACS–EFMC, Siena (Italy), October 7–9, 2007,
pp. COMC-010.
[78] A. Gillet, M. Sanner, D. Stoffler, A. Olson, Structure 2005, 13, 483–491.
[79] E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, D. M. Greenblatt,
E. C. Meng, T. E. Ferrin, J. Comput. Chem. 2004, 25, 1605–1612.
[57] Color versions of Figures 2, 5–8 are available in the Supporting Informa-
tion.
[58] Y.-L. Zhang, Y.-F. Keng, Y. Zhao, L. Wu, Z.-Y. Zhang, J. Biol. Chem. 1998,
273, 12281–12287.
Received: November 7, 2009
Revised: December 11, 2009
Published online on January 20, 2010
414
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 398 – 414