Synthesis and evaluation of carbocyanine dyes as PRMT inhibitors and imaging agents

Article abstract of DOI:10.1016/j.ejmech.2012.06.017

Protein arginine methylation regulates multiple biological processes. Deregulation of protein arginine methyltransferase (PRMT) activities has been observed in many disease phenotypes. Small molecule probes that target PRMTs with strong affinity and selectivity can be used as valuable tools to dissect biological mechanisms of arginine methylation and establish the role of PRMT proteins in a disease process. In this work, we report synthesis and evaluation of a class of carbocyanine compounds containing indolium, benz[e]indolium or benz[c,d]indolium heterocyclic moieties that bind to the predominant arginine methyltransferase PRMT1 and inhibit its methyltransferase activity at low micromolar potencies. In particular, the developed molecules have long wavelength colorimetric and fluorometric photoactivities, which can be used for optical and near-infrared fluorescence imaging in cells or biological tissues. Together, these new chemical probes have potential application in PRMT studies both as enzyme inhibitors and as fluorescent dyes for microscope imaging.

Full text of DOI:10.1016/j.ejmech.2012.06.017

European Journal of Medicinal Chemistry 54 (2012) 647e659
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis and evaluation of carbocyanine dyes as PRMT inhibitors and imaging
agents
Sarmistha Halder Sinha a, Eric A. Owens a, You Feng a, Yutao Yang a, Yan Xie b, Yaping Tu b,
,
a,
**
Maged Henary a , Yujun George Zheng
*
a Department of Chemistry, Georgia State University, PO Box 4098, Atlanta, GA 30302, USA
b Department of Pharmacology, Creighton University, 2500 California Plaza, Omaha, NE 68178, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Protein arginine methylation regulates multiple biological processes. Deregulation of protein arginine
methyltransferase (PRMT) activities has been observed in many disease phenotypes. Small molecule
probes that target PRMTs with strong affinity and selectivity can be used as valuable tools to dissect
biological mechanisms of arginine methylation and establish the role of PRMT proteins in a disease
process. In this work, we report synthesis and evaluation of a class of carbocyanine compounds
containing indolium, benz[e]indolium or benz[c,d]indolium heterocyclic moieties that bind to the
predominant arginine methyltransferase PRMT1 and inhibit its methyltransferase activity at low
micromolar potencies. In particular, the developed molecules have long wavelength colorimetric and
fluorometric photoactivities, which can be used for optical and near-infrared fluorescence imaging in
cells or biological tissues. Together, these new chemical probes have potential application in PRMT
studies both as enzyme inhibitors and as fluorescent dyes for microscope imaging.
Received 10 March 2012
Received in revised form
4 June 2012
Accepted 9 June 2012
Available online 21 June 2012
Keywords:
Arginine methylation
PRMT
Inhibitor
Near-IR
Fluorescence
Carbocyanine
Ó 2012 Elsevier Masson SAS. All rights reserved.
1. Introduction
PRMT member, PRMT1, has been shown to impact a number of
disease pathways. PRMT1 is an essential component of the mixed
Arginine methylation is a prevalent protein post-translational
modification in cells, found in both nuclear and cytoplasmic
proteins. This biochemical reaction is mediated by protein arginine
methyltransferases (PRMTs) that utilize S-adenosyl-L-methionine
(AdoMet, SAM) as the methyl donor to modify the guanidinium
lineage leukemia (MLL) oncogenic transcriptional complex and
confers an aberrant transcriptional activation property critical for
the induction of leukemia [5]. Prominent roles of PRMT1 include
the coactivation of nuclear receptors and regulation of hormone-
dependent cancers [9,10]. Indeed, PRMT1 was found up-regulated
in breast cancer concomitantly with a change in substrate meth-
ylation [11]. PRMT1 variant 1 expression may be used as a marker of
unfavorable prognosis for breast cancer patients [12]. PRMT1
variant 2 was regarded as a marker of unfavorable prognosis in
colon cancer patients [13]. Thus, targeting PRMT activities with
small molecule inhibitors could be an effective therapeutic strategy
in new drug development [3,14,16]. Indeed, quite a few research
efforts have been investigated in the past years in attempt to
develop potent and selective PRMT inhibitors [15e28].
Carbocyanine dyes are unique organic molecules that contain
a conjugated electron-deficient system between two heterocyclic
nitrogen atoms that provides characteristically long absorption and
emission wavelengths; the associated wavelengths of carbocyanine
dyes have been exploited in many applications[29,30]. Using
various synthetic methods, the conjugated system of these
compounds can be altered to assume specific absorption and
fluorescence spectra in the range of 400e1000 nm[31,32].The
side chain of specific arginine residues, resulting in mono and di-
methylated arginine residues and the coproduct S-adenosyl-L-
homocysteine (AdoHcy, SAH). PRMTs are involved in the regulation
of diverse biological processes ranging from structural remodeling
of chromatins, gene transcription, RNA processing, DNA repair,
translation, nucleocytoplasmatic protein shuttling, nuclear
hormone receptor-mediated signaling, to cell proliferation and
differentiation [1e3]. Deregulation of PRMTs and arginine meth-
ylation has been reported to associate with a variety of human
diseases including several types of tumors (e.g. breast cancer,
prostate cancer, and leukemia) [4,5], pulmonary disorders [6]
cardiovascular diseases [7,8] etc. In particular, the predominant
* Corresponding author. Tel.: þ1 404 413 5566; fax: þ1 404 413 5505.
** Corresponding author. Tel.: þ1 404 413 5491; fax: þ404 413 5505.
E-mail addresses: mhenary1@gsu.edu (M. Henary), yzheng@gsu.edu
(Y.G. Zheng).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.06.017

648
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
number of carbon atoms between the heterocyclic nitrogen atoms
in the polymethine chain determines the common name for the
dyes. In particular, penta- and heptacarbocyanine dyes (5 and 7
carbon atoms, respectively) have been extensively utilized for
binding analyses; however, trimethine cyanine dyes (3 carbon
atoms) remain, until recently, unexplored [33]. Trimethine cyanine
dyes commonly absorb and emit light in the visible region;
however, varying the conjugated heterocyclic moieties greatly
shifts the maximum absorption and emission wavelengths. Their
binding properties to biomolecules can be studied using relative
spectrophotometric changes either in absorption or emission
spectra [34]. In this study, we synthesized and identified a set of
trimethine cyanine dyes that specifically inhibit the enzymatic
activity of PRMT1. Also, a proof-of-principle study has demon-
strated that these molecules are well suited for colorimetric and
fluorescent imaging of PRMT activities in living cells. Overall, the
identified carbocyanine molecules with unique photoactive prop-
erties could be powerful chemical probes in elucidating mecha-
nisms and functions of PRMTs underlying different biological or
pathological contexts.
indol-1H-2-ylidene)-1,3-dioxane-4,6-dione 4. The alkylation of 4
provides an easy access to various 1-alkyl-substituted 2-
methylbenz[c,d]-indolium salts, the important precursors to a set
of various carbocyanine dyes.
Benz[c,d]indole derivatives 2e4 were synthesized starting with
the commercial substrate 1. Compound 4 was reacted with various
alkyl iodides under basic conditions to afford esters 5e7, which
were correspondingly transformed into various quaternary salts
8e10. Salts 8e10 were purified by crystallization from methanol.
The condensation of 8e10 with triethyl orthoformate in Ac2O fur-
nished the benz[c,d]indolium trimethine cyanine dyes 11e13 in
very good yield.
Additionally, trimethine cyanines containing slightly varied
heterocyclic end units, indolium 14 and benz[e]indolium 15, were
synthesized as depicted in Scheme 2. Using similar chemistry, as
outlined in Scheme 1, quaternization of commercially obtained
indolium and benz[e]indolium 14e15 proceeds by a SN2 reaction
using corresponding alkyl halides in boiling acetonitrile to yield
salts 16e21 corresponding to starting heterocyclic subunits. Tri-
methine cyanines 22e27 were synthesized in good yield under
identical reaction conditions depicted in Scheme 2. Final trimethine
cyanines were purified by recrystallization from methanol or
column chromatography using 2% MeOH in DCM as eluting solvent.
Experimental procedures for novel compound preparation, chem-
ical and physical spectroscopic data are reported in the
Supplemental information.
2. Results and discussion
2.1. Chemistry
Toward the development of PRMT inhibitors and imaging
agents, we have concentrated on hydrophobic trimethine cyanine
derivatives utilizing ethyl, butyl and phenylpropyl halides for
nitrogen alkylation of the heterocyclic indolium, benz[e]indolium
and benz[c,d]indolium moieties (Schemes 1 and 2). The synthetic
sequence to NIR trimethine cyanine dyes 11e13 involves 6 steps as
shown in Scheme 1. The key step is the formation of 5-(benz[c,d]
2.2. Biochemical identification of the carbocyanine inhibitors of
PRMT1
In an effort to discover potent and selective PRMT inhibitors, we
synthesized and investigated several trimethine cyanine compounds
Scheme 1.

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
649
Scheme 2.
by screening their anti-methyltransferase activity. Our initial
hypothesis developed from the rationale that carbocyanine mole-
cules typically display heteroatomic and cationic structures, which
could potentially resemble positively charged arginine residues in
protein substrates, thereby leading to competitive PRMT inhibitors.
One of the characteristic features of carbocyanine molecules is their
absorption and fluorescence emission in the near-IR range which
offers a great advantage for in vivo fluorescent imaging with minimal
background interference from biomolecular autofluorescence. In the
inhibitor screening study, we selected PRMT1 as the primary target
because it is the predominant PRMT isoform found in mammalian
cells and is involved in many key biological processes. In a typica-
l anti-arginine methyltransferase screening assay, the reaction
mixture involved carbocyanine compounds, recombinantly
expressed His-tagged PRMT1, [14C]-labeled AdoMet, and a histone
H4 peptide containing the 20-amino acid sequence on the N-
terminal tail of histone H4, i.e. H4-20. The reaction was conducted at
containing dyes, this set of compounds with the exception of 24
presented significantly weaker activity. IC50 values of 22, 23, 28 and
29 were all greater than 150
mM. For 30 and 31, at 100 mM
concentration, PRMT1 still retained 80e95% of its activity. These
data demonstrate that reducing the heterocyclic size generally
lowers the binding of the carbocyanine molecules to PRMT1.
Notably, the two molecules that harness additional positive charges
on the N-substitution group, i.e. 28 and 29, do not offer any
advantage for enhancing the inhibition potency. This indicates that
charge alone is not the sole factor that determines the specific
inhibitoreenzyme interactions, as a certain degree of hydropho-
bicity of the N-substitution is required for efficient inhibition.
Contrary to the benz[c,d]indolium trimethine compounds (Group A
in Table 1), longer and bulky substituents on the heterocyclic ring
nitrogen increase inhibition potency. In particular, 24 having
a bulky phenylpropyl substituent showed the strongest potency,
with IC50 of 6.3 mM. This may suggest that a bulky hydrophobic
30 ꢀC in a volume of 30
mL. The degree of diminishment in PRMT1
group either on the heterocyclic ring moiety or on the heterocyclic
ring nitrogen substitution is required for effective binding to
PRMT1 protein.
activity was used as a parameter to evaluate the potency of the
compounds in inhibiting PRMT1-mediated arginine methylation. In
a more quantitative analysis of the inhibition potency, we measured
the methyltransferase activity of the enzyme at a range of concen-
trations of inhibitors and IC50 values were derived from the dose
response curve as the half-inhibition concentration. For all the
enzymatic assays, the methylation reactionwas controlled under the
initial condition so that the reaction yields of the limiting substrates
were lower than 15%, which was to ensure that the concentrations of
AdoMet and peptide substrate did not decrease significantly over the
time course of methylation reaction. This procedure also prevented
potential product inhibition from SAH.
To provide more information on the effect of the heterocyclic
moiety, we synthesized and tested three other trimethine cyanine
analogs bearing the benz[e]indolium end unit (Group C in Table 1)
while maintaining the identical set of nitrogen substituents, i.e.,
ethyl, butyl, and phenylpropyl group. The potency of benz[e]indo-
lium cyanines becomes weaker than the benz[c,d]indolium
cyanines that contain the same substituent. For instance,
comparing the ethyl-substituted compounds bearing the benz[e]
indole 25 (IC50 of 20.3
mM) and the benz[c,d]indole 11 (IC50 of
4.1 M) indicates that the position of the benz group attached on
m
From the biochemical screening, we found three structurally
related trimethine cyanine compounds that show desired anti-
PRMT1 activity; the chemical structures and corresponding IC50
values are shown in Table 1 (Group A). Clearly these compounds are
structurally analogous, all of which contain a delocalized positive
charge on the nitrogen atom and two benz[c,d]indolium moieties at
the ends of the trimethine cyanine main chain. The inhibition
potency is in the order of 11 > 12 > 13. This could possibly suggest
that increasing the size of the N-substituent causes increasing steric
hindrance, which prevents the compound from entering the
inhibitor-binding site of the enzyme. Interestingly, 11 (MHI-21)
shows the smallest size but best inhibition potency with IC50 of
the heterocyclic ring affects inhibitoreenzyme interaction. Also, the
weaker binding of benz[e]indolium is rescued by the incorporation
of a bulky phenylpropyl substituent on the heterocyclic nitrogen;
this is observed as the inhibition potency increases from 25 to 27. In
contrast, in benz[c,d]indolium-containing cyanine compounds, the
end head group seems to dominate the inhibitor-PRMT1 binding
and the bulky substitutions on the heterocyclic nitrogen do not
offer any positive contributions (i.e. 36).
2.3. Noncompetitive inhibition by 11
The finding that certain trimethine cyanine compounds exhibit
low micromolar anti-PRMT1 activity is encouraging. Especially, 11
shows the smallest molecule size but the strongest potency in the
tested compounds; the other one having similar potency to 11 is 24
(E-4). Thus, we next carried out more detailed analyses of PRMT1
inhibition by 11 in order to understand the inhibitory mechanism of
this type of carbocyanine molecules.
4.1
mM, which warrants further investigation.
To further this study, we synthesized and tested another set of
carbocyanine compounds containing an indolium group as the
heterocyclic moiety in order to screen for more potent inhibitors
and explore possible structureeactivity relationship (SAR) (Group B
in Table 1). Compared to the above three benz[c,d]indolium

650
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
Table 1
The names, structures and IC50 data of trimethine cyanine compounds.
Type of compound
Group A
Name
Structure
IC50(mM)
11 [MHI-21]
4.1 ꢁ 0.1
N
N
N
I
I
Trimethine cyanine compounds
containing benzo[c,d]indolium
groups in the two ends
N
12[MHI-06]
13[MHI-36]
22[E-14]
12.0 ꢁ 0.5
N
N
I
15.5 ꢁ 0.2
284.2 ꢁ 3.5
160.8 ꢁ 1.2
Ph
Ph
Group B
N
N
I
Trimethine cyanine compounds
containing indolium groups
in the two ends.
N
I
N
23[E-6]
N
Br
N
24[E-4]
6.3 ꢁ 0.3
N
N
Br
Br
28[A-C3]
368.3 ꢁ 25.9
Br
N
N
Br
Br
N
N
N
Br
Br
29[A-N3]
291.7 ꢁ 40.1
Br
N
+
N
30[MHI-106]
31[MHI-128]
At 100
80% activity retained
mM concentration
-
N
I
Cl
MeO
OMe
At 100
95% activity retained
mM concentration
+
N
-
I
N

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
651
Table 1 (continued )
Type of compound
Name
Structure
IC50(mM)
Group C
25[E-5]
20.3 ꢁ 0.2
N
N
I
Trimethine cyanine compounds
containing benzo[e]indolium
group as the end units
N
N
I
26[E-18]
18.0 ꢁ 0.5
N
Br
N
27[E-8]
14.3 ꢁ 0.2
First, we conducted steady-state enzymatic measurements to
analyze the kinetic pattern of inhibition by 11. The initial velocities
of PRMT1 were measured at several selected concentrations of the
inhibitor over a range of varied concentrations of one substrate
while fixing the concentration of the other. The kinetic inhibition
data points were analyzed by fitting to the linear competitive or
noncompetitive inhibition equations. The data were plotted in the
double reciprocal format with 1/velocity versus 1/concentration of
the varied substrate (Fig.1). It is seen that in both doubleereciprocal
plots, a series of straight lines intersect at a point in the second
quadrant. These data demonstrate that 11 is a noncompetitive
inhibitor with regard to both H4 and AdoMet substrates.
was dialyzed against the assay buffer for 72 h. The methyl-
transferase activity of PRMT1 before and after dialysis was
measured using the filter binding method. As shown in Fig. 2A, in
the presence of 20 mM 11, PRMT1 only retained 10% activity prior to
dialysis. After dialysis, the enzyme activity was restored to 66%
suggesting that PRMT1 inhibition by 11 is reversible. To confirm
this result, we measured time-dependent inhibition at several
concentrations of the inhibitor (0, 1.5, 2.5, 4.0 mM). In the time vs
product plot (Fig. 2B), straight lines were seen for all the
measurements instead of hyperbolic curves. Lack of any nonlinear
time-dependent inhibition indicates that the inhibition is revers-
ible and there is no covalent interaction between PRMT1 and the
inhibitor.
2.4. Reversible vs irreversible inhibition
2.5. Binding model analysis of inhibitoreenzyme interaction
To determine whether 11 reversibly or irreversibly inhibits
PRMT1 enzyme, a dialysis experiment was performed. For each
assay, fixed amounts of enzyme and inhibitor were incubated at
30 ꢀC for 1 h to allow for enzyme inactivation. Then the solution
To understand the structural basis of PRMT1 inhibition by 11, we
performed a docking study of the compound with PRMT1 structure
using the program Autodock. A homology model of human PRMT1
Fig. 1. Steady-state kinetic analysis of PRMT1 inhibition by 11. (A) Doubleereciprocal plot of the initial velocities versus H4 peptide with AdoMet concentration fixed at 5
with varied concentrations of 11 at 2.5 M (O), 5 M ( ), and 7.5 M (>). (B) Doubleereciprocal plot of the initial velocities versus AdoMet with H4-20 concentration fixed at 2
and with varied concentrations of 11 at 0 M (O), 2.5 M ( ), 5 M (>), and 6.5
M (ꢂ).
mM and
m
m
D
m
m
m
mM
m
D
m

652
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
B
A
1
PRMT1
PRMT1
1.1
0.9
0.7
0.5
0.3
0.1
-0.1
PRMT1 + 1.5 µM 11
0.8
0.6
0.4
0.2
0
PRMT1 + 2.5 µM 11
PRMT1 + 4.0 µM 11
PRMT1 + 20 µM 11
Before
dialysis
After
dialysis
0
5
10
15
20
25
30
35
Time [min]
Fig. 2. Test of the reversibility of PRMT1 inhibition by 11. (a) Dialysis experiment in presence and absence of 11. (b) Progression assay of PRMT1 inhibition at different concentrations
(1.5, 2.5, and 4.0 M) of 11.
m
structure was created from rat PRMT3. In the docking, a grid box
was created to cover the whole protein structure. The docking
result was analyzed according to the lowest energy order. As it
turned out, the docked ligand was found to distribute predomi-
nantly in three regions on the protein surface (Fig. 3A); none of
them overlap with the arginine and SAM binding sites. Therefore 11
is likely an allosteric inhibitor. These results are in agreement with
the kinetic results, which showed noncompetitive inhibition in
respect to H4-20 and AdoMet. For a more accurate analysis of the
binding mode, the compound was individually docked in each of
the three regions by choosing a grid box covering only that selected
region. A close inspection on region 1 revealed that 11 binds to
a pocket formed by Glu 19, Tyr 35, Ser 38, Tyr 39, Glu 47, Met 48, Tyr
148, Tyr 152, Val 206, Pro 217, His 293, Lys 295, Thr 297, Arg 327,
Met 352 and Arg 353 residues (Fig. 3B). The inhibitor electrostati-
cally interacts with residues Glu 19 and Glu 47, and makes hydrogen
bond contacts with Ser 38, Tyr 39, Tyr 148, Tyr 152, Lys 295, Thr 297,
Met 352 and Arg 353. The hydroxyl group of Tyr 148 seems to have
a strong H-bonding interaction with two nitrogens of 11 (3.7 Å and
4.3 Å) and a moderate interaction with Tyr 152 (6.7 Å and 7.0 Å). A
detailed study on region 2 shows that the compound binds to the
pocket formed by residues Leu 49, His 45, Gly 80, Thr 81, Gly 82, Ile
83, Met 86, Phe 87, Asn 102, Tyr 107, and Ile111 and makes hydrogen
bond interactions with Phe 42, His 45, Gly 80, Thr 81, Gly 82, Ile 83,
Fig. 3. Docking results. (A) Probable binding regions of 11. Binding mode of the inhibitor in Region 1 (B), in Region 2 (C), and in Region 3 (D).

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
653
Table 2
Met 86 and Tyr 107 residue (Fig. 3C). There is a moderate H-
Comparison of the inhibition of PRMT1, -3, -4, -5 and -6 by 11.a
bonding interaction between the heterocyclic nitrogen of the
compound and the side chain and backbone oxygen atoms of Thr
81, with 5.8 Å, 5.7 Å, respectively. Similar docking study on region 3
revealed that the compound binds to the pocket formed by Glu 47,
Met 48, Leu 49, Lys 50, Glu 52, Val 53, Tyr 148, Pro 290, Tyr 291, Thr
292 and His 293 residue and makes several electrostatic interac-
tions with residues Glu 47, Asp 51, Glu 52, and hydrogen bond
interaction with Lys 50, Met 48, Leu 49, Glu 52, Val 53, Pro 290, Tyr
291, and Thr 292 (Fig. 3D). Together, the inhibitory activity of 11
may come from its single or combined interactions at the three
major binding sites of PRMT1.
Peptide substrate
IC50 (mM)
His6x-rPRMT1
His6x-PRMT3
GST-CARM1
His6x-PRMT5
His6x-PRMT6
H4(1e20)
R4
H3(1e31)
H4(1e20)
H3(1e31)
4.1 ꢁ 0.1
8.3 ꢁ 0.1
z100
9.1 ꢁ 1.0
12.6 ꢁ 1.1
a
Data obtained from the results of the radioactive methylation
assays with varied concentrations of 11. For inhibition of His6x-
PRMT1, 2
were used. For inhibition of His6x-PRMT3, 2
SAM, and 0.1
1 mM H3(1e31), 30
For inhibition of His6x-PRMT6, 10
and 0.5 M enzyme were used.
m
M H4(1e20), 5
m
M [14C]-SAM, and 0.05
m
M enzyme
m
M R4, 5 m
M [14C]-
m
M enzyme were used. For inhibition of GST-CARM1,
M [14C]-SAM, and 5
M enzyme were used.
M H3(1e31), 5
M [14C]-SAM,
m
m
m
m
2.6. Investigation of interaction of 11 and PRMT1 using site-directed
mutagenesis
m
about 2.0 fold higher than PRMT1. 11 inhibits PRMT5 at IC50
9.1 ꢁ1.0 M which is about 2.2 fold higher than PRMT1. The IC50 of
11 for CARM1 is particularly higher, w100 M. 11 inhibits PRMT6
with an IC50 of 12.6 M, 3.1-fold higher than PRMT1. Overall, 11
seems to inhibit all the tested PRMT proteins, with a little prefer-
ence observed for PRMT1. The variation in inhibition potency likely
is caused by the subtle differences between the structures of indi-
vidual PRMT members.
To test the validity of the docking results, site-directed muta-
genesis was conducted to mutate the residues predicted to interact
with the compound in the docking model. Since multiple hydrogen
bonding interactions occur between the side chains of Tyr 148, Tyr
152, Thr 81 and Thr 292 residues and the heterocyclic ring nitrogen
atom of 11, we introduced four mutations, i.e., Y148A, Y152A, T81A,
T292A. All these PRMT1 mutants were produced using QuikChange
site-directed mutagenesis protocol, expressed in Escherichia coli,
and purified by Ni-NTA affinity chromatography. Activity of each
mutant protein was examined by the radioactive methyltransferase
assay in which recombinant PRMT1, [14C] AdoMet, and H4-20 were
m
m
m
2.8. Evaluation of the cellular activity of 11
incubated at 30 ꢀC in the presence or absence of 4
mM 11. The
After the biochemical characterizations, we moved on to
examine possible cellular effects of 11. We first examined subcel-
lular distribution of 11 in HeLa cells. A marked feature of carbo-
cyanine compounds is their spectral absorption and fluorescent
emission in the near-IR region. Measured in methanol, 11 has
absorption maximum at 758 nm and emission maximum at
774 nm. Interestingly, after incubation with 11 for 6 h, cells showed
bright, blue-colored staining and can be viewed directly under
bright field microscope. Microscopic images shown in Fig. 5AeC
demonstrated that 11 was concentrated in the nucleus of HeLa cells
in a dose dependent manner and a small amount of 11 was also
found in the cytoplasmic vesicles. Since carbocyanine compounds
have fluorescent emission in the long wavelength region, we per-
formed near-infrared fluorescent imaging of HeLa cells by confocal
microscopy using a 633 nm laser for fluorescence excitation and
a 715e785 nm Long Pass for emission. As can be seen in Fig. 5E and
F, after 6-h incubation in HeLa cells, the compound was concen-
trated around nucleus or in the nucleus and a low level was in the
cytoplasmic vesicles. The result is consistent with the light micro-
scopic images. As expected, the control HeLa cells treated with
DMSO did not show any fluorescent signal (Fig. 5D).
retained fractional activity was quantified to evaluate the inhibition
potency of the inhibitor on each PRMT mutant (Fig. 4). It is seen that
the activities of Y148A, T81A and T292A protein were inhibited by
11 by z 60%, 65% and 70% respectively, less than that of wild type
PRMT1 (92%). These results support that Y148, T81 and T292 resi-
dues might interact with 11 and thus mutation of these residues
decreased the binding and inhibition. 11 inhibited Y152A mutant
with same potency as is for wt-PRMT1, which signifies that Y152
quite likely is not involved in interaction with 11.
2.7. Selectivity of 11
We then investigated whether 11 is selective for PRMT1 inhi-
bition. For this study, we examined the inhibition by 11 of several
other PRMT isoforms, PRMT3, PRMT4/CARM1, PRMT5, and PRMT6,
using the radioactive methyltransferase assays. The IC50 values are
shown in Table 2. PRMT3 was inhibited at IC50 of 8.3 ꢁ 0.1
mM,
We next studied the growth inhibitory effect of 11 in HeLa cells
by using the MTT assay. The HeLa cells were cultured for 40 h in the
presence of different doses of 11 (Fig. 6). From the data, EC50, the
concentration of the compound required to reduce the cell survival
level to 50% at 48
0.94 ꢁ 0.09 M. This value is remarkably lower than the IC50 value
(i.e. 4.1 M) measured in the biochemical assay, likely suggesting
h of exposure, was determined to be
m
m
that 11 easily penetrates the plasma membrane and subsequently
affects the cell proliferation. To better understand the mechanism
underlying the observed proliferation inhibition, we examined the
effect of 11 on cell cycle of HeLa cells. The subpopulations of pro-
pidium iodide-stained HeLa cell were determined using flow
cytometry at varying concentrations of the compound in a time
dependent manner. Untreated HeLa cells showed a pattern of cell
growth phase distribution at G1 (57.2%), S (12.2%) and G2/M
Fig. 4. Inhibition of the activity of WT- and mutant- PRMT1 by 11. The reaction con-
tained 2 M H4-20, 5 M inhibitor. Activities of all PRMT1
M [14C] AdoMet and 4
mutants were normalized to the same level as that of wt-PRMT1 for clear comparison.
(29.2%). With increasing doses (e.g. 0.3, 1, 2.5 and 5
mM) at 24 h, S
m
m
m
phase arrest was shown to occur with a concomitant decrease of

654
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
Fig. 5. Microscopic imaging of 11 in HeLa Cells. HeLa cells are treated with 0 (control), 2, and 5 mM of 11. Figures A, B and C are the optical images of HeLa cells captured under visible
light by CoolSNAP CF camera attached to a Nikon Ti-80 microscope. Figures D, E and F are fluorescent images of HeLa cells captured by fluorescence confocal microscopy. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
cell populations in the other phases of the cell cycle (Fig. 7) In
particular, S phase cell population increased from 12.2% to 56.5% at
1 mM 11. Under this condition, the G1 cells decreased from 57.2% in
containing different heterocyclic moieties and hydrophobic
substituents, and identified several possessing micromolar potency
for PRMT1 inhibition. In particular, compound 11 presents potent
the control to 23.6% and the G2/M phase cells decreased from 29.2%
to 13.9%. The increase in S phase likely suggests that 11 enters the
nucleus and causes chromatin dysfunction, so that the spindles fail
to assemble which requires DNA damage to be repaired before
proceeding to the G2/M phase. After 48 h of treatment at 2.5 and
inhibitory activity on PRMT1, with IC50 of 4.1 mM. We have
demonstrated that this compound is a noncompetitive inhibitor,
and combined with docking simulation and site mutagenesis
analysis, we suggest 11 likely targets an allosteric site(s) in PRMT1.
Selectivity study showed that the compound has certain specificity
toward PRMT1 inhibition. The cyanine molecules typically possess
long-wavelength absorption and fluorescence properties which
allow them to be used for visualization in cells and tissues. By using
the optical and fluorescent microscopies, cell proliferation, and flow
cytometry assays, we found that 11 is capable of crossing the plasma
membrane, staying in the nucleus and interfering with the chro-
matin function. Together, this study provides unique photoactive
chemical probes for both PRMT inhibition and microscopic imaging.
5 mM 11, appearance of a hypodiploid peak (Sub-G1) was observed,
indicative of apoptosis (data not shown).
3. Conclusions
PRMT proteins play functional roles in a wide variety of diseased
phenotypes, and therefore the development of novel potent small
molecule inhibitors of PRMTs is highly desired for both basic and
applied pharmacology. We synthesized carbocyanine molecules
4. Experimental methods
4.1. Chemicals
1.2
1
Most chemical reagents were obtained from Sigma Aldrich.
Melting points (open Pyrex capillary) were measured on a Mel-
Temp apparatus and are uncorrected. 1H NMR and 13C NMR
spectra were recorded on either Bruker Avance (400 MHz) or
a Varian Unityþ300 (300 MHz) spectrometer using DMSO-d6 or
MeOD-d4 containing tetramethylsilane (TMS) as an internal stan-
dard. Vis/NIR absorption spectrum was recorded on a Varian Cary
50 spectrophotometer. High-resolution accurate mass spectra
(HRMS) were obtained using a Waters Q-TOF micro (ESI-Q-TOF)
mass spectrometer.
0.8
0.6
0.4
0.2
0
0
0.1
0.5
1
2.5
5
11 [µM]
4.2. Benz[c,d]indole-2(1H)-thione (2)
Fig. 6. The antiproliferation effect of 11 in HeLa Cells. Cells were treated with different
concentrations of 11 as indicated for 48 h and O.D. at 530 nm was measured using MTT
assay. Data are presented as the average of three independent experiments.
This compound was obtained in a 93% yield; mp 146e148 ꢀC;
(reported: yield 82%, mp 156 ꢀC.) [35e37].

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
655
A
5µM inhibitor
Control
1µM inhibitor
2.5µM inhibitor
0.3 µMinhibitor
B
70.00%
G1
S
G2
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
control
0.3 µM
1µM
2.5 µM
5µM
Fig. 7. Cell cycle analysis of HeLa cells treated with different doses of 11. Data are presented as the relative fluorescence intensity of cell subpopulations in the FACS profile (panel A)
and as percentage of cell subpopulations under each condition (panel B).
4.3. 2-Methylthiobenz[c,d]indole hydroiodide (3)
J ¼ 7.5 Hz, 2H), 7.79 (t, J ¼ 7.5 Hz, 1H), 7.94 (t, J ¼ 7.8 Hz, 1H), 8.02 (d,
J ¼ 7.2 Hz, 1H), 8.06 (d, J ¼ 8.4 Hz, 1H), 8.45 (d, J ¼ 8.1 Hz, 1H), 8.91
(d, J ¼ 7.2 Hz, 1H).
This compound was prepared using the reported procedures
[36,38]. Since the product was unstable, it is used in the next step
without further purification.
4.5.3. 1-Phenylpropyl-2-(2,2-dimethyl-4,6-dioxo-1,3-dioxan-5-
ylidene)-1H-benz[c,d]indole (7)
Yield 92%, mp 155e157 ꢀC; 1H NMR (400 MHz, DMSO-d6)
d 2.17
4.4. 2-(2,2-Dimethyl-4,6-dioxo-1,3-dioxane-5-yliden)-1H-benz[c,d]
(m, 2H), 2.56 (t, J ¼ 7.0 Hz, 2H), 4.35 (t, J ¼ 7.0 Hz, 2H), 7.17 (m, 5H),
7.71 (d, J ¼ 7.5 Hz, 1H), 7.86 (m, 2 H), 7.88 (t, J ¼ 7.5 Hz, 1H), 8.38 (d,
J ¼ 8.0 Hz, 1H), 8.88 (d, J ¼ 8.0 Hz, 1H).
indole (4)
This compound was obtained in an 85% yield; mp 220 ꢀC;
(reported: yield 94%, mp 223 ꢀC) [36,37].
4.6. General procedures of the synthesis of salts 8e10
4.5. General procedures of the synthesis of compounds 5e7
Ester 5e7 (2.9 mmol) was dissolved in acetic acid (4 mL) and the
mixture was refluxed for 20 min. Concentrated HCl (4 mL) was added
dropwise to the refluxing mixture until the color changed from red
to green. The mixture was cooled to room temperature, and satu-
rated KI solution was added until the product started to precipitate.
The product was filtered off, washed with ether, and dried in vacuo
affording quaternary ammonium salts 8e10 in good yield.
A mixture of compound 4 (10.2 mmol), alkyl iodide (30.6 mmol)
and K2CO3 (30.6 mmol) were heated in DMF (40 mL) at 90 ꢀC for
18 h under a nitrogen atmosphere. The mixture was cooled to room
temperature and then filtered. The filtrate was concentrated under
reduced pressure. The residue was purified on silica gel (flash
column chromatography, using EtOAcehexanes, 1:2 as an eluent).
4.5.1. 1-Ethyl-2-(2,2-dimethyl-4,6-dioxo-1,3-dioxan-5-ylidene)-
4.6.1. 1-Ethyl-2-(2-dimethylaminovinyl)benz[c,d]indolium iodide
1H-benz[c,d]indole (5)
(8)
Yield 65%, mp 183e184 ꢀC; 1H NMR (300 MHz, DMSO-d6)
d
1.46 (t,
Yield 76%, mp 242e244 ꢀC; 1H NMR (300 MHz, DMSO-d6)
d 1.56
J ¼ 7.0 Hz, 3H),1.70 (s, 6 H), 4.39 (q, J ¼ 7.0 Hz, 2H), 7.78 (m,1H), 7.91
(t, J ¼ 7.0 Hz, 3H), 4.46 (s, 3H), 4.71 (q, J ¼ 7.0 Hz, 2H), 8.00 (m, 1H),
8.19 (m, 1H), 8.45 (d, J ¼ 8.0 Hz, 1H), 8.54 (d, J ¼ 8.0 Hz, 1H), 8.69 (d,
J ¼ 8.0 Hz, 1H), 8.79 (d, J ¼ 8.0 Hz, 1H).
(m, 2H), 8.01 (m, 1H), 8.40 (d, J ¼ 8.0 Hz, 1H), 8.90 (d, J ¼ 8.0 Hz, 1H).
4.5.2. 1-Butyl- 2-(2,2-dimethyl-4,6-dioxo-1,3-dioxan-5-ylidene)-
1H-benz[c,d]indole (6)
4.6.2. 1-Butyl-2-(2-dimethylaminovinyl)benz[c,d]indolium iodide, (9)
Yield 74%, 1H NMR (300 MHz, DMSO-d6)
d
0.86 (t, J ¼ 7.5 Hz, 3H),
Yield 82%, 1H NMR (400 MHz, DMSO-d6)
d
0.86 (t, J ¼ 7.5 Hz 3H),
1.27 (q, J ¼ 7.2, 2H), 1.75 (s, 6H), 1.82 (p, J ¼ 6.6 Hz, 2H), 3.39 (t,
1.27 (m, 2H), 1.75 (s, 6H), 1.82 (p, J ¼ 6.6 Hz, 2H), 4.39 (t, J ¼ 7.5 Hz,

656
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
2H), 7.79 (t, J ¼ 8.1 Hz, 1H), 7.94 (t, J ¼ 7.5 Hz, 1H), 8.02 (d, J ¼ 7.2 Hz,
1H), 8.07 (d, J ¼ 8.4 Hz, 1H), 8.45 (d, J ¼ 8.1 Hz, 1H), 8.91 (d,
J ¼ 7.2 Hz, 1H).
4.8.1. 1-Ethyl-2,3,3-trimethyl-3H-indol-1-ium iodide (16)
This compound was obtained in 82% yield; mp 217e221 ꢀC;
(reported: mp 165e166 ꢀC) [39].
4.6.3. 1-Phenylpropyl-2-(2-dimethylaminovinyl)benz[c,d]indolium
4.8.2. 1-Butyl-2,3,3-trimethyl-3H-indol-1-ium iodide (17)
This compound was obtained in 80% yield; mp 116e119 ꢀC;
(reported: yield 83%, mp 102 ꢀC) [40].
iodide (10)
Yield 86%, mp 75e76 ꢀC; 1H NMR (400 MHz, DMSO-d6)
d 2.26
(m, 2H), 2.28 (t, J ¼ 7.0 Hz, 2H), 4.00 (s, 3H), 4.70 (t, J ¼ 7.0 Hz, 2H),
7.14 (m, 5H), 7.96 (t, J ¼ 7.5 Hz, 1H), 8.10 (t, J ¼ 7.5 Hz, 1H), 8.41 (m,
2H), 8.48 (d, J ¼ 8.0 Hz, 1H), 8.92 (d, J ¼ 8.0 Hz, 1H).
4.8.3. 2,3,3-Trimethyl-1-(3-phenylpropyl)-3H-indol-1-ium bromide (18)
81% yield; mp 169e171 ꢀC; 1H NMR (400 MHz, MeOD-d4)
d 1.58
(s, 6H), 2.32 (p, J ¼ 7.6 Hz, 2H), 2.90 (t, J ¼ 7.6 Hz, 2H), 4.53 (t,
J ¼ 8.0 Hz, 2H), 7.23e7.26 (m, 1H), 7.32e7.34 (m, 4H), 7.63e7.66 (m,
2H), 7.72e7.76 (m, 2H).
4.7. General synthetic strategy for final dye compounds 11e13
A mixture of the individual salts 8e10 (0.5 mmol), trie-
thylformate (0.25 mmol), anhydrous sodium acetate (30 mg) and
4 mL of acetic anhydride was boiled for 1 h. The dye was suction
filtered, washed with ethyl ether and dried.
4.8.4. 3-Ethyl-1,1,2-trimethyl-1H-benzo[e]indol-3-ium iodide (19)
This compound was obtained in 81% yield; mp 225e227 ꢀC;
(reported: yield 73%, mp 228 ꢀC) [41].
4.7.1. 1-Ethyl-2-[3-(1-ethylbenz[c,d]indol-2(1H)-ylidene)-1-
propen-1-yl]-iodide (11)
4.8.5. 3-Butyl-1,1,2-trimethyl-1H-benzo[e]indol-3-ium iodide (20)
This compound was obtained in 87% yield; mp 142e145 ꢀC;
(reported: yield 90%, mp 127e129) [42].
Yield 75%; 1H NMR (400 MHz, DMSO-d6)
d
1.40 (t, J ¼ 7.0 Hz, 3H),
4.36 (d, J ¼ 7.0 Hz, 2H), 6.90 (d, J ¼ 13.0 Hz,1H), 7.20 (m, 1H), 7.57 (d,
J ¼ 7.5 Hz, 1H), 7.66 (t, J ¼ 7.5 Hz, 1H), 7.80 (d, J ¼ 8.1 Hz, 1H), 7.97 (t,
J ¼ 7.5 Hz, 1H), 8.27 (d, J ¼ 8.1 Hz, 1H), 8.78 (t, J ¼ 13.0 Hz, 1H), 8.94
4.8.6. 1,1,2-Trimethyl-3-(3-phenylpropyl)-1H-benzo[e]indol-3-ium
iodide (21)
(d, J ¼ 7.5 Hz, 1H). 13C NMR (100 MHz, DMSO-d6)
d
14.56, 111.15,
Yield 81%, mp 169e171 ꢀC; 1H NMR (400 MHz, MeOD-d4)
d 1.58
111.50, 124.20, 124.81, 129.56, 129.94, 130.04, 130.08, 130.17, 130.92,
132.41, 140.64, 148.84, 155.18. TOF HRMS m/z (Mþ) calculated for
C29H25N2 401.2018, found 401.2000. lmax/em ¼ 758/774 nm in
methanol.
(s, 6H), 2.32 (p, J ¼ 7.6 Hz, 2H), 2.90 (t, J ¼ 7.6 Hz, 2H), 4.53 (t,
J ¼ 8.0 Hz, 2H), 7.23e7.26 (m, 1H), 7.32e7.34 (m, 4H), 7.63e7.66 (m,
2H), 7.72e7.76 (m, 2H).
4.9. General synthesis of 22e27
4.7.2. 1-Butyl-2-[3-(1-butyllbenz[c,d]indol-2(1H)-ylidene)-1-
propen-1-yl]- iodide (12)
Quaternized ammonium salts 16e21 were heated to 80 ꢀC in
acetic anhydride. Triethyl orthoformate was added to the reaction
mixture causing an instant color change to deep purple. The reac-
tion mixture was monitored by UV/Vis spectroscopy in methanol
and TLC analysis using DCM as the mobile phase. After 2 h the
reaction was stopped and the flask was cooled in a freezer. After
5e10 h, the product precipitated was filtered and dried. Initial
product purification was performed by recrystallization using
minimum amounts of methanol, reduced temperature, and titra-
tion with ether. Compounds 24, 26 and 27 were additionally puri-
fied using silica gel and 2% MeOH in dichloromethane as eluting
solvent.
Yield 60%, 1H NMR (300 MHz, DMSO-d6)
d
0.96 (t, J ¼ 7.2 Hz, 6H),
1.43e1.50 (m, 4H), 1.85 (p, J ¼ 6.9 Hz, 4H), 4.40 (t, J ¼ 6.6 Hz, 4H),
7.23 (d, J ¼ 13.2 Hz, 4H), 7.72e7.77 (m, 4H), 7.86e7.90 (m, 4H), 8.06
(t, J ¼ 7.5 Hz, 2H), 8.36 (d, J ¼ 8.1 Hz,1H), 8.66 (d, J ¼ 7.2 Hz, 2H), 9.20
(t, J ¼ 13.2 Hz, 1H). 13C NMR (100 MHz, DMSO-d6)
d 14.22, 20.14,
31.31, 44.30, 111.29, 111.74, 124.19, 124.69, 129.56, 129.92, 130.15,
130.27, 130.92, 132.41, 141.28, 148.96, 155.63. TOF HRMS m/z (Mþ)
calculated for C33H33N2 457.2644, found 457.2624. lmax/em ¼ 758/
774 nm in methanol.
4.7.3. 1-Phenylpropyl-2-[3-(1-phenylpropylbenz[c,d]indol-2(1H)-
ylidene)-1-propen-1-yl]iodide (13)
Yield 48%, 1H NMR (400 MHz, DMSO-d6)
d
2.06 (m, 2H), 2.75 (t,
4.9.1. 1-Ethyl-2-((1E,3E)-3-(1-ethyl-3,3-dimethylindolin-2-ylidene)
J ¼ 7 Hz, 2H), 4.27 (t, J ¼ 7.0 Hz, 2H), 6.75 (d, J ¼ 13.6 Hz, 1H), 6.97
(m, 1H), 7.24 (m, 5H), 7.46 (d, J ¼ 7.5 Hz, 1H), 7.63 (t, J ¼ 8.0 Hz, 1H),
7.76 (d, J ¼ 8 Hz, 1H), 7.94 ( t, J ¼ 7.5 Hz, 1H), 8.25 (d, J ¼ 8.0 Hz, 1H),
8.73 (t, J ¼ 13.0 Hz, 1H), 8.87 (d, J ¼ 8.0 Hz, 1H). 13C NMR (100 MHz,
prop-1-en-1-yl)-3,3-dimethyl-3H-indolium iodide (22)
Yield 77%, 1H NMR (400 MHz, MeOD-d4)
d
1.43 (t, J ¼ 8.0 Hz, 6H),
1.77 (s, 12H), 4.22 (q, J ¼ 8.0 Hz, 4H), 6.53 (d, J ¼ 12.0 Hz, 2H), 7.31 (t,
J ¼ 8.0 Hz, 2H), 7.36 (d, J ¼ 8.0 Hz, 2H), 7.45 (t, J ¼ 8.0 Hz, 2H), 7.55
(d, J ¼ 8.0 Hz, 2H), 8.56 (t, J ¼ 12.0 Hz, 1H); 13C NMR (100 MHz,
DMSO-d6) d 30.64, 32.69, 44.16,111.21,111.67,124.22,124.72,126.49,
128.55, 128.68, 128.84, 129.52, 129.90, 130.12, 130.23, 130.93,
132.44, 141.19, 141.31, 148.85, 155.66. TOF HRMS m/z (Mþ) calcu-
lated for C43H37N2 581.2957, found 581.2934. lmax/em ¼ 758/774 nm
in methanol.
MeOD-d4) d 12.69, 28.25, 40.35, 103.44, 112.23, 123.60, 126.77,
130.04, 142.34, 142.95, 152.32, 175.63; TOF HRMS m/z (Mþ) calcu-
lated for C27H33N2 385.2644, found 385.2647. lmax/em ¼ 546/
561 nm in methanol.
4.8. General synthesis of indolium and benz[e]indolium salts 16e21
4.9.2. 1-Butyl-2-((1E,3E)-3-(1-butyl-3,3-dimethylindolin-2-
ylidene)prop-1-en-1-yl)-3,3-dimethyl-3H-indol-1-ium (23)
A mixture of indolenine 14 or 15 (31.4 mmol) and alkyl halide
(94.2 mmol) were refluxed in acetonitrile (25 mL) for 24 h. The
solvent was concentrated and removed in vacuo to afford an
oil residue. The reaction mixture was cooled in an ice bath and
solid was obtained upon addition of ether and acetone. The
resulting crystals were filtered and dried to yield red to light pink
solid which was used without further purification in subsequent
reactions.
Yield 74%, mp 172e175 ꢀC; 1H NMR (400 MHz, DMSO)
d 0.95 (t,
J ¼ 8.0 Hz, 6H), 1.41e1.54 (m, 4H), 1.71 (s, 12H), 1.68e1.80 (m, 4H),
4.14 (t, J ¼ 8.0 Hz, 4H), 6.52 (d, J ¼ 16.0 Hz, 2H), 7.31 (t, J ¼ 8.0 Hz,
2H), 7.46 (q, J ¼ 8.0 Hz, 4H), 7.65 (d, J ¼ 16.0 Hz, 2H), 8.37 (t,
J ¼ 12.0 Hz, 1H); 13C NMR (MeOD-d4)
d 13.81, 19.57, 27.46, 29.21,
43.70, 48.90, 102.50, 111.56, 122.50, 125.20, 128.64, 140.61, 141.87,
173.84; TOF HRMS m/z (Mþ) calculated for C31H41N2 441.3270,
found 441.3279. lmax/em ¼ 546/561 nm in methanol.

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
657
4.9.3. 1-Phenylpropyl-2-((1E,3E)-3-(1-butyl-3,3-dimethylindolin-
(25 mM Na-HEPES, pH 7.0, 300 mM NaCl, 1 mM PMSF, 100 mM
EDTA and 200 mM imidazole, 5% glycerol). Different elution frac-
tions were checked on SDSePAGE. Then, protein fractions were
combined and dialyzed against the storage buffer (25 mM Na-
HEPES, pH 7.0, 500 mM NaCl, 1 mM EDTA, 10 mM DTT and 10%
glycerol) at 4 ꢀC. After dialysis, the protein solution was concen-
trated using Millipore centrifugal filters. His6x-tagged PRMT3 was
expressed from the pReceiver vector. GST-mCARM1 was expressed
from the pGEX-4T1 vector. His6x-tagged PRMT5 and PRMT6 were
expressed from the pET28a vector. All of the proteins were
expressed in E. coli BL21(DE3). All of the His6x-tagged proteins
were purified on Ni-NTA beads. Protein concentrations were
determined using Bradford assay.
2-ylidene)prop-1-en-1-yl)-3,3-dimethyl-3H-indol-1-ium (24)
Yield 67%, mp 255e258 ꢀC; 1H NMR (400 MHz, MeOD-d4)
d 1.74
(s, 12H), 2.18 (p, J ¼ 7.2 Hz, 4H), 2.85 (t, J ¼ 7.5 Hz, 4H), 4.18 (t,
J ¼ 7.8 Hz, 4H), 6.21 (d, J ¼ 13.3 Hz, 2H), 7.45 (t, J ¼ 8.0 Hz, 2H), 7.55
(d, J ¼ 8.0 Hz, 2H), 8.56 (t, J ¼ 12.0 Hz, 1H); 13C NMR (100 MHz,
MeOD-d4) d 24.64, 28.30, 29.97, 33.74, 44.65, 103.69, 112.37, 123.56,
126.82, 127.45, 129.61, 129.71, 129.98, 142.08, 142.21, 143.27, 175.96;
TOF HRMS m/z (Mþ) calculated for C27H33N2 385.2644, found
385.2647. lmax/em ¼ 546/561 nm in methanol.
4.9.4. 3-Ethyl-2-((1E,3E)-3-(3-ethyl-1,1-dimethyl-1H-benzo[e]
indol-2(3H)-ylidene)prop-1-en-1-yl)-1,1-dimethyl-1H-benzo[e]
indol-3-ium iodide (25)
Yield 66%, mp >260 ꢀC; 1H NMR (400 MHz, MeOD-d4)
d
1.50 (t,
4.11. Radioactive methyltransferase assay
J ¼ 8.0 Hz, 6H), 2.10 (s, 12H), 4.34 (q, J ¼ 8.0 Hz, 4H), 6.53 (d,
J ¼ 16.0 Hz, 2H), 7.53 (t, J ¼ 8.0 Hz, 2H), 7.65 (d, J ¼ 8.0 Hz, 2H), 7.67
(d, J ¼ 8.0 Hz, 2H), 8.03 (d, J ¼ 8.0 Hz, 2H), 8.065 (d, J ¼ 12.0 Hz, 2H),
8.295 (d, J ¼ 8.0 Hz, 2H), 8.79 (t, J ¼ 8.0 Hz, 1H); 13C NMR (100 MHz,
The inhibitory activities of small molecule compounds were
examined by using 14C-labeled radioactive methylation assays. The
assays were carried out in 0.6 mL plastic tubes at 30 ꢀC in a reaction
MeOD-d4)
d
13.13, 28.07, 40.68, 52.59, 102.96, 112.14, 123.49, 126.49,
volume of 30
8.0), 0.5 mM dithiothreitol (DTT), 1 mM EDTA, and 50 mM NaCl. In
a typical procedure, peptide substrate, [14C]- S-Adenosyl-
-methi-
onine and inhibitor were preincubated in the reaction buffer for
5 min prior to the initiation by the addition of PRMT1 (0.05
mL. The reaction buffer contained 50 mM HEPES (pH
129.07, 129.55, 131.34, 132.16, 133.82, 140.59; TOF HRMS m/z (Mþ)
calculated for C35H37N2 485.2957, found 485.2960. lmax/em ¼ 586/
605 nm in methanol.
L
mM
4.9.5. 3-Butyl-2-((1E,3E)-3-(3-butyl-1,1-dimethyl-1H-benzo[e]
indol-2(3H)-ylidene)prop-1-en-1-yl)-1,1-dimethyl-1H-benzo[e]
indol-3-ium iodide (26)
final). After it was incubated for 8 min, the reaction was quenched
by spreading the reaction mixture onto P81 filter paper discs
(Whatman). The paper disk was washed with 50 mM NaHCO3, and
dried in air for 2 h. The amount of methylated products was
quantified by liquid scintillation. To determine the IC50 value of
those inhibitors similar experiment was performed with H4-(1e20)
Yield 63%, 1H NMR (400 MHz, MeOD-d4)
d
1.06 (t, J ¼ 8.0 Hz, 6H),
1.55e1.61 (m, 4H), 1.88e1.94 (m, 4H), 2.11 (s, 12H), 4.20 (t,
J ¼ 8.0 Hz, 4H), 6.565 (d, J ¼ 12.0 Hz, 2H), 7.53 (t, J ¼ 8.0 Hz, 2H), 7.64
(d, J ¼ 8.0 Hz, 2H), 7.69 (t, J ¼ 12.0 Hz, 2H), 8.03 (d, J ¼ 8.0 Hz, 2H),
8.06 (d, J ¼ 8.0 Hz, 2H), 8.30 (d, J ¼ 8.0 Hz, 2H), 8.79 (t, J ¼ 12.0 Hz,
(2 m L-methionine (5 mM) and DMSO (2%) and
M), [14C]- S-Adenosyl-
varied concentration of inhibitors. The IC50 value is the concen-
tration of inhibitor at which half of the maximal activity is reached.
The inhibition patterns of 11 were determined by measuring initial
velocities of PRMT1 at a range of varied concentrations of one
substrate, a fixed concentration of the other substrate, and selected
concentrations of the inhibitors. The data were displayed in double
reciprocal formats and fitted to competitive or noncompetitive
kinetic equations.
1H); 13C NMR (100 MHz, MeOD-d4):
d 14.41, 21.11, 21.36, 28.18,
31.13, 52.56, 103.27, 112.37, 123.50, 126.52, 129.09, 129.48, 131.34,
132.10, 133.80, 135.10, 141.02; TOF HRMS m/z (Mþ) calculated for
C39H45N2 541.3583, found 541.3582. lmax/em ¼ 586/605 nm in
methanol.
4.9.6. 2-((1E,3E)-3-(1,1-Dimethyl-3-(3-phenylpropyl)-1H-benzo[e]
indol-2(3H)-ylidene)prop-1-en-1-yl)-1,1-dimethyl-3-(3-
phenylpropyl)-1H-benzo[e]indol-3-ium iodide (27)
4.12. Time-dependent inhibition assay
Yield 55%, 1H NMR (400 MHz, MeOD-d4)
d 2.01 (s, 12H), 2.12 (p,
J ¼ 8.0 Hz, 4H), 2.82 (t, J ¼ 8.0 Hz, 4H), 4.31 (t, J ¼ 8.0 Hz, 4H), 6.395
(d, J ¼ 12.0 Hz, 2H), 7.21e7.33 (m, 12H), 7.55 (t, J ¼ 8.0 Hz, 2H), 7.740
(t, J ¼ 8.0 Hz, 2 H), 7.75 (d, J ¼ 8.0 Hz, 2H), 8.11 (t, J ¼ 8.0 Hz, 4H), 8.30
(d, J ¼ 8.0 Hz, 2H), 8.58 (t, J ¼ 12.0 Hz, 1H); 13C NMR (100 MHz,
Depending upon the IC50 value of the compound, different
concentrations of the compound were incubated with 0.05
enzyme so that in the final reaction mixture 11 was present at 0,1.5,
2.5, 4.0 M. Incubations were conducted for 0, 2, 4, 6, 8, 10, 15, 20,
25, 30 min at 30 ꢀC After each incubation, the reaction was
mM
m
MeOD-d4)
d 26.60, 28.76, 32.32, 101.59, 110.65, 121.92, 125.01,
126.08, 127.55, 127.90, 127.99, 128.07, 128.21, 128.34, 129.78, 130.55,
132.27, 139.29, 140.62. TOF HRMS m/z (Mþ) calculated for C49H49N2
665.3896, found 665.3893. lmax/em ¼ 586/605 nm in methanol.
quenched by spreading 20
discs (Whatman).
mL reaction mixture onto P81 filter paper
4.13. Dialysis experiment
4.10. Protein expression and purification
For the dialysis assay, 1
in a total reaction volume of 520 m
m
M PRMT1 was incubated with 20
mM 11
PRMT1-pET28b plasmid was transformed into E. coli BL21 (DE3)
using heat shock method. Protein expression was induced with 0.3-
mM IPTG at 16 ꢀC for 20 h. After protein expression, cells were
pelleted by centrifuge followed by suspension in the lysis buffer
(25 mM HEPES pH 8.0, 150 mM NaCl, 1 mM PMSF, 1 mM MgSO4, 5%
glycerol, 5% ethylene glycol) and lysed by French Press. The protein
supernatant was purified on the Ni-charged His-tag binding resin
(Novagen). The beads were equilibrated with column buffer
(25 mM Na-HEPES pH 8.0, 500 mM NaCl, 1 mM PMSF, and 30 mM
imidazole, 10% glycerol, 1 mM PMSF) followed by protein loading.
After that the beads were washed thoroughly with the column
buffer, and then the protein was eluted with the elution buffer
L for 1 h at 30 ꢀC to achieve
complete inactivation. Enzyme without any inhibitor was used in
experimental controls. The reaction mixtures were then transferred
to dialysis cassettes (molecular weight cut off of 10 000Da) and
dialyzed for 3 days at 4 ꢀC in the 300 mL reaction buffer contained
50 mM HEPES (pH 8.0), 0.5 mM dithiothreitol (DTT), 1 mM EDTA,
50 mM NaCl and 10% DMSO. The buffer was changed six times
during this period. After finishing dialysis, residual activity assays
were performed on pre- and post-dialysis samples. The activity was
measured in the presence of 2
Adenosyl- -methionine and dialyzed samples were preincubated in
the reaction buffer for 5 min prior to the initiation by the addition
mM peptide substrate, 5 m
M [14C]- S-
L

658
S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
of PRMT1 (0.05
m
M final). After it was incubated for 8 min, the
Appendix A. Supplementary information
reaction was quenched by spreading the reaction mixture onto P81
filter paper discs (Whatman). The paper discs were washed with
50 mM NaHCO3 and dried in air for 2 h. The amount of methylated
products was quantified by liquid scintillation. The amount of
product formed in the control reaction was set as 100% and used to
determine the percent activity remaining in the inhibited reactions.
Supplementary data associated with this article can be found, in
the onlineversion, at http://dx.doi.org/10.1016/j.ejmech.2012.06.017.
References
[1] A. Di Lorenzo, M.T. Bedford, Histone arginine methylation, FEBS Lett. 585
(2011) 2024e2031.
[2] S.S. Wolf, The protein arginine methyltransferase family: an update about
function, new perspectives and the physiological role in humans, Cell. Mol.
Life Sci. 66 (2009) 2109e2121.
[3] C.D. Krause, Z.H. Yang, Y.S. Kim, J.H. Lee, J.R. Cook, S. Pestka, Protein arginine
methyltransferases: evolution and assessment of their pharmacological and
therapeutic potential, Pharmacol. Ther. 113 (2007) 50e87.
[4] S. Pal, S. Sif, Interplay between chromatin remodelers and protein arginine
methyltransferases, J. Cell. Physiol. 213 (2007) 306e315.
[5] N. Cheung, L.C. Chan, A. Thompson, M.L. Cleary, C.W. So, Protein
arginine-methyltransferase-dependent oncogenesis, Nat. Cell. Biol. 9 (2007)
1208e1215.
[6] D. Zakrzewicz, O. Eickelberg, From arginine methylation to ADMA: a novel
mechanism with therapeutic potential in chronic lung diseases, BMC Pulm.
Med. 9 (2009) 5.
4.14. Bright field and fluorescent microscopic imaging
HeLa cells were seeded on coverglass overnight and treated
without or with different concentrations of 11 for 6 h. Cells were
then fixed by 4% paraformaldehyde for 10 min at room tempera-
ture. Then the optical images were taken under visible light by
CoolSNAP CF camera attached to a Nikon Ti-80 microscope. The NIR
fluorescent images were taken by an LSM META NLO confocal
microscope (Creighton university Integrated biomedical imaging
Facility) using a 633-nm laser and a 715e785-nm Long Pass filter.
[7] J. Beltowski, A. Kedra, Asymmetric dimethylarginine (ADMA) as a target for
pharmacotherapy, Pharmacol. Rep. 58 (2006) 159e178.
4.15. Antiproliferative assay of HeLa cells
[8] R. Maas, Pharmacotherapies and their influence on asymmetric
dimethylargine (ADMA), Vasc. Med. 10 (Suppl. 1) (2005) S49eS57.
[9] S.S. Koh, D. Chen, Y.H. Lee, M.R. Stallcup, Synergistic enhancement of nuclear
receptor function by p160 coactivators and two coactivators with protein
methyltransferase activities, J. Biol. Chem. 276 (2001) 1089e1098.
[10] S. Wagner, S. Weber, M.A. Kleinschmidt, K. Nagata, U.M. Bauer, SET-mediated
Hela cells were maintained in Dulbecco’s modified Eagle’s
medium (DMEM) (Cellgro) supplemented with 10% fetal bovine
serum (Cellgro) and 1% penicillinestreptomycin in 5% CO2 at 37 ꢀC.
Cells were washed with PBS and harvested using trypsin. Individual
promoter hypoacetylation is
a prerequisite for coactivation of the
wells of a 96-well tissue culture plate were inoculated with 100 mL
estrogen-responsive pS2 gene by PRMT1, J. Biol. Chem. 281 (2006)
27242e27250.
of complete medium containing 9000 cells and incubated over-
night for attachment at 37 ꢀC in a 5% CO2 incubator. The following
day, the media were removed and the cells were treated with
different concentrations of 11. The final concentration of DMSO in
the culture medium was maintained at 0.2%. Control cells (without
any compound) were treated under the same conditions. After 48 h
of incubation period, cell survival was evaluated using an MTT
[11] I. Goulet, G. Gauvin, S. Boisvenue, J. Cote, Alternative splicing yields protein
arginine methyltransferase
1 isoforms with distinct activity, substrate
specificity, and subcellular localization, J. Biol. Chem. 282 (2007)
33009e33021.
[12] K. Mathioudaki, A. Scorilas, A. Ardavanis, P. Lymberi, E. Tsiambas, M. Devetzi,
A. Apostolaki, M. Talieri, Clinical evaluation of PRMT1 gene expression in
breast cancer, Tumour Biol. 32 (2011) 575e582.
[13] A. Papadokostopoulou, K. Mathioudaki, A. Scorilas, D. Xynopoulos,
A. Ardavanis, E. Kouroumalis, M. Talieri, Colon cancer and protein
arginine methyltransferase 1 gene expression, Anticancer Res. 29 (2009)
1361e1366.
[14] Y.G. Zheng, J. Wu, Z. Chen, M. Goodman, Chemical regulation of epigenetic
modifications: opportunities for new cancer therapy, Med. Res. Rev. 28 (2008)
645e687.
[15] A. Spannhoff, A.T. Hauser, R. Heinke, W. Sippl, M. Jung, The emerging thera-
peutic potential of histone methyltransferase and demethylase inhibitors,
ChemMedChem 4 (2009) 1568e1582.
assay: 10
was added to each well, and after 4 h of incubation, the medium
was removed; after addition of 100 L of DMSO to each well, the
mL of a freshly prepared solution of MTT (5 mg/mL in PBS)
m
optical density values were detected at 530 nm. Cytotoxicity data
were expressed as IC50 values. Data were expressed as mean values
of three individual experiments conducted in triplicate.
[16] A. Spannhoff, R. Heinke, I. Bauer, P. Trojer, E. Metzger, R. Gust, R. Schüle,
G. Brosch, W. Sippl, M. Jung, Target-based approach to inhibitors of histone
arginine methyltransferases, J. Med. Chem. 50 (2007) 2319e2325.
[17] A. Spannhoff, R. Machmur, R. Heinke, P. Trojer, I. Bauer, G. Brosch, R. Schüle,
W. Hanefeld, W. Sippl, M.A. Jung, A novel arginine methyltransferase inhibitor
with cellular activity, Bioorg. Med. Chem. Lett. 17 (2007) 4150e4153.
[18] R. Ragno, S. Simeoni, S. Castellano, C. Vicidomini, A. Mai, A. Caroli,
A. Tramontano, C. Bonaccini, P. Trojer, I. Bauer, G. Brosch, G. Sbardella, Small
molecule inhibitors of histone arginine methyltransferases: homology
modeling, molecular docking, binding mode analysis, and biological evalua-
tions, J. Med. Chem. 50 (2007) 1241e1253.
[19] A. Mai, D. Cheng, M.T. Bedford, M.S. Valente, A. Nebbioso, A. Perrone,
G. Brosch, G. Sbardella, F. De Bellis, M. Miceli, L. Altucci, Epigenetic
multiple ligands: mixed histone/protein methyltransferase, acetyltransferase,
and class III deacetylase (Sirtuin) inhibitors, J. Med. Chem. 51 (2008)
2279e2290.
4.16. Cell cycle assay
For flow cytometric analysis of DNA content, 1 ꢂ 106 cells in
exponential growth were treated with different concentrations of
11 at different concentrations for 24 and 48 h. After the incubation,
the cells were collected, centrifuged and fixed with ice-cold ethanol
(70%) and stored at 4 ꢀC until use. Subsequently, cells were rinsed
with PBS and stained with PBS containing 40
mg/mL propidium
iodide and followed by the addition of RNase A. Samples were
analyzed on a BD-LRFortessa flow cytometer (BD Biosciences). For
cell cycle analysis, DNA histograms were analyzed using Flow Jo
software.
[20] D. Cheng, N. Yadav, R.W. King, M.S. Swanson, E.J. Weinstein, M.T. Bedford,
Small molecule regulators of protein arginine methyltransferases, J. Biol.
Chem. 279 (2004) 23892e23899.
[21] M.T. Bedford, S. Richard, Arginine methylation: an emerging regulator of
protein function, Mol. Cell 18 (2005) 263e272.
Acknowledgments
This work is supported by Georgia Cancer Coalition Distin-
guished Cancer Scholar Award and NIH Grant GM086717 to YGZ. YF
is supported by an American Heart Association Pre-doctoral
Fellowship and a University Doctoral Fellowship at GSU under the
instruction of YGZ. The synthesis of inhibitors was supported by the
GSU Research Initiation Grant and Georgia Cancer Coalition Grant
to MH. Additionally, MH would like to acknowledge the Center for
Diagnostics and Therapeutics at Georgia State University for the
support.
[22] K.K. Li, C. Luo, D. Wang, H. Jiang, Y.G. Zheng, Chemical and biochemical
approaches in the study of histone methylation and demethylation, Med. Res.
Rev. (2010).
[23] E.-M. Bissinger, R. Heinke, A. Spannhoff, A. Eberlin, E. Metzger, V. Cura,
P. Hassenboehler, J. Cavarelli, R. Schüle, M.T. Bedford, W. Sippl, M. Jung, Acyl
derivatives of p-aminosulfonamides and dapsone as new inhibitors of the
arginine methyltransferase hPRMT1, Bioorg. Med. Chem. 19 (2011)
3717e3731.
[24] Y. Feng, M. Li, B. Wang, Y.G. Zheng, Discovery and mechanistic study of a class
of protein arginine methylation inhibitors, J. Med. Chem. 53 (2010)
6028e6039.

S.H. Sinha et al. / European Journal of Medicinal Chemistry 54 (2012) 647e659
659
[25] D. Cheng, S. Valente, S. Castellano, G. Sbardella, R. Di Santo, R. Costi,
M.T. Bedford, A. Mai, Novel 3,5-bis(bromohydroxybenzylidene)piperidin-4-
[34] J.H. Flanagan, S. Khan, S. Menchen, S.A. Soper, R.P. Hammer, Functionalized
tricarbocyanine dyes as near-infrared fluorescent probes for biomolecules,
Bioconjugate Chem. 8 (1997) 751e756.
ones as coactivator-associated arginine methyltransferase
1 inhibitors:
enzyme selectivity and cellular activity, J. Med. Chem. 54 (2011) 4928e4932.
[26] D. Cheng, M.T. Bedford, Xenoestrogens regulate the activity of arginine
methyltransferases, ChemBioChem 12 (2011) 323e329.
[27] R. Heinke, A. Spannhoff, R. Meier, P. Trojer, I. Bauer, M. Jung, W. Sippl, Virtual
screening and biological characterization of novel histone arginine methyl-
transferase PRMT1 inhibitors, ChemMedChem 4 (2009) 69e77.
[28] J. Dowden, R.A. Pike, R.V. Parry, W. Hong, U.A. Muhsen, S.G. Ward, Small
molecule inhibitors that discriminate between protein arginine N-methyl-
transferases PRMT1 and CARM1, Org. Biomol. Chem. 9 (2011) 7814e7821.
[29] L. Strekowski, C.J. Mason, H. Lee, R. Gupta, J. Sowell, G. Patonay, Synthesis of
water-soluble near-infrared cyanine dyes functionalized with [(succinimido)
oxy]carbyonyl group, J. Heterocycl. Chem. 40 (2003) 913e916.
[30] A. Mishra, R.K. Behera, P.K. Behera, B.K. Mishra, G.B. Behera, Cyanines during
the 1990s: a review, Chem. Rev. 100 (2000) 1973e2012.
[31] N. Narayanan, G. Patonay, A new method for the synthesis of heptamethine
cyanine dyes: synthesis of new near-infrared fluorescent labels, J. Org. Chem.
60 (1995) 2391e2395.
[35] N.P. Vasilenko, F.A. Mikhailenko, Y.I. Rozhinskii, 2-Methylbenz[c,d]indole and
its derivatives, Dyes Pigm. 2 (1981) 231e237.
[36] G.E. Ficken, J.D. Kendall, The reactivity of the alkylthio group in nitrogen ring
compounds. III. 2-Methylthiobenz[c,d]indole and its methiodide, J. Chem. Soc.
(1960) 1537e1541.
[37] M.V. Lakshmikantham, W. Chen, M.P. Cava, Thioanhydrides. 3. Synthesis,
properties, and Diels-Alder reactions of sulfur analogues of 1,8-naphthalic
anhydride, J. Org. Chem. 54 (1989) 4746e4750.
[38] T.G. Deligeorgiev, N.I. Gadjev, Styryl dyes containing the benz[c,d]indolium
heterocycle, Dyes Pigm. 15 (1991) 215e223.
[39] R.L. Simmons, R.T. Yu, A.G. Myers, Storable arylpalladium(II) reagents
for alkene labeling in aqueous media, J. Am. Chem. Soc. 40 (2011)
15870e15873.
[40] C.C. Lee, A.T. Hu, Synthesis and optical recording properties of some novel
styryl dyes for DVD-R, Dyes Pigm. 59 (2003) 63e69.
[41] S. Kuster, F. Sauvage, M.K. Nazeeruddin, M. Grätzel, F.A. Nüesch, T. Geiger,
Unsymmetrical squaraine dimer with an extended
p-electron framework: an
[32] M. Mojzych, M. Henary, Synthesis of cyanine dyes, in: L. Strekowski (Ed.),
Heterocyclic Polymethine Dyes (2008), pp. 1e9.
approach in harvesting near infra-red photons for energy conversion, Dyes
Pigm. 87 (2010) 30e38.
[33] M.Y. Losytskyy, K.D. Volkova, V.B. Kovalska, I.E. Makovenko, Y.L. Slominskii,
O.I. Tolmachev, S.M. Yarmoluk, Fluorescent properties of pentamethine
cyanine dyes with cyclopentene and cyclohexene group in presence of bio-
logical markers, J. Fluoresc. 15 (2005) 849e857.
[42] M.V. Kvach, A.V. Ustinov, I.A. Stepanova, A.D. Malakhov, M.V. Skorobogatyi,
V.V. Shmanai, V.A. Korshun,
A convenient synthesis of cyanine dyes:
reagents for the labeling of biomolecules, Eur. J. Org. Chem. 12 (2008)
2107e2117.