Development of GoSlo-SR-5-69, a potent activator of large conductance Ca2+-activated K+ (BK) channels

Article abstract of DOI:10.1016/j.ejmech.2014.01.035

We have designed, synthesised and characterised the effects of a number of novel anthraquinone derivatives and assessed their effects on large conductance, Ca²⁺ activated K⁺ (BK) channels recorded from rabbit bladder smooth muscle cells using the excised, inside/out configuration of the patch clamp technique. These compounds are members of the GoSlo-SR family of compounds, which potently open BK channels and shift the voltage required for half maximal activation (V_1/2) negatively. The efficacy of the anilinoanthraquinone derivatives was enhanced when the size of ring D was increased, since the cyclopentane and cyclohexane derivatives shifted the V _1/2, by -24 ± 6 mV and -54 ± 8 mV, respectively, whereas the cycloheptane and cyclooctane derivatives shifted the V_1/2 by -61 ± 6 mV and -106 ± 6 mV. To examine if a combination of hydrophobicity and steric bulking of this region further enhanced their ability to open BK channels, we synthesised a number of naphthalene and tetrahydro-naphthalene derivatives. The tetrahydro-2-naphthalene derivative GoSlo-SR-5-69 was the most potent and efficacious of the series since it was able to shift the activation V_1/2 by greater than -100 mV when applied at a concentration of 1 μM and had an EC₅₀ of 251 nM, making it one of the most potent and efficacious BK channel openers synthesised to date.

Full text of DOI:10.1016/j.ejmech.2014.01.035

European Journal of Medicinal Chemistry 75 (2014) 426e437
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Development of GoSlo-SR-5-69, a potent activator of large
conductance Ca2þ-activated Kþ (BK) channels
,
,
Subhrangsu Roy b 1, Roddy J. Large b 1, Adebola Morayo Akande a, Aravind Kshatri a,
Tim I. Webb b, Carmen Domene c d, Gerard P. Sergeant a b, Noel G. McHale a
,
,
,
,b
Keith D. Thornbury a b, Mark A. Hollywood a
,
,b,
*
a The Smooth Muscle Research Centre, Dundalk Institute of Technology, Dublin Road, Dundalk, County Louth, Ireland
b Ion Channel Biotechnology Centre, Dundalk Institute of Technology, Dublin Road, Dundalk, County Louth, Ireland
c Chemistry Research Laboratory, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK
d Department of Chemistry, King’s College London, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NH, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
We have designed, synthesised and characterised the effects of a number of novel anthraquinone de-
rivatives and assessed their effects on large conductance, Ca2þ activated Kþ (BK) channels recorded from
rabbit bladder smooth muscle cells using the excised, inside/out configuration of the patch clamp
technique. These compounds are members of the GoSlo-SR family of compounds, which potently open
BK channels and shift the voltage required for half maximal activation (V1/2) negatively. The efficacy of
the anilinoanthraquinone derivatives was enhanced when the size of ring D was increased, since the
cyclopentane and cyclohexane derivatives shifted the V1/2, by ꢀ24 ꢁ 6 mV and ꢀ54 ꢁ 8 mV, respectively,
whereas the cycloheptane and cyclooctane derivatives shifted the V1/2 by ꢀ61 ꢁ 6 mV and ꢀ106 ꢁ 6 mV.
To examine if a combination of hydrophobicity and steric bulking of this region further enhanced their
ability to open BK channels, we synthesised a number of naphthalene and tetrahydro-naphthalene de-
rivatives. The tetrahydro-2-naphthalene derivative GoSlo-SR-5-69 was the most potent and efficacious of
the series since it was able to shift the activation V1/2 by greater than ꢀ100 mV when applied at a
Received 13 September 2013
Received in revised form
13 January 2014
Accepted 16 January 2014
Available online 3 February 2014
Keywords:
Ion channels
Anthraquinone derivatives
Ullmann coupling reaction
BK channel activator
Bladder
concentration of 1
mM and had an EC50 of 251 nM, making it one of the most potent and efficacious BK
channel openers synthesised to date.
Ó 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
The role of these channels in the electrical and resultant me-
chanical activity of the urinary bladder has been extensively stud-
ied by a number of groups in experimental animals [7,11,13,14] and
humans [15e17] and demonstrates that BK channels normally limit
the influx of Ca2þ and thus reduces the contractile activity of the
Large conductance Ca2þ activated (BK) channels are pore form-
ing transmembrane proteins that play numerous roles in the
excitability of neuronal [1e4] and muscle cells from mammals [5e7]
through to nematodes [8e10]. In the lower urinary tract smooth
muscles of mammals, BK channels contribute to the repolarisation
phase of the action potential and thus modulate the influx of Ca2þ
[5,11,12]. Since the mechanical activity of these smooth muscles is
determined by the underlying electrical activity [13,14], the BK
channels can contribute significantly to the resultant mechanical
activity recorded in these tissues.
bladder. Interestingly, mice that have the pore forming BKa subunit
genetically ablated [18], appear functionally incontinent, thus
illustrating the important role these channels play in the control of
bladder function. More recently, Hristov et al. [16], demonstrated
that the expression of BK channels in human urinary bladder was
significantly downregulated in patients suffering from neurogenic
detrusor overactivity and suggested that the development of
potent, efficacious BK channel openers may help offset the down-
regulation of BK channels in these patients.
Over the past two decades, a number of novel BK channel
openers have been developed [reviewed in Refs. [19e23]], but their
efficacy in clinical trials appears limited [20], presumably due to
their lack of effect at physiological membrane potentials (ꢀ80 mV
to 0 mV). Consequently, we recently embarked on a study to
* Corresponding author. The Smooth Muscle Research Centre, Dundalk Institute
of Technology, Dublin Road, Dundalk, County Louth, Ireland. Tel.: þ353 429370475;
fax: þ353 429370509.
E-mail address: mark.hollywood@dkit.ie (M.A. Hollywood).
URL: http://www.smoothmusclegroup.org
These authors contributed equally to this work and should be considered as
1
joint first authors.
0223-5234/$ e see front matter Ó 2014 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2014.01.035

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
427
develop a new family of BK channel openers (which we called the
GoSlo-SR family [24],), in an attempt to produce more efficacious
BK channel openers than those currently available. These anili-
noanthraquinones were reasonably potent (EC50 w2.5
mM) and
shifted the voltage required for half maximal activation (V1/2) of the
BK channels by more than ꢀ100 mV, when applied (at a concen-
tration of 10
mM) to excised patches of rabbit bladder smooth
muscle plasma membranes. In contrast, the prototypical BK chan-
nel openers NS1619 [25] and NS11021 [26] only shifted the acti-
vation V1/2 by approximately ꢀ25 mV and ꢀ50 mV respectively,
when applied at the same concentration.
Our initial study of the structure activity relationship (SAR) of
the GoSlo-SR molecules, suggested that the presence of an acidic
group in 2-position of ring C was essential for BK channel agonist
activity and that bulky and hydrophobic substituents on the phenyl
ring D enhanced the efficacy significantly [24]. For example, the
addition of a trifluoromethyl group in the meta-position of ring D
(GoSlo-SR-5-6, Fig. 1) shifted V1/2 by approximately ꢀ110 mV and
the further addition of a methyl group in the para-position (GoSlo-
SR-5-44, Fig. 1) increased this to approximately ꢀ140 mV, with
little change in potency.
In the present study, we used the inside-out excised patch
configuration of the patch clamp technique to examine the SAR of
these compounds further in an attempt to (i) correlate how the size
of ring D alters efficacy, (ii) ascertain if further alterations of the
phenylamino ring at 4-position of C-ring enhance the efficacy and/
or potency, and (iii) determine if the primary amine group at 1-
position of ring C was essential for the BK channel opening activ-
ity. Our results suggest that the addition of tetrahydro-2-
naphthylamine at the C-4 position of the basic anthraquinone
core (to form GoSlo-SR-5-69) maintained efficacy but increased the
potency by more than ten-fold, compared to the first generation
GoSlo-SR series of molecules.
Scheme 1. Reagents and conditions: (i) ReNH2, phosphate buffer, copper powder,
microwave, 100e120 ꢂC, 5e20 min; (ii) ReNH2, Na2CO3, CuSO4, water, 105 ꢂC, 6e8 h.
room temperature. All the newly synthesized compounds (listed in
Tables 1e3) were characterized by 1H NMR, 13C NMR and high-
resolution mass spectral analysis. The purity of the compounds
was determined by HPLC and their UV visible spectra and purity
determinations are shown in the Supporting Information. With the
exception of GoSlo-SR-5-90, which had a purity of 91.5%, all com-
pounds had a purity of >95%.
3. Results and discussion
3.1. Single channel recording
In this study, we used the inside-out configuration of the patch
clamp technique (see Experimental) with symmetrical 140 mM Kþ
solutions, to evaluate the ability of a series of novel, GoSlo-SR family
members (3e23, Tables 1e3) to open BK channels when applied to
the cytosolic surface of membrane patches. As previously reported
[24], the vast majority of current recorded under these conditions
was due to activation of BK channels, since they (i) had a large
conductance (w330 pS), (ii) shifted their voltage dependence of
activation by approximately ꢀ100 mV, in response to a 10-fold
increase in Ca2þ at the cytosolic face of the patch, (iii) reversed at
the equilibrium potential for Kþ (0 mV), and (iv) were abolished in
the presence of the selective BK channel inhibitors penitrem A [27]
and iberiotoxin.
2. Chemistry
As shown in Scheme 1, the anilinoanthraquinone derivatives
were prepared from commercially available bromaminic acid by
Ullmann coupling reaction using two different methods [28,29].
The compounds 3e5 and 10e19 were obtained by a newly devel-
oped microwave assisted Ullmann coupling reaction in phosphate
buffer using copper powder as a catalyst [28]. However, compounds
7 and 8 were synthesized by classical Ullmann coupling reaction in
water in the presence of sodium carbonate and copper sulphate
[29].
Scheme 2 demonstrates the synthesis of deaminated derivatives
of 1-aminoanthraquinones 20e23 by using an efficient, one-pot,
two-step reaction protocol recently reported in the literature [31].
The primary aromatic amine of 1-aminoanthraquinones was first
diazotized by sodium nitrite and 1 M hydrochloric acid to obtain
the stable, red coloured intermediate diazonium salt. The diazo-
nium salt was further reduced by adding zinc powder in ethanol at
Membrane patches were held at ꢀ60 mV and voltage ramps
(100 mV sꢀ1) were employed in these experiments to examine the
effects of the novel BK channel openers on BK channels. The in-
ternal surface of the patch was continually bathed with 100 nM
Ca2þ in the absence and presence of each test compound.
3.2. Effect of D ring size
We first assessed the effect of altering the cycloaliphatic D ring
size on the efficacy of the molecules designed and synthesised
O
B
NH2
O
NH2
SO3Na
SO3Na
1
2
A
C
4
O
HN
CF3
O
HN
CF3
D
CH3
1 (GoSlo-SR-5-6)
2 (GoSlo-SR-5-44)
Scheme 2. Reagents and conditions: (i) (a) NaNO2, HCl (1 M), 0 ꢂC to rt, 1 h; (b) Zn,
Fig. 1. Structures of the first generation of GoSlo-SR compounds.
EtOH, rt, 5 min.

428
S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
distance between C10 of ring D and the most distal part of the D
ring, as shown by the red arrow in Fig. 2D. As Fig. 2F suggests, when
we plotted the size of the ring D against DV1/2 and fitted these data
Table 1
Structure and effect of GoSlo-SR compounds on activation
each position are given in each column. All compounds were applied at 10
cytosolic surface of the patch in the presence of 100 nM Ca2þ. The change in voltage
required for half maximal activation of the channels ( V1/2) is shown in the right
D
V1/2. The substituents at
m
M to the
by linear regression, we obtained a reasonably good correlation
D
(r2 ¼ 0.88) between D ring size and the
DV1/2.
hand column. Data are quoted as the mean ꢁ standard error of the mean (SEM).
Numbers in parentheses represent the number of replicates. Two of these molecules
[b]
(5 and 6, denoted by ) have been previously reported to inhibit ecto-50-nucleo-
3.3. Effect of a ring fusion at different positions on the phenyl D ring
tidases [30] and the others are novel compounds.
The above experiments suggested that the efficacy of the GoSlo-
SR compounds could be enhanced by increasing the size of the ring
D attached at the N4-position. To examine if a combination of hy-
drophobicity and steric bulking of this region further enhanced the
efficacy, we first assessed the effects of the tetrahydro-1-
naphthalene (10, GoSlo-SR-5-65, Table 2) and 1-naphthalene (11,
GoSlo-SR-5-103, Table 2) derivatives, where the cyclohexyl and
phenyl ring was fused at the ortho- and meta-positions of the
phenyl D ring respectively. These were then compared with the
effects of the commercially available Acid Blue 25 (9, AB-25,
Compound
Name
R
D
V1/2
3
GoSlo-SR-5-10
8 ꢁ 6 (n ¼ 5)
4
GoSlo-SR-5-20
GoSlo-SR-5-8
ꢀ17 ꢁ 6 (n ¼ 5)
ꢀ24 ꢁ 6 (n ¼ 4)
Table 2). As Fig. 3B (blue trace) suggests, application of 10
25 caused negative shift in the activation V1/2 of
approximately ꢀ50 mV. When the data from five similar experi-
ments were summarised (Fig. 3C), 10 M AB-25 shifted the mean
mM AB-
a
5[b]
m
V1/2 by ꢀ54 ꢁ 4 mV (from 176 ꢁ 13 mV to 122 ꢁ 14 mV). However,
compared to AB-25, compound 10 (GoSlo-SR-5-65, Fig. 3DeF) was
twice as efficacious and the V1/2 was shifted by ꢀ116 ꢁ 9 mV
(from ꢀ149 ꢁ 10 mV to 33 ꢁ 10 mV, n ¼ 7) when applied at a
6[b]
Acid Blue 62
ꢀ54 ꢁ 8 (n ¼ 4)
ꢀ61 ꢁ 6 (n ¼ 6)
7
GoSlo-SR-5-32
concentration of 10 mM. Interestingly, as shown in Fig. 3H and I,
compound 11 (GoSlo-SR-5-103, Table 2) was slightly less efficacious
than AB-25, since it only shifted the V1/2 by ꢀ44 ꢁ 4 mV in eighteen
experiments (from 142 ꢁ 4 mV to 98 ꢁ 5 mV) when applied at a
8
GoSlo-SR-5-140
ꢀ106 ꢁ 6 (n ¼ 7)
concentration of 10 mM. These data suggest that the orientation and
the conformation of the additional ring fused at the ortho- and
meta-positions of the phenyl D ring play an important role in
altering their efficacy on BK channels.
To examine if the position of the additional ring fused on phenyl
D ring altered the effects of the compounds, we synthesised the
tetrahydro-2-naphthalene (12, GoSlo-SR-5-69, Fig. 4D) and 2-
naphthalene (13, GoSlo-SR-5-95, Fig. 4G) derivatives. Both of
these compounds were found to be very potent and efficacious
activators of BK channels as they negatively shifted the activation
following the Ullmann coupling reaction (Scheme 1) as reported in
the literature [28,29]. We investigated if the cyclopropyl ring in the
N4-position activated the channels (3, GoSlo-SR-5-10, Table 1), but
found that it had little effect, as illustrated by the typical record
shown in Fig. 2A. In this and subsequent figures, the traces in grey
and blue correspond to 100 nM Ca2þ in the absence and presence of
the drug, respectively. The black trace represents the currents
recorded in the presence of 1
m
M Ca2þ and was used to estimate the
V1/2 by more than ꢀ80 mV when applied at a concentration of 1
(Fig. 4E and H). In comparison, when 1 M AB-25 (Fig. 4B, blue
trace) was applied, the activation V1/2 only shifted by ꢀ25 mV. In
five similar experiments, summarised in Fig. 4C, 1 M of AB-25
mM
number of BK channels in the patch. When a cyclobutyl ring was
introduced to the N4-position (4, GoSlo-SR-5-20) and added to the
cytosolic face of the patch (Fig. 2B, blue trace) the activation V1/2
was shifted negatively from 125 mV to 96 mV, whereas a 10-fold
increase in Ca2þ shifted the V1/2 to 8 mV. Introduction of a cyclo-
heptane ring (7, GoSlo-SR-5-32, Fig. 2C) or a cyclooctane ring (8,
GoSlo-SR-5-140, Fig. 2D) caused larger negative shifts in the acti-
vation V1/2 from 133 mV to 53 mV and from 134 mV to 25 mV,
respectively, suggesting that increasing the size of ring D enhanced
the efficacy of the compounds on BK channels. Fig. 2E shows a
m
m
shifted the V1/2 from 176 ꢁ 13 mV (grey bar) to 150 ꢁ 13 mV
(blue bar), whereas compound 12 (GoSlo-SR-5-69) was approxi-
mately four times as efficacious and shifted the mean V1/2 from
149 ꢁ 7 mV (grey bar) to 46 ꢁ 5 mV (blue bar, n ¼ 10, Fig. 4F). Fig. 4H
illustrates that compound 13 (GoSlo-SR-5-95) was moderately less
efficacious than GoSlo-SR-5-69 and induced a negative shift in the
activation V1/2 of w e80 mV. The lower efficacy was reflected in ten
similar experiments summarised in Fig. 4I, where application of
summary of the shift in activation V1/2 (DV1/2) plotted against the
number of carbons in the D ring. Interestingly, the relationship was
not linear, since increasing ring size from 4 to 5 carbons only shifted
1
m
M of compound 13 shifted the activation V1/2 from 151 ꢁ 9 mV to
74 ꢁ 10 mV. These data suggested that the fusion of either cyclo-
hexyl or phenyl ring at meta- and para-positions of the phenyl D
ring significantly enhanced the efficacy of the compounds.
the mean
D
V1/2 by approximately e7 mV (from ꢀ17 ꢁ 6 mV, n ¼ 5
to ꢀ24 ꢁ 6 mV, n ¼ 4 respectively). Similarly when the effect of the
cyclohexyl D ring was compared to that of the cycloheptyl ring, the
To measure the potency of the compounds, we next performed a
series of experiments in which increasing concentrations of each
compound were applied to the cytosolic face of the BK channels in
the continued presence of 100 nM Ca2þ. A typical experiment with
GoSlo-SR-5-69 is shown in Fig. 5A. Under control conditions (100 nM
Ca2þ, grey trace), the channels were half maximally activated
at ꢀ130 mV. Increasing the concentration of GoSlo-SR-5-69 caused a
concentration dependent, negative shift in the V1/2. To summarise
DV1/2 also changed by only approximately e7 mV (DV1/2 with
cyclohexyl was ꢀ54 ꢁ 8 mV, n ¼ 4 compared to ꢀ61 ꢁ 6 mV, n ¼ 6,
with the cycloheptyl ring). In contrast, changing from a cyclopentyl
to a cyclohexyl ring increased the
to ꢀ54 ꢁ 8 mV and substitution of the cycloheptyl ring with the
cyclooctyl ring changed the
D
V1/2 from ꢀ24 ꢁ 6 mV
D
V1/2 from ꢀ61 ꢁ 6 mV (n ¼ 6)
to ꢀ106 ꢁ 7 mV (n ¼ 7). To examine if these effects could be
correlated to the molecular size of each ring, we calculated the
these data, we obtained the DV1/2 by subtracting the V1/2 obtained in

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
429
Table 2
Structure, efficacy and potency of GoSlo-SR compounds on BK channels. The first column refers to the number of the compound and the second column refers to its family
name. Each substituent at position N4 is given in column 3. Compounds were applied at either 10
M or 1 M (as detailed in column 4) to the cytosolic surface of the patch in the
presence of 100 nM Ca2þ. The change in voltage required for half maximal activation of the channels (
V1/2) and the concentration that produced half maximal effects (EC50) is
m
m
D
shown in the column 5. Data are quoted as the mean ꢁ standard error of the mean (SEM) for V1/2 or mean ꢁ 95% confidence intervals for EC50 (column 5). ‘n’ refers to the
D
number of replicates. Three of these molecules (9, 11 and 13, denoted by [b]) have been previously demonstrated to inhibit P2Y receptors or ecto-50-nucleotidases [30] and the
remainder are novel compounds.
Compound
Name
R
DV1/2
EC50 (95% CI)
9[b]
Acid Blue 25
ꢀ54 ꢁ 4 (n ¼ 5)
ꢀ116 ꢁ 9 (n ¼ 7)
5.9 mM (2.7e12.9 mM)
10
GoSlo-SR-5-65
366 nM (255e525 nM)
11[b]
GoSlo-SR-5-103
GoSlo-SR-5-69
ꢀ44 ꢁ 4 (n ¼ 18)
841 nM (0.3e2.1 mM)
12
ꢀ104 ꢁ 9, (n ¼ 10) (1
m
M)
251 nM (105e597 nM)
13[b]
GoSlo-SR-5-95
GoSlo-SR-5-64
ꢀ85 ꢁ 12 (n ¼ 10) (1
ꢀ88 ꢁ 12 (n ¼ 6)
m
M)
392 nM (0.14e1.1 mM)
14
2.9 mM (1.7e4.8 mM)
15
GoSlo-SR-5-124
ꢀ66 ꢁ 8 (n ¼ 4)
N.D.
16
17
GoSlo-SR-5-7
GoSlo-SR-5-90
ꢀ14 ꢁ 7 (n ¼ 4)
14 ꢁ 8 (n ¼ 5)
N.D.
N.D.
18
19
GoSlo-SR-5-126
GoSlo-SR-5-127
ꢀ26 ꢁ 11 (n ¼ 5)
ꢀ38 ꢁ 6 (n ¼ 5)
N.D.
N.D.
the absence of the drug (100 nM Ca2þ) from the V1/2 measured in
each concentration of the compound. Fig. 5B shows the summary of
comparison, AB-25, shown in Fig. 5D, had an EC50 that was w24-fold
lower (5.9
m
M, Hill coefficient was 0.7 ꢁ 0.11, 9, Table 2) than that of
these data in which the
DV1/2 was plotted against the concentration
GoSlo-SR-5-69 (12, Table 2). As Fig. 5BeE suggest, the order of po-
tency and efficacy was GoSlo-SR-5-69 > GoSlo-SR-5-65 > GoSlo-SR-
5-95 >> AB-25, thus suggesting that the tetrahydro-2-naphthalene
derivative 12 was the most potent and efficacious opener of this
series. However, as shown in Fig. 5F, the 1-naphthalene derivative,11
of GoSlo-SR-5-69 applied. When these data were fitted with the
Langmuir equation, the concentration at which the compound had
its half maximal effect (EC50) was 251 nM (95% confidence intervals
(CI) were 105e597 nM) and the Hill coefficient was 0.69 ꢁ 0.16. In

430
S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
Table 3
Structure, efficacy and potency of deaminated GoSlo-SR compounds on BK channels. The first column refers to the number of the compound and the second column refers to its
family name. Each substituent at position N4 is given in column 3. Compounds were applied at either 10
M or 1 M (as detailed in column 4) to the cytosolic surface of the
patch in the presence of 100 nM Ca2þ. The change in voltage required for half maximal activation of the channels (
V1/2) and the concentration that produced half maximal
effects (EC50) is shown in the column 5. Data are quoted as the mean ꢁ standard error of the mean (SEM) for V1/2 or mean ꢁ 95% confidence intervals for EC50 (column 5). n
Refers to the number of replicates for the V1/2 determination. For EC50 determinations, different concentrations were applied to between 5 and 30 patches for each compound,
m
m
D
D
D
with the exception of 300 nM GoSlo-SR-5-103 and AB-25, which were applied to 4 and 2 patches, respectively. Compound 21 (GoSlo-SR-5-132) has been previously synthesised
[31] but its biological activity has not been determined.
Compound
20
Name
R
D
V1/2
EC50
GoSlo-SR-5-134
ꢀ79 ꢁ 6 (n ¼ 9) (1
ꢀ41 ꢁ 4 (n ¼ 6) (1
m
m
M)
M)
640 nM (460e885 nM)
21[b]
GoSlo-SR-5-132
GoSlo-SR-5-130
984 nM (632e1.52 mM)
22
ꢀ97 ꢁ 4 (n ¼ 7)
ꢀ115 ꢁ 8 (n ¼ 6)
1.8 mM (1.2e3.4 mM)
23
GoSlo-SR-5-138
964 nM (0.56e1.7 mM)
(GoSlo-SR-5-103, Table 2) had a reduced potency (EC50 ¼ 841 nM)
and efficacy compared to the 2-naphthalene derivative 13 (GoSlo-SR-
5-95, EC50 ¼ 392 nM, Table 2), suggesting that the position of the
phenyl ring fused with phenyl D ring significantly altered the po-
tency and efficacy of these molecules. Taken together, the data pre-
sented above suggest that both the orientation and the conformation
of the additional ring fused with ring D markedly altered the efficacy
and potency of these molecules.
potency (964 nM vs. 2.4 mM), but its efficacy was moderately
reduced (ꢀ115 ꢁ 8 mV, p < 0.05) compared to GoSlo-SR-5-44 [24].
When we examined the effects of the deaminated derivative of the
most potent compound 12 (GoSlo-SR-5-69) of this series, com-
pound 20 (GoSlo-SR-5-134, Table 3), we found the efficacy of
GoSlo-SR-5-134 at 1
m
M was significantly decreased (ꢀ79 ꢁ 6 mV
compared to ꢀ104 ꢁ 8 mV, p < 0.005) and its potency was reduced
more than two-fold compared to GoSlo-SR-5-69. The deaminated
derivative of GoSlo-SR-5-95, compound 21 (GoSlo-SR-5-132,
Table 3) was even less effective and shifted the activation V1/2 by
Interestingly, when we reduced the size of the ring, fused with
ring D, from cyclohexane to cyclopentane (14, GoSlo-SR-5-64,
Table 2), the potency (assessed from the EC50) was reduced by
only ꢀ41 ꢁ 4 mV (n ¼ 6) at a concentration of 1
mM, with an EC50 of
more than 10-fold to 2.9
mM and the efficacy was also modestly
984 nM. The data presented in Table 3 suggest that deaminated
derivatives 20e23 had variable effects on the potency and the ef-
ficacy compared to their aminated forms. However, these data
demonstrate that the primary aromatic amine group present at 1-
position in the C-ring of the anilinoanthraquinone is not essential
for opening BK channels.
reduced compared to GoSlo-SR-5-69. Incorporation of either the
polar functional group or polar atom(s) in the additional ring that
attached with ring D, resulted in less efficacious compounds. Thus,
the tetralone derivative 15 (GoSlo-SR-124, Table 2) significantly
reduced the
D
V1/2 to ꢀ66 ꢁ 8 mV (n ¼ 4, p < 0.05) when applied at
10 M. Similarly, the ability to open the BK channels with com-
m
It will now be of interest to examine the effects of these new BK
channel modulators on mechanical activity in the bladder. How-
ever, before we do this, we will also need to carefully characterize
the effects of these compounds on the other ion channels that are
known to contribute to electrical activity in the bladder (ie Ca2þ
channels, voltage-gated Kþ channels, KATP channels and calcium
activated chloride channels). In addition, we will need to examine if
the compounds have any effects on Ca2þ handling in smooth
muscle cells, before we can reliably interpret the effects of the
compounds on mechanical activity.
pounds 16e18 was dramatically reduced (see Table 2). Additionally,
the incorporation of a nitrogen atom into the D ring (19, GoSlo-SR-
5-127, Table 2) also reduced the efficacy to ꢀ38 ꢁ 6 mV (n ¼ 5,
p < 0.001). These data clearly demonstrated that increasing the
hydrophilicity, either of the D-ring or the ring fused with phenyl
ring D reduced their ability to open BK channels significantly.
3.4. Effect of deaminated derivatives
We also examined the contribution of the primary amine pre-
sent at 1-position of the C-ring to BK channel opening activity.
However, compound 22 (GoSlo-SR-5-130, Table 3) shifted the
4. Summary
activation V1/2 by ꢀ97 ꢁ 4 mV at a concentration of 10
m
M and had
We have synthesised the second generation of GoSlo-SR mole-
cules and examined their effects on BK channels in patches of mem-
brane from rabbit bladder smooth muscle cells. The efficacy of these
molecules was enhanced when the size of the ring D was increased.
Furthermore, efficacywasmaintainedandpotencyenhancedbymore
an EC50 of 1.8 M. Neither of these values was significantly different
m
from the aminated form (1, GoSlo-SR-5-6) reported previously by
us [24]. In contrast, the deaminated form of 2, GoSlo-SR-5-44 [24],
compound 23 (GoSlo-SR-5-138, Table 3) showed an increased

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
431
Fig. 2. Increasing D ring size enhances efficacy of the GoSlo-SR molecules as BK channel agonists. Fig. 2AeD shows BK currents evoked by voltage ramps in 100 nM Ca2þ before
(grey traces) and during drug application (blue traces). The black traces show the effect of 1
m
M Ca2þ on BK currents evoked by voltage ramps from ꢀ100 to þ100 mV. Increasing
Ca2þ at the cytosolic surface of the patch increased the number of channel openings and shifted the voltage at which they opened to more negative potentials. The structure of each
molecule is shown in the inset in each panel. As (A)e(D) suggest, increasing the ring size enhanced the efficacy of each compound as evidenced by the negative shift in the current
voltage relationships in the presence of each molecule. The filled lines represent the fit of the data with the Boltzmann equation. Panel E shows summary data in which the
was plotted against ring size and illustrates that the relationship was not linear. Panel F shows summary data in which the size of the D ring (measured as shown by the red arrow in
D) was plotted against V1/2. The solid line in (F) shows the fit of the data obtained by linear regression. (For interpretation of the references to colour in this figure legend, the
DV1/2
D
reader is referred to the web version of this article.)
than an order of magnitude when the D ring was substituted with
tetrahydro-2-naphthalene. Deamination at 1-position on the C ring
slightly reduced efficacy and potency suggesting that the primary
amine group at this position is not obligatory for BK channel opening
activity in rabbit bladder smooth muscle cells.
magnetic resonance (1H NMR) spectra and carbon nuclear mag-
netic resonance (13C NMR) spectra were recorded on Bruker AMX
400 MHz and Bruker AMX 100 MHz NMR spectrometer, respec-
tively, at 27 ꢂC. All 1H NMR chemical shifts are reported in
d units,
parts per million (ppm) downfield of tetramethylsilane (TMS), and
were measured relative to the signal for dimethylsulfoxide
(2.50 ppm). Data for 1H NMR are reported as follows: chemical shift
5. Experimental
(d
ppm), multiplicity (s ¼ singlet, d ¼ doublet, t ¼ triplet,
q ¼ quartet, m ¼ multiplet, br ¼ broad), integration, and coupling
constant (J Hz). Chemical shifts ( ) for carbon are reported in parts
5.1. Chemistry
d
per million (ppm) downfield from tetramethylsilane (TMS) and are
referenced to residual solvent peaks: carbon (DMSO-d6 39.47 ppm).
All reagents and solvents were obtained from commercial
sources and used without further purification. Proton nuclear

432
S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
Fig. 3. Comparison of the effects of 10
m
M AB-25, GoSlo-SR-5-65 and GoSlo-SR-5-103 on the activation of BK channels. Panels A, D and G show the structure of AB-25, GoSlo-SR-5-
65 and GoSlo-SR-5-103, respectively. Panels B, E and H show the effect of 100 nM (grey traces) and 1
SR-5-65 and GoSlo-SR-5-103, respectively, in the presence of 100 nM Ca2þ (blue traces). Solid lines show Boltzmann fits to the data. Panels C, F and I show summary bar-charts from
separate experiments in which the mean V1/2 in the presence of 100 nM Ca2þ (grey bars) and 1 M Ca2þ (black bars) on patches before application of 10
M AB-25 (blue bars, Panel
C), 10 M GoSlo-SR-5-65 (blue bars, Panel F) or 10 M GoSlo-SR-5-103 (blue bars, Panel I). (For interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)
m mM AB-25, GoSlo-
M Ca2þ (black traces) on BK currents before application of 10
m
m
m
m
The purities of isolated products were determined on Agilent 1200
series system (Agilent Technologies, Germany) using the following
procedure: the compounds were dissolved at a concentration of
0.5 mg/mL in purified water:acetonitrile ¼ 90:10 (% v/v), containing
focusedÔ Microwave Synthesis type DiscoverÒ apparatus. Re-
actions were monitored by analytical thin layer chromatography
(TLC), which was performed on aluminium sheets pre-coated with
silica gel 60 F254, or aluminium sheets pre-coated with RP silica gel
18 F254 (Merck, Germany). Column flash chromatography was
performed on Merck Kieselgel 60 (0.040e0.063 mm) or Silica gel
100C18-Reversed phase (Fluka, Switzerland). The columns, packed
with Merck Kieselgel 60, were eluted with various combinations of
ethyl acetate and methanol mixtures. The columns, packed with
Silica gel 100C18-Reversed phase, were eluted with various com-
binations of water and methanol mixtures. A freeze-dryer (CHRIST
ALPHA 1-4 LSC) was used for lyophilisation.
0.1% trifluoroacetic acid (TFA). Sample (5 mL) was then injected into
an Agilent Poroshell 120 EC-C18 column (100 ꢃ 3.0 mm I.D., 2.7
mm
particle size). Elution was performed with a linear gradient of pu-
rified water:acetonitrile (containing for both mobile phases 0.1%
TFA) from 90:10 (% v/v) to 20:80 (% v/v) during 17 min, then return
to original conditions in 1 min, followed by keeping for re-
equilibration at original conditions (90:10% v/v purified water:-
acetonitrile containing 0.1% TFA) for 3 min, at a flow rate of 700 mL/
min. UV absorption was detected from 200 to 750 nm using a diode
array detector and the purity of the compounds was determined at
254 nm (see the Supporting Information). The School of Chemistry
and Chemical Biology, University College Dublin recorded high
Resolution Mass Spectra (HRMS) using a Micromass/Waters LCT
instrument. Microwave reactions were carried out using a CEM
5.1.1. General procedure A: preparation of 3e5 and 10e19
To a 30 mL microwave reaction vial, buffer solution of 0.2 M
Na2HPO4 (3 mL) and 0.12 M NaH2PO4 (5 mL), bromaminic acid
sodium salt (0.2 g, 0.495 mmol), a catalytic amount (w4 mg) of
copper powder and the appropriate aniline or amine derivative

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
433
Fig. 4. Comparison of the effects of 1
and GoSlo-SR-5-95, respectively. Panels B, E and H show the effect of 100 nM (grey traces) and 1
5-69 and GoSlo-SR-5-95, respectively, in the presence of 100 nM Ca2þ (blue traces). Solid lines show Boltzmann fits to the data. Panels C, F and I show summary bar-charts from
separate experiments in which the mean V1/2 in the presence of 100 nM Ca2þ (grey bars) and 1 M Ca2þ (black bars) on patches before application of 1
M AB-25 (blue bars, Panel C),
M GoSlo-SR-5-69 (blue bars, Panel F) or 1 M GoSlo-SR-5-95 (blue bars, Panel I). (For interpretation of the references to colour in this figure legend, the reader is referred to the
web version of this article.)
m
M AB-25, GoSlo-SR-5-69 and GoSlo-SR-5-95 on the activation of BK channels. Panels A, D and G show the structure of AB-25, GoSlo-SR-5-69
m
M Ca2þ (black traces) on BK currents before application of 1
mM AB-25, GoSlo-SR-
m
m
1
m
m
(0.99 mmol) were added. The resulting suspension was capped and
irradiated in the microwave oven for 5e20 min at 100e120 ꢂC. The
reaction mixture was then cooled to room temperature, filtered and
extracted with ethyl acetate (3 ꢃ 50 mL). The combined organic
layers were dried over anhydrous Na2SO4 and concentrated under
reduced pressure. The crude residue was purified by silica gel col-
umn chromatography (eluted with methanol/ethyl acetate, 1:49 (v/
v)) to furnish the anilinoanthraquinone derivatives as blue solids.
5.1.1.2. Sodium 1-amino-4-(cyclobutylamino)-9,10-dioxo-9,10-
dihydroanthracene-2-sulfonate, 4. Yield: 40%; 1H NMR (400 MHz,
DMSO-d6, 27 ꢂC):
d
10.74 (d, J ¼ 6.0 Hz, 1H), 10.10 (brs, 1H), 8.27e
8.23 (m, 2H), 7.83e7.80 (m, 2H), 7.64 (s, 1H), 7.41 (brs, 1H), 4.22 (m,
1H), 2.49e2.46 (m, 2H), 2.06e1.96 (m, 2H), 1.90e1.83 (m, 2H); 13C
NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 181.67, 181.09, 143.71, 143.07,
143.04, 133.88, 133.78, 132.62, 132.58, 125.89, 125.69, 121.45, 109.11,
108.84, 47.08, 30.86, 15.03; HRMS (ES): m/z for C18H15N2O5S
[M ꢀ Naþ], calcd. 371.0702, found 371.0692.
5.1.1.1. Sodium 1-amino-4-(cyclopropylamino)-9,10-dioxo-9,10-
5.1.1.3. Sodium 1-amino-9,10-dioxo-4-((5,6,7,8-tetrahydronaphthalen-
dihydroanthracene-2-sulfonate, 3. Yield: 39%; 1H NMR (400 MHz,
1-yl)amino)-9,10-dihydroanthracene-2-sulfonate, 10. Yield: 50%; 1H
DMSO-d6, 27 ꢂC):
d
10.51 (d, J ¼ 2.4 Hz, 1H), 10.09 (brs, 1H), 8.27e
NMR (400 MHz, DMSO-d6, 27 ꢂC):
d 12.03 (s, 1H), 10.15 (brs, 1H),
8.24 (m, 1H), 8.22e8.20 (m, 1H), 8.17 (s, 1H), 7.82e7.80 (m, 2H),
8.31e8.28 (m, 2H), 7.88e7.83 (m, 2H), 7.83 (s, 1H), 7.50 (brs, 1H), 7.19
(t, J ¼ 7.6 Hz,1H), 7.09 (d, J ¼ 7.6 Hz,1H), 7.0 (d, J ¼ 7.2 Hz,1H), 2.80 (t,
J ¼ 5.6 Hz, 2H), 2.68 (t, J ¼ 5.6 Hz, 2H), 1.81e1.74 (m, 4H); 13C NMR
2.80e2.67 (m, 1H), 0.94e0.90 (m, 2H), 0.65e0.62 (m, 2H); 13C NMR
(100 MHz, DMSO-d6, 27 ꢂC):
d 181.67, 181.44, 145.84, 143.43, 143.21,
133.96, 133.73, 132.62, 132.55, 125.89, 125.71, 122.10, 109.19, 109.05,
24.10, 7.63; HRMS (ES): m/z for C17H13N2O5Na2S [M þ Naþ], calcd.
403.0341, found 403.0359.
(100 MHz, DMSO-d6, 27 ꢂC):
d 182.24, 181.71, 143.97, 142.37, 141.96,
138.81, 137.15, 134.03, 133.51, 133.11, 132.79, 130.94, 126.16, 126.01,
125.93, 122.81, 121.40, 110.78, 109.03, 29.16, 24.66, 22.35, 22.12;

434
S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
Fig. 5. Comparing potency of the GoSlo-SR molecules with AB-25. (A) Typical BK currents obtained from voltage ramps in response to increasing concentrations of GoSlo-SR-5-69
(blue traces). The solid lines show the Boltzmann fits to the data. (B) A summary plot of the
D
V1/2 (ꢁSEM) obtained from patches bathed in 100 nM Ca2þ (Control) before and in the
continued presence of increasing concentrations of GoSlo-SR-5-69. The solid lines in (BeF) show the fit to the data obtained with the Langmuir equation. The structure of each
compound and its EC50 are shown in the inset. For all EC50 determinations, different concentrations of each compound were applied to between 5 and 30 patches, with the
exception of 300 nM GoSlo-SR-5-103 and AB-25, which were applied to 4 and 2 patches, respectively. As (B) suggests, GoSlo-SR-5-69 was the most potent and efficacious compound
of this series and was approximately 24-fold more potent than AB-25 (D). (C) and (E) show that GoSlo-SR-5-65 and GoSlo-SR-5-95 possessed similar potencies and efficacies. As (F)
suggests, GoSlo-SR-103 was approximately seven-fold more potent than AB-25, but was less efficacious. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
HRMS (ES): m/z for C24H19N2O5Na2S [M þ Naþ], calcd. 493.0810,
125.99, 125.89, 123.91, 122.80, 121.10, 110.68, 109.01, 28.76, 28.27,
22.66, 22.48; HRMS (ES): m/z for C24H19N2O5Na2S [M þ Naþ], calcd.
493.0810, found 493.0821.
found 493.0797.
5.1.1.4. Sodium 1-amino-9,10-dioxo-4-((5,6,7,8-tetrahydronaphthalen-
2-yl)amino)-9,10-dihydroanthracene-2-sulfonate, 12. Yield: 76%; 1H
5.1.1.5. Sodium 1-amino-4-((2,3-dihydro-1H-inden-5-yl)amino)-9,10-
NMR (400 MHz, DMSO-d6, 27 ꢂC):
d
12.07 (s, 1H), 10.15 (brs, 1H),
dioxo-9,10-dihydroanthracene-2-sulfonate, 14. Yield: 66%; 1H NMR
8.30e8.27 (m, 2H), 7.97 (s, 1H), 7.88e7.82 (m, 2H), 7.49 (brs, 1H), 7.13
(d, J ¼ 8.0 Hz,1H), 7.02e6.99 (m, 2H), 2.75 (brs, 4H),1.77 (t, J ¼ 2.8 Hz,
(400 MHz, DMSO-d6, 27 ꢂC):
d 12.10 (s,1H),10.15 (brs,1H), 8.30e8.26
(m, 2H), 7.96 (s, 1H), 7.88e7.81 (m, 2H), 7.50 (brs, 1H), 7.29 (d,
J ¼ 8.0 Hz, 1H), 7.15 (s, 1H), 7.04 (dd, J ¼ 2.0, 8.0 Hz, 1H), 2.92e2.88
(m, 4H), 2.11e2.03 (m, 2H); 13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
4H); 13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 182.04, 181.72, 144.04,
142.43, 141.67, 138.15, 136.11, 134.02, 133.54, 133.07, 132.77, 130.05,

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
435
d
181.98, 181.65, 145.48, 144.14, 142.77, 141.90, 140.59, 137.06, 134.09,
at 105 ꢂC. The reaction mixture was then cooled to room temper-
ature, filtered and extracted with ethyl acetate (3 ꢃ 50 mL). The
combined organic layers were dried over anhydrous sodium sul-
phate and concentrated under reduced pressure to obtain the crude
residue. The residue was purified by silica gel column chromatog-
raphy (eluted with methanol/ethyl acetate, 1:49 (v/v)) to furnish
the compounds 7 and 8 as blue solids.
133.61, 132.97, 132.67, 125.98, 125.88, 125.12, 122.70, 121.94, 119.90,
110.62, 108.96, 32.44, 31.84, 25.18; HRMS (ES): m/z for C23H17N2O5
-
Na2S [M þ Naþ], calcd. 479.0654, found 479.0677.
5.1.1.6. Sodium 1-amino-9,10-dioxo-4-((5-oxo-5,6,7,8-
tetrahydronaphthalen-2-yl)amino)-9,10-dihydroanthracene-2-
sulfonate, 15. Yield: 36%; 1H NMR (400 MHz, DMSO-d6, 27 ꢂC):
d
11.70 (s, 1H), 10.01 (brs, 1H), 8.28e8.22 (m, 2H), 8.16 (s, 1H), 7.90e
5.1.2.1. Sodium 1-amino-4-(cycloheptylamino)-9,10-dioxo-9,10-
dihydroanthracene-2-sulfonate, 7. Yield: 49%; 1H NMR (400 MHz,
7.83 (m, 3H), 7.51 (brs,1H), 7.19e7.16 (m, 2H), 2.93 (t, J ¼ 6.0 Hz, 2H),
2.59e2.55 (m, 2H), 2.08e2.02 (m, 2H); 13C NMR (100 MHz, DMSO-
DMSO-d6, 27 ꢂC):
d
10.95 (d, J ¼ 7.6 Hz, 1H), 10.18 (brs, 1H), 8.27e
d6, 27 ꢂC):
d 196.04, 183.54, 182.27, 146.72, 144.85, 144.75, 141.67,
8.24 (m, 2H), 7.81e7.79 (m, 2H), 7.75 (s, 1H), 7.40 (brs, 1H), 3.91e
3.86 (m, 1H), 2.02e1.96 (m, 2H), 1.70e1.59 (m, 10H); 13C NMR
137.12, 133.99, 133.65, 133.23, 133.02, 128.36, 127.02, 126.06, 124.14,
119.07, 118.29, 114.45, 109.80, 38.37, 29.14, 22.77; HRMS (ES): m/z
for C24H17N2O6Na2S [M þ Naþ], calcd. 507.0603, found 507.0623.
(100 MHz, DMSO-d6, 27 ꢂC):
d 181.45, 180.58, 144.25,143.54, 142.96,
133.97, 133.91, 132.45, 132.33, 125.83, 125.69, 121.40, 109.08, 108.67,
52.08, 34.67, 27.57, 23.51; HRMS (ES): m/z for C21H21N2O5Na2S
[M þ Naþ], calcd. 459.0967, found 459.0970.
5.1.1.7. Sodium 1-amino-4-((2,3-dihydrobenzo[b] [1,4]dioxin-6-yl)
amino)-9,10-dioxo-9,10-dihydroanthracene-2-sulfonate,
16.
Yield: 60%; 1H NMR (400 MHz, DMSO-d6, 27 ꢂC):
d
11.99 (s, 1H),
5.1.2.2. Sodium 1-amino-4-(cyclooctylamino)-9,10-dioxo-9,10-
10.15 (brs, 1H), 8.30e8.27 (m, 2H), 7.90 (s, 1H), 7.86e7.84 (m, 2H),
7.49 (brs, 1H), 6.94 (d, J ¼ 8.4 Hz, 1H), 6.82 (d, J ¼ 2.4 Hz, 1H), 6.77
(dd, J ¼ 2.4, 8.8 Hz, 1H), 4.29 (s, 4H); 13C NMR (100 MHz, DMSO-d6,
dihydroanthracene-2-sulfonate, 8. Yield: 57%; 1H NMR (400 MHz,
DMSO-d6, 27 ꢂC):
d
10.95 (d, J ¼ 7.6 Hz, 1H), 10.18 (brs, 1H), 8.27e
8.24 (m, 2H), 7.82e7.79 (m, 2H), 7.76 (s, 1H), 7.40 (brs, 1H), 3.92e
27 ꢂC):
d 181.93, 181.64, 144.10, 143.89, 142.86, 142.06, 141.06,
3.86 (m, 1H), 1.96e1.90 (m, 2H), 1.77e1.59 (m, 12H); 13C NMR
134.09, 133.61, 132.98, 132.68, 132.26, 125.99, 125.87, 122.56, 117.80,
117.37, 112.89, 110.48, 108.92, 64.16, 63.96; HRMS (ES): m/z for
C
(100 MHz, DMSO-d6, 27 ꢂC):
d 181.45, 180.54, 144.21, 143.48, 142.94,
133.97, 133.90, 132.46, 132.33, 125.83, 125.69, 121.50, 109.11, 108.69,
51.28, 32.21, 26.62, 25.14, 23.06; HRMS (ES): m/z for C22H23N2O5S
[M ꢀ Hþ], calcd. 427.1328, found 427.1327.
22H15N2O7S [M ꢀ Hþ], calcd. 451.0600, found 451.0619.
5 .1.1. 8 . S o d i u m 1 - a m i n o - 9 ,10 - d i o x o - 4 - ( ( 3 - o x o - 1, 3 -
dihydroisobenzofuran-5-yl)amino)-9,10-dihydroanthracene-2-
sulfonate, 17. Yield: 38%; 1H NMR (400 MHz, DMSO-d6, 27 ꢂC):
5.1.3. General procedure C: preparation of 20e23
To
a
suspension of 1-amino anthraquinone derivative
d
11.89 (s, 1H), 10.04 (brs, 1H), 8.29e8.25 (m, 2H), 7.98 (s, 1H), 7.88e
7.85 (m, 2H), 7.74e7.68 (m, 3H), 7.48 (brs, 1H), 5.44 (s, 2H); 13C NMR
(100 MHz, DMSO-d6, 27 ꢂC):
183.04, 182.09, 170.31, 144.40, 142.70,
(0.15 mmol) in 1 M HCl (10 mL) was added 1.5 mL aqueous solution
of sodium nitrite (0.60 mmol) at 0 ꢂC. The reaction mixture was
stirred for 5 min at same temperature and then allowed to stir at
room temperature for 1 h. Ten millilitres of ethanol followed by zinc
dust (1.5 mmol) were added into the reaction mixture and further
stirred at room temperature for 5 min. Reaction mixture was then
quenched with 0.5 M sodium bicarbonate solution and extracted
with ethyl acetate (2 ꢃ 50 mL). Organic layers were dried over
anhydrous sodium sulphate and concentrated in vacuo. The crude
residue was purified by C18-reversed phase silica gel column
chromatography (eluted with water and methanol mixtures).
Elution was performed with increased polarity of purified water:-
methanol mixtures from 90:10 (% v/v) to 50:50 (% v/v) to furnish
the pure deaminated derivative in excellent yield as purple solid.
d
142.09, 140.61, 139.50, 134.00, 133.46, 133.31, 132.96, 129.16, 126.41,
126.05, 125.99, 124.26, 122.67, 117.50, 112.71, 109.51, 69.92; HRMS
(ES): m/z for C22H13N2O7S [M ꢀ Naþ], calcd. 449.0443, found
449.0430.
5.1.1.9. Sodium 1-amino-9,10-dioxo-4-(quinolin-6-ylamino)-9,10-
dihydroanthracene-2-sulfonate, 18. Yield: 41%; 1H NMR (400 MHz,
DMSO-d6, 27 ꢂC):
d 11.96 (s, 1H), 10.08 (brs, 1H), 8.97 (s, 1H), 8.31e
8.26 (m, 3H), 8.14 (s, 1H), 8.07 (d, J ¼ 9.2 Hz, 1H), 7.90e7.83 (m, 2H),
7.80 (d, J ¼ 2.4 Hz, 1H), 7.73 (dd, J ¼ 2.4, 9.2 Hz, 1H), 7.54 (q,
J ¼ 4.0 Hz, 1H); 13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 182.97,
182.01, 149.27, 144.55, 142.34, 139.63, 137.69, 135.27, 134.06, 133.41,
133.38, 132.89, 130.32, 126.37, 126.05, 125.99, 122.99, 118.12, 112.58,
109.44; HRMS (ES): m/z for C23H15N3O5NaS [M þ Hþ], calcd.
468.0630, found 468.0630.
5.1.3.1. 9,10-Dioxo-4-((5,6,7,8-tetrahydronaphthalen-2-yl)amino)-
9,10-dihydroanthracene-2-sulfonic acid, 20. Yield: 96%; 1H NMR
(400 MHz, DMSO-d6, 27 ꢂC):
d
11.22 (s, 1H), 8.26 (d, J ¼ 7.6 Hz, 1H),
8.20 (d, J ¼ 7.2 Hz, 1H), 7.96e7.87 (m, 2H), 7.79 (s, 1H), 7.64 (s, 1H),
7.17 (d, J ¼ 8.0 Hz, 1H), 7.08e7.04 (m, 2H), 2.76 (s, 4H), 1.78 (s, 4H);
5.1.1.10. Sodium 1-amino-9,10-dioxo-4-(quinolin-3-ylamino)-9,10-
dihydroanthracene-2-sulfonate, 19. Yield: 42%; 1H NMR (400 MHz,
13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 184.18, 182.43, 154.08,
DMSO-d6, 27 ꢂC):
d 12.01 (s, 1H), 10.06 (brs, 1H), 8.91 (s, 1H), 8.31e
149.50,138.24, 135.78, 134.62,134.30,134.28, 134.07,133.78, 132.49,
130.14, 126.60, 126.46, 124.85, 121.92, 115.83, 114.62, 112.65, 28.78,
28.33, 22.64, 22.46; HRMS (ES): m/z for C24H18NO5S [M ꢀ Hþ],
calcd. 432.0906, found 432.0901.
8.28 (m, 2H), 8.18 (d, J ¼ 2.4 Hz, 1H), 8.04 (s, 1H), 8.03 (d, J ¼ 7.2 Hz,
1H), 7.93 (d, J ¼ 8.4 Hz, 1H), 7.90e7.87 (m, 2H), 7.73e7.61 (m, 2H),
7.50 (brs, 1H); 13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 183.17, 182.10,
147.45, 144.62, 142.42, 139.36, 134.10, 133.49, 133.36, 132.95, 128.47,
128.21, 127.45, 126.09, 126.01, 125.96, 122.49, 113.01, 109.57; HRMS
(ES): m/z for C23H14N3O5S [M ꢀ Naþ], calcd. 444.0654, found
444.0653.
5.1.3.2. 9,10-Dioxo-4-((3-(trifluoromethyl)phenyl)amino)-9,10-
dihydroanthracene-2-sulfonic acid, 22. Yield: 98%; 1H NMR
(400 MHz, DMSO-d6, 27 ꢂC):
d
11.24 (s, 1H), 8.26 (dd, J ¼ 1.2, 7.8 Hz,
5.1.2. General procedure B: preparation of 7 and 8
1H), 8.21 (dd, J ¼ 1.2, 7.4 Hz, 1H), 7.98e7.89 (m, 2H), 7.89 (d,
J ¼ 1.2 Hz, 1H), 7.76 (d, J ¼ 1.2 Hz, 1H), 7.74e7.68 (m, 3H), 7.58 (d,
To a solution of bromaminic acid sodium salt (0.2 g, 0.495 mmol)
in water (5 mL) was added copper sulphate (0.1 mmol) and sodium
carbonate (0.4 mmol), followed by cycloheptylamine or cyclo-
octylamine (0.99 mmol). The reaction mixture was stirred for 6e8 h
J ¼ 6.4 Hz, 1H); 13C NMR (100 MHz, DMSO-d6, 27 ꢂC):
d 184.49,
182.30, 154.28, 147.61, 140.13, 134.71, 134.36, 134.16, 134.02, 132.49,
130.83, 127.09, 126.67, 126.53, 119.92, 115.90, 115.67, 114.24; HRMS

436
S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
(ES): m/z for C21H11NO5F3S [M ꢀ Hþ], calcd. 446.0310, found
or ꢀ50 to þ150 mV over 2 s before returning to ꢀ100 mV. These
voltage ramps were usually repeated 15 times and the currents
were averaged. The currents were divided by the single channel
conductance at each voltage and thus corrected for driving force to
obtain the number of channels (n) multiplied by the open proba-
bility (n.Po). Maximum n.Po was determined by bathing the patch
446.0305.
5.1.3.3. 4-((4-Methyl-3-(trifluoromethyl)phenyl)amino)-9,10-dioxo-
9,10-dihydroanthracene-2-sulfonic acid, 23. Yield: 96%; 1H NMR
(400 MHz, DMSO-d6, 27 ꢂC):
d
11.19 (s, 1H), 8.26 (d, J ¼ 6.8 Hz, 1H),
8.21 (d, J ¼ 7.2 Hz, 1H), 7.97e7.89 (m, 2H), 7.85 (d, J ¼ 1.2 Hz, 1H),
in 1
m
M and repeating the voltage ramps. In some examples shown
7.65 (d, J ¼ 1.2 Hz, 2H) 7.59e7.54 (m, 2H), 2.46 (s, 3H); 13C NMR
in the presence of 1
m
M Ca2þ, voltage ramps were only repeated 5e
(100 MHz, DMSO-d6, 27 ꢂC):
d
184.38, 182.35, 154.25, 148.30, 137.24,
10 times but it was still possible from this to estimate the maximum
number of channels in the patch. The data were then fitted with the
Boltzmann equation of the form:
134.70, 134.27, 134.18, 133.96, 133.62, 132.48, 127.67, 126.64, 126.51,
121.34, 121.28, 115.68, 115.28, 113.73, 18.27; HRMS (ES): m/z for
C
22H13NO5F3S [M ꢀ Hþ], calcd. 460.0467, found 460.0481.
ꢀ
ꢀꢀ
ꢁ. ꢁꢁ
n:Po ¼ 1= 1 þ exp V1=2 ꢀ Vpatch
K
5.2. Computational details for calculating D ring size
where V1/2 was the membrane potential at which there was half
maximal activation, K the slope factor and Vpatch the patch potential
(mV). The change in activation V1/2 (DV1/2) due to the drugs was
obtained by subtracting the V1/2 in control from that measured in
the presence of the drug. The V1/2 is given as the mean ꢁ SEM. The
voltage ramps were sufficiently slow (100 mV sꢀ1) so that the
activation curves were not distorted by the time constants of acti-
To estimate the size of ring D shown in Fig. 2F, calculations were
performed within the density functional theory using the B3LYP
exchange-correlation functional as implemented in the Gaussian09
program package. Geometry optimizations were performed
without any symmetry constraints using the 6-311þþg(d,p) basis
set. Optimized geometries were characterized as minima by
computing the harmonic vibrational frequencies. Coordinates of all
optimized compounds are available upon request.
vation or deactivation. This analysis provided
a continuous
recording of the n.Po over the entire voltage range [33]. The ex-
periments were carried out between 35 and 37 ꢂC. BK channels
were observed in almost every patch and were judged to be BK
channels on the basis of their (i) large conductance, (ii) reversal
potential at EK (0 mV), (iii) their calcium and voltage sensitivity, and
(iv) their blockade by the selective BK channel blockers, penitrem A,
(100 nM) [27] and iberiotoxin.
5.3. Rabbit bladder smooth muscle cell preparation
All experiments were approved by the Dundalk Institute of
Technology (DkIT) Animal Care and Use Committee and were in
accordance with EU legislation. The urinary bladder was removed
from both male and female New Zealand White rabbits after they
had been killed by intravenous injection of pentobarbitone. The
bladder was immediately placed in Krebs’ solution, opened up
longitudinally and the urothelium removed by sharp dissection.
Strips of tissue, 0.5 cm in length, were cut into 1 mm3 pieces and
stored in Hanks Ca2þ free solution for 30 min at 4 ꢂC prior to cell
dispersal. Smooth muscle cells were isolated from the tissue pieces
by w5 min incubation in a dispersal medium containing (per 5 ml
of Ca2þ free Hanks solution): 1 mg protease (Sigma type XXIV),
10 mg trypsin inhibitor (Sigma), 10 mg BSA (Sigma) and 15 mg
collagenase (Sigma type 1a) while stirring at 35e37 ꢂC.
5.5. Drug delivery
The patches of cell membrane under study were continuously
perfused by means of a gravity-fed close delivery system for
application of pharmacological agents.
(w100e200 m in diameter) was attached at the end of the de-
livery system and its tip placed w150 m from the patch. Using this
A glass micropipette
m
m
system the flow could be changed over to a solution containing the
drug within 5 s. During experimentation, excised patches were
bathed with symmetrical Kþ solutions (solution 4) containing
100 nM Ca2þ, with or without the relevant drug concentration.
Although the physicochemical properties of the GoSlo-SR mole-
cules have not been thoroughly determined, we saw no evidence of
aggregation of the drugs in solution when used at concentrations
The tissue suspension was centrifuged at 1000 revolu-
tions minꢀ1 for approximately 1 min and the supernatant removed.
The tissue pieces were resuspended in Ca2þ free Hanks solution and
stirred for 6e8 min to release single relaxed smooth muscle cells.
The isolated cells were then plated in Petri dishes containing
up to 10 mM.
100 m
M Ca2þ (which enhanced adherence of the cells to the bottom
of the dish) and were stored at 4 ꢂC until use within 8 h.
5.6. Solutions for electrophysiology experiments
5.4. Electrophysiology recordings
The following solutions were used. Concentrations in mM are
given in parentheses.
To screen molecules for biological activity, we utilised voltage
ramps to record BK channel openings in inside/out patches of
membrane excised from smooth muscle cells isolated from rabbit
bladders. Electrodes were pulled from Corning borosilicate glass
(1.5 mm O.D. ꢃ 0.86 mm I.D.) using a Sutter P-97 pipette puller and
were fire polished using a Narashige MF 83 microforge. Standard
single channel patch clamp recording methods were used in the
inside-out patch conformation [32]. Voltage clamp commands
were delivered via either Axopatch 1D or Axopatch 200B patch
clamp amplifiers (Axon Instruments) connected to Digidata 1322A
AD/DA converters (Axon Instruments) interfaced to a computer
running pClamp software (Axon Instruments). Data was acquired at
10 KHz and filtered at 2 kHz. Patches were usually held at ꢀ60 mV
or ꢀ100 mV (as denoted in the text) and the patch potential was
ramped from either ꢀ150 mV to þ 50 mV, e100 mV to þ100 mV
A. Krebs’ solution: NaCl (120), KCl (5.9), NaHCO3 (25), NaH2
-
PO4$2H2O (1.2), Glucose (5.5), MgCl2 (1.2), CaCl2 (2.5). pH was
adjusted to 7.4 by bubbling the solution with 95% O2e5% CO2
continuously.
B. Ca2þ free Hanks solution for cell dispersal: NaCl (125), KCl (5.4),
Glucose (10), Sucrose (2.9), NaHCO3 (4.2), KH2PO4 (0.4),
NaH2PO4 (0.3), Hepes (10). pH was adjusted to 7.4 using NaOH.
Solutions 1 and 2 were made up in double distilled water.
C. 100 m
M Ca2þ H. solution: CaCl2$2H2O (0.1 M) in 100 ml of Ca2þ
free Hanks solution.
D. Single channel recording solutions: all recordings were carried
out with symmetrical KCl solutions of the following composi-
tion (mM): KCl (140), Glucose (10), HEPES (10), EGTA (1) [for

S. Roy et al. / European Journal of Medicinal Chemistry 75 (2014) 426e437
437
calcium free and 100 nM Ca2þ], H-EDTA (1) [for 1 M Ca2þ], pH
m
of their NMR facility and Billie McIlveen for excellent technical
assistance.
adjusted to 7.2 using KOH, made up in de-ionised, filtered water
obtained from a Milli-Q water purification system. Solutions
containing 100 nM Ca2þ were used as the pipette solution and
the Chelator program [34], was used to calculate the concen-
tration of calcium required to give the desired free [Ca2þ]i.
Available for download at http://www.ru.nl/organphy/chelator/
Chelmain.html
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.01.035.
5.7. Statistics
References
[1] L. Salkoff, A. Butler, G. Ferreira, C. Santi, A. Wei, Nat. Rev. Neurosci. 7 (2006)
921e931.
[2] H. Hu, L.R. Shao, S. Chavoshy, N. Gu, M. Trieb, R. Behrens, P. Laake, O. Pongs,
H.G. Knaus, O.P. Ottersen, J.F. Storm, J. Neurosci. 21 (2001) 9585e9597.
[3] G. Raffaelli, C. Saviane, M.H. Mohajerani, P. Pedarzani, E. Cherubini, J. Physiol.
557 (2004) 147e157.
[4] Z.W. Wang, Mol. Neurobiol. 38 (2008) 153e166.
[5] Y. Imaizumi, S. Henmi, Y. Uyama, K. Atsuki, Y. Torii, Y. Ohizumi, M. Watanabe,
Am. J. Physiol. 271 (1996) C772eC782.
[6] T.B. Bolton, Y. Imaizumi, Cell. Calcium 20 (1996) 141e152.
[7] U. Klokner, G. Isenberg, Pflugers Arch. 405 (1985) 329e339.
[8] A.G. Davies, J.T. Pierce-Shimomura, H. Kim, M.K. VanHoven, T.R. Thiele,
A. Bonci, C.I. Bargmann, S.L. McIntire, Cell 115 (2003) 655e666.
[9] M. Guest, K. Bull, R.J. Walker, K. Amliwala, V. O’Connor, A. Harder, L. Holden-
Dye, N.A. Hopper, Int. J. Parasitol. 37 (2007) 1577e1588.
[10] L. Holden-Dye, V. O’Connor, N.A. Hopper, R.J. Walker, A. Harder, K. Bull,
M. Guest, Invert. Neurosci. 7 (2007) 199e208.
[11] T.J. Heppner, A.D. Bonev, M.T. Nelson, Am. J. Physiol. 273 (1997) C110eC117.
[12] B. Kyle, E. Bradley, S. Ohya, G.P. Sergeant, N.G. McHale, K.D. Thornbury,
M.A. Hollywood, Am. J. Physiol. Cell. Physiol. 301 (2011) C1186eC1200.
[13] H. Hashitani, H. Fukuta, H. Takano, M.F. Klemm, H. Suzuki, J. Physiol. 530
(2001) 273e286.
Experiments on freshly dispersed rabbit bladder smooth muscle
cells were usually carried out on a minimum of three animals. In all
experiments ‘n’ refers to the number of patches studied. Summary
data are presented as the mean ꢁ SEM and statistical comparisons
were made on raw data using students paired or unpaired t-test as
appropriate, taking p < 0.05 level as significant. Concentratione
effect data were fitted with a HilleLangmuir equation of the form:
ꢀ
ꢁ
^
Y ¼ Bottom þ ðTop ꢀ BottomÞ 1 þ 10ðð½Drugꢄ ꢀ pEC50Þ*HÞ
where DV1/2 was recorded in the presence of the drug under study,
pEC50, was the ꢀlog of the half maximal effective concentration,
[Drug] is the agonist concentration variable and H was the Hill
coefficient.
Conflict of interest statement
[14] H. Hashitani, A.F. Brading, Br. J. Pharmacol. 140 (2003) 159e169.
[15] K.L. Hristov, M. Chen, W.F. Kellett, E.S. Rovner, G.V. Petkov, Am. J. Physiol. Cell.
Physiol. 301 (2011) C903eC912.
[16] K.L. Hristov, S.A.Y. Afeli, S.P. Parajuli, Q. Cheng, E.S. Rovner, G.V. Petkov, PLoS
One 8 (2013) e68052.
K. D. Thornbury, N. G. McHale, G. P. Sergeant, S. Roy and M. A.
Hollywood have submitted a Patent application (IPN WO 2012/
035122 A11) on this family of molecules.
[17] K.L. Hristov, S.P. Parajuli, R.P. Soder, Q. Cheng, E.S. Rovner, G.V. Petkov, Am. J.
Physiol. Cell. Physiol. 302 (2012) C1632eC1641.
[18] A.L. Meredith, K.S. Thorneloe, M.E. Werner, M.T. Nelson, R.W. Aldrich, J. Biol.
Chem. 279 (2004) 36746e36752.
Author contributions
SR performed synthesis and characterization of the compounds.
RL, AMA and AK performed the electrophysiology experiments. CD
performed atomic calculations. RL, AK, AMA & MAH analysed the
data. All authors designed experiments, contributed to and
approved the final manuscript.
[19] T.M. Argentieri, J.A. Butera, Expert Opin. Ther. Pat. 16 (2006) 573e585.
[20] M.L. Garcia, D.M. Shen, G.J. Kaczorowski, Expert Opin. Ther. Pat. 17 (2007)
831e842.
[21] A. Nardi, V. Calderone, S.P. Olesen, Lett. Drug. Des. Dis. 3 (2006) 210e218.
[22] A. Nardi, S.P. Olesen, Exp. Opin. Ther. Pat. 17 (2007) 1215e1226.
[23] A. Nardi, S.P. Olesen, Curr. Med. Chem. 15 (2008) 1126e1146.
[24] S. Roy, A.M. Akande, R.J. Large, T.I. Webb, C. Camarasu, G.P. Sergeant,
N.G. McHale, K.D. Thornbury, M.A. Hollywood, Chem Med Chem 7 (2012)
1763e1769.
Acknowledgements
[25] D. Strøbaek, P. Christophersen, N.R. Holm, P. Moldt, P.K. Ahring, T.E. Johansen,
S.P. Olesen, Neuropharmacology 35 (1996) 903e914.
[26] B.H. Bentzen, A. Nardi, K. Calloe, L.S. Madsen, S.P. Olesen, M. Grunnet, Mol.
Pharmacol. 72 (2007) 1033e1044.
[27] H. Knaus, O. McManus, S. Lee, W. Schmalhofer, M. Garcia-Calvo, L. Helms,
M. Sanchez, K. Giangiacomo, J. Reuben, A. Smith, G. Kaczorowski, M. Garcia,
Biochemistry 33 (1994) 5819e5828.
[28] Y. Baqi, C.E. Müller, Org. Lett. 9 (2007) 1271e1274.
[29] Y. Baqi, S. Weyler, J. Iqbal, H. Zimmermann, C.E. Müller, Pur. Signal 5 (2009)
91e106.
[30] Y. Baqi, S.Y. Lee, J. Iqbal, P. Ripphausen, A. Lehr, A.B. Scheiff, H. Zimmermann,
J. Bajorath, C.E. Müller, J. Med. Chem. 53 (2010) 2076e2086.
[31] Y. Baqi, C.E. Müller, Tetrahedron Lett. 53 (2012) 6739e6742.
[32] O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Pflügers Arch. 391
(1981) 85e100.
[33] A. Carl, K.M. Sanders, J. Neurosci. Methods 33 (1990) 157e163.
[34] T.J. Schoenmakers, G.J. Visser, G. Flik, A.P. Theuvenet, Biotechniques 12 (1992)
870-874e876-879.
This work was funded by Enterprise Ireland’s Applied Research
Enhancement Scheme. A. M. Akande was funded by a Council of
Directors Strand 1 Award and DkIT Research Office. A.K was sup-
ported by Science Foundation Ireland under their Research Fron-
tiers Programme. C. Domene would like to acknowledge the use of
the EPSRC UK National Service for Computational Chemistry Soft-
ware (NSCCS) at Imperial College London in carrying out the
computational work. We would also like to thank Shanghai
YoungLong Chem-Tech Co., Ltd. for supplying Acid Blue 62 as a gift.
The funding agencies did not participate in study design; in the
collection, analysis and interpretation of data; in the writing of the
report; or in the decision to submit the article for publication. We
would like to thank the Royal College of Surgeons in Ireland for use