Inhibitors of potassium channels KV1.3 and IK-1 as immunosuppressants

Article abstract of DOI:10.1016/j.bmcl.2009.02.077

New structural classes of K_V1.3 and IK-1 ion channel blockers have been identified based on a virtual high throughput screening approach using a homology model of KcsA. These compounds display inhibitory effects on T-cell and/or keratinocyte pr

Full text of DOI:10.1016/j.bmcl.2009.02.077

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2299–2304
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Inhibitors of potassium channels KV1.3 and IK-1 as immunosuppressants
Stefano Pegoraro a, Martin Lang a, Tobias Dreker a, Jürgen Kraus a, Svetlana Hamm a, Cathal Meere a,
b
b
b
b,
*
Juliane Feurle a, Stefan Tasler a, , Sylvia Prütting , Zerrin Kuras , Violeta Visan , Stephan Grissmer
*
a 4SC AG, Am Klopferspitz 19a, 82152 Planegg-Martinsried, Germany
b Institute of Applied Physiology, University of Ulm, Albert-Einstein-Allee 11, 89081 Ulm, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
New structural classes of KV1.3 and IK-1 ion channel blockers have been identified based on a virtual high
throughput screening approach using a homology model of KcsA. These compounds display inhibitory
effects on T-cell and/or keratinocyte proliferation and immunosuppressant activity within a DTH animal
model.
Received 23 December 2008
Revised 19 February 2009
Accepted 20 February 2009
Available online 25 February 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Potassium channel
IK-1
KV1.3
Homology model
vHTS
Autoimmune diseases
In human T-cells the Ca2+-dependent potassium channel IK-1
(KCa3.1) and the voltage-gated potassium channel KV1.3 have been
shown to play a pivotal role during cell proliferation. They are cru-
cial mediators for the repolarization of the cell membrane upon
cell stimulation, maintaining an intracellular Ca2+ signal decisive
for the proliferation process. IK-1 and KV1.3 channels display dif-
ferent distributions among certain T-cell subsets, which addition-
ally varies depending on the activation state: Upon activation, IK-
1 is predominant in naïve and central memory T-cells and KV1.3
in effector memory T-cells (TEM cells).1 Blockade of these channels
is linked to inhibition of proliferation, which was proven using
small molecule specific inhibitors such as TRAM-342 for IK-1 and
PAP-13 for KV1.3 (Fig. 1).1 Consequently, both channels are consid-
ered as potential targets for modulating immune response, which
would be desirable for treating autoimmune diseases like multiple
sclerosis, rheumatoid arthritis, diabetes type 1 and the skin disease
psoriasis,1,4 in which IK-1 might also play a role in the proliferation
of keratinocytes (HaCaT cell line).5
tion has to be accepted as a price for higher throughput.6 Conse-
quently, a combination of a rational pre-selection of screening
compounds along with a patch clamp assay was expected to cir-
cumvent the throughput problems without sacrificing the high
quality of data points. For such a purpose, virtual high throughput
screening (vHTS) composed of a docking experiment using 4SCanÒ
technology7 was envisaged. Based on the data published for dis-
ease relevance of these potassium channels at that time,4,8 primary
focus was laid on the discovery of potent small molecule inhibitors
for KV1.3. Back then, a crystal structure of a mammalian potassium
channel was not available—this changed not before 20059—thus
initial modeling was based on a homology model for KV1.3 assem-
bled from the crystal structure of the bacterial KcsA channel.10 The
inner vestibule (i.e., the internal cavity just below the K+-selectivity
pore) was selected as binding pocket for potential inhibitors, as
this binding site was identified as such for TRAM-34 in IK-1 and
is assumed to be addressed by PAP-1 in KV1.3 as well.1,11 Within
this region, both KV1.3 and IK-1 display around 45% homology with
KcsA. The KV1.3 model generated was applied to the evaluation of a
data base of 3.3 Mio virtual and commercially available com-
pounds. Out of the list of best 500 compounds, a selection was
made combining a certain structural diversity with good docking
scores within the binding pocket. Current blocking activity on
KV1.3 was determined for 37 compounds using patch clamp tech-
nology,12,13 for which a hit rate of around 11% was found when
High throughput screening for ion channel modulators is ham-
pered by technical difficulties (even though decreasingly): Either
the screening method provides data of good quality and informa-
tion content but does not allow for a high throughput (patch clamp
assay), or a loss of reproducibility, quality or functional informa-
* Corresponding authors. Tel.: +49 89 7007630; fax: +49 89 700763 29 (S.T.); fax:
+49 731 5023260 (S.G.).
E-mail addresses: stefan.tasler@4sc.com (S. Tasler), stephan.grissmer@
uni-ulm.de (S. Grissmer).
defined as IC50(KV1.3) below 10 lM. A few structurally signifi-
cantly different intriguing compounds were identified, among
which dihydrophenanthridine 1 displayed already good inhibitory
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.077

2300
S. Pegoraro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2299–2304
activation of the imine structure in phenanthridine (3) with acyl
chlorides to give an intermediary imminium ion enabling in situ
nucleophilic attack at C6 with indoles or pyrroles as well as with
Grignard reagents (Scheme 1).14 An inverse strategy was realized
for the synthesis of derivative 6, for which the phenanthridine
(3) was first treated with a Grignard reagent, followed by N-acyla-
tion.15 Oxalylic acid derivative of entry 6 (Table 1) was attained by
saponification of the corresponding ester [entry 7; LiOH, MeOH/
H2O (3:1), 90 °C, 2 h, 48%].
Based on the promising results for KV1.3 blockade (Table 1, dis-
cussion below), further structural variations were scheduled,
including alterations of the bridging unit of the biaryl core towards
the second molecule depicted in Figure 1, dibenzocycloheptanedi-
one 2. Phenanthridinone derivatives were simply N-alkylated in
DMF starting with phenanthridin-6(5H)-one (Table 2, entry 19)
using the corresponding alkyl bromides (KOH, K2CO3, cat. TBAB,
110 °C, 1 h) (e.g., entry 20, exemplarily shown in Table 2). N-acyl
derivatives of phenanthridinone—as closer analogs for the series
depicted in Table 1—proved to be rather unstable (isomerization
and/or decomposition)16 and were consequently not suitable for
evaluation of biological effects. Extending the bridging unit from
a six-membered to a seven-membered ring required a multi-step
synthesis of 5H-dibenzo[b,d]azepin-7(6H)-ones 9 (Table 2, entries
21–25) (Scheme 2). The sequence involved an intramolecular
Friedel–Crafts acylation in 8 as crucial step.17
Due to the already decent level of activity on KV1.3, the screen-
ing concentration for phenanthridine derivatives within the patch
clamp assay was chosen to be 2
lM (Table 1), as compared to
20 M for the other de-novo scaffold variations in Table 2. Out of
l
the numerous derivatives synthesized for the former class of com-
pounds, a selection is given in Table 1. Starting with the initial hit
molecule 1, the N-acetyl unit was replaced by other aliphatic N-
acyl variants (entries 2–5), but a clear SAR was not deducible. A
cyclobutyl residue at the N-carbonyl unit (entry 4) gave rise to a
comparable IC50 value as observed for the methyl residue in the
parent compound 1, 2-propyl was worse, 2-pentyl better (entries
2 and 3 vs entry 1). From entries 5 and 8 it can be concluded, that
further aliphatic extension for an attachment of a cyclic residue
(cyclopentyl or phenyl, respectively) is not tolerated.
Oxalylic derivatives (entries 6 and 7) resulted in quite decent
activities on KV1.3 with IC50 values around 1–3 lM. Direct aro-
matic substitution at the N-carbonyl linker bears some potential
O
O
i,
Cl
R
N
N
R
R'
R'
ii,
X
3
R''
N
R''
H
4
or iii,
BrMg
iv
S
Figure 1. Reference compounds for KV1.3 and IK-1 inhibition, hit molecule 1 and
analog 2 identified based on vHTS; compound 1 is shown within the inner vestibule
of the KV1.3 homology model (closed state of the channel); avalues determined in-
house.
O
O
v
NH
N
O
O
O
O
potential on KV1.3 (Fig. 1). In addition, another bridged biaryl, the di-
benzocycloheptanedione derivative 2, showed some moderate
activity as well, already indicating a possible direction for struc-
tural variations around the hit molecule 1.
A quick approach for substitutional variations at N5 and C6 of
the 5,6-dihydrophenanthridine scaffold was established, including
5
6
Scheme 1. Reagents and conditions: (i) toluene or DMF (THF, if second step is
Grignard reaction), rt, 1 h; (ii) NEt3 or DIEA, rt, 20 h; 5–70%; (iii) rt, 20 h; 29–42%;
(iv) benzo[d][1,3]dioxol-5-ylmagnesium bromide, THF, 65 °C, 7 h, 50%; (v) 2-furoyl
chloride, THF, NEt3, rt, 2 h, 19%; reactions performed in inert atmosphere.

S. Pegoraro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2299–2304
2301
Table 1
KV1.3 activities for selected phenanthridine derivatives
Entry
1a
R
IC50(KV1.3) [
lM] (inhib. @ 2 lM)
O
5.0
O
2
7.4
O
O
3
4
1.9
4.2
O
O
5
— (11%)
O
O
N
O
6
7
X = H
— (41%)
1.4
X
Me
O
R
O
8
— (15%)
O
9
10
11
X = 2-OMe
— (51%)
— (30%)
— (13%)
3-OMe
4-OMe
X
O
O
12
13
X = S
— (43%)
1.5
X
N
O
14
15
0.71
N
O
O
— (15%)
HN
N
R
O
O
16
17
— (8%)
O
— (59%)
R
N
O
F
18
— (27%)
S
F
a
Commercial sources.
for further optimization of KV1.3 blocker capability: incorporation
of a 2-anisolyl as well as 2-thienyl or 2-furanyl resulted in IC50 val-
ther optimizations. A substitution of the 3-indolyl moiety at C6 of
the dihydrophenanthridine by 2-pyrrolyl or benzo[d][1,3]dioxol-5-
yl was detrimental to activity (entries 15 and 16 vs entry 13),
whereas the 2-thienyl group at C6 resulted in virtually identical
activity for the N-(2-methylpentanoyl) derivative with an IC50
ues between 1.5 and 3
lM (entries 9, 12 and 13; inhibition of 43%
at 2 M should result in an IC50 around 3
l
l
M)—comparable to the
2-pentyl derivative of entry 3—but now offering the possibility for
amelioration of activity by tuning the heterocyclic group or further
aromatic substituents. Keeping a five-membered heterocycle in
place, the dimethylpyrazole derivative gave rise to a sub-micromo-
lar activity (entry 14) and thus represents an excellent lead for fur-
somewhat below 2
Variations of the bridging unit started with the elimination of
the aromatic C6-substituent, now utilizing phenanthridin-
6(5H)-one scaffold. Besides the parent scaffold, one derivative is
lM (entry 17 vs entry 3).
a

2302
S. Pegoraro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2299–2304
Table 2
With proliferation of T-cells being mediated by either KV1.3 or
KV1.3 activities for differently bridged biaryl derivatives
IK-1 depending on the T-cell subtype (vide supra) and a sequence
homology for the inner vestibule of the two channels of around
55%, selected derivatives out of Tables 1 and 2 were cross-evalu-
ated on the latter ion channel (patch clamp assay).18 As can be seen
in Table 3, phenanthridines and phenanthridinones displayed vir-
tually no activity on IK-1 (entries 3, 13, 14 and 20), whereas com-
pounds of the 5H-dibenzo[b,d]azepin-7(6H)-one series (entries 21,
23 and 25–27) possessed some potency on IK-1. Independent from
the nature of the substituent at the sulfon moiety (entries 21, 23
Entry
19a
R
IC50(KV1.3) [
l
M)
M]
(inhib. @ 2
l
H
— (12%)
R
O
N
20
— (57%)
N
21a
22
23
24
X = 4-Me
4-NO2
4-Cl
5.8
>10
6.1
3.6
and 25), IC50(IK-1) values were identified around 2
sponding carboxamide derivatives, the 4-Me variant lost signifi-
cantly inhibitory potency at IK-1 (entry 26 vs entry 21, 16 M vs
2.1 M), the 3-chlorinated variant, however, displayed good activ-
ity on this potassium channel comparable to the 4-chlorophenyl-
sulfonamide of entry 23, IC50 values were determined to be
lM. For corre-
O
O
X
S
l
3-Cl
l
R
N
O
S
O
around 1.3 lM.
S
25
13
Upon identification of effective potassium channel blockers—
entry 14 for KV1.3, entries 21, 23, 25 and 27 for IK-1—their effect
on the proliferation of the TEM cell fraction of PBMCs was investi-
gated (Table 4). For this purpose, PBMCs were stimulated using
anti-CD3 antibody,19 upon which neither a potent KV1.3 blocker
(PAP-1) nor an IK-1 blocker (TRAM-34) alone were able to inhibit
O
O
X
26
27
X = 4-Me
3-Cl
4.0
3.4
a
proliferation significantly up to a concentration of above 20
lM.
Commercial sources.
With a background of an IK-1 blocker supplied at 5 M, dose–re-
l
sponse curves were attained for the KV1.3 blocker in question, indi-
cating that proliferation of TEM cells according to this protocol is
mediated by two components, an IK-1 and a KV1.3 based mecha-
nism. Within this setup, PAP-1 was determined to have an EC50
of 430 nM (TRAM-34 background), with a significant deviation in
dependence on the PBMC donor ( 350 nM). For the phenanthridine
Ar
X = OEt
Y
iii
iv
i, ii
NH2
N
of entry 14, an EC50 of 12.5 lM was identified (TRAM-34 back-
X = OH
X = Cl
X
ground), which represents a slightly improved factor towards the
EC50 of PAP-1 (EC50 ratio ꢀ30) as compared to the factor present
for activity at the KV1.3 (factor ꢀ70). However, the activity of this
compound was not significantly altered by the presence of an IK-1
O
7
8
v
X = Cl
blocker: EC50 = 15 vs 12.5
(12% current blocking at 20
l
M. As a virtual inactivity at the IK-1
M) cannot account for this finding,
l
influences of this compound on calcium signaling in anti-CD3 stim-
ulated PBMCs was investigated.20 These experiments revealed a
dose dependent reduction of the thapsigargin-induced increase
in the intracellular calcium concentration unlike that of potassium
channel blockers tested (PAP-1, TRAM-34), thus indicating an addi-
tional interference with the calcium signaling, ultimately adding
another potential effect on T-cell proliferation.
Y
Ar
Y = SO2, CO
N
O
9
Scheme 2. Reagents and conditions: Variant A: (i) ArSO2Cl, pyridine, 140 °C,
30 min, quant.; (ii) ethyl bromoacetate, Na2CO3, toluene, 110 °C, 4 h, 15–30%; (iii)
NaOH, MeOH/H2O (3:1), 30 °C, 4 h, quant.; (iv) SOCl2, 75 °C, 1 h; (v) AlCl3, CHCl3,
À78 °C to rt, 2 h, 5–50% over two steps. Variant B: (i) ethyl bromoacetate, Na2CO3,
toluene, 110 °C, 4 h, 14%; (ii) ArCOCl, pyridine, 125 °C, 3 h; (iii) NaOH, MeOH/H2O
(3:1), 70 °C, 4 h; (iv) SOCl2, rt, 2 h; (v) AlCl3, CHCl3, À78 °C to rt, 20 h, 16–25% over
four steps.
An identical concept as pursued for KV1.3 blockers was realized
for the evaluation of entries 21 and 23 as IK-1 inhibitors. Thus,
Table 3
Activities on IK-1 and HaCaT
Compound entry IC50(KV1.3) [
l
M]
IC50(lK-1) [
l
M]
EC50(HaCaT) [
lM]
shown exemplarily in Table 2 (entry 20), which reflects the general
loss of activity on KV1.3 (best value: 57% current blocking @
20 lM). Extension of the bridging unit by an additional methylene
3
1.9
No effect
No effect
12% inhib. at 20
No effect
2.1
1.4
14
13
1.5
17% inhib. at 25 lM
nd
14
0.71
lM
group was evaluated initially for compound entry 21, in which the
carbonyl attachment to the endocyclic nitrogen as present in deriv-
atives of Table 1 was simultaneously replaced by sulfonyl simply
based on commercial availability of such a compound. As a moder-
ate activity on the KV1.3 channel was identified, further derivatives
were synthesized either as sulfonamides (entries 22–25) or now
also as carboxamides (entries 26 and 27) (only selected derivatives
shown in Table 2). Best IC50 values were determined between 3.4
20
57% inhib. at 20
lM
2.3
8.4
n.d.
n.d.
n.d.
n.d.
n.d.
40
21
5.8
23
6.1
25
13
2.4
16
26
4.0
27
3.4
1.3
PAP-1
TRAM-34
Clotrimazole
0.01
5.0b
6.0b
10a
0.02b
0.07b
15
and 6.1 lM with carboxamides and sulfonamides displaying virtu-
ally identical inhibition (entries 21 and 24 vs entries 26 and 27,
respectively).
n.d. = not determined.
a
Data taken from Ref. 26.
b
Data taken from Ref. 2.

S. Pegoraro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2299–2304
2303
Table 4
Inhibition of anti-CD3 stimulated PBMC proliferation; values are means of n P 2
individual experiments
KV1.3 blocker
lK-1 blocker
EC50 of
EC50 [lM]
—
—
—
TRAM-34
Entry 21
Entry 23
—
TRAM-34 [5
Entry 21 [5
Entry 23 [5
Entry 23 [16.7
—
TRAM-34
Entry 21
Entry 23
PAP-1
PAP-1
PAP-1
No effect
No effect
No effect
ꢀ25
PAP-1
PAP-1
PAP-1
PAP-1
PAP-1
Entry 14
Entry 14
PAP-1 [5
PAP-1 [5
PAP-1 [5
l
l
l
M]
M]
M]
0.43
0.4
1.2
0.5
PAP-1
PAP-1
lM]
Entry 14
Entry 14
TRAM-34
Entry 21
Entry 23
ꢀ15
TRAM-34 [5
TRAM-34
Entry 21
lM]
12.5
lM]
lM]
lM]
0.025
2.5
8.0
Entry 23
Figure 2. Relative inhibition of ear swelling in a DTH mouse model (oxazolone
sensitization/challenge) for clotrimazole and compound of entry 21 (each 1% in
ethanol–DMSO 4:1) in comparison with ProtopicÒ (0.1%); values are means of n = 5
animals per group.
dose–response curves for these compounds were measured with a
M background of PAP-1 present, which proved their activity as
IK-1 blockers. Their EC50 values of 2.5 and 8.0 M were found to be
5
l
l
within the same range compared to TRAM-34 as were IC50 patch
clamp data for these compounds (factor 80–320). A final prove of
potent IK-1 blockade was achieved by substituting the TRAM-34
background for either one of the compounds of entry 21 and 23
in PAP-1 EC50 determinations. With a background of IK-1 blocker
chosen appropriately high (concn > EC50 of the respective com-
pound with 5 lM PAP-1 background), PAP-1 activity was identical
for all compounds, EC50s(PAP-1) were determined between 400
and 500 nM (same PBMC donor).
As mentioned above, with regard to targeting psoriasis, IK-1
was discussed to be involved in the proliferation process of kerat-
inocytes,5 so that inhibition of this potassium channel was pro-
posed to have a potential for the inhibition of proliferation of
human HaCaT cells. However, no correlation between inhibition
of HaCaT proliferation21 and IK-1 activity can be deduced from
the data shown (Table 3). The best EC50 value on these cells was at-
tained for the pyrrolylethyl phenanthridinone derivative of entry
and eventually proliferation is not dependent on a functional
KV1.3 unit.1 On the other hand, IK-1 mRNA expression was iden-
tified in CD4+ T lymphocytes in wild-type mice and its essential
role within these cells was proven by IK-1 inhibition using clo-
trimazole and also in Kcnn4 knock-out mice.22,23 Therefore, an
anti-inflammatory effect possibly detectable in this mouse model
working via inhibition of T-cell proliferation might be mediated
by a dominant IK-1 effect of the compounds. Both compounds
were administered once topically in ethanol–DMSO as vehicle
(compound concentration of 1%) 30 min after challenge (oxazo-
lone) and resulted in comparable reductions of ear swelling of
ca. 20 percentage points towards the vehicle treated group
(Fig. 2). ProtopicÒ (0.1%) was used as positive control. Its active
ingredient interferes with Ca2+ signaling in T-cells by binding to
calcineurin. As 5H-dibenzo[b,d]azepin-7(6H)-one of entry 21
showed an analogous, even though slightly weaker decrease of
intracellular calcium levels in thapsigargin-treated T-cells20 as ob-
served for phenanthridine of entry 14 (vide supra) and unlike that
of a potent IK-1 blocker, an additional effect on calcium signaling
might explain the similar activity in this DTH model as compared
to clotrimazole besides its weaker inhibition of IK-1.
Based on a vHTS approach, two structurally new classes of
potassium channel inhibitors have been identified as new leads
for further optimization processes.24 The phenanthridine of entry
14 displayed an IC50 of 710 nM for KV1.3 inhibition. Besides several
peptidic toxins, only a few small molecules have been reported as
sub-micromolar inhibitors of the KV1.31,25—that is mainly PAP-1
and its derivatives,3,26 PAC derivatives27 and khellinones19—so that
this scaffold represents a promising lead for a new effective class of
such inhibitors, being non-cytotoxic and not displaying any issues
on the hERG channel. Likewise intriguing are the IK-1 blockers of
the 5H-dibenzo[b,d]azepin-7(6H)-one type, entries 21, 23 and 27,
20 with 2.3 lM, which displayed no activity on IK-1. Vice versa,
the potent IK-1 inhibitors TRAM-34 and clotrimazole displayed
only a weak effect on HaCaT cells with EC50 values of 40 and
15
l
M, respectively.
A general cytotoxic effect in either assay, inhibition of PBMC or
HaCaT proliferation, could be excluded for selected compounds:
cell viability was determined on either HaCaT cells or PBMC for
compounds of entry 14 and 21 in comparison with clotrimazole.
Whereas clotrimazole displayed an LD50 of ꢀ40
effect was measurable for both compounds mentioned up to
50 M (highest assay concentration). Furthermore, a caspase 3/7
lM, no cytotoxic
l
assay was performed for compounds of entries 7, 20 and 21 in or-
der to check for apoptosis induction, which could be excluded in a
concentration range up to 30 lM for all three compounds. Finally,
clotrimazole, TRAM-34 and the compound of entry 21 were tested
for proliferation inhibition on HEK-293, the proliferation of which
should be IK-1 independent, only to identify the EC50 values to be
rather similar. Thus, inhibition of HaCaT proliferation should be
neither based on cytotoxic effects and apoptosis nor on IK-1 inhi-
bition—with the latter being in good agreement with the data col-
lection presented (Table 3).
with IC50 values of 2.1, 1.4 and 1.3 lM, respectively. The former
compound (entry 21) displayed anti-inflammatory potential in a
mouse DTH model comparable to clotrimazole. These potassium
channel inhibitors were effective inhibitors of TEM cell proliferation
in an appropriate assay setup.
5H-Dibenzo[b,d]azepin-7(6H)-one of entry 21 was next se-
lected for a comparison study with clotrimazole for the anti-
inflammatory potential in a mouse DTH model (delayed-type
hypersensitivity; ear swelling). Generally, mouse TEM cells do
not up-regulate KV1.3 expression upon activation and—besides
KV1.3 and IK-1—express further members of the KV1.x family in
T-cells. Consequently adjustment of their membrane potential
Acknowledgements
The authors want to thank Dr. Wael Saeb for synthetic assis-
tance and Drs Andrea Aschenbrenner, Andreas Wuzik, Udo Sinks
and Daniel Vitt for scientific input. Funding by the BMBF
(BioChance Plus) was highly appreciated.

2304
S. Pegoraro et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2299–2304
amplitudes before and after compound application, with 1 mM 1-EBIO and
200 nM clotrimazole as controls for current activation and inhibition,
respectively. Intracellular Ca2+ concentration was adjusted to 1
M. Whole
cell patch clamp recording was performed with n P 2 individual experiments
at each compound concentration using different cells. Four or more different
concentrations were determined per dose–response curve.
References and notes
l
1. Wulff, H.; Zhorov, B. S. Chem. Rev. 2008, 108, 1744.
2. Wulff, H.; Miller, M. J.; Hänsel, W.; Grissmer, S.; Cahalan, M. D.; Chandy, K. G.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8151.
3. Vennekamp, J.; Wulff, H.; Beeton, C.; Calabresi, P. A.; Grissmer, S.; Hänsel, W.;
Chandy, K. G. Mol. Pharmacol. 2004, 65, 1364.
4. (a) Jensen, B. S.; Strobaek, D.; Olesen, S.-P.; Christophersen, P. Curr. Drug Targets
2001, 2, 401; (b) Jensen, B. S.; Hertz, M.; Christophersen, P.; Madsen, L. S. Exp.
Opin. Ther. Targets 2002, 6, 623.
19. (a) Baell, J. B.; Gable, R. W.; Harvey, A. J.; Toovey, N.; Herzog, T.; Hänsel, W.;
Wulff, H. J. Med. Chem. 2004, 47, 2326; (b) Harvey, A. J.; Baell, J. B.; Toovey, N.;
Homerick, D.; Wulff, H. J. Med. Chem. 2006, 49, 1433; (c) Cianci, J.; Baell, J. B.;
Flynn, B. L.; Gable, R. W.; Mould, J. A.; Paul, D.; Harvey, A. J. Bioorg. Med. Chem.
Lett. 2008, 18, 2055. A general PBMC assay setup is described in (a), the readout
was performed using BrdU Cell proliferation ELISA kit (Roche) according to
manufacturer’s instructions.
20. Thapsigargin-induced calcium increase: Fresh isolated PBMCs were stimulated
for 48 h with 50 ng/mL anti-CD3 antibody (NatuTec). Non-adherent cells were
harvested, suspended in 107 cells/mL concentration in culture medium (RPMI-
5. Koegel, H.; Alzheimer, C. FASEB J. 2001, 15, 145.
6. (a) Dabrowski, M. A.; Dekermendjian, K.; Lund, P.-E.; Krupp, J. J.; Sinclair, J.;
Larsson, O. CNS Neurol. Disord. Drug Targets 2008, 7, 122; (b) Southan, A.; James,
I. F.; Cronk, D. Drug Discovery World 2005, 6, 17; (c) Xu, J.; Wang, X.; Ensign, B.;
Li, M.; Wu, L.; Guia, A.; Xu, J. Drug Discovery Today 2001, 6, 1278.
7. Seifert, M. H. J.; Wolf, K.; Vitt, D. Biosilico 2003, 1, 143.
8. (a) Beeton, C.; Wulff, H.; Barbaria, J.; Clot-Faybesse, O.; Pennington, M.;
Bernard, D.; Cahalan, M. D.; Chandy, K. G.; Beraud, E. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 13942; (b) Wulff, H.; Calabresi, P. A.; Allie, R.; Yun, S.; Pennington, M.;
Beeton, C.; Chandy, K. G. J. Clin. Invest. 2003, 111, 1703.
9. Long, S. B.; Campbell, E. B.; MacKinnon, R. Science 2005, 309, 897.
10. Zhou, Y.; Morais-Cabral, J. H.; Kaufman, A.; MacKinnon, R. Nature 2001, 414, 43.
11. Wulff, H.; Gutman, G. A.; Cahalan, M. D.; Chandy, K. G. J. Biol. Chem. 2001, 276,
32040.
12. For a general description of patch clamp methodology, see: Hamill, O. P.;
Marty, A.; Neher, E.; Sakmann, B.; Sigworth, F. J. Pflügers Arch. 1981, 391, 85.
13. For a description of a KV1.3 patch clamp assay, see: Grissmer, S.; Nguyen, A. N.;
Aiyar, J.; Hanson, D. C.; Mather, R. J.; Gutman, G. A.; Karmilowicz, M. J.; Auperin,
D. D.; Chandy, K. G. Mol. Pharmacol. 1994, 45, 1227; whole cell patch clamp
recording was performed with n P 2 individual experiments at each
compound concentration using different cells. Four or more different
concentrations were determined per dose–response curve.
1640, 10% FCS) and incubated with 3 lM Fluo3AM (Fischer Scientific), 0.1%
Pluronic F-127 (Sigma–Aldrich) and anti-CD4-APC-antibody (BD Bioscience)
for 30 min at rt. Cells were then washed twice with PBS, adjusted in culture
medium to a final concentration of 5 Â 105 cell/mL and incubated with the
compounds or the corresponding amount of vehicle (DMSO) for 30 min.
Intracellular calcium increase was induced with 25 lM thapsigargin (Sigma–
Aldrich) and was measured using FACSCalibur (BD Bioscience). For the
importance of calcium signaling in T-cell proliferation, see: (a) Vig, M.; Kinet,
J.-P. Nat. Immunol. 2009, 10, 21; (b) Lewis, R. S. Annu. Rev. Immunol. 2001, 19,
497.
21. HaCaT keratinocytes were seeded in KBM/10% FCS and incubated for 24 h at
37 °C. Compounds were diluted in KBM/FCS with a final concn of 1% DMSO,
added to the HaCaTs in triplicate and incubated for 48 h. For a readout on cell
numbers, the Cell Titer Viability Assay from Promega was used.
22. Begenisich, T.; Nakamoto, T.; Ovitt, C. E.; Nehrke, K.; Brugnara, C.; Alper, S. L.;
Melvin, J. E. J. Biol. Chem. 2004, 279, 47681.
23. Lang, F. J. Am. Coll. Nutr. 2007, 26, 613S.
14. von Dobeneck, H.; Goltzsche, W. Chem. Ber. 1962, 95, 1484.
15. (a) Gilman, H.; Eisch, J.; Soddy, T. J. Am. Chem. Soc. 1957, 79, 1245; (b) Adam, W.;
Groer, P.; Mielke, K.; Saha-Möller, C. R.; Hutterer, R.; Kiefer, W.; Nagel, V.;
Schneider, F. W.; Ballmaier, D.; Schleger, Y.; Epe, B. Photochem. Photobiol. 1997,
66, 26.
16. (a) Curtin, D. Y.; Engelmann, J. H. J. Org. Chem. 1972, 37, 3439; (b) Leenders, R.
G. G.; Scheeren, H. W. Tetrahedron Lett. 2000, 41, 9173.
17. Paterson, W.; Proctor, G. R. J. Chem. Soc. 1962, 3468.
24. Pegoraro, S.; Lang, M.; Feurle, J.; Kraus, J. U.S. Patent 7,276,606, 2005.
25. Chandy, K. G.; Wulff, H.; Beeton, C.; Pennington, M.; Gutman, G. A.; Cahalan, M.
D. Trends Pharmacol. Sci. 2004, 25, 280.
26. Schmitz, A.; Sankaranarayanan, A.; Azam, P.; Schmidt-Lassen, K.; Homerick, D.;
Hänsel, W.; Wulff, H. Mol. Pharmacol. 2005, 68, 1254.
27. Schmalhofer, W. A.; Bao, J.; McManus, O. B.; Green, B.; Matyskiela, M.;
Wunderler, D.; Bugianesi, R. M.; Felix, J. P.; Hanner, M.; Linde-Arias, A.-R.;
Ponte, C. G.; Velasco, L.; Koo, G.; Staruch, M. J.; Miao, S.; Parsons, W. H.;
Rupprecht, K.; Slaughter, R. S.; Kaczorowski, G. J.; Garcia, M. L. Biochemistry 2002,
41, 7781.
18. Compound effects on currents through IK-1 channels were tested in whole-cell
patch clamp experiments according to Ref. 11 by comparison of the current