Synthesis of Small-Molecule Fluorescent Probes for the In Vitro Imaging of Calcium-Activated Potassium Channel KCa3.1

Article abstract of DOI:10.1002/anie.202001201

Small-molecule probes for the in vitro imaging of K_Ca3.1 channel-expressing cells were developed. Senicapoc, showing high affinity and selectivity for the K_Ca3.1 channels, was chosen as the targeting component. BODIPY dyes 15–20 were

Full text of DOI:10.1002/anie.202001201

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Synthesis of novel small-molecule fluorescently labeled probes
for the in vitro imaging of KCa3.1 channels
Authors: Bernhard Wünsch, Kathrin Brömmel, Ivan Maisuls, Christian
Paul Konken, Marius Rieke, Zoltan Pethő, Cristian A.
Strassert, Sarah Maskri, Oliver Koch, and Albrecht Schwab
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202001201
Angew. Chem. 10.1002/ange.202001201
Link to VoR: http://dx.doi.org/10.1002/anie.202001201
http://dx.doi.org/10.1002/ange.202001201

10.1002/anie.202001201
Angewandte Chemie International Edition
1
Synthesis of novel small-molecule fluorescently labeled probes for the in vitro
imaging of KCa3.1 channels
Kathrin Brömmel,a,e Sarah Maskri,a Ivan Maisuls,b Christian Paul Konken,c Marius
Rieke,d Zoltan Pethő,d Cristian A. Strassert,b,e Oliver Koch,a Albrecht Schwab,d
Bernhard Wünsch* a,e
a Institute for Pharmaceutical and Medicinal Chemistry, Westphalian Wilhelms-
University Münster, Corrensstraße 48, D-48149 Münster, Germany
Tel.: +49-251-8333311; Fax: +49-251-8332144; E-mail: wuensch@uni-muenster.de
b Center for Nanotechnology, Center for Soft Nanoscience, Institute for Inorganic and
Analytical Chemistry, Westphalian Wilhelms-University Münster, Heisenbergstraße
11, D-48149 Münster, Germany
c Department of Nuclear Medicine, University Hospital Münster, Albert-Schweitzer-
Campus 1, Building A1, D-48149 Münster, Germany
d Institute for Physiology II, University Hospital Münster, Robert-Koch-Straße 27b, D-
48149 Münster, Germany
e Cells-in-Motion Interfaculty Center, Westphalian Wilhelms-University Münster,
Waldeyerstraße 15, D-84149 Münster, Germany
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
2
Abstract
In order to facilitate the in vitro visualization of KCa3.1 channel-expressing cells, novel
small-molecule imaging probes were designed and developed. Senicapoc showing
high affinity and excellent selectivity towards the KCa3.1 channels was selected as
targeting component. Different BODIPY dyes (15 - 20) were synthesized and
connected by a Cu-catalyzed azide alkyne [3+2]cycloaddition with propargyl ether
derivative 8 of senicapoc yielding fluorescently labeled ligands 21 - 26 targeting KCa3.1
channels. The novel dimethylpyrrole-based imaging probes 25 and 26 allow staining
of KCa3.1 ion channels in NSCLC cells following a simple, fast and efficient protocol.
The specificity was shown by removing the punctate staining pattern by pre-incubation
with senicapoc. The density of KCa3.1 channels detected with fluorescent probe 25 and
by immunostaining was identical. The punctate structure of the labeled channels could
be observed in living cells as well. Molecular modeling studies showed binding of the
senicapoc targeting component towards the binding site within the ion channel and
orientation of the linker with the dye along the inner surface of the ion channel.
Keywords
KCa3.1 channel; non-small cell lung cancer cells; labeled ligands; cycloaddition;
fluorescent probes; imaging agents; molecular modelling
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
3
1. Introduction
Ion channels play a critical role in the progression of cancer and contribute to features
of essentially all “cancer hallmarks”.[1] The calcium activated potassium channel 3.1
(KCa3.1) is an intensively studied ion channel in this context. It is involved in critical
steps of the metastatic cascade, such as proliferation, migration, invasion and
extravasation.[2],[3] Inhibition of the KCa3.1 channel led in many different tumor entities
to reduced proliferation, migration and metastasis.[4],[5],[6] Overexpression of this
channel directly correlates with tumor grade and metastatic status and is often related
to poor prognosis for tumor patients and high lethality rates.[7] Moreover, expression of
KCa3.1 channels is dysregulated in many tumor entities. When patients are stratified
according to KCa3.1 channel expression, patient survival is worse for those with
elevated KCa3.1 expression. This points to the predictive power of analyzing KCa3.1
channel expression with respect to patient survival and has predictive power with
respect to patient survival.[8],[9]
Therefore, developing tools for the visualization of KCa3.1 expressing cells would be a
great asset for a better understanding its (diagnostic) role and as a predictive value in
cancer. Currently, only fluorescently labeled antibodies are available for optical
imaging of the KCa3.1 channel, making the process time-consuming and expensive.
Herein, we present the development and synthesis of novel small-molecule probes,
which show first promising results in in vitro imaging of the KCa3.1 expressing cells via
fluorescence microscopy. Simple, fast and efficient staining protocols are the
advantages of these novel imaging probes, which can be instrumental for elucidating
the mechanisms by which KCa3.1 channels contribute to cancer progression and
metastasis.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
4
2. Synthesis
For the development of small-molecule fluorescently labeled probes of the KCa3.1
channel, senicapoc (1) served as starting point. Senicapoc (1) has a high affinity
towards the KCa3.1 channel in the low nanomolar range (IC50 = 11 ± 2 nM; measured
on human erythrocytes) and excellent selectivity over related ion channels.[10] In order
to introduce a fluorescent dye one F-atom of senicapoc should be replaced by a
hydroxy moiety. The OH moiety of the senicapoc derivative 2 should allow broad
chemical modifications including the introduction of a fluorescent dye. (Figure 1)
F
OH
O
O
NH2
NH2
F
1
F
2
(Senicapoc)
Figure 1: Chemical structure of senicapoc (1) and its derivative 2
The phenol 2 was obtained in five steps starting with the protection of 4-bromophenol
(3) with tert-butyldiphenylsilyl chloride (TBDPS-Cl). The protected 4-bromophenol 4
was reacted with n-butyllithium in THF to obtain the lithiated intermediate, which was
trapped with 4-fluorobenzophenone yielding the tertiary alcohol 5. Trimethylsilyl
cyanide and catalytic amounts of InCl3 were used to convert the alcohol 5 into the nitrile
6. The silyl protective group was cleaved under basic conditions resulting in the
intermediate 7 with a free hydroxy group in p-position. In the last step, the nitrile was
hydrolyzed partially under basic conditions to obtain the senicapoc derivative 2.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
5
Although the partial hydrolysis of the nitrile to an amide has been described for the
synthesis of senicapoc in high yields, in case of the phenol 7 the transformation was
rather low and incomplete resulting in a yield of the amide 2 of 23 %. However, a large
amount of the starting material 7 could be re-isolated, which increased the efficiency
of the last reaction step considerably. (Scheme 1)
Scheme 1: Synthesis of the senicapoc derivative 2. Reagents and reaction
conditions: (a) imidazole, TBDPS-Cl, DMF, 24 h, rt, 96 %. (b) 1. n- butyllithium, THF,
1h, -78 °C; 2. 4-fluorobenzophenone, THF, 3h, rt, 58 %. (c) 1. InCl3, trimethylsilyl
cyanide, CH2Cl2, 1 h, 20 – 40 °C, 1.5 h, 40 - 50 °C; 2. KOH, KNa-tartrate, H2O, 12 h,
0 °C – rt, 89 %. (d) KOH, KNa-tartrate, 2-methylbutan-2-ol ,5 h, 100 °C, 88 %. (e)
KOH, EtOH, H2O, 90 °C, 48 h, 23 %.
For the introduction of the fluorescent label a Cu(I)-catalyzed azide-alkyne
[3+2]cycloaddition was envisaged. Therefore, an alkyne moiety had to be introduced
into the phenol 2. For this purpose, phenol 2 was alkylated with propargyl bromide in
the presence of Cs2CO3 to afford the propargyl ether 8. (Scheme 1)
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
6
In order to visualize the senicapoc derived targeting component by introduction of a
fluorescent dye, different boron dipyrromethene (BODIPY) dyes were designed
bearing an azide moiety suitable for the planned [3+2]cycloaddition reaction with the
alkyne 8.
The photochemical properties of the BODIPY dyes are easily adjustable by variation
of the substitution pattern of the dipyrromethene core.[11],[12] In terms of medicinal
research, the BODIPY dyes show some important advantages: The emission intensity
derived from high quantum yields is almost independent from the pH value and the
polarity of the solvent. Moreover, low toxicity was reported. On the other hand, low
solubility of some dyes in water and physiological water-based buffer systems may
cause problems concerning their application in biological systems.
R1
R1
(a)
(b)
R
R
R
R
H
O
NH HN
R1
NH
HN
R1
R
R
R
(c)
R
N
N
NH
N
B
F F
Scheme 2: General concept for the synthesis of BODIPY dyes. (a) H+. (b) oxidant,
e.g. p-chloranil. (c) NEt3, BF3 OEt2. (modified according to literature[13])
.
For the synthesis of BODIPY based dyes a general one-pot three-step approach was
pursued, which includes the acid catalyzed condensation of two equivalents of a
pyrrole with one equivalent of a benzaldehyde followed by oxidation of the
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
7
dipyrrolylmethane with 2,3,5,6-tetrachloro-1,4-benzoquinone (p-chloranil). Then, the
oxidized product was deprotonated with a base and complexed with BF3.OEt2 to give
the BODIPY dye.[13] (Scheme 2)
Due to the limited conjugation with the dipyrrolomethyne core, the phenyl moiety has
only minor effects on the excitation and emission wavelength of the resulting BODIPY
dye. Herein, benzaldehyde derivatives were used bearing an azide moiety within the
substituent in p-position. This azido moiety should be used for the connection of the
BODIPY dye with the targeting component via a Cu(I)-catalyzed azide-alkyne
[3+2]cycloaddtion. Consequently, the benzaldehyde derivatives contain the linker to
the targeting component. Benzaldehyde derivatives with different lengths and polarity
of the linker, i.e. an ethoxy- (9), a 3,6,9-trioxaundecyloxy (10) and a dodecyloxy linker
(11) between the phenyl ring and the azido moiety were synthesized as described in
literature [13],[14],[15] and further used for the synthesis of the BODIPY dyes.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
8
A
N3
O
N3
O
O
N3
O
3
11
S
N
H
N
N
H
H
H
O
9
H
O
H
O
10
11
12
13
14
N3
B
C
N3
O
B
O
B
(a)
(a)
(a)
(a)
12 + 9
14 + 9
14 + 10
13 + 9
S
S
N
N
N
N
F
F
F F
15
16
O
N3
O
N3
3
D
O
(a)
13 + 10
N
N
N
N
B
B
F
F
F F
19
17
O
N3
O
N3
O
3
11
(a)
13 + 11
N
N
N
N
B
B
F
F
F
F
20
18
Scheme 3: A: Structures of benzaldehyde derivatives 9-11 and pyrrole derivatives 12-
14 used for the BODIPY synthesis. B-D: Synthesis of BODIPY dyes with azide moiety
in the side chain. Series B is based on the thiophene annulated pyrrole 12, series C is
based on the ethyldimethylpyrrole 13 and series D on the dimethylpyrrole 14. Reagents
and reaction conditions: (a) 1. TFA, CH2Cl2, molecular sieves 3Å, rt, 24 h; 2. p-chloranil,
CH2Cl2, rt, 30 min; 3. Net3, CH2Cl2, rt, 15 min; 4. BF3ꞏOEt2, CH2Cl2, rt, 12 h. Yields: 15:
7 %. 16: 14 %. 17: 30 %. 18: 22 %. 19: 16 %. 20: 27 %. Yields were calculated over
three reaction steps for >90 % purity compounds.
The dipyrromethene core, which defines the optical properties of the dye, can be varied
by starting from different pyrrole derivatives. An enlarged conjugated π-electron
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
9
system resulting from annulation of an aromatic ring at the pyrrole ring shifts the
emission wavelength towards the near-infrared region. To achieve this effect, the
annulated thienopyrrole 12 was reacted with benzaldehyde 9 according to the general
one-pot procedure outlined in Scheme 3 to obtain BODIPY dye 15. Due to stability
issues with the BODIPY 15 containing the thienopyrrole system, alkyl substituents
were taken into consideration. In general, most of the reactive pyrrole positions should
be blocked by suitable substituents to avoid side reactions during the condensation
with benzaldehydes and to enhance the stability of the final dye. Therefore, 3-ethyl-
2,4-dimethylpyrrole (13) was reacted with the benzaldehydes 9 - 11 to provide the
stable and storable BODIPY dyes 16 - 18. In order to reduce the size and lipophilicity
of the dyes, the reaction of 2,4-dimethylpyrrole (14) with benzaldehydes 9 and 10 was
carried out to afford two additional stable and storable BODIPY dyes 19 and 20.
Since fluorescence microscopy should be performed with the designed fluorescently
labeled ligands, the synthesized dyes need to be suitable for this imaging technique.
Due to lack of conjugation with the dipyrrolomethyne core, electronic effects were not
expected regarding the photophysical properties in relation to the conjugation of the
phenyl moiety, its substituents or the targeting component. Therefore, the optical
characteristics of the six BODIPY dyes 15 - 20 were analyzed before attachment of the
targeting component. High quantum yields resulting in high intensity of the emitted
light are desirable. Excitation and emission wavelengths suitable for commercially
available filter sets (FITC, TRITC) are necessary. The photophysical properties of the
BODIPY dyes 15 - 20 are summarized in Tables 1-3 and discussed thereafter (vide
infra).
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
10
Table 1: Absorption and emission maxima of the dyes 15 - 20.
Absorption
[nm] in
CH3CN
Emission
[nm] in
CH3CN
λem, max
Absorption
[nm] in
CH2Cl2/CH3OH
λabs, max
Emission
[nm] in
CH2Cl2/CH3OH
λem, max
Compound
λabs, max
15
16
17
18
19
20
553
521
521
521
497
497
571
533
533
532
507
508
559
524
524
524
500
500
574
537
537
535
510
510
Wavelength of the FITC and TRITC filter sets: excitation(FITC) = 450 – 490 nm;


emission(FITC) = 515 - 565 nm; excitation(TRITC) = 546/12 nm (bandpass filter);
emission(TRITC) = 590 nm (longpass filter).
Thus, the absorption and emission wavelengths of BODIPY dyes 16 - 20 fit to the FITC
filter set. The TRITC filter set can be used for fluorescence microscopy of Bodipy 15.
Table 2: Photoluminescence quantum yields (ΦF), fluorescence lifetimes (F),
fluorescence (kF) and non-radiative (knr) rate constants in fluid CH3CN at 298 K of
compounds 15 - 20.
compound
ΦF (±0.02)
kF [106 s-1]
knr [106 s-1]
F [ns]
0.891 ± 0.006
6.32 ± 0.05
6.37 ± 0.05
6.60 ± 0.06
3.96 ± 0.03
4.09 ± 0.03
40 ± 20
128 ± 4
127 ± 4
124 ± 4
134 ± 6
122 ± 6
1077 ± 30
30 ± 5
15
16
17
18
19
20
0.04
0.81
0.81
30 ± 5
27 ± 6
0.82
0.53
120 ± 8
122 ± 8
0.50
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
11
Table 3: Photoluminescence quantum yields (ΦF), fluorescence lifetimes (F),
fluorescence (kF) and non-radiative (knr) rate constants in CH2Cl2/CH3OH (1:1) of
compounds 15 - 20.
ΦF
kF
knr
kF
knr
ΦF
F [ns]
F [ns]
[106 s-1] [106 s-1]
[106 s-1]
77K
[106 s-1]
77K
(±0.1)
77K
compound
(±0.02)
298 K
298 K
77K
298 K
61 ± 30
122 ± 4
96 ± 4
298 K
0.818 ±
0.004
6.00 ±
0.02
6.54 ±
0.03
5.83 ±
0.03
3.810 ±
0.016
4.408 ±
0.018
1160 ±
40
1.252 ±
0.002
5.99 ±
0.02
5.46 ±
0.02
15
16
17
18
19
20
0.05
0.73
0.63
0.64
0.41
0.44
0.1
<160
720 ± 80
< 30
45 ± 5
57 ± 5
62 ± 6
0.9
0.9
150 ± 20
160 ± 20
< 40
0.9
0.9
0.9
110 ± 4
5.6 ± 0.2 160 ± 20
< 40
5.06 ±
180 ± 20
0.02
108 ± 6 154 ± 8
100 ± 5 127 ± 8
< 40
5.8 ± 0.2 160 ± 20
< 30
The solvatochromic shift in the absorption and in the emission maxima is modest for
all dyes when increasing the polarity of the solvent, which points to a predominant *
character of the excited states that also accounts for the small Stokes shifts and the
vibrational shoulders observed in all cases (see Table 1 and the corresponding spectra
in the ESI). However, the absorption and the emission spectra both in CH3CN and in
the more polar CH2Cl2/CH3OH mixture clearly show a batochromic shift for compound
15 and a blue-shift for 19 and 20, if compared with 16 - 18 and indicating that the
energy of the excited states can be fine-tuned by judicious choice of the substitution
pattern. In particular, the fused thiophene rings extend the delocalization of the 
system, causing the observed red-shifted absorption and emission maxima.
Interestingly, the fluorescence rate constants in CH3CN are comparable for all
exemplars except for compound 15, which is significantly smaller (Table 2). Moreover,
the radiationless deactivation rates are drastically higher for the latter and significantly
enhanced for 19 and 20, if compared with compounds 16 - 18. This is surprising since
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
12
16
-
18
have
additional
side
chains
providing
extra
degrees of freedom and therefore a higher density of rotovibrational states
that could favor non-radiative relaxation pathways. This points to
radiationless transitions that are mostly governed by electronic effects
associated to the variable substitution patterns. The latter is evident for
compounds 19 - 20 but particularly for 15, as can be also observed in the
more polar CH2Cl2/CH3OH mixture where in all cases the radiative relaxation
rates are slightly slower than in CH3CN while the radiationless processes
are somewhat faster than in the less polar environment (Table 3). Interestingly, in
frozen matrices at 77K, all dyes show practically identically high radiative rate
constants, whereas the radiationless process is only (yet drastically) boosted for
compound 15. This observation points to an enhanced intersystem crossing that could
be ascribed to the additional lone pairs from the sulphur atoms at the thiophene rings
(El Syed rule). It is also evident that the substitution at the methyne bridge has no
significant effect on the excited state energies and their relaxation rate constants.
In order to obtain the desired fluorescently labeled ligands, the synthesized BODIPY
dyes 15 - 20 were coupled with the alkyne 8 via Cu(I)-catalyzed azide-alkyne
[3+2]cycloaddition reaction. (Scheme 4).
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
13
Scheme 4: Synthesis of fluorescently labeled tool compounds. Reagents and reaction
conditions: (a) CuSO4, sodium ascorbate, DMF, H2O, rt, 24 h, 21: 58 %, 22: 57 %, 23:
61 %, 24: 2 %, 25: 19 %, 26: 68 %.
Figure 2: Fluorescently labeled ligands 21 - 26 designed to image KCa3.1 channels.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
14
The six novel fluorescently labeled ligands 21 - 26 shown in Figure 2 were tested in
first simple staining experiments to analyze their suitability for in vitro imaging of the
KCa3.1 channel.
3. In vitro imaging of NSCLC cells
The A549 cell line, used for all experiments, is a hypotriploid epithelial cell line from a
non-small-cell lung cancer (NSCLC) taken from a 58-year-old Caucasian in 1972. The
suffix "3R" refers to the repeated process of intravenous administration of parental cells
(also initially "0R") into immunocompromised mice to obtain lung metastases. Cells
from these metastases were isolated and cultured in vitro followed by another round of
reinjection to select for tumor cells with high metastatic potential.[16]
A549-3R cells overexpress the KCa3.1 channel.[7] Thus, they are predestined for testing
of the novel small-molecule imaging probes. We aim at the development of a fast and
simple staining protocol. Therefore, a drop (30 µL) of a previously prepared staining
solution (10 µM in PBS) of the respective fluorescently labeled ligand was pipetted
onto Parafilm. The cover slip with the adherent cells was placed carefully upside down
onto the drop and the NSCLC cells were incubated in a humidified dark chamber for
10 min. (Figure 3)
Figure 3: Setup for the staining of NSCLC cells.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
15
The NSCLC cells are ready for microscopy after only 10 min incubation time and
subsequent washing with PBS-buffer. This newly developed protocol using the
fluorescently labeled probes is very fast and simple in comparison to the much more
complicated indirect immunofluorescence assay (approx. 4 h). An inverted
fluorescence microscope was used to image NSCLC cells labeled with the fluorescent
senicapoc derivative.
After performing the staining protocol with the fluorescently labeled thiophene
annulated ligand 21, staining of the A549-3R cells could not be observed. To ensure
that this result is not due to the low quantum yield of BODIPY 15, a drop of the staining
solution was examined under the microscope and an orange-red color was observed.
Thus, imaging probe 21 is not suitable for the visualization of the KCa3.1 channel.
The same staining protocol was performed with the fluorescently labeled ligands 22
and 23 bearing an ethyl and two methyl moieties at the pyrrole rings and the NSCLC
cells were observed. Unfortunately, the staining was not specific and no staining
pattern of the KCa3.1 channels was found. (see Figure S1 in the Supporting
Information)
Problems occurred while preparing the staining solution of fluorescently labeled
ligand 24 with the undecyloxy linker. Due to its high lipophilicity, it was not possible to
dilute the DMSO stock solution with PBS-buffer. The molecule precipitated
immediately and neither ultrasound nor heating up the suspension led to a clear
solution. Thus, it was not possible to perform the staining experiments with imaging
probe 24.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
16
In contrast, the fluorescently labeled dimethylpyrrole-based ligands 25 and 26 showed
promising results. In both cases the typical punctate staining pattern of the KCa3.1
channel was observed. [7], [17] (Figure 4, A-D)
As a proof-of-principle, the NSCLC cells were pre-incubated with unlabeled senicapoc
(30 µM) 5 min before staining. Consequently, the typical punctate staining pattern
could not be observed any more. (Figure 4, E-F)
B
Figure 4: A: NSCLC cells incubated for 10 min with a 10 µM staining solution of imaging
probe 25. B: Magnification of A (white box). C: NSCLC cells incubated for 10 min with
a 10 µM staining solution of imaging probe 26. D: Magnification of C (white box). E:
NSCLC cells blocked for 5 min with unlabeled senicapoc (30µM) and subsequently
stained for 10 min with a 10 µM staining solution of imaging probe 25. F: NSCLC cells
blocked for 5 min with unlabeled senicapoc (30 µM) and subsequently for 10 min with
a 10 µM staining solution of imaging probe 26.
Since all the binding regions were blocked with senicapoc, the fluorescently labeled
probes 25 and 26 were no longer able to bind specifically to the KCa3.1 channel. This
experiment indicates that the novel developed fluorescently labeled ligands bind to the
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
17
same region of the KCa3.1 channel as senicapoc does. Therefore, specific binding of
the new probes to the KCa3.1 channel was confirmed. Thus, the two novel imaging
probes 25 and 26 are suitable for fast and simple visualization of the KCa3.1 channels,
which appear as bright spots in the fluorescence images.
Since the best results were obtained using the fluorescently labeled ligand 25 (best
signal-to-noise ratio), this novel imaging probe was further evaluated. In a control
experiment HEK293 cells not expressing the KCa3.1 channel were incubated with a
10 µM staining solution of 25 following the same protocol as used for the NSCLC cell
staining experiments. Staining of the HEK293 cells could not be observed. (see Figure
S2 in the Supporting Information)
Further staining experiments with the fluorescently labeled ligand 25 in living NSCLC
cells showed more background signals due to the incomplete permeation of the cell
membrane. However, the punctate staining pattern of the KCa3.1 channel was still
visible. (see Figure S3 in the Supporting Information)
In order to compare the results obtained by KCa3.1 channel imaging with fluorescently
labeled ligand 25 with the results obtained by the indirect immunofluorescence assay,
both methods were performed and the KCa3.1 channel density was determined for each
experiment. (Figure 5) Therefore, ten squares with a side length of 50 pixels,
corresponding to an area of 9 μm², were randomly located in one cell. Signals were
analyzed with a linescan (MetaVue software) and were counted as one KCa3.1 channel
if the Full Width at Half Maximum (FWHM) was ≤5 pixels (~ 300 nm). In total 15 cells
(150 squares) were analyzed for each staining method.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
18
1225
1150
1075
1000
925
B
signal 1
5
signal 2
850
775
0
10
15
20
distance [pixel]
D
145
130
115
100
85
signal 1
10
70
0
5
15
distance [pixel]
Figure 5: Signal analysis with linescans (blue lines in magnification boxes). 1 pixel ≙
~ 60 nm. A: NSCLC cells incubated for 10 min with a 10 µM staining solution of imaging
probe 25 with 50x50 pixel square, magnification of this square with linescan and 16
signals for KCa3.1 channels. B: Linescan of A: Intensity profil [arbitrary units] of signal 1
and signal 2. C: NSCLC cells after performing the indirect immunofluorescence assay
with 50x50 pixel square, magnification of this square with linescan and 16 signals for
KCa3.1 channels. D: Linescan of C: Intensity profil [arbitrary units] of signal 1.
For both, the imaging with fluorescently labeled ligand 25 and the indirect
immunofluorescence assay an average KCa3.1 channel density of 1.79 µm-2 was found
(compare Figures S4 and S5 in Supporting Information). This analysis shows, that the
same results can be achieved with the imaging probe 25 compared to the indirect
immunofluorescence assay but staining with 25 is more than six-fold faster and more
convenient than performing the indirect immunofluorescence assay.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
19
In order to investigate the binding mode of the fluorescently labeled ligand 25 a
modelling of 25 in the available KCa3.1 Cryo-EM structure (pdb 6cn0)[18] was
performed. (Figure 6)
A
B
C
D
Figure 6: A model of 25 (green) in the KCa3.1 Cryo-EM structure (pdb 6cn0)[18] in the
bound state II (light blue) with bound calmodulin (khaki) . A-B: Overview of 25 binding
in the inner pore. C: Surface representation of 25. D: Molecular representation of 25
and surface representation of ion channel. The given distance of 7.86 Å.(D) indicates
the distance between the O-atoms and thus the length of the linker and the distance
between senicapoc moiety and the fluorescent dye.
The modelling of the fluorescently labeled ligand 25 indicates a linker length that is
nearly perfect to span the distance between the senicapoc moiety and the dye. Figure
6 A-B shows the binding of 25 in the inner pore of the KCa3.1 channel. Figure 6 C-D
shows a more detailed view and how 25 fits onto the surface of the KCa3.1 channel.
The senicapoc moiety (in the upper part) is binding in a similar region as described for
a rosetta homology model of KCa3.1 channel. In this model senicapoc binds also to the
upper inner pore of the channel.[19] The dye is lying outside of the inner pore (lower
part) directly on the surface. Therefore, the linker shows a suitable length to connect
both moieties via the inner pore. (Figure 6 D) The distance between both O-atoms is
~7.86 Å. This analysis confirms a a similar binding mode of the fluorescently labeled
probe 25 and senicapoc in the inner pore of the KCa3.1 channel.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
20
4. Conclusion
A series of fluorescently labeled ligands targeting the KCa3.1 channel was designed,
synthesized and evaluated for their suitability to image KCa3.1 channels in vitro in
NSCLC cells via fluorescence microscopy. Ligands 25 and 26 labeled with
dimethylpyrrole-based BODIPYs and different linkers showed promising results in
staining experiments. The typical punctate staining pattern was observed, which was
reversed after pre-incubation with the competitor senicapoc. Imaging of the KCa3.1
channels with the indirect immunofluorescence assay and ligand 25 led to the same
density of signals, but imaging with 25 was more than six-fold faster. Additionally, a
modelling in the available KCa3.1 Cryo-EM structure showed similar binding modes for
25 and senicapoc. Compounds 22 and 23 labeled with ethyldimethylpyrrole-based
BODIPYs revealed unspecific staining of NSCLC cells without any structure. Whereas
ethyl and 3,6,9-trioxaundecyl moieties are suitable linkers for fluorescently labeled
probes, a dodecyl linker (e.g. 24) led to highly lipophilic compounds resulting in low
water solubility, which caused problems in terms of preparing staining solutions.
Compound 21 labeled with the BODIPY bearing the thienopyrrole moiety was not
suitable as novel small-molecule imaging probe due to lack of staining the NSCLC
cells. Low quantum yields of the connected dye 15 resulting in low intensity of emitted
light is an additional disadvantage of this compound. In summary, dimethylpyrrole-
based BODIPYs connected with the targeting senicapoc moiety via an ethyl linker (25)
represents a promising new small-molecule probe for selective imaging of KCa3.1
channels of NSCLC cells in vitro and in vivo using a simple, fast and efficient staining
protocol.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
21
Acknowledgments
We are grateful for cultivation and preparation of the cells by Sandra Schimmelpfennig
and Sarah Sargin at the Institute of Physiology II. This work was supported by the
Deutsche Forschungsgemeinschaft (DFG), which is gratefully acknowledged.
Moreover, we are grateful to Cells-in-Motion (CiM) Cluster of Excellence for supporting
this project by a Pilot Project. Dr. Ivan Maisuls gratefully acknowledges the Post-
Doctoral Fellowship form the Alexander von Humboldt Foundation. Furthermore,
funding of the research training group “Chemical biology of ion channels (Chembion)”
by the DFG is gratefully acknowledged.
Supporting Information
Supporting Information contains purity data of all compounds, additional staining
experiments, the analysis of the KCa3.1 channel density and details of the molecular
1
modelling studies. Moreover, the complete experimental part including H and 13C
NMR spectra, HPLC chromatograms (purity), absorption and emission spectra and
photophysical data are presented. This material is available free of charge.
Author Information
Corresponding author
*Tel: +49-251-8333311. Fax: +49-8332144. Email: wuensch@uni-muenster.de
Conflict of interests
The authors declare no conflict of interest.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
22
Abbreviations Used
BODIPY
DMSO
FITC
kf
boron dipyrromethane dye
dimethyl sulfoxide
fluorescein isothiocyanate
radiative constant
knr
non-radiative constant
non-small-cell lung cancer
phosphate buffered saline
paraformaldehyde
NSCLC
PBS
PFA
TBDPS-Cl
TRITC
ΦL
tert-butyldiphenylsilyl chloride
tetramethylrhodamine isothiocyanate
quantum yield

fluorescence lifetime
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
23
References
[1] N. Prevarskaya, R. Skryma, Y. Shuba, Physiol. Rev. 2018, 98, 559.
[2] Z. Pethő, K. Najder, E. Bulk, A. Schwab, Cell calcium 2019, 80, 79.
[3] C. J. Mohr, F. A. Steudel, D. Gross, P. Ruth, W.-Y. Lo, R. Hoppe, W. Schroth, H.
Brauch, S. M. Huber, R. Lukowski, Cancers 2019, 11.
[4] K. L. Turner, A. Honasoge, S. M. Robert, M. M. McFerrin, H. Sontheimer, Glia
2014, 62, 971.
[5] M. Rabjerg, A. Oliván-Viguera, L. K. Hansen, L. Jensen, L. Sevelsted-Møller, S.
Walter, B. L. Jensen, N. Marcussen, R. Köhler, PloS one 2015, 10, e0122992.
[6] M. Faouzi, F. Hague, D. Geerts, A.-S. Ay, M. Potier-Cartereau, A. Ahidouch, H.
Ouadid-Ahidouch, Oncotarget 2016, 7, 36419.
[7] E. Bulk, A.-S. Ay, M. Hammadi, H. Ouadid-Ahidouch, S. Schelhaas, A. Hascher,
C. Rohde, N. H. Thoennissen, R. Wiewrodt, E. Schmidt et al., Int. J. Cancer
2015, 137, 1306.
[8] N. Haren, H. Khorsi, M. Faouzi, A. Ahidouch, H. Sevestre, H. Ouadid-Ahidouch,
Histol. Histopath. 2010, 25, 1247.
[9] J.-H. Ko, E. A. Ko, W. Gu, I. Lim, H. Bang, T. Zhou, Mol. Cancer 2013, 12, 106.
[10] J. W. Stocker, L. de Franceschi, G. A. McNaughton-Smith, R. Corrocher, Y.
Beuzard, C. Brugnara, Blood 2003, 101, 2412.
[11] A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891.
[12] T. Kowada, H. Maeda, K. Kikuchi, Chem. Soc. Rev 2015, 44, 4953.
[13] C. P. Konken, G. Haufe, K. Brömmel, B. Wünsch, M. Schäfers, S. Wagner, V.
Hugenberg, Dyes Pigm. 2018, 158, 88.
[14] X. Zhang, Y. Xiao, X. Qian, Org. Lett. 2008, 10, 29.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
24
[15] B. C. Sanders, F. Friscourt, P. A. Ledin, N. E. Mbua, S. Arumugam, J. Guo, T.
J. Boltje, V. V. Popik, G.-J. Boons, J. Am. Chem. Soc. 2011, 133, 949.
[16] A. Hascher, A.-K. Haase, K. Hebestreit, C. Rohde, H.-U. Klein, M. Rius, D.
Jungen, A. Witten, M. Stoll, I. Schulze et al., Clin. Cancer Res. 2014, 20, 814.
[17] V. Nechyporuk-Zloy, C. Stock, H. Schillers, H. Oberleithner, A. Schwab, Am. J.
Physiol. Cell Physiol. 2006, 291, C266-9.
[18] C.-H. Lee, R. MacKinnon, Science (New York, N.Y.) 2018, 360, 508.
[19] H. M. Nguyen, V. Singh, B. Pressly, D. P. Jenkins, H. Wulff, V. Yarov-Yarovoy,
Mol. Pharmacol. 2017, 91, 392.
[20] B. A. Kamino, J. Castrucci, T. P. Bender, Silicon 2011, 3, 125.
[21] R. W. Burry, J. Histochem. Cytochem.2011, 59, 6.
This article is protected by copyright. All rights reserved.

10.1002/anie.202001201
Angewandte Chemie International Edition
25
Table of Contents Graphic
This article is protected by copyright. All rights reserved.