Fluorination of Photoswitchable Muscarinic Agonists Tunes Receptor Pharmacology and Photochromic Properties

Article abstract of DOI:10.1021/acs.jmedchem.8b01822

Red-shifted azobenzene scaffolds have emerged as useful molecular photoswitches to expand potential applications of photopharmacological tool compounds. As one of them, tetra-ortho-fluoro azobenzene is well compatible for the design of visible-light-responsive systems, providing stable and bidirectional photoconversions and tissue-compatible characteristics. Using the unsubstituted azobenzene core and its tetra-ortho-fluorinated analogue, we have developed a set of uni- and bivalent photoswitchable toolbox derivatives of the highly potent muscarinic acetylcholine receptor agonist iperoxo. We investigated the impact of the substitution pattern on receptor activity and evaluated the different binding modes. Compounds 9b and 15b show excellent photochemical properties and biological activity as fluorination of the azobenzene core alters not only the photochromic behavior but also the pharmacological profile at the muscarinic M₁ receptor. These findings demonstrate that incorporation of fluorinated azobenzenes not just may alter photophysical properties but can exhibit a considerably different biological profile that has to be carefully investigated.

Full text of DOI:10.1021/acs.jmedchem.8b01822

Subscriber access provided by ECU Libraries
Article
Fluorination of Photoswitchable Muscarinic Agonists Tunes
Receptor Pharmacology and Photochromic Properties
Luca Agnetta, Marcel Bermudez, Fabio Riefolo, CARLO MATERA, Enrique Claro,
Regina Messerer, Timo Littmann, Gerhard Wolber, Ulrike Holzgrabe, and Michael Decker
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01822 • Publication Date (Web): 03 Mar 2019
Downloaded from http://pubs.acs.org on March 4, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
Fluorination of Photoswitchable Muscarinic Agonists Tunes Receptor
Pharmacology and Photochromic Properties
7
8
9
Luca Agnetta†, Marcel Bermudez$, Fabio Riefolo§#, Carlo Matera§#, Enrique Claro¥, Regina
Messerer†, Timo Littmann‡, Gerhard Wolber$, Ulrike Holzgrabe†, Michael Decker†*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†Pharmaceutical and Medicinal Chemistry, Institute of Pharmacy and Food Chemistry, Julius Maximilian University of
Würzburg, Am Hubland, 97074 Würzburg, Germany.
$Institute of Pharmacy, Freie Universität Berlin, Königin-Luise-Straße 2 + 4, 14195 Berlin, Germany.
§Institute for Bioengineering of Catalonia (IBEC), Barcelona Institute for Science and Technology, Carrer Baldiri Reixac
15-21, 08028 Barcelona, Spain.
#Network Biomedical Research Center in Bioengineering, Biomaterials, and Nanomedicine (CIBER-BBN), 50018 Zaragoza,
Spain.
¥Institut de Neurociències (INc), and, Dept. Bioquímica i Biologia Molecular, Unitat de Bioquímica de Medicina,
Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Barcelona, Spain.
‡Institute of Pharmacy, University of Regensburg, Universitätstraße 31, 93053 Regensburg, Germany.
ABSTRACT: Red-shifted azobenzene scaffolds have emerged as useful molecular photoswitches to expand potential
applications of photopharmacological tool compounds. As one of them, tetra-ortho-fluoro azobenzene is well
compatible for the design of visible light responsive systems, providing stable and bidirectional photoconversions
and tissue-compatible characteristics. Using the unsubstituted azobenzene core and its tetra-ortho-fluorinated analog
we have developed a set of uni- and bivalent photoswitchable toolbox derivatives of the highly potent muscarinic
acetylcholine receptor agonist iperoxo. We investigated the impact of the substitution pattern on receptor activity
and evaluated the different binding modes. Compounds 9b and 15b show excellent photochemical properties and
biological activity, as fluorination of the azobenzene core alters not only the photochromic behavior, but also the
pharmacological profile at the muscarinic M1 receptor. These findings demonstrate that incorporation of fluorinated
azobenzenes may alter not just photophysical properties but can exhibit a considerably different biological profile
that has to be carefully investigated.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 24
1
2
3
4
5
6
INTRODUCTION
In recent years, many different biochemical targets, including ion channels, enzymes and lipids have been effectively
modulated in light-controlled fashion using photoswitchable ligands, expanding the vibrant field of
7
a
8
9
photopharmacology.1 Now, the potential of the application of photopharmacological tool compounds to G protein-
coupled receptors (GPCRs) is being steadily investigated.2 The rapid light-induced isomerization of photochromic
ligands can be directly translated into a change in affinity or activity. This can give important new insights into binding
mode and time course of activation processes enabling precise spatial and temporal resolution of the complex
signaling pathway of GPCRs.3 The muscarinic acetylcholine receptors (mAChRs), which belong to class A GPCRs, have
received special attention in this regard, due to their role as prototypic pharmacological system4 and their therapeutic
potential.5 The muscarinic receptors mediate the excitatory and inhibitory effects of the neurotransmitter
acetylcholine (ACh)6 thus regulating diverse important biological processes. Muscarinic receptors are widely
expressed in the human body including the central nervous system (M1/M4/M5) related to neurological processes of
memory and learning. In the peripheral nervous system (M2/M3) they are related to the so-called “rest and digest”
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures of 1) reference agonists acetylcholine (ACh), carbachol (CCh) and iperoxo (iper) 2) univalent, 3)
homobivalent photoswitchable derivatives with azobenzene- and tetra-ortho-fluoro azobenzene scaffolds and homobivalent
ligands linked by polymethylene linkers.
2
ACS Paragon Plus Environment

Page 3 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
biological functions.7 In this work we present an optimized approach for the application of photochromic ligands to
GPCRs by using the M1 receptor as a prototypic model system, outlining the advantages and challenges of using
redshifted molecular photoswitches.
8
9
Accordingly, we focused on the highly potent orthosteric agonist iperoxo in the design of bidirectional
pharmacological tool compounds with tuned photochemical properties. Iperoxo is a muscarinic ACh receptor agonist
with outstanding potency that is known to tolerate bulky substituents without compromising agonist efficacy.8
Recently, we reported on photoiperoxo (9a), a compound that consists of iperoxo extended with a molecular
photoswitch and a dualsteric photochromic ligand BQCAAI.2c We made use of azobenzene as a molecular photoswitch
to change intrinsically the geometry and polarity of the linking unit and to control its activation by light.1a, 1b, 1d Upon
irradiation with UV- and blue light, and thus triggering a switch from the trans- to the cis-form, we were able to control
the efficacy and to investigate the time course of receptor activation processes.2c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Synthesis of photoiperoxo and F4-photoiperoxoa
aReagents and conditions: (a) Pd/C, EtOH (87%); (b) Oxone, CH2Cl2, water; (c) 2, AcOH/TFA, toluene (43%); (d) NBS,
AIBN, CCl4, 80 °C (50%); (e) nitrosobenzene, AcOH (75%); (f) CBr4, PPh3, CH2Cl2 (60%); (g) 4((4,5-Dihydroisoxazol-3-
yl)oxy)-N,N-dimethylbut-2-yn-1-amine 16, EtOAc/MeCN (39% for 9a, 92% for 9b).
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 24
1
2
3
4
Scheme 2. Synthesis of iper-azo-iper and F4-iper-azo-iperb
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
.
bReagents and conditions: (a) Oxone, CH2Cl2, water; (b) 10, AcOH (36%); (c) KMnO4, FeSO4 7H2O, CH2Cl2 (29%); (d)
NBS, AIBN, CCl4, 80 °C (76% for 14a, 24% for 14b); (e) 4((4,5-Dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-
amine, EtOAc, 60 °C (83% for 15a, 32% for 15b).
Despite this remarkable progress in the development of photopharmacological tool compounds, the poor
photostationary states (PSSs) of our azobenzene-based systems resulted in an unclear correlation between the ratio
of the cis/trans concentration to the actual activity at the receptor. The PSS strongly depends on the electronic
environment of the system and the overlap of transitions of the cis/trans photoisomers and can be influenced by
changing the substitution pattern of the azobenzene moiety. Additionally, the operational wavelengths to trigger
photoisomerization can interfere to a significant degree with the fluorescent readout methods that are commonly
used in GPCR research. To investigate the complex nature of M1 activation, photopharmacological tool compounds
should ideally show superior photochemical properties, such as quantitative and stable bidirectional photoswitching
and be responsive to visible rather than high energy light.1a, 1b, 1d A convincing approach to optimize the photochemical
properties of azobenzenes involving the tetra-ortho-substitution of azobenzene has been described previously by the
groups of Woolley9 and Hecht.10 By choosing the suitable moieties, tetra-ortho-substitution leads to a separation of
the n→* transitions of trans and cis isomers and ultimately to almost complete trans/cis photoconversions.
Importantly, the desired isomer can be selectively formed with wavelengths in the visible light window. Despite their
promising application in photobiology due to the well separated n→* transitions,11 ortho-methoxylated azobenzenes
are sterically demanding and strongly twisted about the N-N double bond. This makes them unlikely to fit into the
narrow receptor binding pocket. Ortho-chlorinated and fluorinated azobenzenes show comparable red-shifted trans
4
ACS Paragon Plus Environment

Page 5 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
n-* transitions, albeit somewhat less than ortho-methoxylated and have been successfully employed in biological
applications.12 However, tetra-ortho-fluoro substituted azobenzenes stand out in their slow thermal cis-to-trans
relaxation rate, representing a truly bistable molecular switch on biological time scale. Moreover, the fluorine
substituents are sterically less demanding compared to the chlorine ones.10a Based on these results we further
investigated the effect of iperoxo-derived photoswitchable ligands by developing a set of tool compounds, containing
the iperoxo motif and the tetra-ortho-fluoro azobenzene or unsubstituted azobenzene, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In modern medicinal chemistry, the bivalent strategy has been extensively applied to GPCR ligands.13 Appropriately
designed bivalent ligands can exhibit higher affinity, potency and selectivity compared with the parent ligand, with
potential therapeutic application.14 Successful examples of the application of this approach to GPCRs include the
human cannabinoid receptor 2 (hCB2),15 opioid receptor (OR dopamine 2 (D2)17 and muscarinic receptors.18 To
this end, we extended the photoiperoxo structure with another iperoxo moiety, creating a homobivalent ligand, to
improve and investigate binding at the M1 receptor.
RESULTS AND DISCUSSION
We now report the design and synthesis of photoswitchable homobivalent iperoxo (15a), tetra-ortho-fluoro-
photoiperoxo (9b) and homobivalent tetra-ortho-fluoro-photoiperoxo (15b), being the corresponding tetra-ortho-
fluorinated analogs of the azobenzene iperoxo derivatives (Figure 1). In addition, we synthesized homobivalent
iperoxo derivatives (18a-d) connected by aliphatic chains as reference compounds to identify the optimal distance
between the two orthosteric moieties for dualsteric binding.
Chemistry. The synthetic routes for the uni- and homobivalent ligands are summarized in Schemes 1 and 2,
respectively. Azobenzene moieties 5 and 7 were accessed starting from the corresponding anilines 2, obtained from
reduction of 1,3-difluoro-5-methyl-2-nitrobenzene 1, and commercially available (4-aminophenyl)methanol 6.
Anilines were used in Baeyer-Mills reactions with 1,3-difluoro-2-nitrosobenzene 4 and nitrosobenzene, respectively,
to afford precursors for bromination, which took place either under radical (for 8b) or nucleophilic substitution
conditions (for 8a). Lastly, 4((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-amine, prepared using the
standard convergent procedure as described previously19, was connected to the photoswitches to afford
photoiperoxo 9a and F4-photoiperoxo 9b.
The synthesis of the homobivalent ligands started from commercially available p-toluidine 10, which was condensed
to compound 13a by means of a Baeyer-Mills reaction. Tetra-ortho-fluorinated analogue 13b was obtained by
oxidative coupling of 12 with potassium permanganate and iron sulfate heptahydrate as oxidizing reagents. Again,
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 24
1
2
3
4
5
6
7
8
9
radical bromination and substitution with 4((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-amine resulted
in formation of target compounds 15a and 15b. Aliphatic derivatives (18a-d) were synthesized reacting the
corresponding double brominated aliphatic chains with two equivalents of 4((4,5-dihydroisoxazol-3-yl)oxy)-N,N-
dimethylbut-2-yn-1-amine (Suppl. Scheme 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
UV/Vis Spectroscopic Characterization. The UV/Vis absorption spectra of the set of compounds 9a,b and 15a,b
were measured at a concentration of 50 M in DMSO. For compounds 9a and 15a (with unsubstituted azobenzene
scaffold), we observed a strong -* transition band at short wavelength (*≈ 320 nm) and a weaker n-*
(n*≈ 422 nm) upon irradiation with blue light (400 nm, trans isomer). Irradiation with UV light (365 nm), and thus
photoconversion to the cis isomer, leads to a decrease in absorbance intensity of the -* and increase of n-* band
Figure 2. Representative (A) absorption spectra of PSS at the dark-adapted state and after illumination with 400 nm (to
trans) and 500 nm (to cis), (B) stability measurement and (C) multiple cis/trans isomerization cycles of compound 9b in
dimethyl sulfoxide at 25°C. Spectra of compound 9a and 15a,b are displayed in Supplementary Information section.
Table 1. UV/VisSpectroscopic data of compounds 9a,b and 15a,b.c
max
(*)
max
(n*)
max
(n*)
Cmpd.
PSStrans PSScis
 (*)
 (n*)
 (n*)
 (n-*)
*103
[1/(M*cm)]
*103
[1/(M*cm)]
[nm]
[nm]
[1/(M*cm)]
[nm]
[nm]
%
%
trans
trans
cis
9a
15a
9b
320
320
307
311
12.2
22.0
16.7
14.8
418
426
443
447
680
920
418
426
412
412
1.04
1.42
1.80
1.38
<10
<10
31
90
99
94
93
62
60
86
98
1440
1180
15b
35
cmax (*, n*) represents the wavelength at the maximal absorption of the * and n* transition band, respectively.
The molar extinction coefficient  (*, n*) were calculated according to Lambert-Beer. (n*) is the difference
between trans and cis regarding max of n* band. PSS percentages after irradiation with operational wavelengths
determined by liquid chromatography using the wavelength at the isosbestic point as detecting wavelength.
6
ACS Paragon Plus Environment

Page 7 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
without affecting the wavelengths of the respective transition bands. Spectra of fluorinated compounds 9b and 15b
were measured using 400 nm for cis→trans and 500 nm for trans→cis photoconversions. Due to the n-orbital
stabilization of electron-withdrawing groups in the cis-state, tetra-ortho-fluoro substitution of the azobenzene core
causes a separation of the transitions of the n-* band, in our case of around 33 nm on average, which makes it
possible to address both isomers selectively with light in the visible region (Figure 1A).20 As a direct consequence,
the PSSs (measured by means of liquid chromatography at the isosbestic point wavelength) are substantially higher
than for non-fluorinated analogues (Table 1), which is beneficial for the application of photoswitchable ligands in
biological systems and allows a clear correlation between the distinct photoisomer and its biological effect. Moreover,
multiple cis/trans switching cycles did not cause noticeable degradation, highlighting the reliability and robustness
of the photochromic conversion of both the tetra-ortho-fluoro and unsubstituted azobenzene scaffolds. This is
confirmed by stability measurements in which the compounds were kept in the dark for at least 120 minutes after
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Competition for specific binding of 200 pM [3H]QNB to rat brain membranes containing high density of all the five
mAChRs by 9a,b and 15a,b. Data points were fitted using the "log(inhibitor) vs. normalized response - Variable slope"
function in GraphPad Prism 6.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 24
1
2
3
4
5
6
7
8
9
switching to the less stable cis-isomer, without significant changes in absorbance and hence in the PSS. Importantly,
this information is required when considering eventual incubation times required in biological assays, making sure
that the photoswitchable ligands do not relax to the trans-isoform during the readout (Figure 1B/C). From the
physicochemical point of view, this set of compounds shows ideal characteristics for stable and bidirectional
photoswitching.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Binding experiments. Firstly, 9a,b and 15a,b were tested for affinity to the M receptors. This was achieved in
competition experiments with the photoswitchable ligands in rat brain membrane preparations. These were
conducted as described previously,21 and contained a high density of all the five mAChR subtypes. [3H]Quinuclidinyl
benzilate ([3H]QNB) is a nonselective muscarinic ligand, which allows the identification of muscarinic receptors22
,
and displays an equilibrium dissociation constant of about 40 pM,23 making it suitable for these experiments. Specific
binding was defined with test compounds at total nominal concentrations in the range of 1 to 100 µM and elaborating
the raw dpm data from the scintillation counter,21 representing total radioactivity.
In general, the compounds 9a,b and 15a,b were found to show good affinity for the M receptors (Figure 3). In
particular, the azobenzene-elongated iperoxo 9a and 9b showed a binding affinity in the low-micromolar range,
although no significant changes could be observed upon irradiation (pIC50 ≃ 5 for both photoisomers, Figure 3A/C).
The affinity increased significantly for the homobivalent ligand 15a and 15b. In good agreement with results from
split-luciferase interaction assay, trans-15a showed an affinity in the high-nanomolar range, with a significant
difference between the two photo isoforms (pIC50 = 6.32 in the dark versus 5.88 under illumination with 365 nm
light, Figure 3B). Thus, we observed that despite the introduction of an azobenzene molecular photoswitch, a good
affinity could be preserved. Moreover, the presence of a second moiety of iperoxo, as in the homobivalent ligand 15a
and 15b, improved binding affinity and created an appreciable difference between trans and cis isomers.
Biological in vitro assay. Our goal was to assess the extent to which the pharmacological profiles of these compounds
change after fluoro-substitution of the azobenzene core. For this purpose, we used a novel split luciferase
complementation technique detecting the interaction between Gq subunit and phospholipase C-3 (PLC-3) and
thus reflecting G protein activation in living HEK 293T cells expressing the human M1 receptor.24 The split luciferase
complementation technique is particularly suitable to detect protein-protein interactions and associated signaling in
living cells.25 Since it is not fluorescence-based, no excitement irradiation is needed and consequently, its readout
does not interfere with the operational wavelengths for photoswitching, providing clear-cut and distinguishable
8
ACS Paragon Plus Environment

Page 9 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentration-response-curves for each photoisomer, which makes it highly suitable for photo-pharmacological
investigations into GPCRs. Technically, HEK 293T cells were engineered to express a fragment of the luciferase at the
Gq subunit of the heterotrimeric G protein and the complementary fragment at the N-terminus of the PLC-3. Upon
binding of the endogenous or synthetic agonists, both fragments are brought in close proximity leading to a
reconstitution of the functional luciferase protein and emission of bioluminescence, in the presence of the substrate
luciferin.24 Also advantageous is the fact that the receptor itself, in this case M1, remains un-engineered in contrast to
e.g. FRET-sensors, where often large constructs at intracellular loops could act as anchors and may affect changes in
conformations. Pharmacological data are depicted in Figure 4 and summarized in Table 2, and show the recorded
potencies and efficacies of compounds 9a,b and 15a,b, reflected by the pEC50, and maximal response Emax for each
isomer. The data are normalized to the maximum response of the synthetic agonist carbachol (CCh) at a concentration
of 100 μM. Measurements of the endogenous agonists acetylcholine (ACh) and the synthetic parent compound
iperoxo as references were not affected by illumination with operational wavelengths.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. G/PLC-3 split-luciferase interaction assay in HEK 293T cells expressing the human muscarinic M1 receptors
(hM1). Concentration-response-curves for A) Reference compounds, B) Iper-linker-Iper derivatives (18a-d), C) 9a and 9b
and D) 15a and 15b at operational wavelength specific for the respective photoconversion. Data represent means ± SEM of
3-4 experiments conducted at least in triplicate.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 24
1
2
3
4
5
6
7
8
9
Firstly, bivalent alkyl-substituted iperoxo compounds were screened to identify the correlation between the distance
of the orthosteric moieties and its corresponding biological activity. For this purpose, homobivalent ligands 18a-d
with different chain lengths were employed. All compounds were characterized as full agonists and longer spacer
proved advantageous for M1 affinity suggesting a dualsteric binding.26 The C10-spacer, which is comparable to the
azobenzene scaffold, shows the highest effect and suggests this distance as optimal for the design of photoswitchable
ligands (Suppl. Figure 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As investigated in previous studies, photoiperoxo 9a was unable to induce a conformational change at the M1 receptor
in FRET studies, neither upon illumination with UV- nor blue light. Instead, it exhibited antagonist behavior in
competition experiments.2c Using the split luciferase complementation assay, 9a shows two distinct curves for each
photoisomer, differing in their efficacy. The maximal response Emax was 48% for the trans and 34% for the cis
photoisomer, indicating partial agonism at the M1 receptor. Surprisingly, substitution of azobenzene with the tetra-
ortho-fluorinated scaffold resulted in full agonist 9b with a significant potency enhancement and pEC50-values almost
two log units higher (comparable to endogenous agonist ACh). However, no significant changes in efficacy or affinity
could be observed for 9b upon irradiation.
The potency at the M1 receptor was modulated effectively by introducing an additional iperoxo moiety resulting in
the homobivalent ligand 15a. By doing so, considerable improvement in the potency was gained and a difference
between the two photoisomers was re-established. Again, replacement of the azo-core with tetra-ortho-fluoro
Table 2. Pharmacological datad
Cmpd.
Iperoxo
CCh
pEC50
%Emax
8.90 ± 0.02
6.34 ± 0.02
6.87 ± 0.03
6.49 ± 0.01
6.77 ± 0.01
6.89 ± 0.01
7.86 ± 0.02
103.5 ± 1.0
99.9 ± 0.9
99.9 ± 1.8
103.6 ± 0.6
92.2 ± 0.6
101.1 ± 0.8
103.7 ± 0.7
ACh
18a
18b
18c
18d
trans
cis
trans
cis
9a
9b
5.49 ± 0.01
7.20 ± 0.02
7.82 ± 0.02
8.97 ± 0.01
5.68 ± 0.02
7.14 ± 0.01
7.49 ± 0.02
8.30 ± 0.02
48.0 ± 0.4
98.3 ± 1.2
96.5 ± 1.1
99.7 ± 0.3
33.5 ± 0.5
99.7 ± 0.6
87.9 ± 0.7
99.1 ± 0.4
15a
15b
dpEC50, concentration of the indicated compounds inducing a half-maximal effect (−log EC50 values); %Emax, maximum effect
as a percentage of ECCh (100 μM); Data were obtained by curve fitting to data from individual experiments shown in Figure
4.
10
ACS Paragon Plus Environment

Page 11 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scaffold (15b) resulted in a distinctive change toward higher potency, being almost 10-fold more pronounced for the
trans isomer. Noteworthy, trans-15b shows a pEC50 concentration in the one-digit nanomolar range, which is
comparable to the agonist iperoxo itself. This is a remarkable observation, as the introduction of a molecular
photoswitch generally causes an overall loss in activity at the receptor. A photoswitch-endowed compound showing
an almost identical potency as the reference compound is hard to achieve. Taken together, these findings suggest, that
binding of trans-form of bivalent derivatives 15a, 15b and 9b stabilizes the M1 receptor in the active conformation to
a greater extent than 9a. Additionally, fluoro substitution leads to pronounced differences in biological activity and is
beneficial for binding to the M1 receptor. As such, we developed a set of photopharmacological GPCR tool compounds
for a better understanding of M1 receptor binding modes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular Modelling. Docking studies using a previously reported homology model of the active M1 receptor
conformation27 reveal a dualsteric (bitopic) binding mode for all investigated photoswitchable iperoxo derivatives in
their trans conformation (Figure 5). We surmise that the photoswitch primarily occurs in solution and that the
pharmacological properties are mainly driven by binding of trans isomers. The active M1 receptor model indicates a
narrow channel between the orthosteric and the allosteric binding sites rendering binding of cis conformations to the
receptor unlikely due to steric interference with the tyrosine lid. Similar to other iperoxo-based dualsteric ligands28
Figure 5. Proposed binding modes and interactions of trans isomers of photoswitchable dualsteric iperoxo derivatives (A)
9b, (B) 9a and (C) 15a in complex with the M1 receptor. Positive ionizable centres are shown as blue stars, yellow spheres
indicate lipophilic contacts, purple disks show aromatic interactions and red arrows indicate hydrogen bonds.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 24
1
2
3
4
5
6
7
8
9
the iperoxo moiety is located in the orthosteric binding pocket showing interactions with D84, N213 and the tyrosine
lid consisting of Y85, Y212 and Y235 (Figure 5). For all compounds, a hydrogen bond between the azo group and the
hydroxy group of Y235 is formed. The sidechain of Y235 shows a reorientation towards the extracellular side, which
allows binding of azobenzene scaffolds through π-π interactions. This sidechain position allows full contraction of the
orthosteric binding site without full closure of the tyrosine lid and thereby stabilizes active receptor conformations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Interestingly, fluorination of the azobenzene scaffold results in a higher potency of 9b and 15b compared to the non-
fluorinated compounds 9a and 15a. This can be rationalized in our model by an optimal geometry of the aromatic
ring in the M1 receptor. Whereas the phenyl ring opposite to Y235 is more flexible in 9a and 15a, it is restrained in
an orientation, optimal for parallel π-π interaction. This is caused by the spatial requirements of the ortho-fluorine
atoms in 9b and 15b and an additional hydrogen bond with the hydroxy group of Y85 (Figure 5A). In addition,
fluorinated compounds show additional lipophilic contacts with the tyrosine lid as well as W231 (Figure 5A/B). Since
the symmetric compounds 15a and 15b are larger, they show secondary interactions in the extracellular vestibule
compared to 9a and 9b. In particular, a charge interaction of the ammonium group in the extracellular receptor
domain with E228 can be observed (Figure 5C). Positive charges in the allosteric vestibule recognized by aromatic
residues or the EDGE sequence have previously been found to be important for the M2 receptor.28b, 29
Interestingly, the M2 receptor lacks the presence of an acidic residue at the beginning of transmembrane domain 7,
suggesting a specific role for E228 in the M1 receptor. This observation is in accordance with the higher potencies for
15a and 15b that can be explained by this specific charge interaction and additional lipophilic contacts with Y158,
which were not observed for 9a and 9b due to the lack of an allosteric ammonium group (Figure 5). The absence of
extensive allosteric interactions and less lipophilic contacts of 9a compared 9b suggest a ligand-receptor complex in
which 9a is not able to fully stabilize the active receptor state. This is supported by previously reported FRET-based
measurements indicating the insensitivity of the M1 receptor upon 9a binding with regard to conformational
changes.2c However, the here applied Gα/PLC-β3 split-luciferase interaction assay unveiled 9a as a weak partial
agonist (Emax 34-48 %).
At first sight, it may appear conflicting that the cis photoisomers bind to the receptor and induce a receptor response,
while the docking data predict only the trans photoisomers to bind. On a closer look, even though the fluorinated
azobenzene scaffold provides higher PSS, a distinct percentage of trans isomer remains (Table 1), which can still bind
and activate the receptor. On a logarithmic scale, as used for the binding and functional studies, even a small trans
percentage shows a pronounced effect. An alternative explanation can be rationalized by a second purely allosteric
12
ACS Paragon Plus Environment

Page 13 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
binding mode as shown for several other dualsteric ligands.26, 28a For such a binding pose the cis orientation is
compulsory, forming - and cation- interactions, which can, to a certain extent, favor the active state of the receptor.
In contrast, the linear nature of the trans photoisomer is not suitable to solely bind to the allosteric binding site. At
this stage, this is still subject of future investigations though.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
We successfully developed a set of photopharmacological tools that allowed detailed investigation of the distinct
effects of fluorination and bivalency on binding properties at the M1 receptor. We designed photoswitchable iperoxo
(9a) and bivalent iperoxo compounds (15a), as well as the redshifted congeners (9b, 15b) by introduction of tetra-
ortho-fluoro scaffold. Bivalent compounds (15a,b) show much higher affinity compared to the univalent compounds
(9a,b) due to additional interactions at allosteric binding sites. Remarkably, the fluorine compounds (9b, 15b) not
only show improved operational wavelengths as shown in a novel, light-independent luciferase complementation
assay, but also increased potency at the M1 receptor. Bivalent and fluorinated photoiperoxo (15b) act as a pronounced
affinity switches, while the univalent photoiperoxo acts as efficacy switch (9a). The work significantly enlarges the
photopharmacological toolbox for mAChRs. We strongly recommend detailed pharmacological evaluation for
redshifted compounds since their biological properties might differ significantly from the parent compound. It will
be interesting to investigate the potential for in vivo photopharmacological control of this series.
EXPERIMENTAL SECTION
General Information. Common reagents and solvents were obtained from commercial suppliers (Aldrich, Steinheim,
Germany; Merck, Darmstadt, Germany) and were used without any further purification. Tetrahydrofuran (THF) was
distilled from sodium/benzophenone under an argon atmosphere. Microwave assisted reactions were carried out on a MLS-
rotaPREP instrument (Milestone, Leutkirch, Germany) using 8-10 weflon disks. Melting points were determined on a Stuart
melting point apparatus SMP3 (Bibby Scientific, UK). Thin-layer chromatography (TLC) was performed on silica gel 60 F254
plates (Macherey-Nagel, Düren, Germany) and spots were detected under UV light (λ=254 nm) or by staining with iodine.
Merck silica gel 60 (Merck, Darmstadt, Germany) was used for chromatography (230-400 mesh) columns or performed on
an Interchim Puri Flash 430 (Ultra Performance Flash Purification) instrument (Montluçon, France). Used columns are:
Silica 25 g – 30 µm, Alox-B 40 g – 32/63 µm, Alox-B 25 g – 32/63 µm (Interchim, Montluçon, France). Nuclear magnetic
resonance spectra were performed with a Bruker AV-400 NMR instrument (Bruker, Karlsruhe, Germany) in [d6]DMSO,
CDCl3, (CD3)2CO. As internal standard, the signals of the deuterated solvents were used (DMSO-d6: 1H 2.50 ppm, 13C
39.52 ppm; CDCl3: 1H 7.26 ppm, 13C 77.16 ppm; (CD3)2CO: 1H 2.05 ppm, 13C 39.52 ppm). Abbreviation for data quoted are:
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 24
1
2
3
4
5
6
7
8
9
s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad; dd, doublet of doublets; dt, doublet of triplets; tt, triplet of
triplets; tq, triplet of quartets. Coupling constants (J) are given in Hz. For purity and reaction monitoring, analytical HPLC
analysis was performed with a system from Shimadzu equipped with a DGU-20A3R controller, LC20AB liquid
chromatograph, and a SPD-20A UV/Vis detector. Stationary phase was a Synergi 4 μm fusion-RP (150×4.6 mm) column
(Phenomenex, Aschaffenburg, Germany). Mobile phase a gradient of MeOH/water with 0.1% formic acid was used.
Parameters: A = water, B = MeOH, V(B)/(V(A) + V(B)) = from 5% to 90% over 10 min, V(B)/(V(A) + V(B)) = 90% for 5 min,
V(B)/(V(A) + V(B)) = from 90% to 5% over 3 min. The method was performed with a flow rate of 1.0 mL/min. Compounds
were only used for biological evaluation if the purity was ≥ 95%. ESI mass spectral data were acquired with a Shimadzu
LCMS-2020. Data are reported as mass-to-charge ratio (m/z) of the corresponding positively charged molecular ions.
2,6-Difluoro-4-methylaniline (2). 1,3-difluoro-5-methyl-2-nitrobenzene 1 (1.0 g, 5.78 mmol) and 10% Pd-C (190 mg) in
EtOH (20 mL) was hydrogenated under atmospheric pressure for 3 h. The catalyst was filtered off, and the filtrate was
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
evaporated to give 2 as a light reddish oil (0.686 g, 4.80 mmol, 87%). H-NMR (CDCl3, 400 MHz): δ [ppm] = 6.63 (m, 2H),
3.49 (s, 2H), 2.23 (s, 3H). 13C-NMR (CDCl3, 400 MHz): δ [ppm] = 153.3 (d, J = 8.3 Hz), 151.0 (d, J = 8.3 Hz), 127.4 (t, J = 8.5
Hz), 121.1 (t, J = 16.6 Hz), 111.9 – 111.3 (m), 20.7 (t, J = 1.7 Hz). ESI-MS: m/z calc. for C7H8F4N2+ [M+H]+: 144.1, found: 144.1.
1-(2,6-Difluoro-4-methylphenyl)-2-(2,6-difluorophenyl)diazene (5). 2,6-Difluoraniline 3 (641 mg, 4.96 mmol, 2.00
equiv.) was dissolved in DCM (30 mL). To this solution, Oxone® (16.8g, 27.3 mmol, 20.00 equiv.) dissolved in water (60 mL)
was added. The solution was under argon at room temperature overnight. After separation of the layers, the aqueous layer
was extracted with DCM twice. The combined organic layers were washed with 1N HCl, saturated sodium bicarbonate
solution, water, brine, dried over MgSO4 and evaporated to dryness. Afterwards toluene (10 mL) and 2,6-difluoro-4-
metylaniline 2 (357 mg, 2.49 mmol, 1.00 equiv.) were added. A mixture of acetic acid (20 mL) and TFA (50 mL) was prepared
and added to the solution. The resulting mixture was stirred at room temperature for 24h. Afterwards it was evaporated to
dryness. Purification by column chromatography over silica gel (DCM:PE,1:4, silica gel) yielded the desired product 5 as a
1
deep orange solid (292 mg, 1.09 mmol, 43%). H-NMR (CDCl3, 400 MHz): δ [ppm] = 7.40 – 7.29 (m, 1H), 7.05 (t, J = 8.6 Hz,
2H), 6.87 (d, J = 10.5 Hz, 2H), 2.41 (s, 3H). ESI-MS: m/z calc. for C13H9BF4N2+ [M+H]+: 269.1, found: 269.1.
(4-(Phenyldiazenyl)phenyl)methanol (7). Nitrosobenzene 4a (1.28 g, 11.9 mmol, 3 equiv.) was dissolved in ethanol
(10 mL). To this solution (4-aminophenyl)methanol 6 (491 mg, 4.00 mmol, 1 equiv.) and acetic acid (20 mL) were added.
The reaction mixture was stirred for 4 h at room temperature. Subsequently, the reaction mixture was poured into 30 mL
of ice-cold water and the precipitate was collected by filtration. Column chromatography (DCM:methanol = 50:1) provided
compound 7 (659 mg, 2.99 mmol, 75%) as bright orange needles. 1H NMR (400 MHz, DMSO-d6): δ [ppm] = 7.88 (m, J = 6.9,
1.8 Hz, 4H), 7.65 – 7.50 (m, 5H), 5.40 (t, J = 5.7 Hz, 1H), 4.61 (d, J = 5.6 Hz, 2H). 13C NMR (101 MHz, DMSO-d6): δ [ppm] =
152.4, 151.3, 146.9, 131.8, 129.93, 127.6, 122.9, 62.9. ESI-MS: m/z calc. for C13H12N2O+ [M+H]+: 213.1, found: 213.1.
14
ACS Paragon Plus Environment

Page 15 of 24
Journal of Medicinal Chemistry
1
2
3
1-(4-(Bromomethyl)phenyl)-2-phenyldiazene (8a). (4-(phenyldiazenyl)phenyl)methanol
7 (300 mg, 1.41 mmol,
4
5
6
7
8
9
1 equiv.) was dissolved in dry THF (15 mL) under argon atmosphere, tetrabromomethane (703 mg, 2.12 mmol, 1.5 equiv.)
and triphenylphosphine (556 mg, 2.12 mmol, 1.5 equiv.) were added and the reaction mixture was stirred for 4 h. The
reaction mixture was filtered and the filtrate was evaporated to dryness and purified by column chromatography (EA:Hex
= 1:10) giving yield to compound 8a as orange needles (232 mg, 0.84 mmol, 60%). 1H NMR (400 MHz, DMSO-d6): δ [ppm] =
7.89 (tt, J = 7.7, 2.0 Hz, 4H), 7.69 – 7.57 (m, 5H), 4.80 (s, 2H).13C NMR (101 MHz, DMSO-d6): δ [ppm] = 151.9, 151.5, 141.5,
131.7, 130.5, 129.5, 122.9, 122.6, 33.6. ESI-MS: m/z calc. for C13H11BrN2+ [M+H]+: 275.0, 277.0, found: 274.9, 276.9.
1-(4-(Bromomethyl)-2,6-difluorophenyl)-2-(2,6-difluorophenyl)diazene (8b). To a solution of 1,2-bis(2,6-difluoro-4-
methylphenyl)diazene 5 (250 mg, 0.93 mmol) in 20 mL of CCl4 was added NBS (0.332 g, 1.86 mmol) and AIBN (11 mg, 0.076
mmol). The resultant solution was stirred overnight at 80 °C. After evaporation of the solvent, the product was purified by
column chromatography over silica gel (DCM:PE,1:4, silica gel) yielding the desired product 8b (163 mg, 0.47 mmol, 50%).
1H-NMR (CDCl3, 400 MHz): δ [ppm] = 7.38 (t, J = 8.5 Hz, 4H), 7.21 (d, J = 8.5 Hz, 11H), 7.13 – 7.03 (m, 14H), 6.87 (dt, J = 14.8,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
7.3 Hz, 42H), 6.65 (d, J = 8.8 Hz, 4H), 4.44 (s, 6H), 4.33 (s, 19H), 2.41 (s, 2H), 2.29 (s, 6H). ESI-MS: m/z calc. for C13H8BrF4N2
[M+H]+: 347.0, 349.0 found: 347.0, 349.0.
4-((4,5-Dihydroisoxazol-3-yl)oxy)-N,N-dimethyl-N-(4-(phenyldiazenyl)benzyl)but-2-yn-1-aminium (9a). 1-(4-
(bromomethyl)phenyl)-2-phenyldiazene 8a (100 mg, 0.36 mmol, 1 equiv.) and 4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-
dimethylbut-2-yn-1-amine 16 (132 mg, 0.73 mmol, 2 equiv.) were dissolved in acetonitrile (10 mL) and stirred at room
temperature for 12 h. Reaction was monitored by LC-MS. The precipitate was filtered off and the solvent was removed under
reduced pressure. The crude product was purified by column chromatography (DCM:MA:NH3 = 20:1:0.1) giving yield to
1
compound 9a as a dark red oil (54 mg, 0.14 mmol, 39%). H NMR (400 MHz, CDCl3): δ [ppm] = 7.99 – 7.86 (m, 6H), 7.56 –
7.46 (m, 3H), 5.24 (s, 2H), 4.88 (d, J = 7.9 Hz, 4H), 4.42 (t, J = 9.6 Hz, 2H), 3.53 – 3.40 (m, 6H), 3.02 (t, J = 9.6 Hz, 2H). 13C NMR
(101 MHz, CDCl3): δ [ppm] = 166.9, 154.1, 152.6, 134.3, 131.9, 129.3, 129.0, 123.7, 123.3, 87.8, 75.9, 70.2, 66.2, 57.5, 54.8,
50.0, 33.1. ESI-MS: m/z calc. for C22H25BrN4O2+ [M+H]+: 377.2, found: 377.2.
N-(4-((2,6-Difluorophenyl)diazenyl)-3,5-difluorobenzyl)-4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-
yn-1-aminium bromide (9b). 1-(4-(bromomethyl)-2,6-difluorophenyl)-2-(2,6-difluorophenyl)diazene 8b (64 mg, 0.184
mmol) and 4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-amine (34 mg, 0.184 mmol, 1 equiv.) in ethyl acetate
(10 mL) were charged in a sealed reaction vessel and stirred for 4 h. The precipitate is collected and washed several times
with ethyl acetate to afford the desired product 9b as a bright yellow powder (90 mg, 0.170 mmol, 92%). 1H-NMR (MeOD,
400 MHz): δ [ppm] = 7.63 – 7.30 (m, 3H), 7.12 (dt, J = 17.0, 8.7 Hz, 2H), 4.97 (d, J = 14.5 Hz, 2H), 4.66 (d, J = 49.0 Hz, 2H),
4.48 – 4.28 (m, 4H), 3.21 (d, J = 30.4 Hz, 6H), 3.05 (dt, J = 15.1, 9.6 Hz, 2H). 13C-NMR (MeOD, 400 MHz): δ [ppm] = 168.82,
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 24
1
2
3
4
5
154.23 (d, J = 5.1 Hz), 151.71 (d, J = 5.5 Hz), 132.40 (t, J = 9.7 Hz), 131.22 (d, J = 9.0 Hz), 118.4, 118.1, 113.5 (d, J = 19.1 Hz), 89.1,
76.7, 71.2, 66.0, 58.3, 55.3, 51.1, 33.6. ESI-MS: m/z calc. for C22H21F4N4O2+ [M]+: 449.16, found: 449.05.
6
7
8
9
1-Methyl-4-nitrosobenzene (11). p-Toluidine 10 (2.00 g, 18.7 mmol) was dissolved in DCM (60 mL), a solution of Oxone®
(11.5 g, 18.7 mmol) in water (60 mL) was added. The resulting biphasic mixture was stirred at room temperature for 30
min. The organic layer was separated, and the aqueous layer was extracted twice with DCM. The combined organic layers
were washed with 1 M aqueous hydrochlorid acid, saturated sodium hydrogen carbonate, and brine, and dried over
anhydrous sodium sulfate. The mixture 11 was concentrated to 10 – 15 ml volume, which is further used without
purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4,4'-Dimethylazobenzene (13a). 1-Methyl-4-nitrososbenzene 11 and p-toluidine 10 (1.00 g, 9.33 mmol) were dissolved
in glacial acetic acid (20 mL) and stirred overnight. The solution was diluted with water and extracted with ethyl acetate.
The organic phase was washed four times with water and once with brine and dried over anhydrous sodium sulfate. The
crude product was purified by flash chromatography (pentane/diethyl ether, 9:1) to obtain the desired product 13a as an
orange crystalline solid (695 mg, 3.31 mmol, 36 %). 1H NMR (400 MHz, CDCl3): δ [ppm] = 7.82 (d, J = 8.3 Hz, 4H), 7.31 (d, J
= 8.6 Hz, 4H), 2.44 (s, 6H). 13C NMR (101 MHz, CDCl3): δ [ppm] = 151.0, 141.3, 129.9, 122.9, 21.6. ESI-MS: m/z calc. for
C14H14N2+ [M+H]+: 211.1, found: 211.0.
1,2-bis(2,6-difluoro-4-methylphenyl)diazene (13b). 2,6-difluoro-4-methylaniline 12 (397m g, 2.77 mmol) and a freshly
ground mixture of potassium permanganate (1.17 g, 4.21 mmol) and iron(II)sulphate heptahydrate (1.17 g, 7.40 mmol)
were dissolved in DCM (10 mL). The solution was refluxed overnight, filtered through celite, dried over anhydrous sodium
sulphate, filtered, and concentrated under reduced pressure. The crude residue was purified by column chromatography
(DCM/petrol ether 1:1) to give the desired product 13b (110 mg, 0.39 mmol, 29%). 1H-NMR (CDCl3, 400 MHz): δ [ppm] =
6.86 (d, J = 10.3 Hz, 4H, arom. trans), 6.65 (d, J = 8.3 Hz, 0.39H, arom. cis), 2.40 (s,6 H,-CH3, trans), 2.29 (s, 0.62H, -CH3, cis).
13C-NMR (CDCl3, 400 MHz): δ [ppm] = 156.8 (d, J = 5.0 Hz), 154.3 (d, J = 5.2 Hz), 143.2 (t, J = 10.3 Hz), 113.2, 113.0, 21.7.
4,4’-Bis(bromomethyl)azobenzene (14a). To a solution of 4,4'-dimethylazobenzene 13a (2.17 g, 10.3 mmol) in 20 mL of
carbon tetrachloride (40 mL) was added N-bromo succinimide (4.22 g, 23.7 mmol) and azobisisobutyronitrile (127 mg, 0.77
mmol). The resultant solution was stirred overnight at 70 °C, then filtered, and washed with chloroform, water and dried
1
under reduced pressure, which yielded 14a as orange crystalline powder ( 2.88 g, 7.81 mmol, 76%). H NMR (400 MHz,
CDCl3): δ [ppm] = 7.90 (d, J = 8.4 Hz, 4H), 7.54 (d, J = 8.5 Hz, 4H), 4.56 (s, 4H). 13C NMR (101 MHz, CDCl3): δ [ppm] = 152.3,
140.8, 129.9, 32.7. ESI-MS: m/z calc. for C12H13Br2N2+ [M+H]+: 368.9, found: 368.8.
1,2-bis(4-(bromomethyl)-2,6-difluorophenyl)diazene
(14b).
To
a
solution
of
1,2-bis(2,6-difluoro-4-
methylphenyl)diazene 13b (226 mg, 0.801 mmol) in 17 mL of CCl4 was added NBS (0.356 g, 2.02 mmol) and AIBN (10 mg,
0.060 mmol). The resultant solution was stirred overnight at 80 °C. After evaporation of the solvent, the product was
16
ACS Paragon Plus Environment

Page 17 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
recrystallized from methanol to afford the desired product 14b as red needles (86 mg, 0.195 mmol, 24%). 1H-NMR (CDCl3,
400 MHz): δ [ppm] = 7.11 (d, J = 9.1 Hz, 4H), 4.43 (s,4H). 13C-NMR (CDCl3, 400 MHz): δ [ppm] = 160.9 (d, J = 7.6 Hz), 158.8
(d, J = 6.7 Hz), 150.4 (t, J = 6.7 Hz), 125.2 (t, J = 27.2 Hz), 114.0 (dd, J = 26.7, 3.7 Hz), 30.6 (t, J = 3.7 Hz). ESI-MS: m/z calc. for
C14H9Br2F4N2+ [M]+: 440.9, found: 440.7.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4-((4,5-Dihydroisoxazol-3-yl)oxy)-N-(4-((4-(((4-((4,5-dihydroisoxazol-3-yl)oxy)but-2-yn-1-yl)dimethyl-l4-
azaneyl)methyl)phenyl)diazenyl)benzyl)-N,N-dimethylbut-2-yn-1-aminium
bromide
(15a).
4,4’-
Bis(bromomethyl)azobenzene 14a (100 mg, 0.272 mmol) and 4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-
amine 16 were dissolved in 30 mL ethyl acetate and stirred at 60°C for 16 h. The precipitate was filtered and washed with
1
cold ethyl acetate to afford the desired product 15a as a yellow powder (166 mg, 0.227 mmol, 83%). H NMR (400 MHz,
DMSO-d6): δ [ppm] = 8.03 (d, J = 8.4 Hz, 1H), 7.83 (d, J = 8.4 Hz, 1H), 4.99 (s, 1H), 4.75 (s, 1H), 4.47 (s, 1H), 4.33 (t, J = 9.6 Hz,
1H), 3.13 (s, 3H), 3.05 (t, J = 9.6 Hz, 1H).13C NMR (101 MHz, DMSO-d6): δ [ppm] = 166.8, 152.8, 134.2, 131.1, 123.0, 86.8,
76.3, 69.6, 64.9, 57.3, 53.3, 49.5, 32.3. ESI-MS: m/z calc. for C32H40N6O42+ [M]2+: 286.2, found: 286.1.
N,N'-((Diazene-1,2-diylbis(3,5-difluoro-4,1-phenylene))bis(methylene))bis(4-((4,5-dihydroisoxazol-3-yl)oxy)-
N,N-dimethylbut-2-yn-1-aminium) bromide (15b). 4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-amine
16 (24.5 mg, 0.134 mmol) in ethyl acetate (40 mL) was slowly added to a stirred solution of 1,2-bis(4-(bromomethyl)-2,6-
difluorophenyl)diazene 14b (118 mg, 0.269 mmol, 2 equiv.) in ethyl acetate (10 mL) at 60°C. The reaction mixture was
stirred for 18 h. The reaction mixture was evaporated to dryness and purified by reverse phase column chromatography
(acetonitrile/water) yielding the desired product 15b as a bright orange powder (35 mg, 0.043 mmol, 32%). 1H-NMR
(MeOD, 400 MHz): δ [ppm] = 7.55 (dd, J = 60.3, 8.6 Hz, 4H), 5.03 (dd, J = 7.6, 6.1 Hz, 4H), 4.77 (d, J = 38.9 Hz, 4H), 4.51 (s,
2H), 4.44 (s, 2H), 3.29 (d, J = 24.8 Hz, 12H), 3.09 (dt, J = 19.3, 9.7 Hz, 4H). 13C-NMR (MeOD, 400 MHz): δ [ppm] = 176.6,
169.90, 168.9 (d, J = 2.4 Hz), 119.0, 118.4, 89.1, 76.7 (d, J = 7.4 Hz), 71.2, 58.4, 55.4 (d, J = 11.4 Hz), 51.2, 33.7. ESI-MS: m/z
calc. for C32H36F4N6O42+ [M]2+: 322.14, found: 322.25.
General procedure for the synthesis of the homobivalent quaternary iperoxo dimers 18a-d. To a solution of 2 equiv.
of 4-((4,5-dihydroisoxazol-3-yl)oxy)-N,N-dimethylbut-2-yn-1-amine 16 in 10 mL acetonitrile, 1 equiv. of the corresponding
bromoalkane 17-C4, 17-C6, 17-C8, and 17-C10 and a catalytic amount of KI/K2CO3 (1:1) were added. The reaction mixture
was heated in the microwave (500 W, 70 °C) for 2-3 h. After cooling to room temperature, the surplus of KI/K2CO3 was
filtered off and the solvent was evaporated to half of the volume. The solution was kept in the fridge overnight. The
precipitate obtained was filtered, washed with Et2O and dried in vacuo.
N1,N4-Bis(4-((4,5-dihydroisoxazol-3-yl)oxy)but-2-yn-1-yl)-N1,N1,N4,N4-tetramethylbutane-1,4-diaminium
bromide (18a). Light, yellow solid; yield 69%. 1H NMR (400 MHz, DMSO-d6): δ [ppm] = 2.73-2.77 (m, 4H), 3.03 (t, 4H, J =
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 24
1
2
3
4
5
6
7
8
9
9.6 Hz), 3.12 (s, 12H), 3.44-3.48 (m, 4H), 4.33 (t, 4H, J = 9.6 Hz), 4.53 (s, 4H), 4.95 (s, 4H). 13C NMR (101 MHz, DMSO-d6): δ
[ppm] = 19.1, 32.2, 49.7, 53.6, 57.2, 62.2, 69.6, 76.0, 86.1, 166.7. ESI-MS: m/z calc. for C22H36N4O42+ [M]2+: 210.1 Found: 210.1
N1,N6-Bis(4-((4,5-dihydroisoxazol-3-yl)oxy)but-2-yn-1-yl)-N1,N1,N6,N6-tetramethylhexane-1,6-diaminium
bromide (18b). Light, yellow solid; yield 49%.1H NMR (400 MHz, DMSO-d6): δ [ppm] = 1.32-1.37 (m, 4H), 1.69-1.78 (m,
4H), 3.03 (t, 4H, J = 9.6 Hz), 3.11 (s, 12H), 3.36-3.40 (m, 4H), 4.33 (t, 4H, J = 9.6 Hz), 4,50 (s, 4H), 4.94 (s, 4H). 13C NMR (101
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2+
MHz, DMSO-d6): δ [ppm] = 21.6, 25.1, 32.2, 49.7, 53.3, 57.1, 62.9, 69.5, 76.1, 85.9, 166.6. ESI-MS: m/z calc. for C24H40N4O4
[M]2+: 224.2 Found: 224.1.
N1,N8-Bis(4-((4,5-dihydroisoxazol-3-yl)oxy)but-2-yn-1-yl)-N1,N1,N8,N8-tetramethyloctane-1,8-diaminium
bromide (18c). Light, yellow solid; yield 15%.1H NMR (400 MHz, DMSO-d6): δ [ppm] = 1.29-1.32 (m, 8H), 1.64-1.72 (m, 4H),
3.02 (t, 4H, J = 9.6 Hz), 3.10 (s, 12H), 3.39-3.41 (m, 4H), 4.32 (t, 4H, J = 9.6 Hz), 4,51 (s, 4H), 4.93 (s, 4H). 13C NMR (101 MHz,
2+
DMSO-d6): δ [ppm] = 21.7, 25.5, 28.1, 32.2, 49.7, 53.2, 57.2, 63.0, 69.5, 76.1, 85.9, 166.7. ESI-MS: m/z calc. for C26H44N4O4
[M]2+: 238.2 Found: 238.1.
N1,N10-Bis(4-((4,5-dihydroisoxazol-3-yl)oxy)but-2-yn-1-yl)-N1,N1,N10,N10-tetramethyldecane-1,10-diaminium
bromide (18d). Light, yellow solid; yield 36%.1H NMR (400 MHz, DMSO-d6): δ [ppm] = 1.30 (s, 12H), 1.62-1.72 (m, 4H),
3.02 (t, 4H, J = 9.2 Hz), 3.09 (s, 12H), 3.36-3.38 (m, 4H), 4.32 (t, 4H, J = 9.2 Hz), 4,51 (s, 4H), 4.93 (s, 4H). 13C NMR (101 MHz,
DMSO-d6): δ [ppm] = 21.7, 25.6, 28.3, 28.6, 32.2, 49.7, 53.2, 57.1, 63.0, 69.5, 76.1, 85.8, 166.6. ESI-MS: m/z calc. for
C28H48N4O42+ [M]2+: 252.2 Found: 252.2.
Photochemical characterization. UV/Vis spectra and experiments were recorded on a Varian Cary 50 Bio UV/Vis
Spectrophotometer using Hellma (Type 100-QS) cuvettes (10 mm light path). Data were plotted using GraphPad Prism 5.0.
For irradiation, high performance LEDs (Mouser Electronics Inc. or Hartenstein) were used as light source. A concentration
of 50 M was prepared for each compound and measured in its dark-adapted state. Next, the probe is illuminated with LEDs
of different wavelengths (254, 365, 380, 400, 430, 500 nm and white light) while gradually increasing the irradiation time
until no change in the spectrum was detectable. Stability measurements were performed irradiating the probe with the light
source, which provided the highest photoconversion to the cis-isomer and kept in dark for at least 120 minutes. During this
time the absorbance at max was recorded every 5 minutes. Lastly, the probe is irradiated with the light source, which
provided the highest photoconversion to the trans-isomer. Photostationary distributions were determined by HPLC analysis
using a 50 M probe in physiological buffer. Absorption was measured at the respective isosbestic point wavelengths.
Analysis was carried out with GraphPad Prism software (GraphPad Software Inc., San Diego CA, www.graphpad.com).
Pharmacology. Cell culture. All experiments were performed with HEK 293T cells stably expressing the novel split
luciferase receptor sensor. Cells were incubated at 37 °C with 5% CO2 and cultivated in DMED with 4500 mg/l glucose, 10%
18
ACS Paragon Plus Environment

Page 19 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
(v/v) FCS, 100 µg/mL penicillin, 100 µg/mL streptomycin sulfate and 2 mM L-glutamine and 600 µg/mL G-418. Every two
to three days the cell lines were routinely passaged.
6
7
8
9
Split Luciferase complementation assay. The assay was performed as described previously,24 except for the following
modifications: A Berthold Mithras LB 940 plate reader was used to quantify the luminescence emitted by the cells, using
white, flat bottomed nunc™ f96 microwell™ polystyrene plates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular Modeling. In brief, all receptor-ligand docking experiments reported in this study were carried out with the
CCDCs software GOLD version 5.1.30 The active M1 receptor homology model used as protein was previously reported.27 All
residues of the extracellular domains and the receptor core region were defined as potential binding pocket. Default settings
were chosen for receptor-ligand docking and GoldScore was used as primary scoring function. All obtained docking poses
and receptor-ligand interactions were analyzed with LigandScout 4.2 using 3D-pharmacophores.31
Binding experiments. [3H]Quinuclidinyl benzilate ([3H]QNB), ethanol solution, was purchased from Amersham
Biosciences (catalog number TRK 604, 42 Ci/mmol, 1 mCi/ml). The experiments were carried out in 10 mM Tris-HCl buffer,
pH 7.0, containing 6 mM MgCl2. Rat brain cortices were used as a source of muscarinic receptors in the assay. After cleaning
the meninges with buffer-soaked filter paper, cortices were dissected, and white matter was carefully trimmed off. This
tissue was homogenized in 40 mL buffer using a Potter homogenizer with a motor-driven Teflon pestle. The homogenate
was centrifuged for 30 min at 50,000 X g, and the resulting pellet was homogenized and centrifuged again. After protein
determination by the Bradford assay, the final pellet was re-suspended at 1 mg of protein/ml, transferred to 1.5-ml
microcentrifuge tubes, and centrifuged once more. After discarding the supernatant, membrane pellets were kept at -80 °C
until use.23 The general procedure consists of the incubation of 20 g membrane protein with 200 pM [3H]QNB (200 pM),
in a total volume of 2 mL and in the presence of varying concentrations of competing compounds. An excess (2 μM) of the
unlabeled muscarinic antagonist atropine was used to define nonspecific binding. After 45 min at 25 oC to reach equilibrium
the reaction mixtures were quickly filtered through glass fiber disks using a semi-automated Brandel harvester allowing the
simultaneous filtration of 24 samples.21 Filters were washed twice with 4 mL of ice-cold Tris-HCl buffer (10mM, pH 7.0),
and counted for radioactivity. The final results were reported as % binding of [3H]QNB in each condition. This analysis was
carried out with GraphPad Prism software (GraphPad Software Inc., San Diego CA, www.graphpad.com).
ASSOCIATED CONTENT
Supporting Information.
UV/VIS spectra, NMR data, HPLC traces, molecular modelling, binding experiments.
Molecular formula strings.
This material is available free of charge via the Internet at http://pubs.acs.org.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 24
1
2
3
Accession codes
4
5
6
7
8
9
PDB code for the crystal structure of the human M2 receptor in complex with agonist iperoxo is 4MQS. This crystal structure
was used for the generation of the homology model for computational studies of the human M1 receptor in complex with
compound 9a, 9b, 15a, 15b, and 18a-d.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Author
*Email: michael.decker@uni-wuerzurg.de
Notes
The authors declare no competing financial interests
ACKNOWLEDGMENT
L.A. and T.L. were supported by the International Doctoral Program “Receptor Dynamics: Emerging Paradigms for
Novel Drugs” funded within the framework of the Elite Network of Bavaria (ENB). We kindly thank Professor Klaus
Mohr from the Institute of Pharmacy and Toxicology at Bonn University for fruitful discussions.
REFERENCES
(1) a) Lerch, M. M.; Hansen, M. J.; van Dam, G. M.; Szymanski, W.; Feringa, B. L. Emerging Targets in Photopharmacology. Angew.
Chem. Int. Ed. 2016, 55, 10978-10999; b) Velema, W. A.; Szymanski, W.; Feringa, B. L. Photopharmacology: Beyond Proof of
Principle. J. Am. Chem. Soc. 2014, 136, 2178-2191; c) Hüll, K.; Morstein, J.; Trauner, D. In Vivo Photopharmacology. Chem. Rev. 2018,
118, 10710-10747; d) Broichhagen, J.; Frank, J. A.; Trauner, D. A Roadmap to Success in Photopharmacology. Acc. Chem. Res. 2015,
48, 1947-1960.
(2) a) Pittolo, S.; Gómez-Santacana, X.; Eckelt, K.; Rovira, X.; Dalton, J.; Goudet, C.; Pin, J.-P.; Llobet, A.; Giraldo, J.; Llebaria, A.;
Gorostiza, P. An Allosteric Modulator to Control Endogenous G protein-coupled Receptors with Light. Nat. Chem. Biol. 2014, 10,
813-815; b) Schönberger, M.; Trauner, D. A Photochromic Agonist for μ-Opioid Receptors. Angew. Chem. Int. Ed. 2014, 53, 3264-
3267; c) Agnetta, L.; Kauk, M.; Canizal, M. C. A.; Messerer, R.; Holzgrabe, U.; Hoffmann, C.; Decker, M. A Photoswitchable Dualsteric
Ligand Controlling Receptor Efficacy. Angew. Chem. Int. Ed. 2017, 56, 7282-7287; d) Bahamonde, M. I.; Taura, J.; Paoletta, S.; Gakh,
A. A.; Chakraborty, S.; Hernando, J.; Fernández-Dueñas, V.; Jacobson, K. A.; Gorostiza, P.; Ciruela, F. Photomodulation of G Protein-
Coupled Adenosine Receptors by a Novel Light-Switchable Ligand. Bioconjug. Chem. 2014, 25, 1847-1854; e) Broichhagen, J.;
Johnston, N. R.; von Ohlen, Y.; Meyer-Berg, H.; Jones, B. J.; Bloom, S. R.; Rutter, G. A.; Trauner, D.; Hodson, D. J. Allosteric Optical
Control of a Class B G-Protein-Coupled Receptor. Angew. Chem. Int. Ed. 2016, 55, 5865-5868; f) Hauwert, N. J.; Mocking, T. A. M.; Da
20
ACS Paragon Plus Environment

Page 21 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Costa Pereira, D.; Kooistra, A. J.; Wijnen, L. M.; Vreeker, G. C. M.; Verweij, E. W. E.; De Boer, A. H.; Smit, M. J.; De Graaf, C.; Vischer, H.
F.; de Esch, I. J. P.; Wijtmans, M.; Leurs, R. Synthesis and Characterization of a Bidirectional Photoswitchable Antagonist Toolbox for
Real-Time GPCR Photopharmacology. J. Am. Chem. Soc. 2018, 140, 4232-4243; g) Westphal, M. V.; Schafroth, M. A.; Sarott, R. C.;
Imhof, M. A.; Bold, C. P.; Leippe, P.; Dhopeshwarkar, A.; Grandner, J. M.; Katritch, V.; Mackie, K.; Trauner, D.; Carreira, E. M.; Frank, J.
A. Synthesis of Photoswitchable Δ9-Tetrahydrocannabinol Derivatives Enables Optical Control of Cannabinoid Receptor 1
Signaling. J. Am. Chem. Soc. 2017, 139, 18206-18212; h) Dolles, D.; Strasser, A.; Wittmann, H.-J.; Marinelli, O.; Nabissi, M.; Pertwee,
R. G.; Decker, M. The First Photochromic Affinity Switch for the Human Cannabinoid Receptor 2. Adv. Therap. 2018, 1, 1700032.
(3) Lohse, M. J.; Hofmann, K. P. Spatial and Temporal Aspects of Signaling by G-Protein-Coupled Receptors. Mol. Pharmacol. 2015,
88, 572-578.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(4) Burger, W. A. C.; Sexton, P. M.; Christopoulos, A.; Thal, D. M. Toward an Understanding of the Structural Basis of Allostery in
Muscarinic Acetylcholine Receptors. J. Gen. Physiol. 2018, 150, 1360-1372.
(5) a) Kruse, A. C.; Kobilka, B. K.; Gautam, D.; Sexton, P. M.; Christopoulos, A.; Wess, J. Muscarinic Acetylcholine Receptors: Novel
Opportunities for Drug Development. Nat. Rev. Drug Discov. 2014, 13, 549-560; b) Svoboda, J.; Popelikova, A.; Stuchlik, A. Drugs
Interfering with Muscarinic Acetylcholine Receptors and Their Effects on Place Navigation. Front. Psychiatry 2017, 8, Article 215;
c) van Koppen, C. J.; Kaiser, B. Regulation of Muscarinic Acetylcholine Receptor Signaling. Pharmacol. Ther. 2003, 98, 197-220; d)
Eglen, R. M. Muscarinic Receptor Subtypes in Neuronal and Non-neuronal Cholinergic Function. Auton. Autacoid. Pharmacol. 2006,
26, 219-233.
(6) Wess, J.; Eglen, R. M.; Gautam, D. Muscarinic Acetylcholine Receptors: Mutant Mice Provide New Insights for Drug
Development. Nat. Rev. Drug Discov. 2007, 6, 721-733.
(7) Caulfield, M. P. Muscarinic Receptors—Characterization, Coupling and Function. Pharmacol. Ther. 1993, 58, 319-379.
(8) a) Antony, J.; Kellershohn, K.; Mohr-Andrä, M.; Kebig, A.; Prilla, S.; Muth, M.; Heller, E.; Disingrini, T.; Dallanoce, C.; Bertoni, S.;
Schrobang, J.; Tränkle, C.; Kostenis, E.; Christopoulos, A.; Höltje, H. D.; Barocelli, E.; De Amici, M.; Holzgrabe, U.; Mohr, K. Dualsteric
GPCR Targeting: A Novel Route to Binding and Signaling Pathway Selectivity. FASEB J. 2009, 23, 442-450; b) Disingrini, T.; Muth,
M.; Dallanoce, C.; Barocelli, E.; Bertoni, S.; Kellershohn, K.; Mohr, K.; De Amici, M.; Holzgrabe, U. Design, Synthesis, and Action of
Oxotremorine-Related Hybrid-Type Allosteric Modulators of Muscarinic Acetylcholine Receptors. J. Med. Chem. 2006, 49, 366-372.
(9) a) Beharry, A. A.; Sadovski, O.; Woolley, G. A. Azobenzene Photoswitching without Ultraviolet Light. J. Am. Chem. Soc. 2011,
133, 19684-19687; b) Samanta, S.; Beharry, A. A.; Sadovski, O.; McCormick, T. M.; Babalhavaeji, A.; Tropepe, V.; Woolley, G. A.
Photoswitching Azo Compounds In Vivo with Red Light. J. Am. Chem. Soc. 2013, 135, 9777-84.
(10) a) Bleger, D.; Hecht, S. Visible-Light-Activated Molecular Switches. Angew. Chem. Int. Ed. 2015, 54, 11338-11349; b) Bléger,
D.; Schwarz, J.; Brouwer, A. M.; Hecht, S. o-Fluoroazobenzenes as Readily Synthesized Photoswitches Offering Nearly Quantitative
Two-Way Isomerization with Visible Light. J. Am. Chem. Soc. 2012, 134, 20597-20600.
(11) Dong, M.; Babalhavaeji, A.; Samanta, S.; Beharry, A. A.; Woolley, G. A. Red-Shifting Azobenzene Photoswitches for in Vivo Use.
Acc. Chem. Res. 2015, 48, 2662-2670.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 24
1
2
3
4
5
6
7
8
9
(12) a) Rullo, A.; Reiner, A.; Reiter, A.; Trauner, D.; Isacoff, E. Y.; Woolley, G. A. Long Wavelength Optical Control of Glutamate
Receptor Ion Channels Using a Tetra-ortho-substituted Azobenzene Derivative. Chem. Commun. (Camb.) 2014, 50, 14613-14615;
b) Wegener, M.; Hansen, M. J.; Driessen, A. J. M.; Szymanski, W.; Feringa, B. L. Photocontrol of Antibacterial Activity: Shifting from
UV to Red Light Activation. J. Am. Chem. Soc. 2017, 139, 17979-17986; c) Konrad, D. B.; Frank, J. A.; Trauner, D. Synthesis of
Redshifted Azobenzene Photoswitches by Late-Stage Functionalization. Chemistry 2016, 22, 4364-4368.
(13) Shonberg, J.; Scammells, P. J.; Capuano, B. Design Strategies for Bivalent Ligands Targeting GPCRs. ChemMedChem 2011, 6,
963-974.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(14) a) Mohr, K.; Schmitz, J.; Schrage, R.; Tränkle, C.; Holzgrabe, U. Molecular Alliance-From Orthosteric and Allosteric Ligands to
Dualsteric/Bitopic Agonists at G Protein Coupled Receptors. Angew. Chem. Int. Ed. 2013, 52, 508-516; b) Halazy, S. G Protein-
coupled Receptors Bivalent Ligands and Drug Design. Expert Opin. Ther. Pat. 2005, 9, 431-446.
(15) Nimczick, M.; Pemp, D.; Darras, F. H.; Chen, X.; Heilmann, J.; Decker, M. Synthesis and Biological Evaluation of Bivalent
Cannabinoid Receptor Ligands Based on hCB(2)R Selective Benzimidazoles Reveal Unexpected Intrinsic Properties. Bioorg. Med.
Chem. 2014, 22, 3938-3946.
(16) Decker, M.; Si, Y. G.; Knapp, B. I.; Bidlack, J. M.; Neumeyer, J. L. Synthesis and Opioid receptor Binding Affinities of 2-
substituted and 3-Aminomorphinans: Ligands for Mu, Kappa, and Delta Opioid Receptors. J. Med. Chem. 2010, 53, 402-418.
(17) Kühhorn, J.; Hübner, H.; Gmeiner, P. Bivalent Dopamine D2 Receptor Ligands: Synthesis and Binding Properties. J. Med.
Chem. 2011, 54, 4896-4903.
(18) a) Piergentili, A.; Quaglia, W.; Tayebati, S. K.; Paparelli, F.; Malmusi, L.; Brasili, L. Synthesis and Muscarinic Receptors affinity
of a Series of Antagonist Bivalent Ligands. Farmaco 1994, 49, 83-87; b) Messer, J. W. Bivalent Ligands for G Protein-Coupled
Receptors. Curr. Pharm. Des. 2004, 10, 2015-2020.
(19) Kloeckner, J.; Schmitz, J.; Holzgrabe, U. Convergent, Short Synthesis of the Muscarinic Superagonist Iperoxo. Tetrahedron
Lett. 2010, 51, 3470-3472.
(20) Knie, C.; Utecht, M.; Zhao, F.; Kulla, H.; Kovalenko, S.; Brouwer, A. M.; Saalfrank, P.; Hecht, S.; Bléger, D. Ortho-
Fluoroazobenzenes: Visible Light Switches with Very Long-Lived Z Isomers. Chem. - Eur. J. 2014, 20, 16492-16501.
(21) Claro, E. Analyzing Ligand Depletion in a Saturation Equilibrium Binding Experiment. Biochem. Mol. Biol. Educ. 2006, 34,
428-431.
(22) a) Yamamura, H. I.; Snyder, S. H. Muscarinic Cholinergic Binding in Rat Brain. Proc. Natl. Acad. Sci. U. S. A. 1974, 71, 1725-
1729; b) Yamamura, H. I.; Kuhar, M. J.; Snyder, S. H. In Vivo Identification of Muscarinic Cholinergic Receptor Binding in Rat brain.
Brain Res. 1974, 80, 170-176.
(23) Salles, J.; Wallace, M. A.; Fain, J. N. Differential Effects of Alkylating Agents on the Multiple Muscarinic Receptor Subtypes
Linked to Activation of Phospholipase C by Carbachol in Rat Brain Cortical Membranes. J. Pharmacol. Exp. Ther. 1993, 264, 521-529.
(24) Littmann, T.; Ozawa, T.; Hoffmann, C.; Buschauer, A.; Bernhardt, G. A Split Luciferase-based Probe for Quantitative Proximal
Determination of Gαq Signalling in Live Cells. Sci. Rep. 2018, 8, Article 17179.
22
ACS Paragon Plus Environment

Page 23 of 24
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(25) Hattori, M.; Ozawa, T. Split Luciferase Complementation for Analysis of Intracellular Signaling. Anal. Sci. 2014, 30, 539-544.
(26) Chen, X.; Klöckner, J.; Holze, J.; Zimmermann, C.; Seemann, W. K.; Schrage, R.; Bock, A.; Mohr, K.; Tränkle, C.; Holzgrabe, U.;
Decker, M. Rational Design of Partial Agonists for the Muscarinic M1 Acetylcholine Receptor. J. Med. Chem. 2015, 58, 560-576.
(27) Bermudez, M.; Rakers, C.; Wolber, G. Structural Characteristics of the Allosteric Binding Site Represent a Key to Subtype
Selective Modulators of Muscarinic Acetylcholine Receptors. Mol. Inform. 2015, 34, 526-530.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(28) a) Bock, A.; Bermudez, M.; Krebs, F.; Matera, C.; Chirinda, B.; Sydow, D.; Dallanoce, C.; Holzgrabe, U.; De Amici, M.; Lohse, M.
J.; Wolber, G.; Mohr, K. Ligand Binding Ensembles Determine Graded Agonist Efficacies at a G Protein-coupled Receptor. J. Biol.
Chem. 2016, 291, 16375-16389; b) Bermudez, M.; Bock, A.; Krebs, F.; Holzgrabe, U.; Mohr, K.; Lohse, M. J.; Wolber, G. Ligand-Specific
Restriction of Extracellular Conformational Dynamics Constrains Signaling of the M2 Muscarinic Receptor. ACS Chem. Biol. 2017,
12, 1743-1748.
(29) Dror, R. O.; Green, H. F.; Valant, C.; Borhani, D. W.; Valcourt, J. R.; Pan, A. C.; Arlow, D. H.; Canals, M.; Lane, J. R.; Rahmani, R.;
Baell, J. B.; Sexton, P. M.; Christopoulos, A.; Shaw, D. E. Structural Basis for Modulation of a G-Protein-coupled Receptor by Allosteric
Drugs. Nature 2013, 503, 295-299.
(30) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. Development and Validation of a Genetic Algorithm for Flexible
Docking. J. Mol. Biol. 1997, 267, 727-748.
(31) Wolber, G.; Seidel, T.; Bendix, F.; Langer, T. Molecule-pharmacophore Superpositioning and Pattern Matching in
Computational Drug Design. Drug Discov. Today 2008, 13, 23-29.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 24
1
2
3
4
Table of Content Graphic:
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24
ACS Paragon Plus Environment