Phenolic Bis-styrylbenzo[ c]-1,2,5-thiadiazoles as Probes for Fluorescence Microscopy Mapping of Aβ Plaque Heterogeneity

Article abstract of DOI:10.1021/acs.jmedchem.8b01681

A fluorescent bis-styryl-benzothiadiazole (BTD) with carboxylic acid functional groups (X-34/Congo red analogue) showed lower binding affinity toward Aβ1-42 and Aβ1-40 fibrils than its neutral analogue. Hence, variable patterns of neutral OH-substituted bis-styryl-BTDs were generated. All bis-styryl-BTDs showed higher binding affinity to Aβ1-42 fibrils than to Aβ1-40 fibrils. The para-OH on the phenyl rings was beneficial for binding affinity while a meta-OH decreased the affinity. Differential staining of transgenic mouse Aβ amyloid plaque cores compared to peripheral coronas using neutral compared to anionic bis-styryl ligands indicate differential recognition of amyloid polymorphs. Hyperspectral imaging of transgenic mouse Aβ plaque stained with uncharged para-hydroxyl substituted bis-styryl-BTD implicated differences in binding site polarity of polymorphic amyloid plaque. Most properties of the corresponding bis-styryl-BTD were retained with a rigid alkyne linker rendering a probe insensitive to cis-trans isomerization. These new BTD-based ligands are promising probes for spectral imaging of different Aβ fibril polymorphs.

Full text of DOI:10.1021/acs.jmedchem.8b01681

Subscriber access provided by Miami University Libraries
Article
Phenolic bis-styrylbenzo[c]-1,2,5-thiadiazoles as probes for
fluorescence microscope mapping of A# plaque heterogeneity
Jun Zhang, Audun Konsmo, Alexander Sandberg, Xiongyu Wu, Sofie Nyström, Ulrike Obermueller,
Bettina M Wegenast-Braun, Peter Konradsson, Mikael Lindgren, and Per Hammarström
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01681 • Publication Date (Web): 01 Feb 2019
Downloaded from http://pubs.acs.org on February 2, 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 43
Journal of Medicinal Chemistry
1
2
3
4
5
Phenolic bis-styrylbenzo[c]-1,2,5-thiadiazoles as probes for
6
7
8
fluorescence microscope mapping of Aβ plaque heterogeneity
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
Jun Zhang,[a] Audun Konsmo,[b]Alexander Sandberg,[a] Xiongyu Wu,[a] Sofie Nyström,[a] Ulrike
Obermüller,[c,d] Bettina M. Wegenast-Braun,[c,d] Peter Konradsson,[a] Mikael Lindgren,[b] and Per
Hammarström,*[a]
[a]Division of Chemistry, Department of Physics Chemistry and Biology, Linköping University,
581 83 Linköping, Sweden
[b]Department of Physics, Norwegian University of Science and Technology, 7491 Trondheim,
Norway
[c]Department of Cellular Neurology, Hertie Institute for Clinical Brain Research, University of
Tübingen, 72076 Tübingen, Germany
[d]DZNE–German Center for Neurodegenerative Diseases, 72076 Tübingen, Germany
*Corresponding author: per.hammarstrom@liu.se
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
4
ABSTRACT
5
6
7
8
A fluorescent bis-styryl-benzothiadiazole (BTD) with carboxylic acid functional groups (X-
34/Congo red analogue) had lower binding affinity towards Aβ1-42 and Aβ1-40 fibrils than its
neutral analogue. Hence, variable patterns of neutral OH substituted bis-styryl-BTDs were
generated. All bis-styryl-BTDs showed higher binding affinity to Aβ1-42 fibrils than to Aβ1-40
fibrils. The para-OH on the phenyl rings was beneficial for binding affinity while a meta-OH
decreased the affinity. Differential staining of transgenic mouse Aβ amyloid plaque cores
compared to peripheral coronas using neutral compared to anionic bis-styryl ligands indicate
differential recognition of amyloid polymorphs. Hyperspectral imaging of transgenic mouse Aβ
plaque stained with uncharged para-hydroxyl substituted bis-styryl-BTD implicated differences in
binding site polarity of polymorphic amyloid plaque. Most properties of the corresponding bis-
styryl-BTD was retained with a rigid alkyne linker rendering a probe insensitive to cis-trans
isomerization. These new BTD-based ligands are promising probes for spectral imaging of
different Aβ fibril polymorphs.
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
2
ACS Paragon Plus Environment

Page 3 of 43
Journal of Medicinal Chemistry
1
2
3
4
INTRODUTION
5
6
7
8
Alzheimer’s disease (AD) is a common neurodegenerative disorder and a leading cause of
dementia among older adults.1-2 Extracellular senile plaques and intracellular neuro-fibrillary
tangles (NFTs), the main two pathological hall-marks of AD, are composed of fibrous amyloid-β
peptides (mainly Aβ1-40 and Aβ1-42) and abnormally hyperphosphorylated tau protein,
respectively.3-5 To date, various small molecular scaffolds targeting the regular cross β-pleated
sheet conformation of protein aggregates have been developed.6-9 In this regard, the pan-amyloid
targeting molecule X-34 [2,5-bis(4’-hydroxy-3’-carboxy-styryl)benzene, see Chart 1], is an
important scaffold for amyloid fibril probe development.10 Congo red, the gold standard for amy-
loid diagnostics, was used as starting scaffold for the design of X-34 by the Klunk group. Congo
red carries anionic sulfonic acid moieties as terminal groups and have been shown to direct specific
interactions with amyloid fibril binding sites.11 In contrast, uncharged analogues K114, DF-9 and
Methoxy-X04, which have been shown to penetrate the blood-brain barrier (BBB) still bind to Aβ
plaques deposited in APP or APP/PS1 mice,12-16 as well as to amyloid fibrils in human tissues.12
These results show that anionic functional groups are not necessary for good Aβ-fibril binding. As
research tools fluorescent amyloid dyes are indispensable. In particular Methoxy-X04 has been
extensively utilized for quantitative 2-photon fluorescence in vivo imaging of amyloid plaque
formation in APP transgenic mice.17-18
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
π-extended 2,1,3-benzothiadiazole (BTD) derivatives are a new class of bio-probes with
attractive photophysical properties: high molar extinction coefficients, large stokes shifts, high
quantum yields, high storage stability, bright emission, good signal-to-noise ratios, efficiency to
cross the cell membranes and low fading after long periods of irradiation.19 Although a large
number of fluorescent small molecule BTD derivatives were applied to bioimaging analyses of
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
several cell types, little is known about this class of bio-probes used to stain Aβ plaques and NFTs.
Nilsson and coworkers reported on L3 (Chart 1), which can intensely stain Aβ plaques,20 and HS-
169 (Figure 1), intensely stain Aβ plaques and NFTs.21 In addition, we recently reported that
BTDSB, a derivative of salicylate bis-styrylbenzene X-34 (Chart 1) displayed emission in the red
region (620 nm) with a long Stokes shift (133 nm). It was also shown to have high binding affinity
(Kd: 93 ± 5 nm) when bound to recombinant Aβ1-42 fibrils and intensely stained both Aβ plaques
and NFTs in AD brain tissue sections.22
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 a starting point for our studies we compared the differential staining of individual plaque
using uncharged DF-9, in comparison with the analogous anionic X-34, known to target the Congo
red binding site. Interestingly, variations in plaque staining implicated differential targeting of
amyloid structures of these ligands. We then redesigned bis-styryl-BTD ligands with modification
of the outer phenyl ring using various hydroxyl motifs investigating Aβ fibril binding affinity (Aβ1-
42 and Aβ1-40) and specificity. The para-substituted bis-styryl-BTD showed superior affinity and
was particularly sensitive to binding site polarity, suggesting additional amyloid polymorph
sensitivity. A further redesign using an alkyne linker retained high affinity and polarity sensing
capacity while removing a potential drawback, i.e. sensitivity towards isomerization of the probe.
Chart 1. Chemical structures of amyloid ligands reported previously.
4
ACS Paragon Plus Environment

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
Comparison of bis-styryl-benzene carboxylate charge moieties for fibril targeting. Anionic
charges have been shown to be important for specific targeting the Congo red binding site of
amyloid fibrils. We took advantage of the different excitation and emission spectra of our
previously reported BTDSB probe and X-34 and DF-9 to compare amyloid binding sites. X-34 and
BTDSB bind to the same binding site 22 and as expected, we observe mainly co-localized staining
of these probes on sections of APP23 mice with amyloid plaque (Figure 1A). In contrast, co-
staining with DF-9 and BTDSB showed vastly different staining patterns with DF-9 bound to the
core of the plaque while BTDSB dominated the periphery (Figure 1B). This is interesting, because
it is known that aged APP23 mice (>18 months) show polymorphic amyloid structures within
individual neuritic plaque.23-24 Co-staining experiments on 5 μM recombinant Aβ1-40 fibrils
showed that in the presence of equimolar concentrations (0.5 μM each) of X-34 and BTDSB, the
latter still retained full binding (Figure 1C). DF-9 on the other hand prevented almost 50% of
BTDSB binding (Figure 1C). Hence together with the imaging data this proposes that two different
binding sites/polymorphs occur on fibrils with varying affinity to uncharged DF-9 in comparison
to the known Congo red-type charge mediated binding site targeted by BTDSB and X-34. The
DF-9 high affinity polymorph appears mainly in the plaque core because it out competes the
negatively charged Congo red analogue BTDSB in the core, but not in the peripheral corona. This
suggests that the DF-9 positive core would be the site preferably probed by the analogous neutral
trans-stilbene BTD-based ligands of the present study.
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
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 43
1
2
3
4
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
Figure 1. (A) Single plane confocal images of APP23 mouse brain sections (female, 19.5 months)
after co-staining with X-34 and BTDSB (top) and the merged image of X-34 and BTDSB (down).
(B) after co-staining with DF-9 and BTDSB (top) and the merged image of DF-9 and BTDSB
(down). Z-depth was selected for the center of the largest plaque. (C) Competition assays using
BTDSB (500 nM) to displace X-34 (or DF-9) in the presence of Aβ1-40 fibrils (5 μM) in vitro in
PBS (pH 7.4).
Synthesis of bis-styryl-benzothiadiazole ligands and in vitro ligand binding with Aβ1-
42/Aβ1-40 fibrils. Compounds 3a to 3j were synthesized via the classic Heck reaction using a new
one-pot procedure as depicted in Scheme 1. Compounds 2b to 2f were prepared via Witting
reaction of the corresponding aldehyde and triphenylphosphine methyl bromide utilizing a previous
procedure.25 Compound 2a was purchased from Sigma-Aldrich. Compound BTDSB was reported
previously by our group.22
6
ACS Paragon Plus Environment

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
Scheme 1. Synthesis of bis-styryl-benzothiadiazole ligands.
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
Conditions and reagents: (a) Pd(OAc)2, triethanolamine, DMF.
We first investigated the photophysical properties of DF-9 and the BTD analogue compound 3c
in solution as shown in Figure 2A. Compared with DF-9, 3c displayed two absorption bands: a
high energy band (blue-shift 35 nm) around 340 nm and a low energy band around 485 nm, likely
arising from the π-π* transition and charge-transfer transition, respectively, similar with that of
BTDSB in PBS buffer.22 DF-9 exhibited an emission spectrum with partially resolved vibronic
substructure and the maximum at around 440 nm, whereas 3c showed a broad featureless emission
with the peak at 660 nm. 3c has larger stokes shift (175 nm) than DF-9 (65 nm). We then generated
emission spectra of 3a-3j and BTDSB, DF-9 as a reference, in the presence or absence of
monomers and fibrils of Aβ1-42 and Aβ1-40 (Figure 2B and Figure S1 in supporting information).
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
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
Figure 2. (A) Normalized absorption (dashed) and emission (solid) spectra of DF-9 (blue) and 3c
(red) with 2 μM in DMSO. (B) Emission spectra of 3c in the absence or presence of Aβ1-42 and
Aβ1-40 peptides (monomers and fibrils) in PBS buffer. (C) Fluorescence intensity enhancements
of 3a to 3j and BTDSB as well as DF-9 as control when bound to Aβ1-42 or Aβ1-40 fibrils
compared to free in solution. Excitation wavelengths: 360 nm for DF-9, 430 nm for 4, 460 nm for
3a, 3h, 3f, 5, 6, and 490 nm for all other ligands.
Overall, we found that these ligands showed larger fluorescence intensities when bound to Aβ1-42
fibrils than when bound to Aβ1-40 fibrils (Figure 2C and Table S1). For several probes weak
fluorescence changes were also observed in the presence of Aβ1-42 and Aβ1-40 monomers in
comparison to free ligand (Figure 2B and Figure S1). More details of the binding properties and
photophysical characterization are summarized in Table S1.
Next, saturation binding assays were performed to quantitatively evaluate the binding affinity of
3a to 3j and BTDSB towards Aβ1-42 fibrils and Aβ1-40 fibrils, using DF-9 and X-34 as references
8
ACS Paragon Plus Environment

Page 9 of 43
Journal of Medicinal Chemistry
1
2
3
4
Table 1. Binding affinity (Kd) and Aβ plaque binding specificity (S/N ratio) to Aβ transgenic
5
6
mouse brain slices.
7
8
9
e
compd.
Kd (nM)a
Aβ1-42
137±10
S/N ratiob
plaque core plaque corona
λem (nm)c
core/corona
cLogPd
6.2
τavg (ns)
Aβ1-40
58±7
DF-9
X-34
BTDSB
3c
13.7 ± 4.3
31.7 ± 7.6
443/443
0.7
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
149±8
50±4
n.d.
n.d.
n.d.
8.2
8.1
6.0
6.0
5.4
6.7
4.8
7.3
5.3
5.4
4.7
6.0
0.8f
5.6
5.4
5.7
5.8
6.2
5.5
5.7
6.5
5.8
6.2
5.7
(299±21)g
83 ± 6
250
23.3 ± 2.6
27.5 ± 1.7
20.2 ± 0.6
5.2 ± 0.5
5.1 ± 0.9
9.3 ± 2.1
5.1 ± 0.7
11.3 ± 1.3
12.6 ± 1.3
42.6 ± 5.4
2.7 ± 0.5
37.4 ± 0.3
17.3 ± 0.9
4.5 ± 0.3
5.3 ± 0.3
9.3 ± 0.3
5.2 ± 1.4
11.5 ± 0.7
3.3 ± 0.8
7.7 ± 0.7
13.4 ± 2.9
11.1 ± 0.7
614/614
574/600
565/577
573/594
575/588
571/590
548/564
594/594
575/575
566/566
556/563
(93 ± 5)g
20 ± 2
29 ± 5
31 ± 4
39 ± 1
39 ± 4
66 ± 8
73 ± 7
77 ± 1
95 ± 1
116± 10
46 ± 2
3j
106 ± 22
110 ± 22
61 ± 17
187 ± 8
109 ± 19
170 ± 13
83 ± 20
284 ± 12
133 ± 10
3e
3b
3d
3a
3g
3i
3f
3h
aValues represent the the average ± SD of 2 determinations. bValues represent the average ± SD of
10 determinations. cEmission maximum of ligand bound to Aβ plaques (cores and coronas). Images
d
were captured at the center of z-depth focal plane of neuritic plaque cores. Calculated cLogP
e
values using ChemDraw Ultra 12.0 of the fully protonated molecules. Measured for solute
molecule in toluene. Measured for solute molecule in MeOH. from reference.22 n.d. = not
f
g
determined.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
(Table 1 and Figure S2). It is noteworthy that these ligands showed higher binding affinity towards
Aβ1-42 fibrils compared to Aβ1-40 fibrils, which is consistent with their higher fluorescence
enhancement with Aβ1-42 fibrils. Therefore, we sorted the new BTD compounds in Table 1
following the increase in Kd values towards the Aβ1-42 fibrils, and thus, with decreasing binding
affinity.
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
We note that 3c had the strongest binding affinity (20 ± 2 nM) of all ligands. The binding affinity
for 3a and 3b was 3.4-fold and 2-fold less compared to 3c, most likely due to the loss of H-bond
stabilization. The Kd data for 3c versus 3j and 3h suggest that the para- and ortho- are marginally
superior to meta-substitution on the outer phenyl rings. As evidenced by the 3.4-fold increase in
affinity between 3c and 3a, 2.3-fold loss of affinity for 3a compared to 3j, the para-OH on the
outer phenyl ring is beneficial to binding affinity while the meta-OH decreases the binding affinity,
supported by the Kd data for 3h versus 3i versus 3d. Although the meta-OH could decrease the
binding affinity, 3f showed higher binding affinity compared to 3h, attributed to the 2 more of H-
bond stabilization by the phenol functional group. As evidenced by the 4.3-fold loss in binding
affinity for BTDSB compared to 3c, the carboxylic acid functional groups decreased the affinity,
suggesting that acidic functional groups are not required for good Aβ binding, which is in line with
X-34 (299 ± 21 nM)22 versus DF-9 (137 ± 10 nM) for Aβ1-42 fibrils and comparable affinities for
Aβ1-40 fibrils.
Synthesis and photophysical properties of benzothiadiazole based amyloid ligands with
various conjugated spacers. A known issue with trans-stilbenes is the cis-trans isomerization of
the vinyl bond. We hence explored para-substituted mono-hydroxyl BTD variants with variable
conjugated spacers. The synthesis of 4 to 7 was performed by means of palladium-catalyzed
coupling reactions of 4,7-dibromo-2,1,3-benzothiadiazole (1) (Scheme 2) Compounds 12, 16, 18
10
ACS Paragon Plus Environment

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 19 were obtained using previously reported procedures.25-27 Compounds 4, 6 and 7, which
possessed distinct spacers such as thiophene and benzene, were obtained vis Suzuki reaction of 1
with arylboronic acid or arylboronic ester. The Sonogashira coupling reaction of 1 with
trimethylsilylacetylene (11) gave 4,7-diethynyl-2,1,3-benzothiadiazole (12), which was derived to
5 by reaction with 4-iodophenol (13).
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 2. Synthesis of BTD derivatives with different conjugated spacersa
a Conditions and reagents: (a) Pd(PPh3)4, 2 M Na2CO3, benzene-EtOH, 90 oC; (b) Pd(PPh3)4, CuI,
triethylamine, 75 oC; (c) 1) KOH in MeOH rt; 2) Pd(OAc)2, PPh3, CuI, triethylamine, THF, rt; (d)
PdCl2(dppf), dppf, KOAc, 1,4-dioxane, 90 oC; (e) NBS, CHCl3, rt.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
9
The photophysical properties of compounds 4 to 7 and 3c were measured in PBS buffer (pH
7.4). The normalized absorption and emission spectra are shown in Figure S3 and the results in
terms of spectroscopic parameters are summarized in Table S2. As expected, all ligands showed
two absorption bands: a high energy band and a low energy band, likely arising from the π-π*
transition and charge-transfer transition, respectively. Compounds 4, 5 and 6 displayed very similar
absorption spectra with the maximum wavelength at 402 nm, 404 nm and 400 nm, respectively,
indicating that C-C single bond, C≡C triple bond and a benzene ring as conjugated spacers gives
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
a similar electronic transition from S0 to S1 localized to the BTD group. Compounds 3c and 7 have
their absorption maxima at 432 and 486 nm, respectively, demonstrating that the transition is
delocalized to the thiophene ring more efficiently than e.g., the C=C double bond. All ligands
showed very weak emission in PBS buffer with emission maxima ranging from 545 nm for 6 up to
660 nm for 7 because of the strong intermolecular interactions of electric dipole and π-π stacking
propensity.28 It is worth noting that all BTD based ligands show longer emission wavelengths (545-
670 nm) than that of DF-9 (440 nm) and have large stokes shift in solution, 146-202 nm, allowing
unambiguous detection without reabsorption effects and not interfering with the background
fluorescence of biomolecules at the blue end of the spectrum (FAD and NADH peaks at approx.
450-520nm ).29
In vitro ligand binding assays of 4 – 7 with Aβ1-42/Aβ1-40 fibrils: We investigated the
emission behavior of 4 to 7 with monomers and fibrillar aggregates of Aβ1-42 and Aβ1-40 peptides
(Figure S4). These probes showed an apparent fluorescence intensity with significant
hypsochromic shift (Figure 2C and Table S3) in the presence of Aβ1-42 fibrils than that when
bound to Aβ1-40 fibrils, demonstrating that these probes are more sensitive to Aβ1-42 fibrils over
than Aβ1-40 fibrils. However, relative weak fluorescence changes were observed in the presence
12
ACS Paragon Plus Environment

Page 13 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
of freshly monomerized Aβ1-42 and Aβ1-40 peptides. It should be noted that upon binding to Aβ1-
42 fibrils, these probes displayed long emission wavelengths, especially 7 (655 nm).
7
8
9
Table 2. Binding affinity (Kd) and Aβ plaque binding specificity (S/N ratio) to Aβ transgenic
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
mouse brain slices
e
compd.
Kd (nM)a
S/N ratiob
λem (nm)c
core/corona
cLogPd
4.4
τavg (ns)
Aβ1-42
Aβ1-40
plaque core
plaque corona
5.9 ± 0.4
4
5
6
7
511 ± 29
250 ± 25
17.4 ± 1.2
17.4 ± 2.4
~ 1
545/557
8.4
36 ± 3
59 ± 5
> 400
> 400
23.2 ± 1.3
~ 1
537/562
5.9
8.2
8.4
3.4
4.4
5.4
40 ± 4
10 ± 1
5.4 ± 0.6
2.9 ± 0.2
630/630
a Values represent the average ± SD of 2 determinations. b Values represent the average ± SD of
c
10 determinations. Excitation wavelength 488 nm. Emission maximum of ligand bound to Aβ
plaques (cores and coronas). d Calculated cLogP values using ChemDraw Ultra 12.0. e Measured
for solute molecule in toluene.
Then, we measured the binding affinity of 4 to 7 towards Aβ1-42 fibrils and Aβ1-40 fibrils. As
shown in Table 2 and Figure S5, the binding affinity of these probes, in conjuncture with 3c,
towards Aβ1-40 fibrils is weaker than towards Aβ1-42 fibrils, which is consistent with the low
fluorescence intensity enhancement upon bound to Aβ1-40 fibrils. Therefore, we will focus on the
binding affinity of these probes to Aβ1-42 fibrils. Compound 4 had the weakest binding affinity all
the ligands, 26-fold weaker than that of 3c and 3-fold less than that of DF-9. This could be attributed
to the shorter distance between phenol functional groups, reducing the effectiveness of 4 to form
15
strong H-bonds with the Aβ1-42 fibril binding pockets . Compared to 3c, the more rigid linear
alkyne analogue 5 had a 2-fold lower Aβ1-42 fibril binding affinity, possibly due to an inability to
attain a conformer orientation complementary to the binding pockets of the fibril but retained high
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
9
affinity (36 ± 3 nM). The order of the binding affinity of 3c, 5, 6 and 7 are consistent with the order
of their emission maxima bound to Aβ1-42 fibrils, demonstrating that the effective conjugation
length will affect the binding affinity30-31. However, compound 6 appear unsuitable for tissue
staining (Figure S6). The same holds essentially true for compound 7. While 7 can stain amyloid
plaque in tissue (Figure S6) and has a >40 fold affinity for Aβ1-42 over Aβ1-40 fibrils, it has
drawbacks. The intensity of compound 7 is rather weak. Its quantum yield is 0.29, it has merely
3.5 and 1.8-fold intensity difference bound to Aβ1-42 and Aβ1-40 fibrils compared to unbound
probe, and small intensity differences over background for amyloid plaque. This property is likely
related to its low affinity towards Aβ1-40 fibrils, as was the case for 6 (Table 2).
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
BTD-ligand staining and spectra implicates sensitivity towards Aβ amyloid polymorphism.
We next went back to tissue sections and assessed the binding specificity of the probes. Brain
sections of an APP23 transgenic mouse (aged 19.5 months) was used for this purpose.32 Most
importantly, APP23 mice at this age show a pronounced intrinsic amyloid plaque structural
polymorphism32-34. We therefore selected this mouse model to assess the possibility of probing
amyloid structural polymorphism by the ligands as implicated in the X-34, DF-9 and BTDSB co-
staining experiments (Figure 1). By measuring the fluorescent intensities of Aβ plaques (specific
signal) and background regions (noise, Figure 4 and S6) using the same fluorescence image, we
first calculated a signal/noise (S/N) ratio, a measure of Aβ binding specificity.14-15 The notion that
amyloid plaques are structurally polymorphic as one goes outward from the center has been
reported previously for plaques formed in humans35 and transgenic mice.36 The binding specificity
data were summarized in Tables 1-2. From the chemical perspective regarding to different spacers,
4, 3c and 7 could apparently stain Aβ plaque cores over Aβ plaque coronas. This is opposite for
DF-9, which when used alone, preferentially stain Aβ plaque coronas rather than Aβ plaque cores
14
ACS Paragon Plus Environment

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(Table 1, Figure S6). Surprisingly, 6 showed merely trace fluorescence signals from all Aβ plaque.
Furthermore, on modifications on the outer phenyl rings using hydroxy and carboxyl motifs: i) 3c,
3d, 3f, 3g, 3i and 3j stain preferably Aβ plaque cores. ii) 3a, 3b, 3h and BTDSB rather
preferentially stain plaque coronas, iii) 3e stain both plaque cores and plaque coronas. Interestingly,
compared to the emission spectra of ligands bound to plaque coronas, the emission spectra of 5,
3a, 3c, 3d and 3e bound to plaque cores were clearly blue- shifted (25, 16, 26, 19 and 21 nm,
respectively). This indicates that the mature Aβ plaque cores have different binding sites or
different binding site structures compared to coronal fibrils. The emission spectra of 4, 7, 3b 3f,
3g, 3h and BTDSB merely slightly or did not show a blue-shift of the core compared to the corona.
The comparison was especially striking for the symmetric 3c versus 3h versus 3j which implicates
a structure activity relationship (SAR)
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
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
Figure 4. (A) Confocal images of mouse brain sections with accompanying normalized emission
spectra (B) and the fluorescence intensity (inset) of 3j and 3h as well as 3c bound to Aβ plaque
core and plaque corona. 500 nM was used for all ligands, excitation 488 nm. Mouse brain sections
are from an APP23 mouse (female, 19.5 months). Scale bars = 50 μm. (C) Chemical structures of
3j, 3h and 3c. (D) Emission maxima dependent on polarity of 3j, 3h and 3c as well as 5 in different
organic solvents (dioxane, toluene, EA, THF, acetone, DMF and DMSO) as well as in water. The
apparent blue shift in water likely stems from limited solubility of the compounds in water.
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: the para- and ortho-OH on the outer phenyl rings are sensitive to stain plaque cores
while the meta-OH is to stain intensely plaque coronas (Figure 4A, 4B and 4C) Interestingly,
fluorescence from 3c bound to plaque core and corona could be easily discriminated by virtue of
their spectra (Figure 4B) 36 This property was less pronounced for 3h and 3j. What is the basis for
this variation? Testing 3c, 3h and 3j in various solvents with decreasing dielectric constants reveals
a linear dependence on the emission peak for all probes. The slope of emission peak versus
dielectric constant was most pronounced for 3c showing that variations in hydrophobicity is best
sensed by 3c. We therefore conclude that at least in part the observed amyloid plaque staining
heterogeneity is explained by amyloid fibril binding site hydrophobicity, sensed most efficiently
by 3c (Figure 4D).
Given that trans-stilbenes are sensitive to cis-trans isomerization we noted that the alkyne
analogue to 3c, compound 5, also showed this variation in spectra comparing the core with the
corona of plaque (Table 2). As expected 5 showed a more pronounced slope of emission peak
versus dielectric constant compared to 3h (Figure 4D), analogous to 3c.
16
ACS Paragon Plus Environment

Page 17 of 43
Journal of Medicinal Chemistry
1
2
3
4
CONCLUSIONS
5
6
7
8
Targeting amyloid by small molecule dyes for diagnostics and research is an intense research area
moving towards new scaffolds to complement or replace Congo red and ThT.22,36-38 Recently the
notion of amyloid targeting has focused on the realization that amyloid fibrils are polymorphic and
behaves in a strain like manner which affect pathology and certainly should have bearing on
amyloid fibril targeting.39 Several studies have reported that combinations of multiple dyes can
reveal Aβ amyloid plaque polymorphic heterogeneity in human AD.40-41 Previous studies have
reported Nile red,42 oligothiophenes,43 ANCA44 as single dyes with striking solvatochromic effects
useful for amyloid fibril discrimination.
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
The structural basis for Aβ amyloid polymorphism has strong support by several structural
models of both Aβ1-40 and 1-42. This shows that Aβ amyloid fibrils generated under in vitro
conditions are polymorphic45 and can populate a variety of structures. These different folds and
their respective assembly forms will generate different binding pockets for amyloid ligands.39 It
has been demonstrated that the bis-azo dye Congo red and anionic oligothiphenes binds in pockets
outlined by lysine side chains.11, 46 We chose here-in to study a derivative inspired by the Congo
red analogue X-34 because bis-styryl-benzene based Florbetaben and Florbetapir are currently
approved for clinical diagnostics. In our study the BTD center motif generates desirable red-shifted
fluorescence properties important for broadening the research tool box. Importantly carboxylic acid
functional groups (X-34 and BTDSB) lower the binding affinity suggesting a diversion towards an
alternative binding site compared to the Congo red type.47 It is difficult to extrapolate in vitro
polymorphic Aβ fibril structures towards in vivo Aβ amyloid plaque. Most ligands of our study
preferred binding to Aβ1-42 over Aβ1-40 fibrils in vitro. The preferred binding site can therefore
be assumed to be present in Aβ1-42 fibrils. Recent model structures of Aβ1-42 fibrils made under
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
similar conditions as used here display protomer folds resulting in a surface located C-terminal
hydrophobic patch45. It is tempting to speculate that this is the high affinity binding site for the
neutral ligands of our study to in vitro fibrils. Amyloid deposits in APP23 mice are however
dominated by Aβ1-40. We have recently shown that in mature APPSwe mice at 18 months (similar
to APP23 mice used here) the diffuse amyloid plaque contains less Aβ1-40 (10:1) compared to
cored plaque (30:1).48 However, most importantly, the core is highly packed. By extrapolation
from DF-9 (neutral) binding in comparison to BTDSB (negatively charged) it appears that the core
binding site is more hydrophobic and the periphery more charged of Congo red type. It has
previously been shown in model fibril structures that the Congo red binding site is defined by a
rack of positively charged lysines11. We therefore speculate that the binding sites for core amyloid
probes 3c and 5 with para-OH on the phenyl rings is beneficial for binding affinity to tightly packed
Aβ fibril polymorphs with buried Lys 16 and 28 and with accessible C-terminal hydrophobic
binding sites. The peripheral plaque binding sites of the corona are in comparison dominated by
polymorphs containing Congo red type binding sites. Variations in packing, called assembly
polymorphism, and complexes with lipids may further contribute to the polarity variations
observed within the core as sensitized by the 3c type of probes reported here. It is noteworthy that
the emission spectra obtained from the amyloid plaque appear exceptionally hydrophobic as
derived from the solvent studies. Our ligands were designed with the aim to obtain planar neutral
red shifted amyloid probe analogues. What we trade-off from the charged ligand is water solubility.
While water solubility is often desirable for research purposes, presumably the neutral molecules
would increase blood brain barrier penetration. For CNS penetrant molecules compared to other
drugs an increased cLogP, less rotatable bonds, and fewer H-bonding partners are desirable49. Such
experiments are planned in future work. These data extend the SAR of BTD-based Aβ fibril
targeting ligands and provide further direction for the development of Aβ fibril ligands for early
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
18
ACS Paragon Plus Environment

Page 19 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
detection of AD. A new prototype based on 4,4'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-
diylbis(ethene-2,1-diyl))diphenol (3c) or 4,4'-(benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethyne-
2,1diyl))diphenol (5) could serve coming designs and discoveries of new early diagnosis tools for
AD and potentially for AD therapeutics.
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
EXPERIMENTAL SECTION
General information. All chemicals and solvents were obtained as reagent grade from conventional
commercial sources and used without further purification unless otherwise noted. Thin-layer
chromatography (TLC) was performed on Merck silica gel 60 F 254 (0.2 mm) on aluminum sheets
using UV-light (λ = 254 nm and 366 nm). Nuclear magnetic resonance (NMR) spectra were
recorded on a Varian Avance 300 MHz with solvent indicated. The H NMR and 13C NMR
1
chemical shifts are reported in parts per million (ppm) relative to tetramethylsilane (TMS) or the
residual solvent protons, 1H NMR abbreviations for NMR data: s = singlet; d = doublet; t = triplet;
dd = doublet of doublets; dt = doublet of triplets; m = multiplet. Coupling constants (J values) are
given in hertz (Hz). Column chromatographic purification was carried out using silica gel 60
(particles size 0.040-0.063 mm). Compound purity is determined by high performance liquid
chromatography (HPLC), and all final test compounds display purity higher than 95%. Analytical
liquid chromatography was run on a Waters system equipped with a Waters 1525 gradient pump,
2998 Photodiode Array Detector, 2424 Evaporative Light Scattering Detector, SQD 2 Mass
Detector and an Xbridge® C18 column (4.6×50 mm, 3.5 μm). A binary linear gradient of A/B
100:0 → 0:100 over 4 min followed by an additional 2 min at 0:100 was used. Flow rate 1.5
mL/min. (A): 95:5 water:acetonitrile, 10 mM NH4OAc pH 5.0; (B) 90:10 acetonitrile:water, 10
mM NH4OAc pH 5.0. Preparative liquid chromatography was run on a Waters system equipped
with a 2535 quaternary gradient pump, 2998 Photodiode Array Detector, 2424 Evaporative Light
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
4
5
6
7
8
Scattering Detector, SQD 2 Mass Detector and an xselect® phenyl-hexyl column (19 × 250 mm,
5.0 μm). Flow rate 25 mL/min. A binary linear gradient of A (95:5 water:acetonitrile, 10 mM
NH4OAc pH 5.0) and B (90:10 acetonitrile:water, 10 mM NH4OAc pH 5.0) was used.
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
Synthesis of Phenolic bis-styrylbenzo[c]-1,2,5-thiadiazoles. A mixture of 1 (100 mg, 0.34
mmol), the corresponding (hydroxy)styrene component one (0.41 mmol), hydroxy-styrene
component two (0.41 mmol), triethanolamine (203 mg, 1.36 mmol) and Pd(II)acetate (5.4 mg,
10%) in dry DMF (10 mL) was stirred under N2 at 110 oC for 24 h. Then the mixture was cooled
to room temperature, quenched by the addition of dil. aq. hydrochloride acid (0.5 N, 10 mL), and
extracted with ethyl acetate (3 x 20 mL). The combined organic layers were concentrated in vacuo,
and the crude product was purified by HPLC to give the target compounds.
1
4,7-di((E)-styryl)benzo[c][1,2,5]thiadiazole (3a)50: 18% yield as red solid. H NMR (300
MHz, Acetone-d6) δ 7.31 –7.36 (m, 2H), 7.42 – 7.47 (m, 4H), 7.71 – 7.80 (m, 6H), 7.92 (s, 2H),
13
8.16 (d, J = 16.5 Hz, 2H). C NMR (75 MHz, Acetone-d6) δ 124.4, 126.8, 127.6, 128.1, 129.3,
133.3, 137.7, 153.8.
4-((E)-2-(7-((E)-styryl)benzo[c][1,2,5]thiadiazol-4-yl)vinyl) phenol (3b): 26% yield as red
solid. 1H NMR (300 MHz, Acetone-d6) δ 6.92 (d, J = 8.7 Hz, 2H), 7.30 – 7.37 (m, 1H), 7.41 – 7.46
(m, 2H), 7.57 – 7.62 (m, 3H), 7.71 – 7.78 (m, 3H), 7.83 – 7.90 (m, 2H), 8.07 (d, J = 16.2 Hz, 1H),
13
8.14 (d, J = 16.2 Hz, 1H), 8.61 (s, OH) C NMR (75 MHz, Acetone-d6) δ 115.7, 115.8, 121.4,
123.7, 124.6, 126.5, 126.7, 126.9, 127.6, 127.8, 128.0, 128.3, 128.5, 128.7, 128.8, 129.2, 129.9,
132.8, 133.3, 134.2, 137.7, 153.8, 158.0. HRMS (ESI, m/z): calcd for [M-H]-, 355.0905, found
355.0932.
20
ACS Paragon Plus Environment

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
4,4'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethene-2,1-diyl))diphenol (3c): 41%
yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.92 (d, J = 8.4 Hz, 4H), 7.55 - 7.61 (m, 6H),
7
8
13
7.82 (d, J = 16.2 Hz, 4H), 8.59 (s, OH). C NMR (75 MHz, Acetone-d6) δ 115.7, 121.6, 126.7,
9
128.2, 129.2, 129.3, 132.8, 153.8. HRMS (ESI, m/z): calcd for [M-H]-, 371.0854, found 371.0815.
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'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethene-2,1-diyl))bis(benzene-1,2-diol)
(3d): 15% yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.88 (d, J = 8.1 Hz, 2H), 7.07 (dd,
J = 8.1, 1.8 Hz, 2H), 7.24 (d, J = 1.8 Hz, 2H), 7.53 (d, J = 16.2 Hz, 2H), 7.81 (s, 2H), 7.97 (d, J =
16.2 Hz, 2H). 13C NMR (75 MHz, Acetone-d6) δ 113.0, 115.5, 119.7, 121.6, 126.6, 129.1, 130.0,
133.0, 145.4, 145.9, 153.8. HRMS (ESI, m/z): calcd for [M-H]-, 403.0753, found 403.0730.
4-((E)-2-(7-((E)-4-hydroxystyryl)benzo[c][1,2,5]thiadiazol-4-yl)vinyl)benzene-1,2-diol
(3e): 29% yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.87 – 6.93 (m, 3H), 7.06 (dd, J =
8.4, 1.8 Hz, 1H), 7.24 (d, J = 1.8 Hz, 1H), 7.50 – 7.60 (m, 4H), 7.81 (s, 2H), 7.96 ( d, J = 16.2 Hz,
13
1H), 8.03 (d, J = 16.2 Hz, 1H). C NMR (75 MHz, Acetone-d6) δ 115.5, 115.6, 115.7, 119.3,
119.7, 121.4, 121.6, 126.1, 126.6, 126.7, 128.2, 128.3, 129.1, 129.9, 130.0, 132.8, 133.1, 133.7,
153.8, 157.7. HRMS (ESI, m/z): calcd for [M-H]-, 387.0803, found 387.0814.
5,5'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethene-2,1-diyl))bis(benzene-1,3-diol)
(3f): 12% yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.37 (s, 2H), 6.71 (d, J = 2.1 Hz,
4H), 7.64 (d, J = 16.5 Hz, 2H), 7.88 (s, 2H), 7.97 (d, J = 16.5 Hz, 2H), 8.30 (s, OH). 13C NMR (75
MHz, Acetone-d6) δ 102.9, 105.3, 124.1, 127.3, 129.3, 133.4, 139.6, 153.8, 158.9. HRMS (ESI,
m/z): calcd for [M-H]-, 403.0753, found 403.0798.
5-((E)-2-(7-((E)-4-hydroxystyryl)benzo[c][1,2,5]thiadiazol-4-yl)vinyl)benzene-1,3-diol
(3g): 26% yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.37 (t, J = 2.1 Hz, 1H), 6.71 (d,
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
4
5
6
7
8
9
J = 2.1 Hz, 2H), 6.92 (d, J = 8.4 Hz, 2H), 7.54 – 7.65 (m, 4H), 7.79 – 7.86 (m, 2H), 7.93 (d, J =
16.5 Hz, 1H), 8.04 (d, J = 16.5 Hz, 1H). 13C NMR (75 MHz, Acetone-d6) δ 115.5, 115.6, 115.7,
119.3, 119.7, 121.4, 121.6, 126.1, 126.6, 126.7, 128.2, 128.3, 129.1, 129.9, 130.0, 132.8, 133.1,
133.7, 153.8, 157.7. HRMS (ESI, m/z): calcd for [M-H]-, 387.0803, found 387.0770.
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
3,3'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethene-2,1-diyl))diphenol (3h): 9%
yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.82 (dd, J = 2.4, 1.2, 1H), 6.84 (t, J = 1.8
Hz, 1H), 7.17 – 7.29 (m, 6H), 7.71 (d, J = 16.2 Hz, 2H), 7.89 (8s, 2H), 8.07 (d, J = 16.2 Hz, 2H),
8.40 (s, OH). 13C NMR (75 MHz, Acetone-d6) δ 113.2, 115.4, 118.5, 124.4, 127.3, 129.2, 129.8,
133.3, 139.0, 153.8, 157.8. HRMS (ESI, m/z): calcd for [M-H]-, 371.0854, found 371.0829.
4-((E)-2-(7-((E)-3-hydroxystyryl)benzo[c][1,2,5]thiadiazol-4-yl)vinyl)benzene-1,2-diol (3i):
37% yield as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.82 (dt, J = 7.8, 1.2 Hz, 1H), 6.89 (d,
J = 8.1 Hz, 1H), 7.07 (dd, J = 8.1, 1.8 Hz, 1H), 7.17 – 7.28 (m, 4H), 7.55 (d, J = 16.5 Hz, 1H), 7.69
(d, J = 16.5 Hz, 1H), 7.82 – 7.89 (m, 2H), 7.99 (d, J = 16.5 Hz, 1H), 8.05 (d, J = 16.5 Hz, 1H). 13C
NMR (75 MHz, Acetone-d6) δ 113.0, 113.1, 115.2, 115.3, 115.6, 118.4, 119.8, 121.5, 124.4, 126.4,
127.6, 128.5, 129.7, 129.8, 130.0, 132.8, 133.5, 139.1, 145.4, 146.0, 153.8, 157.8. HRMS (ESI,
m/z): calcd for [M-H]-, 387.0853, found 387.0872.
2,2'-((1E,1'E)-benzo[c][1,2,5]thiadiazole-4,7-diylbis(ethene-2,1-diyl))diphenol (3j): 38%
1
yield as red solid. H NMR (300 MHz, Acetone-d6) δ 6.93 – 6.99 (m, 4H), 7.14 – 7.19 (m, 2H),
7.73 (d, J = 8.1 Hz, 2H), 7.83 (d, J = 16.5 Hz, 2H), 7.89 (s, 2H), 8.41 (d, J = 16.5 Hz, 2H), 8.79 (s,
13
OH). C NMR (75 MHz, Acetone-d6) δ 115.9, 119.9, 124.1, 124.7, 126.8, 126.9, 128.3, 129.1,
129.7, 153.9, 155.4. HRMS (ESI, m/z): calcd for [M-H]-, 371.0854, found 371.0842.
22
ACS Paragon Plus Environment

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Synthesis of 4,4'-(beno[c][1,2,5]thiadiazole-4,7-diyl)diphenol (4)51. To a mixture of 1 (100
mg, 0.34 mmol) and tetrakis(triphenylphophine)palladium (11.8 mg, 3%) in toluene (10 mL) was
added ethanol (5 mL) solution of 9 (94 mg, 0.68 mmol) and 2 M sodium carbonate solution (5 mL)
9
o
o
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
at 60 C under N2 atmosphere. After the reaction mixture was heated at 90 C for 12 h, it was
poured into water (20 mL) and extracted with ethyl acetate (EA) (30 mL x 3). The combined
organic layers were evaporated in vacuo to dryness. The residue was purified by silica gel column
chroma-tography eluting with EA /n-hexane (1:3, v/v) to give 10 in 89% yield (92 mg, 0.30 mmol)
as yellow solid.
To a mixture of 10 (40 mg, 0.13 mmol) and tetrakis(triphenylphophine)palladium (4.4 mg, 3%)
in toluene (10 mL) was added ethanol (5 mL) solution of 9 (36 mg, 0.26 mmol) and 2 M sodium
o
o
carbonate solution (5 mL) at 60 C under N2 atmosphere. After the mixture was heated at 90 C
overnight, the reaction mixture was poured into water (10 mL) and extracted with EA (20 mL x 3).
The combined organic layers were evaporated in vacuo to dryness. The residue was purified by
silica gel column chromatography eluting with EA /n-hexane (1:2 to 1:1, v/v) to give 4 in 92%
yield (37 mg,0.12 mmol) as yellow solid. 1H NMR (300 MHz, Acetone-d6) δ 7.02 (d, J = 8.7 Hz,
4H), 7.84 (s, 2H), 7.95 (d, J = 8.7 Hz, 4H), 8.59 (s, OH). 13C NMR (75 MHz, Acetone-d6) δ 115.3,
127.2, 128.8, 130.5, 132.1, 154.1, 157.7.
Synthesis of 4,4'-(benzo[c][1,2,5]thiadiazole-4,7 diylbis(ethyne-2,1 diyl))diphenol (5)52. A
mixture of 12 (60 mg, 0.30 mmol) in 1 N potassium hydroxide methanol solution (10 mL) was
stirred at room temperature for 1 h. The reaction mixture was poured into water (10 mL) and
extracted with chloroform (15 mL x 3). The combined organic layers were evaporated in vacuo to
dryness to give 4,7-diethynyl-2,1,3-benzothiadiazole (30 mg, 0.16 mmol) as brown powder. To a
solution of this brown powder, 4-iodophenol (79 mg, 0.36 mmol), and triethylamine (113 μL) in
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 43
1
2
3
4
5
6
7
8
9
THF (10 mL) was added Pd(OAc)2 (2.0 mg, 5%), PPh3 (9.0 mg, 20%) and cupper iodide (1.6 mg,
5%) under N2 atmosphere. After the mixture was stirred at room temperature for 12 h, the reaction
was poured into 1 N hydrochloric acid (10 mL) and extracted with dichloromethane (20 mL x 3).
The combined organic layers were evaporated in vacuo to dryness. The residue was purified by
silica gel column chromatography eluting with EA/n-hexane (1:2, v/v) to give 5 in 32% yield (19
mg, 0. 051 mmol) as red solid. 1H NMR (300 MHz, Acetone-d6) δ 6.95 (d, J = 8.7 Hz, 4H), 7.54
(d, J = 8.7 Hz, 4H), 7.85 (s, 2H), 8.93 (s, OH). 13C NMR (75 MHz, Acetone-d6) δ 83.9, 97.5, 113.4,
115.8, 117.0, 132.0, 133.4.
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
Synthesis of 4',4'''-(benzo[c][1,2,5]thiadiazole-4,7-diyl)bis(([1'',1'''-biphenyl]-4-ol)) (6). To
a mixture of 1 (100 mg, 0.34 mmol) and tetrakis(triphenylphophine)palladium (20 mg, 5%) in
toluene (20 mL) were added ethanol (5 mL) solution of 16 (252 mg, 0.85 mmol) and 2 M sodium
carbonate solution (5 mL) at 60 oC under N2 atmosphere. After the mixture was heated at 85 oC for
overnight, the reaction mixture was poured into water (20 mL) and extracted with EA (30 mL x 3).
The combined organic layers were evaporated in vacuo to dryness. The residue was separated by
silica gel column chromatography eluting with dichloromethane/n-hexane (1:1, v/v) to give 6 in
21% (34 mg, 0.07 mmol) as orange solid. 1H NMR (300 MHz, DMSO-d6) δ 6.90 (d, J = 8.1 Hz,
4H), 7.61 (d, J = 8.1 Hz, 4H), 7.75 (d, J = 8.1 Hz, 4H), 7.96 (s, 2H), 8.06 (d, J = 8.1 Hz, 4H), 9.66
(s, OH). 13C NMR (75 MHz, DMSO-d6) δ 116.3, 126.4, 128.3, 130.0, 130.7, 132.1, 135.2, 140.5,
153.9, 157.9. HRMS (ESI, m/z): calcd for [M-H]-, 471.1167, found 471.1113.
Synthesis
of
4,4'-(5,5'-(benzo[c][1,2,5]thiadiazole-4,7-diyl)bis(thiophene
-5,2-
diyl))diphenol (7). As described above for 4, 19 (100 mg, 0.22 mmol) and 9 (90 mg, 0.65 mmol)
yielded 7 in 46% yield (49 mg, 0.10 mmol) as deep-red solid. 1H NMR (300 MHz, DMSO-d6) δ
6.85 (d, J = 8.7 Hz, 4H), 7.46 (d, J = 3.9 Hz, 2H), 7.59 (d, J = 8.7 Hz, 4H), 8.07 (s, 2H), 8.13 (d, J
24
ACS Paragon Plus Environment

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
= 3.9 Hz, 2H). 13C NMR (75 MHz, DMSO-d6) δ 116.4, 123.3, 125.0, 125.1, 125.6, 127.3, 129.1,
5
6
136.6, 146.1, 152.1 158.1. HRMS (ESI, m/z): calcd for [M-H]-, 483.0296, found 483.0324.
7
8
9
Optical characterization of ligands in organic solvent and water. Absorption and emission
spectra of ligands in different solvents were collected using a Cary win UV 100 spectrophotometer
and a Tecan Infinity M1000 microplate reader (Tecan, Männedorf, Switzerland), respectively.
Quantum yield determination. The quantum yields of DF-9 and BTDSB and its derivatives were
measured in toluene. Dilute solutions (10-6 M) were used to minimize reabsorption effects.
Fluorescence measurement were made three times for each ligand and averaged. Quantum yield
was determined using the following equation:
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
Øsample= Østand (Fsample/Fstand) × (Astand / Asample) × (n2sample/n2
)
stand
Where Ø = Quantum yield, F = Area under fluorescence spectra, A = Absorption maxima, and n =
Refractive index. the reference for ligand 4 and 6 as well as DF-9 was Coumarin 314 in ethanol (Ø
= 0.68) and fluorescein in 0.1M NaOH (Ø = 0.95) was used for all other compounds.
Fluorescence lifetime measurements. Fluorescence decays were performed using an IBH time-
correlated single photo counting (TCSPC) spectrometer. The solution of each compound in toluene
was degassed. After the sample solution was excited at each excitation wavelength, the
fluorescence decay at emission maxima for each probe in toluene was monitored at room
temperature. The instrument response function was determined with a fibril solution. Decay was
fitted with a sum of exponentials (eq. 1) and the quality of the fit was determined from the χ2
parameter and the random distribution of the residuals and the autocorrelation of the residuals.
∑
퐼(푡) = 훼푖exp ( ― 푡/휏푖)
(1)
푖
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 43
1
2
3
4
Where αi are the preexponential factors associated with the decay time τi. The average lifetime of
5
6
fluorescence decay was calculated according to the following equation:
7
8
9
2
∑
훼푖휏푖
휏푎푣푔
=
(2)
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
∑
훼푖휏푖
Characterization of ligands towards recombinant Aβ1-42 peptide (monomers and fibrils)
and Aβ1-40 peptide (monomers and fibrils). Recombinant Aβ1-42 and 1-40 was expressed and
purified using the system described by Linse and coworkers containing an additional methionine
in the N-terminus.53 Before the final step the lyophilized Aβ sample was solubilized in 6 M GuHCl
before gel filtration on a Superdex 75 10/300 column (GE-Healthcare) in PBS buffer. Freshly
monomeric gel filtered Aβ1-42 and 1-40 was used in all experiments. Aβ1-42 protein concentration
was calculated using ε = 1490 M−1 cm−1 at 277 nm. The Aβ1-42 and 1-40 fibrils were generated
o
through incubation of 10 µM Aβ1-42 and 1-40 at 37 C in PBS buffer (pH 7.4) for 48 h under
quiescent conditions and confirmed by thioflavin-T (ThT) staining.
Stock solutions of each ligand were made in DMSO. 1 μM of the respective ligand was added to
5 μM Aβ1-42 peptide (monomers and fibrils) and Aβ1-40 peptide (monomers and fibrils) in PBS
and the excitation and emission spectra for each ligand were collected using a Tecan Saphire
microplate reader (Tecan, Männedorf, Switzerland). Both excitation and emission slit widths were
5 nm unless noted otherwise. The fold increase of the fluorescence intensity after binding to Aβ
monomers or fibrils was calculated by the following equation:
Fold increase (FI) = FItest / (FIprobe - FIPBS
)
Where, FItest, FIprobe and FIPBS represent the fluorescence intensities of the ligands upon binding to
Aβ1-42 or Aβ1-40 fibrils, the ligands in PBS, and PBS alone, respectively.
26
ACS Paragon Plus Environment

Page 27 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vitro binding affinity determination by fluorescence titration. Aβ1-42 or Aβ1-40 binding
affinities (Kd) for 3 (3a-3j), 4, 5, 6, 7 and BTDSB as well as DF-9 and X-34 were determined by
means of fluorescence titrations.14, 22 Increasing concentration of ligand (0-400 nM) was titrated
against a fixed concentration of fibrils (1 μM) and the corresponding fluorescence spectra were
recorded until the fluorescence no longer increased (saturation) by the Infinity M1000 plate reader
(Tecan, Männedorf, Switzerland) using excitation wavelength determined for each compound. The
Kd binding curves were generated by GraphPad Prism 7.0 by using below equation,
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
Y = Bmax * X / (Kd + X)
where X is concentration of ligand and Y is change in fluorescence intensity. Bmax is the maximum
specific binding having the same unit as Y.
Ex vivo plaque binding specificity using APP23 transgenic mouse brain sections. Frozen
brain sections (10 μm) from APP23 mouse brain tissues from a 19.5 months old female54 were
fixed in 98% EtOH, rehydrated in 50% EtOH and dH2O and then incubated in PBS for 10 min.
Probes were diluted to 500 nM in PBS and added to the sections. After 1 h, the sections were
washed with PBS and mounted with DaKo fluorescent mounting medium (DaKo Cytomation,
Glostrup, Denmark). The mounting medium was allowed to solidify over night before collecting
emission spectral images from cortical plaques using an inverted LSM 780 confocal microscope
(Carl Zeiss, Oberkochen, Germany) with 405 nm (DF-9, X-34) or 488 nm (all other probes) as
excitation wavelengths.
ASSOCIATED CONTENT
Supporting Information. The supporting information is available on the ACS Publications
website at: XXX. SI items: Molecular Formula Strings, and Supporting Figures S1-S6, Supporting
Tables S1-S3, and NMR spectra of all target molecules.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
AUTHOR INFORMATION
Corresponding Author
Per Hammarström
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
Division of chemistry, Department of Physics, Chemistry and Biology
Linköping University, 581 83 Linköping, Sweden
E-mail: per.hammarstrom@liu.se
Author Contributions
The study was conceived by JZ, XW, PK and PH. JZ, AK, AS, XW, SN, PK, ML, PH performed
research and analyzed data. UO and BWB provided samples. The manuscript draft was written by
JZ and PH. All authors have given approval to the final version of the manuscript.
Funding Sources
Our work was funded by the China Scholarship Council, the Swedish Research Council (2015-
04521), the Göran Gustafsson Foundation, the Swedish Alzheimer Foundation, the Swedish Brain
foundation, and Linköping University (LiU-Neuro).
Notes
Any additional relevant notes should be placed here.
ACKNOWLEDGMENT
We acknowledge Peter Nilsson for discussions on fluorescence microscopy and Mathias Jucker for
discussions and assistance with APP23 tissue handling.
ABBREVIATIONS
AD, Alzheimer’s disease; TLC, thin layer chromatography.
28
ACS Paragon Plus Environment

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
REFERENCES
(1) Goedert, M.; Spillantini, M. G. A century of Alzheimer's disease. Science 2006, 314, 777-
781.
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
(2) Prince, M.; Bryce, R.; Albanese, E.; Wimo, A.; Ribeiro, W.; Ferri, C. P. The global prevalence
of dementia: a systematic review and metaanalysis. Alzheimers Dement. 2013, 9, 63-75.
(3) Hamley, I. W. The amyloid beta peptide: a chemist’s perspective. role in Alzheimer’s and
fibrillization. Chem. Rev. 2012, 112, 5147-5192.
(4) Nisbet, R. M.; Polanco, J.-C.; Ittner, L. M.; Götz, J. Tau aggregation and its interplay with
amyloid-β. Acta Neuropathol. 2015, 129, 207-220.
(5) Iqbal, K.; Liu, F.; Gong, C. X. Alzheimer disease therapeutics: Focus on the disease and not
just plaques and tangles. Biochem. Pharmacol. 2014, 88, 631-639.
(6) Kung, H. F.; Lee, C. W.; Zhuang, Z. P.; Kung, M. P.; Hou, C.; Plössl, K. Novel stilbenes as
probes for amyloid plaques. J. Am. Chem. Soc. 2001, 123, 12740-12741.
(7) Nesterov, E. E.; Skoch, J.; Hyman, B. T.; Klunk, W. E.; Bacskai, B. J.; Swager, T. M. In vivo
optical imaging of amy-loid aggregates in brain: Design of fluorescent markers. Angew. Chem. Int.
Ed. 2005, 44, 5452-5456.
(8) Shozo, F.; Nobuyuki, O.; Ren, I.; Kazuhiko, Y.; Hiroyuki, A.; Yukitsuka, K. Recent advances
in the development of amyloid imaging agents. Curr. Top. Med. Chem. 2007, 7, 1773-1789.
(9) Nilsson, K. P. R. Small organic probes as amyloid specific ligands – past and recent molecular
scaffolds. FEBS Lett. 2009, 583, 2593-2599.
(10) Scot D. Styren; Hamilton, R. L.; Styren, G. C.; Klunk, W. E. X-34, a fluorescent derivative
of Congo red: a novel histochemical stain for Alzheimer's disease. J. Histochem. Cytochem. 2000,
48, 1223-1232.
29
ACS Paragon Plus Environment