Subscriber access provided by West Virginia University | Libraries
Letter
A Difluoroboron #–Diketonate Probe Shows “Turn– on” Near–Infrared Fluorescence Specific for Tau Fibrils Kwang-su Park, Mi Kyoung Kim, Yujin Seo, Taewoong Ha, Kyeongha Yoo, Seung Jae Hyeon, Yu Jin Hwang, Junghee Lee, Hoon Ryu, Hyunah Choo, and Youhoon Chong ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.7b00224 • Publication Date (Web): 24 Jul 2017 Downloaded from http://pubs.acs.org on July 25, 2017
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 free 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 accessible to all readers and 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.
ACS Chemical Neuroscience 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 36
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
ACS Chemical Neuroscience
1
A Difluoroboron β–Diketonate Probe Shows “Turn–on” Near–Infrared Fluorescence Specific
2
for Tau Fibrils
3 4
Kwang–su Park,1,† Mi Kyoung Kim,1,† Yujin Seo,1 Taewoong Ha,1 Kyeongha Yoo,1 Seung Jae Hyeon,2
5
Yu Jin Hwang,2 Junghee Lee,3,4 Hoon Ryu,2,3,4,* Hyunah Choo,2,5,* and Youhoon Chong1,*
6 7
1
8
Hwayang–dong, Gwangjin–gu, Seoul 143–701, Korea; 2Center for Neuro–Medicine, Korea Institute
9
of Science and Technology, 39–1 Hawolgok–dong, Seoungbuk–gu, Seoul 136–791, Korea; 3Veteran’s
10
Affairs Boston Healthcare System, Boston, MA, USA; 4Boston University Alzheimer’s Disease Center
11
and Department of Neurology, Boston University School of Medicine, Boston, MA, USA; 5Department
12
of Biological Chemistry, Korea University of Science and Technology, Youseong–gu, Daejeon 305–
13
350, Korea
Department of Bioscience and Biotechnology, Bio/Molecular Informatics Center, Konkuk University,
14 15
†
These two authors contributed equally on this work
16 17
*
Corresponding authors
18
Hoon Ryu, Ph.D.; E-mail:
[email protected]; Tel: +1-857-364-5910
19
Hyunah Choo, Ph.D.; E–mail:
[email protected]; Tel: +82–2–958–5157
20
Youhoon Chong, Ph.D.; E–mail:
[email protected]; Tel: +82–2–2049–6100; Fax: +82–2–454–
21
8217
22
-1-
ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
23
ABSTRACT
24 25
Tau aggregation in neuronal cells has recently received significant attention as a robust predictor of
26
the progression of Alzheimer’s disease (AD) because of its proven correlation with the degree of
27
cognitive impairment in AD patients. Accordingly, non–invasive imaging of tau aggregates has been
28
highlighted as a promising diagnostic tool for AD. We have previously identified a tau–specific “turn–
29
on” near–infrared fluorescent (NIRF) probe (1) and, in this study, structural modification was
30
performed to optimize its physicochemical as well as fluorescence properties. Thus, a series of
31
fluorescent dyes (2a–2j) composed of a variously substituted difluoroboron β–diketonate and an N,N–
32
dimethylaniline moiety linked by a length–extendable π–bridge were prepared. Among those,
33
isobutyl–substituted difluoroboron β–ketonate with a π–conjugated 1,4–butadienyl linker (2e) showed
34
the most promising properties as a tau–specific NIRF probe. Compared with 1, the “turn–on”
35
fluorescence of 2e was more specific to tau fibrils, and it showed 8.8– and 6.2–times higher tau–over–
36
Aβ and tau–over–BSA specificity, respectively. Also, the fluorescence intensity of 2e upon binding to
37
tau fibrils was substantially higher (~2.9 times) than that observed from 1. The mechanism for tau-
38
specificity of 2e was investigated, which suggested that the molecular rotor–like property of 2e
39
enables specific recognition of the microenvironment of tau aggregates to emit strong fluorescence. In
40
transgenic cell lines stably expressing GFP–tagged tau proteins, 2e showed good co–localization with
41
tau–GFP. Moreover, the fluorescence from 2e exhibited almost complete overlap with p–Tau antibody
42
staining in the human AD brain tissue section. Collectively, these observations demonstrate the
43
potential of 2e as a tau-specific fluorescent dye in both in vitro and ex vivo settings.
44 45
Key words: Alzheimer’s disease, tau–specific probe, fluorescence imaging, molecular rotor
46
-2-
ACS Paragon Plus Environment
Page 2 of 36
Page 3 of 36
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
ACS Chemical Neuroscience
47
Alzheimer’s disease (AD), the leading cause of dementia, has been highlighted due to an
48
enormous socioeconomic burden imposed by the ever–growing number of elderly patients
49
with dementia.1–2 In the absence of fundamental cure3, delaying the onset or progression of
50
AD has been proposed as a more practical strategy for the management of this devastating
51
disease4. Therefore, early diagnosis of AD has become one of the most fundamental precepts
52
of AD care and, for this purpose, a large body of research has focused on visualization of the
53
major pathological hallmarks of AD5: amyloid plaques and neurofibrillary tangles, which are
54
composed of misfolded amyloid–β (Aβ) and hyperphosphorylated tau proteins, respectively.
55
In particular, a strong correlation between the degree of cognitive impairment in AD patients
56
and the level of aggregated tau6–8 suggests that detection of tau aggregates would provide
57
potential benefit to confirmative diagnosis of AD9–11. Several molecular probes targeting tau
58
fibrils have thus been discovered12–22 and, among those, near–infrared (650–900 nm)
59
fluorescent (NIRF) probes19–22 have gained attention because, compared to visible fluorescent
60
dyes, NIRF dyes provide significant advantages offering higher resolution imaging, a greater
61
imaging depth, and a higher signal–to–background ratio.23
62
Recently, we reported a smart NIRF probe (1, Figure 1) with “turn–on” fluorescence upon
63
binding to the target proteins.24 More interestingly, 1 exhibited selective fluorescence sensing
64
behavior for tau fibrils over Aβ: upon binding to tau fibrils, 1 exhibited fluorescence with an
65
effective quantum yield (Φ) of 16.5%, which is 6.6 times higher than that observed in binding
66
to Aβ fibrils (Φ = 2.5%).24 The “turn–on” fluorescence property as well as tau–specificity of 1
67
were attributed to its molecular rotor–like architecture because fluorescent molecular rotors,
68
characterized by a fluorophore with a rotatable donor–acceptor bond, tend to emit fluorescence
69
which is selectively turned on when the intramolecular rotational relaxation about the donor–
70
acceptor bond is constrained.25 Also, as fluorescence intensity from the molecular rotors is
71
subject to change depending on their microenvironment,26–27 it was conceived that the
72
molecular rotor property of 1 enabled its specific recognition of the tau aggregates to result in
73
tau–specific fluorescence behaviour. -3-
ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
74
Nevertheless, unsatisfactory properties associated with 1 such as poor solubility and
75
stability in aqueous medium, weak fluorescence intensity and suboptimal tau–specificity (1.6
76
times over Aβ) prevented its further development. In addition, synthetic difficulties in
77
preparing the 3,6–dimethoxy–N,N–dimethylanilne functionality hampered optimization of 1
78
through structure–activity relationship study. Scaffold change has thus been attempted by
79
replacing the structural units incorporated in 1. The molecular rotor–like architecture of 1 is
80
composed of three modules: a donor (N,N–dimethylaniln–3,6–dimethoxy–4–yl group; dotted
81
circle, 1, Fig. 1) and an acceptor (α–cyanoester; dotted box, 1) bridged by a π–linker
82
(hexatriene; bold lines, 1). Among those, the acceptor, α–cyanoester, was supposedly
83
responsible for the unfavourable physicochemical as well as fluorescent properties of 1.
84
Therefore, alternatives for the α–cyanoester have been sought, and difluoroboron β–diketonate
85
which is well characterized by intense fluorescence28, solvatochromism29–30 and good aqueous
86
stability31 was proposed as the optimal candidate. A difluoroboron-based dye, styryl-BODIPY,
87
was recently discovered to have favourable photochemical and physicochemical properties for
88
monitoring real-time in vitro tau protein fibrillization32, which supports our reasoning that a
89
difluoroboron β–diketonate dye could be exploited for specific detection of tau fibrils. On the
90
other hand, for facile synthesis and structure–activity relationship study, the donor part of 1
91
was changed into an unsubstituted N,N–dimethylanilne instead of the 3,6–dimethoxy–N,N–
92
dimethylanilne functionality. The two modules, a difluoroboron β–diketonate and an N,N–
93
dimethylanilne moiety, were then combined by using a π–linker to provide a new probe
94
scaffold 2 (Figure 1). The newly designed difluoroboron β–diketonate 2, however, resembles
95
CRANAD–233 (Figure 1), a well–known Aβ–specific NIRF probe. Moreover, difluoroboron
96
β–diketonate dyes have not been utilized for detection of tau fibrils. Notwithstanding these
97
issues, it is worth to note that the mechanism for Aβ–specificity of CRANAD–2 is unexplored
98
and, in particular, contributions of the symmetric nature as well as the length of the π–linker of
99
CRANAD–2 for specific staining of Aβ fibrils are elusive. In this context, a potential role of -4-
ACS Paragon Plus Environment
Page 4 of 36
Page 5 of 36
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
ACS Chemical Neuroscience
100
the π–linker and the α–cyanoacetate functionality in conferring tau–specificity to the resulting
101
difluoroboron β–diketonate dye was assumed, which led to the installation of a length–
102
extendable π–linker (n=1–2, Figure 1) as well as various alkyl substituents (R, Figure 1) to the
103
newly designed probe scaffold 2.
104
105 106
Figure 1. Structure of the title compound 2 in comparison with 1 and CRANAD–2. Dotted circles,
107
dotted boxes and bold lines denote donors, acceptors and π–bridges, respectively.
108 109
Preparation of the title compound is outlined in Scheme 1. Conjugated aldehydes (3–4),
110
obtained from 4–N,N–dimethylaminobenzaldehyde through consecutive Wittig reactions with
111
(1,3–dioxolan–2–ylmethyl)triphenylphosphonium
112
commercially available acetylacetones (5–9) to provide the title compounds (2a–2j) (Scheme
113
1). This aldol–type reaction has been widely applied for the condensation of aromatic
114
aldehydes with acetylacetones, but in our hands, the reaction proceeded in disappointingly low
115
yields. After surveying a variety of reaction conditions, we found out that the reaction
116
proceeds in fair to moderated yields (55%–77%) within a sealed tube.34
bromide,
-5-
ACS Paragon Plus Environment
were
condensed
with
ACS Chemical Neuroscience
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
117 118
Scheme 1. Synthesis of the title compounds (2a–2j)
119 120
The difluoroboron β–ketonates (2a–2j) thus obtained were dissolved in phosphate–
121
buffered saline (PBS) (50 µM) and fluorescence from these compounds was observed before
122
and after mixing with pre–aggregated tau, Aβ fibrils or bovine serum albumin (BSA) (Figure
123
S1; Supporting Information)24. In the absence of the aggregated peptides, none of the title
124
probes showed fluorescence emission (brown lines, Figure S1; Supporting Information).
125
Interestingly, upon exposure to the target peptides, only a limited number of the probes
126
exhibited substantial fluorescence intensity. In particular, the tau–specific “turn–on”
127
fluorescence was observed from the probes with a 1,4–butadienyl π–bridge (n=1) and an
128
aliphatic substituent at R position (2a, 2d and 2e). On the other hand, fluorescence from the
129
probes with a longer π–linker (n=2, 2f–2j) or with a conformationally rigid aromatic R
130
substituent (2b and 2c) was very weak regardless of the type of the aggregated peptides. Under
131
the same conditions with pre–aggregated tau, no measurable fluorescence signal was detected
132
from CRANAD-2, which indicates that it is not capable of staining tau fibrils (Figure S2;
133
Supporting Information). Table 1 collectively shows structure–dependent fluorescence -6-
ACS Paragon Plus Environment
Page 6 of 36
Page 7 of 36
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
ACS Chemical Neuroscience
134
properties as well as tau–specificities of the title probes. Among the compounds prepared,
135
isobutyl–substituted difluoroboron β–ketonate with a π–conjugated 1,4–butadienyl linker (2e)
136
showed the most promising properties as a tau–specific fluorescent probe; fluorescence
137
emission at NIR range (λem = 660 nm) with a large Stokes’ shift (110 nm), 313–fold
138
fluorescence “turn-on” response upon binding to tau–fibrils (FItau, Table 1), and significant
139
specificity for tau (14–fold over Aβ and 7.2–fold over BSA, Table 1).
140 141
Table 1. Fluorescence properties of the synthesized probes (50 µM) in the presence of tau, Aβ fibrils,
142
and BSA FId
SIe
εa (M–1cm–1)
λexb (nm)
λemc (nm)
FItau
FIAβ
FIBSA
SIAβ
SIBSA
2a
35980
560
660
430
130
116
3.3
3.7
2b
68330
600
690
160
22
28
7.3
5.6
2c
52220
610
710
120
40
31
3.0
3.9
2d
59290
550
660
440
76
59
5.8
7.4
2e
53500
550
660
310
22
43
14
7.2
2f
17270
560
720
87
13
7
6.5
13
2g
34870
610
700
–g
–g
–g
–g
–g
2h
23750
610
680
–g
–g
–g
–g
–g
2i
46090
570
700
100
43
6
2.3
17
2j
12840
570
720
79
88
5
0.9
16
1
30711
540
650
106
67
85
1.6
1.2
Cmpd
143 144
a
145
(in PBS). cMaximum emission wavelength of the probe (in PBS). dFold Increase = fluorescence
146
intensity of the probe bound to tau fibrils (or Aβ fibrils or BSA)/fluorescence intensity of the unbound
147
probe. eSelectivity index (SI) = fluorescence intensity of the probe upon binding to tau/fluorescence
148
intensity of the probe upon binding to Aβ fibrils (or BSA). fNot determined due to low fluorescence.
Molar extinction coefficient (in dimethylsulfoxide). bMaximum excitation wavelength of the probe
149 -7-
ACS Paragon Plus Environment
ACS Chemical Neuroscience
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
Page 8 of 36
150
In addition to the fluorescence characteristics, physicochemical properties of the title
151
probes were also determined. The difluoroboron β–diketonate scaffold was anticipated to
152
increase the water solubility of the resulting probes but, due to the planar structure, most of the
153
title probes were shown to be only sparingly to moderately soluble in water (Table 2, Figure
154
S3; Supporting Information). Nevertheless, the isobutyl–substituted difluoroboron β–
155
diketonates 2e showed significantly increased aqueous solubility compared to 1 (Table 2,
156
Figure S3; Supporting Information), which might be attributed to rotation of the methylene-
157
linked substituent (R = CH2CH(CH3)2) out of the difluoroboron β–diketonate plane and
158
thereby decrease in intermolecular forces. Blood-brain barrier (BBB) permeability is another
159
important property of brain-imaging molecular probes. Several physicochemical properties are
160
known to be related to the BBB permeability and, among those, logP value has a major effect
161
on BBB penetration: top 25 CNS drugs are known to have logP values between 2 and 5.35
162
Thus, the logP values of the title probes were assessed, which were shown to be correlated
163
with the solubility profile (Table 2): the probes with sufficient solubility (2c and 2e) showed
164
optimum logP values for BBB penetration while others with low solubility presented less
165
favourable (2b and 1) or non-measurable (2a, 2d, 2f–2j) logP values (Table 2).
166 167
Table 2. Physicochemical properties of the synthesized probes Cmpd
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
1
Solubility (µM)a
40
40
90
40
>100
