A Difluoroboron β-Diketonate Probe Shows “Turn ... - ACS Publications

Jul 24, 2017 - Yujin Seo,. †. Taewoong Ha,. †. Kyeongha Yoo,. †. Seung Jae Hyeon,. ‡. Yu Jin Hwang,. ‡. Junghee Lee,. §,∥. Hoon Ryu,*,‡...
0 downloads 11 Views 2MB Size
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