A Smart Near-Infrared Fluorescence Probe for Selective Detection of

Letter pubs.acs.org/chemneuro

A Smart Near-Infrared Fluorescence Probe for Selective Detection of Tau Fibrils in Alzheimer’s Disease Yujin Seo,†,# Kwang-su Park,†,# Taewoong Ha,† Mi Kyoung Kim,† Yu Jin Hwang,‡ Junghee Lee,§,∥ Hoon Ryu,‡,§,∥ Hyunah Choo,*,‡,⊥ and Youhoon Chong*,† †

Department of Integrative Bioscience and Biotechnology, Bio/Molecular Informatics Center, Konkuk University, Hwayang-dong, Gwangjin-gu, Seoul 143-701, Korea ‡ Center for Neuro-Medicine, Korea Institute of Science and Technology, 39-1 Hawolgok-dong, Seoungbuk-gu, Seoul 136-791, Korea § Veteran’s Affairs Boston Healthcare System, Boston, Massachusetts 02130, United States ∥ Boston University Alzheimer’s Disease Center and Department of Neurology, Boston University School of Medicine, Boston, Massachusetts 02118, United States ⊥ Department of Biological Chemistry, Korea University of Science and Technology, Youseong-gu, Daejeon 305-350, Korea S Supporting Information *

ABSTRACT: Development of a novel, tau-selective smart near-infrared fluorescence (NIRF) probe was attempted by combining the previously identified core scaffold 3,5dimethoxy-N,N-dimethylaniline-4-yl moiety, with the characteristic donor-π-acceptor architecture of the smart NIRF Aβ probes DANIR-2c and MCAAD-3. A series of compounds (2 and 3) were prepared, which were identified as “turn-on” NIRF probes for the visual detection of tau aggregates and Aβ fibrils (λem = 650 nm, Stokes shifts = 70−110 nm). In particular, combination of the 3,5-dimethoxy-N,N-dimethylanilin-4-yl moiety and the donor part of MCAAD-3 endowed the resulting probes, 3g and 3h, with significant selectivity toward tau aggregates (selectivity for tau over Aβ = 5.7 and 3.8); they showed much higher fluorescence intensities upon binding to tau aggregates (FItau = 49 and 108) than when bound to Aβ fibrils (FIAβ = 9 and 28). Quantitative analysis of binding affinities and fluorescence properties of 3g and 3h revealed that microenvironment-sensitive molecular rotor-like behavior, rather than binding affinity to the target, is responsible for their selective turn-on fluorescence detection of tau fibrils. Selective fluorescent labeling of tau fibrils by 3g and 3h was further demonstrated by immunofluorescence staining of human Alzheimer’s disease brain sections, which showed colocalization of the probes (3g and 3h) and phosphorylated tau antibody. KEYWORDS: Near-infrared fluorescence (NIRF), smart probe, tau fibrils, molecular rotor, Alzheimer’s disease

A

PET probes [florbetaben (Neuraceq, Piramal Imaging),9 florbetapir (Amyvid, Eli Lilly and Company),10 and flutemetamol (Vizamyl, GE Healthcare)11], for the estimation of Aβ neuritic plaque density in adult patients with suspected AD and other causes of cognitive decline. Nevertheless, due to a weak correlation between Aβ plaque load and cognition,12 a positive PET scan with these probes cannot be used to provide a confirmative diagnosis of AD. Rather, a negative scan, indicating sparse to no amyloid plaques, suggests that a patient’s cognitive impairment has a low chance of being related to AD. Therefore, the PET probes are approved, not for AD diagnosis or evaluation of cognitive impairment, but for the exclusion of AD in patients with cognitive impairment who have a negative amyloid PET scan result.13 Therefore, confirmatory diagnosis of AD still depends on post-mortem

lzheimer’s disease (AD) is a chronic neurological disorder, which, according to the Alzheimer’s Association, is characterized by three stages: preclinical stage with no symptoms, mild cognitive impairment (MCI), and AD dementia.1−4 Unfortunately, there is no cure and only symptomatic treatment is available for patients with AD. However, as the symptomless preclinical stage is lengthy (10− 20 years),5,6 early diagnosis of AD could provide the possibility of therapeutic intervention for this devastating disease. Senile β-amyloid (Aβ) plaques and neurofibrillary tangles (NFTs) constitute the primary pathological hallmarks of AD,7 and diagnosis has focused on their noninvasive detection, using several molecular imaging modalities. Over the past decade, positron emission tomography (PET) has been the most extensively investigated technique in the search for a reliable diagnostic tool in the field of AD research.8 In conjunction with PET imaging, large-scale screening campaigns of radiolabeled molecular probes have been performed, resulting in United States Food and Drug Administration (FDA) approval of three © XXXX American Chemical Society

Received: June 19, 2016 Accepted: August 30, 2016

A

DOI: 10.1021/acschemneuro.6b00174 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Letter

ACS Chemical Neuroscience assessment of the pathophysiological hallmarks, and current brain imaging technology based on PET imaging of Aβ-plaques needs significant improvement for the noninvasive confirmation of AD. In this context, NFTs, another pathological hallmark of AD, have received more attention as desirable biomarkers for AD. Unlike Aβ-plaques, NFTs composed of aggregated tau protein have shown a significant correlation with the degree of cognitive impairment,12,14,15 and tau imaging has recently been reported as a more robust predictor of AD progression than Aβ imaging.16 In addition, noninvasive detection of tau aggregates in the brain can also be used to diagnose broadspectrum tauopathies,17 characterized by tau inclusions in neuronal and glial cells (Down’s syndrome, progressive supranuclear palsy, corticobasal degeneration, Pick’s disease, and familial FTDP-17 syndrome). To date, several PET tau tracers including [11C]-PBB3,18 [18F]-THK-523,19 [18F]-THK5105,20 [18F]-THK-5117,21 [18F]-T808,22 [18F]-FDDNP23 and [18F]-T80724 have been developed for in vivo detection of neurofibrillary pathology. Despite a number of advantages, PET imaging also has disadvantages including radiation exposure, the use of shortlived radiotracers, and the need for expensive equipment and personnel skilled in its use. Collectively, these drawbacks make PET imaging a difficult treatment to deliver. In contrast, optical imaging modalities do not use isotope labeling, making them attractive alternatives to the classical nuclear imaging techniques represented by PET. In particular, near-infrared fluorescence (NIRF) has emerged as a tool of choice for optical imaging because fluorescent probes with emission wavelengths in the NIR range allow light penetration into tissue and avoid autofluorescence from biological matter in vivo. Nevertheless, the NIR light has only limited depth of penetration in tissue (1−2 cm),25 which is not sufficient for clinical imaging. Recently, this problem was tackled by measuring fluorescence intensities of the Aβ and tau NIRF probes in the cerebrospinal fluids (CSF).26 Thus, an optical fiber coupled to a laser source was inserted near the CSF and, upon laser radiation, the NIRF probes bound to Aβ and tau in the CSF were excited to emit fluorescence. Notably, the concentration ratio of Aβ and tau calculated from their fluorescence intensity ratio showed a good correlation with the risk of developing AD, which demonstrated diagnostic potential of the fluorescence imaging of Aβ and tau. Therefore, in terms of both targets and modalities, NIRF imaging techniques for detecting tau fibrils are a promising tool for the early diagnosis of AD, believed to complement the currently available PET Aβ imaging methods. However, only a few NIRF probes have been identified to detect tau aggregates. Among these, BODIPY-based Zn(II) complex27 and bis(arylvinyl)pyrimidines28 showed good binding affinities to in vitro aggregates of tau; however, they had emission wavelengths outside the NIR range (545 and 581 nm, respectively). Recently, we reported a curcumin-based tau fluorescence probe (1, Figure 1) with interesting fluorescence properties (λem = 620 nm, Stokes shift = 120 nm, quantum yield = 0.321),29 but several suboptimal fluorescence properties limited potential use of 1 as an NIRF probe for noninvasive detection of tau fibrils: (1) fluorescence emission wavelength outside of the NIR range (λem = 620 nm); (2) background autofluorescence; (3) lack of selectivity against Aβ fibrils. Nevertheless, one notable feature of 1 is that it showed significantly enhanced fluorescent properties upon binding to the target molecule. Specifically, the 3,5-dimethoxy substituents

Figure 1. Design of the novel fluorescence probes 2 and 3.

on N,N-dimethylaniline, which restrict rotation of the aromatic ring into a coplanar state in target-free conditions, turned out to play a pivotal role in providing 1 with environment-dependent fluorescence, a typical property of molecular rotors.30 The 3,5dimethoxy-N,N-dimethylaniline-4-yl moiety of 1 (bold lines, Figure 1) was thus proposed as a novel core scaffold for the development of a smart fluorescence probe which elicits a fluorescent signal only after target engagement. In particular, optimization of the fluorescence property as well as tauselectivity of 1 through structural elaboration of the 3,5dimethoxy-N,N-dimethylaniline-4-yl scaffold was anticipated to render a novel tau-specific smart NIRF probe. In this context, the characteristic donor-π-acceptor architecture of the smart NIRF Aβ probes (e.g., DANIR-2c and MCAAD-3) drew our attention because a combination of the 3,5-dimethoxy-N,Ndimethylaniline-4-yl moiety with the π-bridged acceptor part of DANIR-2c31 or MCAAD-332 (Figure 1) was hypothesized to provide the resulting probe molecules with favorable fluorescence properties. It was also of interest that combination of the tau-binding donor (3,5-dimethoxy-N,N-dimethylaniline4-yl) with the Aβ-binding, π-bridged acceptor could provide a selective molecular probe with the ability of detecting tau aggregates. In this study, a series of compounds with a novel donor-πacceptor architecture (2 and 3, Figure 1) were prepared and their fluorescence properties upon binding to tau fibrils were evaluated. Synthesis of the title compounds 2 and 3 are outlined in Scheme 1. First, Wittig reaction of 4-(dimethylamino)-2,6dimethoxybenzaldehyde (4) with (1,3-dioxolan-2-ylmethyl)triphenylphosphonium bromide afforded the conjugated aldehyde 5a after acid hydrolysis. Repeated subjection of 5a to the same reaction conditions afforded 5b with extended conjugation. The key intermediate 5b thus obtained was condensed with malononitrile to give 2 (59% yield). Similarly, condensation of 5b with various cyanoacetates produced 3a− 3h in 49%−77% yields. Evaluation of the synthesized molecules as smart fluorescence probes for tau fibrils was performed by mixing them with preaggregated tau (Figure S1, Supporting Information)19 in phosphate-buffered saline (PBS). To evaluate tau-specificity of the probe molecules, the same experiment was also performed in the presence of Aβ fibrils or bovine serum albumin (BSA), and the fluorescence spectra of the probes before and after B


Letter

ACS Chemical Neuroscience

minimal interference from serum albumin-binding for these probes (Figure 2). For an in-depth analysis of the synthesized compounds as smart NIRF probes for selective detection of tau fibrils, their fluorescence properties are compared in Table 1. The molecular probes prepared in this study showed weak or undetectable autofluorescence but they behaved as selective light-up fluorescent probes upon binding to the tau fibrils (Table 1). Thus, the compounds 2 and 3a−3h, sharing the common donor-π-acceptor architecture, showed fluorescence emission at the same NIR wavelength (λem = 650 nm) and large Stokes shifts (70−110 nm) upon mixing with tau, as well as Aβ, fibrils. Notably, the fluorescence intensity of every molecular probe increased significantly upon mixing with the tau fibrils (FItau = 49−159, Table 1). Although the probe molecules also showed fluorescence upon binding to Aβ fibrils, the fluorescence increases (FIAβ = 9−80, Table 1) were lower than those obtained from binding to tau fibrils. Taken together, the probe molecules (2 and 3a−3h) showed 1.6−5.7-fold tau selectivity over Aβ fibrils in emitting fluorescence signals (Table 1). Of particular interest is that the previously reported NIRF Aβ probe DANIR-2c31 was converted into a tau-selective fluorescence probe by aromatic ortho-dimethoxy substitution (2, selectivity index =3.5; Table 1). The ortho-dimethoxy congeners of MCAAD-332 also exhibited tau-selectivity and, among those, 3g and 3h were the most selective probes, with 5.7 and 3.8-fold selectivity, respectively (Table 1). Considering

Scheme 1. Synthesis of the Fluorescence Probes (2 and 3)

mixing are presented in Figure S2 (Supporting Information). The newly prepared molecular probes were not fluorescent by themselves but, as anticipated, their fluorescence was turned on at 650 nm upon mixing with the aggregated tau or Aβ fibrils. The molecular probes 2, 3a, 3g, and 3h were of considerable interest owing to their favorable fluorescence properties upon binding to the tau fibrils; much larger increases in fluorescence intensity were observed upon binding to tau fibrils than to Aβ fibrils (Figure 2). Significantly lower fluorescence was observed from the probes upon incubation with BSA, which indicates

Figure 2. Fluorescence spectra observed from the fluorescence probes (a) 2, (b) 3a, (c) 3g, and (d) 3h upon binding to tau aggregates (50 μM), Aβ fibrils (50 μM), and bovine serum albumin (BSA) (10 μM) in phosphate-buffered saline (PBS). C


Letter


Table 1. Fluorescence Properties of the Fluorescence Probes (2, 3a-3h) in the Presence of Tau, Aβ Fibrils, and Bovine Serum Albumin (BSA) FId

SIe

compd

ε (M−1cm−1)a

λex (nm)b

λem (nm)c

FItau

FIAβ

FIBSA

SIAβ

SIBSA

clogPf

2 3a 3b 3c 3d 3e 3f 3g 3h

39 075 12 792 9419 35 170 36 595 30 711 37 996 10 212 7173

580 540 550 550 540 540 550 550 550

650 650 650 650 650 650 650 650 650

119 159 124 107 116 106 82 49 108

34 80 59 55 69 67 52 9 28

17 19 22 45 29 85 6 3 10

3.5 2.1 2.1 2.0 1.7 1.6 1.6 5.7 3.8

7.0 8.4 5.6 2.4 4.0 1.2 13.7 16.3 10.8

2.83 4.96 5.67 5.67 5.67 4.88 4.70 3.46 4.13

a

Molar extinction coefficient (in dimethyl sulfoxide). bMaximum excitation wavelength of the probe (in PBS). cMaximum emission wavelength of the probe (in PBS). dFold Increase = fluorescence intensity of the probe bound to tau fibrils (or Aβ fibrils or BSA)/fluorescence intensity of the probe (unbound, free). eSelectivity index (SI) = fluorescence intensity of the probe upon binding to tau/fluorescence intensity of the probe upon binding to Aβ fibrils (or BSA). fclogP was calculated using ChemDraw Ultra 12.0.

concentration (26−66 ng/mL).36 However, Aβ fibrils could only be detected by 3g and 3h at concentrations above 500 ng/ mL and 125 ng/mL, respectively (Table 2). These results strongly indicate that the tau-selective fluorescence of 3g and 3h results from a specific recognition of tau aggregates but not from selective tau-binding affinity. In this context, it is worth to remind that 3g and 3h have the structural characteristics of the molecular rotors, which become fluorescent only if the intramolecular rotational relaxation about the donor−acceptor bond of the fluorophores is constrained.37−39 In other words, binding of the molecular rotors to the target proteins does not guarantee fluorescence emission, but they should adopt the proper conformation for fluorescence. Thus, we reasoned that the probes 3g and 3h bind to tau aggregates and Aβ fibrils in different conformations but only the tau-binding conformation allows fluorescence emission. To support this hypothesis, it was required to prove that 3g and 3h actually behave as molecular rotors. As the emission properties of the molecular rotors are strongly solvent-dependent and the solvent viscosity is the primary determinant of the fluorescent quantum yield, we investigated the molecular rotor properties of 3g and 3h by measuring the viscosity dependence of fluorescence intensity (Figure S7, Supporting Information). The results summarized in Figure 3 clearly demonstrate that the fluorescence emission intensities of the probes increase in a linear fashion with increasing solvent viscosity. Based on these observations, selective turn-on fluorescence of 3g and 3h upon binding to tau aggregates could be attributed to their environment-sensitive molecular rotor-like properties, which would allow them to adopt proper conformations for fluorescence in response to the typical microenvironment of the tau aggregates. The tau-selective NIRF probes 3g and 3h were then evaluated for the staining of tau proteins inside cells.29 Human neuroblastoma cells (SHSY-5Y) transfected with an expression vector for a tau-green fluorescent protein (GFP) fusion protein (pCMV6-htau40-GFP) were treated with the probes (3g and 3h), and the cells were monitored by confocal microscopy at 420 and 650 nm to detect fluorescence from GFP and the probes, respectively (Figure 4). Green fluorescence indicating intracellular expression of tau-GFP fusion protein was observed in the SHSY-5Y cells (“GFP” in Figure 4), while the probe molecules emitted red fluorescence (“Probe” in Figure 4). Merged images (“Merged” in Figure 4)

that 3g and 3h possess heterocycles (4-pyridyl and 2-furyl, respectively) while others have substituted phenyl groups at the R position (Scheme 1), a possible role of the heteroaromatic substituent in selective fluorescence of 3g and 3h upon engagement to the tau aggregates might be proposed. Moreover, ideal fluorescence probes should evade interference by serum albumin,33 and the probes 2, 3a, and 3f−3h showed very low fluorescence in the presence of BSA (FIBSA = 3−19, Table 1). Overall, four compounds (2, 3a, 3g, and 3h) exhibited promising properties as NIRF tau probes (λex = 650 nm, selectivity for tau over Aβ = 2.1−5.7). However, 2 and 3a exhibited unfavorable physicochemical properties; 2 was only sparingly soluble in PBS (solubility = ∼30 μM), while 3a was too lipophilic to cross the blood-brain barrier (clogP = 4.96, Table 1).34,35 As a result, two MCAAD-3 derivatives (3g and 3h) were chosen for further assessment of their fluorescence properties and tau-selectivity. The quantitative binding affinities of the selected probes to tau or Aβ fibrils were determined by in vitro saturation binding assay29 (Figures S3 and S4, Supporting Information), and the apparent binding constants (Kd) thus obtained are shown in Table 2. Interestingly, and unexpectedly, these probes showed Table 2. Apparent Binding Constants (Kd), Fluorescence Quantum Yields (Φ), and Lower Limit of Detection (LOD) of the NIRF Probes (3g and 3h) in the Presence of Tau or Aβ Fibrils Kd (μM)

Φ (%)

LOD (ng/mL)

compd

SI

tau

Aβ

tau

Aβ

tau

Aβ

3g 3h

5.7 3.8

0.89 1.50

0.50 1.02

5.1 16.5

3.8 2.5

31.3 31.3

500 125

similar binding affinities to tau (Kd = 0.89−1.50) and Aβ (Kd = 0.50−1.02) fibrils, which is in contrast to their tau-selective fluorescence properties. Fluorescence quantum yields (the emission efficiency of a given fluorophore; Φ) of the selected probes showed a positive correlation with their fluorescence properties, and 3g and 3h, respectively, exhibited 1.3- and 6.6fold higher quantum yields for tau aggregates than for Aβ fibrils (Table 2). A fluorescence titration study (Figures S5 and S6, Supporting Information) revealed that both 3g and 3h are sensitive enough to detect the tau protein (lower limit of detection =31.3 ng/mL, Table 2) at its physiological D


Letter


Figure 3. Dependence of fluorescence of (a) 3g and (b) 3h on solvent viscosity. Fluorescence emission was measured (λex = 550 nm/λem = 650 nm) in various solvents at 25 °C: (1) H2O, (2) DMSO, (3) glycerol, (4) ethylene glycol (EG), (5) methanol, (6) acetone, and (7) EG/glycerol (1:1, v/ v).

Figure 4. Confocal images of tau-GFP-transfected SH-SY5Y cells after treatment with (a) 3g and (b) 3h. Rows “GFP” and “Probe” show confocal images of the cells obtained at 520 and 650 nm, respectively. The row “Merged” shows merged images of “GFP” and “Probe”. Tau-GFP fusion protein and the probe molecules (3g and 3h) are shown in green and red, respectively, while their colocalization is shown in yellow.

showed colocalization of tau-GFP and the probes in yellow, which indicates specific detection of tau aggregates in live cells by 3g and 3h. Fluorescence labeling of tau aggregates was further examined by incubating tissue slides of human normal and AD brains (Table S1, Supporting Information) with phosphorylated tau (p-tau) antibody (Ser202/Thr205) followed by the probes, 3g or 3h. Fluorescent microscopy was then performed to show the brain tissues stained by the probes in red fluorescence (lower left panels, Figure 5) and those stained by the p-tau antibody in green fluorescence (lower middle panels, Figure 5), respectively. Notably, the merged images (lower right panels, Figures 5) showed colocalization of the probes and p-tau antibody in yellow fluorescence, which demonstrates selective fluorescent labeling of the tau fibrils by the probes 3g and 3h. Very limited

staining by the probes and p-tau antibody was observed in control normal brain tissues (upper rows, Figure 5). In summary, development of a novel tau-selective smart NIRF probe was attempted through combination of the previously identified core scaffold, 3,5-dimethoxy-N,N-dimethylanilin-4-yl moiety, with the characteristic donor-π-acceptor architecture of smart NIRF Aβ probes (DANIR-2c and MCAAD-3). A series of compounds (2 and 3) were prepared, and they were identified as “turn-on” NIRF probes for visual detection of tau aggregates and Aβ fibrils (λem = 650 nm, Stokes shifts =70−110 nm). In particular, probe molecules 3g and 3h showed significant selectivity toward tau aggregates (selectivity for tau over Aβ = 3.8 and 5.7) and they showed much higher fluorescence intensities upon binding to tau aggregates (FItau = 49 and 108) than to Aβ fibrils (FIAβ = 9 and 28) with negligible fluorescence in the presence of serum albumin. Unexpectedly, E



Letter

■

METHODS

■

ASSOCIATED CONTENT

Synthesis of (2E,4E,6E)-Pyridin-4-ylmethyl 2-cyano-7-(4(dimethylamino)-2,6-dimethoxyphenyl)hepta-2,4,6-trienoate (3g). Piperidine (3 μL, 0.03 mmol) was added to a solution of 5b (0.08 g, 0.31 mmol) and pyridine-4-ylmethyl 2-cyanoacetate (0.06 g, 0.36 mmol) in tetrahydrofuran (THF, 8 mL), and the reaction mixture was stirred at 50 °C for 12 h. After the reaction, the mixture was cooled to room temperature and extracted with EtOAc three times. The combined organic layers were dried over MgSO4 and the filtrate was concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (hexanes/acetone = 1:4) to produce the desired compound 3g (0.31 g, 0.74 mmol, 65% yield) as a purple solid: 1H NMR (400 MHz, acetone-d6) δ 8.61 (d, J = 5.6 Hz, 2H), 7.92 (d, J = 12.3 Hz, 1H), 7.44 (d, J = 15.0 Hz, 1H), 7.36−7.30 (m, 3H), 7.16 (dd, J = 14.0, 11.2 Hz, 1H), 6.72 (t, J = 13.5 Hz, 1H), 5.80 (s, 2H), 5.28 (s, 2H), 3.90 (s, 6H), 3.07 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 163.4, 161.4, 157.1, 156.3, 153.2, 150.1, 144.7, 137.9, 125.9, 122.6, 121.7, 115.9, 104.0, 96.2, 88.0, 64.9, 55.5, 40.3; FAB-MS calcd for C24H25N3O4 [M]+ 419, found 419. Synthesis of (2E,4E,6E)-Furan-2-ylmethyl 2-cyano-7-(4-(dimethylamino)-2,6-dimethoxyphenyl)hepta-2,4,6-trienoate (3h). Piperidine (2 μL, 0.03 mmol) was added to a solution of 5b (0.07 g, 0.25 mmol) and furan-2-ylmethyl 2-cyanoacetate (0.05 g, 0.29 mmol) in THF (10 mL), and the reaction mixture was stirred at 50 °C for 8 h. After the reaction, the mixture was cooled to room temperature and extracted with EtOAc three times. The combined organic layers were dried over MgSO4 and the filtrate was concentrated under reduced pressure. The residue was purified by column chromatography on silica gel (hexanes:acetone:CH2Cl2 = 4:1:3) to give the desired compound 3h (0.06 g, 0.14 mmol, 52% yield) as a purple solid: 1H NMR (400 MHz, CDCl3) δ 7.88 (d, J = 12.2 Hz, 1H), 7.43−7.26 (m, 3H), 7.10 (dd, J = 13.9, 11.2 Hz, 1H), 6.67 (t, J = 13.5 Hz, 1H), 6.46 (d, J = 2.6 Hz, 1H), 6.37 (d, J = 2.8 Hz, 1H), 5.80 (s, 2H), 5.21 (s, 2H), 3.88 (s, 6H), 3.06 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 163.3, 161.3, 156.7, 155.5, 153.0, 149.7, 143.3, 137.2, 126.0, 122.8, 115.9, 111.1, 110.6, 103.9, 97.2, 88.0, 58.9, 55.5, 40.3; FAB-MS calcd for C23H24N2O5 [M]+ 408, found 408.

Figure 5. Confocal fluorescence images of human normal (upper panels) and AD (lower panels) brain sections after treatment with the probes, (A) 3g and (B) 3h, are shown in red fluorescence (left panels), while the immunofluorescence staining of the same brain sections with p-Tau antibody are shown in green fluorescence (middle panels). Colocalization of the probes (3g and 3h) with the p-tau antibody are shown as yellow fluorescence in the merged images (right panels). Arrowheads (white) indicate colocalization foci of probes (3g and 3h) with p-tau (Ser202/Thr205). Scale bars (white): 10 um.

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acschemneuro.6b00174. Synthetic procedures and characterization of the new compounds, experimental details for the preparation of aggregated tau, fluorescence-based tau-binding assay, evaluation of the fluorescent properties of the fluorescence probes upon binding to the tau fibrils, detection of the tau aggregates in transfected SHSY-5Y cells, and fluorescence p-tau imaging in human AD brain sections (PDF)

3g and 3h showed similar binding affinities to tau and Aβ fibrils. However, 3g and 3h showed 1.3 to 6.6 times higher quantum yields upon binding to tau aggregates compared with binding to Aβ fibrils, and 4 to 16 times more sensitive detection of tau aggregates than Aβ fibrils. Additionally, in various solvents with different viscosity, 3g and 3h showed microenvironmentdependent fluorescence, which is a typical property of molecular rotors. Based on these observations, selective turnon fluorescence of 3g and 3h upon binding to tau aggregates could be attributed to their characteristic binding conformations in response to the typical microenvironment of the binding sites of tau aggregates. In addition, fluorescence imaging of tau-GFP-transfected SHSY-5Y cells with 3g and 3h confirmed that specific detection of tau aggregates in live cells is achievable by 3g and 3h. Fluorescence imaging in human AD brain sections further confirmed selecive fluorescence staining of tau fibrils by 3g and 3h, which were colocalized with the p-tau antibody. Taken together, combination of the 3,5-dimethoxy-N,N-dimethylanilin-4-yl moiety and the characteristic donor-π-acceptor architecture of MCAAD-3 resulted in molecular rotors 3g and 3h, which exhibited tauselective fluorescence properties in the NIR range.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Tel: +82-2-958-5157. *E-mail: [email protected]. Tel: +82-2-2049-6100. Fax: +82-2-454-8217. Author Contributions #

Y.S. and K.-s.P. contributed equally to this work. Y.S., K.-s.P., T.H., and M.K.K. synthesized the probes and analyzed their fluorescence properties. Y.J.H., J.L., and H.R. performed human brain section staining. H.C. and Y.C. designed the experiments and wrote the paper with input from coauthors. F


Letter

ACS Chemical Neuroscience Funding

(14) Maccioni, R. B., Farías, G., Morales, I., and Navarrete, L. (2010) The revitalized tau hypothesis on Alzheimer’s disease. Arch. Med. Res. 41, 226−2331. (15) Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T., and Hyman, B. T. (1992) Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42, 631−639. (16) Brier, M. R., Gordon, B., Friedrichsen, K., McCarthy, J., Stern, A., Christensen, J., Owen, C., Aldea, P., Su, Y., Hassenstab, J., Cairns, N. J., Holtzman, D. M., Fagan, A. M., Morris, J. C., Benzinger, T. L. S., and Ances, B. M. (2016) Tau and Aβ imaging, CSF measures, and cognition in Alzheimer’s disease. Sci. Transl. Med. 8, 338ra66. (17) Ballatore, C., Lee, V. M. Y., and Trojanowski, J. Q. (2007) Taumediated neurodegeneration in Alzheimer’s disease and related disorders. Nat. Rev. Neurosci. 8, 663−672. (18) Maruyama, M., Shimada, H., Suhara, T., Shinotoh, H., Ji, B., Maeda, J., Zhang, M. − R., Trojanowski, J. Q., Lee, V. M. − Y., Ono, M., Masamoto, K., Takano, H., Sahara, N., Iwata, N., Okamura, N., Furumoto, S., Kudo, Y., Chang, Q., Saido, T. C., Takashima, A., Lewis, J., Jang, M. K., Aoki, I., Ito, H., and Higuchi, M. (2013) Imaging of Tau Pathology in a Tauopathy Mouse Model and in Alzheimer Patients Compared to Normal Controls. Neuron 79, 1094−1108. (19) Fodero-Tavoletti, M. T., Okamura, N., Furumoto, S., Mulligan, R. S., Connor, A. R., McLean, C. A., Cao, D., Rigopoulos, A., Cartwright, G. A., O’Keefe, G., Gong, S., Adlard, P. A., Barnham, K. J., Rowe, C. C., Masters, C. L., Kudo, Y., Cappai, R., Yanai, K., and Villemagne, V. L. (2011) 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer’s disease. Brain 134, 1089−1100. (20) Okamura, N., Furumoto, S., Fodero-Tavoletti, M. T., Mulligan, R. S., Harada, R., Yates, P., Pejoska, S., Kudo, Y., Masters, C. L., Yanai, K., Rowe, C. C., and Villemagne, V. L. (2014) Non-invasive assessment of Alzheimer’s disease neurofibrillary pathology using 18 F-THK5105 PET. Brain 137, 1762−1771. (21) Harada, R., Okamura, N., Furumoto, S., Furukawa, K., Ishiki, A., Tomita, N., Hiraoka, K., Watanuki, S., Shidahara, M., Miyake, M., Ishikawa, Y., Matsuda, R., Inami, A., Yoshikawa, T., Tago, T., Funaki, Y., Iwata, R., Tashiro, M., Yanai, K., Arai, H., and Kudo, Y. (2015) [18F]THK-5117 PET for assessing neurofibrillary pathology in Alzheimer’s disease. Eur. J. Nucl. Med. Mol. Imaging 42, 1052−1061. (22) Chien, D. T., Szardenings, A. K., Bahri, S., Walsh, J. C., Mu, F., Xia, C., Shankle, W. R., Lerner, A. J., Su, M. Y., Elizarov, A., and Kolb, H. C. (2014) Early clinical PET imaging results with the novel PHFtau radioligand [F18]-T808. J. Alzheimer's Dis. 38, 171−184. (23) Small, G. W., Siddarth, P., Kepe, V., Ercoli, L. M., Burggren, A. C., Bookheimer, S. Y., Miller, K. J., Kim, J., Lavretsky, H., Huang, S. C., and Barrio, J. R. (2012) Prediction of cognitive decline by positron emission tomography of brain amyloid and tau. Arch. Neurol. 69, 215− 222. (24) Xia, C. F., Arteaga, J., Chen, G., Gangadharmath, U., Gomez, L. F., Kasi, D., Lam, C., Liang, Q., Liu, C., Mocharla, V. P., Mu, F., Sinha, A., Su, H., Szardenings, A. K., Walsh, J. C., Wang, E., Yu, C., Zhang, W., Zhao, T., and Kolb, H. C. (2013) [18F]T807, a novel tau positron emission tomography imaging agent for Alzheimer’s disease. Alzheimer's Dementia 9, 666−676. (25) Gao, X., Cui, Y., Levenson, R. M., Chung, L. W., and Nie, S. (2004) In vivo cancer targeting and imaging with semiconductor quantum dots. Nat. Biotechnol. 22, 969−976. (26) Harbater, O., and Gannot, I. (2014) Fluorescent robes concentration estimation in vitro and ex vivo as a model for early detection of Alzheimer’s disease. J. Biomed. Opt. 19, 127007. (27) Ojida, A., Sakamoto, T., Inoue, M. − A., Fujishima, S. − H., Lippens, G., and Hamachi, I. (2009) Fluorescent BODIPY-based Zn(II) complex as a molecular probe for selective detection of neurofibrillary tangles in the brains of Alzheimer’s disease patients. J. Am. Chem. Soc. 131, 6543−6548. (28) Boländer, A., Kieser, D., Voss, C., Bauer, S., Schön, C., Burgold, S., Bittner, T., Hölzer, J., Heyny-von Haußen, R., Mall, G., Goetschy, V., Czech, C., Knust, H., Berger, R., Herms, J., Hilger, I., and Schmidt, B. (2012) Bis(arylvinyl)pyrazines, -pyrimidines, and − pyridazines as

This study was supported by a grant from the Korean Health Technology R&D project, Ministry of Health & Welfare, Republic of Korea (HI14C2341). Additional funding was provided by the Korea Institute of Science and Technology (KIST) Institutional Program (2E26650 and 2E26663). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This paper was written as part of Konkuk University’s research support program for its faculty on sabbatical leave in 2014.

■

REFERENCES

(1) Jack, C. R., Albert, M. S., Knopman, D. S., McKhann, G. M., Sperling, R. A., Carrillo, M. C., Thies, B., and Phelps, C. H. (2011) Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer's Dementia 7, 257−262. (2) McKhann, G. M., Knopman, D. S., Chertkow, H., Hyman, B. T., Jack, C. R., Kawas, C. H., Klunk, W. E., Koroshetz, W. J., Manly, J. J., Mayeux, R., Mohs, R. C., Morris, J. C., Rossor, M. N., Scheltens, P., Carrillo, M. C., Thies, B., Weintraub, S., and Phelps, C. H. (2011) The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer's Dementia 7, 263−269. (3) Albert, M. S., DeKosky, S. T., Dickson, D., Dubois, B., Feldman, H. H., Fox, N. C., Gamst, A., Holtzman, D. M., Jagust, W. J., Petersen, R. C., Snyder, P. J., Carrillo, M. C., Thies, B., and Phelps, C. H. (2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on AgingAlzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer's Dementia 7, 270−279. (4) Sperling, R. A., Aisen, P. S., Beckett, L. A., Bennett, D. A., Craft, S., Fagan, A. M., Iwatsubo, T., Jack, C. R., Kaye, J., Montine, T. J., Park, D. C., Reiman, E. M., Rowe, C. C., Siemers, E., Stern, Y., Yaffe, K., Carrillo, M. C., Thies, B., Morrison-Bogorad, M., Wagster, M. V., and Phelps, C. H. (2011) Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer's Dementia 7, 280−292. (5) Braak, H., and Braak, E. (1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 82, 239−259. (6) Braak, H., and Braak, E. (1997) Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol. Aging 18, 351−357. (7) Goedert, M., and Spillantini, M. G. (2006) A century of Alzheimer’s disease. Science 314, 777−781. (8) Ariza, M., Kolb, H. C., Moechars, D., Rombouts, F., and Andrés, J. I. (2015) Tau positron emission tomography (PET) imaging: Past, present, and future. J. Med. Chem. 58, 4365−4382. (9) Sabri, O., Seibyl, J., Rowe, C., and Barthel, H. (2015) Betaamyloid imaging with florbetaben. Clin. Transl. Imaging 3, 13−26. (10) Yang, L., Rieves, D., and Ganley, C. (2012) Brain amyloid imaging  FDA approval of Florbetapir F18 injection. N. Engl. J. Med. 367, 885−887. (11) Hatashita, S., Yamasaki, H., Suzuki, Y., Tanaka, K., Wakebe, D., and Hayakawa, H. (2014) [18F]Flutemetamol amyloid-beta PET imaging compared with [11C]PIB across the spectrum of Alzheimer’s disease. Eur. J. Nucl. Med. Mol. Imaging 41, 290−300. (12) Villemagne, V. L., Furumoto, S., Fodero-Tavoletti, M., Harada, R., Mulligan, R. S., Kudo, Y., Masters, C. L., Yanai, K., Rowe, C. C., and Okamura, N. (2012) The challenge of tau imaging. Future Neurol. 7, 409−421. (13) U.S. Food and Drug Administration. www.fda.gov. G


Letter

ACS Chemical Neuroscience imaging agents for tau fibrils and β-amyloid plaques in Alzheimer’s disease models. J. Med. Chem. 55, 9170−9180. (29) Park, K. − S., Seo, Y., Kim, M. K., Kim, K., Kim, Y. K., Choo, H., and Chong, Y. (2015) A curcumin-based molecular probe for nearinfrared fluorescence imaging of tau fibrils in Alzheimer’s disease. Org. Biomol. Chem. 13, 11194−11199. (30) Ono, M., and Saji, H. (2015) Recent advances in molecular imaging probes for β-amyloid plaques. MedChemComm 6, 391−402. (31) Cui, M., Ono, M., Watanabe, H., Kimura, H., Liu, B., and Saji, H. (2014) Smart near-infrared fluorescence probes with donoracceptor structure for in vivo detection of β-amyloid deposits. J. Am. Chem. Soc. 136, 3388−3394. (32) Fu, H., Cui, M., Tu, P., Pan, Z., and Liu, B. (2014) Evaluation of molecules based on the electron donor−acceptor architecture as nearinfrared β-amyloidal-targeting probes. Chem. Commun. 50, 11875− 11878. (33) Ran, C., Xu, X., Raymond, S. B., Ferrara, B. J., Neal, K., Bacskai, B. J., Medarova, Z., and Moore, A. (2009) Design, synthesis, and testing of difluoroboron-derivatized curcumins as near-infrared probes for in vivo detection of amyloid-β deposits. J. Am. Chem. Soc. 131, 15257−15261. (34) Ametamey, S. M., Honer, M., and Schubiger, P. A. (2008) Molecular imaging with PET. Chem. Rev. 108, 1501−1516. (35) Clark, D. E. (2005) Computational prediction of blood-brain barrier permeation. Annu. Rep. Med. Chem. 40, 403−415. (36) Drubin, D. G., Feinstein, S. C., Shooter, E. M., and Kirschner, M. W. (1985) Nerve growth factor-induced neurite outgrowth in PC12 cells involves the coordinate induction of microtubule assembly and assembly-promoting factors. J. Cell Biol. 101, 1799−1807. (37) Maskevich, A. A., Stsiapura, V. I., Kuzmitsky, V. A., Kuznetsova, I. M., Povarova, O. I., Uversky, V. N., and Turoverov, K. K. (2007) Spectral properties of thioflavin T in solvents with different dielectric properties and in a fibril-incorporated form. J. Proteome Res. 6, 1392− 1401. (38) Friedhoff, P., Schneider, A., Mandelkow, E. M., and Mandelkow, E. (1998) Rapid assembly of Alzheimer-like paired helical filaments from microtubule-associated protein tau monitored by fluorescence in solution. Biochemistry 37, 10223−10230. (39) Voropai, E. S., Samtsov, M. P., Kaplevskii, K. N., Maskevich, A. A., Stepuro, V. I., Povarova, O. I., Kuznetsova, I. M., Turoverov, K. K., Fink, A. L., and Uverskii, V. N. (2003) Spectral properties of thioflavin T and its complexes with amyloid fibrils. J. Appl. Spectrosc. 70, 868− 874.

H


A Smart Near-Infrared Fluorescence Probe for Selective Detection of

Recommend Documents