Real-Time Analysis of Binding Events between Different Aβ1–42

Article pubs.acs.org/ac

Real-Time Analysis of Binding Events between Different Aβ1−42 Species and Human Lilrb2 by Dual Polarization Interferometry Tao Hu,†,‡ Shuang Wang,†,‡ Chuanxia Chen,† Jian Sun,*,† and Xiurong Yang*,†,‡ †

State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, Jilin 130022, China ‡ Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, China S Supporting Information *

ABSTRACT: Abnormal accumulation of 42-residue amyloid-β (Aβ1−42) within the brain triggers the pathogenesis of Alzheimer’s disease (AD). In this paper, we use a dual polarization interferometry (DPI) tool to evaluate the binding events of various Aβ1−42 species such as monomeric Aβ1−42, low molecular weight Aβ1−42 oligomer (LMW Aβ1−42), and high molecular weight Aβ1−42 oligomer (HMW Aβ1−42) with extracellular D1D2 domain of lilrb2 (ED1D2L) receptor that has been proved to be associated with AD. Based on the real-time binding information provided by DPI, the association rate (ka) of ED1D2L receptor with monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42 is individually determined to be 2.85 × 104, 4.52 × 104, and 1.34 × 105 M−1·s−1, and meanwhile, the dissociation rate (kd) corresponds to 1.79 × 10−2, 2.09 × 10−2, and 5.34 × 10−4 s−1, respectively. By analysis of the kinetic parameters of ka and kd values, we discovery that the HMW Aβ1−42 exhibits the fastest rate for ED1D2L receptor in the association phrase, and HMW Aβ1−42 likewise shows the highest affinity with ED1D2L receptor during the dissociation period in contrast to LMW Aβ1−42 and monomeric Aβ1−42. Our findings significantly reveal the different binding behaviors among them from the perspective of kinetics aspect, by which we could indirectly elucidate the malicious impacts in the process of AD triggered by HMW Aβ1−42. Strikingly, this work offers a new exciting clue to explore the dynamic properties associated with interactions of various Aβ1−42 species with other targets and hopefully contributes to drug discovery and screen in the future.

A

oligomers within the brain have been extensively proposed to be the major culprit of AD in past decades,16−19 further mechanisms about why different morphologies of Aβ1−42 exert differential neurotoxicity through binding to corresponding receptors remain obscure. As is well-known that the characterization of protein−ligand interaction plays a pivotal role in the biology system. The precise kinetics parameters, particularly binding constants, are of significant importance for the behavior of biological processes. Their determinations are herein vital to understand the systemic mechanisms of disease evolvement.20,21 Perhaps the most pressing need concerning above-mentioned AD issue is a robust dynamics knowledge about the interactions between different Aβ1−42 species and several cell surface receptors, by which they will greatly contribute to accelerating our in-depth understanding of the etiology of AD and driving the development of drug discovery and screen in terms of preventing toxicity of Aβ1−42 species. Numerous neurotoxic effects have been significantly revealed due to the interactions of Aβ oligomers with several receptors such as PrPC, NMDA receptor, Neuroligin and so forth.22−24 Of particular much interest now is the PrPC, which has been

lzheimer’s disease (AD) is a severe neurodegenerative disease leading to progressive loss of cognitive function, brain tissue destruction, and behavioral abilities.1−3 The major development and pathogenesis of AD are attributed to amyloid deposits (extracellular plaques containing aggregated Aβ) and neurofibrillary tangles (intracellular tangles consisting of hyperphosphorylated tau).4,5 Over the last decades, the amyloid hypothesis has attracted significant attention and been viewed as the predominant framework for research in AD. Successive cleavages of amyloid precursor protein (APP) by β- and γsecretase produce a range of Aβ peptides from 39 to 43 amino acid residues.6,7 It is reported that the hydrophobic nature of Aβ, especially Aβ1−42, is prone to self-association and aggregation and can form a series of species from low molecular weight Aβ1−42 oligomer (LMW Aβ1−42, 2−4 mer) to high molecular weight Aβ1−42 oligomer (HMW Aβ1−42, 14− 48 mer), all of which seem to trigger differential synapse impairment and neurotoxicity in AD.8−12 For instance, HMW Aβ1−42 generally generates excessive reactive oxygen species (ROS) formation in the hippocampal region and transient cognitive deficit, whereas LMW Aβ1−42 gives rise to persistent cognitive impairment.13 As for the identical receptor, LMW Aβ1−42 has been discovered to obviously bind to cellular prion protein (PrPC) receptor, causing acute synaptic plasticity impairment in AD, while HMW Aβ1−42 might be vital for downstream signal cascades transduction.14,15 Although Aβ1−42 © XXXX American Chemical Society

Received: December 13, 2016 Accepted: January 20, 2017 Published: January 20, 2017 A

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

NaCl, 3 mM Na2HPO4·7H2O, pH 7.4) to 100 μM and incubated at 4 °C for 24 h. HMW Aβ1−42 was prepared in PBS to the final concentration of 100 μM and incubated at 25 °C for 24 h, then centrifuged at 16000 g for 20 min, and the supernatant was collected as HMW Aβ1−42.15,31 Atomic Force Microscopy Characterization. A 20 μL aliquot of sample (5 μM) was placed onto freshly cleaved muscovite mica and allowed to dry. Data were acquired from multimode atomic force microscope (Veeco Instruments, U.S.A.).32 Immunoprecipitation. Human embryonic kidney (HEK)293T cells were cultured in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Life Technologies) and 1% antibiotics at 37 °C under a 5% CO2 condition. Immunoprecipitation (IP) assay was performed as described previously.33 Briefly, HEK-293T cells were transfected with the indicated ED1D2L receptor-GFP plasmid using Lipofectamine TM 2000. Meanwhile, 40 μL of protein A/ G PLUS-agarose was incubated with 2 μL of anti-GFP overnight for 6 h at 4 °C on an end-over-end rotator. After being cultured for 36 h, the cells were lysed in IP lysis buffer (Beyotime). The lysates were centrifuged, and the supernatants together with the three Aβ1−42 species (400 nM) were individually added into the precleared tubes, and incubated for 6 h at 4 °C and then washed extensively. Bound Aβ1−42 species were analyzed by Western blotting using antibody (4G8), as well as the GFP. Western Blotting. a 10 μL aliquot of sample was separated by SDS-PAGE (Tanon, Shanghai, China) and electrically transferred to Hybond Enhanced Chemiluminescence Nitrocellulose membrane (Amersham Biosciences, Freiburg). After utilizing the primary antibody, the membrane was fully washed, and the antibody−antigen complexes were then probed using HRP-conjugated secondary antibody. Blots were visualized using enhanced chemiluminescence (Kodak, Rochester, NY). Dual Polarization Interferometry Experiments. The DPI measurements were monitored in real time using an AnaLight Bio200 DPI system (Farfield Scientific Ltd., Crew, U.K.). The details of the technique and theory are given in the Supporting Information. Prior to experiments, the sensor chip was calibrated using an 80% (w/w) ethanol/water solution and ultrapure water to calibrate the waveguide. To create the probe surface, originally amine-functionalized silicon oxynitride AnaChip was modified by 200 μL of His-tagged ED1D2L receptor (0.1 mg/mL) via BS3 linker. Then 200 μL of Aβ1−42 samples with different concentrations were independently injected accompanied with the running buffer (25 mM TrisHCl, 150 mM NaCl, pH 7.4) at a 20 μL/min flow rate. Moreover, 15 mM glycine (pH 2.0) was performed for 2 min after each cycle of ED1D2L receptor/Aβ1−42 binding to regenerate the probe surface. The phase changes of TE and TM were recorded in real time. The measured DPI signals (TE and TM) then can be unambiguously resolved into the thickness, density (proportional to refractive index according to the following eq 3) and mass of the immobilized layer according to the De Feijter’s equations as follows:34,39

extensively investigated since the original report in 2009.15 Very recently, a leukocyte immunoglobulin-like receptor B2 (lilrb2), which is initially found to express in nervous system,25 is responsible for the emergency of toxic effects in AD.26 Given high praise results demonstrate that both monomeric Aβ1−42 and HMW Aβ1−42 show high affinity with extracellular D1D2 domain of lilrb2 (ED1D2L) receptor,27 where HMW Aβ1−42 leads to synaptic plasticity impairment in AD progress.26 However, thus far, further investigations are barely reported with regard to detailed dynamics information on binding properties. In this work, we employ a powerful dual polarization interferometry (DPI) technique to track the real-time binding process between different morphologies of Aβ1−42 and ED1D2L receptor. Compared with other common techniques, such as NMR, X-ray crystallography, and NR, DPI exhibits its own unique advantages, such as untagged reagents, real-time binding information every 20 ms with mass (0.1 pg/mm2), thickness (0.01 nm), and density,28,29 offering higher sensitivity and accuracy. It is reported that both monomeric Aβ1−42 and HMW Aβ1−42 show high affinity with ED1D2L receptor.26 Besides, we further reveal that LMW Aβ1−42 also binds to ED1D2L receptor by immunoprecipitation techniques for the first time. However, the interactions of ED1D2L receptor with above three morphologies of Aβ1−42 are only qualitatively proved through biological tools. Here, based on the new insights into biomolecule behavior onto the biochip surface, DPI not only monitors their real-time structural information on the binding events, but also provides kinetic parameters and binding constants.30 By analysis of the binding curves from changes in mass, the association rate (ka) and dissociation rate (kd) between various Aβ1−42 species and ED1D2L receptor are precisely determined, together with dissociation constant (KD). These outcomes will guide us to understand how HMW Aβ1−42 plays a toxin role in AD from the perspective of kinetics aspect and hopefully contribute to drug discovery and screen. Meanwhile, an accurate binding model between Aβ1−42 species and ED1D2L receptor-mediated cognitive impairment is proposed in order to elucidate the AD mechanisms and construct a biological system.

■

EXPERIMENTAL SECTION Chemicals and Materials. Purified synthetic human βamyloid 1−42 (Aβ 1−42 ) was purchased from Anaspec (Fremont, CA); Antibody to Aβ1−42 (4G8, Covance); Protein A/G PLUS-Agarose and anti-GFP (Santa Cruz, CA); BS3 (Thermo Scientific), bovine serum albumin, and ethanolamine (Sigma); Dulbecco modified Eagle medium (DMEM), fetal bovine serum (FBS), and 1% antibiotics were purchased from Life Technologies (Grand Island, NY, U.S.A.); Lipofectamine 2000 (Invitrogen); Recombinant His-tagged ED1D2L receptor and ED1D2L receptor-GFP plasmid were purchased from Sangon Biotechnology Co. Ltd. (Shanghai, China); Unmodified silicon oxynitride AnaChip (FB 80, Farfield Scientific Ltd.); Ultrapure water (Milli-Q synthesis, Millipore Inc., Bedford, MA) was used throughout. All the buffer solutions were filtered and degassed before use. Aβ1−42 Sample Preparation. Purified synthetic Aβ1−42 peptide was first dissolved in HFIP at the concentration of 1 mg/mL and stored at −20 °C. Before use, the solvent HFIP was removed by evaporation under gentle nitrogen. Monomeric Aβ1−42 was dissolved in NaOH (0.5 mM) and sonicated (30 s). LMW Aβ1−42 was prepared in PBS (1 mM KH2PO4, 155 mM

mL = τL(nL − nbuffer)/(dn/dc)

(1)

ρL = (nL − nbuffer)/(dn/dc)

(2)

Then mL = τL·ρL B

(3) DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry where ρL is density (g/cm3), mL is mass (ng/mm2), and τL is thickness (nm). nL and nbuffer represent the refractive index of layer and buffer, respectively. dn/dc is the refractive-index increment of the solution, which is often a definite value of 0.186. Data collection and analysis were based on the abovementioned calculated parameters and performed using the AnaLight software. The detailed dynamics parameters then can be evaluated. Linear Analysis of Kinetic Parameters. Evaluation of the association and dissociation rate constants were carried out by the following equations:35−37 dR /dt = kaCR max ·(kaC + kd)R

monomeric Aβ1−42 was about 4 kDa; meanwhile, LMW Aβ1−42 and HMW Aβ1−42 individually showed an apparent band at the position of 10−15 kDa (about 2−4 mer) and 50−100 kDa (about 14−48 mer). All the prepared Aβ1−42 species were consistent with previous reports.32,38 According to IP results (Figure S2), furthermore, it was obviously revealed that GFP protein was not capable of binding to the three species, displaying no bands of monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42. However, the (GFP)-ED1D2L receptor fusion protein showed apparent bands at the position of 50−100 and 4 kDa, especially 10−15 kDa, suggesting that LMW Aβ1−42 likewise enabled the interaction with the ED1D2L receptor, together with monomeric Aβ1−42 and HMW Aβ1−42. Immobilization of ED1D2L Receptor Monitored by DPI. The principle of exploring real-time binding behaviors of various morphologies of Aβ1−42 with ED1D2L receptor was displayed in Scheme 1. In order to acquire the recombinant

(4)

where ka and kd represent the rate constant of association and dissociation, respectively; C is the concentration of various Aβ1−42 analytes, Rmax is the maximum binding response of ED1D2L receptor, and R is the relative response at time t. According to eq 1, a plot of dR/dt versus R should be a straight line with a slope (Y value). Y = kaC + kd

Scheme 1. Schematic Representation of DPI Probing Surface for the Real-Time Analysis of Binding Events

(5)

Thus, ka and kd can be obtained by performing linear fitting based on different concentrations of monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42 through eq 2.

■

RESULTS AND DISCUSSION Characterization of Prepared Aβ1−42 Species and Study of LMW Aβ1−42-ED1D2L Receptor Interaction. Very recently, it was reported that HMW Aβ1−42 and monomeric Aβ1−42 exhibited high affinity with ED1D2L receptor.26 We therefore wondered that whether LMW Aβ1−42 was capable of interacting with ED1D2L receptor. To clarify this hypothesis, immunoprecipitation (IP) assay was performed through the bindings of green fluorescent protein (GFP)-ED1D2L receptor fusion protein to different morphologies of Aβ1−42, such as monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42, all of which were prepared according to previously described methods.15,31 The various morphologies of Aβ1−42 characterized by atomic force microscopies (AFM) and Western blotting were displayed in the Figures 1 and S1, respectively. As shown in Figure 1, the monomeric Aβ1−42 was shown to be about 0.8 nm, while both LMW Aβ1−42 and HMW Aβ1−42 were spherical with heights of 1−1.5 nm and ∼17 nm. Derived from Figure S1, we clearly discovered that the band of

His-tagged ED1D2L receptor, we constructed a pET-32a plasmid containing ED1D2L gene and purified it as a secreted protein, showing an apparent molecular weight under denaturing conditions of 22 kDa according to SDS-PAGE characterization (Figure S3). The ED1D2L receptor was then precisely coated onto amino-modified AnaChip sensor (ED1D2L receptor was covalently coupled to the chip via amine coupling using BS3 linker, a homobifunctional crosslinker that contains amine reactive functionalities). Moreover, both BSA and ethanolamine were introduced to block the sensorgram and several remaining amine for the purpose of eliminating nonspecificity adsorption.39−41 Thus, the three Aβ1−42 species were individually captured by ED1D2L receptor and produced differential response signal. The whole in situ formation process was monitored in real time by DPI (Figure 2). As can be seen in Figure 2, the initial loading rate of BS3 onto the sensorgram was quick, during which the mass and thickness sharply increased, while the density showed an opposite phenomenon and then kept a saturated state for several minutes, indicating that large numbers of BS3 were bound to amino-modified AnaChip sensor. When the running buffer was reintroduced to the sensorgram, the thickness and mass considerably decreased, while the density increased, suggesting the majority of BS3 were flowed away, whose binding behavior was in line with a previous report.29 For the immobilization process of His-tagged ED1D2L receptor, BSA, and ethanolamine, they likewise displayed the similar phenomenon and mechanism with BS3. Besides, to clearly quantify the extent of layer real-time changes,

Figure 1. AFM images of monomeric Aβ1−42 (A), LMW Aβ1−42 (B), and HMW Aβ1−42 (C), cross-sectional depict profile corresponding to down picture, respectively. Each image is 2 × 2 μm. C

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Interaction between Three Morphologies of Aβ1−42 and ED1D2L Receptor by DPI. Before achieving real-time binding process among them, probing surface was initially subjected to individual control groups (0 nM, background signal), and then various morphologies of Aβ1−42 with different concentrations were injected into the probe surface. Naturally, several detailed measurements of mass, thickness, and density could be simultaneously collected. As apparently shown in Figure 3A−C, loading mass increased sharply after the addition of monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42, and then introduced targets gradually bound to ED1D2L receptor, respectively. After 600 s, the baseline returned to equilibrium state showing different location sites, strongly indicating that a small amount of three morphologies of Aβ1−42 indeed bound to ED1D2L receptor, respectively. In addition, the variation trend of thickness was similar to that of the mass, and the density exhibited an opposite curve (Figure S5), both of which could indirectly confirm their interactions on the basis of previously reported results.28,43 By comparing the real-time mass variation curves, however, it was noteworthy that the extents of HMW Aβ1−42 and LMW Aβ1−42 mass changes were much greater than that of monomeric Aβ1−42, which could be ascribed to the interactions between per ED1D2L receptor molecule and Aβ1−42 of each species with different mass weight. Naturally, we intended to further illustrate the morphologies of bound Aβ1−42 onto ED1D2L receptor. Prior to investigation, we should explore individual saturation concentration of the three Aβ1−42 species. This is because real-time parameters at nonsaturation state may be greatly influenced by individual binding rates, making them untrue to evaluate bound species, and only the saturation state suggested that the bound amounts of three morphologies of Aβ1−42 would be completely equal. Derived from Figure S6, the saturation concentration of HMW Aβ1−42 (4.4 μM) was larger than monomeric Aβ1−42 (1.7 μM) and LMW Aβ1−42 (2.3 μM),

Figure 2. (A) DPI-based real-time measurements of mass (black), thickness (red), and density (blue) for the whole immobilization process of BS3 (a), His-tagged ED1D2L receptor (b), BSA (c), and ethanolamine (d) on the sensorgram. Changes of density (B) and thickness (C) values as a function of the mass loading during the entire immobilization process of each layer.

the changes of thickness or density were also plotted as a function of the loading mass of BS3, BSA, ED1D2L receptor, and ethanolamine (inset, Figure 2). Those changes could help us know in detail the real-time process of ED1D2L receptor immobilization. More to the point, the mass, thickness, and density values of average immobilization layer characteristics were shown in Table S1. By analysis of the detailed information on layer structure, the thickness of ED1D2L receptor immobilized was close to the extended length of 2 nm. As previously reported in the crystal structure,42 the ED1D2L receptor had approximate dimensions of 62 × 35 × 18 Å. We therefore deduced that the ED1D2L receptor adopted a “flat-on (back)” model (Figure S4) on the sensorgram. Together, these results demonstrated that the layer of mounted ED1D2L receptor for Aβ1−42 species was successfully constructed.

Figure 3. (A−C) Real-time loading mass changes of the interactions of ED1D2L receptor with monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42 (arrow directions: 0, 100, 250, 500, 750, and 1000 nM), respectively. (D−F) Plot of −(kaC + kd) vs C of monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42, attached inset models represent the association rate and dissociation rate corresponding to (A), (B), and (C), respectively. D

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry which could be attributed to the fact that the equivalent amount of HMW Aβ1−42 was prepared using more monomeric Aβ1−42. Under this condition, we began to estimate respective real-time saturation curve through the ratio of LMW Aβ1−42 to monomeric Aβ1−42 and HMW Aβ1−42 to monomeric Aβ1−42. Outcomes of Figure 4A demonstrated that the approximately

Table 1. Determining Data from Kinetics Analysis of Three Aβ1‑42 Species Binding to ED1D2L Receptor of ka (×105 M−1 s−1) and kd (×10−2 s−1), KD (nM) monomeric Aβ1−42 LMW Aβ1−42 HMW Aβ1−42

ka

kd

KD

0.29 ± 0.01 0.45 ± 0.01 1.34 ± 0.14

1.79 ± 0.05 2.09 ± 0.04 0.05 ± 0.002

617 ± 15.21 464 ± 9.42 37.3 ± 0.12

monomeric Aβ1−42 in term of binding rates (ka, Figure 3D−F, inset). Meanwhile, dissociation rate constant can be gained accompanying with ka, and the results were displayed in Table 1. The dissociation rate of HMW Aβ1−42 (5.34 × 10−4· s−1) for ED1D2L receptor was nearly 30-fold lower than that of monomeric Aβ1−42 (1.79 × 10−2 s−1) and LMW Aβ1−42 (2.09 × 10−2 s−1) (kd, Figure 3D−F, inset). The considerably slow dissociation rate of HMW Aβ1−42-ED1D2L receptor indicated that HMW Aβ1−42 showed the strongest affinity with ED1D2L receptor. In this case, the lowest kd of HMW Aβ1−42 was detrimental because a longer time for tight combination with ED1D2L receptor could lead to abnormal downstream signal transduction. Additionally, the association constant (KD) was calculated according to the kd and ka (Table 1), which was much smaller than previously reported.26 Model of Three Aβ1−42 Species with ED1D2L Receptor. Many findings have generated significant advances in knowing of the neurotoxicity triggered by various Aβ1−42 oligomers species, and different Aβ1−42 species are capable of leading to differential neurotoxicity. As far as we all know, however, all these studies were extensively explored in vivo of mouse or relative cell lines by comparing the extent of biological effects or qualitatively demonstrating their interactions. Here, our work has significantly uncovered the binding behaviors of three species with ED1D2L receptor from the perspective of kinetics aspect. On the basis of detailed analysis of association rate and dissociation rate, we proposed a reasonable binding model of three Aβ1−42 species with ED1D2L receptor in vivo, by which the feasible interaction processes can be clearly depicted (Figure 5). During the association period, the binding rates decreased in the order of HMW Aβ1−42 > LMW Aβ1−42 > monomeric Aβ1−42. Namely, HMW Aβ1−42 was prior to

Figure 4. (A) Real-time ratio of binding mass change of LMW Aβ1−42/ monomeric Aβ1−42, HMW Aβ1−42/monomeric Aβ1−42, and HMW Aβ1−42/LMW Aβ1−42. (B) Real-time ratio of binding thickness change of LMW Aβ1−42/monomeric Aβ1−42, HMW Aβ1−42/monomeric Aβ1−42, and HMW Aβ1−42/LMW Aβ1−42.

morphology of LMW Aβ1−42 was about 3 mer, while the HMW Aβ1−42 was about 35 mer. Meanwhile, the ratio of thickness change was also shown in Figure 4B, which was in line with above-mentioned AFM results. Evaluation of Kinetic Information between Aβ1−42 Species and ED1D2L Receptor. Generally, the duration of a receptor−ligand complex will be influenced by the association rate (ka) of ligand and the dissociation rate (kd) of binary complex.44−46 Promoted by this notion, we focused on investigating the durations of the complex between different Aβ1−42 species and ED1D2L receptor. Derived from real-time mass changes (Figure 3A−C), it was clearly found that the whole binding process suggested a reversibility of the interactions of three Aβ1−42 species with ED1D2L receptor. The whole real-time detailed process from the introduction of samples to the dissociation phase was schematically depicted in Figure S7. According to aforementioned robust kinetics information provided by DPI, we moved to the significant event of evaluation of association rate (ka) and dissociation rate (kd), which were bleak in current scenario. The accurate and reliable measurements of ka and kd could be acquired by linear analysis according to the eq 2 in the linear analysis of kinetic parameters of the Experimental Section. As can be seen in Figure 3A−C, every 50 s was selected as a time point from 200 to 400 s, at which a certain R value could be obtained (detailed data shown in Tables S2−S4). The slopes (Y value) were acquired by a plot of dR/dt versus R (eq 4), respectively (Figures S8−S10). Therefore, ka value now could be got by performing linear fitting based on the four different concentrations (the actual concentrations of HMW Aβ1−42 and LMW Aβ1−42 should be about 1/35 and 1/2 of monomeric Aβ1−42 according to Figure 4, respectively) of monomeric Aβ1−42, LMW Aβ1−42, and HMW Aβ1−42 (Figure 3D−F), respectively. The results were displayed in Table 1. By comparing ka values, the association rate of HMW Aβ1−42 (1.34 × 105 M−1·s−1) was nearly 5-fold faster than that of monomeric Aβ1−42 (2.85 × 104 M−1·s−1), which was somewhat lower than that of LMW Aβ1−42 (4.52 × 104 M−1·s−1). These outcomes clearly uncovered that HMW Aβ1−42 preferentially bound to ED1D2L receptor in contrast to LMW Aβ1−42 and

Figure 5. Binding model of ED1D2L receptor with HMW Aβ1−42, LMW Aβ1−42, and monomeric Aβ1−42. Green, black, and yellow arrows show the association process, dissociation period, and the regeneration process of ED1D2L surface, respectively. E

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

■

occupying the unbound sites with the highest rate, in contrast to LMW Aβ 1−42 and monomeric Aβ 1−42 . During the dissociation phase, the calculated rates were arranged by the order of monomeric monomeric Aβ1−42 > LMW Aβ1−42 > HMW Aβ1−42, granting that the whole affinity sites were individually bound by the three species, both monomeric Aβ1−42 and LMW Aβ1−42 quickly separated from the ED1D2L sites, whereas the HMW Aβ1−42 showed the longest residence time due to the highest affinity.

CONCLUSION In summary, we have reported the first demonstration of application using DPI technique for reconstituting interactions between three morphologies of Aβ1−42 and ED1D2L receptor. Based on the in-depth evaluation of binding curves in mass, the kinetics parameters of ka and kd values were precisely measured for the first time. The results of ka values demonstrated that the HMW Aβ1−42 (1.34 × 105 M−1·s−1) quickly bound to ED1D2L receptor, in contrast to monomeric Aβ1−42 (2.85 × 104 M−1· s−1) and LMW Aβ1−42 (4.52 × 104 M−1·s−1). The results of kd values clearly revealed the HMW Aβ1−42 (5.34 × 10−4·s−1) presented the highest affinity for ED1D2L receptor than monomeric Aβ1−42 (1.79 × 10−2 s−1) and LMW Aβ1−42 (2.09 × 10−2 s−1). These outcomes are likely to elucidate the reason why HMW Aβ1−42−ED1D2L receptor binding behavior exerts toxic effects implicated with AD. Together, we have successfully established an analytical platform based DPI for measuring the kinetic and affinity parameters of the tethered captures onto sensorgram with targets. More meaningfully, this work will open a new exciting clue to study the dynamic properties of the interactions of Aβ1−42 oligomers with other targets and even, hopefully, contributes to provide explicit direction for drug or antibody design for AD treatment. ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b04950. The IP results; characterization of prepared Aβ1−42 species by Western blotting; the immobilized model of ED1D2L receptor on the sensor surface; real-time thickness and density changes among them; linear analysis of individual slopes; Tables S1−S4 (PDF).

■

REFERENCES

(1) Wyss-Coray, T. Nat. Med. 2006, 12, 1005−1015. (2) Weiner, H. L.; Frenkel, D. Nat. Rev. Immunol. 2006, 6, 404−416. (3) Hardy, J.; Selkoe, D. J. Science 2002, 297, 353−356. (4) Perrin, R. J.; Fagan, A. M.; Holtzman, D. M. Nature 2009, 461, 916−922. (5) Jakob-Roetne, R.; Jacobsen, H. Angew. Chem., Int. Ed. 2009, 48, 3030−3059. (6) Welander, H.; Franberg, J.; Graff, C.; Sundstrom, E.; Winblad, B.; Tjernberg, L. O. J. Neurochem. 2009, 110, 697−706. (7) Wennmalm, S.; Chmyrov, V.; Widengren, J.; Tjernberg, L. Anal. Chem. 2015, 87, 11700−11705. (8) Bernstein, S. L.; Dupuis, N. F.; Lazo, N. D.; Wyttenbach, T.; Condron, M. M.; Bitan, G.; Teplow, D. B.; Shea, J. E.; Ruotolo, B. T.; Robinson, C. V.; Bowers, M. T. Nat. Chem. 2009, 1, 326−331. (9) Benilova, I.; Karran, E.; De Strooper, B. Nat. Neurosci. 2012, 15, 349−357. (10) Benilova, I.; De Strooper, B. Nat. Neurosci. 2011, 14, 949−950. (11) Hamley, I. W. Angew. Chem., Int. Ed. 2007, 46, 8128−8147. (12) Huang, Y.; Mucke, L. Cell 2012, 148, 1204−1222. (13) Figueiredo, C. P.; Clarke, J. R.; Ledo, J. H.; Ribeiro, F. C.; Costa, C. V.; Melo, H. M.; Mota-Sales, A. P.; Saraiva, L. M.; Klein, W. L.; Sebollela, A.; De Felice, F. G.; Ferreira, S. T. J. Neurosci. 2013, 33, 9626−9634. (14) Gimbel, D. A.; Nygaard, H. B.; Coffey, E. E.; Gunther, E. C.; Lauren, J.; Gimbel, Z. A.; Strittmatter, S. M. J. Neurosci. 2010, 30, 6367−6374. (15) Lauren, J.; Gimbel, D. A.; Nygaard, H. B.; Gilbert, J. W.; Strittmatter, S. M. Nature 2009, 457, 1128−1132. (16) Musiek, E. S.; Holtzman, D. M. Nat. Neurosci. 2015, 18, 800− 806. (17) Haass, C.; Selkoe, D. J. Nat. Rev. Mol. Cell Biol. 2007, 8, 101− 112. (18) Shankar, G. M.; Li, S.; Mehta, T. H.; Garcia-Munoz, A.; Shepardson, N. E.; Smith, I.; Brett, F. M.; Farrell, M. A.; Rowan, M. J.; Lemere, C. A.; Regan, C. M.; Walsh, D. M.; Sabatini, B. L.; Selkoe, D. J. Nat. Med. 2008, 14, 837−842. (19) Lesne, S.; Koh, M. T.; Kotilinek, L.; Kayed, R.; Glabe, C. G.; Yang, A.; Gallagher, M.; Ashe, K. H. Nature 2006, 440, 352−357. (20) Aloy, P.; Russell, R. B. Nat. Rev. Mol. Cell Biol. 2006, 7, 188− 197. (21) Kiel, C.; Beltrao, P.; Serrano, L. Annu. Rev. Biochem. 2008, 77, 415−441. (22) Krafft, G. A.; Klein, W. L. Neuropharmacology 2010, 59, 230− 242. (23) Viola, K. L.; Klein, W. L. Acta Neuropathol. 2015, 129, 183−206. (24) Ferreira, S. T.; Lourenco, M. V.; Oliveira, M. M.; De Felice, F. G. Front. Cell. Neurosci. 2015, 9, 191. (25) Atwal, J. K.; Pinkston-Gosse, J.; Syken, J.; Stawicki, S.; Wu, Y.; Shatz, C.; Tessier-Lavigne, M. Science 2008, 322, 967−970. (26) Kim, T.; Vidal, G. S.; Djurisic, M.; William, C. M.; Birnbaum, M. E.; Garcia, K. C.; Hyman, B. T.; Shatz, C. J. Science 2013, 341, 1399− 1404. (27) Benilova, I.; De Strooper, B. Science 2013, 341, 1354−1355. (28) Zheng, Y.; Yang, C.; Yang, F.; Yang, X. Anal. Chem. 2014, 86, 3849−3855. (29) Coan, K. E.; Swann, M. J.; Ottl, J. Anal. Chem. 2012, 84, 1586− 1591. (30) Escorihuela, J.; Gonzalez-Martinez, M. A.; Lopez-Paz, J. L.; Puchades, R.; Maquieira, A.; Gimenez-Romero, D. Chem. Rev. 2015, 115, 265−294. (31) Ostapchenko, V. G.; Beraldo, F. H.; Mohammad, A. H.; Xie, Y. F.; Hirata, P. H.; Magalhaes, A. C.; Lamour, G.; Li, H.; Maciejewski, A.; Belrose, J. C.; Teixeira, B. L.; Fahnestock, M.; Ferreira, S. T.; Cashman, N. R.; Hajj, G. N.; Jackson, M. F.; Choy, W. Y.; MacDonald, J. F.; Martins, V. R.; Prado, V. F.; Prado, M. A. J. Neurosci. 2013, 33, 16552−16564. (32) Wang, J.; Zhao, C.; Zhao, A.; Li, M.; Ren, J.; Qu, X. J. Am. Chem. Soc. 2015, 137, 1213−1219.

■

■

Article

AUTHOR INFORMATION

Corresponding Authors

*Fax: +86 431 85689278. E-mail: [email protected]. *E-mail: [email protected]. ORCID

Xiurong Yang: 0000-0002-1289-1248 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Nos. 21435005, 21627808) and Key Research Program of Frontier Sciences, CAS (QYZDY-SSWSLH019). F

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (33) Liang, N.; Xu, Y.; Yin, Y.; Yao, G.; Tian, H.; Wang, G.; Lian, J.; Wang, Y.; Sun, F. Endocrinology 2011, 152, 3213−3225. (34) Ricard-Blum, S.; Peel, L. L.; Ruggiero, F.; Freeman, N. J. Anal. Biochem. 2006, 352, 252−259. (35) Tikhonova, E. B.; Dastidar, V.; Rybenkov, V. V.; Zgurskaya, H. I. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 16416−16421. (36) Karlsson, R.; Michaelsson, A.; Mattsson, L. J. Immunol. Methods 1991, 145, 229−240. (37) Edwards, P. R.; Maule, C. H.; Leatherbarrow, R. J.; Winzor, D. J. Anal. Biochem. 1998, 263, 1−12. (38) Chromy, B. A.; Nowak, R. J.; Lambert, M. P.; Viola, K. L.; Chang, L.; Velasco, P. T.; Jones, B. W.; Fernandez, S. J.; Lacor, P. N.; Horowitz, P.; Finch, C. E.; Krafft, G. A.; Klein, W. L. Biochemistry 2003, 42, 12749−12760. (39) Feng, X.; Gao, F.; Qin, P.; Ma, G.; Su, Z.; Ge, J.; Wang, P.; Zhang, S. Anal. Chem. 2013, 85, 2370−2376. (40) Lillis, B.; Manning, M.; Berney, H.; Hurley, E.; Mathewson, A.; Sheehan, M. M. Biosens. Bioelectron. 2006, 21, 1459−1467. (41) Haudenschild, D. R.; Hong, E.; Yik, J. H.; Chromy, B.; Morgelin, M.; Snow, K. D.; Acharya, C.; Takada, Y.; Di Cesare, P. E. J. Biol. Chem. 2011, 286, 43250−43258. (42) Willcox, B. E.; Thomas, L. M.; Chapman, T. L.; Heikema, A. P.; West, A. P., Jr.; Bjorkman, P. J. BMC Struct. Biol. 2002, 2, 1−9. (43) Zheng, Y.; Hu, T.; Chen, C.; Yang, F.; Yang, X. Chem. Commun. 2015, 51, 5645−5648. (44) Eildal, J. N.; Hultqvist, G.; Balle, T.; Stuhr-Hansen, N.; Padrah, S.; Gianni, S.; Stromgaard, K.; Jemth, P. J. Am. Chem. Soc. 2013, 135, 12998−13007. (45) Pisareva, V. P.; Pisarev, A. V.; Hellen, C. U.; Rodnina, M. V.; Pestova, T. V. J. Biol. Chem. 2006, 281, 40224−40235. (46) Lee, C. C.; Liao, Y. C.; Lai, Y. H.; Lee, C. C.; Chuang, M. C. Anal. Chem. 2015, 87, 5410−5416.

G

DOI: 10.1021/acs.analchem.6b04950 Anal. Chem. XXXX, XXX, XXX−XXX