Inhibition of metal-induced amyloid β-peptide aggregation by a blood–brain barrier permeable silica–cyclen nanochelator

Alzheimer's disease (AD) is a neurodegenerative malady associated with amyloid β-peptide (Aβ) aggregation in the brain. Metal ions play important roles in Aβ aggregation and neurotoxicity. Metal chelators are potential therapeutic agents for AD because they could sequester metal ions from the Aβ aggregates and reverse the aggregation. The blood–brain barrier (BBB) is a major obstacle for drug delivery to AD patients. Herein, a nanoscale silica–cyclen composite combining cyclen as the metal chelator and silica nanoparticles as a carrier was reported. Silica–cyclen was characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), Fourier transform infrared (FT-IR) and dynamic light scattering (DLS). The inhibitory effect of the silica–cyclen nanochelator on Zn2+- or Cu2+-induced Aβ aggregation was investigated by using a BCA protein assay and TEM. Similar to cyclen, silica–cyclen can effectively inhibit the Aβ aggregation and reduce the generation of reactive oxygen species induced by the Cu–Aβ40 complex, thereby lessening the metal-induced Aβ toxicity against PC12 cells. In vivo studies indicate that the silica–cyclen nanochelator can cross the BBB, which may provide inspiration for the construction of novel Aβ inhibitors.


Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disorder affecting the memory and cognitive functions of the brain. 1 Although the molecular mechanism of AD pathogenesis is not clearly understood, much research has demonstrated that polymerization of amyloid b-peptides (Ab) into amyloid brils is a critical step in the pathogenesis. 2 The pathological hallmark of AD is the aggregation of Ab, predominantly Ab 40 and Ab 42 generated from the amyloid precursor protein (APP), which lead to the formation of oligomers and neuritic plaques in the brain. 3,4 Metal ions, such as Zn 2+ , Cu 2+ and Fe 3+ , play important roles in the Ab aggregation and neurotoxicity, because they can readily induce Ab nucleation and facilitate the formation of neurotoxic reactive oxygen species (ROS). 5 Thus, metal chelation therapy has been extensively studied as a treatment for AD, which can block the formation of ROS and reduce the Ab aggregation induced by metal ions. 6 Although much research has been directed to the development of AD therapy, 7-10 effective treatments are still unavailable. One of the major reasons is that most of the drug candidates are unable to cross the blood-brain barrier (BBB), 11,12 which is formed primarily by endothelial cells that line the cerebral microvasculature and surrounding perivascular elements. 13 Adjacent endothelial cells form complex tight junctions, creating a physical barrier which severely limits the paracellular transport across the BBB. 14 The BBB allows for the passive diffusion of small lipophilic molecules, whereas limits the passive permeation of hydrophilic substances or molecules with high molecular weight. 15 Since only lipophilic drugs with a molecular weight less than 450 Da can cross the BBB, most of the traditional drug candidates do not meet this requirement. 16 In an attempt to overcome the above limitations, nanocarriers have been investigated as drug delivery vehicles to the central nervous system (CNS). [17][18][19][20] The mesoporous silica (SiO 2 ) nanoparticles can be utilized to carry various drugs and other functional agents due to their unique properties such as large surface area, stable aqueous dispersion, none toxicity, easy surface modication, excellent biocompatibility and in vivo biodegradability. [21][22][23] The organically modied SiO 2 nanoparticles have been used as efficient non-viral vectors to delivery gene therapeutic agent into the CNS in vivo. 24 We and other researchers ever reported that macrocyclic chelator 1,4,7,10-tetraazacyclododecane (cyclen) could reduce the metal-induced Ab aggregation and neurotoxicity. 25,26 However, the hydrophilic cyclen (water solubility: 999 g L À1 ) may be hard to cross the BBB.
In this study, we designed a novel nanoscale chelator SiO 2 -cyclen, which conjugated SiO 2 nanoparticles as delivery carriers with cyclen as a metal chelator, for inhibiting the metal-induced Ab toxicity (Scheme 1). The effect of SiO 2 -cyclen nanochelator on Ab aggregation and neurotoxicity, as well as its BBB permeability were investigated in vitro and in vivo.

Results and discussion
Synthesis and characterization of SiO 2 -cyclen SiO 2 -cyclen nanochelator was designed and fabricated according to a modied literature method. 27 Cetyltrimethylammonium bromide (CTAB) was used as the cationic surfactant, tetraethylorthosilicate (TEOS) was served as the silica source, and ammonium hydroxide was used as the catalyst. In order to attach the metal chelator cyclen on the surface of SiO 2 nanoparticles, 3-chloropropyltriethoxysilane was used as a linker to fabricate the nanoscale SiO 2 -cyclen chelator. The morphology of the acquired SiO 2 -cyclen nanochelator was characterized by SEM (Fig. 1A) and TEM (Fig. 1B). The particles are spherical in shape and their size was smaller than 100 nm, which may enter the cells readily under uid ow conditions. 28 The average hydrodynamic diameter of SiO 2 -cyclen particles was determined by dynamic light scattering, which give a mean diameter of 65.2 AE 4.9 nm (DLS, Fig. 1C). The size of particles is just within the dimension range (40-100 nm) of nanoparticles that is not only suitable for drug carriers and cellular uptake, 29 but also suitable for transporting drugs across the BBB. 30 In the FT-IR spectra, the peak at 1079 cm À1 is attributed to the bond of Si-O, and those at 2931, 1460, 1353 cm À1 are attributed to the bonds of C-H, N-H, C-N, respectively (Fig. 1D). The relative intensity of C-H and N-H increases as the functionalization goes deeper; by contrast, that of Si-O uctuates. The changes manifest that cyclen has been linked to the surface of SiO 2 nanoparticles. The results indicate that the SiO 2 -cyclen nanochelator exists in single particles and disperses separately in aqueous suspension.

Chelation with Cu 2+ or Zn 2+
Cyclen is a metal chelator and has potential to disaggregate the metal-induced Ab aggregates as previously reported. 25,26,31 The chelating ability of SiO 2 -cyclen nanochelator was determined by ICP-MS aer incubation with Cu 2+ and Zn 2+ . The Cu and Zn amounts aer reacting with SiO 2 -cyclen were 22.18 AE 0.33 and 22.48 AE 0.29 mg mg À1 in terms of SiO 2 -cyclen weight, respectively. As a control, no Cu and Zn was detected in SiO 2 -Cl. The results indicate that cyclen tethered to the surface of SiO 2 nanoparticles still retains the chelating ability to Cu 2+ and Zn 2+ .

BCA protein assay
The effect of SiO 2 -cyclen nanochelator on the Zn 2+ -or Cu 2+induced Ab aggregation was investigated by measuring the percentage of soluble Ab in the supernatant of the reaction Scheme 1 Fabrication route to the SiO 2 -cyclen nanochelator.  mixtures, with SiO 2 -Cl and cyclen as the references. As shown in Fig. 2, Ab 40 is almost completely soluble in the absence of metal ions and chelators. However, soluble Ab in the supernatant of Ab reaction mixtures containing Zn 2+ or Cu 2+ decreases to 10% and 24%, respectively, indicating that most Ab is aggregated and deposited by metal ions. In the presence of SiO 2 -cyclen, the solubility of Ab increases obviously, suggesting that the SiO 2 -cyclen nanochelator can inhibit the metal-induced aggregation of Ab. As a comparison, SiO 2 -Cl can hardly inhibit the metal-induced Ab aggregation. These results show that the cyclen-modied mesoporous silica nanoparticles still have metal-chelating function and can inhibit the metal-induced Ab aggregation.

Inhibition of ROS generation
Redox-active metal ions are crucial for the production of ROS and oxidative stress. Ab could promote the production of ROS in the presence of redox-active metal ions, leading to pathological oxidative stress in AD. 32 Chelating agents can reduce the generation of ROS through removing Cu 2+ ions from the Cu-Ab complex. The generation of ROS induced by the Cu-Ab complex was monitored using 2 0 ,7 0 -dichlorouorescein diacetate (DCFH-DA). DCF is a uorescent marker derived from the reaction of non-uorescent DCFH with ROS in the presence of horseradish peroxidase (HRP). 33 The uorescence intensity of DCF correlates with the amount of reactive oxygen radicals. As shown in Fig. 3, strong uorescence of DCF is measured at 522 nm for the Cu-Ab 40 system without the SiO 2 -cyclen nanochelator (b); in the presence of SiO 2 -cyclen, the uorescence intensity decreases obviously (e).
The results indicate that SiO 2 -cyclen can reduce the generation of ROS induced by the Cu-Ab 40 complex. In contrast, SiO 2 -Cl shows no effect on the reduction of ROS (d), because it does not coordinate with the Ab-bound Cu 2+ and hence can hardly inuence the Cu-Ab 40 -mediated redox chemistry. These results indicate that the SiO 2 -cyclen nanochelator reduces the production of ROS induced by Cu-Ab complex almost as effectively as cyclen.

Morphology changes of Ab aggregates
Negative staining TEM was exploited to investigate the effect of the SiO 2 -cyclen nanochelator on the morphology of metal-induced Ab aggregates. The images of Zn 2+ -or Cu 2+ -induced Ab aggregates in the absence and presence of the nanochelator are shown in Fig. 4. Only long unbranched brils, a typical structure for amyloid brils, were observed in the solution of Ab 40 (Fig. 4A). However, aer Zn 2+ or Cu 2+ was added, large amounts of amorphous aggregates were formed in the solution of Ab 40 ( Fig. 4B and C), which are consistent with our previous observations. 34,35 In the presence of SiO 2 -cyclen, the metal-induced Ab 40 aggregates were almost disappeared, and the morphology of the samples was similar to that of Ab 40 samples ( Fig. 4D and G). Cyclen also inhibited the Zn 2+ -or Cu 2+ -induced Ab 40 aggregation and made the morphology similar to that of Ab 40 alone (Fig. 4F and I). More aggregates were observed in the presence of SiO 2 -Cl owing to its inability to chelate Zn 2+ or Cu 2+ (Fig. 4E  and H). The results indicate that the SiO 2 -cyclen nanochelator can inhibit the Zn 2+ -or Cu 2+ -induced Ab 40 aggregation.

Inhibition of neurotoxicity
The neurotoxicity of Zn 2+or Cu 2+ -Ab 40 complexes against PC12 cells was investigated by the MTT assay. The inhibition of SiO 2cyclen nanochelator against the Ab 40 -induced neurotoxicity was shown in Fig. 5, with cyclen and SiO 2 -Cl as the references. Ab 40 in the presence of Zn 2+ or Cu 2+ was quite toxic to the rat pheochromocytoma PC12 cells (cell viability is about 70%), while Zn 2+ , Cu 2+ , and Ab 40 alone are almost nontoxic. In the presence of SiO 2 -cyclen, the cell viability in the Zn 2+ -Ab 40 system increased from 74% to 91%, and that in the Cu 2+ -Ab 40 system increased from 71% to 93%, respectively. Interestingly, SiO 2 -cyclen and its Zn 2+ or Cu 2+ complex is nontoxic toward the cells. In the presence of SiO 2 -Cl, the cell viability is similar to that in the presence of Ab 40 and Zn 2+ or Cu 2+ , indicating that SiO 2 -Cl had no effect on the neurotoxicity of the Zn 2+or Cu 2+ -Ab 40 complex. Aer incubating with cyclen, the cell viability was above 90%, even in the present of Zn 2+or Cu 2+ -Ab 40 complex. These results demonstrate that the SiO 2 -cyclen nanochelator can inhibit the neurotoxicity of Zn 2+or Cu 2+ -Ab 40 complexes and enhance the viability of neuron cells.

Blood-brain barrier permeability
To evaluate the in vivo BBB permeability of the SiO 2 -cyclen nanochelator, animal experiments were carried out by using C57BL/6J mice. The amount of silicon in mice brain pre-or postinjection of SiO 2 -cyclen was listed in Table 1. The amount of Si increased aer 6 h, thus suggesting that the nanochelator could penetrate the BBB of mice. The amount of Si decreased at 24 h probably due to the metabolism of SiO 2 -cyclen in vivo. The SiO 2 -cyclen nanochelator may cross the BBB via the adsorptive or receptor-mediated transportation, which is similar to the situation of insulin, albumin and low density lipoprotein receptor as reported previously. 36

Preparation of SiO 2 -cyclen nanochelator
SiO 2 nanoparticles were synthesized by a modied literature procedure. 37,38 CTAB (0.5 g) was dispersed in water (200 mL) with ultrasonic wave. Ammonium hydroxide (0.75 mL, 28 wt% NH 3 in water) was then added to the solution with strong stirring at room temperature, and TEOS (2.0 mL) was dropped in slowly, giving rise to a white slurry. The resulting product was centrifuged aer 3 h, the CTAB was washed out by ethanol and water, and SiO 2 nanoparticles were obtained aer drying at vacuum. SiO 2 -Cl nanoparticles were prepared according to the modied literature procedure. 39 SiO 2 nanoparticles (200 mg) were dispersed in isopropanol (200 mL) solution and were allowed to react with 3-chloropropyltriethoxysilane (4.0 mL, in excess) at 100 C under nitrogen for 24 h. Excess 3-chloropropyltriethoxysilane was removed by centrifugation and redispersion in ethanol and water, followed by drying at room temperature. Cyclen was tethered to the surface of SiO 2 -Cl nanoparticles according to the literature procedure. 40 In a typical reaction, cyclen (2.0 g, in excess) and triethylamine (6.0 mL) were poured into a ask containing toluene (200 mL) and the SiO 2 -Cl nanoparticles with vigorous stirring under argon atmosphere and reux for 16 h. The sample was washed with water and ethanol by centrifugation and SiO 2 -cyclen nanochelators were obtained aer drying at vacuum. The surface morphology, size, and components of the nanoparticles were investigated by SEM, TEM, DLS, and FT-IR.

Chelation ability of SiO 2 -cyclen
SiO 2 -cyclen (6.0 mg) and SiO 2 -Cl (6.0 mg) nanoparticles were dispersed into CuCl 2 (1.0 mol L À1 , 2.0 mL, in excess) or Zn(Ac) 2 (1.0 mol L À1 , 2.0 mL, in excess) solutions respectively, and cultured at 37 C for 3 h. The samples were cleaned by water to  In vivo BBB penetration assay C57BL/6J mice (8 week, male, 20 g, n ¼ 12) were selected as animal models. SiO 2 -cyclen nanochelators were injected intravenously (8 mg kg À1 body weight) into the mice (3 mice in each group), and the brains of mice were acquired aer 6 h, 12 h and 24 h, with mice without injection as controls. The brain samples were digested by concentrated HNO 3 at 95 C, 30% H 2 O 2 and concentrated HCl at 37 C. The silicon amount in the samples was determined by ICP-MS. All experimental procedures are in accordance with the Guidelines for Care and Use of Laboratory Animals of Nanjing University, and experiments were approved by the Animal Ethics Committee of the Model Animal Research Center of Nanjing University.

Conclusions
In this study, a novel nanoscale chelator, SiO 2 -cyclen, was reported, which is composed by cyclen as the metal-chelating unit and silica nanoparticle as a carrier of cyclen, for inhibiting the toxicity of Ab aggregates. The results show that the SiO 2 -cyclen nanochelator can effectively inhibit Ab aggregation, reduce the generation of reactive oxygen species induced by the Cu-Ab 40 complex, and protect cells from the metal-induced Ab toxicity. Blood-brain barrier is a dynamic barrier protecting the brain against invading organisms and unwanted substances; it is also the most important barrier impeding the drug transport into the brain via the blood circulation. 41 In vivo study demonstrated that the SiO 2 -cyclen nanochelator can overcome the drawbacks of small chemicals (>400 Da) or peptides in passing across the BBB, which may has some reference value for the design of novel Ab inhibitors.

Conflicts of interest
There are no conicts to declare.