Sulforaphane Inhibits the Generation of Amyloid- β Oligomer and Promotes Spatial Learning and Memory in Alzheimer’s Disease (PS1V97L) Transgenic Mice

Abstract

Abnormal amyloid-β (Aβ) aggregates are a striking feature of Alzheimer’s disease (AD), and Aβ oligomers have been proven to be crucial in the pathology of AD. Any intervention targeting the generation or aggregation of Aβ can be expected to be useful in AD treatment. Oxidative stress and inflammation are common pathological changes in AD that are involved in the generation and aggregation of Aβ. In the present study, 6-month-old PS1V97L transgenic (Tg) mice were treated with sulforaphane, an antioxidant, for 4 months, and this treatment significantly inhibited the generation and aggregation of Aβ. Sulforaphane also alleviated several downstream pathological changes that including tau hyperphosphorylation, oxidative stress, and neuroinflammation. Most importantly, the cognition of the sulforaphane-treated PS1V97L Tg mice remained normal compared to that of wild-type mice at 10 months of age, when dementia typically emerges in PS1V97L Tg mice. Pretreating cultured cortical neurons with sulforaphane also protected against neuronal injury caused by Aβ oligomers in vitro. These findings suggest that sulforaphane may be a potential compound that can inhibit Aβ oligomer production in AD.

Keywords

Aβ oligomer Alzheimer’s disease inflammation oxidative stress sulforaphane

INTRODUCTION

Alzheimer’s disease (AD) is the most common progressive neurodegenerative disease causing dementia and is characterized by deposits of abnormal amyloid-β (Aβ) in senile plaques and by intracellular neurofibrillary tangles composed of hyperphosphorylated tau [1]. Abnormal Aβ aggregation was originally shown to contribute to a series of pathological changes in AD [2]. However, fibrillar Aβ aggregates are now known to be poorly correlated with brain dysfunction in humans [3], and clinical trials that have attempted to remove insoluble Aβ in senile plaques have failed [4]. Investigators have gradually recognized that soluble Aβ oligomers are crucial for synaptic dysfunction and that an increase in their concentration corresponds to cognitive decline in animals and humans with this disease [5, 6]. Therefore, soluble oligomers should be the targets of therapies designed to ameliorate or prevent AD [7, 8]. In addition, the AD pathology may develop several decades before the emergence of symptomatic cognitive impairment, and amyloid-based therapies may need to be used preventatively to confer cognitive benefits clinically [4]. Therefore, interventions aimed at reducing the generation of Aβ oligomers at the preclinical stage may constitute a useful approach for AD treatment.

Oxidative stress and inflammation are common pathological changes caused by Aβ oligomers in AD, and both changes can promote the generation of Aβ [9]; moreover, oxidative stress and inflammation are also involved in the aggregation of Aβ [10 –12]. Therefore, we hypothesize that alleviation of oxidative stress and inflammation may attenuate the generation of Aβ oligomers. Here, we investigated the potential effects of sulforaphane (1-isothiocyanato-4-[methylsulfinyl]-butane, SFN) on Aβ oligomer generation; SFN is an isothiocyanate compound that functions as an antioxidant to activate the Nrf2/ARE pathway [13]. Previous studies have found that SFN inhibits Aβ-induced inflammation in human THP-1 macrophages [14] and that SFN reduces oxidant injury in an acute Alzheimer-like model [15]. These data suggest that SFN may affect Aβ oligomer production and associate cascade reactions. In a previous study, we successfully generated a PS1V97L transgenic (Tg) mouse model that forms abundant Aβ oligomers but exhibits no amyloid plaques until 24 months of age [16]. This animal model is ideal for the study of Aβ oligomers. In the present study, we prophylactically treated 6-month-old PS1V97L Tg mice with SFN. We found that this treatment significantly inhibited the generation of Aβ oligomers in PS1V97L Tg mice and maintained the cognition of PS1V97L Tg mice to the normal level compared with that of wild-type (WT) mice. This neuroprotective effect of SFN was confirmed in vitro.

MATERIALS AND METHODS

Ethics statement

All experimental procedures were performed according to the rules in the “Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research”. The study protocol was approved by the Ethics Committee of Capital Medical University, and every effort was made to minimize the number of animals used and their suffering.

Animals

Six-month-old male and female PS1V97L Tg mice were housed in a room at a constant temperature (25±1°C) and humidity (40%–60%) and kept on a 12-h light/dark cycle (lights on at 8:00 AM). The mice had free access to food and water. Expression of the human PSEN1 gene with the V97L mutation was induced as previously described [16]. The PS1V97L Tg mouse lines were maintained by crossing heterozygous Tg mice with wild-type C57BL/6J (WT) animals. Mice were screened by polymerase chain reaction (PCR) to determine their genotypes as previously described [17]. Six-month-old PS1V97L Tg mice (n = 6) were treated with corn oil or 5 mg/kg SFN (Abcam, USA) in corn oil intraperitoneally every day for 4 months until 10 months of age, equal numbers of wild-type mice were treated with corn oil or SFN and used as controls.

Behavioral tests

All groups of mice were tested for spatial learning and memory in the Morris water maze (MWM) after four months of SFN treatment. The mice were kept under a 12-h:12-h light/dark cycle to ensure that the tests were carried out during the active period of the animals. For five consecutive days, all mice were trained to locate a platform that was hidden below the water surface in a 100-cm diameter pool. MWM protocols were carried out by coworkers who were blinded to the genotypes. Training consisted of four trials per day, with a 30-s inter-trial interval. On the sixth day, spatial memory was assessed using a probe trial in which the mice performed a 60-s free search of the place where the platform was previously located in the pool. The frequency of crossing the “platform position” and the time in the target quadrant where the platform had been located was recorded with a DNS-2 type MWM testing set equipped with an online video tracking system (camera, TOTA-450III, Japan).

Protein extraction and immunoprecipitation

Ice-cold Tris-buffered saline (TBS) consisting of 20 mM Tris-HCl, 150 mM NaCl, pH 7.4, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride (PMSF), and a protease inhibitor cocktail (Roche, Switzerland) was added to the frozen hemisphere (excluding the cerebellum), and the tissue was homogenized with a mechanical Dounce homogenizer, followed by centrifugation for 60 min at 13,000 rpm. Soluble Aβ proteins were incubated overnight at 4°C with columns packed with appropriate 6E10 monoclonal antibodies. Columns were created using the Thermo Scientific™ direct immunoprecipitation kit (Thermo Scientific, USA) according to the manufacturer’s protocols. Captured proteins were eluted in elution buffer. The protein content of the brain homogenates was determined using a Pierce bicinchoninic acid (BCA) protein assay kit (Thermo Scientific, USA).

Western blot

The levels of Aβ production-associated molecules or enzymes and phosphorylated tau (p-tau) were measured using western blot. The protein concentrations of the samples were determined using a BCA protein assay kit (Thermo Scientific, USA). Samples were loaded resolved on SDS/PAGE gels, and the separated proteins were transferred to nitrocellulose membranes. The membranes were blocked in a solution of 5% fat-free milk for 30 min at 20°C and incubated overnight at 4°C with one of the following primary antibodies: anti-BACE1 (Abcam 1:1000); anti-PSI (Abcam 1:1000); anti-Aβ (4G8, BioLegend 1:1000); anti-p-tau (pT23l, Invitrogen 1:1000); monoclonal anti-total tau (t-tau) (tau-5, Invitrogen 1:1000); and anti-β-actin (Santa Cruz 1:1000). Primary antibody incubation was followed by incubation at 37°C for 1 h with the HRP-labeled secondary antibody and subsequent visualization using enhanced chemiluminescence reagents (Beyotime Institute of Biotechnology, China). The membranes were scanned (Millipore, USA), and the optical densities (ODs) were determined using ImageJ software (v1.46; National Institutes of Health). For the analysis, the band density was normalized to that of β-actin.

Measurement of oxidative stress parameters and proinflammatory cytokines

The frozen hemispheres (excluding the cerebellum) of the mice were weighed and homogenized in ice-cold phosphate-buffered saline (PBS) containing a Complete Min Protease Inhibitor Cocktail Tablet (Roche, USA) and centrifuged at 5,000×g for 20 min at 4°C to obtain supernatant. Brain tissue homogenates and plasma were diluted with the buffer solution corresponding to the specific assay. The glutathione (GSH) and malondialdehyde (MDA) contents were determined spectrophotometrically with commercially available assay kits (Nanjing Jiancheng Bioengineering Institute, China) used according to the manufacturer’s instructions. The protein concentrations of the brain tissue homogenate samples were determined using a Pierce BCA protein assay kit (Thermo Scientific, USA).

The levels of tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) in plasma and brain tissue homogenates were determined using TNF-α and IL-1β ELISA kits (Nanjing Jiancheng Bioengineering Institute, China) according to the manufacturer’s protocols.

Primary cell cultures and cell treatments

Primary cortical cells were isolated from 18-day-old SD rat embryos. The obtained cells were plated on poly-L-lysine (Sigma, USA)-coated coverslips in 35-mm dishes at a density of 10⁵ cells/dish or in 6-well cell culture clusters at a density of 1.5*10⁵ cell/cm² in DMEM/F-12 medium (Invitrogen, USA) with 10% fetal bovine serum (Invitrogen, Australia). After 4 h, the medium was replaced with neurobasal medium supplemented with 2% B27 (Invitrogen, USA) and 0.5 mM GlutaMAX-I (Invitrogen, USA). Aβ₄₂ monomers (Invitrogen, USA) were added to the medium to concentrations of 5μM, 10μM, and 20μM for 48 h to evaluate neurotoxicity. Normal neurobasal medium was used for the controls. The primary cortical cells cultures were treated with or without SFN at different concentrations followed by incubation with 10μM natural Aβ oligomers (extracted from 15–24-month-old PS1V97L Tg mouse brains) for 48 h to evaluate the protective effect of SFN. The controls were incubated with cell culture medium containing dimethyl sulfoxide (DMSO) or SFN alone.

Immunocytochemistry

The primary cell cultures on coverslips were washed with PBS and then fixed with 4% paraformaldehyde for 30 min at room temperature. Neurons were then washed with PBS three times and permeabilized with 0.3% Triton X-100 for 30 min. After blocking with 10% serum for 1 h, the cells were incubated with primary antibodies overnight at 4°C followed by Alexa Fluor-conjugated secondary antibodies (Thermo Scientific, USA) for 1 h at room temperature. Nuclei were stained with the fluorescent dye 4’,6-diamidino-2-phenylindole (DAPI) (Thermo Scientific, USA). Finally, the cells were mounted onto slides, and the stained sections were examined with a Leica fluorescence microscope.

Cell viability assay and toxicity assays

MTT was used to examine cell viability after Aβ monomer and Aβ oligomer treatments. After stimulation as described above, the cells in the 96-well plates were incubated with 10μl of MTT (5 mg/ml, Sigma-Aldrich, USA) in serum-free medium at 37°C for 4 h. The supernatant was removed and replaced with 100μl of DMSO (Sigma-Aldrich, USA) to dissolve the insoluble formazan crystals. The OD was measured with a microplate reader (Tecan, Switzerland) at 570 nm. The data were represented as MTT reductions relative to the control. Cytotoxicity was evaluated by measuring LDH in the culture medium using the LDH Cytotoxicity Assay Kit (Beyotime Institute of Biotechnology, China) according to the manufacturer’s directions. After collection of the medium, the remaining cells were lysed in 0.9% (w/v) Triton X-100, and the LDH contents of the medium and lysed cells were measured to determine the total LDH content. The OD was measured at 490 nm. The amount of LDH released from the cells was calculated as the percentage of the total LDH level in each sample. The experiments were repeated independently three times in triplicate.

Cell lysates

Following removal of the cell culture medium, 100μl of ice-cold lysis buffer (20 mM Tris pH 7.5, 0.5% Triton X-100, 0.5% deoxycholic acid, 150 mM NaCl, 10 mM EDTA, 30 mM NaPyroP and protease inhibitor, Roche) was added to the dish. The lysed cells were collected using a cell lifter (Costar), transferred to Eppendorf tubes, incubated on ice for 30 min and centrifuged (30 min, 4°C, 12000 g). The supernatants were stored at –80°C until use.

Statistical analyses

The results are presented as the mean±SEM. Statistical comparisons between two groups were performed using Student’s t-test or the Mann–Whitney U test, as applicable. The Comparisons among groups were performed using one-way or two-way ANOVA, and trend analysis was performed when necessary. p < 0.05 was considered significant.

RESULTS

SFN alleviates the cognitive deficits in PS1V97L Tg mice

After treatment, all groups of mice were tested in the MWM for cognitive assessment. The latency to reach the escape platform was not different between the groups during the first two days (Fig. 1A, p > 0.05). The PS1V97L Tg mice without SFN treatment began to show a significantly longer escape latency after the third training day (Fig. 1A, p < 0.01), while the escape latency was significantly shorter for the PS1V97L Tg mice treated with SFN (Fig. 1A, p < 0.01). No statistically significant differences were observed between the PS1V97L Tg mice treated with SFN and the wild-type mice during the five days of the MWM test (p > 0.05).

Fig.1

SFN improved the cognitive deficits in the PS1V97L Tg mice (A-C). Escape latency during the platform trials in the MWM (A). Time spent in the target quadrant (B). Number of platform location crossings in the probe trials (C). The data are shown as the mean±SEM (n = 6/group), ^*p < 0.05, ^**p < 0.01.

Spatial memory was further evaluated in the probe trial, which was performed by removing the platform after 5 training days. We calculated the swimming time in the target quadrant and found that among the four groups, the PS1V97L Tg mice without SFN treatment spent the least amount of time in the target quadrant of the MWM (Fig. 1B, p < 0.01). However, the swimming time was much longer for the PS1V97L Tg mice treated with SFN than that for the control PS1V97L Tg mice (Fig. 1B, p < 0.01). We also found that the PS1V97L Tg mice not treated with SFN crossed the platform location less often than that for the wild-type littermates (Fig. 1C, p < 0.01), and the number of platform crossings was significantly increased in the SFN-treated PS1V97L Tg mice compared to that in the control PS1V97L Tg mice (Fig. 1C, p < 0.01). No differences in the swimming time over the target quadrant or the number of platform crossings were evident between the wild-type mice and the SFN-treated PS1V97L Tg mice (p > 0.05). The wild-type mice treated with or without SFN showed similar performance in this test (p > 0.05). Moreover, no difference in the swimming speed was evident on the first training day (p > 0.05), which excluded any potential influence of motor dysfunction on the escape latency (data not shown).

SFN inhibits the generation of all types of Aβ oligomers in PS1V97L Tg mice

As shown in Fig. 2A, we found a set of apparent assemblies of Aβ oligomers in all groups of mice. In addition to a faint 4-kDa band corresponding to Aβ monomers, 4G8-immunoreactive protein bands were detected at molecular masses that theoretically corresponded to trimeric (14 kDa), tetrameric (18 kDa), hexameric (27 kDa), nonameric (40 kDa), and dodecameric (56 kDa) Aβ oligomers. Different types of Aβ oligomers were also present in the wild-type mice, although at very low levels, and the contents of the different types of Aβ oligomers were not significantly different (Fig. 2A, B, p > 0.05). Compared to the wild-type mice, the PS1V97L Tg mice showed clearly higher Aβ oligomer content but similar Aβ monomer content (Fig. 2A, B, p < 0.01). Among all of the Aβ oligomer types, the tetramer (18 kDa), nonamer (40 kDa), and dodecamer (56 kDa) were the most abundant assemblies; the nonamer (40 kDa) and dodecamer (56 kDa) were especially abundant (Fig. 2A, B, p < 0.01). In the PS1V97LTg mice, SFN treatment reduced the abundance of all types of Aβ oligomers to a level that was not different from that in the wild-type mice (Fig. 2A, B, p < 0.01). Meanwhile, the Aβ monomer content was notably decreased in the SFN-treated PS1V97L Tg mice compared to that in both the wild-type mice and the non-SFN-treated PS1V97L Tg mice (Fig. 2A, B, p < 0.01).

Fig.2

SFN inhibits the generation of all Aβ oligomer types. Western blotting of the assemblies of Aβ oligomers probed with the 4G8 antibody. The blots were reprobed for β-actin (A). The band intensities were quantitated after scanning (B). Values are expressed as the mean±SEM (n = 6/group); APPfl (APP full length); ^*p < 0.05, ^**p < 0.01.

SFN inhibits the amyloidogenic processing of AβPP in PS1V97L Tg mice

We examined the expression levels of BACE1 and PS1 to determine whether SFN could inhibit the amyloidogenic processing of AβPP. As shown in Fig. 3A-C, the expression levels of BACE1 and PS1 increased by 4.4-fold and 2.4-fold, respectively, in the PS1V97LTg mice relative to those in the wild-type mice (p < 0.01), but they decreased significantly after SFN treatment, although the levels were still higher in the SFN-treated PS1V97LTg mice than in the wild-type mice (p < 0.05).

Fig.3

SFN attenuates the amylidogenic processing of AβPP. Western blotting analysis of the BACE1 and PS1 levels in brain tissue (A). The bar graphs present the protein expression levels of BACE1 (B) and PS1 (C). The data are shown as the mean±SEM (n = 6/group); ^*p < 0.05, ^**p < 0.01.

SFN inhibits oxidative stress and inflammation in PS1V97L Tg mice

As shown in Fig. 4A, the GSH content in the brain tissues was significantly decreased in the PS1V97L Tg mice compared to that in the wild-type mice (p < 0.01), but it was increased significantly in the SFN-treated PS1V97L Tg mice (p < 0.05). Conversely, the plasma MDA content in brain tissues was significantly increased in the PS1V97L Tg mice compared to that in the wild-type littermates (p < 0.01) and was significantly decreased by SFN treatment in PS1V97L Tg mice (Fig. 4B). IL-1β and TNF-α are biomarkers of neuroinflammation. As shown in Fig. 4C, D, the levels of IL-1β and TNF-α in the PS1V97L Tg mice increased prominently compared to those of the wild-type mice, and these levels were significantly decreased by SFN treatment in the PS1V97L Tg mice (p < 0.05).

Fig.4

SFN inhibits oxidative stress and inflammation in PS1V97L Tg mice. Measurement of GSH and IL-β expression in brain tissue (A, C). Measurement of MDA and TNF-α levels in plasma (B, D). The data are shown as the mean±SEM (n = 6/group); ^*p < 0.05, ^**p < 0.01.

SFN reduces natural Aβ oligomer-induced cytotoxicity in primary cortical neuron cultures

We first examined the neurotoxicity of the Aβ monomers in cortical neuron cultures, as shown in Fig. 5A, B. No differences in cell viability or LDH release were evident between the Aβ monomer-stimulated groups and the control group. We then extracted natural Aβ oligomers from aged PS1V97L Tg mice to imitate an AD model in our research. As shown in Fig. 5C, D, the natural Aβ oligomers significantly decreased cortical neuron viability and increased LDH release in the neuron cultures. The SFN-pretreated neuron cultures showed significantly improved cell viability and decreased LDH release. Furthermore, this tendency was dose dependent (Fig. 5C, D, p < 0.01). We then examined whether SFN could protect the dendritic integrity of neurons. Cortical neurons on day 7 were treated with 10μM extracted Aβ oligomers with or without 0.1μm SFN. After 48 h of incubation, the Aβ oligomers reduced the dendrite length of cortical neurons by nearly 65% (Fig. 5E, F). However, 0.1μm SFN significantly rescued dendritic integrity (Fig. 5E, F).

Fig.5

SFN reduced natural Aβ oligomer-induced cytotoxicity in primary cortical neuron cultures. The toxicity of the Aβ monomers was tested at 3 concentrations (5μm, 10μm, 20μm). Cell viability is presented as a percentage of the control. LDH release is presented as a percentage of the total. Each value represents the mean±SEM of three independent experiments (A, B). Cortical neurons were pretreated with SFN for 30 min and subsequently stimulated with extracted Aβ oligomers at different concentrations. Cell viability is presented as the percentage relative to the DMSO-treated group. LDH release is presented as a percentage of the total. Each value represents the mean±SD of three independent experiments (C, D). The lengths of the neuronal dendrites were analyzed (E). For the assessment of the lengths of the neuronal dendrites, cells treated or not treated with SFN treatment stained with anti-MAP2 antibodies (red) after a 48-h incubation (F); scale bar: 100μm. ^*p < 0.05, ^**p < 0.01.

SFN reduces hyperphosphorylated tau levels in vivo and in vitro

To determine whether SFN could reduce the hyperphosphorylation of tau, we examined total tau (t-tau) and phosphorylated tau (p-tau) concentrations in brain tissues by western blot. As shown in Fig. 6A-C, the brain concentration of p-tau was reduced significantly in the PS1V97L Tg mice treated with SFN (p < 0.01), while the concentration of p-tau was significantly increased in the PS1V97L Tg mice compared to that in the wild-type mice. We repeated the experiment in vitro, and the results showed that the extracted Aβ oligomers induced an obvious increase in p-tau levels (p < 0.01). However, the p-tau level was reduced in the SFN-pretreated primary cortical neurons after Aβ oligomer exposure, and this decrease occurred in a dose-dependent manner (Fig. 6D-F, p < 0.01).

Fig.6

SFN attenuates the hyperphosphorylation of tau. Western blotting analysis of the t-tau and p-tau levels in brain tissue (A). Western blotting analysis of the t-tau and p-tau levels in primary neuron cell cultures (D). Bar graphs indicate the protein expression levels of p-tau (B, E) and t-tau (C, F). The data are shown as the mean±SEM (n = 6/group); ^*p < 0.05, ^**p < 0.01.

DISCUSSION

The amyloid-based strategy used in clinical trials has long been considered the best method for curing AD. However, the drug candidates that target at Aβ, which are either immunotherapies or β- or γ-secretase inhibitors, have not been successful due to serious side effects or the incomplete removal of preexisting Aβ aggregates [8 , 19]. These results prompted us to identify a compound that could effectively reduce the levels of the toxic Aβ oligomer without disturbing physiological functions. SFN is an isothiocyanate compound that is produced in cruciferous vegetables, such as broccoli sprouts, and SFN functions as an antioxidant, which suggests a potential effect in AD [20]. In the present study, we prophylactically treated 6-month-old PS1V97L Tg mice with SFN when their cognition was normal. After 4 months of treatment, we found that SFN effectively inhibited the generation and aggregation of Aβ, lowered the concentration of Aβ oligomers to a normal level and significantly improved cognition. We also found that SFN had a neuroprotective effect on primary cortical neurons exposed to Aβ oligomers in vitro.

We found that SFN can significantly reduce the levels of all types of Aβ oligomer, including the trimeric, tetrameric, hexameric, nonameric, and dodecameric types in PS1V97L Tg mice when administered from 6 months to 10 months of age. Moreover, the Aβ oligomer levels were decreased to the levels in the wild-type mice. Additionally, when assessed at 10 months of age, the cognition of the PS1V97L Tg mice treated with SFN remained normal compared with that of the wild-type mice. This observation indicates a close correlation between the Aβ oligomer level and cognition. Aβ oligomers are known to be neurotoxic in vivo and in vitro, and Aβ oligomers can rapidly inhibit long-term potentiation and impair synaptic function and plasticity [21, 22]. Aβ oligomers can also damage mitochondrial function, cause endoplasmic reticulum stress, disrupt calcium homeostasis, induce tau phosphorylation, evoke chronic inflammation, and ultimately promote neuronal death [23]. Based on these results, we suggest that the improved cognition of the PS1V97L Tg mice treated with SFN is attributable to the prominent reduction in the Aβ oligomer level. We also suggest that the timing of SFN treatment in the PS1V97L Tg mice is important since the generation of abnormal Aβ oligomers in these mice begin at 6 months of age [16], when neuronal function is only slightly damaged and can be repaired. At this time point, inhibiting Aβ oligomer generation may preserve neuronal function in whole or in part. This idea is supported by the observation that in vitro, pretreatment with SFN can also directly help neurons resist injury caused by Aβ oligomers extracted from PS1V97L Tg mice. Therefore, SFN may be a potential compound that targets at Aβ oligomers.

SFN has been proven to activate the Nrf2 pathway in neuronal cells especially in astrocytes [24, 25]. SFN can release Nrf2 from Keap1 by directly interacting with Keap1, the released Nrf2 can then translocate to the nucleus and bind to ARE sequences, thus increasing the transcription of the phase II antioxidant enzymes such as GSH and peroxiredoxins that can resist the injury of oxidative stress [26]. In addition, the released Nrf2 can also antagonize the transcription factor-nuclear factor-κB (NF-κB) which regulates the expression of inflammatory genes [27]. Otherwise, it is also reported that SFN can inhibited the inflammation induced by Aβ via STAT-1 dephosphorylation except for activation of Nrf2/HO-1 [14]. Oxidative stress and inflammation are recognized as common pathological changes in AD and are involved in the aggregation of Aβ. The oxidative stress metabolite 4-hydroxynonenal and aldehydes had been certified to promote Aβ peptide aggregate to fibrils in vitro [28, 29]. Similar results were also found in vivo that severe oxidative stress accelerates Aβ oligomerization in SOD1-deficient Tg2576 animal models [30], and reduced neuroinflammation inhibits Aβ oligomerization in the brain via interleukin-4 (IL-4) overexpression in AD animal models [11]. In our study, we found that SFN could alleviate oxidative stress and inflammation, as indicated by decreased MDA, IL-1β and TNF-α levels and increased GSH levels in PS1V97L Tg mice. We suggest that the reduced oxidative stress and inflammation may inhibit the aggregation of Aβ. Meanwhile, another explanation for the reduction in Aβ oligomers in our study might be the downregulated expression of BACE1 and PS1, which are involved in the sequential proteolytic processing of AβPP into Aβ peptides in neuronal cells; downregulation of these enzymes leads to decreased Aβ production in PS1V97L Tg mice after SFN treatment. The downregulation of these enzymes in our study may also be due to the antioxidant and anti-inflammatory effect of SFN. Aβ oligomer can induce the generation of reactive oxygen species and inflammation factors; meanwhile, the overproduction of reactive oxygen species and inflammation factors can drive the upregulation of BACE1 and PS1 by activating specific pathway that accelerate the generation of Aβ in turn, such a vicious cycle can accelerate the progress of AD [31 –33]. SFN may reduce the generation of BACE1 and PS1 in our study by interrupt the vicious cycle. Whether SFN can directly inhibit the transcription of BACE1 and PS1 requires further investigation. In addition, it was reported that SFN could covalently bind to Aβ₄₀ at three different NH₂ groups of N-terminal aspartic acid and the ɛ-amino group of lysine at positions 16 and 28 in vitro by LC/MS, and the combination can inhibit the aggregation of Aβ₄₀ from 36% to 15% after 24 h incubation [34]. This may be another reason for SFN to inhibit the generation of Aβ oligomer in our study.

Hyperphosphorylated tau is the major component of neurofibrillary tangles which is another hallmark of AD. In our research, we observed that the levels of p-tau were decreased significantly in SFN-treated PS1V97L Tg mice and cortical neurons. Based on the current understanding, hyperphosphorylation and subsequent mislocalization of tau are downstream steps in AD pathology and are induced by Aβ oligomers [35, 36]. Meanwhile, oxidative stress and inflammation are presumed to facilitate the phosphorylation of tau [37]. Therefore, the reduced levels of p-tau may be due to SFN-induced reductions in Aβ oligomer levels, oxidative stress, and inflammation. Tau is highly expressed in the axons of healthy neurons and has been found in both the presynaptic and postsynaptic compartments of the human brain [38, 39]. Hyperphosphorylated tau aberrantly accumulates in dendrites and affects postsynaptic function [40, 41]. These studies indicate that dysfunctional tau directly affects synaptic signaling networks, leading to synaptic dysfunction. Conceivably, the significantly reduced levels of p-tau that were observed in our study preserved neuronal dendritic integrity and maintained normal cognition at a level equivalent to that of the wild-type mice.

Conclusion

In summary, we found that SFN reduced Aβ oligomer generation, tau phosphorylation, oxidative stress, and inflammation, and improved cognition in PS1V97L Tg mice. This study provides well supported evidence for SFN as a potential anti-Aβ oligomer compound that is warrants further study.

Footnotes

ACKNOWLEDGMENTS

This study was supported by the Key Project of the National Natural Science Foundation of China (81530036); the National Key Scientific Instrument and Equipment Development Project (31627803); Mission Program of Beijing Municipal Administration of Hospitals (SML20150801); Beijing Scholars Program, and Beijing Brain Initiative from Beijing Municipal Science & Technology Commission (Z161100000216137); CHINA-CANADA Joint Initiative on Alzheimer’s Disease and Related Disorders (81261120571); Beijing Municipal Commission of Health and Family Planning (PXM2017_026283_000002).

Authors’ disclosures available online ().

References

Hardy

, Selkoe

(2002) The amyloid hypothesis of Alzheimer’s disease: Progress and problems on the road to therapeutics. Science 297, 353–356.

Karran

, Mercken

, De Strooper

(2011) The amyloid cascade hypothesis for Alzheimer’s disease: An appraisal for the development of therapeutics. Nat Rev Drug Discov 10, 698–712.

Villemagne

, Pike

, Chetelat

, Ellis

, Mulligan

, Bourgeat

, Ackermann

, Jones

, Szoeke

, Salvado

, Martins

, O’Keefe

, Mathis

, Klunk

, Ames

, Masters

, Rowe

(2011) Longitudinal assessment of Abeta and cognition in aging and Alzheimer disease. Ann Neurol 69, 181–192.

Galimberti

, Scarpini

(2016) Emerging amyloid disease-modifying drugs for Alzheimer’s disease. Expert Opin Emerg Drugs 21, 5–7.

Larson

, Lesne

(2012) Soluble Abeta oligomer production and toxicity. J Neurochem 120(Suppl 1), 125–139.

Lesne

, Sherman

, Grant

, Kuskowski

, Schneider

, Bennett

, Ashe

(2013) Brain amyloid-beta oligomers in ageing and Alzheimer’s disease. Brain 136, 1383–1398.

Hefti

, Goure

, Jerecic

, Iverson

, Walicke

, Krafft

(2013) The case for soluble Abeta oligomers as a drug target in Alzheimer’s disease. Trends Pharmacol Sci 34, 261–266.

Rosenblum

(2014) Why Alzheimer trials fail: Removing soluble oligomeric beta amyloid is essential, inconsistent, and difficult. Neurobiol Aging 35, 969–974.

Hunter

, Brayne

(2012) Relationships between the amyloid precursor protein and its various proteolytic fragments and neuronal systems. Alzheimers Res Ther 4, 10.

10.

Murakami

, Irie

, Ohigashi

, Hara

, Nagao

, Shimizu

, Shirasawa

(2005) Formation and stabilization model of the 42-mer Abeta radical: Implications for the long-lasting oxidative stress in Alzheimer’s disease. J Am Chem Soc 127, 15168–15174.

11.

Kiyota

, Okuyama

, Swan

, Jacobsen

, Gendelman

, Ikezu

(2010) CNS expression of anti-inflammatory cytokine interleukin-4 attenuates Alzheimer’s disease-like pathogenesis in APP+PS1 bigenic mice. FASEB J 24, 3093–3102.

12.

Philippens

, Ormel

, Baarends

, Johansson

, Remarque

, Doverskog

(2017) Acceleration of amyloidosis by inflammation in the amyloid-beta marmoset monkey model of Alzheimer’s disease. J Alzheimers Dis 55, 101–113.

13.

de Figueiredo

, Binda

, Nogueira-Machado

, Vieira-Filho

, Caligiorne

(2015) The antioxidant properties of organosulfur compounds (sulforaphane). Recent Pat Endocr Metab Immune Drug Discov 9, 24–39.

14.

, Jhang

, Woo

, Kang

, Chong

(2016) Sulforaphane exerts its anti-inflammatory effect against amyloid-beta peptide via STAT-1 dephosphorylation and activation of Nrf2/HO-1 cascade in human THP-1 macrophages. Neurobiol Aging 38, 1–10.

15.

Zhang

, Zhang

, Fang

, Li

, Zhao

, Shi

, An

(2014) Neuroprotective effects of sulforaphane on cholinergic neurons in mice with Alzheimer’s disease-like lesions. Int J Mol Sci 15, 14396–14410.

16.

Zhang

, Lu

, Jia

, Geula

, Pei

, Xu

, Qin

, Liu

, Li

, Pan

(2014) A lifespan observation of a novel mouse model: In vivo evidence supports abeta oligomer hypothesis. PLoS One 9, e85885.

17.

Wang

, Cheng

, Qin

, Jia

(2012) Val97Leu mutant presenilin-1 induces tau hyperphosphorylation and spatial memory deficit in mice and the underlying mechanisms. J Neurochem 121, 135–145.

18.

Yan

, Vassar

(2014) Targeting the beta secretase BACE1 for Alzheimer’s disease therapy. Lancet Neurol 13, 319–329.

19.

Golde

, Koo

, Felsenstein

, Osborne

, Miele

(2013) gamma-Secretase inhibitors and modulators. Biochim Biophys Acta 1828, 2898–2907.

20.

Tarozzi

, Angeloni

, Malaguti

, Morroni

, Hrelia

(2013) Sulforaphane as a potential protective phytochemical against neurodegenerative diseases. Oxid Med Cell Longev 2013, 415078.

21.

Benilova

, Karran

, De Strooper

(2012) The toxic Abeta oligomer and Alzheimer’s disease: An emperor in need of clothes. Nat Neurosci 15, 349–357.

22.

Shankar

, Li

, Mehta

, Garcia-Munoz

, Shepardson

, Smith

, Brett

, Farrell

, Rowan

, Lemere

, Regan

, Walsh

, Sabatini

, Selkoe

(2008) Amyloid-beta protein dimers isolated directly from Alzheimer’s brains impair synaptic plasticity and memory. Nat Med 14, 837–842.

23.

Viola

, Klein

(2015) Amyloid beta oligomers in Alzheimer’s disease pathogenesis, treatment, and diagnosis. Acta Neuropathol 129, 183–206.

24.

Kraft

, Johnson

(2004) Nuclear factor E2-related factor 2-dependent antioxidant response element activation by tert-butylhydroquinone and sulforaphane occurring preferentially in astrocytes conditions neurons against oxidative insult. J Neurosci 24, 1101–1112.

25.

Liddell

(2017) Are astrocytes the predominant cell type for activation of Nrf2 in aging and neurodegeneration? Antioxidants (Basel) 6, E65.

26.

de Oliveira

, de Bittencourt Brasil

, Furstenau

(2017) Sulforaphane promotes mitochondrial protection in SH-SY5Y cells exposed to hydrogen peroxide by an Nrf2-dependent mechanism. Mol Neurobiol. doi: 10.1007/s12035-017-0684-2

27.

Ahmed

, Luo

, Namani

, Wang

, Tang

(2017) Nrf2 signaling pathway: Pivotal roles in inflammation. Biochim Biophys Acta 1863, 585–597.

28.

Siegel

, Bieschke

, Powers

, Kelly

(2007) The oxidative stress metabolite 4-hydroxynonenal promotes Alzheimer protofibril formation. Biochemistry 46, 1503–1510.

29.

Chen

, Kazachkov

, Yu

(2007) Effect of aldehydes derived from oxidative deamination and oxidative stress on beta-amyloid aggregation; pathological implications to Alzheimer’s disease. J Neural Transm (Vienna) 114, 835–839.

30.

Murakami

, Murata

, Noda

, Tahara

, Kaneko

, Kinoshita

, Hatsuta

, Murayama

, Barnham

, Irie

, Shirasawa

, Shimizu

(2011) SOD1 (copper/zinc superoxide dismutase) deficiency drives amyloid beta protein oligomerization and memory loss in mouse model of Alzheimer disease. J Biol Chem 286, 44557–44568.

31.

Tamagno

, Guglielmotto

, Aragno

, Borghi

, Autelli

, Giliberto

, Muraca

, Danni

, Zhu

, Smith

, Perry

, Jo

, Mattson

, Tabaton

(2008) Oxidative stress activates a positive feedback between the gamma- and beta-secretase cleavages of the beta-amyloid precursor protein. J Neurochem 104, 683–695.

32.

Chen

, Zhou

, Liu

, Deng

, Cai

, Tone

, Tong

, Song

(2012) Increased NF-kappaB signalling up-regulates BACE1 expression and its therapeutic potential in Alzheimer’s disease. Int J Neuropsychopharmacol 15, 77–90.

33.

Sultana

, Butterfield

(2010) Role of oxidative stress in the progression of Alzheimer’s disease. J Alzheimers Dis 19, 341–353.

34.

Nagaveni

, Lakshmi

, Prabhakar

(2014) Sulforaphane interaction with amyloid beta 1-40 peptide studied by electrospray ionization mass spectrometry. Rapid Commun Mass Spectrom 28, 2171–2180.

35.

Chabrier

, Blurton-Jones

, Agazaryan

, Nerhus

, Martinez-Coria

, LaFerla

(2012) Soluble abeta promotes wild-type tau pathology in vivo. J Neurosci 32, 17345–17350.

36.

Amar

, Sherman

, Rush

, Larson

, Boyle

, Chang

, Gotz

, Buisson

, Lesne

(2017) The amyloid-beta oligomer Abeta*56 induces specific alterations in neuronal signaling that lead to tau phosphorylation and aggregation. Sci Signal 10, eaal2021.

37.

Escudero-Lourdes

(2016) Toxicity mechanisms of arsenic that are shared with neurodegenerative diseases and cognitive impairment: Role of oxidative stress and inflammatory responses. Neurotoxicology 53, 223–235.

38.

Tai

, Serrano-Pozo

, Hashimoto

, Frosch

, Spires-Jones

, Hyman

(2012) The synaptic accumulation of hyperphosphorylated tau oligomers in Alzheimer disease is associated with dysfunction of the ubiquitin-proteasome system. Am J Pathol 181, 1426–1435.

39.

Tai

, Wang

, Serrano-Pozo

, Frosch

, Spires-Jones

, Hyman

(2014) Frequent and symmetric deposition of misfolded tau oligomers within presynaptic and postsynaptic terminals in Alzheimer’s disease. Acta Neuropathol Commun 2, 146.

40.

Ittner

, Ke

, Delerue

, Bi

, Gladbach

, van Eersel

, Wolfing

, Chieng

, Christie

, Napier

, Eckert

, Staufenbiel

, Hardeman

, Gotz

(2010) Dendritic function of tau mediates amyloid-beta toxicity in Alzheimer’s disease mouse models. Cell 142, 387–397.

41.

Hoover

, Reed

, Su

, Penrod

, Kotilinek

, Grant

, Pitstick

, Carlson

, Lanier

, Yuan

, Ashe

, Liao

(2010) Tau mislocalization to dendritic spines mediates synaptic dysfunction independently of neurodegeneration. Neuron 68, 1067–1081.