Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Functional Structure/Activity Relationships

The effects of astaxanthin and docosahexaenoic acid-acylated
astaxanthin on Alzheimer's disease in APP/PS1 double transgenic mice
Hongxia Che, Qian Li, Tiantian Zhang, Dandan Wang, Lu Yang, Jie Xu,
Teruyoshi Yanagita, Changhu Xue, Yaoguang Chang, and YuMing Wang
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.8b00988 • Publication Date (Web): 25 Apr 2018
Downloaded from http://pubs.acs.org on April 26, 2018

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 30 Journal of Agricultural and Food Chemistry

1 The effects of astaxanthin and docosahexaenoic acid-acylated astaxanthin on Alzheimer's disease

2 in APP/PS1 double transgenic mice

3 Hongxia Che, † Qian Li,† Tiantian Zhang,† Dandan Wang,† Lu Yang,† Jie Xu,† Teruyoshi
⊥ ⊥,*
4 Yanagita,‡ Changhu Xue,†, Yaoguang Chang,†, * Yuming Wang†,

†
6 College of Food Science and Engineering, Ocean University of China, Qingdao, 266003,

7 Shandong, China

‡
8 Laboratory of Nutrition Biochemistry, Department of Applied Biochemistry and Food Science,

9 Saga University, Saga, 840-8502, Japan

⊥
10 Qingdao National Laboratory for Marine Science and Technology, Laboratory of Marine Drugs

11 & Biological products, Qingdao 266237, Shandong, China

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 2 of 30

12 ABSTRACT

13 Alzheimer's disease (AD) is a progressive neurodegenerative disorder with the characteristics of

14 senile plaques, neuro-inflammation, neurofibrillary tangles and destruction of synapse structure

15 stability. Previous studies have verified the protective effects of astaxanthin (AST). However,

16 whether synthesized docosahexaenoic acid-acylated AST diesters (AST-DHA) could delay AD

17 pathogenesis remains unclear. In the present study, APP/PSEN1 (APP/PS1) double transgenic

18 mice were administrated with AST and AST-DHA for 2 months. The results of radial 8-arm maze

19 and Morris water maze tests showed that AST-DHA exerted more significant effects than AST in

20 enhancing learning and memory levels of APP/PS1 mice. Further mechanical studies suggested

21 that AST-DHA was superior to AST in regulating the parameters of oxidative stress, reducing Tau

22 hyper-phosphorylation, suppressing neuro-inflammation and regulating inflammasome expression

23 and activation in APP/PS1 mice. The findings suggested AST-DHA attenuated cognitive disorders

24 by reducing pathological features in APP/PS1 mice, suggesting AST-DHA might be a potential

25 therapeutic agent for Alzheimer's disease.

26 KEYWORDS: Alzheimer's disease, astaxanthin, cognitive disorder, DHA-acylated AST esters,

27 neuro-inflammation

28

ACS Paragon Plus Environment

Page 3 of 30 Journal of Agricultural and Food Chemistry

29 INTRODUCTION

30 The main pathological features of Alzheimer's disease (AD) are senile plaques,

31 neuro-inflammation, and destruction of synapse structure stability 1. There are very few AD drugs

32 on the present market, and these drugs only provide minimal symptomatic relief rather than

33 changes in disorder progression. Therefore, there is a great need for the therapeutic agents to

34 modify AD.

35 Astaxanthin (AST, Fig.1A), one of natural carotenoids, is widely present in marine organisms

36 such as shrimp, crab, krill, salmon, and microalgae 2. Importantly, the hydroxyl and keto moieties

37 on each ionone ring in AST imply its unique properties, especially, the ability to be esterified with

38 fatty acids to increase stability. Usually, AST occurs in esterified forms in nature, which is

39 chemically bound to various types of fatty acids such as oleic, eicosanoic, palmitic and stearic acid.

40 The AST in red crab langostilla (Pleuroncodes planipes) is comprised of about 70% diesters, 12%

41 monoesterified and 10% unesterified AST 3. Interestingly, it has been reported that the main

42 existing form of AST in Atlantic salmon is unesterified AST 4.

43 The bioactivity of AST is usually related to the reduced markers of oxidative damage.

44 Notably, recent evidence has emerged to indicate AST has a broad range of bioactivities including

45 anti-inflammatory, anti-apoptotic properties5. Moreover, AST plays a vital role in reducing

46 neurotoxicity in cell models of AD. In addition, AST could protect PC12 cells from Aβ-induced

47 cytotoxicity by up-regulating heme oxygenase-1 expression via ERK1/2 pathway5-6. Lobos et al.

48 provided further evidence to indicate that AST protected neurons from the noxious effects of Aβ

49 on mitochondrial ROS production and calcium dysregulation 7. Moreover, Cheong et al. have

50 verified that krill oil could enhance cognitive capability and modulate proteomic alterations in

51 brain of D-galactose induced aging mice 8.

52 Docosahexaenoic acid (DHA) is well known for various bioactivities 9. However, few study

53 focused on the protective effects of DHA-acylated AST ester. The APP/PSEN1 (APP/PS1)

54 transgenic mice, co-expressing the mutated Swedish APP gene and the exon-9-deleted variant of
10
55 the presenilin-1 (PS1) gene, are a successfully established transgenic animal models for AD .

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 4 of 30

56 This model displays age-related plaque pathology, inflammatory response, oxidative damage,

57 age-related memory deficits11-12. In the present study, DHA-acylated AST diesters (AST-DHA)

58 was prepared, and APP/PS1 mice were used to compare the neuroprotective actions of AST and

59 AST-DHA and illustrate the possible underlying mechanisms.

60 MATERIALS AND METHODS

61 Chemicals and Reagents. Astaxanthin was obtained from Xinxiang Kesen Food Additives

62 Corporation (Tongxiang, Zhejiang, China). Aβ42 and Aβ40 ELISA kits were purchased from

63 Wuhan Uscn Life Science, Inc. (Wuhan, Hubei, China). The superoxide dismutase (SOD) assay

64 kit and nitric oxide (NO) assay kit were obtained from Nanjing Jiancheng Bioengineering Institute

65 (Nanjing, Jiangsu, China). The Nitric Oxide Synthase ELISA kit was from R&D System

66 (Minneapolis, MN, USA). p-Tau (Ser396) antibody, Tau antibody, p- GSK-3β (Y216+Y279)

67 antibody, GSK-3β antibody, CD11b antibody, GFAP antibody, IL-1β antibody, TNF-α antibody,

68 NLRP3 antibody, Caspase-1 antibody, Bax antibody, Bcl-2 antibody, caspase-9 antibody and

69 β-actin antibody were from Abcam (Cambridge, UK). Caspase-3 antibody and cleaved caspase-3

70 antibody were purchased from Cell Signaling Technology (Boston, USA).

71 Preparation of AST-DHA. AST-DHA (Fig.1A) was synthesized by 4-dimethylaminopyridine

72 catalyzed reaction of AST and DHA in the presence of 1-(3-Dimethylaminopropyl)-

73 3-ethylcarbodiimide hydrochloride system with nitrogen protection and light isolation according
13
74 to our previous study . Dichloromethane was added to the system after the reaction for 3 h at

75 25 ℃. The organic phase is recovered following the subsequent cleaning by 1 M hydrochloric acid,

76 saturated sodium bicarbonate and saturated sodium chloride. The organic mixture was evaporated

77 to dryness to obtain the AST-DHA (the purity > 90%), which was confirmed by HPLC-DAD

78 according to our previously published study.

79 Animals and Treatments. The animal study protocol was proved by the Animal Ethics

80 Committee of College of Food Science and Engineering of Ocean University of China. All the

81 animals were housed at the Laboratory Animal Facility at the Ocean University of China. The

82 research was conducted in accordance with the Guide for the Care and Use of Laboratory Animals

ACS Paragon Plus Environment

Page 5 of 30 Journal of Agricultural and Food Chemistry

83 (8 th edition, Institute of Laboratory Animal Resources on Life Sciences, National Research

84 Council, National Academy of Sciences, Washington DC). APP/PS1 transgenic mice (half male

85 and half female, weight of 20-25 g, aged 3 months) and eight of their wild-type littermates as

86 normal control group (Control) were obtained from Vital River Laboratories (Beijing, China). All

87 the mice were acclimatized for 1 week under a 12 h/12 h light/dark cycle at 23 °C with 60 ± 10%

88 humidity and provided with food and water ad libitum. The APP/PS1 transgenic mice were

89 randomly divided into Model group, AST group and AST-DHA group. All the mice were

90 supplemented with AIN-93G diet. The mice in AST and AST-DHA group were supplemented

91 with 0.2% AST and AST-DHA for 60 days, respectively. Then the water maze and radial 8-arm

92 maze tests were used to determine learning and memory levels. After that, the mice were

93 sacrificed by rapid decapitation. The cerebral cortex and hippocampus were separated from the

94 whole brain and weighed, then frozen with liquid nitrogen and stored at -80 °C until use.

95 Radial 8-Arm Maze Test. Spatial learning ability was determined by the radial 8-arm maze test
14
96 according to a previous method . The apparatus composed of a central octagonal plate and 8

97 radiating arms was placed at 1 m above the floor. A food cup was located at the end of each

98 radiating arm. Before training, the mice were deprived of half of the diet during the test. The baits

99 were restricted to the food cups. On the first two training days, the food pellet (45 mg) were

100 located at the food cups (eight trails) and central octagonal plate, the mice of the same group were

101 allowed to explore for food for 10 minutes together. On the third and fourth days, every mouse

102 was allowed to explore for food for 10 minutes. On the last day, four identical arms were baited

103 with a single 45 mg food pellet. Each trial continued until all four baits had been consumed or

104 until 10 minutes had elapsed. The numbers of reference memory errors and working memory

105 errors were conducted.

106 Water Maze Test. A circular stainless-steel pool with water (21-23 °C) was divided into four

107 quadrants. A circular black escape platform was located in the center of one quadrant. The mice

108 were trained to find the platform for five consecutive days. The time of finding the platform was

109 recorded as latency. The swim paths, distances, and latencies taken to the platform were

110 monitored with a video camera. Probe test without platform on the sixth day was used to

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 6 of 30

111 determine spatial memory retention. The mice were placed in a position opposite the platform

112 location and allowed to swim for 60 sec. The number crossing over the previous position of the

113 platform and the time spent in the target quadrant were recorded as measures for spatial memory.

114 Immunohistochemistry. For paraffin sections, mice were killed and systemically perfused with

115 phosphate buffered saline (PBS) and 4% buffered paraformaldehyde through the left ventricle to

116 wash out blood cells. Then brain samples were collected, ehydrated and embedded in paraffin

117 using standard techniques. Sections (5 µm) were cut and deparaffinized. After incubation in

118 methanol containing 3% H2O2 for 15 minutes to block endogenous peroxidase activity,

119 nonspecific binding was done with universal blocking reagent for 30 minutes at room temperature.

120 The sections were incubated with the primary antibody against anti-Aβ (1:100; Servicebio),

121 anti-Iba1 (1:500; Servicebio) and anti-GFAP (1:500; Servicebio) at 4 °C for overnight followed by

122 several washing steps in PBS. Following incubation with biotinylated goat anti-rabbit IgG (1/200)

123 for 50 minutes, staining was done through incubation with peroxidase streptoavidin and

124 diaminobenzidine (DAB) at room temperature. Specific primary antibody was omitted in negative

125 control of the reactions. After counter staining nuclei with Mayer’s haematoxylin, the

126 immunopositive amyloid plaques in cortex and hippocampus were observed and counted. The

127 images were taken using a 5 MP Canon A95 camera integrated to the microscope and were

128 evaluated using image-J analysis.

129 Determination of Aβ Concentration by ELISA. Soluble and insoluble Aβ were extracted by the

130 previous method 15. Briefly, the right hippocampus was homogenized in TBS (pH8.0) containing

131 protease inhibitors. Samples were sonicated and centrifuged. The supernatant was used to

132 determine soluble Aβ40 and Aβ42 concentrations, whereas the TBS-insoluble pellet was firstly

133 sonicated in 2% SDS. To analyze the insoluble Aβ content, the SDS-insoluble pellet was

134 dissolved and sonicated in 70% formic acid. The extract was neutralized with 0.5 M Tris before

135 loading on the ELISA plate. The soluble/insoluble Aβ40 and Aβ42 levels were determined by

136 ELISA kits according to the manufacture’s instruction.

137 The Measurement of Oxidative Stress Parameters. The brain was prepared as a tissue

138 homogenate in 0.9% saline solution for the determination of protein concentration using a BCA

ACS Paragon Plus Environment

Page 7 of 30 Journal of Agricultural and Food Chemistry

139 protein assay kit. The concentrations of nitric oxide (NO), inducible nitric oxide synthase (NOS)

140 and the activity of superoxide dismutase (SOD) were measured by the manufacture’s instruction.

141 Western Blot Analysis. The total protein and RNA of hippocampus were extracted by the

142 total-DNA-RNA-Protein kit. Equal amounts of protein were separated on 5-12% SDS-PAGE gels

143 and transferred to poly membranes. The membranes were incubated with antibodies against p-Tau

144 antibody (1:5000), Tau antibody (1:2000), p-GSK3β antibody (1:1000), GSK-3β antibody (1:500),

145 CD11b antibody (1:500), GFAP antibody (1:5000), IL-1β antibody (1:5000), TNF-α antibody

146 (1:200), NLRP3 (1:2000), Caspase 1 (1:1000) at 4 °C for overnight. After this, membranes were

147 incubated with specific horse radish peroxidase (HRP)-conjugated secondary antibodies (1:3000)

148 at room temperature 2 h and the blots were evaluated with chemiluminescent horseradish

149 peroxidase substrate. Then the blots were visualized by enhanced chemiluminescence (ECL)

150 substrate with UVP Auto Chemi Image system. Protein loading was evaluated by anti-β-actin

151 antibody (1:2000)..

152 Statistical Analysis. Data were expressed as mean ± standard deviation (SD). Statistical analyses

153 were evaluated by Student’s t test and Tukey’s test using SPSS 18.0. P < 0.05 was considered

154 statistically significant. Different letters indicated significant differences between each group.

155 RESULTS

156 AST and AST-DHA Improved Cognitive Disorder in APP/PS1 Mice. The spatial learning was

157 measures by radial 8-arm maze and Morris water maze tests. The radial 8-arm maze results

158 showed that APP/PS1 transgenic mice remarkably enhanced the number of reference memory

159 errors and working memory errors in comparison with non-APP/PS1 transgenic mice (Fig 1B and

160 C). Interestingly, AST and AST-DHA obviously reduced the number of reference memory errors

161 and working memory errors in APP/PS1 transgenic mice (Fig 1B and C), and no remarkable

162 difference was observed between these two groups.

163 The results of escape latency showed that the mice of control group had better performance

164 than APP/PS1 transgenic mice from day 1 to day 5, suggesting APP/PS1 transgenic mice

165 exhibited significant deficiency in spatial learning ability (Fig. 1D). The treatment with AST and

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 8 of 30

166 AST-DHA significantly improved the behaviors of APP/PS1 transgenic mice. Interestingly,

167 AST-DHA had a more significant effect than AST in alleviating spatial deficits of APP/PS1

168 transgenic mice.

169 The spatial memory results showed that APP/PS1 transgenic mice in model group spent less

170 time in the target quadrant and less crossing number than the control group (Fig. 1E and F). The

171 mice treated with AST and AST-DHA exerted similar effects with no statistical difference in

172 improving the time spent in the target quadrant. Interestingly, AST-DHA was superior to AST in

173 improving the number of crossing platform.

174 AST and AST-DHA Regulated Aβ Levels in APP/PS1 Mice. Brain sections were

175 immunostained with anti-Aβ antibody to show the Aβ deposition in the cortex and hippocampus

176 (Fig.2A), which was analyzed the number and area of plaques (Fig.2B and C). Compared with

177 control group, the amyloid plaques number in the cortex and hippocampus of APP/PS1 transgenic

178 mice were remarkably increased. Interestingly, AST and AST-DHA treatment could reduce

179 amyloid plaques number in cortex and hippocampus of APP/PS1 transgenic mice. Importantly,

180 AST-DHA was superior to AST in suppressing the number of amyloid plaques in cortex. The

181 results of quantitative analysis of Aβ showed that AST and AST-DHA suppressed the Aβ load.

182 AST-DHA exhibited more statistical effects than AST in reducing the Aβ load in hippocampus.

183 The soluble/insoluble Aβ40 and Aβ42 in the hippocampus were determined by ELISA kits,

184 and the results were shown in Fig. 3 A-D. The APP/PS1 transgenic mice exhibited higher soluble

185 and insoluble Aβ40 and Aβ42 levels than control group. Supplementation of AST and AST-DHA

186 exhibited a remarkable decrease of soluble and insoluble Aβ40 and Aβ42 levels in a certain degree.

187 Notably, AST-DHA exhibited more statistical effects than AST in reducing soluble Aβ40/Aβ42

188 levels. Unexpectedly, AST showed a better effect in reducing insoluble Aβ40 level than

189 AST-DHA group. No significant difference between AST and AST-DHA was observed in

190 inhibiting insoluble Aβ42 generation.

191 Following the soluble/insoluble Aβ40 and Aβ42 measurement, ADAM10, BACE1 and

192 Nicastrin were detected by western blotting. Compared with control group, the ADAM10 level of

ACS Paragon Plus Environment

Page 9 of 30 Journal of Agricultural and Food Chemistry

193 model group was significantly decreased, and BACE1 and Nicastrin expression were obviously

194 increased. Notably, AST and AST-DHA treatment exerted similar effects in decreasing Nicastrin

195 level. However, there were no significant improvement for AST and AST-DHA in protein

196 expressions of ADAM10 and BACE1.

197 Effects of AST and AST-DHA on Oxidative Stress. Oxidative stress plays a central role in the

198 physiopathology of AD. Thus, we further investigated the effects of AST and AST-DHA on

199 oxidative stress to confirm the neuroprotective effects of AST and AST-DHA against AD. The

200 indexes of oxidative stress including SOD, NO, NOS were detected (Fig.4). Compared with

201 control group, the SOD activity in model group was significantly decreased, meanwhile, NOS

202 activity and NO level were obviously increased. Importantly, AST and AST-DHA treatment

203 significantly recovered the activity of SOD as well as declining NO and NOS levels, in which

204 AST-DHA was superior to AST in up-regulating SOD activity and down-regulating NO and NOS

205 levels.

206 The Effects of AST and AST-DHA on the Expression of p-GSK-3β and p-Tau. The effects of

207 AST and AST-DHA treatment on GSK-3β activity and p-Tau protein expression were evaluated

208 by Western blotting (Fig. 5A). Compared with the control group, the expression of p-GSK-3β

209 level in model group was obviously increased (p < 0.05, Fig. 5B). AST-DHA significantly

210 suppressed the expression of GSK-3β phosphorylation. Notably, AST treatment had no effects in

211 regulating the p-GSK-3β expression.

212 Following GSK3β activity, we also investigated the effects of dietary supplements of AST

213 and AST-DHA on p-Tau protein expression. The expression of p-Tau in model group was

214 remarkably increased compared with control group (Fig. 5C). Notably, dietary supplementation of

215 AST and AST-DHA could obviously reverse these changes, and AST-DHA exerted more

216 significant effects than AST in reducing the expression of p-Tau.

217 Inhibitory Effects of AST and AST-DHA on Neuro-inflammation in APP/PS1 Mice. The

218 neuro-inflammation induced by activated microglia and astrocytes is related to AD development.

219 To test the effects of AST and AST-DHA on neuro-inflammatory processes in APP/PS1

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 10 of 30

220 transgenic mice, immunohistochemical analysis and western blotting assay of astrogliosis and

221 microgliosis were performed using the astroglial marker (GFAP) and the microglial marker (Iba1

222 and CD11b, Fig. 6 A-D). The results showed that GFAP, Iba1 and CD11b were markedly

223 increased in the APP/PS1 transgenic mice but were significantly reduced in the AST and

224 AST-DHA treated mice both in immunohistochemical analysis and western blotting assay.

225 Importantly, AST-DHA was superior to AST in regulating the activation of microglia and

226 astrocytes.

227 Cytokines secreted by activated microglia and astrocytes are crucial in the inflammatory

228 processes of AD. The levels of IL-1β and TNF-α in AD mice were determined to investigate the

229 effects of AST and AST-DHA on cytokine production (Fig. 6 B, E and F). Compared with

230 no-transgenic mice, the expression of TNF-α in the APP/PS1 transgenic mice was markedly

231 increased. Notably, AST and AST-DHA supplementation could significantly reduce the

232 expression of TNF-α, and AST-DHA was superior to AST. Only AST-DHA significantly

233 suppressed the expression of IL-1β, and no statistical difference was found between AST group

234 and model group.

235 Inhibitory Effects of AST and AST-DHA on Inflammasome Activation in APP/PS1 Mice.

236 The NLRP3 inflammasome activation initiates an inflammatory response through caspase-1

237 activation, resulting in inflammatory cytokine IL-1β℃maturation and secretion. Therefore,

238 western blotting assay was performed to detect the NLRP3 inflammasome activation-related

239 proteins (Fig. 7). Unexpectedly, no statistical difference was observed in the expression of NLRP3

240 in these four groups. Compared with the control group, the expression of ASC in APP/PS1

241 transgenic mice was significantly reduced. However, AST and AST-DHA supplementation further

242 decreased the expression of ASC protein. Compared with the non-transgenic mice, the expression

243 of Caspase 1 and Pro-IL-1β in APP/PS1 transgenic mice was remarkably increased. Interestingly,

244 AST and AST-DHA supplementation obviously reversed the increase of Caspase 1 and Pro-IL-1β,

245 and AST-DHA exerted more significant effects than AST.

246 AST and AST-DHA Suppressed Apoptosis in APP/PS1 Mice. Western blotting assay was

247 performed to detect the apoptosis-related proteins in APP/PS1 mice (Fig. 8). Unexpectedly, AST

ACS Paragon Plus Environment

Page 11 of 30 Journal of Agricultural and Food Chemistry

248 and AST-DHA supplementation had no corresponding improvement in regulating Bcl-2 and Bax

249 protein expression. The relative densities of Caspase-9 and Caspase-3 in APP/PS1 transgenic mice

250 were remarkably increased compared with the mice in the control group. Following AST and

251 AST-DHA administration, the protein levels of Caspase-9 and Caspase-3 were obviously reduced.

252 Notably, AST-DHA was superior to AST in reducing the protein expressions of Caspase-9/-3. Both

253 AST and AST-DHA could obviously suppress cleaved Caspase-3 expression with a similar degree.

254 DISCUSSION

255 The pathological characteristics of AD brains mainly include extracellular Aβ plaques,

256 intracellular neurofibrillary tangles, massive neuronal cell and synapse loss, and

257 neuro-inflammation 1. APP/PS1 transgenic mouse is one of well-established AD animal model16-17.

258 To illustrate the possibility of false positivity, we determined the efficacy of AST and AST-DHA in

259 8-Arm maze and Morris water maze test in the present study. Both AST and AST-DHA treatment

260 could improve spatial learning ability in APP/PS1 mice by 8-arm maze apparatus, which were

261 further corroborated by the Morris water maze test. The results showed that both AST and

262 AST-DHA could improve the learning and memory skills, in which the AST-DHA was superior to

263 AST in Morris water maze test. The different result of these two behavior tests may be attributed

264 to that 8-Arm maze test is appetitive reinforcement, and Morris water maze test is aversive

265 reinforcement. Furthermore, the learned behavior in these two tests is also different among

266 learning tasks. The Morris water maze test usually learns a spatial task, in which rats have to learn

267 complex behavioral strategies to recognize the platform to escape from water.

268 The accumulation and deposition of Aβ are considered to play an important role in the
18
269 pathogenesis of AD, which is the dominant theory of AD in the past few decades . Aβ is

270 produced by the proteolytic processing of amyloid precursor protein (APP). An important way to

271 process APP is the nonamyloidogenic pathway, in which α-secretase cleaves the Aβ domain in

272 APP, thereby precluding the formation of intact Aβ. However, under normal circumstances, a

273 small amount of APP is processed via the amyloidogenic pathway, in which Aβ is released from

274 APP by β-site amyloid precursor protein cleaving enzyme 1 (BACE1) and γ-secretases. ADAM10

275 and Nicastrin is one of the components of α-secretase and γ-secretases, respectively. Aβ as a

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 12 of 30

276 monomer readily aggregates to form multimeric complexes 19. These complexes are composed of

277 soluble Aβ ranging from oligomers to protofibrils and insoluble Aβ such as amyloid plaques. The

278 immunohistochemistry results showed that AST-DHA treatment showed better effects than AST in

279 decreasing the number of senile plaques and Aβ plaques load in the cortex and hippocampus of

280 APP/PS1 mice, which was consistent with the results of ELISA kits. Interestingly, AST and

281 AST-DHA only showed notable effects in regulating the expression of Nicastrin instead of

282 BACE1 and ADAM10. The Aβ decrease might partly depend on the clearance mechanism, as it

283 has been reported that n-3 LCPUFAs significantly promoted interstitial Aβ clearance from the

284 brain to resist Aβ injury 20.

285 AST treatment is usually related to the reduced markers of oxidative damage. AST may

286 increase the levels of endogenous antioxidant enzymes including superoxide dismutase and
21
287 catalase . AST-DHA exhibited more effective antioxidant capacity than AST, indicating the

288 indispensable role of DHA in AST-DHA.

289 Increasing evidence has indicated Aβ induced hyper-phosphorylation of Tau protein by

290 GSK3β activation 22-23. The results showed that AST and AST-DHA significantly decreased p-Tau

291 and p-GSK3β, and AST-DHA was superior to AST. The cognitive improvement of AST-DHA on

292 APP/PS1 transgenic mice are mainly attributed to the neuroprotective effects on GSK3β activation

293 and tau hyper-phosphorylation.

294 Uncontrolled microglia and astrocytes activation, and sustained inflammatory responses may

295 contribute independently to neurodegeneration 24-25. It can not only be a consequence but also be a
26
296 trigger of pathology . The activation of microglia and astrocytes were found in the brain of

297 APP/PS1 transgenic mice, which was accompanied by a strong increase of TNF-α rather than

298 IL-1β compared with the non-transgenic mice. Following AST and AST-DHA treatment,

299 pro-inflammatory cytokine levels were greatly reduced, especially in the AST-DHA group,

300 indicating that the anti-inflammatory effects of AST and AST-DHA may account for the reduced

301 Aβ level and cognitive improvement in APP/PS1 transgenic mice.

ACS Paragon Plus Environment

Page 13 of 30 Journal of Agricultural and Food Chemistry

302 Neuro-inflammatory cascades depend on the activation of NLRP3 inflammasome, which was
27
303 crucial in neurodegenerative diseases . It has been proved that the toxicity of Aβ can activate

304 NLRP3 inflammasome, process IL-1β and IL-18, and finally induce AD pathology and tissue

305 damage 28. Moreover, in AD transgenic mouse model, the inhibition of NLRP3 can largely protect

306 memory loss and decrease Aβ deposition. A chronic administration of AST and AST-DHA in

307 APP/PS1 transgenic mice led to a remarkable alteration in NLRP3 inflammasome, in which

308 AST-DHA was superior to AST. The effects of AST and AST-DHA on glial activation and NLRP3

309 inflammasome confirmed the importance of AST-DHA in regulating inflammatory responses.

310 Numerous studies on mitochondria dysfunction revealed that the mitochondria were the

311 central of oxidative stress induced apoptosis29-30. Bcl-2 is an anti-apoptotic protein, while Bax has

312 the opposite function. Caspase-9/-3 are the initiator and executioner, respectively, which typically
31
313 predominate in neurodegenerative diseases . Both AST and AST-DHA could significantly

314 decrease the expression Caspase-9/-3 and cleaved Caspase-3, and the effect of AST-DHA was

315 superior to AST. However, no expected effects of AST and AST-DHA on reducing Bax and

316 improving Bcl-2 were observed in APP/PS1 transgenic mice. Apoptosis is a complex and precisely

317 controlled process, which is regulated by multiple pathways. We speculated that the decreased

318 Caspase-9 and Caspase-3 were not only regulated by the Bcl-2/Bax, other pathways might be

319 related to regulating Caspase family.

320 In summary, both AST and AST-DHA could improve AD in different degrees. AST-DHA

321 exhibited better actions than AST in improving learning and memory abilities by the behavior

322 experiments. The further mechanical research indicated AST-DHA exerted more remarkable

323 functions than AST on inhibiting Aβ generation, regulating oxidative stress, suppressing the

324 hyperphosphorylation of Tau and GSK-3β, and reducing neuroglial activation and

325 neuro-inflammation (Fig. 9). Therefore, AST and AST-DHA may be applied as food supplements

326 and/or functional ingredients to relieve neurodegenerative disease.

327 Author Information

328 * E-mail: changyg@ouc.edu.cn; phone: +86 0532 82032597; fax: +86 0532 82032468

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 14 of 30

329 *E-mail: wangyuming@ouc.edu.cn; phone: +86 0532 82032597; fax: +86 0532 82032468

330 Author contributions

331 Hongxia Che and Qian Li designed and conducted the research. Qian Li, Dan-dan Wang and Lu

332 Yang analyzed the data; Hongxia Che wrote the manuscript; Tian-tian Zhang, Jie Xu and

333 Teruyoshi Yanagita revised the manuscript. Yuming Wang, Changhu Xue and Yaoguang Chang

334 had primary responsibility for the final content. All authors read and approved the final

335 manuscript.

336 Funding

337 This work was supported by grants from State Key Program of National Natural Science of China

338 (No. 31330060) and National Natural Science Foundation of China-Shandong Joint Fund for

339 Marine Science Research Centers (U1606403), and the Fundamental Research Funds for the

340 Central Universities (No. 201762028).

341 Conflicts of interest

342 All the authors declare that there is no conflict of interest for any of them.

343 List of Abbreviations

344 AD, Alzheimer’s disease; AST, Astaxanthin; AST-DHA, Docosahexaenoic acid-acylated AST

345 diesters; APP, amyloid precursor protein; PS1, presenilin; Aβ, β-amyloid; SOD, superoxide

346 dismutase; NOS, inducible nitric oxide synthase; NO, nitric oxide; GFAP, glial fibrillary acidic

347 protein; TNF-α, tumor necrosis factor α; IL-1β, interleukin 1beta; Bcl-2, B-cell lymphoma 2; Bax,

348 Bcl-2-associated X Protein; NLRP3, nucleotide-binding oligomerization domain (NOD)-like

349 receptor containing pyrin domain 3; ASC, apoptosis-associated speck-like protein containing a

350 CARD

ACS Paragon Plus Environment

Page 15 of 30 Journal of Agricultural and Food Chemistry

351 REFERENCES

352 1. Cai, H.; Wang, Y.; He, J.; Cai, T.; Wu, J.; Fang, J.; Zhang, R.; Guo, Z.; Guan, L.; Zhan, Q.

353 Neuroprotective effects of bajijiasu against cognitive impairment induced by amyloid-β in APP/PS1

354 mice. Oncotarget. 2017, 54, 92621.

355 2. Ambati, R. R.; Moi, P. S.; Ravi, S. Astaxanthin: sources, extraction, stability, biological activities

356 and its commercial applications--a review. Mar. Drugs. 2014, 1, 128-52.

357 3. Coral-Hinostroza, G. N.; Bjerkeng, B. Astaxanthin from the red crab langostilla (Pleuroncodes

358 planipes): optical R/S isomers and fatty acid moieties of astaxanthin esters. Comp. Biochem. Phys. B.

359 2002, 3, 437.

360 4. Sheehan, E. M.; O'Connor, T. P.; Pja, S.; Buckley, D. J.; Fitzgerald, R. Stability of astaxanthin and

361 canthaxanthin in raw and smoked Atlantic salmon (Salmo salar) during frozen storage. Food Chem.

362 1998, 3, 313-317.

363 5. Hussein, G., Sankawa, U., Goto, H., Matsumoto, K., Watanabe, H. Astaxanthin, a carotenoid

364 with potential in human health and nutrition. J. Nat. Prod. 2006, 3, 443-449.

365 6. Wang, H. Q.; Sun, X. B.; Xu, Y. X.; Zhao, H.; Zhu, Q. Y.; Zhu, C. Q. Astaxanthin upregulates

366 heme oxygenase-1 expression through ERK1/2 pathway and its protective effect against

367 beta-amyloid-induced cytotoxicity in SH-SY5Y cells. Brain Res. 2010, 1, 159-167.

368 7. Lobos, P.; Bruna, B.; Cordova, A.; Barattini, P.; Galaz, J. L.; Adasme, T.; Hidalgo, C.; Munoz, P.;

369 Paula-Lima, A. Astaxanthin Protects Primary Hippocampal Neurons against Noxious Effects of

370 Aβ-Oligomers. Neural. Plast.. 2016, 1, 3456783.

371 8. Cheong, L. Z.; Sun, T.; Li, Y.; Zhou, J.; Lu, C.; Li, Y.; Huang, Z.; Su, X. Dietary krill oil enhances

372 neurocognitive functions and modulates proteomic changes in brain tissues of d-galactose induced

373 aging mice. Food Funct. 2017, 5, 2038-2045.

374 9. Shahidi, F.; Ambigaipalan, P. Omega-3 Polyunsaturated Fatty Acids and Their Health Benefits.

375 Annu. Rev. Food Sci. T. 2018, 9, 345-381.

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 16 of 30

376 10. Reiserer, R. S.; Harrison, F. E.; Syverud, D. C.; Mcdonald, M. P. Impaired spatial learning in the

377 APPSwe + PSEN1DeltaE9 bigenic mouse model of Alzheimer's disease. Genes Brain Behav. 2007, 1,

378 54-65.

379 11. Xu, Z.; Na, X.; Chen, Y.; Huang, H.; Marshall, C.; Gao, J.; Cai, Z.; Wu, T.; Gang, H.; Ming, X.

380 Deletion of aquaporin-4 in APP/PS1 mice exacerbates brain Aβ accumulation and memory deficits.

381 Mol. Neurodegener. 2015, 1, 58.

382 12. Huang, H.; Nie, S.; Cao, M.; Marshall, C.; Gao, J.; Xiao, N.; Hu, G.; Xiao, M. Characterization of

383 AD-like phenotype in aged APPSwe/PS1dE9 mice. Age. 2016, 4, 1-20.

384 13. Yang, L.; Xuemin, L. I.; Zhou, Q.; Zhaojie, L. I.; Jie, X. U.; Xue, C. Synthesis and Mass

385 Spectrometry Analysis of Docosahexaenoic Acid Astaxanthin Ester. Food Sci. 2017, 2, 220-226.

386 14. Enomoto, T.; Ishibashi, T.; Tokuda, K.; Ishiyama, T.; Toma, S.; Ito, A. Lurasidone reverses

387 MK-801-induced impairment of learning and memory in the Morris water maze and radial-arm maze

388 tests in rats. Behav. Brain res. 2008, 2, 197.

389 15. Che, H.; Zhou, M.; Zhang, T.; Zhang, L.; Ding, L.; Yanagita, T.; Xu, J.; Xue, C.; Wang, Y.

390 Comparative study of Phosphatidylcholine rich in DHA or EPA on Alzheimer's Disease and the

391 possible involved mechanisms in CHO-APP/PS1 cell and SAMP8 mice. Food Funct. 2017, 2, 197-207.

392 16. Deng, M.; Huang, L.; Ning, B.; Wang, N.; Zhang, Q.; Zhu, C.; Fang, Y. β-asarone improves

393 learning and memory and reduces Acetyl Cholinesterase and Beta-amyloid 42 levels in APP/PS1

394 transgenic mice by regulating Beclin-1-dependent autophagy. Brain Res. 2016, 1652, 188-194.

395 17. Jiang, T.; Yu, J. T.; Zhu, X. C.; Tan, M. S.; Wang, H. F.; Cao, L.; Zhang, Q. Q.; Shi, J. Q.; Gao, L.;

396 Qin, H. Temsirolimus promotes autophagic clearance of amyloid-β and provides protective effects in

397 cellular and animal models of Alzheimer's disease. Pharmacol. Res. 2014, 81, 54-63.

398 18. Takahashi, R. H.; Nagao, T.; Gouras, G. K. Plaque formation and the intraneuronal accumulation

399 of β-amyloid in Alzheimer's disease. Pathology. International. 2017, 4, 185.

400 19. Riqiang Yan, R. V. Targeting the β secretase BACE1 for Alzheimer's disease therapy. Lancet

401 Neurology. 2014, 3, 319-29.

ACS Paragon Plus Environment

Page 17 of 30 Journal of Agricultural and Food Chemistry

402 20. Ren, H.; Luo, C.; Feng, Y.; Yao, X.; Shi, Z.; Liang, F.; Kang, J. X.; Wan, J. B.; Pei, Z.; Su, H.

403 Omega-3 polyunsaturated fatty acids promote amyloid-β clearance from the brain through mediating

404 the function of the glymphatic system. Faseb J. 2017, 1, 282.

405 21. Grimmig, B.; Kim, S. H.; Nash, K.; Bickford, P. C.; Shytle, R. D. Neuroprotective mechanisms of

406 astaxanthin: a potential therapeutic role in preserving cognitive function in age and neurodegeneration.

407 Geroscience. 2017, 1, 1-14.

408 22. Vossel, K. A.; Xu, J. C.; Fomenko, V.; Miyamoto, T.; Suberbielle, E.; Knox, J. A.; Ho, K.; Kim, D.

409 H.; Yu, G. Q.; Mucke, L. Tau reduction prevents Aβ-induced axonal transport deficits by blocking

410 activation of GSK3β. J. Cell. Biol. 2015, 3, 419-433.

411 23. Simic, G.; Leko, M. B.; Wray, S.; Harrington, C.; Delalle, I.; Natasa, J. M.; Bazadona, D.; Buee,

412 L.; Silva, R. D.; Giovanni, G. D.; Wischik, C. Tau Protein Hyperphosphorylation and Aggregation in

413 Alzheimer's Disease and Other Tauopathies, and Possible Neuroprotective Strategies. Biomol. 2016, 1,

414 6.

415 24. Heppner, F. L.; Ransohoff, R. M.; Becher, B. Immune attack: the role of inflammation in

416 Alzheimer disease. Nat. Rev. Neurosci. 2015, 6, 358-72.

417 25. McGeer, E. G.; McGeer, P. L. Inflammatory processes in Alzheimer's disease. Prog.

418 Neuro-Psychoph. 2003, 5, 741-749.

419 26. Liao, W.; Liu, Z.; Zhang, T.; Sun, S.; Ye, J.; Li, Z.; Mao, L.; Ren, J. Enhancement of

420 Anti-Inflammatory Properties of Nobiletin in Macrophages by a Nano-Emulsion Preparation. J. Agr.

421 Food Chem. 2017, 1, 91-98.

422 27. Zhou, K.; Shi, L.; Wang, Y.; Chen, S.; Zhang, J. Recent Advances of the NLRP3 Inflammasome in

423 Central Nervous System Disorders. J. Immunol. Res. 2016, 2, 1-9.

424 28. Tan, M. S.; Yu, J. T.; Jiang, T.; Zhu, X. C.; Tan, L. The NLRP3 inflammasome in Alzheimer's

425 disease. Mol. neurobio. 2013, 3, 875-82.

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 18 of 30

426 29. Liao, W.; Lai, T.; Chen, L.; Fu, J.; Sreenivasan, S. T.; Yu, Z.; Ren, J. Synthesis and

427 Characterization of a Walnut Peptides Zinc Complex and Its Antiproliferative Activity against Human

428 Breast Carcinoma Cells through the Induction of Apoptosis. J. Agr. Food Chem. 2016, 7, 849-859.

429 30. Liao, W.; Zhang, R.; Dong, C.; Yu, Z.; Ren, J. Novel walnut peptide-selenium hybrids with

430 enhanced anticancer synergism: facile synthesis and mechanistic investigation of anticancer activity. Int.

431 J. Nanomed. 2016, 11, 1305-1321.

432 31. Che, H.; Fu, X.; Zhang, L.; Xiang, G.; Min, W.; Lei, D.; Xue, C.; Jie, X.; Wang, Y.

433 Neuroprotective Effects of n-3 Polyunsaturated Fatty Acid-Enriched Phosphatidylserine Against

434 Oxidative Damage in PC12 Cells. Cell. Mol. Neurobio. 2017, 1-12.

ACS Paragon Plus Environment

Page 19 of 30 Journal of Agricultural and Food Chemistry

435 Figure Captions

436 Fig.1 Structure and effects of AST and AST-DHA on spatial learning and memory deficiency.

437 Structure of AST and AST-DHA (A). The number of reference memory errors in radial 8-Arm

438 maze (B), number of working memory errors in radial 8-Arm maze (C). Time needed to reach the

439 hidden platform in the Morris maze (D). The time spent in the target quadrant. (E) and the number

440 of crossing platform (F) were measured for analysis of spatial memory function. Data were

441 presented as mean ± SD (n = 8), *P< 0.05 was considered statistically significant. Different letters

442 indicated significant difference between each group.

443 Fig.2 Represent photo and quantitative analysis of amyloid plaques in cortex and hippocampus of

444 APP/PS1 transgenic mice (A). Plaques number of Aβ in cortex and hippocampus (B). Relative of

445 quantitative analysis of Aβ load in cortex and hippocampus (C). Scale bar = 100 µm.

446 Fig.3 Effects of AST and AST-DHA on the regulation of Aβ concentration in APP/PS1 transgenic

447 mice. Levels of insoluble Aβ40 (A), soluble Aβ40 (B), insoluble Aβ42 (C) and soluble Aβ42 (D)

448 were measured by ELISA. Representative western blots (E) and densitometry of ADAM10 (F),

449 BACE1 (G) and Nicastrin (H). Data were presented as mean ± SD (n=8); *P< 0.05 was

450 considered statistically significant. Different letters indicated significant difference between each

451 group.

452 Fig.4 Effects of AST and AST-DHA on the SOD activity (A), NO concentration (B) and NOS

453 activity (C) in brain. Data were expressed as mean± SD (n = 8), *P< 0.05 was considered

454 statistically significant. Different letter indicated significant difference between each group.

455 Fig.5. Effects of AST and AST-DHA on tau and GSK-3β hyper-phosphorylation. (A) Western

456 blot analysis of p-GSK3β and p-Tau. Densitometry analysis of p-GSK3β (B) and p-Tau (C).

457 Values were indicated as the mean ± SD (n = 8), *P< 0.05 was considered statistically significant.

458 Different letter indicated significant difference between each group APP/PS1 transgenic mice.

459 Fig.6 Effects of AST and AST-DHA on neuroglial activation and neuro-inflammation. (A) Effects

460 of AST and AST-DHA on glial markers were analyzed by immunohistochemistry mmunostaining

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 20 of 30

461 of the microglial marker Iba1astroglial marker GFAP and the Scale bar = 10 µm. (B) Western blot

462 analysis of CD11b, GFAP, IL-1β and TNF-α. Densitometry analysis of CD11b (C), GFAP (D),

463 IL-1β (E) and TNF-α (F). Values were indicated as the mean ± SD (n = 8), *p< 0.05 was

464 considered statistically significant. Different letter indicated significant difference between each

465 group.

466 Fig.7 Effects of AST and AST-DHA on inflammasome expression and activation. (A) Western

467 blot analysis of NLPR3, ASC, Caspase 1 and Pro-IL-1β. Densitometry analysis of NLRP3 (B),

468 ASC (C), Caspase 1 (D), and Pro-IL-1β (D). Values were indicated as the mean ± SD (n = 8), *p<

469 0.05 was considered statistically significant. Different letter indicated significant difference

470 between each group.

471 Fig.8 Effects of AST and AST-DHA on mitochondria-dependent apoptosis in rat hippocampus. (A)

472 Western blot analysis of Bcl-2, Bax, Caspase 9, Caspase 3 and Cleaved Caspase 3. Densitometry

473 analysis of Bcl-2 (B), Bax (C), Caspase 9 (D), Caspase 3 (E) and Cleaved Caspase 3 (F). Values

474 were indicated as the mean ± SD (n = 8), *p< 0.05 was considered statistically significant.

475 Different letter indicated significant difference between each group.

476 Fig.9 The possible underlying mechanism of AST-DHA on learning and memory abilities in

477 APP/PS1 double transgenic mice.

ACS Paragon Plus Environment

Page 21 of 30 Journal of Agricultural and Food Chemistry

478

479

A B C

D E F

480 Fig.1

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 22 of 30

481

B
C

482 Fig.2

ACS Paragon Plus Environment

Page 23 of 30 Journal of Agricultural and Food Chemistry

483

A B C

D E F

G H

484 Fig.3

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 24 of 30

485

A B C

486 Fig.4

ACS Paragon Plus Environment

Page 25 of 30 Journal of Agricultural and Food Chemistry

487

A B C

488 Fig.5.

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 26 of 30

489

B C D

E F

490 Fig.6

491

ACS Paragon Plus Environment

Page 27 of 30 Journal of Agricultural and Food Chemistry

492

A B C

D E

493 Fig.7

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 28 of 30

494

A C
B

D E F

495 Fig.8

ACS Paragon Plus Environment

Page 29 of 30 Journal of Agricultural and Food Chemistry

496 Fig.9

ACS Paragon Plus Environment

 Journal of Agricultural and Food Chemistry Page 30 of 30

497 TOC Graphic

498

ACS Paragon Plus Environment