Lycium barbarum, a classic Chinese medicine, has a large variety of biological activities, including improvements in immunity, as well as anti-aging and anti-oxidation activities. It has been used to improve or restore deteriorating functions related to aging and diseases. Although its nerve protection effects also have been proved in vitro and in vivo, the molecular mechanism of action is not clear. Here, we report on the effect and possible mechanisms of Lycium extract-ediated protection of Aβ-induced paralysis in Caenorhabditis elegans. Lycium extracts effectively reduced Aβ accumulation and delayed Aβ-induced paralysis in a transgenic C. elegans (CL2006) model that expresses human Aβ1–42. By evaluating the expression of genes related to the proteostasis network, we found that the expression of UPRmt, UPRER and autophagy-related genes was induced by Lucium extracts in CL2006 transgenic strains but not in the wild-type stains. Further RNAi experiments demonstrated that knock down of the UPRmt-related genes could reduce levels of down-regulation induced by Lucium extracts, suggesting that UPRmt is necessary for Lucium to prevent Aβ aggregation and maintain protein stabilization. Therefore, our studies provide more insights into the action and molecular mechanism of Lucium barbarum as a potential neuroprotective agent.

Lycium barbarum

• Alzheimer’s disease

• Aβ

• UPRm

 Meng J, Lv Z, Li X, Sun C, Jiang Z, Zhang W, Chen C (2017) Lycium extracts protect against β amyloid-induced pathological behaviors through UPRmt in transgenic Caenorhabditis elegans. J Aging Age Relat Dis 1(1): 1001 (2017)

Aβ: β Amyloid; AD: Alzheimer’s disease; LBP: Lucium Barbarum Polysaccharide; UPRmt: Mitochondrial Unfolded Protein Response; UPRER: Endoplasmic Reticulum Unfolded Protein Response; PQ: Paraquat; TG: Thapsigargin

Lycium barbarum berries are a traditional Chinese herbal medicine. Many functional components in L. barbarum fruits, including flavonoids, carotenoids, polysaccharides, glycolipids, glycopeptides, anthocyanins, essential oils, organic acids and trace minerals, are responsible for many health-related activities of this plant. In addition to China, the medicinal value of Lycium barbarum berries has been widely recognized in many Asian and Arabian countries. For many years, Chinese and foreign scientists have carried out extensive research on the biological function of Lycium barbarum berries, mainly analyzing the effect of Lycium barbarum extracts in normal physiological conditions or disease models. They found that extracts from Lucium species possess a range of biological activities, such as nourishing the liver and kidney, improving eyesight, delaying aging, improving immunity, decreasing blood-glucose and blood-lipids, and acting as an antitumor and anti-fatigue factor [1]. Its nerve protection effects also have been proved in vitro and in vivo. For example, Lycium barbarum polysaccharide (LBP) can inhibit 6-hydroxy dopamine (6-OHDA)-induced apoptosis in PC12 cells [2], and the extracts of Lycium barbarum have a protective effect in Aβ1-42- and Aβ25- 35-induced neuron injury [3,4]. In a whole animal model, LBP can protect middle cerebral artery occlusion (MCAO)-induced nerve injuries in mice [5,6] and improve the learning and memory abilities in scopolamine-induced brain damage in Sprague- Dawley (SD) rats [7].

Alzheimer’s disease (AD) is widely recognized as a common and devastating neurodegenerative disorder characterized by progressive impairment in memory, cognitive function and personality [8,9]. As the most common form of irreversible dementia, AD has become one of the major diseases to harm the health of the aged population, and AD exerts a great influence on families and society [9-11].The formation of the intracellular neurofibrillary tangle (NFT) and extracellular plaques are two major neuropathological features used for the diagnosis of AD [12]. AD is thought to be caused by the production and deposition of neurotoxic Aβ-peptide in the brain, leading to many consequences such as the formation of neurofibrillary tangles, oxidative stress, and neuronal cell death. Therefore, the focus of research in toxic Aβ production and clearance in the brain of AD patients is one approach for treatment of AD [13]. It has been reported the Luciumextracts can dramatically improve the Morris maze learning ability in the APP/PS1 double transgenic mouse model of Alzheimer’s disease [14]. Elevated homocysteine levels in the serum will increase the risk of AD, and it is found that Lycium barbarum polysaccharides can also inhibit apoptosis in homocysteine-induced neuronal injury [15]. Although increasing data confirm that Lucium barbarum can be used to treat AD in animal models, the molecular mechanism is not clear and requires further study.

In the present study, we used the Aβ-expressing nematode model strain CL2006 to investigate the molecular mechanism ofLuciumfunction. This transgenic nematode strain expresses the human 42 amino acid sequence of Aβ under the control of the muscle-specific unc-54 promoter/enhancer of C. elegans and responds to Aβ expression with increased paralysis [16]. It has been reported that neuromuscular synaptic transmission is specifically impaired by Aβ in this model [17]. Because of its short lifespan and its ease of culturing, the nematode is an advantageous animal model. Therefore, the strain CL2006 provides a good model for important insight into the mechanisms of Aβ toxicity.

Our studies demonstrated that the extracts of Lucium could protect the pathological behaviors in CL2006 transgenic worms by reducing the Aβ level. The Lucium extracts promoted Aβ degradation through UPRmt.

Caenorhabditis elegans strains and culture conditions

The C. elegans strains used in this study are Bristol N2, CL2006, hsp6pr::gfp, hsp4pr::gfp and sqst-1::gfp. The Bristol N2 strain was obtained from the Caenorhabditis Genetics Center (CGC) at the University of Minnesota, USA. The Cl2006 and sqst-1::gfp strains are gifts from Hong Zhang’s lab at the Chinese Academy of Sciences. The hsp4pr::gfp [18] and hsp6pr::gfp [19] strains are gifts from Ying Liu's lab at Peking University with the permission of Professor Cole M.Haynes. All strains were maintained at 20°C on nematode growth medium (NGM) seeded with the Escherichia coli OP50 feeding strain.

Lycium extract preparation and treatment

Lycium barbarum berries were kindly provided by the Hospital of Traditional Chinese Medicine in Zhongning County, Yinchuan, Ningxia, China. The dried berries (100 g) were soaked in water (1 L) at room temperature after being washed five times. The soaked berries were decocted with neutral water (2 L) at a boiling temperature twice, and decocting times were 2.0 h and 1.5 h. The combined concentrated decoctions were filtered by a hollow fiber membrane. The above filtrates were merged and evaporated under a vacuum (1 KPa) at 45°C to remove water and obtain the concentrate. The constant volume of the resulting concentrate was 100 mL, and it was used for the following experiments and stored at -20 °C. The Lucium extracts include mainly water-soluble Lycium barbarum polysaccharides, flavonoids, carotenoids, anthocyanins, referenced the published papers which used the similar protocol to extract Lucium barbarum [20,21].

Lycium extracts were mixed into the OP50 bacteria according to an indicated dilution ratio. The transgenic worms were given the treatment from the L4 stage and the treatment was lasted for 5 days. In particular, to assay the function of Lucium barbarum extracts at different age stages, the worms were givenLuciumat different stages.

Worm synchronization

Worm synchronization was implemented by alkaline hypochlorite treatment of gravid adults. Worms were first washed with M9 buffer (3 g of KH2PO4, 6 g of Na2HPO4, 5 g of NaCl, 1 mL of 1 mol/L MgSO4, in H2O to 1 L) and pelleted by centrifugation (2000 g). Then, the worms were incubated in hypochlorite solution (1 mL of 2 N NaOH, 800 μL of sodium hypochlorite solution, 2.2 mL of dH2O) for 3-5 min to homogenize the large worm particles. Eggs were pelleted by centrifugation and washed at least three times with M9 buffer; then, they were incubated in M9 buffer and allowed to hatch overnight at 20oC. The synchronized L1-stage worms were put on standard NGM plates coated with OP50 at 20oC.

Paralysis assay

Transgenic C. elegans strain CL2006 was egg-synchronized and transferred onto the culture plates containing OP50 with or without Lucium extracts at the L4 stage. To identify paralysis, each worm was gently touched with a platinum loop. The worm was considered paralyzed if it did not move or moved only its head after being touched. The worms were tested for paralysis every day.

Body bends

The control or Lucium extract-treated adult worms were placed on unseeded NGM plates and allowed to acclimatize for 5–10 min. The number of body bends was counted for 20 s. A complete body bend was defined as the bending of the head region across the central-line of the animal [22].

Pumping assay

Pumping assays were operated on NGM plates with bacterial lawns and Lycium-bacteria mixed lawns at a 20°C temperature. To assay for pumping rate, we measured the time required to complete 20 pumps. Four to six measurements were recorded for every animal, and 16 animals were tested per experiment. Three independent experiments were performed.

Quantification of Aβ

The Aβ aggregation was determined using a thioflavin T (ThT) fluorescence assay. The nematodes were harvested with M9 buffer and washed three times, then treated with lysis buffer (HEPES 50 mM, NaCl 150 mM, EDTA 5 mM, DTT 2 mM) and repeatedly frozen and thawed to extract proteins. Subsequently, the concentration of extracted proteins was determined by performing a Bradford Protein Assay (CW biotech). Proteins were incubated at room temperature with ThT (final concentration: 20 μM; Sigma) in PBS buffer. The fluorescence intensity was measured using an excitation wavelength of 440 nm and an emission wavelength of 482 nm in an automatic microplate reader (Thermo).

Feeding RNA interference

For the feeding RNA interference experiment, the RNAi bacterial clones used were from previously published libraries [23]. Each RNAi colony was grown in LB medium with carbenicillin (50 μg/mL) overnight and then, 1 mmol/L isopropylthiogalactoside (IPTG) was added to induce dsRNA expression for 1 h. A volume of 200 μL of the bacterial was applied to a 60-mm plate, to which approximately 500 synchronized L1 larvae were added. Exceptionally, the RNAi interference of lmp-2 was given to the worms at L4 stage, because interference at L1 stage will inhibit the growth of the worms. TheLuciumextract treatment was given to the worms from the L4 stage.

Real-time qPCR

Total RNA was extracted using TRIzol reagent according to the manufacturer’s (Invitrogen) protocol.RNA samples were then reverse transcribed using M-MuLV reverse transcriptase (Promega), and the mRNA levels were measured by RT-qPCR using a 7500 real time PCR system (Applied Biosystems), as described previously [24].

Fluorescence microscopy

The strains hsp6pr::gfp, hsp4pr::gfp and sqst-1::gfp were used to analyze the effect of Lucium extracts on UPRmt, UPRER and autophagy by detecting the intracellular expression of hsp6, hsp4 and sqst-1 in living nematodes. Paraquat (PQ, 100μM) or thapsigargin (TG, 1μM) treatment was used as a positive control for UPRmt and UPRER, respectively. After treatment with Lycium> extracts for 24 h, approximately 20 worms were placed onto 3% agarose on a glass slide. Fluorescence images were taken using a confocal laserscanning microscope (LSM750) (Carl Zeiss).

The proteasome activity

For protein extraction, the worms were treated with lysis buffer (HEPES 50 mM, NaCl 150 mM, EDTA 5 mM, DTT2 mM); the concentration of proteins was determined by the Bradford Protein Assay (CW biotech). Quantification of proteasome activity was accomplished using a fluorogenic peptide substrate assay. The proteins were incubated with Suc-LLVY-AMC (final concentration100 μM; Sigma) in proteasome activity assay buffer (50 mM HEPES pH 7.4, 150 mM NaCl, 5 mM EDTA, and 5 mM ATP) at room temperature. The fluorescence intensity was measured in triplicate over 1h every 10 min with excitation at 355 nm and emission at 460 nm using an automatic microplate reader (Thermo).

Statistical analysis

The results are presented as the mean ± standard error of the mean (SEM) of at least 3 independent experiments. The statistical significance of the difference between two means was calculated using Student’s t-test. For all analyses, p<0.05 was considered statistically significant.

Lycium extracts suppress the pathological behaviors in CL2006 by down-regulating Aβ

To investigate whether Lucium extracts specifically protect against Aβ-induced toxicity in vivo, the transgenic C. elegans strain CL2006 was used as an AD model. In this transgenic C. elegans model, human Aβ42 protein was expressed and aggregated in the body wall muscle, leading to progressive paralysis [25].The worms were treated with Lucium extracts at dilution ratios of 1:100, 1:50 and 1:20 for 5 days.Lucium extracts at dilution ratios of 1:50 and 1:20 significantly delayed Aβ-induced paralysis in this transgenic worm (Figure 1A).

Figure 1. Lycium extracts protect against the Aβ-induced pathological behaviors in C. elegans strain CL2006. (A) Paralysis was analyzed in worms treated with Lycium at dilution ratios of 1:100, 1:50 and 1:20. (B) The rates of paralysis on Day 13are displayed separately. (C) The number of body bends over 20 s was counted. (D) Pumping rates of the worms were assessed. (E) The Aβ aggregation was determined using a thioflavin T fluorescence assay (EGCG was used as a positive control). (F) The different effects of Lycium on Aβ aggregation were measured at different stages of worms (L1, Day 1, Day 3 and Day 6). Error bars represent means ± SEMs (*P<0.05, **P<0.01, ***P<0.001).

In particular, the rates of paralysis at Day 13 decreased by 20% at the 1:100 dilution and decreased by 30% at the 1:50 and 1:20 dilutions (Figure 1B). To further confirm the protective effects of Lucium extracts on Aβ- induced toxicity, the number of body bends over 20 s was counted in control and Lucium extract (1:50 dilution)-treated groups. Lycium extracts also significantly improved the frequency of body bends (Figure 1C). To exclude the effect of food preference of C. elegans, pumping rates of the worms fed with or without Lycium extracts were assessed, and there was no significant difference (Figure 1D). Therefore, Aβ-induced pathological behaviors in the transgenic C. elegans was alleviated by Lucium extracts.

Because the paralysis behavior of the transgenic stain CL2006 is the result of over-expression and aggregation of Aβ, to investigate how Luciumfunctions, we first checked the Aβ level in N2 and CL2006 nematodes with or without Lucium treatment, using EGCG as a positive control. EGCG, Epigallocatechin gallate, is a major component of green tea polyphenols. It was reported that EGCG could reduce beta amyloid (Abeta) deposits and inhibited Abeta oligomerization in transgenic C. elegans (CL2006) [26]. CL2006 showed a 6-fold increase in aggregated proteins versus N2 wild-type nematodes, and the amount of aggregated protein was dose-dependently reduced by Lycium, to a similar to that of the positive control EGCG (Figure 1E). Moreover, the level of Aβ was reduced by half at the 1:50 and 1:20 dilutions (Figure 1E), which was associated with a concomitant reduction of paralysis in the nematodes. Due to Aβ aggregation increasing with age, we also detected the different effects of Lucium at different age stages. At the 1:50 dilution, Aβ was reduced by half if Lucium was given at the L1 stage, and the reduction decreased gradually as the treatment time was delayed (Figure 1F). After Day6, Lycium treatment no longer had a significant effect (Figure 1F). In conclusion,Lucium extracts inhibited paralysis by reducing Aβ levels in CL2006. Given that enhanced aggregation of Aβ might be the result of impaired quality control of protein homoeostasis [27], we focused on determining the mechanism by which Lycium extracts might induce some pathways for degrading damage proteins.

Lycium extracts induce UPRmt- and UPRER-related gene expression in CL2006

The maintenance of protein homeostasis is essential to preserve cell function. The major players in the maintenance of proteostasis include the mitochondrial unfolded protein response (UPRmt), the endoplasmic reticulum unfolded protein response (UPRER) and two proteolytic systems, the ubiquitin-proteasome and the autophagy systems.

To investigate whether Lucium extracts can stimulate the UPR pathways, hsp6pr::gfp and hsp4pr::gfp transgenic strains were used to evaluate the UPRmt and UPRER, respectively, because the chaperone homologs HSP6 and HSP4 are considered markers of UPRmt and UPRER, respectively. Compared to the positive control paraquat (PQ) that could induce UPRmt significantly, no GFP expression could be induced in theLuciumextract-reated hsp6pr::gfp nematodes (Figure 2A)

Figure 2. Lycium extracts induce UPRmt- and UPRER-related gene expression in CL2006 strain. (A, B) hsp6pr::gfp and hsp4pr::gfp transgenic strains were used to evaluate the effect of Lycium on UPRmt and UPRER, with PQ or TG as a positive control, respectively. (C, D) The effect of different dosages of Lycium on the expression ofdve-1 and ubl-5 in CL2006 was detected by qRT-PCR. (E) Expression of hsp60, clpp and haf-1 in CL2006 worms treated with Lycium extracts (1:50) was also detected by qRT-PCR. (F, G) The effect of different dosages of Lycium on the expression of xbp1and abu-1 in CL2006 was detected by qRT-PCR. Error bars represent means ± SEMs (*P<0.05, **P<0.01, ***P<0.001).

while in the hsp4pr::gfp nematodes, Lucium extracts could also not induce the UPRER, in which thapsigargin (TG), a specific inhibitor of the sarcoplasmic/ endoplasmic reticulum Ca2+-ATPase, was used as a positive control (Figure 2B). These data suggested thatLuciumextracts did not cause stress leading to the accumulation of unfolded proteins in the normal nematodes without pathological behaviors. To further study whetherLuciumextracts have functions in the nematode disease model, the expression of UPR-related genes was detected in CL2006 in the presence and absence ofLuciumextracts. dve-1, encoding a transcription factor that binds to the hsp-6 and hsp-60 promoters upon mitochondrial stress, was significantly increased byLuciumextracts at the 1:20 dilution (Figure 2C). UBL-5, a ubiquitin-like protein homolog, is essential for UPRmt, with increased nuclear levels following induction of the UPRmt to help promote the interaction between DVE-1 and the DNA. The mRNA level of ubl-5was up-regulated byLuciumextracts in a dose-dependent manner (Figure 2D). To strengthen the argument of the specific role of UPRmt induction related to the antagonistic effect of Lucium extracts in C. elegans, the expression of some other markers of UPRmt, hsp60, clpp and haf-1 was also detected after Lucium extracts (1:50) treatment.

The data suggested thatLucium barbarum extracts could induce hsp60, clpp and haf-1 expression in CL2006 (Figure 2E), providing further evidence that UPRmt is involved. xbp-1, encoding a transcription factor involved in ER stress, was also increased afterLuciumextracts treatment (Figure 2F). abu-1 is a gene activated when UPRER is blocked to compensate for ER stress, so it was down-regulated by Lucium extracts (Figure 2G). In conclusion, Lucium extracts increase UPRmt- and UPRER-related gene expression in the transgenic stain CL2006. It is possible that Lycium extracts reduce the Aβ level through these pathways.

Lycium extracts up-regulate autophagy-related genes but have no effect on proteasome activity in CL2006 strain

Autophagy and the ubiquitin-proteasome are two major proteolytic systems responsible for cytosolic protein degradation. To monitor whether autophagy is involved in Lucium extractinduced Aβ down-regulation, the sqst-1::gfp transgenic strain was employed. Sqst-1 (the homolog of human p62) is a substrate that is degraded during autophagy. GFP was highly expressed in the control nematodes and dramatically degraded by PQ stress-induced autophagy (Figure 3A, 3B)

Figure 3. Lycium extracts up-regulate autophagy-related genes but have no effect on proteasome activity in CL2006 strain. (A) The sqst-1::gfp transgenic strain was used to evaluate the effect of Lycium on autophagy, with PQ as a positive control. (B) The number of the GFP+ puncta was counted to estimate the sqst-1 expression level. (C) The effect of different dosages of Lycium on the expression ofbec-1 in CL2006 was detected by qRT-PCR. (D) The expression of pink-1 and pdr-1 in CL2006 treated with or without Lycium was detected by qRT-PCR. (E) The effect of different dosages of Lycium on the expression ofuba-1 in CL2006 was detected by qRT-PCR. (F) The proteasome activity in CL2006 treated with different dosages of Lycium was measured using a fluorogenic peptidesubstrate assay. Error bars represent means ± SEMs (*P<0.05, **P<0.01, ***P<0.001).

 The high level of GFP expressed in the Luciumextract-treated nematodes suggested that autophagy was not induced by Lucium extracts in this strain without pathological behavior (Figure 3B). Similarly, the expression of bec-1, which is indispensable for the formation of autophagosomes, was detected in CL2006 with or without Lycium extract treatment. The dose-dependent up-regulation of bec-1 expression suggested that Lucium extracts might also activate the autophagy pathway in CL2006 (Figure 3C). In order to investigate whether mitophagy also occurs, two mitophagy markers, pink-1 and pdr-1, were measured (Figure 3D). We found that Lycium extracts could not up-regulate the expression of these two genes. On the other side, Lucium extracts did not change the level of uba- 1, which is the ubiquitin-activating enzyme in C. elegans (Figure 3E). Further measuring the proteasome activity in CL2006, we found that there was no significant difference after Lycium extract treatment (Figure 3F). Therefore, autophagy rather than the proteasome may be involved in Lucium extract-mediated Aβ down-regulation.

Lycium extracts reduce Aβ aggregation through UPRmt

Given the induction of UPRmt, UPRER and autophagy in CL2006 nematodes treated with Lucium extracts, to determine their function in the prevention of Aβ aggregation, RNA-interference (RNAi) was used, and the knockdown efficiency for each gene was detected (Figure 4A)

Figure 4. Lycium extracts reduce Aβ aggregation through UPRmt. (A) The knockdown efficiency for each gene was checked by qRT-PCR. (B) Aβ aggregation levels were assessed in the conditions where UPRmt-, UPRER- and autophagy-related gene were knocked down by RNAi. Error bars represent means ± SEMs (*P<0.05, **P<0.01, ***P<0.001).In the vector control

(L4440), Lycium extracts could reduce the Aβ level by 60% compared to the control (Figure 4B). When dve-1 and atfs-1, the key transcriptional factors in UPRmt, wereknocked down by RNAi, the Aβ level could not be reduced byLuciumextracts (Figure 4B). When hsp6 and hsp60 were knocked down by RNAi, although Lucium extracts could slightly reduce the Aβ level, there was no significant difference. These results indicate that UPRmt is necessary to prevent Aβ aggregation and maintain protein stabilization. To estimate the influence of UPRER on Aβ-induced toxicity, xbp-1 was knocked down, and in this condition, Aβ levels could still be decreased by 50% by Lucium extracts. Thus, RNAi for xbp-1 did not prevent the reduction of Aβ aggregation by Luciumextracts. Similar to the knockdown of xbp-1 for UPRER, RNAi for bec-1 also did not prevent the decrease in the Aβ level by the Lucium extracts. To further confirm this result, lmp-2, encoding a lysosomal receptor homolog, was also knocked down. Aconsistent result was obtained, suggesting that the autophagy pathway is not a target of  the Lucium extracts (Figure 4B). In conclusion, only UPRmt is involved in Lucium extract-induced Aβ down-regulation.

In this study, we used the transgenic strain CL2006 to monitor the function of Lucium barbarum in AD. We found that the amount of aggregated Aβ and paralysis were dose-dependently reduced by the Lucium extracts. To investigate whether Lucium extracts impact the proteostasis network, UPRmt, UPRER, autophagy and the proteasome system were evaluated in the wild-type and CL2006 transgenic strains treated with or without Lycium extracts. The expression of UPRmt, UPRER and autophagy-related genes was induced by Lucium extracts in the CL2006 transgenic strains but not in the wild-type stains. Furthermore, to identify which member of the proteostasis network is required for the Lycium extract-induced Aβ down-regulation, RNAi was exploited. Knock down of UPRmt-related genes could prevent Lucium extract-induced Aβ down-regulation, suggesting that UPRmt is necessary to prevent Aβ aggregation and maintain protein stabilization.

It is very interesting that Lucium extracts could induce UPRmt, UPRER and autophagy in the CL2006 transgenic strains that have aggregated Aβ but that do not exert effects in the wild-type strains. CL2006 exhibited a 6-fold increase in aggregated proteins compared with N2 wild-type nematodes; therefore, Lycium extracts may function by enhancing the aggregated proteinmediated activation of UPRmt, UPRER or autophagy but without inducing these processes directly. Thus, Lucium extracts have no influence on proteostasis under normal physiological conditions. Although the genes related to UPRmt, UPRER and autophagy were increased by Lucium extracts in CL2006, only UPRmt is required for Lucium extract-induced Aβ down-regulation according to the RNAi experiments. Therefore, UPRER and autophagy are only concomitant results, not causes. UPRmt, which functions through the sensing of mitochondrial stress to coordinate the appropriate response, plays a significant role in Lycium-mediated proteostasis maintenance. It is known that UPRmt decreases with age [28], which is consistent with our result that Lucium extracts have more significant effects at the early stage in CL2006. After Day6, Lycium treatment no longer had a significant effect, which may be due to decreased UPRmt.

According to the published paper, Lucium extracts have high anti-oxidative activity, and we also found that Lucium extracts could reduce the ROS level in both wild-type and transgenic strains (Data is not shown). So was it possible that the restorative effect of the extracts was due to anti-oxidative property? If Aβ first disrupted the redox balance in the mitochondria that in turn affected UPRmt induction and Lucium extracts function through their anti-oxidative property, the result would be decreased ROS level and reduced UPRmt after Lucium extract treatment. However, in our results, Lucium extracts significantly increased UPRmt, and induced UPRmt is necessary for the restorative effect of the extracts. Therefore, the anti-oxidative ability of the extracts was not necessary for decreased Aβ aggregation. This new mechanism is very important for explaining the function of Lucium barbarum as a potential neuroprotective agent.

Our studies provide evidence that Lucium extracts reduce Aβ-induced toxicity and protect from pathological behavior in C. elegans through regulating UPRmt, and this effect is more significant at an early stage. Although the protective effect of Lycium on Aβ-induced cytotoxicity has been reported in vitro [3,4], the mechanism has not been explained. Since intracellular Aβ is cytotoxic and an early causative event in the development of AD, inhibition of Aβ aggregation is one approach for treatment of AD. Moreover, many studies indicate that diverse neurodegenerative diseases might have a common cause and pathological mechanism - the misfolding and aggregation of proteins in the brain, resulting in neuronal apoptosis. Impaired proteostasis is one of the main features of all amyloid diseases. Lucium extract-induced Aβ down-regulation is mediated by UPRmt, which plays an important role in the proteostasis network. Thus, our studies provide more insights into the action of Lucium barbarum as a potential neuroprotective agent

We thank Prof. Hong Zhang (Institute of Biophysics, Chinese Academy of Sciences) for providing sqst-1::gfp transgenic strains, and Prof. Ying Liu (Peking University) for providing the hsp4pr::gfp and hsp6pr::gfp transgenic strains.

This study was supported by National Key R&D Program of China (2017YFA0504000, 2016YFC0903100), the National Natural Science Foundation of China (Grant No. 31500693, No. 31570857), Ningxia Key Research and Development Program Grant, Goji Berry Technological Cooperation Project in Agricultural Comprehensive Development of Ningxia Autonomous Region (No. NTKJ-2015-05) and Regional key projects of science and technology service network plan (STS plan) of Chinese Academy of Sciences.

We are also grateful to three anonymous reviewers for their comments and peer-review.

Jiao Meng, Zhenyu Lv, Xiaopeng Li, Chuanxin Sun, Zhengguo Jiang, Wanchang Zhang and Chang Chen declare that they have no conflict of interest.

1. Cheng J, Zhou ZW, Sheng HP, He LJ, Fan XW, He ZX, et al. An evidencebased update on the pharmacological activities and possible molecular targets of Lucium barbarum polysaccharides. Drug Des Devel Ther. 2015; 9: 33-78.
2. Gao K, Liu M, Cao J, Yao M, Lu Y, Li J, et al. Protective effects of Lycium barbarum polysaccharide on 6-OHDA-induced apoptosis in PC12 cells through the ROS-NO pathway. Molecules. 2014; 20(1): 293-308.
3. Yu MS, Leung SK, Lai SW, Che CM, Zee SY, So KF, et al. Neuroprotective effects of anti-aging oriental medicine Lycium barbarum against beta-amyloid peptide neurotoxicity. Exp Gerontol. 2005; 40(8-9): 716-27.
4. Yu MS, Lai CS, Ho YS, Zee SY, So KF, Yuen WH, et al. Characterization of the effects of anti-aging medicine Fructus lycii on beta-amyloid peptide neurotoxicity. Int J Mol Med. 2007; 20(2): 261-8.
5. Wang HB, Li YX, Hao YJ, Wang TF, Lei Z, Wu Y, et al. Neuroprotective effects of LBP on brain ischemic reperfusion neurodegeneration. Eur Rev Med Pharmacol Sci. 2013; 17(20): 2760-5.
6. Wang T, Li Y, Wang Y, Zhou R, Ma L, Hao Y, et al. Lycium barbarum polysaccharide prevents focal cerebral ischemic injury by inhibiting neuronal apoptosis in mice. PLoS One. 2014; 9(3): e90780
7. Chen W, Cheng X, Chen J, Yi X, Nie D, Sun X, et al. Lycium barbarum polysaccharides prevent memory and neurogenesis impairments in scopolamine-treated rats. PLoS One. 2014; 9(2): e88076.
8. Price DL, Sisodia SS, Borchelt DR. Alzheimer disease--when and why? Nat Genet. 1998; 19(4): 314-6.
9. Selkoe DJ. Alzheimer's disease: genes, proteins, and therapy. Physiol Rev. 2001; 81(2): 741-66.
10. Sloane PD, Zimmerman S, Suchindran C, Reed P, Wang L, Boustani M, et al. The public health impact of Alzheimer's disease, 2000-2050: potential implication of treatment advances. Annu Rev Public Health. 2002; 23:213-31.
11. Hickman RA, Faustin A, Wisniewski T. Alzheimer Disease and Its Growing Epidemic: Risk Factors, Biomarkers, and the Urgent Need for Therapeutics. Neurol Clin. 2016; 34(4): 941-953.
12. Gray EG, Paula-Barbosa M, Roher A. Alzheimer's disease: paired helical filaments and cytomembranes. Neuropathol Appl Neurobiol. 1987; 13(2): 91-110.
13. Singh SK, Srivastav S, Yadav AK, Srikrishna S, Perry G. Overview of Alzheimer's Disease and Some Therapeutic Approaches Targeting Abeta by Using Several Synthetic and Herbal Compounds. Oxid Med Cell Longev. 2016; 2016: 7361613.
14. Zhang Q, Du X, Xu Y, Dang L, Xiang L, Zhang J. The effects of Gouqi extracts on Morris maze learning in the APP/PS1 double transgenic mouse model of Alzheimer's disease. Exp Ther Med. 2013; 5(5): 1528-1530.
15. Ho YS, Yu MS, Yang XF, So KF, Yuen WH, Chang RC. Neuroprotective effects of polysaccharides from wolfberry, the fruits of Lycium barbarum, against homocysteine-induced toxicity in rat cortical neurons. J Alzheimers Dis. 2010; 19(3): 813-27.
16. Ewald CY, Li C. Understanding the molecular basis of Alzheimer's disease using a Caenorhabditis elegans model system. Brain Struct Funct. 2010; 214(2-3): 263-83.
17. Rebolledo DL, Aldunate R, Kohn R, Neira I, Minniti AN, Inestrosa NC. Copper reduces Abeta oligomeric species and ameliorates neuromuscular synaptic defects in a C. elegans model of inclusion body myositis. J Neurosci. 2011; 31(28): 10149-58.
18. Calfon M, Zeng H, Urano F, Till JH, Hubbard SR, Harding HP, et al. IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA. Nature. 2002; 415(6867): 92-6.
19. Yoneda T, Benedetti C, Urano F, Clark SG, Harding HP, Ron D. Compartment- specific perturbation of protein handling activates genes encoding mitochondrial chaperones. J Cell Sci. 2004; 117(Pt 18): 4055-66.
20. Ahn M, Park JS, Chae S, Kim S, Moon C, Hyun JW, et al. Hepatoprotective effects of Lycium chinense Miller fruit and its constituent betaine in CCl4-induced hepatic damage in rats. Acta Histochem. 2014; 116(6): 1104-12.
21. Ha KT, Yoon SJ, Choi DY, Kim DW, Kim JK,Kim CH. Protective effect of Lycium chinense fruit on carbon tetrachloride-induced hepatotoxicity. J Ethnopharmacol. 2005; 96(3): 529-35.
22. Bull K, Cook A, Hopper NA, Harder A, Holden-Dye L,Walker RJ. Effects of the novel anthelmintic emodepside on the locomotion, egg-laying behaviour and development of Caenorhabditis elegans. Int J Parasitol. 2007; 37(6): 627-36.
23. Kamath RS, Fraser AG, Dong Y, Poulin G, Durbin R, Gotta M, et al. Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature. 2003; 421(6920): 231-7.
24. Meng J, Lv Z, Qiao X, Li X, Li Y, Zhang Y, et al. The decay of Redox-stress Response Capacity is a substantive characteristic of aging: Revising the redox theory of aging. Redox Biol. 2017; 11: 365-374.
25. Link CD. Expression of human beta-amyloid peptide in transgenic Caenorhabditis elegans. Proc Natl Acad Sci U S A. 1995; 92(20): 9368- 72.
26. Abbas S, Wink M. Epigallocatechin gallate inhibits beta amyloid oligomerization in Caenorhabditis elegans and affects the daf-2/insulinlike signaling pathway. Phytomedicine. 2010; 17(11): 902-9.
27. Cohen E. Aging, protein aggregation, chaperones, and neurodegenerative disorders: mechanisms of coupling and therapeutic opportunities. Rambam Maimonides Med J. 2012; 3(4): e0021.
28. Jensen MB, Jasper H. Mitochondrial proteostasis in the control of aging and longevity. Cell Metab. 2014; 20(2): 214-25.