Royal jelly peptides: potential inhibitors of β-secretase in N2a/APP695swe cells

Royal jelly (RJ) is a type of natural health product with a long history of use. Royal jelly peptides (RJPs) obtained from RJ have numerous bioactivities. To study the neuroprotective effect of RJPs, major royal jelly proteins were digested into crude RJPs and subsequently purified by RP-HPLC. Purified RJP fractions were evaluated in N2a/APP695swe cells. Our results indicated that purified royal jelly peptides (RJPs) (1–9 μg/mL) could inhibit external beta-amyloid 40 (Aβ1-40) and beta-amyloid 42 (Aβ1-42) production through the down-regulation of β-secretase (BACE1) in N2a/APP695 cells. The modulation of BACE1 may be related to histone acetylation modification. Our results demonstrated a neuroprotective function of RJPs, which indicates that RJPs may serve as potential β-secretase inhibitors in ameliorating Aβ-related pathology in Alzheimer’s Disease.

Effects of fraction 1 C I RJP on the mRNA levels of Aβ metabolism related genes. In order to investigate how 1 C I RJP down-regulates extracellular Aβ 1-40 and Aβ 1-42 accumulation in N2a/APP cells, we performed real-time PCR (qPCR) to determine the change of mRNA levels for Aβ metabolism related genes. We treated the cells with 9 µg/ml 1 C 1 RJP for 24 or 12 hours after the cells were cultured for 24 hours. Several genes related to Aβ metabolism were examined, including APP, BACE1, Presenilin 1 (PS1), Presenilin 2 (PS2),    5A,B). However, with a 12-hour 1 C 1 RJP treatment, gene expression of IDE and LRP1 was significantly up-regulated, whereas expression of APOE was down-regulated; the expression of other genes was still not significantly changed (Fig. 5C,D). Furthermore, we investigated BACE1 expression at 3 hours, 6 hours, 12 hours and 24 hours' treatment with 9 µg/ml 1C 1 RJP, results indicated that BACE1 mRNA level was significantly down-regulated after cells were treated for 6 hours (Fig. 5E).

Fraction 1 C I RJP down-regulates BACE1 protein expression and BACE1 activity. The generation
of Aβ depends on the cleavage of amyloid precursor protein (APP) by β-secretase (BACE1) and γ-secretase. For this reason, we thought that RJP may affect BACE1 or γ-secretase expression or their catalytic activities. The γ-secretase complex contains four components: Presenilin 1 (PS1), Presenilin 2 (PEN2), APH-1 and Nicastrin, of which PS1 provides the catalytic function and PEN2 facilitates the autocatalytic cleavage of PS1 28 . We evaluated APP, BACE1, PS1 and PEN2 protein expression by western blot analysis after 24 hours of treatment with 1 C I RJP in both N2a and N2a/APP695a cells. N2a cells and N2a/APP695 cells were regarded as the control group and model group, respectively. Our data showed that APP expression was significantly higher in the model group than in the normal N2a cells (P < 0.01 vs model, Fig. 6A), whereas 1 C I RJP treatment did not change the APP expression in either group. These results indicate that our cell model successfully expressed high levels of APP. Moreover, BACE1 protein expression significantly elevated in N2a/APP695 cells (P < 0.01 vs control, Fig. 6B). When N2a/APP695 cells were treated with 1 C I RJP for 24 hours, the BACE1 level decreased in a dose-dependent manner and was significantly attenuated at 1 C I RJP concentrations of 3 μg·ml −1 and 9 μg·ml −1 (P < 0.01 vs model, Fig. 6B). However, 1 C I RJP did not affect BACE1 expression in N2a cells (P < 0.01 vs control, Fig. 6B). For PS1 and PSEN2 protein expression, 1 C I RJP did not significantly influence their expression in N2a/APP695 cells (data not shown). We further determined BACE1 activity using a beta-secretase fluorometric assay kit in N2a and N2a/APP695a cells treated or not with 1 C I RJP. Data revealed that BACE1 activity was greatly enhanced in the model cells than control cells, which may be responsible for greater Aβ generation (P < 0.01 vs control, Fig. 6C). Treatment with 1 C I RJP for 24 hours could weaken its activity in model cells but not N2a cells in a dose-dependent manner, and the activity was significantly down-regulated when treated with 9 μg·ml −1 of 1 C I RJP (P < 0.05 vs model, Fig. 6C).
Histone deacetylation is involved in the 1 C I RJP-induced down-regulation of BACE1. The roles of HDACs in cognitive function as well as in neurological disorders and diseases have been demonstrated for years 29 . Considering RJP could down-regulate BACE1 expression, we next determined whether 1 C I RJP is related to histone deacetylation. Thus, we detected HDAC1 and HDAC2 protein expression after treatment of N2a and N2a/APP695 cells with 1 C I RJP for 24 hours. Our data showed that 9 μg/ml 1 C I RJP could significantly enhance HDAC1 protein expression (P < 0.05 vs model, Fig. 7A), whereas 1 C I RJP did not affect HDAC2 expression (data not shown).
To investigate whether HDAC1 plays an important role in 1 C I RJP induced down-regulation of BACE1 expression, we designed HDAC inhibitor experiment. TSA and VPA are HDAC inhibitors, which can restrain the down-regulation of BACE1 through boosting histone H3 acetylation 30,31 . We pretreated N2a/APP695 cells with TSA and VPA before 1 C I RJP treatment. We found that the BACE1 protein level increased (P < 0.05 vs control, Fig. 7B) and decreased significantly upon treatment with 1 C I RJP (P < 0.001 vs model, Fig. 7B), which was the same as the above results. Meanwhile, the data indicated that BACE1 expression was higher in the TSA and RJP co-treatment group than in the RJP treatment only group (P < 0.01, Fig. 7B). In the same line, VPA displayed a similar function as TSA (P < 0.05, Fig. 7B). These results suggest that TSA and VPA could strongly prevent the down-regulation of BACE1 by 1 C I RJP treatment. However, BACE1 expression in the TSA or VAP and RJP co-treatment group was significantly lower than in cells only treated with TSA or VAP, which implies that there may be other mechanisms by which RJP regulates BACE1 (P < 0.05, Fig. 7B).

Discussion
Royal jelly contains several types of proteins, including major royal jelly protein 1 to 9 (MRJP1-9), and MRJP1-MRJP5 account for 82-90% of total royal jelly protein 7 . MRJP1 is the main component of WSRJ protein, with a molecular weight of approximately 55 kDa. Additionally, MRJP2, MRJP3, MRJP4 and MRJP5 are approximately 49 kDa, 60-70 kDa, 60 kDa and 80 kDa, respectively 32 . Royal jelly is the life-long food of queen bees and first-three-day food for bee larvae. Queen bees can live much longer and have better fertility than normal bees because of eating Royal jelly. It has been demonstrated that major royal jelly protein 1 was the component which plays the main role in bee larvae differentiation 33 . And larvae bees digest and absorbed it by intestinal. As for human being, royal jelly is a kind of health care products. , and LDLR after treatment with 9 μg·mL −1 1 C 1 RJP for 12 hours. (E) Relative mRNA levels of BACE1 after treatment with 9 µg/ml 1 C 1 RJP for 3, 6, 12, and 24 hours. Non-treated N2a cells were considered as control. Data are expressed as the mean ± SD of three independent experiments. * P < 0.05 and ** P < 0.01 vs N2a/APP treated with vehicle, # P < 0.05, ## P < 0.01 and ### P < 0.001 vs control. Statistical analyses were performed by a one-way ANOVA test and the addition of Bonferroni multiple comparisons test. People used royal jelly by oral administration. In order to imitate the process of bee larvae digestion, we used larvae intestinal enzymes complex to enzymolysis royal jelly protein and separate effective peptides. Royal jelly peptides (RJPs) obtained from royal jelly or major royal jelly protein (MRJP) have been studied in several publications, but few have focused on the neuroprotective effect of RJPs. In this study, we obtained crude RJPs (MW < 1 kDa, 1-3 kDa, and 3-5 kDa) from MRJP digest. Afterwards, 1-3-kDa RJPs were chosen to be purified using an RP-HPLC method, and several RJP fractions were obtained from the original RJPs. Based on preliminary bioactivity selection, we chose fraction 1 C I RJP as the active component to perform the subsequent investigation.
N2a/APP695 cells produced large amounts of Aβ1-40 and Aβ1-42 extracellularly, and no exogenous Aβ1-40 or Aβ1-42 was detected in N2a cell medium after they were cultured for 48 hours. These results were similar to those in a previous study 21 . Unquestionably, 1 C I RJP was purer than crude RJPs after several steps of RP-HPLC purification. Considering the output of 1 C I RJP and its preliminary bioactive concentration, we chose 1, 3, and 9 μg/ml as the suitable concentrations for the subsequent investigation. Our results showed that 1 C I RJP treatment for 48 hours could significantly inhibit the accumulation of extracellular Aβ1-40 and Aβ1-42 of N2a/ APP695 cells in a dose-dependent manner. These results suggest that 1 C I RJP affects Aβ metabolism in N2a/ APP695 cells. Numerous genetic and biochemical studies have indicated that Aβ overexpression and aggregation can be toxic and have been regarded as one of the key pathological factors of the onset of AD 17 . Thus, RJPs obtained from royal jelly may be beneficial in alleviating the progression of AD.
Multiple gene groups are involved in the complex Aβ metabolism, including its production (APP, PS1, and PS2), degradation (IDE) and transportation (APOE, LRP1, and LDLR). Several key genes were examined in our experiment using real-time PCR. There was no significant change for them in expression in the groups treated with RJP for 24 hours. Perhaps 24-hour time point was not the best for measuring the gene transcriptional levels with RJP treatment, and the transcription process of these genes had already ended. However, when we measured gene expression after 12-hour RJP treatment, the relative gene expression level of IDE and LRP1 were significantly up-regulated. The up-regulation of IDE could accelerate the degradation of Aβ, and the elevation of LRP1 may enhance the efflux of Aβ across the blood brain barrier (BBB) and reduce the Aβ burden in the brain. Aβ is generated from the serial cleavage of amyloid precursor protein (APP). APP is cleaved by β-secretase (BACE1) and generates the secreted derivative sAPPβ and β-C-terminal fragment (β-CTF). Afterwards, β-CTF is cleaved within the membrane by γ-secretase, leading to the release of Aβ 34 . BACE1 is the key rate-limiting enzyme of Aβ production. The expression level of BACE1 can reflect the production level of Aβ to some extent. The basal transcriptional level of BACE1 in N2a/APP cells after 6 and 12 hours of culturing revealed that BACE1 was highly expressed after 6 hours and inhibited after 12 hours through a cellular stress reaction (Fig. 5E). After treatment with 1C 1 RJP for 6 hours, mRNA level of BACE1 was down-regulated significantly (Fig. 5E). These results indicated that 1 C 1 RJP can inhibit BACE1 expression at a specific time point. Interestingly, both a reduction in BACE1 protein expression and enzyme activity induced by 1C 1 RJP were observed at a later time point as well, which in parallel with inhibition effect of RJP on BACE1 transcriptional level, hence contribute to the reduction of APP cleavage, leading to less Aβ release. The process from transcription to translation may need some time. So the changes of BACE of RNA and protein have shown at different timing. Anyway, all the above results imply that 1C I RJP can inhibit the extracellular Aβ accumulation of N2a/APP695 cells through the down-regulation of BACE1 protein expression and activity. As BACE1 is the key rate-limiting enzyme in the generation of Aβ, blocking BACE1 proteolytic activity will suppress Aβ generation, thereby targeting Aβ pathology for AD therapy. Currently, many potential BACE1 inhibitor drugs have been discovered, such as MK8931, AZD-3293, JNJ-54861911, E2609 and CNP520 35 . Few natural inhibitors have been reported and we take ginsenoside Re for an example, ginsenoside Re in the range of 25-100 μM/l (24 mg/l-95mg/l) can decrease BACE1 protein and mRNA expression in a dose dependent manner 36 . In comparison with it, our RJ peptides can decrease the BACE1 protein in lower concentration despite the peptides are mixed peptides. The single peptide has been analyzed and further experiments are in process now. Considering the bioactivity of 1 C I RJP, it could serve as a new natural BACE1 inhibitor.
Gene expression can be controlled by epigenetic modification. Recent studies have shown that histone acetylation activates BACE1 expression, and there is an increase in histone H3 acetylation in the BACE1 promoter 26,27 . Histone deacetylases (HDACs) can remove acetyl groups from histones, strengthening the binding of DNA with histones, thereby regulating gene expression by inhibiting the binding of DNA and transcription factors 24 . Our data showed that 9 μg/ml 1 C I RJP could significantly enhance HDAC1 protein expression, whereas 1 C I RJP did not affect HDAC2 expression. The HDAC1 inhibitor experiment results suggest that TSA and VPA could strongly prevent the down-regulation of BACE1 induced by 1 C I RJP treatment, but we cannot exclude the possibility that there may be other mechanisms by which RJP regulates BACE1.
Thus, our study found that 1 C I RJP could down-regulate the BACE1 protein level and proteolytic activity to eliminate exogenous Aβ accumulation in N2a/APP695 cells. Additionally, 1 C I RJP may affect BACE1 expression through a histone deacetylation modification mechanism. However, the exact mechanism by which RJP regulates BACE1 expression should be investigated further. Even though we verified that the relatively pure RJP functioned well in the present study, we haven't elucidated its sequence and structure information. Obviously, we should explore how to obtain monomer peptides that act on BACE1. In addition, animal model experiments should also be considered to test RJP activity in future study. In summary, all the above results suggest that RJPs are neuroprotective, and royal jelly could be a source of RJPs, which could be regarded as a potential natural product to relieve neurodegenerative diseases such as AD in the elderly.

Materials and Methods
Chemicals and materials. Ultrapure water was produced by a Milli-Q system (Millipore, USA

Preparation of the Apis mellifera larva intestinal enzyme.
The procedure was performed as following steps: Briefly, 2-3 days old Apis mellifera larvae were washed with cold 0.9% normal saline, and the intestinal canal was isolated from the larvae (Mentougou district, Beijing, China). The intestinal canal was mixed with 50 mM PBS (pH = 7.0) at a ratio of 1: 1, and was ground using an electric homogenizer in an ice bath. Then, the mixture was centrifuged at 20000 g for 20 min at 4 °C, and the middle layer solution was collected. Centrifuge the middle layer solution again in the same conditions, and a yellow transparent liquid was obtained, which was the intestinal enzyme solution. The mixture was stored at −20 °C. The WSRJ was about 42.7 mg/ml while intestinal enzyme mixture was about 31.9 mg/ml. The total volume was 10 ml, except for those two components, we added PBS to 10 ml. The optimized digestion condition for the intestinal enzyme solution and water-soluble royal jelly protein (1:3 mg/mg) mixture was 37 °C, pH 8.5 for 24 hours in a 2000-mL container. The reaction was stopped by placing the mixture in an ice bath, and the crude royal jelly peptides were obtained. This solution was centrifuged at 10000 g at 4 °C for 10 min to remove the insoluble components. Additionally, a 10 μm (Millipore, USA) microfiltration membrane was used to eliminate impure contents. Afterwards, crude RJPs were preliminarily purified through a 5-kDa, 3-kDa and 1-kDa ultra-filtration column and RJPs with MW < 1 kDa, 1-3 kDa, and 3-5 kDa were obtained (MSM-2008, Shanghai Mosu Science Equipment, China). Then, the digests were lyophilized and reconstituted in distilled water. In addition, the digests were filtered using a 0.22 μm (Millipore, USA) membrane, and the protein concentration was measured using a BCA assay.

Preparation
Crude peptides were purified using the RP-HPLC method. The obtained 1-3 kDa RJP was purified using the HPLC system (LC2000, CXTH, Beijing, China) with different sizes of chromatography columns. The mobile phases were as follows: (A) deionized water with 0.5% TFA and (B) acetonitrile with 0.5% TFA. The HPLC system consisted of two pumps and a UV detector. We used a 220-nm detection wave length in all the purification processes.
For the first separation, a 35 × 50 mm low-pressure glass column was employed with the C18 material, 20 μm, 300 Å (ODS-A, YMC, Japan). The elution condition was 100% A for 30 min, 25% B for 20 min and 55% B for 20 min at a flow rate of 20 mL/min.
Then, a 250 × 20 mm, S-10 μm, 12 nm C18 column (ODS-A, YMC, Japan) was used for the next stage of preparation, and the mobile phase, 10% B for 16 min and 60% B for 20 min at a flow rate of 10 mL/min, was applied for the purification process.
Cell viability assay. Cells (3 × 10 4 cells/ml) were plated in 96-well plates and allowed to grow for 24 hours.
Statistical analysis. All assays were repeated at least three times, and all values are represented as the mean ± SD. Statistical analysis was performed using a one-way analysis of variance (ANOVA), and multiple comparisons were made using the Bonferroni's test. A P value less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS 20.