Royal jelly and related substances may extend the lifespan of the worm Caenorhabditis elegans

We first investigated the effects of RJ on the lifespan of the wild-type N2 strain of C. elegans. RJ treatment was begun at the young adult stage with concentrations ranging from 1 to 100 µg/ml. RJ treatment at 10 µg/ml extended the mean lifespan by 7–9%, whereas either 1 or 100 µg/ml RJ had little or no effects on the lifespan (Fig. 1, Table S1), indicating that there is an optimal dose of RJ for lifespan extension. In contrast, pRJ prolonged the mean lifespan at all concentrations tested (1, 10 and 100 µg/ml). The maximal effect was observed at 10 µg/ml, at which concentration the mean lifespan was increased by 7–18% (Fig. 2, Table S1). These results suggest that both RJ and pRJ contain the lifespan-extending agents and that these agents are not proteinaceus.

To gain insights into the nature of the lifespan-extending agents, we performed fractionations of both RJ and pRJ. RJ was divided into EtOH-soluble (RJ-Fr.1) and water-soluble (RJ-Fr.2) fractions. Neither of these fractions extended the lifespan (Fig. S1, Fig. S2, Table S1). pRJ was fractionated by Octadecyl-silica (ODS) column chromatography and eluted with water (pRJ-Fr.4) and subsequently with 30% MeOH (pRJ-Fr.5). pRJ-Fr.4 at concentrations from 5 to 100 µg/ml increased the mean lifespan (Fig. 3, Table S1). The maximal effect was observed at 10 µg/ml, at which concentration the mean lifespan was increased by 9%. In contrast, pRJ-Fr.5 at concentrations of 10, 25 and 100 µg/ml increased the mean lifespan by 8–9%, 18–19% and 17–19%, respectively (Fig. 4, Table S1). These results indicated that the lifespan-extending agents in pRJ were enriched in the 30% MeOH-eluted fraction more than in the water-eluted fraction.

To understand the mechanism underlying the lifespan extension by pRJ-Fr.5, we analyzed genome-wide changes in gene expression during treatment of C. elegans N2 with pRJ-Fr.5. Using the Agilent C. elegans (V2) Gene Expression Microarray, we monitored the expression of 20,000 genes. To identify differentially regulated genes, we eliminated all probes with absent or marginal flags and then performed a t-test with the significance level set at p<0.05. Of these genes, 733 were further selected using the criterion of at least a 1.8-fold change (Table S2). Further analysis of these 733 genes revealed that pRJ-Fr.5 down-regulated ins-9 and up-regulated ins-20 and ins-23, all of which encode insulin-like peptides (Table S2). Among these insulin-like peptide genes, real-time RT-PCR confirmed down-regulation of ins-9 gene expression after pRJ-Fr.5 treatment (Fig. 5). These results are consistent with previous findings implicating reduced insulin/IGF-1 signaling (IIS) in lifespan extension [1]. pRJ-Fr.5 also down-regulated dod-19, dao-4, and fkb-4 and up-regulated dod-3 (Table S2). These expression changes were all verified by real-time RT-PCR analysis (Fig. 5) and, more importantly, correlated with the changes observed when IIS is reduced in C. elegans[52], [53]. Certain DNA motifs were previously reported to be associated with the FOXO transcription factor DAF-16 [53], [54], the activation of which is known to promote longevity in C. elegans upon reduction of IIS [1]. The DAF-16-binding element (DBE: TTGTTTAC) [54] and the DAF-16-associated element (DAE: CTTATCA) [53] were overrepresented in the upstream regions of ins-9, dod-3, dod-19, dao-4 and fkb-4 (Table 1), suggesting that their gene expression is controlled by DAF-16 activity.

To clarify whether the IIS-DAF-16 pathway is involved in pRJ-Fr.5-induced extension of lifespan, we examined the effects of pRJ-Fr.5 on the lifespan of a daf-16 deletion mutant. The findings that pRJ-Fr.5 extended the mean lifespan of this mutant by 8–12% (Fig. 6, Table S1) and that this effect was smaller than that observed in wild-type N2 (18–19%) (Fig. 4, Table S1) indicated that pRJ-Fr.5 extends the lifespan by both IIS-DAF-16 pathway-dependent and -independent mechanisms.

To ascertain whether pRJ-Fr.5 acts on the IIS-DAF-16 pathway, we examined the effects of pRJ-Fr.5 treatment on DAF-16 nuclear localization, which has been shown to be augmented when IIS is abrogated [55], [56]. We found that pRJ-Fr.5 treatment induced DAF-16 nuclear localization (Fig. S3), suggesting that pRJ-Fr.5 acts on the IIS-DAF-16 pathway.

Next, we analyzed the components of pRJ-Fr.5. The amount of sugars contained in pRJ-Fr.5 was estimated to be 20%(w/w) in terms of glucose. Peptides accounted for more than 60%(w/w) of pRJ-Fr.5. The molecular weight measurement indicated that pRJ-Fr.5 contained low-molecular weight peptides, such as dipeptides and tripeptides, as well as 16.5% (w/w) 10-HDA. We also measured the 10-HDA content in RJ, pRJ and the other fractions derived from them. The 10-HDA concentrations were as follows: RJ: 1.7%, pRJ: 5.3%, RJ-Fr.1: 8.9%, RJ-Fr.2: 2.1% and pRJ-Fr.4: <0.1%. These results showed that 10-HDA was enriched especially in pRJ-Fr.5.

To elucidate whether 10-HDA is a lifespan-extending agent, we assessed its effects on lifespan. Worms treated beginning at the young adult stage with concentrations of 10-HDA ranging from 10 to 100 µM all showed extensions of the mean and maximum lifespans (Fig. 7, Table S3). The largest increase was observed at 25 µM, at which concentration the mean lifespan was increased by 12%.

We next tested whether 10-HDA extends lifespan through the IIS-DAF-16 pathway by measuring the effects of this compound on the lifespan of the daf-16 deletion mutants. 10-HDA at concentrations from 10 to 100 µM extended the mean lifespan of this mutant by 6–15% (Fig. 8, Table S3). The ability of 10-HDA to extend the lifespan of the daf-16 deletion mutants and the wild-type N2 to similar extents clearly indicates that its lifespan-extending effect is mediated through a mechanism independent of the IIS-DAF-16 pathway.

To examine the contribution of 10-HDA to pRJ-Fr.5-induced lifespan extension, we tested the effect of combining pRJ-Fr.5 and 10-HDA on lifespan. The lifespan extension achieved by the combination of pRJ-Fr.5 and 10-HDA was greater than that induced by each treatment alone, but the effect was less than additive (Fig. 9, Table S3). These results suggest that pRJ-Fr.5 and 10-HDA do not extend lifespan independently of each other. Therefore, part of the lifespan extension by pRJ-Fr.5 is probably due to its 10-HDA component.

The C. elegans strains were maintained at 20°C on nematode growth medium (NGM) agar with Escherichia coli OP50 as a food source, as previously described [94]. The N2 Bristol strain was used as the wild-type C. elegans. The mutant strain used in this study was CF1038: daf-16(mu86) I and TJ356: zIs356 [Ex(daf-16::gfp + rol-6)].

Fresh RJ, which was produced by honeybees (Apis mellifera L.) foraging on Brassica sp. in China, was obtained from Api Co., Ltd., Gifu, Japan. RJ hydrolyzed by Protease N (pRJ) was prepared as previously described [95]. The following drugs and chemicals were purchased and used: 10-HDA (Alfresa Pharma Co., Ltd., Osaka, Japan) and Protease N “Amano” (from Bacillus subtilis; Amano Enzyme Inc. Aichi, Japan).

Fresh RJ (1 kg) was mixed with water (1 L), hexane (2 L) and EtOH (4 L), and then shaken slowly overnight at room temperature. This mixture was filtered through No. 2 filter paper and then the extracts were concentrated under pressure until they became a dark yellow material (RJ-Fr.2). This residue was dried by heating under reduced pressure and then mixed with 5% EtOH. The supernatant from this suspension was then freeze-dried (RJ-Fr.1). The yields of RJ-Fr.1 and RJ-Fr.2 were 40.9% and 16.0%, respectively.

pRJ (20 g) was mixed with water and then chromatographed on an ODS column. The column was eluted stepwise with water and 30% (v/v) aqueous MeOH. Each fraction (1 L each) was collected and freeze-dried. These fractions were designated as pRJ-Fr.4 (15 g in the H₂O phase) and pRJ-Fr.5 (3.8 g in the 30% MeOH phase).

Eggs that were isolated with hypochlorite were placed on fresh NGM agar plates containing UV-killed E. coli strain OP50, unless otherwise stated. UV-killing was used to avoid any effects of live E. coli on the compounds examined in this study and any effects of these compounds on growth of live E.coli. To kill the OP50, plates covered with OP50 were UV-irradiated as previously described [96]. Worms were raised until the L4 molt and were subsequently transferred onto a new plate containing 40 µM 5-fluoro-2′-deoxyuridine (FUdR, Sigma Aldrich, St. Louis, MO, USA) to prevent self-fertilization. The day of transfer at the L4 molt was counted as 0-day adult in the lifespan assay. The worms were transferred to fresh plates daily, and the number of surviving worms was monitored until death unless otherwise stated. Worms were judged to be dead when they did not respond to a mechanical stimulus. To focus on aging, worms that had become desiccated on the side of the plate after crawling off, that displayed extruded internal organs or that died because of progeny hatching inside the uterus (matricidal death) were excluded from our analysis. The results of the survival assays were analyzed using the Kaplan-Meier method, and significance was measured with the log-rank test using the statistical analysis package StatMate III (ATMS, Tokyo, Japan).

EtOH solutions of RJ, pRJ and RJ-Fr.1, aqueous solutions of RJ-Fr.2, pRJ-Fr.4 and pRJ-Fr.5 as well as 10-HDA in DMSO, were added to liquid NGM that had been autoclaved and cooled to 50°C. The media were immediately dispensed into Petri dishes. Experiments involving RJ, pRJ and RJ-Fr.1 were performed in parallel with those involving a control group treated with 0.1% EtOH; and experiments involving 10-HDA were conducted in parallel with those involving a control group treated with 0.03% DMSO.

The C. elegans N2 strains were treated with pRJ-Fr.5 (0 (control) or 25 µg/ml) for 24 h beginning at the L4 stage. A total of 8,000-10,000 worms were collected for each sample. The worms were homogenized in TRIzol® Reagent (Invitrogen™, Carlsbad, CA) using a Precellys 24 (Bertin Technologies, Montigny-le-Bretonneux, France). Total RNA was extracted with a PureLink™ RNA Mini kit (Invitrogen™). The Agilent C. elegans (V2) Gene Expression Microarray, 4x44K (G2519F-020186) was used for global gene expression analysis. This microarray contains 43,803 C. elegans complementary DNA (cDNA) probes, each consisting of a single 60-oligomer oligonucleotide sequence. Target RNA labeling and hybridization were performed according to the protocol for one-color microarray-based gene expression analysis using the Quick Amp Labeling Kit (Agilent Technologies, Santa Clara, CA). In brief, 500 ng of RNA was transcribed using the oligo(dT)-based T7 promoter primer and MMLV-RT in the first- and second-strand cDNA synthesis reactions. The double-stranded cDNAs were used as templates for the preparation of fluorescent complementary RNAs (cRNAs) in the presence of T7 RNA polymerase and cyanine 3-CTP dye in an in vitro transcription reaction. The labeled cRNAs were purified, fragmented, and hybridized to microarrays in a rotating hybridization oven at 10 rpm for 17 h at 65°C. After hybridization, the microarrays were washed according to the manufacturer's instructions and scanned using an Agilent DNA Microarray Scanner with Scan Control software (Agilent Technologies). The resulting images were processed, and the raw data were collected using the Agilent Feature Extraction software. The gene expression data were analyzed using GeneSpring GX 11 (Agilent Technologies). The signal intensity of each probe was normalized by a percentile shift, in which each value was divided by the 75th percentile of all the values in its array. The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus (GEO) and are accessible through the GEO Series accession number GSE26094↗ (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE26094↗). To identify the genes with biological significance, we applied flags attributed to the signal intensity of each probe, the fold change, and the Student's t-test values. All the data are MIAME compliant.

Total RNA was reverse-transcribed to cDNA using a High Capacity cDNA Reverse Transcription kit (Applied Biosystems), and subjected to real-time PCR using the SYBR Premix Ex Taq II (Perfect Real Time) (TaKaRa) and the Thermal Cycler Dice Real Time System (TaKaRa). The following primers were used: ins-9 forward, 5′-GGCGAGAAGAACCTTGGAAAC-3′; ins-9 reverse, 5′-ACAGCACAGCTTAGAGAGATCCTG-3′; ins-20 forward, 5′-TCATCATCACAGGCACAAAGG-3′; ins-20 reverse, 5′-GCAAAATATCATCATCCGTCAGG-3′; ins-23 forward, 5′-CAGAGCTTCACGTTCGTAGGG-3′; ins-23 reverse, 5′-GAACAGTACTCGGTTGGACTTGG-3′; dod-3 forward, 5′-AAGCCATGTTCCCGAATGAG-3′; dod-3 reverse, 5′-GCTGCGAAAAGCAAGAAAATG-3′; dod-19 forward, 5′-ACCGTTCCCAGTTTTACAGTCC-3′; dod-19 reverse, 5′-TATTTTGAGGCGCGGATACAC-3′; dao-4 forward, 5′-GCACATTACAAATGCTTCAAGGAC-3′; dao-4 reverse, 5′-TGACACCCTCATCCCCATAAC-3′; fkb-4 forward, 5′-CTATGCGAGGAATGTGTATTGGAG-3′; fkb-4 reverse, 5′-TGGACAGTGTAATAGAGTGGCTGAC-3′; rla-1 forward, 5′-ACCGGCGAGAAGATCGCTAC-3′; rla-1 reverse, 5′-CGGAAGAGACAGAAGTGATGAGG-3′. The relative expression level of each gene was calculated using the comparative Ct method. Ribosomal protein, Large subunit, Acidic (P1) family member (rla-1) was used as an internal control gene.

Sugar content was estimated by orcinol/sulfuric acid analysis. Peptide content was estimated by the Lowry method. The molecular weights of peptides were estimated by HPLC analysis on a Superdex Peptide HR 10/30 column (Pharmacia Biotech, Uppsala, Sweden). The 10-HDA content was determined using HPLC with a Consmosil 5C18-MS-II column (Nacalai Tesque, Tokyo, Japan) at 40°C. The column was eluted with a mobile phase of 10 mM phosphate buffer (pH 2.5) and MeOH (1960:1540, v/v) at a flow rate of 1.0 ml/min.