Ethanol‐soluble proteins from the royal jelly of Xinjiang black bees

Abstract

Royal jelly is a nutritious food that has beneficial effects to human health. However, the functional substances remain unclear. Herein, we fractioned the royal jelly proteins of Xinjing black bees according to the Osboren method. Two main proteins from the ethanol‐soluble fraction were purified and identified. RJG‐1 was determined as glucosylceramidase, and RJG‐2 was major royal jelly protein 1 (MRJP1). RJG‐1 showed potent cytotoxicity toward various mammalian cells, and caused quick disruption of cell membranes. With glucosylceramidase activity, RJG‐1 may degrade the glucosylceramide of the cell membranes and disrupt the membrane structure, thereby resulting in cell necrosis. This study extends our knowledge about the composition and function of royal jelly, and is significant for the application of royal jelly.

Royal jelly is a nutritious food mainly containing proteins, which have broad biological functions, such as anti‐oxidation, immunity regulation, anti‐hypertension, antitumor, and so on. ¹ , ² , ³ It is reported that more than 80% of the soluble proteins belong to the major royal jelly protein (MRJP) family, including MRJP1–MRJP9. ⁴ , ⁵ , ⁶ However, due to the low abundance and complexity of royal jelly protein components, other minor proteins are difficult to be isolated and identified. Proteomic analysis has been extensively used to identify the proteins in royal jelly. Scarselli et al. identified more than 20 proteins from the royal jelly of Apis mellifera, and they all belong to the Apis mellifera genome. ⁷ Schönleben et al. identified 20 proteins from royal jelly by trying to exclude the influence of high content of MRJP family proteins. ⁵ Furusawa et al. revealed 46 novel proteins and identified some potential phosphorylation and glycosylation sites using one‐ and two‐dimensional proteomics platforms. ⁸ Li's group did an extensive work in exploring the proteome of royal jelly of honey bees, and has identified over a hundred proteins from Apis mellifera, and dozens of proteins from Apis cerana. ⁹ , ¹⁰ , ¹¹ However, except the MRJPs, the functions of royal jelly proteins are rarely determined due to the low abundance.

Even for the MRJPs, their functions are only partially revealed. It has been reported that MRJP1 is important for both honeybee development and human health. For honeybees, MRJP1 plays a dominant role in determining whether the larvae develop into a queen or worker bees. ¹² . In regard to human health, MRJP1 showed a potential effect on regulating blood pressure by in vitro assay, ¹³ and Fan et al. indicated that it could regulate vascular smooth muscle cell contraction, migration and proliferation. ² MRJP1 and MRJP2 are able to stimulate mouse macrophages to release tumor necrosis factor (TNF‐α). ¹⁴ MRJP2 also shows inhibitory effect on Paenibacillus larvae. ¹⁵ Kohno et al. found that MRJP3 might have anti‐inflammatory effects by inhibiting the production of proinflammatory cytokines by macrophages. ¹⁶ In addition to these large amount major proteins, some low molecular weight antimicrobial peptides have also been identified, including royalisin ¹⁷ and jelleines. ¹⁸ These functional proteins are mainly water‐soluble and major proteins, but the activities of most low‐abundant proteins are still unknown. In this study, we focused on the components of the ethanol‐soluble proteins in the royal jelly of Xinjiang black bees, which is a sub‐species of Apis mellifera mellifera, and has adapted to the harsh local natural environment. ¹⁹

Royal jelly proteins have complicated components. In order to trace the bioactive protein components, the royal jelly proteins were firstly fractioned according to the Osboren method. ²⁰ Different fractions were subjected to MTT assay for the detection of cytotoxicity using HeLa cells. The water‐soluble protein, globulin, and glutelin fractions showed moderate cytotoxicity, while the ethanol‐soluble protein fraction showed high cytotoxicity against HeLa cells (Figure 1a). SDS‐PAGE showed that the ethanol‐soluble protein fraction had relatively simple components compared to the other fractions. There were two main protein bands in this fraction, and their molecular weights were about 50 and 60 kDa, respectively (Figure 1b). Thus, these two proteins were temporarily named as royal jelly gliadin‐1 (RJG‐1, the ~60 kDa protein) and royal jelly gliadin‐2 (RJG‐2, the ~50 kDa protein), respectively.

FIGURE 1.

Bioassay guided separation of royal jelly proteins using the Osboren method. (a) The cytotoxicity of different fractions of royal jelly proteins from Xinjiang black bees on HeLa cells. (b) Different fractions of royal jelly proteins from Xinjiang black bees were analyzed by SDS‐PAGE. The ethanol‐soluble proteins showed two main bands, which were assigned as RJG‐1 and RJG‐2, respectively. ESP, ethanol‐soluble protein; WSP, water‐soluble protein; GLB, globulin; GLL, glutelin. (c) The IC50 values of RJG‐1 on various cell lines measured by MTT assay. The IC50 values were computed and expressed as an average of at least three independent experiments. *p<.05, comparison of the IC50 values of RJG‐1 on 22Rv1 and WPMY‐1 cell lines

Due to the relatively simple components and higher cytotoxicity of the ethanol‐soluble proteins, we tried to purify the two proteins RJG‐1 and RJG‐2, but left the other three fractions, as their compositions are too complicated. The ethanol‐soluble proteins were further separated with an ÄKTA pure chromatography system using a Sephadex G‐200 column, according to their difference in molecular weight. Although two peaks were observed from the chromatogram (Figure S1), only purified RJG‐1 was obtained (Figure 1b), which was used for further activity test. RGJ‐2 was not purified, and it might be due to the effect of MRJPs, as their concentrations were too high, ⁵ and several MRJPs have similar molecular weights to MRJP1. The protein RJG‐1 was then subjected to SDS‐PAGE, and a single clear band was observed (Figure 1b). RJG‐1 displayed quite different properties from the water‐soluble proteins, as it has a low solubility in water. A subsequent cytotoxicity assay for RJG‐1 was performed by treating the cells with RJG‐1 protein (0–3 mg/ml) for 4 hours, and measured using the MTT method. The results showed that RJG‐1 had a potent cytotoxicity against various cell lines (Figure 1c). WPMY‐1 is an immortalized normal cell from human prostate, and RJG‐1 had relatively lower cytotoxicity on it compared to the prostate cancer cell line 22Rv1 (p <.05). Toward HeLa and RAW264.7 cells, RJG‐1 also showed a high cytotoxicity (Figure 1c). As for the following protein identification that RJG‐2 was determined as MRJP1, which has never been reported for cytotoxicity, thus, RJG‐1 is the main protein responsible for the cytotoxicity of the ethanol‐soluble protein fraction, and this protein may have a potential application in cancer treatment.

The RJG‐1 and RJG‐2 protein bands of the SDS‐PAGE were in‐gel trypsinized and identified by LC‐MS/MS. Based on the Sequest searching for RJG‐1, the protein with the highest score was glucosylceramidase (score was 1,662.36) with a molecular weight of 59.3 kDa and a calculated isoelectric point (pI) of 5.57. A specific tryptic peptide FFSAAWSAPTWM_(oxidation)K from glucosylceramidase was detected by the high‐resolution LC‐MS/MS. As shown in the mass spectrum of the ion, the monoisotope peak at the m/z of 773.36 in double charge state was consistent with the molecular weight of this peptide (Figure 2a). The fragmented b and y ions of the precursor ion also matched the peptide sequence well (Figure 2b), indicating RJG‐1 is glucosylceramidase (Accession no. A0A088APM4). This protein was isolated from royal jelly for the first time, though it had been predicted as a glucosylceramidase‐like protein in the Gene database. Similarly, RJG‐2 was identified as major royal jelly protein 1 (MRJP1) as it had the highest score (2,257.66), which has a molecular weight of 48.9 kDa and a calculated pI of 5.34. A specific tryptic peptide TSDYQQNDIHYEGVQnILDTQSSAK from MRJP1 was detected, whose monoisotope peak at the m/z of 952.11 in triple charge state was exactly consistent with its molecular weight (Figure 2c). The fragmented b and y ions of the precursor ion also matched the peptide sequence (Figure 2d), indicating the SDS‐PAGE band of RJG‐2 is MRJP1 (Accession no. O18330). MRJP1 is the most abundant royal jelly protein, playing pivotal roles in honeybee nutrition and larvae development. Its crystal structure has been identified in an oligomer form in complex with apisimin and methylenecholesterol, in which MRJP1 shows a unique six‐bladed β‐propeller fold with three disulfide bonds. ²¹

FIGURE 2.

Identification of RJG‐1 and RJG‐2. (a) A specific tryptic peptide FFSAAWSAPTWM_(oxidation)K of glucosylceramidase was detected by LC‐MS/MS from the RJG‐1 protein band. (b) The fragmented b and y ions were assigned to the specific peptide sequence. (c) A specific tryptic peptide TSDYQQNDIHYEGVQnILDTQSSAK of MRJP1 was detected by LC‐MS/MS from the RJG‐2 protein band. (d) The fragmented b and y ions were assigned to the specific peptide sequence

RJG‐1 showed potent cytotoxicity against mammalian cells (Figure 1c), but no antibacterial activity (data not shown). The activity of RJG‐1 on HeLa cells was further investigated by flow cytometry after staining the RJG‐1 treated cells with the cell apoptosis kit (Annexin V/propidium iodide). It showed that the number of necrotic cells increased after treated with RJG‐1 at IC₅₀ for different time (Figure 3a). However, the number of apoptosis cells showed no obvious increase. This suggests that RJG‐1 induces cell necrosis instead of cell apoptosis. Direct observation of the cell morphology under the confocal microscope was performed by staining the cells with a Live/Dead staining kit after the treatment of RJG‐1. The untreated cells were stained green with smooth boundary of the cell membranes, while the cells started to turn red after a short incubation period (1 min) with RJG‐1. As the incubation time increased, the cell membranes showed obvious roughness and some small particles appeared around the cells (Figure 3b), which could be membrane fragments. This indicates that RJG‐1 could cause serious cell membrane damage and kill the cell. The observed cell disruption was much more than that detected by the flow cytometry, which might be because the cell fragments cannot be detected by the flow cytometer. Because RJG‐1 was determined as glucosylceramidase, the enzymatic activity of this protein was detected using a substrate 4‐methylumbeliferone (4MU)‐β‐d‐glucopyranoside that can generate fluorescence when 4MU was released. ²² The results showed that RJG‐1 had a high glucosylceramidase enzyme activity (Figure 3c).

FIGURE 3.

The action mechanism of RJG‐1. (a) Flow cytometry was carried out to detect cell apoptosis and necrosis after the cells were treated with blank medium for 4 hr, RJG‐1 at IC50 for 1 or 4 hr. The cell apoptosis kit (Annexin V/PI) was used to stain cells after the treatment. No obvious increase in cell apoptosis but an increase of necrotic cells was observed after the treatment with RJG‐1. (b) The morphological changes of cells after treatment with RJG‐1 were observed under a confocal microscope. HeLa cells were stained with Live/Dead staining kit and then treated with RJG‐1 at IC50 for different time, and then photographed using the confocal microscope. The live and dead cells were indicated by green and red, respectively. (c) The glucosylceramidase enzymatic activity was detected by 4‐methylumbeliferone‐β‐d‐glucopyranoside, which can generate fluorescence after treated by glucosylceramidase

The cytotoxicity of royal jelly proteins had not been reported, even for those antimicrobial proteins and peptides, such as MRJP2, royalisin and jelleines. ¹⁵ , ¹⁷ , ¹⁸ On the contrary, the oligomer of MRJP1 showed obvious proliferative activity. ²³ RJG‐1 caused cell necrosis, which may be due to the quick cell membrane disruption (Figure 3a,b). This phenomenon is quite similar to that induced by some antimicrobial lytic peptides, ²⁴ but no obvious cell membrane blebbing was observed for RJG‐1 treated cells. RJG‐1 has been confirmed with glucosylceramidase enzymatic activity. It is a kind of enzyme that hydrolyses the β‐glucosidic linkage of glucosylceramide, ²⁵ which is a membrane sphingolipid and the core structure of many glycolipids. RJG‐1 treatment may degrade the glucosylceramide of cell membrane and disrupt the membrane structure. This could explain the cytotoxicity of RJG‐1 and the observed cell morphological changes.

For the two main components of the ethanol‐soluble proteins, MRJP1 existed in all fractions as shown in the SDS‐PAGE (Figure 1b), while RJG‐1 existed mainly in the ethanol‐soluble proteins, which was indicated by the enzymatic assay (Figure 3c, S2). The content of MRJP1 in the ethanol‐soluble proteins was much lower than that in the other fractions, suggesting the Osboren method greatly reduced the content of MRJPs, thus glucosylceramidase (RJG‐1) could be purified from royal jelly for the functional tests. Additionally, it also indicates that this method is suitable for the separation of royal jelly proteins and helpful for the study on the function of royal jelly. Therefore, this study extends our knowledge about the components of royal jelly and will be helpful for the functional study.

AUTHOR CONTRIBUTIONS

Yanyan Yuan: Data curation; investigation; methodology; writing‐original draft. Wujun Wang: Investigation; methodology. Ruru Fan: Investigation; visualization. Jianhui Jiang: Funding acquisition; project administration; resources; writing‐review and editing. Shan Feng: Data curation; investigation; software. Huiwei Yin: Investigation. Shi‐Zhong Luo: Funding acquisition. Long Chen: Data curation; formal analysis; funding acquisition; project administration; supervision; writing‐review and editing.

CONFLICT OF INTEREST

The authors declare no competing financial interest.

Supporting information

Appendix S1: Supporting Information

Yuan Y, Wang W, Fan R, et al. Ethanol‐soluble proteins from the royal jelly of Xinjiang black bees. Protein Science. 2021;30:291–296. 10.1002/pro.3985