The mechanism of Andrena camellia in digesting toxic sugars (2024)

PMCID: PMC11152697

Zhen Li,^1,2,3,5 Shiqing Zhong,^1,2,5 Qiang Huang,^1,2 Yong Zhang,^1,4 Tianyu Xu,^1,2 Wenkai Shi,^1,2 Dongsheng Guo,³ and Zhijiang Zeng^1,2,6,∗

Author information Article notes Copyright and License information PMC Disclaimer

Table1

Compositions of manninotriose, raffinose, and stachyose and their metabolites in gut of Andrena camellia, Apis cerana, and Apis mellifera at different time points (n= 3)

Table2

Composition of galactose and its metabolites in the gut of A.camellia, A.cerana, and A.mellifera after foraging for mannitritose, raffinose, and stachyose at different time points (n= 3)

Lead contact

For information and requests for resources, please contact Zhi Jiang Zeng (moc.anis@5691seeb).

Materials availability

This study did not generate new unique reagents.

Data and code availability

A. camellia genome at the chromosomal level have been deposited in NCBI under accession number PRJNA921724. All other data are included in the supplemental information.

Insects collection

The samples of wild female A.camellia were collected from the C.oleifera plantation in Xicun Town, Yuanzhou District, Yichun City (114°20′ N, 27°73′ E). Apis cerana and Apis mellifera were placed in the same C.oleifera plantation, with five experimental colonies of each bee species.

Enzyme activity of GAL, GALK, GALT and GALE in the gut of A.camellia, A cerana and A.mellifera

The gut of the foragers from A.camellia, A.cerana, or A.mellifera was dissected (discarding the honey sacs, 5 bees’ guts made up one sample, 8 biological replicates for each species) in 1.5mL sterile EP tubes, 1 000μL ultrapure water was added and well grinded. The supernatant was collected after 5min of standing, centrifuged at 12 000rpm for 10min and assessed for the enzyme activity of galactosidase (GAL), galactokinase (GALK), galactose-1-phosphate uridyltransferase (GALT) and UDP-galactose-4-epimerase (GALE) by ELISA (Mlbio Biotechnology Co., Shanghai, China) according to per the manufacturer’s instructions.

Sample treatment and analysis of oligosaccharides and breakdown products in the gut of three types of bees by high-performance liquid chromatography (HPLC)

Before dietary treatment, bees (A.camellia, A.cerana, and A.mellifera foragers) were starved for 2h and then fed mannotriose, raffinose, or stachyosedissolved in 50% (w/w) fructose solution. The three oligosaccharides were added in the same amount as in the C.oleifera pollen. After 48h and 72h of treatment, the entire gut was dissected from. Three oligosaccharide solutions were prepared by dissolving mannitritose, raffinose and stachyose (Sigma, Missouri, US, LC-MS, >99% pure) in 50% fructose solution according to the mass ratio (w/w) in C.oleifera pollen, individually.11 For detailed diet ratios, describe as TableS2. Collecting foragers of A.camellia, A.cerana and A.mellifera, respectively. All three types of bees were divided into three treatment groups, and each of the three groups was adequate supplied with a 50% fructose solution containing three individual oligosaccharide diets. Each treatment group was set up in three biological replicates, comprising 30 honey bees per replicate (reared in a plastic cup cage). The plastic cup cages with A.cerana and A.mellifera were placed in an incubator at a temperature of 34.5°C and a relative humidity of 75%. The cup cages with A.camellia were incubated at a temperature of 25°C and 45% relative humidity. At the beginning of the individual oligosaccharide feeding treatment, all honey bees were starved for 2 h.23 The gut from the three treatment groups of the three honey bees (with the honey sacs removed) was dissected in 1.5mL sterile EP tubes at 48th h and 72nd h after the start of feeding, separately. The gut of 15 honey bees from each cup cage was dissected each time to comprise one sample, and three samples were taken from each treatment group each time. Before dissection, the 15 honey bees in each group were weighed, and the gut dissected out of each tube was also weighed for subsequent weight conversions. Meanwhile, the foraging amounts of the three types of honey bees at the 48th h and 72nd h were also weighed and recorded.

After weighing 15 kinds of saccharide standards separately and accurately (see TableS3 for 15 saccharide standards information), add ultrapure water to prepare 10mg/mL single standard master batch, and then take the appropriate amount of single standard master batch to prepare 40μg/mL standard mixed standard (for 15 types of saccharide concentration, refer to TableS4). Accurately weigh 50mg of freeze-dried gut powder sample (three samples per treatment group) in a 2.0mL centrifuge tube and add 700μL of 80% ethanol. The sample was shaken at 50°C for 2h and then diluted with 700μL H₂O and centrifuged at 10 000rpm for 3min. Lastly, the supernatant was transferred to a 1.5mL injection vial.

A Thermo ICS5000 ion chromatograph (Thermo Fisher Technology Co., Ltd., Waltham, MA, USA) equipped with an electrochemical detector and a CarboPac™ PA1 (250×4.0mm, 4μm, Thermo Fisher Technology, Waltham, MA, USA) chromatographic column was used to construct standard curves for 15 saccharides (working solution concentration as the x-axis and peak area as the y-axis) and to simultaneously perform absolute quantification of saccharides in honey bee gut samples. The mobile phases were A: H₂O, B: 100mM NaOH; the injection volume was 10μL, the flow rate was 1.0mL/min, and the column temperature was 30°C. Elution gradient: 0.0∼12.0min, 5% ∼10%B; 12.1∼15.0min, 5%∼100% B; 15.1∼25.0min, 100% B; 25.1∼40.0min, 100%∼5% B; 40.1∼60.0min, 5% B. Fitting information for the 15 saccharides was presented in TableS5.

Appropriate amounts of galactose 1-phosphate, uridine diphosphate galactose (UDP-galactose) and uridine diphosphate glucose (UDP- glucose) standards (Sigma, Missouri, US, LC-MS,>99% pure) were accurately weighed and prepared into a single standard master batch of 10mg/mL with methanol, individually. This was followed by aspirating three appropriate amounts of the saccharide single standard master batch to compose mixed standard solutions of different concentrations as shown in TableS6.

50mg of freeze-dried A.camellia, A.cerana or A.mellifera gut sample (three samples per treatment group, all honey bees were foragers) was weighed into a 5mL sterile centrifuge tube and 1000μL of extraction solution (methanol: acetonitrile: water= 2:2:1, v/v) was added. Ultrasonic treatment in the ice water bath for 10min, liquid nitrogen flash freezing for 1min, repeated three times. Samples were placed at -20°C for 1h and later centrifuged at 13 000rpm for 15min at 4°C. The supernatant was removed by aspirating 800mL into a new 2mL sterile centrifuge tube, blowing dry with a nitrogen blower and adding 600μL of 50% acetonitrile aqueous solution (v/v) for re-dissolution. Dissolved samples were shaken for 30s and then sonicated in an ice water bath for 10min. Following sonication, the samples were centrifuged at 3 000rpm for 15min at 4°C and 400μL of supernatant was injected into a 1.5mL injection vial.

High-performance liquid chromatography (LC-30A, Shimaduzu Co. Tokyo, Japan) tandem mass spectrometry (AB Sciex, 5600 Q-TOF/6500 Q-TRAP, Framingham, MA, USA) was applied to construct standard curves for the three saccharides (working solution concentration as the x-axis and peak area as the y-axis) and to concurrently detect the absolute compositions of galactose 1-phosphate, UDP-galactose and UDP-glucose in all honey bee gut samples. Separations were carried out using an ACQUITY UPLC® BEH C18 chromatographic column (2.1×100mm, 1.7μm, Waters, Milford, MA, USA). Analytes were separated by ultrapure water (containing 25mM CH₃COONH₄ and 25mM NH₄OH) (A)and Acetonitrile (B)at a flow-rate of 0.3mL/min. The linear gradient elution program was: 0.0∼1.0min, 85% B; 1.1∼12.0min, 65% B; 12.1∼15.0min, 40% B; 15.1∼20.0min, 85% B. The column was thermostated at 40°C and injection volume 5μL.

Mass spectrometry conditions: electrospray ionization source (ESI). Ion-source temperature 600°C, ion-source voltage 5500 V, curtain gas 20 psi, nebulizer gas and auxiliary gas 60 psi. Scanning was performed using multiple reaction monitoring (MRM).

The standard curves for the galactose 1-phosphate, uridine diphosphate galactose and uridine diphosphate glucose were provided in TableS7.

The relative expression level of four genes in the heads and thorax of A.camellia, A.cerana, and A.mellifera

To explore the NAGA-like, GALK, GALT, and GALE genes expression of the head and thorax of the A.camellia, A.cerana and A.mellifera. The head and thorax of the three species were dissected at the 48th h of feeding. Total RNA was extracted from the head and thorax tissues after feed the three mixed oligosaccharides. The head and thorax of A.camellia, A.cerana or A.mellifera were dissected in 1.5mL RNase centrifuge tubes as one sample (eight biological replicates were set for each honey bee species) and total RNA was extracted according to the instructions of the TransZol™ Up Plus RNA extraction kit (Beijing TansGen Biotechnology Co., Beijing, China). The final volume of complementary DNA (cDNA) prepared with a cDNA synthesis kit (Takara, Dalian, China) was 30μL. Primers were designed by Primer 5.0 using the CDS sequences of A.camellia obtained from our lab sequence results, and the CDS sequences of A.cerana (ApisCC1.0) and A.mellifera (Amel_HAv3.1) genes downloaded from National Center for Biotechnology Information (NCBI). Sequences of primers for the GAL, GALK, GALT, and GALE genes of A.camellia, A.cerana and A.mellifera were listed in TableS8. All the primers were synthesized by Sangon Biotech (Shanghai, China). RT-qPCR amplification systems include 5μL of SYBR® Premix ExTaqTM II (Takara Co. Beijing, China), 0.2μL ROX (Takara Co. Beijing, China), 0.4μL of forward primer (10μM), 0.4μL of reverse primer (10μM), 3μL nuclease-free water, and 1μL of diluted template cDNA (400ng/μL) in an RT-qPCR machine (ABI Q5, Fisher Scientific Inc., Waltham, MA, USA). Cycling conditions were 95°C for 45s and 60°C for 1min, followed by 40 cycles of 50°C heated to 90°C (1°C increase per 6 s). At the end of the reaction, according to the CT value of each target gene and the reference gene (GAPDH), the relative expression of the target gene was calculated by the 2^–ΔΔCT method.24

Clustering relationship analysis of NAGA-like, GALK, GALT, and GALE genes in different honey bee genomes

hom*ologous Coding Sequence of NAGA-like, GALK, GALT, and GALE in the genomes of 15 bee species were selected for cluster analysis, with Drosophila melanogaster as the outgroup. Phylogenetic trees were constructed using the MEGA 7.0 software packages.

Invitro expression of four enzymes and invitro digestion of three oligosaccharides and galactose

NAGA-like, GALK, GALT, and GALE were successfully expressed using the constructed plasmids with the coding regions of four genes into E.coli (E.coli BL21 (DE3)-PET-28A-(NAGA-like; GALK; GALT; GALE)). Finally, the conversion of galactose to UDP-glucose was simulated invitro using these four enzymes. Synthesize NAGA-like, GALK, GALT or GALE genes respectively, prepare the following system according to TableS9 in a 0.2mL EP tube, dilute the template 20-fold, and take 0.5μL to amplify NAGA-like, GALK, GALT or GALE. Mix well, and then put it into the GeneAmp PCR System 2400 PCR equipment (Fisher Scientific Inc., Waltham, MA, USA) for amplification. Amplification conditions were 94°C for 1min, followed by 30 cycles (98°C for 15 s, 58°C for 15s and then 68°C 1min). After storage at 68°C for 5min, the final instrument temperature was reduced to 4°C to preserve the samples.

The PCR of NAGA-like, GALK, GALT or GALE products were recovered using the DNA Gel Recovery Kit (Dongsheng Biotech, Guangzhou, China), and the experimental procedures were referred to the reagent manufacturer's instructions. Photographs of PCR products during cut gel recovery are shown in FigureS1. Then, the PCR recovery products of NAGA-like, GALK, GALT or GALE and pET-28a+ plasmid (Fisher Scientific Inc., Waltham, MA, USA) was each taken in 15μL, respectively, and were double digested with XhoI/BamHI (Takara, Dalian, China), and the enzymatic system was shown in TableS10. After mixing, the reaction was done at 37°C for about 3 h. Likewise, the enzymatic products of GAL, GALK, GALT or GAKLE were recovered using the DNA Gel Recovery Kit (Dongsheng Biotech, Guangzhou, China). Linking of the target fragment to the plasmid was performed using T4 DNA Ligase (Takara, Dalian, China), with the reagent addition system as described in TableS11, and sustained ligation at 16°C for 1 h.

5μL of ligation product (GAL, GALK, GALT or GALE) was added to a 1mL centrifuge tube with 50μL of DH5α receptor cells in an ice bath. The tubes were then gently spun and mixed and incubated in an ice bath for 30min. Heat shock was applied to the tubes in a water bath at 42°C for 90 s. Following this, the tubes were immediately transferred to an ice bath and incubated for 2min. 200μL of LB medium was added to each of the tubes, mixed, and incubated with shaking at 37°C for 1h at 200rpm. The liquid was spread uniformly on a Luria-Bertani (LB) medium containing kanamycin (100μg/mL) in an ultra-clean bench at room temperature until the liquid was absorbed. Finally, the plates were inverted and transferred to a biochemical incubator at 37°C for overnight incubation (12 h).

Several single clones were picked from the plates and cultured in centrifuge tubes containing 3mL of LB medium overnight (12 h) on a shaker to extract the plasmids, which were digested at 37°C for 2h (the enzymatic reaction system as described in TableS12). The digested products (GAL, GALK, GALT, or GALE) were separated by electrophoresis on 1% agarose gel containing ethidium bromide (EB) and imaged by UVP gel imaging system (FigureS2).

The constructed expression plasmid (PET-28A-( GAL, GALK, GALT, GALE)) was transformed into E. coli BL21 (DE3) expressing bacteria, coated on LB plates containing the appropriate antibiotic (Kan, final concentration 50μg/mL) and incubated in an inverted incubator at 37°C overnight (12 h).Next, single colonies were picked from the plates and inoculated in 3mL of LB liquid medium (containing Kan) in an ultra-clean bench and pre-incubated overnight (12 h) at 37°C on a 230rpm shaker. In a 1L triangular flask, 2mL of pre-culture bacterial solution was added to 200mL of LB medium (containing Kan) at a ratio of 1: 100 with a micropipette and incubated at 37°C on a shaking bed at 230rpm for about 2-3 h, so that its OD_600nm reached 0.4-0.8.

Also in the ultra-clean bench, 1mL of 100mM inducer IPTG was added to the culture solution to give a final concentration of 0.5mM. Incubation was induced overnight (12 h) at 18°C on a shaker at 180rpm. After incubation, the triangular vials were removed from the shaker and placed on ice for 10min, during which time they were shaken two or three times. Using a pipette, the bacterial solution was pipetted into a clean 50mL centrifuge tube and centrifuged at 4 000rpm for 10min at 4°C. The supernatant was discarded. Repeat centrifugation until all organisms were collected.

Bacteria were resuspended in ice pre-cooled lysis buffer (50mM NaH₂PO₄, 300mM NaCl, 10mM imidazole, adjusted to pH 7.4 with NaOH), trying to avoid air bubbles as much as possible, and 200mL of bacterial solution was finally resuspended in 16mL lysis buffer and shaken slowly for 1h on ice. A total of 5 bottles were shaken, totaling 1 L, and finally resuspended in 80mL lysis buffer, which was divided into two 50mL centrifuge tubes, 40mL in each tube, and then ultrasonicated on ice for 6 s, with 6s intervals, 60 times, at 300 W. After ultrasonication, centrifugation was performed at 4° C for 30min at 10,000 g. Transfer the supernatant to a clean centrifuge tube, add 1mL of Ni-NTA beads (Kingsley Biotechnology, Shanghai, China) pretreated with lysis buffer, and shake slowly on ice for 1h to fully bind the beads to the protein.

Mixed solution of beads-proteins was transferred to a Poly-Prep chromatography column (Bio-Rad, Hercules, CA, US) to allow beads to settle naturally. Washing with 8mL wash buffer (50mM NaH₂PO₄, 300mM NaCl, 20mM imidazole, adjusted to pH 7.4 with NaOH) was performed two times. Finally, the lower target proteins were eluted with elution buffer (50mM NaH2PO4, 300mM NaCl, 250mM imidazole, adjusted to pH 7.4 with NaOH) into 1.5mL EP tubes 3 times, 1mL each time. Bradford assay was used to determine the protein concentration (TableS13) in each tube and then 12% SDS-PAGE was applied to detect the purification effect (FigureS3).

Analysis the efficiency of hydrolysis with three types of oligosaccharides with GAL invitro by HPLC

Firstly, 100mg each of mannotriose, raffinose and stacyose standards (Sigma, Missouri, US, LC-MS,>99% pure) were dissolved in 5mL of LC-MS-grade water, and then transferred to a 10mL volumetric flask and fixed to scale line with LC-MS-grade water. Before testing, each standard solution was diluted according to the gradient shown in TableS14 and loaded into a 1.5-mL injection vial. The fitting degrees of the 3 saccharide standards were all ≥ 0.9992, indicating a strong linear relationship. The results are shown in TableS15.

The mother liquor containing two times the content of single oligosaccharide was prepared with reference to the content of manninotriose and raffinose and stachyose in Camellia oleifera nectar,11 respectively, and then 1mL of the mother liquor was added to each of the two 10mL centrifuge tubes. One tube was used as a control with 1mL of PBS, and the other tube was added with 1mL of GAL solution (concentration with 0.8mg/mL), and two tubes both to shake and mix. Three replicates were set up for each group. The oligosaccharide content in control and treatment groups were detected after 48h (the digestion was stopped by heating the tubes at 80°C for 15min to inactivate the enzyme) of digestion at room temperature using HPLC techniques (Agilent, Santa Clara, CA, USA). Samples were diluted 4-fold prior to testing.

Analyses were performed in a ZORBAX NH₂ column (4.6×250mm, 5μm, Agilent, USA). Mobile phase A: 67% acetonitrile in LC-MS-grade water (v/v); gradient elution program: 0.0-17.0min, 100% A; flow rate: 1.0mL/min; column temperature 35°C; post-run for 5min. The amount injected into the liquid chromatography system was 10μL.

To prepare mixed standards including galactose, galactose 1-phosphate, UDP-galactose and UDP-glucose standards (Sigma, Missouri, US, LC-MS,>99% pure) were accurate weighed. Then, LC/MS-grade water was used to prepare mixed standard santonin and caffeine solutions of 0.16, 0.8, 4, 20 and 40mg/mL, respectively, which were used to form the standard working curves of both (TableS16).

The masterbatch was prepared based on the total amount of galactose in C. oleifera nectar and the solvent was mass spectrometry grade water.11 Take two 10mL centrifuge tubes, add 1mL of mother liquor to one tube and then add 3mL PBS solution as control group. After adding 1mL of mother solution to the other tube, 1mL each of GALK, GALT and GALE solutions (GALK concentration with 1.5mg/mL, GALt concentration with 0.9mg/mL, GALE concentration with 1.0mg/mL) were added sequentially as treatment groups. For each group of experimental sets, four biological replicates were established. The two groups of tubes were shaken and mixed and then digested until 48h (the digestion was stopped by heating the tubes at 80°C for 15min to inactivate the enzyme) and LC-MS (AB Sciex, 5600 Q-TOF/6500 Q-TRAP, Framingham, MA, USA) was applied to detect the concentration of galactose, galactose 1-phosphate, UDP-galactose and u UDP-glucose in all samples.

Separations were carried out using an ACQUITY UPLC® BEH C18 chromatographic column (2.1×100mm, 1.7μm, Waters, Milford, MA, USA). Analytes were separated by LC-MS-grade water (containing 25mM CH₃COONH₄ and 25mM NH₄OH) (A)and Acetonitrile (B)at a flow-rate of 0.3mL/min. The linear gradient elution program was: 0.0∼1.0min, 85% B; 1.1∼12.0min, 65% B; 12.1∼15.0min, 40% B; 15.1∼20.0min, 85% B. The column was thermostated at 40°C and injection volume 5μL.

Mass spectrometry conditions: electrospray ionization source (ESI). Ion-source temperature 600°C, ion-source voltage 5500 V, curtain gas 20 psi, nebulizer gas and auxiliary gas 60 psi. Scanning was performed using multiple reaction monitoring (MRM).

Data access

A.camellia genome at the chromosomal level have been deposited in NCBI under accession number PRJNA921724. All other data are included in the supplemental information.

Data analysis

All data are expressed as mean ± S.E.. For statistical analysis, one-way analysis of variance (ANOVA) was used to compare the relative gene expression level, enzyme activity and saccharide composition among the three honey bees followed by appropriate post hoc test was used to determine statistical significance (P<0.05).

Author contributions

Z.L. and Z.J.Z. designed research; Z.L., S.Q.Z., and T.Y.X. performed research; Z.J.Z., Q.H., and D.S.G. provided guidance for data; Z.L., Y.Z., and W.K.S. analyzed data; Z.L. and Z.J.Z. wrote the paper.

Declaration of interests

The authors declare no conflicts of interests.

Click here to view.^{(528K, pdf)}