Sampling of honeybees
To identify changes in the intestinal bacteria composition and abundance of honeybees (Apis mellifera) in winter (i.e., winter bees) across colonies, we selected a colony (Anhui Agricultural University, Hefei, Anhui province) that has successfully overwintered to spring reproduction for more than 3 years in normal management mode. According to the experience of colony management, honeybees begin to overwinter and stay in hive for about two months in Hefei, Anhui province, China. Taking this into account, we collected samples on the 10th of December, 2019 (before overwintering, BO), the 25th of December, 2019 (prophase of overwintering, PO), the 10th of January, 2020 (metaphase of overwintering, MO), the 25th of January, 2020 (anaphase of overwintering, LO), 10 February, 2020 (telophase of overwintering, EO), and the 25th of February, 2020 (after overwintering, AO). We sampled overwintering bees (PO, MO, LO and EO) on top of the frames inside the hive and before or after overwintering bees (include BO and AO) from the outside of the hive. A total of 18 samples containing 270 honeybees were collected. Each biological sample comprised 15 honeybees to decrease the effect of the differences between individual honeybees and three biological replicates were analyzed to decrease the experimental error.
Bees were anesthetized at −20 °C for 5 mins. Then, the surface bacteria of honeybees were cleaned off using 70% absolute alcohol, 80% absolute alcohol, 90% absolute alcohol, 0.1 M PBS (Phosphate-buffered Saline, purchased from Sangon Biotech, Shanghai, China), and sterile water 59. After cleaning, the whole gut of the bees were carefully removed into 1.5 ml sterile centrifuge tubes using sterile forceps, immediately frozen using liquid nitrogen, and stored at −80 °C until further analysis.
DNA extraction
From each biological sample, 15 guts were pooled together into a 1.5 ml sterile centrifuge tube. Total genomic DNA of intestinal bacteria was collected from guts sample using an E.Z.N.A. Stool DNA Kit (Omega Biotek, USA). Next, the intestinal bacteria 16S rRNA gene was amplified using PCR based on total genomic DNA and specific primers (Forward primer: 341F (5'-CCTACGGGNGGCWGCAG-3'; Reverse primer: 805R (5'-GACTACHVGGGTATCTAATCC-3'). The target V3 and V4 regions of the bacterial 16S rRNA gene were acquired 60. The 25 µL PCR reaction mixtures containing 25 ng of total DNA, 12.5 µL of PCR Premix, 5 µL of primer, and PCR-grade water. The PCR conditions consisted of an initial denaturation at 98 °C for 30 seconds; 32 cycles of denaturation at 98 °C for 10 seconds, annealing at 54 °C for 30 seconds, and extension at 72 °C for 45 seconds; and then a final extension at 72 °C for 10 minutes. The purified and quantified PCR products was assessed on an Agilent 2100 Bioanalyzer (Agilent, USA) and with a Library Quantification Kit for Illumina (Kapa Biosciences, USA). The sequences of libraries were acquired by NovaSeq PE250 platform (Illumina, USA).
Diversity analysis and statistical analysis
Based on the sequences of libraries, we obtained feature tables and feature sequences according to the fqtrim software (v0.94), Vsearch software (v2.3.4) and DADA2. Alpha diversity was applied to analyze the complexity of species diversity for a sample using five indices, including Chao1, Observed species, Goods coverage, the index Shannon, and the Simpson index. All these indices were calculated using QIIME261. Beta diversity was also calculated using QIIME2, and the graphs of the analysis results by principal component analysis, principal coordinate analysis, Anosim, nonmetric multidimensional scaling and linear discriminant analysis effect size were drawn using the R package. BLAST was used for sequence alignment, and the feature sequences were annotated using the SILVA database for each representative sequence. Other diagrams were implemented using the R package (v3.5.2). In addition, we analyzed some data using charts created by by Microsoft Excel 2016 and using IBM SPSS Statistics v24 (IBM Corp., Armonk NY, USA). Certain images were constructed using GraphPad Prism 5 (GraphPad Inc., La Jolla, CA, USA) and edited using Adobe Illustrator CS6.