Comparison of Gut Microbiota in Overwintering Bees: Apis cerana vs. Apis mellifera

Abstract

Bees play important roles in socio-economic development, biodiversity conservation, and ecosystem stability. However, during the cold season, resources become limited, leading to significant losses in bee colonies. Although many studies have described the characteristics of winter bees and demonstrated that notable changes occur in their gut microflora, the underlying mechanisms remain yet to be fully elucidated. Therefore, this study was conducted to compare the gut microbiota dynamics of overwintering bees. Sample acquisition involved randomly selecting ten colonies each from three bee farms containing Apis cerana (AC) and Apis mellifera (AM), followed by dissection for further analysis. DNA was extracted, and 16S rDNA sequencing, along with various bioinformatics tools, was used to assess microbial diversity, functional differences, and species comparisons between AC and AM gut microbiota. AC exhibited lower β diversity in the gut microbiota than AM during winter. Moreover, Gilliamella and Apibacter were relatively more abundant in AC. Regarding microbial functions, key pathways included the phosphotransferase system, galactose metabolism, the pentose phosphate pathway, and carbohydrate transport and metabolism. These results suggest the presence of microbial diversity differences between AC and AM, with the differential microbial functions mainly enriched in metabolic pathways that facilitate adaptation to cold environmental stress.

1. Introduction

Pollinators are vital for socioeconomic development, biodiversity conservation, and ecosystem stability [1]. In particular, bees pollinate over 75% of vegetable crops and plants for seed production [2], with Apis cerana (AC) and Apis mellifera (AM) being the most commonly used species. During the cold season, resources become limited, and the overwintering period leads to high colony losses [3]. Winter bees survive the cold winter by forming a “winter cluster” in the hive and generating heat through strong vibrations of their flight muscles [4,5]. The health of these winter bees is crucial to the survival of the whole colony and determines their ability to reproduce successfully in the following spring [6]. During winter, AC demonstrates superior cold resistance and foraging abilities, rendering them the main honey bee species for honey collection and plant pollination during this period [7]. At outside temperatures below 14 °C, the number of foraging bees in AC colonies is significantly higher than that in AM colonies, additionally, the cooling point of AC is significantly lower than that of AM, indicating that AC exhibits stronger cold resistance [8].

Research on the mechanisms underlying cold resistance in honey bees includes the screening of genes associated with this trait, which has revealed that the p38 MAPK gene plays an important physiological role in honey bees during the overwintering period. During winter, juvenile hormone levels decrease, whereas yolk protein levels increase [9]. In addition, trehalose concentration in winter wasps has been demonstrated to be 1.9 times higher than that during summer, with the concentration of all amino acids except proline and alanine showing a downward trend [10]. Moreover, total protein, glucose, glycogen, and lipid levels are higher in winter bees [11,12]. AC is actively involved in pathways related to sugar and amino acid biosynthesis and metabolism, as well as calcium channel activity, in response to cold stress [8].

Although many studies have described the characteristics of winter bees, the underlying mechanisms remain yet to be fully elucidated. Increasing evidence suggests that the gut microbiome of bees significantly affects their central metabolism, which can improve health and longevity [13,14,15,16]. Honeybees harbor relatively simple but crucial lineages of five core gut bacteria—Gilliamella, Snodgrassella, Lactobacillus Firm 4, Lactobacillus Firm 5, and Bifidobacterium—along with some non-core bacteria, such as Frischella, Commensalibacter, and Bartonella. These microbes serve as valuable biological models for exploring microbiota functions [15,16,17,18]. Previous studies have demonstrated that seasonal shifts in temperate bee communities lead to notable changes in gut microbiota [19,20], with winter bees exhibiting the highest bacterial load and lowest community alpha diversity, characterized by increased levels of Bartonella and Commensalibacter [19]. Bartonella may play a crucial role in supplying essential amino acids to bees during winter, facilitating health maintenance and the synthesis of proteins for their larvae [6]. Therefore, seasonal changes in microbiota may improve the adaptability of bees.

AC and AM exhibit significant differences in cold resistance during the overwintering period, and changes occur in their gut microbiota during this period [19]. Investigating the differences in the gut microbiota of AC and AM during winter may elucidate the relationship between the gut microbiota and the seasonal physiological adaptation of bees in temperate regions. In this study, we employed 16S rRNA gene amplicon and metagenomic sequencing to assess the differences in the gut microbiota of AC and AM during winter, aiming to establish a connection between differential microbes and cold tolerance. The findings of this study may provide important evidence for improving the survival of bees during winter.

2. Materials and Methods

2.1. Sample Acquisition

Three bee farms with both AC and AM were selected, with ten colonies randomly selected from each bee farm (five AC and AM colonies each). From each colony, 25 worker bees were captured for dissection and further examination. The bees were soaked in ethanol and stored at −20 °C until dissection. During the dissection process, the abdomen was disinfected with an alcohol soaked cotton ball, and sterile scissors were used to cut open the abdomen, allowing access to the inner tissues for the removal of the midgut and hindgut. To eliminate differences between individual bees, the guts of bees from each of the five bees colonies per species were concentrated in an EP tube, with five samples for each colony providing five samples. These tissues were frozen in liquid nitrogen and stored at −80 °C. Sample collection was completed in January 2022, i.e., in the middle of the winter.

2.2. DNA Extraction

Each colony was regarded as a biological sample, from which five intestines were collected and placed into a 1.5 mL sterile centrifuge tube. Microbial DNA was extracted using the HiPure Soil DNA Kits (or HiPure Stool DNA Kits) (Magen, Guangzhou, China) following the manufacturer’s protocols. After obtaining the total genomic DNA from the intestinal bacterial samples, the V3 + V4 region of 16S rDNA was amplified using specific primers with barcodes. The primer sequences used were 341F: CCTACGGGNGGCWGCAG and 806R: GGACTACHVGGGTATCTAAT. The purified amplification products were connected to a sequencing connector, and a sequencing library was constructed before sequencing was performed on an Illumina platform.

2.3. Bioinformatics Analysis

FASTP (version 0.18.0) was used to filter the original Illumina platform data, cluster the sequences, and remove chimeras, ultimately yielding representative sequences for operational taxonomic units (OTUs) along with their abundance information. The representative OTU or amplicon sequence variant (ASV) sequences were classified into organisms using a naïve Bayesian model with the RDP classifier (version 2.2) based on the SILVA database (version 138.1), applying a confidence threshold value of 0.8. Diversity indices, including Chao1, ACE, Shannon, Simpson, Good’s coverage, and Pielou’s evenness indices, were calculated using QIIME (version 1.9.1) [8]. The phylogenetic diversity whole-tree index was calculated using Picante software (version 1.8.2). Comparisons of alpha diversity indices between groups were conducted using Welch’s t-test and the Wilcoxon rank test within the Vegan package in R (version 2.5.3). Principal component analysis was also performed using the Vegan package. Multivariate statistical techniques, including principal coordinates analysis (PCoA) and non-metric multi-dimensional scaling based on (un)weighted UniFrac, Jaccard, and Bray–Curtis distances, were generated using the Vegan package and plotted using the ggplot2 package in R (version 2.2.1). Venn analysis between groups was performed using the VennDiagram package in R (version 1.6.16), whereas UpSet plots were generated using the UpSetR package (version 1.3.3) to identify unique and common species, OTUs, or ASVs. Species comparisons between groups were performed using Welch’s t-test and the Wilcoxon rank test in the Vegan package. Species comparisons among groups were analyzed using Tukey’s HSD test and the Kruskal–Wallis H test in the Vegan package.

Several complementary approaches were used to annotate the assembled sequences. The unigenes were annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG) and the evolutionary genealogy of genes: non-supervised orthologous groups (eggNOG). Based on gene abundance, the functional abundance for each type was calculated, and a functional abundance histogram was generated using the ggplot2 package. Welch’s t-test was performed to assess differences in species/functional means between the two groups using the Vegan package.

3. Results

3.1. Structural Diversity of the Gut Microbiota

To investigate the differences in the structural diversity of the intestinal flora between overwintering bees AC and AM, we used the Chao, Simpson, and Shannon indices to evaluate the richness, evenness, and diversity of samples, respectively (Figure 1A–C). Our analysis of the alpha diversity of the intestinal microbes revealed no significant differences between AC and AM. These results indicate the absence of differences in the gut microbial diversity between AC and AM during the overwintering period.

PCoA was performed based on the Bray–Curtis data results, followed by a weighted UniFrac distance analysis of similarities (ANOSIM) grouping test. The PCoA revealed significant differences in the microbial communities between the two sample groups, as indicated by the distance between AC and AM along the PCoA axes (Figure 1D). In addition, the intestinal flora of AC was assessed using weighted Uni-Frac distance (ANOSIM R = 0.3614, p = 0.001 (Figure 1E)) which revealed significant differences compared to AM. These results further suggest that the β diversity of AC is different from that of AM, highlighting the presence of significant differences in the structural diversity of the intestinal flora between AC and AM during the wintering period.

3.2. Composition of the Gut Microbiota

The horizontal abundance analysis of phyla showed that Proteobacteria and Bacteroidetes were more dominant in the intestinal tract of bees during the overwintering period (Figure S1). Based on the top 10 species abundance ranking, the taxonomic analysis showed that Lactobacillus, Gilliamella, Bifidobacterium, Commensalibacter, and Apibacter were the dominant genera in AC, whereas Lactobacillus, Commensalibacter, Gilliamella, Bifidobacterium, and Hafnia-Obesumbacterium were predominant in AM (Figure 2A and Figure S1).

The Wilcoxon rank-sum test was performed to further compare the differences in gut microbes between the two groups at the genus level. The results indicated significant differences in four genera: Gilliamella, Commensalibacter, Apibacter, and Frischella. Gilliamella and Apibacter were more abundant in AC, whereas Commensalibacter and Frischella were significantly more abundant in AM (Figure 2B). To determine the differential enrichment of specific bacterial taxa between AC and AM, we used linear discriminant analysis effect size analysis with a logarithmic linear discriminant analysis score cutoff of 2.5. This analysis identified four discriminant genera, namely Gilliamella, Commensalibacter, Apibacter, and Frischella, as key discriminant factors, consistent with the Wilcoxon rank-sum test results. A cladistic map illustrating the taxonomic hierarchy of gut microbes from phylum to species also showed significant differences in the phylogenetic distribution of microbiota between AC and AM (Figure 2C,D).

3.3. Metagenomic Analysis Revealed Differential Functional Features

To further explore the different functional characteristics of intestinal microorganisms in AM and AC during the overwintering period, samples were randomly selected from Yangzhou and Rongchang, with six samples selected from each location (three samples were randomly selected from AM and three from AC). PCoA revealed a significant separation between the AM and AC samples, with each group clustering in specific areas (Figure S2).

The metagenomic data were annotated using the KEGG and eggNOG databases to explore the differences in functional characteristics and metabolic pathways between the two bee species during the overwintering period. At level 1 of the KEGG database, metabolism related genes were the most abundant in the gut microbiota of the two bee species, followed by genes associated with environmental and genetic information processing (Figure S3). Subsequently, the distribution of functional genes related to metabolism at level 2 of the KEGG pathways was analyzed. The top five pathways identified in AC and AM were global and overview maps, carbohydrate metabolism, membrane transport, amino acid metabolism, and metabolism of carbohydrate cofactors and vitamins (Figure S4). To identify the KEGG pathways with significant differences in abundance between AC and AM, the top 10 pathways with the highest relative abundance were analyzed based on a significance level of p < 0.05, as illustrated in the figure. The analysis revealed that the phosphotransferase system (PTS), galactose metabolism, and pentose phosphate pathway were significantly enriched in AC compared to those in AM (Figure 3A).

Welch’s t-test was used to compare the functional differences in eggNOG annotations between the two groups at each function level, with results considered significant at a p-value threshold of <0.05. According to the annotated results from the eggNOG database, 10 significantly different functions were identified between AC and AM: G: carbohydrate transport and metabolism; L: replication, recombination, and repair; C: energy production and conversion; O: post-translational modification, protein turnover, J: translation, ribosomal structure and biogenesis; P: inorganic ion transport and metabolism; chaperones; E: amino acid transport and metabolism; D: cell cycle control, cell division, chromosome partitioning; U: intracellular trafficking, secretion, and vesicular transport; Q: secondary metabolite biosynthesis, transport and catabolism. However, AC was only significantly enriched for the carbohydrate transport and metabolism function (Figure 3B and Figure S5).

4. Discussion

Winter in temperate regions poses significant challenges for bees, with winter colony losses often exceeding 20% in many areas [21,22]. Sufficient, successful overwintering colonies are crucial for beekeeping to meet the pollination and production demands in the following spring [23]. During the annual management of honey bees, the intestinal flora of overwintering bees was demonstrated to exhibit the largest bacterial load and the lowest colony α diversity [19,20]. However, the correlation between these factors and the cold resistance of overwintering bees has not been reported. Given the differences in cold tolerance between AC and AM, modern high throughput 16S rDNA and metagenomic sequencing techniques were used to analyze the association between honey bee gut microbiota and cold tolerance in this study.

To eliminate differences caused by geographical distance, we selected AC and AM raised in the same apiary across three regions. Gut microbiota is strongly influenced by the host diet, and our results demonstrated that AC or AM from different geographical locations clustered together [24,25]. The 16S rRNA gene sequencing results indicated no differences in the intestinal microbial diversity between AC and AM during winter, consistent with the results of previous studies [20]. Honeybees possess a relatively simple microbiome, and the core microbiome of AC is essentially the same [26]. As social insects, worker bees with different social roles also have highly consistent microbiomes [27,28]. In addition, studies have demonstrated that worker bees from different hives, states, countries, and subspecies exhibit highly consistent microbiomes [29,30,31]. However, significant differences were observed in the diversity of the intestinal flora structures between AC and AM.

Gilliamella and Apibacter were significantly enriched in AC compared to those in AM. Notably, the winter associated microorganism Bartonella, identified in previous studies [19], was not detected in our samples, and only a small amount was found in AM. The gut microbiota in bees is mainly concentrated in the hindgut, which is divided into two regions: the ileum and rectum [27]. Gilliamella is mainly distributed in the ileum [32], whereas Apibacter is distributed in the midgut, ileum, and rectum and is more abundant in the rectum [33]. Apibacter spp., belonging to the phylum Bacteroidetes, are ubiquitous in AC [34], Apis dorsata, and bumblebee species (genus Bombus) but are rarely observed in AM [15,34]. Notably, in our study, the 16S rRNA results indicated the absence of Apibacter in AM. In bumblebees, Apibacter is associated with reduced infection of the trypanosome gut parasite Crithidia bombi and is considered a beneficial symbiont [35]. Apibacter possesses metabolic pathways for glucose, fructose, and mannose degradation. It can utilize microaerobic respiration and fermentation to break down a limited range of monosaccharides and dicarboxylic acids but is unlikely to participate in the digestion of complex polysaccharides [34,36,37]. In AC and Apis dorsata, which are widely distributed in South and East Asia, over 80% of adult worker bees are colonized by Apibacter [15]. Notably, Apibacter has acquired genes encoding for nitrate respiration (NAR) [33], which has been demonstrated to benefit the colonization of the gut microbiome in bees [38]. The NAR pathway may provide Apibacter with specific metabolic advantages and promote host specific adaptation to the environment. Gilliamella, classified under Proteobacteria and Gammaproteobacteria, is one of the most important sugar decomposing and fermenting bacteria in the bee gut [39,40]. As a facultative anaerobe, Gilliamella lacks many genes for the tricarboxylic acid cycle and electron transport chain; however, it possesses a complete glycolysis pathway in its genome, as well as numerous genes encoding the PTS. This process allows Gilliamella to produce ATP and biosynthetic precursors directly from carbohydrate catabolism via glycolysis and the pentose phosphate pathway [41,42]. In addition, metagenomic analysis revealed that Gilliamella contains genes for pectin degrading enzymes, which may facilitate the breaking down of the rigid polysaccharide walls of pollen grains and the release of the constituent monosaccharides. This function is crucial for regulating the intestinal nutrition of honey bees, which ultimately affects their overall health [43,44]. The increased abundance of Gilliamella may represent a physiological adaptation necessary for bees to survive during the cold season.

Metagenomic sequencing enables a more comprehensive analysis of the relationship between microbial function and host physiology. The results of the KEGG functional analysis revealed that the functional differences between the gut microbiota of AC and AM were primarily associated with variations in metabolic pathways. Specifically, the PTS, galactose metabolism, and pentose phosphate pathways were significantly enriched in AC. The PTS is an intermembrane transport system that specializes in the transport of sugars and derivatives [45]. Gilliamella can synthesize multiple variants of the PTS, allowing it to ingest a variety of sugars and derivatives [46]. Rapid energy metabolism plays an important role in determining the lifestyle of an organism, particularly in bees, which obtain carbohydrates from nectar and amino acids from pollen. These nutrients are subsequently degraded and fermented by microbial enzymes, resulting in the production of short chain fatty acids that can be used by the host [47]. Sugar consuming insects generate energy (ATP) required for nearly all physiological processes by diverting dietary sugars to the pentose phosphate pathway [48]. During the cold winter months, bees form colonies to produce heat, leading to an increased demand for ATP. The eggNOG analysis results revealed that the enriched function in AC was carbohydrate transport and metabolism. Honey bees use sugar almost exclusively as a flight substrate, and their hemolymph contains the highest total sugar content among all insect species. This high sugar concentration is crucial for maintaining an adequate fuel supply [49]. AC exhibits higher cold tolerance, and the significant enrichment of the carbohydrate transport and metabolism pathway indicates that its microbial community has a strong capability to metabolize these carbon rich food sources [50], thereby generating a greater energy supply.

5. Conclusions

Differences were observed in the composition and function of the gut microbiota of AC and AM during the wintering period. Multiple metabolic pathways were found to be associated with the gut bacteria, with Gilliamella identified as a potential functional microorganism that improves the cold tolerance of honey bees. Moreover, several metabolic pathways were significantly enriched in AC, including the PTS, galactose metabolism, and pentose phosphate pathways. Although the composition and function of the gut microbiota of AC and AM during the wintering period were analyzed based on 16S rRNA and metagenomic sequencing, absolute quantitative references regarding the microbial composition structure were absent. Overall, our analysis revealed characteristic differences in the gut microbiota of AC and AM during winter, which may significantly affect their cold tolerance. Therefore, future studies will focus on exploring the functional roles of individual bacterial species in winter bees and their implications for overall bee colony health.