Gut Bacterial Inhabitants of Open Nested Honey Bee, Apis Florea

Honey bees are complex social system, which are highly dynamic having close interactions with their surrounding environment. Gut microbiota of honey bees has a major role in interaction behavior with its environment and health. Apis orea is the primitive among all the honey bees and are indigenous to Indian subcontinent. The study reports the identication and analysis of bacteria in the gut of wild species of honey bee, Apis orea, by culture-based and culture-independent methods. Cultured bacteria were identied and characterized by MALDI-TOF MS and 16S rRNA sequencing. A comprehensive analysis and identication of non-culturable bacteria were performed by 16S rRNA amplicon next generation sequencing. The sequence analysis approach classied gut bacteria into 5 bacterial phyla, 8 families and 10 genera in major. The dominant bacterial taxa identied in Apis orea belonged to Prevotellaceae (52.1%), Enterobacteriaceae (42.7%) and Halobacteriaceae (1.3%). The dominant bacteria belonged to genera of Prevotella, Escherichia-Shigella, Natronomonas, Methylobacterium, Pantoea, Bidobacterium, Enterobacter, Klebsiella, Lactobacillus and Nitrobacter belonging to phyla Bacteroidetes, Proteobacteria, Euryarchaeota, Actinobacteria, and Firmicutes. Many of these bacteria identied herewith are not reported for their occurrence in others species of Apis genus making this study of highly relevance with respect to bee microbiome.


Introduction
Honey bees are complex social system, which are highly dynamic having close interactions with their surrounding environment making them superorganisms. Fossil records indicate that among all other living species of Apis, Apis orea is the most primitive. This is substantiated by their colony size, open nest and behavioral pattern (Kaspi and Sha r 2013; Radloff et al. 2005). The evolutionary divergence of Apis orea from the common ancestor is much earlier from the remaining Apis clades (Biewer et al. 2016). Apis orea is prevalent and restricted to south and southeastern part of Asia. It is commonly known as dwarf bee or little bee and survives temperature up to 50 °C (Biewer et al. 2016). The honey production by these little bees is less in quantity and often consumed by them hence, not reared for commercial purpose. Contribution of Apis orea towards pollination is agriculturally and ecologically signi cant rather than role in producing honey. Depending upon the forage availability, A. orea often migrates between plains and adjacent low hills during seasonal variations. The species prefer warmer climate for their foraging activity (Balachandra et al. 1999). A. orea species functions as an important pollinator and recent times have experienced a drastic decline in the population. Various reasons for extreme honey bee losses have been proposed, which include indiscriminate use of toxic pesticides, poor nourishment, genetic diversity, parasites and microbial pathogens (Anjum et al. 2018).
Accumulating evidence suggests a crucial role between the host-microbe interactions (Saraithong et al. 2015). Microbes contribute to functional capabilities to the host by providing essential nutrients and immunological defense to the honey bee health. The diversity and sociality affect an organism's physiological and behavioral adaptations.
Gut microbiome in the honey bee gut help to protect against attacking pathogens (Kwong and Moran 2016). Some studies showed that relative abundance of core gut species is having a direct impact on the susceptibility to various pathogens (Raymann and Moran 2018 For majority of the wild bee species, composition and function of microbiome is largely unknown and remains to be elucidated (Engel et al. 2016). The need of such study arises as the gut microbiome has key roles in host health which will extricate the relationship between host tness in both managed and wild pollinator bees (Engel et al. 2016

Culturing of bacteria
The bee gut samples (20) were homogenized using micropestle. Different dilutions (i.e., 1/10, 1/100 and 1/1000) of this composite homogenate were made and 100 µl aliquots each of the diluted sample were inoculated into six different media procured from HiMedia, Mumbai, India: Nutrient Agar (NA), MRS Agar (MRS), Brain Heart Infusion Agar (BHI), Eosin-methylene blue (EMB), Luria Bertani (LB) and Gluconobacter agar (GB) and incubated for 24-72 h at 30 ºC. The bacterial colonies grown on the plates were enumerated and selected, based on different morphologies. The separated colonies in master plates were repetitively sub-cultured to obtain pure colonies of bacteria.

MALDI-TOF MS Based Characterization
A simple extraction protocol was employed to analyze the bacterial sample. Loopful of bacterial cultures were mixed thoroughly with ethanol (70% v/v) and the suspended cells were centrifuged at 12,000 rpm for 5 min. The pellet was recovered by discarding the supernatant carefully. The pellet was air dried at room temperature and resuspended in formic acid (70% v/v) by vigorous mixing followed by the addition of acetonitrile. The mixture was centrifuged at 12,000 rpm to separate the pellet and 1 µl of clear supernatant was placed on a MALDI target plate. The bacterial smear was overlaid with 1 ml saturated solution of alpha-cyano-4-hydroxycinnamic acid (HCCA) matrix prepared in acetonitrile (50%) and tri uoroacetic acid (2.5%) and allowed to dry at room temperature. The extracted samples were analyzed using Auto ex speed system (Bruker Daltonik GmbH, Germany).Mass spectra were obtained in a mode of linear positive ion extraction at a laser frequency of 1000 Hz within a mass range from 2k to 20kDa. The ion source 1 voltage was 19.5 kV, ion source 2 voltage was maintained at 18.2 kV, lens voltage at 7 kV, and the extraction delay time was 240 ns. Calibration of spectra was done externally by using the standard calibration mixture (E.coli extracts including RNase A and myoglobin as additional proteins, Bruker Daltonics). The MALDI Biotyper software 3.0 (Bruker Daltonik) was used to identify the bacterial isolates and to visualize the mass spectra. Species-level identity has been considered for the isolates with biotyper score value ≥2.0, while the analysis for the isolates with score value ranging from 1.7 to 1.99 was repeated to achieve the higher score values (Kurli et al. 2018). The isolates which were not identi ed up to species level and with biotyper score value <2.0 were subjected for 16S rRNA sequencing.

Sequence analysis
The sequence data obtained were assembled and analyzed using DNA sequence assembling software Lasergene SeqMan Pro (DNASTAR Inc.). The 16S rRNA sequence of each bacterial isolate was compared using BLAST (Camacho et al. 2009) against 16S ribosomal RNA sequences (Bacteria and Archaea) database (a subdivision of GenBank).

Next generation sequencing
Community DNA extraction The honey bees were surface sterilized and, the alimentary tract of worker honey bees (60) were collected in 1.5 mL of phosphate buffered saline. To this 200 µL of cell lysis buffer (ATL buffer, Genomic DNA extraction, Qiagen Tissue kit) was added and subjected to homogenization in the presence of glass beads using the temperature controlled (~22°C ) vortex shaker for 30 min. Twenty microlitre of proteinase K was added and further steps followed were according to manufacturer's protocol. The concentration of DNA was measured using Qubit HS DNA kit (Invitrogen, USA) and stored at −20 °C till further processing.

Amplicon sequencing
To investigate bacterial diversity of honey bee gut, NGS library preparation was carried out by targeting V4 region of

Bioinformatics analysis
The raw reads obtained from the high-throughput sequencing were analyzed using QIIME1 pipeline ( 2010). The taxonomic assignment was performed using SILVA (v132) raw taxonomy. Alpha diversity indices were assessed by richness like Chao1, ACE and diversity (Shannon) were calculated via QIIME, which predisposes sample rarefaction to the same sequencing depth (Bokulich 2012).

Results And Discussion
The global decline in the population of pollinator bees (Potts et al. 2010) has attracted the researchers for a comprehensive study of host microbiome community. As major contributors in pollination, honey bees are very crucial organisms in securing the agricultural produce and the maintaining ecosystem. Unlike western bees, A. orae are not reared for commercial honey or wax; rather they occur in wild functioning as major pollinators (Balachandra et al. 1999). These little dwarf bees differ from western bees in their defense against pathogens (Suwannapong et al. 2011). In the present study, Asiatic bees, A. orea were selected to study and analyze the gut microbiome. The bee symbionts are likely to play a vital role in self defense and metabolism. Apis orea, (dwarf bee) is naturally distributed in Indian subcontinent throughout south-east Asia. The most important contribution of this honeybee is its valuable pollination of many fruit plants and diversi ed ora in tropical ecosystems (Soman and Chawda 1996).
The defensive and metabolic capabilities of bees are highly correlated with gut microbiome interaction and to further add knowledge on role and interaction of microbial community with host, the analysis of gut bacteria of Apis orea honey bees was carried out. Healthy bees were obtained directly from single hive across Western Ghats forest area of Kodagu district, which is recognised as a global biodiversity hotspot, India. The entire thorax and abdomen was processed for analysis, thus including gut microbes and organisms attached to hemolymph or tissues. The analysis of gut bacteria was performed by culture dependent and culture-independent technique from a total of 80 worker dwarf honey bees.
In culture dependent method, the identi cation and characterization of culturable diversity of bee gut bacteria was done by MALDI-TOF-MS and 16S rRNA gene sequencing analysis. A total of 91 aerobic and facultative anaerobic bacteria were isolated from guts of worker A. orae bees. Based on colony characteristics, the bacterial isolates were initially subjected to MALDI-TOF-MS. Fifty six isolates were identi ed up to species level. The isolates with biotyper score value <2.0 could not be identi ed up to species level. Remaining thirty-ve such isolates which were not identi ed up to species level from MALDI-TOF-MS were further subjected to 16S rRNA sequencing. Collectively from both MALDI-TOF-MS and 16S rRNA sequencing of culture dependent analysis, the gut bacterial isolates belonging to three bacterial phyla/classes were identi ed; alpha-Proteobacteria (1%), Firmicutes (19%) and Gamma-Proteobacteria (80%) (Fig.1). The percentage of isolates belonging to genera, Klebsiella, Enterobacter, Bacillus, Citrobacter, Staphylococcus and Lactobacillus are represented in Fig. 2. These isolates were found to be common in Apis clade as Strains of Bacillus safensis is reported to produce many industrially relevant enzymes, such as amylase, protease, lipase, inulinase and chitinase. The occurrence of Brevundimonas nasdae was prominent in our study. The pH of honey bee midgut is around 8 and this pH favors the optimal growth of Bacillus species and Brevundimonas nasdae and these may aid in degrading of carbohydrate fed by the bees. In addition to Brevundimonas nasdae (Phylum: Proteobacteria), Bacillus species and, Solibacillus silvestris (Phylum: Firmicutes) were predominant in the gut of A. orea. These bacterial species are rarely or no where reported in the available literature on the Apis gut microbiota of other species, whereas interestingly occur in Apis mellifera. The differences in the occurrence of these bacteria might be due to geographical location or existing characteristics of the environment or may be due to the feature of A. orea species itself, which requires further studies.
Prevotella, Escherichia-Shigella, Natronomonas, Methylobacterium, Pantoea, Bi dobacterium, Enterobacter, Klebsiella and Lactobacillus are the dominant genera found in the gut of A. orea worker bees. The bacterial genera Prevotella, Natronomonas etc. are uncultivable under typical laboratory conditions as they require strict anaerobic conditions or haloalkaliphilic conditions. Prevotella, Natronomonas and Methylobacterium are not reported in the available literature of honey bee gut microbiota, however, Methylobacterium and Prevotella are the predominant inhabitants of gut in bark beetle, (Dendroctonus rhizophagus) (Briones-Roblero et al. 2017). Species of Prevotella, are non-cellulolytic carbohydrate degrading bacteria, which bring about digestion of cell wall polysaccharides like xylan (Flint et al.2012). A relatively lower proportion of Brevundimonas, Staphylococcus, Streptococcus, Gluconobacter and Gilliamella genera was observed in our culture independent studies, whereas many of these are predominant in Apis mellifera (Kwong and Moran 2016). Functional redundancy and crosstalk among the microbes, and host has huge metabolic and physiological impact and, the signi cance of the presence of these bacterial communities can be untangled by further metagenomic and metatranscriptomic studies.
High throughput sequencing and quality trimming of 16S rRNA gene yielded ~0.118 million quality reads which were used for subsequent analysis. Taxonomic assignment of sequences with the reference database resulted in 589 operational taxonomic units (OTUs). Alpha diversity estimation of gut of A. orea using species richness and nonparametric Shannon index suggested higher bacterial diversity in A. orea worker bees. The alpha diversity index which is an indicator of bacterial diversity, were calculated for A. orea and is given in Table 1. Shannon index for bacterial communities was 3.121 and this observation is suggestive of richness in bacterial diversity in the gut of A. orea.
16S rRNA gene sequences of the bacterial isolates of A. orea was used to construct the phylogenetic tree showing relationship among the bacteria with reference strains of GenBank (Fig. 5). The bacterial populations in the gut were diverse among forager bees of A. orea which belonged to phyla Firmicutes, alpha and beta-Proteobacteria. A plethora of bacterial abundance in any niche suggests their signi cance in ecological diversity; in A. orea, Prevotella was a signi cant member accounting for 59% of the total gut microbe indicating the possible ecological importance.  (Kwong and Moran 2012). In our study, Snodgrassella alvi was not detected whereas, Gilliamella apicola were found in traces. Acinetobacter was found in our study similar to the reports of Kim et al (2014). Acinetobacter apis spp. nov., was isolated from the intestinal tract of a honey bee, A. mellifera (Kim et al. 2014). In the metagenomic survey, the class Alpha Proteobacteria and Gamma Proteobacteria dominated the gut environment of A. mellifera (Engel and Moran 2013a

Conclusion
In conclusion, the gut communities of A. orea are more diverse in composition and the sequence analysis approach classi ed gut bacteria into Bacteroidetes, Proteobacteria, Euryarchaeota, Actinobacteria, and Firmicutes as the major phyla. Most of the isolates found are opportunistic and bene cial gut inhabitants. Prevalence of members of genera Prevotella (59%) is observed in our study which is different in abundance against other species of Apis clade. Sequencing the whole gut using NGS has allowed us to analyze the importance of microbiome role in A. orea host to some extent. Likewise, with other microbiome project (human, animal and insects), this work provides additional information and data pertaining to the microbiome of A. orea, a primitive clade of honey bee, revealing some of the bacterial cobionts not found in earlier reports on honey bee microbiome project. However, further genomic studies are required to study the relevance for diversi ed microbial occurrence and a comparative analysis of managed and wild species may offer insight about host-microbe interaction.
Declarations Table   Table.1. Alpha diversity estimation. Non-parametric alpha diversity was calculated for A. orea gut bacteria.

Chao1
Observed OTUs Shannon ACE  Next generation sequencing of culture independent gut bacterial ora of A. orea. Taxa distribution of genus at different phylogenetic level of honey bee gut bacterial ora Figure 5 Molecular Phylogenetic analysis by Maximum Likelihood method The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model (Tamura and Nei, 1993). The tree with the highest log likelihood (-5509.8000) is shown. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. The analysis involved 39 nucleotide