Metagenomics of the human gut microbiota is a well-established field, and it is now apparent that this microbial community comprises hundreds of species, which are involved in many complex interactions. By contrast, the microbiota of the bee gut is simpler, with 95% of gut bacteria in adult worker honey bees represented by only eight phylotypes1. The commercial and environmental importance of bees is well-reported, but their value as a model for host–microbiota interactions is only now being realized using whole-genome sequencing.

Credit: NPG

The genome sequences of gut symbionts from various bee species were recently published2,3,4. Using the Illumina HiSeq2000 and PacBio RS, Kwong et al. sequenced three isolates of Gilliamella apicola and of Snodgrassella alvi to a depth of >500x coverage2. One type strain from each species was sequenced to completion and circularized, and the other four strains were sequenced to draft status. Both species are dominant members of the gut microbiota of honey bees and bumble bees, and annotation of the type strains provided evidence that each occupies a different, but complementary, metabolic niche. G. apicola is heavily dependent on carbohydrate metabolism, whereas S. alvi has lost this ability, and the metabolic networks of each species suggest that they might provide each other with nutrients. Intriguingly, this study also provided the first experimental evidence that S. alvi strains are only able to colonize bees from their native host genus and can outcompete non-native species even when they are initially at a tenfold numerical disadvantage. Comparative genomic analysis also revealed that little or no horizontal gene transfer (HGT) occurs between S. alvi and G. apicola strains from two different bee species, which is indicative of specific, long-term host–microbiota associations. However, HGT seems to occur between isolates that reside in the same host, which highlights the importance of interactions between gut symbionts.

Bees are social insects, and the transmission of symbionts between individuals that live in close proximity probably helps to maintain a similar microbiota.McFrederick et al. focused on the less well-studied halactid bee species Megalopta centralis and Megalopta genalis to determine whether social structure has a role in shaping their bacterial communities3. Both species are found in solitary and social nests, which enabled the role of social structure to be distinguished from that of host species. Using 454 pyrosequencing, a total of 12 nests of M. genalis (four social nests and eight solitary nests) and six nests of M. centralis (three social nests and three solitary nests) were sampled, and the results were collated with publicly available data. Data analysis showed no evidence of bacterial community clustering by social structure, although clustering by host species was strong, which is consistent with the findings of Kwong et al2.

In large-scale sequencing projects, there is a high risk of capturing contaminating sequences from unintended sources. While sequencing the bumble bee Bombus impatiens, Martinson et al. also sequenced a novel gammaproteobacterial member of the B. impatiens gut microbiota4. At just less than 2 Mb, the genome is at least 500 kb smaller than that of the G. apicola and S. alvi type strains, which is consistent with a highly restricted host range, owing to specific long-term co-evolution with its host. Indeed, the authors found that this new species, which is known as 'Candidatus Schmidhempelia bombi' and is a close relative of G. apicola, has not been found in any other insect so far.

These newly sequenced bee gut symbionts have revealed multiple genes that are candidates for determining host specificity. These include genes that mediate interactions with the host epithelium, such as those that encode type IVa pili in G. apicola and S. alvi, as well as type VI secretion system (T6SS) proteins in G. apicola and T1SS proteins in S. alvi. Proteins from T1SSs, T5SSs and T6SSs were also found in 'Ca. Schmidhempelia bombi'. Flp pili systems, which are involved in adhesion to the gut wall and biofilm formation, were also identified in some strains.

Given the simplicity of the bee gut microbiota and the widespread availability of affordable genome sequencing methods, it seems to be only a matter of time before we have a more comprehensive understanding of the function and community structure of the microbiota that is associated with these invertebrates. This should benefit many disciplines, from host–bacteria co-evolution and bacterial symbiosis to bee conservation.