Endosymbionts and honey bee colony losses?

Honey bees, Apis mellifera, are essential pollinators for the maintenance of natural biodiversity and agriculture [1]. Colony losses witnessed throughout the Northern hemisphere are therefore worrying [2], especially because no single driver has yet emerged as the definitive cause [3]. Interactions between viruses, ectoparasitic mites and microsporidian endoparasites are most likely key factors [3–5], but the underlying mechanisms are not well understood. Although it is known that maternally-inherited, facultative bacterial endosymbionts such as Wolbachia or Rickettsia can significantly interfere with viral and fungal infections of arthropods [6], they have so far been neglected in this regard. Here we propose to evaluate the potential role of such endosymbionts for colony losses. Endosymbionts are widespread [7] in arthropods and transmitted vertically [8], but can only spread in host populations when infected females have a higher fitness, e.g. via providing protection against viruses or fungi [6]. For example, Wolbachia can protect the host against several vectored RNA viruses [9] and can be regarded as part of host immunity [6]. However, endosymbionts such as Spiroplasma and Hamiltonella can also be beneficial for their host’s vectorial capacity, e.g. in the whitefly Bemisia tabaci–Tomato yellow leaf curl virus system, Hamiltonella protects viral particles in the vector [6,10]. To shed light on the potential influence of endosymbionts on losses, we here suggest an investigation of symbiontmediated host protection against viruses transmitted by parasitic mites and/or associated with microsporidians (e.g. Nosema ceranae) [5], which could contribute to the tolerance of honey bee populations, e.g. against the mite Varroa destructor [11]. Moreover, endosymbionts carried by parasitic mites might favour virus transmission to and virulence in honeybees, which could explain regional differences in the impact of mites [11]. A combination of metagenomics and laboratory experiments appears suitable to compare the bacterial and viral communities associated with honey bees and their parasites in host populations with or without elevated losses [2]. In conclusion, it seems as if endosymbionts play a role in honey bee pathology [12] and should therefore be investigated as a potential key to our understanding of major colony losses.