Gut microbiota structure differs between honey bees in winter and summer

Lucie Kešnerová; Olivier Emery; Michaël Troilo; Joanito Liberti; Berra Erkosar; Philipp Engel

an important characteristic of different bee types (e.g. foragers, nurses, winter bees), 87 season, or environmental exposure. An example is the experimental exposure of bees to 88 antibiotics which did not result in a strong shift in the relative composition, but in an overall 89 reduction of bacterial load, rendering bees more susceptible to pathogen invasion [11] . In 90 addition to the limitations of current studies using relative abundance data, almost nothing 91 is known about the gut microbiota of winter bees as compared to foragers or nurses. This 92 is surprising, as winter bees are critical for colony health and survival during the cold season 93 of the year, when resources are limited and most colony losses occur [33, 34] . 94 Characterizing the gut microbiota of winter bees and identifying factors that shape its 95 community composition may help to understand the physiological adaptations that honey 96 bees need to survive the cold season in temperate regions. In this study, we used qPCR and 97 16S rRNA gene amplicon sequencing to assess differences in the gut microbiota of nurses, 98 foragers, and winter bees. We analyzed bacterial loads of major community members in 99 566 individual worker bees sampled from a single hive over two years. We then expanded 100 our analysis to the entire community and analyzed pooled samples from 14 different hives 101 to test if similar community changes occur in winter bees across hives. Finally, we 102 performed experiments with gnotobiotic bees to test the influence of diet on differences 103 in gut microbiota composition. Our study reveals major differences in total bacterial load 104 and in the abundance of specific gut community members in the gut microbiota of nurses, 105 foragers, and winter bees and identifies dietary pollen as a major contributing factor. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/703512 doi: bioRxiv preprint Page 6 of 44 Over a period of two years, we sampled ~24 adult worker bees of A. mellifera each month 109 from a single hive located on the Dorigny campus of the University of Lausanne, 110 Switzerland. These bees were used to determine seasonal changes in the absolute 111 abundance of seven major community members of the honey bee gut microbiota using qPCR and to determine the overall community structure by 16S rRNA community analysis. 129 Bees were anesthetized with CO2 for 10 s and the gut including crop, midgut, hindgut, and Microbiota-depleted bees were generated and colonized as described in Kešnerová et al. 142 [6]. The treatment group of bees fed on pollen and sugar water was the same as in the 143 previous study. The other treatment group (bees fed on sugar water only) was carried out 144 in parallel with the same batch of bees. Bees were fed ad libitum with sterilized bee pollen 145 (P) and sterilized sugar water (SW, 50% sucrose, w/v) (SW+P treatment), or with only 146 sterilized sugar water (SW treatment). Bees were sampled 10 days after colonization and 147 the guts were dissected as described before. with the Nucleospin PCR Clean-up kit (Macherey-Nagel, Germany) according to the 154 manufacturer's instructions and the resulting DNA was used for qPCR. For the pooled gut 155 samples, 2 ml of glass beads and 15 ml of CTAB lysis buffer were added to each Falcon tube. 156 Samples were then homogenized in a Fast-Prep24 TM 5G homogenizer at 6 m/s for 40 s, 157 briefly centrifuged, and an aliquot