SRP311505 PRJNA715941 Differences in the gut microbiota of larvae and adults of a highly polyphagous insect developing in ancestral and exotic hosts We tested the gut microbial host use facilitation hypothesis, which proposes that variation among herbivores in their ability to consume chemically defended plants can be due, in part, to variation in their associated microbial communities. To reach this purpose, we studied the microbiota of a highly polyphagous insect, Anastrepha ludens, developing in six of its hosts, including two ancestral, three exotic and one occasional host. Criollo peach is rife with polyphenols and the occasional host with capsaicinoids exerting high fitness costs on the larvae, thus nicely setting the stage to test the GMFH. We surveyed the microbiota of the pulp of each host fruit, as well as the gut microbiota of larvae and adult flies and found that the gut of A. ludens larvae lacks a stable microbiota, since it the latter was invariably associated with the composition of the pulp-associated microbiota of the various host plant species studied and was also different from the microbiota of adult flies. The microbiota of adult males and females was similar between them, independent of host plant and was dominated by bacteria within the Enterobacteriaceae. We found that, in the case of C. pubescens, the microbiota is enriched in potentially deleterious on fitness host AM genera that were not identified in any of the other five hosts studied. In contrast, the pulp of the ancestral host C. edulis is enriched in several bacterial groups that can be beneficial for larval development. We also report for the first time the presence of bacteria within the Arcobacteraceae family in the gut microbiota of A. ludens stemming from C. edulis. Based on our findings related to C. pubescens, we propose that the pulp-associated DDmicrobiota likely contributes to defining the host range of the highly polyphagous A. ludens as it could limit the performance of larvae developing in this pulp. Similar scenarios can be predicted for other plant-insect interactions