Autophagy has both tumor-promoting and -suppressing effects in cancer, including colorectal cancer (CRC), with transformed cells often exhibiting high autophagic flux. In established tumors, autophagy inhibition can lead to opposite responses resulting in either tumor cell death or hyperproliferation. The functional mechanisms underlying these differences are poorly understood. The present study aimed to investigate the relationship between the autophagic capacities of CRC cells and their sensitivities to autophagy inhibition. All studied CRC cell lines showed high basal autophagic flux. However, only HCT116 and Caco-2/15 cells displayed regulated autophagic flux upon starvation. Knockdown of ATG5 (which disrupts autophagosome elongation) or RAB21 (which decreases autophagosome/lysosome fusion) had little effect on CRC cell proliferation in vitro . Nonetheless, inhibition of autophagy in vivo had a substantial cell line-dependent impact on tumor growth, with some cells displaying decreased (HCT116 and Caco-2/15) or increased (SW480 and LoVo) proliferation. RNA sequencing and Western blot analyses in hyperproliferative SW480 tumors revealed that the mTORC2 and AKT pathways were hyperactivated following autophagy impairment. Inhibition of either mTOR or AKT activities rescued the observed hyperproliferation in autophagy-inhibited SW480 and reduced tumor growth. These results highlight that autophagy inhibition can lead, in specific cellular contexts, to compensatory mechanisms promoting tumor growth.

Aedes aegypti develop in aquatic habitats in which mosquito larvae are exposed to physicochemical elements and microorganisms that may influence their life cycle and their ability to transmit arboviruses. Little is known about the natural bacterial communities associated with A. aegypti or their relation to the biotic and abiotic characteristics of their aquatic habitats. We characterized the physicochemical properties and bacterial microbiota of A. aegypti breeding sites and larvae on Guadeloupe and in French Guiana. In addition, we explored whether geographic location, the type of breeding site and physicochemical parameters influenced the microbiota associated with this mosquito species. We used large-scale 16S rRNA gene sequencing of 160 breeding sites and 147 pools of A. aegypti larvae and recorded 12 physicochemical parameters at the sampled breeding sites. Ordination plots and multiple linear regression were used to assess the influence of environmental factors on the bacterial microbiota of water and larvae. We found territory-specific differences in physicochemical properties (dissolved oxygen, conductivity) and the composition of bacterial communities in A. aegypti breeding sites that influenced the relative abundance of several bacteria genera (e.g., Methylobacterium, Roseoccocus) on the corresponding larvae. A significant fraction of the bacterial Electronic supplementary material The online version of this article (

Studies have demonstrated the importance of breeding site, few had disentangled the role of microbiome, physico-chemical and biological factors of water as well as landuse on larval microbial communities and their recruitment in mosquito. A quantitative exploration of the interplay of multiple factors on mosquito microbiome was performed using a dataset obtained through a field survey undertaken in French Guiana. Two complementary hypotheses were tested (i) the most dissimilar larval microbiome structures in breeding sites displayed the most contrasting water properties and land-use, (ii) a higher specificity level of environmental parameters have an incidence on larval microbiome. Variance partitioning approach validated the two hypothesis by providing evidence that water bacterial community is a most significant driver shaping the structure of the bacteriome in mosquito than other environmental parameters from the breeding sites. However, land-use does not play such important role to explain variance. Our results consolidate and complement the knowledge shaping mosquito microbiota but also highlighted the large unknown in understanding the ecology of the recruitment into host.