also in agreement with the study from Steinberg and colleagues with respect to effects on the trafficking of retromer cargo proteins such as Glut1 but their study was focused on the effect of increased Rab7a activity on mitophagy – the autophagic clearance of mitochondria – where they showed that loss of TBC1D5 could enhance Rab7a-dependent mitophagy. Our study and that of Steinberg and colleagues do differ with respect to how loss of TBC1D5 affects the endosomal localisation of the retromer CSC as they do not observe an increase in endosomal VPS35 staining in TBC1D5 knockout (KO) cells. This difference could be due to their use of knockout cells versus our knockdown cells because a genetic KO can be compensated for through adaptations in the cells used for the KO. As KOs take days to generate along with weeks for generating clonal cell lines, there is scope for genetic adaptation to occur in a KO that would be less likely to occur in an RNAi-mediated knockdown over a period of 72 h. For example, expression of other GAPs such as TBC1D15 or Armus may increase to mitigate some of the effects of TBC1D5 KO, but may not do so in the
