A tubulovesicular network in Gemmata

Damn it, aren't those bacteria surprising? The report by Acehan et al., in JCS 2014 reveals the presence of a network made of membrane tubules and vesicles in the periplasm of the planctomycetes Gemmata obscuriglobus. This is very surprising by itself, but even more because this tubulovesicular network (TVN) seems to connect the outer membrane to the inner one, so the cytoplasm is directly connected to the outside. How is that possible? In addition, this TVN is in close contact with the membrane coat proteins previously identified in the proteome of this organism based on structural similarities with proteins like clathrin or some nucleoporins. This is most likely directly related to the phenomenon of protein internalisation and degradation previously reported by Lonhienne et al., in PNAS 2010 in the same organism. All together, the presence of a TVN in contact with membrane coat proteins and related to endocytosis is pretty close to one of the possibilities of what is expected to have been present in the ancestral eukaryote.
All this is nicely covered in a JCS 'in this issue' highlight.