ER swaps C-terminal signal for membrane tether: attaches
protein (loosely) to membrane.
ER adds glycosylation: glucose ‘fuse’ marks
proteins as misfolded.
ER contains chaperonins and PDI: ensures proteins are correctly
folded, crosslinked, and do not aggregate.
ER dislocates broken proteins via BiP: ensures terminally
misfolded proteins are destroyed.
Golgi modifies glycosylation: prepares protein for ECM, by
making it more acidic, hydrophilic and protease-resistant, and also
able to bind appropriate receptors.
Golgi phosphorylates mannose: required for lysosomal
transport.
Small hydrophobic molecules may enter the nucleus by diffusion.
Proteins enter via the nuclear pores in the Ran cycle: these must have
a suitable signal peptide such as PPKKKRKV.
Both mitochondria and chloroplast use TIM/TOM-like systems, with an
amphipathic N-terminal signal. However, plastids require
further signals for lumenal import, and the sequences must be
distinguishable.
Both the nucleus and peroxisomes import their proteins pre-folded.
However, peroxisomes import them through relatively small proteins
complexes, whilst the nucleus has very large and complex pores.
Nuclear-directed proteins (including the nuclear pore proteins
themselves) must be able to relocate into the nucleus after mitosis,
because they are released into the cytoplasm during the breakdown of
the nuclear envelope during prometaphase. Likewise the KDEL sequence
for RER retention: proteins are continually re-trafficked from the CGN
to the RER. However, the RER signal can be cleaved because proteins
won’t be re-imported.
Having the RER import signal at the amino terminus allows proteins
to be imported into the RER cisternae as they are translated: this
saves energy, because the physical force of translation can be used to
push the protein through. Note that all other import mechanisms require
either GTP, ATP or a proton motive force.
The peroxisome is highly oxidising. Isomery of disulfide bonds
would be difficult under such circumstances, so perhaps this is why
proteins have to be imported prefolded and fully crosslinked.
Vesicular trafficking could be demonstrated using GFP fusion
proteins. Cisternal maturation could be demonstrated by showing that
the Golgi traffics molecules too big to fit in a vesicle.
Cells and their endosymbionts have plenty of machinery for
translocating proteins, but the only machinery for translocating RNAs
across membranes is the nuclear pore, which would be a rather more
difficult system to subvert for use by endosymbionts.