PROTEIN TRAFFIC ACROSS MEMBRANES 
GuNTER Blobel, M.D., PH.D., Investigator 
Research in Dr. Blobel's laboratory is focused on 
four protein translocation systems (translocons) 
that function in the translocation of cytoplasmically 
synthesized proteins across (or integration into) 1) 
the endoplasmic reticulum (ER), 2) the prokaryotic 
plasma membrane, 3) the chloroplast envelope, 
and 4) the two mitochondrial membranes. 
Each of the four translocons is likely to consist of 
at least four entities. 1) A cytosolic signal-recogni- 
tion factor that recognizes and binds to a signal se- 
quence, thereby segregating the signal sequence 
from the rest of the chain and preventing inactiva- 
tion of the signal sequence (inactivation might 
occur rapidly if the signal sequence were free to 
fold with the rest of the chain and be no longer ac- 
cessible on the surface of the folded preprotein). 
2) A "homing" receptor that is part of the mem- 
brane-bound components of a given translocon 
and functions as the cognate receptor for a given 
signal-recognition factor. After binding of signal-rec- 
ognition factor to its homing receptor, the signal- 
recognition factor is dissociated from the signal 
sequence. These steps target the signal sequence to 
the membrane and permit it to interact with a 
second membrane-bound signal-recognition sys- 
tem, which is part of a protein-conducting channel. 
3) A ligand-gated protein-conducting channel 
that would possess a cytosol-exposed signal se- 
quence-binding domain and that would open in 
response to ligand binding, i.e., binding of the 
signal sequence to this domain. The channel would 
close after completion of translocation and open 
again only after binding of another signal sequence. 
This channel (unlike ligand-gated ion channels) 
could also open in a second dimension, namely to 
the lipid bilayer, as a result of encountering a stop- 
transfer sequence. In this case the portion of the 
chain that is located in the channel would be 
extruded, through this opening, from the channel 
to the lipid bilayer, followed by closing of the chan- 
nel, in both dimensions. As a result a portion of the 
chain would be integrated into the lipid bilayer. 
Accessory proteins could function on either side of 
the channel, perhaps as protein antifolding factors. 
4) A signal peptidase removes the signal peptide, in 
many cases. This enzyme is part of the trans compo- 
nents of the translocon (the cis components are the 
cytosolic or cytosol-exposed components). 
Data were published from this laboratory in the 
past year on all four translocons. 
I. Endoplasmic Reticulum Translocons. 
A. £R translocon of canine pancreas. By fusing 
rough microsomes with plasma lipid bilayers, Dr. 
Sanford Simon, in collaboration with Dr. Joshua 
Zimmerberg (National Institutes of Health) has de- 
tected large aqueous channels in the ER membrane. 
These channels were mostly open at negative mem- 
brane potential at the cytoplasmic site of the mem- 
brane and closed at positive voltages. There was a 
dramatic increase in the number of open channels 
when 100 |jlM GTP was added, whereas 100 |jiM 
GTP"yS caused closing of channels. A similar channel 
was also detected when inverted vesicles derived 
from the plasma membrane ot Escherichiacoli were 
fused. Because both membranes share the ability to 
translocate secretory proteins, it is likely that this 
channel is the long-sought-after protein-conducting 
channel. 
In 1986, signal peptidase was isolated in Dr. 
Blobel's laboratory as a complex of at least five 
polypeptide chains. Drs. Gregory Shelness and 
Yashpal Kanwar have sequenced the cDNA for the 
glycoprotein subunit of the complex. The deduced 
amino acid sequence shows one classical trans- 
membrane segment and one glycosylation site. It is 
not similar to any other protein in the data bank. 
Dr. Christopher Nicchitta has used chemical al- 
kylation to distinguish the process of nascent chain 
targeting and signal sequence insertion from subse- 
quent chain translocation across the membrane. 
Translocation across the membrane is mediated by 
an A^-ethylmaleimide-sensitive membrane protein. 
This protein might be part of the protein-conduct- 
ing channel. 
A trimeric G protein of the a-, (3-, 7-subunit type 
was detected in the ER by Drs. Yves Audigier and 
Sanjay Nigam. The function of this G protein, 
which does not appear to be involved in protein 
translocation, remains to be investigated. 
B. ER translocon of yeast. In 1986 Dr. Blobel's lab- 
oratory, and independently two other laboratories, 
developed yeast cell-free systems that faithfully re- 
produce protein translocation across yeast micro- 
somes. Components of the yeast ER translocon 
have not been isolated. However, Drs. William 
Chirico and Gerald Waters succeeded in isolating 
two closely related 70 kDa heat-shock proteins 
(98% homologous) of yeast cytosol that are re- 
Continued 
27 
