by
Abstract
Participants at a recent meeting in France witnessed the scientific method in action. The prevailing theory of protein transport within cells has been tested and found wanting in many ways. Biologists at the meeting learned how the model is being reconstructed to suit the latest evidence - and where evidence is still lacking for parts of it.
How do secretory proteins get from their site of synthesis in the endoplasmic reticulum (ER) to the outside world? In quest of answers, researchers gathered in Giens, France, from September 26 to October 1, 1998, at a meeting cosponsored by the European Science Foundation and the European Molecular Biology Organization. At the conference, "Membrane Dynamics in Exocytosis: Molecular Mechanisms," the consensus was that the main dogma, the forward vesicular transport model, had been or was about to be scrapped.
The forward vesicular transport model had predicted that newly synthesized proteins, after they exit from the endoplasmic reticulum via molecular containers known as COP II vesicles, move from compartment to compartment (cisterna to cisterna) via transport containers called COP I vesicles. Each cisterna was thought to be preexistent and equipped with its own unique set of enzymes. This would allow for stepwise assembly and maturation of oligosaccharides, as well as other modifications of the newly synthesized proteins. Another prediction was that the proteins in transit moved in and out of these compartments unrestricted. This "bulk flow" hypothesis stated that unless they were endowed with positive sorting signals, proteins would travel the pathway by default.
This model was a useful framework for identifying key molecules in the pathway, and for formulating hypotheses about their functions. Unfortunately, it failed to live up to its predictions. For example, the idea of preexistent cisternae had to give way to a more dynamic view, allowing for the possibility that each compartment is transient, and capable of forming de novo.
The assumption that each compartment has its own set of modifying enzymes has turned out to be wrong. Most, if not all, resident proteins of the pathway show gradient-like distributions, occupying not one but two or more cisternae. In addition, COP I, the key component of the vesicle thought to bring newly synthesized proteins forward, is involved instead in recycling proteins that are residents of the pathway, not commodities for transport.
The observation that newly synthesized proteins leaving the ER do not do so unrestricted has seriously undermined the "bulk flow" concept. Instead, the proteins actively concentrate into COP II vesicles. These vesicles bud, uncoat, and fuse with each other, forming an intermediate membrane carrier that brings cargo forward from the ER to the Golgi apparatus.
These shortcomings in the old forward vesicular model of protein transport have impelled a switch to more progressive concepts, such as the cisternal maturation model. A realization among the participants of the failings of the previous model has prompted an attitude more open to new ones.
Newly synthesized proteins bound to exit the cell pass first into the ER, where they undergo service and quality control. Here they receive N-linked oligosaccharides, fold, and oligomerize, assisted by a series of proteins that "chaperone" the nascent polypeptide into a more mature state, one that is ready to leave the ER. In addition, some chaperons exert a "quality control," ensuring that misfolded proteins do not leave the ER but instead are diverted for degradation. This probably explains the role of what were identified originally as "heat shock" or "stress" proteins. Crucial factors for protein folding are temperature, oxidative levels, and pH. Elevated temperature causes increased misfolding, and the cell responds by producing more chaperons.
Peter Walter of the University of California at San Francisco described an elegantly simple system to explain how the cell mobilizes more chaperons in response to stress or change. Sensing change, a transmembrane kinase resident in the ER signals to a transcription factor known as HAC1p, which induces transcription of response genes that encode products such as folding chaperons. How this signal/activation is transmitted has remained an open question until now. Surprisingly, the transcript for HAC1p is present as a pre-mRNA, and is acted upon directly by the membrane-bound kinase serving as a nuclease. In much the same way that pre-tRNA is subjected to splicing, the kinase catalyzes the removal of a 250nt intron from the HAC1 pre-mRNA to produce an active HAC1 mRNA. So far, no one knows whether HAC1p performs this function in the nucleus (the nuclear membrane is continous with the ER) or regulates splicing in the cytoplasm.
Chris Kaiser of the Massachusetts Institute of Technology showed how ERO-1, a 96 kDa protein also resident in the ER, is required for the formation of disulfide bridges. In fact, ERO-1 is crucial for disulfide formation, to stabilize the folding of intermediates as well as that of the final product, and precedes that of protein disulfide isomerise (PDI). Its overexpression caused increased resistance to exogenous agents which would arrest disulfide bond formation and therefore increased folding. However, how ERO-1 actually works is not clear. It may form a disulfide linkage using its well-conserved cysteins, then donate this directly to the newly synthesized protein. Alternatively, ERO-1 may donate the disulfide linkage to other chaperons, which in turn donate this to the folding protein.
In the separate work of Bill Balch of the Scripps Research Institute and Randy Schekman of the University of California at Berkeley, we now have a glimpse of an active complex of machinery for the segregation and concentration of proteins bound for export. Taken together, their findings deal a potentially fatal blow to the concept of "bulk flow," which had postulated that proteins destined for export enter into transport vehicles unrestricted.
Balch summarized his group's work on a di-acidic cytoplasmic motif that enhances the export of a viral glycoprotein called VSV-G, comparing it with mutants that had been altered in ways that have allowed him to reconstruct a detailed image of events that lead up to VSV-G's entrance into and concentration in COP II vesicles.
Schekman's parallel experiments suggest that COP II components sample
proteins actively, interacting directly with their cytoplasmic domains.
This sampling causes those proteins that are recognized to segregate
themselves away from the others, to concentrate and finally become
incorporated into the COP II vesicles. The author of this text presented
evidence showing that COP I also samples the membrane, in this case
searching for proteins resident in the Golgi complex. This activity seems
to depend on hydrolysis of GTP by a small GTPase.
Several of the speakers explored candidate cargo-receptor molecules in the ER. Hans-Peter Hauri and Howard Riezman, both of the University of Basel, outlined their arguments for p53/58 (a lectin-like transmembrane molecule) and the p24 family (small, abundant transmembrane proteins). The activities of both show some correlation with export of cargo from the ER. In the case of p53/58, this function seems restricted to a few types of cargo molecules, and the same can be said for p24 proteins. Therefore, it's unlikely they serve as general cargo receptors. It is not really clear that such receptors are necessary in the first place. Whatever their roles, the p24 proteins are highly conserved from yeast to mammals, suggesting that they carry out some essential function. Jean Gruenberg of the University of Geneva proposed, based on the behavior of one p24 protein (p23) when overexpressed, that the role of p24s may be in providing structural stability to membranes between the ER and the Golgi apparatus.
Rainer Pepperkok of the European Molecular Biology Laboratory (EMBL) spoke about the microtubule highway running from the ER to the Golgi apparatus and beyond. Intermediate cisternal carriers formed by the fusion of COP II and COP I vesicles deliver newly synthesized proteins from the 100 or so peripheral sites to the central and juxtanuclear Golgi complex. These intermediates are COP I coated, and move along microtubules with the help of dynein and dynactin. The possibility was raised that during transit, the intermediate would not only exchange coat but also composition in terms of resident proteins. A continuous counter-current of retrograde COP I vesicles would be the likely mechanism to allow for this. These vesicles would also be traveling along microtubules, but with the help of kinesins, to take them away from the center and back to the periphery.
The meeting's poster session was the site of several debates. One
concerned the existence of transport vesicles
leaving the trans-Golgi network for the plasma membrane. It is easy to
detect large structures moving toward the periphery of the cell on
microtubuli tracts, but smaller and faster moving objects also appear in
abundance. These are not readily picked up by conventional imaging
techniques such as confocal microscopy. Sensitive high-speed video imaging,
however, does reveal fast-moving but faint vesicles leaving the TGN. Not
until these are compared to the larger, slower-moving structures will we be
able to determine their relative contribution to the actual transport
event.
Jennifer Lippincott-Schwartz of the National Institutes of Health highlighted the usefulness of computer-aided modeling to elucidate the complex dynamics of transport through the entire pathway. She showed how simple first order kinetics reasonably fit the observed ER-to-Golgi to plasma membrane (PM) transport. Although preliminary, the results underscore the potential power of computer modeling for cell function studies.
As more and more evidence accumulates to show extensive intracellular transport, the issue of detection becomes crucial. Bruno Goud of the Curie Institute and Jamie White of EMBL spoke about the small GTPase rab6. Goud's group found that rab proteins actively attract a set of energy-generating complexes of molecules known as motors. In this case, rab6 attracts an enzyme known as rab6 kinesin, allowing for plus-ended motility (movement along microtubules from their so-called "minus" ends to their "plus" ends). For the first time, this discovery has clearly provided an important role for rab proteins. Moreover, it opens a new field in cell biology, providing the long-sought link between membranes and the cytoskeleton. In the opinion of this author, it is likely that rab proteins, strategically placed all around the cell, attract motor proteins to mediate motility and positioning of proteins destined for transport. This is not far-fetched, as there are enough motors to match the large number of rab proteins present in the cell. It will be a puzzle for future studies to figure out how rab proteins are regulated, along with their corresponding motors.
It is clear from several studies that microtubules are not needed to maintain the Golgi stack, nor for directed transport through the secretory pathway. Under these conditions, the juxtanuclear collection of Golgi stacks disassemble and rebuild themselves into functional but peripheral stacks close to ER exit sites. This ability to rebuild in the absence of microtubules suggests the existence of a first level of organization. Sorting and directed transport fall within this. Higher eukaryotes have imposed a second level of organization on top of this by collecting all stacks into a central juxtanuclear Golgi apparatus. Presumably, this allows for sorting of proteins after transport in polarized cells. This second level requires both microtubules and motor proteins and, one assumes, other cytoskeletal components. In this respect, rab proteins are poised to be the key components in regulating and coordinating the two levels.
How does the complex Golgi apparatus reconstruct itself after cell division? Using live cell imaging, Graham Warren of the Imperial Cancer Research Fund in London touched upon this questions. He showed how the Golgi apparatus segregates with the help of the mitotic spindle into the two daughter cells, followed by a breakdown and dispersal of the Golgi elements, which subsequently re-coalesce at several peripheral points. One issue of debate has been whether proteins normally resident in the Golgi enter the ER during mitosis; the existence of such a pathway is supported by the notion that in interphase cells, there is a slow but significant routing of Golgi resident glycosylation enzymes into the ER. The extent of this pathway in interphase cells could not account for the rapid dispersal of Golgi enzymes seen during mitosis. It is possible, though, that a small but significant fraction of the enzymes are trapped in the ER during mitosis, and that these play a role in the subsequent assembly of Golgi stacks at the ER exit sites."
The buildup of peripheral Golgi stacks following mitosis or microtubuli depolymerization shows that proteins resident in the Golgi can arrange themselves and their membranes into functional Golgi stacks. It is clear that Golgi components have the capacity for self-organization. Whether or not this means the Golgi can form de novo is unclear. To show this, one would need to demonstrate that the Golgi apparatus can form itself from newly synthesized proteins. This may turn out to be difficult to prove, which could prevent some researchers from taking the full step away from the idea of preexistent compartments and being willing to ponder the secretory pathway in a temporal rather than spatial sense.
In summary, a more fitting title for the meeting might be "The Renaissance of the Exocytic Pathway." Many old ideas/dogmas are being thrown out at the moment. For anyone looking to join this field, the time could not be better. All in all, the meeting featured a variety of great talks and poster presentations that depicted new and intersting aspects of the secretory pathway. In light of this, it was truly sad that Thomas Kreis was not there to see and be part of the meeting he helped to organize, making it the success it was.
Tommy Nilsson is a group leader at the European Molecular Biology Laboratory in Heidelberg.
Caleb Brown is an illustrator and biologist living in Montana. By day he drives a delivery van, and by night he draws pictures with his computer.
Dictionary of Cell Biology - employs a graphical interface and search engine to access more than 6,500 entries. From the Institute of Biomedical and Life Sciences at the University of Glasgow.
AAA Protein Superfamily - a database maintained by Kai-Uwe Fröhlich at the University of Tuebingen. Includes the NSF and SEC 18 proteins involved in secretion.
Neurotransmission Machinery Visualized for the First Time - a Howard Hughes Medical Institute news article summarizing a recent structural study of synaptic vesicle fusion.
Syntaxin 1A - the Flybase entry for a component of the membrane fusion complex. A concise overview.
ER to Golgi Traffic Visualized in Living Cells - includes movies related to Jennifer Lippincott-Schwartz's recent publication in Nature.
Web sites mentioned in this column: