Chapter_13_Solutions

Chapter_13_Solutions - Chapter 13 Intracellular Vesicular...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Chapter 13 Intracellular Vesicular Traffic THE MOLECULAR MECHANISMS OF MEMBRANE TRANSPORT AND THE MAINTENANCE OF COMPARTMENT DIVERSITY DEFINITIONS 13–1 13–2 13–3 13–4 13–5 13–6 13–7 13–8 13–9 13–10 13–11 13–12 Transport vesicle Rab protein Adaptor protein Dynamin NSF Clathrin-coated vesicle Arf protein Rab effector SNARE protein (SNARE) Retromer Coated vesicle Sar1 protein 13 In This Chapter THE MOLECULAR A281 MECHANISMS OF MEMBRANE TRANSPORT AND THE MAINTENANCE OF COMPARTMENT DIVERSITY TRANSPORT FROM THE A288 ER THROUGH THE GOLGI APPARATUS TRANSPORT FROM THE A294 TRANS GOLGI NETWORK TO LYSOSOMES TRANSPORT INTO THE A298 CELL FROM THE PLASMA MEMBRANE: ENDOCYTOSIS TRANSPORT FROM THE A303 TRANS GOLGI NETWORK TO THE CELL EXTERIOR: EXOCYTOSIS TRUE/FALSE 13–13 True. The cytosolic leaflets of the two membrane bilayers are the first to come into contact and fuse, followed by the noncytosolic leaflets. It is this pattern of leaflet fusion that maintains the topology of membrane proteins, so that protein domains that face the cytosol always do so, regardless of what compartment they occupy. False. Although the organelle-specific distribution of Rab proteins made such a hypothesis attractive at one time, it is now clear that the Rab proteins do not bind to complementary Rab proteins. Instead, Rab proteins bind to specific Rab effectors to accomplish the docking of appropriate vesicles to the target membrane. 13–14 THOUGHT PROBLEMS 13–15 If the flow of membrane between cellular compartments were not balanced in a nondividing liver cell, some compartments would grow in size and others would shrink (in the absence of new membrane synthesis.) Keeping all A281 A282 Chapter 13: Intracellular Vesicular Traffic Figure 13–25 Identity of compartments and pathways involved in the biosynthetic–secretory, endocytic, and retrieval pathways (Answer 13–17). Biosynthetic–secretory pathways are indicated with black arrows; endocytic pathways are indicated with thick gray arrows; retrieval pathways are indicated with light gray arrows. CYTOSOL lysosome late endosome nuclear envelope endoplasmic reticulum early endosome CGN TGN EXTRACELLULAR SPACE plasma membrane cisternae cis medial trans Golgi apparatus secretory vesicle the membrane compartments the same relative size is essential for proper functioning of a liver cell. The situation is different in a growing cell such as a gut epithelial cell. Over the course of a single cell cycle, all of the compartments must double in size to generate two daughter cells. Thus, there will be an imbalance in favor of the outward flow, which will be supported by new membrane synthesis equal to the sum total of all the cell’s membrane. 13–16 Coat proteins serve two functions: they concentrate specific membrane proteins, and they bend the membrane into a sphere. 13–17 A. The intracellular compartments involved in the biosynthetic–secretory pathway and endocytic pathway are labeled in Figure 13–25. B. The biosynthetic–secretory pathway (black arrows), endocytic pathway (thick gray arrows), and retrieval pathway (light gray arrows) are shown in Figure 13–25. 13–18 This is an apt analogy. Cargo receptors can be incorporated into a coated vesicle only if they can bind to adaptor proteins—‘have a ticket for a particular ride’—which allows them to enter a coated pit and be incorporated into a vesicle—‘the cable car.’ Cargo receptors—‘like the skiers’—are mixed together without any guaranteed traveling companions, but all get to the next compartment—‘station.’ Reference: Pearse BMF, Smith CJ & Owen DJ (2000) Clathrin coat construction in endocytosis. Curr. Opin. Struct. Biol. 10, 220–228. 13–19 The position of one triskelion is shown in Figure 13–26A. A triskelion must be flexible at its vertex to be able to accommodate different sizes of coated vesicles (Figure 13–26B). As the size of the coat increases, its radius of curvature decreases requiring individual triskelions to flatten out slightly. To accommodate the different angles required to fit into a pentagon and a hexagon, a triskelion needs to be flexible at its ‘knees’ (Figure 13–26B). References: Kirchhausen T (2000) Clathrin. Annu. Rev. Biochem. 69, 699–727. Ybe JA, Brodsky FM, Hofmann K, Lin K, Liu SH, Chen L, Earnest TN, Fletterick RJ & Hwang RK (1999) Clathrin self-assembly is mediated by a tandemly repeated superhelix. Nature 399, 371–375. 13–20 The specificity for both the transport pathway and the transported cargo come not from the clathrin coat, but from the adaptor proteins that link the clathrin to the transmembrane receptors for specific cargo proteins. The clathrin-coated vesicle triskelion (A) (B) knee vertex Figure 13–26 Formation of a clathrin coat (Answer 13–19) (A) The location of a single triskelion in a coated vesicle. (B) The sites of maximum flexibility of a triskelion. THE MOLECULAR MECHANISMS OF MEMBRANE TRANSPORT several varieties of adaptor proteins allow different cargo receptors, hence different cargo proteins, to be transported along specific transport pathways. Incidentally, humans are different from most other organisms in that they have two heavy-chain genes. Like other mammals, they also have two lightchain genes. In addition, in the neurons of mammals the light chain transcripts are alternatively spliced. Thus, there exists the potential in humans for additional complexity of clathrin coats; the functional consequences of this potential variability are not clear. References: Kirchhausen T (2000) Clathrin. Annu. Rev. Biochem. 69, 699–727. Pearse BMF, Smith CJ & Owen DJ (2000) Clathrin coat construction in endocytosis. Curr. Opin. Struct. Biol. 10, 220–228. 13–21 A. Clathrin-coated vesicles cannot assemble in the absence of adaptor proteins, which link the clathrin to the membrane. At high clathrin concentrations, and under the appropriate ionic conditions, clathrin cages assemble in solution, but they are empty shells, lacking membranes and other proteins. Self-assembly of clathrin into baskets shows that the information for clathrin baskets is contained in the clathrin molecules themselves. B. Without clathrin, adaptor proteins still bind to receptors in the membrane, but no clathrin coat can form. Thus, no clathrin-coated pits or vesicles will form. C. In the absence of dynamin, clathrin-coated pits can form and proceed toward vesicle formation, but the last step—membrane fusion—is blocked without dynamin. As a result, deeply invaginated coated pits will be observed in the absence of dynamin. D. Procaryotic cells do not perform endocytosis. A procaryotic cell does not contain any receptors with appropriate cytosolic tails that could mediate the binding of adaptor proteins. Therefore, no clathrin-coated vesicles will form. 13–22 Assuming that the altered dynamin functioned normally in all ways except for the ability to hydrolyze GTP, this result supports the action of dynamin as a mechanochemical pinchase, perhaps acting as shown in Figure 13–27. In the absence of GTP hydrolysis, the mutant dynamin would not be expected to complete the final step in vesicle formation since that step requires GTP hydrolysis to power it. According to the alternative hypothesis, the inability of dynamin to hydrolyze GTP would have locked the mutant dynamin into its active ‘ON’ state, which, if anything, might have been expected to increase vesicle formation. Thus, the observed inhibition of vesicle formation argues against this hypothesis. Reference: Marks B, Stowell MHB, Vallis Y, Mills IG, Gibson A, Hopkins CR & McMahon HT (2001) GTPase activity of dynamin and resulting conformation changes are essential for endocytosis. Nature 410, 231–235. 13–23 Most of the statements in this description are correct. The GAP, however, is not located in the Golgi membrane. The COPII subunits—probably in association with another protein—act as the GAP, stimulating hydrolysis of A283 clathrincoated pit clathrincoated vesicle dynamin plasma membrane Figure 13–27 One model for how dynamin might pinch a vesicle off the membrane (Answer 13–22). In response to GTP hydrolysis dynamin, in this model, uncoils like a spring under tension and pops the vesicle from the membrane. A284 Chapter 13: Intracellular Vesicular Traffic GTP by Sar1. Thus, the GAP is associated with the vesicle membrane, not the Golgi membrane. Sar1–GDP then causes disassembly of the coat almost as soon as the vesicle has formed. The uncoated vesicle carries other proteins—a Rab protein and a vSNARE—that orchestrate docking and fusion with the Golgi membrane. 13–24 If Arf1 were mutated so that it could not hydrolyze GTP, Arf1 would exist in a cell as Arf1–GTP. Since Arf1–GTP promotes assembly of COPI-coated vesicles, you would expect such vesicles to form readily, but they might not form at the right place in the cell. Normally, Arf1 is delivered specifically to the Golgi membrane by a Golgi-bound Arf1–GEF, which converts the cytosolic Arf1–GDP to Arf1–GTP, exposing a fatty acid tail that allows it to bind to the membrane. The mutant Arf1, which would always have GTP bound and its fatty acid tail exposed, might bind inappropriately to other cell membranes, thus promoting COPI-coated vesicle formation at inappropriate places in a cell. Disassembly of the COPI coat requires hydrolysis of GTP by Arf1. Thus, in the presence of the mutant form of Arf1, the COPI coat would not be able to disassemble. Since the uncoated vesicle is the substrate for the fusion reaction with the target membrane, the mutant Arf1–GTP would be expected to block Arf1-mediated transport. If the mutant Arf1 were the only form of Arf1 in the cell, it is likely that it would prove lethal. Under these conditions, all Arf1-mediated transport involving COPI-coated vesicles should be blocked. Furthermore, there should be an accumulation of COPI-coated vesicles that cannot be uncoated, which might reduce the availability of free COPI subunits necessary for other transport pathways. What is unclear is the extent to which other members of the Arf family of proteins might substitute for Arf1. It would be necessary to do the experiment to know the result for certain. There will always be some v-SNAREs in the target membrane. Immediately after fusion, the v-SNAREs will be in inactive complexes with t-SNAREs. Once NSF pries the complexes apart, v-SNAREs may be kept inactive by binding to inhibitory proteins. Accumulation of v-SNAREs in the target membrane beyond some minimal population is thought to be prevented by active retrieval pathways that incorporate v-SNAREs into vesicles for redelivery to the original donor membrane. 13–25 13–26 A. The problem is that vesicles having two different kinds of v-SNAREs in their membranes could dock, in principle, on either of two different target membranes. B. The answer to this puzzle is not presently known, but it seems likely that cells have ways of turning the docking ability of SNAREs on and off. Regulation of docking might be achieved, for example, through other proteins that are copackaged with SNAREs into ER transport vesicles and that facilitate the interactions of the correct kind of v-SNARE with its partner t-SNAREs in the cis Golgi network. 13–27 The cell’s SNAREs are all bound to the cytosolic surface of whatever membrane they are in. They function by juxtaposing the cytosolic surfaces of the two membranes to be fused. By contrast, enveloped viruses must fuse with a cell membrane by bringing together its external surface with an external surface of a cell membrane. Thus, enveloped viruses cannot make use of a cell’s SNAREs because they are located on the wrong side of the membrane. It is for this reason that enveloped viruses make their own fusion proteins, which are properly situated on their external surface. CALCULATIONS 13–28 The volume of a cylinder 1.5 nm in diameter and 1.5 nm in height is 2.65 nm3 [3.14 ¥ (0.75 nm)2 ¥ 1.5 nm], which equals 2.65 ¥ 10–24 L [2.65 nm3 ¥ (cm/107 THE MOLECULAR MECHANISMS OF MEMBRANE TRANSPORT nm)3 ¥ (L/103 cm3)]. There are about 88 water molecules in this volume. water molecules = 2.65 10–24 L ¥ 55.5 mole ¥ 6 ¥ 1023 molecules L cylinder cylinder mole = 88.2 In each monolayer in a circle of membrane 1.5 nm in diameter, there are about 9 phospholipids [3.14 ¥ (0.75 nm)2 ¥ (PL/0.2 nm2) = 8.8 PL]. Thus, there are about 5 water molecules per phospholipid in the area of close approach of the two membranes. This number is slightly less that half the number (10–12) estimated to be associated with phospholipid head groups under normal circumstances. This means that when a vesicle and its target membrane are drawn together in preparation for fusion, somewhat more than half of the water molecules that would normally bind to the membranes must be squeezed out. Reference: Meuse CW, Krueger S, Majkrzak CF, Dura JA, Fu J, Connor JT & Plant AL (1998) Hybrid bilayer membranes in air and water: infrared spectroscopy and neutron reflectivity studies. Biophys. J. 74, 1388–1398. A285 DATA HANDLING 13–29 These observations indicate that brefeldin A blocks COPI-coated vesicle formation by interfering with the exchange of GTP for GDP, which is essential for Arf to bind to the Golgi membrane and initiate formation of coated vesicles. Brefeldin A does not affect Arf-GTP-mediated formation of COPIcoated vesicles. Observation 1 shows that brefeldin A does not block coated vesicle formation if Arf is first locked into its active form by GTPgS. Thus, the assembly of the COPI coats and formation of vesicles are not affected by brefeldin A, if the active membrane-bound form of Arf is present. Observation 2 shows that a protein in the Golgi membrane—a GEF—catalyzes the exchange of GTP for GDP. This exchange reaction is blocked by brefeldin A. Observation 2, by itself, does not distinguish whether the effect of brefeldin A is on the GEF, or on Arf: binding of brefeldin A to either protein could interfere with the exchange reaction. In fact, recent experiments show that brefeldin A binds to the complex of Arf and GEF, locking Arf into a nonproductive GDP-bound conformation. References: Donaldson JG, Finazzi D & Klausner RD (1992) Brefeldin A inhibits Golgi membrane-catalyzed exchange of guanine nucleotide into Arf protein. Nature 360, 350–352. Helms JB & Rothman JE (1992) Inhibition by brefeldin A of a Golgi membrane enzyme that catalyzes exchange of guanine nucleotide bound to Arf. Nature 360, 352–354. Chardin P & McCormick F (1999) Brefeldin A: the advantage of being uncompetitive. Cell 97, 153–155. 13–30 These results indicate that all three components—Arf1, Arf1–GAP, and COPI subunits—are required for efficient hydrolysis of GTP. Based on the effects of other GAPs on small GTPases, it is tempting to speculate that COPI subunits bind to the complex of Arf1 and Arf1–GAP and provide the arginine ‘finger’ that seems to be critical for GTP hydrolysis. It would be very informative to have a crystal structure of the assembly; however, COPI is too complex to make this a simple task. You might test your conclusion in other ways. If COPI subunits provide the catalytic arginine, then site-directed mutagenesis of arginines in COPI might identify one that is essential for GTP hydrolysis. A286 Chapter 13: Intracellular Vesicular Traffic Reference: Goldberg J (1999) Structural and functional analysis of the Arf1–ArfGAP complex reveals a role for coatomer in GTP hydrolysis. Cell 96, 893–902. 13–31 To generate maximal alkaline phosphatase activity, vesicles from each strain must carry both v-SNAREs and t-SNAREs (see Figure 13–6B, experiment 1). If either vesicle is lacking v-SNAREs or t-SNAREs, phosphatase activity is reduced to 30–60% of the maximum (see experiments 3, 4, 6, 7, 8, and 9). If both vesicles are missing either v-SNAREs (see experiment 2) or t-SNAREs (see experiment 5), phosphatase activity is very low, as it is if one vesicle is missing both SNAREs (see experiments 10 and 11). For a reasonable level of fusion, complementary SNAREs must be present on the vesicles. It does not matter which kind of SNARE is on vesicles from strain A so long as vesicles from strain B carry a complementary SNARE (compare experiments 3 and 4, experiments 6 and 7, and experiments 8 and 9). You might have wondered why there is a low background of phosphatase activity, even where no fusion is expected (see experiments 2, 5, 10, and 11). If a few vesicles were to break, releasing small amounts of pro-Pase and protease, then a small amount of active alkaline phosphatase could be generated in the absence of vesicle fusion. Reference: Nichols BJ, Undermann C, Pelham HRB, Wickner WT & Haas A (1997) Homotypic vacuolar fusion mediated by t- and v-SNAREs. Nature 387, 199–202. 13–32 A. As designed, this experiment does not allow you to detect synaphin binding to either Snap25 or synaptobrevin in the absence of syntaxin. Because antibodies against syntaxin were used in the detection step, only syntaxin was detected. Antibodies against Snap25, or against synaptobrevin, could have been used to detect such binding in mixtures that did not contain syntaxin. B. Binding of synaphin to syntaxin in the absence of Snap25 and synaptobrevin is weak, giving a barely detectable band at the highest concentration of syntaxin that was used (1.5 mM). Binding is improved very slightly in the presence of synaptobrevin alone, but much more in the presence of Snap25 alone. The best binding occurs when syntaxin, Snap25, and synaptobrevin are all present. This result suggests that synaphin binds best to the form of syntaxin that exists in the complete SNARE complex. C. The results in Figure 13–7 suggest that the natural target for synaphin in nerve terminals is the complete SNARE complex. Since this complex of two t-SNAREs and a v-SNARE only occurs when a synaptic vesicle is docked at the plasma membrane, the results suggest that synaphin could be involved in the next step; that is, fusion of the vesicle with the membrane. Reference: Tokumaru H, Umayahara K, Pellegrini LL, Ishizuka T, Saisu H, Betz H, Augustine GJ & Abe T (2001) SNARE complex oligomerization by synaphin/complexin is essential for synaptic vesicle exocytosis. Cell 104, 421–432. 13–33 A. The larger complexes (at apparent sizes of 100 kd, 120 kd, and 190 kd) are oligomers of the SNARE complex that are formed by the binding of synaphin, which presumably links the complexes together. The apparent sizes are not exact multiples of the 60-kd SNARE complex because the gels are run under nondenaturing conditions, where rates of migration are especially sensitive to shape. B. The synaphin-derived peptide prevents oligomerization of SNARE complexes by competing with synaphin for binding to syntaxin. If the synaphin binding site on syntaxin is already occupied by the peptide, synaphin will be unable to bind. This is analogous to the use of a competitive inhibitor of an enzyme to block its activity. C. The ability of the peptide to interfere with synaptic transmission in the squid giant axon means that the binding of synaphin to syntaxin is critical THE MOLECULAR MECHANISMS OF MEMBRANE TRANSPORT for synaptic vesicle fusion. This is a key experiment because it tells you that synaphin plays an important biological role in synaptic vesicle fusion. In any biochemical experiment such as that depicted in Figure 13–8, it is a concern that the interactions observed in the test tube may not reflect the true situation in the cell. The observation in squid axons indicates that the biochemical results are truly relevant to the cellular process of synaptic fusion. (The squid giant axon is a favored tool for studies of nerve function because its large size simplifies injection experiments of the type done here.) D. The biochemical results indicate that synaphin promotes oligomerization of SNARE complexes, and the injection studies indicate that the binding of synaphin to syntaxin is essential for vesicle fusion. Taken together, these results suggest that oligomerization of SNARE complexes may be a critical step in synaptic vesicle fusion. By forming oligomers of SNARE complexes— perhaps even a circle around the site of fusion—synaphin may bring the vesicle membrane and the plasma membrane into close enough proximity that they can fuse. These experiments do not rule out the possibility that additional proteins are required for fusion after the synaphin-promoted oligomerization of the SNARE complexes. Reference: Tokumaru H, Umayahara K, Pellegrini LL, Ishizuka T, Saisu H, Betz H, Augustine GJ & Abe T (2001) SNARE complex oligomerization by synaphin/complexin is essential for synaptic vesicle exocytosis. Cell 104, 421–432. 13–34 Syntaxin and Snap25 are t-SNAREs, and synaptobrevin is a v-SNARE. NSF and its accessory proteins recognize complexes of t- and v-SNAREs, and use the energy of ATP hydrolysis to pry them apart. Binding of NSF and its accessory proteins to a SNARE complex depends on the presence of ATP. In the absence of ATP the complex does not form; thus, the beads did not bring down anything other than NSF. In the presence of ATP the complex forms, but NSF then hydrolyzes ATP to separate the SNAREs and release them. Thus, in the presence of ATP, only NSF is attached to the beads. In the presence of ATPgS, the complex can form, but it cannot dissociate because NSF cannot hydrolyze ATPgS. As a result, all members of the complex remain attached to NSF and are brought down by the beads. Reference: Söllner T, Whiteheart SW, Brunner M, Erdjument-Bromage H, Geromanos S, Tompst P & Rothman JE (1993) SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318–324. 13–35 A. Since vesicles form and accumulate when the function of Sec4 is impaired (as in Sec4ts and Sec4N133I), Sec4 cannot be involved in vesicle formation. Accumulation of vesicles in these mutants suggests that the vesicles can no longer deliver their cargo to the growing bud when Sec4 is not working properly. Thus Sec4 seems to be involved in vesicle targeting and fusion. Functionally, Sec4 resembles mammalian Rab proteins, which are also required for proper delivery of transport vesicles to their target membrane. Indeed, Sec4 was the first identified member of the Rab family of proteins. Sec4 is unlike mammalian Sar1 and Arf proteins, which are required for formation of coated vesicles. B. From the description of the defects in the presence of the mutant Sec4 proteins, and by analogy to Rab proteins, it is possible to outline the way normal Sec4 functions in delivery of vesicles to the bud membrane (Figure 13–28). The presence of some Sec4 (20% of total) in the cytosol of wild-type cells represents Sec4 that is recycling after delivery of vesicles to the bud membrane. Removal of the C-terminal cysteines prevents attachment of a lipid that is essential for the binding of Sec4 to the forming vesicle. If Sec4 does not bind to the vesicle, it cannot carry out its function. C. The inhibitory properties of Sec4N133I are very interesting and not altogether easy to interpret. Since Rab proteins function as monomers, it is A287 A288 Chapter 13: Intracellular Vesicular Traffic transport vesicle GTP PO4 GAP Figure 13–28 Outline of Sec4 function in delivery of transport vesicles from an internal membrane to the bud membrane (Answer 13–35). Additional proteins that are involved are not shown. GTP Sec4 active vesicle delivery and GTP hydrolysis internal membrane GTP GTP GDP GEF GDP inactive Sec4 plasma membrane of bud unlikely that there is a direct effect of Sec4N133I on normal Sec4. More likely, there is an indirect effect that prevents normal Sec4 from carrying out its function. For example, if a vesicle component such as v-SNARE were present in limiting amounts, the accumulation of vesicles carrying Sec4N133I might deplete the supply, and thereby interfere with the proper fusion of vesicles carrying normal Sec4. Alternatively, Sec4N133I may bind too tightly to its Rab-like effector on the target membrane, preventing normal Sec4 from gaining access to the docking machinery. References: Walworth NC, Goud B, Kabcenell AK & Novick PJ (1989) Mutational analysis of SEC4 suggests a cyclical mechanism for the regulation of vesicular traffic. EMBO J. 8, 1685–1693. Guo W, Roth D, Walch-Solimena C & Novick P (1999) The exocyst is an effector for Sec4p, targeting secretory vesicles to sites of exocytosis. EMBO J. 18, 1071–1080. TRANSPORT FROM THE ER THROUGH THE GOLGI APPARATUS DEFINITIONS 13–36 13–37 13–38 13–39 13–40 13–41 13–42 Cisternal progression model Proteoglycan Cis-face Complex oligosaccharide Golgi apparatus (Golgi complex) High-mannose oligosaccharide Trans Golgi network (TGN) TRUE/FALSE 13–43 True. A misfolded protein is selectively retained in the ER by binding to chaperone proteins such as BiP and calnexin. Only after it has been released from such a chaperone protein—and thus approved as properly folded— does a protein become a substrate for exit from the ER. TRANSPORT FROM THE ER THROUGH THE GOLGI APPARATUS 13–44 True. The oligosaccharide chains are added in the lumens of the ER and Golgi apparatus, which are topologically equivalent to the outside of the cell. This basic topology is conserved in all membrane budding and fusion events. Thus, oligosaccharide chains are always topologically outside the cell, whether they are in a lumen or on the cell surface. True. Addition of very long unbranched chains of repeating disaccharide units in the Golgi apparatus can create proteoglycans in which the mass of sugar far exceeds that of protein. A289 13–45 THOUGHT PROBLEMS 13–46 Soluble ER proteins that are destined to reside in other membrane organelles or to be secreted are bound by transmembrane cargo receptors. The cytosolic domains of these cargo receptors bind to the COPII coats on the vesicles that form on the ER membrane, incorporating the cargo receptors, along with their cargo, into COPII-coated vesicles. The gene that is mutated in patients with cystic fibrosis encodes a protein that functions as a Cl– channel in the plasma membrane. Many of the mutations that cause cystic fibrosis produce a protein that is only slightly misfolded. Although the protein would function perfectly normally—and would prevent the disease phenotype—if it reached the plasma membrane, it is retained in the ER and degraded. Since yeast vacuolar vesicles normally carry a mixture of t-SNAREs and vSNAREs, you might imagine that they would bind to one another on the same vesicle, rendering them unavailable for vesicle docking and fusion. In that case, NSF and ATP would be required to dissociate the complex so that individual SNAREs would be available to dock the vesicles in preparation for vesicle fusion. If prying apart t- and v-SNARE complexes were the only role of NSF and ATP you might not expect them to be required for fusion of vesicles that , each had a single type of SNARE. Surprisingly other experiments showed that treatment of vesicles with NSF and ATP was also required for mixtures in which one vesicle carried just a t-SNARE and the other vesicle carried just a v-SNARE. This result suggests that NSF may have additional roles beyond that of untangling complexes of v-SNAREs and t-SNAREs. One possibility is that NSF is required to remove inhibitor proteins that bind to individual SNAREs to keep them inactive till needed. Reference: Nichols BJ, Undermann C, Pelham HRB, Wickner WT & Haas A (1997) Homotypic vacuolar fusion mediated by t- and v-SNAREs. Nature 387, 199–202. 13–49 Calnexin and HMG CoA reductase are transmembrane proteins and calreticulin is soluble. You can deduce this by examining the C-termini of the proteins for the ER retrieval signal for soluble proteins. Calreticulin has at its Cterminus the classic ER retrieval signal, KDEL. The KDEL receptor binds the retrieval signal and returns calreticulin to the ER whenever it escapes to the Golgi apparatus. Although not discussed in MBoC, HMG CoA reductase and calnexin each bear a C-terminal ER retrieval signal for a transmembrane protein: KKXX for HMG CoA reductase and KXRXX for calnexin. Reference: Zerangue N, Malan MJ, Fried SR, Dazin PF, Jan YN, Jan LY & Schwappach B (2001) Analysis of endoplasmic reticulum trafficking signals by combinatorial screening in mammalian cells. Proc. Natl Acad. Sci. U.S.A. 98, 2431–2436. 13–50 The modified PDI would be located outside the cell. If PDI were missing the ER retrieval signal, its gradual flow out of the ER to the Golgi apparatus would not be countered by its capture and return to the ER, as normally occurs. Similarly, it would be expected to leave the Golgi apparatus by the 13–47 13–48 A290 Chapter 13: Intracellular Vesicular Traffic default pathway, mixed with the other proteins the cell is secreting. It would not be expected to be retained anywhere else along the secretory pathway because it presumably has no signals to promote such localization. Reference: Munro S & Pelham HR (1987) A C-terminal signal prevents secretion of luminal ER proteins. Cell 48, 899–907. 13–51 The KDEL receptor binds its ligands more tightly in the Golgi apparatus, where it captures proteins that have escaped the ER, so that it can return them. The receptor binds its ligands more weakly in the ER, so that those proteins that have been captured in the Golgi apparatus can be released upon their return to the ER. The basis for the different binding affinities is thought to be the slight difference in pH; the lumen of the Golgi apparatus is slightly more acidic than that of the ER, which is neutral. Since the primary job of the KDEL receptor is to capture proteins that have escaped from the ER, it would be reasonable to design the system so that the receptors are found in the highest concentration in the Golgi apparatus. This is, in fact, the way it is in the cell. You would be correct if you predicted that the KDEL receptor does not have a classic ER retrieval signal; after all, the receptor is designed to spend most of its time in the Golgi apparatus, and a classic signal would ensure its efficient return to the ER. It does, however, have a ‘conditional’ retrieval signal; upon binding to an ER protein in the Golgi apparatus, its conformation is altered so that a binding site for COPI subunits is exposed. That signal allows it to be incorporated into COPIcoated vesicles, which are destined to return to the ER. Reference: Teasdale RD & Jackson MR (1996) Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the Golgi apparatus. Annu. Rev. Cell Dev. Biol. 12, 27–54. 13–52 If the KDEL signal and the KDEL receptor were all that was required to retain a protein in the ER, then addition of KDEL to a secreted protein should result in its retention in the ER. Clearly, addition of KDEL to rat growth hormone or human chorionic gonadotropin did not result in their efficient retention in the ER. Presumably, their slower rate of secretion was due to the KDEL system, since changing KDEL to KDEV abolished the effect. A comparable effect is also seen for ER residents that have had their KDEL signals removed; they are secreted, but at significantly slower rates than true secreted proteins. One explanation that might account for both these effects is kin recognition, which embodies the idea that residents of the ER might have a general affinity for one another, making it more difficult for any of them to leave the compartment. According to this idea, ER proteins that are missing their KDEL signal are secreted slowly because they still retain their affinity for other ER residents. Similarly, secreted proteins with an added KDEL signal would not be expected to have a general affinity for ER residents, and thus, would escape the ER at a higher rate than true ER residents. Reference: Zagouras P & Rose JK (1989) Carboxy-terminal SEKDEL sequences retard but do not retain two secretory proteins in the endoplasmic reticulum. J. Cell Biol. 109, 2633–2640. 13–53 If therapeutic proteins with N-linked oligosaccharides were produced in nonprimate cells, they would carry occasional oligosaccharides with Gal(a1–3)Gal linkages. Since such linkages are not present on normal human proteins, the protein might be recognized as foreign by the immune system, triggering production of antibodies against the protein. Such antibodies would eliminate the protein, along with any potential therapeutic benefit. Humans, who are periodically infected with microorganisms that contain Gal(a1–3)Gal linkages, already have circulating antibodies to this disaccharide, and are thus likely to eliminate the protein even more quickly. Reference: Takeuchi Y, Porter CD, Strahan KM, Preece AF, Gustafsson K, Cosset FL, Weiss RA & Collins MK (1996) Sensitization of cells and retroviruses to human serum by (a1-3) galactosyltransferase. Nature 379, 85–88. TRANSPORT FROM THE ER THROUGH THE GOLGI APPARATUS 13–54 (1) Attached carbohydrates promote protein folding by making the intermediates more soluble and mediating their binding to chaperones. (2) Attached carbohydrates serve as a recognition marker for transport from the ER and for protein sorting in the trans Golgi network. (3) Oligosaccharides on proteins provide protection against proteases. (4) Oligosaccharides on cell surface proteins can function in cell–cell adhesion. (5) Oligosaccharides on the cell surface provide a protective coat against pathogens. (6) Attached carbohydrates can play a regulatory role by modulating protein–protein interactions. In the vesicle transport model, different rates of protein movement might be accommodated by the ease with which proteins enter transport vesicles. Normal size proteins may enter vesicles readily and be transported rapidly. By contrast, very large proteins may not fit easily into standard vesicles, or may require a rare class of megavesicle. As a consequence, very large proteins might be expected to move across the Golgi complex more slowly. Different rates of protein movement are more difficult to account for in the cisternal maturation model. All proteins—small and large—that share the same Golgi compartment would be expected to move across the Golgi stack at the same rate as the compartment, itself. In principle, different rates might be accommodated in this model if very large proteins encountered delays at entering the Golgi stack, or leaving it. The difficulty in accommodating different rates of protein movement in either model, alone, is one reason why it is thought that both types of movement may operate: a fast track by vesicular transport, and a slow track by cisternal maturation. Reference: Pelham HRB & Rothman JE (2000) The debate about transport in the Golgi—two sides of the same coin? Cell 102, 713–719. A291 13–55 DATA HANDLING 13–56 A. The altered VSV G proteins with ‘membrane-spanning’ segments that are 12, 8, or 0 amino acids long do not make it to the plasma membrane; they remain in an intracellular location (see Table 13–2). The presence of oligosaccharides (endo H sensitivity) indicates that each of these proteins was inserted into the ER membrane, as expected since the signal peptide was not altered, and was subsequently modified by addition of oligosaccharide chains. The presence of the small C-terminal domain (protease sensitivity) on the proteins with segments 12 and 8 amino acids long indicates that these proteins are anchored in the membrane much like the normal G protein. By contrast, the complete protease resistance of the G protein with a zero amino acid transmembrane segment indicates that it passed all the way through the ER membrane into the lumen. Thus, the VSV G proteins with segments 12 and 8 amino acids long are in an internal membrane, but the one that is missing the membrane-spanning segment entirely is in an internal lumen. The partial endo H resistance of the G protein with a membrane-spanning segment of 12 amino acids suggests that some fraction of this G protein makes it as far as the medial compartment of the Golgi, which is where the relevant sugar modification occurs. The remainder of this protein is either in the membrane of the ER or the cis compartment of the Golgi. The endo H sensitivities of the G proteins with 8- and 0- amino acid segments indicate that they do not make it past the cis compartment of the Golgi and may not make it out of the ER. B. For the VSV G protein, the minimum length of the membrane-spanning segment appears to be 8 amino acids or less. G proteins with modified membrane-spanning segments only 8 amino acids long are anchored in the membrane much like the normal G protein. This result is surprising since 8 amino acids arranged in an a helix are not thought to be long enough to A292 Chapter 13: Intracellular Vesicular Traffic span the membrane. There are several possibilities: the short membranespanning segments may be arranged as extended chains rather than as a helices; the membrane may be less than 3 nm thick at the point where these segments penetrate the membrane; or adjacent portions of the G protein, including at least one basic amino acid (K or R), may be pulled into the membrane. C. The minimum length of a membrane-spanning segment consistent with proper sorting of the VSV G protein is 13 or 14 amino acids. VSV G proteins with segments 14 amino acids long are sorted to the plasma membrane like normal G proteins, whereas those with segments 12 amino acids long are not (see Table 13–2). It is curious that shorter membrane-spanning segments anchor the protein in the membrane perfectly well but interfere with sorting. This is thought to be the case because the vesicles that leave the Golgi apparatus have thicker membranes than those that come from the ER. The difference in thickness of vesicle membranes is due to the high concentration of cholesterol in Golgi-derived vesicles. Reference: Adams GA & Rose JK (1985) Structural requirements of a membrane-spanning domain for protein anchoring and cell surface transport. Cell 41, 1007–1015. 13–57 A. The mutant cell lines are arranged in Table 13–8 in the order that corresponds to the steps in the processing pathway for N-linked oligosaccharides. The numbers and kinds of sugars in N-linked oligosaccharides from the mutant cells define their position in the processing pathway by reference to Figure 13–11. Mutant C, for example, has lost no glucoses, therefore it must carry a defect in the first processing enzyme; namely, glucosidase I. Mutant H has lost one glucose, but retains the other two, therefore it must be defective in the second step of the pathway; namely, the one controlled by glucosidase II. Similar reasoning allows all the mutants to be identified with individual steps in the pathway in Figure 13–11, and thus ordered as shown in Table 13–8. A more straightforward way to approach this problem is to begin by writing out the numbers of each kind of sugar that are present at each step in the pathway. You will find that the distribution of sugars at a step will match the distribution in a mutant, thereby allowing you to order the mutants in the pathway. B. The site of oligosaccharide processing at which each of the mutants is defective is indicated in Table 13–8. Note that the oligosaccharide in mutant G was generated in the ER, but the step at which processing is blocked (that is, the next step) occurs in the Golgi. Table 13–8 Mutant and wild-type cell lines arranged in the order corresponding to the steps in the pathway for oligosaccharide processing (Answer 13–57). CELL LINE Mutant C Mutant H Mutant D Mutant G Mutant E Mutant B Mutant F Mutant A Mutant I Wild type Man 9 9 9 8 5 5 3 3 3 3 GlcNAc 2 2 2 2 2 3 3 5 5 5 Gal 0 0 0 0 0 0 0 0 3 3 NANA 0 0 0 0 0 0 0 0 0 3 Glc 3 2 0 0 0 0 0 0 0 0 SITE ER ER ER Golgi Golgi Golgi Golgi Golgi Golgi – ENZYME glucosidase I glucosidase II ER mannosidase Golgi mannosidase I GlcNAc transferase I Golgi mannosidase II GlcNAc transferase II galactose transferase NANA transferase ‘Site’ indicates the location of the processing step that is defective. The listed enzymes are the ones that are directly responsible for adding or removing sugars at the steps that are blocked. As indicated in the answer, some mutations in these cells might be in other enzymes. TRANSPORT FROM THE ER THROUGH THE GOLGI APPARATUS C. The processing enzymes modify the N-linked oligosaccharide in one of two ways: they remove sugars or they add them. Mutants C, H, D, G, and B are defective in steps at which carbohydrate is removed; they are likely to be defective in the processing enzymes themselves. Mutants E, F, A, and I are defective in steps at which carbohydrate is added. These mutants may be defective in the processing enzymes; however, they could be defective in one of the enzymes responsible for synthesizing the sugar monomer, or the enzyme responsible for activating the sugar in preparation for addition (for example, synthesizing UDP–GlcNAc, the activated form of GlcNAc), or in the proteins responsible for transporting the sugar monomers into the lumen of the ER or Golgi. 13–58 In the vesicular transport model, vesicles carry proteins across the stack by budding from one cisterna and fusing with the next. It is this role in the forward movement of proteins that is the critical difference between the two models. Vesicles are also required to maintain the identity of each cisterna by capturing resident proteins that have escaped and returning them to the appropriate cisterna. This retrograde flow is also used to capture ER-resident proteins that have escaped into the Golgi apparatus and return them to the ER. In the cisternal maturation model, vesicles are not required to move proteins across the Golgi apparatus. Movement of the stacks themselves accomplishes the forward movement of proteins. Vesicles are still required to maintain the identity of individual cisternae, but in this model they are not returning escaped proteins, but rather are transferring proteins in a retrograde direction to a new residence because their old residence has changed identities, from a cis cisterna to a medial cisterna, for example. In this model, as in the vesicular transport model, vesicles are responsible for returning escaped ER proteins to the ER. The critical difference between the two models is that the forward movement of proteins is accomplished by vesicles in the vesicular transport model and by movement of the cisternae themselves in the cisternal maturation model. A293 13–59 A. The radioactive label (GlcNAc) is added in the medial compartment, and the lectin precipitation depends on the presence of galactose, which is added in the trans compartment. Therefore, this experiment follows the movement of material between the medial and the trans compartments of the Golgi apparatus. B. If proteins moved through the Golgi apparatus by cisternal maturation, then a protein that entered the Golgi in a mutant cell should remain with that stack and mature as the newly formed cisterna moves through the stack. Thus, the cisternal maturation model predicts that none of the labeled G protein (which was labeled in the medial compartment of the Golgi apparatus in the mutant cell) should have galactose attached to it (which could only have been added in the Golgi apparatus from the wild-type cell). For this model, the fusion of the infected mutant cells to uninfected wild-type cells (see Table 13–4, line 1) should be the same as the fusion of infected mutant cells to uninfected mutant cells (line 2). By contrast, if material moved through the Golgi apparatus by vesicular transport, there is the possibility that proteins might move between separated Golgi stacks inside transport vesicles. The vesicular transport model predicts that some labeled G protein may acquire galactose in this way. For this model the fusion of infected mutant cells to uninfected wild-type cells (line 1) should yield more radioactive precipitate than fusion of infected mutant cells to uninfected mutant cells (line 2) but less than fusion of infected wild-type cells to uninfected wild-type cells (line 3). C. The results in Table 13–4 support the vesicular transport model, since nearly half the labeled G protein acquired galactose. The extent of galactose addition is surprising because it suggests that once a vesicle leaves a cisterna, it has roughly an equal chance of fusing with a cisterna in the same A294 Chapter 13: Intracellular Vesicular Traffic or different Golgi stack. A number of other control experiments showed that the morphology of the Golgi stacks was unaltered by the fusion procedure, that the mutant and wild-type Golgi stacks remained distinct from one another, and that G protein did move into the wild-type Golgi stack. Reference: Rothman JE, Miller RL & Urbani LJ (1984) Intercompartmental transport in the Golgi complex is a dissociative process: facile transfer of membrane protein between two Golgi populations. J. Cell Biol. 99, 260–271. 13–60 A. Antibodies specific for assembled PC and for individual chains were critical to the interpretation of the results. Fully assembled PC is a rigid rod that is too big to fit into transport vesicles, but individual chains—because they are more flexible—might not be excluded from transport vesicles. The specific concern of the authors was that PC might move between Golgi cisternae by disassembly, incorporation into transport vesicles, delivery to the next cisterna, and reassembly into PC. If antibodies specific only for the assembled form had been used, such a cycle of disassembly and reassembly would have been missed. It would have appeared that PC molecules were moving with the cisternae, when they were actually being transported by vesicles in their unassembled form. B. These results support the cisternal maturation model. They show convincingly that PC moves from cisterna to cisterna across the Golgi apparatus, and they do not support the idea that transport vesicles are involved in its movement. It is important to note that these experiments do not address the possibility that other proteins in these cells may move through the Golgi stack via transport vesicles. Reference: Bonfanti L, Mironov Jr. AA, Martinez-Menarguez JA, Martella O, Fusella A, Baldassarre M, Buccione R, Geuze HJ, Mironov AA & Luini A (1998) Procollagen traverses the Golgi stack without leaving the lumen of cisternae: evidence for cisternal maturation. Cell 95, 993–1003. TRANSPORT FROM THE TRANS GOLGI NETWORK TO LYSOSOMES DEFINITIONS 13–61 13–62 13–63 Autophagy Vacuole Lysosome TRUE/FALSE 13–64 13–65 False. The proton pump in lysosomes pumps protons into the lysosome to maintain a low pH. True. Endosomal membrane proteins are selectively retrieved from late endosomes by transport vesicles that deliver the proteins back to endosomes or to the trans Golgi network. The interior of late endosomes is mildly acidic (about pH 6), and as they mature into lysosomes the pH drops to the lysosomal value of pH 5.0. False. Addition of a weak base would cause M6P receptors to accumulate in late endosomes. M6P receptors, which bind lysosomal enzymes quite well at neutral pH, normally release bound enzymes at the lower pH of the late endosome and are then recycled to the Golgi. If the pH of the late endosome 13–66 TRANSPORT FROM THE TRANS GOLGI NETWORK TO LYSOSOMES Figure 13–29 Engulfment of a mitochondrion by ER membrane to form an autophagosome (Answer 13–68). A295 mitochondrion were raised, M6P receptors could not release their bound enzymes, and because they could not be recycled, they would become trapped in the late endosome. endoplasmic reticulum THOUGHT PROBLEMS 13–67 The lysosomal enzymes are all acid hydrolases, which have optimal activity at the low pH (about 5.0) in the interior of lysosomes. If a lysosome were to break, the acid hydrolases would find themselves at pH 7.2, the pH of the cytosol, and would therefore do little damage to cellular constituents. As shown in Figure 13–29, an autophagosome formed by engulfment of a mitochondrion by the ER membrane will have four layers of membrane that separate the matrix of the mitochondrion from the cytosol. From outside to inside, the sources of membranes and spaces are ER membrane, ER lumen, ER membrane, cytosol, outer mitochondrial membrane, intermembrane space, inner mitochondrial membrane, and matrix. AUTOPHAGY cytosol ER membrane ER lumen cytosol outer membrane matrix intermembrane space inner membrane 13–68 13–69 A. If the pH in late endosomes were raised to pH 6.6, the M6P receptor would bind hydrolases in the normal way in the trans Golgi network and transport them to late endosomes. At the higher endosomal pH, the receptor would not release the hydrolases and thus could not be recycled back to the trans Golgi network. Under these circumstances, the receptor might be dragged into mature lysosomes and destroyed. B. If the pH in the trans Golgi network were lowered to pH 6, the M6P receptor would not bind to the lysosomal hydrolases, and thus could not deliver them to late endosomes via the principal transport pathway. Under these conditions, the hydrolases would exit the cell via the default pathway. Once outside the cell, where the pH is around 7, some hydrolases would bind to M6P receptors that tour through the plasma membrane and then be delivered to late endosomes, via endocytosis to early endosomes. In the late endosomes, the M6P receptors would release the bound hydrolases and recycle to the trans Golgi network in the normal way. 13–70 This striking result indicates that there must be a lysosomal delivery pathway that is independent of M6P and the M6P receptor. The nature of the pathway is unknown. The M6P-independent pathway might operate inside the cell to accomplish sorting—presumably—from the trans Golgi to lysosomes, or as a scavenger pathway that picks up lysosomal enzymes from outside the cell and delivers them to lysosomes, where they are perfectly happy. Studies with M6P-receptor deficient mice indicate that both types of pathways may operate. In thymocytes from such mice, lysosomal enzymes appear to be delivered via an intracellular route, whereas liver and skin cells can pick them up via an extracellular route. Reference: Dittmer F, Ulbrich EJ, Hafner A, Schmahl W, Meister T, Pohlmann R & von Figura K (1999) Alternative mechanisms for trafficking of lysosomal enzymes in mannose 6-phosphate receptor-deficient mice are cell type specific. J. Cell Sci. 112, 1591–1597. 13–71 Adapor proteins in general mediate the incorporation of specific cargo proteins into clathrin-coated vesicles by linking the clathrin coat to specific cargo receptors. Because melanosomes are specialized lysosomes, it would seem reasonable that the defect in AP3 affects the pathway for delivery of pigment granules from the trans Golgi network, which involves clathrin-coated vesicles. AP3 localizes to coated vesicles budding from the trans Golgi network, which is consistent with a function in transport from the Golgi to lysosomes. Interestingly, humans with the genetic disorder Hermansky–Pudlak A296 Chapter 13: Intracellular Vesicular Traffic syndrome have similar pigmentation changes, and they also have bleeding problems and pulmonary fibrosis. These symptoms are all thought to reflect deficiencies in production of specialized lysosomes, which result from just a single biochemical defect. References: Kantheti P, Qiao X, Diaz ME, Peden AA, Meyer GE, Carskadon SI, Kapfhamer D, Sufalko D, Robinson MS, Noebels JL & Burmeister M (1998) Mutation in AP-3 delta in the mocha mouse links endosomal transport to storage deficiency in platelets, melanosomes, and synaptic vesicles. Neuron 21, 111–122. Zhen L, Jiang S, Feng L, Bright NA, Peden AA, Seymour AB, Novak EK, Elliott R, Gorin MB, Robinson MS & Swank RT (1999) Abnormal expression and subcellular distribution of subunit proteins of the AP-3 adaptor complex lead to platelet storage pool deficiency in the pearl mouse. Blood 94, 146–155. CALCULATIONS 13–72 A. In the hypothetical situation in which both the hydrolase and the M6P receptor are soluble, the rate of association will increase in direct proportion to the number of M6P groups. Each additional M6P group gives the hydrolase one additional way to bind to the receptor. Although the concentration of the hydrolase ([H]) remains the same, the concentration of M6P groups (which is what the receptor binds to) increases by a factor of four when the hydrolase has four M6P groups attached to it instead of one, thereby increasing the rate of association by a factor of four. The rate of dissociation of a hydrolase from a receptor is the same whether the hydrolase has one M6P group or four. The rate of dissociation is related to the stability of the interaction between a single M6P group and the M6P receptor. That interaction is unaffected by other, unbound M6P groups on the hydrolase. If the rate of association increases by a factor of four while the rate of dissociation remains unchanged, the affinity constant for the binding of a hydrolase with four M6P groups must be four times larger than the affinity constant for binding of a hydrolase with one M6P group. B. If the first receptor is assumed to be locked in place and does not interfere with binding of a second receptor to other M6P groups on the hydrolase, the affinity constant for binding a second receptor will be three-quarters that of the affinity constant calculated in part A. The presence of one receptor on each hydrolase covers one M6P group, reducing by one-quarter the number available for subsequent binding to a second receptor. This would reduce the rate of association with the second receptor by one-quarter, giving an affinity constant that is also one-quarter less. C. In the real situation with a soluble hydrolase and a membrane-bound receptor, binding to the first receptor would cause the hydrolase to become localized to a thin layer of the lumen adjacent to the membrane. This would have the effect of substantially increasing the concentration of the hydrolase in the immediate neighborhood of the M6P receptors. The increased local concentration would increase the rate of association with a second receptor correspondingly (but would not affect the rate of dissociation). As a result, the affinity constant for binding to a second receptor would increase substantially. The actual magnitude of the increase would depend on the volume of the lumen versus the volume of the thin layer adjacent to the membrane. In framing this problem we have skirted several important issues that are essential to a detailed understanding of the true situation. For example, a hydrolase with four M6P groups can interact with as many as four M6P receptors (provided they do not interfere with one another), and at equilibrium (which the real system may never achieve) there would be hydrolases in the population with zero to four bound receptors, with the various forms TRANSPORT FROM THE TRANS GOLGI NETWORK TO LYSOSOMES interconvertible by appropriate rate constants. We have avoided this complexity to try to bring out two conceptual points. The presence of multiple M6P groups on lysosomal hydrolases increases their affinity for M6P receptors (and improves the efficiency of lysosomal targeting) in two distinct ways. First, multiple M6P groups increase the rate of association of the hydrolases with M6P receptors by providing more opportunities for binding. Second, multiple M6P groups increase the concentration of hydrolases near the membrane, giving rise to a much tighter overall binding than could be achieved if hydrolases had only a single M6P group. The situation is not unlike that of a climber on a sheer rock face: one toe- or finger-hold is good, but four are better. A297 DATA HANDLING 13–73 A. The corrective factors are the lysosomal enzymes themselves. Hurler’s cells supply the enzyme missing from Hunter’s cells, and Hunter’s cells supply the enzyme missing from Hurler’s cells. These enzymes are present in the medium because of inefficiency in the sorting process. Since they carry M6P, which normally should direct them to lysosomes, they presumably escaped capture by the lysosomal pathway and were secreted. They are taken into cells and delivered to lysosomes by receptor-mediated endocytosis, which operates due to a small number of M6P receptors on the cell surface. The degradative enzymes, bound to receptors, are taken up through coated pits into endosomes and are eventually delivered to lysosomes. Since lysosomes are the normal site of action for these degradative enzymes, the defect is thereby corrected. B. Protease treatment destroys the lysosomal enzymes themselves. Periodate treatment and alkaline phosphatase treatment both remove the M6P signal that is required for binding to the receptor, thus preventing the enzymes (which are still active) from entering the cell. C. Such a scheme is unlikely to work for defects in cytosolic enzymes. External proteins normally do not cross membranes; thus, even when they are taken into cells, they remain in the lumen of a membrane-bounded compartment. In addition, foreign proteins are usually delivered to lysosomes and degraded. Reference: Kaplan A, Achord DT & Sly WS (1977) Phosphohexosyl components of a lysosomal enzyme are recognized by pinocytosis receptors on human fibroblasts. Proc. Natl Acad. Sci. U.S.A. 74, 2026–2030. 13–74 The results with the hypothetical I-cell mutants indicate that cells from mutant A are defective in the receptor for M6P, that cells from mutant B are defective in GlcNAc phosphotransferase, and that cells from mutant C are defective in GlcNAc phosphoglycosidase. These defects can be deduced from the experimental observations as indicated below. Cells from mutant A were unable to take up lysosomal enzymes from wildtype cells, indicating that they have defective M6P receptors (observation 1). They are not defective in the M6P marker, since the lysosomal enzymes they secrete can correct the defect in Hurler’s cells (observation 2). Cells from mutant B were able to take up lysosomal enzymes from wildtype cells (observation 1), indicating that they have functional M6P receptor. Since their secreted enzymes could not correct the defect in Hurler’s cells (observation 2), they were not properly modified. The improper modification was not corrected by treatment with the enzyme that removes GlcNAc (observation 3), suggesting (by elimination) that the defect is in GlcNAc phosphotransferase. Cells from mutant C were able to take up lysosomal enzymes from wildtype cells (observation 1), indicating that they have a functional M6P receptor. Since their secreted enzymes could not correct the defect in Hurler’s cells (observation 2), they were not properly modified. The A298 Chapter 13: Intracellular Vesicular Traffic improper modification was corrected by treatment with the enzyme that removes GlcNAc (observation 3), indicating that the cells are defective in GlcNAc phosphoglycosidase. 13–75 From the position of the melanosomes in Ashen, Dilute, and Leaden melanocytes, it appears that these mice have defects in transporting melanosomes to the tips of the branches so that the pigment can be properly released. The proteins that are missing in these mice normally form a complex, Rab27a/Mlph/MyoVa, which links melanosomes (via Rab27a) to a microtubule-based motor (MyoVa). That linkage allows melanosomes to be transported along microtubule tracks to the tips of the melanocyte branches. Once delivered, and prior to release, the melanosomes may be anchored to the cell cortex by the binding of Mlph to actin filaments. Defects in any of the individual proteins in the Rab27a/Mlph/MyoVa complex would prevent melanosome transport to the cell extremeties, leaving them clustered together in the interior of the cell. References: Wilson SM, Yip R, Swing DA, O’Sullivan TN, Zhang Y, Novak EK, Swank RT, Russell LB, Copeland NG & Jenkins NA (2000) A mutation in Rab27 causes the vesicle transport defects observed in ashen mice. Proc. Natl. Acad. Sci. USA 97, 7933–7938. Hume AN, Tarafder AK, Ramalho JS, Sviderskaya EV & Seabra MC (2006) A coiled-coil domain of melanophilin is essential for myosin Va recruitment and melanosome transport in melanocytes. Mol. Biol. Cell 17, 4720–4735. TRANSPORT INTO THE CELL FROM THE PLASMA MEMBRANE: ENDOCYTOSIS DEFINITIONS 13–76 13–77 13–78 13–79 13–80 13–81 13–82 13–83 13–84 13–85 13–86 Endocytosis Multivesicular body Macrophage Pinocytosis Transcytosis Caveola Clathrin-coated pit Receptor-mediated endocystosis Caveolin Early endosome Phagocytosis TRUE/FALSE 13–87 False. Not all particles that bind are ingested. Phagocytes have a variety of specialized surface receptors that are functionally linked to the phagocytic machinery of the cell. Only those particles that bind to these specialized receptors can be phagocytosed. False. The LDL receptor and many other receptors enter coated pits irrespective of whether they have bound their specific ligands. 13–88 TRANSPORT INTO THE CELL FROM THE PLASMA MEMBRANE: ENDOCYTOSIS 13–89 False. Many molecules that enter early endosomes are specifically diverted from the journey to late endosomes and lysosomes; they are recycled instead from early endosomes back to the plasma membrane via transport vesicles. Only those molecules that are not retrieved from endosomes are delivered to lysosomes for degradation. False. During transcytosis, vesicles that form from either the apical or basolateral surface first fuse with early endosomes, then move to recycling endosomes, where they are sorted into transport vesicles bound for the opposite surface. A299 13–90 THOUGHT PROBLEMS 13–91 Since the surface area and volume of a macrophage do not change significantly over this time, the rate of exocytosis must also equal 100% of the plasma membrane each half hour. Extracellular space Cytosol Plasma membrane Clathrin coat Membrane of deeply invaginated clathrin-coated pit Captured cargo particles Lumen of deeply invaginated clathrin-coated pit Because lipids rafts are thicker than other areas of the plasma membrane, the membranes inside caveolae, which form from lipid rafts, are presumably also thicker. Thus, transmembrane proteins that collect passively in caveolae might be expected to have longer transmembrane segments than normal. Influenza virus enters cells by endocytosis and is delivered to endosomes, where it encounters an acidic pH that activates its fusion protein. The viral membrane then fuses with the membrane of the endosome, releasing the viral genome into the cytosol (Figure 13–30). NH3 is a small molecule that readily penetrates membranes. Thus it can enter all intracellular compartments, including endosomes, by diffusion. Once in a compartment that has an acidic pH, NH3 binds H+ to form NH4+, which is a charged ion and therefore cannot diffuse across the membrane. NH4+ ions therefore accumulate in acidic compartments, raising their pH. When the pH of the endosome is raised, viruses are still endocytosed, but because the viral fusion protein cannot be activated, the viruses cannot enter the cytosol. Remember this the next time you have the flu and are near a stable. In the absence of bound Fe, transferrin does not interact with its receptor and circulates in the bloodstream until it catches an Fe ion. Once iron is bound, the iron–transferrin complex can bind to the transferrin receptor on EXTRACELLULAR SPACE CYTOSOL virus plasma membrane 13–92 A. B. C. D. E. F. G. 13–93 13–94 13–95 endocytosis fusion with endosome H+ pH-activated virus envelope fusion H+ endosome H+ viral genome enters cytosol Figure 13–30 Pathway for entry of the influenza virus into cells (Answer 13–94). A300 Chapter 13: Intracellular Vesicular Traffic the surface of a cell and be endocytosed. Under the acidic conditions of the endosome, the transferrin releases its iron, but the transferrin, itself, remains bound to the transferrin receptor, which is recycled back to the cell surface. The neutral pH of the blood causes the receptor to release the transferrin, itself, into the circulation, where it can pick up another Fe ion to repeat the cycle. (The iron released in the endosome moves on to lysosomes, and from there it is transported into the cytosol.) This system allows cells to take up iron efficiently, even though the concentration of iron in the blood is extremely low. The iron bound to transferrin is concentrated at the cell surface by binding to transferrin receptors; it becomes further concentrated in clathrin-coated pits, which collect the transferrin receptors. In this way, transferrin cycles between the blood and endosomes, delivering the iron that cells need to grow. CALCULATIONS 13–96 A. HRP does not bind to a specific cellular receptor and is taken up only by fluid-phase endocytosis. Since endocytosis is a continuous process, HRP gets taken up steadily at a rate that depends only on its concentration in the medium; thus, its uptake rate does not saturate. By contrast, EGF binds to a specific EGF receptor and is internalized by receptor-mediated endocytosis. The limit to the amount of EGF that gets taken up is set by the number of EGF receptors on the cells; when the receptors are saturated, no further increase in uptake occurs (except at enormously high concentrations, where fluid-phase endocytosis becomes significant). B. At 40 nM, EGF is taken up at a rate of 16 pmol/hr, while at a 1000-fold higher concentration (40 mM), HRP is taken up at 2 pmol/hr. Since the uptake of HRP is linear, the rate at a 1000-fold lower concentration is expected to be 2 ¥ 10–3 pmol/hr. Thus, at equal concentrations of 40 nM, EGF should be taken up 8000 times faster than HRP [(16 pmol/hr)/(2 ¥ 10–3 pmol/hr)]. If EGF and HRP were present at 40 mM, both would be taken up by pinocytosis at the same rate (2 pmol/hr). EGF, however, would also be taken up by receptor-mediated endocytosis at the saturation rate of 16 pmol/hr. Thus, EGF would be taken up 9 times faster than HRP [(2 pmol/hr + 16 pmol/hr)/(2 pmole/hr)]. C. An endocytic vesicle 20 nm (2 ¥ 10–6 cm) in radius contains 3.4 ¥ 10–17 mL of fluid. 3 vesicle volume = 4pr 3 = (4/3) ¥ 3.14 ¥ (2 ¥ 10–6 cm)3 = 3.4 ¥ 10–17 cm3 = 3.4 ¥ 10–17 mL A solution of 40 mM HRP contains 2.4 ¥ 1016 molecules/mL of HRP. 17 L HRP = 40 mmol HRP ¥ ¥ 6 ¥ 10 molecules 1000 mL L mmol = 2.4 ¥ 1016 molecules/mL Hence each vesicle contains, on average, 0.8 molecule of HRP [(2.4 ¥ 1016 molecules/mL)(3.4 ¥ 10–17 mL/vesicle)]. D. These calculations, as alluded to by the authors, make the point that by having specific tight-binding receptors on the cell surface, cells can take up molecules from their surroundings at much higher rates—several orders of magnitude higher—than they could simply by taking in fluid, especially at the low concentrations that are typical in biology. Fishing provides an analogy. You could fish by taking random net-fulls from a stream, and occasionally you might catch a fish. But if you put bait where you cast your net, you increase your chances of success enormously. Each time a molecule of EGF hits a receptor, it sticks and subsequently makes its way to a coated pit to be TRANSPORT INTO THE CELL FROM THE PLASMA MEMBRANE: ENDOCYTOSIS internalized. If the EGF were simply trapped like HRP, its rate of uptake would be infinitesimal at the usual in vivo concentrations. Reference: Haigler HT, McKanna JA & Cohen S (1979) Rapid stimulation of pinocytosis in human A-431 carcinoma cells by epidermal growth factor. J. Cell Biol. 83, 82–90. 13–97 If all the receptors were bound to ligand, there would be 10 ligands in a vesicle with a volume of 1.66 ¥ 10–18 L, which is 10 mM [(10 ligands/1.66 ¥ 10–18 L) ¥ (mole/6 ¥ 1023 ligand) = 10–5 mole/L or 10 mM]. This concentration is 10,000 times higher than the circulating concentration (1 nM) in the extracellular fluid. In order to concentrate the ligand 1000-fold in the vesicle, only 1 in 10 of the receptors would need to carry a bound ligand. If the Kd were 1 nM (10–9 M), half the receptors would be occupied by ligand. From the rule-of-thumb relationships worked out in Problem 3–103 (see Table 3–6), if the Kd were 10fold higher than the ligand concentration, only 10% of the ligand would be bound to the receptor, which in this case corresponds to 1 in 10 of the receptors in the vesicle having a bound ligand. Thus, the Kd for the receptor–ligand binding would need to be 10 nM (10–8 M) to concentrate the ligand 1000-fold above the ligand concentration in the extracellular fluid. A301 13–98 A. At 0°C, endocytosis is blocked and the labeled transferrin receptors are trapped on the cell surface and accessible to trypsin treatment. After 1 hour at 37°C, most of the receptors in intact cells (~70%) are not sensitive to trypsin because they are inside the cell (presumably in endosomes) and, therefore, are not accessible. When cells are incubated at 37°C, the labeled receptors are endocytosed and cycle through the endosomal compartment of the cell, thereby becoming inaccessible to trypsin. B. Both trypsin treatment and antibody binding indicate that 30% of the total transferrin receptor is on the cell surface after 1 hour at 37°C. When the transferrin receptors are allowed to recycle by incubation at 37°C, 30% is accessible to trypsin treatment of intact cells; therefore, 30% is on the surface. Similarly, antibody binds to 30% of the total receptor in the absence of detergent (0.54%/1.76% = 30%). Recycling of transferrin receptors is very fast, and this distribution between the surface and internal compartments turns out to be the equilibrium distribution for transferrin receptors. Reference: Bleil JD & Bretscher MS (1982) Transferrin receptor and its recycling in HeLa cells. EMBO J. 1, 351–355. 13–99 A. At 37°C transferrin receptors recycle between the plasma membrane and the endosomal compartment. In the beginning, all the plasma membrane receptors are labeled and all the endosomal receptors are unlabeled. Therefore, every receptor that is initially internalized is labeled. At the start, whenever a labeled receptor is internalized, it is replaced on the cell surface by an unlabeled receptor, thereby diluting the label on the surface. For that reason, the rate of entry of labeled receptors is rapid at first but then slows down, even though the flow of receptors into the cell in constant. At later times, labeled receptors begin to reappear on the surface as they are cycled back to the membrane. Eventually, the labeled and unlabeled receptors in the plasma membrane and endosomes become thoroughly mixed. At that point the rates of internalization and reappearance of labeled receptors become equal and a plateau level is reached. B. Initially, all the cell-surface receptors are labeled. The initial rate of internalization of labeled receptors is 100% in 7 minutes, or 14% of the cell-surface receptors per minute. C. At the plateau, 30% of the labeled receptors are sensitive to trypsin, indicating that 30% of the receptors are on the cell surface. Since 14% of the cellsurface receptors are internalized per minute, 4.2% of the total receptor population (30% ¥ 14%) are internalized per minute. A302 Chapter 13: Intracellular Vesicular Traffic D. At 4.2% of the total receptors per minute, an equivalent of the entire population of receptors would be internalized in 24 minutes (100%/4.2% per minute). Since the rates of internalization and reappearance are equal, an equivalent of the entire population of receptors would also be returned to the cell surface in the same interval. Thus, 24 minutes is the average time it takes for transferrin receptors to cycle from the cell surface through the endosomal compartment and back to cell surface. E. If the mean cycle time is 24 minutes and 30% of the receptors are on the cell surface at any given time, then each receptor spends, on average, about 7 minutes (24 minutes ¥ 30%) on the surface (and about 17 minutes inside the cell). Reference: Bleil JD & Bretscher MS (1982) Transferrin receptor and its recycling in HeLa cells. EMBO J. 1, 351–355. DATA HANDLING 13–100 A. Binding of LDL by normal cells and JD’s cells reaches a plateau because there are a limited number of LDL receptors per cell and they become saturated at high levels of LDL. The slope of the binding curve gives a measure of the binding affinity and the plateau gives a measure of the total number of binding sites (about 20,000 to 50,000, though you could not calculate this from the data shown here). JD has slightly fewer receptors on his cells, but they have an affinity similar to those in normal cells. Cells from patient FH bind essentially no LDL, even at saturating external LDL levels. Either these cells completely lack the LDL receptor, or the receptor is defective so that its affinity for LDL is drastically reduced. It could also be that the cells do contain receptors, but for some reason they fail to appear on the surface of the cell. B. Cells from the hypercholesterolemic patients take up LDL at a very low rate. Lack of entry is readily explained for patient FH because no LDL bound to the cells. This result indicates that the receptor is crucial for LDL cholesterol to enter cells. Since LDL is not taken up by JD’s cells, his LDL receptors must also be defective, but in a different way from FH’s LDL receptors. JD’s cells bind LDL with the same affinity as normal and almost to the same level. Although his receptors are normal as far as LDL binding is concerned, the bound LDL does not get in at the normal rate. Thus, mere possession of a receptor on the cell surface is no guarantee of entry. C. LDL must enter cells in order for the cholesterol esters to be released and hydrolyzed to cholesterol, which causes inhibition of cholesterol synthesis. In the affected patients, LDL enters the cells very slowly and, therefore, inhibits cholesterol synthesis only slightly. D. If the defects in the hypercholesterolemic patients are due to defects in their LDL receptors, then free cholesterol should inhibit cholesterol synthesis in their cells as well as in normal cells. Free cholesterol does inhibit cholesterol synthesis in all these cells, strongly supporting the idea that the defects in the patients are due solely to problems with their LDL receptors. Reference: Brown MS & Goldstein JL (1979) Receptor-mediated endocytosis: insights from the lipoprotein receptor system. Proc. Natl Acad. Sci. U.S.A. 76, 3330–3337. 13–101 A. JD’s mother evidently has one normal gene and one defective gene that encodes a receptor that cannot bind LDL (and hence does not ‘show up’ in any of the assays). With a single good gene, her cells synthesize half the usual number of functional LDL receptors, as confirmed by electron microscopy. These receptors are distributed about 50% in and 50% outside of coated pits, as are the receptors in a normal person, and they can internalize bound LDL. JD’s mother has a mild hypercholesterolemia because, with only half the usual number of functional receptors, she cannot clear LDL from the circulation as efficiently as normal individuals. Since JD can internalize none of TRANSPORT FROM THE TRANS GOLGI NETWORK TO THE CELL EXTERIOR: EXOCYTOSIS his LDL receptors, she must have passed the defective gene to her son. B. Several peculiar observations must be accounted for to understand the behavior of JD’s father’s LDL receptors: (1) His cells bind more LDL than normal cells, but internalize less than half the bound LDL; (2) His cells carry about 50% more LDL receptors on their surface than normal cells; and (3) Only about 20% of the receptors are associated with coated pits. These observations can be understood if JD’s father carries one normal gene for the LDL receptor and one gene that encodes an LDL receptor that cannot be internalized. The one normal copy of the gene allows JD’s father to clear some circulating LDL, but not as efficiently as a person with two functional genes, accounting for his mild hypercholesterolemia. The father’s defective gene, unlike the mother’s, encodes an LDL receptor that can bind LDL, but cannot internalize it. This defective receptor is more numerous on the cell surface precisely because it cannot be internalized. A normal LDL receptor is constantly cycling from the cell surface to the interior and back, carrying any bound LDL inside. Thus, at any given time a portion of the normal LDL receptor population is in the interior and unavailable for surface binding. By contrast, the entire population of internalizationdefective LDL receptors will be on the cell surface, available for LDL binding. This explains why the father’s cells have more LDL receptors on their surface and bind more LDL than normal cells, but internalize less than half. Because the excess LDL receptors on the cell surface are located outside of coated pits, the defective receptors must not be able to enter coated pits and bind there. C. JD’s inability to metabolize LDL is a direct consequence of acquiring two defective forms of the LDL-receptor gene from his parents. JD received one binding-defective LDL-receptor gene from his mother and one internalization-defective LDL-receptor gene from his father. Since both genes are defective, JD has severe hypercholesterolemia. The behavior of the receptors on JD’s cells is like his father’s defective population of receptors. The near normal number of receptors and amount of LDL binding occur because greater numbers of the internalization-deficient receptors are present on the cell surface. The inability of these receptors to associate with coated pits, which was inferred from the behavior of the mixture of receptors in his father, is clear in JD. In JD’s cells, only 2.8% of his LDL receptors are found in coated pits, close to what might be expected from a purely random distribution (coated pits occupy about 2% of the cell surface). This distribution suggests that JD’s LDL receptors are defective in the domain that is necessary to localize the receptor in coated pits. D. These studies show clearly that JD’s defective LDL metabolism is due to defective receptors, not to defective internalization machinery. The critical observations are the ones with JD’s father. The behavior of the defective population of receptors in the father is very much like the behavior of JD’s receptors. The defect in this population of receptors in JD’s father cannot be due to a defect in the internalization machinery. If the defect were in the internalization machinery, all of the bound LDL in JD’s father would behave the same. Instead, half behaves normally, and half never seems to get in. Reference: Brown MS & Goldstein JL (1979) Receptor-mediated endocytosis: insights from the lipoprotein receptor system. Proc. Natl Acad. Sci. U.S.A. 76, 3330–3337. A303 TRANSPORT FROM THE TRANS GOLGI NETWORK TO THE CELL EXTERIOR: EXOCYTOSIS DEFINITIONS 13–102 Synaptic vesicle A304 Chapter 13: Intracellular Vesicular Traffic 13–103 Constitutive secretory pathway (or default pathway) 13–104 Secretory vesicle 13–105 Exocytosis 13–106 Regulated secretory pathway TRUE/FALSE 13–107 False. Secretory proteins, even those that are not normally expressed in a given secretory cell, are appropriately packaged into secretory vesicles. For this reason, it is thought that the sorting signal, which is not yet defined, is common to proteins in this class. 13–108 False. Once positioned beneath the plasma membrane, a secretory vesicle waits until the cell receives an appropriate signal—often a rise in Ca2+ concentration—before fusing with the membrane and releasing its contents. THOUGHT PROBLEMS 13–109 In a cell capable of regulated secretion, the three main classes of protein that must be sorted before they leave the trans Golgi network are (1) those destined for lysosomes, (2) those destined for secretory vesicles, and (3) those destined for immediate delivery to the cell surface. 13–110 A. Vesicles on the endocytic pathway will contain transferrin, and thus be labeled with colloidal gold; vesicles on the exocytic pathway will contain albumin, and thus be labeled with ferritin. B. Clathrin-coated vesicles are rapidly uncoated after they pinch off from the plasma membrane, so some will be caught with their coats off, while others will still have their coats on. 13–111 Aggregates of the secretory proteins would be expected to form in the ER, just as they do in the trans Golgi network. Because the aggregation is specific for secretory proteins, ER proteins would be largely excluded from the aggregates. It is likely that such aggregates would eventually be degraded by the quality-control mechanisms that operate in the ER. 13–112 The actual explanation is that the single amino acid change causes the protein to misfold slightly so that, although it is still active as a protease inhibitor, it is prevented by chaperone proteins in the ER from exiting the cell. It therefore accumulates in the ER lumen and is eventually degraded. Alternative interpretations might have been: (1) the mutation affects the stability of the protein in the bloodstream so that it is degraded faster than the normal protein; (2) the mutation inactivates the ER signal sequence and prevents the protein from entering the ER; or (3) the mutation created an ER (or Golgi) retrieval signal so that the protein was continually returned to the ER (or Golgi). One could distinguish among these possibilities by using fluorescently tagged antibodies against the protein to follow its transport in the cells. 13–113 Synaptic transmission involves the release of neurotransmitters by exocytosis. During this event, the membranes of the synaptic vesicles fuse with the plasma membrane of the nerve terminals. To make new synaptic vesicles, membrane must be retrieved from the plasma membrane by endocytosis. This endocytosis step is blocked if dynamin is defective, as expected since dynamin is required to pinch off the clathrin-coated endocytotic vesicles. Thus, the Shibire mutant flies are paralyzed at the elevated temperature because they cannot recycle their synaptic vesicle membranes. Reference: Koenig JH & Ikeda K (1999) Contribution of active zone subpopulation of vesicles to evoked and spontaneous release. J. Neurophysiol. 81, 1495–1505. TRANSPORT FROM THE TRANS GOLGI NETWORK TO THE CELL EXTERIOR: EXOCYTOSIS A305 DATA HANDLING 13–114 The slow step in the constitutive secretion of transferrin occurs in the ER. The slow step in the constitutive secretion of albumin occurs in the Golgi. From Figure 13–22, it is clear that most of the transferrin in the cell is in the ER and most of the albumin is in the Golgi. The steady-state distribution of proteins along the constitutive pathway tells you where the proteins spend the majority of their time. As with any pathway, an accumulation occurs at the slow step. Therefore, the location of the majority of material corresponds to the slow step. The constitutive secretion of transferrin is slow relative to albumin because it is delayed in the ER. This result appears to be general: if the constitutive secretion of a protein is slow, the protein is delayed for some reason in the ER. References: Fries E, Gustafsson L & Peterson PA (1984) Four secretory proteins synthesized by hepatocytes are transported from the endoplasmic reticulum to Golgi complex at different rates. EMBO J. 3, 147–152. Lodish HF, Kong N, Snider M & Strous GJAM (1983) Hepatoma secretory proteins migrate from rough endoplasmic reticulum to Golgi at characteristic rates. Nature 304, 80–83. 13–115 Your experiments show that vesicles transport G protein without concentrating their contents. The concentration of G protein in the cisternal space was actually slightly higher than in the vesicles and vesicle buds, as measured by linear and surface density of labeled G proteins (see Table 13–6). If the vesicles were transporting G protein in a selective way (like clathrincoated vesicles), the concentration of G protein in the vesicles should have been substantially higher than in the Golgi cisternae. Reference: Orci L, Glick BS & Rothman JE (1986) A new type of coated vesicular carrier that appears not to contain clathrin: its possible role in protein transport within the Golgi stack. Cell 46, 171–184. 13–116 The pro-peptide is removed from pro-insulin in immature secretory vesicles. The red fluorescence in compartments from the ER through the trans Golgi network indicates that they contain only pro-insulin. The green fluorescence in mature secretory vesicles indicates that they contain insulin. The yellow fluorescence, which arises when both the red and green fluorophores are excited in the same place—the combination of red and green light is yellow—indicates that pro-insulin and insulin are both present in immature secretory vesicles. Thus, immature secretory vesicles must be where the pro-peptide is removed. The absence of label in lysosomes, mitochondria, and nuclei (and other compartments) provides assurance that you are indeed following the secretory pathway. 13–117 A. If the mechanism of sorting in polarized cells involved a signal-dependent pathway to one domain of the plasma membrane and a default pathway to the other, a foreign protein would be expected to follow one or the other pathway. Since foreign proteins would not be expected to contain the signal responsible for specific sorting, they would be more likely to follow the default pathway. B. The equal delivery of foreign proteins to the apical and the basolateral surfaces of polarized MDCK cells is not in agreement with the simplest expectations of the proposed sorting mechanism. Thus, the hypothesis that there is one signal-dependent pathway and one default pathway, as formulated, seems to be incorrect for polarized MDCK cells. Whatever pathway is being detected by the foreign proteins in these experiments is indifferent to which part of the cell surface it delivers its cargo. References: Gottlieb TA, Beaudry G, Rizzolo L, Colman A, Rindler MJ, Adesnik M & Sabatini DD (1986) Secretion of endogenous and exogenous A306 Chapter 13: Intracellular Vesicular Traffic proteins from polarized MDCK monolayers. Proc. Natl Acad. Sci. U.S.A. 83, 2100–2104. Kondor-Koch C, Bravo R, Fuller SD, Cutler D & Garoff H (1985) Protein secretion in the polarized epithelial cell line MDCK. Cell 43, 297–306. 13–118 Antibodies specific for the cytoplasmic domain of synaptotagmin do not stain the nerve terminals because the cytoplasmic domain is never exposed on the outside of the cell. By contrast, the lumenal domain is exposed to the outside of the cell when the synaptic vesicle fuses with the plasma membrane to release neurotransmitter molecules into the synaptic cleft. At that time, the antibody can bind to the lumenal domain of synaptotagmin. The membrane of the synaptic vesicle is quickly retrieved from the plasma membrane and reused to form new synaptic vesicles that contain bound antibodies within them. When the fusion of synaptic vesicles with the plasma membrane is stopped by lowering the temperature to 0°C, no labeling is observed. Reference: Matteoli M, Takei K, Perin MS, Südhof TC & DeCamilli P (1992) Exo-endocytotic recycling of synaptic vesicles in developing processes of cultured hippocampal neurons. J. Cell Biol. 117, 849–861. 13–119 A. The rapid component of the fusion response is due to vesicles that are already docked at the membrane and waiting for the signal to fuse and release their contents. The slow component of the fusion process is due to those vesicles that are not already docked and waiting; they are in various states of preparation for the ‘ready-to-go’ state. B. NSF in some way must be required for a step in the preparation for the ready-to-go state. Thus, when NSF is inhibited, the slow component of the fusion process is also inhibited. Interference with a preparation step would also explain why inhibition of NSF blocks the rapid component only after the second step. Nearly all the ready-to-go vesicles fuse after the first flash. In the absence of NEM (active NSF) the pool of ready-to-go vesicles is repopulated in the 2 minutes between flashes. But when NSF is inhibited, the pool remains depleted, giving rise to a much reduced rapid component in response to the second flash. C. These results show very clearly that NSF does not control the final step in fusion, otherwise its inhibition would have blocked the initial rapid fusion in response to the first flash (see Figure 13–24B, +NEM). They also point to a role of NSF in a preparatory (early) step by showing that the slow component is inhibited after both flashes, and that the rapid component is inhibited after the second flash. D. NSF-mediated ATP hydrolysis is required to disentangle v- and t-SNAREs after the fusion event. Vesicles that are docked and waiting already have their SNAREs paired, but are probably kept from fusing by a regulatory protein. In response to Ca2+ the regulatory protein is thought to release the brake on the paired SNAREs so they can complete the fusion event. For a vesicle to re-form so it can dock and fuse again, its SNAREs must be pried apart. If they are not, the vesicle cannot dock with the membrane. It is this step in the recycling of SNAREs that requires NSF. Reference: Xu T, Ashery U, Burgoyne RD & Neher E (1999) Early requirement for a-SNAP and NSF in the secretory cascade in chromaffin cells. EMBO J. 18, 3293–3304. ...
View Full Document

This note was uploaded on 01/07/2011 for the course BIOLOGY 7.012 taught by Professor Ericlander during the Spring '04 term at MIT.

Ask a homework question - tutors are online