Coordinated regulation of bidirectional COPI transport at the Golgi by CDC42

Abstract

The Golgi complex has a central role in the intracellular sorting of secretory proteins. Anterograde transport through the Golgi has been explained by the movement of Golgi cisternae, known as cisternal maturation. Because this explanation is now appreciated to be incomplete, interest has developed in understanding tubules that connect the Golgi cisternae. Here we show that the coat protein I (COPI) complex sorts anterograde cargoes into these tubules in human cells. Moreover, the small GTPase CDC42 regulates bidirectional Golgi transport by targeting the dual functions of COPI in cargo sorting and carrier formation. CDC42 also directly imparts membrane curvature to promote COPI tubule formation. Our findings further reveal that COPI tubular transport complements cisternal maturation in explaining how anterograde Golgi transport is achieved, and that bidirectional COPI transport is modulated by environmental cues through CDC42.

Figure 1. Characterizing VSVG transport

a, Cytoplasmic sequence of VSVG with residues critical for binding by coatomer highlighted. b, Effect of point mutations in the VSVG tail on the in vitro binding of this tail by coatomer, n = 5. c, FLIM assessing interaction between VSVG and coatomer. Representative images that are pseudo-colored based on τ₁ values are shown (out of 10), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m. d, Effect of mutations in the VSVG tail on its transport from the ER to the trans-Golgi. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. e, Effect of mutations in the VSVG tail on its transport from the ER to the cis-Golgi. Representative images of colocalization are shown (out of 5), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m.

Figure 2. Characterizing VSVG-LDLR transport

a, Cytoplasmic sequence of LDLR with basic residues critical for binding by coatomer highlighted. b, Effect of mutations in the LDLR tail on the transport of VSVG-LDLR from the ER to the trans-Golgi. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um, c, Effect of mutations in the LDLR tail on the transport of VSVG-LDLR from the ER to the cis-Golgi. Representative images of colocalization are shown (out of 5), bar = 5 um. Quantitation is also shown, n = 4, mean +/− s.e.m., d, Effect of mutations in cargo tails on the intra-Golgi transport of various cargoes. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um.

Figure 3. Further characterizing cargo transport

a, FLIM assessing interaction between different forms of VSVG and coatomer. Representative images that are pseudo-colored based on τ₁ values are shown (out of 10), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m., b, Effect of mutations in the VSVG tail on its transport from the ER to the trans-Golgi. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m.

Figure 4. Characterizing how cdc42 affects VSVG transport

a, Effect of expressing active cdc42 on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. b, Immunoblotting of whole cell lysates to assess efficiency of siRNA treatment, n = 3. c, Effect of siRNA against cdc42 on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. Quantitation is also shown, n = 4, mean +/− s.e.m. d, Effect of expressing active cdc42 on VSVG transport from the ER to the cis-Golgi. Representative images of colocalization are shown (out of 5), bar = 5 um. Quantitation is also shown, n = 5, mean +/− s.e.m. e, Effect of siRNA against cdc42 on VSVG transport from the ER to the cis-Golgi. Representative images of colocalization are shown (out of 5), bar = 5 um. Quantitation is also shown, n = 4, mean +/− s.e.m.

Figure 5. Further characterizing the effects of cdc42

a, Effect of various conditions on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. b, Effect of expressing active cdc42 on the transport of VSVG-KDELR. Representative images of colocalization is shown (out of 5), bar = 5 um. c, Effect of siRNA against cdc42 on the transport of VSVG-KDELR. Representative images of colocalization are shown (out of 5), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m. d,e, FLIM assessing the effect of active cdc42 on the interaction between different VSVG forms and coatomer. Representative images that are pseudo-colored based on τ₁ values are shown (out of 10), bar = 5 um. Quantitation is also shown as table and as graph, n= 3, mean +/− s.e.m., ***P<0.001, NS indicating non-significance (two-tailed Mann-Whitney test).

Figure 6. Characterizing how cdc42 affects COPI cargo sorting and carrier formation

a, b, Effect of different small GTPases on the in vitro binding of cargo tails by coatomer. Representative blots are shown (out of 3). Quantitation is also shown, n = 3, mean +/− s.e.m., *P<0.05, NS indicating non-significance (two-tailed Student’s t-test). c, d, Effect of different active forms of cdc42 on tubule and vesicle formation in the COPI reconstitution system, n = 4, mean +/− s.e.m., **P<0.01, NS indicating non-significance (two-tailed Student’s t-test).

Figure 7. Delineating the role of di-arginine residues in cdc42

a, Amino acid sequence at the carboxyl terminus of cdc42. The last three residues are cleaved upon prenylation. b, The di-arginine residues in cdc42 are required for cdc42 dimerization. Representative result from gel filtration (out of 2) is shown. c, Effect of different conditions on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. d, Effect of mutations in cdc42 on its ability to bind coatomer in vitro. Representative blot (out of 2) is shown. e, Effect of mutations in cdc42 on its ability to compete with cargo tails for binding to coatomer in vitro. Representative blot (out of 3) is shown. f, Effect of different forms of cdc42 to compete with the KDELR tail for binding to coatomer in vitro. Representative blot (out of 2) is shown. g, FLIM examining VSVG-KDELR interacting with coatomer, n= 3, mean +/− s.e.m., ***P<0.001, NS indicating non-significance (two-tailed Mann-Whitney test).

Figure 8. Delineating how external stimuli regulate bidirectional Golgi transport

a, Effect of different conditions on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m. b, Effect of serum on the transport of VSVG-KDELR, n = 3, mean +/− s.e.m. c, Effect of EGF on the intra-Golgi transport of VSVG, n = 3, mean +/− s.e.m. d, Effect of EGF on the transport of VSVG-KDELR, n = 3, mean +/− s.e.m. e, f, Effect of different conditions on VSVG transport from the ER to the cis-Golgi, n = 3, mean +/− s.e.m.

Figure 9. Delineating how Src regulates bidirectional Golgi transport

a, c, Effect of different conditions on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. Quantitation is also shown, n = 3 experiments, mean +/− s.e.m. b, d, Effect of different conditions on the transport of VSVG-KDELR, n = 3, mean +/− s.e.m.

Figure 10. Further characterizing how bidirectional COPI transport at the Golgi is regulated

a, Summarizing how external stimuli regulates bidirectional COPI transport through a signaling cascade. b, Effect of Src activation on the intra-Golgi transport of VSVG. Colocalization is exemplified by line scanning across representative images (out of 5), bar = 5 um. Quantitation is also shown, n = 3, mean +/− s.e.m.

Figure 1. Coatomer binds VSVG to promote its transport through the Golgi

a-c, In vitro binding of the VSVG tail (a, full-length, b, truncations, c, point mutants) by coatomer, n = 3, d, Using the Wbp1 peptide that contains either the retrograde di-lysine motif (KK) or not (SS) for competition in cargo binding by coatomer, n = 4, e, FLIM examining VSVG interacting with coatomer, n = 3, mean +/− s.e.m., ***P<0.001 (two-tailed Mann-Whitney test), f, Effect of mutating the VSVG tail on its transport from ER to trans-Golgi, n = 4, mean +/− s.e.m.

Figure 2. Coatomer also binds to the LDLR tail to promote the transport of VSVG-LDLR through the Golgi

a, In vitro binding of the LDLR tail (full-length and point mutants) by coatomer, n = 4, b, Using the Wbp1 peptides for competition in cargo binding by coatomer, n = 3, c, Effect of mutating the LDLR tail on the transport of VSVG-LDLR from ER to trans-Golgi, n = 4, mean +/− s.e.m. d, Effect of mutations in cargo tails on the intra-Golgi transport of different cargoes, n= 3, mean +/− s.e.m. e, Titrating coatomer level for in vitro binding to cargo tails, n = 2, f, FLIM examining different cargoes interacting with coatomer, n = 3, mean +/− s.e.m., **P<0.01, ***P<0.001 (two-tailed Mann-Whitney test).

Figure 3. Cdc42 modulates cargo sorting by COPI

a,b, Effect of different conditions on the intra-Golgi transport of VSVG, n = 5, mean +/− s.e.m. c, Effect of expressing active cdc42 on VSVG-KDELR transport, n = 3, mean +/− s.e.m. d, e, Immunogold labeling of reconstituted COPI tubules, with representative (out of 15) EM images shown (bar = 50 nm), and also quantitation shown (n = 4, mean +/− s.e.m., ***P<0.001, NS indicating non-significance, two-tailed Student’s t-test), f, g, Effect of adding different cdc42 forms on the binding of cargo tails by coatomer, n = 2.

Figure 4. Cdc42 modulates carrier formation by COPI

a, Effect of active cdc42 on COPI tubule formation, with representative (out of 15) EM images shown (bar = 50 nm), and also quantitation shown (n = 4, mean +/− s.e.m., **P<0.01, two-tailed Student’s t-test), b, Effect of active cdc42 on COPI vesicle formation, n = 6, mean +/− s.e.m., **P<0.01 (two-tailed Student’s t-test), c,d, Effect of mutations in cdc42 on its ability to induce liposome tubulation, c, representative (out of 10) EM images (bar = 500 nm), d, quantitation (n = 5, mean +/− s.e.m., ***P<0.001, two-tailed Student’s t-test), e, Effect of mutations in cdc42 on its ability to induce COPI tubule formation, n = 4, mean +/− s.e.m., **P<0.01 (two-tailed Student’s t-test), f, Effect of mutations in cdc42 on its ability to promote the intra-Golgi transport of VSVG, n = 3, mean +/− s.e.m.