|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. Neither palmitoylation-deficient GAD65(C30,45A)-GFP nor wt GAD65-GFP colocalize with the cis-Golgi protein GM130. Correlation coefficient (r) for colocalization of palmitoylation-deficient and wt GAD65-GFP with GM130. Neither protein colocalizes significantly with GM130. Results are presented as mean ± s.e. for 10 cells for each protein.
Fig. S2. The spatial localization of endogenous CASP in Golgi membranes is similar to that of transfected palmitoylation-deficient GAD65-GFP in primary rat hippocampal neurons. High-resolution projected confocal images (A, B) and selected confocal stacks (C) of rat hippocampal neurons at DIV10 double labeled with polyclonal antibodies to endogenous CASP and monoclonal antibodies against either the trans-Golgi network protein TGN38 (A and C) or cis-Golgi matrix protein GM130 (B). Endogenous CASP protein segregates from both TGN38 (A, enlarged frame; C) and GM130 (B, enlarged frame), and appears to be oriented more proximal to the cytosol than the two markers. The spatial localization of the CASP protein in Golgi membranes is similar to that shown for transfected palmitoylation-deficient GAD65-GFP in rat hippocampal neurons in Fig. 3.
Fig. S3. The spatial localization of two synaptic vesicle membrane proteins, VAMP2-GFP and VGAT-HA, and the dominant negative mutant of Rab5a, GFP-Rab5a(S34N), in Golgi membranes is similar to that of wt GAD65-GFP in rat hippocampal neurons. High-resolution projected confocal images of rat hippocampal neurons at DIV10 transiently expressing either VAMP2-GFP (A), HA-tagged vesicular GABA transporter (VGAT-HA) (B), or GFP-Rab5a(S34N) (C). Neurons were double immunolabeled for GM130 and either GFP (A and C) or HA-tag (B). The orientation of these proteins in Golgi membranes compared to GM130 is similar to that shown for transfected wt GAD65-GFP in rat hippocampal neurons in Fig. 3.
Fig. S4. Live-cell imaging of wt GAD65-GFP, GAD65(C30,45A)-GFP and Arf1-mRFP1 in BFA-treated COS-7 cells. Live-cell imaging of COS-7 cells coexpressing wt Arf1-mRFP1 and either wt GAD65-GFP (A, B) or palmitoylation-deficient GAD65(C30,45A)-GFP (C, D). Representative images at 0, 7.5 and 15 minutes following addition of BFA (5 µg/ml) are shown. Arf1-mRFP1 has dissociated completely at the 7.5 minute timepoint (B, D, see also Movies 6 and 7) and palmitoylation-deficient GAD65 has dissociated completely at the 15 minute timepoint following addition of BFA (C, see also Movie 6). By contrast, a significant fraction of wt GAD65-GFP remains clustered in the Golgi region at the 7.5 and 15 minute timepoints following addition of BFA (A and Movie 7).
Fig. S5. Palmitoylated GAD65-GFP colocalizes with CI-M6PR in BFA-treated cells. Confocal images of COS-7 cells transiently expressing wt GAD65-GFP and incubated without (A) or with (B) BFA (2 µg/ml) for 1 hour at 37°C. Cells were fixed and immunolabeled for GFP (green) and endogenous CI-M6PR (red). In untreated cells, GAD65-GFP is detected primarily in a perinuclear Golgi localization and in a few puncta (arrowheads) where it colocalizes with CI-M6PR. (B) BFA treatment results in a loss of the typical juxtanuclear localization of GAD65-GFP and redistribution into small perinuclear puncta where it colocalizes with endogenous CI-M6PR (arrowheads, enlarged frames a and b), indicating localization in the TGN.
Fig. S6. Recycling of GAD65(C30,45A)-GFP and GAD65-GFP to the Golgi complex occurs by a microtubule-independent pathway. Kinetics of fluorescence recovery after photobleaching of the entire Golgi-associated pool of GAD65(C30,45A)-GFP (squares) and wt GAD65-GFP (circles) in COS-7 cells subjected to a brief treatment with nocodazole (5 µg/ml, closed symbols), which affects microtubules but not the Golgi compartment (see Materials and Methods). Control cell cultures (open symbols) were treated with vehicle (DMSO) alone. The Golgi-associated pool of wt and palmitoylation-deficient GAD65-GFP were selectively photobleached in living COS-7 cells and fluorescence recovery was monitored over time at 37°C. Recovery of GFP-fluorescence into photobleached Golgi compartment was recorded at low laser power and normalized taking into consideration the loss of GFP fluorescence in the whole cell during the recording. Data are the mean ± s.e. for five cells.
Movie 1. GAD65-GFP-containing vesicles in neurons move bi-directionally in somatodendritic and axonal compartments. Cultured rat hippocampal neuron at DIV10 expressing wt GAD65-GFP imaged live at 37°C. Frames are maximum intensity projections of eight stacks acquired at 1.2 second intervals. Fluorescence is inverted so that fluorescent structures appear dark.
Movie 2. Palmitoylated GAD65-GFP-containing vesicles move randomly in the cytosol of MDCK cells. MDCK cell expressing wt GAD65-GFP imaged live at 37°C. Frames are maximum intensity projections of eight stacks acquired at 1.26 second intervals. In addition to the Golgi localization of wt GAD65-GFP, numerous GAD65-GFP-containing vesicles move randomly in the cytosol.
Movie 3. Palmitoylation-deficient GAD65-GFP is not detected in vesicles in live MDCK cells. MDCK cell expressing palmitoylation-deficient GAD65-GFP imaged live at 37°C. Frames are maximum intensity projections of eight stacks acquired at 3.6 second intervals. Palmitoylation-deficient GAD65-GFP was not detected in vesicles and instead remained confined to the Golgi and ER compartments of live MDCK cells.
Movie 4. BFA treatment results in the dissociation of the Golgi pools of both GAD65(C30,45A)-GFP and GalT-mRFP1 in live COS-7 cells. COS-7 cells coexpressing palmitoylation-deficient GAD65-GFP and GalT-mRFP1 imaged live at 37°C immediately after the addition of BFA (5 µg/ml). Imaging was performed as described in the Materials and Methods. Images were line averaged (eight iterations), which led to ∼16 second image scan time. Images were collected at 30 second intervals every 10 images. BFA treatment resulted in the complete dissociation of the Golgi pools of both GAD65(C30,45A)-GFP and GalT-mRFP1 at the 15 minute time point following treatment with BFA.
Movie 5. BFA treatment results in the dissociation of only a fraction of the Golgi pool of wt GAD65-GFP, whereas a prominent fraction is still detected in perinuclear punctate structures. COS-7 cells coexpressing wt GAD65-GFP and GalT-mRFP1 imaged live at 37°C immediately after the addition of BFA (5 µg/ml). Imaging was performed as described in the Materials and Methods. Images were line averaged (eight iterations), which led to ∼16 second image scan time. Images were collected at 30 second intervals every 10 images. The Golgi pool of GalT-mRFP1 has redistributed in a ‘blink-out’ fashion to the ER at ∼15 minutes following treatment with BFA. By contrast, only a fraction of the Golgi pool of wt GAD65-GFP has dispersed at this time point and a prominent fraction of the protein was still detected in punctate structures in the perinuclear region at the end of the movie (30 minutes).
Movie 6. BFA treatment results in the complete dissociation of the Golgi pool of Arf1-mRFP1 within 3 minutes and that of GAD65(C30,45A)-GFP within 15 minutes. COS-7 cells coexpressing GAD65(C30,45A)-GFP (left panel) and Arf1-mRFP1 (right panel) imaged live at 37°C immediately after the addition of BFA (5 µg/ml). Images were line averaged (8 iterations), which led to ∼16 second image scan time. Images were collected at 30 second intervals every 10 images.
Movie 7. A significant fraction of wt GAD65-GFP remains clustered in the perinuclear region at the 15 minute time point following addition of BFA. COS-7 cells coexpressing wt GAD65-GFP (left panel) and Arf1-mRFP1 (right panel) imaged live at 37°C immediately after the addition of BFA (5 µg/ml). Images were line averaged (eight iterations), which lead to ∼16 second image scan time. Images were collected at 30 second intervals every 10 images. BFA treatment resulted in the complete redistribution of the Golgi pool of Arf1-mRFP1 to the cytosol within 3 minutes. By contrast, a significant fraction of the Golgi pool of wt GAD65-GFP remained clustered in the perinuclear region at the end of the movie (15 minutes).
Movie 8. Fluorescence recovery after photobleaching the entire Golgi pool of the palmitoylation-deficient GAD65-GFP in live MDCK cells. The Golgi pool of GAD65(C30,45A)-GFP in live MDCK cells was photobleached as described in the Materials and Methods. Images were line averaged (four iterations), which led to ∼4 second image scan time. Fluorescence emission was collected at ∼4 second intervals. Note the rapid recovery of the palmitoylation-deficient GAD65-GFP into the bleached area and the absence of the protein in vesicular structures.
Movie 9. Fluorescence recovery after photobleaching the entire Golgi pool of wt GAD65-GFP in live MDCK cells. The Golgi pool of wt GAD65-GFP in live MDCK cells was photobleached as described in the Materials and Methods. Images were line averaged (four iterations), which led to ∼4 second image scan time. Fluorescence emission was collected at ∼4 second intervals. The recovery of the wt GAD65-GFP into the bleached area was slower than that observed for the palmitoylation-deficient mutant (Movie 8). Note the presence of the wt GAD65-GFP-positive structures, which move randomly in the cytosol.
| ||||||||||||||||||||
