QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online August 26, 2004
doi: 10.1242/10.1242/jcs.01314

This Article Summary Full Text Full Text (PDF) Supplemental Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hasegawa, H. Articles by Hay, J. C. Search for Related Content PubMed PubMed Citation Articles by Hasegawa, H. Articles by Hay, J. C.

Intramolecular protein-protein and protein-lipid interactions control the conformation and subcellular targeting of neuronal Ykt6

Haruki Hasegawa¹, Zhifen Yang¹, Leif Oltedal², Svend Davanger³ and Jesse C. Hay^1,*,

¹ University of Michigan, Department of Molecular, Cellular and Developmental Biology, Ann Arbor, MI 48109-1048, USA
² University of Bergen, Department of Anatomy and Cell Biology, Locus on Neuroscience, 5009 Bergen, Norway
³ University of Oslo, Centre for Molecular Biology and Neuroscience, PO Box 1105 Blindern, 0317 Oslo, Norway

View larger version (46K): [in a new window]   Fig. 1. Space-filling model of Ykt6p longin-domain structure. Surface-exposed residues that are conserved from yeast to human are highlighted with colors. White residues are not conserved. Views of two sides of the domain are shown. Structural coordinates are obtained from the Protein Data Bank (number 1H8M). Yellow indicates nonpolar residues, green indicates polar residues, red indicates acidic residues and blue indicates basic residues. The figure was generated using Deep View Swiss-PDB Viewer (The Swiss Institute of Bioinformatics).

View larger version (129K): [in a new window]   Fig. 2. Mutant Ykt6p longin domains are targeted normally. Mutations were introduced into N-terminally Myc-tagged yeast Ykt6p longin domain (residues 1-139) and transfected into PC12 cells. See Table 1 for the complete listing of mutations. Subcellular localization of each mutant and endogenous rat ykt6 was examined by double-labeling fluorescent microscopy using anti-Myc antibody (9E10) and anti-rat-Ykt6 antibody. Whereas colocalization of mutants with endogenous rat Ykt6 is shown for R2E/V8N, F39E and I59EA (top three rows; 12%, 18% and 16% overlap, respectively), only the subcellular localization of Myc-tagged mutant proteins is shown for E100K/Y101N, R50E/R56E and F42 (bottom row). Arrows point to a subset of overlapping punctate structures. Scale bar, 10 µm.

View larger version (98K): [in a new window]   Fig. 3. Normal targeting of Ykt6 requires both conserved and Ykt6-specific targeting signals in the longin domain. (A) Several distinct longin domains, when removed from their normal protein context, can localize to the Ykt6 vesicular structures. Longin (lgn) domains from Sec22b, VAMP7, Sec22a and sedlin were N-terminally Myc-tagged and expressed in PC12 cells. Subcellular localization of each Myc-longin domain and endogenous rat Ykt6 was examined by double-label fluorescence microscopy using anti-Myc antibody (9E10) and anti-rat-Ykt6 antibody. In each row, the left-hand panel shows the localization of a longin domain of different origin, the middle panel shows endogenous Ykt6 and the right-hand panel is the merged image. Arrows point to punctate structures where the Myc-longin domain and endogenous rat Ykt6 colocalize (22%, 19%, 22% and 14% overlap, respectively). Scale bars, 10 µm. (B) Only the Ykt6 longin domain can direct normal subcellular targeting of the Ykt6 SNARE and prenylation motifs. The longin domain of rat Ykt6 was replaced by exogenous longin domains to create N-terminally Myc-tagged chimeric Ykt6 proteins. Each chimeric rat Ykt6 was expressed in PC12 cells and the subcellular localization was examined by fluorescent microscopy using anti-Myc antibody and FITC-labeled secondary antibody. Scale bars, 10 µm.

View larger version (127K): [in a new window]   Fig. 4. Surface mutations on the Ykt6p longin domain disrupt the targeting of full-length Ykt6p. The same complete set of mutations (Fig. 2, Table 1) were introduced into N-terminally Myc-tagged yeast full-length Ykt6p. Mutants were expressed in PC12 cells as before. Localization of each mutant was examined by co-staining with anti-rat-Ykt6 or anti-syntaxin-1 antibodies. (Top row) The subcellular localization of Myc-tagged yeast Ykt6p mutant proteins E100K/Y101N, F42E and R50E/R56E. (Second row) The localization of the V8N mutant stained with anti-Myc (polyclonal) and anti-syntaxin 1 (47% overlap) antibodies. (Third and fourth rows) Double labeling of mutant-construct-expressing cells with anti-Myc (E10) and anti-Ykt6 antibodies. These constructs displayed 1.6% and 2.6% overlap with endogenous Ykt6, respectively. In the second case, the bright perinuclear area of the mutant was excluded from the calculation owing to obviously coincidental overlap. Scale bar, 10 µm.

View larger version (61K): [in a new window]   Fig. 5. Rat Ykt6 localization is regulated by similar longin-domain surface features. Mutations to the longin domain were introduced to full-length rat Ykt6 and truncated rat Ykt6 composed of only a longin domain (residues 1-137). All the mutants were expressed in PC12 cells and the subcellular localization of each mutant was examined by fluorescent microscopy using anti-Myc antibody (9E10) and FITC-labeled secondary antibody. (Top) Mutants in the full-length context. (Bottom) Mutants in the isolated Ykt6 longin constructs. Scale bar, 10 µm.

View larger version (101K): [in a new window]   Fig. 6. Mistargeting is rescued by removal of C-terminal lipids. (A) Effect of C-terminal deletions on yeast Ykt6p mislocalization mutants. The last five amino acid residues (CCIIM) were deleted from yeast Ykt6p mislocalization mutants and expressed in PC12 cells (Table 1). The subcellular localization of each mutant and endogenous rat Ykt6 was examined by double-label fluorescence microscopy using anti-Myc antibody (9E10) and anti-rat-Ykt6 antibody. (Top left) Myc-tagged yeast Ykt6p F39E with CCIIM deletion. (Bottom left) R50E/R56E with CCIIM deletion. Arrows point to punctate structures where the mutants and endogenous rat Ykt6 colocalize (overlaps were 31% and 18%, respectively). Identical results were obtained when CCIIM was deleted from other mislocalization mutants (Table 1). Scale bar, 10 µm. (B) Effect of C-terminal mutations on rat-Ykt6 mislocalization mutants. Rat Ykt6 mutants that have mutations in both the longin domain and the C-terminal lipidation motif were created (Table 2). Mutants were expressed in PC12 cells and the subcellular localization of each mutant was examined by fluorescence microscopy using anti-Myc antibody and FITC-labeled secondary antibody. Mutations to the longin domain are indicated at the left and mutations to the C-terminus are indicated at the top. Scale bar, 10 µm.

View larger version (87K): [in a new window]   Fig. 7. 2-Bromopalmitate suppresses the mislocalization phenotype of selected mutants. (A) PC12 cells were transfected with full-length yeast Ykt6p mutants. At 5 hours after transfection, the growth medium was changed to a medium containing 2-bromopalmitate and incubated until the cells were fixed. After fixation, subcellular localization of each mutant and endogenous rat Ykt6 was examined by double-labeling fluorescent microscopy using anti-Myc antibody and anti-Ykt6 antibody. Top panels show co-staining of endogenous Ykt6 (left) and Myc-Ykt6p-V8N (middle), and merged (right) images after drug treatment. Arrows point to puncta where colocalization is observed (27% overlap). The same degree of colocalization was observed for Myc-Ykt6p-K100E/Y101N (bottom left) and endogenous Ykt6 (not shown). Mislocalization phenotypes of Myc-Ykt6p-F42E (bottom center), Myc-Ykt6p F39E (bottom right), Myc-Ykt6p R50E/R56E (not shown) and Myc-Ykt6p I59E (not shown) were not altered by the 2-bromopalmitate treatment. Scale bar, 10 µm. (B) GAP-43 translocates from membrane to cytosol in the presence of 2-bromopalmitate. Localization of endogenous GAP-43 was analysed by membrane fractionation followed by immunoblotting. Abbreviations: P, pellet fraction; S, soluble fraction; T, total fraction. Presence or absence of 2-bromopalmitate (2-BP) is indicated at the top.

View larger version (31K): [in a new window]   Fig. 8. Localization mutants display disrupted protein-protein and protein-lipid intramolecular interactions. (A) Results of bimolecular binding assay. Bead-immobilized GST-longin domains were mixed with a Triton-X-100 cell lysate prepared from PC12 cells overexpressing either lipidated (Myc-rYkt6 residues 137-198) or non-lipidated (Myc-rYkt6 residues 137-193) SNARE motif and incubated at 4°C for 1 hour. Beads were washed, boiled in SDS sample buffer, subjected to SDS-PAGE and immunoblotted to detect the bound SNARE motif. The lipidated (residues 137-198) and non-lipidated (residues 137-193) Myc-SNARE motifs are indicated by arrows. (B) Equivalent protein loading to the beads. Proteins bound to 1.25 µl bed volume of glutathione-Sepharose beads were resolved on a 15% gel by SDS-PAGE and stained with Coomassie blue. (C) Equivalent amount of SNARE motif was used for the binding assay. 10% of the PC12 extracts containing the SNARE motifs used in the binding assay were analysed by SDS-PAGE and immunoblotting. (D) Myc-rYkt6-137-193 actively assembles with ER/Golgi SNAREs. Cell lysate containing Myc-rYkt6-137-193 was incubated with buffer alone or with purified, recombinant, ER/Golgi SNAREs (2 µM each), followed by Superose-12 gel filtration chromatography. Column fractions were immunoblotted with anti-Ykt6 antibody. Elution positions of molecular weight standards are indicated above.

View larger version (91K): [in a new window]   Fig. 9. Transmembrane-anchored Ykt6 constructs cannot localize normally to punctate Ykt6 structures. The C-terminal lipidation motif (CCAIM) of rat Ykt6 was replaced with the true transmembrane domain (TM) of Sec22b, VAMP7 or GOS-28. The chimeric proteins were expressed in PC12 cells and their subcellular localizations were examined by fluorescence microscopy using anti-Myc antibody and the indicated co-stained marker. Arrows point to gigantic aberrant membrane structures with acidic lumens induced by Ykt6 containing a VAMP7 transmembrane domain. Scale bar, 10 µm.

View larger version (175K): [in a new window]   Fig. 10. Ykt6 exhibits a very broad vesicular cytoplasmic localization in cultured hippocampal neurons. Hippocampal neurons and glia from 1-4-day-old rats were cultured for 15 days in vitro before fixation and staining with anti-rat-Ykt6 (red) and anti-Sec6 (green) antibodies, and analysed by confocal microscopy. (B-E) Individual synaptic terminals with the plasma membrane clearly demarcated by Sec6 staining. Scale bar, 25 µm (A), 2 µm (E).

View larger version (24K): [in a new window]   Fig. 11. Model for Ykt6 intramolecular interactions (left), targeting (middle) and activation upon membrane insertion (right). In this model, the Ykt6 longin domain (Lgn) interacts with the lipid groups (black) and the SNARE motif (red) such that the lipids are shielded from solvent and from premature insertion into membranes and the SNARE motif is sequestered and unavailable to other SNAREs. However, the longin domain is still available for interaction with a targeting component on the Ykt6 compartment. This allows inactive Ykt6 to be targeted to its specialized location, where binding between the targeting component and the longin domain results in: (1) interaction of the lipid group(s) with the membrane; (2) release of the longin domain from its interaction with the C-terminal lipid(s); and (3) release of the Ykt6 SNARE motif from sequestration to allow potential SNARE interactions.