XB-ART-4637Development November 1, 2003; 130 (22): 5425-36.
GBP binds kinesin light chain and translocates during cortical rotation in Xenopus eggs.
In Xenopus, axis development is initiated by dorsally elevated levels of cytoplasmic beta-catenin, an intracellular factor regulated by GSK3 kinase activity. Upon fertilization, factors that increase beta-catenin stability are translocated to the prospective dorsal side of the embryo in a microtubule-dependent process. However, neither the identity of these factors nor the mechanism of their movement is understood. Here, we show that the GSK3 inhibitory protein GBP/Frat binds kinesin light chain (KLC), a component of the microtubule motor kinesin. Upon egg activation, GBP-GFP and KLC-GFP form particles and exhibit directed translocation. KLC, through a previously uncharacterized conserved domain, binds a region of GBP that is required for GBP translocation and for GSK3 binding, and competes with GSK3 for GBP. We propose a model in which conventional kinesin transports a GBP-containing complex to the future dorsal side, where GBP dissociates and contributes to the local stabilization of beta-catenin by binding and inhibiting GSK3.
PubMed ID: 14507779
Article link: Development
Species referenced: Xenopus laevis
Genes referenced: dvl1 dvl2 frat1 gsk3b kif5b klc1 myc pias1 tpr
Article Images: [+] show captions
|Fig. 1. Translocation of dorsal determinants and GBP-GFP during cortical rotation. (A) Diagram of translocation events during cortical rotation. Before cortical rotation (left), dorsal determinants are present in the vegetal shear zone. After cortical rotation (top right), the cortex has rotated 30° relative to the core cytoplasm toward the prospective dorsal side, and dorsal determinants have translocated in the same direction beyond the extent of cortical displacement. In an experimental situation (bottom right), the cortex is immobilized, and eggs are imaged from the vegetal pole with an inverted scanning laser confocal microscope. From the perspective of the viewer, the core cytoplasm appears to undergo a reverse rotation following activation, and the dorsal determinants translocate in the opposite direction. Note that the relative displacement of the core cytoplasm, cortex and determinants after rotation is the same in the normal and experimental situations. (B) GBP-GFP particles exhibit fast transport opposite the direction of the yolk during cortical rotation. Time-lapse images of the vegetal shear zone of an activated egg showing movements of GBP-GFP (green) during cortical rotation. Position of yolk platelets was inferred from dark areas in the GFP channel. First frame of sequence shows starting position of three GBP-GFP particles (yellow circles) and three yolk platelets (asterisks). (B′) Same field of view 90 seconds later showing paths (yellow arrows) taken by the GBP-GFP particles moving from left to right and yolk platelets (Y) moving from right to left. See Movie 1 (http://dev.biologists.org/supplemental/) to view time-lapse movie of GBP-GFP particle translocation during cortical rotation.|
|Fig. 2. GBP and Frat associate with kinesin light chain (KLC) in vivo. (A) KLC1 and Frat1 co-immunoprecipitate in COS-7 cells. Cells were transfected with full-length FLAG-tagged mouse Frat1 (mFrat1) and a fragment of mouse KLC1 obtained in the two-hybrid screen that consists of the N-terminal 198 residues of KLC1 fused to a Glu-Glu epitope tag (mKLC1). Cell extracts were immunoprecipitated with anti-GG or anti-FLAG antibody as indicated (IP), and the lysates were analyzed by Western blot as indicated (Blot). A portion of the cell lysate was reserved before immunoprecipitation and analyzed by Western blot to confirm expression of the transfected constructs (TL). Lanes 3 and 7 show the level of background binding of mFrat1-FLAG and mKLC1-GG, respectively, to the protein A/G beads used for immunoprecipitation. (B) XKLC4 co-immunoprecipitates GBP in Xenopus embryos. Embryos were injected with XKLC4-HA and GBP-myc RNAs at the 2-4-cell stage, lysed after 4 hours and immunoprecipitated (IP) with anti-HA antibody or no antibody (no Ab) as a negative control (left panel). A portion of each sample was taken prior to immunoprecipitation to show expression of injected RNAs (total lysates, right panel). Samples were analyzed by Western blot with anti-myc and anti-HA antibodies (Blot).|
|Fig. 4. GBP binds within the first 44 amino acids of XKLC4. (A) Schematic diagram of the XKLC4 deletion constructs. The wild-type (WT) protein is depicted on top, with the previously identified domains shown as boxes, and with numbers indicating the first and last amino acid residues of the domains. In the case of the heptad repeat-containing and TPR domain constructs, the numbers indicate the first and last residues of XKLC4 that are included. For XKLC4Δ 1-4, the numbers indicate the residues flanking the deletions. (B) Xenopus embryos were injected at the 2-4-cell stage with GBP-myc and WT or mutant XKLC4-HA RNAs as indicated. After 4-5 hours the tagged proteins were immunoprecipitated with anti-HA antibody or anti-FLAG antibody as a negative control and detected by Western blotting with anti-HA and anti-myc antibodies (left panel). An aliquot of each sample was taken before immunoprecipitation to show relative expression levels of injected RNAs (total lysates, right panel). Note that the XKLC4-TPR construct runs with the Ig heavy chain from the antibody used in the immunoprecipitation. (C) Embryos were injected with GBP-myc RNA and WT or deletion mutant XKLC4-HA RNA as indicated and processed as in B.|
|Fig. 5. XKLC4 translocates during cortical rotation and associates with GBP-GFP in particles. (A,B) XKLC4-GFP particles exhibit fast transport opposite the yolk during cortical rotation. Shown here are time-lapse images of the vegetal shear zone of an activated egg showing movements of XKLC4-GFP (green) and yolk platelets (red) during peak cortical rotation. (A) First frame of sequence showing starting position of four XKLC4-GFP particles (yellow circles) and three yolk platelets (asterisks). (B) Same field of view approximately 38 seconds later showing paths (yellow arrows) taken by the XKLC4-GFP particles moving from left to right and yolk platelets (Y) moving from right to left. Please also see Movies 2-4 (http://dev.biologists.org/supplemental/) to view time-lapse movies of XKLC4-GFP particle translocation just prior to cortical rotation, during peak cortical rotation and during late cortical rotation. (C-E) Localization of XKLC4-HA and GBP-GFP during cortical rotation. (C) Confocal image of GBP-GFP particles in the vegetal shear zone of an egg fixed during peak cortical rotation. (D) Localization of XKLC4-HA stained with anti-HA antibodies and labeled with Alexa Fluor 568 in the same confocal section. (E) Merged images of C and D. Arrowheads indicate examples of particles containing both GBP-GFP and XKLC4-HA.|
|Fig. 6. XKLC4 interacts with domain III of GBP. (A) Schematic diagram of GBP deletion constructs. The wild-type (WT) protein is shown on top, with the conserved domains represented as boxes, and numbers indicating the first and last amino acid residues of the domains. For the deletion constructs, the numbers indicate the residues that flank the deletions. (B,C) Association of XKLC4 with GBP deletion mutants in vivo. (B) Xenopus embryos were injected at the 2-4-cell stage with XKLC4-HA RNA and WT GBP-myc, ΔN-I-myc, Δ-II-myc orΔ C-III-myc RNA as indicated above each lane. After a 4-5-hour incubation, lysates were immunoprecipitated (IP) with anti-HA antibody or no antibody (no Ab) as a negative control. A portion of each sample was taken prior to immunoprecipitation (IP) to show expression of injected RNAs (total lysates, right panel). Samples were immunoblotted with anti-HA and anti-myc antibodies (Blot). (C) As in B, embryos were injected with XKLC4-HA RNA and WT GBP-myc, ΔC-I-myc,Δ N-III-myc or ΔC-III-myc RNA, as indicated above each lane. Samples were processed and immunoblotted as in B. A portion of each sample was taken prior to immunoprecipitation to show expression of injected RNAs (total lysates, right panel). (D) Binding of XKLC4 and GSK3 to GBP in vivo is mutually exclusive. Xenopus embryos were injected at the 2-4-cell stage with XKLC4-HA RNA, WT GBP-myc RNA and WT GSK3-myc or kinase-dead (kd) GSK3-myc RNA as indicated above each lane. After a 4-5-hour incubation, lysates were immunoprecipitated with anti-HA antibody or no antibody (no Ab) as a negative control (left panel). A portion of each sample was taken prior to immunoprecipitation to show expression of injected RNAs (total lysates, right panel). Samples were immunoblotted with anti-HA and anti-myc antibodies (Blot).|
|Fig. 7. Domain III of GBP is important for normal particle formation in Xenopus eggs. (A-F) Activated eggs expressing wild-type (WT) GBP-GFP (A,D), ΔC-III-GFP (B), GFP alone (C), ΔN-III-GFP (E) orΔ -II-GFP (F) were fixed during peak cortical rotation and the shear zone was imaged from the vegetal pole with a scanning laser confocal microscope. Scale bars in A,D: 10 μm. A and B are from one experiment, and C-F are from a second experiment. (G-H) GBP-GFP constructs are expressed at similar levels. (G) Eggs were injected with WT GBP-GFP or ΔC-III-GFP RNAs and prepared as for the live imaging experiments. Egg lysates were then Western blotted with an anti-GFP antibody. (H) Eggs were injected with WT GBP-GFP, Δ-II-GFP, ΔN-III-GFP or GFP RNAs, and processed as in G.|
|Fig. 8. Model for involvement of XKLC and GBP in translocation of dorsal determinants. At the onset of cortical rotation (A), KLC bound to KHC on the subcortical microtubule array nucleates particles that include GBP and its binding partner Dsh. As cortical rotation progresses (B), kinesin transports these particles along the rapidly aligning microtubule bundles towards their plus ends, which are oriented toward the prospective dorsal marginal zone (d). Upon reaching the prospective dorsal region (C), Dsh recruits GBP to theβ -catenin degradation complex by binding to Axin (horizontal yellow oblong), which is bound to APC (vertical mauve oblong). GBP dissociates from KLC in favor of binding to GSK3, thereby removing GSK3 from the Axin complex by competing with Axin for its binding. The removal of GSK3 from the degradation complex allows β-catenin to accumulate in the dorsal region, where it later activates the transcription of dorsal organizer genes.|