Glycogen synthase kinase 3 phosphorylates kinesin light chains and negatively regulates kinesin-based motility - PubMed (original) (raw)

Glycogen synthase kinase 3 phosphorylates kinesin light chains and negatively regulates kinesin-based motility

Gerardo Morfini et al. EMBO J. 2002.

Abstract

Membrane-bounded organelles (MBOs) are delivered to different domains in neurons by fast axonal transport. The importance of kinesin for fast antero grade transport is well established, but mechanisms for regulating kinesin-based motility are largely unknown. In this report, we provide biochemical and in vivo evidence that kinesin light chains (KLCs) interact with and are in vivo substrates for glycogen synthase kinase 3 (GSK3). Active GSK3 inhibited anterograde, but not retrograde, transport in squid axoplasm and reduced the amount of kinesin bound to MBOs. Kinesin microtubule binding and microtubule-stimulated ATPase activities were unaffected by GSK3 phosphorylation of KLCs. Active GSK3 was also localized preferentially to regions known to be sites of membrane delivery. These data suggest that GSK3 can regulate fast anterograde axonal transport and targeting of cargos to specific subcellular domains in neurons.

PubMed Disclaimer

Figures

Fig. 1. Kinesin heavy and light chain isoforms show different phosphorylation patterns in vivo. (A) Autoradiographs of in vivo phosphorylation of axonal kinesin heavy (HC) and light chains (LC). Kinesin was immunoprecipitated from optic nerves labeled by fast axonal transport with 35S (lanes 2 and 3) or 32P (lanes 5 and 6). Proteins were visualized by Coomassie Blue (CB), fluorography (35S) or autoradiography (32P). Arrowheads indicate the position of IgG heavy and light chains. Purified bovine brain kinesin is included for comparison (lanes 1 and 4). (B) Peptide maps of 35S- and 32P-labeled KLC (lanes 1 and 2) and KHC (lanes 3 and 4) from optic nerve confirm the identification of kinesin phosphopeptides.

Fig. 2. Kinesin light chains released from vesicles are phosphorylated. GSK3 is a kinase that specifically binds to kinesin and phosphorylates KLCs. (A) Washing vesicles with control buffers (–ATP) has little effect on the amount of kinesin (HC and LC, left) recovered in vesicle fractions (P), but incubation of vesicles with 1 mM ATP (+ATP) releases most kinesin (HC and LC, right) into the supernatant (S). Kinesin is released from vesicles in an ATP-dependent process, and released light chains (LC) migrate at a higher apparent molecular weight. (B) Increased apparent molecular weight of released KLC is due to phosphorylation. Immunoblots of kinesin vesicle fractions incubated without (lane 1) and with (lane 2) ATP. Lane 3 is with ATP but was treated with alkaline phosphatase prior to electrophoresis. Removal of phosphates eliminates the shift. (C) GSK3 specifically phosphorylates KLC subunits. SyproRed (SR) staining and western blotting (WB) with antibodies against KHCs (H2) and KLCs (63-90) indicate purity of rat brain kinesin. KLCs are phosphorylated by recombinant GSK3 (32P left lane). GSK3 is also autophosphorylated (32P right lane, no kinesin added). (D) Phosphorylation of the KLC2 C-terminal tail (KLC2t) requires pre-phosphorylation by CKII. Recombinant KLC2t was incubated with GSK3 and [γ-32P]ATP before (left lanes) or after (right lanes) incubation with CKII and unlabeled ATP. GSK3 phosphorylated KLC2t only after pre-phosphorylation with CKII. This confirms that KLC2t is a substrate for GSK3 and that kinesin purified from brain must be pre-phosphorylated at a priming site (see Table I). (E) GSK3 and kinesin co-immunoprecipitate. Brain lysate (lane 1), control immunoprecipitates with beads only, and beads/secondary antibody (lanes 2 and 3, respectively) and kinesin antibodies (lane 4) were immunoblotted with antibodies against KHC, GSK3 and PKA.

Fig. 3. GSK3 and kinesin co-localize on a subpopulation of MBOs. (A) Immunoblots show kinesin, GSK3β, PKB, PKA and synaptic vesicle markers (Syn = synapsin and p38 = synaptophysin) in three vesicle fractions (V0, V1 and V2) and remaining supernatant (Cyt). GSK3 and kinesin are highly enriched in V1. (B) The upper panel shows immunoblots of V1 vesicles in continuous sucrose gradient fractions (0.32–2 M sucrose). Each fraction was diluted and pelleted to ensure that only vesicle-bound proteins were analyzed. Kinesin and GSK3β were present on a subset of V1 membranes of similar density. Scans show partial overlap of peaks. Arrows indicate peaks for kinesin (dark) and GSK3β (lighter). (C) KHC and GSK3β antibodies each labeled a single band in rat brain (R) and squid optic lobe (S). (D–F) Kinesin and GSK3 co-localize in double label immunofluorescence. Punctate immunostaining patterns for GSK3β (D) and kinesin (E) extensively overlap in the same cellular regions (F shows merge of D and E) in BHK21 cells. Scale bar: 25 µm. For higher resolution localization, GSK3β (G) and kinesin (H) distributions were determined in squid axoplasm. Both show punctate distributions characteristic of vesicles in squid axoplasm (Pfister et al., 1989; Stenoien and Brady, 1997). Most vesicles stained for both kinesin and GSK3β (I shows merge of G and H). However, some contained only kinesin (asterisk within inserts) or only GSK3β. Inserts in top right corners (G–I) are 4× digital magnifications of areas delineated by white squares.

Fig. 4. GSK3 was enriched in microtubule domains. Primary cultured hippocampal neurons (A–C) and 3T3 fibroblasts (D–F) were triple stained for GSK3β (A and D), tubulin (B and E) and actin (C and F). Perikarya and proximal neurites of a hippocampal neuron (A–C) and a 3T3 fibroblast (D–F) are shown. Scale bars in (C), (F) and (H) are 25 µm. GSK3β immunostaining is punctate and localized preferentially in MT-rich regions. (G–H) In detergent-permeabilized BHK21 cells, GSK3β staining aligned with MTs. Higher magnification (inserts of areas marked by arrows) revealed punctate staining closely matching MT distribution (asterisk). Scale bar for the insert is 3 µm. (I) GSK3β co-pellets with taxol-stabilized MTs. Total rat brain homogenate (lane 1) was centrifuged at 100 000 g. The supernatant (lane 2) was incubated with taxol and GTP then recentifuged to obtain an MT pellet (lane 3) and a final supernatant (lane 4). Immunoblots show GSK3α/β and MAP1A enriched in the MT pellet. PKB was absent from MT fractions.

Fig. 5. Active GSK3β inhibited fast anterograde transport in isolated axoplasm. Each data point represents an average rate for particles moving in the specified direction in a field as described previously (see Brady et al., 1985, 1990a; McGuinness et al., 1989; Ratner et al., 1998). Anterograde (black arrows) and retrograde (gray arrows) transport were measured concurrently. Retrograde transport was unaffected by any of these treatments. (A) Perfusion of axoplasm with 10 nM GSK3β selectively inhibits anterograde axonal transport. The number of organelles that appear to move in anterograde transport was also reduced. (B) Perfusion with inactive, autophosphorylated GSK3 did not affect transport. (C) Co-perfusion of active GSK3 with millimolar CREB phosphopeptide (a specific GSK3β substrate that acts as a competitive inhibitor) abolished the inhibitory effects of GSK3β on transport. (D) Perfusion of axoplasm with PKA catalytic subunit had no effect on fast axonal transport in either direction. (E) Mean rates of particle movement for all measurements between 30 and 40 min (30m) with control buffers or PKA were not significantly different from preperfusion rates (0m) in either anterograde (gray bars on left) or retrograde (black bars on right) directions. Addition of GSK3β reduced anterograde transport at 30–40 min by 44% (#, significant at P ≤0.0001). Addition of GSK3 with 1 mM CREB phosphopeptide, a competitive inhibitor of GSK3, abolished the effect of GSK3 on anterograde transport. No other differences between experimental and matched control were statistically significant (P ≤0.05).

Fig. 6. GSK3β phosphorylation of kinesin light chain releases kinesin from membranes without altering MT binding or ATPase activity. (A) Phosphorylation of kinesin by GSK3β did not affect kinesin interaction with MTs. GSK3β alone (lane 1) or GSK3β and kinesin (lanes 2–5) were labeled by incubation with [32P]ATP, then 80 ng aliquots were incubated with taxol-stabilized bovine brain MTs and either ATP (duplicate experiments shown in lanes 2 and 3) or AMP-PNP (duplicate experiments shown lanes 4 and 5), then pelleted. No effects on MT binding were seen between kinesin with KLCs phosphorylated by GSK3 and control kinesin fractions. (B) MT-activated ATPase activity of kinesin was also unaffected by GSK3 phosphorylation. Bars represent the percentage of ATP hydrolyzed by control, non-phosphorylated (white bars) and GSK3β-phosphorylated (gray bars) rat brain kinesin (n = 3). (C and D) Perfusion of active GSK3 in squid axoplasm dramatically reduced kinesin bound to membranes. (C) Membrane fractions from squid axoplasms perfused with buffer (Control), and with active GSK3 (GSK3) were immunoblotted for kinesin (KHC) and actin. (D) Quantitative immunoblots for kinesin and actin with [125I]protein A; values were quantitated by phosphoimager and kinesin levels normalized relative to actin (n = 4). GSK3 treatment reduced membrane-bound kinesin by 70%.

Fig. 7. GSK3 was increased at sites of active membrane delivery. A hippocampal neuron (A–C) and a differentiating PC12 cell (D–F) were double immunostained for GSK3β (A and D) and tubulin (B and E). In both cell types, the level of GSK3 was high in cell bodies and lower amounts were distributed along neurites. Merged images (C and F) highlight relative enrichment of GSK3β relative to tubulin at neurite tips (arrows). GSK3 immunoreactivity appears concentrated distal to MT ends. (G–J) High magnification of a triple-stained PC12 growth cone. Anti-GSK3β labels vesicle-like structures in the growth cone core (G), with a pattern distinct from MTs (H) and actin (I). In the merged image (J), the relationship of GSK3β to cytoskeletal elements is emphasized. Asterisks serve as reference points. Immunolocalization with an antibody that recognizes GSK3β independently of phosphorylation (active and inactive forms) shows enrichment of immunoreactivity in growth cones (K and G). In contrast, an antibody that only recognizes GSK3β phosphoSer9 (inactive kinase) (Wang et al., 1994a) is not enriched in growth cones (L). PhosphoSer9 immunoreactivity is abolished by phosphatase treatment (M), while it has no effect on staining with the phosphorylation-independent GSK3β antibody (K). Immunoblotting on proteins isolated from growth cones (N) confirmed the results of immunostaining. Lane 1 contains homogenate from E18 rat brain, lane 2 a 3000 g supernatant from E18 brain and lane 3 purified growth cone particles. The antibodies used on each of the four blots are indicated on the right.

Fig. 8. A simple model for the action of GSK3 on kinesin function. KLCs on many MBOs are pre-phosphorylated. When GSK3 is activated by action of one or more phosphatases, it phosphorylates KLCs further. Phosphorylation by GSK3 increases the accessibility of the J-domain motifs on KLCs to hsc70 (Tsai et al., 2000) and leads to removal of kinesin from a vesicle. Released kinesin appears to be degraded rapidly (Li et al., 1999) and vesicles become available for insertion into the plasma membrane. Given that membrane receptor pathways can regulate GSK3 activity locally, this would act as a targeting mechanism for membrane proteins to regions of local GSK3 activation. Local changes in kinase/phosphatase activity have been implicated in specific targeting of ion channels to nodes of Ranvier (de Waegh et al., 1992).

References

1. Baum L., Hansen,L., Masliah,E. and Saitoh,T. (1996) Glycogen synthase kinase 3 alteration in Alzheimer disease is related to neurofibrillary tangle formation. Mol. Chem. Neuropathol., 29, 253–261. - PubMed
1. Bloom G.S., Richards,B.W., Leopold,P.L., Ritchey,D.M. and Brady,S.T. (1993) GTPγS inhibits organelle transport along axonal microtubules. J. Cell Biol., 120, 467–476. - PMC - PubMed
1. Brady M.J., Bourbonais,F.J. and Saltiel,A.R. (1998) The activation of glycogen synthase by insulin switches from kinase inhibition to phosphatase activation during adipogenesis in 3T3-L1 cells. J. Biol. Chem., 273, 14063–14066. - PubMed
1. Brady S.T. (1985) A novel brain ATPase with properties expected for the fast axonal transport motor. Nature, 317, 73–75. - PubMed
1. Brady S.T. (1991) Molecular motors in the nervous system. Neuron, 7, 521–533. - PubMed

Glycogen synthase kinase 3 phosphorylates kinesin light chains and negatively regulates kinesin-based motility - PubMed (original) (raw)