XB-ART-11149J Biol Chem. April 28, 2000; 275 (17): 12701-11.
Ca(2+)/calmodulin-dependent protein kinase II is stimulated by Wnt and Frizzled homologs and promotes ventral cell fates in Xenopus.
Wnt ligands working through Frizzled receptors have a differential ability to stimulate release of intracellular calcium (Ca(2+)) and activation of protein kinase C (PKC). Since targets of this Ca(2+) release could play a role in Wnt signaling, we first tested the hypothesis that Ca(2+)/calmodulin-dependent protein kinase II (CamKII) is activated by some Wnt and Frizzled homologs. We report that Wnt and Frizzled homologs that activate Ca(2+) release and PKC also activate CamKII activity in Xenopus embryos, while Wnt and Frizzled homologs that activate beta-catenin function do not. This activation occurs within 10 min after receptor activation in a pertussis toxin-sensitive manner, concomitant with autophosphorylation of endogenous CamKII. Based on data that Wnt-5A and Wnt-11 are present maternally in Xenopus eggs, and activate CamKII, we then tested the hypothesis that CamKII participates in axis formation in the early embryo. Measurements of endogenous CamKII activity from dorsal and ventral regions of embryos revealed elevated activity on the prospective ventral side, which was suppressed by a dominant negative Xwnt-11. If this spatial bias in CamKII activity were involved in promoting ventral cell fate one might predict that elevating CamKII activity on the dorsal side would inhibit dorsal cell fates, while reducing CamKII activity on the ventral side would promote dorsal cell fates. Results obtained by expression of CamKII mutants were consistent with this prediction, revealing that CamKII contributes to a ventral cell fate.
PubMed ID: 10777564
Article link: J Biol Chem.
Genes referenced: camk2g gsc myc prl.1 tbxt wnt11b wnt5a
Article Images: [+] show captions
|FIG. 1. Activation of Ca21/calmodulin-dependent protein kinase II by members of the Wnt and Frizzled family. Panel A, Ca21- independent CamKII activity in Xenopus cytoplasmic extracts at stage 7. Xenopus embryos were injected with prolactin (control), Xwnt-8, Xwnt-5A, Xwnt-11, Rfz-1, Rfz-2, Mfz-3, Mfz-4, Mfz-6, Mfz-7, or Mfz-8 RNA as indicated and cytoplasmic extracts were assayed as described under “Experimental Procedures” for their ability to phosphorylate Syntide-2 in the presence or absence of specific CamKII inhibitors. The specific Ca21-independent CamKII activity determined in the presence of EGTA is shown. n 5 number of experiments. Panel B, total activity of CamKII in cytoplasmic extracts at stage 7. Total CamKII activity of cytoplasmic extracts of injected embryos was measured in the presence of exogenous Ca21. n 5 number of experiments. Inset of panel B, Xwnt-5A and Xwnt-8 are equally expressed in injected Xenopus embryos. Prolactin (lane 1), Xwnt-8-myc (lane 2), or Xwnt-5A-myc (lane 3) (1 ng), respectively, were injected into embryos, and cytoplasmic extracts were immunoblotted with an anti-c-Myc monoclonal antibody, 9E10. Panel C, Xwnt-5A and Rfz-2 synergistically activate CamKII. Xenopus embryos were injected with low doses of Xwnt-5A or Rfz-2 and the specific CamKII activity was measured at stage 7. In combination, both factors gave a more than additive activation (predicted activation shown by dotted line) of CamKII, n 5 number of experiments. Panel D, Autophosphorylation of CamKII is increased in Xwnt-5A injected embryos. Xenopus embryos were injected with prolactin or Xwnt-5A RNAs as in A, then CamKII was immunoprecipitated with a CamKII-specific monoclonal antibody. Immunoprecipitates were blotted and probed with a polyclonal antibody recognizing the Thr286-phosphorylated form of CamKII (left panel). Reprobing this blot with a CamKII specific antibody shows equal loading of different lanes (right panel).|
|FIG. 2. Direct activation of CamKII by inducible Frizzled chimeras. Panel A, the cartoon denotes that cytoplasmic domains were derived from Frizzled homologs. Panel B, treatment of cells expressing Rfz-2/b2-AR with an agonist (isoproterenol) but not antagonist (propranolol) for 10 min induces CamKII activity in a manner blocked by pertussis toxin (Ptx). Rfz-1/b2-AR does not activate CamKII even in the presence of agonist.|
|FIG. 3. Effect of injected RNA probes on endogenous CamKII activity. Panel A, influence of pertussis toxin, kinase dead CamKII (CamKII K42M), dominant negative Xwnt-11 (dnXwnt-11), and dominant negative Xwnt-8 (dnXwnt-8) RNA on Wnt induced CamKII activity. Embryos were injected with RNAs and doses (ng) as indicated and the specific Ca21-independent CamKII activity was determined. n, number of experiments. The increase in CamKII activity (above dotted baseline) upon injection of Xwnt-5A is sensitive to pertussis toxin, kinase dead CamKII, and dnXwnt-11 but not dnXwnt-8. Panel B, influence of constitutively active CamKII (CamKII T286D, 1 ng), kinase dead CamKII (CamKII K42M, 1–2 ng), dnWnt-11 (1 ng), pertussis toxin A protomer (Ptx, 0.375 ng), and control prolactin (1 ng) RNA injections on endogenous CamKII activity. Embryos were injected with RNAs at the 4-cell stage into both dorsal (upper panel) or both ventral blastomeres (lower panel). At stage 7 embryos were cut into dorsal (M) and ventral (u) halves and immediately assayed for specific CamKII activity as described in the legend to Fig. 1. Endogenous levels of CamKII activity in untreated dorsal (d) or ventral (v) halves of stage 7 embryos are given as comparison after normalization to protein content. n 5 number of experiments. Panel C, dominant negative Xwnt-11 (dnWnt-11), but not dominant negative Xwnt-8 (dnWnt-8), can block Xwnt-5A induced relocalization of PKC. Confocal image of stage 8 animal cap tissue from Xenopus embryos injected at the 2-cell stage into both cells with XPKCa-myc (0.25 ng), Xwnt-5A (0.13 ng), prolactin (1 ng), dnXwnt-8 (1 ng), or dnXwnt-11 (1 ng) RNA as indicated. Detection of the c-myc epitope to monitor PKC localization was as described (12).|
|FIG. 4. Dorso-ventral differences in endogenous CamKII activity in Xenopus embryos. Panel A, embryos were cut into ventral (u) and dorsal (M) halves at stage 7 and 10, respectively, and assayed for CamKII activity as described in the legend to Fig. 1. Ventral halves of stage 7 and stage 10 Xenopus embryos show a higher Ca21-independent CamKII activity than dorsal halves (left panel). Total kinase activity at stage 7, determined in the presence of Ca21, is the same in dorsal and ventral halves (right panel). Given CamKII activities were normalized to protein content. n 5 number of experiments. Panel B, spatial asymmetry in CamKII activity is accompanied by an enrichment of the autophosphorylated form of CamKII on the ventral side of the embryo at stage 7. Xenopus embryos were cut into dorsal (d) and ventral (v) halves before CamKII was immunoprecipitated with a CamKII-specific monoclonal antibody. Immunoprecipitates were blotted and probed with a polyclonal antibody recognizing the Thr286-phosphorylated form of CamKII (upper P-CamKII blot). Reprobing these blots with a CamKII specific antibody shows equal amounts of CamKII in dorsal and ventral halves (lower CamKII blot).|
|FIG. 5. Effects of expressing CamKII mutants in Xenopus embryos. Panel A, dorsal but not ventral injection of the constitutively active mutant CamKII T286D (1 ng) results in ventralization of embryos which is accompanied by loss of endogenous goosecoid (gsc) expression at stage 10, and a loss of notochord formation at stage 30 (see Table II). Ventral but not dorsal injection of the kinase dead mutant CamKII K42M (1 ng) results in axis duplication at a low incidence accompanied by a second domain of gsc expression at stage 10 (see Tables II and III for details). Injection of prolactin (PL) control RNA (1 ng) had no effects on either axis formation or on gsc expression. Arrowheads denote dorsal lips. Panel B, dorsal injection of the constitutively active mutant results in up-regulation of ventral marker genes (compare arrowheads of Xvent-2, Xvent-1, Xwnt-8, and GATA-2 relative to arrowheads in control lacZ embryos) but not of the pan-mesodermal marker Xbra in stage 10 embryos. Ventral injection of the kinase dead mutant of CamKII results in decrease or loss of staining for ventral markers but not of Xbra. lacZ injections had no effect.|
|FIG. 6. Effects of overexpressing dnXwnt-11 in Xenopus embryos. Panel A, phenotypic analysis of normal embryos injected with dnXwnt-11 RNA (1.0 ng) into both ventral blastomeres at the 4-cell stage (upper panel) or into two cells of the 4-cell stage embryos following UV ventralization (lower panel). Arrows indicate the formation of ectopic cement glands. Injection of lacZ RNA served as control. Panel B, ventral injection of dnXwnt-11 RNA results in down-regulation of the ventral marker gene Xvent-1 but not of Xvent-2 in stage 10 embryos. lacZ served as control.|