Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi

Glycogen Synthase Kinase-3

Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_378

Synonyms

Historical Background

Glycogen synthase kinase-3 (GSK-3) is a highly conserved protein-serine/threonine kinase that was first isolated from skeletal muscle in 1980 as one of five enzymes capable of phosphorylating glycogen synthase (Embi et al. 1980). In resting tissues, GSK-3 phosphorylation inhibits glycogen synthase, the rate-limiting enzyme of glycogen synthesis. In subsequent work, insulin was found to cause inactivation of GSK-3 (via induction of PKB/Akt) and this relieves the suppression of glycogen synthase, leading to enhanced glucose conversion into glycogen in response to insulin. In mammals GSK-3 is encoded by two genes that generate highly related proteins termed GSK-3α and GSK-3β that have molecular masses of 51 and 46 kDa, respectively. In brain, the GSK-3β gene is alternatively spliced to generate a version with a 13 amino acid insert termed GSK-3β2. In addition to its initial role in glycogen metabolism, GSK-3 has since been demonstrated to regulate many cellular processes by suppressing substrates and to be regulated itself by several upstream pathways. As exemplified by insulin, extracellular signal-dependent control of GSK-3 then relieves its blockade and allows targets to become active. Hence, this protein kinase acts as a regulatory switch that throttles the flux through numerous signaling pathways initiated by diverse stimuli (reviewed in: Forde and Dale 2007; Doble and Woodgett 2007). Deregulation of GSK-3 has been implicated in the development of cancer, diabetes, Alzheimer’s disease, schizophrenia, and bipolar disorder. Indeed, GSK-3 is a validated target of lithium, which is frequently prescribed to patients with bipolar disorder (Klein and Melton 1996). Given its involvement in many pathophysiological processes, GSK-3 is a tempting therapeutic target and several pharmaceutical companies have active GSK-3 inhibitor discovery programs. However, its proclivity for involvement in multiple pathways also raises issues of selectivity. It is therefore imperative to assess the full spectrum of GSK-3 cell functions to predict clinical efficacy and avoid surprises.

Signaling Pathways Involving GSK-3

  Phosphoinositide 3-kinase (  PI3K ): The first pathway biochemically shown to directly regulate GSK-3 activity was  phosphoinositide 3-kinase ( PI3K). This pathway is induced by a wide variety of stimuli, including insulin and polypeptide mitogens. Activating mutations in  PI3K are frequently observed in human cancers and inactivation of  PTEN, the lipid phosphatase that reverses the action of  PI3K, is also found, albeit less commonly, in tumors (and is responsible for the cancer predispositions of Cowden’s disease).  PI3K activation of PKB/Akt leads to phosphorylation of GSK-3α and β at serine 21 and 9, respectively, leading to their partial inactivation. Although rarely mutated in human cancers, PKB/Akt was first identified as an oncogene. Cancer-associated  PI3K mutations have also been linked to activation of serum and glucocorticoid-activated kinase 3 (SGK3), which also phosphorylates GSK-3.  p38 mitogen-activated protein kinase (p38 MAPK) can also inactivate GSK-3β via phosphorylation of Ser389 and Thr390 and this mode of regulation has been observed in neurons and thymocytes (Thornton et al. 2008).

Cyclic AMP: This second messenger is induced primarily in response to seven-transmembrane pass receptors that signal through G proteins that couple to  adenylyl cyclase, as exemplified by the β-adrenergic/epinephrine receptors. Generated cyclic AMP binds to the regulatory subunits of cyclic AMP-dependent protein (PKA), causing dissociation from and relief of inhibition of the catalytic subunits. These phosphorylate a number of substrates, including GSK-3α and β at the same sites as PKB/Akt (Li et al. 2000). Complicating matters further, certain other protein-serine kinases such as p70 S6 kinase (S6 K) are also capable of phosphorylating GSK-3 at these same sites.

Wnt: The canonical Wnt pathway acts primarily by raising cytoplasmic levels of β-catenin that then translocates to the nucleus to regulate gene expression in concert with LEF/TCF DNA-binding proteins. In unstimulated cells, cytoplasmic β-catenin levels are kept very low via phosphorylation by GSK-3 at three residues that target the protein for ubiquitinylation by the E3 ligase, βTrCP and subsequent destruction by the 26S proteasome. A small fraction (<10%) of cellular GSK-3 (α and β) and a priming kinase (casein kinase 1; CK1) are specifically associated with a “destruction complex” that comprises the adenomatous polyposis coli (APC) tumor suppressor and the Axin scaffolding protein. This complex constantly recruits newly synthesized β-catenin and targets it for degradation. In the presence of a suitable Wnt ligand, the LRP5/6 co-receptors become phosphorylated by GSK-3 and CK1, which creates a high-affinity binding site for Axin (Zeng et al. 2005). Through a poorly understood mechanism, the destruction complex reorganizes as a consequence allowing β-catenin to escape phosphorylation and, hence, accumulate. One of the primary transcriptional targets of β-catenin is the gene for Axin2, the polypeptide product of which acts to restore the function of the destruction complex leading to only transient Wnt signaling. APC is inactivated in ~70% of colon cancers and a further 15% show mutations in the phosphorylation site region of β-catenin. Axin is also mutated in several cancers including hepatocellular carcinomas, as is β-catenin. In each of these tumors, β-catenin is no longer subject to negative control and its levels and transcriptional activity increase dramatically (20-fold or more). Inactivation of both alleles of both GSK-3 genes results in a similar effect but disruption of only 3 of the 4 GSK-3 alleles has little effect on β-catenin regulation highlighting the fact that only a small fraction of GSK-3 is associated with the destruction complex and is this relevant to Wnt signaling (Doble et al. 2007). This selective sequestration/secondment of small fractions of GSK-3 to specific compartments underlies the principle by which GSK-3 participates in several distinct signaling pathways while maintaining signal specificity.

Notch: Activation of  Notch receptors via their interaction with specific Delta-like and Jagged ligands leads to ectodomain truncation of  Notch by Adam-like metalloproteinases followed by juxtamembrane cleavage by a γ-secretase complex and formation of an intracellular domain fragment ( NotchICD). This signaling fragment binds to a DNA-binding protein termed CSL or RJBPκ, which transactivates genes including members of the Hes/Hey and Sonic Hedgehog (Shh) family. Activated  Notch signaling in ES cells (such as induced by expression of  NotchICD) can increase stem cell proliferation, drive neural cell fates and, in later stages of development, modulate differentiation of endothelial and epithelial cell structures as well as hematopoietic stem cell differentiation and maintenance of myogenic satellite cells (Androutsellis-Theotokis et al. 2006). Deregulated expression of  Notch1 is also associated with T-cell acute lymphoblastic leukemia. Inhibition of GSK-3 leads to stabilization of  NotchICD. Thus, GSK-3 likely acts to downregulate Notch signaling once it has been initiated.

Hedgehog: The mammalian Hedgehog pathway is initiated by three polypeptide ligands (Sonic, Desert, and Indian Hedgehog) that activate Smoothened (Smo) serpentine protein to regulate the Gli family of transcriptional activators (reviewed in Jiang and Hui 2008). Briefly, in resting cells, a transmembrane protein termed Patched inhibits Smo allowing several protein kinases including GSK-3 to phosphorylate Gli and target it for cleavage into a repressor form (GliR). When present, processed Hedgehog ligand binds to Patched and relieves inhibition of Smo. This consequently promotes the activation of Gli (GliA) through a complex process that includes interference of Gli phosphorylation/cleavage and inactivation of a suppressor protein termed Sufu. GliA then induces expression of various target genes (including itself and snail). Ectopic activation of Hedgehog signaling can lead to basal cell carcinomas and medulloblastomas.

These pathways each play important, fleeting, and often temporally interdependent roles in normal development. All share the commonality of frequent deregulation in cancer via chronic activation and loss of feedback control. The fact that GSK-3 plays a fundamental role in suppressing signaling in each of these pathways is remarkable and, at the same time, rather puzzling – representing a significant potential vulnerability for loss of control without obvious rationale for burdening a single protein kinase with the responsibility of keeping multiple important pathways in check.

Regulatory Quirks of GSK-3

While the catalytic domains of GSK-3α and β products are virtually identical in their protein kinase domains (excluding the possibility of isoform-selective small molecule ATP-binding site inhibitors), they share only 36% identity in the last 76 C-terminal residues (Woodgett 1990). GSK-3α also has a glycine-rich N-terminal extension compared to the β-isoform. In addition to mammals, the genomes of many other species such as fish, amphibians, and lizards encode genes for both isoforms, although bird genomes only contain GSK-3β and appear to have selectively lost GSK-3α (Allon et al. 2011).

Although GSK-3 phosphorylates its many substrates on serine or threonine residues, it transiently expresses tyrosine kinase activity toward itself upon initial folding (Lochhead et al. 2006). The phosphorylated tyrosine lies within the “T loop” in kinase subdomain VIII (Tyr279 for GSK-3α and Tyr216 for GSK-3β). The role of this phosphotyrosine is to allow the kinase domain to adopt an active conformation for exogenous substrates. Analysis of the crystal structure of GSK-3β revealed that unphosphorylated Tyr216 is in a conformation that interferes with substrate access whereas the phosphorylated tyrosine moves out of the path. Hence, unlike serine phosphorylation of the N-terminal region of GSK-3, tyrosine phosphorylation promotes activity. This posttranslational modification appears quite stable and unregulated. Treatment of cells with GSK-3 inhibitors causes slow loss of the phosphotyrosine at a rate consistent with its half-life. Hence, in normal conditions, the residue is phosphorylated stoichiometrically and remains so for its natural life (Lochhead et al. 2006).

GSK-3 has a strong preference for substrates that are already phosphorylated at a proximal serine/threonine lying C-terminal to the GSK-3 target residue. The so-called priming site of the substrate slots into a pocket comprising of three basic amino acids – Lys205, Arg96, and Arg180. Binding of the priming phosphate into this pocket on GSK-3 induces a conformational change, aligning the substrate for subsequent phosphorylation. Most substrates of GSK-3 demonstrate an absolute requirement for prior phosphorylation by another kinase at the “priming” residue located C-terminal to the site of subsequent phosphorylation by GSK-3 (at least 500-fold preference for the phosphorylated form). GSK-3 phosphorylates these substrates at the fourth or fifth serine or threonine residue N-terminal to the primed site (pS/TXXXpS/T), where the first pS/T (Ser or Thr) is the target residue, X is any amino acid (often Pro), and the last pS/T is the site for priming phosphorylation. Several protein kinases act as priming enzymes for GSK-3 substrates, including CDK-5, PAR-1, casein kinase I,  casein kinase II, PKA, and PKC. In several cases, the residue initially phosphorylated by GSK-3 acts as a priming phosphate to induce phosphorylation of a second residue N-terminal to it. This leads to a series of phosphorylated residues as observed in glycogen synthase (priming kinase:  casein kinase II) and β-catenin (priming kinase: casein kinase I).

Many substrates have been identified for GSK-3 to various degrees of rigor. These fall into three general categories: metabolism and regulation; structural/cytoskeletal; transcriptional regulators (Table 1). It is important to note that individual signals that regulate GSK-3 do not change the phosphorylation state of all of its targets. Rather, subsets of substrates are impacted due to local sequestration and other mechanisms.
Glycogen Synthase Kinase-3, Table 1

GSK-3 substrates arranged into functional subgroups

Metabolism

ATP-citrate lyase

Protein phosphatase 1

Glycogen synthase

Protein phosphatase 1 inhibitor-2

Pyruvate dehydrogenase

Acetyl CoA carboxylase

Signaling proteins

Axin

Cyclic-AMP-dependent protein kinase – RII subunit

Adenomatous polyposis coli

Cyclin D1 and E

Cubitus interruptus/Gli

Eukaryotic initiation factor 2B

Per2

Cry2

Presenilin-1

Nucleoporin p62

p21 cdk inhibitor

Lipoprotein receptor-related protein 5/6

Insulin receptor substrate 1 and 2

Tuberous sclerosis 2

NGF receptor

Mcl-1

Structural proteins

 Ncam

Collapsin response mediator proteins 2 and 4

Dynamin-like protein

Neurofilament heavy subunit

Kinesin light chain

Microtubule-associated protein 1B and 2

Ninein

Telokin (KRP) (kinase-related protein)

Tau

CLIP-associated protein 1 and 2

Paxillin

DF3/MUC1 (mucin-like glycoprotein)

Transcription factors

β-catenin

Nuclear factor of activated T cells

Bcl3

CCAAT/enhancer-binding proteins a and β

Snail

Cyclic AMP response element-binding protein

GATA4

Microphthalmia-associated transcription factor

hypoxia-inducible factor 1

Glucocorticoid receptor

Heat shock factor-1

Bcl-2 interacting transcriptional repressor

c- Myc and L- Myc

SMAD1

c-Jun

 NF-κB (p65 and p105)

JunD

Notch

 P53

c-Myb

Genetic Analysis of GSK-3 Functions

As mentioned above, mammals have two GSK-3 genes that encode two distinct proteins: GSK-33α and GSK-3β (Woodgett 1990) (Fig. 1). GSK-3β is also differentially spliced in the brain to generate a minor (~15% of total) isoform termed GSK-3β2, which contains a 13 amino acid insert within the kinase domain (exon 8A). GSK-3 orthologues have been found in all eukaryotes examined to date, including genetically tractable organisms. While structurally highly related, GSK-3α and GSK-3β are not functionally equivalent. This became obvious upon the generation of GSK-3β knockout mice (Hoeflich et al. 2000). Embryos carrying homozygous deletions of exon 2 of GSK-3β die between embryonic day 16.5 and birth due to liver degeneration caused by defects in NF-κB responses to maternal infection or (at birth) to patterning defects in the heart that preclude lung circulation of oxygenated blood. GSK-3α null animals, by contrast, are viable and demonstrate that GSK-3β can substitute for a but not vice versa (MacAulay et al. 2007). GSK-3α mutant mice do exhibit a variety of phenotypes including insulin sensitization and behavioral abnormalities. Conditional alleles of GSK-3α and β have also been generated. The first analysis of these animals involved rat insulin promoter-driven Cre that expresses the recombinase only in β-islet cells. These animals only lacked GSK-3β in the β-cells and had normal insulin responses. However, β-islet cell inactivation of GSK-3β rescued insulin resistance of insulin receptor substrate-2 mutant animals as well as insulin resistance caused by haploinsufficiency of the insulin receptor itself. Although the GSK-3α global knockout mice were insulin-sensitized, this effect was largely restricted to the liver, which exhibited significantly higher efficiency in glycogen deposition and was also dependent on strain background. Since GSK-3β null animals are inviable, floxed GSK-3β animals with either albumin-Cre or myosin light chain kinase-Cre have been generated to inactivate the gene in liver or skeletal muscle, respectively. These mice revealed inverse effects to the GSK-3α mutants, such that knocking out GSK-3β in the liver had no impact, whereas the animals lacking the enzyme in skeletal muscle were insulin-sensitized. These unanticipated results demonstrated that not only are the two isoforms of GSK-3 nonredundant, they also have tissue-specific roles.
Glycogen Synthase Kinase-3, Fig. 1

Structural architecture of GSK-3 genes and proteins. Panel A. Chromosomal organization of mouse GSK-3 genes. Coding exons are red, noncoding exons blue. The exon flanked by LoxP sites (exon2) is indicated. In the middle is a schematic topology of the two protein isoforms. Pale yellow boxes indicate the kinase domain, green boxes indicate the unconserved regions. The PKB/Akt phosphorylation site (Ser21/Ser9 in GSK-3α/β respectively) is shown by a red-circled “P.” Below this are protein sequence alignments of mouse GSK-3α and GSK-3β, color-coded as above

An allelic series of embryonic stem cells have been engineered that are deficient in the various alleles of GSK-3α and GSK-3β. GSK-3 nullizygous ES cells proliferate normally but were highly refractive to differentiation cues and maintained expression of the pluripotency-associated transcription factors nanog and oct-4 even when grown as embryoid bodies in the absence of LIF (Doble et al. 2007). Likewise, treatment of mouse or human ES cells with GSK-3 inhibitors maintains pluripotency (Sato et al. 2004). GSK-3 therefore plays a critical role in supporting the capacity of embryonic stem cells to differentiate and, in its absence, these cells are effectively locked in a “ground state” (Ying et al. 2008).

Consequences of total inactivation of GSK-3: The block to differentiation in ES cells upon inactivation of GSK-3 is recapitulated upon genetic inactivation of the kinase in more differentiated or lineage-committed cells. For example, expression of Cre recombinase in presumptive neuronal precursor cells (via nestin-Cre) causes greatly enhanced neuronal precursor populations (as marked by Sox2) at the expense of differentiated neurons (Kim et al. 2009). These animals died at birth. Activation of the Wnt,  Notch, Hedgehog, and PI-3 K pathways downstream of GSK-3 was observed in the nullizygous neurons. Treatment of primary neuronal cultures with inhibitors (dominant-negative TCF4, γ-secretase inhibitor, cyclopamine and wortmannin, respectively) to these individual pathways revealed partial dependence on each for proliferation, with only a combination of the inhibitors achieving full suppression. This study also noted a loss of polarized cell division in dividing neurons in the subventricular zone, a process that is required for efficient stratification of the neuronal layers, suggesting an important role for GSK-3 in regulating spindle pole orientation. Similar suppression of differentiation to that observed in ES cells and neuronal precursors has been observed in other cell types and tissues in which all alleles of GSK-3 has been selectively genetically inactivated (Woodgett lab, Unpublished observations).

GSK-3 and Cancer Etiology

These animal models each share a similar property: differentiation is inhibited and there is expansion of precursor populations. These findings have implications for in vitro expansion of adult stem/progenitor cells in regenerative medicine but also provide insight into the early stages of tumor development. GSK-3 has been implicated in cancer at various levels, primarily as a tumor suppressor. Its inhibitory role in the major signaling pathways described above that are most commonly activated in human cancers is well documented. However, the kinase has also been positively linked to pancreatic cancer and regulation of Snail, a transcription factor key to epithelial to mesenchymal transitions (EMT), a process associated with metastasis. A splice mutant of GSK-3β has recently been identified in a population of leukemic stem cells derived from blast crisis chronic myelogenous leukemia (CML) (Abrahamsson et al. 2009). Reintroduction of wild-type GSK-3β into the CML cells reduced engraftment efficiency suggesting a role in promoting stem cell expansion. By contrast, GSK-3 was found to play a tumor-promoting role in MLL-driven leukemias where its inhibition led to regression of the leukemic cells through destabilization of the CDK inhibitor, p27Kip1. In addition, inhibition of GSK-3 has been reported to induce glioma apoptosis through inhibition of  NF-κB and to enhance sorafenib-stimulated apoptosis in melanoma cell lines. Hence, GSK-3’s role in cancer progression remains to be properly elucidated.

Summary

GSK-3 is clearly an unusual protein kinase. It has an inhibitory influence in multiple signaling pathways that are critical for normal development as well as in metabolic control. Its high activity in “resting” tissue suggests its role is generally and suppressive. Signals must relieve the inhibitory action of GSK-3 and there are several means to achieve this including phosphorylation and sequestration. There are many inhibitors of GSK-3, a number of which are potent and relatively specific. Of note, these inhibitors do not discriminate between the isoforms or subcellular fractions. Chemical inhibition of GSK-3 is therefore a blunt tool for investigating the functions of the kinase. It is also recognized that the consequences of GSK-3 inhibition depend on the degree to which activity is suppressed. For example, mice lacking a single GSK-3 allele (equivalent to 25% inhibition) have clearly measurable behavioral phenotypes whereas almost complete inhibition by drugs is required to induce β-catenin stabilization. Lastly, research into this protein kinase has largely focused on the GSK-3β isoform. A search of PubMed identifies over 1300 articles with GSK-3β in the title but only 66 for GSK-3α. This is in spite of the fact that the two isoforms are largely, although not completely, redundant and no available small molecule inhibitors are selective to either isoform. Hence, much of the literature is overly focused on GSK-3β and many data are erroneously attributed to effects of GSK-3β without consideration of the role of GSK-3α. This “β bias” is both unnecessary and unfortunate in that much of the data pertain equally to the less appreciated isoform which likely pulls as much weight in cellular regulation as its sibling.

References

  1. Abrahamsson AE, Geron I, Gotlib J, Dao KH, Barroga CF, Newton IG, Giles FJ, Durocher J, Creusot RS, Karimi M, Jones C, Zehnder JL, Keating A, Negrin RS, Weissman IL, Jamieson CH. Glycogen synthase kinase-3beta missplicing contributes to leukemia stem cell generation. Proc Natl Acad Sci U S A. 2009;106:3925–9.PubMedCrossRefPubMedCentralGoogle Scholar
  2. Allon LT, Pietrokovski S, Barkan S, Avrahami L, Kaidanovich-Beilin O, Woodgett JR, Barnea A, Eldar-Finkelman H. Selective loss of glycogen synthase kinase-3a in birds reveals distinct roles for GSK-3 isozymes in tau phosphorylation. FEBS Lett. 2011;585:1158–62.CrossRefGoogle Scholar
  3. Androutsellis-Theotokis A, Leker RR, Soldner F, Hoeppner DJ, Ravin R, Poser SW, Rueger MA, Bae SK, Kittappa R, McKay RD. Notch signalling regulates stem cell numbers in vitro and in vivo. Nature. 2006;442:823–6.PubMedCrossRefGoogle Scholar
  4. Doble BW, Woodgett JR. Role of glycogen synthase kinase-3 in cell fate and epithelial-mesenchymal transitions. Cells Tissues Organs. 2007;185:73–84.PubMedCrossRefGoogle Scholar
  5. Doble BW, Patel S, Wood GA, Kockeritz LK, Woodgett JR. Functional redundancy of GSK-3alpha and GSK-3beta in Wnt/beta-catenin signaling shown by using an allelic series of embryonic stem cell lines. Dev Cell. 2007;12:957–71.PubMedCrossRefPubMedCentralGoogle Scholar
  6. Embi N, Rylatt DB, Cohen P. Glycogen synthase kinase-3 from rabbit skeletal muscle. Separation from cyclic-AMP dependent protein kinase and phosphorylase kinase. Eur J Biochem. 1980;107:519–27.PubMedCrossRefGoogle Scholar
  7. Forde JE, Dale TC. Glycogen synthase kinase 3: a key regulator of cellular fate. Cell Mol Life Sci. 2007;64:1930–44.PubMedCrossRefGoogle Scholar
  8. Hoeflich KP, Luo J, Rubie EA, Tsao M-S, Jin O, Woodgett JR. Requirement for glycogen synthase kinase-3beta in cell survival and NF-kappaB activation. Nature. 2000;406:86–90.PubMedCrossRefGoogle Scholar
  9. Jiang J, Hui CC. Hedgehog signaling in development and cancer. Dev Cell. 2008;15:801–12.PubMedCrossRefGoogle Scholar
  10. Kim WY, Wang X, Wu Y, Doble BW, Patel S, Woodgett JR, Snider WD. GSK-3 is a master regulator of neural progenitor homeostasis. Nat Neurosci. 2009;12:1390–7.PubMedCrossRefPubMedCentralGoogle Scholar
  11. Klein PS, Melton DA. A molecular mechanism for the effect of lithium on development. Proc Natl Acad Sci U S A. 1996;93:8455–9.PubMedCrossRefPubMedCentralGoogle Scholar
  12. Li M, Wang X, Meintzer MK, Laessig T, Birnbaum MJ, Heidenreich KA. Cyclic AMP promotes neuronal survival by phosphorylation of glycogen synthase kinase 3beta. Mol Cell Biol. 2000;20:9356–63.PubMedCrossRefPubMedCentralGoogle Scholar
  13. Lochhead PA, Kinstrie R, Sibbet G, Rawjee T, Morrice N, Cleghon V. A chaperone-dependent GSK3beta transitional intermediate mediates activation-loop autophosphorylation. Mol Cell. 2006;24:627–33.PubMedCrossRefGoogle Scholar
  14. MacAulay K, Doble BW, Patel S, Hansotia T, Sinclair EM, Drucker DJ, Nagy A, Woodgett JR. Glycogen synthase kinase 3 alpha-specific regulation of murine hepatic glycogen metabolism. Cell Metab. 2007;6:329–37.PubMedCrossRefGoogle Scholar
  15. Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlou AH. Maintenance of pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat Med. 2004;10:55–63.PubMedCrossRefGoogle Scholar
  16. Thornton TM, Pedraza-Alva G, Deng B, Wood CD, Aronshtam A, Clements JL, Sabio G, Davis RJ, Matthews DE, Doble B, Rincon M. Phosphorylation by p38 MAPK as an alternative pathway for GSK3beta inactivation. Science. 2008;320:667–70.PubMedCrossRefPubMedCentralGoogle Scholar
  17. Woodgett JR. Molecular cloning and expression of glycogen synthase kinase-3/Factor A. EMBO J. 1990;9:2431–8.PubMedPubMedCentralGoogle Scholar
  18. Ying QL, Wray J, Nichols J, Batle-Morera L, Doble B, Woodgett J, Cohen P, Smith A. The ground state of embryonic stem cell renewal. Nature. 2008;453:519–23.PubMedCrossRefPubMedCentralGoogle Scholar
  19. Zeng X, Tamai K, Doble B, Li S, Huang H, Habas R, Okamura H, Woodgett J, He X. A dual kinase mechanism for Wnt co-receptor phosphorylation and activation. Nature. 2005;438:873–7.PubMedCrossRefPubMedCentralGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Samuel Lunenfeld Research Institute, Mount Sinai Hospital and Department of Medical BiophysicsUniversity of TorontoTorontoCanada