 |
Early Pattern Formation in the Vertebrate Embryo
Summary: Peter Klein's laboratory studies cell signaling in vertebrate embryogenesis and the mechanisms of lithium action in bipolar disorder, hematopoiesis, and embryonic development.
Embryogenesis in vertebrates involves the transformation of a symmetrical egg into multiple cell types organized into a body plan with dorsal-ventral, anterior-posterior, and left-right axes. Our laboratory studies the mechanisms acting shortly after fertilization that establish the body plan of the frog
Xenopus laevis, which offers unique advantages for studying this early period in vertebrate development. We also study the mechanism of lithium action in multiple settings, including its effects on developing embryos and on the behavior of adult mice.
Lithium Action and Dorsal Axis Formation
Lithium has pronounced effects on the development of diverse organisms, from cellular slime molds to mammals, in addition to its well-known use in the therapy of bipolar disorder, its insulin-like activities, and its ability to stimulate white blood cell production in the bone marrow. In frog and fish embryos, lithium causes expansion of dorsal and anterior structures, often yielding embryos with two heads and a bifurcated spinal cord.
Many hypotheses have been advanced to explain the mechanism of lithium action. One of the most widely accepted models is the inositol-depletion hypothesis, which holds that lithium acts through inhibition of the enzyme inositol monophosphatase (IMPase), depleting cells of inositol. Our laboratory has shown, however, that alternative inhibitors of IMPase have no effect on embryonic development, suggesting that the effects of lithium on developing embryos are mediated through a different target. While other enzymes related to IMPase are also inhibited by lithium, our work has focused on the protein kinase glycogen synthase kinase-3 (GSK-3), a novel direct target of lithium action.
Activation of the Wnt Pathway by Lithium
The Wnt pathway is a highly conserved signaling mechanism found in vertebrates and invertebrates. In vertebrates, Wnt genes have been implicated in dorsal axis formation as well as patterning of the nervous system, kidney development, limb patterning, and other developmental processes. When Wnt mRNAs are artificially introduced into early frog embryos, they cause dorsalization similar to the effects of lithium. Since the Wnt pathway functions by inhibiting GSK-3, an antagonist of the pathway, we predicted that lithium might also activate this pathway through inhibition of GSK-3. Indeed, we showed that lithium inhibits GSK-3 (but not multiple other protein kinases) in vitro and in vivo. Furthermore, inhibition of GSK-3 by lithium mimics Wnt signaling, leading to stabilization of β-catenin protein and activation of Wnt-dependent gene expression in Xenopus embryos and a variety of mammalian cell types. Recent work from several laboratories, including ours, has shown that lithium also inhibits GSK-3 in vivo in the mammalian brain.
Thus, lithium has been shown to activate a signaling pathway that plays a central role in embryonic development. The possibility that this also explains the therapeutic action of lithium in bipolar disorder is being investigated. Ongoing research in transgenic mouse models is focused on the role of GSK-3 in mediating the behavioral effects of lithium. Understanding how lithium works in the treatment of bipolar disorder will allow the development of new, safer drugs for treating this common psychiatric illness and may help to elucidate the molecular pathogenesis of mania. (This work has been supported in part by grants from the National Institutes of Health and the EJLB Foundation.)
New Regulators of GSK-3 and Wnt Signaling
Novel genes that regulate GSK-3 activity directly have been identified in a yeast two-hybrid screen using GSK-3β as bait. One of these genes is the Xenopus homolog of axin (also reported by several other laboratories), which has been shown to be an antagonist of Wnt signaling. We have defined a 23–amino acid domain within axin responsible for interaction with GSK-3. Surprisingly, expression of the GSK-3 interaction domain (GID) alone completely inhibits GSK-3 activity in vivo and in vitro. Thus, the GID provides a novel and specific alternative inhibitor of GSK-3 that should mimic the physiological effects of lithium.
Consistent with this hypothesis, the GID peptide inhibits tau protein phosphorylation in neuronal cells and causes stabilization and accumulation of β-catenin protein, axis duplication in Xenopus embryos, and activation of LEF/Tcf-dependent transcription in multiple cell types, all effects similar to lithium. Thus, the GID peptide mimics lithium action biochemically and phenotypically. This peptide is now being used to examine the effects of inhibition of GSK-3 in neuronal cells of adult, transgenic mice to ask whether the GID peptide mimics the effects of lithium on mouse behavior.
Valproic acid (VPA) is an antiepileptic therapy that, like lithium, is widely used to treat bipolar disorder. Its mechanism of action is also unknown. We have found that VPA can mimic lithium in selected assays, including activation of Wnt-dependent gene expression, but that VPA acts through a distinct target and is a more general activator of transcription than lithium. Our data show that VPA, at therapeutic concentrations, potently inhibits histone deacetylases (HDACs), which generally repress transcription. As a consequence, VPA causes hyperacetylation of endogenous targets of HDACs, such as histones and p53, and leads to activation of transcription of a large number of genes. VPA is also a potent teratogen in humans and other vertebrates, and we find that developmental defects caused by VPA are remarkably similar to those caused by trichostatin A, a well-known HDAC inhibitor. Our data support the hypothesis that VPA-induced birth defects arise through inhibition of HDACs.
Wnt Receptors (Frizzleds)
The frizzled genes, first identified for their roles in tissue polarity in Drosophila, encode transmembrane receptors for Wnt/wingless proteins. A large family of frizzled genes has now been identified, but the role of frizzleds in vertebrate species remains unclear. We have therefore identified a family of Xenopus frizzled homologs (Xfzs) expressed in early embryos and are characterizing their functions in early vertebrate development.
Using a functional assay for synergy between Wnts and frizzleds, we have identified a strong and specific interaction between Wnt-1 and Xfz3. Both are highly expressed in dorsal neural tissues that give rise to the neural crest, a population of cells that contribute to a large variety of tissues, including the sensory ganglia, the sympathetic nervous system, the adrenal medulla, craniofacial structures, and melanocytes. We find that Xfz3 is required for Wnt-1 activity in neural crest induction assays. Furthermore, depletion of Xfz3 blocks endogenous neural crest formation, while overexpression of Xfz3 leads to ectopic formation of the neural crest. Since Xfz3 is also expressed in dorsal neural tissues in mouse embryos, it may play a similar role in the development of the neural crest in mammals.
We have also identified a novel protein named kermit that binds to the carboxyl terminus of Xfz3. The kermit gene is expressed in a pattern that overlaps Xfz3 expression. Kermit is recruited to cell membranes when Xfz3 is overexpressed, and does not bind to Xfz8, indicating a level of selectivity in its binding activity. Depletion of kermit blocks the activity of Xfz3, as well as Wnt-1, in neural crest induction assays, indicating that kermit is required for Xfz3 signaling.
Xfz8 is highly expressed in the Spemann organizer at the beginning of gastrulation, suggesting a role for Xfz8 in patterning the embryo during gastrulation. We tested this hypothesis by expressing an inhibitory form of Xfz8 (Nxfz8) encoding only the ligand-binding domain. Expression of Nxfz8 in cells that normally express Xfz8 causes a marked reduction in the length of the anterior-posterior axis and broadening of the mediolateral axis, an effect that appears to be due to inhibition of convergent extension, the coordinated cell movements that drive morphogenesis during gastrulation and neurulation.
Last updated December 01, 2003
|
|
|
|
