GSK3β—Not Just for Tauists Anymore
Article appears by special arrangement with Alzheimer Research Forum. See original article with additional links/commentary.
30 January 2004. Alzheimer's researchers sneak regular peeks over the shoulders of colleagues working on Parkinson's and other neurodegenerative diseases. Why not do the same with schizophrenia researchers, who are working in a disease that features hallucinations—an occasional feature of AD—and obvious deficits in working memory? Here, then, is such a peek: A paper in Nature Genetics, wherein researchers implicate impaired signaling in the AKT1-GSK3β pathway in schizophrenia.
The tauophiles in the Alzheimer's research community have a longstanding interest in glycogen synthase kinase (GSK)3β because it is known to phosphorylate tau (see Alzheimer Research Forum related news story), and those more drawn to the amyloid camp have also begun to show interest in the GSK3α variant, which could be involved in the production of Aβ (see Alzheimer Research Forum related news story). GSK3β is well-studied due to its role in regulating insulin and β-catenin in the Wnt pathway.
The major factor controlling GSK3β activation is the protein-serine/threonine kinase AKT1, which has already come to interest AD researchers for its role in inhibiting apoptosis. And the researchers have been quick to note that AKT1 mediates phoshatidylinositol 3 (PI3) kinase signaling, which in turn is susceptible to Aβ control. Psychiatric disease researchers have turned their attention to AKT1, which appears to be a principal target of lithium and other drugs used in bipolar disorder. There is some evidence that the AKT gene increases risk for bipolar disorder. Given the evidence for overlapping genetic and biochemical abnormalities in bipolar disorder and schizophrenia, AKT1 is also of great interest in schizophrenia research.
In the current study, a multi-institutional team led by Joseph Gogos of Columbia University and Maria Karayiorgou of the Rockefeller University, both in New York City, went on an interesting "fishing expedition." They write, "We speculated that alterations in brain levels or activity of protein kinases and phosphatases may contribute to schizophrenia susceptibility in humans and that this might be observed in the peripheral tissues of individuals with schizophrenia." Assessing protein levels in peripheral blood lymphocytes, they focused on kinases implicated in synaptic plasticity, and were rewarded with the finding that levels of AKT1 protein were 68 percent lower in subjects with schizophrenia (P = 0.014). This went hand-in-hand with significant decreases in phosphorylation of the AKT1 target GSK3β both in peripheral lymphocytes and in postmortem tissue from frontal cortex of schizophrenia patients. In addition, AKT1 levels were reduced in the cortex of patients.
The site of the AKT1 gene, in the cytogenetic band 14q32, has never been fingered as a susceptibility locus in linkage scans of kindreds with schizophrenia. However, the authors found a significant association between schizophrenia and a haplotype of the AKT1 gene associated with lower protein levels. In another experiment, the authors turned to AKT1 -/- mice to investigate whether these show any abnormalities that might be relevant to schizophrenia. They indeed found that this knockout strain produced deficits in a measure of sensorimotor gating, a defect characteristic of schizophrenia. In addition, the authors showed in-vivo evidence that haloperidol, much like lithium, boosts phosphorylation of AKT1, evidence that the AKT1-GSK3β pathway could be a prime site of action for the antipsychotic medication.
As catalogued by the authors, AKT1 turns up as a player in many cellular processes, with links to important molecules (e.g., GABRA, BDNF, NRATC, calcineurin, neuregulin), and thus is poised to have wide influence. "This influence may be additive, perhaps due to small impairment of several processes, or may be restricted to a small number of key processes that are particularly AKT1 dosage-dependent. In any case, our data is consistent with a model in which impairment of AKT1-GSK3β signaling increases the liability of these neuronal circuits to additional genetic or environmental insults that ultimately lead to the disease," the authors conclude.—Hakon Heimer (Alzheimer Research Forum).
Emamian ES, Hall D, Birnbaum MJ, Karayiorgou M, Gogos JA. Convergent evidence for impaired AKT1-GSK3beta signaling in schizophrenia. Nat Genet. 2004 Jan 25 [Epub ahead of print]. Abstract
Comments on Related News
Related News: Playing on Without AKT1: Subtle Cortical Deficits Suggest VulnerabilitiesComment by: Takeo Yoshikawa
, Akihiko Takashima
Submitted 30 November 2006
Posted 30 November 2006
I recommend the Primary Papers
In this study, Karayiorgou and Gogos’s group have conducted a meticulous anatomical analysis of pyramidal cell dendritic structures in the prefrontal layer V cortex, as well as genome-wide expression and pharmaco-behavioral analyses, focusing on prefrontal functions in Akt1-deficient mice. The study examines the reduced (or altered) AKT1-GSK3β signalling theory of schizophrenia, proposed by this (Emamian et al., 2004) and other groups.
AKT1 as a genetic susceptibility gene for schizophrenia shows promise in the Caucasian population but this is not reflected in Asian populations as evidenced by our results (Ide et al., 2006). In addition, even in Caucasians, true causal variants have not been identified. Because of this, schizophrenia researchers are interested in observing disease-relevant phenotypes in Akt1-deficient mice. In this study, they have detected morphological and functional alterations of frontal cortex-related traits in mutant mice using state-of-the-art techniques.
To further strengthen AKT1 as a candidate disease gene in schizophrenia, several issues need to be addressed in the near future. For instance, if a reduction of AKT1 signalling occurs in the brain, tau should be hyper-phosphorylated by activated GSK3β, which in turn will lead to the formation of neurofibrillary tangles (NFTs) as seen in Alzheimer’s disease. Therefore, it would be interesting to determine whether Akt1-deficient mice show a similar pattern of tau phosphorylation. Accumulating evidence suggests that hyper-phosphorylated tau may affect a variety of neuronal functions. Our recent biochemical analyses failed to reveal any significant reduction of AKT-mediated signalling in the prefrontal cortex of schizophrenic brains or the expected inverse correlation between phosphorylation levels of AKT and tau (Ide et al., 2006). This highlights the difficulty of examining protein phosphorylation status using postmortem brains, where results are often confounded by multiple, uncontrollable factors.
Another important but poorly understood point is the functional relationship among subspecies of the AKT family (at least AKT1, AKT2 and AKT3) and GSK3 (GSK3α and GSK3β) (for example see Sale et al., 2005). We look forward to continuing multidisciplinary studies aimed at unravelling the role of the AKT cascade, including the clarification of downstream pathways (Datta et al., 1999; (O’Mahony et al., 2006) in schizophrenia pathology.
View all comments by Takeo Yoshikawa
View all comments by Akihiko Takashima
Related News: DISC1: A Matter of Life or Death for Neural Progenitors
Comment by: Khaled Rahman
Submitted 26 March 2009
Posted 26 March 2009
Mao and colleagues present an impressive body of work implicating GSK3β/β-catenin signaling in the function of Disc1. However, several key experimental controls are missing that detract from the impact of their study, and it is unclear whether this function of Disc1 among its many others is the critical link between the t(1;11) translocation and psychopathology in the Scottish family.
The results of Mao et al. suggest that acute knockdown of Disc1 in embryonic brain causes premature exit from the proliferative cell cycle and premature differentiation into neurons. In fact, they observe fewer GFP+ cells in the VZ/SVZ and greater GFP+ cells within the cortical plate. This is in contrast to the study by Kamiya et al. (2005), in which they find that knocking down Disc1 caused greater retention of cells in the VZ/SVZ and fewer in the cortical plate, suggesting retarded migration. Although the timing of electroporation (E13 vs. E14.5) and examination (E15 vs. P2) differed between the two studies, these results are not easily reconciled.
The authors also suggest that they can rescue the deficits in proliferation by overexpressing human wild-type DISC1, stabilizing β-catenin expression, or inhibiting GSK3β activity, and thus conclude that Disc1 is acting through this pathway. This conclusion, however, rests on an error in logic. If increasing X causes an increase in Y, and decreasing Z causes a decrease in Y, this does not mean that X and Z are operating via the same mechanism. In fact, overexpressing WT-DISC1, stabilizing β-catenin, or inhibiting GSK3β activity all increase proliferation in control cells. Thus, the fact that these manipulations also work in progenitors with Disc1 silenced only tells us that these effects are independent or downstream of Disc1. What are needed are studies that show a differential sensitivity of Disc1-silenced cells to manipulations of β-catenin or GSK3β. In other words, is there a shift in the dose response curves? This is what is to be expected given that Mao et al. show changes in β-catenin levels and changes in the phosphorylation of GSK3β substrates in Disc1 silenced cells.
Furthermore, it is surprising that a restricted silencing of Disc1 in the adult dentate gyrus produces changes in affective behaviors, when total ablation of dentate neurogenesis in the adult produces little effects on depression-related behaviors (Santarelli et al., 2003; Airen et al., 2007). The fact that inhibiting GSK3β increases proliferation in both control and Disc1 knockdown animals to a similar degree suggests that the “rescue” of any behavioral deficits is independent of the drug’s effects on proliferation. Correlating measures of proliferation with behavioral performance would help address this issue.
How this study will lead to new or improved therapeutic interventions is also an open question. Lithium is well known for its mood-stabilizing properties, and this study may point to better, more efficient ways to address these symptoms. However, it is also known that lithium does little for, if not worsens, cognitive symptoms in patients (Pachet and Wisniewski, 2003), and it is this symptom domain that is in dire need of drug development.
It is also important to keep in mind that acute silencing of Disc1 in a restricted set of cells will not necessarily recapitulate the pathogenetic process of a disease-associated mutation. It remains to be seen if similar results are obtained in animal models of the Disc1 mutation (Clapcote et al., 2007; Hikida et al., 2007; Li et al., 2007).
Kamiya A, Kubo K, Tomoda T, Takaki M, Youn R, Ozeki Y, Sawamura N, Park U, Kudo C, Okawa M, Ross CA, Hatten ME, Nakajima K, Sawa A. A schizophrenia-associated mutation of DISC1 perturbs cerebral cortex development. Nat Cell Biol. 2005 Dec 1;7(12):1167-78. Abstract
Santarelli, L. et al. Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 301, 805–809 (2003). Abstract
Airan, R.D. et al. High-speed imaging reveals neurophysiological links to behavior in an animal model of depression. Science 317, 819-23 (2007). Abstract
Pachet AK, Wisniewski AM. The effects of lithium on cognition: an updated review. Psychopharmacology (Berl). 2003 Nov;170(3):225-34. Review. Abstract
Clapcote SJ, Lipina TV, Millar JK, Mackie S, Christie S, et al. (2007) Behavioral phenotypes of Disc1 missense mutations in mice. Neuron 54: 387–402. Abstract
Hikida T, Jaaro-Peled H, Seshadri S, Oishi K, Hookway C, et al. (2007) Dominant-negative DISC1 transgenic mice display schizophrenia-associated phenotypes detected by measures translatable to humans. Proc Natl Acad Sci U S A 104: 14501–14506. Abstract
Li W, Zhou Y, Jentsch JD, Brown RA, Tian X, et al. (2007) Specific developmental disruption of disrupted-in-schizophrenia-1 function results in schizophrenia-related phenotypes in mice. Proc Natl Acad Sci U S A 104: 18280–18285. Abstract
View all comments by Khaled Rahman
Related News: DISC1: A Matter of Life or Death for Neural Progenitors
Comment by: Simon Lovestone
Submitted 27 March 2009
Posted 27 March 2009
This is an intriguing paper that builds on a growing body of evidence implicating wnt regulation of GSK3 signaling in psychotic illness (Lovestone et al., 2007).
It is interesting that the authors report that binding of DISC1 to GSK3 results in no change in the inhibitory Ser9 phosphorylation site of GSK3 but a change in Y216 activation site and that this resulted in effects on some but not all GSK3 substrates. This poses a challenge both in terms of understanding the role of GSK3 signaling in schizophrenia and other psychotic disorders and in drug discovery.
The authors cite some of the other evidence for regulation of GSK3 signaling in psychosis, including, for example, the evidence for a role of AKT signaling alteration in schizophrenia and lithium, an inhibitor of GSK3, as a treatment for bipolar disorder. But in both cases, AKT (Cross et al., 1995) and lithium (Jope, 2003), the effect on GSK3 is predominantly via Ser9 phosphorylation and not via Y216. The unstated implication is at least two, possibly three, mechanisms for regulation of GSK3 are all involved in psychotic illness—the auto-phosphorylation at Y216, the exogenous signal transduction regulated Ser9 site inhibition and, if the association of schizophrenia with the wnt inhibitor DKK4 we reported is true (Proitsi et al., 2008), also via the wnt signaling effects on disruption of the macromolecular complex that brings GSK3 together with β-catenin. On the one hand, this might be taken as positive evidence of a role for GSK3 in psychosis—all of its regulatory mechanisms have been implicated; therefore, the case is stronger. On the other hand, GSK3 lies at the intersection point of very many signaling pathways and so is likely to be implicated in many disorders (as it is), and the fact that in cellular and animal models related to psychosis there is no consistent effect on the enzyme is troublesome.
From a drug discovery perspective, those with GSK3 inhibitors in the pipeline will be watching this space carefully. However, it is worth noting that Mao et al. find very selective effects of DISC1 on GSK3 substrates. Despite convincing evidence of an increase in Y216 phosphorylation, which one would expect to increase activity of GSK3 against all substrates, the authors find no evidence of effects on phosphorylation of the GSK3 substrates Ngn2 or C/EBPα. This is somewhat puzzling and merits further attention, especially as in vitro direct binding of a DISC1 fragment to GSK3 inhibited the action of GSK3 on a range of substrates. Might there be more to the direct interaction of DISC1 with GSK3 than a regulation of Y216 autophosphorylation and activation? If, however, GSK3 regulation turns out to be part of the mechanism of schizophrenia or bipolar disorder, then identifying which of the substrates and which of the many activities of GSK3, including on plasticity and hence cognition (Peineau et al., 2007; Hooper et al., 2007), are important in disease will become the critical task.
Lovestone S, Killick R, Di Forti M, Murray R. Schizophrenia as a GSK-3 dysregulation disorder. Trends Neurosci. 2007 Apr 1 ; 30(4):142-9. Abstract
Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature . 1995 Dec 21-28 ; 378(6559):785-9. Abstract
Jope RS. Lithium and GSK-3: one inhibitor, two inhibitory actions, multiple outcomes. Trends Pharmacol Sci . 2003 Sep 1 ; 24(9):441-3. Abstract
Proitsi P, Li T, Hamilton G, Di Forti M, Collier D, Killick R, Chen R, Sham P, Murray R, Powell J, Lovestone S. Positional pathway screen of wnt signaling genes in schizophrenia: association with DKK4. Biol Psychiatry . 2008 Jan 1 ; 63(1):13-6. Abstract
Peineau S, Taghibiglou C, Bradley C, Wong TP, Liu L, Lu J, Lo E, Wu D, Saule E, Bouschet T, Matthews P, Isaac JT, Bortolotto ZA, Wang YT, Collingridge GL. LTP inhibits LTD in the hippocampus via regulation of GSK3beta. Neuron . 2007 Mar 1 ; 53(5):703-17. Abstract
Hooper C, Markevich V, Plattner F, Killick R, Schofield E, Engel T, Hernandez F, Anderton B, Rosenblum K, Bliss T, Cooke SF, Avila J, Lucas JJ, Giese KP, Stephenson J, Lovestone S. Glycogen synthase kinase-3 inhibition is integral to long-term potentiation. Eur J Neurosci . 2007 Jan 1 ; 25(1):81-6. Abstract
View all comments by Simon Lovestone
Related News: DISC1: A Matter of Life or Death for Neural Progenitors
Comment by: Nick Brandon (Disclosure)
Submitted 27 March 2009
Posted 30 March 2009
I recommend the Primary Papers
Li-huei Tsai and colleagues have identified another pathway in which the candidate gene DISC1 looks to have a critical regulatory role, namely the wnt signaling pathway, in progenitor cell proliferation. In recent years we have seen that DISC1 has a vital role at the centrosome (Kamiya et al., 2005), in cAMP signaling (Millar et al., 2005), and in multiple steps of adult hippocampal neurogenesis (Duan et al., 2007). They have shown a pivotal role for DISC1 in neural progenitor cell proliferation through regulation of GSK3 signaling using a spectacular combination of cellular and in utero manipulations with shRNAs and GSK3 inhibitor compounds. These findings clearly implicate DISC1 in another “druggable” pathway but at this stage do not really identify new approach/targets, except perhaps to confirm that manipulating adult neurogenesis and the wnt pathway holds much potential hope for therapeutics. Perhaps understanding the mechanism of inhibition of GSK3 by DISC1 in more detail might reveal more novel approaches or encourage more innovative work around this pathway. In addition, I have read the other comment (by Rahman), and though I agree that this work still leaves many questions to be answered, the paper is much more significant and likely reconcilable with previous papers than appreciated. The commentary from Lovestone was very insightful and brings up additional gaps and issues with the present work. Additional experimentation I am sure will tease out more key facets of the DISC1-wnt interaction in the near future.
There are many avenues now to proceed with this work. In particular, from the DISC1-centric view, a GSK3 binding site on DISC1 overlaps with one of the critical core PDE4 binding site. Mao et al. show that residues 211 to 225 are a core part of a GSK3 binding site. Previously, Miles Houslay had shown very elegantly that residues 191-230 form a common binding site (known as common site 1) for both PDE4B and 4D families (Murdoch et al., 2007). It will be important to understand the relationship between GSK3 and PDE4 related signaling in reference to the activity of DISC1 starting at whether a trimolecular complex among DISC1-PDE4-GSK3 can form. Then it will be critical to understand the regulatory interplay among these molecules. For example, it is known that PKA can regulate GSK3 activity (Torii et al., 2008) and the interaction between DISC1 and PDE4, while both GSK3 and PKA can phosphorylate β-catenin (Taurin et al., 2006). The output of these relationships on progenitor proliferation will further deepen insights into the role of DISC1 complexes in neuronal processes. This type of situation is not really surprising for a molecule (DISC1) which has been shown to interact with >100 proteins (Camargo et al., 2007). The context of these interactions in both normal development and disease is likely to be critical to allow understanding of its complete functional repertoire.
Another area where these new findings need to be exploited is in the study of additional animal models. Though the two behavioral endpoint models used in the paper (amphetamine hyperactivity and forced swim test) provide a tantalizing glimpse of the behavioral importance of the complex, it would be critical to look in additional models relevant for schizophrenia and mood disorders. Furthermore, it will be very interesting to look at the effects of GSK3β inhibitors in some of the DISC1 animal models already available and to see if they can reverse all or a subset of reported behaviors. In reviewing a summary of the phenotypes available to date (Shen et al., 2008) there is clearly a number of lines which share the properties with mice injected with DISC1 shRNA into the dentate gyrus and would be of value to look at.
A very exciting paper which I am sure will drive additional research into understanding the role of DISC1 in psychiatry and hopefully encourage drug discovery efforts around this molecular pathway (Wang et al., 2008).
1. Kamiya A, Kubo K, Tomoda T, Takaki M, Youn R, Ozeki Y, Sawamura N, Park U, Kudo C, Okawa M, Ross CA, Hatten ME, Nakajima K, Sawa A. A schizophrenia-associated mutation of DISC1 perturbs cerebral cortex development. Nat Cell Biol . 2005 Dec 1 ; 7(12):1167-78. Abstract
2. Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR, Malloy MP, Chubb JE, Huston E, Baillie GS, Thomson PA, Hill EV, Brandon NJ, Rain JC, Camargo LM, Whiting PJ, Houslay MD, Blackwood DH, Muir WJ, Porteous DJ. DISC1 and PDE4B are interacting genetic factors in schizophrenia that regulate cAMP signaling. Science . 2005 Nov 18 ; 310(5751):1187-91. Abstract
3. Duan X, Chang JH, Ge S, Faulkner RL, Kim JY, Kitabatake Y, Liu XB, Yang CH, Jordan JD, Ma DK, Liu CY, Ganesan S, Cheng HJ, Ming GL, Lu B, Song H. Disrupted-In-Schizophrenia 1 regulates integration of newly generated neurons in the adult brain. Cell . 2007 Sep 21 ; 130(6):1146-58. Abstract
4. Murdoch H, Mackie S, Collins DM, Hill EV, Bolger GB, Klussmann E, Porteous DJ, Millar JK, Houslay MD. Isoform-selective susceptibility of DISC1/phosphodiesterase-4 complexes to dissociation by elevated intracellular cAMP levels. J Neurosci . 2007 Aug 29 ; 27(35):9513-24. Abstract
5. Torii K, Nishizawa K, Kawasaki A, Yamashita Y, Katada M, Ito M, Nishimoto I, Terashita K, Aiso S, Matsuoka M. Anti-apoptotic action of Wnt5a in dermal fibroblasts is mediated by the PKA signaling pathways. Cell Signal . 2008 Jul 1 ; 20(7):1256-66. Abstract
6. Taurin S, Sandbo N, Qin Y, Browning D, Dulin NO. Phosphorylation of beta-catenin by cyclic AMP-dependent protein kinase. J Biol Chem . 2006 Apr 14 ; 281(15):9971-6. Abstract
7. Camargo LM, Collura V, Rain JC, Mizuguchi K, Hermjakob H, Kerrien S, Bonnert TP, Whiting PJ, Brandon NJ. Disrupted in Schizophrenia 1 Interactome: evidence for the close connectivity of risk genes and a potential synaptic basis for schizophrenia. Mol Psychiatry . 2007 Jan 1 ; 12(1):74-86. Abstract
8. Shen S, Lang B, Nakamoto C, Zhang F, Pu J, Kuan SL, Chatzi C, He S, Mackie I, Brandon NJ, Marquis KL, Day M, Hurko O, McCaig CD, Riedel G, St Clair D. Schizophrenia-related neural and behavioral phenotypes in transgenic mice expressing truncated Disc1. J Neurosci . 2008 Oct 22 ; 28(43):10893-904. Abstract
9. Wang Q, Jaaro-Peled H, Sawa A, Brandon NJ. How has DISC1 enabled drug discovery? Mol Cell Neurosci . 2008 Feb 1 ; 37(2):187-95. Abstract
View all comments by Nick Brandon
Related News: DISC1: A Matter of Life or Death for Neural Progenitors
Comment by: Akira Sawa, SRF Advisor
Submitted 8 April 2009
Posted 8 April 2009
Mao and colleagues’ present outstanding work sheds light on a novel function of DISC1. Because DISC1 is a multifunctional protein, the addition of new functions is not surprising. Thus, for the past several years, the field has focused on how DISC1 can have distinct functions in different cell contexts (for example, progenitor cells vs. postmitotic neurons, or developing cortex vs. adult dentate gyrus). In addition to Mao and colleagues, I understand that several groups, including ours, have obtained preliminary, unpublished evidence that DISC1 regulates progenitor cell proliferation, at least in part via GSK3β. Thus, I am very supportive of this new observation.
If there might be a missing point in this paper, it is unclear whether suppression of GSK3β occurs in several different biological contexts in brain in vivo. In other words, it is uncertain whether DISC1’s actions on GSK3β are constitutive or context-dependent. How can we reconcile differential roles for DISC1 in progenitor cells in contrast to postmitotic neurons? We have already obtained a preliminary promising answer to this question, which is currently being validated very intensively. These two phenotypes (progenitor cell control and postmitotic migration) may compensate for each other in cortical development; thus, overall cortical pathology looks milder in adults, at least in our preliminary unpublished data using DISC1 knockout mice. We are not sure how this novel function of DISC1 may account for the pathology of Scottish cases. Although I have great respect for the Scottish pioneers of DISC1 study, such as St. Clair, Blackwood, and Muir (I believe that the St. Clair et al., 1990 Lancet paper is one of the best publications in psychiatry), now is the time to pay more and more attention to the question of the molecular pathway(s) involving DISC1 in general schizophrenia (see 2009 SRF roundtable discussion). Unlike the role of APP in Alzheimer’s disease, DISC1 is not a key biological target in general schizophrenia, instead being an entry point to explore much more important targets for schizophrenia. There may be no more need to stick to DISC1 itself in the unique Scottish cases in schizophrenia research. In sum, although there may still be key missing points in this study, I wish to congratulate the authors on their outstanding work.
St Clair D, Blackwood D, Muir W, Carothers A, Walker M, Spowart G, Gosden C, Evans HJ. Association within a family of a balanced autosomal translocation with major mental illness. Lancet . 1990 Jul 7 ; 336(8706):13-6. Abstract
View all comments by Akira Sawa