The papers by Chen et al., 2003 and Emamian et al., 2003 offer compelling and complementary evidence for the significance of a single phosphorylation event—serine 776 of ataxin-1—on the pathogenesis of the polyglutamine-induced neurodegenerative disease spinocerebellar ataxia type 1 (SCA1). Emamian et al. showed that, while wild-type ataxin-1[82Q] induced profound nuclear inclusions, the A776 mutant failed to form nuclear inclusions in transfected cells. Remarkably, the ataxin-1[82Q]-A776 transgenic mice also exhibited reduced nuclear inclusions in Purkinje cells, and concomitantly displayed very mild, if any, degeneration of these cells, compared to mice expressing wild-type ataxin-1[82Q].

In the same order of ideas, Chen et al. provided a molecular mechanism underlying the difference in the pathogenesis of mutant and wild-type ataxin-1. They found that 14-3-3e and z selectively bound to S776 phosphorylated ataxin-1, but not to A776...  Read more

The papers by Chen et al., 2003 and Emamian et al., 2003 offer compelling and complementary evidence for the significance of a single phosphorylation event—serine 776 of ataxin-1—on the pathogenesis of the polyglutamine-induced neurodegenerative disease spinocerebellar ataxia type 1 (SCA1). Emamian et al. showed that, while wild-type ataxin-1[82Q] induced profound nuclear inclusions, the A776 mutant failed to form nuclear inclusions in transfected cells. Remarkably, the ataxin-1[82Q]-A776 transgenic mice also exhibited reduced nuclear inclusions in Purkinje cells, and concomitantly displayed very mild, if any, degeneration of these cells, compared to mice expressing wild-type ataxin-1[82Q].

In the same order of ideas, Chen et al. provided a molecular mechanism underlying the difference in the pathogenesis of mutant and wild-type ataxin-1. They found that 14-3-3e and z selectively bound to S776 phosphorylated ataxin-1, but not to A776 ataxin-1, which resulted in stabilization of ataxin-1. The length of the polyglutamine tract is likely to enhance the stabilization of 14-3-3 with mutant ataxin-1. Chen et al. then identified Akt as the responsible kinase for this phosphorylation event. As an ultimate experiment, they utilized Drosophila genetics to suggest a synergistic role of the PI3K/Akt pathway and ataxin-1[82Q] in causing neurodegeneration. They propose that 14-3-3 stabilized Akt-phosphorylated mutant ataxin-1, thereby competing with factors mediating its degradation. These studies are exciting, as they provide the first insight for a dual role for Akt and 14-3-3 in the adult CNS and further shed light on the cell death processes in SCA1 pathogenesis.

Several issues remain to be determined in the future.

1. Is serine776 phosphorylation of ataxin-1 upregulated in SCA1 patients?

2. If so, how prevalent is it among the patients? What about the activity of the PI3K and/or Akt in the patients?

3. What is the role of nuclear inclusions in the disease? The A776 mutant fails to induce nuclear inclusions and is impaired in causing neurodegeneration. Thus, there is a perfect parallel between the appearance of inclusions and the pathogenesis of the disease. These findings somehow contradict the previous published results.

View all comments by Li-Huei Tsai

Phosphorylation of proteins is known to be associated with neurodegeneration, but the causal relationship between phosphorylation and neurodegeneration is unclear. Two articles by the laboratories of Harry Orr and Huda Zoghbi in the current Neuron and Cell, respectively, highlight the importance of phosphorylation in neurodegeneration. These labs have investigated the role of phosphorylation in neurodegeneration induced by expanded polyglutamine repeats in ataxin-1. Orr’s group noted that phosphorylation of Serine 776 of ataxin-1 was associated with inclusion formation and neurodegeneration. To test the causative role, they generated a mouse that carried an ataxin-1 gene that lacked Serine 776,

but contained an expanded polyglutamine repeat. The mice had substantially reduced toxicity. In their article, Zoghbi’s group looked at 14-3-3 protein, which binds phosphorylated proteins, and showed that 14-3-3 binds ataxin-1. Binding of 14-3-3 to ataxin-1 appears to slow its degradation. Together, these articles suggest that phosphorylation plays an important role in the accumulation...  Read more

Phosphorylation of proteins is known to be associated with neurodegeneration, but the causal relationship between phosphorylation and neurodegeneration is unclear. Two articles by the laboratories of Harry Orr and Huda Zoghbi in the current Neuron and Cell, respectively, highlight the importance of phosphorylation in neurodegeneration. These labs have investigated the role of phosphorylation in neurodegeneration induced by expanded polyglutamine repeats in ataxin-1. Orr’s group noted that phosphorylation of Serine 776 of ataxin-1 was associated with inclusion formation and neurodegeneration. To test the causative role, they generated a mouse that carried an ataxin-1 gene that lacked Serine 776,

but contained an expanded polyglutamine repeat. The mice had substantially reduced toxicity. In their article, Zoghbi’s group looked at 14-3-3 protein, which binds phosphorylated proteins, and showed that 14-3-3 binds ataxin-1. Binding of 14-3-3 to ataxin-1 appears to slow its degradation. Together, these articles suggest that phosphorylation plays an important role in the accumulation and toxicity of ataxin-1.

The significance of this work extends well beyond work on ataxin-1 because increased phosphorylation is associated with many proteins that accumulate in neurodegenerative diseases. For instance, the tau protein that accumulates to form neurofibrillary tangles shows increased phosphorylation at Ser 396, 404 and 202. Phosphorylation at Ser 129 is associated with accumulation of αa-synuclein in Lewy bodies (Fujiwara et al. 2002). The ability of phosphorylation to regulate the turnover of ataxin-1 suggests the possibility that phosphorylation also regulates the turnover of tau and α-synuclein. Indeed, 14-3-3 has been shown to bind α-synuclein, and 14-3-3 stimulates phosphorylation of tau protein Agarwal-Mawal, 2003; Ostrerova, 1999; Xu et al., 2002). In addition, 14-3-3 is present in both neurofibrillary tangles and Lewy bodies (Ubl et al. 2002; Layfield et al., 1996; Kawamoto et al., 2002). Taken together, these results suggest that protein phosphorylation plays a pivotal role in the pathophysiology of protein aggregation, perhaps regulating degradation of aggregation-prone proteins, and that 14-3-3 is a key regulator of this process.

References:
Fujiwara H, Hasegawa M, Dohmae N, Kawashima A, Masliah E, Goldberg MS, Shen J, Takio K, Iwatsubo T. alpha-Synuclein is phosphorylated in synucleinopathy lesions. Nat Cell Biol. 2002 Feb;4(2):160-4. Abstract

Agarwal-Mawal A, Qureshi HY, Cafferty PW, Yuan Z, Han D, Lin R, Paudel HK. 14-3-3 Connects Glycogen Synthase Kinase-3beta to Tau within a Brain Microtubule-associated Tau Phosphorylation Complex. J Biol Chem. 2003 Apr 11;278(15):12722-8. Abstract

Ostrerova N, Petrucelli L, Farrer M, Mehta N, Choi P, Hardy J, Wolozin B. alpha-Synuclein shares physical and functional homology with 14-3-3 proteins. J Neurosci. 1999 Jul 15;19(14):5782-91. Abstract

Xu J, Kao SY, Lee FJ, Song W, Jin LW, Yankner BA. Dopamine-dependent neurotoxicity of alpha-synuclein: a mechanism for selective neurodegeneration in Parkinson disease. Nat Med. 2002 Jun;8(6):600-6. Abstract

Ubl A, Berg D, Holzmann C, Kruger R, Berger K, Arzberger T, Bornemann A, Riess O. 14-3-3 protein is a component of Lewy bodies in Parkinson's disease-Mutation analysis and association studies of 14-3-3 eta. Brain Res Mol Brain Res. 2002 Dec 16;108(1-2):33-9. Abstract

Layfield R, Fergusson J, Aitken A, Lowe J, Landon M, Mayer RJ. Neurofibrillary tangles of Alzheimer's disease brains contain 14-3-3 proteins. Neurosci Lett. 1996 May 3;209(1):57-60. Abstract

Kawamoto Y, Akiguchi I, Nakamura S, Honjyo Y, Shibasaki H, Budka H. 14-3-3 proteins in Lewy bodies in Parkinson disease and diffuse Lewy body disease brains. J Neuropathol Exp Neurol. 2002 Mar;61(3):245-53. Abstract

View all comments by Benjamin Wolozin

Akt-1: The Good Guy Takes a Knock but Stays the Course

Opening scene: Akt protects its king (i.e., neuron) from dark forces. Protein kinase B (PKB or Akt) is a family of serine-threonine kinases with three isoforms. Following activation by either of the numerous receptor tyrosine kinases, Akt phosphorylates substrates bearing the R-x-R-x-x-S/T-F/L consensus motif. The first step in Akt’s activation is a conformational change upon binding of its Pleckstrin homology domain to the PI3K product, membrane phospholipid phosphatidylinositol 3,4,5-P3). Recruitment of Akt to the membrane is then a signal for the sequential phosposrylations of threonine 308 and serine 473 by the phosphoinositide-dependent protein kinases PDK1 and 2, respectively. Phosphorylation of both sites is required for Akt to become fully active.

Akt is a multifunctional gatekeeper molecule to many signaling events downstream of growth factor stimulation (Datta et al., 1999;   Read more

Akt-1: The Good Guy Takes a Knock but Stays the Course

Opening scene: Akt protects its king (i.e., neuron) from dark forces. Protein kinase B (PKB or Akt) is a family of serine-threonine kinases with three isoforms. Following activation by either of the numerous receptor tyrosine kinases, Akt phosphorylates substrates bearing the R-x-R-x-x-S/T-F/L consensus motif. The first step in Akt’s activation is a conformational change upon binding of its Pleckstrin homology domain to the PI3K product, membrane phospholipid phosphatidylinositol 3,4,5-P3). Recruitment of Akt to the membrane is then a signal for the sequential phosposrylations of threonine 308 and serine 473 by the phosphoinositide-dependent protein kinases PDK1 and 2, respectively. Phosphorylation of both sites is required for Akt to become fully active.

Akt is a multifunctional gatekeeper molecule to many signaling events downstream of growth factor stimulation (Datta et al., 1999; Lawlor et al., 2001). Cell proliferation, survival, and apoptosis are intimately regulated by Akt activity. Other cell processes under Akt control are glucose metabolism (e.g., through insulin signaling and PKBβ action) Cho et al., 2001), and vascular homeostasis (e.g., through integrin signaling and eNOS production) Shiojima et al., 2002). Downstream targets of Akt-orchestrated, inactivating phosphorylations are the pro-apoptosis molecules BAD, Forkhead, IKKα, GSK3-β and caspase 9, which account for the "good guy" role of Akt in cell survival. Frequently, Akt’s partner in sequestering these molecules is the chaperone protein 14-3-3.

In neurons (for example, those stressed by trophic factor withdrawal or excitotoxic injury), Akt’s position as a survival mediator is firmly established by these mechanisms, as well as another, novel one that involves hindrance of JNK activation by AKT (Dudek et al., 1997; Kim et al., 2002 ). It is relevant to note that mammalian cells have three Akt genes encoding corresponding isoforms, whereas Drosophila has but one. Thus, in fruit flies, mutation of Akt is lethal, whereas in mice, Akt-1 disruption causes growth retardation and organ-specific apoptosis, or a partial lethal phenotype. A large number of studies demonstrate, in various cell types, that constitutively active (CA)-Akt is sufficient to block cell death, and that a dominant-negative construct inhibits growth factor-induced survival.

Scene Two/New Villains Appear. Akt has just been cast as master regulator of antiapoptotic defenses, but the neurodegenerative disease plot thickens as intracellular and presumably misfolded proteins enter the stage as toxic "gain-of-function" attackers of cell survival. How will our hero fare?

In elegant back-to-back papers from the laboratories of Huda Zoghbi (Chen et al., 2003) and Harry Orr (Emamian et al., 2003), we learn that the PI3K/Akt signaling pathway enhances, not suppresses, the neurotoxicity of ataxin-1. This surprising role reversal pertains to 14-3-3 function, as well. In SCA-1, an inherited polyglutamine tract expansion in ataxin-1, leads to autosomal-dominant degeneration of cerebellar Purkinje cells. As in Huntington’s disease, accumulations of ataxin-1 are found as insoluble intranuclear inclusions. The nuclear inclusions per se are not required for disease progression (Cummings et al., 1999). In the Emamian paper, transfected CHO cells were used to isolate ataxin-30Q with a single phosphoserine residue (S776). Mutation of this site (A776) eliminated nuclear inclusions in cultured cells and transgenic mice, promoted ataxin partitioning into the soluble fraction, and resulted in disease-free mice, despite the persistent presence in the nucleus of ataxin-polyQ. The story here is complicated, but phosphorylation of serine appears to be unequivocally required for disease expression. Although P-S776 is linked to nuclear transport of ataxin-1, nuclear localization is not itself causative. Similarly, the polyQ tract is necessary but not sufficient for pathogenesis. The authors hypothesize that another protein, interacting specifically with P-S776, drives disease progression.

The Chen et al. paper is very convincing. Immunoprecipitation and protein purification techniques were applied in COS1 cells transfected with ataxin-1-82Q-S776 to identify 14-3-3 as its binding partner. Ataxin-1 has a 14-3-3 binding motif in which the critical S776 residue is embedded, and mutant of polyQ-ataxin that can’t be phosphorylated also was defective in 14-3-3 binding. In the presence of ataxin-S776-polyQ, 14-3-3 was recruited to yet larger nuclear inclusions. Increased steady-state levels of ataxin accompanied the aggravation of inclusions by 14-3-3. The authors demonstrated the effects of this interaction in vivo by showing that double-transgenic flies (14-3-3/ataxin polyQ) developed more severe retinal degeneration than did either transgene alone. Toward identifying the kinase responsible for ataxin-1 phosphorylation, the authors noted that this motif corresponded to a putative Akt consensus motif. In an in-vitro assay, purified Akt kinase phosphorylated a GST-ataxin-30Q fusion protein. The reaction was confirmed in cotransfected HeLa cells with constitutively active Akt. When GST-ataxin-30Q is incubated with immobilized 14-3-3, they associate only when ataxin is phosphorylated by Akt-1. Finally, co-IP of 14-3-3 was increased from transfections containing ataxin-1 and constitutively active Akt. Crossings of the SCA-1/ataxin-transgenic flies with Akt and PI3K markedly exacerbated the retinal phenotype. Interestingly and surprisingly, double-transgenic flies harboring either dPDK1 or dGSK3β genes showed no added effect. The authors conclude that 14-3-3 stabilizes Akt-phosphorylated, mutant ataxin levels, apparently sequestering it from protein degradation and thus accelerating disease. Since Akt accelerates SCA-1-type neurodegeneration, suppressing it may hold therapeutic value in this and possibly some pathogenic protein disorders, but could be deleterious in others.

To identify the kinase responsible for ataxin-1 phosphorylation, it was noted that the 14-3-3 motif corresponded to a putative Akt consensus motif. Using an in-vitro phosphorylation assay, a GST-ataxin-30Q fusion protein was phosphorylated by purified Akt kinase. The reaction was confirmed in HeLa cells transfected with CA-Akt. When GST-ataxin-30Q was incubated with immobilized 14-3-3, the two associated only when ataxin was phosphorylated by Akt-1. Finally, coimmunoprecipitation of 14-3-3 was increased in transfections containing ataxin-1 and CA-Akt.

These papers are important and thoroughly researched. They raise several interesting questions. If PI3K produces the same exacerbation of SCA-1 phenotype as Akt, why did the PKB-kinase, PDK1, have no effect? Since the PDK-1-catalyzed phosphorylation of threonine 308 is essential to Akt activation, and PDK-1 activity (with a putative PDK-2) is also required for the serine 473 step, one would expect it to have some effect in this system. The use of phospho-specific antibodies to these Akt sites in future work could show that it is the S473 phosphorylation that is pivotal.

It also remains to be worked out in which cellular compartments the soluble 14-3-3/ataxin complexes are toxic, since only a small fraction of 14-3-3 resides in the nuclear inclusions. The Chen et al. results will have even greater impact if reproduced in the transgenic SCA-1 mouse model of cerebellar involvement. For instance, crossing the SCA-1 with an Akt-1 knockout strain should lead to a straightforward answer.

Postlude or postmortem? In spite of these important studies, more articles have since continued to affirm the protective effect of Akt in other neurodegenerative models where a toxic protein is expressed. For instance, Fred Gage’s and Jeff Rothstein’s laboratory (see ARF related news story) delayed the progression of motor neuron disease (ALS) in the SOD1 mutant G93A mouse by delivering the neurotrophic factor IGF to the CNS. Intramuscular injection, followed by nervous uptake and retrograde transport of adeno-associated.virus (AAV)-IGF was associated with decreased apoptosis of spinal motor neurons and greater Akt activity.

Our own work in endothelial cells (Suhara et al., 2003) shows that viral-mediated expression of β-amyloid42 results in apoptotic death, and that this apoptosis is rescued by constitutively active Akt and aggravated by dominant-negative Akt. We chose endothelial cells because the Akt pathway is well-characterized in them, but have obtained the same results in other cell types. The mechanism of intracellular Aβ-induced death was in part through inhibition of Akt activation on both serine and threonine residues. The downstream effector of Aβ-induced death was activated GSK-3β, an Akt substrate and confirmed pro-apoptotic actor in a number of neurodegenerative conditions (Lucas et al., 2001; Jackson et al., 2002; Hetman et al., 2000; Takashima et al., 1998). Conversely, in the Chen work, GSKβ3 overexpression had no effect.

The work by Chen et al. does not exclude a normal neuroprotective role for Akt in the background; however, when a specific protein pathogen is involved (e.g., ataxin-1 polyQ), the detrimental action of Akt to stabilize the toxic molecule overwhelms its gentler side. The matter may indeed by simple: If the toxic protein has an Akt phosphorylation consensus site, the kinase will be pro-apoptotic by directly affecting the protein’s degradation. If, as suspected of most neurotoxic proteins (e.g., β-APP, Aβ, α-synuclein), there is no such site, Akt will promote survival. This hypothesis could be put to the test. For instance, the protein tau has a putative Akt site (Ksiezak-Reding et al., 2000). Actually, several regulatory molecules toggle between survival and degeneration depending on whether the cell is proliferating, quiescent, or differentiated. Likewise, proteasome inhibition may be pro- or antiapoptotic depending on what stage of the cell cycle the neuron is in. Even 14-3-3, which classically neutralizes pro-apoptotic proteins, may fall into the "ataxin camp" by abetting dopamine-mediated α-synuclein toxicity (Xu et al., 2002).

A major challenge for cell biologists working in neurodegeneration is to understand these signaling relationships as they change with specific cell subtype and in the context of misfolded proteins. Detailed road maps are needed to base therapy on this strategy so as not to inadvertently target bystander cell types for uncontrolled proliferation, developmental derangement, or death.

View all comments by Henry Querfurth