It has been known for some time that abnormalities of the proteasome system occur in neurons in several neurodegenerative disorders, including Alzheimer’s disease (AD) and Lewy body (LB) diseases. While a number of proteasome abnormalities have been described in these diseases, little has been published about the mechanistic implications of these abnormalities. It has been shown experimentally that proteasome inhibitors can cause apoptosis of postmitotic neurons, but the pathway by which this apoptosis occurs remains incompletely described. Thus, this paper by Rideout and colleagues is a major step forward in understanding how proteasome abnormalities might cause neurodegeneration.

The authors tie together two disparate lines of research on neurodegenerative diseases: the one showing dysfunction of the proteasome in these disorders and the one showing aberrant cell-cycle activation in some of the same disorders, most notably AD. Rideout et al. show convincingly and elegantly that aberrant activation of cyclin-dependent kinases (Cdks) that act at the G1-to-S phase of the cell...  Read more

It has been known for some time that abnormalities of the proteasome system occur in neurons in several neurodegenerative disorders, including Alzheimer’s disease (AD) and Lewy body (LB) diseases. While a number of proteasome abnormalities have been described in these diseases, little has been published about the mechanistic implications of these abnormalities. It has been shown experimentally that proteasome inhibitors can cause apoptosis of postmitotic neurons, but the pathway by which this apoptosis occurs remains incompletely described. Thus, this paper by Rideout and colleagues is a major step forward in understanding how proteasome abnormalities might cause neurodegeneration.

The authors tie together two disparate lines of research on neurodegenerative diseases: the one showing dysfunction of the proteasome in these disorders and the one showing aberrant cell-cycle activation in some of the same disorders, most notably AD. Rideout et al. show convincingly and elegantly that aberrant activation of cyclin-dependent kinases (Cdks) that act at the G1-to-S phase of the cell cycle is required for proteasomal inhibition-induced apoptosis of primary cortical neurons. The work is remarkable in using not only pharmacological inhibitors, but also expression of specific genes that inhibit Cdks and expression of dominant-negative Cdks, to show the participation of Cdk2, Cdk4, and Cdk6 in the apoptosis. The results are also notable in their agreement with two recent papers from Karl Herrup’s group, in which it is shown that cell-cycle aberrations precede neuronal cell death in AD brain (Yang et al., 2003; Yang et al., 2001).

Rideout and colleagues also demonstrate that cyclins D1 and E translocate to the nucleus and accumulate in ubiquitinated inclusions in neurons in which the proteasome is inhibited pharmacologically. In this regard, our laboratory has reported that increases in the levels of the AβPP-interacting protein AβPP-BP1 can also cause neuronal apoptosis (Chen et al., 2000). Interestingly, AβPP-BP1, which is elevated in AD brain (unpublished data from our laboratory), initiates a ubiquitin-like pathway that interfaces with the ubiquitin pathway to cause accumulation of cyclin E. It is exciting to anticipate that the literatures on proteasomal dysfunction, cell-cycle dysfunction, and AβPP dysfunction in neurodegenerative diseases may be beginning to converge.

View all comments by Rachael Neve

This is an excellent study that delineates the precise role of cdks in neuronal death using a combination of pharmacological and molecular tools. It is particularly interesting that both the proteasome and cdks are obligatory mediators of apoptosis, but that cdks are not involved in formation of ubiquitin-protein inclusions. In a sense, these results support previous suggestions that inclusions are a byproduct of degeneration, but are not necessary for neuronal death.

While there is no doubt that neurons die independently of lesion formation in many different diseases, it is hard to imagine that the presence of a cytoplasmic or nuclear inclusion would not disrupt normal cellular processes, and eventually promote death. Nevertheless, studies such as this one, which elucidate the temporal and spatial relationships between various death markers, are crucial in designing appropriate targets for treating neurodegenerative diseases.

View all comments by Inez Vincent

This is a great paper and an important step forward in two regards. Firstly, it makes all of us working on parkin reevaluate an important but unstated assumption, namely, that parkin acts as a single protein enzyme. The paper clearly shows that parkin can act as part of a complex in concert with additional proteins. Although previous results using recombinant parkin protein have suggested that parkin has activity as a single protein in vitro (e.g., Shimura et al., 2001; see also ARF related news story), perhaps its in-vivo activity is more complex, with adaptor proteins controlling activity towards specific proteins. Secondly, this is another example of the protective role that parkin plays in neuronal survival. Imai and colleagues demonstrated that parkin overexpression protects cells against ER stress (Imai et al., 2000) or an unfolded ER protein (  Read more

This is a great paper and an important step forward in two regards. Firstly, it makes all of us working on parkin reevaluate an important but unstated assumption, namely, that parkin acts as a single protein enzyme. The paper clearly shows that parkin can act as part of a complex in concert with additional proteins. Although previous results using recombinant parkin protein have suggested that parkin has activity as a single protein in vitro (e.g., Shimura et al., 2001; see also ARF related news story), perhaps its in-vivo activity is more complex, with adaptor proteins controlling activity towards specific proteins. Secondly, this is another example of the protective role that parkin plays in neuronal survival. Imai and colleagues demonstrated that parkin overexpression protects cells against ER stress (Imai et al., 2000) or an unfolded ER protein (Imai et al., 2001; see also ARF related news story). We have shown that parkin protects against mutant α-synuclein or proteasome inhibition (Petrucelli et al., 2002; see also ARF related news story). Although unable to confirm the protective effect of parkin on ER stress (Darios et al., 2003) showed a cell survival benefit of parkin in response to C2-ceramide, a trigger for mitochondrially mediated apoptosis. The common link between these studies is that the ability of parkin to protect cells is dependent on ubiquitination.

The mechanistic step forward that Staropoli et al. identify is that a specific parkin substrate, cyclin E, seems to be involved. By controlling cyclin E levels in the cell, parkin may prevent specific triggers of cell death, such as abortive reentry into the cell cycle. Our previous results may be related, as α-synuclein mutations impair proteasomal function, and cyclin E is a known proteasome substrate. Therefore, a key experiment now is to see if mutant α-synuclein induces cyclin E accumulation and whether this contributes to cellular toxicity. I suspect that all of these pathways converge on a relatively small number of proteasome substrates, of which cyclin E would be a great candidate. Parkin would, therefore, control cell death by ubiquitinating and/or promoting degradation of the key substrate(s), perhaps even in situations where the proteasome is partially inhibited.

Like all good papers, this study raises a few more questions for the field. Why is it that dopaminergic cells are selectively vulnerable to these processes? Is apoptosis critical here? For example, our results in postmitotic cells indicate a nonapoptotic cell death compared to the apoptosis induced in embryonic cells, which are primed for apoptosis due to their developmental stage. Finally, is cyclin E the critical substrate, or do the other suggested substrates (PaelR1, synphilin, synuclein, and others) also play a role?

View all comments by Mark Cookson

The findings by Staropoli and colleagues (2003) provide strong evidence that an altered cell cycle machinery plays a crucial role in the pathogenesis of Parkinson’s disease (PD), especially the autosomal recessive, early-onset form of PD (ARPD). However, the study lacks direct evidence that accumulated cyclin E contributes to the neuronal cell death evoked by excitotoxicity. A study that demonstrates the inhibitory effect of a cyclin-dependent kinase (CDK) inhibitor on the excitotoxicity-mediated neuronal cell death in the parkin-deficient neurons would provide more solid evidence for the involvement of accumulated cyclin E in neuronal cell death. In addition, given the proposed pivotal role of parkin in the regulation of cyclin E, it is surprising that almost the same level of cyclin E is observed in sporadic PD, where parkin is intact, compared to ARPD. In any event, this paper provides a new aspect that may help us understand the mechanism of accumulation of cell-cycle markers in the vulnerable neurons, not only in PD, but also in other neurodegenerative diseases (reviewed in...  Read more

The findings by Staropoli and colleagues (2003) provide strong evidence that an altered cell cycle machinery plays a crucial role in the pathogenesis of Parkinson’s disease (PD), especially the autosomal recessive, early-onset form of PD (ARPD). However, the study lacks direct evidence that accumulated cyclin E contributes to the neuronal cell death evoked by excitotoxicity. A study that demonstrates the inhibitory effect of a cyclin-dependent kinase (CDK) inhibitor on the excitotoxicity-mediated neuronal cell death in the parkin-deficient neurons would provide more solid evidence for the involvement of accumulated cyclin E in neuronal cell death. In addition, given the proposed pivotal role of parkin in the regulation of cyclin E, it is surprising that almost the same level of cyclin E is observed in sporadic PD, where parkin is intact, compared to ARPD. In any event, this paper provides a new aspect that may help us understand the mechanism of accumulation of cell-cycle markers in the vulnerable neurons, not only in PD, but also in other neurodegenerative diseases (reviewed in Bowser and Smith, 2002).

View all comments by Mark A. Smith

The article by Martinez and colleagues identifies the calcium-binding protein calmodulin as a new binding partner for α-synuclein. This intriguing observation adds to the growing list of proteins that bind α-synuclein. This list includes phospholipase D, 14-3-3, protein kinase C, ERK, GRK, parkin, the DA transporter, tyrosine hydroxylase, and the proteasomal protein S6’ [1-8]. Binding to calmodulin is particularly intriguing because of a growing body of literature suggesting that α-synuclein either binds, or is modulated by, divalent ions. Martinez cites an article by Lansbury’s group showing that calcium does not alter the CD spectrum of α-synuclein, which suggests that calcium does not induce α-helical structure or β-pleated sheet structure in α-synuclein. However, α-synuclein has four tyrosines whose physical behavior can be monitored by measuring the fluorescence emitted by these tyrosines [9]. Both Jensen’s group and my group have shown that calcium alters this fluorescence spectrum, which suggests that calcium directly interacts with α-synuclein [10]. Jensen’s group...  Read more

The article by Martinez and colleagues identifies the calcium-binding protein calmodulin as a new binding partner for α-synuclein. This intriguing observation adds to the growing list of proteins that bind α-synuclein. This list includes phospholipase D, 14-3-3, protein kinase C, ERK, GRK, parkin, the DA transporter, tyrosine hydroxylase, and the proteasomal protein S6’ [1-8]. Binding to calmodulin is particularly intriguing because of a growing body of literature suggesting that α-synuclein either binds, or is modulated by, divalent ions. Martinez cites an article by Lansbury’s group showing that calcium does not alter the CD spectrum of α-synuclein, which suggests that calcium does not induce α-helical structure or β-pleated sheet structure in α-synuclein. However, α-synuclein has four tyrosines whose physical behavior can be monitored by measuring the fluorescence emitted by these tyrosines [9]. Both Jensen’s group and my group have shown that calcium alters this fluorescence spectrum, which suggests that calcium directly interacts with α-synuclein [10]. Jensen’s group confirmed this interaction by examining equilibrium dialysis. Other groups have also observed that α-synuclein regulates calcium-mediated reactions [11, 12]. In addition, we have shown that α-synuclein interacts with a select group of other divalent cations, including zinc, magnesium and iron [9]. Each ion appears to exhibit a different effect on α-synuclein behavior. Martinez examined whether calcium or calmodulin affects α-synuclein fibrillization, but found no effect. While this is true, magnesium does inhibit α-synuclein fibrillization, as do β and γ-synuclein [9, 13, 14]. The current observation that calmodulin binds α-synuclein deepens the link between α-synuclein and divalent cations. Binding of calmodulin to calcium is known to regulate many signal transduction processes. Increasing evidence suggests that α-synuclein also regulates signal proteins, such as PKC and phospholipase D [1, 3]. Binding of calmodulin to α-synuclein suggests that calcium might be an important link regulating the interaction of α-synuclein with its substrates.—Benjamin Wolozin, Loyola University, Maywood, Illinois.

References:
1. Ostrerova N et al. alpha-Synuclein shares physical and functional homology with 14-3-3 proteins. J Neurosci. 1999 Jul 5;19(14):5782-91. Abstract

2. Jenco JM et al. Regulation of phospholipase D2: selective inhibition of mammalian phospholipase D isoenzymes by alpha- and beta-synucleins. Biochemistry. 1998 Apr 7;37(14):4901-9. Abstract

3. Ahn BH et al. alpha-Synuclein interacts with phospholipase D isozymes and inhibits pervanadate-induced phospholipase D activation in human embryonic kidney-293 cells. J Biol Chem. 2002 Apr 5;277(14):12334-42. Abstract

4. Pronin AN et al. Synucleins are a novel class of substrates for G protein-coupled receptor kinases. J Biol Chem. 2000 Aug 25;275(34):26515-22. Abstract

5. Choi P et al. Co-association of parkin and alpha-synuclein. Neuroreport. 2001 Sep 17;12(13):2839-43. Abstract

6. Lee FJ et al. Direct binding and functional coupling of alpha-synuclein to the dopamine transporters accelerate dopamine-induced apoptosis. FASEB J. 2001 Apr;15(6):916-26. Abstract

7. Perez RG et al. A role for alpha-synuclein in the regulation of dopamine biosynthesis. J Neurosci. 2002;22(8):3090-9. (No abstract available)

8. Snyder H et al. Aggregated Synuclein binds to the S6' Proteasomal Protein and Inhibits Proteasomal Function. J. Biol. Chem. 2002. (Submitted)

9. Golts N et al. Magnesium inhibits spontaneous and iron-induced aggregation of alpha-synuclein. J Biol Chem. 2002 May 3;277(18):16116-23. Abstract

10. Nielsen MS et al. Ca2+ binding to alpha-synuclein regulates ligand binding and oligomerization. J Biol Chem. 2001 Jun 22;276(25):22680-4. Abstract

11. Lee D et al. alpha-Synuclein exhibits competitive interaction between calmodulin and synthetic membranes. J Neurochem. 2002 Sep;82(5):1007-17. Abstract

12. Park SM et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood. 2002 Oct 1;100(7):2506-14. Abstract

13. Uversky VN et al. Biophysical properties of the synucleins and their propensities to fibrillate: inhibition of alpha-synuclein assembly by beta- and gamma-synucleins. J Biol Chem. 2002 Apr 5;277(14):11970-8. Abstract

14. Hashimoto M et al. beta-synuclein inhibits alpha-synuclein aggregation: a possible role as an anti-parkinsonian factor. Neuron. 2001 Oct 25;32(2):213-23. Abstract

View all comments by Benjamin Wolozin