|
|
|
|
|
|
|
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
First published online 20 March 2007
doi: 10.1242/jcs.006924
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Short Report |
Department of Molecular Cell Biology, Max-Planck-Institute of Biochemistry, Am Klopferspitz 18, 82152 Martinsried, Germany
* Author for correspondence (e-mail: stemmann{at}biochem.mpg.de)
Accepted 14 February 2007
Summary
The highly abundant AAA-ATPase p97 is required for diverse cellular processes, of which ER-associated protein degradation (ERAD) is understood best. Previously, a new role of p97 in spindle disassembly at the end of mitosis has been reported. However, we show that neither addition of dominant-negative p97 mutants nor depletion of crucial p97 adaptors impairs transition of meiotic spindles into interphase arrays of microtubules. The dominant-negative approach is validated by inhibition of ERAD, which we reconstitute for the first time in the powerful biochemical system of Xenopus egg extracts. The role of p97 in spindle disassembly during meiotic exit should therefore be reconsidered.
Key words: Microtubules, Mitotic exit, p97, Spindle disassembly, Xenopus egg extracts
Introduction
The transformation of an interphase microtubule (MT) array into a mitotic spindle involves dramatic changes in dynamics and movements of MTs. Continuous growth and shrinkage lie within the nature of MTs themselves but are also influenced by MT-stabilizing and -destabilizing factors. Additionally, motor proteins contribute to self-organization of the mitotic spindle (Heald et al., 1996
). These regulators of MT-behaviour are themselves regulated by cyclin-dependent kinase 1 (Cdk1)-dependent phosphorylation (Verde et al., 1992). The transition of mitotic spindles into interphase MT arrays during mitotic exit has hardly been studied so far. Although Cdk1 inactivation is clearly essential for this process (Murray et al., 1989), Zheng and co-workers have recently reported that, surprisingly, it is not sufficient (Cao et al., 2003). Spindle disassembly additionally required the activity of an AAA-ATPase, called p97 in vertebrates (the unfavorable name VCP, for valosin-containing protein, was given after an artefact) and Cdc48 in yeast.
p97 is an abundant, conserved protein involved in a variety of cellular functions, such as membrane fusion or ER-associated protein degradation (ERAD) (Ye, 2006). The unifying scheme in these processes is that p97 extracts ubiquitylated target proteins from their environment, such as protein complexes or membrane compartments. One p97 monomer is composed of an N-domain responsible for substrate binding and two AAA-domains (called D1 and D2). In its native conformation, p97 exists as a hexamer that appears like two rings on top of each other (DeLaBarre and Brunger, 2003), with one ring built out of the N and D1 domains, the other of the D2 domains. Adaptor proteins, such as p47 and the heterodimeric Ufd1-Npl4, play crucial roles in recruiting specific substrates to p97.
The majority of the established roles of p97 lie outside mitosis. However, p97 was first discovered in S. cerevisiae as a cell-division-cycle mutant cdc48-1, which causes arrest in mitosis with undivided nuclei and medium-sized spindles (Frohlich et al., 1991). Likewise, knockdown of p97 in HeLa cells leads to transient accumulation in a prometaphase-like state (Wojcik et al., 2004). These observations prompted us to develop tools for the investigation of vertebrate p97 in cell-free extracts from Xenopus eggs. Studying p97 in the same model system, Cao et al. reported an essential role of p97 for spindle disassembly during exit from meiosis II, which closely resembles mitosis (Cao et al., 2003). Inactivating p97 by addition of a dominant-negative mutant or immunodepletion of Ufd1, these researchers observed persistence of MT-asters despite normal inactivation of Cdk1. Trying to establish an assay for p97 activity we repeated the experiments of Cao and colleagues. Surprisingly, we find that spindle disassembly in Xenopus egg extracts takes place normally in the absence of functional p97. We support our results by two independent approaches to abrogate p97 activity and by establishing ERAD as a positive control.
Results and Discussion
Depletion of p97 adaptors has no effect on spindle disassembly
Given the high concentration of p97 in Xenopus egg extract (0.25 mg/ml or 2.8 µM, data not shown), we interfered with p97 function by depleting the less abundant adaptor proteins Ufd1, Npl4 or p47. Using purified polyclonal antibodies against corresponding full-length Xenopus proteins for two rounds of depletions, we were able to reduce the amount of Ufd1 by 
90% in cytostatic factor-arrested extract (CSF-extract) that represents a metaphase-like state (Fig. 1A, left). At the same time >95% of the Ufd1-binding Npl4 were co-depleted. With anti-Npl4 antibody, 99% of Npl4 and 90% of Ufd1 were removed from the extract (Fig. 1A, middle). However, the most complete depletion (>95% and >99% for Ufd1 and Npl4, respectively) was achieved with a combination of both antibodies (Fig. 1A, right). Similarly, reduction of p47 was 99% complete (Fig. 1B). p97 co-precipitated with Ufd1, Npl4 and p47 (data not shown). However, because of its vast excess overall p97 levels remained unchanged.
|
Depleted CSF-extracts were combined with sperm nuclei and Rhodamine-labeled tubulin, and incubated to allow for assembly of MT-asters – which occasionally fuse into bipolar, spindle-like structures. Meiotic exit was then triggered by Ca2+ and the state of the MT-cytoskeleton was controlled every 10 minutes by fluorescence microscopy over a total period of 60 minutes. Using the same assay, Cao et al. reported that spindle disassembly was completely blocked in Ufd1-depleted extract for up to 80 minutes (Cao et al., 2003). But when we repeated this experiment, spindle disassembly occurred normally within 40 minutes after Ca2+ addition, despite efficient removal of Ufd1-Npl4 (Fig. 1C). To clarify this issue, we additionally assayed spindle disassembly in CSF-extracts depleted with anti-Npl4 or a combination of anti-Ufd1 and anti-Npl4 antibodies. We even supplemented extracts, which had been depleted of >95% and >99% of Ufd1 and Npl4, respectively, with soluble anti-Ufd1 or anti-Npl4 antibodies to 0.25 mg/ml before triggering meiotic exit. Finally, we also tested the effect of depleting p47, the adaptor known to be involved in p97-mediated membrane fusions. In no case did we observe impairment of spindle disassembly (Fig. 1C), despite the fact that our depletions – at least for Npl4 – were more efficient than those of Cao et al. (Cao et al., 2003) and that we supplemented depleted extract with additional, soluble antibody instead of simply adding antibody to untreated extract as did Cao and co-workers. However, MT-asters persisted when high Cdk1 activity was maintained by addition of non-degradable cyclin B1
90, showing that depolymerization of MT-asters depends on meiotic exit and is not simply due to prolonged incubation (Fig. 1C). Thus, neither p97-Ufd1-Npl4 nor p97-p47 are necessary to bring about the transition of spindle-like structures into interphase MT-arrays in Xenopus egg extracts.
|
). This dominant-negative mutant was reported to block spindle disassembly in Xenopus egg extract at 0.5-0.6 mg/ml (Cao et al., 2003). To test this prediction, we generated the analogous dominant-negative version of Xenopus p97 by changing the conserved glutamate residues in D1 and D2 at positions 305 and 578, respectively, to glutamines (resulting in the mutant protein p97-QQ). Inactivation of the AAA-domains was confirmed by ATPase assay in comparison with wild-type p97 (p97-wt) (Fig. 1G). CSF-extract was pre-incubated with p97-wt, p97-QQ (both at 0.5 mg/ml) or reference buffer and spindle disassembly was assayed as before. Remarkably, it was again normal in all samples, i.e. astral structures disappeared within 40 minutes (Fig. 1C). Spindles disassembled on schedule even when the final concentration of p97-QQ was raised to 2 mg/ml or when pre-incubation with the dominant-negative mutant was extended from 30 to 60 minutes (Fig. 1C and data not shown).
To test the quality of our recombinant p97-QQ, we established ERAD as an assay for p97 activity in Xenopus egg extracts. To this end, we monitored steady-state levels of CFTRF508, an unstable mutant (F508) of the human cystic fibrosis transmembrane conductance regulator (CFTR), which is degraded via ERAD in mammalian cell culture (Varga et al., 2004; Weihl et al., 2006). In contrast to standard rabbit reticulocyte lysate, the low-speed Xenopus extracts used throughout this study give rise to anti-CFTR-reactive proteins (with the apparent molecular mass of 160 and 140 kDa) upon addition of CFTRF508 mRNA (Fig. 1D). These translational products probably represent fully and partially glycosylated CFTRF508 properly inserted into ER membranes. At concentrations as low as 0.25 mg/ml the QQ-mutant impaired ERAD, resulting in a 2.7-fold elevated steady-state level of CFTRF508 (Fig. 1E). We also followed the disappearance of CFTRF508 in a cycloheximide shut-off. In control extract, 90% of the ERAD substrate was degraded with a half-life of 45 minutes (Fig. 1F). Degradation was, however, greatly impaired in presence of p97-QQ. Inhibition of ERAD shows that our dominant-negative mutant efficiently interferes with p97 activity in Xenopus egg extracts, validating our earlier observation that it has no effect on spindle disassembly. Whereas ERAD has been analyzed in reticulocyte lysate before (Xiong et al., 1999), this is the first report of its reconstitution using just CFTRF508-encoding mRNA and standard Xenopus egg extracts. This simple system should prove valuable for future investigations of this important degradation pathway.
Of the two AAA domains, D2 is responsible for catalytic activity of p97, whereas D1 stabilizes the quaternary structure (DeLaBarre and Brunger, 2003). Thus, the E305Q mutation might impair hexamer formation and weaken the dominant-negative effect although dissociation-reassembly assays argue otherwise (Wang et al., 2003). We generated versions of Xenopus p97 with an intact D1 but inactivated D2 domain by introducing an E578Q point mutation and a double K524A-E578Q mutation (Kitami et al., 2006), resulting in the p97-Q and p97-AQ mutants, respectively. Despite reduction or absence of ATPase activity (Fig. 1G), these p97-Q and p97-AQ mutants (like the triple mutant p97-QAQ) again had no influence on conversion of asters into interphase MT-arrays at the tested concentrations of 0.5-2 mg/ml (Fig. 1C). Finally, we tested p97 mutants from other species to exclude the possibility that deviations from the sequence of Xenopus p97 enhance their dominant-negative character. We examined the murine version of p97-QQ (Ye et al., 2001; Cao et al., 2003) as well as the human version of p97-QQ at concentrations of 0.25-5 µg/µl. All these mutants lacked ATPase activity but, importantly, none affected spindle disassembly relative to corresponding p97-wt or buffer controls (Fig. 1C,G). Thus, experiments using a multitude of dominant-negative mutants, at concentrations up to ten times higher than those used by Cao et al. (Cao et al., 2003) and with the highest concentration exceeding endogenous p97 levels 20-fold, lead us to conclude that p97 is dispensable for spindle disassembly in Xenopus egg extracts.
|
To exclude any effect of sperm density on the timing of microtubule depolymerization, we tested various concentrations ranging from 200-2000 nuclei/µl, including the concentration of 500 nuclei/µl used by Zheng and co-workers. In no case did we observe any impact on spindle disassembly relative to the corresponding control of the same sperm density. Thus, interference with p97 function neither blocks nor delays disassembly of sperm-induced spindles in Xenopus egg extracts that exit meiosis.
Spindle disassembly in the absence of functional p97 occurs independently of sperm chromatin
Isolated sperm nuclei contain p97 and its adaptors in low concentrations (data not shown). The amounts added back with sperm nuclei to immunodepleted extracts are negligible. For Npl4, which – relative to extract – exhibits the highest concentration, 0.036% is replenished by addition of 2000 nuclei/µl of depleted extract (data not shown). Nevertheless, to fully exclude any influence of added chromatin, we triggered aster formation by adding purified centrosomes instead. We then compared disappearance of MT-asters in extracts supplemented with Xenopus p97-QQ, p97-wt (both at 1 mg/ml) or reference buffer by analyzing aliquots every 10 minutes after release. As before, all samples were indistinguishable in the way spindle-like structures were replaced by interphase MT-arrays (Fig. 3A). Similarly, disappearance of centrosome-dependent asters was not delayed but, rather, slightly accelerated in Ufd1-Npl4-depleted extract relative to a mock-depleted control, consistent with our earlier experiments (Fig. 3B,C; data not shown). Thus, irrespective of whether their formation has been triggered by chromatin or centrosomes, MT-asters disassemble normally when Xenopus egg extracts leave meiosis in the absence of functional p97.
In this study we have carefully reinvestigated in Xenopus egg extracts the involvement of the AAA-ATPase p97 in spindle disassembly. In contrast to a previous study performed using a similar approach (Cao et al., 2003), we find that p97 and its adaptors proteins Ufd1 and Npl4 are not required for spindle disassembly during exit from meiosis, but are crucial for the ERAD pathway. Given that we have taken great care to repeat the Xenopus experiments of Cao and colleagues under very similar conditions, it is difficult to reconcile their data with ours. Antibody-mediated protein depletion can be prone to errors if the antibody crossreacts with other proteins as, perhaps in this case, a factor required for spindle disassembly. However, Cao et al. reported that the observed defect in the Ufd1-depleted extract could be rescued by bacterially expressed Ufd1-Npl4, suggesting antibody crossreactivity was not a problem in their experiments. Therefore, an alternative explanation for our differing results is that, under their conditions, p97 depletion causes a secondary or indirect effect on spindle stability due to perturbations in other p97-dependent processes such as ERAD. Our findings, although highlighting the importance of p97 in ERAD, demonstrate that the role of p97 and its adaptor proteins in spindle disassembly should be reconsidered.
Materials and Methods
Recombinant proteins and antibodies
His-tagged, bacterially expressed p97 proteins were purified over Ni2+-NTA-agarose (Qiagen) and dialyzed against sperm dilution buffer (SDB; 5 mM K-HEPES pH 7.7, 1 mM MgCl2, 100 mM KCl, 150 mM sucrose). ATPase activity of p97 proteins was assayed as follows. First, 2 µg of p97 in 40 µl SDB plus 0.4 mM ATP were incubated at 37°C. After 0, 15 and 30 minutes, reactions in 10-µl aliquots were stopped with 40 µl perchloric acid (1 M). After adding 40 µl of Malachite Green solution (4.2 g ammonium molybdate in 100 ml 4 M HCl mixed with 0.135 g Malachite Green in 300 ml H2O) and incubation for 20 minutes at room temperature, extinctions at 615 nm were determined. The Xenopus cDNAs of Ufd1, Npl4, p47 and UT6 (a p97-binding fragment of Ufd1) were cloned into modified pGEX (GE Healthcare) or pQE80 (Qiagen) vectors. Bacterially expressed GST-tagged or His-tagged proteins were purified over glutathione sepharose (GE Healthcare) or Ni2+-NTA-agarose columns, respectively, and dialyzed against SDB or 0.2 M NaHCO3 pH 8.3, 0.5 M NaCl. Antibodies against Xenopus Ufd1, Npl4 and p47 were raised in rabbits by injection of His-tagged proteins. For affinity purification, corresponding GST-tagged proteins were immobilized to NHS-columns (GE Healthcare). Only in case of anti-Ufd1, a mixture of His-tagged Ufd1 (in complex with GST-Npl4) and GST-UT6 was used for preparation of the affinity column. Immunopurified antibodies were dialyzed against SDB. Monoclonal antibodies were from PROGEN Biotechnik (Xenopus p97), Upstate (CFTR, M3A7) and the Developmental Studies Hybridoma Bank (
-tubulin).
Analysis of spindle disassembly in Xenopus egg extracts
CSF-arrested Xenopus egg extracts and sperm nuclei were prepared as described (Philpott et al., 1991; Murray, 1991). Recombinant p97 or human cyclin B190 were added to extracts at 0.25-5 mg/ml or 0.03 mg/ml final concentration, respectively, and incubated for 30-60 minutes at 4°C. For depletions, antibodies were coupled to magnetic protein-A beads (DYNAL, 0.25 mg/ml) in suspension before performing two rounds of 30-minute depletions at 12°C with 15 µg of each antibody per 50 µl of CSF-extract. In experiments shown in Fig. 1C and Fig. 2B, depleted extract contained additional soluble anti-Npl4 or anti-Ufd1 antibody (0.25 mg/ml each). CSF-extracts were combined with Rhodamine-labelled tubulin (Cytoskeleton) and 200-2000 sperm nuclei per µl (500 nuclei for experiments shown in Fig. 2). After 30 minutes at 20°C, extracts were released by adding Ca2+ (0.6 mM) and analyzed every 5-10 minutes to assess chromatin condensation and associated asters for 100-150 sperm nuclei.
ERAD and shut-off experiments
The CFTRF508 mutant of human CFTR was amplified from a plasmid provided by P. J. Thomas (The University of Texas Southwestern Medical Center, Dallas, TX) using the oligonucleotids 5'-TAATACGACTCACTATAGGGACTACCATGCAGAGGTCGCCTCT-3' and 5'-CTAAAGCCTTGTATCTTGCACCTCTTC-3'. mMESSAGEmMachineTM T7Ultra kit and MEGAclear colums (Ambion) were used to generate capped, poly-adenylated CFTRF508 mRNA.
For shut-off experiments CSF-extracts were released at 20°C by 0.6 mM CaCl2 for 10 minutes, supplemented with CFTRF508 mRNA to 0.05 mg/ml and incubated for 90 minutes. After adding Xenopus p97-QQ (0.25-1 mg/ml) or SDB, extracts were incubated 15 minutes before translation was stopped by 0.2 mg/ml cycloheximide. Aliquots were taken 0, 30, 45, 60, 75 and 90 minutes thereafter and subjected to immunoblotting.
Acknowledgments
We thank T. U. Mayer, J.-M. Peters, H. Richly, P. J. Thomas, and Y. Ye for reagents and A. Strasser and S. Ebener for technical assistence. We are grateful to members of the Stemmann lab, A. Buchberger, C. Schuberth and S. Jentsch for critical reading of the manuscript and continuous support. This work was supported by a fellowship from Boehringer Ingelheim Fonds to S.H. and the Emmy Noether Program of the Deutsche Forschungsgemeinschaft.
References
Cao, K., Nakajima, R., Meyer, H. H. and Zheng, Y. (2003). The AAA-ATPase Cdc48/p97 regulates spindle disassembly at the end of mitosis. Cell 115, 355-367.[CrossRef][Medline]
DeLaBarre, B. and Brunger, A. T. (2003). Complete structure of p97/valosin-containing protein reveals communication between nucleotide domains. Nat. Struct. Biol. 10, 856-863.[CrossRef][Medline]
Frohlich, K. U., Fries, H. W., Rudiger, M., Erdmann, R., Botstein, D. and Mecke, D. (1991). Yeast cell cycle protein CDC48p shows full-length homology to the mammalian protein VCP and is a member of a protein family involved in secretion, peroxisome formation, and gene expression. J. Cell Biol. 114, 443-453.
Heald, R., Tournebize, R., Blank, T., Sandaltzopoulos, R., Becker, P., Hyman, A. and Karsenti, E. (1996). Self-organization of microtubules into bipolar spindles around artificial chromosomes in Xenopus egg extracts. Nature 382, 420-425.[CrossRef][Medline]
Kitami, M. I., Kitami, T., Nagahama, M., Tagaya, M., Hori, S., Kakizuka, A., Mizuno, Y. and Hattori, N. (2006). Dominant-negative effect of mutant valosin-containing protein in aggresome formation. FEBS Lett. 580, 474-478.[CrossRef][Medline]
Murray, A. W. (1991). Cell cycle extracts. Methods Cell Biol. 36, 581-605.[Medline]
Murray, A. W., Solomon, M. J. and Kirschner, M. W. (1989). The role of cyclin synthesis and degradation in the control of maturation promoting factor activity. Nature 339, 280-286.[CrossRef][Medline]
Philpott, A., Leno, G. H. and Laskey, R. A. (1991). Sperm decondensation in Xenopus egg cytoplasm is mediated by nucleoplasmin. Cell 65, 569-578.[CrossRef][Medline]
Varga, K., Jurkuvenaite, A., Wakefield, J., Hong, J. S., Guimbellot, J. S., Venglarik, C. J., Niraj, A., Mazur, M., Sorscher, E. J., Collawn, J. F. et al. (2004). Efficient intracellular processing of the endogenous cystic fibrosis transmembrane conductance regulator in epithelial cell lines. J. Biol. Chem. 279, 22578-22584.
Verde, F., Dogterom, M., Stelzer, E., Karsenti, E. and Leibler, S. (1992). Control of microtubule dynamics and length by cyclin A- and cyclin B-dependent kinases in Xenopus egg extracts. J. Cell Biol. 118, 1097-1108.
Wang, Q., Song, C. and Li, C. C. (2003). Hexamerization of p97-VCP is promoted by ATP binding to the D1 domain and required for ATPase and biological activities. Biochem. Biophys. Res. Commun. 300, 253-260.[CrossRef][Medline]
Weihl, C. C., Dalal, S., Pestronk, A. and Hanson, P. I. (2006). Inclusion body myopathy-associated mutations in p97/VCP impair endoplasmic reticulum-associated degradation. Hum. Mol. Genet. 15, 189-199.
Wojcik, C., Yano, M. and DeMartino, G. N. (2004). RNA interference of valosin-containing protein (VCP/p97) reveals multiple cellular roles linked to ubiquitin/proteasome-dependent proteolysis. J. Cell Sci. 117, 281-292.
Xiong, X., Chong, E. and Skach, W. R. (1999). Evidence that endoplasmic reticulum (ER)-associated degradation of cystic fibrosis transmembrane conductance regulator is linked to retrograde translocation from the ER membrane. J. Biol. Chem. 274, 2616-2624.
Ye, Y. (2006). Diverse functions with a common regulator: Ubiquitin takes command of an AAA ATPase. J. Struct. Biol. 156, 29-40.[Medline]
Ye, Y., Meyer, H. H. and Rapoport, T. A. (2001). The AAA ATPase Cdc48/p97 and its partners transport proteins from the ER into the cytosol. Nature 414, 652-656.[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
