- Open Access
Anaphase B spindle dynamics in Drosophila S2 cells: Comparison with embryo spindles
© de Lartigue et al; licensee BioMed Central Ltd. 2011
Received: 13 February 2011
Accepted: 8 April 2011
Published: 8 April 2011
In the Drosophila melanogaster syncytial blastoderm stage embryo anaphase B is initiated by a cell cycle switch in which the suppression of microtubule minus end depolymerization and spatial reorganization of the plus ends of outwardly sliding interpolar microtubules triggers spindle elongation. RNA interference in Drosophila cultured S2 cells may present a useful tool for identifying novel components of this switch, but given the diversity of spindle design, it is important to first determine the extent of conservation of the mechanism of anaphase B in the two systems.
The basic mechanism, involving an inverse correlation between poleward flux and spindle elongation is qualitatively similar in these systems, but quantitative differences exist. In S2 cells, poleward flux is only partially suppressed and the rate of anaphase B spindle elongation increases with the extent of suppression. Also, EB1-labelled microtubule plus ends redistribute away from the poles and towards the interpolar microtubule overlap zone, but this is less pronounced in S2 cells than in embryos. Finally, as in embryos, tubulin FRAP experiments revealed a reduction in the percentage recovery after photobleaching at regions proximal to the pole.
The basic features of the anaphase B switch, involving the suppression of poleward flux and reorganization of growing microtubule plus ends, is conserved in these systems. Thus S2 cells may be useful for rapidly identifying novel components of this switch. The quantitative differences likely reflect the adaptation of embryonic spindles for rapid, streamlined mitoses.
Mitosis is mediated by the mitotic spindle, a cellular machine composed of microtubules (MTs) and mitotic motors [1–5]. The critical function of mitosis is revealed at its climax, during anaphase, at which point the spindle coordinates separation of the sister chromatids to opposite spindle poles (anaphase A) and spindle elongation (anaphase B) in preparation for cytokinesis [6–8]. In the Drosophila melanogaster syncytial blastoderm stage embryo, highly dynamic MTs drive remarkably rapid movements of the chromosomes and spindle poles. Anaphase B spindle elongation is proposed to depend on an interpolar (ip) MT sliding filament mechanism generated by homotetrameric kinesin-5 motors and an "on-off" switch orchestrated by the suppression of poleward MT flux [9–13]. The current model of the mechanism underlying this anaphase B "switch" postulates that the pre-anaphase B spindle is maintained at a steady state length by the balance between ipMT sliding and ipMT depolymerization at the poles via kinesin-13-dependent depolymerization, generating poleward flux. In response to cyclin B degradation (i) a MT catastrophe gradient causes ipMT plus ends to invade the overlap zone where outward ipMT sliding occurs; and (ii) kinesin-13 (KLP10A)-dependent depolymerization is switched off, tipping the balance of forces to allow outward ipMT sliding to push apart the spindle poles .
It is now recognized that the mechanisms of mitosis can vary significantly in different cell types, even within the same organism . For example, in the Drosophila melanogaster syncytial blastoderm stage embryo multiple spindles progress rapidly and synchronously through mitosis. Cultured S2 cells, in contrast, do not contain hundreds of spindles progressing synchronously through mitosis, making them less amenable to the quantitation of spindle dynamics. However, these cells have emerged as a very useful model system for studying mitosis using RNA interference (RNAi) techniques to probe the function of candidate proteins in mitosis (and other subcellular processes) [12, 15–19]. S2 cell spindles differ from those of the embryo in a number of different ways. Firstly, mitosis is much slower, about 40-50 mins from nuclear envelope breakdown (NEB) through to cytokinesis. S2 cell spindles are also less centrosome-dependent and can be formed by centrosome-independent mechanisms . Despite these differences, it is possible that some of the underlying molecular characteristics of mitosis may be conserved between embryos and S2 cells, including aspects of the anaphase B switch. For example, Matos et al. observed a suppression of MT poleward flux at anaphase B onset . In order to determine the extent of conservation of the anaphase B switch between embryos and S2 cells, and to evaluate the suitability of S2 cells for identifying novel components of the anaphase B switch, we used various S2 cell lines to examine MT dynamics in S2 cell mitotic spindles and compared the results with those obtained using embryo spindles.
Anaphase A and B are synchronized in Drosophila S2 cells
Summary of comparative data
Drosophila syncytial embryo (cycle 12) 10,11
Drosophila S2 cells
Pre-anaphase B steady
Pre-anaphase B spindle
Anaphase B spindle length
MT flux rates:
0.053 ± 0.013 μm/s
0.05 ± 0.006 μm/s
0.008 ± 0.019 μm/s
0.03 ± 0.007 μm/s
Anaphase B rate
0.08 ± 0.015 μm/s
0.11 ± 0.019 μm/s
0.0098 ± 0.0047 μm/s
MT plus end distribution
at anaphase B
Redistribution to the equator
Partial redistribution to the equator
MT turnover at pole:
81.0 ± 14.3%
113.1 ± 12.3%
45.8 ± 11.2%
71.8 ± 21.2%
In the embryo anaphase A chromosome segregation proceeds at an average rate of 0.1 μm/s during cycle 12 (Table 1) and comes almost to completion prior to the initiation of anaphase B spindle elongation . In other organisms anaphase A and B appear to be much more synchronized and previous reports have indicated that this may also be the case in Drosophila S2 cells [18, 22]. In order to investigate the timing and rate of anaphase A and B in S2 cells, we used a cell line in which the chromosomes were labelled by the stable expression of GFP-tagged histone 2B, in addition to the stable expression of RFP-α-tubulin to label the spindle. Using time-lapse microscopy we imaged concanavalin A-flattened S2 cells from NEB through to cytokinesis (Figure 1A). In the majority of cases (60%) we found that chromosome to pole motility occurred at the same time as anaphase B spindle elongation in these cells. In the remaining cells chromosome to pole motility began prior to spindle elongation in the same manner as the syncytial embryo. Chromosome to pole motility occurred at an average rate of 0.0098 μm/s ± 0.0047 μm/s (Figure 1B; n = 10; table 1), approximately 10 times more slowly than in the embryo.
The suppression of MT poleward flux at anaphase B onset is linked to spindle elongation during anaphase B
Partial suppression of microtubule poleward flux at anaphase B onset
Average MT flux rate
(± standard deviation)
0.05 ± 0.006
0.032 ± 0.007
MT plus ends redistribute at anaphase B onset
Anaphase B spindles exhibit a difference in MT turnover
Microtubule turnover during pre-anaphase B and anaphase B
16.6 ± 2.8
98.5 ± 12.2
16.8 ± 5.14
71.8 ± 21.2
10.7 ± 2.2
107.5 ± 11.4
17.5 ± 8.72
113.1 ± 12.3
The Drosophila syncytial blastoderm stage embryo is an essential tool in the study of mitosis. Injecting antibody and dominant negative protein inhibitors into the embryo, often in combination with the use of mitotic mutants permits detailed study of the role of mitotic proteins [21, 23]. However, the generation and characterization of reagents used in these embryo studies is often difficult and time consuming and it would be extremely useful to be able to rapidly identify potential novel components of the anaphase B switch prior to making this investment. Drosophila S2 cells, though less amenable to quantitation of spindle dynamics, are very useful for RNAi-mediated depletion of mitotic proteins. Indeed, they have previously been used to probe the functions of several mitotic proteins in S2 cell spindle assembly [12, 15, 16, 24, 25]. In order to assess the utility of S2 cell RNAi in identifying novel components of the anaphase B switch, it is important to understand whether this switch and the molecular mechanisms underlying it are conserved between embryos and S2 cells.
There is variation between and even among different species in the timing of anaphase B spindle elongation relative to anaphase A chromosome segregation . Spindles of the syncytial embryo undergo a distinct anaphase A chromosome segregation whilst maintaining a constant spindle length, which is followed just prior to anaphase A completion, by anaphase B spindle elongation. In S2 cells, however, anaphase A and B are more synchronized, with chromosome segregation occurring just prior to or at the same time as spindle elongation and completing at approximately the same time. Chromosome to pole motility also occurs at a much slower rate in S2 cells than in the embryo, at an average of 0.0098 μm/s (Table 1).
A key feature of the switch from constant spindle length pre-anaphase B to spindle elongation during anaphase B in the embryo is the suppression of poleward microtubule flux. Our data confirm and extend previous findings that this component of the anaphase B switch is conserved in S2 cells  and we further observe that there is an inverse correlation between the rate of poleward flux and the rate of spindle elongation, as in embryos. However, while flux is normally completely suppressed in embryonic spindles, it is only partially suppressed in S2 cell spindles (Table 2) and to a variable extent (Figure 3). In embryos, mitotic spindles are adapted to carry out the multiple rapid and synchronous mitoses that occur as the fertilized egg develops into the multicellular larva - here the complete suppression of poleward flux allows the spindle to elongate at its maximal rate. In S2 cells there is no such requirement for rapid, synchronous divisions, and here, the incomplete suppression of flux is consistent with a more stately, less streamlined pace of progression through mitosis.
An additional factor influencing this difference may be the contribution of poleward flux to chromosome segregation in S2 cells, which continues throughout spindle elongation. There is some debate in the literature as to the role of flux in chromosome segregation in S2 cells. Buster et al. reported that MT minus-end depolymerization associated with flux was the main driving force behind chromosome to pole motility . However, a subsequent paper by Matos et al. argued that Pacman activity, namely kinetochore motility coupled to MT plus end depolymerization, drives anaphase A in S2 cells, since the reduction in flux at anaphase B onset does not affect the mean velocity of kinetochore (k) MT shortening . The inverse linear relationship we observed between the rates of poleward flux and anaphase B indicates that suppression of flux is linked to spindle elongation at anaphase B onset. It is possible that flux is partially suppressed at anaphase B onset in S2 cells in order to permit spindle elongation, but that the lower rate of MT flux in combination with Pacman mechanisms continues to drive chromosome to pole motility.
The second component to the proposed anaphase B switch in Drosophila embryos is the redistribution of ipMT plus ends to the overlap zone at the spindle equator. We examined the distribution of the MT plus end tracking protein EB1 to determine if this phenomenon was conserved in S2 cells. While there was a difference in the EB1 distribution at anaphase B onset, only 25% of cells displayed concentrated EB1 fluorescence at the overlap region as in the embryo. In the majority of cells there was a general reduction in EB1 fluorescence across the spindle and redistribution away from the poles and towards the equator, but fluorescence was not as tightly redistributed to the midzone as in the embryo (Figure 4). A tight distribution of MT plus ends at the overlap region in S2 cells may not be necessary, since such a rapid and coordinated spindle elongation is not needed. It is also possible that EB1 binds to the polymerizing kMT ends on segregating sister chromatids , which would be moving away from the spindle midzone and therefore make the EB1 distribution appear more broadly spread across the spindle in S2 cells. Indeed, EB1 fluorescence becomes progressively more concentrated at the midzone over the course of anaphase B, by which point the chromatids should have fully separated (Figure 4B).
We also examined MT dynamics at regions proximal to the pole and the equator using FRAP and, as observed in the embryo, found that there was a reduction in fluorescence recovery close to the pole (Figure 5; Table 3). This further suggests that although the redistribution of EB1 fluorescence to the spindle midzone is not so pronounced in S2 cells, the alteration in spindle MT dynamics at anaphase B onset is conserved, such that MTs proximal to the pole stabilize. In future work it would be interesting to determine whether the role of cyclin B degradation in these changes is also conserved and to model the S2 cell data in a similar manner to the embryo. This would allow us to model the potential outcomes of RNAi depletion of candidate proteins involved in the anaphase B switch and interpret further experimental data.
There is a substantial level of conservation in the basic mechanisms underlying anaphase B in both the Drosophila syncytial blastoderm stage embryo and cultured S2 cells. Therefore, it would seem that RNAi studies might provide a useful means of identifying novel components of the anaphase B switch. However, there are a number of features of S2 cells that we feel may compromise their usefulness for such studies. Firstly, it is much more difficult to achieve the same level of quantitation of spindle dynamics in S2 cells as in the embryo since the mitotic index is much lower, with only 2-3% of cells undergoing mitosis at any one time. Furthermore, mitosis is not synchronized in S2 cells and, although there have been reports of methods to synchronize S2 cells and increase their mitotic index, we were unable to successfully achieve this goal over the course of our study. Since anaphase A and B occur much more synchronously than in the embryo, it is therefore important to remember that kMTs will still be present along with ipMTs in the S2 cell spindle during anaphase B and may contribute to any measurements of MT dynamics that are undertaken - because of the temporal separation of anaphase A and B in embryos, it is possible to specifically focus on ipMTs. In addition, S2 cells require flattening to the coverslip or culture dish prior to imaging. Concanavalin A represents the optimal means of achieving flattened cells, however it is important to minimise the amount of time that cells are imaged after flattening since it can begin to interfere with cytokinesis. We found that the optimal imaging time was from 20 minutes up to 2 hours after flattening (data not shown). Furthermore, S2 cells are very sensitive to light and prolonged exposure to the imaging laser can damage the cells and arrest mitosis. We therefore only imaged cells that had entered anaphase less than 40 minutes after NEB, and added Vitamin C (ascorbic acid) to the medium to act as an antioxidant and avoid build up of cell damage. Finally, compared to untransformed cells in other wild-type organisms, mitosis in the S2 cell line is more variable and often looks abnormal even in untreated cells. It is therefore of great importance to establish robust control samples and to perform objective imaging and image analysis in order to accurately discriminate bona fide RNAi-induced effects from simple cellular variation.
In conclusion, we have demonstrated that the basic components of an anaphase B switch previously observed in the Drosophila syncytial embryo are conserved in cultured S2 cells, at least qualitatively. Thus S2 cells may provide a useful model for identifying novel molecular components of the anaphase B switch via RNAi, bearing in mind that a number of technical difficulties peculiar to the study of anaphase B in this cell line will need to be surmounted first. The quantitative differences, which can be generalized to an "all-or-none" versus a "partial" switch in embryo versus S2 cell spindles, respectively, are likely to reflect differences in the speed and synchronicity of mitosis and the structure of the spindles in the two systems.
All experiments reported here were conducted on Drosophila S2 cells cultured in 1× Schneider's Dosophila medium (Gibco) supplemented with 10% FBS, 1 unit/ml penicillin, and 1 μg/ml streptomycin, in 75 cm2 flasks at 27°C. Prior to microscopy, S2 cells were flattened onto glass-bottomed culture dishes (MatTek Corporation) coated with 0.25 mg/ml concanavalin A (Sigma Aldrich) for 20 minutes. Vitamin C was also added to the medium to act as an antioxidant and minimise cell death during imaging.
Time lapse microscopy
Time-lapse images were collected using an Olympus microscope equipped with an UltraView spinning disk confocal head (Perkin Elmer) with a 100× 1.35 NA objective. A single confocal plane was acquired at a rate of 5-10 s/frame at 24-26°C and recorded using a charge-coupled device camera (Orca II; Hamamatsu Photonics).
Fluorescent speckle microscopy and kymography
To measure flux rates we used S2 cells stably expressing GFP-α-tubulin under the control of a leaky inducible metallothionein promoter (a gift from Prof. Helder Maiato, University of Porto, Portugal), which express low levels of GFP-α-tubulin without induction. Images were analyzed using MetaMorph Imaging Software (Universal Imaging). The Sharpen and Low-Pass Filter commands were applied and speckle movement was quantified using kymography analysis, in which moving speckles appeared as oblique lines whose slope corresponds to their rate of movement. All calculations were performed in Microsoft Excel and flux rate was calculated relative to the movement of the spindle poles over time, such that the movement of the pole was subtracted from that of the speckle.
Fluorescence recovery after photobleaching
For FRAP experiments we used S2 cells stably expressing GFP-α-tubulin (a gift from Dr. David Sharp) and a laser-scanning confocal microscope (FV1000; Olympus) with a 60× 1.40 NA objective. Image acquisition was performed using the Fluoview software (version 1.5; Olympus). Cells were imaged using the 488 nm line from an argon laser, whilst a separate 405 nm laser was used to photobleach GFP-α-tubulin, allowing simultaneous imaging and bleaching. The spindle was bleached in rectangles of 1 μm diameter proximal to either the pole or equator and images were acquired every 3 s. The spindles were corrected for movement using Matlab (Mathworks) and the fluorescence intensity of the bleached regions over time were measured using MetaMorph imaging software. All data were normalized by I(t) = [F(t)Tpre]/[T(t)Fpre], where F(t) and T(t) are the mean fluorescence in the FRAP region and in the entire spindle at time t, and Fpre and Tpre are the mean fluorescence in the FRAP region and in the entire spindle just before the bleach. The percentage recovery was obtained by using the calculation (Finf-F0)/(Fpre-F0), where F0 is the mean intensity in the FRAP region just after photobleaching and Finf is the final fluorescence obtained from the single exponential fit. The fit to the data also yielded the recovery half time.
The distribution of EB1 across the spindle over time was analyzed using S2 cells stably expressing RFP-EB1 (also a gift from Dr Steve Rogers). A time lapse movie was acquired as described above, the spindle was aligned using Matlab and then a whole spindle kymograph was generated using MetaMorph imaging software.
Chromosome to pole motility
Chromosome-to-pole motility rates were measured using S2 cells stably expressing GFP-histone2B and RFP-α-tubulin (a gift from Prof. Ronald D. Vale, University of California, San Francisco, USA).
We thank Prof. Helder Maiato, Prof. Ron Vale, Dr. David Sharp and Dr. Steve Rogers for the generous gifts of S2 cell lines and advice on working with S2 cells. Thanks also to the cytoskeleton group at the University of California, Davis and other members of the Scholey laboratory for useful discussion. The work was supported by National Institute of Health Grant GM55507 to JMS.
- Walczak CE, Heald R: Mechanisms of mitotic spindle assembly and function. Int Rev Cytol 2008, 265: 111–58. full_textView ArticlePubMedGoogle Scholar
- Karsenti E, Vernos I: The mitotic spindle: a self-made machine. Science 2001, 294: 543–7. 10.1126/science.1063488View ArticlePubMedGoogle Scholar
- Mitchison TJ, Salmon ED: Mitosis: a history of division. Nat Cell Biol 2001, 3: E17–21. 10.1038/35050656View ArticlePubMedGoogle Scholar
- Wittmann T, Hyman A, Desai A: The spindle: a dynamic assembly of microtubules and motors. Nat Cell Biol 2001, 3: E28–34. 10.1038/35050669View ArticlePubMedGoogle Scholar
- Goshima G, Scholey JM: Control of mitotic spindle length. Annu Rev Cell Dev Biol 2010, 26: 21–57. 10.1146/annurev-cellbio-100109-104006View ArticlePubMedGoogle Scholar
- Scholey JM, Brust-Mascher I, Mogilner A: Cell division. Nature 2003, 422: 746–52. 10.1038/nature01599View ArticlePubMedGoogle Scholar
- Gadde S, Heald R: Mechanisms and molecules of the mitotic spindle. Curr Biol 2004, 14: R797–805. 10.1016/j.cub.2004.09.021View ArticlePubMedGoogle Scholar
- Wadsworth P, Khodjakov A: E pluribus unum: towards a universal mechanism for spindle assembly. Trends Cell Biol 2004, 14: 413–9. 10.1016/j.tcb.2004.07.004View ArticlePubMedGoogle Scholar
- Brust-Mascher I, Civelekoglu-Scholey G, Kwon M, Mogilner A, Scholey JM: Model for anaphase B: role of three mitotic motors in a switch from poleward flux to spindle elongation. Proc Natl Acad Sci USA 2004, 101: 15938–43. 10.1073/pnas.0407044101PubMed CentralView ArticlePubMedGoogle Scholar
- Brust-Mascher I, Scholey JM: Microtubule flux and sliding in mitotic spindles of Drosophila embryos. Mol Biol Cell 2002, 13: 3967–75. 10.1091/mbc.02-05-0069PubMed CentralView ArticlePubMedGoogle Scholar
- Civelekoglu-Scholey G, Sharp DJ, Mogilner A, Scholey JM: Model of chromosome motility in Drosophila embryos: adaptation of a general mechanism for rapid mitosis. Biophys J 2006, 90: 3966–82. 10.1529/biophysj.105.078691PubMed CentralView ArticlePubMedGoogle Scholar
- Rogers GC, Rogers SL, Schwimmer TA, Ems-McClung MC, Walczak CE, Vale RD, Scholey JM, Sharp DJ: Two mitotic kinesins cooperate to drive sister chromatid separation during anaphase. Nature 2004, 427: 364–70. 10.1038/nature02256View ArticlePubMedGoogle Scholar
- Brust-Mascher I, Scholey JM: Kinesin-5-dependent poleward flux and spindle length control in Drosophila embryo mitosis. Mol Biol Cell 2009, 20: 1749–62. 10.1091/mbc.E08-10-1033PubMed CentralView ArticlePubMedGoogle Scholar
- Cheerambathur DK, Civelekoglu-Scholey G, Brust-Mascher I, Sommi P, Mogilner A, Scholey JM: Quantitative analysis of an anaphase B switch: predicted role for a microtubule catastrophe gradient. J Cell Biol 2007, 177: 995–1004. 10.1083/jcb.200611113PubMed CentralView ArticlePubMedGoogle Scholar
- Goshima G, Wollman R, Goodwin SS, Zhang N, Scholey JM, Vale RD, Stuurman N: Genes required for mitotic spindle assembly in Drosophila S2 cells. Science 2007, 316: 417–21. 10.1126/science.1141314PubMed CentralView ArticlePubMedGoogle Scholar
- Morales-Mulia S, Scholey JM: Spindle pole organization in Drosophila S2 cells by dynein, abnormal spindle protein (Asp), and KLP10A. Mol Biol Cell 2005, 16: 3176–86. 10.1091/mbc.E04-12-1110PubMed CentralView ArticlePubMedGoogle Scholar
- Maiato H, Khodjakov A, Rieder CL: Drosophila CLASP is required for the incorporation of microtubule subunits into fluxing kinetochore fibres. Nat Cell Biol 2005, 7: 42–7. 10.1038/ncb1207PubMed CentralView ArticlePubMedGoogle Scholar
- Matos I, Pereira AJ, Lince-Faria M, Cameron LA, Salmon ED, Maiato H: Synchronizing chromosome segregation by flux-dependent force equalization at kinetochores. J Cell Biol 2009, 186: 11–26. 10.1083/jcb.200904153PubMed CentralView ArticlePubMedGoogle Scholar
- Clemens JC, Worby CA, Simonson-Leff N, Muda M, Maehama T, Hemmings BA, Dixon JE: Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways. Proc Natl Acad Sci USA 2000, 97: 6499–503. 10.1073/pnas.110149597PubMed CentralView ArticlePubMedGoogle Scholar
- Maiato H, Rieder CL, Khodjakov A: Kinetochore-driven formation of kinetochore fibers contributes to spindle assembly during animal mitosis. J Cell Biol 2004, 167: 831–40. 10.1083/jcb.200407090PubMed CentralView ArticlePubMedGoogle Scholar
- Brust-Mascher I, Scholey JM: Microinjection techniques for studying mitosis in the Drosophila melanogaster syncytial embryo. J Vis Exp 2009, 15: 1382.Google Scholar
- Hayashi T, Sano T, Kutsuna N, Kumagai-Sano F, Hasezawa S: Contribution of anaphase B to chromosome separation in higher plant cells estimated by image processing. Plant Cell Physiol 2007, 48: 1509–13. 10.1093/pcp/pcm117View ArticlePubMedGoogle Scholar
- Wang H, Brust-Mascher I, Cheerambathur D, Scholey JM: Coupling between microtubule sliding, plus-end growth and spindle length revealed by kinesin-8 depletion. Cytoskeleton (Hoboken) 2010, 67: 715–28.View ArticleGoogle Scholar
- Kwon M, Morales-Mulia S, Brust-Mascher I, Rogers GC, Sharp DJ, Scholey JM: The chromokinesin, KLP3A, dives mitotic spindle pole separation during prometaphase and anaphase and facilitates chromatid motility. Mol Biol Cell 2004, 15: 219–33. 10.1091/mbc.E03-07-0489PubMed CentralView ArticlePubMedGoogle Scholar
- Goshima G, Wollman R, Stuurman N, Scholey JM, Vale RD: Length control of the metaphase spindle. Curr Biol 2005, 15: 1979–88. 10.1016/j.cub.2005.09.054View ArticlePubMedGoogle Scholar
- Buster DW, Zhang D, Sharp DJ: Poleward tubulin flux in spindles: regulation and function in mitotic cells. Mol Biol Cell 2007, 18: 3094–104. 10.1091/mbc.E06-11-0994PubMed CentralView ArticlePubMedGoogle Scholar
- Tirnauer JS, Canman JC, Salmon ED, Mitchison TJ: EB1 targets to kinetochores with attached, polymerizing microtubules. Mol Biol Cell 2002, 13: 4308–16. 10.1091/mbc.E02-04-0236PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.