|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online November 26, 2007
doi: 10.1242/10.1242/dev.010983
1 Max Planck Institute of Molecular Cell Biology and Genetics,
Pfotenhauerstrasse 108, 01307 Dresden, Germany.
2 Liverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA,
UK.
3 Institut de Recerca Biomèdica (IRB) and Institució Catalana de
Recerca i Estudis Avançats (ICREA), Parc Cientific Barcelona, Josep
Samitier 1-5, 08028 Barcelona, Spain.
4 CNRS UMR144, Institut Curie, Membrane and Cytoskeleton Dynamics Group, 26 Rue
d'Ulm 75005 Paris, France.
5 Hybrigenics SA, 3-5 Impasse Reille, 75014 Paris, France.
6 Département de biochimie, Sciences II, 30, Quai Ernest Ansermet CH-1211
Genève 4, Switzerland.
Author for correspondence (e-mail:
marcos.gonzalez{at}biochem.unige.ch)
Accepted 20 September 2007
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key words: Drosophila, Meiosis, Spermatogenesis, Testis
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Cytokinesis requires coordination between the central
spindle and the actomyosin contractile ring
(Adams et al., 1998;
Jantsch-Plunger et al., 2000;
Somers and Saint, 2003).
Recently, membrane trafficking has also been implicated in cytokinesis
(Albertson et al., 2005;
Glotzer, 2005). Membrane
trafficking is necessary for the massive increase in plasma membrane surface
area, and can also locally enrich specific components in the cleavage furrow
plasma membrane (Bluemink and de Laat,
1973, VerPlank and Li,
2005).
The centralspindlin complex, composed of Pavarotti (Pav), RacGAP50C, and
the associated Rho guanine nucleotide exchange factor (GEF) Pebble, is
essential for communication between central spindle microtubules and the
actomyosin contractile ring, probably by regulating the activity of the small
GTPase RhoA (Somers and Saint,
2003). Active RhoA regulates actin polymerization, myosin II and
citron kinase (Amano et al.,
1996; Matsui et al.,
1996; Yamashiro et al.,
2003). Around 20 highly conserved proteins, including central
spindle, actin myosin ring and RhoA pathway machinery are required for
cytokinesis in multiple systems (Glotzer,
2005). However, recent RNAi screens have repeatedly identified
membrane-trafficking components necessary for cytokinesis
(Echard et al., 2004;
Eggert et al., 2004;
Skop et al., 2004).
The three main classes of trafficking factors implicated in cytokinesis are
components of the secretory pathway, endocytic and/or recycling factors and
membrane fusion machinery. Golgi proteins required for the secretory pathway
such as Cog5 and Syntaxin5 are required for cytokinesis
(Farkas et al., 2003;
Xu et al., 2002). Endocytic
recycling factors necessary for cytokinesis include Rab11 and
Rab11FIP3/Arfophilin, which associate with the central spindle and furrow
cortex (Skop et al., 2001;
Wilson et al., 2005). Fusion
machinery such as the exocyst complex and t- and v-SNAREs localizes at the
mitotic midbody and furrow cortex, and is necessary for cytokinesis
(Fielding et al., 2005;
Finger et al., 1998;
Gromley et al., 2005;
Jantsch-Plunger and Glotzer,
1999; Low et al.,
2003).
One membrane-trafficking component implicated in cytokinesis is the class
III ADP ribosylation factor, ARF6. A constitutively active, GTPase-defective
ARF6 mutant, ARF6Q67L, concentrated at the central spindle and midbody of HeLa
cells, and ARF6Q67L overexpression caused late cytokinesis defects
(Schweitzer and D'Souza-Schorey,
2002). Knockdown of ARF6 in HeLa cells using siRNA caused a late
cytokinesis block (Schweitzer and
D'Souza-Schorey, 2005). Possible effectors for ARF6 during
cytokinesis are the Rab11/ARF6 binding proteins Rab11FIP3 and Rab11FIP4, which
ARF6 recruits to the central spindle in HeLa cells
(Fielding et al., 2005;
Schweitzer and D'Souza-Schorey,
2002; Schweitzer and
D'Souza-Schorey, 2005). ARF6 regulates endocytosis, recycling and
actin remodelling (D'Souza-Schorey et al.,
1995; Radhakrishna and
Donaldson, 1997; Song et al.,
1998). Via these mechanisms, ARF6 affects processes such as
adherens junction disassembly and cell migration, including the formation of
cord-like structures by hepatocytes in the mouse liver in response to
hepatocyte growth factor (D'Souza-Schorey
and Chavrier, 2006; Suzuki et
al., 2006). In Drosophila, the ARF6-GEF Loner/Schizo is
necessary for myoblast fusion (Chen et
al., 2003) and midline crossing of axons
(Onel et al., 2004).
In contrast to the central spindle microtubules and the actomyosin
contractile ring, little is known about temporal and spatial coordination of
membrane trafficking during cytokinesis. Many membrane-trafficking components
are localized to the central spindle (Skop
et al., 2004), but the molecular machinery connecting the central
spindle to membrane trafficking is unclear. Here we show that cytokinesis
requires ARF6 in the Drosophila male germ line. ARF6 localizes to the
plasma membrane and a population of early and recycling endosomes. In dividing
cells, ARF6 is specifically enriched on recycling endosomes associated with
the Pav central spindle. ARF6 is not required to target recycling endosomes to
the central spindle, but is required for rapid membrane addition during
cytokinesis. We suggest that ARF6 enrichment on recycling endosomes at the
central spindle increases the rate of recycling to the plasma membrane, thus
coordinating membrane recycling with the central spindle and cleavage furrow
invagination during cytokinesis.
|
MATERIALS AND METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
;
Giansanti et al., 1998;
Salzberg et al., 1994).
Pav-GFP, GFP-
-tubulin, His2AvDGFP and DE-cad-GFP under polyubiquitin
promoter control (Lee et al.,
1988), and Sqh-GFP with the sqh promoter were previously
described (Clarkson and Saint,
1999; Minestrini et al.,
2003; Oda and Tsukita,
2001; Rebollo et al.,
2004; Royou et al.,
2004). 
-Tubulin GFP flies were a gift from S. Llamazares
(IRB and ICREA, Barcelona, Spain).
Transgenics
Transgenic flies were generated by injecting the following vectors:
P(Ubi:arf6-HA), a PCR product containing ARF6 coding sequence from the LD22876
clone (BDGP EST Project) with SacI and XbaI sites in primer
overhangs and including a C-terminal HA epitope tag was cloned between
SacI and XbaI sites of pSRalpha. A
SacI-XbaI fragment excised from PSRalpha was ligated between
SacI and XbaI sites of TOPO. A KpnI-XbaI
fragment from the resulting vector was cloned between the polyubiquitin vector
KpnI and SpeI sites. P(w+arf6+)
arf6 rescue construct, a 3.8 kb PCR product from genomic DNA
containing arf6 and flanking sequences was cloned into TOPO-XL. The
3.8 kb fragment from XbaI, NotI and SphI digestion
was inserted between NotI and XbaI sites of pCasper4.
P(UbiGFP-Rab5) and P(UbiGFP-Rab11) were generated from pUAST-GFP-Rab5
(Wucherpfennig et al., 2003),
and pUAST-GFP-Rab11 (Emery et al.,
2005). P(UbiGFP-Rab4), PCR product containing rab4-coding
sequence from pOT2-GH18176 with primer overhang XhoI-sites was cloned
into the XhoI site of pEGFP-C3. The NheI-XbaI
fragment from pEGFP-C3-rab4 was inserted in the pUAST XbaI-site. In
all cases, GFP-Rab containing NotI-XbaI fragments from pUAST
were cloned between polyubiquitin vector NotI and XbaI
sites.
Antibodies and microscopy
A rabbit polyclonal antibody generated (Eurogentec) against amino acids
99-112 of Drosophila ARF6 (ARTELHRIINDREM) binds ARF6 at 20 kDa in
western blots. Rabbit anti actin (Sigma A2066) was used 1:400. Anti-ARF6 was
used 1:50 for western blotting, but was unsuitable for immunofluorescence.
Embryo immunofluorescence staining was performed using standard techniques.
Mouse antibody BP102 (Hybridoma Bank) was used 1:30, and Rabbit anti MHC
(Kiehart and Feghali, 1986) at
1:500. arf6 embryos zygotically rescued by CyO, hb-lacZ were
identified using rabbit anti β-galactosidase (Cappel) 1:500. Testes
dissected in PBS were fixed for 20 minutes in PBS containing 4%
paraformaldehyde, and a further 20 minutes after the addition of 0.2% Triton
X-100. After washing with PBS, subsequent staining and washing steps were
performed in PBS containing 0.1% Triton X-100. A 2-hour block with 0.5% BSA
was followed by overnight incubation at 4°C with 0.5% BSA and primary
antibodies: rat anti-HA (clone 3F10, Roche), 1:500, rabbit anti-Pav 1:250
(Adams et al., 1998). After
three 20-minute washes, primary antibodies were detected using Alexa Fluor
546-conjugated anti-rat (Molecular Probes) and Cy5-conjugated anti-rabbit
(Jackson ImmunoResearch Laboratories) antibodies at 1:500, with 2% normal goat
serum for 2 hours at room temperature.
Confocal images were acquired using Zeiss LSM 510 and LeicaDMIRE2, TCS SP2 SP2 microscopes, with 63x (NA 1.4) and 100x (NA 1.4) objective lenses. Colocalization (± s.e.m.) was quantified in unprocessed images in the Zeiss LSM image browser by manually counting punctae. In dividing cells, punctae within 3 µm of Pav staining were classified as `central spindle' localized, other punctae as `non central spindle'. Images were processed for contrast/brightness, levels and `dust and scratches' with Adobe Photoshop 7.0 (Adobe Systems).
Live imaging
Spermatocyte imaging was carried out as described
(Rebollo and Gonzalez, 2004).
Cell perimeter and diameter were measured in ImageJ
(http://rsb.info.nih.gov/ij/).
Furrow ingression, perimeter and surface area rates are linear regression line
slopes. Four to six confocal sections were maximally projected, except
Sqh-GFP.
Germ line clones of arf6 mutants
Germline clones were generated using the flp/FRT systems described
(Chou and Perrimon, 1992).
Females (genotype y w hsflp;
FRTG13OvoD1/FRTG13arf61) raised at 25°C were
heat shocked for 2 hours at 38°C as third instar larvae to activate the
flippase, generating germ line clones (genotype y w hsflp;
FRTG13arf61/FRTG13arf61). y w hsflp;
FRTG13OvoD1/FRTG13arf61 females were crossed for 3
days to wild-type males in vials supplemented with fresh yeast before
collecting eggs. The same procedure was used for
arf63.
Yeast two-hybrid analysis
PCR from clone LD22876 generated a Glu67-Leu substitution of the ARF6 ORF.
NotI and SpeI overhang sites were generated by PCR and the
resulting fragment cloned into pB27 bait plasmid derived from pBTM116
(Vojtek and Hollenberg, 1995).
A random-primed cDNA library from 0-24 hour Drosophila embryo
poly(A+) RNA was constructed into the pP6 plasmid derived from
pGADGH (Bartel et al., 1993).
The two-hybrid system was used to detect protein-protein interactions
(Bartel, 1993). The library was
transformed into the Y187 yeast strain. Around 10 million independent yeast
colonies were collected, pooled and stored at -80°C, and over 50 million
interactions tested using a previously described mating protocol
(Fromont-Racine et al., 1997).
Prey fragments of positive clones were PCR amplified and sequenced at 5'
and 3' junctions. Corresponding genes were identified in the GenBank
database (NCBI) using an automated procedure
(Formstecher et al.,
2005).
Pav-binding assay
DNA encoding Pav655-865 was cloned into pGEX4T1 at the GST C-terminus.
pGEX4T1Pav655-865 and pGEX4T1 were transformed into E. coli strain
B21, and expression induced by 1 mM isopropyl β-D-thiogalactopyranoside
for 5 hours at 20°C. GST proteins were affinity purified using
glutathione-Sepharose beads (Amersham Biosciences), eluted using glutathione,
dialyzed against 20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2 mM EDTA, 2 mM
β-mercaptoethanol and 10% glycerol, and stored at -80°C. HeLa cells
were transfected with pSRalpha(ARF6HA) and pSRalpha(ARF6Q67LHA) using
Effectene (Qiagen), lysed after 20-24 hours in 50 mM Tris-HCl pH 5.5, 137 mM
NaCl, 1% Triton X-100, 10 mM MgCl2, 10% glycerol (Buffer B) with
complete protease inhibitor cocktail tablets (Roche), and centrifuged for 15
minutes at 13,000 rpm (15,000 g) at 4°C. Supernatants were
incubated with 20 µg of GST fusion protein for 15 minutes at 4°C with
0.5% BSA, and for 1-2 hours at 4°C after adding glutathione-Sepharose
beads. Beads were washed three times in Buffer B, once in Buffer B containing
0.1% SDS and once in PBS. Bound proteins were eluted using 4x NuPage LDS
sample buffer (Invitrogen) and assayed by western blotting with rabbit anti-HA
antibody at 1:500 (Roche) and affinity-purified polyclonal rabbit anti-GST at
1:10,000 (Protein Expression and Purification Facility, MPI-CBG, Dresden).
|
 |
Let P1 be the perimeter, and S1 the
surface area at time t1, and P2 be the
perimeter and S2 the surface area at time
t2.The perimeter change between t1 and
t2 is:
 |
 |
Cell volume was calculated as the product of the cell area in each slice and the slice separation, from stacks of optical slices taken at 0.51 µm intervals through spermatocytes expressing DE-cad-GFP.
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ARF6 is not essential for fly viability. Homozygous
arf61 progeny from homozygous mutant mothers
(maternal/zygotic mutants) were viable until adulthood, presenting no overt
external morphological phenotype. Recent reports show that expression of a
GDP-bound dominant negative ARF6 protein (ARF6TN) impairs myoblast fusion and
axon path finding during embryogenesis
(Chen et al., 2003;
Onel et al., 2004). Both
developmental events occurred normally in arf6 null mutant embryos
(see Fig. S1B-E in the supplementary material), indicating that the ARF6TN
protein causes secondary defects beyond the suppression of ARF6 function.
Female arf61 flies showed reduced fertility because of
a partially penetrant requirement for ARF6 during chorion formation in the
germ-line (see Fig. S2 and Table S1 in the supplementary material). Male
arf61 flies were completely sterile. Mutant spermatids
showed a `four-wheel-drive' phenotype, indicating a cytokinesis defect during
spermatocyte meiosis (Fuller,
1993) (Fig. 1B-E):
over 90% of spermatids contained more than one nucleus, and 41% had four
nuclei per mitochondrial Nebenkern derivative
(Fig. 1E), corresponding to 79%
failure in cytokinesis during the two meiotic divisions (see Fig. S3 in the
supplementary material). 1.9% of cells showed 8:1 nuclei-to-Nebenkern
ratios, suggesting that cytokinesis of gonial cell mitosis prior to meiosis is
also occasionally affected, as previously suggested for other cytokinesis
mutants (Brill et al., 2000;
Giansanti et al., 2004).
arf62 and arf63 showed phenotypes
indistinguishable from arf61. Male sterility and the
mutant cytokinesis phenotype are due to the arf61
mutation, because an arf6+ genomic transgene rescued the
defects, yielding fertile males (Fig.
1D,E).
|
-Tubulin GFP showed that chromosome
segregation and centrosome behavior occurs normally (not shown).
-Tubulin GFP fusion (GFP--tubulin) indicated that spindle
initially forms normally in arf61 mutant spermatocytes
(see Movies 1, 2 in the supplementary material). In arf6 mutant
cells, a cleavage furrow was established, but later regressed. We therefore
analyzed furrow ingression kinetics in wild-type and arf6 mutant
spermatocytes.
In control cells, cell shape and diameter were constant until anaphase
onset (see Fig. 4B and Movies
1, 3, 4, 7 and 9 in the supplementary material). After anaphase onset, cells
elongated, decreasing in equatorial diameter at 0.4 µm/minute
(Fig. 3B,
Fig. 4C). During anaphase B,
which starts 2 minutes after anaphase onset (min AA), Pav accumulated at the
central spindle (Minestrini et al.,
2003) (Fig. 2A,
Fig. 3A,
Fig. 4C; see Movie 3 in the
supplementary material). The centralspindlin complex subsequently signals to
the cortex, and the actomyosin contractile ring forms. Myosin regulatory light
chain (Sqh-GFP) accumulated at the future cleavage furrow 1 minute after the
onset of Pav accumulation at the central spindle
(Fig. 2D,
Fig. 3A; see Movie 4 in the
supplementary material) (Adams et al.,
1998; Royou et al.,
2004; Somers and Saint,
2003). Shortly after Pav and Sqh accumulation, equator contraction
accelerated to 1 µm/minute (Fig.
3, Fig. 4C). Five
minutes after Pav accumulation, a plasma membrane indentation, the `cleavage
furrow', appeared (Fig. 2A,
Fig. 3A,
Fig. 4C). The furrow progressed
at 1 µm/minute, finally decelerating to stop around 35 minutes after
anaphase onset at width 3-5 µm (Fig.
2A, Fig. 3A,
Fig. 4C; see Fig. S4A in the
the supplementary material). This narrow opening between the two daughter
cells differentiates into the ring canal
(Hime et al., 1996).
|
In `early regressors', anaphase B-cell elongation occurred. The equator contracted at 0.3 µm/minute, only slightly slower than in the wild type (Fig. 3B, Fig. 4C). Pav and Sqh central spindle and contractile ring targeting occurred only slightly later than in the wild type (Fig. 2B,E, Fig. 4C; see Movies 5 and 6 in supplementary material). However, fast equator contraction did not occur, only accelerating to 0.5 µm/minute, and indentation occurred 10 minutes, instead of 5 minutes after central spindle Pav accumulation (Fig. 2B, Fig. 3A, Fig. 4C). Shortly after the cleavage furrow indentation appeared, when it is around 15 µm wide, the furrow regresses. After cleavage furrow collapse, Pav dissociates from the central spindle, although Sqh remains at the cortex during regression (Fig. 2E; see Movie 6 in the supplementary material).
These observations indicate that ARF6 is not necessary for targeting Pav to the central spindle, or for actomyosin ring formation. The mutant phenotypes reveal two critical phases for ARF6 function during cytokinesis: (1) an early role during cleavage furrow progression after furrow initiation, and (2) a later role in ring canal establishment at the end of cytokinesis. Since ring canal stabilization may be a specialized event restricted to germ cells, and may be due to the higher frequency of the early regressors, we decided to concentrate on the early regressors and the early role of ARF6 in furrow progression.
ARF6 is required for rapid plasma membrane addition during cytokinetic cleavage furrow progression
Spermatocytes divide, producing two daughter cells with half the volume of
the mother (volume change 0.8±1.4%, n=5 cells). For spherical
cells, this implies that the membrane surface increases by 26% during
cytokinesis. We therefore investigated whether the arf6 phenotype is
caused by a defect in membrane addition to the cell surface. The absence of
surface increase could lead to an increase in membrane tension, which would
counteract the forces generated by the contractile ring. This hypothesis was
prompted by the established role of ARF6 during endocytic membrane recycling
(D'Souza-Schorey et al., 1998;
Prigent et al., 2003;
Radhakrishna and Donaldson,
1997) which might be essential for rapid membrane addition from an
endosomal, ARF6-dependent membrane store.
The kinetics of plasma membrane increase during meiosis I was studied by measuring cell perimeter of spermatocytes in confocal images, as well as the total surface area calculated for 3D-reconstructed cells (Fig. 4; see Fig. S4 in the supplementary material). Experiments on the relationship between perimeter and surface area changes indicated that perimeter increase correlates well with surface area increase (see Fig. S4C in the supplementary material). Plasma membrane growth during meiosis I was negligible prior to anaphase. Membrane increase started during anaphase B cell elongation at 0.6 µm/minute. This `perimeter rate' corresponds to a membrane addition of around 8 µm2/minute (see models in the supplementary material). The perimeter rate increased greatly (2.5-fold, corresponding to around 22 µm2/minute) directly after the onset of Pav accumulation, peaking around 15 min AA at the time of maximum furrow ingression rate and the appearance of membrane indentation (Fig. 4B,C). Subsequently, the rate decreased until the completion of cytokinesis. In arf6 mutants, slow membrane addition characteristic of early cytokinesis was maintained after furrow membrane indentation, and the rapid membrane addition phase never occurred (Fig. 4B,C; see Fig. S4B in the supplementary material). These data suggest that ARF6 is involved in rapid membrane addition to the plasma membrane, which is necessary during the rapid contraction of the actin ring during cytokinesis.
Membrane addition to the plasma membrane is uncoupled from actomyosin ring contraction
The arf6 mutant phenotype reveals a link between cleavage furrow
progression and rapid membrane surface increase. To find out whether cleavage
furrow progression defects lead to a defect in surface increase, or vice
versa, we studied furrow progression and membrane addition rates in profilin
chickadee (chic) mutants
(Cooley et al., 1992;
Giansanti et al., 1998). In
chic13E mutants, the central spindle initially formed
normally and Pav was targeted properly
(Fig. 2C), but actomyosin
contractile ring formation fails
(Giansanti et al., 1998). As a
consequence, furrow ingression kinetics are even more affected than in
arf6 mutants (Fig. 3).
However, in these chic13E cells, membrane addition
initially occurred with kinetics similar to control cells, until the premature
disassembly of the central spindle and Pav-GFP disappearance from the central
spindle area (Fig. 2C,
Fig. 4A,B; see Fig. S4B in the
supplementary material).
|
ARF6 endosomes are associated with the Pav central spindle during cytokinesis
We then studied the subcellular localization of ARF6, its possible
association with intracellular endosomal membranes, and the Pav central
spindle. We used GFP-Rab4 as a marker for early endosomes along the fast
recycling route to the plasma membrane, and GFP-Rab11 to label recycling
endosomes along the kinetically slower recycling route
(Sheff et al., 1999;
van der Sluijs et al., 1992).
Functional HA-tagged ARF6 was expressed from the polyubiquitin promoter. This
expression rescued the arf6 mutant cytokinesis and chorion phenotypes
(Fig. 1E; see Fig. S2D in the
supplementary material). ARF6-HA was present in the cytosol and enriched at
endosomes and the plasma membrane, as previously reported in mammalian cells
(Fig. 5; see Fig. S5 in the
supplementary material) (D'Souza-Schorey
and Chavrier, 2006).
The results of the localization analysis are summarized in Table 1. Early during meiosis I cytokinesis, ARF6-positive endosomes (66% of punctae) associated with Pav-positive central spindle microtubules including the cortical microtubule population where the cleavage furrow forms (Table 1, Fig. 5A-B,D, arrowheads; see Fig. S5D in the supplementary material). This contrasts with 52% of Rab4 endosomes and 40% of Rab5 endosomes associated with the central spindle during meiosis I (Table 1). The ARF6 central spindle endosomal population corresponds mainly (87%) to Rab4-labelled endosomes.
|
In summary, recycling endosomes at the central spindle contain ARF6. Is ARF6 specifically enriched in the central spindle population of Rab4 and Rab11 recycling endosomes? Most central spindle Rab4 endosomes (73%) were decorated by ARF6 whereas only 22% of the Rab4 endosomes not localized to the central spindle contained ARF6 (Table 1, Fig. 5C). Similarly, 85% of Rab11 endosomes at late central spindles contained ARF6, versus 10% elsewhere in the cell (Table 1, Fig. 5C). These data indicate that ARF6 targeting to recycling endosomes is specifically biased towards the central spindle endosomal population.
Since central spindle recycling endosomes are enriched in ARF6, we asked whether ARF6 itself targets recycling endosomes to the central spindle. We therefore observed Rab4 and Rab11 endosome distribution in time-lapse movies. In arf61 mutants, rab4 endosomes were targeted to the central spindle as in wild-type controls (Fig. 6A,B; see Movies 7 and 8 in the supplementary material). Therefore, ARF6 does not target endosomes to the spindle, but instead functions downstream of endosomal targeting. ARF6 does not seem to play a direct role in Rab11 targeting either. In the arf6 late regressors, the regression occurred around the time when Rab11 accumulation was clearly observed. However, we observed arf6 late regressors in which Rab11 endosome localization does not appear to be affected (Fig. 6C,D; see Movies 9 and 10 in the supplementary material).
|
ARF6 binds the centralspindlin component Pavarotti
What targets ARF6 to central spindle endosomes? Using a Drosophila
embryo cDNA library, Pav was identified in a two-hybrid screen for proteins
interacting with Drosophila ARF6Q67L mutant
(Fig. 7A). Five clones
corresponding to the Pav ORF define the ARF6-binding domain: a region adjacent
to the coiled-coil domain of Pav (amino acids 727-844;
Fig. 7B). Binding assays
confirmed this interaction (Fig.
7C). These results suggest that Pav might contribute to ARF6
recruitment to central spindle endosomes.
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasma membrane addition during cytokinesis: secretory versus endocytic trafficking
Four solutions exist to the demand for rapid surface area increase during
cytokinesis: (1) decreasing cell volume; (2) stretching existing membrane; (3)
resolving membrane microvilli; and (4) delivering membrane to the surface.
There are no reports of cell volume decrease during cytokinesis. Under
tension, the surface of biological membranes stretches by only 2-3% before
lysis, (Needham and Hochmuth,
1989). In P815Y mastoma cells, unfolding of microvilli accumulated
during interphase is sufficient to account for the surface area increase
during cytokinesis (Knutton et al.,
1975). By contrast, microvilli in some ascidian eggs show the
converse behavior, increasing in number during cleavage furrow progression and
disappearing during interphase, suggesting that this mechanism is not
conserved (Satoh and Deno,
1984). Cells that do not produce sufficient extra surface area
between G1 and cytokinesis, must increase membrane surface during cytokinesis
by delivering membrane to the surface.
|
|
). Both secretory and endocytic factors are implicated
in cytokinesis (Echard et al.,
2004; Eggert et al.,
2004; Skop et al.,
2004). The relative contribution of the endocytic versus the
secretory pathway to cytokinesis may depend on the rate of membrane deposition
that the pathway can deliver, relative to the speed of cytokinesis. Meiosis I
cytokinesis in Drosophila spermatocytes is very demanding, requiring
500 µm2 expansion in 20 minutes. Endocytic recycling can make
large stores of membrane available rapidly without de novo synthesis
(Pelissier et al., 2003).
Indeed, in BSC1 cells, endocytic recycling, but not the addition of newly
synthesised membrane from the Golgi, is sufficient for the increase in plasma
membrane surface during cytokinesis
(Boucrot and Kirchhausen,
2007).
The analysis of membrane addition kinetics in wild-type and arf6
cells reveals two components of membrane addition: a slow ARF6-independent
process and, after central spindle formation, a 2.5-fold faster addition
process boosted by ARF6 (Fig.
4). The accelerated addition rate coincides with the positioning
of Rab4/ARF6 endosomes at the cleavage furrow early during cytokinesis.
Rab11/ARF6 recycling endosomes might be involved later for stabilization of
the ring canals. The slow component might correspond to secretory trafficking,
or other ARF6-independent recycling routes. Indeed, in addition to ARF6, Golgi
factors such as Cog5 and Syntaxin 5 have been implicated in the process of
cytokinesis in Drosophila testes
(Farkas et al., 2003;
Xu et al., 2002). It will be
interesting to see the rate effects of mutants of these factors in comparison
to arf6.
Why are some arf6 mutant cells impaired in the rate of plasma membrane addition causing early regression of the cleavage furrow, whereas other cells show a normal addition rates and only a late regression phenotype? We favor the possibility that the lack of ARF6 uncovers a natural variation in the activities of other components involved in plasma membrane insertion, or in the amount of endocytic membrane available for recycling. Late regressors might be those cells in which these other components or amounts of available membrane are above a certain threshold level that, if not reached, would lead to early regression. In the late regressors, although membrane addition proceeds at a normal rate, the membrane inserted independently of ARF6 might lack key components that are essential for the stability of the ring canal, and thereby for completion of cytokinesis. Such a defect, which might actually occur throughout cytokinesis, would only manifest itself later by leading to late furrow regression,
Membrane recycling from the central spindle
Is central spindle targeting of recycling endosomes functionally
significant? In cleaving Xenopus embryos, most membrane insertion
occurs next to the furrow (Bluemink and de
Laat, 1973). As in Drosophila spermatocytes, machinery
for fusion of intracellular vesicles with the plasma membrane, including the
exocyst complex and syntaxin, is localized to the cleavage furrow and central
spindle in many cell types (Fielding et
al., 2005; Gromley et al.,
2005; Jantsch-Plunger and
Glotzer, 1999; Low et al.,
2003; VerPlank and Li,
2005). Central spindle proteins may assemble the relevant
endocytic and/or secretory factors to facilitate efficient membrane addition.
The central spindle might therefore function as a sensor during cytokinesis,
implementing membrane trafficking at the right time and, perhaps, at the right
place.
Our data showing that ARF6 binds to Pav suggest a possible molecular link
between the central spindle and the trafficking machinery. If they are not at
the central spindle, both Rab4 and Rab11 recycling endosomes show low levels
of ARF6 colocalization during cytokinesis, whereas most of them contain ARF6
when at the spindle. Pav might ensure local enrichment of ARF6 in central
spindle endosomes. Indeed, mammalian ARFs bind MKLP1, suggesting Pav-mediated
ARF recruitment (Boman et al.,
1999). Yeast two-hybrid and GST pull-down experiments confirmed
this interaction in Drosophila, suggesting that this might be a
conserved mechanism in cytokinesis (Fig.
7).
Our data show that in the absence of ARF6, Rab4 recycling endosomes are
still targeted to the spindle. Similarly, Rab11 recycling endosomes also reach
the central spindle in late arf6 regressors. It therefore seems that
ARF6 and Rab11 are recruited independently, with ARF6 acting downstream of
Rab4/Rab11 endosome localization to mediate rapid membrane recycling. Rab11,
recruited late to the central spindle, may act in cytokinesis completion as
previously suggested (Fielding et al.,
2005; Wilson et al.,
2005). It has been proposed that ARF6 recruits Rab11 recycling
endosomes to the central spindle (Fielding
et al., 2005). The contrasting situation in Drosophila
spermatocytes may be due to the fact that the interaction between ARF6 with
human FIP3/4 is not conserved for the Drosophila FIP3/4 homologue
Nuclear fallout and ARF6 (Wilson et al.,
2005).
ARF6-dependent rapid recycling at the central spindle
How does ARF6 boost the recycling rate? One possibility is that ARF6
connects the recycling endosomes concentrated at the central spindle with
exocyst-defined fusion sites at the plasma membrane of the cleavage furrow.
The exocyst complex localizes to vesicular structures at the central spindle
and cleavage furrow, which would be adjacent to the cortical central spindle
ARF6 endosomes shown in this report
(Fielding et al., 2005;
Gromley et al., 2005).
ARF6-GTP interacts with the exocyst complex subunit Sec10
(Prigent et al., 2003). ARF6
interaction with the exocyst complex may therefore mediate targeted recycling
of membrane to discrete plasma membrane domains
(D'Souza-Schorey and Chavrier,
2006). ARF6 might alternatively influence recycling endosome or
plasma membrane phospholipid metabolism using the effector phospholipase D, a
mechanism frequently implicated in regulated recycling and secretion
(Brown et al., 1993;
Caumont et al., 1998;
Jovanovic et al., 2006;
Vitale et al., 2002). Our data
suggest that in the absence of ARF6, Rab4/Rab11 endosomes still contribute to
a basic rate of membrane recycling, but ARF6 recruitment contributes to more
efficient membrane insertion by endowing recycling vesicles with a label to
perform directed exocytosis.
Life without ARF6
ARF6 is essential for meiotic cytokinesis in the testes. Occasional
spermatids containing more than four nuclei in arf6 mutants are
consistent with cytokinesis failure during the mitosis prior to meiosis in the
spermatocytes (Fig. 1).
Additionally, karyotyping of third instar homozygous arf6 larval
brains revealed a low but significant frequency of tetraploidy:
4.2±2.2% (n=5 brains, 511 mitosis) in
arf6L51b/arf6L51b mutants
versus 0% (n=4 brains, 385 mitosis) in
arf6L51b/+ heterozygotes
and 0% (n=4 brains, 150 mitosis) in wild-type animals. This suggests
a cytokinesis failure during mitosis in this tissue. Furthermore, there is an
incompletely penetrant germ line ARF6 requirement during chorion morphogenesis
(see Fig. S2 in the supplementary material). However, most somatic mitosis and
other developmental processes occur normally in individuals completely lacking
ARF6.
Many other Drosophila cytokinesis mutants (e.g. fwd, gio,
klp3A, fws) preferentially affect spermatocyte cytokinesis, with little
or no effect on somatic cells (Brill et
al., 2000; Farkas et al.,
2003; Giansanti et al.,
2006; Williams et al.,
1995). However, it is surprising that many previously proposed
ARF6-dependent processes are not affected in arf6 maternal/zygotic
null mutants. For example, Loner/Schizo, which plays a role during myoblast
fusion and axon path finding in Drosophila has a specific GEF
activity on ARF6, but not ARF1 in vitro, and overexpression of a dominant
negative GDP-bound ARF6 mutant partially phenocopies loner/schizo
mutants (Chen et al., 2003;
Onel et al., 2004). The lack
of myoblast/neuronal phenotypes of the arf6 null mutant suggests that
the real target of Loner/Schizo is another GTPase or second redundantly acting
target.
In mammalian cultured cells, ARF6 mediates essential processes including
cell migration, cell-cell adhesion and phagocytosis
(D'Souza-Schorey and Chavrier,
2006). The mouse arf6 knockout shows a developmental
phenotype consistent with impaired cell migration during hepatic cord
formation (Suzuki et al.,
2006). Although we have not analyzed cell adhesion, migration or
phagocytosis in detail, arf6 null mutants survive to the adult stage
with no overt morphological defects, which requires all these processes.
Drosophila has no second arf6 gene
(Lee et al., 1994). The
closest homologue encoded in the genome, ARF1 (68% identical), is involved in
secretory, but not endocytic trafficking (reviewed in
D'Souza-Schorey and Chavrier,
2006). The mouse arf6 knockout
(Suzuki et al., 2006) will
tell us in the future whether these functions of ARF6 are vertebrate
specific.
Supplementary material
Supplementary material for this article is available at
http://dev.biologists.org/cgi/content/full/134/24/4437/DC1
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adams, R. R., Tavares, A. A., Salzberg, A., Bellen, H. J. and
Glover, D. M. (1998). pavarotti encodes a kinesin-like
protein required to organize the central spindle and contractile ring for
cytokinesis. Genes Dev.
12,1483
-1494.
Albertson, R., Riggs, B. and Sullivan, W. (2005). Membrane traffic: a driving force in cytokinesis. Trends Cell Biol. 15,92 -101.[CrossRef][Medline]
Amano, M., Ito, M., Kimura, K., Fukata, Y., Chihara, K., Nakano,
T., Matsuura, Y. and Kaibuchi, K. (1996). Phosphorylation and
activation of myosin by Rho-associated kinase (Rho-kinase). J.
Biol. Chem. 271,20246
-20249.
Ang, A. L., Taguchi, T., Francis, S., Folsch, H., Murrells, L.
J., Pypaert, M., Warren, G. and Mellman, I. (2004). Recycling
endosomes can serve as intermediates during transport from the Golgi to the
plasma membrane of MDCK cells. J. Cell Biol.
167,531
-543.
Bartel, P. L. (1993). Using the two-hybrid system to detect protein-protein interactions. In Cellular Interactions in Development: A Practical Approach (ed. D. A. Hartley), pp. 153-179. Oxford: Oxford University Press.
Bartel, P., Chien, C. T., Sternglanz, R. and Fields, S. (1993). Elimination of false positives that arise in using the two-hybrid system. Biotechniques 14,920 -924.[Medline]
Bluemink, J. G. and de Laat, S. W. (1973). New
membrane formation during cytokinesis in normal and cytochalasin B-treated
eggs of Xenopus laevis. I. Electron microscope observations. J.
Cell Biol. 59,89
-108.
Boman, A. L., Kuai, J., Zhu, X., Chen, J., Kuriyama, R. and Kahn, R. A. (1999). Arf proteins bind to mitotic kinesin-like protein 1 (MKLP1) in a GTP-dependent fashion. Cell Motil. Cytoskeleton 44,119 -132.[Medline]
Boucrot, E. and Kirchhausen, T. (2007).
Endosomal recycling controls plasma membrane area during mitosis.
Proc. Natl. Acad. Sci. USA
104,7939
-7944.
Brill, J. A., Hime, G. R., Scharer-Schuksz, M. and Fuller, M. T. (2000). A phospholipid kinase regulates actin organization and intercellular bridge formation during germline cytokinesis. Development 127,3855 -3864.[Abstract]
Brown, H. A., Gutowski, S., Moomaw, C. R., Slaughter, C. and Sternweis, P. C. (1993). ADP-ribosylation factor, a small GTP-dependent regulatory protein, stimulates phospholipase D activity. Cell 75,1137 -1144.[CrossRef][Medline]
Caumont, A. S., Galas, M. C., Vitale, N., Aunis, D. and Bader,
M. F. (1998). Regulated exocytosis in chromaffin cells.
Translocation of ARF6 stimulates a plasma membrane-associated phospholipase D.
J. Biol. Chem. 273,1373
-1379.
Chen, E. H., Pryce, B. A., Tzeng, J. A., Gonzalez, G. A. and Olson, E. N. (2003). Control of myoblast fusion by a guanine nucleotide exchange factor, loner, and its effector ARF6. Cell 114,751 -762.[CrossRef][Medline]
Chou, T. B. and Perrimon, N. (1992). Use of a yeast site-specific recombinase to produce female germline chimeras in Drosophila. Genetics 131,643 -653.[Abstract]
Clarkson, M. and Saint, R. (1999). A His2AvDGFP fusion gene complements a lethal His2AvD mutant allele and provides an in vivo marker for Drosophila chromosome behavior. DNA Cell Biol. 18,457 -462.[CrossRef][Medline]
Cooley, L., Verheyen, E. and Ayers, K. (1992). chickadee encodes a profilin required for intercellular cytoplasm transport during Drosophila oogenesis. Cell 69,173 -184.[CrossRef][Medline]
D'Souza-Schorey, C. and Chavrier, P. (2006). ARF proteins: roles in membrane traffic and beyond. Nat. Rev. Mol. Cell Biol. 7,347 -358.[CrossRef][Medline]
D'Souza-Schorey, C., Li, G., Colombo, M. I. and Stahl, P. D.
(1995). A regulatory role for ARF6 in receptor-mediated
endocytosis. Science
267,1175
-1178.
D'Souza-Schorey, C., van Donselaar, E., Hsu, V. W., Yang, C.,
Stahl, P. D. and Peters, P. J. (1998). ARF6 targets recycling
vesicles to the plasma membrane: insights from an ultrastructural
investigation. J. Cell Biol.
140,603
-616.
Echard, A., Hickson, G. R., Foley, E. and O'Farrell, P. H. (2004). Terminal cytokinesis events uncovered after an RNAi screen. Curr. Biol. 14,1685 -1693.[CrossRef][Medline]
Eggert, U. S., Kiger, A. A., Richter, C., Perlman, Z. E., Perrimon, N., Mitchison, T. J. and Field, C. M. (2004). Parallel chemical genetic and genome-wide RNAi screens identify cytokinesis inhibitors and targets. PLoS Biol. 2, e379.[CrossRef][Medline]
Emery, G., Hutterer, A., Berdnik, D., Mayer, B., Wirtz-Peitz, F., Gaitan, M. G. and Knoblich, J. A. (2005). Asymmetric Rab 11 endosomes regulate delta recycling and specify cell fate in the Drosophila nervous system. Cell 122,763 -773.[CrossRef][Medline]
Farkas, R. M., Giansanti, M. G., Gatti, M. and Fuller, M. T.
(2003). The Drosophila Cog5 homologue is required for
cytokinesis, cell elongation, and assembly of specialized Golgi architecture
during spermatogenesis. Mol. Biol. Cell
14,190
-200.
Fielding, A. B., Schonteich, E., Matheson, J., Wilson, G., Yu, X., Hickson, G. R., Srivastava, S., Baldwin, S. A., Prekeris, R. and Gould, G. W. (2005). Rab11-FIP3 and FIP4 interact with Arf6 and the exocyst to control membrane traffic in cytokinesis. EMBO J. 24,3389 -3399.[CrossRef][Medline]
Finger, F. P., Hughes, T. E. and Novick, P. (1998). Sec3p is a spatial landmark for polarized secretion in budding yeast. Cell 92,559 -571.[CrossRef][Medline]
Formstecher, E., Aresta, S., Collura, V., Hamburger, A., Meil,
A., Trehin, A., Reverdy, C., Betin, V., Maire, S., Brun, C. et al.
(2005). Protein interaction mapping: a Drosophila case study.
Genome Res. 15,376
-384.
Fromont-Racine, M., Rain, J. C. and Legrain, P. (1997). Toward a functional analysis of the yeast genome through exhaustive two-hybrid screens. Nat. Genet. 16,277 -282.[CrossRef][Medline]
Fuller, M. T. (1993). Spermatogenesis. In The Development of Drosophila melanogaster (ed. A. Martinez-Arias and M. Bate), pp. 71-147. Cold Spring Harbor: Cold Spring Harbour Laboratory Press.
Giansanti, M. G., Bonaccorsi, S., Williams, B., Williams, E. V.,
Santolamazza, C., Goldberg, M. L. and Gatti, M. (1998).
Cooperative interactions between the central spindle and the contractile ring
during Drosophila cytokinesis. Genes Dev.
12,396
-410.
Giansanti, M. G., Farkas, R. M., Bonaccorsi, S., Lindsley, D.
L., Wakimoto, B. T., Fuller, M. T. and Gatti, M. (2004).
Genetic dissection of meiotic cytokinesis in Drosophila males. Mol.
Biol. Cell 15,2509
-2522.
Giansanti, M. G., Bonaccorsi, S., Kurek, R., Farkas, R. M., Dimitri, P., Fuller, M. T. and Gatti, M. (2006). The class I PITP giotto is required for Drosophila cytokinesis. Curr. Biol. 16,195 -201.[CrossRef][Medline]
Glotzer, M. (2005). The molecular requirements
for cytokinesis. Science
307,1735
-1739.
Gromley, A., Yeaman, C., Rosa, J., Redick, S., Chen, C. T., Mirabelle, S., Guha, M., Sillibourne, J. and Doxsey, S. J. (2005). Centriolin anchoring of exocyst and SNARE complexes at the midbody is required for secretory-vesicle-mediated abscission. Cell 123,75 -87.[CrossRef][Medline]
Hime, G. R., Brill, J. A. and Fuller, M. T. (1996). Assembly of ring canals in the male germ line from structural components of the contractile ring. J. Cell Sci. 109,2779 -2788.[Abstract]
Jantsch-Plunger, V. and Glotzer, M. (1999). Depletion of syntaxins in the early Caenorhabditis elegans embryo reveals a role for membrane fusion events in cytokinesis. Curr. Biol. 9,738 -745.[CrossRef][Medline]
Jantsch-Plunger, V., Gonczy, P., Romano, A., Schnabel, H.,
Hamill, D., Schnabel, R., Hyman, A. A. and Glotzer, M.
(2000). CYK-4: A Rho family gtpase activating protein (GAP)
required for central spindle formation and cytokinesis. J. Cell
Biol. 149,1391
-1404.
Jovanovic, O. A., Brown, F. D. and Donaldson, J. G.
(2006). An effector domain mutant of Arf6 implicates
phospholipase D in endosomal membrane recycling. Mol. Biol.
Cell 17,327
-335.
Kiehart, D. P. and Feghali, R. (1986).
Cytoplasmic myosin from Drosophila melanogaster. J. Cell
Biol. 103,1517
-1525.
Knutton, S., Sumner, M. C. and Pasternak, C. A. (1975). Role of microvilli in surface changes of synchronized P815Y mastocytoma cells. J. Cell Biol. 66,568 -576.
