Download Identification and localization of a β‐COP‐like protein involved in the

Journal of Experimental Botany, Vol. 54, No. 390, pp. 2053±2063, September 2003
DOI: 10.1093/jxb/erg230
RESEARCH PAPER
Identi®cation and localization of a b-COP-like protein
involved in the morphodynamics of the plant Golgi
apparatus
Isabelle Couchy, Susanne Bolte, Marie-TheÂreÁse Crosnier, Spencer Brown and
BeÂatrice Satiat-Jeunemaitre*
Laboratoire de Dynamique de la Compartimentation Cellulaire, Institut des Sciences du VeÂgeÂtal,
CNRS UPR2355, 91198 Gif-sur-Yvette cedex, France
Received 18 March 2003; Accepted 29 May 2003
Abstract
Introduction
This paper examines the molecular machinery
involved in membrane exchange within the plant
endomembrane system. A study has been undertaken
on b-COP-like proteins in plant cells using M3A5, an
antibody raised against the conserved sequence of
mammalian b-COP proteins. In mammalian cells,
b-COP proteins are part of a complex named the coatomer, which probably recruits some speci®c areas of
the endomembrane system. Immuno¯uorescence
analyses by confocal laser scanning microscopy
showed that b-COP-like proteins marked predominantly the plant Golgi apparatus. Other proteins known
to be part of a potential machinery for COPI vesicle
formation (g-COP, b¢-COP and Arf1 proteins) were
immunolocalized on the same membraneous structures as b-COP. Moreover, b-COP and other COPI antibodies stained the cell plate in dividing cells. It is
further shown that, in maize root cells, and in contrast
to observations upon mammalian cells, the drug
Brefeldin A (BFA) does not induce the release of
b-COP and Arf1 proteins from the Golgi membrane
into the cytosol. These data clearly demonstrate that
the antibody M3A5 is a valuable marker for studies on
traf®cking events in plant cells. They also report for
the ®rst time the location of COP components in plant
tissue at the light level, especially on a model well
known for secretion, i.e. the maize root cells. They
also suggest that the membrane recruitment
machinery may function in a plant-speci®c way.
In all eukaryotic cells, the Golgi apparatus (GA) is a
central organelle in the secretory processes. In higher plant
cells, the GA is characterized by more than a hundred
Golgi stacks dispersed throughout the cytoplasm (SatiatJeunemaitre and Hawes, 1992). These stacks consist of
membrane-bounded cisternae, organized in a polarized
manner and surrounded by populations of vesicles
(Robinson et al., 1998; Dupree and Sherrier, 1998;
Hawes et al., 1999).
The recent visualization of Golgi stacks in vivo has
underlined its tremendous ability to exchange membranes
and proteins with other components of the endomembrane
system in mammalian cells rapidly (Cole et al., 1996;
Simpson et al., 2001; Presley et al., 2002) as in plant cells
(Boevink et al., 1998; NebenfuÈhr et al., 1999, 2000;
Saint-Jore et al., 2002). The intensity of this membrane
exchange contrasts with the apparent structural stability of
the plant GA. A fundamental question is to understand the
mechanisms by which the Golgi morphology, its spatial
distribution and function are acquired, preserved and
regulated concomitantly with membrane ¯ow. There is a
previous report on a family of Rho proteins which could be
involved in the spatial distribution of the plant GA
(Couchy et al., 1998). This paper focuses on proteins
that may be involved in the regulation of the Golgi
membrane exchanges.
The integrity of the GA implies maintenance both of
lipid composition in its bounding membranes and of the
protein complement for processing, packaging and targeting biosynthetic molecules. Therefore, in parallel with the
secretory membrane ¯ow which goes through the GA, cells
have to operate a complex membrane recycling machinery
Key words: b-COP, Brefeldin A, coatomer, Golgi apparatus,
plant cells.
* To whom correspondence should be addressed. Fax: +33 1 69 82 33 55. E-mail: [email protected]
Journal of Experimental Botany, Vol. 54, No. 390, ã Society for Experimental Biology 2003; all rights reserved
2054 Couchy et al.
to preserve and regulate the membrane economy of the
Golgi. Occurrence of a general retrograde/recycling
pathway throughout the plant GA was initially suggested
by studies on endocytosis in plant cells (Tanchak et al.,
1984; Hawes et al., 1996), where internalized cationized
ferritin was observed within Golgi cisternae. Such retrograde transport may use a speci®c population of transport
vesicles. Protein-coated vesicles, different from
clathrin-coated vesicles, have been described in the
vicinity of the GA (Hawes et al., 1996; Pimpl et al.,
2000). They share some structural features with the
mammalian COPI-coated vesicles. These latter vesicles
are thought to be able selectively to co-opt and transport
membrane proteins and soluble molecules in a retrograde
direction and, some believe, in an anterograde direction as
well through mammalian Golgi stacks (Nickel and
Wieland, 1997; Harter, 1999). They are described as
75 nm vesicles, covered with a 10 nm protein coat (COP =
Coat Protein), made of seven peptides (a-COP, b-COP,
b¢-COP, d,-COP, e-COP, g-COP, and z-COP) (Kreis et al.,
1995). These peptides are present in the cytosol as a
protein complex, termed the coatomer. The coatomer is
recruited on the Golgi membrane by a small GTP binding
protein, Arf1 (ADP-ribosylation factor) (Nickel and
Wieland, 1997).
The question is whether the `COPI-like vesicles'
observed in plant cells share some protein identity and
function with their mammalian homologues, and how they
are involved in the morphodynamics of the plant Golgi
apparatus. Genes coding for homologues of most of these
mammalian proteins have now been identi®ed in plant
cells (Memon et al., 1993; Regad et al., 1993; Szopa and
MuÈller-RoÈber, 1994; Kahn, 1995; Andreeva et al., 1998;
Takeuchi et al., 2002; Pasqualato et al., 2003). To study
COPI protein homologues in plant tissues further, antibodies raised against mammalian b-COP and b¢-COP
proteins have been used for biochemical and immunocytochemical studies. Immunolocation of the targeted
epitopes was observed by confocal laser scanning microscopy (CLSM). In order to identify putative protein
partners co-localizing with the epitopes recognized by
b-COP and b¢-COP antibodies, comparative immuno¯uorescence studies using antibodies raised against Arf and
g-COP homologues cloned from A. thaliana (Pimpl et al.,
2000) were performed. Potential association of these
proteins with the plant GA was studied by co-localization
experiments using a Golgi marker (JIM84: Horsley et al.,
1993). Brefeldin A (BFA), a potent disrupting agent for
plant Golgi morphology (Satiat-Jeunemaitre and Hawes,
1992; Satiat-Jeunemaitre et al., 1996) and, in mammalian
cells, an inhibitor of COPI coat recruitment onto Golgi
membranes was used to investigate the dynamic association and the cycling of these proteins with Golgi
membranes.
Materials and methods
Plant material
Nicotiana tabacum Bright Yellow-2 (BY-2) suspension-cultured
cells were grown in the dark at 26 °C with 100 rpm and subcultured
every 7 d at 5/80 ml, in a Murashige and Skoog medium
supplemented with sucrose and thiamine. For immuno¯uorescence
studies and protein extracts, cells were sampled 3 d after
subculturing (middle of their exponential phase of growth).
Maize caryopses (Zea mays, LG20.80, Limagrain, France) were
immersed in tap water for 3 h and then allowed to germinate in Petri
dishes on moist ®lter paper in the dark at 26 °C. Root apices were
excised from 3-d-old shoots.
Mung bean (Vigna radiata) seeds were immersed for 4 h in
distilled water and allowed to germinate on moistened ®lter paper in
plastic boxes at 26 °C in darkness; biochemical analysis and
immunocytolocalization studies on the epidermal strips were
performed from the growing part of 3-d-old hypocotyl (5 mm
below the hook).
Pollen tubes from Nicotiana sylvestris were grown for 3 h at room
temperature in a medium containing 10% sucrose, 0.01% boric acid,
and calcium nitrate 3 mM.
BFA treatment
Brefeldin A (BFA, Sigma) stock was 20 mg ml±1 in dimethylsulphoxide (DMSO). Roots were immersed in 100 mg ml±1 aqueous BFA
at 26 °C in darkness and ®xed after 1 h as previously described
(Satiat-Jeunemaitre and Hawes, 1992). To treat the suspension
cultures, BFA was added at a ®nal concentration of 100 mg ml±1
(Couchy et al., 1998). Parallel analyses were made upon cells treated
with DMSO at the same concentration.
Antibodies
The b-COP antibodies (mouse monoclonal M3A5) were initially
kindly provided by the late TE Kreis (Universite de GeneÁve, Suisse).
They are now commercially produced by Sigma.
In order to identify putative protein partners co-localizing with the
epitopes recognized by b-COP antibodies, a panel of antibodies were
used in co-localization studies. Anti g-COP antibody and anti-Arf1p,
raised against homologues cloned from A. thaliana, were kindly
provided by D Robinson (Heidelberg, Germany); anti-b¢-COP
antibody by R Pepperkok (Heidelberg, Germany). A rat monoclonal
antibody JIM84 (IgM) was used as neat supernatant as a Golgi
marker (Horsley et al., 1993).
Secondary antibodies were purchased either from Sigma (antirabbit, anti-mouse, or anti-rat IgGs conjugated with FITC, respectively used at 1:60, 1:40 and 1:40) or Interchim (anti-mouse, anti-rat,
or anti-rabbit IgG conjugated with Cy3, used at 1:800).
After washing, cells were mounted in Citi¯uor antifade-agent.
Images were collected either with a Sarastro 2000 (Molecular
Dynamics) or a Leica upright laser scanning confocal microscope
TCS SP2 (Leica Microsystems, Heidelberg, Germany). Different
¯uorochromes were detected sequentially frame-by-frame with the
acousto-optical tunable ®lter system (AOTF) using laser lines 488
and 543 nm. The images were coded green (¯uorescein-isothiocyanate) and red (cyanidine3.18) giving yellow co-localization in
merged images. Oil objectives used were 403 NA 1.25 and 633 NA
1.30, giving resolution of ~200 nm in the XY-plane and 400 nm
along the Z-axis (pinhole 1 Airy unit). Images were processed using
Adobe Photoshop (Adobe Systems).
Protein extraction
5 g of material were ground in a mortar with 6 ml of homogenization
buffer (0.5 M sorbitol, 10 mM KH2PO4, 2 mM salicylhydroxamic
acid (a chaotropic agent), 5% (w/v) polyvinylpyrrolidone 40 000
COPI components in plant cells
2055
MW (PVP40); 10 mM EGTA, pH 8.2, in the presence of protease
inhibitors: 1 mM PMSF, 10 mM leupeptin, 10 mM E-64, and 1 mM
pepstatin A (Boehringer). Cell lysates were spun 15 min at 15 000 g.
The supernatant was then spun 1 h at 150 000 g. This new
supernatant containing soluble proteins was kept for analysis. The
pellet containing microsomes was suspended in 0.5 M sorbitol and
10 mM KH2PO4, pH 8.2, and kept for analysis.
Immunoblot analysis
Protein samples were treated by a modi®ed SDS-lysis buffer
(Laemmli, 1970) containing 5 mM dithiothreitol (DTT), heated at
100 °C for 3 min, and separated on a 10% or 12% SDSpolyacrylamide gel at 25 mA in a Bio-Rad system.
Relative molecular mass markers were run in parallel with the
protein extracts (unstained markers, 14±97.4 kDa; stained markers
20±105 kDa, Bio-Rad). Gels were either stained with Coomassie
Brilliant Blue (Serva) or proteins on gels were transferred to
nitrocellulose sheets (0.45 mm; Schleicher and Schuell) by the
method of Towbin et al. (1979), and stained with Ponceau S (Sigma)
in 1% acetic acid to verify equal loading in each lane. After
destaining in water, the sheets were blocked with a solution
containing 5% milk powder, 10 mM TRIS-buffered saline, TBS
0.1 M, pH 7.4 and 0.05% (v/v) Tween 20 for 90 min prior to
incubation with the primary antibody overnight at 4 °C in the buffer
(M3A5, 1:50; Arf1p, 1:1000; g-COP, 1:1000; b¢-COP, 1:200 and
1:50; JIM84, 1:10). Primary antibodies were detected by two
methods. (a) Using alkaline phosphatase conjugated anti-rabbit, antirat IgG or anti-mouse antisera (Promega) at 1:5000 dilution. Colour
development was carried out by standard nitro-blue tetrazolium/
BCIP procedures. (b) Chemiluminescence, where primary antibodies were detected using horseradish peroxidase (HRP) conjugated anti-rat (Santa-Cruz), anti-rabbit (Amersham) or anti-mouse
(Santa-Cruz) antibodies at 1:5000. The chemiluminescence reaction
was detected using a Kodak ®lm.
Fig. 1. Detection of b-COP-like-proteins by immunoblotting of
protein extracts from BY-2 cells (A), mung bean hypocotyls (B), or
Nicotiana sylvestris pollen tubes (C). (A, B) 50 mg of proteins were
loaded; (C), 15 mg of proteins were loaded. Gel, 10% SDS-PAGE; S,
soluble proteins; M, membraneous proteins; T, total extracts.
Membrane probed with mouse anti-b-COP (M3A5) antibody (1:50)
and revealed by HRP-coupled secondary antibody. These results are
typical of repetitions. The positions of prestained commercial
molecular markers are shown to the left, and the deduced molecular
mass of an essential band is shown to the right. (A) »100 kDa band is
clearly detected in all plant extracts.
Results
625), recognized a polypeptide of ~100 kDa (Fig. 1). That
is the expected molecular mass for b-COP. The speci®c
activity of antigens was higher in the soluble than the
membrane fractions, being stronger in the soluble fraction
(Fig. 1A). This 100 kDa band was also identi®ed in protein
extracts from mung bean hypocotyls (Fig. 1B) and from
Nicotiana sylvestris pollen tubes (Fig. 1C), suggesting that
a b-COP homologue was present in various plant cell
types.
The idea was to look for other COPI components, in
particular, the proteins g-COP (98 kDa) and b¢-COP
(102 kDa). Western blots on BY-2 protein extracts were
therefore realized with the corresponding antibodies.
Blotting with the plant g-COP antibody (Pimpl et al.,
2000) revealed a unique band of ~100 kDa (Fig. 2A).
Similarly to what was previously described for b-COP, the
band was broader in the soluble fraction, suggesting a
higher protein concentration. These results are in accordance with data obtained for cauli¯ower (Pimpl et al.,
2000). The b¢-COP antibodies used here did not cross-react
on western blots with various plant extracts, probably due
to an insuf®cient loading of proteins. The dif®culty of
clearly detecting speci®c plant proteins on western blots
using other types of b¢-COP antibodies has been outlined
by Contreras et al. (2000).
Western blot analysis on various plant extracts, using
antibodies against COPI components
M3A5 antibody stains Golgi stacks, the cell plate and
the plasma membrane
In western blots of BY-2 cell protein extracts, the antibody
M3A5, raised against a peptide sequence in the C-terminal
moiety of the mammalian b-COP protein (24 AA, 620±
Immunolocalization experiments were performed either on
maize root squashes, mung bean epidermal strips or BY-2
suspension cells with substantially similar results. In
Immuno¯uorescence staining
Immunostaining procedures were performed on partially digested
tissues or cells as described in Couchy et al. (1998) or SatiatJeunemaitre and Hawes (2001). Brie¯y, biological material (i.e.
BY-2 cells, 3 mm root apices, mung bean epidermal strips) was ®xed
with paraformaldehyde 3% in PBS, pH 6.9. Cell walls were partially
digested by incubation in enzyme solution: 1% cellulase R10
(Onozuka), 1% pectinase (Sigma) in PBS, pH 6.9. Immuno¯uorescence was performed on Vectabond coated multiwell slides
(Vector Laboratories), after having layered cells or gently squashed
root apices on the slides. Cell membranes were permeabilized with
0.5% Triton X-100 for 20 min and washed with buffer. Non-speci®c
binding was blocked by 1% bovine serum albumin (BSA). Primary
antibodies were applied overnight at 4 °C. Slides were rinsed with a
stream of PBS supplemented with 1% ®sh gelatine (Sigma), and then
secondary antibodies conjugated with ¯uorochrome were applied for
1 h at room temperature and in darkness. After a 1 h wash
comprising ®ve baths of PBS supplemented with 1% ®sh gelatin,
slides were either mounted with Citi¯uor AF1 or retained for a
second immunostaining series.
2056 Couchy et al.
In control cells, similar 1 mm organelles were stained by
JIM84 (Figs 4A, 5A) and M3A5 antibodies (Fig. 5B).
Merged pictures showed a clear co-localization of the two
antibodies (Fig. 5C). The cell plate of dividing cells was
also stained by the two antibodies (Fig. 5D±F). Similar
results were obtained in BY-2 tobacco cells, except that the
plasma membrane was never stained by either of the two
antibodies. These results con®rmed that b-COP-like
proteins are associated with Golgi membranes.
BFA effects on the reorganization of M3A5
immunostaining pattern
Fig. 2. Western blot with anti-g-COP and anti-Arf1p on BY-2 cell
protein extracts. 50 mg of proteins were loaded. S, soluble proteins; M,
membraneous proteins; Gel, 10% SDS-PAGE. Membrane probed with
rabbit anti-g-COP and rabbit anti-Arf1p antibodies (1:1000) and
revealed by HRP-conjugated secondary antibody. These results are
typical of repetitions. The positions of prestained commercial
molecular markers are shown to the left, and deduced molecular
masses of essential bands are shown to the right.(A) Anti-g-COP
antibody revealed a unique band of ~100 kDa, both in M and S
fractions, with a higher density in the S fraction. (B) Arf1p antibody
revealed a major protein band at 20 kDa as expected, and a light band
at 40 kDa, which may represent a dimer.
interphase cells, the M3A5 antibody revealed numerous
~1 mm structures dispersed throughout the whole cell
(Fig. 3A, B). In dividing cells, the M3A5 antibody
decorated the new cell plate as well (Fig. 3C), and the
plasma membrane may appear slightly stained (not
shown). To check the consistency of the immunolabelling
throughout the root tissue, methacrylate sections of the
whole root were immunostained with M3A5. The punctate
¯uorescent pattern was found in each cell type (Fig. 3D).
Staining of the plasma membrane was also clearly
observed. The observed variability of plasma membrane
staining by M3A5 in maize root squash techniques was
observed for other antibodies known to stain the root
plasma membrane such as JIM84 (compare Fig. 4A and
Fig. 5). It may be related to the 0.5% Triton treatment
needed to permeabilize the membranes.
This M3A5 ¯uorescent pattern resembled in many
respects the typical immunostaining of plant Golgi stacks
by JIM84 (Satiat-Jeunemaitre and Hawes, 1992). Indeed,
JIM84 staining revealed a ¯uorescent punctuated pattern
of ~1 mm structures throughout the cytoplasm and a plasma
membrane staining (Figs 4A, 5A). This resemblance
suggested either an association or a co-localization of
b-COP with Golgi stacks.
To check this, double immunostaining of maize root
cells with JIM84 (a Golgi marker, Horsley et al., 1993) and
M3A5 antibody was performed. Observations were made
on both control (Fig. 5A±F) and BFA-treated cells
(Fig. 5G±I).
BFA induced typical Golgi morphological changes in
maize root cells. After 1 h of BFA treatment (100 mg ml±1),
the 1 mm Golgi stacks stained with JIM84 coalesced in the
cell to form larger ¯uorescent domains typical of BFA
compartments (Figs 4B, 5G). When BFA-treated cells
were immunostained with the b-COP antibody M3A5
(Fig. 5H), the ¯uorescent pattern was changed in the same
manner as that observed for Golgi membranes with JIM84
(Fig. 5G): membranes carrying the epitopes coalesced in
two or three ¯uorescent areas. Dual immunostaining
patterns showed that the proteins recognized by M3A5
were associated with Golgi membranes and they were not
released upon BFA treatment into the cytosol, as already
suggested by our biochemical data (Fig. 5I).
Immunostaining pattern of other COPI components
machinery in maize root cells
g-COP and b¢-COP location was investigated by immuno¯uorescence. In maize root cells, antibodies raised against
plant g-COP and mammalian b¢-COP proteins stained
punctate structures similar to M3A5 or JIM84 (Fig. 6A, B).
They also stained the cell plate (Fig. 6B; and data not
shown), but not the plasma membrane. These staining
patterns suggest that both of these proteins may be
associated with the Golgi apparatus. This hypothesis was
further reinforced by the two proteins' behaviour after
BFA treatment: epitopes recognized either by g-COP or
b¢-COP antibodies were gathered in ¯uorescent aggregates
recalling BFA compartments (Fig. 6C, D), as were
epitopes recognized by the b-COP antibody. However,
small differences in Atg-COP and b¢-COP staining were
observed when compared to M3A5 (anti-b-COP) staining.
Both stained weaker and their ¯uorescent patterns were
more irregular than those of JIM84 or M3A5.
The location of Arf1p (p=plant), another potential
protein partner for b-COP, was also investigated in this
study. An antibody against Arf1p cloned from Arabidopsis
thaliana (Pimpl et al., 2000) was tested on maize roots and
BY-2 cells. The speci®city of the antibody was checked on
western blots. It revealed a major protein band at 20 kDa as
expected, and a minor band at 40 kDa (Fig. 2B), which
may represent a dimer (Pasqualato et al., 2003).
COPI components in plant cells
2057
Fig. 3. Immuno¯uorescence studies with M3A5 antibody. In maize root cells (A, C, D) as in BY-2 cells (B), a group of 8 cells, a punctate pattern
throughout the cytoplasm was revealed by the anti-b-COP antibody, within numerous structures of ~1 mm. (C) In dividing cells, M3A5 antibody
decorated the cell plate. (D) This ¯uorescent pattern was not dependent upon cell type or position, as seen on this root section (methacrylate
embedding). Plasma membrane was slightly stained on methacrylate sections, and in some root cells. Bar units in mm.
Fig. 4. BFA effects on Golgi distribution in maize: immuno¯uorescence by JIM84. Golgi staining in maize root cells by JIM84. (A) The Golgi
apparatus is comprised of numerous units dispersed throughout the cytoplasm. Plasma membrane is usually stained. (B) After BFA treatment
(100 mg ml±1 for 1 h), typical BFA compartments formed from Golgi membranes; plasma membrane remained positive. Bar units in mm.
In maize root cells, immunostaining with Arf1p antibody gave a punctate pattern (Fig. 7A). Double immunostaining with JIM84 con®rmed that Arf1p was partly
associated with Golgi membranes (Fig. 7D±F). Moreover,
it also appeared to be associated with other membraneous
structures: a stained radiating network can be seen, with
2058 Couchy et al.
Fig. 5. Double immunolocalization of epitopes recognized by JIM84 (anti-Golgi) and M3A5 (anti-b-COP) in control (A±F) and BFA-treated (G±
I) maize root. When roots were probed with both JIM84 (A, D, G) and M3A5 (B, E, H), a clear co-localization of the antigens within the cell was
observed at the level of the Golgi units (C, F, I) both in interphase cells (A±C), in dividing cells (D±F) as well as in BFA-treated cells at the level
of the BFA compartments (G±I). Co-localization at the plasma membrane was also detected in some cases (F). Bar units in mm.
staining of the plasma membrane and a light staining of the
nuclear membrane (Fig. 7A, E); the cell plate was also
stained in dividing cells (Fig. 7B). BFA treatment induced
a redistribution of Arf1p into two or three ¯uorescent
compartments (Fig. 7C). Double immunostaining con®rmed that Arf1p was mostly associated with BFA
compartments where Golgi membranes concentrate
(Fig. 7G±I), even in dividing cells (Fig. 7J±L).
These results suggest that, in maize roots, epitopes
recognized by b¢-COP, g-COP and Arf1p antibodies are
indeed associated with Golgi membranes as is the b-COPlike protein. The four proteins behave in the same way:
redistribution concomitant with Golgi membranes into
BFA compartments and decoration of the cell plate.
Discussion
The aim of this work was to identify a speci®c tool to study
plant b-COP proteins, to examine their cellular location
and BFA reactivity, and to question the role of these
proteins in the morphodynamics of the plant Golgi
apparatus.
For mammalian cells, it is claimed that the coatomer is
involved in the recruitment of Golgi membrane areas,
forming COPI vesicles. COPI-coated structures have been
shown to be involved in the bi-directional membrane
exchange between the endoplasmic reticulum (ER) and
Golgi complex, and within the Golgi complex (Nickel and
Wieland, 1997; Orci et al., 1997; Harter, 1999). It is
unclear whether these models, derived from genetic,
biochemical and morphological dissection of mammalian
and yeast cells, also pertain to plant cells. In favour of this
later claim lie the facts that: (i) pro®les of budding COPlike vesicles have been described on Golgi membranes
(Hawes et al., 1996; Pimpl et al., 2000); (ii) homologues of
COPI vesicular transport machinery have been identi®ed
in various databases or cloned (Regad et al., 1993;
Andreeva et al., 1998; Movafeghi et al., 1999; Pimpl
COPI components in plant cells
2059
Fig. 6. Immuno¯uorescence study with plant g-COP and mammalian b¢-COP antibodies in control and BFA-treated maize root. (A, C) Staining
with anti-g-COP antibody. (A) In control cells, a punctate pattern was observed within the cytoplasm. (C) After BFA treatment (100 mg ml±1, 1 h),
¯uorescence was redistributed into two or three zones recalling Golgi-derived BFA compartments. (B, D) Staining with anti-b¢-COP antibody.
(B) Numerous ~1 mm structures were revealed in the cytoplasm, as also the cell plate in dividing cells. No staining of the plasma membrane was
observed. (D) After BFA treatment, ¯uorescence was redistributed into areas recalling Golgi-derived BFA compartments.
et al., 2000); and (iii) g-COP and Arf1p, two components
of COPI vesicle formation (Kreis et al., 1995), have been
localized on plant Golgi membranes (Pimpl et al., 2000;
Ritzenthaler et al., 2002; Takeuchi et al., 2002). However,
plant cell compartmentation, plant Golgi apparatus organization, functions, and reactivity to pharmacological
agents are different from those of other eukaryotic cells
(Satiat-Jeunemaitre et al., 1996; MeÂrigout et al., 2002)
and, therefore, the role of putative COP vesicles in plant
Golgi morphodynamics has still to be de®ned.
Plant b-COP is associated with Golgi membrane
dynamics
These biochemical and immuno¯uorescence studies suggest that the M3A5 antibody recognizes a b-COP-like
protein in plant cells. The hypothesis was further
reinforced by MALDI-TOF mass spectrometry analysis
of the proteins immunoprecipitated with M3A5, showing
signi®cant homology with rat b-COP (data not shown;
Couchy, 2001). Such `b-COP like' protein, having strong
similarities with the rat b-COP, has recently been cloned in
A. thaliana (GenBank accession number AL161259.2).
In addition, it was shown by immunolabelling that
putative b-COP, b¢-COP, g-COP, and Arf1 proteins are
localized on Golgi membranes of maize root cells. These
data complement the analysis of Contreras et al. (2000) on
rice extracts where all COPI subunits seem to coimmunoprecipitate with antibodies against b¢-COP. They
are also in agreement with the in vitro and in vivo data
recently obtained with other plant cells (Pimpl et al., 2000;
Ritzenthaler et al., 2002; Takeuchi et al., 2002).
Therefore, all the potential machinery for COPI vesicle
formation (i.e. coatomer complex and Arf1 proteins) are
co-detected on Golgi membranes, suggesting a role for
COPI vesicles in plant Golgi morphodynamics. Whether
the function for COPI vesicles essentially concerns the
2060 Couchy et al.
Fig. 7. Immuno¯uorescence study with anti-Arf1p antibody and Golgi marker JIM84 on maize root cells, in control and BFA-treated cells. In
interphase cells (A, cell on the right; E), the anti-Arf1p antibody stains 1 mm structures dispersed throughout the cytoplasm, but also labels the
nuclear membrane and the plasma membrane. Moreover, in dividing cells (B, telophase), staining revealed the cell plate. After BFA treatment, the
anti-Arf1p antibody was redistributed into ¯uorescent aggregates; plasma membrane remained positive (C, H, K). Double localization with the
Golgi marker JIM84 (D, G, J) and Arf1p (E, H, K) shows a clear co-localization between the recognized epitopes as seen on the merged images
(F, I, L), in control cells (F) as in BFA-treated cells (I, L). This co-localization concerns plasma membrane and the cell plate as well.
recycling of Golgi membrane components through a
retrograde movement (trans to cis cisternae recycling
pathway) or not has still to be studied. Furthermore, it may
be noted that the vesicular or tubular form of coatomercoated structures is, in fact, still the subject of vigorous
debate beyond the scope of this discussion, but suggesting
that the roles for the coatomer in Golgi membrane
transformation has still to be precisely established.
Is there an additional function for b-COP proteins in
plant cells?
Besides Golgi staining, two novel staining features with
COPI antibodies were described in this study: in dividing
cells, proteins recognized by Arf1, b-COP, b¢-COP, and
g-COP antibodies were localized on cell plates. Moreover,
in maize root cells, b-COP and Arf1p antibodies also
COPI components in plant cells
lightly stained the plasma membrane. Proteins recognized
by anti-b-COP have been found in a maize plasma
membrane fraction (P Moreau et al., personal communication). Arf1 is not found on the plasma membrane of
suspension cells (Ritzenthaler et al., 2002; Takeuchi et al.,
2002). However, Arf1p was found in several membrane
fractions of yeast cells, suggesting an association with
multiple membrane compartments as well as an enrichment at the plasma membrane (Yahara et al., 2001). Does
the presence of Arf on the plasma membrane or other
membraneous fractions bear any relation to the COPI
machinery?
The fact that both Arf1 and b-COP proteins are localized
on the cell surface and that all components of the COPI
machinery observed in this study stain the new cell plate
suggest additional functions for COPI components in plant
cells. It has been shown that Golgi stacks accumulate near
the growing edge of the cell plate (NebenfuÈhr et al., 2000).
(i) The localization of COPI components on the cell plate
may represent an ultimate step of the recycling movement
by COPI vesicles between the new plasma membrane
being formed (made of Golgi-derived membranes) and the
Golgi stacks (part of the trans-to-cis cisternae recycling
pathway). This continuous cycling of COPI vesicles would
ensure a retrieval of Golgi membrane domains from the
newly forming plasma membrane and would help the
plasma membrane to acquire its identity. (ii) The cell plate
is known to be rich in clathrin coated vesicles, involved in
endocytosis, i.e. recycling events between plasma membrane and endosomal compartments (Samuels et al., 1995).
The staining of the cell plate by COPI antibodies may be
linked to these endocytic events. In animal and yeast cells
there is a clear involvement of Arf1 in endocytosis
(Gaynor et al., 1998) and a function for b-COP in the
endocytotic pathway has been suggested as well (Aniento
et al., 1996).
As a whole, b-COP may be part of the molecular
machinery recycling Golgi proteins, but also determining
the identity of the new cell surface.
Molecular implications of the reactivity of Golgi and
coatomer components to Brefeldin A
The reactivity of the plant Golgi to BFA varies according
to the plant cell type studied and can differ from what is
usually described for mammalian cells (Satiat-Jeunemaitre
et al., 1996; Ritzenthaler et al., 2002; MeÂrigout et al.,
2002; Saint-Jore et al., 2002). Is this diversity in Golgi
reactivity to BFA mirrored by differential behaviour of
COPI type vesicles? What might be learnt from these BFA
data concerning the function of COPI components?
In mammalian and yeast cells, the prime molecular
target for BFA is a guanine nucleotide exchange factor
(GEF) responsible for the conversion of Arf-GDP to ArfGTP (Donaldson et al., 1992; Pasqualato et al., 2003).
Therefore, in the presence of BFA, Arf is not attached to
2061
the membranes, and the recruitment of the coatomer to the
membrane mediated by Arf anchoring is no longer
achieved. Consequently, b-COP proteins are released
into the cytosol within minutes of treatment and, later,
most of the Golgi apparatus is redistributed within the ER
compartment (see review in Scales et al., 2000; Presley
et al., 2002). So, in that case, alteration of the Golgi
morphology appears to be directly caused by the release of
Arf and COP components into the cytosol.
In maize root cells, the situation may be different: BFA
induced a reorganization of the Golgi in a tissuespeci®c manner, as Golgi markers formed aggregates
that were previously named `BFA compartments' (SatiatJeunemaitre and Hawes, 1993). These BFA compartments
have been observed in various root materials (Wee et al.,
1998; Baldwin et al., 2001; Jin et al., 2001). They are made
of Golgi-derived vesicles and accumulated secretory
products (Satiat-Jeunemaitre and Hawes, 1993; SatiatJeunemaitre et al., 1996). Arf1p, b-COP, g-COP, and
b¢-COP proteins stay associated (or continue to cycle) with
the Golgi-derived membranes trapped in these BFA
compartments. Therefore, in contrast to what has been
described for mammalian cells, they do not appear to be
released into the cytosol.
Regarding g-COP staining, the results on maize root
cells slightly differ from those decribed for the BY-2 cell
line, where g-COP was dissociated from the Golgi
membrane after 5 min of BFA treatment (Ritzenthaler
et al., 2002). This difference in g-COP reactivity to BFA is
probably due to tissue organization, as in Arabidopsis roots
it is also distinct from observations upon BY-2 cells
(Geldner et al., 2003). Further studies are needed to
understand g-COP behaviour and function in maize root
cells better.
The observations that substantial amounts of Arf1
remain associated with the Golgi stacks after BFA has
already been reported in biochemical studies with other
plant material (Fig. 9 in Ritzenthaler et al., 2002; Fig. 7 in
Pimpl et al., 2000), and in immunocytological studies in
transgenic BY-2 cell lines (Fig. 10 in Ritzenthaler et al.,
2002). Surprisingly, this fact has sometimes been neglected when discussing BFA action. For instance it is claimed
that `one of the ®rst detectable effects of BFA treatment of
BY-2 cells is the nearly complete loss of COPI coat
proteins (as judged by Atg-COP and AtArf1) from Golgi
stacks, an observation that is identical to that from BFAtreated mammalian cells' (Ritzenthaler et al., 2002). This
claim implies that `BFA causes the release of AtArf1 and
COPI from the GA, with a subsequent alteration of the GA
morphology' (see quotation in Takeuchi et al., 2002). By
contrast, for maize roots, data suggest that alteration of GA
morphology is not subsequent to AtArf1 release. Based on
b-COP and Arf1 immunolocalization after BFA treatment,
it is deduced that, in maize root cells, the Golgi apparatus
was reorganized even when a signi®cant amount of Arf1p
2062 Couchy et al.
and b-COP remained attached to the membranes. This
situation contrasts with what happens in animal cells,
where Arf release occurs in 15±30 s after BFA,
concomitantly or before the Golgi reorganization
(Presley et al., 2002).
The failure of BFA to release b-COP and Arf proteins
from the Golgi does not necessarily mean that the drug's
target or its mechanism of action differ from that in
animals. However, it suggests that additional molecular
targets may be involved in the disruption of plant Golgi
dynamics in response to BFA. BFA may well have
multiple molecular actions in plants, as it also inhibits the
synthesis of speci®c lipids of the plant Golgi (MeÂrigout
et al., 2002). An alteration of membrane lipids by BFA
may alter the anchoring of Arf1p, and/or the steady state of
coatomer attachment/recycling to the membrane.
Acknowledgements
We are very grateful to Dr Reiner Pepperkok (Heidelberg) for
providing antibodies against b'-COP, and to Professor David
Robinson (Heidelberg) for providing antisera against g-COP and
Arf1p. The IFR87 (FR-W2251) `La Plante et son Environnement'
provided the confocal microscopy facility. We are particularly
appreciative of the expert technical assistance provided by Dr ValeÂrie
Labas for MALDI-TOF analysis (Laboratoire de Neurobiologie et
diversite cellulaire, Ecole SupeÂrieure de Physique et de Chimie
Industrielle de Paris (ESPCI)) and by Mrs Nathalie Mansion for
photography assistance.
References
Andreeva AV, Kutuzov MA, Evans DE, Hawes CR. 1998.
Proteins involved in membrane transport between the ER and the
Golgi apparatus: 21 putative homologues revealed by dbEST
searching. Cell Biology 22, 145±160.
Aniento F, Gu F, Parton RG, Gruenberg J. 1996. An endosomal
b-COP is involved in the pH-dependent formation of transport
vesicles destined for late endosomes. Journal of Cell Biology 133,
29±41.
Baldwin TC, Handford MG, Yuseff MI, Orellana A, Dupree P.
2001. Identi®cation and characterization of GONST1, a Golgi
localized GDP-mannose transporter in Arabidopsis. The Plant
Cell 13, 2283±2295.
Boevink P, Oparka K, Santa Cruz S, Martin B, Betteridge A,
Hawes C. 1998. Stacks on tracks: the plant Golgi apparatus
traf®cs on an actin/ER network. The Plant Journal 15, 441±447.
Cole NB, Smith CL, Sciaky N, Terasaki M, Edidin M,
Lippincott-Schwartz J. 1996. Diffusional mobility of Golgi
proteins in membranes of living cells. Nature 273, 797±801.
Contreras I, Ortiz-Zapater E, Castilho LM, Aniento F. 2000.
Characterization of COPI coat proteins in plant cells. Biochemical
and Biophysical Research Communications 273, 176±182.
Couchy I. 2001. Recherche de proteÂines associeÂes aÁ la
morphodynamique de l'appareil de Golgi dans les cellules
veÂgeÂtales. Thesis, University Paris XI, Orsay.
Couchy I, Minic Z, Laporte J, Brown S, Satiat-Jeunemaitre B.
1998. Immunodetection of Rho-like plant proteins with Rac1 and
Cdc42Hs antibodies. Journal of Experimental Botany 49, 1647±
1659.
Donaldson JG, Finazzi D, Klausner RD. 1992. Brefeldin A
inhibits Golgi membrane-catalysed exchange of guanine
nucleotide onto ARF protein. Nature 360, 350±352.
Dupree P, Sherrier DJ. 1998. The plant Golgi apparatus.
Biochimica et Biophysica Acta 1404, 259±270.
Gaynor EC, Chen CY, Emr SD, Graham TR. 1998. ARF is
required for maintenance of yeast Golgi and endosome structure
and function. Molecular Biology of the Cell 9, 653±670.
Geldner N, Anders N, Wolters H, Keicher J, Kornberger W,
Muller P, Delbarre A, Ueda T, Nakano I, JuÈrgens G. 2003.
The Arabidopsis GNOM ARF-GEF mediates endosomal
recycling, auxin transport, and auxin-dependent plant growth.
Cell 112, 219±230.
Harter C. 1999. COPI proteins: a model for the role in vesicle
budding. Protoplasma 207, 125±132.
Hawes CR, Brandizzi F, Andreeva AV. 1999. Endomembranes
and vesicles traf®cking. Current Opinion in Plant Biology 2, 454±
461.
Hawes CR, Faye L, Satiat-Jeunemaitre B. 1996. The Golgi
apparatus and pathways of vesicule traf®cking. In: Smallwood M,
Knox JP, Bowles DJ, eds. Membranes: specialized functions in
plants. BIOS publishers, 337±365.
Horsley D, Coleman J, Evans D, Crooks K, Peart J, SatiatJeunemaõÃtre B, Hawes C. 1993. A monoclonal antibody, JIM84,
recognizes the Golgi apparatus and plasma membrane in plant
cells. Journal of Experimental Botany 44, 223±229.
Jin JB, Kim YA, Kim SJ, Lee SH, Kim DH, Cheong G-W,
Hwang I. 2001. A new dynamin-like protein, ADL6, is involved
in traf®cking from the trans-Golgi network to the central vacuole
in Arabidopsis. The Plant Cell 13, 1511±1525.
Kahn RA. 1995. Arf1. In: Zerial M, Huber LA, eds. Guidebook of
small GTPases. Oxford: Oxford University Press.
Kreis TE, Lowe M, Pepperkok R. 1995. COPs regulating
membrane traf®c. Annual Review of Cell Devevlopment Biology
11, 677±706.
Laemmli UK. 1970. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680±685.
Memon A, Clark GB, Thompson Jr GA. 1993. Identi®cation of
an ARF type low molecular mass GTP-binding protein in pea
(Pisum sativum). Biochemical and Biological Research
Communications 193, 809±813.
MeÂrigout P, KeÂpeÁs F, Perret A-M, Satiat-Jeunemaitre B,
Moreau P. 2002. Effects of brefeldin A and nordihydroguaiaretic acid on endomembrane dynamics and lipid synthesis
in plant cells. FEBS Letters 518, 88±92.
Movafeghi A, Happel N, Pimpl P, Tai G-H, Robinson DG. 1999.
AtSec21p and AtSec23p homologues: probable coat proteins of
plant COP vesicles. Plant Physiology 119, 1437±1445.
NebenfuÈhr A, Frohlick J, Staehelin LA. 2000. Redistribution of
Golgi stacks and other organelles during mitosis and cytokinesis
in plant cells. Plant Physiology 124, 135±152.
NebenfuÈhr A, Gallagher LA, Dunahay TG, Frohlick JA,
Mazurkiewicz AM, Meehl JB, Staehelin LA. 1999. Stop-andgo movements of plant Golgi stacks are mediated by the actomyosin system. Plant Physiology 121, 1127±1141.
Nickel W, Wieland FT. 1997. Biogenesis of COPI-coated transport
vesicles. FEBS Letters 413, 395±400.
Orci L, Stamnes M, Ravazzola M, Amherdt M, Perrelet A,
SoÈllner TH, Rothman JE. 1997. Bidirectional transport by
distinct populations of COPI-coated vesicles. Cell 90, 335±349.
Pasqualato S, Renault L, Cher®ls J. 2003. Arf, Ar1, Arp and Sar
proteins: a family of GTP-binding proteins with a structural
device for `front-back' communication. EMBO Reports 3, (in
press).
Pimpl P, Movafeghi A, Coughlan S, Denecke J, Hillmer S,
Robinson DG. 2000. In situ localization and in vitro induction of
plant COP-I-coated vesicles. The Plant Cell 12, 2219±2236.
COPI components in plant cells
Presley JF, Ward TH, Pfeifer AC, Siggia ED, Phair RD,
Lippincot-Schwartz J. 2002. In vivo dissection of COPI and
Arf1 dynamics and roÃle in Golgi membrane transport. Nature
417, 187±193.
Regad F, Bardet C, Tremousaygue D, Moisan D, Lescure B,
Axelos M. 1993. cDNA cloning and expression of an Arabidopsis
GTP-binding protein of the ARF family. FEBS Letters 316, 133±
136.
Ritzenthaler C, NebenfuÈr A, Movafeghi A, Stussi-Garaud C,
Benhia L, Pimpl P, Staehelin LA, Robinson DG. 2002.
Reevaluation of the effects of Brefeldin A on plant cells using
tobacco bright yellow 2 cells expressing Golgi-targeted Green
Fluorescent Protein and COPI antisera. The Plant Cell 14, 237±
261.
Robinson DG, Hinz G, Holstein SEH. 1998. The molecular
characterization of transport vesicles. Plant Molecular Biology
38, 49±76.
Saint-Jore CM, Evins J, Batoko H, Brandizzi F, Moore I, Hawes
C. 2002. Redistribution of membrane proteins between the Golgi
apparatus and endoplasmic reticulum in plants is reversible and
not dependent on cytoskeletal networks. The Plant Journal 29,
661±678.
Samuels AL, Giddings Jr TH, Staehelin LA. 1995. Cytokinesis in
tobacco BY-2 and root tip cells: a new model of cell plate
formation in higher plants. The Journal of Cell Biology 130,
1345±1357.
Satiat-Jeunemaitre B, Cole L, Bourett T, Howard R, Hawes CR.
1996. Brefeldin A effects in plant and fungal cells: something
new about vesicle traf®cking? Journal of Microscopy 181, 162±
177.
Satiat-Jeunemaitre B, Hawes CR. 1992. Redistribution of a Golgi
glycoprotein in plant cells treated with Brefeldin A. Journal of
Cell Science 103, 1153±1166.
2063
Satiat-Jeunemaitre B, Hawes CR. 1993. The distribution of
secretory products in plant cells is affected by Brefeldin A. Cell
Biology International 17, 183±193.
Satiat-Jeunemaitre B, Hawes C. 2001. Immunocytochemistry for
light microscopy. In: Hawes C, Satiat-Jeunemaitre B, eds. Plant
cell biology. A practical approach. Oxford: Oxford University
Press, 207±235.
Scales SJ, Gomez M, Kreis TE. 2000. Coat proteins regulating
membrane traf®c. International Review of Cytology 195, 67±144.
Simpson JC, Neubrand VE, Wiemann S, Pepperkok R. 2001.
Illuminating the human genome. Histochemistry and Cell Biology
115, 23±29.
Szopa J, MuÈller-RoÈber B. 1994. Cloning and expression analysis
of ADP-ribosylation factor from Solanum tuberosum L. Plant
Cell Reports 14, 180±183.
Takeuchi M, Ueda T, Yahara N, Nakano A. 2002. Arf1 GTPase
plays roles in the protein traf®c between the endoplasmic
reticulum and the Golgi apparatus in tobacco and Arabidopsis
cultured cells. The Plant Journal 31, 499±515.
Tanchak MA, Grif®ng LR, Mersey BG, Fowke LC. 1984.
Endocytosis of cationized ferritin by coated vesicles of soybean
protoplasts. Planta 162, 481±486.
Towbin H, Staehelin T, Gordon J. 1979. Electrophoresis transfer
of proteins from polyacrylamide gels to nitrocellulose sheets:
procedure and some applications. Proceedings of the National
Academy of Sciences, USA 76, 4350±4354.
Wee EG-T, Sherrier DJ, Prime TA, Dupree P. 1998. Targeting of
active sialyltransferase to the plant Golgi apparatus. The Plant
Cell 10, 1759±1768.
Yahara N, Ueda T, Sato K, Nakano A. 2001. Multiple roles of
Arf1 GTPase in the yeast exocytic and endocytic pathways.
Molecular Biology of the Cell 12, 221±238.