Cell, Vol. 112, 219–230, January 24, 2003, Copyright 2003 by Cell Press
The Arabidopsis GNOM ARF-GEF
Mediates Endosomal Recycling, Auxin Transport,
and Auxin-Dependent Plant Growth
(Haj et al., 2002). Some receptors even need to be endocytosed
in order to signal through downstream modules
that can only be recruited on endomembranes (Defea
et al., 2000). Also, interference with endocytosis in Drosophila
wing disks alters distribution of morphogenic
Niko Geldner,1
Nadine Anders,1
Hanno Wolters,1
Jutta Keicher,1
Wolfgang Kornberger,1
Philippe Muller,2
Alain Delbarre,2
Takashi Ueda,3
Akihiko Nakano,3
and Gerd Ju¨ rgens1,
*
signals, such as dpp (which is thought to be taken up1
ZMBP, Entwicklungsgenetik
and to be transcytosed during migration through theUniversita¨ t Tu¨ bingen
tissue) (Teleman et al., 2001). In contrast to these exam-Auf der Morgenstelle 3
ples from the animal literature, virtually nothing is knownD-72076 Tu¨ bingen
in higher plants about links between vesicle transportGermany
and the establishment of polarity, growth factor distribu-2
Institut des Sciences Vegetales
tion, or signal transduction pathways.CNRS, Batiment 23
A focus of interest in plant research has been to eluci-Avenue de la Terrasse
date the transport mechanism for the plant growth regu-91198 Gif-sur-Yvette
lator auxin. The active, polar transport of auxin throughFrance
the plant is thought to arise from the coordinated polar3
Molecular Membrane Biology Laboratory
distribution of carriers in the plasma membrane and isRIKEN, 2-1 Hirosawa, Wako
held responsible for a plethora of developmental pro-Saitama 351-0198
cesses (Friml and Palme, 2002). Candidates for auxinJapan
carriers were identified in the past few years and have
indeed been shown to localize in a strictly polar fashion
(Ga¨ lweiler et al., 1998; Mu¨ ller et al., 1998). One of theseSummary
candidates, the multispan membrane protein PIN1, continuously
recycles through endomembrane compart-Exchange factors for ARF GTPases (ARF-GEFs) reguments
(Geldner et al., 2001), suggesting that its polarlate vesicle trafficking in a variety of organisms. The
localization is highly dependent on directed vesicle traf-Arabidopsis protein GNOM is a brefeldin A (BFA) sensificking.tive
ARF-GEF that is required for the proper polar loThe
GNOM(GN) gene (also called EMB30) of Arabi-calization of PIN1, a candidate transporter of the plant
dopsis thaliana has been defined through loss-of-func-hormone auxin. Mutations in GNOM lead to develoption
mutants displaying striking embryonic phenotypesmental defects that resemble those caused by interfer(Mayer
et al., 1993). Mutant embryos show a loss ofing with auxin transport. Both PIN1 localization and
cell-to-cell alignment along the embryonic axis, lack theauxin transport are also sensitive to BFA. In this paper,
embryonic root, and display variable fusion or deletionwe show that GNOM localizes to endosomes and is
of cotyledons and hypocotyl. In severe cases, a com-required for their structural integrity. We engineered
plete loss of the apical-basal axis was observed. Thesea BFA-resistant version of GNOM. In plants harboring
genetic defects have been mimicked by treating in vitrothis fully functional GNOM variant, PIN1 localization
cultured Brassica juncea embryos with high doses of
and auxin transport are no longer sensitive to BFA,
auxin or with inhibitors of its polar transport (Hadfi et
while trafficking of other proteins is still affected by
al., 1998), suggesting a link between GNOM function
the drug. Our results demonstrate that GNOM is reand
auxin transport or perception. GNOM encodes a
quired for the recycling of auxin transport components
GDP/GTP exchange factor for small G proteins of the
and suggest that ARF-GEFs regulate specific endosoARF
class (ARF-GEF) (Busch et al., 1996; Shevell et al.,
mal trafficking pathways.
1994; Steinmann et al., 1999). ARF proteins play a role
in the recruitment of vesicle coats necessary for vesicle
Introduction
budding and cargo selection (Donaldson and Jackson,
2000). Thus, GNOM acts as a regulator of intracellular
Intracellular vesicle trafficking maintains the compart- vesicle trafficking. However, it was not at all clear how
mentalized organization of the eukaryotic cell and me- this cellular activity relates to auxin and to the embryonic
diates exchange of information with its environment. axis formation defects of gnom mutants.
Polarization of cells, the correct transduction of extra- By several criteria, gnom embryos lack any apparent
cellular signals, or the establishment of morphogen phenotype of a secretory mutant. gnom seedlings do
gradients during development all depend on highly regu- not become necrotic after germination, and gnom cell
lated intracellular vesicle trafficking events. Establish- suspension cultures are easily established and mainment
and maintenance of epithelial cell polarity is de- tained. Moreover, gnom mutants are neither affected in
pendent on directed vesicle transport to distinct plasma pollen tube growth nor in cytokinesis, two processes
membrane subdomains (Mostov et al., 2000). Receptor that require very active, regulated vesicle trafficking
levels at the surface are controlled by ligand-induced (Bednarek and Falbel, 2002; Parton et al., 2001) and
endocytosis and subsequent degradation or recycling which are often defective in knockout mutants of central
components of the vesicle transport machinery (Lukowitz
et al., 1996; Sanderfoot et al., 2001). Finally, a number*Correspondence: gerd.juergens@zmbp.uni-tuebingen.de
Cell
220
of subcellular markers were investigated and found to in the response to BFA are to be expected between
localize essentially normally in gnom (Shevell et al., the plant and mammalian endomembrane systems. The
2000; Steinmann et al., 1999). The only abnormality ob- internalization of PIN1 into perinuclear compartments
served was the localization of the putative auxin efflux upon BFA treatment is an example for such a difference
carrier PIN1 during embryogenesis. Its strictly coordi- because rapid and strong internalization of plasma
nated polar localization in the adult plant arises gradu- membrane markers in response to BFA has not been
ally during embryogenesis from an initially nonpolar dis- observed in mammals. BFA-induced intracellular accutribution.
This process is disrupted in gnom embryos, mulation of PIN1 can be explained by continued endocywhich
show a largely randomized polarity of cells with tosis in the presence of blocked resecretion (Figure 1B).
respect to each other (Steinmann et al., 1999). This led However, since BFA is predicted to inhibit at least 5
to the hypothesis that GNOM might be involved in traf- ARF-GEFs simultaneously, it cannot be deduced which
ficking of auxin efflux carriers. In this view, their disorga- and how many ARF-GEFs might be involved in this recynization
or improper functioning in gnom mutants would cling process.
abolish polar auxin transport, which could account for To examine whether GNOM is the ARF-GEF responsithe
observed phenotypes. ble for PIN1 recycling, we analyzed PIN1 localization in
PIN1 is rapidly internalized in response to brefeldin A transgenic plants expressing an engineered BFA-resis(BFA;
Geldner et al., 2001), which has been widely used tant GNOM variant. Although other ARF-GEFs remained
as a reversible inhibitor of vesicle trafficking in yeast, BFA sensitive, BFA-resistant GNOM was sufficient for
mammalian, and plant cells. A subclass of large ARF- rendering PIN1 recycling to the plasma membrane inGEFs
is the primary molecular target of BFA. They are sensitive to BFA and, in addition, led to BFA-resistance
defined by a number of critical residues in their central of auxin transport and auxin-mediated growth recatalytic
Sec7 domain (Peyroche et al., 1999; Sata et al., sponses. Double localization experiments of tagged
1999). GNOM carries these hallmarks of BFA sensitivity, GNOM variants with a number of markers both in the
and its activity has indeed been shown to be inhibited presence and absence of BFA revealed that GNOM loby
the drug (Steinmann et al., 1999). BFA traps sensitive calizes to endosomes and gnom loss-of-function muARF-GEFs
in an abortive complex, preventing ARF ef- tants display altered endosome morphology.
fector activation necessary for vesicle budding and
cargo selection (Peyroche et al., 1999; Robineau et al., Results
2000). The structural changes of the endomembrane
system in a given cell upon BFA treatment are thus
An Engineered BFA-Resistant GNOM ARF-GEF
contingent on the differential contributions of BFA-senRescues
gnom Mutant Plants
sitive versus BFA-resistant ARF-GEFs to the overall netBrefeldin
A (BFA) inhibits activation of ARF proteins by
work of membrane traffic.
binding to an ARF-GDP/ARF-GEF complex (Peyroche
In mammals, the sites of BFA action often do not
et al., 1999; Robineau et al., 2000). Naturally occurring
correlate with the subcellular localization of BFA-sensiBFA-resistant
or BFA-sensitive ARF-GEF variants differ
tive ARF-GEFs. For example, the only two clearly BFAin
a number of amino acid residues consistently consensitive
ARF-GEFs involved in Golgi trafficking do not
served in the exchange domain. Some of these residues
localize to the cis-Golgi, which is strongly sensitive to the
were shown to be critical for BFA resistance of yeast and
drug (Zhao et al., 2002). The same is true for endosomal
mammalian ARF-GEFs (Figure 1A). Exchanging those
trafficking. Endosomes have been known for a long time
residues renders BFA-resistant ARF-GEFs sensitive and
to be structurally affected by BFA, and transcytosis was
vice versa, both in biochemical assays and in vivo (Peyshown
to be inhibited by this drug (Hunziker et al., 1991;
roche et al., 1999; Sata et al., 1999). We reasoned thatLippincott-Schwartz et al., 1991; Wood et al., 1991).
if the BFA-sensitive GNOM ARF-GEF of Arabidopsis wasHowever, the only ARF-GEFs known to localize to endoalso
rendered BFA-resistant by introducing homologoussomes are animal specific BFA-resistant exchange
amino acid exchanges into its catalytic domain, wefactors such as EFA6, ARNO, or ARF-GEP100 (Franco et
could determine which BFA-sensitive trafficking path-al., 1999; Frank et al., 1998; Someya et al., 2001).
ways are specifically regulated by GNOM.These ARF-GEFs act on animal-specific ARF6, which is
A 7.5 kb genomic fragment of the GNOM gene con-involved in endocytosis and endosomal trafficking
taining 2.1 kb upstream of the ATG and about 0.6 kb(D’Souza-Schorey et al., 1998). ARFs often localize to
downstream of the stop codon complemented the gnoma wide range of membrane compartments, regulating
mutant phenotype (data not shown). This fragment wasdistinct trafficking processes, and ARF-GEFs are thought
used to insert coding sequences for a 3ϫ myc tag orto confer specificity to ARF action (Donaldson and JackGFP
at the 3Ј end of the open reading frame (Figureson, 2000). Although three different exchange factors
1C). A BFA-resistant variant was derived from the myc-are known for ARF6, their relative contributions to spetagged
construct by introducing an M to L mutation atcific ARF6-regulated transport events remain to be demamino
acid position 696 in the Sec7 domain (Figure 1C).onstrated.
We chose this mutation amongst a number of possibleIn comparison to mammals, Arabidopsis has a higher
ones because it was shown to confer good, albeit notnumber of predominantly BFA-sensitive, large ARFcomplete,
BFA resistance to yeast Gea1p without affect-GEFs. By sequence analysis, only three out of the eight
ing its ability to complement the gea1 knockout (Pey-Arabidopsis ARF-GEFs can be predicted to be BFA reroche
et al., 1999). Furthermore, the closest homolog ofsistant (Figure 1A), and except for ARF1, no clear orGNOM,
GNOM-LIKE 1 (GNL1), carries L instead of Mthologs of mammalian ARFs, including ARF6, can be
identified (Ju¨ rgens and Geldner, 2002). Thus, differences at this position (Figure 1A), suggesting that an M to L
GNOM-Dependent Endosomal Recycling
221
Figure 1. Engineering of a BFA-Resistant GNOM Variant
(A) Sequence alignment of the region determining BFA resistance of ARF-GEFs. All large ARF-GEFs from human (H.s.), yeast (S.c.), and
Arabidopsis (A.t.) are shown. At the top, ARNO, as an example of a BFA-resistant, small, mammalian ARF-GEF. Residues known to be involved
in BFA sensitivity are boxed. The ones determining resistance are written in bold. Asterisks indicate that BFA sensitivity/resistance was
determined experimentally, and unmarked ones are predictions inferred from this data. The residue boxed in black indicates the amino acid
exchange chosen to be introduced into GNOM (bottom line).
(B) Schematic model to explain internalization of PM markers upon BFA treatment. BFA is thought to block a sensitive ARF-GEF responsible
for recycling, while ongoing endocytosis might be mediated by a resistant ARF-GEF or, alternatively, by ARF-independent endocytosis.
(C) Overview of the BFA-resistant GNM696L
-myc construct used. Positions indicated are relative to the translational start. Black boxes indicate
translated regions, light gray boxes UTRs, lines indicate introns, and white boxes intergenic regions (promoter). The box marked “Sec7 domain”
indicates the region of the central catalytic domain. A 3ϫ myc-tag was translationally fused to the 3Ј end of the ORF of the complementing
genomic XbaI fragment. The resulting construct was then mutated.
mutation will not interfere with the exchange activity comitant strong increase in apparent membrane signal
that coalesced into large perinuclear aggregates (Figureof plant ARF-GEFs. The three constructs GNM696L
-myc,
GNwt
-myc, and GN-GFP were transformed into segregat- 2E). These observations were consistent with earlier differential
centrifugation experiments of cell extracts,ing populations of gnom heterozygous Arabidopsis
plants. For each construct, gnom mutant plants from which gave a predominantly cytosolic and a minor membrane-bound
form of GNOM, with the latter increasingseveral independent lines were shown to be rescued.
Thus, GNM696L
-myc, GNwt
-myc, and GN-GFP fusion pro- upon BFA treatment (Steinmann et al., 1999). We sometimes
also observed a weak plasma membrane labelteins appeared to be fully functional variants of the
GNOM protein. appearing upon BFA treatment, which was not observed
in untreated tissue (Figures 2D and 2E). Neither the increase
in membrane label nor aggregation into patchesPIN1 Recycling Is BFA Resistant in GNM696L
-myc
was seen for the GNM696L
-myc protein, illustrating its BFATransgenic Seedlings
resistance (Figure 2F).Root tips of GNOMwt
-myc seedlings displayed polar loIn
untreated roots of GNwt
-myc seedlings, no colocali-calization of PIN1, which strongly accumulated in interzation
of GNOM and PIN1 was observed (Figure 2G).nal compartments upon BFA treatment for 60 min (FigHowever,
double labeling of GNwt
-myc after BFA treat-ures 2A and 2B), as previously described for wild-type
ment revealed a complete colocalization of PIN1 and(Geldner et al., 2001). By contrast, root tips of
GNOM in perinuclear aggregates (Figure 2H), previouslyGNOMM696L
-myc seedlings showed nearly complete redescribed
as “BFA compartments” (Geldner et al., 2001;sistance to BFA-induced intracellular accumulation,
Satiat-Jeunemaitre and Hawes, 1992; Steinmann et al.,with no or only very few intracellular patches visible after
1999). Double labeling of GNOM and PIN1 in GNM696L
-60 min (Figure 2C).
myc lines after BFA treatment revealed no colocalizationImmunolocalization of untreated GNwt
-myc seedlings
of the two proteins, with GNOM remaining mainly cyto-with monoclonal myc-antibody revealed strong cytosolic
and PIN1 staying at the plasma membrane (Figuresolic “background” staining and a number of small in-
2I). The BFA insensitivity of PIN1 accumulation at thetracellular patches (Figure 2D). Upon BFA treatment,
cytosolic GNwt
-myc signal was decreased, with a con- plasma membrane in GNM696L
-myc lines, together with
Cell
222
Figure 2. Brefeldin A Responses of PIN1 and GNOM Are Altered in Engineered GNOM Lines
Confocal images of seedling root tips stained with PIN1 antibody (green) and monoclonal myc antibody (red) on GNOM-myc transgenic lines.
(A–C) PIN1, (D–F) GNOM-myc, and (G–I) PIN1 and GNOM-myc. (A, D, and G) Control treatment on GNwt
-myc line, (B, E, and H) BFA 50 ␮M
for 60 min on GNwt
-myc line, and (C, F, and I) BFA 50 ␮M 60 min on GNM696L
-myc line. Note yellow intracellular dots in (H), indicating colocalization
of PIN1 and GNOM.
the colocalization of PIN1 and BFA-sensitive GNOM to ARF GTPases recruit COPI coats as well as clathrin
coats to membranes for vesicle budding (Boman, 2001,the same endomembranes after BFA treatment, strongly
suggests that GNOM ARF-GEF is responsible for recy- and references therein). Whereas COPI coats mediate
intra-Golgi transport and recycling to the ER, clathrincling PIN1 through an intracellular compartment.
coats have been implicated in the formation of endocytic
vesicles and vesicle budding from the trans-Golgi net-The GNOM ARF-GEF Localizes to Endosomes
To identify the intracellular site of GNOM action, we work (for review, see Holstein, 2002). An early observation
regarding BFA action was the inhibition of COPIperformed double-labeling experiments of GNOM with
available markers for endomembrane compartments. recruitment to membranes (Orci et al., 1991). A similar
effect on the ␥-COP subunit has recently been describedAlthough a number of specific antibodies are available
for plants, very few have been shown to function in in tobacco BY-2 cells (Ritzenthaler et al., 2002). We reasoned
that if GNOM were involved in the recruitment ofconfocal microscopy imaging of Arabidopsis seedling
root tips. The AtSEC12 antiserum detects an ER resi- COPI coats, ␥COP and GNOM should colocalize and
␥-COP localization should become BFA resistant indent, transmembrane protein (Bar-Peled and Raikhel,
1997). This antiserum labeled membrane compartments GNM696L
-myc roots. In untreated cells, ␥-COP appeared
to be associated with Golgi stacks, as judged from theresembling previous descriptions of ER membranes
(Figure 3A). GNwt
-myc did not colocalize with AtSEC12 number, size, and characteristic shape of the signal (Figure
3C), which is consistent with previous descriptions(Figure 3A). Upon BFA treatment, the ER network appeared
unchanged, and the AtSEC12 signal did not colo- of ␥-COP label in tobacco (Ritzenthaler et al., 2002).
However, upon BFA treatment, the reported abolish-calize with the GNOM-labeled patches (Figure 3B). Thus,
GNOM-positive membranes were different from the ER ment of ␥-COP membrane association was not observed
in our system. Instead, ␥-COP stayed at the membrane,network.
GNOM-Dependent Endosomal Recycling
223
Figure 3. Colocalization with Intracellular
Markers Define the GNOM Compartment as
Endosomal
Confocal pictures of seedling root tips with
either myc antibody (red) and Sec12p, ␥COP,
TLG2a antibodies (green) or GN-GFP (green)
and FM4-64 (red), DAPI staining of nuclei
(blue). (A and B) AtSec12 and GNwt
-myc, (C
and D) ␥COP and GNwt
-myc, (F–G) TLG2a and
GNwt
-myc, and (H–M) GN-GFP and FM4-64.
(A, C, and F) Control treatment on GNOMwt
myc
line, (B and D) BFA 50 ␮M for 60 min
teated GNwt
-myc line, (G) BFA 200 ␮M for 4
hr, and (E) BFA 50 ␮M for 60 min on GNM696L
myc
line. (H–J) untreated cells after 30 min of
FM4-64 uptake, (H) and (I) single channels,
(J) overlay, (K–M) BFA 50 ␮M for 45 min with
equal time of FM4-64 uptake, (K) and (L) single
channels, and (M) overlay. Note tightly
associated, sometimes overlapping green
and red dots in (J) and completely overlapping
intracellular dots in (M).
but the individual Golgi stacks aggregated into larger clude that GNOM does not regulate budding of Golgiassociated,
COPI-coated vesicles.clusters, without completely losing their integrity (Figure
3D). Such discrepancies have previously been reported The trans-Golgi network (TGN) can be viewed as the
station of the secretory pathway at which COPI-depen-between different mammalian cell lines (Hunziker et al.,
1991). GNwt
-myc did not appear to colocalize with ␥-COP dent membrane trafficking is taken over by non-COPI
coats, such as clathrin (Kirchhausen, 2000). Antibodiesin untreated cells (Figure 3C). Upon BFA-treatment,
␥-COP-labeled clusters surrounded a GNOM-positive raised against TGN-localized AtTLG2a (SYP41) (Bassham
et al., 2000) showed strong staining of elongated,central patch, without showing any obvious colocalization
(Figure 3D). In BFA-resistant GNOM lines, ␥-COP often sickle-shaped structures (Figure 3F) that appeared
to be different from structures labeled with Golgi stacklabel still aggregated to enclose a “core” which, however,
was GNOM negative (Figure 3E). Thus, ␥-COP and markers (Wee et al., 1998). Again, no colocalization was
observed between TLG2a and GNwt
-myc (Figure 3F). InGNOM did not colocalize nor did ␥-COP compartments
become resistant in BFA-resistant GNOM lines. We con- contrast to GNOM, TLG2a label did not change upon
Cell
224
treatment with BFA, even when high doses were applied
for prolonged time periods, and GNwt
-myc did not colocalize
with TLG2a (Figure 3G). We conclude that GNOMpositive
endomembranes are also different from the
TGN and that GNOM probably does not act along the
secretory pathway.
The fluorescent styryl dye FM4-64 is used as an endocytic
tracer in yeast and mammalian cells. It fluoresces
upon insertion into the plasma membrane and is taken
up by the cell exclusively through membrane trafficking
in yeast (Vida and Emr, 1995). This eventually leads to
labeling of all the compartments along the endocytic
pathway down to the vacuole. Recently, there have been
a number of reports about the use of FM dyes as endocytic
markers in plants (Emans et al., 2002; Parton et
al., 2001; Ueda et al., 2001). Because FM4-64 can only
be used for labeling of live cells, we performed double
labeling with GN-GFP (Figures 3H–3M). Although weak,
GN-GFP labeling of live cells resembled GN-myc immunofluorescence
labeling of fixed cells (Figure 3H; compare
with Figure 2D). Upon uptake of FM4-64 for 30 min,
GNOM-GFP appeared to be closely associated, and partially
overlapping with, the dye-labeled compartments
(Figures 3I and 3J). In the presence of BFA, FM4-64
accumulated in larger patches and, in this case, GN-GFP
completely colocalized with the FM4-64 label to those
patches (Figures 3K–3M). At this time point, vacuoles
were not yet labeled by FM4-64. These data strongly
suggest that GNOM acts at an endosomal compartment
that partially overlaps with, but is not entirely identical
to, the FM4-64-labeled compartment.
Endosomes Are Abnormal in gnom Mutant Cells
If GNOM acts at an endosomal compartment, gnom
mutant cells might show some abnormality in endosoFigure
4. Endosomal ARA7-GFP Labeling in Wild-Type and gnom
mal structure or function. To test this idea, gnom mutant
Arabidopsis Protoplasts
cell cultures were established and transfected with
Projections of confocal sections. Representative examples of ARA7ARA7-GFP
constructs for transient expression. By seGFP
label in (A) wild-type and (B) gnom mutant cells.
quence similarity, ARA7 is a member of the Rab5 family
of endosomal Rab GTPases (Ueda et al., 2001). ARA7
was shown to localize to structures very similar to such as pectins (Baluska et al., 2002; Friml et al., 2002;
GNOM-positive compartments and to also colocalize Geldner et al., 2001; Grebe et al., 2002; Heese et al.,
with compartments of early FM4-64 accumulation (Ueda 2001). To examine whether GNOM is specifically inet
al., 2001, and T.U. and A.N., unpublished data). Upon volved in PIN1 trafficking, we analyzed the effects of
transient expression in cultured cells, ARA7-positive BFA treatment on the localization of KNOLLE, PMcompartments
appeared as larger patches in gnom than ATPase, and the PIN1 homolog PIN2 in our lines.
in wild-type cells (Figures 4A and 4B). In addition, ARA7 The cytokinesis-specific KNOLLE syntaxin localizes
compartments were often ring shaped, which was only to intracellular patches and the newly forming cell plate,
rarely observed in wild-type cells. Similar alterations of destined to become a new plasma membrane (Lauber
ARA7 compartments were observed when treating wild- et al., 1997). Upon BFA treatment, KNOLLE coalesced
type cell cultures with BFA (T.U. and A.N., unpublished into large patches, similar to the described BFA comdata).
These observations support the subcellular local- partments (Figures 5A and 5B; Geldner et al., 2001).
ization data and strongly suggest that GNOM ARF-GEF BFA-induced accumulation was also observed for PMplays
a role in regulating endosome structure and ATPase and PIN2 (Figures 5D and 5E, and 5G and 5H,
function. respectively). GNOM colocalized with all three proteins
upon BFA treatment (data not shown). However, in the
GNM696L
-myc line, KNOLLE aggregation into large patchesRecycling of Other Plasma-Membrane Proteins
in BFA-Resistant GNOM Seedlings still occurred (Figure 5C), suggesting that KNOLLE trafficking
in cytokinetic cells is dependent on other BFA-BFA has been shown to induce intracellular accumulation
of PIN1 and to alter the distribution of other plasma sensitive GEFs. For PM-ATPase and PIN2 localization,
the situation was more complex. First of all, not all PM-membrane localized proteins, including KNOLLE (SYP111),
AtSNAP-33, PM-ATPase, the auxin influx carrier AUX1, ATPase or PIN2-labeled cells responded to BFA, and
the variability between individual roots was higher thanthe PIN1 homolog PIN3, or even cell-wall components
GNOM-Dependent Endosomal Recycling
225
Figure 5. BFA Resistance of Other Plasma
Membrane Markers in BFA-Resistant Lines
(A–C) KNOLLE, (D and E) PM-ATPase, and (G
and H) PIN2. (A, D, and G) Untreated GNwt
myc
line, (B, E, and H) BFA 50 ␮M for 60
min GNwt
-myc, and (C) BFA 50 ␮M for 60 min
GNM696L
-myc line. (F and I) Percentage of cells
showing intracellular accumulation of label
after BFA treatment in sensitive versus resistant
lines. “Sens” is GNwt
-myc line (black) and
“res” is GNM696L
-myc (gray). Each bar is an
average of five root tips, representing about
1000 cells in total.
in the case of PIN1. This might be due to differences in trafficking, which inhibits fundamental processes such
as cell-plate expansion during cytokinesis (Yasuhararecycling rates or transport routes used between individual
cells and roots. When comparing effects of BFA in and Shibaoka, 2000). In contrast, low concentrations of
BFA have surprisingly specific effects on auxin trans-GNM696L
-myc lines, a similar variability was observed,
ranging from almost complete to very weak increases port-related processes such as root hair cell polarity,
gravitropism, or initiation of lateral root primordia (Geld-in resistance compared to GNOMwt
-myc lines. In order
to determine if there is a significant partial resistance ner et al., 2001; Grebe et al., 2002). We tested whether
GNOM is the ARF-GEF responsible for the inhibition ofdespite strong variability, we counted several thousand
cells from z axis scans of a number of root tips. This these processes by BFA. We exposed seedlings to a
change in gravity vector and recorded their realignmentrevealed a partial BFA-resistance of PIN2 and PMATPase
in BFA-resistant lines (Figures 5F and 5I). Thus, after 36 hr. Treatment with 5 ␮M BFA strongly interfered
with the gravitropic response of GNwt
-myc lines (FiguresGNOM action does not mediate trafficking of all plasma
membrane proteins to the same extent. 6B and 6E). GNM696L
-myc seedlings, in contrast, showed
absolutely no difference in their gravitropic response in
the absence or presence of BFA (Figures 6B and 6E).GNM696L
Conferred BFA Resistance of Auxin
Transport and Auxin-Mediated The same was true for lateral root formation. Treatment
with 10 ␮M BFA led to complete inhibition of lateral rootPhysiological Responses
The observation that cycling of PIN proteins became initiation in BFA-sensitive lines whereas, resistant lines
showed no decrease in the number of lateral rootsBFA-resistant in GNM696L
-myc lines led us to investigate
if this would be reflected in BFA insensitivity of polar formed (Figure 6C). Additionally, this concentration of
BFA significantly reduced primary root elongation,auxin transport. Reliable measurements of polar transport
in Arabidopsis can be done on inflorescence stems. which was also less affected in BFA-resistant GNOM
lines (Figure 6D). Thus, the sole alteration of GNOM ARF-Exposing the upper end of inflorescence stems of GNwt
myc
lines to a pulse of radioactively labeled auxin led GEF rendered auxin-related developmental processes
insensitive to inhibition by low concentrations of BFA.to a peak of radioactive auxin at a distance of about 13
mm after 90 min of transport. Upon concomitant BFA
treatment, this peak was nearly abolished (Figure 6A, Discussion
left). By contrast, GNM696L
-myc lines showed no difference
of auxin peak intensity with or without BFA treat- GNOM Is a Large ARF-GEF that Localizes
to Endosomes and Regulates Theirment (Figure 6A, right). Thus, BFA-resistant GNOM confers
BFA-resistance of both PIN1 cycling and polar auxin Structure and Function
Our results indicate that GNOM is an ARF-GEF involvedtransport activity in planta.
High concentrations of BFA lead to growth arrest of in recycling of plasma membrane proteins from endosomes.
This was surprising because GNOM is a membercells, probably due to the severe block of intracellular
Cell
226
Figure 6. BFA Resistance of Auxin Transport
and Auxin-Dependent Growth Responses in
BFA-Resistant Lines
(A) Polar auxin transport of a sensitive and a
resistant line in the absence (gray circles) or
presence (black triangles) of 20 ␮M BFA.
Arrows indicate the peak of polarly transported
auxin, the slope to the left is diffusive
transport of auxin. Note that there is no difference
in the radioactive auxin peak upon BFA
treatment in the resistant line.
(B) Gravitropic root growth response of BFA
sensitive or resistant GNOM lines in the absence
(Ϫ) or presence (ϩ) of 5 ␮M BFA. 4-dayold
seedlings were turned by 135Њ and grown
for another 36 hr. Degree deviation from the
gravitropic vector was measured. Each root
was assigned one of twelve 30Њ sectors.
About 80 seedlings were counted per histogram.
Note that sensitive lines display less
gravitropic curvature in the presence of BFA
(90Њ Ϯ 15Њ) than resistant lines (139.5Њ Ϯ 8Њ).
(C) Lateral root numbers of GNOM BFA-sensitive
and -resistant lines in the absence (Ϫ) or
presence (ϩ) of 10 ␮M BFA. 5-day-old seedlings
were treated for 6 days with or without
BFA. Lateral root primordia visible under the
binocular were counted. About 16 seedlings
were counted for each value. Note that lateral
root formation is abolished in GNOM BFAsensitive
lines in the presence of BFA.
(D) Primary root elongation in the absence (Ϫ)
or presence (ϩ) of 10 ␮M BFA of GNOM BFAsensitive
or resistant lines. 5-day-old seedlings
were grown for 4 days. About 20 seedlings
were counted for each value.
(E) Example of 11-day-old seedlings grown
for 7 days on 5 ␮M BFA. GNOM BFA-sensitive
line (left), GNOM BFA-resistant line (right).
Note lack of lateral roots and gravitropic
growth in the GNOM BFA-sensitive line.
of the Gea/GNOM/GBF1 (GGG) subfamily of large ARF- large ARF-GEFs, neither the GGG-type nor the Sec7/BIG
type, has been shown previously to act on endosomes.GEFs. GGG-type ARF-GEFs have been implicated in ERGolgi
or intra-Golgi traffic, in yeast as well as in animals Our conclusion that GNOM is not involved in transport
through the secretory pathway is based on two lines of(Peyroche et al., 2001; Zhao et al., 2002). This class of
ARF-GEFs can be distinguished from other subfamilies, evidence. First it does not colocalize with several markers
of the secretory system, neither before nor after BFAsuch as the Sec7/BIG class, by overall homology and
differences in size and domain structure. None of the treatment. Second, the compartments investigated are
GNOM-Dependent Endosomal Recycling
227
either naturally resistant to BFA treatment and thus can- ARF-GEFs such as GNL1 or GNL2 might be candidates
not be regulated by GNOM or, if sensitive, they do not for regulating such alternative, perhaps more general,
become resistant in BFA-resistant GNOM lines. In con- pathways.
trast to the secretory markers, the endocytic tracer FM4- While in yeast, ARF-GEFs for the endosomal system
64 showed significant colocalization with GNOM. While are ill-defined, in animals, a number of specific ARFthe
two signals were closely associated and partially GEF classes have evolved for endosomal trafficking
overlapping in the absence of BFA, they completely col- (Donaldson and Jackson, 2000). In Arabidopsis, there
ocalized in the presence of BFA. Although the organiza- are only large ARF-GEFs, which, however, are more nution
of plant endosomes is essentially unknown, mam- merous than in yeast and mammals (Ju¨ rgens and Gelmalian
endosomes are thought to be largely contiguous dner, 2002). Thus, increased complexity of vesicle trafmembrane
compartments that are subdivided through ficking pathways in plants was achieved by a different
combinatorial action of partially overlapping rab do- functional diversification of ARF-GEFs than in animals.
mains (Zerial and McBride, 2001). It is thus conceivable In this view, the subcellular specificity of GNOM action
that GNOM localizes to a subdomain of a more or less apparently reflects the evolution of plant-specific ways
contiguous endosomal system, which partially breaks to regulate endosomal trafficking.
down in response to the drug.
In addition to its endosomal localization and role in The Importance of GNOM-Regulated Vesicle
recycling, GNOM also appears to regulate endosome Trafficking for Polar Auxin Transport
structure, as evidenced by structural alterations of ARA7 A specific feature of plants is the regulation of diverse
compartments in gnom loss-of-function cell lines. Inter- developmental processes through the polar flow of the
estingly, the ring-shaped structures observed in gnom growth regulator auxin, achieved by the coordinated
cells very much resemble endosomes observed in ARF polar distribution of auxin efflux components of the PIN
loss-of-function mutants of yeast (Gaynor et al., 1998; family. The fact that the polar distribution of PIN1 results
Yahara et al., 2001). Our findings indicate that loss of from a continuous recycling process suggests that reguARF-GEF
function has a similar effect on endosomal lation of vesicle trafficking is central to establishment
structure. Although the endosomes observed in gnom and maintenance of polar auxin transport (Geldner et
cells may seem dysfunctional, there is precedence from al., 2001).
animal cells that BFA-induced gross morphological al- BFA-induced internal accumulation of PIN1 was posterations
of the endosomal system do not prevent con- tulated to reflect inhibition of exocytosis in the presence
tinued recycling of markers through this system (Hun- of continued endocytosis (Figure 1B), which is consisziker
et al., 1991). This suggests that structural integrity tent with other recent findings (Emans et al., 2002). Here,
and function can be uncoupled to some degree. we have identified the responsible molecular target by
What are the ARF substrates regulated by GNOM? The demonstrating that a BFA-resistant GNOM ARF-GEF is
Arabidopsis genome encodes six class I ARF proteins
sufficient to maintain PIN1 cycling dynamics in the presrelated
to ARF1 and three more divergent ARFs that
ence of BFA. Moreover, this single alteration also leads
cannot be grouped unambiguously with mammalian or
to BFA-resistant polar auxin transport, which indicates
yeast non-ARF1 classes (Ju¨ rgens and Geldner, 2002).
that GNOM is an important in vivo mediator of auxin
Arabidopsis ARF1 acts in ER/Golgi trafficking in a similar
efflux.
fashion to yeast or mammalian ARF1 (Ritzenthaler et al.,
Low concentrations of BFA cause growth and devel-
2002; Lee et al., 2002; Takeuchi et al., 2002). Whereas
opmental defects that can be interpreted as consenonplant
ARF1 also plays a role in endosomal trafficking
quences of auxin transport inhibition, consistent with
(Gaynor et al., 1998; Gu and Gruenberg, 2000), this has
reported effects of BFA on auxin efflux (Delbarre et al.,
not been investigated for Arabidopsis ARF1, and the
1998; Morris and Robinson, 1998). This effect of BFA
more divergent ARFs are functionally uncharacterized.
was abrogated by exclusively rendering the endosomeThus,
the ARF substrate in GNOM-regulated endosomal
localized GNOM ARF-GEF resistant to BFA. This is in
trafficking remains to be determined.
contrast to the current view that BFA primarily affects
GNOM might act only on a subset of endosomes, as
Golgi-based secretion (Nebenfu¨ hr et al., 2002). The apsuggested
by the differential effects of BFA on plasmaparent
specificity of low concentrations of BFA to plantmembrane protein recycling in BFA-resistant GNOM
endosomal trafficking is surprising. It is conceivable thatplants. Whereas trafficking of basally localized PIN1 is
endosomal trafficking is more sensitive to BFA inhibitionregulated by GNOM, recycling of other proteins appears
than is Golgi-based secretion. Alternatively, partial inhi-to be partially or completely independent of GNOM. The
bition of endosomal trafficking might more readily leadlatter proteins include apically localized PIN2, apolarly
to phenotypic consequences than partial inhibition oflocalized PM-ATPase, and the cytokinesis-specific synthe
secretory pathway, which might reflect the impor-taxin KNOLLE. These findings suggest the existence of
tance of endosomes in auxin transport and possiblyadditional recycling routes. In animals, distinct recycling
other signaling events.pathways have been described (Gruenberg, 2001). Polarized
cells, in particular, make use of specialized apical
Engineered BFA Sensitivities of ARF-GEFsand basolateral recycling routes (Mostov et al., 2000).
Are a New Approach to Dissect VesiclePlants endosomes are essentially undefined. Yet the
Transport Pathwaysimpressively high number of putative endosomal rab
Brefeldin A (BFA) has been extensively used to probeGTPases in Arabidopsis suggests a rather complex envesicle
trafficking pathways in eukaryotic cells, and itdosomal system, which may easily consist of several,
has been established that the catalytic domains of ARFfunctionally independent endosomes (Rutherford and
Moore, 2002; Ueda and Nakano, 2002). GNOM-related GEFs are primary targets of BFA action (Peyroche et
Cell
228
BFA Treatmentsal., 1999; Sata et al., 1999). Since eukaryotic genomes
BFA incubation of seedlings for immunofluorescence labeling wasencode several BFA-sensitive ARF-GEFs that may act
done in cell-culture dishes containing 1 ml basal medium (BM) (0.5
at different stations within the network of intracellular
MS, 1% sucrose [pH 5.8]) and the indicated concentration of BFA
trafficking, the cellular responses to BFA treatment are (BFA stock solution was 50 mM). Control treatments contained equal
manifold, depending on the organism and cell type un- amounts of BFA solvent (DMSO/ethanol 1:1). BFA treatments for
growth assays were done by transferring seedlings germinated onder study, and are often difficult to interpret. Here, we
BM plates onto BFA-containing plates. BFA plates were stored athave taken the converse approach by assessing specific
4ЊC for no longer than 2–3 days before use.differences in response to BFA in cells that express a
single ARF-GEF rendered BFA-resistant on the backTransient
Transfection of Protoplasts
ground of all other sensitive ARF-GEFs. This approach Transient expression of GFP-ARA7 in protoplasts of cultured cells
might also prove useful to investigate in vivo activities were done as described (Ueda et al., 2001), except that suspension
of mammalian ARF-GEFs and to identify the molecular cultured cells were incubated in enzyme solution for 2 hr at 30ЊC
and then passed through the nylon mesh twice (125 ␮m pore at firsttarget for the BFA effects on endosomal trafficking in
and then 40 ␮m pore).mammals. The comparative study of BFA effects on
cells with only one resistant versus sensitive ARF-GEF
Measurement of Polar Auxin Transport
allows for a much more precise dissection of vesicle Polar auxin transport was measured in inflorescence axes of transtrafficking
pathways than was possible before. The genic Arabidopsis plants as described (Parry et al., 2001), with the
same would apply for the reverse experiment of render- following modifications. A concentration of 2 ␮M IAA were used for
all incubations. BFA treatment was done by immersion for 30 mining a BFA-resistant ARF-GEF BFA-sensitive. A systemin
a solution containing IAA and either 20 ␮M BFA or the sameatic extension of this approach might be a means to
volume of DMSO (0.05% v/v). The specific activity of [3
H]IAA used
map specificity of ARF-GEF-dependent trafficking pathwas
962 TBq moleϪ1
.
ways in higher eukaryotic cells.
Growth Measurements
Experimental Procedures Gravitropic growth was assessed by marking the gravity vector on
plates. Pictures of plates were taken and angles measured from
Plant Growth Conditions digital images with Adobe Illustrator.
Plants on soil or plates were grown in growth chambers under long Lateral root formation was assessed by inspecting the primary
day conditions at 21ЊC. root under a binocular for lateral root primordia. In the case of
absence of visible primordia, chloral hydrate-cleared preparations
of roots were also inspected microscopically for absence of pericy-Plasmid Construction, Generation of Transgenic Plants,
cle divisions, indicative of early stages of lateral root development,and Complementation Analysis
not visible in the binocular.An AvrII site was introduced at the 3Ј end of the GNOM-ORF in the
GNOM cDNA Vector c96 (Busch et al., 1996) by primer-extension
AcknowledgmentsPCR (Ausubel et al., 2002; Sata et al., 1999). A 3ϫ myc tag or GFP
tag was synthesized by oligonucleotide hybridization (myc) or PCR
We thank S. Richter for technical assistance and T. Hamann, M.from pSMGFP vector (U70495) and inserted into this site. A Bpu10I,
Heese, A. Schnittger, and D. Weijers for critically reading the manu-PacI fragment encompassing the myc-tag or GFP-tag was then
script. We are indebted to W. Michalke, K. Palme, and D.G. Robinsontransferred into the complementing 7,5 GNOM genomic fragment
for generously providing antibodies. We also wish to thank M.T.(Col ecotype) in pBluescript. The resulting pBlue-GNXbaI-myc was
Morita and M. Tasaka from the Nara Institute of Science and Tech-further modified by introducing a fragment containing the mutagennology
for establishment of the gnom suspension culture cell line.ised Sec7-domain encoding region. Site-directed mutagenesis of
This work was supported by the Deutsche Forschungsgemeinschaftthe Sec7 domain was done in c96 through primer-extension PCR,
(SFB 446, A9).introducing an ATG (M) to CTG (L) change into the GNOM ORF, and
the mutated region was transferred to pBlue-GNXbaI-myc as an
Received: October 28, 2002MscI-Bpu10I fragment. All PCR-derived fragments were sequenced.
Revised: December 16, 2002The resulting GNXbaIM696L
-myc, GNXbaIwt
-myc and GNXbaI-GFP genomic
fragments were then transferred as XbaI restricted fragments
Referencesinto pBAR A (AJ251013) and transformed into plants of a gnom
heterozygous population (F2 population of emb30-1Col
/wtLer
F1 hyAusubel,
F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G.,brids). Complementation analysis was done by PCR screening of
Smith, J.A., and Struhl, K.A. (2002). Current Protocols in MolecularT1 plants using a Ler/Col polymorphism in the GNOM coding region
Biology (New York: John Wiley & Sons).(Busch et al., 1996). Mutant complementation by the constructs was
then confirmed by segregation analysis of seedling phenotypes of Baluska, F., Hlavacka, A., Samaj, J., Palme, K., Robinson, D.G.,
homozygous and heterozygous gnom mutant lines in the F2. In order Matoh, T., McCurdy, D.W., Menzel, D., and Volkmann, D. (2002).
to exclude effects in transgenic plants that are not linked to the F-actin-dependent endocytosis of cell wall pectins in meristematic
introduced GNOM locus itself, but to differences in the lines or root cells. Insights from brefeldin A-induced compartments. Plant
the genetic background of the hybrids used for transformation, we Physiol. 130, 422–431.
always compared independent BFA-resistant transgenic lines to
Bar-Peled, M., and Raikhel, N.V. (1997). Characterization of AtSEC12
independent sensitive ones.
and AtSAR1. Proteins likely involved in endoplasmic reticulum and
Golgi transport. Plant Physiol. 114, 315–324.
Antibody Staining and Confocal Laser Scanning Microscopy Bassham, D.C., Sanderfoot, A.A., Kovaleva, V., Zheng, H., and RaikImmunofluorescence
preparations and confocal microscopy were hel, N.V. (2000). AtVPS45 complex formation at the trans-Golgi netdone
as described (Lauber et al., 1997). Antibodies and dilutions work. Mol. Biol. Cell 11, 2251–2265.
used were: anti-PIN1 (1:200) and anti-PIN2 (1:600), both kindly proBednarek,
S.Y., and Falbel, T.G. (2002). Membrane trafficking during
vided by K. Palme; anti-KNOLLE (1:4000), 9E10 anti-myc (1:600)
plant cytokinesis. Traffic 3, 621–629.
(SantaCruz), anti-TLG2a (1:200) (Rosebiotech), anti-AtSec12 (1:50)
Boman, A.L. (2001). GGA proteins: new players in the sorting game.(Rosebiotech), and anti-At␥COP (1:1000), kindly provided by D.G.
J. Cell Sci. 114, 3413–3418.Robinson; and anti-PM-ATPase (1:1000), kindly provided by W. Michalke.
Busch, M., Mayer, U., and Ju¨ rgens, G. (1996). Molecular analysis of
GNOM-Dependent Endosomal Recycling
229
the Arabidopsis pattern formation of gene GNOM: gene structure Lauber, M.H., Waizenegger, I., Steinmann, T., Schwarz, H., Mayer,
U., Hwang, I., Lukowitz, W., and Jurgens, G. (1997). The Arabidopsisand intragenic complementation. Mol. Gen. Genet. 250, 681–691.
KNOLLE protein is a cytokinesis-specific syntaxin. J. Cell Biol. 139,Defea, K.A., Zalevsky, J., Thoma, M.S., Dery, O., Mullins, R.D., and
1485–1493.Bunnett, N.W. (2000). beta-arrestin-dependent endocytosis of proLee,
M.H., Min, M.K., Lee, Y.J., Jin, J.B., Shin, D.H., Kim, D.H., Lee,teinase-activated receptor 2 is required for intracellular targeting of
K.H., and Hwang, I. (2002). ADP-ribosylation factor 1 of Arabidopsisactivated ERK1/2. J. Cell Biol. 148, 1267–1281.
plays a critical role in intracellular trafficking and maintenance ofDelbarre, A., Muller, P., and Guern, J. (1998). Short-lived and phosendoplasmic
reticulum morphology in Arabidopsis. Plant Physiol.phorylated proteins contribute to carrier-mediated efflux, but not to
129, 1507–1520.influx, of auxin in suspension-cultured Tobacco cells. Plant Physiol.
Lippincott-Schwartz, J., Yuan, L., Tipper, C., Amherdt, M., Orci,116, 833–844.
L., and Klausner, R.D. (1991). Brefeldin A’s effects on endosomes,Donaldson, J.G., and Jackson, C.L. (2000). Regulators and effectors
lysosomes, and the TGN suggest a general mechanism for regulat-of the ARF GTPases. Curr. Opin. Cell Biol. 12, 475–482.
ing organelle structure and membrane traffic. Cell 67, 601–616.
D’Souza-Schorey, C., van Donselaar, E., Hsu, V.W., Yang, C., Stahl,
Lukowitz, W., Mayer, U., and Ju¨ rgens, G. (1996). Cytokinesis in theP.D., and Peters, P.J. (1998). ARF6 targets recycling vesicles to the
Arabidopsis embryo involves the syntaxin-related KNOLLE geneplasma membrane: insights from an ultrastructural investigation. J.
product. Cell 84, 61–71.Cell Biol. 140, 603–616.
Mayer, U., Buettner, G., and Juergens, G. (1993). Apical-basal pat-Emans, N., Zimmermann, S., and Fischer, R. (2002). Uptake of a
tern formation in the Arabidopsis embryo: studies on the role of thefluorescent marker in plant cells is sensitive to brefeldin A and wortgnom
gene. Development 117, 149–162.mannin. Plant Cell 14, 71–86.
Morris, D.A., and Robinson, J.S. (1998). Targeting of auxin carriersFranco, M., Peters, P.J., Boretto, J., van Donselaar, E., Neri, A.,
to the plasma membrane: differential effects of brefeldin A on theD’Souza-Schorey, C., and Chavrier, P. (1999). EFA6, a sec7 domaintraffic
of auxin uptake and efflux carriers. Planta 205, 606–612.containing exchange factor for ARF6, coordinates membrane recyMostov,
K.E., Verges, M., and Altschuler, Y. (2000). Membrane trafficcling and actin cytoskeleton organization. EMBO J. 18, 1480–1491.
in polarized epithelial cells. Curr. Opin. Cell Biol. 12, 483–490.Frank, S., Upender, S., Hansen, S.H., and Casanova, J.E. (1998).
Mu¨ ller, A., Guan, C., Ga¨ lweiler, L., Tanzler, P., Huijser, P., Marchant,ARNO is a guanine nucleotide exchange factor for ADP-ribosylation
A., Parry, G., Bennett, M., Wisman, E., and Palme, K. (1998). AtPIN2factor 6. J. Biol. Chem. 273, 23–27.
defines a locus of Arabidopsis for root gravitropism control. EMBOFriml, J., and Palme, K. (2002). Polar auxin transport–old questions
J. 17, 6903–6911.and new concepts? Plant Mol. Biol. 49, 273–284.
Nebenfu¨ hr, A., Ritzenthaler, C., and Robinson, D.G. (2002). BrefeldinFriml, J., Wisniewska, J., Benkova, E., Mendgen, K., and Palme, K.
A: deciphering an enigmatic inhibitor of secretion. Plant Physiol.(2002). Lateral relocation of auxin efflux regulator PIN3 mediates
130, 1102–1108.tropism in Arabidopsis. Nature 415, 806–809.
Orci, L., Tagaya, M., Amherdt, M., Perrelet, A., Donaldson, J.G.,Ga¨ lweiler, L., Guan, C., Mu¨ ller, A., Wisman, E., Mendgen, K.,
Lippincott-Schwartz, J., Klausner, R.D., and Rothman, J.E. (1991).Yephremov, A., and Palme, K. (1998). Regulation of polar auxin
Brefeldin A, a drug that blocks secretion, prevents the assembly oftransport by AtPIN1 in Arabidopsis vascular tissue. Science 282,
non-clathrin-coated buds on Golgi cisternae. Cell 64, 1183–1195.2226–2230.
Parry, G., Delbarre, A., Marchant, A., Swarup, R., Napier, R., Perrot-Gaynor, E.C., Chen, C.Y., Emr, S.D., and Graham, T.R. (1998). ARF
Rechenmann, C., and Bennett, M.J. (2001). Novel auxin transportis required for maintenance of yeast Golgi and endosome structure
inhibitors phenocopy the auxin influx carrier mutation aux1. Plantand function. Mol. Biol. Cell 9, 653–670.
J. 25, 399–406.
Geldner, N., Friml, J., Stierhof, Y.D., Ju¨ rgens, G., and Palme, K.
Parton, R.M., Fischer-Parton, S., Watahiki, M.K., and Trewavas, A.J.(2001). Auxin transport inhibitors block PIN1 cycling and vesicle
(2001). Dynamics of the apical vesicle accumulation and the ratetrafficking. Nature 413, 425–428.
of growth are related in individual pollen tubes. J. Cell Sci. 114,
Grebe, M., Friml, J., Swarup, R., Ljung, K., Sandberg, G., Terlou, 2685–2695.
M., Palme, K., Bennett, M.J., and Scheres, B. (2002). Cell polarity
Peyroche, A., Antonny, B., Robineau, S., Acker, J., Cherfils, J., and
signaling in Arabidopsis involves a BFA-sensitive auxin influx pathJackson,
C.L. (1999). Brefeldin A acts to stabilize an abortive ARFway.
Curr. Biol. 12, 329–334.
GDP-Sec7 domain protein complex: involvement of specific resiGruenberg,
J. (2001). The endocytic pathway: a mosaic of domains. dues of the Sec7 domain. Mol. Cell 3, 275–285.
Nat. Rev. Mol. Cell Biol. 2, 721–730.
Peyroche, A., Courbeyrette, R., Rambourg, A., and Jackson, C.L.
Gu, F., and Gruenberg, J. (2000). ARF1 regulates pH-dependent (2001). The ARF exchange factors Gea1p and Gea2p regulate Golgi
COP functions in the early endocytic pathway. J. Biol. Chem. 275, structure and function in yeast. J. Cell Sci. 114, 2241–2253.
8154–8160.
Ritzenthaler, C., Nebenfuhr, A., Movafeghi, A., Stussi-Garaud, C.,
Hadfi, K., Speth, V., and Neuhaus, G. (1998). Auxin-induced develop- Behnia, L., Pimpl, P., Staehelin, L.A., and Robinson, D.G. (2002).
mental patterns in Brassica juncea embryos. Development 125, Reevaluation of the effects of brefeldin A on plant cells using to-
879–887. bacco Bright Yellow 2 cells expressing Golgi-targeted green fluoresHaj,
F.G., Verveer, P.J., Squire, A., Neel, B.G., and Bastiaens, P.I. cent protein and COPI antisera. Plant Cell 14, 237–261.
(2002). Imaging sites of receptor dephosphorylation by PTP1B on Robineau, S., Chabre, M., and Antonny, B. (2000). Binding site of
the surface of the endoplasmic reticulum. Science 295, 1708–1711. brefeldin A at the interface between the small G protein ADP-ribosyHeese,
M., Gansel, X., Sticher, L., Wick, P., Grebe, M., Granier, F., lation factor 1 (ARF1) and the nucleotide-exchange factor Sec7 doand
Ju¨ rgens, G. (2001). Functional characterization of the KNOLLE- main. Proc. Natl. Acad. Sci. USA 97, 9913–9918.
interacting t-SNARE AtSNAP33 and its role in plant cytokinesis. J. Rutherford, S., and Moore, I. (2002). The Arabidopsis Rab GTPase
Cell Biol. 155, 239–249. family: another enigma variation. Curr. Opin. Plant Biol. 5, 518–528.
Holstein, S.E. (2002). Clathrin and plant endocytosis. Traffic 3, Sanderfoot, A.A., Pilgrim, M., Adam, L., and Raikhel, N.V. (2001).
614–620. Disruption of individual members of Arabidopsis syntaxin gene famiHunziker,
W., Whitney, J.A., and Mellman, I. (1991). Selective inhibi- lies indicates each has essential functions. Plant Cell 13, 659–666.
tion of transcytosis by brefeldin A in MDCK cells. Cell 67, 617–627. Sata, M., Moss, J., and Vaughan, M. (1999). Structural basis for
Ju¨ rgens, G., and Geldner, N. (2002). Protein secretion in plants: from the inhibitory effect of brefeldin A on guanine nucleotide-exchange
the trans-Golgi network to the outer space. Traffic 3, 605–613. proteins for ADP-ribosylation factors. Proc. Natl. Acad. Sci. USA 96,
2752–2757.Kirchhausen, T. (2000). Three ways to make a vesicle. Nat. Rev. Mol.
Cell Biol. 1, 187–198. Satiat-Jeunemaitre, B., and Hawes, C. (1992). Redistribution of a
Cell
230
Golgi glycoprotein in plant cells treated with Brefeldin A. J. Cell Sci.
103, 1153–1166.
Shevell, D.E., Leu, W.M., Gillmor, C.S., Xia, G., Feldmann, K.A., and
Chua, N.H. (1994). EMB30 is essential for normal cell division, cell
expansion, and cell adhesion in Arabidopsis and encodes a protein
that has similarity to Sec7. Cell 77, 1051–1062.
Shevell, D.E., Kunkel, T., and Chua, N.H. (2000). Cell wall alterations
in the Arabidopsis emb30 mutant. Plant Cell 12, 2047–2060.
Someya, A., Sata, M., Takeda, K., Pacheco-Rodriguez, G., Ferrans,
V.J., Moss, J., and Vaughan, M. (2001). ARF-GEP(100), a guanine
nucleotide-exchange protein for ADP-ribosylation factor 6. Proc.
Natl. Acad. Sci. USA 98, 2413–2418.
Steinmann, T., Geldner, N., Grebe, M., Mangold, S., Jackson, C.L.,
Paris, S., Ga¨ lweiler, L., Palme, K., and Ju¨ rgens, G. (1999). Coordinated
polar localization of auxin efflux carrier PIN1 by GNOM ARF
GEF. Science 286, 316–318.
Takeuchi, M., Ueda, T., Yahara, N., and Nakano, A. (2002). Arf1
GTPase plays roles in the protein traffic between the endoplasmic
reticulum and the Golgi apparatus in tobacco and Arabidopsis cultured
cells. Plant J. 31, 499–515.
Teleman, A.A., Strigini, M., and Cohen, S.M. (2001). Shaping morphogen
gradients. Cell 105, 559–562.
Ueda, T., and Nakano, A. (2002). Vesicular traffic: an integral part
of plant life. Curr. Opin. Plant Biol. 513–517.
Ueda, T., Yamaguchi, M., Uchimiya, H., and Nakano, A. (2001). Ara6,
a plant-unique novel type Rab GTPase, functions in the endocytic
pathway of Arabidopsis thaliana. EMBO J. 20, 4730–4741.
Vida, T.A., and Emr, S.D. (1995). A new vital stain for visualizing
vacuolar membrane dynamics and endocytosis in yeast. J. Cell Biol.
128, 779–792.
Wee, E.G., Sherrier, D.J., Prime, T.A., and Dupree, P. (1998). Targeting
of active sialyltransferase to the plant Golgi apparatus. Plant
Cell 10, 1759–1768.
Wood, S.A., Park, J.E., and Brown, W.J. (1991). Brefeldin A causes
a microtubule-mediated fusion of the trans-Golgi network and early
endosomes. Cell 67, 591–600.
Yahara, N., Ueda, T., Sato, K., and Nakano, A. (2001). Multiple roles
of Arf1 GTPase in the yeast exocytic and endocytic pathways. Mol.
Biol. Cell 12, 221–238.
Yasuhara, H., and Shibaoka, H. (2000). Inhibition of cell-plate formation
by brefeldin A inhibited the depolymerization of microtubules
in the central region of the phragmoplast. Plant Cell Physiol. 41,
300–310.
Zerial, M., and McBride, H. (2001). Rab proteins as membrane organizers.
Nat. Rev. Mol. Cell Biol. 2, 107–117.
Zhao, X., Lasell, T.K., and Melancon, P. (2002). Localization of large
ADP-ribosylation factor-guanine nucleotide exchange factors to different
Golgi compartments: evidence for distinct functions in protein
traffic. Mol. Biol. Cell 13, 119–133.