Current Biology 20, 2144–2149, December 7, 2010 ª2010 Elsevier Ltd All rights reservedDOI 10.1016/j.cub.2010.11.016
SPIKE1 Signals Originate
from and Assemble Specialized
Domains of the Endoplasmic Reticulum
Chunhua Zhang,1Simeon O. Kotchoni,1A. Lacey Samuels,3
and Daniel B. Szymanski1,2,*
1Department of Agronomy
2Department of Biological Sciences
Purdue University, West Lafayette, IN 47907-2054, USA
3Department of Botany, University of British Columbia,
Vancouver, BC V6T 1Z4, Canada
In the leaf epidermis, intricately lobed pavement cells use
Rho of plants (ROP) small GTPases to integrate actin and
microtubule organization with trafficking through the secre-
tory pathway [1–5]. Cell signaling occurs because guanine
nucleotide exchangefactors (GEFs) promoteROP activation
and their interactions with effector proteins that direct the
cell growth machineries . In Arabidopsis, SPIKE1 (SPK1)
is the lone DOCK family GEF [7, 8]. SPK1 promotes polarized
its mode of action in cells is not known. Vertebrate DOCK
proteins are deployed at the plasma membrane [9, 10]. Like-
wise, current models place SPK1 activity and/or active ROP
at the plant plasma membrane and invoke the localized
patterning of the cortical cytoskeleton as the mechanism
for shape control [1, 4, 6, 11]. In this paper, we find that
SPK1 is a peripheral membrane protein that accumulates
at, and promotes the formation of, a specialized domain of
the endoplasmic reticulum (ER) termed the ER exit site
(ERES). SPK1 signals are generated from a distributed
network of ERES point sources and maintain the homeo-
stasis of the early secretory pathway. The ERES is the loca-
tion for cargo export from the ER . Our findings open up
unexpected areas of plant G protein biology and redefine the
ERES as a subcellular location for signal integration during
Results and Discussion
Arabidopsis SPK1 Is a Peripheral Membrane Protein that
Colocalizes with Rho of Plants at Intracellular Puntae
Cotyledon pavement cells undergo a synchronized lobe
formation and cell expansion process that is clearly altered
ing (Figures 1A and 1B) and nongrowing (Figure 1C) pavement
cells, a specific SPIKE1 (SPK1) antibody (Figures S1A–S1C,
available online) detected punctate structures with a mean
diameter of 515 6 271 nm (mean 6 standanrd deviation [SD])
that were often broadly distributed over faint clouds of
diffuse SPK1 signal. SPK1 was detected in the nucleus (Fig-
ures 1D–1F) and as less abundant, larger cylindrical structures
ofunknown functionthatprotruded towardthecentralvacuole
(Figure 1A, inset). At three days after germination (DAG), pave-
ment cells had a stronger diffuse signal in trans-vacuolar
strands and along the cell perimeter that may reflect the
greater cytosolic volume of these less vacuolated cells. In
leaf pavement cells, SPK1 had both a punctate and peripheral
localization (Figures S1D–S1F), the latter partially colocalized
with the plasma membrane marker PHOT1-GFP (Figures
S1O–S1Q) . Surprisingly, SPK1 punctae were quite
uniformly distributed in pavement cells. For example, the
mean density of SPK1 particles in the central apical domain
of pavement cells was 0.9 6 0.1 particles/mm2and this value
was not significantly different from the particle densities
measured in the lobes and in the basal cortex of expanding
pavement cells (Figures S1D–S1F and Table S1). Although
we cannot rule out a pool of plasma membrane-associated
SPK1, the punctate localization of SPK1 is not an artifact of
disruption of the plasma membrane or endoplasmic reticulum
(ER) systems because image planes of the upper and medial
cell surface reveal that GFP-fusion proteins that mark these
compartments retain their overall structure during our labeling
protocol (Figures S1G–S1Q).
We found that SPK1 punctae were the result of a peripheral
association with cell membranes. In cell fractionation experi-
ments using leaf extracts, more than 99% of SPK1 was distrib-
uted between organelle and crude microsome fractions, with
only trace amounts being detected in the soluble fraction (Fig-
ure 1G). SPK1 was tightly associated with microsomes but
had the properties of a peripheral membrane protein because
detected in microsomal but not cytosolic fractions , one
would expectROP tobepresent atSPK1punctae.Inpavement
cells, ROP localization is complicated. Type I ROPs are preny-
lated and their C termini are posttranslationally modified on
after departure from the ER. To detect ROPs in Arabidopsis
pavement cells, we affinity purified two different ROP anti-
bodies, both of which labeled the cell periphery and the same
punctate structures (Figures S1R–S1T). The peripheral ROP
signal in leaf pavement cells was consistent with a plasma
membrane localization (Figure 1L). However, intracellular pools
of punctate ROP that appear to define internal organelle
surfaces have also been reported in antibody labeling and
GFP-ROP localization studies [16–19]. Double labeling of
SPK1 and ROP (Figures 1I–1K) detected a 56% colocalization
of colocalization was highly significant (p < 2 3 10232) when
compared to randomized SPK1 and ROP particles with an
that we used to calculate chance colocalization subtracted
11% 6 5% (mean 6 SD) of the overlap between two identical
SPK1 images that were analyzed in two separate channels.
The SPK1 punctae are abundant and contain a large cellular
otide exchange factor (GEF)-independent functions for SPK1
domain in vivo . The simplest explanation is that SPK1
punctae correspond to subcellular locations of ROP activation.
SPK1 Is Localized to Subdomains of the ER
To learn more about potential endomembrane functions for
SPK1, weseparatedcrude leafcellmicrosomes oncontinuous
sucrose gradients and analyzed the distribution of SPK1
relative to known organelle markers (Figure 2A). In leaf micro-
somes, the known ER resident proteins SEC12 and PDI were
located in the densest fractions near the bottom of the
gradient. The ER fractions were well resolved from other com-
partment markers, including those for the plasma membrane
and the Golgi (Figure 2A). SPK1 signal overlapped significantly
with the ER markers (Figure 2A). As further proof for ER
localization, EDTA-stripping of ribosomes from rough ER
reproducibly caused a clear shift of both SEC12 and SPK1 to
less dense fractions on the gradient (Figure 2B and Figures
S2A and S2B).
The cell fractionation and localization data suggested that
SPK1 punctae defined subdomains of the ER. The mean size
and density of SPK1 punctae were very similar to the values
reported for endogenous ER exit site (ERES) in tobacco BY-2
cells . At the ERES, plant cells use a conserved group of
signaling and structural proteins to assemble coat protein
complex II (COPII)-coated vesicles that traffic protein and lipid
to the Golgi apparatus . Antibodies raised against known
Arabidopsis ERES proteins SEC12 and SAR1 are specific
[21, 23] and label hundreds of punctae in developing Arabido-
posis pavement cells (Figures 2D and 2G). Fluorescent protein
fusions to SEC12 are distributed throughout the ER [21, 24].
The punctate distribution of SEC12 that we observed probably
reflected a subpool of SEC12 that was resistant to detergent
extraction during our localization protocol. As expected for
known ERES proteins, SAR1 and SEC12 clearly colocalized
(Figures S2F–S2H). A pool of SPK1 localized to putative
ERES, because SPK1 punctae showed a highly significant
level of overlap with SEC12 (Figures 2C–2E, 52% 6 12%,
mean 6 SD, p < 1 3 10223), SAR1 (Figures 2F–2H, 34% 6
8%, p < 1 3 10224), and SEC13 (Figures S2C–S2E, 31% 6
11%, p < 1 3 10215). Colocalization of SPK1 with ERES
markers was also observed in leaf mesophyll cells (data not
shown). We also found that ROP was localized to these ER
subdomains because 51% of the ROP signal colocalized
with SAR1 (Figures S2I–S2K).
In plant cells, the ERES is a subdomain of the ER that
generates COPII-coated vesicles . Plant cells do not
appear to utilize an intermediate compartment to traffic vesi-
cles between the ER and the Golgi. Instead, ERES may exist
as mobile secretory units that physically associate and move
with Golgi bodies . However, immunolocalization data
suggest that ERES are more abundant than the Golgi and
that the two compartments only partially overlap . We
examined the extent to which SPK1 punctae overlapped with
the Golgi, using the YFP-tagged soybean a-1,2 Mannosidase I
(YFP:ManI)  as a cis-Golgi marker. We found that SPK1
punctae were more abundant than the Golgi, and there was
no significant colocalization (Figures 2I–2K); however, we did
observe numerous occasions in which SPK1 punctae were
adjacent to Golgi bodies (Figure 2K, arrows). These results
Figure 1. A Punctate Colocalization of SPK1
and ROP in Expanding Pavement Cells
(A–C) SPK1 localization pattern is not signifi-
cantly changed at different stages of epidermal
cell growth. SPK1 localization in 3 DAG (A),
5 DAG (B), and 12 DAG (C) cotyledon pavement
cells. A maximum Z projection of the upper half
of the cell is shown in each panel. (A, inset) An
xz view of a large SPK1 puncta is indicated by
the arrow and reveals a cylindrical structure
oriented toward the central vacuole.
(D–F) SPK1 is localized to the nucleus. (D) Immu-
nolabeling of SPK1 in the nucleus. (E) Staining of
the nucleus by propidium iodide (PI). (F) Overlay
of (D) and (E).
(G) Cellular fractionation of SPK1. Different
fractions of cell extracts were loaded by equal
proportion (except lane 6) and blotted with anti-
SPK1. The same membrane was reprobed with
anti-PEP carboxylase antibody.
(H) Solubilization of membrane-associated SPK1
protein by different reagents. Pellet fractions
pavement cells. Single image plane from the
cortical region of a pavement cell labeled with
anti-SPK1 (I), anti-ROP (J), and the overlay (K).
(L) Peripheral localization of ROP. Single optical
section from the medial plane of a pavement
cell. The abbreviation vac denotes the large
The scale bars represent 10 mm in (A–C) and (L),
2 mm in (A inset); and 5 mm in (I–K). See also
Figure S1 for antibody characterizations and
Table S1 for SPK1 particle density measure-
with ROPs inleaf
ROPGEF Signaling from Subdomains of the ER
are consistent with our sucrose gradient fractions that were
probed with Golgi and ER marker antibodies and clearly indi-
cate that SPK1 is an ER-localized protein. As previously
observed , we found that SPK1, SAR1, and SEC12 punctae
were adjacent to but distinct from the ER that is labeled with
the GFP-HDEL marker (Figures S1G–S1I and S2L to S2Q).
Importantly, there are numerous specialized domains of the
ER , and it remains to be determined whether all of the
SPK1 punctae correspond to ERES that participate in vesicle
trafficking between the ER and the Golgi.
SPK1 Promotes ERES Assembly but Does Not Block ER
Given that a pool of SPK1 colocalized with SEC12 and SAR1,
two proteins with regulatory functions during COPII-coated
vesicle formation , we tested for an effect of SPK1 on the
Figure 2. SPK1 Localizes to a Subdomain of the
ER Termed the ER Exit Site
(A) Western blots of microsomal fractions sepa-
rated on sucrose density gradients. Each fraction
was blotted with antibodies against proteins that
localize to specific organelles and anti-SPK1
antibody. Mannosidase, VPS45, RabA-4b, ASP,
H+-ATPase, and SYP21 were used as markers
for Golgi, TGN, TGN and post-Golgi vesicles,
vacuole, plasma membrane, and prevacuolar
(B) Dissociation of ribosomes from the ER using
EDTA shifts the distribution of SPK1 and SEC12
from high (24 to 29) to lower (20 to 26) sucrose
(C–E) Arabidopsis wild-type leaf pavement cell
labeled with anti-SPK1 and anti-SEC12 anti-
bodies. Single image plane from the cortical
region of a pavement cell labeled with anti-
SPK1 (C), anti-SEC12 (D), and the overlay (E).
(F–H) Arabidopsis wild-type leaf pavement cell
labeled with anti-SPK1 and anti-SAR1 anti-
bodies. Single image plane from the cortical
region of a pavement cell labeled with anti-
SPK1 (F), anti-SAR1 (G), and the overlay (H).
(I–K) Arabidopsis pavement cell expressing YFP:
ManI labeled with anti-SPK1 antibody. Single
slice image from the cortex region of the epi-
dermal cell labeled with anti-SPK1 (I), expressing
YFP:ManI (K), and the overlay (K). Arrows in (I–K)
indicate example regions of SPK1 punctae close
for the effect of Mg2+on SPK1 distribution on
sucrose gradient; the double labeling of SPK1
with SEC13 in pavement cells; double labeling of
nolabeling of SAR1 and SEC12 in epidermal cells
ERES labeled with SAR1. Compared to
the particulate distribution of SEC12
and SAR1 in the wild-type (Figures 3A
and 3C), spk1 cells displayed a more
diffuse signal with a reduced number
of discernable punctae (Figures 3B
and 3D). To rule out the trivial explana-
tion that SEC12 was solubilized from
the membrane or that the GDP-bound
nucleotide status of SAR1 released it
into a soluble fraction , we measured the microsome
association of both proteins in cell fractionation experiments.
We found that in both wild-type and spk1 cell extracts, SEC12
and SAR1 partitioned completely into the microsome fraction,
unlike PEPC, a known cytosolic enzyme (Figure 3E). There-
fore, spk1 does not affect the membrane association of
either COPII component; rather, it promotes their local
concentration into specialized membrane domains. This result
is important because ERES number/cell varies greatly
between organisms and cell types, and in general the genetic
and biochemical basis of its control are poorly understood.
Future experiments will examine the possibility that SPK1
locally concentrates COPII proteins at the ER surface, either
through direct physical interactions or by defining a special-
ized membrane environment in the ER that promotes COPII
Current Biology Vol 20 No 23
The punctate ER localization of SPK1 and the COPII locali-
zation defect of the mutant suggested that SPK1 could partic-
ipate in some aspect of ERES assembly and/or trafficking
between the ER and Golgi. However, nonessential functions
of SPK1 at ERES are expected because, unlike spk1, muta-
tions that efficiently block COPII are lethal . We used
measure the exchangeable pool of Golgi YFP:ManI and the
rate at which it accumulates to a steady state in the Golgi.
In 3 DAG cotyledons, photobleached Golgi from wild-type
and spk1 cells displayed a clear time-dependent increase in
YFP:ManI fluorescence compared to neighboring unbleached
controls (Figures S3A and S3B). The normalized time-depen-
dent FRAP signal of wild-type and mutant cells was reproduc-
ible (Figures S3E and S3F) and could be fitted equally well to
a single phase exponential association equation that was
used to calculate the mean half-times of recovery (t1/2) and
the size of the exchangeable pool. In spk1 cells, we detected
a small decrease in the exchangeable pool of Golgi-localized
YFP:ManI (Figures S3A, S3B, S3E, S3F, and S3J), that became
obvious at later developmental stages (Figures S3C, S3D,
S3G, S3H, and S3J). The t1/2values for spk1 (154 6 26 s,
mean 6 SD) and wild-type cells (170 6 43 s, mean 6 SD)
did not differ (Figure S3I), and the values for both were similar
to those reported for another Golgi resident enzyme in
tobacco leaf cells . Apparently, the SEC12 and SAR1
localization defects in spk1 were not sufficient to cause
a global and easily detectable ER export defect. Perhaps
SPK1 affects the trafficking and/or recycling of subsets of
cargo between the ER and Golgi, and our arbitrary choice of
YFP:ManI for the trafficking assay fails to reveal the nature
of this activity. It may be that the reduced exchangeable Golgi
pool in spk1 reflects a clog in the Golgi that is caused by
defective retrograde transport and the failure to efficiently
recycle specific cargo or cargo receptors at the ER-Golgi
ER Functions for SPK1 that Are Independent
of the WAVE/ARP2/3 Pathway
To better understand the importance of SPK1 in the early
secretory pathway, we tested for additional defects in the
physiology of the ER. In leaf epidermal cells, the cortical ER
exists as a network of interconnected tubules and cisternae
(Figure 4A). In spk1, the ER network was less tubulated and
In yeast, dilated ER cisternae are characteristic of secretion
mutants, some of which affect COPII-dependent trafficking
. High-pressure freeze fixation and electron microscopy
of thin sections from wild-type and spk1 pavement cells
detected highly variable diameters of the ER lumen in the
mutant (Figures 4C and 4D). The mean diameter of transverse
sections through the rough ER of spk1 (189 6 137 nm, mean 6
SD) was significantly greater than that of the wild-type (132 6
43 nm, mean 6 SD) (p < 0.01). We did not detect a difference
in the diameter of the smooth ER. Although the vacuoles of
spk1 pavement cells were occasionally smaller than those
of the wild-type, we did not detect any clear defects in
the morphology of the Golgi or other endomembrane
Plant cells utilize conserved ER stress response signaling
pathways to sense imbalanced protein export, attenuate
translation, and increase the protein folding capacity of the
ER. Dilation of the ER could result from a failure to export or
degrade unfolded proteins or subsets of secreted proteins. If
so, one might expect to detect an activation of the ER stress
response. We measured the steady state mRNA levels of
several known ER stress response genes as well as two cold
and osmotic stress response genes, CBF1 and COR47 .
In spk1 seedlings, the ER stress response genes were specif-
ically elevated by factors ranging from w4 to w9 compared to
wild-type plants (Figure 4E). Another known marker for ER
stress and translational attenuation is hyperphosphorylation
of eIF2a . Consistent with the notion that the ER stress
response is constitutively activated in spk1, eIF2a was
hyperphosphorylated compared to the wild-type (Figure 4F).
However, the transcriptional ER stress response was not
sylation inhibitor tunicamycin, both wild-type and mutant cells
had similar levels of marker gene induction compared to
untreated controls (Figures S4A and S4B). To test for the
involvement of ROP in the ER stress phenotype, we measured
the panel of ER stress marker genes in transgenic lines that
overexpress dominant-negative (DN) and constitutively active
(CA) forms of ROP . The ER stress markers were strongly
upregulated in DN-rop2 lines but varied randomly in the
CA-rop2 lines (Figure S4C). These data provide additional
support for the hypothesis that spk1 ER phenotypes are
caused by defective activation of ROP.
If SPK1 maintains efficient protein trafficking in the early
secretory pathway, one might expect that the mutant would
Figure 3. SEC12 and SAR1 Localization Are Altered in spk1 Cells
(A and B) Single image plane from the upper cortex of 3 DAG cotyledon pavement cells labeled with anti-SEC12 in either punctate (Col) (A) or more diffuse
patterns (spk1) (B).
(C and D) Single image plane from the upper cortex of 3 DAG cotyledon pavement cells labeled with anti-SAR1 antibody in either punctate (Col) (C) or more
diffuse patterns in spk1 (D).
(E)MembraneassociationofSAR1 andSEC12in3DAGwild-type(lane1and2)andspk1(lane3and4)cotyledon cells.a-PEPCispositive controlforsoluble
fraction and a-SPK control demonstrates lack of protein in mutant.
The scale bars represent 10 mm. See also Figure S3 for the FRAP analysis of YFP:ManI trafficking in wild-type and spk1 pavement cells.
ROPGEF Signaling from Subdomains of the ER
be hypersensitive to tunicamycin. The wild-type plants were
not noticeably affected but when treated under identical
conditions, 100% of the spk1 plants died (Figure 4G). The
WAVE-ARP2/3 branch of SPK1 signaling was not involved in
ER homeostasis because the ARP2/3 null mutant dis2 
was not hypersensitive to tunicamycin (Figure 4G) and did
not display constitutive upregulation of ER stress response
genes (Figure 4E). Therefore the ER-related phenotypes of
spk1 define a clear bifurcation of SPK1 signaling activities.
Implications for the Subcellular Control of ROP Signaling
In conclusion, we find that SPK1 functions at discrete punctae
on the surface of the ER. The ER morphology and trafficking
phenotypes of spk1 point to its importance in the early
secretory pathway. Genetic and biochemical data suggest
that ROP activation is the primary function of SPK1 . Our
analyses of ROP localization and the ER stress response
caused by DN-rop2 suggest that SPK1 activities at ERES are
linked to its GEF activity and ROP signaling. These results
expand cellular models of ROP signaling that presume the
plasma membrane to be the sole location for ROP activation
and signaling. SPK1 also appears to promote ERES formation.
This activity could reflect a form of positive autoregulation in
which SPK1 promotes the formation of its own subcellular
signaling domain. Further research into the interactions
between SPK1 and components of the ERES may help to
clarify the poorly understood cellular control of the number,
spacing, and activity of this specialized domain of the ER
[35, 36]. The plant ERES may be a strategic location where
dynamics. Although the SPK1-dependent cargo at the ER
surface is not known, SPK1 may monitor the flux of misfolded
proteins , the assembly of ribonucleoprotein complexes
, or the general export load of the ER [21, 35].
In tip growing plant cells  and polarized nonplant cells
, the activation of ROP/Rac small GTPases and their
effectors is often restricted to specialized cortical domains in
the cell. However, in growing pavement cells, we find no
obvious correlation between the density and location of
SPK1 punctae and cell shape. Given that the cellular control
and geometry of pavement cell shape change is poorly under-
stood, one can only speculate about how SPK1 localization
relates to morphogenesis. Perhaps ERES form a latent distrib-
uted network of point sources for ROP activation and only
subsets of SPK1 punctae are actively signaling. However,
the colocalization of SPK1 and ROP is extensive, and if this
defines active pools of SPK1, ERES could be locations for
constitutive SPK1 signaling. In this scenario SPK1 could
have general effects on the partitioning of ROP-GTP into
specialized membrane domains . In addition, different
classes of effectors with ROP-GTP-dependent subcellular
targeting could convert distributed SPK1 signals into localized
cellular responses. Further experiments that integrate SPK1
signaling and shape change in different epidermal cell types
willhelp todefinehow ER-localized signaling complexes coor-
Supplemental Information includes Supplemental Experimental Proce-
dures, four figures, and two tables and can be found with this article online
This research was supported by National Science Foundation Molecular
and Cell Biology Grant 0640872 to D.B.S. and Natural Sciences and
Engineering Research Council of Canada Discovery Grant to A.L.S. Thanks
to Eileen Mallery and members of the Samuels lab for editorial assistance
and to David Robinson and Natashia Raikhel for generous antibody gifts.
We are also grateful to Zhenbiao Yang for the kind gift of the CA- and
Figure 4. Activation of the ER Stress Response in spk1
(A and B) Single image plane from the upper cortex of 3 DAG wild-type (A) or
spk1 (B) cotyledon pavement cells expressing the GFP-HDEL ER marker.
(C and D) High-pressure freeze transmission electron microscopy images of
4 DAG wild-type (C) or spk1 (D) cotyledon pavement cells.
(E) Real-time PCR quantification of the expression level of ER and abiotic
stress induced genes in spk1 and dis2(arpc2) relative to the wild-type in
3 DAG cotyledons. ACT2 gene was used as the internal control.
(F) eIF2a is hypherphosphorylated in 3 DAG cotyledons of spk1 (lane 2)
compared with the wild-type (lane 1).
(G) spk1 is hypersensitive to tunicamycin. Recovery of wild-type, spk1, and
dis2(arpc2) seedlings after treatment with 0.17% DMSO or 2.5 mg/ml
tunicamycin for 1 hr.
Arrowheads in (A) and (B) indicate ER bodies. In both (C) and (D), arrows
indicate ER tubules and CW denotes cell wall. The scale bars represent
10 mm in (A) and (B) and 500 nm in (C) and (D). See also Figure S4 for
the induction of ER stress response genes by tunicamycin and the
expression of ER stress response genes in CA-rop2 and DN-rop2 mutants.
See Table S2 for the list of the oligonucleotide primer sequences.
Current Biology Vol 20 No 23
Received: July 28, 2010
Revised: October 8, 2010
Accepted: November 4, 2010
Published online: November 24, 2010
1. Fu, Y., Gu, Y., Zheng, Z., Wasteneys, G., and Yang, Z. (2005).
Arabidopsis interdigitating cell growth requires two antagonistic path-
ways with opposing action on cell morphogenesis. Cell 120, 687–700.
2. Bloch, D., Lavy, M., Efrat, Y., Efroni, I., Bracha-Drori, K., Abu-Abied, M.,
Sadot, E., and Yalovsky, S. (2005). Ectopic expression of an activated
RAC in Arabidopsis disrupts membrane cycling. Mol. Biol. Cell 16,
3. Lavy, M., Bloch, D., Hazak, O., Gutman, I., Poraty, L., Sorek, N.,
Sternberg, H., and Yalovsky, S. (2007). A Novel ROP/RAC effector links
cell polarity, root-meristem maintenance, and vesicle trafficking. Curr.
Biol. 17, 947–952.
4. Yalovsky, S., Bloch, D., Sorek, N., and Kost, B. (2008). Regulation of
membrane trafficking, cytoskeleton dynamics, and cell polarity by
ROP/RAC GTPases. Plant Physiol. 147, 1527–1543.
5. Szymanski, D.B. (2009). Plant cells taking shape: new insights into
cytoplasmic control. Curr. Opin. Plant Biol. 12, 735–744.
6. Berken, A., and Wittinghofer, A. (2008). Structure and function of
Rho-type molecular switches in plants. Plant Physiol. Biochem. 46,
7. Qiu, J.L., Jilk, R., Marks, M.D., and Szymanski, D.B. (2002). The
Arabidopsis SPIKE1 gene is required for normal cell shape control
and tissue development. Plant Cell 14, 101–118.
8. Basu, D., Le, J., Zakharova, T., Mallery, E.L., and Szymanski, D.B.
(2008). A SPIKE1 signaling complex controls actin-dependent cell
morphogenesis throughthe heteromeric
complexes. Proc. Natl. Acad. Sci. USA 105, 4044–4049.
9. Co ˆte ´, J.F., Motoyama, A.B., Bush, J.A., and Vuori, K. (2005). A novel and
evolutionarily conserved PtdIns(3,4,5)P3-binding domain is necessary
for DOCK180 signalling. Nat. Cell Biol. 7, 797–807.
10. Nishikimi, A., Fukuhara, H., Su, W., Hongu, T., Takasuga, S., Mihara, H.,
Cao, Q., Sanematsu, F., Kanai, M., Hasegawa, H., et al. (2009).
Sequential regulation of DOCK2 dynamics by two phospholipids during
neutrophil chemotaxis. Science 324, 384–387.
11. Brembu, T., Winge, P., Bones, A.M., and Yang, Z. (2006). A RHOse by
any other name: a comparative analysis of animal and plant Rho
GTPases. Cell Res. 16, 435–445.
12. Bassham, D.C., Brandizzi, F., Otegui, M.S., and Sanderfoot, A.A. (2008).
The secretory system of Arabidopsis. In The Arabidopsis Book.
13. Szymanski, D.B. (2005). Breaking the WAVE complex: the point of
Arabidopsis trichomes. Curr. Opin. Plant Biol. 8, 103–112.
14. Wan, Y.L., Eisinger, W., Ehrhardt, D., Kubitscheck, U., Baluska, F., and
Briggs, W. (2008). The subcellular localization and blue-light-induced
movement of phototropin 1-GFP in etiolated seedlings of Arabidopsis
thaliana. Mol Plant 1, 103–117.
15. Bracha-Drori, K., Shichrur, K., Lubetzky, T.C., and Yalovsky, S. (2008).
Functional analysis of Arabidopsis postprenylation CaaX processing
enzymes and their function in subcellular protein targeting. Plant
Physiol. 148, 119–131.
16. Lin, Y., Seals, D.F., Randall, S.K., and Yang, Z. (2001). Dynamic localiza-
tion of rop GTPases to the tonoplast during vacuole development. Plant
Physiol. 125, 241–251.
17. Bischoff, F., Vahlkamp, L., Molendijk, A., and Palme, K. (2000).
Localization of AtROP4 and AtROP6 and interaction with the guanine
nucleotide dissociation inhibitor AtRhoGDI1 from Arabidopsis. Plant
Mol. Biol. 42, 515–530.
18. Fowler, J.E., Vejlupkova, Z., Goodner, B.W., Lu, G., and Quatrano, R.S.
(2004). Localization to the rhizoid tip implicates a Fucus distichus
Rho family GTPase in a conserved cell polarity pathway. Planta 219,
19. Sorek, N., Segev, O., Gutman, O., Bar, E., Richter, S., Poraty, L., Hirsch,
J.A., Henis, Y.I., Lewinsohn, E., Ju ¨rgens, G., and Yalovsky, S. (2010).
An S-acylation switch of conserved G domain cysteines is required
for polarity signaling by ROP GTPases. Curr. Biol. 20, 914–920.
20. Delcroix, J.D., Valletta, J.S.,Wu, C., Hunt, S.J., Kowal, A.S., and Mobley,
W.C. (2003). NGF signaling in sensory neurons: evidence that early
endosomes carry NGF retrograde signals. Neuron 39, 69–84.
21. Yang, Y.-D., Elamawi, R., Bubeck, J., Pepperkok, R., Ritzenthaler, C.,
and Robinson, D.G. (2005). Dynamics of COPII vesicles and the Golgi
apparatus in cultured Nicotiana tabacum BY-2 cells provides evidence
fortransientassociationofGolgi stackswithendoplasmic reticulum exit
sites. Plant Cell 17, 1513–1531.
22. Matheson, L.A., Hanton, S.L., and Brandizzi, F. (2006). Traffic between
the plant endoplasmic reticulum and Golgi apparatus: to the Golgi
and beyond. Curr. Opin. Plant Biol. 9, 601–609.
23. Bar-Peled, M., and Raikhel, N.V. (1997). Characterization of AtSEC12
and AtSAR1. Proteins likely involved in endoplasmic reticulum and
Golgi transport. Plant Physiol. 114, 315–324.
24. daSilva, L.L.P., Snapp, E.L., Denecke, J., Lippincott-Schwartz, J.,
Hawes, C., and Brandizzi, F. (2004). Endoplasmic reticulum export sites
and Golgi bodies behave as single mobile secretory units in plant cells.
Plant Cell 16, 1753–1771.
25. Staehelin, L.A., and Kang, B.H. (2008). Nanoscale architecture of endo-
plasmic reticulum export sites and of Golgi membranes as determined
by electron tomography. Plant Physiol. 147, 1454–1468.
26. Nebenfu ¨hr,A.,Gallagher,L.A.,
Mazurkiewicz, A.M., Meehl, J.B., and Staehelin, L.A. (1999). Stop-and-
go movements of plant Golgi stacks are mediated by the acto-myosin
system. Plant Physiol. 121, 1127–1142.
27. Staehelin, L.A. (1997). The plant ER: a dynamic organelle composed of
a large number of discrete functional domains. Plant J. 11, 1151–1165.
28. Faso, C., Chen, Y.N., Tamura, K., Held, M., Zemelis, S., Marti, L.,
Saravanan, R., Hummel, E., Kung, L., Miller, E., et al. (2009). A missense
mutation in the Arabidopsis COPII coat protein Sec24A induces the
formation of clusters ofthe endoplasmic reticulumand Golgiapparatus.
Plant Cell 21, 3655–3671.
29. Saint-Jore, C.M., Evins, J., Batoko, H., Brandizzi, F., Moore, I., and
Hawes, C. (2002). Redistribution of membrane proteins between the
Golgi apparatus and endoplasmic reticulum in plants is reversible and
not dependent on cytoskeletal networks. Plant J. 29, 661–678.
30. Novick, P., Field, C., and Schekman, R. (1980). Identification of 23
complementation groups required for post-translational events in the
yeast secretory pathway. Cell 21, 205–215.
31. Koizumi, N., Martinez, I.M., Kimata, Y., Kohno, K., Sano, H., and
Chrispeels, M.J. (2001). Molecular characterization of two Arabidopsis
Ire1 homologs, endoplasmic reticulum-located transmembrane protein
kinases. Plant Physiol. 127, 949–962.
32. Lageix, S., Lanet, E., Pouch-Pe ´lissier, M.-N., Espagnol, M.-C., Robaglia,
C., Deragon, J.-M., and Pe ´lissier, T. (2008). Arabidopsis eIF2alpha
kinase GCN2 is essentialfor growth instress conditions and isactivated
by wounding. BMC Plant Biol. 8, 134.
33. Fu, Y., Li, H., and Yang, Z. (2002). The ROP2 GTPase controls the
formation of cortical fine F-actin and the early phase of directional cell
expansion during Arabidopsis organogenesis. Plant Cell 14, 777–794.
34. El-Din El-Assal, S., Le, J., Basu, D., Mallery, E.L., and Szymanski, D.B.
(2004). DISTORTED2 encodes an ARPC2 subunit of the putative
Arabidopsis ARP2/3 complex. Plant J. 38, 526–538.
35. Hanton, S.L., Chatre, L., Renna, L., Matheson, L.A., and Brandizzi, F.
(2007). De novo formation of plant endoplasmic reticulum export sites
is membrane cargo induced and signal mediated. Plant Physiol. 143,
36. Spang, A. (2009). On vesicle formation and tethering in the ER-Golgi
shuttle. Curr. Opin. Cell Biol. 21, 531–536.
37. Caldwell, S.R., Hill, K.J., and Cooper, A.A. (2001). Degradation of
endoplasmic reticulum (ER) quality control substrates requires trans-
port between the ER and Golgi. J. Biol. Chem. 276, 23296–23303.
38. Wilhelm, J.E., Buszczak, M., and Sayles, S. (2005). Efficient protein
trafficking requires trailer hitch, a component of a ribonucleoprotein
complex localized to the ER in Drosophila. Dev. Cell 9, 675–685.
39. Berzat, A., and Hall, A. (2010). Cellular responses to extracellular
guidance cues. EMBO J. 29, 2734–2745.
ROPGEF Signaling from Subdomains of the ER
Current Biology, Volume 20
SPIKE1 Signals Originate
from and Assemble Specialized
Domains of the Endoplasmic Reticulum
Chunhua Zhang, Simeon O. Kotchoni, A. Lacey Samuels, and Daniel B. Szymanski
Figure S1. Characterization of the Anti-SPK1 and Anti-ROP Antibodies and Localization
Protocols. Related to Figure 1
(A) Characterization of anti-SPK1 antibody. Equal amounts of cell extracts from wild-type (Col-
0) (lane 1) and spk1 (lane 2) null mutant were separated on SDS-PAGE gels and probed in
western blots with anti-SPK1 antibody and the same membrane was re-probed with anti-PEPC
(B) and (C) Immunolabeling of spk1 cells with affinity purified anti-SPK1 antibody (B) and anti-
ACTIN antibody (C).
(D) to (F) Immunolabeling of SPK1 protein in 14 DAG wild-type (Col-0) leaf pavement cells.
(D) Maximum Z-projection of a whole cell labeled with anti-SPK1 antibody. (E) Single image
plane from the upper cortical region of a lobe (box 1) in (D). (F) Single image plane from the
cortex near the center of the cell (box 2) shown in (D).
(G) to (I) Immunolabeling of SPK1 protein (H) in a GFP-HDEL-expressing (G) leaf pavement
cell. (I) Overlay of (G) and (H). Inset in (I) is the enlargement of the boxed region.
(J) A maximum Z-projection from the upper cortex of a PHOT1-GFP living cell.
(K) Single optical section of the medial plane of a vacuolated PHOT1-GFP-expressing living
(L) to (Q) Immunolabeling of SPK1 protein (M and P) in a PHOT1-GFP (L and O) leaf
pavement cell. (N) Overlay of (L) and (M). (Q) Overlay of (O) and (P). (L) to (M) are maximum
Z-projections of confocal images from the upper half of the cell. (O) to (Q) are single optical
sections from the medial plane of the cell.
(R) to (T) Immunolabeling of ROP by affinity purified anti-ROP antibody from rabbit (R) and
anti-ROP antibody from rat (S). (T) Overlay of (R) and (S). (R) to (T) are single optical sections
from the upper cortex of the cell. Bar= 5 µm in (E), (F), (R), (S) and (T). Bar = 10 µm in all
Figure S2. Quantification for the Effect of Mg2+ on SPK1 Distribution on Sucrose Density
Gradients and Immunolabeling of SPK1 and ROP with ERES Markers SEC13, SAR1 and
SEC12 in Arabidopsis Wild-Type and GFP-HDEL Pavement Cells. Related to Figure 2
(A) to (B) Quantification for the effect of Mg2+ on SPK1 distribution on sucrose gradient. (A)
Quantification of the western blot signal intensity of SPK1 and SEC12 proteins in gradient
fractions in the presence of Mg2+. (B) Quantification of the intensity of SPK1 and SEC12
proteins in different fractions in the absence of Mg2+.
(C) to (E) Immunolabeling of SPK1 and SEC13 in a wild-type leaf pavement cell. (C) Single
confocal image plane from the cortex region of an epidermal cell labeled with anti-SPK1. (D)
Single confocal image plane from the cortex region of an epidermal cell labeled with anti-
SEC13. (E) Overlay of (C) and (D)
(F) to (H) Immunolabeling of SAR1 and SEC12 in wild-type leaf pavement cell. (F) Single
confocal image plane showing the labeling of SAR1 in a pavement cell. (G) Single confocal
image plane showing the labeling of SEC12 in a pavement cell. (H) Overlay of (F) and (G).
(I)-(K) Immunolabeling of SAR1 and ROP in a wild-type leaf pavement cell. (I) Single confocal
image plane showing the labeling of SAR1 in a pavement cell. (J) Single confocal image plane
showing the labeling of ROP in a pavement cell. (K) Overlay of (I) and (J).
(L) to (N) Immunolabeling of SAR1 in a GFP-HDEL pavement cell. (L) Single confocal image
plane from GFP-HDEL. (M) Single confocal image plane showing the immunolabeling of
SAR1. (N) Overlay of (L) and (M).
(O) to (Q) Immunolabeling of SEC12 in a GFP-HDEL pavement cell. (O) Single confocal image
plane from GFP-HDEL. (P) Single confocal image plane showing the immunolabeling of
SEC12. (Q) Overlay of (O) and (P).
Bar = 5 µm in (C), (D) and (E). Bar = 10 µm in all other panels.
Figure S3. Altered Protein Trafficking in the Early Secretory Pathway in spk1 cells. Related to
(A) to (D) Fluorescence recovery after photobleaching (FRAP) analysis of YFP:Man I
trafficking from ER to the Golgi in wild-type (Col) and spk1. (A) FRAP of YFP:Man I in a wild-
type (Col-0) pavement cell at 3 DAG. (B) FRAP of YFP:Man I in a spk1 mutant pavement cell at
3 DAG. (C) FRAP of YFP:Man I in a wild-type (Col-0) pavement cell at 8 DAG. (D) FRAP of
YFP:Man I in a spk1 pavement cell at 8 DAG. The circled regions represent the region of interest
(ROI). One ROI was bleached and the other one was used as control. The numbers in each
image identify the time point from the FRAP experiment (seconds). The first image in each panel
represents the pre-bleach conditions. The blue circles label bleached regions and red circles
represents unbleached regions.
(E) to (H) Representative fluorescence recovery curves of YFP:Man I in wild-type and spk1
pavement cells at 3 DAG and 8 DAG. Traces in blue color represent the bleached regions and
traces in red color represent the un-bleached regions. (E) Fluorescence recovery curve of
YFP:Man I in 3 DAG wild-type (Col) cotyledon pavement cells. (F) Fluorescence recovery
curve of YFP:Man I in 3 DAG spk1 cotyledon pavement cells. (G) Fluorescence recovery curve
of YFP:Man I in 8 DAG wild-type (Col) cotyledon pavement cells. (H) Fluorescence recovery
curve of YFP:Man I in 8 DAG spk1 cotyledon pavement cells.
(I) The half-time for fluorescence recovery in different stages of cotyledon pavement cells in
wild-type (Col) and spk1. There are no significant differences in the half-time of fluorescence
recovery between wild-type (Col) and spk1 at 3 DAG and 8 DAG (t-test, p>0.05).
(J) The maximum fluorescence recovery and exchangeable protein pools in wild-type (Col) and
spk1 at 3 DAG and 8 DAG. The maximum fluorescence recovery in wild-type (Col) and spk1 3
DAG cotyledons is not significantly different. However, at 8 DAG, the maximum fluorescence
recovery for spk1 is about 50% of the pre-bleach levels, which is significantly different than
wild-type (Col) (indicated by asterisk) (t-test, p<0.01).
Figure S4. ER Stress Genes Response to Tunicamycin Treatment in spk1 and the Expression of
ER Stress Genes in CA-rop2 and DN-rop2 Mutants. Related to Figure 4
(A) Expression of ER stress genes in Col and spk1 after treatment by tunicamycin or DMSO.
The expression level of each gene is normalized to ACT2.
(B) Fold induction of different genes by tunicamycin in Col and spk1.
(C) Expression of ER stress genes in CA-rop2 (G15V), a mutation that stabilizes bound GTP and
DN-rop2 (D121A), a mutation that reduces the affinity of ROP for nucleotide. The expression
level of each gene is normalized to ACT2.
Table S1. SPK1 Particle Density in Different Subcellular Domains of Expanding Pavement Cells
Cell domain Central apical domain
particles/µm2 (mean ±
Table S2. Oligonucleotide Primers Used for Quantitative Real-Time PCR
0.9 ± 0.2
0.8 ± 0.2 0.9 ± 0.1
Supplemental Experimental Procedures
Plant strains and growth conditions
The Arabidopsis thaliana ecotype Col-0 was used as the wild-type. The spk1-1 allele is as
previously described  and was used in all the mutant analyses except the FRAP experiment.
The spk1-3 allele is as previously described  and was used in the FRAP experiment to avoid
silencing of the YFP:ManI transgene in spk1-1. The GFP-HDEL ER marker was obtained from
the Arabidopsis Biological Resource Center (Columbus, Ohio) . The YFP:ManI plant line was
a gift from Dr. Andreas Nebenfuhr . PHOT1-GFP is as described in . The CA-rop2 and
DN-rop2 mutants were gifts from Dr. Zhenbiao Yang . The plants that were used for
immunolocalization and microsomal fraction preparation were grown on half-strength Murashige
and Skoog (½ MS) media with 1% sucrose at 22 ºC with constant illumination. To test the
expression of ER stress genes in wild-type and spk1, 3 DAG cotyledons were used. To test the
expression of ER stress genes in CA-rop2 and DN-rop2, 9 DAG shoots were used because the
DN-rop2 mutants could not be identified from other plants in a segregating population at early
stages. To test the sensitivity of wild-type (Col) and spk1 to ER stress induction, 3 DAG
seedlings grown on normal ½ MS medium with agar were treated with ½ MS liquid medium
containing 2.5 g/ml tunicamycin or 0.17% DMSO for 1 hour. After rinsing with ½ MS, the
treated seedlings were placed on solid ½ MS medium and were allowed to recover for 6 days
with continuous light illumination.
Immunolocalization by freeze shattering and confocal microscopy
The protocol for immunolocalization is similar to what has been previously described . Plant
materials were incubated in fixative solution (100 mM PIPES, 10 mM EGTA, 4 mM MgCl2, 2%
paraformaldehyde, 0.5% glutaraldehyde, 0.1% triton x-100, pH 6.9) for 1 hour at room
temperature with gentle shaking. The samples were washed for 2 times (10 minutes each) in
PEMT buffer (100 mM PIPEs, 10 mM EGTA, 4 mM MgCl2 , 0.1% triton x-100, pH 6.9) and
then freeze shattered in liquid nitrogen. The shattered samples were permeabilized in buffer (1 x
PBS, 1% triton x-100, pH 6.9) for 1 hour and then washed with PBSTG (1 x PBS, 25 mM
Glycine, 0.1% triton x-100, PH 6.9) for 2 times (5 minutes each). The permeabilized samples
were then used for immunolabeling with different antibodies. For the double labeling of SPK1
with SEC12, SPK1 with SAR1, and SPK1 with ROP, SPK1 antibody was converted to goat
antibody according to the supplier's recommendation
(http://www.jacksonimmuno.com/technical/exampleb.asp). In the double labeling of SAR1 with
SEC12 and SAR1 with ROP, SAR1 antibody was converted to goat antibody. The samples were
first incubated with a primary antibody (anti-SPK1 or anti-SAR1) that was converted to goat
antibody and normal mouse serum and then the samples were incubated with goat anti-rabbit Fab
fragment after washing steps. Another primary antibody (anti-SAR1 or anti-SEC12 antibodies or
anti-ROP) was applied to the sample after anti-rabbit Fab fragment incubation. The FITC-mouse
anti-goat antibody was used to detect the converted primary antibody and Rhodamine red X-
mouse anti-rabbit was used to detect the other primary antibody. The conversion of anti-SPK1 or
anti-SAR1 antibody was complete because we did not detect any Rhodamine red X-mouse anti-
rabbit fluorescence in the samples in the absence of anti-SAR1 or anti-SEC12 or anti-ROP
antibodies. Anti-SAR1 and anti-SEC13 were gifts from Dr. David Robinson’s lab (University of
Heidelberg) and were used at 1:400 and 1:100 dilutions respectively . Anti-SEC12 antibody
was a gift from Dr. Natasha Raikhel’s lab (University of California, Riverside) and was used at
1:400 dilution . The mouse monoclonal anti-actin (JLA20) antibody (Calbiochem, San Diego,
CA) was used at 1:400 to detect actin. Alexa488 conjugated goat anti-rabbit, Alexa546
conjugated goat anti-rabbit, Alexa488 conjugated goat anti-rat and Rhodamine conjugated goat
anti-mouse (IgM) secondary antibodies were from Invitrogen (Carlsbad, CA) and were used at
1:400 dilutions. Normal mouse serum, goat anti-rabbit Fab fragment, FITC mouse anti-goat
antibody, and Rhodamine red X mouse anti-rabbit antibody were from Jackson Immunoresearch
(West Grove, PA). The normal mouse serum and the goat anti-rabbit Fab fragment were used at
1:200 dilutions. The FITC mouse anti-goat antibody and the Rhodamine red X mouse anti-rabbit
antibody were used at 1:400 dilutions. Affinity purified anti-SPKN, affinity purified anti-ROP
from rabbit and affinity purified anti-ROP from rat were used at 1:18, 1:20 and 1:5 dilutions
A Bio-Rad Radiance 2100 Laser Scanning Confocal Microscope (Bio-Rad, Hercules, CA), was
used for imaging. A 488 nm laser line and a 500-560 nm emission filter were used for imaging in
the green channel and a 543 nm laser line and a 555-625 nm emission filter were used for
imaging in the red channel. A 100x 1.4 NA oil objective or a 40x 1.3 NA oil objective were used
to take images after immunostaining. For live cell imaging, the GFP-HDEL and YFP:ManI lines
were imaged using a 60x 1.2 NA water objective.
Quantification for SPK1 particle size distribution and co-localization
To quantify the size distribution of SPK1, 5 representative images from each labeling were
selected. In each selected image, 6 regions with the size of 40 pixel by 40 pixel were used. The
images were thresholded and the particles in the selected regions were analyzed using Image J
software (http://rsb.info.nih.gov/ij/). The particles with feret diameter of 1 pixel were discarded.
To quantify the co-localization between two proteins, at least 5 representative confocal images
from each double labeling experiment were selected. From each selected image, 6 regions of 40
pixel by 40 pixel in size from the apical region of the cell cortex were used. The selected regions
in red channel images were rotated 90 and 180 degrees. After thresholding, the co-localizations
between the green channel and the red channel (g/r), the green channel and the 90 degree-rotated
red channel image (g/r90), and the green channel and the 180 degree-rotated red channel images
(g/r180) were measured using Metamorph 6.0 (Molecular Devices, Sunnyvale, CA). The co-
localization values were calculated using the equation co-localization=g/r-(g/r90+g/r180)/2.
Cell fractionation assay and sucrose density gradient separation of membrane proteins
For the cell fractionation assay, the microsomal fraction was isolated from 21 DAG plants using
a homogenization buffer containing 20 mM Hepes/KCl pH 7.2, 50 mM KOAc, 2 mM
Mg(OAc)2, 250 mM Sorbitol, 1 mM EDTA, 1 mM EGTA, 1mM DTT, 1 mM PMSF and 1 %
(v/v) protease inhibitor cocktail (Sigma, St.Louis, Missouri) . Two grams of shoots were
homogenized in 5 ml homogenization buffer on ice, and then filtered through a pre-wetted
double layer of Miracloth to remove the cell debris. A small portion of the flow through was
saved as total protein. Then the remainder was spun at 1000 g for 30 minutes at 4 C. The
supernatant was then centrifuged at 10,000 g for 30 minutes at 4 C. The resulting 10,000 g
supernatant was then spun at 200,000 g for 30 minutes at 4 C. The supernatant fraction after
200,000 g spin was defined as 200k S. The 1000 g pellet (1k P), 10,000 g pellet (10k P) and the
200,000 g pellet (200k P) fractions were resuspended in the same volume of homogenization
buffer as the supernatant. To concentrate the soluble fraction, 1 ml of 200k S was incubated with
0.02% sodium deoxycholate for 30 minutes on ice, then the mixture was incubated with 10%
trichloroacetic acid overnight on ice at 4°C. The mixture was spun at 20,000 g for 15 minutes
and the resulting pellet was resuspended in 100 l of homogenization buffer. To test the
solubilization of SPK1 by different salts and chaotropes, the 20,000 g pellet fractions were
incubated as indicated in the figure with gentle rocking at room temperature for 30 minutes.
Then the solublized and membrane-associated proteins were separated by centrifugation at
200,000 g for 30 minutes at 4 C. The resulting pellet fractions were blotted with anti-SPK1 and
For sucrose gradients, the microsomal fraction was isolated from Arabidopsis shoots (21
DAG) using the same homogenization buffer as above. Two grams of shoots were homogenized
in 10 ml homogenization buffer on ice, filtered through a pre-wetted double layers of Miracloth
and centrifuged at 1,000 x g, 4 °C, for 30 min. The resulting supernatant was centrifuged at
200,000 g. The resulting freshly prepared pellet, P-200k, was resuspended in 200 l 10 mM
Tris•HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 10% (v/v) glycerol, PMSF and protease
inhibitors (suspension buffer) and directly loaded on sucrose density gradients.
For sucrose gradient centrifugation, 20-50 % and (w/w) sucrose density gradients were made in
10 mM Tris•HCl pH 7.6, 2 mM EDTA, 1 mM DTT, 1 mM PMSF, and protease inhibitors
(centrifugation buffer). For sucrose gradient analysis performed in the presence and absence of
free Mg2+, 20-55 % (w/w) sucrose gradients were used. The fractions were analyzed using
antibodies to known organelle marker proteins as described in the figure legend. Western blot
signal intensities were measured and normalized using ImageQuant 5.2 software (Molecular
dynamics, Sunnyvale, CA). To assay the membrane association of SAR1, SEC12 and ROP
proteins, 3 DAG cotyledons were harvested from wild-type and spk1 seedlings. The microsomal
fractions were prepared and the 1000 g supernatant fractions were centrifuged at 200,000 g. The
resulting pellet (P) was resuspended in the same volume of homogenization buffer as the
supernatant (S). Equal proportions of pellet and supernatant fractions were loaded on SDS-
PAGE gels and blotted. To detect the level of phosphorylation of eukaryotic initiation factor 2 ,
equal amounts of 1000 g supernatants from wild-type (Col) and spk1 were blotted with an
antibody against phospho-eIF-2 (pS51) (Epitomics, Burlingame, CA). A HRP conjugated goat
anti-rabbit secondary antibody (IgG) and SuperSignal West Pico Chemiluminescence Substrate
(Thermo Scientific, Rockford, IL) was used to detect the protein.
RNA isolation, reverse transcription and quantitative RT-PCR
Total RNA was isolated with Trizol (Invitrogen, Carlsbad, CA) reagent. Two micrograms of
total RNA was random primed and reverse transcribed using M-MLV reverse transcriptase
(Promega, Madison, WI) and oligonucleotides mixtures (Random Primer 6, New England
Biolabs, Ipswich, MA). 1/50 of the total cDNA was used in each q-RT-PCR reaction. Gene
specific primers for AtBiP1, AtBiP2, AtCNX1, AtCNX2, AtPDIL3, CBF1, and COR47 are listed
in Table S2. ACT2 was used as the reference gene for normalization. Real-time PCR primers
were designed using Primer Express Version 3.0 (Applied Biosystems Inc, Foster City, CA). The
qPCR reactions were performed using the ABI 7900 system. In each reaction, 10 l SYBR
Green PCR Master Mix (2x) (Applied Biosystems Inc, Foster City, CA) was used in 20 l total
reactions. Triplicate reactions were performed for each cDNA preparation and two biological
replicates were performed for each sample. Ct (cycles to threshold) was calculated by
substracting the Ct value of the ACT2 gene from the Ct value of Candidate genes ( Ct = Ct
(Candidate)- Ct (ACT2)). To compare the gene expression level in wild-type and mutants, the
Ct value was calculated by substracting Ct mutant from Ct wild-type. The relative
expression level in wild-type compared to the mutant is calculated as: fold change = 2- Ct. The
data shown are representative of two independent biological replicates. To test the induction of
ER stress genes by tunicamycin, wild-type and spk1 plants were grown in ½ MS medium with
1% sucrose and 0.8% agar. Three day old seedlings were transferred to a liquid version of the
same media containing either 5 g/ml tunicamycin or 0.17% DMSO. After incubation in liquid
medium for 5 hours, cotyledons from wild-type and spk1 plants were harvested separately and
were used for RNA isolation. Expression levels of the ER stress and other marker genes were
calculated relative to ACT2.
High Pressure Freeze Fixation and Transmission Electron Microscopy (HPF-TEM)
Whole cotyledons of 4 DAG seedlings were processed for HPF-TEM as previously described 
and TEM images containing cross sections of rough or smooth ER were used. To measure the
diameter of the ER cisternae, ER domains were distinguished based on the presence or absence
of ribosomes. For each ER fragment mean diameter was measured based on several
measurement for each cisternae. The first width measurement was done 50 nm from the end of
the tubule. After the first measurement, the ER diameter was measured every 100 nm along its
length. Images were taken from 2 wild-type and 3 spk1 cotyledons that were well preserved. 20
images from at least 10 cells for each genotype were analyzed. A Student’s T-test was used to
test for significant differences in ER diameter.
Fluorescence Recovery After Photobleaching (FRAP)
Multiple time course function of Lasersharp software in a Bio-Rad Radiance 2100 Laser
Scanning Confocal Microscope was used to do the FRAP. To immobilize the Golgi during the
timecourse , 3 and 8 DAG seedlings were treated with 5µM latrunculin B for 1 hour and the
FRAP experiments were done within 1 hour after treatment. Three Regions Of Interest (ROIs) of
identical size and shape were selected for each FRAP experiment. One ROI is the bleached
region, one is the control region without bleaching and one is the background that has no Golgi
apparatus. Before photo bleaching, the signal intensity was sampled for two data points with
laser power lower than 5% and the iris 4-6. To bleach, 100% transmittance power for the 474,
488 and 514 laser lines was used. After bleaching, the fluorescence intensity recovery was
monitored every 5 seconds using the pre-bleach microscope settings. The mean fluorescence
intensity of the bleached region and the unbleached region was calculated by subtracting the
mean fluorescence intensity of the background region. Then calculated fluorescence intensity
was normalized to the pre-bleaching fluorescence. The FRAP data was plotted with GraphPad
Prism 5.0c (La Jolla, CA) and the curve was fitted with a single phase exponential association
equation Y=Y0 + (Plateau-Y0)*(1-exp(-K*x) (Y, fluorescence intensity; x, time of recovery; Y0,
fluorescence intensity right after photobleaching; Plateau, fluorescence intensity at equilibrium).
The half time for fluorescence recovery and size of the exchangeable pool was calculated from
the fitted curve. Data from 8-10 individual Golgi from different cells were used to calculate these
Qiu, J.L., Jilk, R., Marks, M.D., and Szymanski, D.B. (2002). The Arabidopsis SPIKE1
gene is required for normal cell shape control and tissue development. Plant Cell 14, 101-
Basu, D., Le, J., Zakharova, T., Mallery, E.L., and Szymanski, D.B. (2008). A SPIKE1
signaling complex controls actin-dependent morphogenesis through the WAVE and
ARP2/3 complexes. PNAS USA 105, 4044-4049.
Nelson, B.K., Cai, X., and Nebenfuhr, A. (2007). A multicolored set of in vivo organelle
markers for co-localization studies in Arabidopsis and other plants. Plant J. 51, 1126-
Nebenfuhr, A., Gallagher, L.A., Dunahay, T.G., Frohlick, J.A., Mazurkiewicz, A.M.,
Meehl, J.B., and Staehelin, L.A. (1999). Stop-and-go movements of plant golgi stacks are
mediated by the acto-myosin system. Plant Physiol. 121, 1127-1142.
Wan, Y.L., Eisinger, W., Ehrhardt, D., Kubitscheck, U., Baluska, F., and Briggs, W.
(2008). The subcellular localization and blue-light-induced movement of phototropin 1-
GFP in etiolated seedlings of Arabidopsis thaliana. MOl. Plant 1, 103-117.
Fu, Y., Li, H., and Yang, Z. (2002). The ROP2 GTPase controls the formation of cortical
fine F-actin and the early phase of directional cell expansion during Arabidopsis
organogenesis. Plant Cell 14, 777-794.
Yang, Y.-D., Elamawi, R., Bubeck, J., Pepperkok, R., Ritzenthaler, C., and Robinson,
D.G. (2005). Dynamics of COPII vesicles and Golgi apparatus in cultured Nicotiana
tabacum BY-2 cells provides evidence for transient association of Golgi stacks with
endoplasmic reticulum exit sites. Plant Cell 17, 1513-1531.
20 Download full-text
8. Bar-Peled, M., and Raikhel, N. (1997). Characterization of AtSEC12 and AtSAR1:
proteins likely involved in endoplasmic reticulum and golgi transport. Plant Physiol. 114,
Young, R., McFarlane, H.E., Hahn, M., Western, T., Haugnn, G., and Samuels, A.L.
(2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized
secretion of pectin-rich mucilage. Plant Cell 20, 1623-1638.
Brandizzi, F., Snapp, E.L., Roberts, T.H., Lippincott-Schwartz, J., and Hawes, C. (2002).
Membrane protein transport between the endoplasmic reticulum and the Golgi in tobacco
leaves is energy dependent but cytoskeleton independent: Evidence from selective
photobleaching. Plant Cell 14, 1293-1309.