Two Leucine-Rich Repeat Receptor Kinases Mediate Signaling,
Linking Cell Wall Biosynthesis and ACC Synthase
Shou-Ling Xu,aAbidur Rahman,b,1Tobias I. Baskin,band Joseph J. Kiebera,2
aDepartment of Biology, University of North Carolina, Chapel Hill, North Carolina 27599-3280
bDepartment of Biology, University of Massachusetts, Amherst, Massachusetts 01003
The plant cell wall is a dynamic structure that changes in response to developmental and environmental cues through poorly
understood signaling pathways. We identified two Leu-rich repeat receptor-like kinases in Arabidopsis thaliana that play a
role in regulating cell wall function. Mutations in these FEI1 and FEI2 genes (named for the Chinese word for fat) disrupt
anisotropic expansion and the synthesis of cell wall polymers and act additively with inhibitors or mutations disrupting
cellulose biosynthesis. While FEI1 is an active protein kinase, a kinase-inactive version of FEI1 was able to fully complement
the fei1 fei2 mutant. The expansion defect in fei1 fei2 roots was suppressed by inhibition of 1-aminocyclopropane-1-
carboxylic acid (ACC) synthase, an enzyme that converts Ado-Met to ACC in ethylene biosynthesis, but not by disruption of
the ethylene response pathway. Furthermore, the FEI proteins interact directly with ACC synthase. These results suggest
that the FEI proteins define a novel signaling pathway that regulates cell wall function, likely via an ACC-mediated signal.
The regulation of cell expansion plays a fundamental role in plant
growthand development. Despite thiscritical role,the regulatory
inputs that control this process are poorly understood. Cell
expansion is regulated primarily by turgor pressure and by the
properties of the plant cell wall, which is composed of a poly-
saccharide network of cellulose microfibrils cross-linked by
hemicelluloses in a pectin matrix, along with numerous proteins
(Somerville, 2006). The primary load-bearing elements of the cell
linking are key factors that determine both the direction and
extent of cell expansion (Darley et al., 2001). In longitudinally
in an orientation perpendicular to the axis of expansion, thus
constricting radial expansion (Green, 1980; Taiz, 1984; Baskin,
2005). Consistent with this, disruption of cellulose biosynthesis
by treatment with various chemical inhibitors results in a rapid
loss of growth anisotropy (Scheible et al., 2001; Desprez et al.,
Cellulose microfibrils are synthesized by cellulose synthase,
an enzyme that is present at the plasma membrane as a hexa-
meric protein complex called the rosette (reviewed in Somerville,
2006). Genetic analysis and inhibitor studies indicate that cyto-
plasmic microtubules play an important role in guiding the
orientation of the deposition of cellulose microfibrils (reviewed
in Baskin, 2001), and the cellulose synthase rosette was found to
move along the plasma membrane in tracks that largely coin-
cided with the cortical microtubules (Paredez et al., 2006).
Additional components involved in regulating cell wall biosyn-
thesis have been identified in genetic screens for mutations that
alter root or hypocotyl elongation in Arabidopsis thaliana. The
cobra (cob) mutant displays radial expansion in the elongation
zone of the root, and this is correlated to a disorganization of
microfibrils and a reduction in the level of crystalline cellulose in
cells in the root elongation zone. COB encodes a putative
glycosylphosphatidylinositol (GPI)-anchored extracellular pro-
tein that is localized to the longitudinal sides of root cells in a
banding pattern transverse to the longitudinal axis (Schindelman
et al., 2001). The sos5 mutant is a conditional mutant that
displays arrested root growth and a swollen root phenotype in
the presence of salt stress (Shi et al., 2003). SOS5 encodes a
GPI-anchored extracellular protein with two arabinogalactan
to play a role in cell adhesion.
Several members of the receptor-like Ser/Thr protein kinase
(RLK) family in Arabidopsis have been implicated in regulating
cell growth in different contexts (He ´maty and Ho ¨fte, 2008). The
RLKs are a large, diverse family of transmembrane signaling
elements in plants, only a few of which have been functionally
characterized (Morillo and Tax, 2006). The Arabidopsis protein
THE1, which belongs to the Cr RLK1L (for Catharanthus roseus
protein kinase1–like) subfamily, has been hypothesized to sense
cell wall integrity (He ´maty et al., 2007). A second group of RLKs,
the WAKs, are tightly bound to the cell wall and likely play an
important rolein regulating itsfunction (He etal.,1996; Anderson
et al., 2001). Here, we describe two Arabidopsis leucine-rich
1Current address: Cryobiofrontier Research Center, Iwate University,
Iwate 020-8550, Japan.
2Address correspondence to firstname.lastname@example.org.
The author responsible for distribution of materials integral to the
findings presented in this article in accordance with the policy described
in the Instructions for Authors (www.plantcell.org) is: Joseph J. Kieber
WOnline version contains Web-only data.
The Plant Cell, Vol. 20: 3065–3079, November 2008, www.plantcell.org ã 2008 American Society of Plant Biologists
repeat (LRR) RLKs in a distinct RLK clade whose disruption
links 1-aminocyclopropane-1-carboxylic acid (ACC) synthase
(ACS) to this pathway, as well as SOS5, which together define a
novel pathway regulating cell wall biosynthesis.
Disruption of FEI1 and FEI2 Alters Cell Expansion
The Arabidopsis genome encodes >200 predicted LRR-RLKs,
most of which have unknown functions (Morillo and Tax, 2006).
We identified two highly similar LRR-RLKs (82% amino acid
identity) (Figure 1A; see Supplemental Figure 1 online) that when
both disrupted caused a swollen-root phenotype (Figures 1 and
2). We named these kinases FEI, after the Chinese word for fat.
FEI1 (At1g31420) and FEI2 (At2g35620) are in the same RLK
subfamily XIII as ERECTA (Shiu and Bleecker, 2001), which is
distinct from the THE1 and WAK subfamilies. The insertions in
fei1, fei2-1, and fei2-2 (Figure 1B) result in the elimination of the
corresponding full-length transcript (Figure 1E). In the case of
fei1, there is a truncated transcript present in the mutant plants
corresponding to the region of the gene upstream of the T-DNA
insertion site (Figure 1E). The single fei1 and fei2 mutants were
indistinguishable from the wild type in all aspects of growth and
development (Figure 1). The double fei1 fei2 mutant was nearly
(Figures 1C and 1F), but in the presence of 4.5% (high) sucrose,
(Figures 1D, 1F, and 2). Root elongation was reduced in the fei1
fei2 mutant 2 d after transfer compared with wild-type seedlings
(Figure 1G), and swelling was visible 3 d after transfer (see
Supplemental Figure 2 online). Four days after transfer to non-
permissive conditions, the diameter of the mutant root was
greater than twofold larger compared with the wild type (wild
type, 163 6 11 mm, n= 8;fei1 fei2, 3166 68 mm, n =8). The F2 of
a cross between fei1/fei1 and fei2/fei2 segregated seedlings
displaying the mutant phenotype in a ratio consistent with two
recessive loci (653 wild type, 39 swollen roots, x2= 0.45 for the
C-myc epitope tag (FEI1-myc or FEI2-myc) was able to fully
complement the root-swelling phenotype of fei1 fei2 (Figures 3B
and 3C), confirming that the phenotype was the result of disrup-
tion of these genes.
Wild-type Arabidopsis root cellsundergo primarily longitudinal
expansion. The increased diameter and reduced elongation
observed in fei1 fei2 double mutant roots suggests that aniso-
this, examination of transverse sections of root apices revealed
inner layers, displayed a high degree of radial swelling (Figure 2).
The root cells of the single fei mutants appeared indistinguish-
able from wild-type rootcells (seeSupplemental Figure 3online).
The number of cells in each of the layers of the root was not
appreciably altered in the fei1 fei2 mutants (i.e., there were 23 to
26 [n = 5] epidermal cells in fei1 fei2 versus 20 to 27 [n = 5] for the
wild type). We conclude that the fei1 fei2 mutations cause the
cells in the elongation zone to undergo a shift in expansion from
longitudinal to isotropic. The fei1 fei2 mutants also displayed
swollen roots on medium that contains an elevated concentra-
tion of NaCl (see Supplemental Figure 4B online). However, fei1
fei2 roots do not swell in the presence of 1 to 6% mannitol or
sorbitol (see Supplemental Figure 5 online), indicating that the
effect of sucrose and NaCl was not the result of a response to
Intrinsic Kinase Activity Is Not Required for FEI Function
The sequences of the C-terminal domains of FEI1 and FEI2 have
all of the features of a Ser/Thr protein kinase catalytic domain,
including all 11 conserved subdomains of eukaryotic protein
kinases (see Supplemental Figure 1 online) (Hanks and Quinn,
1991). To test if the FEI1 kinase has intrinsic protein kinase
activity, we expressed the kinase domain of FEI1 in Escherichia
coli as a glutathione S-transferase (GST) fusion protein. Purified
recombinant FEI1 was active in in vitro protein kinase assays; it
was able to autophosphorylate and to phosphorylate myelin
loss of kinase activity (Figure 3A), as has been observed in other
protein kinases. Interestingly, thiskinase-inactive version of FEI1
(or FEI2) was able to complement a fei1 fei2 mutant (Figures 3B
and 3C), although complementation was not as consistent as
that observed with the wild-type FEI1 or FEI2 gene: 10 of 10
independent transformants displayed full complementation
when transformed with wild-type FEI1 or FEI2, whereas 3 of 10
and2of10independent transformants werefullycomplemented
with the respective mutant versions. This indicates that kinase
activity is not essential for FEI function in vivo, although it con-
tributes to optimal FEI function.
FEI Is Localized to the Plasma Membrane and Is
Analysis of the deduced amino acid sequence of FEI1 and FEI2
predicts a single transmembrane domain, similar to other RLKs.
Consistent with this, both FEI1-myc and FEI2-myc fusion pro-
teins were present in a microsome fraction (Figure 4I). Further-
was able to complement the fei1 fei2 mutant (see Supplemental
Figure 6 online), localized to the periphery of the cell in a pattern
consistent with a plasma membrane localization (Figure 4J).
Both FEI1 and FEI2 are most highly expressed in the root
meristem and elongation zone, as revealed by promoter–b-
glucuronidase (GUS) fusions (Figure 4). Published microarray
analysis revealed that FEI1 and FEI2 are expressed at approx-
imately equal levels in the different radial layers of the root tip,
including the epidermis (Birnbaum etal., 2003).Extendedstaining
of FEI promoter–GUS lines revealed a broader staining pattern
for these two genes (Figure 4), similar to the pattern obtained
from publicly available array data (Zimmermann et al., 2005).
FEI1 and FEI2 Function in Hypocotyls and Flowers
FEI1 and FEI2 both are expressed in the hypocotyls of etiolated
3066The Plant Cell
Figure 1. fei1 fei2 Mutants Display Conditional Root Anisotropic Growth Defects.
(A) Structures of the predicted FEI and ERECTA proteins. The percentage identities between the kinase or LRR N-terminal domains of FEI1 and FEI2 or
FEI2 and ERECTA are indicated. aa, amino acids.
(B) Cartoon of fei1, fei2-1, and fei2-2 alleles. Boxes represent exons (blue area represents the kinase domain), and triangles indicate the positions of T-
(C) and (D) Phenotypes of indicated seedlings grown on MS medium plus 1% sucrose (C) or plus 4.5% sucrose (D) for 9 d. Bars = 1 cm in top panels
and 1 mm in bottom panels.
59 to the site of the T-DNA insertion (see Supplemental Table 1 online) were used in PCR for 30 cycles from cDNA derived from the indicated line.
(F) Quantification of root growth after transfer to permissive or nonpermissive conditions. The indicated seedlings were grown on MS medium
containing 0% sucrose for 4 d and then transferred to MS medium containing either 0% or 4.5% sucrose as indicated. Root growth from the time of
transfer until day 9 is indicated on the y axis. Error bars show SE (n > 30).
(G) Kinetics of root elongation of wild-type and fei1 fei2 mutant seedlings. Wild-type or fei1 fei2 mutant seedlings were grown on MS medium containing
0% sucrose for 4 d and then transferred to MS medium containing either 0% or 4.5% sucrose as indicated. Root lengths were measured each day after
transfer, and the amount of root growth that occurred each day after transfer was then calculated. Error bars show SE (n > 15).
(H) Phloroglucinol staining for lignin (red) of seedlings grown on MS medium containing 0% sucrose for 4 d and then transferred to MS medium
containing 4.5% sucrose for 5 d. Bar = 0.5 cm.
The FEI LRR-RLKs Regulate Cell Expansion3067
cells that undergo primarily longitudinal expansion. Thus, we
examined if the fei1 fei2 double mutant had defects in hypocotyl
growth. The hypocotyls of etiolated seedlings of fei1 fei2 were
(Figures 4K to 4M). However, contrary to the root phenotype of
fei1 fei2, this was not accompanied by a decrease in the over-
all length of the hypocotyl (see Supplemental Figure 7 online),
and this occurred in either low or high sucrose. This swollen-
hypocotyl phenotype was complemented by transgenes contain-
ing genomic copies of either FEI1 or FEI2. The modest increase
substantially less than the increased width observed in the
mutant roots. Examination of transverse sections of wild-type
and mutant etiolated seedlings revealed that the increased
diameter of the fei1 fei2 hypocotyls was associated with an
increase in cell size, not cell number (Figure 4L). We did not
observe a significant change from the wild type in the level or
spatial distribution of lignin in the fei1 fei2 etiolated hypocotyls,
as revealed by phloroglucinol staining (see Supplemental Figure
8 online). There was no obvious swelling in any other tissues of
the fei1 fei2 mutant. However, the fei1 fei2 cob triple mutant (see
below), but neither fei1 fei2 nor cob, had shortened stamen
filaments, and this triple mutant was partially infertile (Figure 4N),
indicating a role for the FEI kinases in these tissues. Consistent
with this, analysis of the promoter-GUS fusions reveals expres-
The fei1 fei2 Mutant Is Defective in Cellulose Biosynthesis
The altered pattern of cell expansion in the fei1 fei2 mutants
suggests a defect in cell wall function. As cortical microtubules
have been implicated in regulating anisotropic growth, we ex-
amined theirarrangement in epidermal cells ofwild-type and fei1
fei2 roots using an anti-a-tubulin antibody. In both wild-type and
fei1 fei2 double mutant root cells, the microtubules in the
elongation zone were aligned primarily transversely to the axis
of growth 3 d after transfer to nonpermissive conditions (see
Supplemental Figure 9 online). This indicates that growth an-
isotropy in the fei1 fei2 mutants is not the result of disruption of
the pattern of microtubules.
the mutant, we examined the effect of isoxaben, an inhibitor of
cellulose synthase, on fei1 fei2. Growth in the presence of high
sucrose rendered wild-type roots hypersensitive to isoxaben
(Figure 5A), which indicates that elevated sucrose sensitizes
roots to perturbations in cellulose synthesis. In the presence of
low sucrose, both the prc1 mutant, which disrupts a catalytic
subunit of cellulose synthase (CESA6) (Fagard et al., 2000), and
fei1 fei2 displayed increased sensitivity to isoxaben (Figure 5C).
This suggests that fei1 fei2 perturbs the biosynthesis or function
of cellulose. Consistent with this, the roots of fei1 fei2 seedlings
grown in nonpermissive conditions produced ectopic lignin
(Figure 1H), which is generally correlated with a decreased level
of crystalline cellulose (Humphrey et al., 2007). We further
analyzed cellulose synthesis by measuring the incorporation of
fractions of excised root tips. In permissive conditions, fei1 fei2
roots were similar to wild-type roots. However, in nonpermissive
conditions, fei1 fei2 mutant roots displayed a striking defect in
cellulose biosynthesis, as measured by the incorporation of
Peng et al., 2000) and acid-soluble material (noncrystalline
cellulose and other wall polymers; Heim et al., 1998) (Figure 5D).
When viewed with a transmission electron microscope, the
walls from the swollen root cells of the fei1 fei2 mutant were not
appreciably altered in thickness compared with wild-type cell
walls. However, the size of the intercellular spaces in the outer
cell layer layers of the fei1 fei2 mutant roots was reduced in
Figure 2. Analysis of Wild-Type and fei1 fei2 Mutant Roots 4 d after
Transfer from Medium Containing 0% Sucrose to Medium Containing
(A) and (B) Cleared whole mount of wild-type (A) and fei1 fei2 (B) roots
viewed with Nomarski optics. Note that abnormal lateral expansion in the
mutant root is most apparent in the epidermis.
(C) and (D) Transverse sections through the meristems of wild-type (C)
and fei1 fei2 (D) mutant roots.
(E) and (F) Transverse sections through the elongation zones of wild-type
(E) and fei1 fei2 (F) mutant roots. Bar = 100 mm.
3068 The Plant Cell
nonpermissive conditions (Figure 5B), similar to that in the sos5
and rsw1-20 mutants (Beeckman et al., 2002; Shi et al., 2003).
of unknown function. Null cob mutants are extremely deficient in
cellulose, are strongly dwarfed, and are sterile (Roudier et al.,
in this study, result in fertile plants that displayed a sucrose-
dependent swollen-root phenotype (Figure 6). prc1-1, which is a
likely null allele of CESA6, also displayed a sucrose-dependent
swollen-root phenotype (Figure 6). We examined the genetic
interactions of fei1 fei2 with cob and prc1. The fei1 fei2 cob and
fei1 fei2 prc1 triple mutants displayed an enhanced root pheno-
type compared with the parental lines; the triple mutant roots
were significantly shorter and more swollen in nonpermissive
conditions (Figures 6B and 6C). Moreover, the fei1 fei2 cob and
fei1 fei2 prc1 triple mutants displayed swollen roots even in
permissive conditions, in which the single or double mutants did
not display significant swelling (Figures 6A and 6C). These
synergistic interactions suggest that FEI1 and FEI2 act in a
pathway independent from COB or PRC1 to regulate cell wall
sos5 was isolated as a mutant that displayed a swollen root tip
in the presence of moderately high salt (Shi et al., 2003). The
AGP-like and fasciclin-like domains. As the phenotype of sos5 is
similar to that of the fei1 fei2 double mutant, we tested the effect
of high sucrose on sos5 seedlings. Similar to fei1 fei2, growth of
sos5-2 (a novel T-DNA insertion allele that is a null transcript; see
Supplemental Figure 4 online) in the presence of high sucrose
also resulted in a swollen-root phenotype (Figure 6B). By con-
trast, we did not observe a swollen-root phenotype in other sos
mutants (sos1, sos2, sos3, and sos4) in response to elevated
sucrose (see Supplemental Figure 10 online). Furthermore, eti-
olated sos5-2 seedlings displayed swollen hypocotyls similar to
were indistinguishable from those of the fei1 fei2 double mutant
in their response to increasing levels of NaCl (Figure 6). Likewise,
the hypocotyl width of the sos5-2 fei1 fei2 triple mutant etiolated
seedlings was comparable to that of the fei1 fei2 double mutant
(Figure 4M). The nonadditive nature of sos5-2 and fei1 fei2
suggests that these gene products act in a linear pathway to
regulate cell elongation.
Figure 3. Intrinsic Kinase Activity Is Not Required for FEI Function.
(A) Kinase activity of FEI1. Wild-type and FEI1K334Rproteins were expressed in E. coli as GST fusion proteins, purified by glutathione-affinity
chromatography, and then subjected to an in vitro kinase assay and analyzed by SDS-PAGE. Purified GST was included as a control, and myelin basic
protein (MBP) was the substrate. Top, staining of the gel with Coomassie blue; bottom, autoradiograph of the gel. The positions of molecular mass
markers are shown at right.
(B) Complementation of the fei1 fei2 mutant phenotype by introduction of a wild-type (gFEI1 or gFEI1) or kinase-inactive (gFEI1K334Ror gFEI2K332R)
version of FEI1 or FEI2. Two independent lines (a and b) are shown for each. Seedlings were grown for 4 d on MS medium containing 0% sucrose and
then transferred for 4 d to MS medium containing 4.5% sucrose, and representative seedlings were photographed.
(C) Quantification of root elongation from (B). The mean (n > 15) 6 SE of seedling growth from days 4 to 8 is shown.
The FEI LRR-RLKs Regulate Cell Expansion3069
ACS Plays a Role in FEI1/FEI2-Mediated Cell Expansion
Ethylene plays an important role in regulating expansion in many
plant cells, and inhibition of ethylene biosynthesis or perception
can partially revert the swollen phenotypes of certain root mor-
phology mutants, such as sabre (Aeschbacher et al., 1995) and
cev1 (Ellis et al., 2002). We determined the effect of blocking
ethylene biosynthesis on the fei1 fei2 swollen-root phenotype.
that blocks ACC oxidase activity by acting as a competitive
inhibitor, reverted fei1 fei2 mutant roots grown in the presence of
high sucrose or elevated NaCl to a nearly wild-type morphology
(Figures 7Aand 7B, Table 1;seeSupplemental Figure 4B online).
AIB also reverted the defect in cellulose synthesis in fei1 fei2
of fei1 fei2. Aminooxyacetic acid (AOA), which inhibits enzymes
that require pyroxidal phosphate, including ACS, reverted the
fei1 fei2 swollen-root phenotype (Figures 7A and 7B, Table 1). As
it is unlikely that this phenotypic reversion of fei1 fei2 is due to
Figure 4. FEI1 and FEI2 Expression, Localization, and Function in Hypocotyls and Flowers.
(A) to (H) Staining (blue) of transgenic lines harboring the promoter of FEI1 ([A] to [D]) or FEI2 ([E] to [H]) fused to GUS. (A), (C), (E), and (G) are from
seedlings grown on MS media for 7 d. (B) and (F) show 3-d-old etiolated seedlings. (D) are (H) are flowers from plants grown in soil under long days for 3
weeks. Bars in (A) and (E) = 100 mm.
(I) Root tissue from plants expressing FEI1-myc or FEI2-myc was fractionated into soluble and microsome fractions. The total (T), soluble (S), and
microsome (P) fractions were subjected to protein gel blotting and probed with an anti-C-myc (top) or anti-Hsc70 (bottom) antibody.
(J) Localization of FEI2-GFP fusion proteins. Top, Differential interference contrast (DIC) and GFP images of root cells from MS-grown seedlings;
bottom, images from seedlings plasmolyzed in 0.8 M mannitol. Red arrows indicate regions of membrane that have detached from the cell wall.
(K) Images of hypocotyls from wild-type (left) and fei1 fei2 mutant (right) 3-day-old etiolated seedlings. Note that the fei1 fei2 hypocotyls are thicker.
(L) Transverse sections through hypocotyls of wild-type and fei1 fei2 mutant etiolated 3-day-old seedlings. Bar = 50 mm.
(M) Quantification of hypocotyl widths from etiolated seedlings. Asterisks indicate significant differences from the wild type (Student’s t test, P < 0.05,
n = 20). Error bars show SE (n = 20).
(N) Stage 12 flowers from the indicated genotypes. Several petals and sepals were removed from each flower to reveal the inner parts. Note that the fei1
fei2 cob triple mutant has shorter stamen filaments. Bar = 1 mm.
3070The Plant Cell
Figure 5. The fei Mutants Affect Cell Wall Function.
(A) High sucrose or NaCl enhance the effect of isoxaben. Wild-type seedlings were grown on MS medium containing 0% sucrose for 4 d and then
transferred to MS medium plus the indicated supplement in the presence of 0 (control) or 1 nM isoxaben as indicated. At 24 h after transfer, root tips
were imaged. Bar = 1 mm.
(B) Transmission electron microscopy of cell wall junctions from wild-type and fei1 fei2 mutant epidermal root cells from seedlings grown on 4.5%
sucrose. Bar = 1 mm.
(C) Response of the indicated seedlings to isoxaben. Seedlings of the indicated genotypes were grown on MS medium containing 0% sucrose for 4 d
and then transferred to medium containing 1% sucrose and the indicated level of isoxaben. At 24 h after transfer, root tips were imaged. Bar = 1 mm.
(D) Incorporation of [14C]Glc into acid-soluble or acid-insoluble fractions from excised root tips from wild-type and fei mutant seedlings grown in 0%
sucrose for 4 d and then transferred to 0% or 4.5% sucrose as indicated for 3 d. Asterisks indicate statistical differences between fei1 fei2 and the
respective wild-type sample (Student’s t test, P < 0.05). Values are means 6 SE from three biological replicates, and the experiments were repeated at
least three times with similar results.
(E) Incorporation of [14C]Glc into acid-soluble or acid-insoluble fractions from excised roots from wild-type and fei mutant seedlings that were grown in
0% sucrose for 4 d and then transferred to 4.5% sucrose in the absence (control) or presence of AIB (1 mM) as indicated for 3 d. Asterisks indicate
statistical differences between fei1 fei2 and the respective wild-type sample (Student’s t test, P < 0.05). Values are means 6 SE from three biological
replicates, and the experiments were repeated at least three times with similar results.
off-target effects. Furthermore, this is not a general effect of AIB,
as it did not revert the root-swelling phenotype of the cob mutant
(see Supplemental Figure 11 online), even at higher concentra-
(1-MCP) nor silver ion (silver thiosulfate), both of which block
ethylene perception, had any appreciable effect on the root
phenotype of fei1 fei2 mutants (Table 1). Likewise, neither etr1,
which disrupts an ethylene receptor, nor ein2, a strong ethylene-
insensitive mutant that acts downstream of ETR1, suppressed
the fei1 fei2 root phenotype (Figure 7A, Table 1).
Consistent with the other similarities to the fei1 fei2 mutant,
root swelling in sos5-2 seedlings grown in the presence of either
high sucrose or elevated NaCl was reversed by AIB and AOA but
not by blocking the response to ethylene (Figures 7A and 7B,
Table 1; see Supplemental Figure 4 online). This suggests either
that swelling in the absence of FEI depends on a hitherto
undiscovered pathway for ethylene perception or that ACC itself
acts as a signaling molecule.
We tested if the FEIs interacted with ACS using a yeast two-
hybrid assay. The kinase domains of both FEI1 and FEI2 in-
teracted with both ACS5 and ACS9, two type 2 ACS proteins
(Chae and Kieber, 2005). By contrast, neither FEI1 nor FEI2
interacted with ACS2 (Figure 7C), which belongs to a distinct
subclade of ACS proteins (type 1) that have divergent C-terminal
domains (Chae and Kieber, 2005). Likewise, the eto2 and eto3
mutations, which alter the C-terminal domains of ACS5 and
ACS9, respectively, and which block the rapid degradation of
these proteins in vivo, disrupted the interaction with FEI1 and
FEI2 in the yeast two-hybrid interaction (Figure 7C). Disruption of
thekinase activitydidnotaffecttheinteraction withACS, asboth
FEI1K334Rand FEI2K332Rinteracted with ACS5. By contrast, the
ERECTA kinase domain did not interact with ACS5 in a yeast
two-hybrid assay, indicating that there is specificity in the inter-
action with ACS5. We failed to detect an interaction between
We were not able to detect phosphorylation of ACS5 in vitro by
in this analysis was enzymatically active and could be phosphor-
ylated in our conditions by a partially purified soybean (Glycine
been shown previously to phosphorylate ACS (Tatsuki and Mori,
2001; Sebastia ` et al., 2004). Thus, the lack of phosphorylation of
Figure 6. Genetic Interaction of fei1 fei2 with Other Mutants Affecting Cell Elongation.
(A) and (B) Phenotypes of wild-type and various mutant seedlings grown in medium containing 0% sucrose for 4 d and then transferred to MS medium
containing no (A) or 4.5% (B) sucrose for 4 d. Bars = 1 cm (top panels) and 1 mm (bottom panels; close-ups of root tips).
(C) Quantification of root elongation of various mutants grown and transferred as in (A) and (B). Values represent means of growth at 4 d after transfer to
the indicated conditions. Error bars show SE (n > 15).
(D) Quantification of total root elongation of the indicated lines at 4 d after transfer from MS medium containing 1% sucrose to the same medium with
various levels of NaCl added. Error bars show SE (n > 15).
3072 The Plant Cell
Figure 7. Role of ACC/Ethylene on the fei Phenotype.
(A) Phenotypes of seedlings grown on MS medium containing 0% sucrose for 4 d and then transferred to MS medium containing 4.5% sucrose plus
nothing, AOA (0.375 mM), or AIB (1 mM), as indicated. Bar = 1cm. Note that the distribution of lateral rootsin thefei1 fei2 mutants in the presence of high
sucrose is variable; the architecture of the fei1 fei2 ein2 triple mutant is not substantially different from that of the fei1 fei2 parent.
(B) Close-ups of root tips from (A). Bar = 1 mm.
(C) Yeast two-hybrid interactions among the FEIs and ACSs. Bait and prey vectors containing the soluble kinase domains of the wild type or mutant FEI1
and FEI2 were cloned into a yeast two-hybrid bait vector and cotransformed into yeast with the indicated wild-type and eto mutant ACS preys. Positive
interactions result in Leu prototrophy (growth on –Leu). The soluble kinase domains of ERECTA empty bait (pEG202) and prey (pJG4-5) vectors were
used as controls.
(D) FEI1 does not phosphorylate ACS5 in vitro. Top, Coomassie blue–stained gel of purified GST-FEI and/or ACS5 protein; bottom, autoradiograph
following an in vitro kinase assay. The arrows indicate the position of ACS5.
Measurements of ethylene production revealed that root tis-
sues from wild-type and fei1 fei2 mutant seedlings grown on low
or high sucrose in the light made comparable amounts of
ethylene (8.9 6 0.8 pL·cm21root segment21·d21for the wild
type versus 11.9 6 0.2 pL·cm21root segment·d21for fei1 fei2).
Likewise, ethylene production in dark-grown fei1 fei2 seedlings
was similar to that in the wild type (5.6 6 0.3 pL·seedling21·d21
for wild-type seedlings versus 5.8 6 0.7 pL·seedling21·d21for
catalytic activity of ACS.
FEIs Are Required for Anisotropic Growth in the Root
We show that the FEI1 and FEI2 LRR-RLKs are necessary for
anisotropic cell expansion in Arabidopsis root cells and also play
a role in cell expansion in stamen filamentsand the hypocotyls of
etiolated seedlings. Biochemical studies and genetic analyses
with other cellulose-deficient mutants reveal that these FEI
kinases modulate cell wall function, including positively regulat-
anisotropic expansion. Two other divergent RLKs have been
implicated in cell wall function: the WAK and THE1 kinases. The
WAK kinases are involved in cell expansion in various Arabidop-
cell wall (Anderson et al., 2001; Wagner and Kohorn, 2001).
Interestingly, wak2 mutants display reduced cell expansion that
is sensitive to the level of sugar and salt in the medium (Kohorn
et al., 2006). However, in contrast with fei1 fei2, high sugar levels
suppress the cell expansion defect in wak2, and it is the extent,
has been hypothesized to be involved in monitoring the cell wall
integrity, as the1 mutations suppress the short-hypocotyl phe-
notype, but not the cellulose-deficient phenotype, of prc1
(He ´maty et al., 2007). This is distinct from the FEIs, as the fei1
fei2 double mutant significantly impairs cellulose biosynthesis.
The the1 mutation also suppresses some, but not all, other
mutants affecting cell expansion. Alteration of THE1 function
does not have an effect in the wild-type background, suggesting
perhaps genetic redundancy or that it plays a role only in
conditions in which cell wall integrity is compromised. While it
would be interesting to determine the interaction between the1
and fei1 fei2, the lack of suppression of the root-elongation
phenotype of prc1 by the1 may render this genetic interaction
turn provide feedback to the cellulose biosynthesis machinery.
SOS5 encodes a putative cell surface adhesion protein that is
required for normal cell expansion (Shi et al., 2003). Several lines
of evidence suggest that SOS5 functions in a linear pathway with
the FEIs: (1) sos5 mutants have a very similar root-elongation
salt; (2) the root-swelling phenotypes of both fei1 fei2 and sos5-2
are suppressed by AIB and AOA but not by blocking the known
thickened-hypocotyl phenotype; (4) the fei1 fei2 and sos5-2
mutations show a nonadditive genetic interaction; and (5) the
patterns of expression of the FEIs and SOS5 are largely over-
lapping (Figure 4) (Shi et al., 2003). Thus, SOS5 acts on the same
pathway as the FEIs to mediate the function of the cell wall. As
SOS5 encodes an extracellular protein, it is possible that it acts
In addition to fei1 fei2 and sos5-2, the cob and prc1 mutants
also display root swelling that is dependent on the concentration
conditional phenotype reflects defects that are apparent only at
high rates of cell elongation, such as in the presence of sucrose
(Benfey et al., 1993). However, our data do not support this
hypothesis, as increasing sucrose above 1% actually led to a
conditions, but the root-swelling phenotype of both fei1 fei2 and
sos5-2 continued to intensify. Furthermore, low levels of NaCl,
which reduce the rate of root elongation, also caused swelling in
medium, as high levels of sorbitol or mannitol do not induce the
phenotype. Our results indicate that wild-type plants are more
susceptible to perturbation of cellulose biosynthesis in the pres-
ence of high sucrose or salt. How these conditions affect the
function of the cell wall remains to be determined.
Kinase Activity Is Dispensable for FEI Function
Consistent with their sequences, the FEIs have intrinsic kinase
activity; however, kinase activity is not essential for FEI function,
at least for the phenotypes that we observed. There are many
examples of so-called pseudokinases (reviewed in Kroiher et al.,
2001; Boudeau et al., 2006), which display clear homology to
kinases but lack conservation of one or more of the catalytic
residues in the kinase core. Pseudokinases are especially prev-
alent in plant genomes, and it has been estimated that;20% of
2007). For example, STRUBBELIG (SUB), a member of the
Table 1. Root Elongation in the Absence or Presence of
fei1 fei2 sos5-2 1.50 6 0.10 3.40 6 0.04 0.77 6 0.07 1.06 6 0.11
eto21.90 6 0.05 2.78 6 0.05 4.03 6 0.06 3.49 6 0.13
cob0.12 6 0.01 0.13 6 0.01 0.12 6 0.01 nd
etr1-3 fei1 fei21.59 6 0.09 nd
ein2 fei1 fei20.98 6 0.06 nd
etr1-34.82 6 0.05 nd
ein24.84 6 0.12 nd
4.63 6 0.07 3.67 6 0.05 4.42 6 0.05 4.03 6 0.12
1.62 6 0.09 3.76 6 0.05 0.99 6 0.07 1.12 6 0.08
2.26 6 0.12 3.68 6 0.04 1.16 6 0.09 1.35 6 0.11
Values represent means 6 SE of at least 15 root elongation measure-
ments (in centimeters) between days 4 and 9. nd, not determined.
3074 The Plant Cell
organs, includes two alterations in residues that are highly
conserved in functional kinases, and genetic and biochemical
analyses indicate that the SUB kinase domain is catalytically
inactive (Chevalier etal.,2005).TheArabidopsis homolog ofCR4
RLK encodes an active kinase, but disruption of the kinase
catalytic domain by site-directed mutagenesis does not disrupt
its function in vivo (Gifford et al., 2005), similar to what we
are then transphosphorylated by, a kinase-active member of the
same protein family. An alternative possibility it that FEI signaling
does not involve phosphorylation but, rather, the proteins act as
scaffolds to localize other components in aprotein complex orto
a particular place in the cell. An example of this is the human
Kinase Suppressor of Ras (KSR) protein, which is similar in
sequence to protein kinases but which acts as a scaffold protein
that coordinates the assembly of a multiprotein mitogen-acti-
vated protein kinase complex at the membrane (Claperon and
Therrien, 2007). In any case, the kinase activity of the FEIs, while
not essential, is clearly required for optimal function, as only a
catalytically inactive version of the FEIs were fully complemented.
As kinase activity is not required for function, it is possible that
truncated FEI1transcriptpresent. Thesimilarityinthestrength of
the phenotype of fei1 fei2 to sos5-2, a null allele in a gene acting
on the same pathway as the FEIs, argues somewhat against this.
Role of ACS5 in the SOS5/FEI Pathway
What role do ACS5 and other type 2 ACS enzymes play in
regulating cell wall function in the root? ACS5 has been shown to
be an enzymatically active ACS (Yamagami et al., 2003), the
product of which is ACC, the immediate precursor for ethylene.
Ethylene has been shown to play a role in regulating anisotropic
growth. In hypocotyls, ethylene inhibits elongation primarily by
altering the orientationof cellelongation, which iscorrelated with
a change in the orientation of the microtubules (Steen and
but radial expansion is only modestly increased and microtu-
bules appear to be unaffected (Baskin and Williamson, 1992).
Thus, in the root, ethylene appears to primarily inhibit the overall
amount of cell expansion, not its orientation. One potential
mechanism for this is the elevation of ROS levels in the elonga-
the cross-linking of Hyp-rich glycoproteins and callose deposi-
tion in the cell wall, both of which may contribute to reduced cell
expansion (De Cnodder et al., 2005).
There are several mutants that affect growth anisotropy in the
root that are linked to ethylene, including sabre, cev1, and lue1
(Aeschbacher et al., 1995; Ellis et al., 2002; Bouquin et al., 2003).
elevated levels of ethylene, and its phenotype is partially sup-
pressed by mutations that disrupt ethylene signaling (Ellis et al.,
2002). Similar to fei1 fei2, the swollen-root phenotype of the
sabre mutant can be partially rescued by blocking ethylene
action through the use of ethylene biosynthesis inhibitors, and
the sabre mutant does not display an increase in ethylene
biosynthesis (Aeschbacher et al., 1995). However, in contrast
with fei1 fei2, sabre also can be rescued by inhibition of ethylene
perception or by etr1. The authors propose that ethylene and
SABRE counteract each other to regulate the degree of radial
expansion of root cells. However, neither ethylene-overproduc-
ing mutants nor constitutive ethylene-signaling mutants have
such a dramatic swollen-root phenotype, which would be
predicted from such a model.
The interaction of type 2 ACSs with the FEIs and the reversion
of the fei1 fei2 mutant by inhibitors of ethylene biosynthesis
strongly suggest a link between ACS function and altered cell
wall function in fei mutant roots. However, several lines of
evidence indicate that this is not the result of altered ethylene
levels: (1) mutants that increase or decrease ethylene biosyn-
(2) the fei phenotype cannot be reversed by blocking ethylene
perception; and (3) in nonpermissive conditions, ethylene pro-
duction isnotsubstantially altered infei1fei2mutantroots.Thus,
we conclude that the FEIs do not alter ACS activity or levels and
that the FEIs do not act via ethylene. How, then, does ACS
function in the FEI pathway, and how do the FEIs affect ACS
One possibility is that the ACS protein may perform a function
if this were the case, it would not explain the reversion of fei1 fei2
by AIB, which is a structural analog of ACC that should not
directly affect ACS function. A second model is that perhaps fei1
fei2 alter ethylene biosynthesis in a small number of critical cells,
which may not be detectable in our analysis, and this elevated
ethylene may be perceived by a second, independent ethylene
response pathway that functions in this developmental context.
This model is possible, but two lines of evidence argue some-
what against it. First, it would not explain the lack of root swelling
in various ethylene biosynthesis mutants; second, it is probable
that, similar to ETR1 and its paralogs, any additional ethylene
receptor would be blocked by silver ion (Burg and Burg, 1967),
and thus silver should, but does not, revert the fei1 fei2 pheno-
type. A final model that is consistent with the data is that ACC
itself, rather than ethylene, acts as a signaling molecule to
regulate cell expansion in the FEI/SOS5 pathway. In such a
competitive inhibitor to block binding to a hypothetical ACC
receptor. Disruption of ethylene binding would not affect this
response, and there would be no alteration in ethylene levels in
the mutant. The data are most consistent with this model, in
What is the nature of the interaction of the FEI and ACS
proteins? The FEI proteins do not appear to phosphorylate
ACS5, which is consistent with the lack of requirement for kinase
activity for FEI function. Furthermore, ethylene levels are not
altered in fei1 fei2 mutants, suggesting that there is no change in
ACS levels or activity. One model consistent with the data is that
the FEIs actas ascaffold to localize afraction of ACS protein to a
subdomain of the plasma membrane and/or to assemble ACS
The FEI LRR-RLKs Regulate Cell Expansion3075
into a protein complex. This would be similar to KSR, a protein
kinase that acts as a scaffold in a mitogen-activated protein
kinase cascade. This localized ACS would then generate a
localized signal to regulate cell wall biosynthesis.
We propose that the FEI kinases play a role in regulating cell
wall architecture, possibly mediating interactions between the
cell wall and intracellular signaling pathways. The FEI RLKs may
act as a scaffold to localize ACS or may complex ACS with other
proteins. The extracellular SOS5 protein also feeds into this
pathway. Exactly how ACS functions in this pathway, and how
this pathway interacts with the biosynthetic machinery of the cell
wall and with other regulatory inputs into cell wall function, are
important questions for the future.
The Columbia (Col-0) ecotype of Arabidopsis thaliana was used in this
to position +2599 (relative to the translational start site). The fei2-1 inser-
tion was isolated by PCR screening (using primers FEI2-S5, FEI1-A5, and
T-DNA left border primer; see Supplemental Table 1 online) of a T-DNA
insertion library made in a Col-0 gl1 line (http://www.dartmouth.edu/
~tjack/et.html). The fei2-1 insertion was localized to position +2012. The
fei2-2 insertion (SALK_044226) (Alonso et al., 2003) was localized to
of PCR-amplified products using gene-specific and left border primers
double mutant line was used in all experiments unless noted otherwise.
(Alonso et al., 2003) alleles were obtained from the SALK T-DNA insertion
collection. The cob-1 and prc1-1 alleles used in this study were obtained
(Chang et al., 1993) mutants have been described previously.
Growth Conditions and Measurements
growth in vitro, seeds were surface-sterilized and cold-treated at 48C for
3 d in the dark and then treated with white light for 3 h. Seedlings were
grown on vertical plates containing 13 Murashige and Skoog (MS) salts,
1% sucrose, and 0.6% phytagel (Sigma-Aldrich) at 228C in ;100 mE
constant light. For measurements of root elongation, seedlings were
grown for 4 d on vertical plates containing no sucrose or in some cases
1% sucrose, as noted in the figure legends, and then transferred to MS
medium supplemented with the indicated additions. For the ethylene
inhibitor studies, AIB (1 mM), AOA (0.375 mM), and MCP (20 mg
Ethylbloc; Floralife) were added to a 6-liter container or silver thiosulfate
(0.02 mM) was added to the high-sucrose MS agar.
Total RNA was isolated from 7-d-old seedlings using the RNeasy kit
(Qiagen). First-strand cDNA was synthesized from 1 mg of the total RNA
pretreated with RNase-free DNase (Promega) using the SuperScript II kit
(Invitrogen) with random hexamers, according to the manufacturer’s
instructions. Quantitative RT-PCR was performed with SYBR Premix Ex-
specific primers (see Supplemental Table 1 online).
FEI Constructs and Transgenic Plants
Genomic fragments comprising the entire coding region of FEI1 or FEI2
and 1 kb ofthe respective 59 flanking DNA were amplified from BAC T8E3
or T20F21 DNA, respectively, by PCR (primers FEI1-S7/FEI1-A3 and
FEI2-S7/FEI2-A4; see Supplemental Table 1 online) using Pfu DNA
polymerase as described by the manufacturer (Stratagene), and the
fragments were cloned into pENTR-TOPO-D (Invitrogen). The resultant
entry plasmid was used in an LR reaction (as described by the manufac-
turer; Invitrogen) to introduce the respective genes into the binary
domain of FEI1 were amplified from cDNA by RT-PCR using first-strand
cDNA generated from wild-type Col RNA and gene-specific primers
(FEI1-C2/FEI1-A5; see Supplemental Table 1 online). Kinase-deficient
versions of FEI1 or FEI2 were obtained by site-directed mutagenesis
using primers containing the desired point mutation (FEI1-M2F/FEI1-
M2R and FEI2-M2F/FEI2-M2R; see Supplemental Table 1 online). For
expression of a GFP fusion protein, a FEI2 genomic fragment (amplified
using primers FEI2-S/FEI2-A4; see Supplemental Table 1 online) was
cloned into pENTR-TOPO-D (Invitrogen) and then introduced into the
binary vector pGWB5 (Nakagawa et al., 2007). For promoter-GUS fu-
sions, genomic fragments comprising 3 kb of 59 flanking DNA of FEI1 or
FEI2 were amplified from wild-type genomic DNA (using primers FEI1-
PROM-F1/FEI1-PROM-R1 and FEI2-PROM-F2/FEI2-PROM-R2; see
Supplemental Table 1 online), cloned into pENTR4 vector, and then
introduced into the binary vector pGWB2 (Nakagawa et al., 2007). All
clones were confirmed by DNA sequencing. The resulting plasmids were
transformed into Agrobacterium tumefaciens strain GV3101. Transgenic
plants were generated by the floral dip method (Clough and Bent, 1998)
and selected on MS medium containing 50 mg/L kanamycinand 30 mg/L
hygromycin. All destination binary vectors were kindly provided by
Tsuyoshi Nakagawa from the Research Institute of Molecular Genetics
in Matsue, Japan.
Protein Kinase Assays
The FEI1 and FEI1K334Rkinase domains in pENTR-TOPO-D (see above)
were introduced into the plasmid pDEST15 by Gateway cloning (Invitro-
gen). The respective GST fusion proteins were isolated using Glutathione
Sepharose 4 Fast Flow medium according to the manufacturer’s direc-
tions (Amersham Biosciences). ACS5 was purified as described (Chae
in vitro kinase assays were performed in kinase reaction buffer (50 mM
Tris-HCl, pH 7.5, 10 mM MgCl2, 10 mM MnCl2, 10 mM DTT, 10 mM ATP,
and 5 mCi of [g-32P]ATP [2 mCi/mL; Perkin-Elmer Life Science]). The
reaction was incubated at room temperature for 1 h and then terminated
by adding 10 mL of 63 SDS sample buffer. The reaction was then
incubated at 978C for 5 min and run on 12% SDS-PAGE. The gel was
stained with Coomassie Brilliant Blue G 250, dried, and subjected to
Phloroglucinol staining was performed as described (Can ˜o-Delgado
et al., 2003). Seedlings were fixed in a solution of three parts ethanol to
one part acetic acid and then cleared in a solution of chloral hydrate:
glycerol:water (8:1:2). The seedlings were then stained with lignin in a 2%
Analysis of FEI Expression Patterns
Tissue from transgenic lines harboring the FEI1 or FEI2 promoter–GUS
fusions was stained in 100 mM sodium phosphate buffer (pH 7.0) with 10
mM EDTA (pH 8.0), 0.5 mM potassium ferricyanide, 0.5 mM potassium
3076 The Plant Cell
ferrocyanide, 1 mM 5-bromo-4-chloro-3-indolyl-b-glucuronic acid, and
0.1% Triton X-100. The tissue was stained either for 1 h or overnight at
378C as indicated. Chlorophyll was removed with 95% ethanol. Ten
independent transgenic lines were analyzed, and a representative line
Localization of FEI2-GFP
Root apices from 7-d-old transgenic plants harboring 35S:FEI2-GFP
were used for confocal analyses. A Zeiss LSM510 confocal microscope
nm and 530 to 560 nm) was used for this analysis. Mannitol (0.8 M) was
applied to the root tip on the slides for plasmolysis.
Membrane Fractionation of FEI1-myc Fusion Proteins
sucrose MS plates for 7 d. Membrane proteins were fractionated by
grinding 200 mg of root tissue per 500 mL of buffer (20 mM Tris [pH 8.0],
0.33 M sucrose, 1mM EDTA,and plant protease inhibitor cocktail [Roche
Applied Science]), and insoluble debris was pelleted by centrifugation at
2000g for 10 min at 48C. The supernatant from the spin was designated
the total fraction. Some of the total fraction (150 mL) was further centri-
fuged at 20,000g for 45 min at 48C. The supernatant from this spin was
SDS-PAGE and analyzed by protein gel blotting. The anti-myc antibody
was obtained from Roche Applied Science. Anti-Hsc antibody used as a
loading control was obtained from Stressgen, and chicken anti-mouse
secondary antibody was obtained from Santa Cruz Biotechnology.
Cellulose Synthesis Assays
Cellulose synthesis was determined by [14C]Glc labeling as described
(Fagard et al., 2000) with the following modifications. Seedlings were
grown on 0% sucrose MS plates for 4 d and then transferred to MS
medium containing various supplements (as indicated in the figure
legends) for 3 d. Root tips (1.5 cm) were cut and washed three times
with 3 mL of glucose-free MS medium. Forty root tips were then
incubated in 1 mL of MS medium containing 0.1 mCi/mL [14C]Glc (NEN
Research) for 1 h in the dark at 228C in glass tubes. After treatment, the
roots were washed three times with 3 mL of glucose-free MS medium.
Next, the roots were extracted three times with 3 mL of boiling absolute
ethanol for 20 min, and total aliquots were collected (ethanol-soluble
fraction). Roots were then resuspended in 3 mL of chloroform:methanol
(1:1, v/v), extracted for 20 min at 458C, and finally resuspended in 3 mL of
acetone for 15 min at room temperature with gentle shaking. The
remaining material was resuspended in 500 mL of an acetic acid:nitric
acid:water solution (8:1:2, v/v/v) for 1 h in a boiling-water bath. Acid-
soluble material and acid-insoluble material were separated by glass
were washed with 5 mL of water. The acid wash and water wash
constitute the acid-soluble fraction. The filters yield the acid-insoluble
fraction. The amount of label in each fraction was determined by scin-
tillation countingusingliquidscintillation fluid(ScintiverseBDcocktail; SX
18-4; Fisher). The percentage of label incorporation was expressed as
label (ethanol plus acid-soluble plus acid-insoluble fractions).
Arabidopsis root tips were fixed in 2% paraformaldehyde and 2.5%
glutaraldehyde in phosphate buffer (0.1 M sodium phosphate, pH 7.4).
After rinsing with phosphate buffer, the samples were postfixed with
1% osmium tetroxide in sodium phosphate buffer for 30 min. Samples
were dehydrated through an increasing ethanol series followed by
propylene oxide and infiltrated and embedded in Polybed 812 epoxy
resin (Polysciences). For light microscopy, 1-mm cross sections of the
root tips were cut using a glass knife and a Leica Ultracut S ultrami-
crotome (Leica Microsystems), mounted on glass slides, and stained
with 1% toluidine blue in 1% borax. For transmission electron micros-
copy, selected blocks were further trimmed and ultrathin sections (70
nm) were cut using a diamond knife. Ultrathin sections were mounted
on 200-mesh copper grids and stained with 4% uranyl acetate and
Reynolds’ lead citrate. Sections were examined using a LEO EM-910
transmission electron microscope operating at 80 kV (Carl Zeiss), and
digital images were taken using an Orius SC1000 CCD camera
Whole root tips were visualized by first fixing in an ethanol:acid (9:1)
solution overnight, followed by two washes in 90 and 70% ethanol. Roots
were then cleared with a chloral hydrate:glycerol:water solution (8:1:2),
and the tips were visualized using Nomarski optics using a Nikon Eclipse
Analysis of Microtubules
Seedlings were grown for 5 d on 1% sucrose and then transferred onto
plates containing 1% sucrose, 4.5% sucrose, or 1% sucrose plus 50
mM NaCl for 3 d. Seedlings were fixed, stained for microtubules, and
imaged, all as described (Bannigan et al., 2006). Briefly, the fixative
contained 4% paraformaldehyde, 1% glutaraldehyde, 50 mM PIPES,
and 1 mM CaCl2. Seedlings were permeabilized by mild digestion of
pectin and brief incubation in ice-cold methanol. After rehydration in
PBS, roots were incubated with 1:1000 mouse monoclonal anti-tubulin
antibody (Sigma-Aldrich) at 378C overnight. The secondary antibody
used was Cy3-conjugated goat anti-mouse antibody (1:200; Jackson
ImmunoResearch). The imaging of whole roots was performed using a
confocal microscope (510 Meta; Carl Zeiss) equipped with a 633 oil-
immersion objective. Projections were assembled using Zeiss soft-
Measurement of Ethylene Production
Approximately30 seedlings weregrown on 1% sucrose MS plates for 3d
and then transferred to 4.5% sucrose plates for 3 d. Root tips (1 cm) were
excised and placed in 22-mL gas chromatography vials that contained
24 h at 238C in the dark, and the accumulated ethylene was measured as
described by Vogel et al. (1998). For etiolated tissue, seedlings (;40 per
vial) were grown in 22-mL gas chromatography vials containing 3 mL of
MS medium in the dark for 4 d. The accumulated ethylene was measured
by gas chromatography as described (Vogel et al., 1998).
Yeast Two-Hybrid Analysis
The open reading frames corresponding to thevarioustested geneswere
cloned into the bait plasmid (pEG202) or prey plasmid (pJG4-5) by
Gateway cloning from the respective entry clones made with the primers
shown in Supplemental Table 1 online. The plasmids were transformed
into the yeast strain EGY48 via LiOAc transformation as described (Chen
et al., 1992).
Sequence data from this article can be found in the Arabidopsis Genome
Initiative or GenBank/EMBL databases under the following accession
numbers: FEI1, At1g31420; FEI2, At2g35620; and SOS5, At3g46550.
The FEI LRR-RLKs Regulate Cell Expansion3077
The following materials are available in the online version of this article.
Supplemental Figure 1. Sequence Alignment of FEI1 and FEI2.
Supplemental Figure 2. Time Course of Root Swelling following
Transfer from 0 to 4.5% Sucrose Media.
Supplemental Figure 3. Transverse Sections through the Elongation
Zone of the Root from Various Single, Double, and Triple Mutants.
Supplemental Figure 4. Analysis of the sos5-2 Mutant.
Supplemental Figure 5. The fei1 fei2 Mutant Phenotype in Response
to Sucrose Is Not the Result of Increased Osmoticum.
Supplemental Figure 6. The FEI2-GFP Fusion Is Functional.
Supplemental Figure 7. Hypocotyl Length Is Not Affected in the fei
Supplemental Figure 8. Phloroglucinol Staining for Lignin in Wild-
Type and fei1 fei2 Seedlings Grown on MS Medium for 3 d in the
Supplemental Figure 9. Organization of Microtubules Is Not Altered
in the fei1 fei2 Mutant.
Supplemental Figure 10. Growth in the Presence of Elevated
Sucrose Does Not Affect Other sos Mutants.
Supplemental Figure 11. Effect of Inhibition of Ethylene Biosynthesis
on the cob Mutant.
Supplemental Table 1. Primers Used in This Study.
This work was supported by National Science Foundation Grant IOB-
0444347 to J.J.K. We thank Victoria Madden and Elena Davis for help in
transmission electron microscopy and Jason Reed, Jayson Punwani,
and Cris Argueso for critically reading the manuscript.
Received September 18, 2008; revised October 23, 2008; accepted
October 29, 2008; published November 18, 2008.
Aeschbacher, R., Hauser, M.-T., Feldmann, K.A., and Benfey, P.N.
(1995). The SABRE gene is required for normal cell expansion in
Arabidopsis. Genes Dev. 9: 330–340.
Alonso, J.M., et al. (2003). Genome-wide insertional mutagenesis of
Arabidopsis thaliana. Science 301: 653–657.
Anderson, C.M., Wagner, T.A., Perret, M., He, Z.H., He, D., and
Kohorn, B.D. (2001). WAKs: Cell wall-associated kinases linking the
cytoplasm to the extracellular matrix. Plant Mol. Biol. 47: 197–206.
Bannigan, A., Wiedemeier, A.M.D., Williamson, R.E., Overall, R.L.,
and Baskin, T.I. (2006). Cortical microtubule arrays lose uniform
alignment between cells and are oryzalin resistant in the Arabidopsis
mutant, radially swollen 6. Plant Cell Physiol. 47: 949–958.
Baskin, T.I. (2001). On the alignment of cellulose microfibrils by cortical
microtubules: A review and a model. Protoplasma 215: 150–171.
Baskin, T.I. (2005). Anisotropic expansion of the plant cell wall. Annu.
Rev. Cell Dev. Biol. 21: 203–222.
Baskin, T.I., and Williamson, R.E. (1992). Ethylene, microtubules and
root morphology in wild-type and mutant Arabidopsis seedlings. Plant
Biochemistry and Physiology Symposium 11: 118–130.
Beeckman, T., Przemeck, G.K.H., Stamatiou, G., Lau, R., Terryn, N.,
De Rycke, R., Inze, D., and Berleth, T. (2002). Genetic complexity of
cellulose synthase A gene function in Arabidopsis embryogenesis.
Plant Physiol. 130: 1883–1893.
Benfey, P.N., Linstead, P.J., Roberts, K., Schiefelbein, J.W., Hauser,
M.-T., and Aeschbacher, R. (1993). Root development in Arabidop-
sis: Four mutants with dramatically altered root morphogenesis.
Development 119: 57–70.
Birnbaum, K., Shasha, D.E., Wang, J.Y., Jung, J.W., Lambert, G.M.,
Galbraith, D.W., and Benfey, P.N. (2003). A gene expression map of
the Arabidopsis root. Science 302: 1956–1960.
Boudeau, J., Miranda-Saavedra, D., Barton, G.J., and Alessi, D.R.
(2006). Emerging roles of pseudokinases. Trends Cell Biol. 16: 443–452.
Bouquin, T., Mattsson, O., Naested, H., Foster, R., and Mundy, J.
(2003). The Arabidopsis lue1 mutant defines a katanin p60 ortholog
involved in hormonal control of microtubule orientation during cell
growth. J. Cell Sci. 116: 791–801.
Burg, S., and Burg, E. (1967). Molecular requirements for the biological
activity of ethylene. Plant Physiol. 42: 144–152.
Can ˜o-Delgado, A., Penfield, S., Smith, C., Catley, M., and Bevan, M.
(2003). Reduced cellulose synthesis invokes lignification and defense
responses in Arabidopsis thaliana. Plant J. 34: 351–362.
Castells, E., and Casacuberta, J.M. (2007). Signalling through kinase-
defective domains: The prevalence of atypical receptor-like kinases in
plants. J. Exp. Bot. 58: 3503–3511.
Chae, H.S., Faure, F., and Kieber, J.J. (2003). The eto1, eto2, and eto3
mutations and cytokinin treatment increase ethylene biosynthesis in
Arabidopsis by increasing the stability of ACS protein. Plant Cell 15:
Chae, H.S., and Kieber, J.J. (2005). Eto Brute? The role of ACS
turnover in regulating ethylene biosynthesis. Trends Plant Sci. 10:
Chang, C., Kwok, S.F., Bleecker, A.B., and Meyerowitz, E.M. (1993).
Arabidopsis ethylene-response gene ETR1: Similarity of product to
two-component regulators. Science 262: 539–544.
Chen, D.C., Yang, B.C., and Kuo, T.T. (1992). One-step transformation
of yeast in stationary phase. Curr. Genet. 21: 83–84.
Chevalier, D., Batoux, M., Fulton, L., Pfister, K., Yadav, R.K.,
Schellenberg, M., and Schneitz, K. (2005). STRUBBELIG defines a
receptor kinase-mediated signaling pathway regulating organ devel-
opment in Arabidopsis. Proc. Natl. Acad. Sci. USA 102: 9074–9079.
Claperon, A., and Therrien, M. (2007). KSR and CNK: Two scaffolds
regulating RAS-mediated RAF activation. Oncogene 26: 3143–3158.
Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method for
Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant
J. 16: 735–743.
Darley, C.P., Forrester, A.M., and McQueen-Mason, S.J. (2001). The
molecular basis of plant cell wall extension. Plant Mol. Biol. 47:
De Cnodder, T., Vissenberg, K., Van Der Straeten, D., and Verbelen,
J.P. (2005). Regulation of cell length in the Arabidopsis thaliana root
by the ethylene precursor 1-aminocyclopropane-1-carboxylic acid: A
matter of apoplastic reactions. New Phytol. 168: 541–550.
Desprez, T., Vernhettes, S., Fagard, M., Refre ´gier, G., Desnos, T.,
Aletti, E., Py, N., Pelletier, S., and Ho ¨fte, H. (2002). Resistance
against herbicide isoxaben and cellulose deficiency caused by dis-
tinct mutations in same cellulose synthase isoform CESA6. Plant
Physiol. 128: 482–490.
Ellis, C., Karafyllidis, I., Wasternack, C., and Turner, J.G. (2002). The
Arabidopsis mutant cev1 links cell wall signaling to jasmonate and
ethylene responses. Plant Cell 14: 1557–1566.
Fagard, M., Desnos, T., Desprez, T., Goubet, F., Refregier, G.,
Mouille, G., McCann, M., Rayon, C., Vernhettes, S., and Hofte,
3078The Plant Cell
H. (2000). PROCUSTE1 encodes a cellulose synthase required for Download full-text
normal cell elongation specifically in roots and dark-grown hypocotyls
of Arabidopsis. Plant Cell 12: 2409–2424.
Gifford, M.L., Robertson, F.C., Soares, D.C., and Ingram, G.C. (2005).
ARABIDOPSIS CRINKLY4 function, internalization, and turnover are
dependent on the extracellular crinkly repeat domain. Plant Cell 17:
Green, P.B. (1980). Organogenesis—A biophysical view. Annu. Rev.
Plant Physiol. 31: 51–82.
Hanks, S.K., and Quinn, A.M. (1991). Protein kinase catalytic domain
sequence database: Identification of conserved features of primary
structure and classification of family members. Meth. Enzymol. 200:
He, Z.-H., Fujiki, M., and Kohorn, B.D. (1996). A cell wall-associated,
receptor-like protein kinase. J. Biol. Chem. 271: 19789–19793.
Heim, D.R., Larrinua, I.M., Murdoch, M.G., and Roberts, J.L. (1998).
Triazofenamide is a cellulose biosynthesis inhibitor. Pestic. Biochem.
Physiol. 59: 163–168.
He ´maty, K., and Ho ¨fte, H. (2008). Novel receptor kinases involved in
growth regulation. Curr. Opin. Plant Biol. 11: 321–328.
He ´maty, K., Sado, P.-E., Van Tuinen, A., Rochange, S., Desnos, T.,
Balzergue, S., Pelletier, S., Renou, J.-P., and Ho ¨fte, H. (2007). A
receptor-like kinase mediates the response of Arabidopsis cells to the
inhibition of cellulose synthesis. Curr. Biol. 17: 922–931.
Humphrey, T.V., Bonetta, D.T., and Goring, D.R. (2007). Sentinels at
the wall: Cell wall receptors and sensors. New Phytol. 176: 7–21.
Kieber, J.J., Rothenburg, M., Roman, G., Feldmann, K.A., and Ecker,
J.R. (1993). CTR1, a negative regulator of the ethylene response
pathway in Arabidopsis, encodes a member of the Raf family of
protein kinases. Cell 72: 427–441.
Kohorn, B.D., Kobayashi, M., Johansen, S., Riese, J., Huang, L.F.,
Koch, K., Fu, S., Dotson, A., and Byers, N.R. (2006). An Arabidopsis
cell wall-associated kinase required for invertase activity and cell
growth. Plant J. 46: 307–316.
Kroiher, M., Miller, M.A., and Steele, R.E. (2001). Deceiving appear-
ances: signaling by “dead” and “fractured” receptor protein-tyrosine
kinases. Bioessays 23: 69–76.
Lang, J., Eisinger, W., and Green, P. (1982). Effects of ethylene on the
orientation of microtubules and cellulose microfibrils of pea epicotyl
cells with polylamellate cell walls. Protoplasma 110: 5–14.
Moore, B. (2004). Bifunctional and moonlighting enzymes: Lighting the
way to regulatory control. Trends Plant Sci. 9: 221–228.
Morillo, S.A., and Tax, F.E. (2006). Functional analysis of receptor-like
kinases in monocots and dicots. Curr. Opin. Plant Biol. 9: 460–469.
Nakagawa, T., et al. (2007). Improved Gateway binary vectors: High-
performance vectors for creation of fusion constructs in transgenic
analysis of plants. Biosci. Biotechnol. Biochem. 71: 2095–2100.
Paredez, A.R., Somerville, C.R., and Ehrhardt, D.W. (2006). Visuali-
zation of cellulose synthase demonstrates functional association with
microtubules. Science 312: 1491–1495.
Peng, L., Hocart, C.H., Redmond, J.W., and Williamson, R.E. (2000).
Fractionation of carbohydrates in Arabidopsis root cell walls shows
that three radial swelling loci are specifically involved in cellulose
production. Planta 211: 406–414.
tubule reorientations: Mediation by helical arrays. Planta 164: 439–447.
Roudier, F., Fernandez, A.G., Fujita, M., Himmelspach, R., Borner,
G.H.H., Schindelman, G., Song, S.,Baskin, T.I., Dupree,P.,Wasteneys,
G.O., and Benfey, P.N. (2005). COBRA, an Arabidopsis extracellular
glycosyl-phosphatidyl inositol-anchored protein, specifically controls
highly anisotropic expansion through its involvement in cellulose
microfibril orientation. Plant Cell 17: 1749–1763.
Scheible, W.R., Eshed, R., Richmond, T., Delmer, D., and Somerville,
C. (2001). Modifications of cellulose synthase confer resistance to
isoxaben and thiazolidinone herbicides in Arabidopsis ixr1 mutants.
Proc. Natl. Acad. Sci. USA 98: 10079–10084.
Schindelman, G., Morikami, A., Jung, J., Baskin, T.I., Carpita, N.C.,
Derbyshire, P., McCann, M.C., and Benfey, P.N. (2001). COBRA
encodes a putative GPI-anchored protein, which is polarly localized
and necessary for oriented cell expansion in Arabidopsis. Genes Dev.
Sebastia `, C.H., Hardin, S.C., Clouse, S.D., Kieber, J.J., and Huber, S.
C. (2004). Identification of a new motif for CDPK phosphorylation in
vitro that suggests ACC synthase may be a CDPK substrate. Arch.
Biochem. Biophys. 428: 81–91.
Shi, H., Kim, Y., Guo, Y., Stevenson, B., and Zhu, J.-K. (2003). The
Arabidopsis SOS5 locus encodes a putative cell surface adhesion
protein and is required for normal cell expansion. Plant Cell 15: 19–32.
Shiu, S.-H., and Bleecker, A.B. (2001). Receptor-like kinases from
Arabidopsis form a monophyletic gene family related to animal
receptor kinases. Proc. Natl. Acad. Sci. USA 98: 10763–10768.
Somerville, C. (2006). Cellulose synthesis in higher plants. Annu. Rev.
Cell Dev. Biol. 22: 53–78.
Steen, D.A., and Chadwick, A. (1981). Ethylene effects in pea stem
tissue.Evidence for microtubule mediation. PlantPhysiol. 67: 460–466.
Taiz, L. (1984). Plant cell expansion: Regulation of cell wall mechanical
properties. Annu. Rev. Plant Physiol. 35: 585–657.
Takahashi, H., Kawahara, A., and Inoue, Y. (2003). Ethylene promotes
the induction by auxin of the cortical microtubule randomization
required for low-pH-induced root hair initiation in lettuce (Lactuca
sativa L.) seedlings. Plant Cell Physiol. 44: 932–940.
Tatsuki, M., and Mori, H. (2001). Phosphorylation of tomato 1-amino-
cyclopropane-1-carboxylic acid synthase, LE-ACS2, at the C-terminal
region. J. Biol. Chem. 276: 28051–28057.
Vogel, J.P., Woeste, K.W., Theologis, A., and Kieber, J.J. (1998).
Recessive and dominant mutations in the ethylene biosynthetic gene
ACS5 of Arabidopsis confer cytokinin insensitivity and ethylene over-
production, respectively. Proc. Natl. Acad. Sci. USA 95: 4766–4771.
Wagner, T.A., and Kohorn, B.D. (2001). Wall-associated kinases are
expressed throughout plant development and are required for cell
expansion. Plant Cell 13: 303–318.
Yamagami, T., Tsuchisaka, A., Yamada, K., Haddon, W.F., Harden,
L.A., and Theologis, A. (2003). Biochemical diversity among the
1-amino-cyclopropane-1-carboxylate synthase isozymes encoded by
the Arabidopsis gene family. J. Biol. Chem. 278: 49102–49112.
Zimmermann, P., Hennig, L., and Gruissem, W. (2005). Gene-expres-
sion analysis and network discovery using Genevestigator. Trends
Plant Sci. 10: 407–409.
The FEI LRR-RLKs Regulate Cell Expansion3079