Current Biology 17, 922–931, June 5, 2007 ª2007 Elsevier Ltd All rights reserved
018
A Receptor-like Kinase Mediates
the Response of Arabidopsis Cells
to the Inhibition of Cellulose Synthesis
Kian Hématy,1 Pierre-Etienne Sado,1,3
Ageeth Van Tuinen,1,4 Soizic Rochange,1,5
Thierry Desnos,1,6 Sandrine Balzergue,2
Sandra Pelletier,1 Jean-Pierre Renou,2
and Herman Höfte1,*
1
Laboratoire de Biologie Cellulaire, UR501
Institut Jean-Pierre Bourgin
Institut National de la Recherche Agronomique Centre
de Versailles
Route de St Cyr
78026 Versailles
France
2
Unité mixte de Recherche de Génomique Végétale
2, rue Gaston Crémieux
CP 5708
91057 Évry cedex
France
Summary
Background: A major challenge is to understand how
the walls of expanding plant cells are correctly assem-
bled and remodeled, often in the presence of wall-
degrading micro-organisms. Plant cells, like yeast, react
to cell-wall perturbations as shown by changes in gene
expression, accumulation of ectopic lignin, and growth
arrest caused by the inhibition of cellulose synthesis.
Results: We have identified a plasma-membrane-bound
receptor-like kinase (THESEUS1), which is present in
elongating cells. Mutations in THE1 and overexpression
of a functional THE1-GFP fusion protein did not affect
wild-type (WT) plants but respectively attenuated and
enhanced growth inhibition and ectopic lignification in
seedlings mutated in cellulose synthase CESA6 without
influencing the cellulose deficiency. A T-DNA insertion
mutant for THE1 also attenuated the growth defect
and ectopic-lignin production in other but not all cellu-
lose-deficient mutants. The deregulation of a small num-
ber of genes in cesA6 mutants depended on the pres-
ence of THE1. Some of these genes are involved in
pathogen defense, in wall crosslinking, or in protecting
the cell against reactive oxygen species.
Conclusions: The results show that THE1 mediates the
response of growing plant cells to the perturbation of
cellulose synthesis and may act as a cell-wall-integrity
sensor.
*Correspondence: hofte@versailles.inra.fr
3
Present address: Unité Biopolymères, Interactions, Assemblages,
Institut National de la Recherche Agronomique, rue de la Gérau-
dière, BP 71627 F-44316 Nantes Cedex 3, France.
4
Present address: Laboratory for Plant Physiology, Arboretumlaan 4
6703 BD Wageningen, The Netherlands.
5
Present address: UMR 5546 CNRS-UPS, Signaux et messages cel-
lulaires, Pole de Biotechnologies végétales, 24, chemin de Borde
rouge, BP17 Auzeville, 31326 Castanet-Tolosan, France.
6
Present address: CEA cadarache, DEVM/LBDP, 13108 St Paul-
lez-Durance cedex, France.
DOI 10.1016/j.cub.2007.05.
Article
Introduction
The plant cell wall is a complex composite material com-
posed of polysaccharides, proteins, and phenolic com-
pounds [1]. Thanks to its remarkable architecture, the
thin walls of growing cells are resistant to extreme ten-
sile forces imposed by turgor pressure (around 1 Mpa)
of the cell. A major question in plant biology is how plant
cells can expand without losing the coherence of their
cell walls. Plant cell expansion is promoted by the turgor
pressure and, under physiological conditions, is limited
by the remodeling of load-bearing cell-wall polymers
[2]. Cell-wall remodeling involves cell-wall-modifying
enzymes and a family of lubricating proteins, the expan-
sins. Precise coordination of cell-wall deposition and
remodeling is required to maintain a constant cell-wall
thickness as observed in most mature cells and to pre-
vent wall rupture during growth. So far, nothing is known
about the signaling involved in this coordination in
plants.
In contrast to plants, a cell-wall-integrity-signaling
pathway has been reasonably well characterized in
yeast [3, 4]. Wall damage is sensed by membrane pro-
teins, such as Wsc1p and Mid1p, which operate during
vegetative growth and pheromone-induced morpho-
genesis, respectively. Their signal activates via Rho1p,
a MAP-kinase cascade leading to the transcription fac-
tors Rlm1p and the Swi4p/Swi6p complex, SBF, which
in turn control a suite of cell-wall biosynthetic genes.
The activation of this pathway in response to cell-wall
stress results, among other things, in a massive deposi-
tion of chitin in lateral walls of the mother cell and the
growing bud.
Plants most likely also have the ability to sense the
integrity of their cell walls [5]. For instance, the inhibition
of cellulose synthesis by chemical inhibitors or in mutant
backgrounds leads to the accumulation of ectopic lignin
and callose and to changes in the composition of matrix
polysaccharides [6, 7]. In mutant alleles of the cellulose-
synthase catalytic subunit (CESA) CESA3, ectopic ligni-
fication1 (eli1 [8, 9]) and constitutive expression of
vegetative storage proteins1 (cev1 [10, 11]) jasmonate-
and ethylene-signaling pathways are activated and
stress-response genes are upregulated. It remains to
be shown whether cell-wall-integrity sensors are in-
volved in these responses. Feedback signaling also
may be involved in linking cell growth to cellulose syn-
thesis. Cellulose microfibrils play a key role in the control
of growth anisotropy of plant cells [12]. In growing cells,
they are oriented perpendicular to the growth axis like
hoops around a barrel and thus are thought to provide
resistance to turgor pressure in the radial direction.
Consistent with this idea is the observation that per-
turbation of cellulose synthesis leads to cell swelling.
The dwarf phenotype of cellulose-deficient mutants
therefore is commonly interpreted as resulting from
the loss of growth anisotropy in cells with weakened
cell walls. Closer examination, however, reveals a more
Cell-Wall-Integrity Sensing in Arabidopsis
923
complicated picture. Radial-cell swelling is a relatively
late phenotype that is observed for only 24 hr after treat-
ment with cellulose-inhibiting herbicides [6] or after the
temperature shift in temperature-sensitive cellulose-
deficient mutants [13, 14]. A much earlier response is
the arrest of cell elongation, which can be observed in
the hypocotyl [15] or in the root (unpublished data),
within 3 hr after treatment with the cellulose inhibitor iso-
xaben and before any detectable change in the orienta-
tion of microfibrils. These observations suggest that cell
elongation is coupled to cellulose synthesis through
feedback signaling.
In this report, we set out to study how cell-wall alter-
ations are perceived and translated into changes in
gene expression, ectopic-lignin production, and the
inhibition of cell elongation. Null mutants for the cellu-
lose-synthase catalytic subunit CESA6, when grown in
the dark, are cellulose deficient, show a short hypocotyl,
accumulate ectopic lignin and callose, and show altered
transcript levels compared to those of the wild-type for
some 900 genes. We have isolated second-site muta-
tions that attenuate some of these pleiotropic effects
without rescuing cellulose synthesis. Here, we describe
three allelic mutants for a gene encoding a novel
plasma-membrane receptor kinase that plays a role in
monitoring the integrity of the cell wall.
Results
Second-Site Mutants Partially Restore Hypocotyl
Growth but Not Cellulose Deficiency in prc1
The cellulose-deficient mutant procuste1-1 (prc1-1)
shows a short-hypocotyl phenotype when grown in the
dark [16, 17]. To identify factors involved in the coordi-
nation of cell-wall synthesis and cell expansion, we
screened M2 seedlings of an ethyl-methanesulfonate
(EMS) mutagenized population of prc1-1 in the dark for
individuals with longer hypocotyls. We identified five in-
dependent partial-suppressor mutants with hypocotyls
Figure 1. Phenotype of prc1 and prc1/the1
Mutants
(A) Five-day-old dark-grown seedlings of the
WT, prc1-1, and two suppressor alleles, the1-
1 and the1-2, in a Col0 background. The scale
bar represents 5 mm.
(B) Cellulose accumulation as measured by
(14C)-D-glucose incorporation into hot sulfu-
ric-acid-insoluble cell-wall (CW) fraction,
expressed per seedling.
(C and D) Hypocotyl length (C) and growth
rate (D) of dark-grown WT (Col0), prc1-1,
and its two suppressor alleles.
Error bars represent the standard deviation
[SD]. Forty to fifty hypocotyls were measured
for the growth kinetics, and cellulose quanti-
fication was performed on three replicates
of 50 seedlings.
intermediate in length between prc1-1 mutants and the
wild-type (WT). Two allelic semidominant suppressors,
designated theseus1-1 (the1-1) and theseus1-2
(the1-2) after the Greek mythological figure Theseus,
who slew the brigand Procustes, were selected for fur-
ther study (Figure 1A). We previously showed that WT
hypocotyls grow slowly up to 48 hr after seed imbibition,
at which time the elongation rate accelerates abruptly in
cells at the hypocotyl base [15]. This acceleration prop-
agates subsequently toward the top of the hypocotyl.
Figures 1C and 1D show that the growth rate of WT
hypocotyls further increased beyond day two to reach
a peak at day five, after which the growth rate abruptly
dropped. Instead, hypocotyl cells in cellulose-deficient
prc1-1 mutants initially show the same elongation rate
as the WT but subsequently fail to accelerate their
growth [15]. This is confirmed in Figures 1C and 1D,
which show that the hypocotyl growth rate of prc1-1
remained more or less constant before slowing down
from day six on. Instead, the hypocotyl growth rate of
both the1/ prc1-1 alleles, which was initially slower
than that of prc1-1, strongly increased beyond day three
to reach over half the growth rate of WT hypocotyls
between day four and day five. Beyond day five, the
growth rate dropped, but more gradually than that of
the WT. The number of cells in epidermal-cell files
was unaltered in prc1-1/the1-1 compared to prc1-1
hypocotyls (data not shown), indicating that the1-1
affected only cell elongation and not cell division in the
hypocotyl.
We next compared the cell-wall composition of dark-
grown hypocotyls of, respectively, the WT, prc1-1, and
double homozygous the-1/prc1-1 with Fourier-
transform infrared microscopy (FTIR). Hierarchical clus-
ter analysis of the spectra allows the classification of
mutants according to their cell-wall defects [18]. The
two the-1/prc1-1 mutants clustered together but
remained within the cellulose-deficient cluster, suggest-
ing that both mutants were still cellulose deficient
Current Biology
924
despite their increased hypocotyl length (data not
shown). A Student’s t test was not able to distinguish
the spectra of the two the-1/prc1-1 mutants and prc1-
1, indicating that the increased cell elongation in the-1/
prc1-1 did not involve major changes in the ratio be-
tween cell-wall polysaccharides. Cellulose synthesis
was analyzed by measuring in vivo incorporation of
14
C-glucose into the crystalline cellulosic cell-wall
fraction of 4-day-old dark-grown seedlings. When ex-
pressed per seedling, the 14C-glucose incorporation in
the cellulosic fraction was reduced 2.5-fold in prc1-1
compared to the WT and did not differ significantly
from that of the1-1/prc1-1 mutants (Figure 1B), and
this confirmed that cellulose biosynthesis was not
restored in prc1/the1-1.
THE1 Encodes a Plasma-Membrane Receptor-like
Kinase of the CrRLK1L Family
Map-based cloning showed that both the1-1 and the1-2
carried a mutation in open reading frame (ORF)
At5g54380 (the1-1: G110A and the1-2: G448A) causing
amino acid changes in the predicted protein sequence
(the1-1: G37D and the1-2: E150K, Figure 2A) of a mem-
brane-bound receptor-like protein kinase (RLK). A
T-DNA insertion line in a Arabidopsis ecotype Wassilev-
skija (WS) background was isolated for this gene and
crossed to prc1-8, an allele in the same background.
This mutant allele (the1-3) also partially restored the hy-
pocotyl growth defect of prc1-8 (see below). Reverse
transcriptase-polymerase chain reaction (RT-PCR)
analysis revealed the presence of a truncated version
of the THE1 mRNA encoding a protein lacking the entire
kinase domain in the the1-3 mutant (data not shown). It
therefore remains to be shown whether this mutant
behaves as a complete null allele. As shown in Figure 2D,
expression from the CaMV 35S promoter of the full-
length At5g54380 cDNA fused at its C terminus to GFP
in the1/prc1-1 restored the Prc1-1 phenotype. The
shared phenotype of three independent alleles and the
complementation of the mutant phenotype with the chi-
meric cDNA indicate that At5g54380 indeed corre-
sponds to THE1.
The Arabidopsis genome encodes over 600 predicted
RLKs (PlantsP Kinase Classifaction: http://plantsp.
genomics.purdue.edu/html/) [19, 20]. THE1 belongs
to the CrRLK1L (Catharanthus roseus protein-kinase-
1-like) subfamily named after a founding member of
unknown physiological function, with in vitro-character-
ized kinase activity [21]. This subfamily, which carries
a predicted extracytoplasmic domain with no sequence
similarity to known protein domains, has 17 members in
Arabidopsis, all of which are encoded by intron-less
genes. THE1 encodes a protein of 855 amino acids
with a molecular weight of 93.3 kDa and a theoretical
pI of 5.7. THE1 is a predicted type I transmembrane pro-
tein (Figure 2A); the two allelic mutations occurred at
100% conserved residues (the1-1: G37D and the1-2:
E150K, Figure 2A) within the predicted extracytoplasmic
N-terminal domain. The N terminus contains a predicted
19-residue cleavable secretory signal peptide. The cyto-
solic C-terminal region is predicted to contain a protein
kinase domain (between residues 510 and 783; prosite
pattern PS50011), which contains the 11 characteristic
kinase subdomains [22] with the predicted ATP-binding
site (between residues 516 and 538; prosite pattern
PS00107) and the serine/threonine kinase active-site
signature (between residues 630 and 642; prosite
pattern PS00108, Figure 2A). The amino acid sequence
contains 18 potential N-linked glycosylation sites, 16
of which are in the N-terminal domain.
To validate the protein-kinase activity of THE1, we
produced the C-terminal cytosolic domain (415 amino
acids; residues 441 to 855) using an in vitro transcrip-
tion-translation system (Figure 2B). Incubation of the
unlabeled protein with (g32P)-ATP led to the incorpora-
tion of radioactivity in the in vitro-synthesized protein,
indicating that THE1 indeed has autophosphorylation
activity (Figure 2B).
Phosphoproteomics data from Nühse et al. [23] fur-
ther confirmed that THE1 is a major phosphorylated
protein in plasma-membrane fractions isolated from
Arabidopsis cell-suspension cultures. One of the tryptic
peptides generated from the C terminus of THE1 was
shown to carry one, two, or three phosphates (peptide
2 in Figure 2C) toward its N terminus or a single phos-
phate on one residue of a Thr pair at a position closer
to the C terminus. Interestingly, no peptides were found
that were simultaneously phosphorylated at both posi-
tions, suggesting that the two phosphorylation sites
were mutually exclusive. Another phosphopeptide (pep-
tide 1, Figures 2A and 2C) was also identified. This pep-
tide, however, which maps within the conserved kinase
signature domain, also could have originated from ten
other members of the CrRLK1L family (data not shown).
These phosphopeptides display an unusually high de-
gree of Thr phosphorylation (70%), compared to the
only 10% threonine phosphorylation sites identified by
Nühse et al. [23].The observation that THE1 possesses
an active kinase domain, which is phosphorylated
primarily on Thr residues at the C terminus of the protein,
is consistent with the in vitro-phosphorylation data for
CrRLK1 [21].
We next used confocal laser-scanning microscopy
to study the intracellular localization of a functional
THE1-GFP fusion protein in epidermal cells of the hypo-
cotyl of a prc1-1/the1-1::35S-THE1-GFP transformant
(see below, Figures 2E and 2F). All tissues investigated
were homogeneously labeled at the cell surface. Upon
mild plasmolysis, the fluorescence retracted from
the cell wall and stained Hechtian strands, showing
that the fusion protein was primarily associated with
the plasma membrane (Figure 2F). The plasma-
membrane location confirms the above-mentioned
phophoproteomics results on suspension-cultured
Arabidopsis cells [23].
THE1 Is Expressed in Elongating Cells
and in Vascular Tissue
To study the THE1 expression pattern, we used the UidA
reporter gene expressing b-glucuronidase (GUS) placed
under the control of a 2 kb upstream sequence contain-
ing the putative promoter region and 50 -UTR of THE1.
Seedlings from 20 independent transgenic lines were
stained for GUS activity. Fourteen lines showed a very
similar staining pattern, whereas six lines did not stain.
Five lines harboring a homozygous T-DNA insertion at
a single locus were established and used for detailed ex-
pression analysis (Figure 3). GUS staining was observed
Cell-Wall-Integrity Sensing in Arabidopsis
925

Figure 2. THESEUS1 Is a Functional Plasma-Membrane-Anchored Receptor-like Kinase of the CrRLK1L Family

(A) THE1 protein structure with characteristic domains boxed: signal peptide (yellow), CrRLK1-like extra-cytoplasmic domain (white), mem-
brane-anchor domain (orange), juxta-membrane segment (light gray), kinase catalytic domain (blue) with predicted ATP-binding site (purple)
and Ser/Thr signature (light blue), and phosphorylated C-terminal domain (dark gray). Phosphopeptides detected in Nühse et al. [23] are under-
lined in red. The histogram at the top of the figure was generated with WebLogo 2.8.2 [45] and represents the amino acid sequence conservation
(in bits: 3.5 = 100%) among the 17 Arabidopsis CrRLK1L members (for a full protein alignment, see Figure S2).
(B) In vitro phosphorylation of THE1 kinase domain. (35S)-Met-labeled in vitro-translated THE1-kinase-domain fragment (415–855aa) (middle
lane) and (g-32P)-ATP incorporation into in vitro-translated THE1 kinase domain through autophosphorylation (right lane). Molecular-weight
markers are from the BenchMark Prestained Protein Ladder (Invitrogen).
(C) Sequence of peptide 1 and 2 shown in (A). Phosphorylated residues are highlighted in red [23]. Peptide 1 is conserved in 11 Arabidopsis
CrRLK1L members (Figure S1), whereas peptide 2 is specific for THE1. Peptide 2 carries two mutually exclusive phosphorylation domains phos-
phorylated at one, two, or three positions (above) or at a single position (below), respectively [23].
(D) Five-day-old dark-grown seedlings. Genotypes, mean, and standard deviation (SD) of hypocotyl lengths are indicated. THE1ox lines with
comparable GFP fluorescence in Col0 and prc1-1/the1-1 were selected. The scale bar represents 5 mm.
(E and F) Plasma-membrane localization of THE1-GFP expressed from the 35S promoter in hypocotyl epidermal cells of stably transformed
3-day-old light-grown Col0 seedlings before (E) or after (F) plasmolysis with 5% NaCl for 10 min. Nomarski (above), single confocal optical sec-
tion visualizing GFP fluorescence (500–520 nm, middle) and autofluorescence (620–700 nm, bottom). Arrowhead indicates GFP-labeled Hechtian
strands. The scale bar represents 50 mm.

in all organs investigated. Staining was absent from the
root tip and root hairs and present in the elongation zone
of both main and lateral roots. In mature roots, only the
central cylinder was stained (Figure 3B). When grown in
the dark, GUS staining was homogeneous throughout
2-day-old seedlings and gradually disappeared starting
from the hypocotyl base in cells that ceased their
elongation, and it remained primarily in cotyledons after
7 days (Figure 3D). In the light, GUS expression was ini-
tially observed in the hypocotyl and the shoot apical
meristem and leaf primordia and became restricted later
to the meristem and leaf base and petiole (Figure 3A). In
summary, GUS was expressed primarily in expanding
cells and in vascular tissue. Genevestigator-microarray
Current Biology
926
expression data [24] were consistent with the GUS
expression pattern (data not shown).
Overexpression of THE1-GFP Enhances the Prc1
Phenotype
The similar mutant phenotype of the two missense
alleles and the T-DNA insertion allele implies that THE1
is required for normal hypocotyl growth inhibition in
a CESA6-deficient background. We next studied the
Figure 4. THE1-Dependent Growth Inhibition in Several but Not All Cellulose-Deficient Mutants and Not in a WT Background
(A) Five-day-old dark-grown seedlings. Genotypes, mean, and SD of hypocotyl lengths are indicated. The scale bar represents 5 mm.
(B) Greenhouse-grown 4-week-old plants. The scale bar represents 5 mm.
Figure 3. GUS Expression Pattern in Light-
and Dark-Grown Seedlings Expressing the
UidA Gene from the THE1-Upstream Se-
quence
(A–B) GUS staining during seedling develop-
ment in the light is associated with elongation
zone of organs (hypocotyl, root, and leaf pet-
iole). The age is shown on the right of each
seedling.
(C) Close-up view of the root apex of a 3-day-
old light-grown seedling showing GUS stain-
ing in the vasculature and the elongation zone
but not in root tip.
(D) Lateral roots of an 8-day-old light-grown
seedling show an expression pattern similar
to that of the primary root, including the
expression in the vasculature.
(E) GUS staining in the elongation zone of
dark-grown hypocotyls.
(F) GUS staining in a 4-week-old rosette leaf.
Scale bars in (A)–(D) represent 1 mm; scale
bars in (E) and (F) represent 5 mm.
effect of its overexpression. Transgenic lines were
constructed expressing THE1-GFP from the constitutive
CaMV 35S promoter (hereafter referred to as THE1ox) in
both a WT and a prc1-1/the1-1 background. Among 18
independent prc1-1/the1-1 transformants, 14 showed
a hypocotyl length similar to or shorter than that of
prc1-1. Instead, all of the 18 transformants in Col0
showed a WT phenotype (Figure 4A) despite the pres-
ence of comparable levels of GFP fluorescence in Col0
Cell-Wall-Integrity Sensing in Arabidopsis
927

and prc1-1/the1-1 backgrounds (data not shown). The

presence in itself of THE1-GFP in the plasma membrane
was therefore not sufficient to inhibit hypocotyl elonga-
tion. It required activation of THE1-GFP, and this activa-
tion only occurred in the absence of CESA6.

the1-3 Attenuates the Hypocotyl Growth Inhibition

in Other but Not All Cell-Wall Mutants
and Not in a Cytoskeleton Mutant
Using the T-DNA insertion allele the1-3, we evaluated the
involvement of THE1 in the response to other wall-dam-
aging mutations. Double-mutant combinations of the1-3
with other dwarf mutants were analyzed for their dark-
grown hypocotyl length (Figure 4B). The hypocotyl
length of the1-3 seedlings was comparable to, or even
slightly shorter than, that of WT controls. the1-3 partially
reverted not only the hypocotyl-elongation defect of
prc1-8 but also that of other cellulose-deficient mutants,
such as cesA3eli1-1 [9], cesA1rsw1-10 [17], and pompom1-
2 [25]. The effect of the1-3 on cellulose-deficient mutant
korrigan1-1 [26] was less pronounced and undetectable
on botero1, which lacks the catalytic P60 subunit of the
microtubule-severing enzyme, katanin [27, 28]. kor1-1
and bot1-1 already showed a hypocotyl length compara- Figure 5. THE1-Dependent Ectopic Lignification in cellulose-
ble to that of the prc1-8/the1-3 double mutant, so the ab- deficient mutants but Not in a WT Background
sence of an effect of the1-3 may somehow reflect an up- Five-day-old dark-grown without sucrose (A) or light-grown on 4.5%
per growth limit for the1-3 mutants. We therefore also sucrose (B) seedlings were stained for lignin with Phloroglucinol-
assessed the effect of the1-3 or THE1ox on the light- HCl. Pictures were taken under a binocular. The inset shows an
grown phenotype of cell-wall mutants (Figure 4C). As ex- enlarged image of the hypocotyl base and collet of (from left to right)
pected, the1-3 or THE1ox greenhouse-grown plants prc1-1, prc1-1/the1-1, and prc1-1/the1-1::THE1ox.
were indistinguishable from WT controls. The same
was true for the1-3 or THE1ox in a prc1-8 background,
light (Figure 5B). No ectopic lignin was observed in the
both of which remained indistinguishable from the WT.
WT, the1-3, or prc1-1 (data not shown), and THE1ox pre-
This is not surprising because mutations in CESA6 do
sented patches of hyperlignification at the base of the
not cause a light-grown phenotype [16]. Instead, the
hypocotyl in a WT background. eli1-1 accumulated ec-
CESA3 mutant eli1-1 was severely dwarfed, as de-
topic lignin as described [9]. Again, the1-3 and THE1ox
scribed previously [9]. Interestingly, the introduction of
respectively repressed and very strongly enhanced the
the1-3 in this background strongly attenuated the Eli1-
accumulation of ectopic lignin in eli1-1. These observa-
1 phenotype, and THE1ox greatly enhanced the dwarf
tions convincingly show that THE1 is required for the
phenotype. Greenhouse-grown pom1-1 mutants and
dwarf phenotype and ectopic-lignin accumulation in
cesA1rsw1-10 were also slightly dwarfed. Again, this phe-
the greenhouse-grown cellulose-deficient mutants eli1-
notype was attenuated by the1-3 and enhanced by
1, rsw1-10, and pom1-2. Instead, the1-3 did not revert
THE1ox. Finally, we did not observe an effect of the1-3
the light-grown phenotype of kor1-1 or bot1-1, indicating
on the light-grown phenotype of both kor1-1 and bot1-1.
that the developmental effect of the latter mutations
does not involve THE1.
THE1 Controls Ectopic-Lignin Accumulation
in Cellulose-Deficient Mutant Backgrounds
Dark-grown prc1-1 hypocotyls accumulate ectopic Potential Target Genes for THE1
lignin [6, 17]. We investigated whether this phenotype To identify potential target genes for THE1, we carried
persisted in the1/ prc1-1 double mutants. Phloroglucinol out a transcriptomic analysis with complete Arabidopsis
staining in prc1-1 showed patches of cells that accumu- transcriptome microarray (CATMA) chips [29, 30]. The
lated ectopic lignin, particularly in the endoderm within following comparisons were made with RNA prepared
the hypocotyl (Figure 5A) and the mature part of the from 5-day-old dark-grown seedlings: prc1-8 versus
root. Instead, in the1-1/prc1-1, the amount of ectopic lig- WS, prc1-8/the1-3 versus prc1-8 (in WS), and prc1-1/
nin was reduced in hypocotyls and entirely absent in the the1-1 versus prc1-1 (in Col0). In total, 887 genes were
root (Figure 5A). In THE1ox lines, ectopic-lignin accumu- significantly deregulated in prc1-8, compared to the
lation was restored in the hypocotyl, and it was strongly WT (602 up and 285 down). Such a large effect of the
enhanced throughout the root in prc1-1/the1-1 but not in absence of a cellulose-synthase catalytic subunit on
a WT background. These results demonstrate that ec- gene expression underscores the pleiotropic effects of
topic-lignin accumulation in prc1 is at least in part de- the mutation. Comparably large numbers of genes
pendent on functional THE1 and confirm that the protein were differentially expressed in the comparisons of
requires activation, which only occurs in a genetic back- prc1-8 with prc1-8/the1-3 (342; 200 up and 142 down)
ground deficient for CESA6. We also assessed the accu- and prc1-1 with prc1-1/the1-1 (873; 355 up and 518
mulation of ectopic lignin in a subset of the mutants in the down). Thirty potentially positively THE1-regulated
Current Biology
928

Figure 6. THE1-Dependent Deregulation of Gene Expression in prc1

(A) Venn diagrams showing intersections between sets of genes upregulated (top) or downregulated (bottom) in prc1-8 compared to the WT (WS)
and sets of genes with opposite changes, respectively, in prc1-1/the1-1 (in Col0) and prc1-8/the1-3 (in WS) compared to corresponding single
prc1 mutants.
(B) List of THE1-dependent differentially expressed genes in (A). The color code depicts relative transcript levels on the CATMA microarray and
was generated with Genesis 1.7.0 [46].
(C) Real-time qPCR analysis of selected genes from intersections shown in (A) and listed in (B). Relative transcript levels compared to the WT
control are shown. Backgrounds are Col0 and WS, respectively, for the five lanes to the left and the four lanes to the right. Analysis was per-
formed on two independent RT reactions from RNA different from those used in the microarray analysis (A and B).

genes were found in the intersections between prc1-8
and WS up, prc1-8/the1-3 and prc1-8 down, and prc1-
1/the1-1 and prc1-1 down, repectively. Similarly, six po-
tentially negatively THE1-regulated genes were found in
the complementary intersection (Figures 6A and 6B, and
Figure S1 in the Supplemental Data available online).
Among these genes, we selected six (five from the first
intersection and one from the second intersection) for
validation by quantitative RT-PCR (Figure 6C). Consis-
tent with the microarray data, transcript levels for five
genes were higher in two prc1 alleles, compared to the
WT, and lower in the two prc1/the1 combinations, com-
pared to the corresponding prc1 alleles. Interestingly,
the transcript levels for these genes were also greatly
enhanced in prc1-1/the1-1::THE1ox, compared to
prc1-1, with no significant differences between
WT::THE1ox and the WT. As expected, the gene
At5g53190 showed a transcript profile exactly opposite
to that of the five other genes. In conclusion, we have
identified a small number of genes, the deregulation of
which, in the absence CESA6, depends on the presence
of THE1. Among the 30 genes that were upregulated
upon THE1 activation, genes related to defense, oxida-
tive stress, and cell-wall metabolism were largely over-
represented (Figure 6B, Figure S1, and Discussion).
Discussion
The inhibition of cellulose synthesis either with her-
bicides or because of mutations triggers a set of
characteristic cellular changes, including growth inhibi-
tion, lignin accumulation, changes in matrix polysaccha-
ride composition, and altered transcript levels for hun-
dreds of genes. These changes presumably reflect the
ability of the plant to perceive changes in the cell wall.
We showed that THE1 mediates a subset of these re-
sponses. Indeed, hypocotyl-growth inhibition, ectopic-
lignin accumulation, and changes in the expression of
a small set of genes in a prc1 mutant background were
diminished in the1 loss-of-function mutants and en-
hanced upon overexpression of THE1-GFP. Other re-
sponses did not depend on THE1. Indeed, callose
accumulation (data not shown) and the deregulation of
many other genes were not affected by mutations in
the1. Other THE1 family members might have partially
redundant roles in mediating these responses.
We showed that THE1 encodes a functional plasma-
membrane-bound Ser/Thr kinase. Phosphoproteomics
on the plasma membrane of Arabidopsis cell cultures
showed that the C terminus of THE1 is phosphorylated
in vivo [23]. One peptide contained two mutually exclu-
sive phosphorylation sites, suggesting that they may
act as a molecular switch. We isolated two mutations
in highly conserved residues within the predicted extra-
cellular domain. These mutations might abolish the
binding of THE1 to an extracellular ligand. Interestingly,
these point mutations or a T-DNA insertion in THE1, as
well as overexpression of THE1-GFP in the plasma
membrane, did not show any phenotypic effect in
a WT background. This suggests that THE1 requires
Cell-Wall-Integrity Sensing in Arabidopsis
929
activation through some posttranslational mechanism.
What is the signal? In principle, it could be the altered
cellulose-synthase complex itself in CESA1 (rsw1-10),
CESA3 (eli1-1), or CESA6 (prc1-1) mutants, and perhaps
in pom1-1. In this scenario, the absence of an effect of
the1-3 on kor1-1 might indicate that kor1-1 does not
affect the CESA complex in the same way as the other
mutations. Alternatively, the factor that activates THE1
might be a change in the cell-wall composition, the
release of particular cell-wall fragments in mutant walls,
or perhaps a mechanical strain at the interface between
the cell wall and plasma membrane in the mutants.
Finally, our data do not exclude the possibility that
THE1 recognizes a secondary messenger that is re-
leased upon cell-wall damage.
Microarray analysis identified 36 genes that were
differentially regulated by THE1 (Figure S1). Among
30 positively regulated genes, two encode putative tran-
scription factors (TFs) WRKY45 and an AP2-like protein
(AP2-EREBP B4). Both might control the transcription of
other THE1-regulated genes. Several upregulated genes
encode proteins with a role in protection against reac-
tive oxygen species: glutaredoxin (Grx) A2, thioredoxin
(Trx) H5 [31], PMSR2 [32], and tyrosine aminotransfer-
ase (TAT3), the first committed enzyme in the synthesis
of radical scavengers tocopherols from tyrosine. Genes
encoding cell-wall-related proteins were also overrepre-
sented among the upregulated genes. This included
putative cell-wall peroxidase 59 [33] and two closely
related extensins. Extensins are hydroxyproline-rich
proteins that form covalent networks involving di-
isodityrosin crosslinks, the formation of which is cata-
lyzed by cell-wall peroxidases [34]. Other cell-wall
proteins were the cell-wall-loosening-protein expansin
1 (AtEXPA1) and two small, 99% identical, glycine-rich
proteins (GRP3 and GRP3S). Interestingly, these GRP
proteins have been shown to bind to wall-associated
kinase1 (WAK1) [35], a receptor kinase that also might
play a role in wall-to-cytoplasm signaling. An interesting
possibility is that THE1 and WAK signaling pathways are
connected through these GRPs. Other upregulated
genes encode potential pathogen-defense proteins: an
inhibitor of fungal xyloglucan endoglucanases (EDGP),
a trypsin inhibitor, a putative defense-related protein,
and SUR2/CYP83B1, which is a cytochrome-P450 en-
zyme involved in the synthesis of defense molecules, in-
dole glucosinolates. Activation of the latter enzyme also
indirectly leads to reduced auxin levels through the de-
pletion of the pool of indole-3-acetaldoxime, a common
precursor for glucosinolates and auxin [36], and these
reduced auxin levels also might contribute to reduced
growth. Interestingly, despite the observed THE1-
dependent accumulation of ectopic lignin, no monoli-
gnol biosynthetic genes were upregulated, suggesting
that the observed ectopic lignin was the result of cross-
linking of pre-existing monolignols. Among the downre-
gulated genes, two genes encode nodulin-like proteins
[37]: MtN21, a predicted polytopic integral membrane
protein, and MtN3, which does not show structural sim-
ilarity to proteins of known function. A rice homolog
(Os8N3) is a disease-susceptibility gene [38].
In conclusion, the following scenario emerges from
our observations. THE1 is part of a cell-wall surveillance
system in growing cells. Inhibition of cellulose synthesis
leads to the activation of THE1, and this activation might
involve a change in the kinase activity of the protein,
through a so-far-unknown mechanism. THE1 also may
be activated during other cell-wall perturbations that
occur during abiotic stress or pathogen attack. Some
pathogens indeed produce cellulose synthesis inhibi-
tors, such as thaxtomin A by Steptomyces scabiei [39];
others attack the cell wall with hydrolytic enzymes
[40]. This in turn might cause the activation of gene
expression. This response involves the synthesis of de-
fense proteins, among which are an enzyme involved in
glucosinolate synthesis (CYP83B1 [41, 42]) and inhibi-
tors of microbial proteases and fungal Xyloglucan-
specific endoglucanases [43]. The latter enzymes frag-
ment xyloglucans into oligosaccharides. Interestingly,
Takeda et al. [44] showed that the incorporation of xylo-
glucan oligosaccharides in the cell wall by endogenous
xyloglucan endotransglycosylase/hydrolase promoted
cell-wall loosening and cell elongation, whereas the in-
corporation of larger xyloglucan polysaccharides led
to growth inhibition. It is therefore conceivable that the
production of specific inhibitors for fungal XEGases is
crucial to prevent further loosening and weakening of
the wall. The response also involves the synthesis of
structural cell-wall proteins and the activation of their
oxidative crosslinking in the cell wall. These cell-wall
modifications presumably provide protection against
further attack by hydrolytic enzymes and might cause
growth inhibition, which also might be an adaptive
response to both biotic and some abiotic stresses.
The activation of oxidative crosslinking also would
lead to the accumulation of ectopic lignin in cells that
already produce monolignols. Future work will aim at
the understanding of the THE1 signaling pathway, the
identification of the ligand, and the function of the down-
stream genes in the control of cell elongation and the
defense against pathogens.
Supplemental Data
Experimental Procedures, two figures, and two tables are available
at http://www.current-biology.com/cgi/content/full/17/11/922/
DC1/.
Acknowledgments
T. Nühse is thanked for making his data available before publication.
G. Mouille is thanked for micro-FTIR-spectroscopy and scientific
advice. O. Grandjean and L. Gissot are thanked for their help with
confocal microscopy. M.-T. Hauser, S. Vernhettes, M. Gonneau, T.
Desprez, and T. Nühse are thanked for stimulating discussions. M.
Da Costa is thanked for helping with the RTS100 in vitro translation
and kinase assay. D. Tolleter, A. Lours, C. Vandervennet, and
J. Delacourt are thanked for their technical assistance.
This work was supported by grants from the French Ministery of
Research, Actions Concertées Incitatives Biologie Moléculaire, Cel-
lulaire et Structurale (grant no. 195 to H.H.) with a PhD fellowship to
K.H., and the European Economic Community framework program 5
‘‘GEMINI.’’
Received: March 19, 2007
Revised: April 22, 2007
Accepted: May 3, 2007
Published online: May 31, 2007
References
1. Somerville, C., Bauer, S., Brininstool, G., Facette, M., Hamann, T.,
Milne, J., Osborne, E., Paredez, A., Persson, S., Raab, T., et al.
Current Biology
930
(2004). Toward a systems approach to understanding plant
cell walls. Science 306, 2206–2211.
2. Cosgrove, D.J. (2005). Growth of the plant cell wall. Nat. Rev.
Mol. Cell Biol. 6, 850–861.
3. Heinisch, J.J. (2005). Baker’s yeast as a tool for the development
of antifungal kinase inhibitors–targeting protein kinase C and the
cell integrity pathway. Biochim. Biophys. Acta 1754, 171–182.
4. Levin, D.E. (2005). Cell wall integrity signaling in Saccharomyces
cerevisiae. Microbiol. Mol. Biol. Rev. 69, 262–291.
5. Pilling, E., and Hofte, H. (2003). Feedback from the wall. Curr.
Opin. Plant Biol. 6, 611–616.
6. Desprez, T., Vernhettes, S., Fagard, M., Refregier, G., Desnos,
T., Aletti, E., Py, N., Pelletier, S., and Hofte, H. (2002). Resistance
against herbicide isoxaben and cellulose deficiency caused by
distinct mutations in same cellulose synthase isoform CESA6.
Plant Physiol. 128, 482–490.
7. His, I., Driouich, A., Nicol, F., Jauneau, A., and Hofte, H. (2001).
Altered pectin composition in primary cell walls of korrigan,
a dwarf mutant of Arabidopsis deficient in a membrane-bound
endo-1,4-beta-glucanase. Planta 212, 348–358.
8. Cano-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.
9. Cano-Delgado, A.I., Metzlaff, K., and Bevan, M.W. (2000). The
eli1 mutation reveals a link between cell expansion and second-
ary cell wall formation in Arabidopsis thaliana. Development
127, 3395–3405.
10. Ellis, C., Karafyllidis, I., Wasternack, C., and Turner, J.G. (2002).
The Arabidopsis mutant cev1 links cell wall signaling to jasmo-
nate and ethylene responses. Plant Cell 14, 1557–1566.
11. Ellis, C., and Turner, J.G. (2001). The Arabidopsis mutant cev1
has constitutively active jasmonate and ethylene signal path-
ways and enhanced resistance to pathogens. Plant Cell 13,
1025–1033.
12. Somerville, C. (2006). Cellulose synthesis in higher plants. Annu.
Rev. Cell Dev. Biol. 22, 53–78.
13. Arioli, T., Peng, L., Betzner, A.S., Burn, J., Wittke, W., Herth, W.,
Camilleri, C., Hofte, H., Plazinski, J., Birch, R., et al. (1998).
Molecular analysis of cellulose biosynthesis in Arabidopsis.
Science 279, 717–720.
14. Williamson, R.E., Burn, J.E., Birch, R., Baskin, T.I., Arioli, T.,
Betzner, A.S., and Cork, A. (2001). Morphology of rsw1, a cellu-
lose-deficient mutant of Arabidopsis thaliana. Protoplasma 215,
116–127.
15. Refregier, G., Pelletier, S., Jaillard, D., and Hofte, H. (2004). Inter-
action between wall deposition and cell elongation in dark-
grown hypocotyl cells in Arabidopsis. Plant Physiol. 135, 959–
968.
16. Desnos, T., Orbovic, V., Bellini, C., Kronenberger, J., Caboche,
M., Traas, J., and Hofte, H. (1996). Procuste1 mutants identify
two distinct genetic pathways controlling hypocotyl cell elonga-
tion, respectively in dark- and light-grown Arabidopsis seed-
lings. Development 122, 683–693.
17. Fagard, M., Desnos, T., Desprez, T., Goubet, F., Refregier, G.,
Mouille, G., McCann, M., Rayon, C., Vernhettes, S., and Hofte,
H. (2000). PROCUSTE1 encodes a cellulose synthase required
for normal cell elongation specifically in roots and dark-grown
hypocotyls of Arabidopsis. Plant Cell 12, 2409–2424.
18. Mouille, G., Robin, S., Lecomte, M., Pagant, S., and Hofte, H.
(2003). Classification and identification of Arabidopsis cell wall
mutants using Fourier-Transform InfraRed (FT-IR) microspec-
troscopy. Plant J. 35, 393–404.
19. Shiu, S.H., and Bleecker, A.B. (2003). Expansion of the receptor-
like kinase/Pelle gene family and receptor-like proteins in Arabi-
dopsis. Plant Physiol. 132, 530–543.
20. Shiu, S.H., and Bleecker, A.B. (2001). Plant receptor-like kinase
gene family: Diversity, function, and signaling. Sci. STKE 2001,
RE22.
21. Schulze-Muth, P., Irmler, S., Schroder, G., and Schroder, J.
(1996). Novel type of receptor-like protein kinase from a higher
plant (Catharanthus roseus). cDNA, gene, intramolecular auto-
phosphorylation, and identification of a threonine important for
auto- and substrate phosphorylation. J. Biol. Chem. 271,
26684–26689.
22. Hanks, S.K., and Hunter, T. (1995). Protein kinases 6. The eu-
karyotic protein kinase superfamily: Kinase (catalytic) domain
structure and classification. FASEB J. 9, 576–596.
23. Nuhse, T.S., Stensballe, A., Jensen, O.N., and Peck, S.C. (2004).
Phosphoproteomics of the Arabidopsis plasma membrane and
a new phosphorylation site database. Plant Cell 16, 2394–2405.
24. Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L., and Gruis-
sem, W. (2004). GENEVESTIGATOR. Arabidopsis microarray
database and analysis toolbox. Plant Physiol. 136, 2621–2632.
25. Hauser, M.T., Morikami, A., and Benfey, P.N. (1995). Conditional
root expansion mutants of Arabidopsis. Development 121,
1237–1252.
26. Nicol, F., His, I., Jauneau, A., Vernhettes, S., Canut, H., and
Hofte, H. (1998). A plasma membrane-bound putative endo-
1,4-beta-D-glucanase is required for normal wall assembly and
cell elongation in Arabidopsis. EMBO J. 17, 5563–5576.
27. Bichet, A., Desnos, T., Turner, S., Grandjean, O., and Hofte, H.
(2001). BOTERO1 is required for normal orientation of cortical
microtubules and anisotropic cell expansion in Arabidopsis.
Plant J. 25, 137–148.
28. Burk, D.H., Liu, B., Zhong, R., Morrison, W.H., and Ye, Z.H.
(2001). A katanin-like protein regulates normal cell wall biosyn-
thesis and cell elongation. Plant Cell 13, 807–827.
29. Crowe, M.L., Serizet, C., Thareau, V., Aubourg, S., Rouze, P.,
Hilson, P., Beynon, J., Weisbeek, P., van Hummelen, P.,
Reymond, P., et al. (2003). CATMA: A complete Arabidopsis
GST database. Nucleic Acids Res. 31, 156–158.
30. Hilson, P., Allemeersch, J., Altmann, T., Aubourg, S., Avon, A.,
Beynon, J., Bhalerao, R.P., Bitton, F., Caboche, M., Cannoot,
B., et al. (2004). Versatile gene-specific sequence tags for
Arabidopsis functional genomics: Transcript profiling and
reverse genetics applications. Genome Res. 14, 2176–2189.
31. Laloi, C., Mestres-Ortega, D., Marco, Y., Meyer, Y., and Reich-
held, J.P. (2004). The Arabidopsis cytosolic thioredoxin h5
gene induction by oxidative stress and its W-box-mediated
response to pathogen elicitor. Plant Physiol. 134, 1006–1016.
32. Bechtold, U., Murphy, D.J., and Mullineaux, P.M. (2004). Arabi-
dopsis peptide methionine sulfoxide reductase2 prevents cellu-
lar oxidative damage in long nights. Plant Cell 16, 908–919.
33. Valerio, L., De Meyer, M., Penel, C., and Dunand, C. (2004).
Expression analysis of the Arabidopsis peroxidase multigenic
family. Phytochemistry 65, 1331–1342.
34. Held, M.A., Tan, L., Kamyab, A., Hare, M., Shpak, E., and Kielis-
zewski, M.J. (2004). Di-isodityrosine is the intermolecular cross-
link of isodityrosine-rich extensin analogs cross-linked in vitro.
J. Biol. Chem. 279, 55474–55482.
35. Park, A.R., Cho, S.K., Yun, U.J., Jin, M.Y., Lee, S.H., Sachetto-
Martins, G., and Park, O.K. (2001). Interaction of the Arabidopsis
receptor protein kinase Wak1 with a glycine-rich protein,
AtGRP-3. J. Biol. Chem. 276, 26688–26693.
36. Woodward, A.W., and Bartel, B. (2005). Auxin: Regulation,
action, and interaction. Ann. Bot. (Lond.) 95, 707–735.
37. Gamas, P., Niebel Fde, C., Lescure, N., and Cullimore, J. (1996).
Use of a subtractive hybridization approach to identify new
Medicago truncatula genes induced during root nodule devel-
opment. Mol. Plant Microbe Interact. 9, 233–242.
38. Yang, B., Sugio, A., and White, F.F. (2006). Os8N3 is a host dis-
ease-susceptibility gene for bacterial blight of rice. Proc. Natl.
Acad. Sci. USA 103, 10503–10508.
39. Scheible, W.R., Fry, B., Kochevenko, A., Schindelasch, D.,
Zimmerli, L., Somerville, S., Loria, R., and Somerville, C.R. (2003).
An Arabidopsis mutant resistant to thaxtomin A, a cellulose
synthesis inhibitor from Streptomyces species. Plant Cell
15, 1781–1794.
40. Vorwerk, S., Somerville, S., and Somerville, C. (2004). The role of
plant cell wall polysaccharide composition in disease resis-
tance. Trends Plant Sci. 9, 203–209.
41. Bak, S., Tax, F.E., Feldmann, K.A., Galbraith, D.W., and Feyer-
eisen, R. (2001). CYP83B1, a cytochrome P450 at the metabolic
branch point in auxin and indole glucosinolate biosynthesis in
Arabidopsis. Plant Cell 13, 101–111.
42. Barlier, I., Kowalczyk, M., Marchant, A., Ljung, K., Bhalerao, R.,
Bennett, M., Sandberg, G., and Bellini, C. (2000). The SUR2
gene of Arabidopsis thaliana encodes the cytochrome P450
Cell-Wall-Integrity Sensing in Arabidopsis
931

CYP83B1, a modulator of auxin homeostasis. Proc. Natl. Acad.

Sci. USA 97, 14819–14824.
43. York, W.S., Qin, Q., and Rose, J.K. (2004). Proteinaceous inhib-
itors of endo-beta-glucanases. Biochim. Biophys. Acta 1696,
223–233.
44. Takeda, T., Furuta, Y., Awano, T., Mizuno, K., Mitsuishi, Y., and
Hayashi, T. (2002). Suppression and acceleration of cell elonga-
tion by integration of xyloglucans in pea stem segments. Proc.
Natl. Acad. Sci. USA 99, 9055–9060.
45. Crooks, G.E., Hon, G., Chandonia, J.M., and Brenner, S.E.
(2004). WebLogo: A sequence logo generator. Genome Res.
14, 1188–1190.
46. Sturn, A., Quackenbush, J., and Trajanoski, Z. (2002). Genesis:
Cluster analysis of microarray data. Bioinformatics 18, 207–208.

Accession Numbers

The data from the CATMA microarray experiment have been depos-
ited in the Complete Arabidopsis Transcriptome database (CATdb)
(http://urgv.evry.inra.fr/catdb) with the accession number RA05-
05_Procuste.