Download Cloning of a cDNA Encoding a Plasma Membrane

The Plant Cell, Vol. 8, 2265-2276, December 1996 O 1996 American Society of Plant Physiologists
Cloning of a cDNA Encoding a Plasma
Membrane-Associated, Uronide Binding Phosphoprotein
with Physical Properties Similar to Vira1 Movement Proteins
Philippe Reymond,a Béatrice Kunz,a Kalanethee Paul-Pletzer,a Rudolf Grimm,b Christoph Eckerskorn,c
E. Farmera,'
and Edward
lnstitut de Biologie et de Physiologie Végétales, Bâtiment de Biologie, Universite de Lausanne, 1015 Lausanne,
Switzerland
Hewlett-Packard, Hewlett-Packard Strasse, 76337 Waldbronn, Germany
Max-Planck-lnstitut für Biochemie, 82152 Martinsried, Germany
a
Oligogalacturonides are structural and regulatory homopolymers from the extracellular pectic matrix of plants. In vitro
micromolar concentrations of oligogalacturonatesand polygalacturonateswere shown previously to stimulate the phosphorylation of a small plasma membrane-associated protein in potato. lmmunologically cross-reactive proteins were
detected in plasma membrane-enriched fractions from all angiosperm subclasses in the Cronquist system.
Polygalacturonate-enhancedphosphorylation of the protein was observed in four of the six dicotyledon subclasses but
not in any of the five monocotyledon subclasses. A cDNA for the protein was cloned from potato. The deduced protein
is extremely hydrophilic and has a proline-rich N terminus. The C-terminal half of the protein was predicted to be a coiled
coil, suggesting that the protein interacts with other macromolecules. The recombinant protein was found to bind both
simple and complex galacturonides. The behavior of the protein suggests severa1 parallels with vira1 proteins involved
in intercellular communication.
INTRODUCTION
Unlike many types of animal cells, plant cells are surrounded
by a massive extracellular matrix that defines their shape and
prohibits their movement within the organism. Like the extracellular matrices of animal cells, plant cell walls contain important
regulatory information necessary for the control of defense and
developmental gene expression (Aldington and Fry, 1993; Cote
and Hahn, 1994) and probably for cell fate determination
(Berger et al., 1994). Among the regulatory molecules of the
plant cell wall, the best characterized are the oligogalacturonic
acids (OGAs) and polygalacturonic acids (PGAs). These molecules form a quantitatively important part of the pectic matrix
(ONeill et al., 1990) and are probably the most abundant polymeric regulatory molecules on the earths surface.
Originally, OGAs were discovered to be defense gene
regulators (Darvill and Albersheim, 1984; West et al., 1984;
Ryan, 1988); more recent evidence also suggests that they
have a role in development, including morphogenesis (Mohnen
et al., 1990). In particular, it seems that OGAs can affect severa1 processes influenced by or depending on the hormone
auxin (Bellincampi et al., 1996). Most OGA-responsivegenes
are sensitive only to a narrow range of OGAs whose degree
To whom correspondence should be addressed.
of polymerization (DP) is between 10 and 15 (Ryan, 1988). It
is likely that as a result of size and charge, larger OGAs (of
DP > 15) do not easily penetrate the plant cell wall under experimental conditions. Much regulatory information residing
in the plant extracellular matrix almost certainly remains undiscovered as a result of this restraint (Reymond et al., 1995).
Currently, all that is clear is that there is no single OGA regulator but rather a spectrum of unmodified and modified OGA
signals.
Although the biological activity of OGAs has been characterized in many systems (reviewed in Coté and Hahn, 1994),
little is known about how signaling from these molecules leads
to gene expression. In common with many other classes of
defense gene elicitors (Nürnberger et al., 1994), OGAs are
known to stimulate K+ efflux and increase cytosolic (H+] and
[Ca2+]in plant cells (Thain et al., 1990; Mathieu et al., 1991).
A G protein has been implicated in OGA signal transduction
(Legendre et al., 1992), as have inositol phosphates (Messiaen
and Van Cutsem, 1994).With the diverse effects of exogenous
OGAs on defense and development, the possibility for multiple signal pathways exists.
Binding sites for oligouronides appear to be of very low abundance on the plant cell surface (Diekmann et al., 1994).
Currently, no receptors for OGAs are known, and their
2266
The Plant Cell
discovery remains a challenge. The fact that OGAs form intermolecular complexes in solution hinders biochemical
approaches designed to characterize binding sites because
labeled pectic ligands bind to themselves (Kohn, 1985). Many
types of OGAlpectin binding proteins are expected to exist in
the plant cell. These include various proteins and enzymes
anchored to the pectic matrix, chaperone-likeproteins involved
in wall assembly or rearrangement, receptors for uronide signals, and perhaps other proteins involved in cell-to-cell
communication. In all these cases, it would be interesting to
isolate and characterize such uronide binding proteins from
plant tissues.
We have been studying the interaction of OGAs and PGAs
with proteins in isolated plasma membranes. Galacturonides,
at micromolar concentrations, enhance the in vitro threonine
phosphorylation of a small protein (pp34) that copurifies with
potato and tomato plasma membranes (Farmer et al., 1989,
1991). Large OGAs with a DP > 13 are necessary to stimulate
the phosphorylation of this protein in vitro (Reymond et al.,
1995). The OGA size requirement for stimulated phosphorylation, together with its association with the plasma membrane,
suggests that the protein may play an as yet unknown role in
signal transduction for larger uronides. We studied the distribution of the polypeptide and its in vitro phosphorylation in
all of the angiosperm subclasses defined by Cronquist (1988).
The results obtained encouraged us to attempt to clone the
protein with a view to investigating its potential direct interaction with uronides. The physical properties of the protein were
similar to those of vira1 cell-to-cell movement proteins, leading us to speculate that the protein somehow may be involved
in intercellular communication.
RESULTS
PGA Enhances the in Vitro Phosphorylation of a
Protein from Numerou Dicotyledonous Plants
The OGAIPGA-enhanced phosphorylation of a plasma membrane-associated protein allows the investigation of pectic
matrix-cell surface interactions. This rapid and simple assay
based on labeling proteins with radioactive phosphate permits
the study of the interaction of any uronide, modified or unmodified, with the plasma membrane (Farmer et al., 1991). The
PGA-stimulatedin vitro phosphorylation of plant plasma membrane-associated proteins has been reported in only three
species to date: tomato and potato from the subclass Asteridae (Farmer et al., 1989, 1991) and soybean from the widely
separate subclass Rosidae (Reymond et al., 1995).
We decided to test whether this phenomenon could be observed in plasma membrane preparations isolated from
throughout the Angiospermae. Therefore, we extended our
search to the other four dicotyledon (Magnolopsid) subclasses
and the five monocotyledon (Liliopsid) subclasses. In each
case, we chose at least one plant per subclass and attempted
to prepare plasma membranes from at least one tissue type.
We probed the plasma membrane preparations with the antipotato pp34 antibodies produced previously in rabbit granulomafluid (Jacinto et al., 1993). This was designed to indicate
whether the protein occurred in a particular plasma membrane
preparation. We also tested each membrane batch for enhanced in vitro phosphorylation in the presence of tomato leaf
PGA. Although not fully characterized, this preparation is the
uronide that is most active in the phosphorylation assay in
potato and tomato plasma membranes (Farmer et al., 1991;
Reymond et al., 1995).
Results of a typical experiment are shown in Figure 1. In
this case, plasma membrane preparationsfrom potato and buttercup leaves were shown to contain one major protein
cross-reactingwith the anti-potato pp34 antibodies (Figure 1C)
as well as the PGA-enhancedphosphorylation(quantitated with
a Molecular Imager; Figure 1B) of one protein of the same apparent molecular mass, ~ 3 kD
4 in the case of potato and ~ 3 1
kD in the case of buttercup (Figure 1A). All dicotyledon subclasses were found to contain proteins cross-reacting with the
anti-potato pp34 antibodies, as described in Table 1. However,
in the case of endive (Cichorium endivia, subclass Asteridae),
no cross-reactivity with the anti-potato pp34 antibodies was
observed. In Arabidopsis, avery weak cross-reactivity was observed in leaf plasma membranes, whereas a stronger signal
at -36 kD was detected in inflorescence plasma membrane
preparations.The apparent masses of dicot plasma membrane
proteins detected by the anti-potato pp34 antibodies fel1 within
a narrow range of -30 to 36 kD. In all five monocotyledon subclasses, cross-reactivity with anti-potato pp34 antibodies was
detected at apparent masses between 30 and 32 kD.
Having detected the presence of proteins cross-reactingwith
the anti-potato pp34 antibodies in all angiosperm subclasses,
we were curious to determine whether PGA-enhanced phosphorylation of these proteins was equally widespread. We
observed PGA-enhanced protein phosphorylation in plasma
membranes from the four dicotyledon subclasses, Magnoliidae, Hamamelidae, Rosidae, and Asteridae, but not in
subclasses Dilleniidae and Caryophyllidae (Table 1). PGAstimulated phosphorylation was quantitated by using a Molecular Imager. Between 1.5- and 8.4-fold increases in specific
protein phosphorylation were observed (Table 1). In every case
but one, endive, the apparent mass of the phosphorylated protein correlated with that of the protein recognized by the
anti-potato pp34 antibodies. In endive, no cross-reactive proteins were detected. In all monocotyledonous representatives
tested, spanning a large evolutionary spectrum of plants, no
PGA-enhanced in vitro protein phosphorylation was observed.
Cloning Remorin
To clone the cDNA encoding the potato plasma membraneprotein, we purified the polypeptide by following a previously
established protocol (Jacinto et al., 1993). The purification be-
Uronide Binding Protein
R
B
>
6
CO
UJ
Q
CD
UJ
LU
DC
R
R
Figure 1. Detection and Quantitation of PGA-Stimulated Protein Phosphorylation in Plasma Membrane Preparations from Potato and
Buttercup.
(A) In vitro phosphorylation of leaf plasma membranes (2 |jg of protein) from potato (S) and buttercup (R) in the presence (+) and absence
(-) of 0.2 ng of tomato leaf PGA. Samples were run on an SDS-polyacrylamide gel and analyzed by using a Molecular Imager. Bands
showing increased phosphorylation are indicated by arrowheads. Numbers at left refer to molecular mass standards (in kilodaltons).
(B) Quantitation of the ~35-kD band from (A) in the absence (closed
bars) and presence (open bars) of 0.2 ng of tomato leaf PGA. Band
intensity was integrated in samples of potato (S) and buttercup (R) leaf
plasma membranes
(C) Immunoblot of potato (S) and buttercup (R) leaf plasma membranes
(10 ng of protein) probed with the anti-potato pp34 antibodies. Bands
cross-reacting with the antibody are indicated by arrowheads.
2267
gan with 7 kg of potato leaves, from which 63 mg of plasma
membrane (protein) was isolated by two-phase partitioning.
The final yield of protein was too low to permit accurate quantitation; however, when comparing the density of known
amounts of Coomassie blue-stained BSA with that of the protein, we estimated a recovery of ~4 to 8 u.g. Initial N-terminal
sequencing was not possible as a result of blocking (data not
shown), and the remaining polypeptide (~150 pmol) was used
for internal sequencing. Eight peptides were successfully sequenced (Figure 2A). Peptide sequences were then used to
design polymerase chain reaction (PCR) primers that permitted us to isolate a 330-bp DMA fragment from a potato leaf
cDNA library. Using this PCR fragment as a probe, a full-length
cDNA was isolated from the same library. The correct cloning
of the cDNA was established by aligning sequenced peptides
with the cDNA: the eight peptides represent 41% of the protein (Figure 2A). Furthermore, the protein expressed in
Escherichia coli was recognized by the anti-potato pp34 antibodies (Figure 3).
The potato cDNA (Figure 2A) predicts a protein of 198 amino
acids, with a calculated mass of 21,769 D. The protein is particularly rich in glutamic acid (16.7%) and lysine (17%) residues.
Of the first 50 N-terminal residues, 11 (22%) are prolines. The
predicted pi of 6.14 agrees with our own measurement of the
potato leaf protein (data not shown). Notable features predicted
from the cDNA are four extremely hydrophilic domains (Figure 2B). No signal sequence for cotranslational insertion into
the endoplasmic reticulum is present. No features indicative
of transmembrane domains or membrane anchor signature
sequences were predicted, and no sequences indicative of
organellar localization were detected. The protein sequence
contains no N-glycosylation sites but has several potential serine and threonine phosphorylation sites. The C-terminal half
of the protein is predicted to be a coiled coil. The protein contains nine partially conserved sequences of seven amino acids
based on the sequence kA[E/D]ekak, where uppercase letters represent conserved residues and lowercase letters
represent the most common amino acid found in the sequence.
The full-length cDNA was found to be homologous to a cDNA
for an Arabidopsis DNA binding protein (Dbp; Alliotte et al.,
1989). The two proteins share 67% amino acid identity and
80% similarity (Figure 4). The protein is also homologous to
a rice expressed sequence tag. Comparison of the predicted
sequence of the potato protein with Arabidopsis Dbp and the
rice expressed sequence tag (Figure 4) indicates particularly
high homology in the central region of the proteins. This region is very hydrophilic and contains several pairs of charged
residues. Like the Arabidopsis Dbp, our protein migrates
anomalously ~12 kD above its predicted mass.
We propose the name remorin for this potato protein. A
remora is a fish that attaches itself to the surface of other larger
organisms; the name alludes to the interaction of the protein
with the plasma membrane and with uronides (discussed below) but does not imply a function. The designation used
previously, pp34, is no longer useful because it suggests a
mass much higher than the calculated mass (21.8 kD) and
2268
The Plant Cell
Table 1. Presence and PGA-StimulatedPhosphorylation of Remorin in Angiosperms
~
Subclass
Family
Species
Magnoliidaee
Hamamelidaee
Caryophyllidaee
Dilleniidaee
Ranunculaceae
Fagaceae
Chenopodiaceae
Brassicaceae
Ranunculus ficaria
Fagus sylvatica
Spinacia oleracea
Arabidopsis thaliana
Rosidaee
Asteridaee
Fabaceae
Solanaceae
Pisum sativum
Alismatidae'
Arecidae'
Commelinidae'
Liliidae'
Zingiberidae'
Asteraceae
Hydrocharitaceae
Areceae
Gramineae
Asphodelaceae
Marantaceae
~
~
~
~~
A n t 1 - p ~ ~ ~Relative
CrossMass
PGA-Stimulated
Tissuea Reactivityb (kD)C
Phosphorylationd
L
L
L
L
I
L
Solanum tuberosum
Lycopersicon esculentum
Cichorium endivia
€geria densa
Monstera deliciosa
Zea mays
Chlorophytum elatum variegatum
Ctenanthe oppenheimiana
L
L
L
L
L
L
L
L
Y es
Yes
Y es
Weak
Yes
Y es
Yes
Yes
No
Weak
Y es
Yes
Yes
Y es
31
31
30
36
36
34
34
32
28
30
32
32
31
30
1.5
3.8
O
O
O
8.4
5.3
2.5
2.0
O
O
O
O
O
For growth conditions, see Methods. L, leaf; I, inflorescence
The antibody used is characterized in Jacinto et al. (1993).
Relative mass of a cross-reactive polypeptide that was recognized by the anti-potato pp34 antibody on an immunoblot. In all cases except
for Cichorium endivia, the cross-reactive band corresponded to the PGA-stimulated phosphoprotein.
Fold increase in phosphorylation in the presence of tomato leaf PGA.
e Class Dicotyledoneae.
Class Monocotvledoneae.
a
'
because the protein from other species migrates during electrophoresis within a relatively broad range of masses.
Remorin Binds Uronides
To understand the uronide-enhanced in vitro phosphorylation
of remorin, we decided to test whether the interaction between
uronides and the protein was direct or indirect. To test for a
direct interaction between pectic polysaccharides and remorin, we developed a binding assay based on biotinylated
uronides. We employed a homogeneous OGA with a DP of
20, purified according to Spiro et al. (1993), and a fractionated
pectin (G-50 PIIF; Bishop et al., 1984). These uronides were
labeled at the reducing ends with biotin and purified by affinity chromatographyon avidin columns to remove unlabeled
uronides (see Methods). The number of biotin molecules per
uronide was not quantitated, but it is likely to be one because
labeled OGAs retain their discrete mobilities close to the parent OGA in gel electrophoresis (data not shown). Remorin was
expressed as a fusion protein with glutathione S-transferase
(GST) in E. co/i(Figure 5A). To test whether GST-remorin could
bind uronides, the fusion protein-containing bacterial extracts
were electrophoresed, blotted onto nitrocellulose, and probed
with a biotinylated pectic fragment. Biotin-horseradish peroxidase does not bind to GST-remorin fusion protein, nor does
biotin displace biotinylated OGA bound to GST-remorin (data
not shown).
Figure 5 shows that GST-remorin binds to both eicosagalacturonic acid (Figure 56,lane 1) and fractionated pectin (Figure
5B, lane 2). As a control, extracts containing overexpressed
GST were also probed in the same manner with the biotinylated
pectic fragment (Figure 58, lane 3). No uronide binding proteins were found in these extracts. GST-remorin was not labeled
when protein blots were treated simply with avidin-horseradish peroxidase in the absence of ligand (Figure 58, lane 4;
see Methods). To demonstrate clearly that labeled uronides
bind to GST-remorin in the absence of other proteins, we performed a similar experiment with purified GST-remorin. The
fusion protein was probed with biotinylated fractionated pectin. This ligand bound to purified GST-remorin (Figure 6),
whereas no signal was detected with purified GST.
Table 2 presents the results of preliminary competition
studies that were performed in an effort to begin to understand
the nature and specificity of ligand binding to remorin. Ligands
chosen as competitors were a series of uronides that included
tri-a-l,4-galacturonic acid and poly-p-l,4-mannuronic acid,
which are inactive in stimulating the phosphorylation of remorin (Farmer et al., 1991; Reymond et al., 1995), as well as
poly-~-a-l,4-guluronicacid, a molecule closely related to polyu-1,4-galacturonic acid. A number of other polyanions were
tested, including carboxymethylcellulose,poly-L-glutamate, and
heparin. Molecules that were ineffective in competing with
fractionated pectin were poly-a-1,4-guluronic acid, tri-a-l,4galacturonic acid, poly-&l ,Cmannuronic acid, carboxymethylcellulose, and poly-i-glutamate. Fractionated pectin and the
Uronide Binding Protein
2269
1 2 3
i
TTACCTTCCTCCAACTACTAAATATCTTCTTCTTTGAAGAATTCTCTGTTTTCTTGATTC
61
TGTTTGTAGCCATGGCAGMTTGGAAGCTAAGAAAGTAGAAATTGTGGACCCTGCACCCC
M A E L E A K K V E I V D P A P P
121
CTGCGCCAGGACCTGTTGAAGCTCCTAAAGAAGTGGTGGCTGATGAGAAAGCCATAGTTG
A P G P V E A P K E V V A D E K A I V A
181
CACCAGCTCTGCCTCCTCCTGCAGAAGAAAAAGAAAAACCCGATGACTCGAAAGCATTAG
P A L P P P A E E K E K P D D S K A L V
241
TTGTCGTTGAAACTAAAGCACCAGAACCTGCTGATGAGAAAAAAGAGGGATCTATTGACA
V
V
E
T
K
A
P
E
P
A
D
E
K
K
E
n
.<!
I
P
GAGATGCTGTGCTTGCTCGCGTTGCAACAGAGAAGAGAGTATCACTCATCAAAGCATGGG
D A V L A H V A T E K R V S L I K
361
AGGAAAGTGAGAAATCAAAAGCCGAAAACAAAGCTCAGAAGAAGGTATCTGCAATTGGTG
421
CATGGGAGAACAGCAAGAAAGCTAACCTAGAGGCTGAGCTCAAAAAGATGGAGGAACAGT
M E N S K K A N L E A E L K K M E E O L
481
TGGAGAAAAAGAAGGCCGAATATACTGAGAAAATGAAAAACAAAATTGCTCTACTCCACA
E
K
K
K
S
A
K
E
A
Y
E
T
N
E
K
K
A
M
Q
K
K
N
K
V
K
I
S
A
A
L
I G
L
H
K
AGGAAGCAGAAGAAAAGAGAGCGATGATTGAAGCTAAACGTGGAGAAGATCTTCTCAAGG
E A E E K R A M I E A K
R G E D L L K A
601
CAGAGGAGCTTGCAGCAAAATACCGTGCCACTGGAACTGCTCCAAAGAAAATCCTTGGAA
661
TATTTTGAAGCAGCAAGCACCAGGTCTGCATCGGTAGTGATTGGGAGTTACATnTGTAA
721
781
841
AGTTTGTGCAATTGGAATTTTGTTTCTTGTTAGATTACATTTGTGTGATTATGTATTTTA
F
901
E
L
A
A
K
Y
R
A
T
G
T
A
P
K
K
I
L
G
51
36 ». ———
29
A
541
E
83
R
301
£_S
102
I
*
Figure 3. Recombinant Remorin Is Recognized by the Anti-Potato
pp34 Antibodies.
A protein gel blot was developed with the anti-potato pp34 antibodies. Lane 1 contains the positive control, potato plasma membrane;
lane 2, extract from E. co//-containing expression vector pGEX2T; lane
3, the E. co//-expressing remorin-GST fusion in pGEX2T. In all cases,
20 ng of protein was probed. The arrowheads show cross-reactive remorin. Numbers at left refer to molecular mass standards (in kilodaltons).
GAACCATTTATTGTTTATTGTTACGTGTGCATAGTGATGTATTTCCAGTGTATATAGCAC
CTGGACAAATTAACTTTGTGGGATTGTATGAAAAAAAATGTTGAAGGAAATCTTCATGTT
AGTACACAACTCTTGCAGAAAAAAAAAAAAAAAAAAAAAAAAAAAA
related molecule tomato leaf PGA reduced pectin binding by
^50%. Only a large excess (1000-fold) of tomato leaf PGA completely abolished binding of labeled pectin (data not shown).
Heparin appeared to fully displace fractionated pectin. DNA
and RNA were weak competitors of pectin binding to
GST-remorin, reducing pectin binding by only 5 and 8%,
respectively. A single-stranded DNA was completely inactive
in reducing labeled uronide binding in the competition assay
(Table 2).
B
50
100
150
200
Amino acid
Figure 2. Deduced Sequence of Remorin.
(A) The cDNA sequence and the derived amino acid sequence of potato
remorin. The predicted amino acid sequence is provided below the
corresponding nucleic acid sequence. The asterisk indicates the stop
codon. The GenBank accession number is U72489. Peptides found
by sequencing the plant protein are underlined.
(B) Hydropathy analysis of the remorin sequence. A Kyte-Doolittle
hydropathy plot (Kyte and Doolittle, 1982) of the predicted amino acid
sequence was generated by using DNA Strider software (C. Marck,
Department of Cell and Molecular Biology, Gif-sur-Yvette, France), using
a window value of 11. Increased hydrophilicity is indicated by negative values.
DISCUSSION
The uronide-stimulated phosphorylation of plasma membrane
proteins may provide useful clues to the nature of the interaction of part of the plant extracellular matrix with plasma
membrane-associated proteins. Is the uronide-stimulated
phosphorylation of proteins widespread, or is it an exception
observed in a few distinct species? In this study, we observed
the stimulated in vitro phosphorylation of remorin-like proteins
in plasma membranes isolated from four subclasses of
dicotyledons: Asteridae, Rosidae, Magnoliidae, and Hamamelidae (Table 1). In only two subclasses of dicotyledons,
Dilleniidae and Caryophyllidae, were we unable to detect PGAenhanced phosphorylation. Thus, the interaction between uronides and remorin is widely conserved in dicotyledonous
plants.
How do these results correlate with what we know of biological responses to uronides in plants? Biological responses to
OGAs have been reproducibly observed in three of the six
2270
The Plant Cell
remorin
DbP
rice EST
26
29
13
remorin
Dbp
rice EST
56
47
37
remorin
Dbp
rice EST
85
77
60
remorin
Dbp
rice EST
remorin
Dbp
rice EST
g
H
t
mN
lw:8:E:m::ISBBHM:E
e -v -.
A
a
-
-
-
W
r
l
T
*
a
t
g
145
137
104
remorin
DbP
remorin
DbP
Figure 4. Remorin 1s Related to the Arabidopsis Dbp and to a Rice Expressed Sequence Tag
Sequence alignment of the Arabidopsis Dbp (Alliotte et al., 1989) and a rice expressed sequence tag (rice EST; GenBank accession number
D40002)with potato remorin are shown. ldentical amino acids are boxed with black squares, and conserved amino acids are shown in lowercase
letters. The asterisk indicates the end of the partia1 rice expressed sequence tag. Dashes indicate gaps to maximize alignment.
subclasses of dicotyledonous plants: Asteridae, Rosidae, and
Dilleniidae (Coté and Hahn, 1994). In the case of Dilleniidae,
biological responses to uronides have been characterized in
cucumber (Robertsen and Svalheim, 1990) and cotton (Davis
and Essenberg, 1995). Currently, we are unaware of any published data concerning the biological effects of OGAs on
Arabidopsis. Failure to observe PGA-enhancedphosphorylation of the protein in Arabidopsis (Table 1) and in a second
species from the Dilleniidae, cauliflower (data not shown), may
reflect genuine differences in the physiology of these plants
from other representatives of the Dilleniidae and other subclasses. However, this seems to be unlikely and needs further
investigation. Trivial reasons, such as suboptimal assay conditions and/or lability of the system, are equally plausible. The
uronide ligand we used, tomato leaf PGA, was probably an
appropriate choice because the Arabidopsis cell wall, like that
of tomato, appears to be a typical type I wall (Zablackis et al.,
1995).
The other dicotyledon subclass in which no PGA-stimulated
in vitro phosphorylation was observed was the Caryophyllidae; we chose spinach as representative. Spinach leaves
contain a protein of apparent mass m30 kD that is recognized
by the anti-potato pp34 antibodies (Table 1). We are unaware,
however, of any reports of biological activity of OGAs in the
Caryophyllidae. In summary, it is clear that remorin-like pro-
teins are widespread in dicotyledons and that in four of six
dicotyledon subclasses, in vitro PGA-enhanced phosphorylation of this protein was observed.
After examining the presence proteins that cross-react with
the anti-pp34 antibodies and their phosphorylation in
dicotyledons, we turned to the other major division of the
angiosperms, the monocotyledons. Proteins that cross-react
with the anti-potato pp34 antibodies are found in each monocotyledon subclass (Table 1); however, no PGA-enhanced
protein phosphorylation was observed in any monocotyledonderived plasma membrane. Our choice of monocotyledons
representing all five subclasses in the Cronquist system included plants as diverse as dense-leaved pondweed, €geria
densa (subclass Alismatidae), representing more ancient
monocotyledons, and maize (subclass Commelinidae), a modern monocotyledon with a type II cell wall. We also tested a
second ligand, maize leaf PGA (from a type II cell wall), in the
phosphorylation assay. This preparation was inactive in our
assays of monocotyledon-derived plasma membranes but
functioned in dicotyledon plasma membranes (data not shown).
In monocotyledons, no confirmed biological responses to
OGAs have been reported. Thus, from our perspective, it
appears that there may be real differences between monocotyledons and dicotyledons regarding in vitro and in vivo
responses to OGAs and PGAs. Monocotyledons, however, have
Uronide Binding Protein
almost certainly received less attention than dicotyledons with
respect to signal transduction studies, and it is premature to
conclude that monocotyledons will not respond to exogenous
OGAs. Moreover, biological activity of OGAs has been reported
in the Gymnospermae (Asiegbu et al., 1994), and it will be
worthwhile to test for the presence of remorin in these plants.
In summary, data on the occurrence of uronide-enhanced
remorin phosphorylation in dicotyledons correlate broadly with
the fact that to date, only this class of angiosperms is known
to display biological responses to these molecules. These considerations suggest a potential role of the remorin in the
biological acitivity of OGAs and convinced us to try to clone
the remorin cDNA.
Remorin Is a Hydrophilic, OGA Binding Protein That Is
Associated with the Plasma Membrane
kD
2271
1 2 3 4 5 6
105
82
49
33
29
Figure 6. Purified GST-Remorin Binds Fractionated Pectin.
A protein gel blot containing affinity-purified GST-remorin (lanes 1,
3, and 5) and affinity-purified GST (lanes 2, 4, and 6) as a control was
probed with either biotinylated fractionated pectin (30 nM; lanes 1 and
2). the anti-potato pp34 antibody (lanes 3 and 4), or an anti-GST antibody (lanes 5 and 6). Each lane contains 5 ng of protein. The positions
of the recombinant proteins are indicated with arrowheads. Numbers
at left refer to molecular mass standards (in kilodaltons).
A cDNA for remorin was cloned from a potato leaf cDNA library. Because the predicted protein has a mass of 22 kD and
kD
1
2
105
82
-=
49 *" ~~~
33
29
B
kD
102
83
51
36
,
29
Figure 5. Binding of Simple and Complex Galacturonides to Bacterial
Extracts Containing Recombinant GST-Remorin
(A) GST-remorin (lane 1) and GST (lane 2) proteins were expressed
in E. co//. Total soluble proteins (20 ng per lane) were separated by
SDS-PAGE and stained with Coomassie blue. The positions of recombinant proteins are indicated by arrowheads. Numbers at left refer to
molecular mass standards (in kilodaltons).
(B) GST-remorin (arrowhead) was expressed in £. co//. Bacterial extracts (20 n9 of soluble protein per lane) were probed with either
biotinylated oligogalacturonide (DP of 20; 30 nM; lane 1), biotinylated
fractionated pectin (30 nM; lane 2), or in the absence of ligand (lane
4) As a control, bacterial extracts expressing GST alone were probed
with biotinylated fractionated pectin (30 nM; lane 3). Numbers at left
refer to molecular mass standards (in kilodaltons).
the plant protein has a relative mass of 35 kD, we were concerned that we had sequenced another protein. This was not
the case because all eight recovered peptides correspond to
the deduced remorin sequence, and the protein expressed in
E. coli was recognized by the anti-potato pp34 antibodies (Figure 3) as well as by an antibody raised against an internal
peptide predicted from the remorin cDNA (data not shown).
Another strong indication of correct sequencing is the identical electrophoretic behavior of remorin and Arabidopsis Dbp
(Alliotte et al., 1989). Both proteins migrate at ~34 kD, although
their predicted mass is ~22 kD.
The hydrophilicity of potato remorin (Figure 2B) is remarkable. This is surprising in view of the relatively tight association
of the protein with plasma membranes. Nothing in the sequence data provides clues as to the nature of this association.
We have examined the interaction of remorin with plasma membranes in a variety of ways, including washing the membranes
with sodium carbonate solutions, which is an established
method for removing peripheral proteins (Fujiki et al., 1982).
Low concentrations of sodium carbonate (10 mM) were insufficient to solubilize the protein, whereas 100 mM sodium
carbonate solubilized ~30°/o of the protein (data not shown).
We would have expected a higher efficiency of solubilization
of the protein by sodium carbonate if it were a peripheral membrane protein and somewhat less if the protein were integral
(Fujiki et al., 1982). Detergents were found to be the best means
of solubilizing remorin: concentrations of at least 0.5% (v/v)
Triton X-100 are necessary to remove the protein from the
plasma membrane.
A striking feature of remorin is the proline-rich N terminus.
The configuration of prolines in this region is reminiscent of
sites involved in protein-protein interactions; for example, those
found in signaling proteins having SH3 binding domains
(Pawson, 1995). In this case, the motif X-P-p-X-P, where X tends
to be an aliphatic residue and p is a semiconserved praline,
is found at residues 15 to 19 (APPAP) and 37 to 41 (APALP)
2272
The Plant Cell
Table 2. Binding of Biotinylated Pectin to GST-Remorin in the
Presence of Various Polyanions
Polyanion
No competitor
Fractionated pectin
Tomato leaf PGA
a-1,4-D-Trigalacturonicacid
a-l,4-~-Polyguluronicacid
~-1,4-~-Polymannuronic
acid
Carboxymethylcellulose
Poly-L-glutamate
Heparin
Single-stranded DNA (26 nucleotides)
DNAb
RNAC
Binding
Activity (VO)"
1O0
45
64
1O0
1 O0
1 O0
1O0
1O0
O
1 O0
95
92
Binding of biotinylated fractionated pectin (30 nM) to GST-remorin (see Figure 5B) was determined in the presence of a 500-fold excess of competitor. The intensity of the remorin band was integrated
by using a Molecular Imager.
One milligram of low molecular weight DNA from herring.
One milligram of total yeast RNA.
a
(Figure 2A). Whether these motifs play any role@)in the association of remorin with other molecules, for example,
signaling proteins, remains to be tested. The C-terminal half
of the protein is predicted to be a coiled coil. These structures
tend to imply protein-protein interaction (Branden and Tooze,
1991)but may also be involved in binding polysaccharides,
such as heparin (Margalit et al., 1993).Based on the predicted
structure of remorin (the SH3-like domains [Pawson, 1995)and
the coiled coil structure), we believe that the protein interacts
with other macromolecules.
To investigate the potential interactionof uronides with remorin, we decided to test the ability of the recombinant protein
to bind to oligogalacturonides. During the course of our studies,
we never observed an enzymatic activity of remorin toward
PGA or pectin-derived polymers; thus, the protein is unlikely
to degrade the galacturonic acid-containing polymers we used
as probes. Binding of biotinylated uronide fragments to the
GST-remorin fusion was observed (Figure 58). We chose two
ligands that represented either simple OGAs (a fragment with
a DP of 20; eicosagalacturonic acid) or fractionated pectin
(Bishop et al., 1984),a molecule likely to be representative of
structures existing in the extracellular matrix. A purified
GST-remorin fusion protein also binds uronides (Figure 6), indicating that other components of the E. coli extracts are not
responsible for the binding.
lnitial competition studies were performed with crude bacteria1extracts containing the GST-remorin fusion protein. The
results obtained (Table 2) were complex, and only limited conclusions can be drawn. Tomato leaf PGA and fractionated pectin
were only partially effective in reducing the binding of the biotinylated pectic probe. Why were these ligands not better
competitors? The reason is not clear, but the most probable
explanation comes from the established fact that OGAs readily interact with themselves to form intermolecular complexes
(Kohn, 1985).If intermolecular complex formation occurs in
the binding assays, the competing ligand might interact with
the bound label without displacing it. That is, the remorin itself may bind to intermolecular PGA complexes as well as to
the free ligand. The possibility that intermolecular complexes
of OGA are themselves regulatory ligands has already been
proposed (Farmer et al., 1991).
Heparin seems to be a strong competitor for the pectic ligand.
Direct interaction between heparin and remorin would be consistent with a previous report of the heparin in stimulated in
vitro phosphorylation of remorin (Farmer et al., 1991).A PGA
binding peroxidase was recently shown to bind tightly to heparin, whereas other peroxidases that did not bind to PGA also
did not bind to heparin (Pene1 and Greppin, 1996).
As demonstrated previously with the Dbp protein (Alliotte
et al., 1989),labeled DNA was found to bind to remorin (data
not shown). The failure of DNA and RNA to displace labeled
uronides (Table 2) may suggest that remorin has two separate
binding sites for these molecules, but again, the complexity
of binding studies as a result of ligand-ligand interaction (Kohn,
1985)necessitates further work. The copurification of remorin with plasma membranes from all subclasses of
angiosperms (Table l), together with our inability to detect the
protein in potato and tomato nuclei by immunoblotting (data
not shown), leads us to expect that the principal cellular pools
of remorin in the tissues we have examined are unlikely to be
associated with DNA. In summary, the competition results provide a correlation between the ability of closely related
molecules to stimulate in vitro phosphorylation of remorin and
their ability to bind the recombinant protein. Most importantly,
this study indicates that the interaction between remorin and
OGAs is direct. The direct binding of OGA to remorin was not
expected because typically it is protein kinases that are ligandactivated, not their substrates. Although the significance of
ligand-stimulated phosphorylation of remorin is not known,
there is no reason that such mechanisms may not exist or play
roles in regulation. Severa1 possible mechanisms whereby
OGA binding might lead to increased phosphorylation of remorin can be imagined. OGA might cause a conformational
change in remorin, leading to its interaction with a protein kinase and subsequent phosphorylation. We cannot rule out
the simultaneous binding of OGA to remorin and to its kinase,
which at present remains uncharacterized.
How does remorin bind uronides? To examine whether the
predicted remorin protein sequence provided clues, we compared this sequence with the crystal structure of a protein
known to bind PGA: pectate lyase C (PelC) from the bacterial
plant pathogen Erwinia chrysanthemi (Yoder and Jurnak, 1995).
Uronide Binding Protein
remorin
PdC
D-2-R-5-KKVX-36-EIKE-22-KK-18-K-5-R-l-D-25-
rm"n
-E-lO-K-19-KK-3O-R-6-R-2l-K
PdC
-D- 9-K-16-KK- 2 7-R-4 -R-2 1-R
KKVE-34-EEKE-21-KK-16-K-5-K-2-E-
2-
Figure 7. Alignment of the PelC PGA Binding Residuesand Predicted
Remorin Amino Acid Sequence Shows Similarity.
Amino acid residues thought to be involved in uronide binding in PelC
(Yoder and Jurnak, 1995) are compared with residues in the remorin
primary sequence that show similarity in spacing. Otherwise, PelC
and remorin show no overall sequence homology.
Residues thought to be important for uronide binding have
been identified in this enzyme. We were surprised to find a
similar distribution of these residues in remorin and in the PelC
PGA binding groove (Figure 7). Although the two proteinsshow
no obvious sequence homology, the charged residue placement and spacing indicate that these residues may represent
convergently evolved uronide binding sites. Directed mutagenesis of the remorin sequence, together with structural studies,
is necessary to test whether the charged residues indicated
in Figure 7 are dispensable for the binding of uronides to remorin. It is noteworthy that PelC is a (3-strand protein (Yoder and
Jurnak, 1995). whereas remorin is predicted to be a-helical.
This profounddifference in structure is suggestiveof very different biological roles for the two proteins.
A Potential Role for Remorin
Remorin copurifies tightly with plasma membranes.This membrane association would not be predicted from the cDNA
encoding a hydrophilic protein. The protein is phosphorylated
in vitro (Farmer et al., 1991) and in vivo (E.E. Farmer, unpublished data) and can bind polyanionic ligands. We are aware
of only one other protein reported in the literature that clearly
displays these same features: the 30-kD cell-to-cell movement
protein of tobacco mosaic virus (TMV-MP), a protein involved
in macromolecular trafficking through plasmodesmata (Lucas
and Gilbertson, 1994). Despite the absence of predicted
membrane-spanning regions, the TMV-MP associateswith the
plasma membrane like an integral membrane protein (Moore
et al., 1992). This behavior is very similar to that of remorin.
TMV-MP binds single-stranded DNA (Lucas and Gilbertson,
1994) and thus is similar to remorin, which binds severa1polyanions. The TMV-MP can be phosphorylated in vitro in the
presence of cell wall-associated proteins (Citovsky et al., 1993)
under conditions similar to those we used to phosphorylate
remorin. Citosky et al. (1993) propose that the phosphorylation of TMV-MP is a mechanism of deactivating the protein by
sequestering it in cell walls. By analogy, we could imagine that
2273
the phosphorylation of remorin is involved in localization into
the wall. At this stage, however, we must also consider phosphorylation as a possible regulatory mechanism involved in
OGA signaling. Despite no apparent sequence homology between the two proteins, the intriguing similarities discussed
above lead us (tentatively) to a hypothesis that we are testing.
The apparently hydrophobic behavior of remorin could be
explained by its association with plasmodesmata or similar
structures present in our plasma membrane preparations. Our
working hypothesis is that remorin is a plasmodesmaassociated protein potentially involved in cell-to-cell signaling
and/or molecular transport. The interaction with uronides, and
particularly their effect on remorin phosphorylation, provides
candidatesignaling molecules with which to probe these potentia1 functions. To test this hypothesis as well as to investigate
other possible roles of remorin, transgenic plants with altered
levels of this protein are envisaged. Recombinant remorin will
also be employed as a tool for the isolation of its kinase(s),
with the long-term goal of the in vitro reconstruction of what
may be part of the plant cell's (inter)cellular signaling
mach inery.
METHODS
Plants
The classification used was described by Cronquist (1988) and has
been revised (Soltis and Soltis, 1995). Ranunculus ficaria (creeping
buttercup) and Fagus sylvatica (beech) came from wild populations
in Lausanne. Pisum sativum (garden pea), Solanum tuberosum cv
Desirée (potato), Lycopersicon esculentum cv Bonny Best (tomato),
Cichorium endivia (endive), and Zea mays (maize) were grown under
standard greenhouse conditions (Farmer et al., 1991). Arabidopsis
thaliana (mouse-earcress) was grown in a 16-hr-lightl8-hr-darkcycle
at 26OC/18OC in a growth chamber. Spinacia oleracea (spinach) came
from a local market, and €geria densa (dense-leavedpondweed) came
from an aquarium supplier. Monstera deliciosa (cheese plant), Chlorophytum elatum variegatum (spider plant), and Ctenanthe
oppenheimiana (ctenanthe)were grown under tropical climate conditions (>95% humidity) with no supplementary lighting.
In Vitro Phosphorylation
Plasma membrane was purified by zona1 centrifugation followed by
the dextran-polyethyleneglycol two-phase partition method (Yoshida
et al., 1983),with both polymers at a concentration of 5.9% (wlv). Potato
and tomato plasma membranes were previously characterized(Farmer
et al., 1991). Plasma membranes purified from other species were not
biochemically characterized. Plasma membrane was thiophosphorylated at 3OoC (Farmer et al., 1991) by using y-35S-adenosine-5'
thiophosphate (Y-~~S-ATP).
To ensure that the use of thiophosphate
did not hinder uronide-stimulated protein phosphorylationin plasma
membranes from monocotyledons,an alternative substrate, y-=P-ATP,
was used. In each case, comparable results were obtained, and all
data presented are for Y-~~S-ATI?
Tomato leaf polygalacturonate used
2274
The Plant Cell
in the phosphorylation assay was the fraction TFA-PIIF, as described
by Bishop et al. (1984). Uronides were quantitated according to
Blumenkrantz and Asboe-Hansen (1973). Proteins were quantitated
with the BCA reagent from Pierce Corporation (Rockford, IL).
Protein Purification and Sequencing
One- to 2-month-old greenhouse-grownpotatoplants(cv Desirée)were
used as a source of plasma membrane (Farmer et al., 1991). Plasma
membranevesicles (correspondingto 63 mg of protein)were prepared
from 7 kg of leaves.Remorinwas purified from the plasma membranes,
as described by Jacinto et al. (1993), and the partially purified protein
was electrophoresed in a 12%SDS-polyacrylamide gel. After Coomassie Brilliant Blue R 250 staining and destaining, the -34-kD band
(apparentmass) was excisedand exhaustivelywashed with water. The
band was then minced in an Eppendorf tube and dehydratedto near
completion using a Speed-vac (Zivy, Basel, Switzerland). Digestion
was performed by the addition of an endoproteinase Lys-C solution
for 16 hr. For sequencing, automated Edman degradation was performed using a protein sequencing system (model G1005A;
Hewlett-Packard,Palo Alto, CA).
were usedto amplify the completeremorin coding region. The BamHIEcoRI-amplified fragment was cloned into the expression vector
pGEX2T (Pharmacia Biotech) digested with BamHl and EcoRI. The
fusion protein was produced in Escherichia coli JM109 cells, according to Bernasconi et al. (1994). An overnight culture of E. coli cells
that carried pGEX-remorin was diluted 10-foldin 250 mLof Luria-Bertani
broth, and the cells were grown at 37OC with vigorous shaking. After
1 hr, isopropyl P-D-thiogalactopyranosidewas added to a final concentration of 0.5 mM. After incubation for another 6 hr at 3OoC,cells
were collected by centrifugation, resuspended in 3 mL of extraction
buffer (50 mM Tris-HCI. pH 8.0, 5 mM MgC12, 2.5 mM DTT, 100 mM
NH,CI, 20% glycerol), subjected to five cycles of freezing and thawing, and sonicated.The cell lysate was centrifuged to removebacterial
debris, and proteinsfrom the supernatant were analyzed by SDSPAGE.
When necessary, the supernatant was applied to a 5-mL glutathione
Sepharose 48 column (Pharmacia Biotech).The retained fusion protein, GST-remorin, was eluted with 10 mM glutathione,50 mM Tris-HCI,
pH 8.0, and concentrated by using Centricon-10concentrators (Amicon). The purity of GST-remorin preparation was assessed by
SDS-PAGEand silvei staining of the gel. As controls, crude and affinitypurified preparations of GST were obtained by following the same
protocol.
cDNA Cloning
Binding Studies
Polymerasechain reaction (PCR) of DNA from a potato leaf (cv Bintje)
cDNA iibrary (in A. ZAP; Stratagene;a kind gift from D. Johnston, Federal Agricultura1 Research Station of Changins, Nyon, Switzerland)
was performedusing oligonucleotideprimersof low degeneracy based
on the codon usage preference of potato (from J.M. Cherry, Harvard
University,Cambridge, MA).The reaction mixture consisted of 10 mM
Tris, pH 8.3, 50 mM KCI, 1.5 mM MgCI2,0.2 mM each deoxynucleotide triphosphate, 20 milliunits of Taq polymerase per pL, and 1 pM
each primer. Cycling conditions were as follows: 2 min at 94OC then
1 min at 94OC, 1.5 min at 46OC and 1.5 min at 72OC for 35 cycles, then
2 min at 72OC. A PCR product of 330 bp was obtained by using the
primers 5'-GTTGAAATTGTKGAY-3'and 5'-TCCCAAGCWCCAATW-3:
where K IS T or G, Y is T or C, and W is T or A. The primers were
derived from the sequence VEIVDP and AIGAWE of the sequenced
peptides KVEIVDPAPPAPGPVEAand VSAIGAWENSK, respectively
(Figure 2A). The PCR product was cloned into the pGEM-T vector
(Promega. Madison. WI) and sequenced using the T7 Sequencing
Kit (PharmaciaBiotech, Uppsala, Sweden), according to the manufacturer's instructions. The partia1 PCR product was labeled with
digoxigenin(DIG system; BoehringerMannheim. Mannheim,Germany)
and used to probe the potato cDNA library. Labeling, hybridization,
washing, and detection by chemiluminescence were performed according to the supplier's instructions. 60th strands of positive cDNA
clones were sequenced in their entirety. Sequence alignments and
analyses were performed using software from Genetics Computer
Group(version 8; Madison, WI; Devereux et al.. 1984)and the BLAST
network service from the National Center for Biotechnology Information (Altschul et al.. 1990). Coiled coil prediction was done by using
Coils (version 2.2) program (A. Lupas, Max-Planck-lnstitut für Biochemie, Martinsried. Germany).
Expression of Glutathione S-Transferase Remorin
To generate a glutathione S-transferase (GST)fusion gene, PCR primers
containing additional BamHl or EcoRl restriction sites at their 5'end
Uronides used in binding studies included purified oligogalacturonic
acids (OGAs) with a degree of polymerization (DP) of 20. The OGAs
were purified according to Spiro et al. (1993) and as described in
Reymond et al. (1995). Fractionated pectin (G5O-PIIF; Bishop et al.,
1984)was a gift from C.A. Ryan(WashingtonState University, Pullman).
Uronideswere biotinylated by reductive amination (Prakash and Vijay,
1983). A solution of uronide (500 pg) in dimethylformamide (360 pL)
containing 0.5 M HCI (10 pL), glacial acetic acid (10 pL), and biotinLC-hydrazide (200 pg; Pierce) was incubated at 4OoC for 3 hr.
NaCNBH3(20 pL; 0.1 M in dimethylformadide)was then added to the
reaction mixture, and incubation at 4OoC was continued for 16 hr. The
reaction mixture was then dialyzed against two changes of 5 L of water in 1000 D molecularweight cut-offdialysistubing. Labeleduronides
were purified on a monomeric avidin column (Pierce) according to the
manufacturer's instructions. The eluate from this column was dialyzed
first against 2 L of PBS and then exhaustively against distilled water.
Electrophoretic separation of the labeled ligands (Farmer et al., 1991)
was used to asses their purity. The uronide content of the purified
ligands was estimatedaccording to Blumenkrantzand Asboe-Hansen
(1973).
Binding assays were performed based on Silva et al. (1987). Proteins were separated in a 12% SDS-polyacrylamide gel. The gel was
then washed twice for 1 hr in renaturation buffer (50mM NaCI, 10 mM
Tris-HCI, pH 7.0, 20 mM EDTA, 0.1 mM DTT, 4 M urea) before electroblotting onto nitrocellulose. The blot was probed for 3 hr at room
temperature with gentle agitation in binding buffer (50mM NaCI, 10
mM Tris-HCI, pH 7.4, 1 mM EDTA, 5% nonfat dry milk, 100 pM CaCI,)
containing 30 nM labeled ligand. After four 30-min washes in binding
buffer, the blot was incubated for 1 hr in binding buffer containing 2
pglmL avidin-horseradish peroxidaseconjugate (Pierce), followed by
five 5-min washes in binding buffer and five 5-min washes in binding
buffer without nonfat dry milk. Bound ligand was detected with a
chemiluminescence assay (ECL; Amersham), and the intensity of the
binding was quantified using a Molecular lmager (Bio-Rad). A very
weak affinity of avidin-horseradish peroxidase for GST-remorin was
Uronide Binding Protein
noted, and all bindingexperiments were conductedwith acontrol using
avidin-horseradish peroxidase in the absence of any labeled ligand.
Polyanionsusedfor the competitionexperimentswere trigalacturonic
acid (Sigma),polymannuronicacidwith a DPof -23, and polyguluronic
acid with a DP of -19 (a gift from M. Saxton, Universityof California
at Davis). Carboxymethylcellulose. poly-L-glutamate, and heparin were
from Sigma. We took the fact that heparin binds avidin(Pierce technical information)into account during the experiments.Oligonucleotides
were low molecular weight DNA from herring (Fluka), total yeast RNA
type VI from rorula (Sigma), and a synthetic 26-mer single-stranded
DNA (5'-CGCGGATCCGAGAAGAAGAAGTATCC-3?.
ACKNOWLEDGMENTS
This work was supported by Swiss National Science FoundationGrant
No. 31-36472-92and by the canton of Vaud. We thank Boris Künstner
for expert preparation of plant material, Pierre Hainard for the gift of
plants from the University of Lausanne tropical greenhouse, David
Johnston and Pia Malnoe for the potato cDNA library, ClarenceA. Ryan
for stimulatingconversation and the gift of tomato leaf PGA, Michael
Saxton for alginic acids, Antonio Conconifor tomato nuclei, and Dean
DellaPenna for help in designing PCR probes. We thank Severine
Frütigerfor N-terminalsequencing, Rose-MarieHofer for administrative and scientific support, and Roland Beffaand Yves Poirierfor critical
comments.
Received August 2, 1996; accepted October 7, 1996.
REFERENCES
Aldington, S., and Fry, S.C. (1993).Oligosaccharins.Adv. Bot. Res.
19, 1-101.
Alliotte, T., Tire, C., Engler, G., Peleman, J., Caplan, A,, Van
Montagu, M., and Inzé, D. (1989). An auxin-regulated gene of
Arabidopsis thaliana encodes a DNA-bindingprotein. Plant Phys-
iol. 89, 743-752.
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J.
(1990).Basic local alignment search tool. J. MOI.Biol. 215,403-410.
Asiegbu, F.O., Daniel, G., and Johansson, M. (1994).Defence related
reactions of seedling roots of Norway spruce to infection by Heterobasidion annosum (Fr.) Bref. Physiol. MOI.Plant Pathol. 45, 1-19.
Bellincampi, D., Cardarelli, M., Zaghi, D., Serino, G., Salvi, G., Gatz,
C., Cervone, F., Altamura, M.M., Constantino, P., and De Lorenzo,
G. (1996). Oligogalacturonides prevent rhizogenesis in rol6-
transformed tobacco explants by inhibiting auxin-induced expression of the rol6 gene. Plant Cell 8,477-487.
Berger, F., Taylor, A., and Brownlee, C. (1994).Cell fate determination by the cell wall in early Fucus development. Science 253,
1421-1423.
Bernasconi, F?, Walters, E.W., Woodworth, A.R., Siehl, D.L., Stone,
T.E., and Subramanian, M.V. (1994). Functional expression of
Arabidopsis thaliana anthranilate synthasesubunit I in Escherichia
coli. Plant Physiol. 106, 353-358.
2275
Bishop, P.D., Pearce, G., Bryant, J.E., and Ryan, C.A. (1984). Isola-
tion and characterizationof proteinaseinhibitor inducing factor from
tomato leaves. J. Biol. Chem. 259, 13172-13177.
Blumenkrantz, N.J., and Asboe-Hansen, B. (1973). New methods
of quantitative determination of uronic acids. Anal. Biochem. 54,
484-489.
Branden, C., and Tooze, J. (1991). lntroductionto ProteinStructure.
(New York: Garland Publishing).
Citovsky, V., McLean, B.G., Zupan, J.R., and Zambryski, P. (1993).
Phosphorylationof tobacco mosaic virus cell-to-cellmovement protein by a developmentally regulated plant cell wall-associated protein
kinase. Genes Dev. 7, 904-910.
Cote, F., and Hahn, M.G. (1994). Oligosaccharins: Structures and signal
transduction. Plant MOI.Biol. 26, 1379-1411.
Cronquist, A. (1988).The Evolution and Classificationof Flowering
Plants, 2nd'ed. (Bronx, NY: New York Botanical Gardens).
Darvill, A., and Albersheim, P. (1984). Phytoalexins and their elicitors: A defense against microbial infectionin plants. Annu. Rev. Plant
Physiol. 35, 243-275.
Davis, G.D., and Essenberg, M. (1995).(+)-6-Cadineneis a product
of sesquiterpene cyclase activity in cotton. Phytochemistry 39,
553-567.
Devereux, J., Haeberli, P., and Smithies, O. (1984).A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids
Res. 12, 387-395.
Diekmann, W., Herkt, E., Low, P.S., Niirenberger, T., Scheel, D.,
Terschiiren, C., and Robinson, D. (1994). Visualisationof elicitor
binding loci at the plant cell surface. Planta 195, 126-137.
Farmer, E.E., Pearce, G., and Ryan, C.A. (1989). In vitro phosphorylation of plant plasma membrane associated proteins in response
to the proteinase inhibitor inducingfactor. Proc. Natl. Acad. Sci. USA
86, 1539-1542.
Farmer, E.E.,Moloshok, T.D., Saxton, M.J., and Ryan, C.A. (1991).
Oligosaccharide signaling in plants: Specificity of oligouronideenhancedplasma membrane protein phosphorylation.J. Biol. Chem.
266, 3140-3145.
Fujiki, Y., Hubbard, A.L., Fowler, S., and Lazarow, P.B. (1982). Isolation of intracellular membranes by means of sodium carbonate
treatment: Application to endoplasmic reticulum. J. Cell Biol. 93,
97-102.
Jacinto, T., Farmer, E.E., and Ryan, C.A. (1993). Purification of potato
leaf plasma membrane protein pp34, a protein phosphorylated in
response to oligogalacturonide signals for defense and development.
Plant Physiol. 103, 1393-1397.
Kohn, R. (1985).lon binding on polyuronates-alginateand pectin.Pure
Appl. Chem. 42, 371-397.
Kyte, J., and Doolittle, R.F. (1982). A simple method for displaying
the hydropathic character of a protein. J. MOI.Biol. 157, 105-132.
Legendre, L., Heinstein, P.F., and Low, P.S. (1992). Evidencefor participationof GTP-bindingproteins in elicitation of the rapidoxidative
burst in cultured soybean cells. J. Biol. Chem. 267, 20140-20147.
Lucas, W.J., and Gilbertson, R.L. (1994). Plasmodesmata in relation to vira1 movement within leaf tissues. Annu. Rev. Phytopathol.
32, 387-411.
Margalit, H., Fischer, N., and Een-Sasson, S.A. (1993). Comparative analysis of structurally defined heparin binding sequences
reveals a distinct spatial distribution of basic residues. J. Biol. Chem.
268, 19928-19931.
2276
The Plant Cell
Mathieu, Y., Kurkdjian, A., Xia, H., Guern, J., Spiro, M.D., ONeill,
M.D., and Albersheim, I? (1991). Membrane responses induced
by oligogalacturonidesin suspension-culturedtobacco cells. Plant
J. 1, 333-343.
Messiaen, J., and Van Cutsem, P.V. (1994). Pectic signal transduction in carrot cells: Membrane, cytosolic and nuclear responses
induced by oligogalacturonides. Plant Cell Physiol. 35, 677-689.
Mohnen, D., Eberhard, S., Marfa, V., Doubrava, N., Toubart, P.,
Gollin, D.J., Gruber, T.A., Nuri, W., Albersheim, I?, and Darvill,
A. (1990).The control of root, vegetative shoot and flower morphogenesis in tobaccothin cell layer explants (TCLs). Development108,
191-201.
Moore, RJ., Fenczik, C.A., Deom, C.M., and Beachy, R.N. (1992).
Developmentalchanges in plasmodesmata in transgenic tobacco
expressing the movement protein of tobacco mosaic virus. Protoplasma 170, 115-127.
Nürnberger, T., Nennstiel, D., Jabs, T., Sacks, W.R., Hahlbrock,
K., and Scheel, D. (1994). High affinity binding of afungal oligopeptide elicitor to parsley plasma membranestriggers multipledefense
responses. Cell 78, 449-460.
ONeill, M., Albersheim, I?, and Darvill, A. (1990).The pectic polysaccharides of primary cell walls. In Methods in Plant Biochemistry,
Vol. 2, PM. Dey, ed (London: Academic Press), pp. 415-441.
Pawson, T. (1995). Protein modules and signalling networks. Nature
373, 573-579.
Penel, C., and Greppin, H. (1996). Pectin binding proteins: Characterization of the binding and comparisonwith heparin. Plant Physiol.
Biochem. 34, 479-488.
Prakash, C., and Vijay, I.K. (1983).A new fluorescent tag for labelling
saccharides. Anal. Biochem. 128, 41-46.
Reymond, P., Griinberger, S., Paul, K., Miiller, M., and Farmer, E.E.
(1995). Oligogalacturonide defensesignals in plants: Largefragments
interact with the plasma membrane in vitro. Proc. Natl. Acad. Sci.
USA 92, 4145-4149.
Robertsen, B., and Svalheim, O. (1990). The nature of lignin-likecompounds in cucumber hypocotyls induced by a-1,4-linked
oligogalacturonides. Physiol. Plant. 79, 512-518.
Ryan, C.A. (1988).Oligosaccharides as recognition signals for the expression of defensivegenes in plants. Biochemistry 27, 8879-8883.
Silva, C.M., Tully, D.B., Petch, L.A., Jewell, C.M., and Cidlowski,
J.A. (1987).Application of a protein-blottingprocedureto the study
of humanglucocorticoid receptor interactions with DNA. Proc. Natl.
Acad. Sci. USA 84, 1744-1748.
Soltis, P.S., and Soltis, D.E. (1995).Plant molecular systematics: Inferences of phylogeny and evolutionary processes. Evol. Biol. 28,
139-194.
Spiro, M.D., Kates, K.A., Koller, A.L., ONeill, M.A., Albersheim,
I?, and Darvill, A.G. (1993). Purificationand characterizationof biologically active IA-linked a-o-oligogalacturonides after partia1
digestion of polygalacturonicacid with endopolygalacturonase.Carbohydr. Fies. 247, 9-20.
Thain, J.F., Doherty, H.M., Bowles, D.J., and Wildon, D.C. (1990).
Oligosaccharidesthat induce proteinase inhibitor activity in tomato
plantscause depolarizationof tomato leaf cells. Plant Cell Environ.
13, 569-574.
West, C.A., Bruce, R.J., and Jin, D.F. (1984). Pectic fragments of plant
cell walls as mediators of stress responses. In Structure, Function
and Biosynthesisof Plant Cell Walls, W.M. Dugger and S.BartnickGarcia, eds (Baltimore, MD: Waverley Press), pp. 569-574.
Yoder, M.D., and Jurnak, F. (1995). The refined three-dimensional
structure of pectate lyase C from Erwinia chrysantbemi at 2.2
resolution. Plant Physiol. 107, 349-364.
Yoshida, S., Uemura, M., Niki, T., Sakai, A., and Gusta, L.V. (1983).
Partition of membrane particles in aqueous two-polymerphase system and its practicaluse for purificationof plasma membranesfrom
plants. Plant Physiol. 72, 105-114.
a
Zablackis, E., Huang, J., Müller, B., Darvill, A., and Albersheim,
I? (1995). Characterization of the cell-wall polysaccharides of
Arabidopsis tbaliana leaves. Plant Physiol. 107, 1129-1138.
Cloning of a cDNA encoding a plasma membrane-associated, uronide binding phosphoprotein
with physical properties similar to viral movement proteins.
P Reymond, B Kunz, K Paul-Pletzer, R Grimm, C Eckerskorn and E E Farmer
Plant Cell 1996;8;2265-2276
DOI 10.1105/tpc.8.12.2265
This information is current as of June 18, 2017
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw153229
8X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY