Structure–Function Analysis of Cf-9, a ... - Plant CellFigure 1. Plant eLRR Proteins with Known...

Structure–Function Analysis of Cf-9, a Receptor-Like Proteinwith Extracytoplasmic Leucine-Rich Repeats W

Renier A.L. van der Hoorn, a,1 Brande B.H. Wulff,b,c,1 Susana Rivas,b,d,1 Marcus C. Durrant,e Anke vander Ploeg,a Pierre J.G.M. de Wit,a and Jonathan D.G. Jonesb,2

aWageningen University, Laboratory of Phytopathology, 6709 PD, Wageningen, The Netherlandsb Sainsbury Laboratory, John Innes Centre, Norwich Research Park, Norwich NR4 7UH, United Kingdomc Instituto de Biologıa Molecular y Celular de Plantas, Universidad Politecnica de Valencia, 46022 Valencia, Spaind Laboratoire des Interactions Plantes-Microorganismes, Unite Mixte de Recherche, Centre National de la Recherche

Scientifique/Institut National de la Recherche Agronomique 2594/441, 31326 Castanet-Tolosan Cedex, Francee Computational Biology Group, John Innes Centre, Norwich Research Park, Norwich NR4 7UH, United Kingdom

The tomato (Lycopersicon pimpinellifolium) resistance protein Cf-9 belongs to a large class of plant proteins with

extracytoplasmic Leu-rich repeats (eLRRs). eLRR proteins play key roles in plant defense and development, mainly as

receptor-like proteins or receptor-like kinases, conferring recognition of various pathogen molecules and plant hormones.

We report here a large-scale structure–function analysis of an eLRR protein. A total of 66 site-directed mutants of Cf-9 were

analyzed for activity in Avr9 recognition and for protein stability and the results interpreted with the help of a homology

model of the Cf-9 structure. Conserved Trp and Cys pairs in the N-terminal LRR-flanking domain appear to be important for

Cf-9 activity and are probably exposed at the putative concave inner surface of the Cf-9 protein, where recognition

specificity also resides. Removal of each of the 22 putative N-linked glycosylation sites (PGS) revealed that many PGSs

contribute to Cf-9 activity and that the PGSs in the putative a-helices of the LRR modules are essential. Immunoblot analysis

and mass spectrometry showed that all but one of the PGSs are N-glycosylated. Introduction of glycosylation at the putative

concave b-sheet surface blocks Cf-9 activity, in some cases probably by disturbing specific recognition, and in another

case by steric hindrance with existing N-glycans. The glycosylation pattern and several other features are conserved in

other eLRR proteins, where similar mutations show similar phenotypes.

INTRODUCTION

Proteins with extracytoplasmic Leu-rich repeats (eLRRs) play

a crucial role in plant defense and development by perceiving

extracellular signals that can be pathogen-derived molecules or

plant hormones, respectively. The LRR domain is proposed to

act as a versatile recognition surface where specific protein–

protein interactions occur. This LRR domain is usually flanked by

small domains containing Cys residues and fused to other

domains, through which the eLRR proteins can be classified

(Figure 1).

A function has been assigned to many eLRR proteins (refer-

ences in Table 1). PGIP is a polygalacturonase inhibiting protein,

AFP inhibits ice crystallization, and LRX1 is required for root hair

morphogenesis. Many receptor-like proteins (RLPs) with eLRRs

confer disease resistance through recognition of pathogen-

derived molecules (e.g., Cf-9, RPP27, and EIX1), whereas other

RLPs function in meristem development (FEA2 and CLV2) or

distribution of stomata (TMM). Some receptor-like kinases (RLKs)

with eLRRs also confer pathogen recognition (e.g., FLS2 and

Xa21), whereas others act in perception of steroid or peptide

hormones (e.g., BRI1, PSKR, and SR160) or play a role in other

developmental processes, such asmeristemdevelopment, stem

elongation, abscission, pollination, and nodulation (e.g., CLV1,

ER, HAESA, LePRK, and SYMRK). It can be expected that addi-

tional functions will be assigned to eLRR proteins, given the fact

that the Arabidopsis thaliana genome encodes two PGIP-like

proteins, 11 LRX-like proteins, 59 eLRRRLPs, and 216 eLRRRLKs

(Arabidopsis Genome Initiative, 2000; Shiu and Bleecker, 2001;

Baumberger et al., 2003; Ferrari et al., 2003; Tor et al., 2004).

Several interactors of eLRR proteins have been identified.

Ligands can be proteins (e.g., polygalacturonase, CLV3, LAT52,

and xylanase) or peptides (e.g., systemin, flagellin, and phyto-

sulfokine) (references in Table 1). Many eLRR proteins are also

found to interact with other components of signaling complexes.

BRI1, for example, interacts with BAK1 (Li et al., 2002; Nam and

Li, 2002), CLV1 is expected to interact with CLV2 (Jeong et al.,

1999), and LePRK2 associates with LePRK1 (Wengier et al.,

2003).

Specific protein–protein interactions can be mediated by

the LRR domain. Much information on the LRR structure and

1 These authors contributed equally to this work.2 To whom correspondence should be addressed. E-mail [email protected]; fax 44-01603-450011.The authors responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy describedin the Instructions for Authors (www.plantcell.org) are: Jonathan D.G.Jones ([email protected]) and Pierre J.G.M.de Wit ([email protected]).WOnline version contains Web-only data.Article, publication date, and citation information can be found atwww.plantcell.org/cgi/doi/10.1105/tpc.104.028118.

The Plant Cell, Vol. 17, 1000–1015, March 2005, www.plantcell.orgª 2005 American Society of Plant Biologists

function is derived from studieswith nonplant LRRproteins, such

as the porcine ribonuclease inhibitor (RI). The crystal structure of

RI revealed a horseshoe-like structure, where multiple LRRs are

stacked to form a parallel b-sheet at the concave (inner) side of

the protein and a series of a-helices on the convex (outer) side

of the protein (Kobe and Deisenhofer, 1993). Binding of RI to

ribonuclease occurs through solvent-exposed residues (x in

xxLxLxx) at the concave b-sheet surface (Kobe andDeisenhofer,

1996). The variation of displayed residues at the b-sheets con-

tributes to the diverse recognition specificities of LRR proteins.

Plant eLRR domains are typified by a distinct LRR consensus

sequence, Cys-containing LRR-flanking domains, and a large

number of putative glycosylation sites (Kajava, 1998). In addition,

the crystal structure of PGIP revealed, when compared with

RI, the presence of a second parallel b-sheet on one side of the

LRR domain and a twisted, decreased curvature of the b-sheet

surface (Di Matteo et al., 2003). Domain-swap experiments

showed that the concave b-sheet surface of PGIP contains the

residues that determine specificity for different polygalacturo-

nases (Leckie et al., 1999). Similar domain-swap experiments

revealed that specificity determinants in Cf-4 and Cf-9 reside in

only a few residues exposed at the putative b-sheet of the central

LRRs (Van der Hoorn et al., 2001; Wulff et al., 2001).

In summary, there is a large class of plant-specific proteins

with eLRRs for which diverse functions in defense and develop-

ment have been described, but little is known about how they

function. To provide more information on eLRR mechanisms,

we undertook a large scale structure–function analysis on con-

served features of eLRR proteins. Cf-9 was chosen as a pro-

totype for the analysis because this RLP contains conserved

features of eLRR proteins, and a reliable, convenient quantitative

agroinfiltration assay is available to analyze Cf-9 function (Van

der Hoorn et al., 2000). This quick assay was shown to repre-

sent the phenotypes for resistance in tomato (Lycopersicon

esculentum) plants (Wulff et al., 2001). We monitored expressed

mutant proteins, which provided additional knowledge of the

glycosylation pattern. The phenotypes arising from mutagene-

sis are discussed in the light of a homology model for the Cf-9

structure. These data will provide useful insights for our un-

derstanding of how proteins with eLRRs function.

RESULTS

Conserved Trp and Cys Residues Are Essential

for Cf-9 Activity

To identify amino acid residues important for function of pro-

teins with eLRRs, we compared LRR-flanking domains of

eLRR proteins with known functions (Figure 1, Table 1) and

used directed mutagenesis of Cf-9 to investigate the importance

of these residues for Cf-9 activity.

Several conserved motifs could be distinguished in the

LRR-flanking domains. The N-terminal LRR-flanking domain

(B-domain) contains the conserved motifs LLxxK, LssW,

and CxWxGVxC, whereas the C-terminal LRR-flanking domain

(D-domain) contains the conserved motif GNxGLCGxPLxxxC

(Figure 2A). Cf-9 also carries these motifs, except that Cf-9

carries an Arg instead of the first Cys in the D-domain (Figure 2A).

Furthermore, Cf-9 carries three additional Cys residues in the

B-domain, some of which can also be found in other eLRR

proteins (Figure 2A). Cys residues are of particular interest

because it has been proposed that they can be involved in the

formation of intramolecular and/or intermolecular disulfide

bridges (Dievart and Clark, 2003).

Nine conserved residues in the Cf-9 LRR-flanking domains

were selected for mutagenesis, namely the Lys in the LLxxK

motif (designated K1), the Trp in the LssW and CxWxGVxC mo-

tifs (W1 and W2, respectively), and the six Cys residues (called

C1 to C6, of which C4 and C5 are in the conserved CxWxGVxC

motif and C6 is in the D-domain). For each of these residues, an

Ala substitution mutant of Cf-9 was generated and its activity

determined by coexpression with Avr9 in at least three indepen-

dent agroinfiltration assays (Van der Hoorn et al., 2000). These

experiments show that W1A and W2A mutations completely

abolish Cf-9 activity (Figure 2C). This severe phenotype was also

observed for the additional mutant W2T (Figure 2C). By contrast,

mutations K1A, C1A, and C6A have no effect on Cf-9 activity

(Figures 2B and 2C, curves a, b, and g). Furthermore, mutants

C2A and C3A show only slightly reduced activity (Figures 2B and

2C, curves c and d), whereas mutants C4A and C5A are nearly

inactive (Figures 2B and 2C, curves e and f). Thus, substitution of

the conserved Trp (W1 andW2) and the twomost conserved Cys

residues (C4 and C5) abolish or severely reduce Cf-9 activity,

respectively. None of the active mutants induce necrosis in the

absence of Avr9 (data not shown), indicating that themutants are

not autoactivating.

To verify that the expressed mutant proteins are still stable,

cMyc-epitope-tagged versions of the mutant proteins were

constructed using a new agroinfiltration vector (Figure 2D).

Significantly, all myc-tagged Cf-9 mutant proteins accumulate

Figure 1. Plant eLRR Proteins with Known Function.

Plant proteins with eLRRs can be divided into four classes. The number

of genes predicted from the Arabidopsis genome, as well as represen-

tatives for which a function is known, are shown. TM, transmembrane

domain.

Structure–Function Analysis of Cf-9 1001

to similar levels as myc-Cf-9 wild-type protein (Figure 2E),

indicating that none of these residues are essential for Cf-9

protein stability. The activities of the myc-tagged Cf-9 mutants,

however, are lower than those without myc-tag (Figure 2C). This

is probably due to the fact that the N-terminal myc-tag slightly

reduces the overall Cf-9 activity, compared with that of the

wild-type Cf-9 (Figure 2B, curve h). As a result, phenotypes for

myc-tagged C2A, C3A, C4A, and C5A mutants are even more

pronounced than without a myc-tag. Significantly, myc-K1A,

myc-C1A, and myc-C6A mutants remain as active as myc-Cf-9,

indicating that these mutations have not even a weak effect on

Cf-9 activity. Taken together, these data indicate that the order of

reduction of Cf-9 activity is as follows: (W1A�W2A/T)� (C5A$

C4A) > (C3A $ C2A) > (C1A � C6A � K1A � Cf-9).

The reduced activities of the Cf-9 Cys mutants also suggest

that intramolecular disulfide bridges may exist between Cys C2

and C3 and between C4 and C5 because substitutions of these

Cys result in similar phenotypes within each couple. To investi-

gate this further, double mutants C2,3A and C4,5A were con-

structed and their activities determined. Mutant C2,3A is less

active than C2A or C3A, suggesting that C2A and C3A pheno-

types do not result from a broken putative disulfide bridge alone,

but from additive effects. Double mutant C4,5A has a weak

activity that is too low to compare with the weak activities of the

C4A and C5A mutants.

We also constructed mutants C1-5A and C1-6A, in which

five or all six of the Cys were substituted. These mutants are

all inactive, indicating that the phenotypes of the single Cys

mutations are additive (Figure 2C). Myc-tagged versions of

thesemutants, including the C1-6Amutant accumulate to similar

levels as wild-type myc-Cf-9 (Figure 2E), demonstrating that

these proteins are stable, despite being unable to form disulfide

bridges through these Cys.

Putative Glycosylation Sites Are Important for Cf-9 Activity

Another distinct feature of plant eLRR proteins is their large

number of putative N-linked glycosylation sites (PGSs; N in

NxS/T motifs). Cf-9 contains 22 PGSs (Figure 3A) in its extracel-

lular domain, of which 19 reside in the LRR domain. Previous

N-deglycosylation studies using PNGase F showed that Cf-9

is heavily N-glycosylated (Piedras et al., 2000) and that this

glycosylation is evenly distributed over three parts of Cf-9 (Van

der Hoorn et al., 2003). The relevance of PGSs in Cf-9 became

Table 1. Plant eLRR Proteins with Known Function

Name Typea Species Function Reference

AFP PGIP-like Carrot Antifreeze Worrall et al. (1998)

PGIP PGIP-like Bean Polygalacturonase inhibition Toubart et al. (1992)

SHY PGIP-like Petunia Pollen tube growth Guyon et al. (2004)

LRX1 LRX-like Tomato Root hair morphogenesis Baumberger et al. (2001)

Cf-2 RLP Tomato Leaf mold resistance Dixon et al. (1996)

Cf-9 RLP Tomato Leaf mold resistance Jones et al. (1994)

CLV2 RLP Arabidopsis Meristem development Jeong et al. (1999)

FEA2 RLP Maize Meristem development Taguchi-Shiobara et al. (2001)

LeEix2 RLP Tomato Xylanase elicitor perception Ron and Avni (2004)

RPP27 RLP Arabidopsis Downy mildew resistance Tor et al. (2004)

TMM RLP Arabidopsis Stomatal distribution Nadeau and Sack (2002)

Ve1 RLP Tomato Fungal resistance Kawchuk et al. (2001)

Vf2 RLP Apple Apple scab resistance Vinatzer et al. (2001); Belfanti et al. (2004)

BAK1 RLK Arabidopsis Brassinosteroid signaling Li et al. (2002); Nam and Li (2002)

BRI1 RLK Arabidopsis Brassinosteroid perception Li and Chory (1997); Wang et al. (2001)

CLV1 RLK Arabidopsis Meristem development Clark et al. (1997)

EMS1/EXS RLK Arabidopsis Anther cell development Zhao et al. (2002) Canales et al. (2002)

ER RLK Arabidopsis Stem elongation Torii et al. (1996)

FLS2 RLK Arabidopsis Flagellin perception Gomez-Gomez and Boller (2000);

Gomez-Gomez et al. (2001)

HAESA RLK Arabidopsis Leaf abscission Walker (1993) Jinn et al. (2000)

HAR1 RLK Lotus Root development and nodulation Krusell et al. (2002); Nishimura et al. (2002)

LePRK2 RLK Tomato Pollen LAT52 perception Muschietti et al. (1998); Tang et al. (2002)

LKA RLK Pea Brassinosteroid perception Nomura et al. (2003)

NARK RLK Soybean Nodule autoregulation Searle et al. (2003)

OsBRI1 RLK Rice Internode elongation Yamamuro et al. (2000)

PSKR RLK Carrot PSK perception Matsubayashi et al. (2002)

SR160 RLK Tomato Systemin perception Scheer and Ryan (2002)

SYMRK RLK Lotus Bacterial and fungal symbiosis Endre et al. (2002); Stracke et al. (2002)

VH1 RLK Arabidopsis Vascular development Clay and Nelson (2002)

Xa21 RLK Rice Rice blast resistance Song et al. (1995)

Xa26 RLK Rice Rice blast resistance Sun et al. (2004)

a Classification as represented in Figure 1.

1002 The Plant Cell

Figure 2. Targeted Mutagenesis of LRR-Flanking Domains of Cf-9.

(A) Alignment of LRR-flanking domains. The B-domain is the N-terminal LRR-flanking domain and starts after the predicted signal peptide and


clear when three out of four inactive Cf-9 ethyl methanesulfonate

(EMS) point mutants appeared to carry a mutation that likely

affectsN-glycosylation (Wulff et al., 2004). In mutant M2, PGS 18

in LRR 24 was removed (here called mutant 18SL), whereas in

EMS mutants F965 and M140, one additional PGS was intro-

duced in LRRs 12 and 18, respectively (here called mutants

DN12 and DN18) (Wulff et al., 2004).

To investigate the relevance of these and other PGSs in Cf-9,

an extensive series of mutant TAP-tagged Cf-9 proteins was

generated (Figure 3B). For nearly all PGSs, Asn residues were

substituted with Asp (N into D), which are common at similar

positions in Cf-9 and other eLRR proteins (Figure 3A; data not

shown). For similar reasons, PGS1 and 2 in the B-domain were

mutated intoNxA. Accumulation of themutant Cf-9-TAPproteins

was investigated by immunoblot analysis of agroinfiltrated

tobacco tissues. All mutant Cf-9-TAP proteins accumulated

to similar levels as wild-type Cf-9-TAP (Figure 3B), except for

mutant 13ND, which we could not detect even when expressed

in the presence of P19 silencing inhibitor (Voinnet et al., 2003) in

Nicotiana benthamiana (data not shown).

Agroinfiltration into Avr9-transgenic tobacco in at least three

independent experiments revealed that most PGSmutants have

reduced activities (Figure 3B). In addition to EMS mutant 18SL,

mutants 6ND, 9ND, and 11ND also appeared inactive. Mutant

13ND is also inactive, which correlates with the absence of

protein of this mutant. Activity of mutants 12ND and 19ND could

just be detected, whereas all other PGS removal mutants are

active, although at different levels (Figure 3B). The activity

depends on the presence of Avr9 because no necrosis was

induced in wild-type tobacco (data not shown).

To verify the most severely reduced activities, PGSs 6, 9, 11,

12, 13, and 19 were also mutated by changing NxS/T into NxA

and PGS18 by changing NxS into DxS. All these mutant proteins

accumulate to levels similar to wild-type Cf-9-TAP (Figure 3B).

Mutant 13SA also accumulates, indicating that the instability of

the 13ND mutant is not caused by the removal of glycosylation

alone and that PGS13 is not required for Cf-9 activity. Signifi-

cantly, the activities of the PGS removal mutants caused by

S-into-A and T-into-A substitutions confirm the activities of the

PGS removal mutants caused by N-into-D substitutions. Thus,

mutants 6SA and 11TA are inactive, like mutants 6ND and 11ND.

Mutant 12SA is inactive, whereas 12ND already had low activity,

and mutant 9SA has low activity, whereas 9ND was inactive.

Exceptions are the full activities of 18ND and 19SA, which

contrast with the inactivity of EMS mutant 18SL and weak

activity of mutant 19ND (Figure 3B).

Thus, for four out of seven PGS inactive mutants, the low

activity of the PGS removal mutation was confirmed with an

independent mutation. When summarized by plotting the activ-

ities onto the Cf-9 sequence, it becomes obvious that these four

PGS (6, 9, 11, and 12) all cluster in the putative a-helices of LRRs

4 to 12 (Figure 3C).

Nearly all PGSs of Cf-9 Are Glycosylated

Subsequently, we analyzed the PGS removal mutants on

high-resolution SDS-PAGE gels to determine their size,

when compared with wild-type Cf-9-TAP protein. Previously,

N-deglycosylation ofCf-9withPNGaseF resulted in an increased

mobility corresponding to a loss in apparent molecular mass of

60 kD (Piedras et al., 2000; Van der Hoorn et al., 2003).

Theoretically, if each of the 22 PGSs is N-glycosylated, then on

average, each N-glycosylation site could contribute 3 kD to this

apparent molecular mass. To detect such small size differences,

we separated mutant proteins, interspaced by wild-type Cf-9-

TAPprotein, on 7%protein gels run overnight at low voltage. This

revealed a small, but significantly increased mobility for most

PGS mutants, when compared with wild-type Cf-9-TAP (Figures

3B and 3D). Exceptions are mutants 18SL and 18ND, which both

migrate as wild-type Cf-9-TAP, and 2SA, for which the increased

mobility was not always detected.

Two pieces of data show that the observed increased mobil-

ites resulted from elimination of N-glycans and not from the

amino acid substitutions themselves. First, the increased mobil-

ity was observed for both N-into-D and S/T-into-A substitutions.

Second, N-deglycosylation of these mutant proteins by PNGase

Figure 2. (continued).

terminates two amino acids beyond the conserved Cys pair. The D-domain is the C-terminal LRR-flanking domain (only the conserved region is shown).

Conserved residues are indicated in bold, and mutagenized residues are indicated below the Cf-9 sequence. Underlined numbers indicate the number

of amino acid residues omitted to simplify the alignment.

(B) Comparative quantitative analysis of activities of a selection of nontagged Cf-9 mutants (indicated with superscript a to h in [C]). Agrobacterium

cultures carrying a plasmid that encodes a (mutant) Cf-9 protein were mixed in different ratios with a culture of equal density carrying an Avr9-encoding

plasmid and infiltrated into opposite tobacco leaf halves. At 7 d post-infiltration (dpi), the percentage of infiltrated area that had become necrotic was

measured and plotted against the percentage of culture carrying (mutant) Cf-9.

(C) Summary of hypersensitive response–inducing activities of the all nontagged or myc-tagged Cf-9 (mutant) proteins when coexpressed with Avr9.

Leaves of tobacco plants were infiltrated with Agrobacterium carrying a binary plasmid encoding (mutant) Cf proteins. Avr9was coexpressed by mixing

Agrobacterium cultures with strains carrying Avr9 on binary plasmids. Necrosis was scored at 7 dpi. þþþ, necrosis of entire infiltrated area; þþ, >50%

necrosis of infiltrated area;þ, 10 to 50% necrotic area;6, 0 to 10% necrotic area;�, no necrosis. Superscripts a to h refer to (B). The data represent an

average of at least three agroinfiltration assays.

(D) Myc-tagged Cf-9 construct designed for agroinfiltration. The four N-terminal myc-tags (top) are interrupted by an intron in the expression cassette

(bottom) to exclude bacterial expression. 35S, constitutive Cauliflower mosaic virus promoter; SP, signal peptide of tobacco PR1a gene to ensure

extracellular targeting; term, polyadenylation signal of PI-II gene.

(E) Protein accumulation of (mutant) myc-tagged Cf-9 proteins at 1 dpi in tobacco with Agrobacterium cultures. Proteins were extracted from infiltrated

leaves and analyzed on immunoblots using anti-myc antibody.

1004 The Plant Cell

Figure 3. Effects of Removal and Addition of Putative Glycosylation Sites in Cf-9.

(A) Position of the putative glycosylation sites (N in NxS/T) in the Cf-9 protein. The consensus LRR sequence is shown on top, with the putative solvent-

exposed residues of the putative b-sheet underlined.


F treatment resulted in identical electrophoretic mobility for all

tested mutants (data not shown).

As an independent confirmation for the presence or absence

of N-glycosylated sites, we analyzed purified Cf-9-TAP proteins

by mass spectrometry (MS). Cf-9-TAP was produced by large

scale agroinfiltration of N. benthamiana and purified from solu-

bilizedmembranes by immunoprecipitation (seeMethods). Cf-9-

TAP was isolated from an SDS-PAGE gel, digested with trypsin,

and analyzed three times (MS1-3) with matrix-associated laser

desorption and ionization time-of-flight (MALDI-TOF) to maxi-

mize coverage of the Cf-9-TAP protein.

In total, 15 peptides were detected in independent experi-

ments (Figure 4). Although much of the Cf-9-TAP protein se-

quencewas coveredwith identified peptides, no peptide (except

one, see below) containing a PGSwas detected (Figure 4B). This

is consistent with N-glycosylation of all PGSs (except one) as

indicated by analysis of the PGS removal mutants. The addi-

tional, unassigned molecular masses that were detected (data

not shown) could not, however, be assigned to certain glycosy-

lated peptides without knowing the exact structure of the

N-glycan. Significantly, a peptide carrying PGS18 was detected

twice (peptide i, Figure 4). This peptide was also detected using

quadrupole-time-of-flight (Q-TOF), which also confirmed the

sequence of this peptide (Figure 4B). These data confirm that

PGS18 is not N-glycosylated.

In conclusion, mutant protein electrophoresis and MS anal-

ysis showed that, except for PGS18, all PGSs of Cf-9 are

N-glycosylated. This confirms earlier observations that glyco-

sylation is evenly distributed over the different LRRs of Cf-9 (Van

der Hoorn et al., 2003) and is consistent with the slightly in-

creased mobility for each of the PGS removal mutants.

Adding Glycosylation at the Putative Concave b-Sheet

Surface Affects Cf-9 Activity

In addition to PGS removal mutants, we also created and studied

mutants with introduced PGSs at putative solvent-exposed

positions of the putative concave b-sheet in LRRs 6, 12, 18,

and 26 of Cf-9 (mutants QN6, DN12, DN18, and DN26, re-

spectively; Figure 3B). The rationale for this was given by the

inactivity of the EMSmutants where a PGSwas introduced at the

putative concave b-sheet surface of LRRs 12 and 18 (mutants

F965 and M140, here called mutants DN12 and DN18, respec-

tively; Wulff et al., 2004). Furthermore, we introduced a PGS in

LRR12, outside the putative solvent-exposed region, by intro-

ducing amino acids RSW, an insert found in nearly all homologs

of Cf-9 (Parniske et al., 1997).

Agroinfiltration of Avr9-transgenic tobacco confirmed the in-

activity of EMS mutants DN12 and DN18 and showed that

mutant DN26 is also inactive, whereas mutants DN6 and RSW

have only slightly reduced activity (Figure 3B). The mutant

proteins accumulate to similar levels as wild-type Cf-9-TAP

(Figure 3D), indicating that the reduced activities are not due to

protein instability.

The inactivity of the D-into-N mutants can result from an

introduced N-linked glycan or from the intrinsic property of the

amino acid substitution itself. To discriminate between these

two possibilities, we made D-into-Q mutants DQ12, DQ18,

and DQ26. These mutants have a similarly changed charge as

D-into-N mutants but do not carry an additional PGS. Agro-

infiltration demonstrated that these D-into-Q mutants are fully

active (Figure 3B), indicating that the phenotype of the D-into-N

mutants results from an introduced N-glycan, rather than from

the amino acid substitution itself. Expression of the mutant in the

absence of Avr9 did not trigger necrosis (data not shown).

On immunoblots, PGS addition mutants QN6, DN12, DN26,

and RSW have a significantly lower mobility than wild-type Cf-9-

TAP (Figures 3B and 3D), consistent with additional glycosyla-

tion. Asexpected, controlmutantsDQ12andDQ26have asimilar

mobility as Cf-9-TAP (Figure 3B). The molecular mass of the

DN18 and DQ18 mutants could not consistently be measured

relative towild-typeCf-9-TAP, but DN18 always had adecreased

mobility when compared with DQ18, indicating that DN18 also

carries additional glycosylation. The phenotypes of the PGS

addition mutants are summarized in Figure 3C.

Glycosylation Patterns Are Conserved in Other

Proteins with eLRRs

The glycosylation pattern and its role for Cf-9 activity may also

hold for other eLRR proteins. We therefore examined the pattern

of PGSs in other eLRR proteins (Table 1). The number of PGSs in

LRR domains of eLRR proteins differs significantly, ranging from

0 for LePRK2 to 29 for Cf-2 (Figure 5A). Despite this variation, the

overall distribution pattern of PGSs is similar for all these proteins

(Figure 5A). PGSs dominate at three major and four minor

Figure 3. (continued).

(B) Table showing activities and sizes of all Cf-9 PGS removal and addition mutants. First column: acronym of Cf-9 mutant (*, originally identified as EMS

mutant). Second column: hypersensitive response–inducing activity when agroinfiltrated into Avr9-transgenic tobacco (þþ, activity similar to wild-type

Cf-9-TAP; þ, activity less than Cf-9-TAP; 6, threshold activity; �, no detectable activity). These data represent an average of at least three

agroinfiltration assays. Third column: protein levels of agroinfiltrated PGSmutants analyzed by immunoblots compared with wild-type Cf-9-TAP (wt). All

TAP-tagged mutant proteins except mutant 13ND accumulate to similar levels as wild-type Cf-9-TAP. Fourth column: size of PGS mutants on

immunoblots when compared with wild-type Cf-9-TAP. Samples from at least two independent agroinfiltrations were analyzed multiple times on

immunoblots next to Cf-9-TAP samples. Size differences compared with wild-type Cf-9-TAP were scored as <, #, ¼, $, or > wild-type Cf-9-TAP and

averaged in the right column. #, shown in immunoblots in (D).

(C) Summary of the positions and activities of the Cf-9 PGS removal (rectangles) and addition (circles) mutants. The gray scale indicates the effect of the

mutations on Cf-9 activity as indicated in the accompanying legend.

(D) Representative immunoblots for some of the Cf-9 PGS mutants. Both PGS removal and PGS addition mutants are shown. Lanes marked with a C

were loaded with the Cf-9-TAP control.

1006 The Plant Cell

positions in the LRR consensus (Figure 5B). Asn in the conserved

Asn ladder (N in xxLxLxxN) are never part of NxS/T signatures,

presumably because these amino acids are buried in the

structure. PGSs in the a-helix region of LRRs are common in

many eLRR proteins containing PGSs (Figure 5A) and may be

essential for eLRR function, as we found for Cf-9.

When plotted on the LRR module, it is evident that a PGS at

the middle position of the putative b-sheet of adjacent LRRs

does not occur, with Cf-9 being the exception (Figure 5C). In the

case of Cf-9, however, we found that PGS18 in LRR24 is not

glycosylated, whereas PGS19 in LRR25 is. By contrast, PGSs in

adjacent putative a-helices are common among eLRR proteins

(Figure 5D). This difference can be explained because glycosyl-

ation of adjacent putative b-sheets may cause steric hindrance,

as this is the concave side of the curved LRR structure, whereas

glycosylation at adjacent putative a-helices is at the convex,

more spacious side of the LRR curvature. Another interesting

feature to note is that, as in Cf-9, most eLRR proteins have at

least 8 to 12 adjacent LRRswithout a PGS at the putative b-sheet

(Figure 5C), suggesting that these nonglycosylated regions pro-

vide a recognition surface for specific protein–protein interac-

tions (Kobe andDeisenhofer, 1996). PGSs at the putativeb-sheet

are confined to the four most C-terminal LRRs.

DISCUSSION

Cf-9 carries several amino acid motifs that could contribute to its

structure and function. In this investigation, we focused our

attention on conserved amino acids in LRR-flanking domains

and on putative glycosylation sites. The N-terminal LRR flanking

B-domain contains two Trp that are essential for Cf-9 activity and

four Cys that synergistically contribute to Cf-9 activity. We also

found that most of the putative glycosylation sites contribute to

Cf-9 activity and that those in putative a-helices are essential.

Figure 4. Mass Spectrometric Analysis of the Cf-9-TAP Protein.

Cf-9-TAP protein was immunopurified from a large-scale agroinfiltration of N. benthamiana leaves, separated on SDS gel and stained, with colloidal

Coomassie blue. The protein at 170 kD, corresponding to Cf-9-TAP, was excised from gel, digested with trypsin, and analyzed three times by MALDI-

TOF (MS1 to 3) and once by Q-TOF.

(A) Assigned peaks (peptides a to o) from the MALDI analysis. Mox, oxidized Met.

(B) Bar representation of Cf-9-TAP and derived tryptic peptides. Trypsin sites are indicated in white lines on the black Cf-9-TAP bar. The 22 PGSs are

indicated with circles and numbered 1 to 22. Closed circles indicate decreased molecular mass of the PGS removal mutant protein as observed on

immunoblots (see Figure 2B). Peptides found by three independent MALDI-TOF analyses (MS1 to 3) are named a to o andmatch those mentioned in (A).

Gray peptides m and n can originate from two regions within the TAP tag. Note that only peptides lacking a PGS were found, with the exception of

a peptide containing PGS18. Q-TOF analysis confirmed the identity of this fragment.


Cf-9 is glycosylated at all PGSs, except for PGS18, and adding

glycosylation sites at the putative concave b-sheet surface can

abolish Cf-9 activity.

A Model for the Cf-9 Structure

To understand the various effects of the mutations, a homology

model was generated for the structure of Cf-9, based on the

crystal structures of bean (Phaseolus vulgaris) PGIP (Di Matteo

et al., 2003) and bacterial LRRprotein internalin A (InlA) (Schubert

et al., 2002). Although Cf-9 displays only 34 and 28% sequence

identity to PGIP and InlA, respectively, the repetitive nature of the

LRR motif was of considerable help in generating a sequence

alignment for homology modeling of the N-terminal LRR flanking

B-domain and LRRs of Cf-9. However, the first 50 amino acids

(including 27 amino acids from the B-domain) and the last 68

amino acids of Cf-9 (domains E to G) were omitted from the

model because no suitable template structures could be found.

Also, because neither of the templates contains a feature com-

parable to the loop-out region, we explored several alternative

conformations for this region, generated by loop searching

techniques, before selecting a model consisting of two a-helices

joined by a short loop, based on criteria of chemical feasibility.

The structural model of Cf-9 shows a typical nonglobular LRR

fold (Figures 6A and 6B). At the concave (inner) side, the parallel

putative b-sheet surface is curved and twisted in a right-handed

superhelix, like the template structures fromwhich it is derived. It

is at this putative b-sheet surface where specificity is determined

Figure 5. Distribution of Putative N-Glycosylation Sites in the LRR Domains of Proteins with eLRRs.

(A) Distribution of PGSs (N in NxS/T motifs) for each eLRR protein. For each position in the LRR consensus sequence (at the top, putative solvent-

exposed positions are underlined), the number of PGSs was counted and plotted in a matrix, with the number of PGSs indicated in the different gray

scales (indicated in key). The total number of PGSs and the number of PGSs in the LRR domain are indicated at the left. Not all putative a-helices in the

LRR domain have the same length (represented as fading matrix at the right).

(B) Distribution of PGSs in LRR consensus of LRR proteins. PGSs given in the matrix of (A) were counted for each position in the LRR consensus

sequence (at the bottom). The total number of PGSs at the seven major positions is indicated at the top of the bars.

(C) Distribution of PGSs in putative b-sheets of LRRs in eLRR proteins. Each block represents the putative solvent-exposed amino acid residues of

putative b-sheets of each LRR module, from the N-terminal side (top) to the C-terminal side (bottom). Black or gray boxes indicate the presence of

a PGS at these putative solvent-exposed positions, as explained by the key (top). Proteins are ordered by their glycosylation patterns.

(D) Distribution of PGSs in putative a-helices of LRRs in eLRR proteins. Each block represents residues in putative a-helices of each LRR module, from

the N-terminal side (top) to the C-terminal side (bottom). The gray scale indicates the position of the PGS within the putative a-helix, as explained by the

key (top). Proteins are ordered as in (C).

1008 The Plant Cell

by solvent-exposed residues (Van der Hoorn et al., 2001; Wulff

et al., 2001). The putative a-helices of the LRRs provide the outer

surface of the structure and act as wedges that cause curvature

of the putative b-sheet surface. The B-domain is compact and

caps the hydrophobic core of the LRR.

The proposed nonglobular fold of Cf-9 is supported by

experimental data. In gel filtration experiments, purified 160-kD

myc-tagged Cf-9 protein was found to migrate with an apparent

molecularmass of 420 kD (Rivas et al., 2002; Van derHoorn et al.,

2003). Deglycosylated Cf-9 still migrates with a large apparent

molecular mass, indicating that this effect is due to the non-

globular shape of the Cf-9 protein itself, rather than to its

glycosylation (Van der Hoorn et al., 2003). The apparent molec-

ular mass of Cf-9 is similar to the 440-kD ferritin that was used as

Figure 6. Model of the Cf-9 Structure.

Homology model of the Cf-9 structure with studied amino acid residues colored as indicated in the legend.

(A) and (B) Illustration of the full Cf-9 model from side view (A) and top view (B). Note the position of the essential PGSs (red) and the introduced PGSs

(purple).

(C) to (F) Illustration (C) and space-filling ([D] to [F]) representations of the B-domains of Cf-9 ([C] and [D]), PGIP (E), and FLS2 (F). The orientation of the

B-domain is the same as in (A), with the concave (inner) side of the LRR domain turned toward the viewer on the lower part of the structures. Note the

orientation of the coplanar Trp (green) and clustered Cys bridges (yellow) in the B-domain and that the Cys bridge is exposed in Cf-9 and FLS2.

(G) Illustration of LRRs 24 to 27 of Cf-9, showing PGSs at the putative b-sheet surface. PGS18 (blue) is not glycosylated, whereas PGSs 19 to 22 are.

Introduction of glycosylation between PGS19 and PGS21 in PGS addition mutant DN26 abolishes Cf-9 function.


a calibration marker in these gel filtration assays. Ferritin is

a globular protein complex with a physical radius of ;74 A

(Granier, 1997). The Cf-9model would fit in a sphere with a radius

of 51 A, but the model lacks the N and C termini, the N-glycans,

and detergent molecules. However, although these predictions

may match the experimental data, extreme caution should be

taken when interpreting gel filtration experiments (Van der Hoorn

et al., 2003).

The extended spring-like structure of Cf-9 is probably quite

flexible, and the curvature of the LRR domain and exposure of

residues at the concave inner surface may change upon binding

to other proteins. These changes were detected in LRR protein

RI upon binding with ribonuclease A (Kobe and Deisenhofer,

1996). Even more flexibility is expected to reside in the loop-out

region, which may take different conformations and perhaps act

as a hinge between the two LRRdomains. Manymutations in this

region were found to inactivate the BRI1 eLRR protein (Li and

Chory, 1997; Friedrichsen et al., 2000). The inactivity of Cf-9

mutant 18SL may also result from interference with function of

the loop-out region.

Role of Conserved Residues in the N-Terminal

LRR-Flanking B-Domain

The N-terminal B-domain of Cf-9 preceding the LRRs contains

two conserved Trp (W1andW2) that are essential for Cf-9 activity

and four Cys (C2 to C5) that synergistically contribute to Cf-9

activity. In the PGIP crystal structure, the Trp side chains are

adjacent and coplanar, lying in the core of the B-domain,

whereas the four Cys are linked into two Cys pairs by disulfide

bonds. These features are preserved in our model. Thus, four

Cys of Cf-9 are connected by two putative disulfide bridges, C2-

C3 and C4-C5 (Figure 6C). These putative disulfide connections

comply with our phenotypic data because C2A andC3Amutants

have a similar reduced activity, and C4A and C5A mutants also

have a similar weak activity. The weak activity of the double

mutant C2,3A indicates that these Cys have additive roles

besides holding the putative disulfide bridge. Intramolecular

disulfide bridges in LRR-flanking domains are common, as they

are also present in structures of PGIP (Mattei et al., 2001) and

glycoprotein Iba (Huizinga et al., 2002). The Trp, however, are

only conserved amongst plant eLRR proteins. It is likely that both

the coplanar Trp and the two clustered Cys bridges render this

domain unusually rigid. Disturbance of this rigidity in the sub-

stitution mutants probably causes the severe phenotypes.

Alternatively, the LRR-flanking Cys could also be involved in

intermolecular disulfide bridges, connecting to other interacting

proteins. This has been suggested for CLV1, which migrates at

a higher molecular weight in the absence of reducing agent

(Trotochaud et al., 1999). Notably, the conserved and important

C4-C5 pair is probably exposed at the B-domain of Cf-9, at the

putative concave inner surface of the protein (Figure 6D). This

disulfide bridge is less exposed in PGIP (Figure 6E) but clearly

exposed in a model for the B-domain of FLS2 that we generated

(Figure 6F).

However, although the putative C4-C5 Cys bridge is probably

exposed, in contrast with CLV1 studies, no disulfide-linked

complexes have been observed for Cf-9 (Rivas et al., 2002) nor

has it been reported for any of the other well-characterized eLRR

proteins. Nevertheless, disulfide bridges may exist transiently

when complexes are assembled. Transient assembly of signal-

ing complexes occurs for the TGFb receptor, resulting in

a (TbRI)2(TbRII)2 heterotetramer complex (Schlessinger, 2000).

A similar transient assembly of signaling complexes appears to

occur for CLV1 and possibly also for BRI1 andBAK1 (Trotochaud

et al., 1999; Li et al., 2002; Nam and Li, 2002). LRR-flanking Cys

may contribute to the stabilization of these complexes through

transient intermolecular disulfide bridges. Redox-induced disul-

fide rearrangements play essential roles in the function of NPR1

(Mou et al., 2003) and integrins (Yan andSmith, 2001) andmay be

common in regulation of proteins (Hogg, 2003).

Distribution and Role of N-Glycosylation

We also showed that nearly all of the 22 PGSs are glycosylated

in Cf-9. These PGSs are evenly distributed on the Cf-9 pro-

tein (Figures 6A and 6B, red and orange residues). The shift in

molecular mass of the PGS removal mutants indicates that each

Cf-9molecule is fully glycosylated at each of these PGSs. Little is

known about the identity of theN-glycans in Cf-9, but it has been

shown that PGIP is N-glycosylated with a typical plant N-glycan

composed of threeN-acetylglucosamines, three mannoses, one

xylose, and one fucose (Mattei et al., 2001). Because Cf-9 and

PGIP were both produced in Nicotiana species, it is conceivable

that Cf-9 carries similar N-glycans for each PGS.

Immunoblot experiments and MS of Cf-9 tryptic digests

showed that PGS18 is not glycosylated. In PGIP, one PGS at

the b-sheet surface of the LRR domain also lacks glycosylation

(Mattei et al., 2001). It is unclear why specifically these sites in

PGIP and Cf-9 are not glycosylated. In the Cf-9 model, PGS18 is

not exposed (Figure 6G, blue residue), but this does not explain

the absence of an N-glycan because N-glycosylation would

occur cotranslationally (Glabe et al., 1980). Studies in mammals

showed that glycosylation can be inhibited by some amino acid

residues, especially by Pro, at positions x and y in NxS/Ty

(Shakin-Eshleman et al., 1996; Mellquist et al., 1998), but the two

PGS sequences of Cf-9 and PGIP do not show this signature.

Remarkably, most of the 22 PGSs contribute to Cf-9 activity,

and we identified four PGSs where the presence of N-glycans is

essential for Cf-9 activity. These four are all in putative a-helices,

exposed at the putative convex surface of the LRR domain

(Figures 6A and 6B, red residues).Many other eLRRproteins also

carry PGSs at these positions (Figure 5), presumably because

these residues are exposed at the surface of the protein. There

are several ways to explain the loss of activity of the many PGS

removal mutants of Cf-9. Glycosylation may protect proteins

from degradation (Gahring et al., 2001), but the Cf-9 PGS

removal mutants are all stable proteins. N-glycans may also

force a conformation of the LRR domain, facilitate interactions

with the cell wall, serve as a quality control checkpoint for

extracellular targeting, or play other roles (Leconte et al., 1994).

Glycosylation at the Putative Concave b-Sheet Surface

The putative concave LRR b-sheet is thought to provide a rec-

ognition surface where specific protein–protein interactions

1010 The Plant Cell

occur (Kobe and Deisenhofer, 1996). Consistent with this, it was

found that specificity determinants of PGIP, Cf-9, andCf-4 reside

in amino acids at putative solvent-exposed positions at the

putative concave inner side of the eLRR proteins (Leckie et al.,

1999; Van der Hoorn et al., 2001; Wulff et al., 2001).

Remarkably, we found that the C-terminal four LRRs of Cf-9

carry two N-glycans at the middle position of the putative

concave b-sheet surface, indicating that putative solvent-

exposed amino acids at this surface are unable to mediate

specific protein–protein interactions. The presence of a PGS at

these middle positions in C-terminal LRRs occurs often in other

eLRR proteins (Figure 5), although removal of these N-glycans in

Cf-9 does not affect Cf-9 activity (Figures 3B and 3C).

The middle position of the putative b-sheet surface is not

glycosylated in adjacent LRRs. Thus, PGSs 19 and 21 are at the

putative b-sheet surface of LRRs 25 and 27. PGS18 at the

putative b-sheet surface of LRR24 is not glycosylated. Adding

glycosylation at the putative b-sheet surface of LRR26 disrupts

Cf-9 function (mutant DN26), possibly because of steric hin-

drance with the adjacent N-glycans in LRRs 25 and 27 (Figure

6G), which may cause a significantly disturbed LRR structure in

this region.

In the first 22 LRRs before the loop-out domain, Cf-9 carries

only oneN-glycan at theb-sheet surface, but this is at the edge of

the b-sheet surface of LRR14 (Figure 6B, orange residue). The

activities of PGS addition mutants correlate with the conclusion

that specificity of Cf-9 resides in LRRs 13 to 17 (Wulff et al., 2001).

Introducing N-glycans in the adjacent LRR12 or LRR18 likely

disturbs specific interactions, which renders the mutants in-

active. By contrast, glycosylation is tolerated when more distant

in LRR6.

Other Mutant eLRR Proteins Have Similar Phenotypes

The investigated Trp and Cys in the B-domain, as well as the

glycosylation pattern, are well conserved amongst other eLRR

proteins. This suggests that similar phenotypes would result if

similar mutations are introduced in RLPs and RLKs with eLRRs.

Indeed, this prediction already holds for two mutants that

resulted from EMS screens.

In the bri1-5mutant, the second Cys of the conserved Cys pair

in the B-domain of BRI1 has been replaced by Tyr (C69Y). This

mutation has a weak phenotype, resembling the weak activity of

the corresponding C5A mutant of Cf-9. The weak phenotype of

the bri1-5 allele proved useful for activation tagging experiments

that resulted in the identification of BAK1, an RLK that interacts

with BRI1 and acts in BRI1 signaling (Li et al., 2002), and of BRS1,

a Ser protease that probably processes a protein involved in an

early event in BRI1 signaling (Li et al., 2001). Overexpression of

each of these genes can complement the bri1-5 phenotype.

In the clv1-8 mutant, the D295N mutation results in the

introduction of a PGS at the middle position of the putative

b-sheet (LRR9 in CLV1) in the middle of a region of 15 LRRs that

do not contain a PGS in the putative b-sheet (Clark et al., 1997).

The inactivity of thismutant resembles that of Cf-9mutants DN12

and DN18 that also contain an additional N-glycan on the

putative b-sheets of LRRs 12 and 18, respectively. Interestingly,

the clv1-8 mutation is dominant negative, probably because it

interferes with other eLRR proteins that have a functional overlap

with CLV1 (Dievart et al., 2003). This suggests that PGS addition

mutants might still assemble in signaling complexes that are not

active.

Taken together, this extensive structure–function analysis of

Cf-9 has provided valuable details on eLRR protein function and

creates important leads for further investigation of this function-

ally diverse protein family.

METHODS

Cloning Procedures

All DNA manipulations were performed using standard protocols

(Sambrook et al., 1989). PCR was performed with Pfu polymerase

(Stratagene, La Jolla, CA), and the authenticity of all cloned PCR frag-

ments was confirmed by sequencing. pMOG410 is described elsewhere

(Hoodet al., 1993) andcarries theb-glucuronidasegene interruptedby the

second intron of the potato (Solanum tuberosum) ST-LS1 gene (Vancan-

neyt et al., 1990). Binary vector pRH109 (Cf-9) was described previously

(Van der Hoorn et al., 2001). Oligonucleotide primers were synthesized by

Amersham-Pharmacia Biotech (Buckinghamshire, UK) or Sigma-Aldrich

(St. Louis, MO) and are summarized in Supplemental Table P online.

Construction of (Mutant) myc-Tagged Cf-9 Vectors

Binary vector pRH385 (encoding myc-tagged Cf-9) was generated as

follows. The fragment containing the 35S promoter and part of the

tobacco (Nicotiana tabacum) Pr1a gene encoding the signal peptide was

amplified from pRH87 (containing 35S-SP-Avr4; Van der Hoorn et al.,

2000) with primers XF5 and SR1 and cloned into the pBluescript SK�(pBS; Stratagene) using XbaI and SalI restriction sites, resulting in pRHx1.

The fragment encoding four myc epitope tags interrupted by the intron of

the potatoST-LS1 gene (Eckes et al., 1986) was amplified frompMOG410

using primers MF4 and MR4 (both encoding a double cMyc epitope tag)

and cloned into pRHx1 using SalI and XhoI restriction sites, resulting in

pRHx2. The fragment carrying the 35S promoter and encoding the A- and

B-domains of Cf-9 protein from pRH109was replaced by a PCR fragment

encoding only the B-domain, using primers XFX and CR1 and XbaI and

ClaI restriction sites. This resulted in pRHx3, containing an introduced

XhoI site encoding two Ser, followed by the N terminus of the mature Cf-9

protein. pRH385 was subsequently generated from this construct by

cloning the 35S-SP-2myc-intron-2myc fragment from pRHx2 into

pRHx3, using XbaI and XhoI restriction sites.

Fragments containing the mutations C1A, C3A, C4A, W1A, and W2A

were generated by amplification with primer pairs C1F þ CR1, XFX þC3R, XFX þ C4R, XFX þW1R, and XFX þW2R, respectively. Fragments

containing the mutations K1A and C2A were generated by PCR overlap

extension using border primer pair XFX þ CR1 and overlap primer pairs

K1f þ K1r and C2f þ C2r, respectively. These mutant fragments were

generated using Cf-9 as a template and cloned into pBS using XhoI and

ClaI restriction sites. These fragments were subsequently used to replace

the fragments encoding the B-domain in plasmid pRH385, resulting in

mutated myc-Cf-9. Plasmid names for the myc-Cf-9 mutants are as

follows: pRH406 (C1A), -407 (C2A), -408 (C3A), -409 (C4A), -445 (C2,3A),

-494 (K1A), -477 (W1A), -468 (W2A), and -459 (W2T).

Fragments carrying the C5Amutations were generated by PCR onCf-9

templates, using primer pair C5F þ BR1, cloned into pBS, and used to

replace the ClaI-BamHI fragment encoding LRRs 1 to 17 in pRH385,

resulting in C5A-mutated myc-Cf-9 (pRH410). The fragment carrying the

C6A mutation was amplified by PCR overlap extension from Cf-9 using

border primers BF2þ PR1 and overlap primers C6fþ C6r. This fragment


was cloned and used to replace the fragment encoding LRRs 17-domain

G of pRH385, resulting in C6A-mutated myc-Cf-9 (pRH451).

For construction of the C1-5A and C1-6A mutants, the C1-4A mutant

B-domain was amplified using primers CF1 and C34R on template

pRH407, which already contains the C2A mutation in Cf-9. This C1-4A

mutated B-domain was used to replace the B-domain of the C5A mutant

(pRH410), using XhoI andClaI restriction sites, generating C1-5Amutants

of myc-Cf-9 (pRH483). The C6A mutation was subsequently introduced

into pRH483 by replacing the fragment encoding theC-terminal half of the

Cf proteins, resulting in a C1-6A mutated version of myc-Cf-9 (pRH520).

The C4,5A mutant of myc-Cf-9 was constructed by replacing the

ClaI-BamHI fragment pRH409 (C4A), encoding the C-terminal part of

the B-domain and 17 LRRs, with that of pRH410 (C5A), thereby creating

pRH527 (C4,5A).

Cf-9 mutants lacking the myc-epitope tag were generated by replacing

the 35S-SP-2myc-intron-2myc fragment from the vectors mentioned

above by a cloned PCR fragment of the 35S promoter followed by

a fragment encoding the signal peptide of Cf-9, amplified from pRH109

using primers XF5 and XR1, using XbaI and XhoI restriction sites. Plasmid

names for the nontagged Cf-9 mutants are pRH440(C1A), -441(C2A),

-442(C3A), -443(C4A), -444(C5A), -468(C6A), -500(K1A), -480(W1A), and

-473(W2A).

Mutagenesis of Putative Glycosylation Sites

The mutagenesis of the putative glycosylation sites in Cf-9 was per-

formed in pBluescript toolkit constructs containing Cf-9 cassettes using

the Stratagene Quick Change kit according to the manufacturer’s

instructions. The pBS construct SLJ13950 (Rivas et al., 2002) contains

the full-length Cf-9 open reading frame fused to the TAP tag. SLJ20781

(pBS 59 Cf-9) was made by digesting SLJ13950 with HincII and ScaI and

cloning the 2101-bp 59 Cf-9 fragment in the EcoRV site of pBS KSþ.

SLJ20784 (pBS 39 Cf-9) was made by digesting SLJ13950 with ClaI and

BstBI and self-ligating to remove the 1992-bp 59 Cf-9 fragment. In the

case of mutants 19ND, 21ND, and DN26, the toolkit construct pSLJ9595

(Benghezal et al., 2000) was used as a template for mutagenesis, and in

the case of 18SL, the mutation was amplified from genomic DNA of the

mutant M2 (Wulff et al., 2004). Mutated fragments were cloned into

derivatives of the binary expression vector pSLJ14070 (Rivas et al., 2002)

behind the 35S promoter of Cauliflower mosaic virus using unique

restriction sites: In pSLJ20734, the full-length Cf-9:TAP fragment has

been replaced by a ClaI-BamHI cassette encoding part of the jellyfish

green fluorescent protein. In pSLJ20808, the 59 coding region of Cf-9

extending from the SnaBI site to the BsiWI site has been replaced by

a linker derived from oligos SnaBsiTOP and SnaBsiBOT (see Supple-

mental TableP online). In SLJ20809, theCf-9 39 coding sequence fused to

TAP and the 39 untranslated region have been replaced by a linker derived

from oligos BsiBamTop and BsiBamBot. All mutants were sequenced

from the binary constructs to exclude secondary PCR- or cloning-derived

mutations.

Agroinfiltration and Biochemical Procedures

Agroinfiltration of tobacco (N. tabacum) and N. benthamiana plants was

done as described previously (Van der Hoorn et al., 2000; Wulff et al.,

2001). Hypersensitive response–inducing activity was scored at 7 dpi

from at least three independent experiments by comparing activities with

that of the positive control (Cf-9, myc-Cf-9, or Cf-9-TAP). Activities are

summarized in the figures (withþþþ,þþ,þ,6, or�) as explained in the

legends. A more extensive quantitative analysis was performed for the

Cf-9 B-domain mutants in Figure 2B as described earlier (Van der

Hoorn et al., 2000).

For detection of myc- and TAP-tagged Cf proteins, infiltrated leaf

sectors were harvested at 1 dpi and ground in liquid nitrogen. Proteins

were dissolved in extraction buffer (50 mM Tris-HCl, pH 7.5, 150 mM

NaCl, 1% Triton X-100, 1 mM DTT, 0.5% polyvinyl polypyrrolidone, and

protease inhibitors [PI Complete tablet; Roche, Indianapolis, IN]) and

centrifuged for 15 min at 10,000g at 48C, and SDS sample buffer was

added to the supernatant. Protein gel blotting using anti-myc and

peroxidase antiperoxidase and ECL detection were done as described

previously (Van der Hoorn et al., 2003).

Mass Spectrometric Analysis of Cf-9-TAP Protein

Agrobacterium tumefaciens GV3101 pMP90 containing plasmid

SLJ14190 was used to transiently express TAP-tagged Cf-9 under the

control of its genomic promoter in N. benthamiana (Rivas et al., 2002).

Two days after agroinfiltration, 72 leaves were harvested and protein

extracts prepared as described (Rivas et al., 2002). Cf-9:TAP was

subjected to immunoprecipitation with IgG beads (Sigma-Aldrich) for 2 h

at 48C, after a preclearing step with agarose beads for 1 h at 48C. After

incubation, beadswere washed three timeswith extraction buffer [50mM

Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM 4-(2-aminoethyl)-benzenesulfonyl

fluoride (Pefablock; Sigma-Aldrich), 2 mg mL�1 of antipain, 2 mg mL�1

of leupeptin, and 2 mg mL�1 of aprotinin] supplemented with 40 mM

octylglucoside (Rivas et al., 2002). Immunoprecipitated proteins were

finally resuspended in SDS-PAGE sample buffer and separated on

a 7.5% SDS-polyacrylamide gel (Laemmli, 1970). The gel was fixed in

50% ethanol and 2% phosphoric acid for at least 3 h and washed three

times with water. Staining was performed in a solution of 34% methanol,

17% (NH4)2SO4, 3% phosphoric acid, and 0.66% Coomassie Brilliant

Blue G-250 for at least 1 d. A band of ;185 kD, corresponding to Cf-9:

TAP, was excised from the Coomassie-stained gel and subjected to in-

gel tryptic digestion (Shevchenko et al., 1996). Tryptic peptides were

eluted and analyzed by MALDI-TOF-MS and Q-TOF-MS/MS (John Innes

Centre Proteomics Facility, Norwich, UK). Data were analyzed with

Mascot and MS-Fit from Protein Prospector (http://prospector.ucsf.

edu/) and subjected to database searches against the Cf-9-TAP protein

sequence.

Homology Modeling

Homology modeling was performed using the Homology module of

Insight II (Insight II, Release 2000.1; Accelrys, Cambridge, UK). The x-ray

crystal structures of PGIP (accession code 1OGQ) and InlA (1o6V) were

retrieved from the Protein Data Bank (Berman et al., 2000) and used as

templates to build the individual structures. Modeling was restricted to

residues 51-795 of Cf-9 and 24-115 of FLS2. The B-domains were

constructed using 1OGQ as a template, ensuring that the conserved Trp

and Cys residues common to both 1OGQ and Cf-9 were matched in the

sequence alignment. Disulfide linkages were constructed between res-

idue pairs 53-77 and 78-85 of Cf-9 and 61-68 of FLS2. Tomodel the large

number of LRRs, a composite template was assembled using overlap-

ping copies of 1OGQ and the LRR section of 1O6V by performing head-

to-tail local structural alignments on the first and last few LRRs of each

template as appropriate. The sequence alignment of Cf-9 with its

template is included in the supplemental data online. Loops were

generated using loop searching facility within Insight II and a database

of loops taken from a parsed subset of release 103 of the Protein Data

Bank. The model structures were energy minimized to a derivative of 1.0

using the consistent valence force field within the Discover module and

validated using Procheck plus SwissPDBviewer (http://www.expasy.ch/

spdbv/mainpage.html) for inspection of the Ramachandran plots. Figures

were generated usingMolscript (Kraulis, 1991) and Raster 3D (Merritt and

Bacon, 1997). The Cf-9 and FLS2 models are provided in the supple-

mental data online.

1012 The Plant Cell

ACKNOWLEDGMENTS

We thank David Studholme for rendering Cf-9 structures; David Jones

for providing templates for mutants 19ND, 21ND, and DN26; Lilian Fritz-

Laylin and Matthieu Joosten for useful suggestions; and the John Innes

Centre Proteomic Facility (Norwich, UK) for sample preparation and

analysis. This work was supported by the Dutch Organization for

Scientific Research (Dutch Organization for Scientific Research, Veni,

and Talent grants to R.v.d.H.), the UK Biotechnology and Biological

Sciences Research Council (Grant 83/P13272 to S.R.), and the Gatsby

Charitable Foundation.

Received September 30, 2004; accepted December 22, 2004.

REFERENCES

Arabidopsis Genome Initiative (2000). Analysis of the genome

sequence of the flowering plant Arabidopsis thaliana. Nature 408,

796–815.

Baumberger, N., Doesseger, B., Guyot, R., Diet, A., Parsons, R.L.,

Clark, M.A., Simmons, M.P., Bedinger, P., Goff, S.A., Ringli, C.,

and Keller, B. (2003). Whole-genome comparison of leucine-rich

repeat extensins in Arabidopsis and rice. A conserved family of cell

wall proteins form a vegetative and a reproductive clade. Plant

Physiol. 131, 1313–1326.

Baumberger, N., Ringli, C., and Keller, B. (2001). The chimeric leucine-

rich repeat/extensin cell wall proteins LRX1 is required for root hair

morphogenesis in Arabidopsis thaliana. Genes Dev. 15, 1128–1139.

Belfanti, E., Silfverberg-Dilworth, E., Tartarini, S., Patocchi, A.,

Barbieri, M., Zhu, J., Vinatzer, B.A., Gianfranceshi, L., Gessler,

C., and Sansavini, S. (2004). The HcrVf2 gene from a wild apple

confers scab resistance to a transgenic cultivated variety. Proc. Natl.

Acad. Sci. USA 101, 886–890.

Benghezal, M., Wasteneys, G.O., and Jones, D.A. (2000). The

C-terminal dilysine motif confers endoplasmic reticulum localization

to type I membrane proteins in plants. Plant Cell 12, 1179–1201.

Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N.,

Weissig, H., Shindyalov, I.N., and Bourne, P.E. (2000). The protein

data bank. Nucleic Acids Res. 28, 235–242.

Canales, C., Bhatt, A.M., Scott, R., and Dickinson, H. (2002). EXS,

a putative LRR receptor kinase, regulates male germline cell number

and tapetal identity and promotes seed development in Arabidopsis.

Curr. Biol. 12, 1718–1727.

Clark, S.E., Williams, R., and Meyerowitz, E.M. (1997). The CLAVATA1

gene encodes a putative receptor kinase that controls shoot and floral

meristem size in Arabidopsis. Cell 89, 575–585.

Clay, N.K., and Nelson, T. (2002). VH1, a provascular cell-specific

receptor kinase that influences leaf cell patterns in Arabidopsis. Plant

Cell 14, 2707–2722.

Dievart, A., and Clark, S.E. (2003). Using mutant alleles to determine

the structure and function of leucine-rich repeat receptor kinases.

Curr. Opin. Plant Biol. 6, 507–516.

Dievart, A., Dalal, M., Tax, F.E., Lacey, A.D., Huttly, A., Li, J., and

Clark, S.E. (2003). CLAVATA1 dominant-negative alleles reveal func-

tional overlap between multiple receptors kinases that regulate

meristem and organ development. Plant Cell 15, 1198–1211.

Di Matteo, A., Federici, L., Mattei, B., Salvi, G., Johnson, K.A.,

Savino, C., De Lorenzo, G., Tsernolou, D., and Cervone, F. (2003).

The crystal structure of polygalacturonase-inhibiting protein (PGIP),

a leucine-rich repeat protein involved in plant defense. Proc. Natl.

Acad. Sci. USA 100, 10124–10128.

Dixon, M.S., Jones, D.A., Keddie, J.S., Thomas, C.M., Harrison, K.,

and Jones, J.D.G. (1996). The tomato Cf-2 disease resistance locus

comprises two functional genes encoding leucine-rich repeat pro-

teins. Cell 84, 451–459.

Eckes, P., Rosahl, S., Schell, J., and Willmitzer, L. (1986). Isolation

and characterisation of a light-inducible, organ-specific gene from

potato and the analysis of its expression after tagging and transfer

into tobacco and potato shoots. Mol. Gen. Genet. 199, 216–224.

Endre, G., Kereszt, A., Kevel, Z., Mihacea, S., Kalo, P., and Kiss, G.B.

(2002). A receptor kinase gene regulating symbiotic nodule develop-

ment. Nature 417, 962–966.

Ferrari, S., Vairo, D., Ausubel, F.M., Cervone, F., and De Lorenzo, G.

(2003). Tandemly duplicated Arabidopsis genes that encode

polygalacturonase-inhibiting proteins are regulated coordinately by

different signal transduction pathways in response to fungal infection.

Plant Cell 15, 93–106.

Friedrichsen, D.M., Joazeiro, C.A.P., Li, J., Hunter, T., and Chory, J.

(2000). Brassinosteroid-insensitive-1 is a ubiquitously expressed

leucine-rich repeat receptor serine/threonine kinase. Plant Physiol.

123, 1247–1255.

Gahring, L.C., Carlson, N.G., Meyer, E.L., and Rogers, S.W. (2001).

Granzyme B proteolysis of a neuronal glutamate receptor generates

an autoantigen and is modulated by glycosylation. J. Immunol. 166,

1433–1438.

Glabe, C.G., Hanover, J.A., and Lennarz, W.J. (1980). Glycosylation of

ovalbumin nascent chains. The spatial relationship between trans-

lation and glycosylation. J. Biol. Chem. 255, 9236–9242.

Gomez-Gomez, L., Bauer, Z., and Boller, T. (2001). Both the extra-

cellular leucine-rich repeat domain and the kinase activity of FLS2 are

required for flagellin binding and signaling in Arabidopsis. Plant Cell

13, 1155–1163.

Gomez-Gomez, L., and Boller, T. (2000). FLS2: An LRR receptor-like

kinase involved in the perception of the bacterial elicitor flagellin in

Arabidopsis. Mol. Cell 5, 1003–1011.

Granier, T. (1997). Comparison of the structures of cubic and tetrag-

onal forms of horse-spleen apoferritin. Acta Crystallogr. D Biol.

Chrystallogr. 53, 580–587.

Guyon, V., Tang, W.-H., Monti, M.M., Raiola, A., De Lorenzo, G.,

McCormick, S., and Taylor, L.P. (2004). Antisense phenotypes

reveal a role for SHY, a pollen-specific leucine-rich repeat protein,

in pollen tube growth. Plant J. 39, 643–654.

Hogg, P.J. (2003). Disulfide bonds as switches for protein function.

Trends Biochem. Sci. 28, 210–214.

Hood, E.E., Gelvin, S.B., Melchers, L.S., and Hoekema, A. (1993).

New Agrobacterium helper plasmids for gene transfer to plants.

Transgenic Res. 2, 208–218.

Huizinga, E.G., Tsuji, S., Romijn, R.A.P., Schiphorst, M.E., De Groot,

P.G., Sixma, J.J., and Gros, P. (2002). Structures of glycoprotein Iba

and its complex with von Willebrand factor A1 domain. Science 297,

1176–1179.

Jeong, S., Trotochaud, A.E., and Clark, S.E. (1999). The Arabidopsis

CLAVATA2 gene encodes a receptor-like protein required for the

stability of the CLAVATA1 receptor-like kinase. Plant Cell 11, 1925–

1933.

Jinn, T.-L., Stone, J.M., and Walker, J.C. (2000). HAESA, an Arabi-

dopsis leucine-rich repeat receptor kinase, controls floral organ

abscission. Genes Dev. 14, 108–117.

Jones, D.A., Thomas, C.M., Hammond-Kosack, K.E., Balint-Kurti,

P.J., and Jones, J.D.G. (1994). Isolation of the tomato Cf-9 gene for

resistance to Cladosporium fulvum by transposon tagging. Science

266, 789–793.

Kajava, A.V. (1998). Structural diversity of leucine-rich repeat proteins.

J. Mol. Biol. 277, 519–527.

Kawchuk, L.M., Hachey, J., Lynch, D.R., Kulcsar, F., van Rooijen, G.,

Waterer, D.R., Robertson, A., Kokko, E., Byers, R., Howard, R.J.,


Fischer, R., and Prufer, D. (2001). Tomato Ve disease resistance

genes encode cell surface-like receptors. Proc. Natl. Acad. Sci. USA

98, 6511–6515.

Kobe, B., and Deisenhofer, J. (1993). Crystal structure of porcine

ribonuclease inhibitor, a protein with leucine-rich repeats. Nature 366,

751–756.

Kobe, B., and Deisenhofer, J. (1996). Mechanism of ribonuclease

inhibition by ribonuclease inhibitor protein based on the crystal

structure of its complex with ribonuclease A. Mol. Biol. 264, 1028–

1043.

Kraulis, P.J. (1991). Molscript: A program to produce both detailed

and schematic plots of protein structures. J. Appl. Crystallogr. 24,

946–950.

Krusell, L., et al. (2002). Shoot control of root development

and nodulation is mediated by a receptor-like kinase. Nature 420,

422–426.

Laemmli, U.K. (1970). Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227, 680–685.

Leckie, F., Mattei, B., Capodicasa, C., Hemmings, A., Nuss, L.,

Aracri, B., De Lorenzo, G., and Cervone, F. (1999). The specificity of

polygalacturonase-inhibiting protein (PGIP): A single amino acid sub-

stitution in the solvent-exposed b-strand/b-turn region of the leucine-

rich repeats (LRRs) confers new recognition capability. EMBO J. 18,

2352–2363.

Leconte, I., Carpentier, J.-L., and Clauser, E. (1994). The functions of

the human insulin receptor are affected in different ways by mutation

of each of the four N-glycosylation sites in the b subunit. J. Biol.

Chem. 269, 18061–18071.

Li, J., and Chory, J. (1997). A putative leucine-rich repeat receptor

kinase involved in brassinosteroid signal transduction. Cell 90,

929–938.

Li, J., Lease, K.A., and Walker, J.C. (2001). BRS1, a serine carboxy-

peptidase, regulates BRI1 signaling in Arabidopsis thaliana. Proc.

Natl. Acad. Sci. USA 98, 5916–5921.

Li, J., Wen, J., Lease, K.A., Doke, J.T., Tax, F.E., and Walker, J.C.

(2002). BAK1, an Arabidopsis LRR receptor-like protein kinase,

interacts with BRI1 and modulates brassinosteroid signaling. Cell

110, 213–222.

Matsubayashi, Y., Ogawa, M., Morita, A., and Sakagami, Y. (2002).

An LRR receptor kinase involved in perception of a peptide plant

hormone, phytosulfokine. Science 296, 1470–1472.

Mattei, B., Bernalda, M.S., Federici, L., Roepstorff, P., Cervone,

F., and Boffi, A. (2001). Secondary structure and post-translational

modifications of the leucine-rich repeat protein PGIP (polygalactur-

onase-inhibiting protein) from Phaseolus vulgaris. Biochemistry 40,

569–576.

Mellquist, J.L., Kasturi, L., Spitalnik, S.L., and Shakin-Eshleman,

S.H. (1998). The amino acid following an Asn-X-Ser/Thr sequon is an

important determinant of N-linked core glycosylation efficiency. Bio-

chemistry 37, 6833–6837.

Merritt, E.A., and Bacon, D.J. (1997). Raster3D photorealistic molec-

ular graphics. Methods Enzymol. 277, 505–524.

Mou, Z., Fan, W., and Dong, X. (2003). Inducers of plant systemic

acquired resistance regulate NPR1 function through redox changes.

Cell 113, 935–944.

Muschietti, J., Eyal, Y., and McCormick, S. (1998). Pollen tube

localization implies a role in pollen-pistil interactions for the tomato

receptor-like protein kinases LePRK1 and LePRK2. Plant Cell 10,

319–330.

Nadeau, J.A., and Sack, F.D. (2002). Control of stomatal distribution on

the Arabidopsis leaf surface. Science 296, 1697–1700.

Nam, K.H., and Li, J. (2002). BRI1/BAK1, a receptor kinase pair

mediating brassinosteroid signaling. Cell 110, 203–212.

Nishimura, R., Hayashi, M., Wu, G.-J., Kouchi, H., Imaizumi-Anraku,

H., Murakami, Y., Kawasaki, S., Akao, S., Ohmori, M., Nagasawa,

M., Harada, K., and Kawaguchi, M. (2002). HAR1 mediates systemic

regulation of symbiotic organ development. Nature 420, 426–429.

Nomura, T., Bishop, G.J., Kaneta, T., Reid, J.B., Chory, J., and

Yokota, T. (2003). The LKA gene is a brassinosteroid insensitive-1

homolog of pea. Plant J. 36, 291–300.

Parniske, M., Hammond-Kosack, K.E., Golstein, C., Thomas, C.M.,

Jones, D.A., Harrison, K., Wulff, B.B.H., and Jones, J.D.G. (1997).

Novel disease resistance specificities result from sequence exchange

between tandemly repeated genes at the Cf4/9 locus of tomato. Cell

91, 821–832.

Piedras, P., Rivas, S., Droge, S., Hillmer, S., and Jones, J.D.G. (2000).

Functional, c-myc-tagged Cf-9 resistance gene products are plasma-

membrane localized and glycosylated. Plant J. 21, 529–536.

Rivas, S., Romeis, T., and Jones, J.D.G. (2002). The Cf-9 disease

resistance protein is present in an ;420-kilodalton heteromultimeric

membrane-associated complex at one molecule per complex. Plant

Cell 14, 689–702.

Ron, M., and Avni, A. (2004). The receptor for the fungal elicitor

ethylene-inducing xylanase is a member of a resistance-like gene

family in tomato. Plant Cell 16, 1604–1615.

Sambrook, J., Fritsch, E.F., and Maniatis, T.T. (1989). Molecular

Cloning: A Laboratory Manual, 2nd ed. (Cold Spring Harbor, NY: Cold

Spring Harbor Laboratory Press).

Scheer, J.M., and Ryan, C.A., Jr. (2002). The systemin receptor SR160

from Lycopersicon peruvianum is a member of the LRR receptor

kinase family. Proc. Natl. Acad. Sci. USA 99, 9585–9590.

Schlessinger, J. (2000). Cell signalling by receptor tyrosine kinases.

Cell 103, 211–225.

Schubert, W.D., Urbanke, C., Ziehm, T., Beier, V., Machner, M.P.,

Domann, E., Wehland, J., Chakraborty, T., and Heinz, D.W. (2002).

Structure of internalin, a major invasion protein of Listeria monocyto-

genes, in complex with its human receptor E-cadherin. Cell 111,

825–836.

Searle, I.R., Men, A.E., Laniya, T.S., Buzas, D.M., Iturbe-Ormaetxy,

I., Caroll, B.J., and Gresshoff, P.M. (2003). Long-distance signaling

in nodulation directed by a CLAVATA1-like receptor kinase. Science

299, 109–112.

Shakin-Eshleman, S.H., Spitalnik, S.L., and Katuri, L. (1996). The

amino acid at the X position of an Asn-X-Ser sequon is an important

determinant of N-linked core-glycosylation efficiency. J. Biol. Chem.

271, 6363–6366.

Shiu, S.-H., and Bleecker, A.B. (2001). Receptor-like kinases from

Arabidopsis form a monophyletic gene family related to animal

receptor kinases. Proc. Natl. Acad. Sci. USA 98, 10763–10768.

Shevchenko, A., Jensen, O.N., Podtelejnikov, A.V., Sagliocco, F.,

Wilm, M., Vorm, O., Mortensen, P., Shevchenko, A., Boucherie, H.,

and Mann, M. (1996). Linking genome and proteome by mass

spectrometry: Large-scale identification of yeast proteins from two

dimensional gels. Proc. Natl. Acad. Sci. USA 93, 14440–14445.

Song, W.-Y., Wang, G.-L., Chen, L.-L., Kim, H.-K., Pi, L.-Y., Holsten,

T., Gardner, J., Wang, B., Zhai, W.-X., Zhu, L.-H., Fauquet, C., and

Ronald, P. (1995). A receptor-like protein encoded by the rice disease

resistance gene, Xa21. Science 270, 1804–1806.

Stracke, S., Kistner, C., Yoshida, S., Mulder, L., Sato, S., Kaneko, T.,

Tabata, S., Sandal, N., Stougaard, J., Szczyglowski, K., and

Parniske, M. (2002). A receptor-like kinase required for both bacterial

and fungal symbiosis. Nature 417, 959–962.

Sun, X., Cao, Y., Yang, Z., Li, X., Wng, S., and Zhang, Q. (2004). Xa26,

a gene conferring resistance to Xanthomonas oryzae pv. oryzae

in rice, encodes an LRR receptor kinase-like protein. Plant J. 37,

517–527.

1014 The Plant Cell

Taguchi-Shiobara, F., Yuan, Z., Hake, S., and Jackson, D. (2001). The

fasciated ear2 gene encodes a leucine-rich repeat receptor-like

protein that regulates shoot meristem proliferation in maize. Genes

Dev. 15, 2755–2766.

Tang, W., Ezcurra, I., Muschietti, J., and McCormick, S. (2002). A

cysteine-rich extracellular protein, LAT52, interacts with the extracel-

lular domain of the pollen receptor kinase LePRK2. Plant Cell 14,

2277–2287.

Tor, M., Brown, D., Cooper, A., Woods-Tor, A., Sjolander, K., Jones,

J.D.G., and Holub, E.B. (2004). Arabidopsis downy mildew resistance

gene RPP27 encodes a receptor-like protein similar to CLAVATA2

and tomato Cf-9. Plant Physiol. 135, 1100–1112.

Torii, K.U., Mitsukawa, N., Oosumi, T., Matsuura, Y., Yokoyama, R.,

Whittier, R.F., and Komeda, Y. (1996). The Arabidopsis ERECTA

gene encodes a putative receptor protein kinase with extracellular

leucine-rich repeats. Plant Cell 8, 735–746.

Toubart, P., Desiderio, A., Salvi, G., Cervone, F., Daroda, L., and De

Lorenzo, G. (1992). Cloning and characterization of the gene encod-

ing the endopolygalacturonase-inhibiting protein (PGIP) of Phaseolus

vulgaris L. Plant J. 2, 367–373.

Trotochaud, A.E., Hae, T., Wu, G., Yang, Z., and Clark, S.E. (1999).

The CLAVATA1 receptor-like kinase requires CLAVATA3 for its

assembly into a signaling complex that includes KAPP and a rho-

related protein. Plant Cell 11, 393–405.

Vancanneyt, G., Schmidt, R., O’Connor-Sanchez, A., Willmitzer, L.,

and Rocha-Sosa, M. (1990). Construction of an intron-containing

gene: Splicing of the intron in transgenic plants and its use in

monitoring early events in Agrobacterium-mediated plant transforma-

tion. Mol. Gen. Genet. 220, 245–250.

Van der Hoorn, R.A.L., Laurent, F., Roth, R., and De Wit, P.J.G.M.

(2000). Agroinfiltration is a versatile tool that facilitates comparative

analysis of Avr9/Cf-9-induced and Avr4/Cf-4-induced necrosis. Mol.

Plant Microbe Interact. 13, 439–446.

Van der Hoorn, R.A.L., Rivas, S., Wulff, B.B.H., Jones, J.D.G., and

Joosten, M.H.A.J. (2003). Rapid migration in gel filtration of the Cf-4

and Cf-9 resistance proteins is an intrinsic property of Cf proteins and

not because of their association with high-molecular-weight proteins.

Plant J. 35, 305–315.

Van der Hoorn, R.A.L., Roth, R., and De Wit, P.J.G.M. (2001).

Identification of distinct specificity determinants in resistance protein

Cf-4 allows construction of a Cf-9 mutant that confers recognition of

avirulence protein AVR4. Plant Cell 13, 273–285.

Vinatzer, B.A., Patocchi, A., Gianfranceschi, L., Tartarini, S.,

Zhang, H.-B., Gessler, C., and Sansavini, S. (2001). Apple contains

receptor-like genes homologous to the Cladosporium fulvum resis-

tance gene family of tomato with a cluster of genes cosegregating with

Vf apple scab resistance. Mol. Plant Microbe Interact. 14, 508–515.

Voinnet, O., Rivas, S., Mestre, P., and Baulcombe, D. (2003). An

enhanced transient expression system in plants based on suppres-

sion of gene silencing by the p19 protein of tomato busy stunt virus.

Plant J. 33, 949–956.

Walker, J.C. (1993). Receptor-like protein kinase genes of Arabidopsis

thaliana. Plant J. 3, 451–456.

Wang, Z.-W., Seto, H., Fuijoka, S., Yoshida, S., and Chory, J. (2001).

BRI1 is a critical component of a plasma-membrane receptor for plant

steroids. Nature 410, 380–383.

Wengier, D., Valsecchi, I., Cabanas, M.L., Tang, W.H., McCormick,

S., and Muschietti, J. (2003). The receptor kinases LePRK1 and

LePRK2 associate in pollen and when expressed in yeast, but

dissociate in the presence of style extract. Proc. Natl. Acad. Sci.

USA 100, 6860–6865.

Worrall, D., Elias, L., Ashford, D., Smallwood, M., Sidebottom, C.,

Lillford, P., Telford, J., Holt, C., and Bowles, D. (1998). A carrot

leucine-rich-repeat protein that inhibits ice recrystallization. Science

282, 115–117.

Wulff, B.B.H., Thomas, C.M., Parniske, M., and Jones, J.D.G. (2004).

Genetic variation at the tomato Cf-4/Cf-9 locus induced by EMS

mutagenesis and intralocus recombination. Genetics 167, 459–470.

Wulff, B.B.H., Thomas, C.M., Smoker, M., Grant, M., and Jones,

J.D.G. (2001). Domain swapping and gene shuffling identify se-

quences required for induction of an Avr-dependent hypersensitive

response by the tomato Cf-4 and Cf-9 proteins. Plant Cell 13, 255–272.

Yamamuro, C., Ihara, Y., Wu, X., Noguchi, T., Fujioka, S., Takatsuto,

S., Ashikari, M., Kitano, H., and Matsuoka, M. (2000). Loss of

function of a rice brassinosteroid insensitive 1 homolog prevents

internode elongation and bending of the lamina joint. Plant Cell 12,

1591–1605.

Yan, B., and Smith, J.W. (2001). Mechanism of integrin activation by

disulphide reduction. Biochemistry 40, 8861–8867.

Zhao, D.-Z., Wang, G.-F., Speal, B., and Ma, H. (2002). The EXCESS

MICROSPOROCYTES1 gene encodes a putative leucine-rich repeat

receptor protein kinase that controls somatic and reproductive cell

fates in the Arabidopsis anther. Genes Dev. 16, 2021–2031.


DOI 10.1105/tpc.104.028118; originally published online February 18, 2005; 2005;17;1000-1015Plant Cell

Pierre J.G.M. de Wit and Jonathan D.G. JonesRenier A.L. van der Hoorn, Brande B.H. Wulff, Susana Rivas, Marcus C. Durrant, Anke van der Ploeg,

Leucine-Rich RepeatsFunction Analysis of Cf-9, a Receptor-Like Protein with Extracytoplasmic−Structure

This information is current as of June 5, 2020

Supplemental Data /content/suppl/2005/02/01/tpc.104.028118.DC1.html

References /content/17/3/1000.full.html#ref-list-1

This article cites 86 articles, 42 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

http://www.plantcell.org/cgi/alerts/ctmain

http://www.plantcell.org/cgi/alerts/ctmain

http://www.aspb.org/publications/subscriptions.cfm

Structure–Function Analysis of Cf-9, a ... - Plant CellFigure 1. Plant eLRR Proteins with Known...

Documents

Transcript of Structure–Function Analysis of Cf-9, a ... - Plant CellFigure 1. Plant eLRR Proteins with Known...