Nematode CLE signaling in Arabidopsis requires CLAVATA2 and CORYNE


(fax +1 573 884 9676; e-mail


Plant-parasitic cyst nematodes secrete CLAVATA3 (CLV3)/ESR (CLE)-like effector proteins. These proteins have been shown to act as ligand mimics of plant CLE peptides and are required for successful nematode infection; however, the receptors for nematode CLE-like peptides have not been identified. Here we demonstrate that CLV2 and CORYNE (CRN), members of the receptor kinase family, are required for nematode CLE signaling. Exogenous peptide assays and overexpression of nematode CLEs in Arabidopsis demonstrated that CLV2 and CRN are required for perception of nematode CLEs. In addition, promoter–reporter assays showed that both receptors are expressed in nematode-induced syncytia. Lastly, infection assays with receptor mutants revealed a decrease in both nematode infection and syncytium size. Taken together, our results indicate that perception of nematode CLEs by CLV2 and CRN is not only required for successful nematode infection but is also involved in the formation and/or maintenance of nematode-induced syncytia.


Biotrophs are pathogens that establish intimate parasitic relationships with the host that they infect. Often these relationships involve some kind of modification or reprogramming of the host cell(s) to accommodate the pathogen’s subsequent growth and development. Plant-parasitic nematodes are biotrophs that mainly attack the roots of plants and cause crop damage of over $100 billion annually (Sasser and Freckman, 1987). The most economically important plant-parasitic nematodes include the cyst-forming nematodes of Heterodera and Globodera spp. These sedentary endoparasitic nematodes form intimate parasitic relationships with their hosts by penetrating the root as motile juveniles and migrating intracellularly until they reach the root vasculature where they select a single cell to initiate a feeding site. The initial syncytial cell undergoes developmental changes to re-differentiate into a syncytium to support subsequent nematode growth and development in later sedentary stages (Davis et al., 2004). The syncytium forms when neighboring cells fuse as a result of partial cell wall degradation (Endo, 1964), creating a permanent feeding cell that shares characteristics with plant cell types including meristematic cells, endosperm cells, transfer cells, and developing xylem (Mitchum et al., 2008). Development and maintenance of the syncytium is dependent on the secretory effector proteins originating in the esophageal gland cells and delivered into the host root through the stylet of the plant-parasitic nematode (Davis et al., 2008). Recently, the CLAVATA3 (CLV3)/ESR (CLE)-like effector proteins secreted by cyst nematodes have been shown to act as ligand mimics of plant CLE peptides, and are required for successful nematode infection (Wang et al., 2005; Patel et al., 2008; Lu et al., 2009; Wang et al., 2010a,b).

Plant CLEs are small peptide ligands involved in regulating a population of specialized cells, called stem cells, which allow post-embryonic organogenesis to occur (Simon and Stahl, 2006). These stem cell pools can be found in the shoot apical meristem (SAM), the root apical meristem (RAM), and the vascular cambium. Whether or not these stems cells remain in an undifferentiated state or differentiate into new plant tissues is tightly controlled by CLE signaling pathways. In Arabidopsis, the population of stem cells which resides in the organizing center (OC) of the SAM is maintained by the expression of the transcription factor WUSCHEL (WUS) (Laux et al., 1996). Differentiation of those stems cells is promoted when the ligand–receptor pair of CLV3, a small extracellular peptide ligand in the CLE family (Fletcher et al., 1999; Rojo et al., 2002), binds to CLV1 (Ogawa et al., 2008), a leucine-rich-repeat receptor-like kinase (LRR-RLK) and downregulates WUS. Previous models have suggested that CLV1 forms a receptor complex with the LRR-receptor-like protein (RLP) CLV2 (Clark et al., 1993; Kayes and Clark, 1998; Jeong et al., 1999; Trotochaud et al., 1999). More recently, it has been suggested that CLV1 acts in parallel or together with the heterodimer receptor complex of CLV2 and CORYNE (CRN) (Miwa et al., 2008; Muller et al., 2008; Bleckmann et al., 2010; Guo et al., 2010; Zhu et al., 2010). In comparison to the SAM, much less is known about the regulation of stem cells in the RAM. The quiescent center (QC) is the equivalent to the OC in the SAM. However, there are significant differences between the OC and the QC. In contrast to the OC, the cells surrounding the QC are maintained as stem cells. In addition, stem cells are differentiated in both proximal and distal directions. This indicates that there is a signaling ligand involved in cell–cell communication to maintain the cells surrounding the QC as stem cells, and a signal to promote differentiation (Sarkar et al., 2007; Stahl et al., 2009). Previous reports have identified that the WUS-related homeobox 5 (WOX5) transcription factor is expressed in the QC of the RAM and is required to maintain the distal stem cell pool (Sarkar et al., 2007). Recently it has been shown that CLE40, the closest homolog to CLV3, is expressed in the columella cells and regulates expression of WOX5 (Stahl et al., 2009). The WOX5/CLE40 signaling pathway appears only to control the distal stem cell pool, indicating that other CLE signaling pathways may exist to control the proximal stem cell pool. Consistent with these observations, a number of Arabidopsis CLEs are expressed in roots (Sharma et al., 2003), and when some of these CLEs are overexpressed they have been shown to cause premature termination of the primary root meristem (Fiers et al., 2004; Strabala et al., 2006; Meng et al., 2010). In addition, the short root phenotype has been shown to be dependent on perception by CLV2 and CRN (Casamitjana-Martinez et al., 2003; Fiers et al., 2005; Miwa et al., 2008; Meng et al., 2010). Taken together this indicates that a CLV-like and a CLE-controlled signaling pathway can act in the root.

CLE-like genes from nematodes have been reported in the soybean cyst nematode (SCN; Heterodera glycines) (Wang et al., 2005, 2010a), the beet cyst nematode (BCN; Heterodera schachtii) (Patel et al., 2008; Wang et al., 2010b), and the potato cyst nematode (PCN; Globodera rostochiensis) (Lu et al., 2009). Beet cyst nematode CLEs have been detected in the dorsal gland ampulla, indicating they are probably secreted from the stylet into host cells (Patel et al., 2008). More recently, SCN CLEs have been shown to be secreted directly to the syncytial cytoplasm where the variable domain is thought to redirect the nematode CLE peptides to the apoplast (Wang et al., 2010a). These findings suggest that when delivered to the apoplast, nematode CLEs would be available to interact with extracellular host receptors to function as ligand mimics of plant CLE signaling pathways. Overexpression studies have shown that nematode CLEs can trigger plant CLE signaling pathways (Wang et al., 2005; Lu et al., 2009; Wang et al., 2010a,b), but the identity of the receptors and downstream signaling pathways that are activated to initiate developmental cascades required for the re-differentiation of root cells to form syncytia are currently unknown.

In this paper we describe the use of synthetic CLE peptides, nematode CLE overexpression lines, promoter–reporter lines, and nematode infection assays of receptor mutants to investigate a role for CLV2 and CRN in nematode CLE signaling. Our results indicate that the CLV2/CRN signaling pathway is required for successful nematode infection and syncytium development.


CLV2 and CRN are required for nematode CLE perception

We have previously shown that exogenously applied 12-amino-acid (aa) peptides corresponding to the CLE motifs of the SCN (HgCLEs) and the BCN (HsCLEs) CLEs can function as plant CLE peptide mimics causing termination of the primary root meristem in a concentration-dependent manner (Wang et al., 2010b). In fact, HsCLE2 was found to share an identical 12-aa CLE motif with Arabidopsis CLEs 5 and 6 (Wang et al., 2010b). In Arabidopsis, it has been shown that the short root phenotype caused by overexpression or exogenous application of some plant CLE peptides is dependent on CLV2 signaling (Fiers et al., 2005; Miwa et al., 2008; Muller, 2008; Meng et al., 2010). More recent evidence indicates that CLV2 forms a complex with CRN and can transmit the signal from CLV3 binding in a CLV1-independent manner (Miwa et al., 2008; Muller et al., 2008; Bleckmann et al., 2010; Zhu et al., 2010). To determine whether or not CLV2 and CRN might play a role in perception of cyst nematode CLEs we screened the Arabidopsis clv2-1 null mutant and the crn-1 amorphic allele for resistance to the HgCLE, HsCLE1, and HsCLE2 12-aa peptides. Seeds were grown on vertical plates in the absence of exogenous peptide or in the presence of 1 μm HgCLE or 10 μm of the HsCLEs and roots were measured 9 days after germination. Wild-type seedlings [Landsberg erecta (Ler)] had statistically shorter roots when grown on plates with any of the CLE peptides in comparison with the no-peptide control (Figure 1a). In contrast, root growth in clv2-1 and crn-1 was relatively unimpaired in the presence of the different CLE peptides (Figure 1a). The same observation was made with sol2-1, another mutant allele of CRN (Miwa et al., 2008) (Figure S1 in Supporting Information). Previous reports have indicated that the short root phenotype can be attributed to a decrease in the number of meristematic cells in the RAM (Fiers et al., 2005). Using Nomarski optics, we confirmed that clv2-1 and crn-1 were insensitive to peptide application, resulting in root meristems that were indistinguishable from the no-peptide control (Figure 1b–d).

Figure 1.

 Effect of cyst nematode CLAVATA3 (CLV3)/ESR (CLE) peptides on receptor mutants.
(a) Average root length of wild-type (Ler), clv2-1, and crn-1 seedlings grown for 9 days on media with or without the synthetic nematode dodecapeptide CLE motif. Data represent the mean ± SE, = 10. Asterisks indicate statistically significant differences in root length of peptide-treated mutants compared with the respective peptide-treated Ler root by Student’s t-test (< 0.05). Data are representative of three independent experiments.
(b–d) Representative roots tips of seedlings grown on media with or without synthetic CLE peptides for 10 days and visualized with differential interference microscopy: (b) no peptide; (c) sensitive to peptide, and (d) resistant to peptide. Brackets indicate the region of the root apical meristem containing dividing cells. Scale bar, 50 μm.

Nematode CLEs function in planta through a CLV2- and CRN-dependent pathway

Overexpression of HgCLE2, HsCLE1, and HsCLE2 in wild-type Arabidopsis has been shown to cause a wus-like phenotype similar to other plant CLEs (Strabala et al., 2006; Meng et al., 2010; Wang et al., 2005, 2010a,b). If CLV2 and/or CRN are involved in perception of nematode CLEs then we would expect the phenotypes to be diminished or abolished when overexpressed in clv2-1 and/or crn-1. Each of the nematode CLE genes were cloned into an overexpression vector and transformed into the mutant backgrounds. Transgenic seedlings in the T1 generation were screened and characterized in soil. In contrast to the overexpression phenotypes seen in wild-type Arabidopsis, where a high percentage of wus-like phenotypes were observed when the peptides were correctly targeted to the apoplast (Wang et al., 2010a,b), no wus-like phenotypes were observed when HgCLE2, HsCLE1, and HsCLE2 were overexpressed in clv2-1 or crn-1 (Table 1). These results demonstrate that mutations in CRN and CLV2 abolish nematode CLE overexpression phenotypes.

Table 1.   Summary of nematode CLE overexpression phenotypes in clv2-1 and crn-1
BackgroundConstructT1 shoot phenotypes
wus-like (%)WT (%)T1 lines (no.)Total T1 (no.)
  1. aWang et al., 2010a.

  2. bWang et al., 2010b.

  3. WT, wild type.


Spatial and temporal relationship between CLV2, CRN, and nematode feeding sites

Cyst nematodes enter the root near the zone of elongation, migrate through root cortical cells using their stylet to puncture through cell walls, and begin feeding from a single cell near the vascular cylinder. CLE peptides, originating from the highly active dorsal esophageal gland cell, are delivered through the feeding stylet to the cytoplasm of the host root cell (Wang et al., 2010a). In order for CLV2 and CRN to be able to perceive the nematode CLEs as ligand mimics they must be expressed in the correct spatial and temporal context.

Using a CRN:GUS transgene in Arabidopsis, CRN expression was previously shown to be expressed throughout the root including the vasculature where the nematode initiates feeding (Figure 2a–c; Muller et al., 2008). To confirm whether CRN is expressed in nematode feeding sites, transgenic Arabidopsis seedlings expressing CRN:GUS were infected with BCN and monitored during nematode development. Expression of GUS was detected in feeding sites as soon as early second-stage juveniles (J2) began to feed. (Figure 2d). The expression of GUS reached its peak once nematodes reached late J2 parasitic stages, but remained detectable in the feeding sites of third-stage juvenile (J3) parasitic nematodes (Figure 2e–g). By the time the nematodes reached the fourth juvenile life stage (J4), GUS expression was either weak or absent in feeding sites (Figure 2h).

Figure 2.

CRN:GUS expression during nematode infection.
(a–c) The GUS expression in uninfected Arabidopsis root tips (a), middle section of the root showing GUS staining in the vascular tissue and a lateral root primordium (b), and GUS staining in the junction of a lateral root in the upper section of the root towards the hypocotyl (c).
(d–h) CRN:GUS expression in response to Heterodera schachtii; early parasitic second-stage juvenile (J2), 2 days post-inoculation (dpi) (d), late parasitic J2, 4 dpi (e), parasitic third-stage juvenile (J3), 6 dpi (f), longitudinal section through the syncytium (g) corresponding to the region of the root highlighted by the black box in (f), parasitic fourth-stage juvenile (J4), 10 dpi (h).
Abbreviations: N, nematode; Syn, syncytium. Scale bar, 50 μm.

Similar to CRN, CLV2 is expressed in many different vegetative tissues (Jeong et al., 1999). However, little is known about the expression pattern of CLV2 in roots. For an ongoing study to visualize CLV2 expression in roots, mCherry was fused to the C-terminus of the Arabidopsis Histone 2B (H2B) gene and placed under the transcriptional control of the CLV2 promoter (AB and RS, unpublished data). The H2B protein has been shown to be a valid marker for chromatin organization in plant nuclei and has been used to describe development of the syncytial endosperm in Arabidopsis (Boisnard-Lorig et al., 2001). In uninfected roots, CLV2:H2B-mCherry fluorescence was detected throughout the root vasculature with the strongest expression detected in lateral root primordia and the zone of elongation extending down to the root apical meristem (Figure S2). Expression of CLV2 under the control of the endogenous promoter, using 1252 bp of the CLV2 5′ region was sufficient to rescue the clv2-1 mutant in all isolated lines (= 20). We utilized the Arabidopsis CLV2:H2B-mCherry transgenic line to evaluate CLV2 expression during nematode infection. Upon nematode infection, increased expression of CLV2:H2B-mCherry fluorescence was detected in the nuclei of syncytia fed upon by parasitic J2s (Figure 3a,b). At the J3 life stage, CLV2:H2B-mCherry continued to be specifically expressed within feeding sites (Figure 3c,d). No autofluorescence was detected in the nuclei of syncytia fed upon by parasitic J2s in wild-type plants (Figure 3e,f).

Figure 3.

 Confocal and differential interference contrast (DIC) images of CLV2:H2B-mCherry expression during nematode infection.
(a) Parasitic second-stage juvenile (J2) parasitic with DIC, 4 days post-inoculation (dpi), CLV2:H2B-mCherry transgenic line.
(b) Parasitic J2 with mCherry fluorescence, 4 dpi, CLV2:H2B-mCherry transgenic line.
(c) Parasitic third-stage juvenile (J3) with DIC, 6 dpi, CLV2:H2B-mCherry transgenic line.
(d) Parasitic J3 with mCherry fluorescence, 6 dpi, CLV2:H2B-mCherry transgenic line.
(e) Parasitic J2 with DIC, 4 dpi, wild-type plants.
(f) Parasitic J2 with mCherry fluorescence, 4 dpi, wild-type plants.
Abbreviations: N, nematode. Outline indicates the area of the syncytium. Scale bars, 50 μm.

Mutant alleles of CLV2 and CRN cause a reduction in nematode infection and defects in syncytial size

By using an RNA interference (RNAi) approach targeting nematode CLE genes, previous reports have shown that nematode CLE peptides are important for successful infection of host plant roots (Bakhetia et al., 2007; Patel et al., 2008). To determine if perception of nematode CLEs by CLV2 or the CLV2/CRN complex is required, root infection assays with nematodes were performed on the clv2-1 and crn-1 single mutants and the crn-1 clv2-1 double mutant. According to Muller et al. (2008), crn-1 clv2-1 is morphologically indistinguishable from either of the single mutants, indicating that they act in the same pathway. The mutant alleles and the wild-type Ler were randomized in 12-well plates and grown on modified Knop’s medium. Two weeks after germination, seedlings were inoculated with infective J2s. J4 females were counted at 14 days post-inoculation (dpi) and adult females were counted at 30 dpi. Both the single and double mutants showed a statistically significant reduction in nematode infection with the exception of crn-1 at 14 dpi (Figure 4a). At 30 dpi nematode infection was reduced by approximately 25% in all receptor mutants tested. A similar reduction in nematode infection across all mutant lines supports the hypothesis that CLV2 and CRN are acting in the same signaling pathway. Using sol2-1, we observed a 40% reduction in nematode infection (Figure S3a). Since the establishment of a feeding site is required for nematode development and reproduction, the above observations motivated us to determine if there were any defects in syncytial size between the receptor mutants and the wild type. The mutant alleles and the wild-type Ler were grown on vertical square plates and inoculated with infective J2s. At 14 dpi, syncytia that were transparent and fed upon by only one nematode were measured. The average area of wild-type Ler syncytia was 1402 ± 147 μm2 (Figure 4b). In contrast, a statistically significant reduction of 40% in syncytium size was observed in the receptor mutants. The average area of crn-1, clv2-1, and crn-1 clv2-1 was 797 ± 89, 745 ± 61, and 808 ± 57 μm2, respectively (Figure 4b). The same reduction in syncytium size was seen in the sol2-1 mutant allele (Figure S3b).

Figure 4.

 Effect of clv2-1 and crn-1 mutant alleles on Heterodera schachtii infection.
(a) Fourth-stage juvenile (J4) females were counted at 14 days post-inoculation (dpi) and adult females were counted at 30 dpi. Data represent mean ± SE, = 35 for Ler, 32 for crn-1, 34 for clv2-1, and 29 for crn-1 clv2-1. Data are representative of three independent experiments.
(b) Seedlings were grown on vertical square plates for 10 days and inoculated with 10 second-stage juveniles (J2s) per root. At 14 dpi, syncytia that fed only one nematode and appeared translucent were microscopically examined and their area was determined. Data represent mean ± SE, = 11 for Ler and crn-1, 14 for clv2-1, and 12 for crn-1 clv2-1.
Asterisks indicate statistically significant differences compared with Ler by Student’s t-test (< 0.05).


Nematode CLE genes have been found to be upregulated in the dorsal esophageal gland cell at the onset of parasitism and remain on through the adult female life stage. CLE genes are turned off in adult males that are no longer feeding (Wang et al., 2005; Patel et al., 2008; Lu et al., 2009; Wang et al., 2010a). In SCN and BCN, immunolocalization studies have localized nematode CLEs along the dorsal gland extension and in the ampulla at the base of the nematode stylet, indicating they are secreted into host plant roots via the stylet (Wang et al., 2005; Patel et al., 2008; Wang et al., 2010a). Consistent with these results an immunofluorescence study found that SCN CLEs are secreted directly into the root cytoplasm of the host plant (Wang et al., 2010a). The variable domain of SCN CLEs is then able to redirect the proteins into the apoplast where they can act as plant CLE ligand mimics by interacting with extracellular membrane-bound plant CLE receptors. However, thus far, host plant receptors that perceive nematode CLE signals have not been identified.

Many studies have used synthetic CLE peptides to help determine the roles that plant CLE peptides play in plant growth and development. Previous studies have shown that nematode CLE peptides cause root growth phenotypes similar to other plant CLEs (Lu et al., 2009; Wang et al., 2010a,b). Other studies have also shown that these peptide screens can identify receptors that may be involved in certain CLE signaling pathways by utilizing receptor mutants (Fiers et al., 2005; Stahl et al., 2009; Meng et al., 2010). To identify potential nematode CLE receptors we tested mutants of plant CLE receptors implicated in CLE signaling in the RAM. In the root, exogenous peptide assays and overexpression studies have shown that CLV2 is required for proper proximal meristem function (Stahl et al., 2009; Meng et al., 2010). In addition, a new member of the receptor kinase family, CRN, forms a heterodimer with CLV2 and is required for proper localization of the CLV2/CRN complex to the plasma membrane (Bleckmann et al., 2010; Zhu et al., 2010). In Arabidopsis, CRN has been found to be widely expressed in both shoot and root tissues, suggesting dual roles in shoot and root development (Muller et al., 2008). CLV2 has been found to be expressed in shoot tissues (Jeong et al., 1999), but less is known about its expression in the root. In this work we screened a null mutant allele of CLV2 and an amorphic mutant allele of CRN for resistance to the nematode CLE peptides. Both clv2-1 and crn-1 were resistant to HgCLE, HsCLE1, and HsCLE2 peptides (Figures 1 and S1). Similar to synthetic peptide assays, overexpression of HgCLE, HsCLE1, and HsCLE2 in the clv2-1 and crn-1 mutant backgrounds abolished the wus-like phenotypes seen when the nematode CLEs are overexpressed in wild-type backgrounds (Wang et al., 2005, 2010a,b). Taken together, the peptide assays and overexpression data indicate that CLV2 and CRN are required for perception of nematode CLE.

In order to serve as a receptor complex for nematode CLE peptides, CLV2 and CRN would most likely need to be expressed in feeding cell initials as well as the developing feeding sites. Using promoter–reporter lines we confirmed that both CLV2 and CRN were expressed in nematode-induced syncytia (Figures 2 and 3), consistent with a role in nematode CLE perception. The recent detection of CLV2 and CRN expression from microaspirated syncytial contents at 5 dpi by microarray analysis (Szakasits et al., 2009) supports this finding. It is also possible that nematode CLE receptors are expressed in the cells adjacent to the expanding syncytium. As the nematode CLEs are redirected to the host root apoplast, extracellular receptors of the adjacent cells that are primed for incorporation could trigger plant CLE signaling pathways needed to fully form the syncytium. In the future it will be interesting to more precisely localize the CLV2 and CRN proteins within syncytia using immunofluorescence techniques. This will aid in determining whether or not these nematode CLE receptors are localized within the cell wall openings that occur during syncytium formation or if they are localized on the outer plasma membrane of the syncytium and/or adjacent cells.

Previous reports have demonstrated that SCN and BCN CLEs are important for nematode parasitism by showing a reduction in nematode infection after knocking down CLE expression in the worm using RNAi approaches (Bakhetia et al., 2007; Patel et al., 2008). To directly test for a role of CLV2/CRN in nematode CLE perception we performed infection assays on the receptor mutants. We showed that a reduction in nematode infection occurs on the receptor mutants (Figures 4a and S3). Concurrently, we also saw a reduction in syncytium size in the receptor mutants (Figures 4b and S3). The fact that we saw a similar reduction in both nematode infection and syncytium size in both the single and double mutants is consistent with genetic and biochemical data that CLV2 and CRN are acting in the same pathway (Muller et al., 2008; Bleckmann et al., 2010; Zhu et al., 2010). These data indicate that not only is nematode CLE perception by CLV2 and CRN important for successful nematode infection, but demonstrates that CLE signaling also plays a role in feeding cell formation.

The involvement of CRN in nematode CLE signaling also opens up the interesting possibility that nematode CLE signaling may be directly or indirectly suppressing host plant defense responses. In root tips of sol2-1, another mutant allele of CRN, plant disease resistance-related and stress responsive genes were upregulated (Miwa et al., 2008). Therefore, when nematode CLEs are secreted they could activate the CLV2/CRN signaling pathway leading to a suppression of plant disease resistance-related and plant stress responsive genes. One might speculate that the main target for nematode CLEs is a signaling pathway which allows developmental programming of root cells for syncytium formation to occur and that suppression of plant defense responses is just an added benefit to the nematode. Alternatively, the nematode may require suppression of plant defense responses through plant CLE signaling in order for the syncytium to form properly. Further studies will need to be performed to investigate this possibility.

Several possibilities exist for why we only see a partial reduction in nematode numbers and syncytium size in the clv2-1 and crn-1 mutant backgrounds. First, besides CLEs, nematodes secrete many different effectors that probably play an important role in feeding cell formation (Wang et al., 2001; Gao et al., 2003). For example, when BCN CLEs were targeted with RNAi a similar partial reduction in nematode infection was observed (Patel et al., 2008), either as a consequence of limited reductions in transcript levels or an indication that the other effectors still active in the nematode allow infection to proceed. A second possibility for the partial reduction in the receptor mutants is that there could be multiple nematode CLE receptors. The nematode CLEs reported so far belong to gene families (Lu et al., 2009; Wang et al., 2010a,b). In addition, PCN CLEs have multiple CLE motifs that may be simultaneously processed to release different CLE peptides (Lu et al., 2009). This leaves the possibility that nematode CLE peptides may activate multiple plant CLE signaling pathways concurrently to function in an antagonistic or synergistic fashion as reported for plant CLEs (Whitford et al., 2008). The current plant CLV3 signaling pathway in the shoot indicates that there are parallel signaling pathways. Genetic evidence indicates that CLV1 acts in a separate pathway from the CLV2/CRN pathway (Muller et al., 2008). In support of the genetic data, recent reports using luciferase complementation assays and FRET analysis have shown that CLV1 forms a homodimer and that CLV2 and CRN form a heterodimer without CLV3 stimulation (Bleckmann et al., 2010; Zhu et al., 2010). There is also evidence for CLV1 interacting with the CLV2/CRN complex, leading to the possibility that different signaling pathways could be activated depending on which receptor in the complex interacts with the CLE ligand (Bleckmann et al., 2010; Guo et al., 2010; Meng and Feldman, 2010; Zhu et al., 2010). Recently, various plant CLE peptides were used as cold competitors for radiolabeled CLV3 CLE binding to CLV1, CLV2, and the CLV1-related BARELY ANY MERISTEM (BAM) 1 and BAM2 (Guo et al., 2010). Arabidopsis CLE5, which has an identical CLE motif to the recently identified nematode CLE, HsCLE2 (Wang et al., 2010b), was included in this study. CLE5 provided either full or partial competition to CLV3 binding for all four receptors (Guo et al., 2010), providing direct evidence that the nematode CLE and CLV2 can form a receptor–ligand complex. However, the binding was not specific to CLV2. Thus it is possible that in the crn-1 clv2-1 double mutants, nematodes are still able to signal through other receptors in the roots.

Additional candidate receptors could include CLV1 and/or BAM1 and BAM2. Unlike CLV2, which has a broad expression pattern in plants, CLV1 expression is thought to be restricted to the center of the SAM and its function is thought to be confined to stem cell specification in the shoot (Clark et al., 1997; Fletcher et al., 1999). Therefore, in order to utilize CLV1 as a receptor, nematodes would have to activate CLV1 expression in the roots. Recently, BAM1 and BAM2 have been shown to act redundantly in the SAM and are widely expressed throughout the plant, including in root tissues (DeYoung et al., 2006; DeYoung and Clark, 2008). We have found that bam1 is also resistant to exogenous application of synthetic nematode CLE peptides (AR, S. Chen, XW and MGM, unpublished data). Furthermore, there are over 200 LRR-RLKs in Arabidopsis and only a few receptor–CLE ligand pairs have been characterized (Shiu and Bleecker, 2001). Thus, further studies using reporter fusions and a combination of mutants will need to be performed to investigate the possible involvement of other host plant receptors in nematode CLE signaling.

To date, the exact function of nematode CLE proteins in syncytium formation is unresolved. However, this paper has shown that nematode CLE signaling through the CLV2/CRN receptor complex is important for proper syncytium formation and ultimately successful nematode infection. These findings open the door for identifying the downstream signaling components regulated by CLV2/CRN to uncover the role that nematode CLE signaling plays in syncytium formation.

Experimental procedures

Peptide assays

Arabidopsis seeds were sterilized using the chlorine gas method (Wang et al., 2010b). Sterilized seeds were germinated on vertical plates in a growth chamber at 22°C under long-day conditions (16 h light/8 h dark) containing synthetic peptides (Sigma-Genosys, as previously described (Wang et al., 2010b). The clv2-1 mutant in the Ler background (Koornneef et al., 1983) was obtained from the Arabidopsis Biological Resource Center ( The crn-1 mutant in the Ler background (Muller, 2008) and the sol2-1 mutant in the Utr background (Miwa et al., 2008) have been described previously. The HgCLEp, HsCLE1p, and HsCLE2p peptides used in this study were as described (Wang et al., 2010b). Two days after germination, root length was marked each day for 9 days. Plates were scanned using an Epson Perfection V200 PHOTO scanner ( and total root length was determined using Scion Image. Primary root tips of Arabidopsis were mounted on glass slides and visualized with an Olympus Vanox AHBT3 microscope ( equipped with Nomarski optics.

Overexpression in mutant backgrounds

The CLE gene sequences from the SCN (HgCLE2ΔSP) and the BCN (HsCLE1 and HsCLE2) used to generate the overexpression constructs were previously described (Wang et al., 2010a,b). Constructs were transformed into the mutant backgrounds using the Arabidopsis floral dip method (Clough and Bent, 1998). Seeds from primary Arabidopsis transformants (T1) were selected on 0.5 × MS medium [MS basal nutrients salts (Caisson Laboratories,, 2% sucrose, 0.8% Type A agar (Sigma,, pH 5.7] containing 50 μg ml−1 timentin (GlaxoSmithKline, to control Agrobacterium contamination and 50 μg ml−1 kanamycin and grown under the same conditions as above. Seedlings resistant to kanamycin were transplanted to soil 7 days after germination. Two weeks after transplanting to soil the shoot phenotypes were observed.

Promoter-reporter lines

CRN:GUS has been previously described and characterized (Muller et al., 2008). To generate CLV2:H2B-mCherry, vector pMDC99 (Curtis and Grossniklaus, 2003) was modified by introducing the CDS of chimeric construct mCherry-H2B at the 3′ site of the gateway cassette using the unique PacI restriction site to give pAB149. To analyze the expression of CLV2 1252 bp of the 5′ region and 9 bp of the CDS were amplified using the primers AB_CLV2_Pro_F (5′ CACCAGACACAAAGCCCTTTCCATTGTC 3′) and AB_CLV2_Pro_R (5′ CTTTATCATAGCTCAGAGGA 3′) to give a CACC-TOPO-containing amplicon, which was cloned into pENTR/D-TOPO® (Invitrogen, This entry clone was used in a LR reaction with pAB149 to give pAB183 (CLV2:H2B-mCherry).

Nematode infection of promoter–reporter lines

The BCN H. schachtii was propagated on greenhouse-grown sugar beets (Beta vulgaris cv Monohi). The eggs of the BCN were isolated and hatched as previously described (Mitchum et al., 2004). After 2 days, second-stage juveniles (J2) were collected and surfaced-sterilized according to Wang et al. (2007), except that 0.004% mercuric chloride, 0.004% sodium azide, and 0.002% Triton X-100 were used. Sterilized seeds were grown on modified Knop’s medium with Daishin agar (Brunschwig Chemie, (Sijmons et al., 1991). Ten days after germination, seedlings were inoculated with 20 sterilized J2 per root.

Histochemical β-glucuronidase (GUS) assays

At the indicated timepoints, freshly excised CRN:GUS tissues were infiltrated with GUS substrate buffer [0.5 mm 5-bromo-4chloro-3-indolyl glucuronide, 100 mm 2-amino-2-(hydroxymethyl)-1,3-propanediol (TRIS), pH 7.0, 50 mm NaCl, 0.06% Triton X-100, 3 mm potassium ferricyanide) and incubated overnight at 37°C (Jefferson et al., 1987). Stained roots were placed in glass Petri dishes and visualized with a Nikon Eclipse TS100 inverted microscope (

Confocal microscopy

CLV2:H2B-mCherry seed was sterilized, grown, and inoculated with nematodes as described above. At the indicated timepoints, infected roots were mounted on glass slides and visualized with a 510 META confocal scanning microscope (Carl Zeiss, excited at 543 nm.

Infection assay with receptor mutants

Sterilized receptor mutants were plated in 12-well Falcon tissue culture plates (BD Biosciences, containing modified Knop’s medium with 0.8% Daishin agar in a randomized block design. Plants were grown at 24°C with a 12-hour photoperiod. Fourteen days after germination, seedlings were inoculated with 200 surface-sterilized BCN J2. The J4 females were counted at 14 dpi and adult females were counted at 30 dpi. The average values were calculated and significant differences were determined by using Student’s t-test (< 0.05). To measure syncytium size, receptor mutants were germinated on modified Knop’s medium in vertical square plates and inoculated at 10 days after germination with 10 surface-sterilized BCN J2. At 14 dpi, syncytia that were transparent and fed upon by only one nematode were visualized with a Nikon Eclipse TS100 inverted microscope and photographed using a Nikon COOLPIX 5000 digital camera. The syncytia were outlined using the Adobe Photoshop CS5 magnetic lasso tool and the area of the longitudinal section was calculated by the software. This is similar to the approach others have recently taken to measure the area of syncytia (Siddique et al., 2009; Hofmann et al., 2010). Significant differences were determined by using Student’s t-test (< 0.05).


The authors would like to thank Robert Heinz for the maintenance of the nematode cultures and the MU Molecular Cytology Core for help with imaging. This work was supported by the USDA-NRI Competitive Grants Program (grant nos 2007-35607-17790 and 2009-35302-05304 to MGM and XW), a USDA Special Grant (grant no. 2008-34113-19420) to MGM, and an MU Life Sciences Graduate Research Fellowship to AR.