Nematode effector proteins originating from esophageal gland cells play central roles in suppressing plant defenses and in formation of the plant feeding cells that are required for growth and development of cyst nematodes. A gene (GrUBCEP12) encoding a unique ubiquitin carboxyl extension protein (UBCEP) that consists of a signal peptide for secretion, a mono-ubiquitin domain, and a 12 amino acid carboxyl extension protein (CEP12) domain was cloned from the potato cyst nematode Globodera rostochiensis. This GrUBCEP12 gene was expressed exclusively within the nematode's dorsal esophageal gland cell, and was up-regulated in the parasitic second-stage juvenile, correlating with the time when feeding cell formation is initiated. We showed that specific GrUBCEP12 knockdown via RNA interference reduced nematode parasitic success, and that over-expression of the secreted GrΔSPUBCEP12 protein in potato resulted in increased nematode susceptibility, providing direct evidence that this secreted effector is involved in plant parasitism. Using transient expression assays in Nicotiana benthamiana, we found that GrΔSPUBCEP12 is processed into free ubiquitin and a CEP12 peptide (GrCEP12) in planta, and that GrCEP12 suppresses resistance gene-mediated cell death. A target search showed that expression of RPN2a, a gene encoding a subunit of the 26S proteasome, was dramatically suppressed in GrΔSPUBCEP12 but not GrCEP12 over-expression plants when compared with control plants. Together, these results suggest that, when delivered into host plant cells, GrΔSPUBCEP12 becomes two functional units, one acting to suppress plant immunity and the other potentially affecting the host 26S proteasome, to promote feeding cell formation.
The cyst-forming nematodes of the genera Heterodera and Globodera are biotrophic pathogens that have evolved intimate parasitic relationships with their host plants. These obligate parasites penetrate host roots as motile juveniles and migrate towards the vasculature, where they transform selected root cells into a specialized feeding structure called a syncytium (Jones and Northcote, 1972). The syncytium is a metabolically active multi-nucleate structure with characteristic features such as a dense cytoplasm, increased numbers of organelles, enlarged nuclei and nucleoli, and thickened cell walls with elaborate wall ingrowths (Jones, 1981). Once the syncytium is initiated, the nematode becomes sedentary. The establishment of a syncytium is absolutely required for the nematode to complete the rest of its life cycle because this unique feeding structure serves as the sole source of nutrients for the developing nematode. Like other biotrophic phytopathogens, cyst nematodes secrete effector proteins into host plant cells to manipulate host cellular processes that enable successful parasitism. The majority of these effectors are synthesized in the esophageal gland cells (one dorsal and two sub-ventral), and are subsequently secreted into root cells through the nematode stylet, a hollow mouth spear (Hussey, 1989). It is now believed that these stylet-secreted effectors play central roles in root invasion, suppression of host defense, and syncytium formation and maintenance (Hussey, 1989; Davis et al., 2004, 2008).
Plants have developed an effective two-layered immune system to defend themselves against pathogen attacks. In addition to basal defense, now referred to as PAMP-triggered immunity (PTI), which involves the recognition of pathogen-associated molecular patterns (PAMPs) by membrane-residing pattern recognition receptors, plants utilize resistance proteins that directly or indirectly perceive specific pathogen effectors to activate effector-triggered immunity (ETI) (Chisholm et al., 2006; Jones and Dangl, 2006). ETI is often associated with development of the hypersensitive response, a type of programmed cell death that is localized to the infection site and effectively restricts pathogen growth (Heath, 2000). Phytopathogens, such as bacteria, oomycetes and fungi, on the other hand, deploy diverse effectors that suppress immune signaling to promote successful infection (Dou and Zhou, 2012). Evidence is emerging that effectors secreted from nematodes also play a direct role in modulating plant immunity (Postma et al., 2012; Jaouannet et al., 2013).
Ubiquitin is a highly conserved 76 amino acid protein that is found in all eukaryotic cells. The best-characterized function of ubiquitin is its role in protein degradation. In the ubiquitin/proteasome system (UPS), ubiquitin serves as a covalent molecular signal that directs target proteins to the 26S proteasome for degradation (Dreher and Callis, 2007). It is becoming obvious that the host UPS is involved in plant defense mechanisms. Interestingly, however, phytopathogens have also evolved mechanisms to manipulate the host UPS for their own benefit (Dielen et al., 2010; Shirsekar et al., 2010).
Ubiquitin proteins may be classified into two groups: polyubiquitin proteins and ubiquitin carboxyl extension proteins (UBCEPs). While polyubiquitins contain tandemly repeated ubiquitin monomers, UBCEPs are comprised of a single ubiquitin monomer fused to a carboxyl extension protein (CEP) of either 52 amino acids (CEP52) or 76-80 amino acids (CEP76-80) (Callis and Viestra, 1989). These ubiquitin precursors are rapidly processed by specific proteases during or soon after translation to release either free ubiquitin monomers or ubiquitin and a CEP (Jonnalagadda et al., 1989; Monia et al., 1989; Callis et al., 1990). The two types of CEPs then become associated with ribosomes (Finley et al., 1989; Redman and Rechsteiner, 1989; Callis et al., 1990). The ubiquitin moiety of UBCEPs was suggested to have a chaperone function that facilitates incorporation of CEPs into ribosomes (Finley et al., 1989).
Genes encoding UBCEPs have been cloned from the cyst nematode species Heterodera glycines (Gao et al., 2003) and Heterodera schachtii (Tytgat et al., 2004). Interestingly, nematode-encoded UBCEPs appear to represent a unique class, because they are secretory proteins and their CEPs have no similarity to ribosomal CEPs. Their specific gland expression indicates that nematode-produced UBCEPs are important effectors (Gao et al., 2003; Tytgat et al., 2004); however, their role in plant parasitism is yet to be elucidated.
Here we report the isolation and characterization of a UBCEP gene (GrUBCEP12; GenBank accession number JX308826) from the potato cyst nematode Globodera rostochiensis. Using multiple approaches, we provide direct evidence that the nematode-produced GrUBCEP12 protein is important for plant parasitism. More importantly, our results reveal that the secreted form of GrUBCEP12 (GrΔSPUBCEP12) is processed into free ubiquitin and a CEP12 peptide (GrCEP12) in planta, and that the derived GrCEP12 peptide suppresses cell death related to ETI. A role for GrCEP12 in suppressing host defense was further confirmed using GrCEP12 over-expression plants, which showed increased susceptibility to two unrelated pathogens. Furthermore, we found that RPN2a, a gene encoding a subunit of the 26S proteasome, is dramatically down-regulated in GrΔSPUBCEP12 but not GrCEP12 over-expression plants. The 26S proteasome is an essential component of the UPS. Taking together, we propose that, when secreted in planta through the nematode stylet, GrΔSPUBCEP12 is processed into two functional peptides that potentially act in different pathways to promote nematode parasitism.
Isolation of GrUBCEP12 from G. rostochiensis
By database searches coupled with PCR, we cloned a cDNA (GrUBCEP12) from G. rostochiensis that showed similarity to the previously identified Heterodera UBCEP genes (Gao et al., 2003; Tytgat et al., 2004). The GrUBCEP12 cDNA contained an open reading frame of 345 bp encoding a 114 amino acid protein that consisted of an N–terminal signal peptide (SP) of 26 amino acids (as predicted by SignalP; Petersen et al., 2011), a ubiquitin domain of 76 amino acids, and a carboxyl extension protein (CEP) domain of 12 amino acids (GrCEP12) (Figure S1). The GrCEP12 differs significantly from Heterodera CEPs and is much shorter, without any similarity to ribosome-associated CEPs (Figure S1).
GrUBCEP12 expression is gland-specific and developmentally regulated
In situ mRNA hybridization was used to determine the spatial expression of GrUBCEP12 in nematode tissues. When using the antisense cDNA probe specific for GrUBCEP12, hybridization signals were observed exclusively within the dorsal gland cell of both pre-parasitc and parasitic nematode life stages (Figure 1a,b). No signal was detected when the sense cDNA probe was used (Figure S2). Quantitative real-time RT–PCR analysis was performed to determine the GrUBCEP12 expression profile through the five nematode developmental stages: egg, pre-parasitic second-stage juvenile (pre–J2), and parasitic second-, third- and fourth-stage juveniles (par–J2, -J3, and -J4). GrUBCEP12 mRNA expression was high in pre–J2 and par–J2 stages, and then quickly diminished in later parasitic stages (Figure 1c).
GrUBCEP12 is secreted, proteolytically processed in planta, and localized in the cytoplasm and nucleus of plant cells
The specific gland expression and up-regulation of GrUBCEP12 during the early stages of nematode infection suggest that GrUBCEP12 is a secreted effector that is probably involved in facilitating syncytium formation. To verify the secretion of GrUBCEP12, pre–J2s were probed using an anti-ubiquitin antibody. Strong fluorescent signals were observed in both the dorsal gland cell body and the dorsal gland extension and ampulla (Figure 2a), where gland secretions are collected before their release through the nematode stylet. No or very faint signals were observed in other tissues except for the tail region (Figure S3), suggesting that ubiquitin itself is potentially present at low levels in most nematode tissues. The signal observed in the tail region may indicate ubiquitin synthesized from ubiquitin precursor genes. However, we believe that the strong staining in the dorsal gland cell and its associated extension and ampulla was due to the presence of the GrUBCEP12 protein because of its exclusive and high expression in the dorsal gland cell (Figure 1a,b). The detection of GrUBCEP12 in the dorsal gland cell in close proximity to the nematode stylet is a strong indication that the protein is secreted into host plant cells.
Polyubiquitin proteins and UBCEPs are expected to be rapidly cleaved by de-ubiquitinating enzymes after translation to release functional ubiquitin (Jonnalagadda et al., 1989; Monia et al., 1989; Callis et al., 1990). To determine whether GrUBCEP12 is processed by endogenous plant enzymes after delivery into plant cells, we created four constructs expressing various forms of GrUBCEP12 fused C–terminally to hemagglutinin (HA)-tagged GFP (green fluorescence protein) under the control of the CaMV 35S promoter: (i) a full-length form containing the SP (GrUBCEP12–GFP:HA), (ii) a secreted form lacking the SP (GrΔSPUBCEP12–GFP:HA), (iii) a secreted form mutated at the predicted ubiquitin cleavage site (GrΔSPUB–mut–CEP12–GFP:HA) (Hanania et al., 2009), and (iv) a truncated form containing only the CEP domain (GrCEP12–GFP:HA) (Figure 2c). Although the SP of GrUBCEP12 is expected to be cleaved off in the endoplasmic reticulum of the dorsal gland cell prior to secretion from the nematode stylet, inclusion of a construct expressing a pre-protein was thought to be helpful to determine whether GrUBCEP12 is processed during translocation through the secretory pathway. The constructs expressing GrCEP12–GFP:HA or GFP:HA were used as controls. These constructs were transiently expressed in leaf tissue of Nicotiana benthamiana using agroinfiltration, and the infiltrated leaves were then subjected to immunoblot analysis. When GrΔSPUBCEP12–GFP:HA was transiently expressed in N. benthamiana leaves, the anti-HA antibody specifically detected an accumulated protein of approximately 29 kDa (Figure 2b), which was much smaller than the expected size of GrΔSPUBCEP12–GFP:HA (approximately 38 kDa) and matched the expected size of GrCEP12–GFP:HA (approximately 29 kDa) (Figure 2b), indicating that the ubiquitin moiety was removed. Interestingly, however, the anti-HA antibody detected a protein that closely matched the estimated size of GrUBCEP12–GFP:HA (approximately 40 kDa) when the corresponding construct was expressed in planta (Figure 2b), showing that, unlike GrΔSPUBCEP12, the full-length GrUBCEP12 protein containing the SP was not processed. As expected, mutation of the ubiquitin cleavage site rendered the protein uncleavable because a protein that matched the estimated size of GrΔSPUB–mut–CEP12–GFP:HA (approximately 38 kDa) was detected when the mutated construct was expressed (Figure 2b). An expression signal was also observed for the GFP:HA control (Figure 2b).
GrUBCEP12 is expected to be delivered into the host cell through the nematode stylet. To examine the subcellular localization of the secreted GrΔSPUBCEP12 protein and the derived GrCEP12 in plant cells, two constructs were generated in which each protein was C–terminally fused to a joint GFP:GUS (β–glucuronidase) protein under the control of a double CaMV 35S promoter (Hewezi et al., 2008). The fusion of GrΔSPUBCEP12 or GrCEP12 with GFP:GUS generates a protein that is big enough to prevent passive protein diffusion between cellular compartments. Each construct was transiently expressed in onion epidermal cells by biolistic bombardment. GFP signals were observed in both the cytoplasm and the nucleus for both constructs (Figure 2d). For the vector control (GFP:GUS alone), the GFP signal was only observed in the cell cytoplasm (Figure S4).
Plant host-derived RNAi of GrUBCEP12 causes impaired nematode parasitism
Host-derived RNAi has been demonstrated to be a useful tool to silence genes within the feeding nematode (Sindhu et al., 2009). To test the potential of host-derived RNAi of GrUBCEP12 within the feeding nematode and to observe the resultant effects on nematode parasitism, we generated transgenic potato lines expressing hairpin dsRNA (Figure S5a) complementary to the unique region (Figure S5b) of GrUBCEP12 under the control of the superpromoter (Lee et al., 2007). Expression of the hairpin dsRNA in the obtained transgenic lines was confirmed by amplifying the single-stranded loop region (a 363 bp GUS fragment) (Figure S5a) in quantitative RT–PCR assays (Figure 3a). The phenotypes of transgenic lines confirmed to have hairpin dsRNA expression were not obviously different from those of the control lines. Five independent transgenic lines (#5, 46, 50, 54 and 55) that show high levels of dsRNA expression, as well as line #85, with a minimal level of transgene expression, and empty vector control lines were inoculated with G. rostochiensis to assess nematode parasitism. At 5 weeks after inoculation, a significantly lower number of nematode females was observed in the five independent RNAi lines compared to the control lines and line #85 (Figure 3b). To determine whether the reduction in female numbers resulted from dsRNA-mediated suppression of GrUBCEP12 within the feeding nematodes, we used quantitative RT–PCR to examine GrUBCEP12 expression in nematodes excised from roots of RNAi lines and the control lines. Nematodes recovered from the five RNAi lines showed a 2.5–5.5-fold reduction in GrUBCEP12 expression compared to those recovered from the control lines (Figure 3c), whereas nematodes recovered from RNAi line #85 showed no obvious reduction of GrUBCEP12 expression (Figure 3c). The suppression was specific for GrUBCEP12 because no reduction was observed for the non-target genes GrCM1 (Lu et al., 2008) and GrUBCEP52, a canonical UBCEP gene, in nematodes recovered from the independent RNAi lines compared to those from control lines (Figure S6).
Over-expression of GrΔSPUBCEP12 in potato affects plant susceptibility to G. rostochiensis
To assess potential effects of the secreted GrΔSPUBCEP12 protein on plant phenotype and nematode susceptibility, we generated transgenic potato lines over-expressing GrΔSPUBCEP12 under the control of the CaMV 35S promoter. Expression of the transgene in independent transgenic lines was confirmed by quantitative RT–PCR assays (Figure 4). No obvious phenotypic differences were observed in shoots and roots of transgenic lines compared with control lines expressing GUS. Three independent transgenic lines showing high levels of transgene expression were tested for G. rostochiensis infection. All three lines exhibited significantly increased numbers of nematode females compared to the control lines (Figure 4).
GrCEP12 suppresses plant immunity
The GrCEP12 peptide derived from GrΔSPUBCEP12 does not have similarity to any known proteins. Evidence is emerging that nematode-secreted effectors have a direct role in suppressing plant immunity (Postma et al., 2012; Jaouannet et al., 2013), and we therefore assessed GrCEP12 for a potential function in this area. The programmed cell death (PCD) induced by recognition of pathogen avirulence (Avr) effectors by corresponding host resistance proteins may be easily observed in an agroinfiltration assay in N. benthamiana leaves, and we thus used this system to investigate whether GrCEP12 suppresses PCD associated with ETI. Co-expression of resistance proteins Gpa2 and Rx2 from potato with their respective elicitors RBP-1 from the potato cyst nematode G. pallida and coat protein (CP) of potato virus X induces PCD in leaves of N. benthamiana (Bendahmane et al.,2000; Sacco et al.,2009). We initially tested whether GrCEP12 could suppress Gpa2/RBP–1-mediated PCD in N. benthamiana. We used the empty vector pMD1 and the construct expressing the nine amino acid HA tag peptide as negative controls. Agrobacterium strains carrying the GrCEP12 construct and negative control constructs were infiltrated into N. benthamiana leaves 24 h prior to infiltration of Agrobacterium cells carrying Gpa2 and RBP–1, and the cell-death phenotype was scored 2–5 days later. The cell-death phenotype started to appear at 2 days when pMD1 or the HA construct was co-expressed with Gpa2/RBP–1 (Figure 5a). However, the cell-death phenotype was suppressed when the GrCEP12 construct was co-expressed with Gpa2/RBP–1 (Figure 5a). Furthermore, we also found that the suppression activity of GrCEP12 on Rx2/CP-mediated cell death was comparable to that of suppressor of necrosis 1 (SNE1), a secreted protein from Phytophthora infestans (Kelley et al., 2010) under our experimental conditions (Figure 5a).
To further evaluate a role for GrCEP12 in suppressing plant immunity, we generated transgenic potato lines over-expressing GrCEP12 and evaluated their susceptibility to infection by G. rostochiensis and Streptomyces scabies, an actinobacterium that causes common scab of potato (Bignell et al., 2010). Two transgenic lines (#3 and 4) confirmed as having high levels of transgene expression (Figure 5b) were selected for infection assays. At 5 weeks after nematode inoculation, more nematode females were recovered from the two transgenic lines compared with the control lines (Figure 5b). Interestingly, a high percentage of the vegetatively multiplied plantlets of each transgenic line showed severe disease symptoms and some even died after S. scabies inoculation, whereas infected control lines and mock-inoculated plantlets of each transgenic line grew into healthy plants (Figure 5c). As the parasitic processes of G. rostochiensis and S. scabies have very little in common, our results suggest that GrCEP12 has a role in suppressing basal plant defense.
The RPN2a gene is dramatically suppressed in GrΔSPUBCEP12 over-expression plants
Many genes encoding UPS components were found to be differentially regulated in syncytia induced by H. glycines on both susceptible and resistant soybean plants (Ithal et al., 2007; Kandoth et al., 2011), suggesting an involvement of the host UPS in syncytium formation as well as defense against nematode infection. We selected 25 UPS-related genes (Table S1) that showed differential regulation in H. glycines-induced syncytia (Ithal et al., 2007; Kandoth et al., 2011), and evaluated their expression in G. rostochiensis-infected potato roots by RT–PCR assays (Table S1). Among these genes, six were found to be significantly up- or down-regulated during the early stages of G. rostochiensis infection (Table S1). As ubiquitin is an essential component of the UPS and is expected to be released from GrΔSPUBCEP12 due to processing, we determined whether these six UPS-related genes also have altered expression in GrΔSPUBCEP12 over-expression lines. Surprisingly, one of the six genes, RPN2a, encoding a subunit of the 26S proteasome, showed a significant reduction of expression (by 86–89%) in all three GrΔSPUBCEP12 over-expression lines compared to the control lines (Figure 6a). In contrast, only one of the two GrCEP12 over-expression lines showed a small degree of suppression of RPN2a compared with the control lines (Figure 6b).
Like UBCEP genes from Heterodera nematode species, GrUBCEP12 from G. rostochiensis also encodes an unusual UBCEP that contains an N–terminal signal peptide for secretion, a mono-ubiquitin domain, and a very short CEP domain showing no similarity to any known peptides. Several lines of evidence indicate that GrUBCEP12 is secreted into plant tissue during nematode parasitism. First, GrUBCEP12 encodes a secretory protein and is expressed exclusively within the nematode's dorsal gland cell. Second, GrUBCEP12 expression was significantly up-regulated in the parasitic J2 stage, which correlates with the time at which syncytia start to be initiated. Third, and most importantly, the anti-ubiquitin antibody detected GrUBCEP12 not only in the dorsal gland cell body where it is synthesized, but also in the dorsal gland extension and ampulla that connect to the nematode esophagus and stylet. The movement of GrUBCEP12 towards the nematode stylet provides direct evidence that GrUBCEP12 is secreted.
The initial translation products of canonical UBCEP genes are rapidly processed to release free ubiquitin and CEPs (Jonnalagadda et al., 1989; Monia et al., 1989; Callis et al., 1990). Unlike the canonical UBCEPs, nematode UBCEPs are secretory proteins containing a SP. Interestingly, when transiently expressed in N. benthamiana leaves, GrUBCEP12 was not processed, in contrast to GrΔSPUBCEP12 (lacking the SP), which was processed in planta. As nematode signal peptides are generally recognized by the plant secretory machinery (Lu et al., 2009; Lozano-Torres et al., 2012), it appears that GrUBCEP12 is not accessible to plant endogenous enzymes for processing during its translocation through the secretory pathway. Therefore, we believe that nematode UBCEPs are protected from processing during their translocation through the nematode secretory pathway, and that it is the ubiquitin carboxyl extension protein that is secreted into host plant cells (the SP is removed once secreted). The detection of a UBCEP in stylet secretions of a root-knot nematode species (Bellafiore et al., 2008) further supports this conclusion. It is significant that the secreted GrΔSPUBCEP12 protein was processed in the same way as canonical UBCEPs in planta, which suggests that, once delivered into host plant cells through the nematode stylet, GrΔSPUBCEP12 is recognized by plant endogenous enzymes and is processed into free ubiquitin and a short GrCEP12 peptide. Due to the cleavage of GrΔSPUBCEP12 in planta, the signals observed in both the cytoplasm and the nucleus of plant cells for the GrΔSPUBCEP12–GFP:GUS fusion protein may represent the site of action of the derived GrCEP12 peptide rather than the GrΔSPUBCEP12 protein.
To further demonstrate a role of GrUBCEP12 in plant parasitism, we utilized plant host-derived RNAi (Sindhu et al., 2009) to specifically suppress GrUBCEP12 expression and evaluated the subsequent effect on nematode parasitism. A significant reduction in the number of nematode females was apparent on independent RNAi lines compared to the control lines. This reduction in nematode parasitic success resulted from specific suppression of GrUBCEP12 expression in the feeding nematodes. Furthermore, transgenic potato lines over-expressing the secreted GrΔSPUBCEP12 protein were more susceptible to nematode infection. Together, these results provide direct evidence that GrUBCEP12 is involved in plant parasitism.
Evidence is mounting that phytonematodes utilize their secreted small peptides to promote parasitism. For example, 16D10, a 13 amino acid peptide that is secreted by the root-knot nematode Meloidogyne incognita plays an important role in nematode parasitism, probably through its interaction with plant SCARECROW-like transcription factors to regulate formation of feeding cells (Huang et al., 2006). Cyst nematodes secrete peptide mimics of plant endogenous CLE peptides to redirect plant developmental pathways necessary for syncytium formation and maintenance (Mitchum et al., 2008). It is striking that the 12 amino acid GrCEP12 peptide suppressed the cell death that is associated with ETI. The increased susceptibility of transgenic potato lines over-expressing GrCEP12 to G. rostochiensis and S. scabies provides additional evidence that the small GrCEP12 peptide has a role in suppressing defense. It has been reported that there is extensive overlap in the downstream signaling pathways between ETI and PTI (Tsuda and Katagiri, 2010). GrCEP12 therefore probably targets a common component in the downstream signaling pathways of ETI and PTI. Diverse small-molecule phytohormones play critical roles in regulating plant immunity (Pieterse et al., 2009; Albrecht et al., 2012). It is worth noting that phytosulfokine (PSK), a 5 amino acid growth-promoting peptide hormone, was recently shown to be required for PTI inhibition mediated by PSK receptors (Igarashi et al., 2012). It is tempting to speculate that GrCEP12 may mimic an as yet identified endogenous plant peptide that is involved in suppressing plant immunity. Our subcellular localization study indicated that GrCEP12 may be a dual nucleo-cytoplasmic-localized effector. Alternatively, GrCEP12 may co-opt host cell nuclear functions to affect defense. Identification of host target(s) of GrCEP12 is necessary to uncover the mechanism that GrCEP plays in suppressing plant immunity.
Studies have indicated that the host UPS probably plays critical roles in syncytium formation as well as resistance against nematode infection (Ithal et al., 2007; Kandoth et al., 2011). Among the six UPS-related genes that were significantly regulated in G. rostochiensis-infected potato roots, the RPN2a gene encoding a subunit of the 26S proteasome was greatly suppressed in GrΔSPUBCEP12 but not GrCEP12 over-expression lines. Hanna et al. (2007) reported that a ubiquitin stress response in yeast (Saccharomyces cerevisiae) induced altered proteasome composition and ubiquitin-dependent regulation of the proteasome-associated de-ubiquitinating enzyme Ubp6 was mediated at the gene expression level. We therefore propose that the suppression of RPN2a detected in the GrΔSPUBCEP12 over-expression lines may be directly related to the derived ubiquitin rather than the GrCEP12 peptide. The free ubiquitin released from nematode UBCEPs was hypothesized not to play a biological role in syncytium formation based on the assumption that ubiquitin is always abundant in plant cells (Gao et al., 2003; Tytgat et al., 2004). Interestingly, however, polyubiquitin genes were found to be up-regulated in developing syncytia during a compatible interaction between H. glycines and soybean (Ithal et al., 2007), which clearly suggests an increased demand for free ubiquitin in the initial stage of syncytium formation. We hypothesize that ubiquitin released from nematode-secreted UBCEPs may further perturb local cellular ubiquitin levels and thus affect the host 26S proteasome, an essential component of the UPS. Phytopathogens have evolved effectors to manipulate the host UPS to promote their survival (Dielen et al., 2010; Shirsekar et al., 2010). For example, bacterial pathogens as well as oomycete and fungal pathogens secrete effectors that mimic or interact with host ubiquitin E3 ligases to promote virulence (Dielen et al., 2010). Viral proteins and bacterial effectors may also inhibit host proteasome activities for their own benefit (Dielen et al., 2010). A recent study provided evidence that some subunits of the 26S proteasome are involved in innate immunity in Arabidopsis (Yao et al., 2012). The up-regulation of RPN2a in initiating syncytia induced by H. glycines on resistant plants (Kandoth et al., 2011) suggested a role for RPN2a in plant defense against nematode infection. It may be possible that nematodes use their secreted ubiquitin to suppress local defense by manipulating the function of the host 26S proteasome to promote syncytium formation. Further studies are needed to elucidate the role of nematode-secreted ubiquitin and a possible involvement of the host 26S proteasome in plant parasitism.
Nematode culture and inoculation
The potato cyst nematode (Globodera rostochiensis) pathotype Ro1 was propagated on potato (Solanum tuberosum cv. Katahdin) and cysts were extracted (Brodie, 1996). Pre-parasitic second-stage juveniles (pre–J2s) were collected by hatching nematode eggs in potato root diffusate containing gentamycin sulfate (1.5 mg ml−1) and nystatin (0.05 mg ml−1), and were used for inoculation on monoxenic potato root cultures grown in Petri dishes or on potato plantlets grown in six-well plates (Greiner Bio-One, http://www.gbo.com/) (Lu et al., 2008).
Plant material for the agroinfiltration assay
Nicotiana benthamiana plants were grown in the greenhouse at 22–28°C for 4-5 weeks. One day before agroinfiltration, plants were transferred into a growth room and maintained at 20–25°C, 35–50% humidity, with a 16 h light/8 h dark cycle.
The primers used for various experiments are listed in Table S2.
GrUBCEP12 gene cloning and sequence analysis
The HgUBI1 protein sequence (AF469060) (Gao et al., 2003) was used to search the Nematode.net database (http://nematode.net), and a cDNA of G. rostochiensis that showed similarity to HgUbi1 was identified. mRNA from pre–J2s was used for first-strand cDNA synthesis (Lu et al., 2008). The corresponding GrUBCEP12 cDNA was obtained by PCR using the first-strand cDNA as template (Lu et al., 2008). An alignment of ubiquitin carboxyl extension protein sequences from cyst nematodes, Saccharomyces cerevisiae, Caenorhabditis elegans and Arabidopsis thaliana was produced using ClustalW2 software (http://www.ebi.ac.uk/Tools/clustalw2/index.html) and BOXSHADE (http://workbench.sdsc.edu).
In situ mRNA hybridization and immunofluorescence labeling
Digoxigenin-labeled antisense and sense cDNA probes targeting the CEP domain and the 3′ untranslated region of GrUBCEP12 (Figure S4b) were synthesized (Wang et al., 2001). Both pre–J2s and parasitic stages of G. rostochiensis were processed for in situ mRNA hybridization (Wang et al., 2001).
Immunofluorescence labeling of pre–J2s using an anti-ubiquitin antibody (Sigma–Aldrich, http://www.sigmaaldrich.com/) was performed (Goellner et al., 2000). The anti-ubiquitin antibody and goat anti-rabbit fluorescein isothiocyanate-conjugated secondary antibody (Sigma–Aldrich) were used at a dilution of 1:100 and 1:300, respectively. The treated nematodes were mounted on glass slides and observed under an epifluorescence microscope (Olympus, http://www.olympus-global.com/).
mRNA isolation and developmental expression profiling
mRNA from various nematode life stages were extracted (Lu et al., 2008) and used for quantifying GrUBCEP12 expression by quantitative RT–PCR, which was performed as described by Lu et al. (2008) in a 25 μl reaction volume containing iQ SYBR Green Supermix (Bio–Rad Laboratories, http://www.bio–rad.com/), 500 nM of both forward and reverse primers, and 1 μl cDNA. mRNA samples for each developmental stage were prepared from two independent experiments and used for cDNA synthesis. All quantitative PCR assays consisted of three technical replicates for each cDNA sampleThe G. rostochiensis β–actin gene (Gr–act–1) (EF437156) was used as an endogenous reference for data analysis using the method (Lu et al., 2009).
In planta processing of GrUBCEP12
To examine GrUBCEP12 processing in planta, gene sequences encoding four forms of GrUBCEP12 were amplified: (i) a full-length form containing the SP (GrUBCEP12), (ii) a secreted form lacking the SP (GrΔSPUBCEP12), (iii) a secreted form mutated at the ubiquitin cleavage site (GrΔSPUB–mut–CEP12) (Hanania et al., 2009), and (iv) a truncated form containing only the CEP domain (GrCEP12) (Figure 2c). The amplified products were cloned into the pBIN61-EGFP:HA binary vector (Sacco et al., 2009) (kindly provided by P. Moffett, Départment de Biologie, Université de Sherbrooke, Canada) to generate EGFP:HA fusion constructs. These constructs, as well as the vector pBIN61-EGFP:HA, were transformed into Agrobacterium tumefaciens strain C58C1. Agrobacterium cells were cultured and then resuspended in a buffer containing 10 mm MgCl2, 0.195% w/v MES (pH 5.6) and 200 μM acetosyringone to a final absorbance at 600 nm of 0.5. The cell suspensions were infiltrated into N. benthamiana leaves using a needleless syringe. The infiltrated leaves were collected 2 days after infiltration and used to prepare protein extracts.
Collected leaves were homogenized in extraction buffer (2% SDS, 80 mm Tris/HCl, pH 6.8, 10% glycerol, 0.002% bromophenol blue and 5% β–mercaptoethanol). Samples of protein extracts were separated by SDS–PAGE and electroblotted onto nitrocellulose membrane (Bio–Rad Laboratories). After washing with TBS (80 mM Tris/HCl, pH 7.5 and 200 mM NaCl) for 5 min and blocking with TBS containing 5% w/v nonfat dry milk for 1 h at room temperature, the membrane was incubated with anti-HA antibody (1:5000 dilution) (Roche Applied Science, http://www.roche-applied-science.com/) and then with horseradish peroxidase-conjugated goat anti-rabbit IgE (1:5000 dilution) (Bio–Rad Laboratories). HA-tagged proteins were visualized using the Immobilon Western Chemiluminescent system (Millipore, http://www.millipore.com/). Extracted protein samples from three independent infiltrations were analyzed and similar results were obtained.
GrΔSPUBCEP12 and GrCEP12 coding sequences were cloned into the modified pRJG23 vector (Grebenok et al., 1997) to generate GrΔSPUBCEP12–GFP:GUS and GrCEP12–GFP:GUS fusion constructs. These constructs were delivered into onion epidermal cells by biolistic bombardment (Hewezi et al., 2008). Epidermal peels were then incubated for 24 h in the dark. The subcellular localization of the fusion proteins was visualized using a Zeiss Axiovert 100 microscope (http://www.zeiss.com/). Three independent experiments were performed and similar results were obtained.
Transgene construction and potato transformation
GrΔSPUBCEP12 and GrCEP12 coding sequences were cloned into the binary vector pMD1 (Tai et al., 1999), and the resulting constructs were transformed into A. tumefaciens strain LBA4404. The GUS protein-encoding gene was also cloned into pMD1 to use as a negative control.
A modified pMD1 vector harboring an RNAi cassette that allows cloning of the sense and the antisense gene fragments interspersed by a 363 bp GUS gene fragment under the control of the superpromoter (Lee et al., 2007) (kindly provided by S.B. Gelvin, Department of Biological Sciences, Purdue University, West Lafayette, IN) was constructed, yielding the pSUPERMD-RNAi vector. The DNA fragment (187 bp) corresponding to the CEP domain and the 3′ untranslated region of GrUBCEP12 (Figure S4b) was amplified and inserted in both sense and antisense orientations into the pSUPERMD-RNAi vector. The resulting RNAi construct was transformed into A. tumefaciens strain LBA4404.
Transgenic potato plants were generated using LBA4404 carrying the above individual constructs or a control construct (pMD1-GUS or pSUPERMD-RNAi). Transformation of potato (S. tuberosum cv. Désirée) was performed as described by Jung et al. (2005) with minor modifications. Briefly, potato internode segments from in vitro-grown Désirée plants were cut and incubated in A. tumefaciens cell suspension for 20 min, then transferred onto callus induction medium (Jung et al., 2005) and incubated for 3 days at 24°C in the dark. The explants were then transferred onto 3C5ZR medium (Jung et al., 2005) containing 50 μg ml−1 kanamycin and 600 μg ml−1 timentin (GlaxoSmithKline, http://www.gsk.com/). After approximately 4 weeks, emerging shoots were dissected and transferred into culture tubes containing propagation medium [1 x MS salts, 1.2 mm NaH2PO4·H2O, 0.01% w/v myo-inositol, 0.4 μg ml−1 thiamine HCl, 3% w/v sucrose, 0.3% w/v Gelrite (RPI, http://www.rpicorp.com/), pH 6.0] containing kanamycin at 50 μg ml−1 and timentin at 119 μg ml−1. Plantlets were cultivated in a growth chamber at 24°C under a 16 h light/8 h dark cycle.
To confirm transgene expression in independent transgenic lines, mRNA was extracted from roots of GrΔSPUBCEP12 and GrCEP12 over-expression lines, and total RNA was extracted from roots of RNAi lines using the NucleoSpin RNA Plant kit (Clontech, http://www.clontech.com/). To check expression of the hairpin dsRNA in RNAi lines, the single-stranded loop region of the 363 bp GUS fragment was targeted. cDNA samples from two independent experiments, consisting of three technical replicates for each sample, were tested by quantitative PCR. The potato polyubiquitin gene was used as an endogenous reference for data analysis (Lu et al., 2009).
GrUBCEP12 expression in G. rostochiensis infecting RNAi lines
To evaluate GrUBCEP12 expression in nematodes infecting RNAi lines and control lines, infected root segments containing nematodes were collected 2 weeks after inoculation and used as parasitic nematode materials (Lu et al., 2008) to examine GrUBCEP12 expression by quantitative RT–PCR. Primers (Table S2) that recognized sequences outside the region complementary to the hairpin dsRNA were used to ensure specific detection of endogenous GrUBCEP12 transcripts in the nematodes. The expression of non-RNAi target genes of GrCM1 (Lu et al., 2008) and GrUBCEP52 (ID# GR11615; http://nematode.net) was also evaluated to determine the specificity of gene silencing. mRNA from roots of two independent extractions was tested by quantitative PCR, with three technical replicates for each sample. The G. rostochiensis β–actin gene (Gr–act–1) was used as an endogenous reference for data analysis.
Cell-death suppression assay in N. benthamiana
Primers targeting GrCEP12 and the HA tag peptide-encoding sequences were synthesized by Sigma-Aldrich. Each primer pair was mixed, denatured and then annealed at room temperature. The annealed DNA fragment was cloned into pMD1 to generate expression constructs pMD1-GrCEP12 and pMD1-HA. These constructs, as well as pMD1, were transformed into A. tumefaciens strain GV3101. Transformed GV3101 was cultured overnight, and the cells were collected, washed and resuspended in 10 mm MES, pH 5.5, 200 μM acetosyringone at a final absorbance at 600 nm of 0.3. Cell suspensions were cultured for 3 h and then infiltrated into N. benthamiana leaves. For co-expression of Rx2/CP (Bendahmane et al., 2000) or Gpa2/RBP–1 (Sacco et al., 2009) (kindly provided by G.B. Martin, Boyce Thompson Institute for Plant Research, Ithaca, NY), equal volumes of bacterial dilutions of the same density (absorbance at 600 nm of 0.2 and 0.3 for Rx2/CP and Gpa2/RBP–1, respectively) were mixed and infiltrated into previously infiltrated areas 1 day after the first infiltration. Symptoms were monitored, and photographs were taken 4 days after the last infiltration. Five and two independent assays for Gpa2/RBP–1 and Rx2/CP, respectively, were performed.
Nematode and Streptomyces infections on transgenic potato lines
Internode segments (0.5–1.0 cm) cut from in vitro-grown potato plantlets were propagated in deep Petri dishes (Fisher Scientific, http://www.fishersci.com/) containing propagation medium. Approximately 2 weeks after propagation, shoot tops were cut and then either cultivated in six-well plates (Greiner Bio-One) or inoculated with Streptomyces scabies. Two weeks after growth in six-well plates, the plantlets (4–8 per line) were inoculated with 150 surface-sterilized pre–J2s and maintained at 20°C for 5 weeks. Nematode female numbers were then counted. Three independent nematode infection assays were performed and similar results were obtained.
For the S. scabies infection assay, S. scabies strain 87–22was cultured at 28°C on ISP4 agar medium (Difco, http://www.bd.com/) for 5–7 days, until the culture sporulated. Spores were collected, and mycelium fragments and agar were removed from the spore suspension by filtering through sterile, non-absorbent cotton wool. Spores were then resuspended in sterile water at a final absorbance at 600 nm of 1.5. The cut ends of the plantlet shoots (6–8 per line) were dipped in the suspension, embedded in propagation medium in culture tubes, and cultivated at 24°C under a 16 h light/8 h dark cycle. Disease symptoms were monitored and photographs were taken at 5–6 weeks post-inoculation. Two independent infection assays were performed and similar results were obtained.
Evaluation of UPS-related gene expression
mRNA from uninfected and infected potato roots (3 days after G. rostochiensis inoculation) were used for RT–PCR to examine the expression of 25 UPS-related genes (Table S1). mRNA from roots of GrΔSPUBCEP12 and GrCEP12 over-expression lines was used for quantitative RT–PCR to examine the expression of the six UPS-related genes that showed significant regulation in nematode-infected potato roots (Table S1). cDNA samples from three (for GrΔSPUBCEP12 lines) or two (for GrCEP12 lines) independent experiments were tested by quantitative PCR, with three technical replicates for each sample. The potato polyubiquitin gene was used as an endogenous reference for data analysis.
We thank Kent Loeffler (Department of Plant Pathology and Plant-Microbe Biology, Cornell University, Ithaca, NY) for photography. This work was supported by funding from US Department of Agriculture, Agricultural Research Service.