SUMMARY
Verticillium wilt, caused by soil-borne fungi of the genus Verticillium, is an economically important disease that affects a wide range of host plants. Unfortunately, host resistance against Verticillium wilts is not available for many plant species, and the disease is notoriously difficult to combat. Host-induced gene silencing (HIGS) is an RNA interference (RNAi) based process in which small RNAs are produced by the host plant to target parasite transcripts. HIGS has emerged as a promising strategy for improving plant resistance against pathogens by silencing genes that are essential for these pathogens. Here, we assessed whether HIGS can be utilized to suppress Verticillium wilt disease by silencing previously identified virulence genes of V. dahliae through the host plants tomato and Arabidopsis. In transient assays, tomato plants were agroinfiltrated with Tobacco rattle virus (TRV) constructs to target V. dahliae transcripts. Subsequent V. dahliae inoculation revealed suppression of Verticillium wilt disease in some, but not all, cases. Next, expression of RNAi constructs targeting V. dahliae transcripts was pursued in stable transgenic Arabidopsis thaliana plants. Also in this host, V. dahliae inoculation revealed reduced Verticillium wilt disease in some cases. Thus, our study suggests that, depending on the target gene chosen, HIGS against V. dahliae is operational in tomato and A. thaliana plants and may act as a plant protection approach that may be used in Verticillium wilt-susceptible crops.
INTRODUCTION
Verticillium wilts are vascular wilt diseases that are caused by soil-borne fungi of the genus Verticillium (Fradin and Thomma, 2006; Klimes et al., 2015). This genus comprises ten species of soil-borne fungi that differ in their morphological features, such as resting structures, as well as in their ability to cause plant diseases (Inderbitzin et al., 2011). Within the Verticillium genus, V. dahliae is the most notorious pathogenic species that can infect hundreds of dicotyledonous hosts, including ecologically important plants and many high-value crops worldwide (Fradin and Thomma, 2006; Klosterman et al., 2009; Inderbitzin et al., 2011). Verticillium wilt diseases are difficult to control due to the long viability of the resting structures, the wide host range of the pathogens, and the inability of fungicides to affect the pathogen once in the plant vascular system. Thus, the most sustainable way to control Verticillium wilt diseases is the use of resistant cultivars. Polygenic resistance to Verticillium spp. has been described for several plant species, including potato, hop, alfalfa, cotton and strawberry (Simko et al., 2004; Bolek et al., 2005; Wang et al., 2008; Yang et al., 2008; Jakse et al., 2013; Antanaviciute et al., 2015), whereas single dominant resistance genes have been identified only in tomato, sunflower, cotton, potato and lettuce (Schaible et al., 1951; Putt, 1964; Barrow, 1970; Lynch et al., 1997; Mert et al., 2005; Hayes et al., 2011; Christopoulou et al., 2015). In tomato (Solanum lycopersicum), a single dominant locus that confers Verticillium resistance has been identified as the Ve locus, which controls Verticillium isolates that are assigned to race 1, whereas race 2 strains escape recognition (Schaible et al., 1951; Pegg, 1974). The Ve locus contains two closely linked and inversely oriented genes, Ve1 and Ve2, both of which encode extracellular leucine rich repeat (eLRR) receptor-like proteins (RLPs) (Kawchuk et al., 2001; Wang et al., 2010). Of these, only Ve1 was found to confer resistance against race 1 isolates of Verticillium in tomato (Fradin et al., 2009). Interestingly, interfamily transfer of Ve1 from tomato to Arabidopsis thaliana has resulted in race-specific Verticillium resistance in the latter species (Fradin et al., 2011; 2014; Zhang et al., 2014), implying that the underlying immune signaling pathway is conserved (Fradin et al., 2011; Thomma et al., 2011). Tomato Ve1 serves as an immune receptor for recognition of the effector protein Ave1 that is secreted by race 1 strains of V. dahliae (de Jonge et al., 2012). More recently, homologs of tomato Ve1 acting as immune receptors that govern resistance against V. dahliae race 1 strains through recognition of the Ave1 effector have been characterized in other plant species including tobacco, potato, wild eggplant and hop, suggesting an ancient origin of the immune receptor Ve1 (Song et al., 2016).
Although the tomato Ve1 gene is still currently deployed in tomato cultivars, isolates of Verticillium that escape Ve1-mediated recognition appeared within a few years after the introduction of the tomato Ve1 (Pegg and Brady, 2002). These race 2 isolates of Verticillium steadily supplanted race 1 strains in various regions because of the extensive use of Verticillium race 1-resistant cultivars (Dobinson et al., 1996). So far, there remains no source of commercially employed resistance to Verticillium race 2 strains.
RNA interference (RNAi) is a conserved regulatory mechanism that affects gene expression in eukaryotic organisms (Baulcombe, 2005). RNA silencing is triggered by the processing of double stranded RNA (dsRNA) precursors into short interfering RNA (siRNAs) duplexes of 21-28 nucleotides in length, and followed by the guided cleavage or translational repression of sequence-complementary single-stranded RNAs by the generated siRNAs duplexes, which are incorporated into a silencing complex called RISC (RNA-induced silencing complex) (Ruiz-Ferrer and Voinnet, 2009). Plants and other eukaryotes have evolved RNAi machineries that not only regulate developmental programs, but also provide protection from invaders, such as viruses. In plants, RNAi has been exploited extensively and has become a powerful functional genomics tool to silence the expression of genes of interest as well as to engineer viral resistance (Duan et al., 2012). Interestingly, organisms that live within, or develop intimate contact with, a host, such as bacteria (Escobar et al., 2001; Escobar et al., 2002), nematodes (Huang et al., 2006), insects (Baum et al., 2007; Mao et al., 2007) and parasitic plants (Tomilov et al., 2008), are sensitive to small RNAs generated by the host and that are targeted to parasite transcripts. This so-called host-induced gene silencing (HIGS) has also emerged as a promising strategy against plant pathogens, including fungi and oomycetes. Initial reports of HIGS against filamentous pathogens were described for the maize kernel and ear rot pathogen Fusarium verticillioides (Tinoco et al., 2010) and the barley powdery mildew fungus Blumeria graminis (Nowara et al., 2010). Subsequent reports demonstrated the functionality of HIGS in suppressing diseases caused by the fungal pathogens Puccinia spp. (Yin et al., 2011; 2015; Zhang et al., 2012; Panwar et al., 2013), Fusarium spp. (Koch et al., 2013; Ghag et al., 2014; Cheng et al., 2015; Hu et al., 2015; Chen et al., 2016), Sclerotinia sclerotiorum (Andrade et al., 2015) and Rhizoctonia solani (Zhou et al., 2016), as well as by the oomycete pathogens Phytophthora infestans (Jahan et al., 2015; Sanju et al., 2015) and Bremia lactucae (Govindarajulu et al., 2015). Many of these pathogens make very intimate contact with host cells, potentially facilitating the occurrence of HIGS. In this study, we assessed whether HIGS can be used to suppress Verticillium wilt disease in tomato and A. thaliana by targeting previously identified virulence factors of V. dahliae, a pathogen that is known not to invade host cells at any stage of disease development.
RESULTS
Tobacco rattle virus-based silencing in tomato compromises V. dahliae Ave1 expression
Tobacco rattle virus (TRV)-based virus-induced gene silencing (VIGS) has extensively been used in various plant species, including tomato (Liu et al., 2002; Senthil-Kumar et al., 2007), and TRV-based VIGS has successfully been used to investigate candidate genes for their involvement in Verticillium wilt resistance in tomato (Fradin et al., 2009). In order to investigate whether HIGS can be established against the xylem-colonizing fungus V. dahliae that is not known to penetrate adjacent living cells, we attempted to exploit TRV-based VIGS to produce dsRNAs that are targeted towards V. dahliae Ave1 transcripts. The experiment was performed in Ve1 tomato plants that are normally immune to infection by Ave1-carrying V. dahliae strains, such that successful HIGS would immediately result in vascular wilt disease that does not occur if Ave1 expression is not compromised (Fradin et al., 2009; de Jonge et al., 2012). To this end, a 1:1 mixture of Agrobacterium tumefaciens cultures carrying TRV1 and TRV2::Ave1 (Figure 1A) was infiltrated into cotyledons of Ve1 tomato plants. A recombinant construct containing a fragment of the GUS gene (TRV2::GUS) was used as a negative control (Figure 1A). At ten days after TRV treatment, plants were challenged with either the V. dahliae race 1 strain JR2 (Faino et al., 2015), or an Ave1 deletion mutant (V. dahliae JR2△Ave1; de Jonge et al., 2012), and inspected for Verticillium wilt symptoms (stunting and wilting) up to 14 days post inoculation (dpi). As expected, no significant disease symptoms were observed on TRV::GUS-treated plants that were inoculated with the wild-type race 1 V. dahliae strain (Figure 2A), indicating that TRV treatment by itself does not compromise Ave1-triggered immunity in Ve1 tomato plants. Furthermore, the Ave1 deletion mutant caused clear Verticillium wilt disease, as Verticillium wilt disease developed on Ve1 plants treated with TRV2::Ave1 and subsequent inoculation with the Ave1 deletion strain (Figure 2A). However, intriguingly, Verticillium wilt disease also developed on Ve1 plants upon TRV2::Ave1 treatment and subsequent inoculation with the wild-type race 1 V. dahliae strain (Figure 2A). This finding suggests that Ave1 expression in V. dahliae is indeed compromised due to TRV-induced HIGS in tomato. The compromised immunity was confirmed by fungal recovery assays by plating stem sections on potato dextrose agar (PDA) plates, and by fungal biomass quantification in stem sections of the inoculated plants (Figure 2).
It was recently demonstrated that Tobacco mosaic virus (TMV) may infect fungi in addition to plants, remaining for up to six subcultures, and also persisted in plants infected by the virus-infected fungus (Mascia et al., 2014). This finding raises the theoretical possibility that also TRV may infect V. dahliae and cause VIGS (directly) rather than HIGS from inside the tomato cells (indirectly). To exclude that the impairment of Ave1-triggered immunity in Ve1 tomato plants is due to TRV-infection of V. dahliae itself, stem sections from Verticillium-inoculated TRV::GUS- and TRV::Ave1-treated tomato plants were placed on PDA plates. A single colony that grew from wild-type V. dahliae-inoculated TRV::GUS-treated plants (V. dahliae JR2TRV::GUS) and Ave1 deletion mutant-inoculated TRV::Ave1-treated plants (V. dahliae JR2△Ave1TRV::Ave1), and three independent colonies that grew from wild-type V. dahliae-inoculated TRV::Ave1-treated plants (V. dahliae JR2TRV::Ave1) were subjected to PCR to detect a viral coat protein gene fragment of TRV (TRV2_CP), showing that TRV2_CP was not detected in all fungal isolates (Figure S1A). Furthermore, the fungal isolates were used to infect Ve1 tomato plants and tomato plants that lack Ve1. This analysis showed that, similar to V. dahliae JR2TRV::GUS, also V. dahliae JR2TRV::Ave1 induced no disease symptoms on tomato plants expressing Ve1, while tomato plants lacking Ve1 showed clear Verticillium wilt disease (Figure S1B-D). These data support the hypothesis that the impairment of Ave1-triggered immunity in Ve1 plants is not caused by TRV-infection of V. dahliae, but genuinely by HIGS through TRV-treatment of tomato.
TRV-based fungal gene silencing in tomato inhibits Verticillium wilt disease
To further investigate the potential of TRV-mediated HIGS against V. dahliae in tomato, two previously identified virulence genes of V. dahliae were targeted. The first target gene is NLP1, encoding a member of the necrosis- and ethylene-inducing-like protein (NLP) family in V. dahliae, and targeted deletion of NLP1 in V. dahliae significantly compromises virulence on tomato as well as A. thaliana plants (Santhanam et al., 2013). The second candidate gene is V. dahliae Sge1, encoding a homolog of the transcription factor Sge1 (SIX Gene Expression 1) in F. oxsporum, and V. dahliae mutants of the Sge1 are non-pathogenic on tomato (Santhanam and Thomma, 2013). To produce dsRNAs of the gene fragments in planta, cotyledons of ten-day-old Moneymaker tomato plants were treated with the silencing constructs TRV::GUS, TRV2::NLP1 and TRV2::Sge1 in combination with TRV1 (Figure 1A), respectively. At ten days after TRV treatment, plants were challenged with either the V. dahliae strain JR2 (Faino et al., 2015), a NLP1 deletion mutant (V. dahliae JR2△NLP1; Santhanam et al., 2013), or a Sge1 deletion mutant (V. dahliae JR2△Sge1; Santhanam and Thomma, 2013), and monitored for Verticillium wilt symptoms on tomato plants at 14 dpi. As expected, significantly compromised Verticillium wilt symptoms were observed on Moneymaker tomato plants upon TRV2::GUS or TRV2::NLP1 treatment and subsequent inoculation with the NLP1 deletion mutant of V. dahliae strain JR2 (Figure 3A). Interestingly, upon inoculation with the wild-type V. dahliae strain JR2, a moderate reduction of Verticillium wilt symptoms was observed on Moneymaker tomato plants treated with TRV2::NLP1 when compared to TRV::GUS-treated plants (Figure 3A). The plants that were treated with TRV::NLP1 and subsequent inoculation with the wild-type V. dahliae strain JR2 showed reduced Verticillium wilt symptoms but were not as diseased as plants upon inoculation with the NLP1 deletion mutant or water (Figure 3A). These data are further supported by fungal biomass quantifications in stem sections of the inoculated plants (Figure 3B). In contrast, no significant Verticillium wilt disease reduction was observed in Moneymaker tomato plants upon the TRV::Sge1 treatment and subsequent inoculation with the wild-type V. dahliae strain JR2, although fungal biomass quantifications revealed that less fungal biomass accumulated in planta in the TRV::Sge1-treated plants than the TRV::GUS-treated plants followed by inoculation with the wild-type V. dahliae strain JR2 (Figure 4A; B). To determine whether TRV-mediated targeting transcripts of V. dahliae Sge1 in tomato results in complete silencing of the V. dahliae Sge1 gene, we performed real-time PCR to measure relative expression level for the Sge1 gene in V. dahliae JR2 inoculating with the TRV::Sge1-treated plants compared to the TRV::Sge1-treated plants. However, only a slight reduction in Sge1 expression in TRV::Sge1-targeted V. dahliae was monitored when compared with TRV::GUS-targeted V. dahliae (Figure 4C). In conclusion, although not all TRV-based RNAi constructs targeting V. dahliae transcripts in tomato suppressed Verticillium wilt disease, TRV-mediated transient HIGS against V. dahliae in tomato can be achieved.
HIGS in Ve1-transgenic A. thaliana does not impair Ave1-triggerred immunity
To assess whether HIGS against V. dahliae can be made operational in stable transgenic plants by expressing dsRNAs, we exploited hairpin RNA-based RNAi to produce dsRNAs to target V. dahliae Ave1 transcripts in Ve1-expressing A. thaliana plants. To this end, a fragment of the V. dahliae Ave1 gene was cloned into the Gateway vector pFAST R03 (Shimada et al., 2010) to obtain the RNAi construct pFAST R03_Ave1 (Figure 1B) that leads to hairpin RNA formation after transcription. A recombinant RNAi construct containing a fragment of the green fluorescent protein (GFP) gene (pFAST R03_GFP) was used as a negative control (Figure 1B). Subsequently, the Ave1 and GFP RNAi constructs were transformed into recipient Ve1-expressing A. thaliana plants (Fradin et al., 2011) (Figure S2A). No obvious developmental alterations were observed in the transgenic plants when compared with the recipient Col-0 and Ve1 plants (Figure 5A), and three independent Ave1 RNAi lines expressing Ve1 (pFAST R03_Ave1 in Ve1-1, pFAST R03_Ave1 in Ve1-2 and pFAST R03_Ave1 in Ve1-3) as well as transgenic and non-transgenic control lines were inoculated with either the V. dahliae race 1 strain JR2 or an Ave1 deletion mutant, and monitored for Verticillium wilt symptoms up to 21 dpi. As expected, upon mock-inoculation or inoculation with the V. dahliae JR2, no disease symptoms were observed in Ve1 plants and GFP-RNAi Ve1 plants (Figure 5A). In contrast, GFP- or Ave1-RNAi Col-0 plants lacking Ve1 were as diseased as non-transformed control lines (Figure 5A). However, despite transcripts for hairpin Ave1 formation in Ave1-RNAi Ve1 plants were detected (Figure S2A), Verticillium wilt symptoms were not observed in Ave1-RNAi Ve1 plants upon inoculation with the wild-type race 1 V. dahliae strain JR2 in repeated assays, while the Ave1 deletion strain caused clear Verticillium wilt symptoms on Ve1 plants (Figure 5A). The phenotypes correlated with the degree of V. dahliae colonization as determined with real-time PCR (Figure 5). These data show that expression of an RNAi construct targeting Ave1 transcripts in A. thaliana plants expressing Ve1 does not compromise Ave1-triggered immunity.
HIGS in A. thaliana can reduce Verticillium wilt
To further investigate whether HIGS against V. dahliae can be established in stable transgenic A. thaliana plants by expressing dsRNAs, V. dahliae NLP1 and Sge1 were targeted. To this end, RNAi constructs pHellsgate 12_NLP1 and pHellsgate 12_Sge1 were generated (Figure 1B). The recombinant RNAi construct carrying a fragment of the GFP gene (pHellsgate 12_GFP) was used as a negative control (Figure 1B). Subsequently, RNAi constructs targeting NLP1, Sge1, or GFP were transformed into A. thaliana ecotype Col-0, and independent NLP1-, Sge1, or GFP-RNAi lines were selected (Figure S2B; C). No phenotypic alterations were observed in NLP1- or Sge1-RNAi plants when compared with the recipient A. thaliana Col-0 plants or GFP-RNAi plants (Figure 6A; 7A). Three independent NLP1-RNAi lines (pHellsgate 12_NLP1-1, pHellsgate 12_NLP1-2 and pHellsgate 12_NLP1-3) as well as GFP-RNAi and non-transgenic control lines were assayed for the development of Verticillium wilt symptoms. As expected, markedly compromised Verticillium wilt symptoms were observed on A. thaliana plants upon inoculation with the NLP1 deletion mutant (Figure 6A). Interestingly, upon inoculation with the V. dahliae JR2, a significant reduction of Verticillium wilt symptoms was observed in NLP1-RNAi plants when compared with GFP-RNAi and non-transgenic controls (Figure 6A). These data are further supported by fungal biomass quantifications in stem sections of the inoculated plants (Figure 6B). Additionally, three independent Sge1-RNAi lines (pHellsgate 12_Sge1-1, pHellsgate 12_Sge1-2 and pHellsgate 12_Sge1-3) as well as GFP-RNAi and non-transformed control lines were assayed for Verticillium wilt disease development. Intriguingly, we observed a marked reduction of Verticillium wilt symptoms in Sge1-RNAi A. thaliana lines inoculated with V. dahliae JR2 (Figure 7A). In contrast, GFP-RNAi A. thaliana lines were as susceptible as non-transgenic control lines (Figure 7A). The phenotypes correlated with the level of V. dahliae biomass as determined with real-time PCR (Figure 7). Collectively, these results suggest that, although not all RNAi constructs targeting V. dahliae transcripts in A. thaliana induced HIGS against V. dahliae, hairpin RNA-mediated HIGS in A. thaliana can reduce Verticillium wilt disease.
DISCUSSION
In this manuscript, we show that HIGS against V. dahliae can be achieved through TRV-based fungal gene silencing in tomato, and through hairpin RNA-mediated fungal gene silencing in stable transgenic A. thaliana lines. We established the TRV-mediated HIGS assay through targeting V. dahliae Ave1 transcripts in Ve1 tomato plants, and further used this approach to assess whether HIGS against V. dahliae in tomato can be established through TRV constructs targeting previously identified V. dahliae virulence factors. We also investigated whether HIGS against V. dahliae can be established in transgenic A. thaliana plants through hairpin RNA-based RNAi constructs targeting transcripts of the same previously identified V. dahliae virulence genes. Our results clearly show that plants transiently (in tomato) or stably (in A. thaliana) expressing RNAi constructs targeting transcripts of genes that are essential for V. dahliae pathogenicity can become protected from Verticillium wilt disease. Our results are in line with, and extend beyond, recent reports on protection of cotton plants stably expressing an RNAi construct against V. dahliae (Zhang et al., 2016), and on bidirectional cross-kingdom RNAi and fungal uptake of external RNAs to confer plant protection (Wang et al., 2016).
Reports on HIGS against fungal pathogen infections have accumulated over recent years (Yin et al., 2011; 2015; Zhang et al., 2012; Koch et al., 2013; Panwar et al., 2013; Ghag et al., 2014; Andrade et al., 2015; Cheng et al., 2015; Hu et al., 2015; Chen et al., 2016; Zhou et al., 2016). Among these reports, disease suppression was observed upon silencing of various types of genes, including those that encode the biosynthesis of structural components such chitin and ergosterol, but also genes involved in developmental regulation, secondary metabolism and pathogenicity. Therefore, the selection of suitable target genes is arguably the most important prerequisite for developing a successful HIGS against fungal pathogens. We selected HIGS target genes based on our previous studies of gene deletion mutants in V. dahliae with significantly compromised virulence (NLP1, Santhanam et al., 2013; and Sge1, Santhanam and Thomma, 2013). On beforehand, it was not clear whether silencing of such genes rather than gene deletion would lead to visible virulence phenotypes, as the protein encoded by the target gene may not be completely absent. Indeed, TRV-mediated HIGS of Sge1 in tomato did not lead to compromised Verticillium wilt symptoms, which may be explained by the incomplete silencing of the Sge1 gene in V. dahliae (Figure 4C). This also explains why RNAi-mediated HIGS of V. dahliae Ave1 did not compromise Ve1-mediated immunity in transgenic A. thaliana plants. Also in the cases where the expected visual phenotypes were obtained, fungal biomass quantifications revealed that more fungal biomass accumulated in the inoculated plants upon HIGS of the fungal target gene than upon inoculation with the corresponding V. dahliae deletion mutant (Figure 3B; Figure 4B; Figure 6B; Figure 7B). Thus, arguably, RNAi may not be appropriate to target genes of which the activity is required early in the infection process when RNAi may not have taken its full effect, or genes of which a low-dose of transcripts is biologically active.
Mobility of small RNAs within organisms is a well-known phenomenon, facilitating gene silencing in adjacent cells and surrounding or even distant tissues (Weiberg et al., 2015). Over recent years, several examples of exchange of small RNAs between host plants and invading pathogens have been described, although the mechanistic details of the actual exchange remains to be elucidated (Knip et al., 2014). Nevertheless, small RNA-based bidirectional cross-kingdom gene silencing has been proposed as a common mechanism for cross-kingdom gene regulation in plant-pathogen interactions (Weiberg et al., 2015; Chaloner et al., 2016; Wang et al., 2016). For example, endogenous small RNAs from the fungus Botrytis cinerea have been proposed to transfer into host plants to target defense-related plant transcripts to promote disease development (Weiberg et al., 2013). In this manner, HIGS taps into a process that naturally occurs between plants and pathogens. A search for pathogen-derived small RNAs matching transcripts of host plants or plant-derived small RNAs targeting transcripts of the invading pathogens may facilitate the development of HIGS strategies to engineer resistance in plants against pathogens for which no natural resistance sources have been identified.
EXPERIMENTAL PROCEDURES
Plant growth conditions and manipulations
Plants were grown at 21°C/19°C during 16 h/8 h light/dark photoperiods, respectively, in the climate chamber or the greenhouse with a relative humidity of ~75%, and 100 W⋅m-2 supplemental light when light intensity dropped below 150 W⋅m-2. A. thaliana transformations were performed as described (Clough and Bent, 1998).
Generation of the constructs
The Gateway-compatible Tobacco rattle virus (TRV) two-component Agrobacterium-mediated expression system was used for gene silencing in tomato as previously described (Liu et al., 2002), while the Gateway-compatible vectors pFAST R03 (Shimada et al., 2010) and pHellsgate 12 (Helliwell and Waterhouse, 2003) for hairpin RNA-mediated gene silencing were used to generate stable Arabidopsis transformants. The three V. dahliae genes Ave1 (de Jonge et al., 2012), NLP1 (Santhanam et al., 2013) and Sge1 (Santhanam and Thomma, 2013) were selected for RNAi-based HIGS. Gene annotations for V. dahliae Ave1, Sge1 and NLP1 were obtained from Ensembl Genomes database (http://fungi.ensembl.org/Verticillium_dahliaejr2/Info/Index). Selected DNA fragments were amplified by PCR from the corresponding plasmids using gene-specific primers listed in Table S1. The DNA fragments were cloned into pDONR207 by using the Gateway® BP Clonase® II Enzyme Mix (Invitrogen, California, USA) to generate entry vectors, and all the entry vectors were verified by sequencing. Subsequently, the entry vector pDONR207 carrying the Ave1 fragment was transferred to TRV2 and pFAST R03 to generate constructs TRV2::Ave1 and pFAST R03_Ave1 (Figure 1), while pDONR207 entry vectors carrying NLP1 or Sge1 fragment were recombined into pTRV2 and pHellsgate 12 to generate constructs TRV2::NLP1, TRV2::Sge1, pHellsgate 12_NLP1 and pHellsgate 12_Sge1 (Figure 1) by using Gateway® LR Clonase® II Enzyme Mix (Invitrogen, California, USA). All constructs were transformed to Agrobacterium tumefaciens strain GV3101 (pMP90) by electroporation.
TRV treatment
TRV vectors were agroinfiltrated as previously described (Liu et al., 2002; Fradin et al., 2011). Briefly, cotyledons of 10-day-old tomato (Solanum lycopersicum cv. Moneymaker (ve1) or 35S::Ve1 tomato (Ve1); Fradin et al., 2009) were infiltrated as 1:1 mixtures of pTRV1 and pTRV2 constructs. 10-15 days after TRV inoculation, plants were inoculated with race 1 V. dahliae strain JR2 (Faino et al., 2015); the corresponding mutants: V. dahliae JR2 ΔAve1 (de Jonge et al., 2012); V. dahliae JR2 ΔNLP1 (Santhanam et al., 2013); V. dahliae JR2 ΔSge1 (Santhanam and Thomma, 2013); or tap water as control. The inoculated plants were evaluated by observing disease symptoms up to 14 days post inoculation (dpi).
Verticillium wilt disease and fungal recovery assays
V. dahliae was grown on potato dextrose agar (PDA) at 22 °C, and conidia were collected from 7- to 10-day-old plates and washed with tap water. Disease assays on tomato plants were performed as previously described (Fradin et al., 2009). Briefly, twenty-day-old Ve1 tomato plants (for Verticillium inoculation after TRV inoculation) or ten-day-old Ve1 and ve1 tomato plants (for inoculation with Verticillium colonies re-isolated from infected tomato plants) were uprooted, the roots were rinsed in water, dipped for 5 min in a suspension of 106 conidiospores/mL water, and transplanted to soil. Verticillium outgrowth assays of Ve1 tomato plants, canopy area measurement and fungal biomass quantification in tomato plants were performed as previously described (Fradin et al., 2009; Santhanam et al., 2013). Verticillium disease assay on A. thaliana, as well as fungal biomass quantification in infected A. thaliana plants were performed as previously described (Ellendorff et al., 2009; Song et al., 2016). The fungus-specific primer ITS1-F, based on the internal transcribed spacer (ITS) region of the ribosomal DNA, in combination with the V. dahliae-specific reverse primer ST-Ve1-R (Ellendorff et al., 2009) were used to measure fungal colonization. Primers for tomato actin and A. thaliana RuBisCo (Table S1) were used as endogenous plant control.
Quantitative Real Time-PCR (qRT-PCR) and Reverse transcription-PCR (RT-PCR)
To determine expression of V. dahliae Sge1 gene for silencing, stems of TRV-treated tomato plants were harvested at 14 days post Verticillium inoculation as described above, and flash frozen in liquid nitrogen, and stored at -80 °C for total RNA isolation.
To check Ave1, NLP1, Sge1, GFP DNA fragment presence in the transcripts of the corresponding transgenic A. thaliana lines, two-week-old transgenic and non-transgenic A. thaliana lines were harvested and ground into a powder in liquid nitrogen. Total RNA extraction, cDNA synthesis and RT-PCR were performed as described earlier (Song et al., 2016). Primers for hairpin expression analysis are listed in Table S1. To analyze expression of Sge1 gene for silencing, qRT-PCR was conducted by using primers Sge1-F(qRT) and Sge1-R(qRT) with V. dahliae GAPDH as an endogenous control (Table S1) as previously described (Santhanam and Thomma, 2013).
ACKNOWLEDGEMENTS
Y.S. acknowledges a PhD fellowship from the China Scholarship Council (CSC). B.P.H.J.T. is supported by a Vici grant of the Research Council for Earth and Life sciences (ALW) of the Netherlands Organization for Scientific Research (NWO). Constantinos Patinios and Yidong Wang are acknowledged for technical assistance and Bert Essenstam for excellent plant care at Unifarm. The authors declare no conflict of interest.