Abstract
Plant parasitic nematodes (PPN) locate host plants by following concentration gradients of root exudate chemicals in the soil. We present a simple method for RNAi-induced knockdown of genes in tomato seedling roots, facilitating the study of root exudate composition, and PPN responses. Knockdown of sugar transporter genes, stp1 and stp2 in tomato seedlings triggers corresponding reductions of glucose and fructose, but not xylose, in collected root exudate. This corresponds directly with reduced infectivity and stylet thrusting of the promiscuous PPN Meloidogyne incognita, however we observe no impact on the infectivity or stylet thrusting of the selective Solanaceae PPN Globodera pallida. This approach can underpin future efforts to understand the early stages of plant-pathogen interactions in tomato, and potentially other crop plants.
RNA interference (RNAi) is widely used for the analysis of plant gene function, primarily through the transgenic production of dsRNA constructs in planta, and secondarily through Virus-Induced Gene Silencing (VIGS) (Watson et al., 2005). Publications from the Wolniak lab have shown that exogenous dsRNA can silence genes of the water fern Marsilea vestita (Klink and Wolniak, 2001), and crude lysate from Escherichia coli expressing virus-specific dsRNA have also been used to protect plants from viral pathology (Tenllado et al., 2003). Here we present a similar approach to triggering RNAi in tomato seedlings, which we term exogenous (exo)RNAi. In this approach, aqueous dsRNA is delivered exogenously to tomato seedlings.
Plant root exudate comprises a complex mixture of compounds including volatile and soluble chemicals which may derive from intact or damaged root cells, or sloughed-off root border cells (Dakora and Phillips, 2002). It has been estimated that 11% of photosynthetically-assimilated carbon is released as root exudate (Jones et al., 2009). The monosaccharides glucose, fructose and xylose represent the major sugar component of tomato root exudates (Kamilova et al., 2006). Plant parasitic nematodes (PPNs) are responsible for an estimated 12.3% loss in crop production globally each year (Sasser and Freckman, 1987), and are attracted to host plants by components of plant root exudate. Here we assess the chemosensory response of the root knot nematode, Meloidogyne incognita (a promiscuous pathogen of flowering plants), and the potato cyst nematode, Globodera pallida (a selective pathogen of Solanaceae plants) to each of the three major monosaccharide sugars of tomato plant root exudate, and the efficacy of exoRNAi against stp1 and stp2, known transporters of monosaccharide sugars in tomato (Gear et al., 2000).
Meloidogyne incognita infective stage juveniles were attracted to glucose (CI: 0.33 ±0.07; P<0.001) and fructose (CI: 0.39 ±0.09; P<0.001), but not xylose (CI: 0.04 ±0.09; P>0.05) as compared to control treated worms (Fig 1A). Glucose (125.1% ±5.5; P<0.001) and fructose (124.8% ±5.4; P<0.001) also triggered an elevated level of serotonin-triggered stylet thrusting in treated juveniles; xylose failed to trigger any significant response (99.36% ±10.87; P>0.05) when compared to control treatments (Fig 1B). Globodera pallida infective stage juveniles were mildly repelled by glucose (CI: -0.23 ±0.09; P>0.05), and did not respond to fructose (CI: 0.15 ±0.08; P>0.05), or xylose (CI: -0.19 ±0.09; P>0.05) as compared to control treated worms (Fig 1C). Glucose (118.6% ±9.7; P>0.05), fructose (107.2% ±7.3; P>0.05), or xylose (119.6% ±8.6; P>0.05) had no significant impact on the frequency of serotonin-triggered stylet thrusting in G. pallida infective juveniles when compared to control treatments (Fig 1D).
Treatment of tomato seedlings with stp1 dsRNA triggered a significant reduction in stp1 transcript abundance (0.17 ±0.05; P<0.001), yet had no impact on stp2 abundance (1.037 ±0.13; P>0.05) relative to neomycin transferase (neo) dsRNA treatment. Likewise, stp2 dsRNA induced significant reductions in stp2 transcript abundance (0.21 ±0.06; P<0.001), but not stp1 (0.94 ±0.05; P>0.05) relative to neo dsRNA treatments (Fig 2A). Corresponding reductions in glucose and fructose exudate concentration were observed for both stp1 (5.10 μg/ml ±1.31; P<0.01 and 3.14 μg/ml ±0.92; P<0.01, respectively) and stp2 (4.90 μg/ml ±1.45; P<0.01 and 10.90 μg/ml ±1.07; P<0.05, respectively) dsRNA treated seedlings. No significant changes in xylose exudate concentration were observed across treatment groups (Fig 2B-D).
Root exudates collected from tomato seedlings which had been treated with either stp1 or stp2 dsRNA were less capable of stimulating stylet thrusting in M. incognita relative to exudates collected from control dsRNA treated seedlings (13.92 ±5.10%, P<0.001; and 17.53 ±8.12%, P<0.001, respectively. Fig 3A). No significant difference in stylet thrusting frequency was observed for G. pallida juveniles when exposed to root exudates from stp1 or stp2 dsRNA-treated seedlings, relative to control treated seedlings (108.2 ±38.87%, P>0.05; and 77.34 ±30.84%, P>0.05, respectively) (Fig 3B).
When exoRNAi-treated seedlings were challenged by M. incognita infection assay, significant reductions in percentage infection levels relative to control (neo) dsRNA treatment were observed for both stp1 (14.15% ±4.77; P<0.01) and stp2 (27.08% ±7.32; P<0.05) dsRNA treatments (Fig 3C). Knockdown of stp1 (14.15% ±4.77; P>0.05) or stp2 (14.15% ±4.77; P>0.05) did not significantly reduce the percentage infection levels of G. pallida relative to neo dsRNA treatment (Fig 3D).
These data demonstrate that the exogenous application of aqueous dsRNA onto tomato seedlings is sufficient to trigger specific gene knockdown. However, we found that different experimental populations of tomato seedlings could display wide variation in the expression of both sugar transporter genes, and reference genes which resulted in high standard error of mean (SEM) values. This made it difficult to resolve gene knockdown levels for an isolated number of experiments. This may be due to variation in the susceptibility of tomato seedlings to exoRNAi, as has been observed for Tobacco Rattle Virus (TRV) VIGS approaches in tomato (Liu et al., 2002), or it could indicate that larger replicates of seedlings are required to consistently resolve gene expression data post exoRNAi. It should also be noted that attempts to silence phytoene desaturase in order to observe a bleaching phenotype in the cotyledons were unsuccessful (data not shown). This may indicate that only genes expressed in the tomato root are susceptible to this approach.
It is well established that plant root exudates mediate both positive and negative interactions with commensal and pathogenic microbes (Badri et al., 2009), insects (Walker et al., 2003), and other plants (Bais et al., 2006). Plant parasitic nematodes also respond to plant root exudates (Teillet et al., 2013). The present study aimed to probe the involvement of monosaccharide sugars of tomato root exudate for involvement in the attraction and activation of parasitic behaviours in the promiscuous root knot nematode M. incognita, and the host-selective potato cyst nematode G. pallida. STP1 and STP2 are known transporters of monosaccharide sugars (Gear et al., 2000), and our data demonstrate that both play a role in regulating the level of glucose and fructose (but not xylose) exudation from tomato seedling roots. exoRNAi knockdown of each transporter significantly reduced the amount of glucose and fructose secreted from plant roots, which corresponded with a decrease in M. incognita infectivity, but not G. pallida infectivity. These results suggest that glucose and fructose are important chemical cues which infective stage M. incognita use to find host plants. These data indicate that glucose and fructose trigger host-finding and stylet thrusting in promiscuous PPNs, as opposed to host-specific PPNs, an observation which is consistent with the ubiquitous nature of monosaccharide sugars in plant root exudates (Kamilova et al., 2006). The demonstration that STP1 and STP2 are specifically involved in the exudation of both monosaccharides from tomato roots is an important finding which can underpin future efforts to study the link between plant root transporters, and chemical constituents of root exudates.
Acknowledgements
This work was supported financially by Queen’s University Belfast. Dalzell is supported by an Early Career Fellowship from The Leverhulme Trust. Warnock was supported by a Gates Foundation Grand Challenges Grant, Wilson was supported by a EUPHRESCO fellowship, Canet-Perez was supported by an Invest Northern Ireland Proof-of-Concept award and Fleming by a Department of Agriculture and Rural Development studentship award.
Footnotes
↵* Joint first authors