Plant-mediated gene silencing restricts growth of the potato late blight pathogen Phytophthora infestans.

Phytophthora infestans is an oomycete that causes severe damage to potato, and is well known for its ability to evolve rapidly in order to overcome resistant potato varieties. An RNA silencing strategy was evaluated here to clarify if small interfering RNA homologous to selected genes in P. infestans could be targeted from the plant host to reduce the magnitude of the infection. As a proof-of-concept, a hairpin RNA (hp-RNA) construct using the GFP marker gene was designed and introduced in potato. At 72 hpi, a 55-fold reduction of the signal intensity of a corresponding GFP expressing P. infestans strain on leaf samples of transgenic plants, compared with wild-type potato, was detected. This suggests that an RNA interference construct in the potato host could be processed and target a transcript of the pathogen. Three genes important in the infection process of P. infestans, PiGPB1, PiCESA2, and PiPEC, together with PiGAPDH taking part in basic cell maintenance were subsequently tested using an analogous transgenic strategy. Out of these gene candidates, the hp-PiGPB1 targeting the G protein β-subunit (PiGPB1) important for pathogenicity resulted in most restricted disease progress. Further, Illumina sequencing of inoculated transgenic potato leaves revealed sRNAs of 24/25 nt size homologous to the PiGPB1 gene in the transgenic plants indicating post-transcriptional silencing of the target gene. The work demonstrates that a host-induced gene-silencing approach is functional against P. infestans but is highly dependent on target gene for a successful outcome. This finding broadens the arsenal of control strategies to this important plant disease.

Introduction

Phytophthora infestans is the oomycete pathogen responsible for the late blight disease on its potato host (Solanum tuberosum) inciting the worldwide most severe potato losses (Haverkort ; Forbes, 2012). Enormous breeding efforts to produce new varieties with improved resistance have been ongoing for more than 100 years. Exploitation of resistance genes from wild Solanum species started with S. demissum (Reddick, 1928, 1934) and has continued ever since (Vleeshouwers ). Besides pyramiding dominant resistance genes, the emphasis has been on introducing quantitative resistance traits to reduce the short ‘shelf life’ of qualitative resistance genes. However, finding durable resistance gene combinations without negative trait drag from donor species such as, for example, late maturity, remains a challenge. The success of P. infestans as a pathogen originates from its effective reproduction in both asexual and sexual forms. Under ideal conditions, the life cycle can be completed on foliage in about five days, where one lesion can generate up to hundreds of thousands of new sporangia (Fry, 2008). Furthermore, the genome of P. infestans is one of the largest among oomycetes (240Mb), containing vast numbers of transposable elements (TE) (Haas ). There are also hundreds of predicted genes coding for disease-promoting effector proteins, predominantly located in the TE-rich genomic regions which together drive the diversification process, all leading to an exceptional high potential to adapt to new control strategies of the potato crop.

Effector proteins in P. infestans belong mainly to two groups that target distinct sites in the host plant. The apoplastic effectors are secreted into the plant extracellular space, whereas cytoplasmic effectors are translocated inside the plant cell where they target different subcellular compartments (Armstrong ; Kamoun, 2006; Whisson ). Both classes of effectors are modular proteins with cleavable amino-terminal secretion signals. Cytoplasmic effectors carry an additional domain after the signal peptide that mediates translocation inside host cells and is defined by conserved motifs, such as the RXLR amino acid sequence. The discovery of many potential effectors in the genomic sequence of P. infestans has enabled high-throughput analysis for additional resistance gene candidates in a variety of species in the Solanaceae (Vleeshouwers ; Oh ).

Gene silencing or RNA interference (RNAi) is a master regulatory mechanism with diverse roles such as the control of gene expression at transcriptional and post-transcriptional levels and chromatin organization of eukaryotic organisms. Central players in RNA silencing are small RNAs (sRNAs), often ranging in size from 19–40 nucleotides, and divided into different classes with diverse roles (Ghildiyal and Zamore, 2009; Malone and Hannon, 2009; Ruiz-Ferrer and Voinnet, 2009). Generally, sRNAs are generated from long double stranded RNA precursors which are digested by the type III RNase called Dicer or Dicer-like (Dcl) into short duplexes known as siRNA (short interfering RNA). The duplex is unwound and the antisense strand incorporated to an Argonaute (Ago) protein, which then binds to homologous mRNA and degrades it or inhibits its translation. This intramolecular hybridization of self-complementary RNA based on hairpin or stem-loop RNA is a key step in RNA silencing. Another important protein associated with RNAi is RNA-dependent RNA polymerase which is responsible for the conversion of RNA into double-stranded RNA and for the amplification of the silencing signal through the generation of secondary siRNA.

P. infestans possesses the components of canonical gene silencing pathways similar to those of other eukaryotes but the involved proteins display unusual protein domain organization (Vetukuri ). First, only one Dicer-like protein, having the expected dual RNase III domains but lacking other typical domains for Dicers, was identified. A second Dicer-like protein (PiDcl2) was later found in the genome trace archive (Fahlgren ). Small RNAs of approximately 21 nt have been associated with partial silencing in P. infestans (Ah-Fong ) and 40 nt sRNAs were associated with TE silencing (Vetukuri ). Based on deep sequencing of sRNAs from P. infestans, distinct classes of sRNAs at 21, 25/26, and 32 nt were found where the biogenesis of 21 nt sRNAs was shown to be PiDcl1-dependent while longer sRNAs were PiAgo-dependent (Vetukuri ). Furthermore, P. infestans lacks DNA methylation associated proteins known to be ubiquitous in plants.

Plants utilize post-transcriptional gene silencing to protect themselves against invasive nucleic acids such as transposons, viruses, and transgenes (Cogoni and Macino, 2000). Knowledge on the basic mechanisms of gene silencing provides new opportunities to explore plant–pathogen interactions and potential strategies for novel disease control. In this area there are promising reports where RNAi-based constructs in plants were designed to target fungal plant pathogens (Nowara ; Koch ; Ghag ). In order to expand the toolbox for potato breeders not just to rely on dominant resistance genes from various Solanum species, it would be interesting to know if a similar gene-silencing approach driven by the plant host could be a functional strategy to target this oomycete plant pathogen.

The main objective of this study was to investigate the possibility of exploiting RNA silencing to target selected genes in P. infestans via the host plant and, thereby, reduce its capacity to initiate and develop disease on potato. The work showed that sRNAs targeting a pathogen gene, generated in the host plant during infection, incites host-induced gene silencing of the corresponding pathogen transcript. However, the choice of target genes and precursor hp-RNA is crucial for a successful outcome of this strategy.

Materials and methods

P. infestans strains and plant inoculation

The eGFP strain under the promoter of Ham34 (Avrova ) and the wild-type 88069 strain were cultured as described earlier by Vetukuri . Each P. infestans strain at a concentration of 5×104 spores ml–1 water was used for leaf inoculations (Grenville-Briggs ) of potato plants at 4 weeks post transfer to soil from in vitro conditions.

Potato transformation and cultivation

For transformation, in vitro potato plants of cv. Desiree were grown at 22 °C with a 16h photoperiod on Murashige–Skoog (MS) medium (Duchefa Biochemie B.V., Amsterdam, Netherlands), supplemented with 2% (w/v) sucrose and 0.3% (w/v) gelrite (Duchefa). The internodes of the plants were used as explants for transformation (Millam, 2007) using the Agrobacterium tumefaciens strain C58. Ten potential transgenic shoots per construct were grown on MS media with 50 μg ml–1 kanamycin. Plantlets were transferred to soil and grown under the same light and temperature conditions in a culture chamber as for the plants grown in vitro. Validation of transgenic plants was done by PCR (the primers are listed in Supplementary Table S1 at JXB online) and Sanger sequencing.

hp-RNA constructs for potato transformation

All the gene-silencing constructs were made with the ubiquitin1 (UBQ1) promoter for constitutive expression and the heat shock protein 18.2 (HSP) terminator from Arabidopsis thaliana. To enable a hairpin (hp) formation of selected sequences, a 71bp intron from the Ste20-like PiC20 gene of P. infestans (Tani ) was used. cDNA of 88069 and the Ham34:eGFP strains of P. infestans were used as templates in the gene PCR amplification step. Besides the green fluorescence protein (GFP) marker gene, sequences of the following genes were used: G-protein β-subunit (PiGPB1) (Latijnhouwers and Govers, 2003), cellulose synthase A2 (PiCESA2) (Grenville-Briggs ), pectinesterase (PiPEC) (Ospina-Giraldo ), and the constitutive glyceraldehyde 3-phosphate dehydrogenase (PiGAPDH) gene. The choice of target genes relied on published data on PiCESA1 and PiGAPDH (Zeng ; Grenville-Briggs ) together with our analysis of PiGPB1 and PiPEC in infected potato plants (see Supplementary Fig. S1 at JXB online). The specificity of each target gene was determined by BLAST and, using sequences from highly conserved nucleotides, from CLUSTAL-W alignments within each gene family. The BP and LR reactions were performed by using MultiSite Gateway® Pro Technology (Invitrogen, Carlsbad, CA, USA), and recombined into the binary vector pGWB1 (Nakagawa ). All primers used for plasmid construction are listed in Supplementary Table S2 at JXB online, and detailed cloning steps are outlined in Supplementary Fig. S2 at JXB online. All constructs were confirmed by sequencing (Macrogen Inc. Seol, Korea).

Quantitative real-time PCR (qRT-PCR)

Leaf samples from wild-type, cv. Desiree, and transgenic plants were collected at 24, 48, and 72h post-inoculation (hpi), snap-frozen in liquid nitrogen, and stored at –70 °C. The leaf materials were used for RNA extraction, cDNA synthesis, and SYBR green qRT-PCR assays (Vetukuri ). To check the transcript level for each gene of interest, specific primers were designed to anneal outside the hairpin sequence. Primers are listed in Supplementary Table S3 at JXB online. Calculations and statistical analyses were carried out according to Avrova . Primers for hairpin expression analysis are listed in Supplementary Table S4 at JXB online. DNA quantification of P. infestans and calculation was as described earlier by Llorente . At least three biological replicates were used for each individual qRT-PCR analysis.

Phenotypic analysis of transgenic potato plants

Colonization and disease progression were monitored up to 30 d after inoculations. The effect of PiGPB1 silencing on spore formation was assayed by analysing the shape and number of sporangia using an epifluorescence microscope (Leica DMI 4000). Sporangia were collected at 24, 48, 72, and 96 hpi from detached leaves and counted 1h after leaf detachment as described by Lehtinen .

Confocal microscopy

The GFP expression levels on leaves of the transgenic and wild-type plants inoculated by the GFP-expressing Ham34:eGFP P. infestans strain were analysed by confocal microscopy (Zeiss LSM510). GFP was excited at 488nm,and detected at 505–530nm. Images were analysed using the LSM 510 software.

Northern blot hybridization

To detect sRNA molecules homologous to the PiGPB1 gene, Northern blot hybridization was performed (Vetukuri ). Primers used for the generation of PiGPB1-specific riboprobes were: Spe-Fw ATATACTAGTAGT TCTCAGCCAATCTTCG and Not-Rev ATATGCGGCCGCTTCAACTTGGT CTAGTTTCCAT.

Western blotting

Total GFP proteins from the GFP-tagged P. infestans strain colonized on hp-GFP transgenic plants, were resolved on SDS-containing 9% polyacrylamide gel. The proteins were transferred to nitrocellulose membrane and incubated with an anti-GFP monoclonal antibody, following the procedure described previously by Moschou . The membranes were subsequently incubated with a chemi-luminescent substrate (ECL Western blotting system kit, GE Healthcare, Uppsala) and exposed using a Fuji Phosphorimager. Ponceau staining of the membrane (Bannur ) was used as a loading control.

Small RNA sequencing

Total RNA was extracted from hp-PiGPB1 transgenic and wild-type potato leaves at 24, 48, and 72 hpi infected by the 88069 strain, the 88069 mycelia control, and non-infected transgenic plants using the mirVanaTM miRNA isolation kit (Ambion, Austin, TX, USA). Eight sRNA libraries were generated using the Illumina small RNA sample preparation kit and sequenced using Illumina HiSeq 2500 at SciLifeLab, Stockholm, Sweden.

Bioinformatic analysis

All sRNA reads were trimmed at any base with a phred-scaled quality score below 10 and adapter sequences were removed using Cutadapt v1.3 (Martin, 2011). Reads shorter than 18 bases in length were excluded from all further analysis. The resulting set of filtered reads for each sample was aligned to the P. infestans reference genome (http://www.broadinstitute.org/annotation/genome/phytophthora_infestans/MultiDownloads.html) using Bowtie2 v2.1.0 (Langmead and Salzberg, 2012) with one mismatch allowed per seed (-N 1), a seed length of 18 (-L 18), and the remainder of alignment options set to their default values. The SAM format alignments produced by Bowtie2 were converted to BAM files, sorted, and indexed using SAMtools v0.1.19 (Li ). The number, locations, lengths, and sense/antisense orientations of reads mapping to the PiGPB1 gene in P. infestans were obtained using a combination of the SAMtools view command and custom python scripts.

Results

Generation and validation of transgenic potato harbouring hp-RNA constructs

A series of gene-cloning steps were performed to accomplish binary vectors containing the target gene sequence in a sense and antisense orientation together with an intron (I) to enable a short hp-RNA formation (see Supplementary Fig. S2 at JXB online). This construction work resulted in five binary vectors: UBQ1:GFP-I-GFP; UBQ1:PiGPB1-I-PiGPB1; UBQ1:PiCesA2-I-PiCesA2; UBQ1:PiPEC-I-PiPEC; and UBQ1:PiGAPDH-I-PiGAPDH, that were used for A. tumefaciens-mediated transformation. Ten transgenic potato plants from each construct were made, and three individual lines L1, L2, and L3 were used in subsequent analysis. The transgenic status of the plant materials was validated by PCR amplification of the sequences matching individual constructs in DNA samples from the different transgenic plants (see Supplementary Fig. S3A at JXB online) and by performing Sanger sequencing to check that each construct was intact in the plants, followed by analysing the transcription levels of each gene by qRT-PCR (see Supplementary Fig. S3B at JXB online). Further, in order to verify that the silencing of the target gene was not an effect of the transformation procedure it self, the PiGPB1 transcript level was analysed in hp-GFP plants and compared with hp-PiGPB1 and wild-type potato plants infected with P. infestans. The PiGPB1 transcript level in the hp-GFP plants compared with wild-type plants remained intact at 24, 48, and 72 hpi while the transcript level was gradually decreasing with time in hp-PiGPB1 plants as expected (see Supplementary Fig. S3C at JXB online).

A GFP-expressing P. infestans strain can be targeted from the host plant

Three transgenic potato lines (hp-GFPL1, hp-GFPL2, and hp-GFPL3) harbouring the hp-RNA construct matching a GFP-tagged P. infestans strain were analysed first. The intensity of the GFP signal in the P. infestans-colonizing transgenic potato leaves was monitored and quantified using confocal microscopy. There was a marked reduction in the GFP signal at 72 hpi in all three transgenic lines compared with wild-type plants (Fig. 1A, B; see Supplementary Figs S4 and S5 at JXB online). At 72 hpi, as much as a 55-fold reduction of the signal intensity of the GFP-expressing P. infestans strain was recorded on leaf samples of transgenic plants compared with wild-type potato (Fig. 1C). In order to monitor the GFP-silencing in P. infestans further, transcript levels were checked. A significant reduction in relative transcript levels of the GFP gene in P. infestans growing on transgenic plants at 48 and 72 hpi was detected compared with wild-type potato (Fig. 1D). To check if the protein level in the hp-GFP potato lines was altered, a Western blot was run (Fig. 1E). A reduction of the GFP protein was seen in the potato transgenic lines compared with wild-type plants upon infection with the GFP-tagged P. infestans isolate. Together, these data suggest that hp-RNA in the potato host could be processed and target a transcript of the invading P. infestans pathogen.

Fig. 1.

Silencing of GFP in P. infestans by hp-RNA. Confocal laser scanning microscopy of P. infestans transformants (Ham34:eGFP) expressing green fluorescent protein. GFP expression in mycelia grown on (A) wild-type and (B) hp-GFPL1 (UBQ:GFP-I-GFP) transgenic potato. Bars=25 μm. (C) Fold change of total intensity of GFP in mycelia grown on hp-GFP plants compared with wild-type plants. AU=arbitrary unit. (D) Transcript abundance of GFP in P. infestans transformants grown on wild-type and transgenic plants at 24, 48, and 72 hpi, quantified by qRT-PCR. Data are normalized to P. infestans actinA mRNA levels and represent means ±SE (n=3 pooled leaves of 3 plants). Asterisks indicate significant difference to the wild type (Student’s t test; *P <0.05; **P <0.01; ***P <0.001). (E) Western blot analysis of GFP protein. hp-GFP plant lines inoculated with: wild type (wt) P. infestans (88069), the GFP-tagged P. infestans, wild-type plants inoculated with: the GFP-tagged P. infestans (+), 88069 (–), and water. Ponceau staining was used as loading control.

Monitoring RNA-mediated inhibition of disease progression

Encouraged by the data on GFP silencing in P. infestans, the next step was to monitor the effects of host-mediated silencing of the PiGPB1, PiCESA2, PiPEC, and PiGAPDH genes. A reduction of P. infestans DNA content over time on these transgenic plant leaves compared with the wild type was generally found but that trend was most obvious in the hp-PiGPB1 lines (Fig. 2A). In the case of the hp-PiPEC plants, a rather rapid decrease of P. infestans DNA was already detected at 24 hpi (Fig. 2B). By contrast, the DNA levels were lowest at 72 hpi compared with earlier time-points in the hp-PiCESA2 plants but not reaching the level seen in the hp-PiGPB1 lines (Fig. 2C). In the leaf samples from the hp-PiGAPDH transgenic lines, the growth of P. infestans seemed to be less at 48 and 72 hpi compared with 24 hpi (Fig. 2D). For all these transgenic plants, colonization by P. infestans occurred as necrotic lesions on the leaves to a limited extent compared with the wild-type plants (see Supplementary Fig. S6 at JXB online).

Fig. 2.

P. infestans DNA in leaves of the wild type and three individual transgenic lines L1, L2, and L3 of (A) hp-PiGPB1, (B) hp-PiPEC, (C) hp-PiCESA2, and (D) hp-PiGAPDH potato plants quantified by qRT-PCR at 24, 48, and 72 hpi. Data are normalized to potato EF1 DNA levels and represent means ±SE (n=3 pooled leaves of 3 plants). Asterisks indicate significant difference to the wild type. (Student’s t test; *P <0.05; **P <0.01; ***P <0.001).

Reduced transcript level in hp-PiGPB1 and hp-PiGAPDH plants

Further studies were done on the hp-PiGPB1 plants where a gene important for sporangia formation was targeted with promising effects on the pathogen. The basic cell maintenance gene in the hp-PiGAPDH plants was included for comparison. In order to monitor the silencing of PiGPB1 and PiGAPDH in P. infestans, transcript levels of these genes were checked in the transgenic material. The relative transcript levels of PiGPB1 at 48 and 72 hpi were significantly down-regulated in all three transgenic lines compared with inoculated wild-type plants (Fig. 3A). In the case of the hp-PiGAPDH plants, a reduced transcript level of PiGAPDH was particularly seen at 48 hpi and 72 hpi (Fig. 3B). These results were in line with a reduced P. infestans DNA content detected in infected leaves. To record overall plant performance, the plants were allowed to grow for a month. After 30 d, the inoculated wild-type plants were almost dead, whereas transgenic hp-PiGPB1 plants showed greatly reduced disease spread disease symptoms (Fig. 3C). A similar performance was seen for the hp-PiGAPDH transgenic plants (Fig. 3D), although the effect was not as prominent compared with the hp-PiGPB1 plants.

Fig. 3.

Analysis of transgenic potato harbouring a hp-PiGPB1 or hp-PiGAPDH construct. (A) Transcript abundance of the PiGPB1 in wild-type and three lines of hp-PiGPB1 plants. (B) Transcript abundance of the PiGAPDH gene in wild-type and hp-PiGAPDH plants inoculated with P. infestans at 24, 48, and 72 hpi quantified by qRT-PCR. Data are normalized to P. infestans actinA mRNA levels and represent means ±SE (n=3 pooled leaves of 3 plants). Asterisks indicate significant difference to the wild type (Student’s t test; *P <0.05: **P <0.01; ***P <0.001). Overall performance of (C) wild-type and hp-PiGPB1 transgenic plants, and (D) wild-type and hp-PiGAPDH transgenic plants at 30 dpi.

Host induced silencing of PiGPB1 gene affects spore formation

PiGPB1 is a gene known to be involved in phosphatase-mediated signalling in mycelia leading to the formation of sporangia (Kasahara ; Latijnhouwers and Govers, 2003). Silencing of this gene causes P. infestans to produce abundant aerial mycelia and very few defective sporangia. In the present study, no sporangia were found on leaves of the hp-PiGPB1 plants. In order to assay any phenotypic changes whatsoever, a detached leaf assay was used where inoculated leaves were taken at 24, 48, 72, and 96 hpi from the hp-PiGPB1 plants. Leaving the leaves detached for 1h in a plastic box under humid conditions promoted sporangia formation. A condition that by time favoured the pathogen over the plant host with declining hp-PiGPB1 expression (Fig. 4B). Yet, a 6-fold reduction in the number of sporangia on colonized transgenic leaves was observed compared with the wild-type at 96 hpi (Fig. 4A). Sporangia produced by P. infestans at 24 hpi, grown on both control and hp-PiGPB1 leaves, were normal with the expected ‘lemon shapes’ and single hyphae. By contrast, sporangia formed on hp-PiGPB1 plants at 48, 72, and 96 hpi were abnormal with an outgrowth of multiple hyphae and the loss of the characteristic phenotype at 48 hpi (Fig. 4C). This reduction in number of sporangia and sporangia with multiple hyphal outgrowths are in agreement with the phenotype of the PiGPB1-silenced mutants of P. infestans observed earlier by Latijnhouwers and Govers (2003).

Fig. 4.

Effects on spore formation in transgenic hp-PiGPB1 plants. (A) Numbers of sporangia produced by the 88069 strain of P. infestans on wild-type and transgenic plants at 24, 48, 72, and 96 hpi. Data represent means ±SE (n=3 pooled samples of 3 plants). (B) Relative expression of hp-PiGPB1 construct in hp-PiGPB1 plants inoculated with P. infestans at 0, 1, and 3h after leaf detachment. Data are normalized to P. infestans actinA mRNA levels and represent means ±SE (n=3 pooled leaves of 3 plants) (C) Phenotypes of sporangia collected at 24, 48, 72, and 96 hpi from wild-type and hp-PiGPB1 transgenic plants. Bars =10 μm.

Detection of sRNAs homologous to PiGPB1 gene in P. infestans

In order to confirm the presence of sRNAs homologous to hp-RNA constructs in transgenic plants, Northern blot hybridizations were used. Probes specific to PiGPB1 detected antisense sRNAs, ranging from 21, 25, 26, 27, and 28 nt in size (see Supplementary Fig. S7 at JXB online). Subsequently, Illumina sequencing of sRNAs generated from wild-type and hp-PiGPB1 transgenic plants resulted in unique but few 24/25 nt size class sRNAs only mapping to the PiGPB1 gene in the transgenic plants and not seen in control plants (Fig. 5). These data suggest that siRNAs are generated from the hp-RNA constructs in transgenic plants during the infection process.

Fig. 5.

Small RNAs complementary to the PiGPB1 gene. Distribution of small RNA reads homologous to the PiGPBI gene in transgenic hp-PiGPB1 and wild-type potato plants at 24, 48, and 72 hpi based on Illumina sequencing.

Discussion

This work demonstrated the possibility that hp-RNA constructs in potato could affect the host colonization and invasion by P. infestans to various degrees. Three genes of importance in the first stages of infection, together with a gene involved in basic cell maintenance, were examined. The current understanding of key genes and signals in early plant infection phases is meagre. It could be anticipated that various receptor classes are main players in these interactions, not least G protein-coupled receptors common in eukaryotes. Heterotrimeric G proteins are known to be key regulators in numerous signalling pathways impacting arrays of downstream targets. However distinct differences are known between animals and plants (Urano and Jones, 2014). P. infestans has only a single α (PiGPA1) and one β (PiGPB1) subunit gene. PiGPB1 is differentially expressed in various stages of the life cycle of P. infestans (Laxalt ; Latijnhouwers and Govers, 2003; Dong ). Primarily, PiGPB1 is important for sporangia formation and is not strongly expressed in mycelia. When the infection progresses from a biotrophic to a necrotrophic mode at 36–48 hpi, the continuous formation of sporangia is required (Judelson and Blanco, 2005), a process which is impaired in the hp-PiGPB1 transgenic plants. Fewer sporangia, and at the same time abnormal phenotypes, impose difficulties to infect new host cells leading to reduced colonization and spread of the pathogen.

Oomycetes have cell walls composed of cellulose, not chitin as fungi do, but both pathogenic organism groups have evolved pectinases to break down plant cell walls, enzymes aiding in the penetration and subsequent establishment of the infection. Phytophthora species encode large sets of different cell wall-degrading enzymes (Ospina-Giraldo ). Thus, influencing cell wall-associated genes, such as hp-PiCESA2 and hp-PiPEC, was also of great interest. In the hp-PiCESA2 plants, P. infestans is supposed to have impaired the formation of appressoria (Grenville-Briggs ), leading to attenuated colonization but such an effect was not consistent over time. On the hp-PiPEC plants, P. infestans encountered difficulty in establishing the infection process but the few spores that managed to penetrate the mesophyll cells continued to colonize the leaves. Some of this ‘incomplete’ inhibition could be attributed to redundancy since CESA, PEC, and GAPDH belong to gene families of varying size in P. infestans. The challenge here was to design hp-gene constructs that specifically targeted P. infestans and not the host plant.

In the Arabidopsis–Phytophthora parasitica interaction, host-driven gene-silencing targeting GFP and the PnPMA gene encoding for a plasma membrane H+ ATPase were not reported to be successful and was explained by the lack of proper machinery required for the uptake of silencing signals in the oomycete pathogen (Zhang ). This is contrary to what is already a well-established methodology in oomycete research; experiments where the treatment of protoplasts with hp-RNA results in transient RNA silencing, which shows beyond doubt that oomycetes do have the machinery needed for the uptake of silencing molecules (Whisson ; Vetukuri ). Nevertheless, the mechanism of RNA uptake from the plant host during the infection process into P. infestans cells has not been well studied. It was shown recently that the fungus Botrytis cinerea delivers sRNAs to hijack the host RNAi machinery and selectively silences host immunity genes during the infection process of Arabidopsis (Weiberg ). Notably, out of the 73 B. cinerea sRNAs with the potential to silence plant genes, 52 were derived from transposable elements (TE) of the fungal genome. If sRNAs are included in the vast list of P. infestans effectors, this implies thousands of new candidates because of its TE-saturated genome constitution, and adds to the enormous capacity to overcome host resistance. The Illumina sequencing data revealed sRNAs of 24/25 nt size homologous to the PiGPB1 gene in the transgenic plants, indicating post-transcriptional silencing of the target gene. The number of sRNA reads detected were considerably less than expected which could be attributed to unknown timing for the required cellular processes, which thereby leads to non-optimal timing of sample collection. However, based on hairpin RNA expression studies in plants, it can be presumed that the sRNAs are initially formed on the plant side (Mansoor ).

The process of effector protein translocation between potato and P. infestans has been explored for many years and, hitherto, led to inconclusive results (Wawra ). Similarly, the sRNA translocation process is poorly understood. No homologous proteins to SID1 and SID2, earlier identified in C. elegans to be responsible for the intercellular communication of the RNA-induced silencing signal, and also environmental RNA interference (Winston , 2007) are neither found in P. infestans nor in potato or Arabidopsis genomes. In the case of the barley–Blumeria graminis interaction, exosomes have been speculated to be involved in the translocation of the silencing signal (Nowara ) but this critical question remains to be answered.

Future potential

The hp-RNA-mediated gene silencing of P. infestans clearly contributes to reduce disease progress in potato and should be regarded as an additional strategy to complement resistance gene deployment. Broader resistance would most likely be achieved by combining several RNAi transgenes, preferably targeting single copy genes of importance in the infection process. Understanding potential sRNA transport from P. infestans to the host plant opens up alternative revenues to control this important plant pathogen. During the processing of this paper one work on the interaction between B. lcatucae and lettuce was published (Govindarajulu ).

Supplementary data

Supplementary data can be found at JXB online.

Supplementary Fig. S1. Data on in planta expression of PiGPB1 and PiPEC gene.

Supplementary Fig. S2. Schematic representation of plasmid constructs.

Supplementary Fig. S3. Data on hp-RNA constructs.

Supplementary Fig. S4. Silencing of GFP in P. infestans by hp-RNA.

Supplementary Fig. S5. Mycelia growth in hp-GFP transgenic potato lines.

Supplementary Fig. S6. Overall picture of necrotic lesions on leaves.

Supplementary Fig. S7. Northern blot analysis.

Supplementary Table S1. PCR primers for transgene detection.

Supplementary Table S2. Primers for plasmid constructs.

Supplementary Table S3. Primers for qRT-PCR.

Supplementary Table S4. Primers for hp-RNA expression analyses.