Transgenic Nicotiana benthamiana plants expressing a hairpin RNAi construct of a nematode Rs-cps gene exhibit enhanced resistance to Radopholus similis

Burrowing nematodes (Radopholus similis) cause severe harm in many agronomic and horticultural crops and are very difficult to manage. Cathepsin S is one of the most important cysteine proteinases and plays key roles in nematodes and many other parasites. To evaluate the effect of in planta RNAi on the control of this nematode, a specific fragment from the protease gene, cathepsin S (Rs-cps), was cloned into the binary vector pFGC5941 in the forward and reverse orientations to construct recombinant plant RNAi vectors. Transgenic Nicotiana benthamiana plants expressing Rs-cps dsRNA were obtained and studied. The transcript abundance of Rs-cps dsRNA appeared to be diverse in the different transgenic lines. Moreover, the bioassay results revealed that Rs-cps transgenic N. benthamiana plants were resistant to R. similis and the transcription level of Rs-cps in R. similis was drastically decreased. In addition, the reproduction and hatching rate of R. similis isolated from the Rs-cps transgenic plants were also significantly reduced. Our results suggest that Rs-cps is essential for the reproduction and pathogenicity of R. similis. This is the first study to employ in planta RNAi approach to target the Rs-cps gene for the control of plant parasitic nematodes.

were introduced into N. benthamiana explants by Agrobacterium-mediated transformation. The transgenic seedlings germinated from transformed N. benthamiana calli were selected for kanamycin resistance, and then, the seedlings were transferred into the rooting medium when they grew to approximately 2 cm. After acclimatization, the transgenic plantlets were transplanted into sterilized nutritive soil in a greenhouse for normal growth. We obtained a total of 26 independent kanamycin-resistant transgenic plants, including 15 Rs-cps transgenic plants and 11 egfp transgenic plants. These transgenic plants had wild-type N. benthamiana morphology and growth (result not shown).

Molecular analysis of Rs-cps transgenic N. benthamiana plants. The independently generated T0
generation transgenic lines were analysed by PCR. A 467-bp DNA fragment of the target gene was amplified from most of the putative T0 generation Rs-cps transgenic plants, and a 315-bp DNA fragment was amplified from most of the putative egfp transgenic plants, while no specific band was amplified from the wild-type N. benthamiana plants ( Fig. 2A,B). These results showed that the target DNA fragment was successfully inserted into the N. benthamiana genomic DNA. Although 13 of the 15 the kanamycin-resistant transgenic plants test positive for the RNAi construct in PCR, some detected plants showed nonspecific bands. Therefore, we only select part of transgenic plants (Nos. 3, 4, 6, 7, 12, 13 and 14) with specific bands for the Southern blot. The results showed that these Rs-cps transgenic plants had one to four insertion loci (Plants No. 4 and 13 had a single insertion locus, plants No. 6 and 14 had two insertion loci, plants No. 7 and 12 had three insertion loci, and plant No. 3 had four insertion loci), but no Rs-cps hybridization bands were observed with genomic DNA from the egfp transgenic plants and wild-type N. benthamiana plants (Fig. 2C). RT-PCR was performed to detect the expression of Rs-cps dsRNA in positive transgenic N. benthamiana lines (Nos. 3, 4, 6, 7, 12, 13 and 14). A 467-bp fragment corresponding to the sequence of Rs-cps was amplified from these positive Rs-cps transgenic plants, while no specific amplification was observed from the wild-type N. benthamiana plants (Fig. 2D). These results indicated that the integrated Rs-cps dsRNA was successfully expressed in transgenic N. benthamiana plants. Genetic stability analysis indicated that the integrated Rs-cps could be inherited steadily in the genomic DNA of the T1 generation transgenic N. benthamiana plants (result not shown).

Expression analysis of Rs-cps in T2 generation transgenic lines. To detect the expression of the
Rs-cps dsRNA transcript and its abundance, qPCR analysis of the selected single copy T2 generation transgenic lines (No. 4 and 13) was carried out. The result showed that Rs-cps dsRNA was expressed in two positive transgenic lines, while the transcript abundance of dsRNA varied between the two lines. The Rs-cps dsRNA expression level in line No. 4 was 3.07 times higher, which was significant (p < 0.05), than that in line No. 13 of the Rs-cps transgenic N. benthamiana plants (Fig. 3). Previous studies have shown that the effectiveness of RNAi is higher in single-copy lines than in other lines 36,37 . Therefore, the single-copy Rs-cps transgenic line (No. 4) with a relatively high expression level of dsRNA was selected for a nematodes feeding bioassay to analyse the reproduction and pathogenicity of R. similis and the efficiency of RNAi.   (Fig. 4). The pot experiment results showed that the plant height, fresh above-ground plant weight and fresh root weight of the Rs-cps transgenic plants (No. 4) were significantly higher than those in the control groups (p < 0.05). There was a significant difference in the fresh root weight between uninoculated wild-type and Rs-cps transgenic N. benthamiana plants (p < 0.05); however, no significant differences in the plant height and fresh above-ground plant weight were observed between them (p > 0.05). There was no significant difference in the three growth parameters between the control groups (p > 0.05) ( Fig. 5A-C). Additionally, the number of nematodes in the rhizosphere of the Rs-cps transgenic plants was 1158, which was significantly lower than that of the egfp transgenic plants (3954) and wild-type N. benthamiana plants (4182) (p < 0.05) (Fig. 5D). The pot inoculation trials clearly demonstrated resistance to R. similis in the T2 generation Rs-cps transgenic N. benthamiana plants was significantly improved.  Transcription of Rs-cps in tested R. similis was dramatically suppressed by T2 generation transgenic N. benthamiana plant-derived Rs-cps dsRNA. We assume that the significant reduction in pathogenicity of R. similis resulted from suppression of its Rs-cps mRNA levels by feeding on transgenic plants expressing Rs-cps dsRNA. To confirm this hypothesis, the Rs-cps mRNA levels in R. similis isolated from the transgenic N. benthamiana roots were detected by qPCR. The results showed that (Fig. 6) the transcription level of Rs-cps in R. similis isolated from transgenic N. benthamiana plants expressing Rs-cps dsRNA was significantly lower (p < 0.05) than that from the control groups (egfp transgenic plants and wild-type N. benthamiana plants) and was as much as 74.5% lower than that from wild-type N. benthamiana plants. There was no significant (p > 0.05) difference in the Rs-cps expression between the two control groups. Taken together, we conclude that the suppression of Rs-cps expression in R. similis by feeding on the roots of transgenic plants expressing Rs-cps dsRNA causes weaker pathogenic ability.
Transgenic N. benthamiana plants expressing Rs-cps dsRNA inhibited the reproduction and hatching of R. similis. As described previously, the reproductive capability of R. similis treated with Rs-cps dsRNA (in vitro RNAi) was significantly decreased compared with the control groups 16 , and the number of nematodes in the rhizosphere of the Rs-cps transgenic plants was significantly lower than that of the control groups. To evaluate the effect of transgenic N. benthamiana plant-derived dsRNAs (in planta RNAi) on the reproduction and hatching of R. similis, the nematodes were inoculated onto carrot callus. After being cultured for 50 d, the nematodes isolated from the roots of Rs-cps transgenic N. benthamiana plants exhibited significantly lower reproduction than those from the roots of egfp transgenic and wild-type N. benthamiana plants. The number of nematodes in the Rs-cps transgenic N. benthamiana groups (CPS) was 2136, which was significantly (p < 0.05) lower than those in the egfp transgenic N. benthamiana groups (16296) and wild-type N. benthamiana groups (15536), but there was no significant (p > 0.05) difference observed between the latter two groups (Fig. 7A). In terms of the infection symptoms, the carrot callus presented brown hygrophanous lesions because of the reproduction of a large number of nematodes derived from the roots of egfp transgenic and wild-type N. benthamiana plants. The carrot callus inoculated with nematodes derived from Rs-cps transgenic N. benthamiana roots only presented a slightly brown infection spot, and the brown hygrophanous lesion was not observed in this experiment ( Fig. 7C-F). The hatching rate of eggs derived from Rs-cps transgenic N. benthamiana plants was 32%, which was significantly lower (p < 0.05) than that in the control groups (eggs derived from egfp transgenic and wild-type N. benthamiana plants). The hatching rates of eggs derived from the egfp transgenic and wild-type N. benthamiana plants were 88% and 90%, respectively, and there was no significant difference (p > 0.05) between these groups (Fig. 7B). All these results suggested that the Rs-cps transgenic plant-derived dsRNAs not only inhibited reproduction but also suppressed the hatching of R. similis.

Discussion
Using RNAi in crop pathogens and pest control to suppress target genes using specific dsRNA has been well documented in recent years. Among the early studies, dsRNA was usually delivered by ingestion 20 , the injection method 38 and feeding via bacteria expressing dsRNAs 39,40 . In spite of the huge costs, the injection method was still widely used for insect pests due to its high efficiency and accuracy. However, using a similar method for plant parasitic nematodes has been a major challenge. The introduction of special dsRNAs into plant parasitic nematodes could not be achieved consistently using microinjection 26 . The ingestion method (performed by soaking with synthesized dsRNA in vitro) and another approach to deliver special dsRNA into nematodes both showed potential for application in nematode control once special dsRNA was verified to down-regulate target gene expression. However, the efficiency of the latter two methods may not be as satisfactory for the instability of dsRNA when exposed to diverse environments. As a result, the above three methods are not applicable for the control of plant parasitic nematodes in farmland conditions. As an alternative, feeding the nematodes with transgenic plants expressing special dsRNA has been developed as an effective method to deliver dsRNA and control plant parasitic nematodes in agriculture. Transgenic plants possess relatively stable expression level of dsRNAs, providing nematodes with continual stress. Using in planta RNAi against plant parasitic nematodes was first described for root-knot nematodes 41 . It has also been used to study the control methods of many nematodes, including M. incognita 19,26,42,43 , M. javanica 44 , H. glycines 45,46 , R. similis 28,29 and Pratylenchus vulnus 27 . However, using an in planta RNAi method to target the cps gene for the control of plant parasitic nematodes has not yet been reported.
In this study, the cps gene from R. similis was selected as the special dsRNA construct, which was then transferred into N. benthamiana plants. The exact length of the dsRNA fragment needed to trigger RNAi in plants and other eukaryotes is still not entirely clear 47 . The targeting sequence for gene silencing in plants is approximately 300 to 700 bp in length 48 . In feeding experiments, most sequences range from 300 to 520 bp 49 . Previous research has shown that long dsRNA sequences may lead to greater RNAi effect compared to short dsRNAs 50 . Therefore, in the present study, a 438-bp partial cDNA sequence of Rs-cps was chosen as the target fragment for RNAi. As reported by Mao et al. 51 , the long dsRNA produced in plants could effectively suppress insect gene expression.
Our study also showed that the transcription of Rs-cps in tested R. similis was dramatically suppressed by T2 generation transgenic N. benthamiana plant-derived Rs-cps dsRNA. Therefore, the result of the Rs-cps expression in R. similis suggests that the use of the 438-bp target fragment was successful in generating RNAi in this study.
Cysteine proteinases play key roles in nematodes and many other animal parasites 31 . Nematode cysteine proteinases mainly include cathepsin B-, L-, S-, K-, and Z-like cysteine proteinases, and cathepsin L and cathepsin B have been extensively studied in recent years 29,31,[33][34][35] . Cathepsin S is one of the most important cysteine proteinases in nematodes and many other parasites. However, the functions of cathepsin S have not been researched using in planta RNAi in plant nematodes. This study was designed to investigate the functions of Rs-cps in R. similis and to find new targets for its control. Therefore, we used the similar method and strategy in this manuscript as described previously. The strategy of in planta RNAi against plant nematodes has been tested in several economically important crops 25,28,33,[44][45][46] . All these works were using in planta RNAi to study gene functions of plant nematodes. However, only one economically important crop has been tested in each study. What's more, the main purpose of this study is to investigate the functions of Rs-cps and to find new targets for its control. Therefore, we used a single model plants (N. benthamiana) for the RNAi research in this study. Previous studies have shown that the effectiveness of RNAi is higher in single-copy lines than in other lines 36,37 . Our results showed that No. 4 and No. 13 Rs-cps transgenic N. benthamiana plants had a single copy insertion. The expression level of Rs-cps dsRNA in line No. 4 was significant (p < 0.05) higher than that in line No. 13 of the Rs-cps transgenic plants. Therefore, we only selected the higher expression single-copy No. 4 Rs-cps transgenic line for the nematodes feeding bioassay.
Rs-cps, the target gene chosen in this study, plays important roles in the reproduction and pathogenesis of R. similis. In our previous studies, we verified the feasibility of RNAi for the Rs-cps gene in R. similis by soaking with in vitro synthesized dsRNA 16 . Therefore, the Rs-cps gene is expected to be a promising target for controlling R. similis. In this study, N. benthamiana plants were transformed to express the special Rs-cps dsRNA and challenged with nematodes to research the Rs-cps function in R. similis using in planta RNAi. Previous studies have shown that delivering dsRNA to plant parasitic nematodes by feeding them with transgenic plants results in the reduction of the target mRNA expression by approximately 65% to 80% 28,29,44 . In this study, the transcription level of Rs-cps in R. similis fed with Rs-cps transgenic N. benthamiana plants was significantly reduced by 74.5% compared with the control group (p < 0.05). However, this dsRNA-mediated RNAi did not completely eliminate expression of the gene products. That is, the target genes can be knocked down, rather than knocked out, through plant-mediated RNAi. This was consistent with the results seen using in vitro RNAi-mediated gene silencing 11,[14][15][16] . There are several possible reasons for this phenomenon. The first reason for the phenomenon may be due to the lower accumulation of dsRNA in the transgenic plants. Dicer-like enzymes in plants could have decreased the expression levels of the generated dsRNA in these transgenic lines 47 . The second reason may be that there are energy metabolic pathways that compensate for the RNAi targeting genes.
In this study, we confirmed that the resistance of T2 generation Rs-cps transgenic N. benthamiana plants to R. similis was significantly improved. The result was consistent with the roles of Rs-cps in R. similis, which was verified by in vitro RNAi 16 . In planta RNAi has been used to research gene functions in cyst nematodes, root-knot nematodes, R. similis and other plant parasitic nematodes. As reported by Klink et al. 46 , in planta RNAi targeting of four genes of H. glycines caused 81-93% fewer females to develop in transgenic soybean roots. In another report, transgenic Arabidopsis expressing 16D10 dsRNA had a wide resistance against four major root-knot nematode species 19 . Steeves et al. demonstrated that MSP transgenic soybean plants significantly reduced the reproductive potential of H. glycines 45 . All these studies were done using in planta RNAi to research gene functions and demonstrate the efficacy of an RNAi-based strategy for controlling sedentary plant parasitic nematodes. There is limited information available regarding in planta RNAi against migratory endoparasites compared to sedentary endoparasites, even though Li et al. reported that the T2 generation transgenic plants expressing special dsRNA show enhanced resistance to R. similis 28,29 . In this study, we first confirmed the feasibility of in planta RNAi in controlling R. similis by targeting the Rs-cps gene. Steeves et al. 45 first confirmed the inheritability of gene silencing induced by in planta RNAi in the sedentary plant nematode H. glycines. In another report, burrowing nematodes infecting transgenic N. benthamiana plants expressing specific dsRNA targeting the Rs-cb-1 gene showed a significant reduction in pathogenicity. Interestingly, the progeny nematodes (F1 generation) that hatched from the eggs also showed a significant reduction in pathogenicity and reproduction when allowed to infect the wild-type N. benthamiana plants. Our study showed that the hatching rate of eggs derived from Rs-cps transgenic plants was 32%, which significantly lower than that in control groups. The result was consistent with the inheritability of gene silencing induced by in planta RNAi described above. In conclusion, this study demonstrated that specific dsRNA targeting Rs-cps was effective at suppressing the reproductive capability and hatching of R. similis, and the resistance to R. similis was significantly improved in Rs-cps transgenic N. benthamiana plants. To the best of our knowledge, this is the first study to report the use of an in planta RNAi approach targeting the Rs-cps gene to control R. similis. The current results will enrich the understanding of the use of RNAi against migratory plant parasitic nematodes and provide a scientific basis for the control of other nematodes and pests.

Methods
Plant materials and growth conditions. Nicotiana benthamiana was used as the background line for transformation in this study. For germination, seeds were soaked in sterile water for 1 h at 26 °C, treated with 75% (v/v) absolute ethanol for 1 min and with 0.75% NaOCl for 30 min, washed several times with sterile water, and then sowed on MS medium (pH 5.8) solidified with 0.3% phytagel 52,53 . The aseptic seeds germinated, and the seedlings were cultured in a growth chamber at 26 ± 1 °C under a photoperiod of 16 h-light /8 h-dark 22 . Culturing of nematodes. Carrot discs were prepared as previously described 54 . The banana burrowing nematode was collected from the roots of Anthurium andraeanum and cultured on carrot discs at 25 °C in a dark incubator 55 . Approximately 50 d after inoculation, the cultured nematodes were extracted from the carrot discs according to the method of Zhang et al. 14 . Cloning and sequencing of the Rs-cps gene. Total RNA was isolated from mixed-stage nematodes using TRIzol reagent (Invitrogen, USA) and digested with DNase (RQ1, Promega) to remove the contaminating DNA molecules. The quality of the extracted RNA was assessed using a spectrophotometer (Nanodrop ND-2000C, Thermo Scientific). The full-length cDNA sequence of Rs-cps was amplified using the specific primers Rs-cps-S1 and Rs-cps-A1 (Table 1), and the purified PCR product was cloned and confirmed by sequencing as previously described 16 . The recombinant pMD20-Rs-cps plasmid was extracted for later use.

Plant expression vector construction and N. benthamiana transformation. For the Rs-cps plant
RNAi expression vector, the partial coding region of Rs-cps was amplified from the pMD20-Rs-cps plasmid containing the full-length Rs-cps cDNA using the specific primers RNAi-F and RNAi-R (Table 1) 16 . XhoI and XbaI restriction enzyme sites were added to the 5′ end of the forward primer RNAi-F. At the same time, NcoI and BamHI restriction enzyme sites were added to the 3′ end of the reverse primer RNAi-R. The digested Rs-cps fragments were inserted into the XbaI/BamHI and XhoI/NcoI enzyme sites of the binary vector pFGC5941 at inverted repeat sequences to form a plant RNAi expression vector, pFGC-Rs-cps2, which can generate a hairpin RNAi construct. A similar non-endogenous control vector, pFGC-egfp2, was constructed using the primers eGFP-F/ eGFP-R (Table 1) 29 . Then, the constructed pFGC-Rs-cps2 and pFGC-egfp2 were introduced into the competent cells of Agrobacterium tumefaciens (EHA105) by the freeze-thaw method 56 . The co-cultivation of a N. benthamiana leaf disc was followed using previously described methods 57,58 . The plant transformation was performed as described previously 59 , and the transformants were selected for kanamycin resistance on MS medium. After one Primer name Sequences (5′ to 3′) Primer use Rs-cps-S1 AGTGCCCCTCCGAAATGT cDNA amplification Rs-cps-A1 TGTCCGTTCTTCCGTTCA month of tissue culture, the well-rooted N. benthamiana plants were transferred into sterilized nutritive soil for normal growth and reproduction under the greenhouse conditions described above.

Molecular analysis of putative transformed N. benthamiana plants. Putative transgenic N. benth-
amiana plant lines were identified by PCR, Southern blot hybridization and RT-PCR. Genomic DNA (gDNA) was extracted from the leaves of T0 generation transgenic and wild-type N. benthamiana plants using the cetyltrimethyl ammonium bromide (CTAB) method 60 . Rs-cps transgenic N. benthamiana plants were confirmed by PCR with the specific primers RNAi-F and RNAi-R ( Table 1). The egfp transgenic plants were similarly checked using the specific primers eGFP-F/eGFP-R as described above (Table 1). To confirm integration of the transformed plants, Southern blot hybridization was performed using the DIG High Prime DNA Labeling and Detection Starter Kit I (Roche). The extracted gDNA (15 μg) of each PCR-positive N. benthamiana line was digested with EcoRI (Thermo Scientific) at 37 °C and then transferred to a Hybond N+ membrane (Amersham) 29 . The DIG-labelled probe was prepared using the specific primers Southern-F/Southern-R (Table 1) 16 . Hybridisation and detection were performed as previously described 16,29 . Equal amounts of gDNA from the T0 generation egfp transgenic and wild-type N. benthamiana plants were used as the controls. For the RT-PCR, total RNAs of PCR and Southern-positive Rs-cps transgenic N. benthamiana leaves were extracted and reverse transcribed using the PrimeScript TM 1st Strand cDNA Synthesis Kit (Takara). The PCR reactions were carried out using the specific primers RNAi-F/RNAi-R (  14,61 . There were 5 replicates for each experiment, and each experiment was conducted twice. Representative data from one round of the experiments were presented.

Expression analysis of Rs-cps in R. similis extracted from T2 generation transgenic N. benthamiana plants.
Approximately 100 mixed-stage nematodes feeding on the roots of the wild-type and transgenic N.
benthamiana plants were isolated. Extracted nematodes were washed with DEPC water, frozen immediately in liquid N2 and stored at −80 °C. Total RNA was extracted using the RNeasy Micro Kit (Qiagen) and reverse transcribed as described above. Transcript accumulation of Rs-cps in R. similis was analysed by qPCR using the specific primers qPCR-F/qPCR-R and Actin-F/Actin-R (Table 1) as previously described 28,29 . All the results were obtained from three independent biological replicates. The relative gene expression data were analysed as described above.
Analysis of reproduction and hatching of nematodes extracted from transgenic N. benthamiana plants. The  Statistical analyses. All statistical analyses were conducted using SAS 9.2 (SAS Institute, Cary, NC, USA) in this study. The data collected from the experiments were subjected to one-way analysis of variance (ANOVA) and tested for differences among treatments at the 5% level using Duncan's Multiple Range Test (DMRT).