This study used the full-length cDNA construct of CWMV RNA1 (pCB-35S-R1) described previously (Yang et al., 2016) together with modified constructs of RNA2 (ᐃRNA2). The earlier RNA2 construct (pCB-35S-R2) was modified to abolish the translation of the N-terminal extension of CP, to delete the entire RT domain of the CP and to insert a multiple cloning site (MCS) by PCR-based site-directed mutagenesis. This modified construct was designated it as pCB-35S-R3 (Figure 1A). The two constructs pCB-35S-R1 and pCB-35S-R3 were individually transformed into Agrobacterium tumefaciens strain GV3101 by electroporation and used to infiltrate leaves of N. benthamiana. By 54 days, the upper leaves of plants inoculated with ᐃCWMV including pCB-35S-R1 and pCB-35S-R3 developed numerous chlorotic local lesions similar to the plants inoculated with wild-type CWMV (Figures 1B,C). Following the method described using BSMV (Yuan et al., 2011), the infiltrated N. benthamiana leaf sap was then used to inoculate wheat. At 40–42 dpi, the newly-emerged leaves of inoculated plants consistently developed typical mosaic symptoms similar to those inoculated with sap containing wild-type virus (Figure 1D). Northern blot analysis confirmed that a shorter form of RNA2 could be detected in the upper leaves of both wheat and N. benthamiana when the modified construct had been used (Figures 1E,F). The results demonstrate successful and typical infection of both N. benthamiana and wheat using the modified RNA construct.

Nicotiana benthamiana is the most widely used experimental host plant for VIGS because of its susceptibility to many plant viruses and the rapidity with which visible phenotypes appear (Goodin et al., 2008). To assess silencing by Agrobacterium mediated CWMV VIGS, partial PDS gene sequences derived from N. benthamiana (300 bp; nt 200–500 of EU165355) and T. aestivum (300 bp; nt 1025–1325 of FJ517553) were inserted into pCB-35S-R3 in the sense orientation to generate pCB-35S-R3:NbPDS and pCB-35S-R3:TaPDS, respectively (Figure 2A). These were co-infiltrated with the RNA1 construct to four-leaf stage N. benthamiana plants which were then maintained at 17°C. The ᐃCWMV:NbPDS (contains pCB-35S-R1 and pCB-35S-R3:NbPDS) caused systemic infection and induced visible photobleaching in the non-inoculated leaves of N. benthamiana at 27 dpi (Figure 2B). Photobleaching could still be observed on the fresh flush of N. benthamiana leaves at 65 dpi (Figure 2D). Although it is impossible to agroinfect cereals directly, N. benthamiana infiltrated leaves provide an excellent source for rub-inoculation to wheat. Leaf extracts from the ᐃCWMV:TaPDS (contains pCB-35S-R1 and pCB-35S-R3:TaPDS) and BSMV:TaPDS treatments caused systemic infection and symptoms when rub-inoculated to wheat. After 30 days, the PDS gene silencing phenotype was also observed in wheat (Figure 2C). Real-time RT-qPCR then showed that the mRNA levels of TaPDS and NbPDS were significantly less in the ᐃCWMV:TaPDS and ᐃCWMV:NbPDS treatments than in controls infected with the empty vector (ᐃCWMV:00) (Figure 2E). Thus, the modified CWMV vectors carrying NbPDS or TaPDS were effective in the silencing endogenous PDS in N. benthamiana and wheat.

To determine the effects of fragment lengths on silencing efficiency, we assessed the phenotypes induced by 500, 800, 1000, and 1500 bp NbPDS and TaPDS fragments in N. benthamiana and wheat, respectively. Based on their photobleaching, 500 bp NbPDS inserts appeared to be the most efficient, and the 1500 bp NbPDS inserts were the least efficient after 27 dpi. It was then demonstrated by qPCR that the NbPDS RNA levels were most strongly reduced (to less than 15%) by the 500 bp NbPDS insertion (Figures 3A,C). Interestingly, the same result also was observed when PDS was VIGS in the wheat after 42 dpi (Figures 3B,D). For VIGS analyses, it is important to maximize foreign fragment stability in the infiltrated leaves and to minimize loss of inserts during virus movement. To evaluate the stability of inserts in infiltrated and systemically infected leaves, N. benthamiana and wheat leaves were infiltrated with PDS⁵⁰⁰, PDS⁸⁰⁰, PDS¹⁰⁰⁰, or PDS¹⁵⁰⁰ and samples were collected at 7 dpi from infiltrated leaves, and at 14, 21, 28, 35, 42, 49, 56, and 63 dpi from systemically infected N. benthamiana and wheat leaves, respectively. These were then tested by RT-PCR using primer pairs P14F/P14R covering the insertion site. The inserted sequence was relatively stable in the infiltrated leaves up to 63 dpi in most treatments, but distinct degradation products were observed in the PDS¹⁵⁰⁰ treatment at all time points both in N. benthamiana and wheat (Figures 4A,B and Supplementary Figure S1).

To test whether the CWMV VIGS vector can also be used to inhibit miRNA activity, miR165/166 was cloned into pCB-35S-R3 to generate pCB-35S-R3: STTMmiR165/166 using the STTM strategy as previously described (Yan et al., 2012; Figure 5A). N. benthamiana plants infiltrated with ᐃCWMV:STTMmiR165/166 (contains pCB-35S-R1 and pCB-35S-R3: STTMmiR165/166) developed typical CWMV symptoms on their upper non-inoculated leaves (Figures 5B,C). Stem-loop RT-PCR together with real time PCR assays confirmed that the STTMmiR165/166 sequences were expressed in these plants (Figure 5D), that the transcript level of mature miR165/166 was reduced and that the expression of the Target gene Class III HOMEODOMAIN-LEUCINE ZIPPER transcriptional factors (HD-ZIPIII) mRNA was increased relative to ᐃCWMV:00 controls (Figure 5E). To determine whether the CWMV VIGS vector also can suppress the expression of miRNAs in wheat, miR3134 a novel miRNA of wheat was cloned into pCB-35S-R3 to obtain pCB-35S-R3:STTMmiR3134 using the STTM strategy (Figure 6A). After confirming that STTMmiR3134 sequences were expressed in the inoculated N. benthamiana leaves, wheat plants were rubbed with infiltrated N. benthamiana leaf sap. Wheat plants infiltrated with bbbCWMV:STTMmiR3134 developed etiolation symptoms on the fourth leaves at 20 dpi (Figure 6B). At 14 dpi, stem-loop RT-qPCR assays showed a decline in the relative transcript level of mature miR3134 in ᐃCWMV:STTMmiR3134-infected wheat (Figure 6C). RT-qPCR results also showed that the mRNA levels of the candidate target gene of miR3134, Genbank accession AK335430 (Jiao et al., 2015) were obviously increased compared to those in ᐃCWMV:00-infected wheat (Figure 6D).

Virus-induced gene silencing vector CWMV- was compared with TRV- for inducing gene silencing in N. benthamiana and BSMV and FoMV in wheat. 500 bp NbPDS was selected and inserted into TRV and CWMV-based VIGS vector to construct TRV:NbPDS⁵⁰⁰ and CWMV:NbPDS⁵⁰⁰, respectively. After 27 dpi, the visible photobleaching could be observed on the fresh flush of N. benthamiana leaves induced by ᐃCWMV:NbPDS⁵⁰⁰ but not any white streaks appeared in the plants rubbed with TRV:NbPDS⁵⁰⁰ at 17°C (Figure 7A). RT-qPCR also showed that the amount of NbPDS mRNA was significantly less in the plants infected with ᐃCWMV:NbPDS⁵⁰⁰ than plants infected with TRV:NbPDS⁵⁰⁰ (Figures 7B,C). Additionally, 500 bp TaPDS was selected and inserted into BSMV-, FoMV-, and CWMV-based VIGS vector to generate BSMV:TaPDS⁵⁰⁰, FoMV:TaPDS⁵⁰⁰ and CWMV:TaPDS⁵⁰⁰, respectively. Two-leaf-stage wheat plants were rubbed with saps of N. benthamiana leaves agroinfiltrated with BSMV:TaPDS⁵⁰⁰, FoMV:TaPDS⁵⁰⁰ or ᐃCWMV:TaPDS⁵⁰⁰, respectively. And these inoculated plants were maintained at 17°C as well. All wheat plants infected with ᐃCWMV:TaPDS⁵⁰⁰ appeared the bleached patches and streaks or albinism fully expanded in the fourth leaf at 42 dpi (Figure 7D). However, neither BSMV:TaPDS⁵⁰⁰ or FoMV:TaPDS⁵⁰⁰ inoculated plants showed any silencing phenotype (Figure 7D). To confirm the silencing of TaPDS genes, RT-qPCR was performed. The mRNA levels of TaPDS in the wheat rubbed with sap of ᐃCWMV:TaPDS⁵⁰⁰ were strikingly reduced compared to those of BSMV:TaPDS⁵⁰⁰ or FoMV:TaPDS⁵⁰⁰-infected plants (Figure 7E). Taken together, it clearly suggests that CWMV VIGS vector can induce more efficient gene silencing both in the N. benthamiana and wheat at 17°C.

Nicotiana benthamiana plants were grown in a cabinet at 17°C, with 16 h light/8 h dark and 70% r.h. Wheat (Yangmai 158) plants used for VIGS experiments were grown in a greenhouse until the four leaf-stage, then inoculated with CWMV and transferred to a climate chamber at 17°C until evaluation.

pCB-35S-R1 and pCB-35S-R2 are full-length cDNA clones of RNA1 and RNA2 of CWMV developed and stored in our lab. Escherichia coli DH5α was used for the construction of all plasmids and all plasmids constructed in this study were verified by sequencing. The primers used in this paper are detailed in Supplementary Table S1.

pCB-35S-R3 was constructed as follows. The mutant construct pCB-35S-R2:M5 which abolished the translation of the N-terminal extension of CP and the RT domain of the CP was developed and stored in our lab and was selected as template. The primer pair P1F/P1R was used to amplify the segment S1 from pCB-35S-R2:M5. This 550 bp PCR product was digested with Spe I/Sal I and ligated into the similarly digested plasmid pCB-35S-R2:M5 to obtain pCB-35S-R3 containing a MCS.

Chinese wheat mosaic virus gene silencing constructs were prepared as follows. The N. benthamiana 300, 500, 800, 1000, and 1500 bp PDS fragments (NbPDS; Genbank accession: EU165355) were obtained by RT-PCR using the P2F/P2R, P3F/P3R, P4F/P4R, P5F/P5R, and P6F/P6R primer pairs (Supplementary Table S1), respectively. The resulting fragments were integrated into pCB-35S-R3 in the sense orientation to generate pCB-35S-R3:NbPDS³⁰⁰, pCB-35S-R3:NbPDS⁵⁰⁰, pCB-35S-R3:NbPDS⁸⁰⁰, pCB-35S-R3:NbPDS¹⁰⁰⁰, and pCB-35S-R3:NbPDS¹⁵⁰⁰, respectively. In a similar fashion, 300 bp Triticum aestivum PDS fragments (TaPDS; Genbank accession: FJ517553) were amplified by RT-PCR with P7F/P7R primer pairs (Supplementary Table S1) and inserted into pCB-35S-R3.

STTM165/166 (for silencing of miR165/166 using the STTM strategy) was constructed as follows. A primer with restriction enzyme cutting site, corresponding to the target mimic of miR165/166 and 48 nt STTM spacer (5′-GTTGTTGTTGTTATGGTCTAATTTAAATATGGTCTAAAGAAGAAGAAT-3′) were used to amplify STTM165/166by PCR using primers P8F/P8R (Figure 5A). The resulting fragments were integrated into pCB-35S-R3 to generatepCB-35S-R3:STTM165/166. STTM3134a (for silencing of miR3134a) was constructed by the same method using primers P9F/P9R (Figure 6A).

Northern blot assays were performed as previously described (Yang et al., 2016). Total RNAs were extracted from leaf tissues using TRIzol reagent (Invitrogen) and treated with DNase (TaKaRa). Three micrograms of total RNAs were separated on a denaturing 2% formaldehyde agarose gel and transferred to Hybond-N⁺ membranes (Amersham Bioscience) using 20× SSC buffer. The RNAs were cross-linked to membrane matrix by UV for 50 s. Northern blotting for assays of CWMV genomic RNAs was carried out using the DIG High Prime DNA Labeling and Detection Starter Kit II (Roche). The DNA oligo nucleotides complementary to the 3′ terminus of the CWMV genome were labeled with digoxigenin (DIG) at their 3′ ends using the DIG Oligo nucleotide Tailing Kit (Roche) and then purified using a G25 Sephadex column (GE). Membranes were pre-hybridized for 2 h and hybridized overnight at 42°C using the DIG Luminescent Detection Starter kit for nucleic acids (Roche). The hybridization signals were visualized by Amersham Imager 600 (GE). All these procedures were performed according to the manufacturers’ instructions.

The pCB-35S-R3 derivatives were transformed into A. tumefaciens strain GV3101. For agroinfiltration, single colonies were cultured overnight at 28°C with constant shaking in 10 ml of LB containing kanamycin (100 μg/ml) and rifampicin (25 μg/ml). Then, 0.1 ml aliquots were used to inoculate 10 ml LB containing the same antibiotics and grown at 28°C for 10 h. Bacterial cells were pelleted at 5000 rpm for 5 min, resuspended in infiltration buffer containing 10 mM 2-(N-morpholino) ethanesulfonic acid (MES), pH 5.2, 10 mM MgCl₂, and 0.1 mM acetosyringone to 0.6–0.7 OD₆₀₀ and incubated at room temperature for 2–3 h. For agroinfiltration, equal amounts of bacteria harboring pCB-35S-R1 and pCB-35S-R3 (or its derivatives) were mixed, and infiltrated into four to six N. benthamiana leaves immediately above the cotyledons with a 1 ml needleless syringe. After maintenance in a growth chamber for 7–10 days post infiltration (dpi), the infiltrated leaves were harvested, ground in 20 mM Na-phosphate buffer (pH 7.2) containing 1% celite, and the sap was mechanically inoculated onto wheat seedlings at the two-leaf stage.

Total RNAs were extracted from three independent biological replicates of N. benthamiana plants, uninoculated wheat plants, CWMV-infected N. benthamiana plants, and CWMV-infected wheat plants with TRIzol reagents as described by the manufacturer (Invitrogen). Real-time quantification of microRNAs was performed by stem-loop RT-qPCR as previously reported (Salone and Rederstorff, 2015). Briefly, 2 μg purified total RNA were reverse transcribed using miRNA-specific stem-loop primers (50 nM) in a 20-μl reaction volume with the Takara RNA PCR Kit to generate cDNA following the manufacturer’s instructions. The expression patterns of the miRNAs were then analyzed by real-time quantitative PCR (qPCR) with a standard SYBR Green I PCR kit (Invitrogen) protocol. Each qPCR reaction contained 10 μl of 2 × SYBR PCR master mix, 0.8 μl of forward primer (10 μM), 0.8 μl of reverse primer (10 μM), 0.5 μl of cDNA, and sterilized ddH₂O added up to 20 μl. The reactions were incubated in a 96-well plate at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Then, the samples were heated from 60 to 95°C to acquire the denaturing curve of the amplified products. All reactions were performed in triplicate on a 7900 Real Time PCR System (Applied Biosystems). Relative expression was calculated using the comparative cycle threshold method (Livak and Schmittgen, 2001; Schmittgen and Livak, 2008) and values for each miRNA were normalized to the expression levels of U6. All the primers used are listed in Supplementary Table S1.

The expression of the Target gene was also analyzed by qPCR. The total RNA was first treated with DNase I (Takara) and then reverse transcribed to generate cDNA using an oligo (dT) primer and a Primescrip RT reagent kit (Takara) according to the manufacturer’s instructions. The primer pair P14F/P14R (see Supplementary Table S1) was then added to amplify the PCR products. The qPCR was performed in triplicate on a 7900 Real Time PCR System (Applied Biosystems) with a standard SYBR Green I PCR kit (Invitrogen) protocol. In each reaction, actin gene for which stabilized expression after virus infection was used as the internal reference (Supplementary Figure S3). The reactions were incubated in a 96-well plate at 95°C for 3 min, followed by 40 cycles of 95°C for 15 s, 60°C for 20 s, and 72°C for 30 s. Each experiment was replicated three times. The relative expression levels were calculated using the 2^-ΔΔCT method (Livak and Schmittgen, 2001). The values of threshold cycle (CT) were calculated by Rotor-Gene 6 software (Corbett Robotics, Australia) and the 2^-ΔΔCT method was used to assess the relative changes in gene expression (Nobuta et al., 2010).