Expression Analysis of Hairpin RNA Carrying Sugarcane mosaic virus (SCMV) Derived Sequences and Transgenic Resistance Development in a Model Rice Plant

Developing transgenic resistance in monocotyledonous crops against pathogens remains a challenging area of research. Sugarcane mosaic virus (SCMV) is a serious pathogen of many monocotyledonous crops including sugarcane. The objective of present study was to analyze transgenic expression of hairpin RNA (hpRNA), targeting simultaneously CP (Coat Protein) and Hc-Pro (helper component-proteinase) genes of SCMV, in a model rice plant. Conserved nucleotide sequences, exclusive for DAG (Aspartic acid-Alanine-Glycine) and KITC (Lycine-Isoleucine-Threonine-Cysteine) motifs, derived from SCMV CP and Hc-Pro genes, respectively, were fused together and assembled into the hpRNA cassette under maize ubiquitin promoter to form Ubi-hpCP:Hc-Pro construct. The same CP:Hc-Pro sequence was fused with the β-glucuronidase gene (GUS) at the 3′ end under CaMV 35S promoter to develop 35S-GUS:CP:Hc-Pro served as a target reporter gene construct. When delivered into rice callus tissues by particle bombardment, the Ubi-hpCP:Hc-Pro construct induced strong silencing of 35S-GUS:CP:Hc-Pro. Transgenic rice plants, containing Ubi-hpCP:Hc-Pro construct, expressed high level of 21–24 nt small interfering RNAs, which induced specific suppression against GUS:CP:Hc-Pro delivered by particle bombardment and conferred strong resistance to mechanically inoculated SCMV. It is concluded that fusion hpRNA approach is an affordable method for developing resistance against SCMV in model rice plant and it could confer SCMV resistance when transformed into sugarcane.


Introduction
Sugarcane mosaic virus (SCMV), belonging to genus Potyvirus, family Potyviridae, is a serious pathogen of many monocotyledonous crops including sugarcane. In Pakistan, approximately 10-32% losses has been estimated in sugarcane yield which results in 6-10% decline in sugar production [1]. Infected plants showed mosaic pattern and are characterized by typical pinwheel shaped inclusion bodies in cell cytoplasm. The 10 kb single stranded RNA genome of SCMV encodes a single polypeptide which is cleaved either co-or post translationally into ten mature proteins (P1, HC-Pro, P3, 6K1, CI, 6K2, VPg, NIa-Pro, NIb, and CP) [2,3]. C-terminus of the polyprotein encodes the coat protein (CP) which encapsidates the viral genome through helical arrangement of its multiple subunits [4]. Hc-Pro is a multifunctional protein and a proteinase, which is responsible for the multiplication of virus genome, systemic virus movement [5], and suppression of plant RNA silencing machinery [6] by interacting with numerous host factors.
A conserved DAG (Aspartic acid-Alanine-Glycine) motif of CP is responsible for aphid mediated transmission of virus in combination with Hc-Pro which has a conserved KITC (Lycine-Isoleucine-Threonine-Cysteine) motif that binds in aphid stylet [7]. There are two hypotheses about aphid mediated Potyvirus transmission, including the bridge hypothesis and the direct hypothesis. The bridge hypothesis reported that the N-terminal region of Hc-Pro (KITC motif) recognizes an unknown receptor in the aphid stylets. At the same time or subsequently, an Hc-Pro downstream motif, presumably containing the PTK motif, undergoes a specific interaction with the DAG motif on the virus CP, thereby mediating retention of the virions at appropriate sites in the vector [8].
Over the last few decades, considerable efforts have been made to control SCMV using diverse approaches. Pathogen derived resistance through coat protein mediated resistance has been used extensively to develop resistance against different virus groups. It has been reported that sense SCMV CP gene construct transformed into sugarcane showed different level of resistance when challenged with SCMV [9]. Furthermore, multiple sugarcane lines were also generated through the transformation of the coding sequence of CP gene of Sorghum mosaic virus (SrMV) strain H into sugarcane. However, some of these sugarcane transgenic lines displayed the mosaic symptoms after being challenged with SCMV-H in field trials [10].
RNA silencing is an evolutionarily conserved gene regulation mechanism in eukaryotes, which plays a fundamental role in controlling both regulation of endogenous gene expression and defense against invasive nucleic acids such as viruses and transposable elements [11,12]. RNA silencing is induced by double stranded RNA (dsRNA) or hairpin RNA (hpRNA), which is processed into 21-24 nucleotide (nt) small interfering RNA (siRNA) duplex by Dicer or Dicer-like (DCL) protein. One strand of the siRNA duplex is incorporated into the Argonaute protein to form RNA-Induced Silencing Complex (RISC) and guides RISC to single-stranded RNA via sequence complementarity, resulting in Argonaute-mediated cleavage of the target RNA [11,13].
hpRNA-induced silencing has been established to be a powerful tool to developing plant viral resistance through the silencing of viral RNA. hpRNA targeting SrMV CP gene showed approximately 87.5% resistance against SrMV in transgenic sugarcane plants [14]. Evidence about long hpRNA construct targeting multiple genes of Rice black streaked dwarf virus was successfully used to regenerate stable and resistant lines against virus in rice [15]. This paper reports the expression analysis of hpRNA, targeting simultaneously CP and Hc-Pro genes of SCMV, in a model rice plant. The approach was designed to generate SCMV resistance by silencing viral genes that play roles in virus transmission, encapsulation and multiplication, and counter defence against RNA silencing. The validation of the approach not only will lead to the development of SCMV resistant sugarcane but also will provide valuable material to developing resistance in other crops.

Preparation of the Ubi-hpCP:Hc-Pro Construct.
Consensus sequences of CP and Hc-Pro genes of SCMV, 240 bp each, were selected and fused into a chimeric fragment for use as the trigger DNA sequence in the hpRNA cassette. These targeted regions were selected by retrieving the multiple sequences of CP and Hc-Pro genes (about 50 sequences for each gene) from NCBI database and aligned by using ClustalW [16]. A consensus sequence (240 bp) from each gene was selected to form the fusion sequence. This CP:Hc-Pro fusion sequence was synthesized by Gene-Art6 Gene Synthesis (Thermo Fisher Scientific, Waltham, MA) ( Figure S1 in Supplementary Material available online at https://doi.org/10.1155/2017/1646140) and assembled into the hpRNA cassette in both sense and antisense gene orientations. For directional cloning into the pStarling vector, restriction sites of KpnI/SpeI and BamHI/SmaI were incorporated at 5 and 3 ends of fusion fragment, respectively. The expression cassette in pStarling was excised with NotI and inserted into pWBVec8 binary plant expression vector [17] ( Figure 1(a)). The vector was transformed into Agrobacterium tumefaciens AGL1 strain by electroporation method for plant transformation. pWBVec8 binary vector [17] was also transformed as an empty vector control.

Preparation of 35S-GUS: CP-Hc-Pro Fusion Target
Construct. The synthesized CP: Hc-Pro fusion fragment was transcriptionally fused with coding sequence of Escherichia coli -glucuronidase gene (GUS) which was subcloned into pART7 [18] at the BamHI and XhoI sites between the CaMV 35S promoter and the octopine synthase (OCS) terminator. The cassette was then excised with NotI and cloned into pWB-Vec2a to form the binary vector suitable for Agrobacteriummediated transformation (Figure 1(b)).

Rice Transformation.
The seed of Nipponbare variety of rice (Oryza sativa) was selected and cultured on N6D medium for callus induction. Calli were transformed by the Ubi-hpCP:Hc-Pro and empty vector constructs using Agrobacterium-mediated transformation according to Hiei et al. [19].

Particle Bombardment of Rice Calli and Transgenic Rice
Leaves. Nipponbare rice seed were germinated on N6D (Rice callus induction) medium [19] for 4-6 weeks. Healthy looking rice calli or leaves of transgenic rice were selected for particle bombardment. Plasmid DNA (1 g) was mixed with gold particles in 0.6 in diameter, and 2.5 M CaCl 2 (16 L) was added, which aids the binding of negatively charged DNA molecules to gold particles. After brief vortexing, 0.1 M spermidine solution (6.4 L) was added into the mixture for protection from endonucleases. These DNA-coated gold particles were bombarded into rice calli or leaves using PDS 1000-He Biolistic particle delivery system at 1100 Psi pressure and 26 inch Hg vacuum at 6 cm optimized target distance. GUS expression was analyzed using histochemical staining 2-3 days after bombardment.

RNA Extraction and Northern Blot Analysis.
Total RNA was extracted from leaf of transgenic rice plants using TRIzol5 Reagent (Thermo Fisher Scientific) following the manufacturer's recommendation.
For Northern blot analysis of target gene expression, total RNA (15 g) was separated in 1.3% formaldehydeagarose gel and blotted onto HyBond-N nylon membrane (GE Healthcare Amersham, Rydalmere, NSW, Australia). 32 P-labelled antisense CP:Hc-Pro RNA probe was prepared by linearizing the CP:Hc-Pro plasmid in the pENTR6/D-TOPO5 vector with SpeI, followed by in vitro transcription using Sp6 RNA polymerase in the presence of 32 P-UTP. Hybridization was performed as previously described [21].
For small interfering RNA (siRNA) detection, approximately 30 g total RNA was separated in 17% denaturing polyacrylamide gel. RNA was electroblotted to HyBond-N + nylon membrane (GE Healthcare Amersham) using 0.5x TBE buffer. After UV cross-linking, the membrane was hybridized at 42 ∘ C with 32 P labeled antisense CP:Hc-Pro RNA probe [22]. Washing of hybridized membrane was carried out for three times at 42 ∘ C with 2x SSC containing 0.2% SDS. Hybridized membranes were visualized and recorded with phosphorimager (FLA-5000, Fuji Photo Film, Tokyo, Japan) after overnight exposure. The intensity of the 21-24 nt siRNA bands and the corresponding U6 RNA bands was determined three times against the background of the gel blots using the Multi Gauge Version 3.0 software (FUJIFILM). The average siRNA band intensity was then normalized against the average U6 RNA band intensity to generate the quantification data.

DNA Extraction, PCR Detection, and Southern Blot
Analysis. Genomic DNA of rice plants was extracted from leaf tissue using CTAB protocol according to [23]. Forward (5 -ATCTCACCGACTACAGCTTAG-3 ) and reverse (5 -GGTGTTACGTGTTTTTCATATGC-3 ) primers were used for PCR detection of the Ubi-hpRNA transgene using the following conditions: initial denaturation at 95 ∘ C for 5 min, followed by 35 cycles of 95 ∘ C for 30 s, 54 ∘ C for 30 s, and extension at 72 ∘ C for 1 min, with a final extension at 72 ∘ C for 5 min.
For Southern blot analysis of transgene copy number, approximately 15 g of RNase A-treated rice genomic DNA was digested with SpeI and separated in 1% agarose gel along with 1 kb plus DNA marker (Gene Ruler, Thermo Fisher Scientific, Walthan, MA) in 1x TBE buffer and transferred onto the Nylon-N + membrane (GE Healthcare Amersham). Probe was prepared by excising the CP:Hc-Pro fragment from the TOPO vector with SpeI and XhoI followed by gel purification. Purified fragment was labeled with 32 P radioactive dCTP using Deca label6 DNA labeling kit (Thermo Fisher Scientific) and hybridized with membrane at 65 ∘ C in hybridization buffer as previously described [24]. After hybridization, membranes were washed in 1x SSC for 20 min first at room temperature, followed by washes at 50 ∘ C and 60-65 ∘ C. Blots were visualized and recorded by FLA-5000 phosphorimager (Fuji Photo Film) after overnight exposure.

Assay of Transgenic Rice Plants for SCMV Resistance.
Transgenic rice plants at 5-10 leaf stage were inoculated with SCMV-strain A by rubbing carborundum-dusted leaves with extract of SCMV-infected sugarcane leaves in 0.1 M sodium phosphate buffer (pH 7.0). Infected leaves were harvested at 25 days after inoculation (DPI), and total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific). The presence of SCMV was analyzed using RT-qPCR using   (Figure 1(a)). The chimera sequence was fused transcriptionally with GUS reporter gene and subcloned into pART7 [18] between CaMV 35S promoter and the octopine synthase (OCS) terminator, to develop the 35S-GUS:CP:Hc-Pro target construct for detecting the hpRNA silencing efficiency (Figure 1(b)). It is worth noting that cobombardment with the Ubi-hpCP:Hc-Pro construct also reduced the number of blue spots from the expression of the 35S-GUS:Y-Sat construct, but at a lesser degree from an average of 27 per square millimetre to 12. It is unclear why this reduction occurred but it could be due to the sharing of the OCS terminator sequence between the hpRNA and the 35S-GUS:Y-Sat constructs that could induce transgene cosuppression.

Transgenic Rice Plants Containing Ubi-hpCP:Hc-Pro
Accumulate siRNAs. The rice plant, a model monocotyledonous species, was used as a surrogate of sugarcane because of its transformation efficiency, compared to the other cereal crops, such as wheat, maize, and barley and the conservation of gene sequences [26][27][28][29][30]. The Ubi-hpCP:Hc-Pro and the empty vector control (VC) constructs were transformed into rice using Agrobacterium-mediated methods, generating the six independent transgenic lines, respectively. One empty vector control and five Ubi-hpCP:Hc-Pro primary transgenic lines (T 0 ) were analyzed for transgene copy number by Southern blot hybridization and for the accumulation of siRNAs by Northern blot hybridization. Empty vector control line showed no CP:Hc-Pro hybridization signals on Southern blot while lines #2, #3, and #4 showed a hybridizing band indicating that each of these lines may contain a single copy of transgene. The size of the hybridizing band for lines #2, #3, and #4 appears to be the same (∼12 kb), which seems to indicate that they came from a single transgenic event. This is despite the fact that plants were derived from different callus lines. Lines #1 and #5 contained multiple copies as indicated by the multiple hybridizing bands on the Southern blot ( Figure 3). Small interfering RNAs (siRNAs) 21-24 nts in size, corresponding to the CP:Hc-Pro sequence of the Ubi-hpCP:Hc-Pro, were readily detected by Northern blot analysis in all the five independent lines (Figure 4(a)), indicating that the Ubi-hpCP:Hc-Pro was expressed and the resulting hpRNA processed by Dicer (DCLs) enzyme which recognizes and cleaves the dsRNA into siRNAs. The size distribution of the siRNA was consistent with that of hpRNA-derived siRNAs in Arabidopsis (Arabidopsis thaliana), consisting of 21, 22, and 24 nt size classes [31], (although the 21 and 22 nt species were not clearly separated) (Figure 4(a)). This indicates that the CP:Hc-Pro hpRNA was processed in transgenic rice by DCL4, DCL2, and DCL3, respectively, corresponding to the pattern observed in Arabidopsis [32]. In a previous report, DCL4 dependent 21 nt siRNAs were the predominant species of siRNAs derived from the SCMV [33]. Moreover, DCL2 and DCL4 function redundantly in maize and Arabidopsis plants to produce higher level of 21 nt and 22 nt siRNAs [33,34]. It is noteworthy that in the different transgenic lines the siRNA hybridization signals showed different intensity, indicating a variable level of siRNA in these lines (Figure 4(b)). Interestingly, the high-copy-number transgenic line #5 did not accumulate more siRNAs; instead, it appears to generate slightly less siRNAs, suggesting that the multiple copy transgene insertion may have resulted in transcription repression of the hpRNA transgene [35].

The Ubi-hpCP:Hc-Pro Transgene Confers Silencing to a
Biolistically Delivered Target Gene. The efficacy of Ubi-hpCP:Hc-Pro transgene against the target viral sequences was shown in a transient expression assay by delivering the 35S-GUS:CP:Hc-Pro target construct into transgenic rice leaves using particle bombardment. Strong GUS expression was observed in leaf tissue of the empty vector control transgene but not in the leaves of the Ubi-hpCP:Hc-Pro transgenic plants (three replicates of all transgenic lines), indicating that siRNAs derived from the Ubi-hpCP:Hc-Pro induced silencing against the 35S-GUS:CP:Hc-Pro target construct by targeting CP:Hc-Pro fusion sequence ( Figure S2).

Ubi-hpCP:Hc-Pro Transgene Stably Inherited in T 1 Generation and Confers Resistance against SCMV.
To investigate if the Ubi-hpCP:Hc-Pro transgenic plants were resistant to SCMV infection, seeds were collected from lines #2, #3, #4, and #5 primary transgenic plants (line #1 did not produce seeds) from which T 1 plants were established and assayed for transgene expression and SCMV resistance. The Ubi-hpCP:Hc-Pro transgene expression, as indicated by the hybridizing signals of the unprocessed CP:Hc-Pro hpRNA on the Northern blot, was clearly detected in the T 1 progeny of the single-copy lines #2, #3, and #4, indicating stable inheritance from the primary transgenic plants (Figure 5(a)). However, for line #5, the highest-copy-number transgenic line, three of the four T 1 plants analyzed showed no Ubi-hpCP:Hc-Pro signal, which may indicate the transcriptional suppression of the Ubi-hpCP:Hc-Pro transgene in these individuals of high copy number transgene insertion. It is also noticeable that plant #5 from line 3 also showed no Ubi-hpCP:Hc-Pro hybridization signals ( Figure 5(a)).
To examine if the Ubi-hpCP:Hc-Pro transgene could confer resistance to SCMV infection, a number of T 1 plants of line #3, shown to express high levels of Ubi-hpRNA in the Northern blot ( Figure 5(a)), together with T 1 plants of an empty vector control line, were inoculated with SCMV. The 12 T 1 Ubi-hpCP:Hc-Pro plants showed visibly higher growth vigor than the 12 empty vector control plants at 6 weeks after inoculation, implying resistance to SCMV (strain A, a mild strain of SCMV) infection ( Figure 5(b)). SCMV strain A infection studies typically showed stunted growth with mild mosaic pattern without necrosis in its natural hosts that are sugarcane, maize, and sorghum [36]. In the present study, no clear mosaic symptoms were observed in the SCMV-inoculated plants possibly because rice is not the natural host of SCMV and very mild mosaic symptoms were reported through artificial inoculation of SCMV [36]. To confirm the successful SCMV resistance, RNA was extracted from a subset of the inoculated plants and analyzed for the presence of SCMV using RT-qPCR with CP and Hc-Pro specific primers. The empty vector control plants showed high levels of SCMV CP and Hc-Pro RNA ( Figure 5(c)). In contrast, no significant amount of SCMV RNA was detectable in all five T 1 transgenic plants analyzed, indicating that the Ubi-hpCP:Hc-Pro transgenic plants (from line #3) were highly resistant to SCMV infection. Both the CP and Hc-Pro target sequences were similarly silenced in the SCMVinfected Ubi-hpCP:Hc-Pro transgenic plants ( Figure 5(c)), indicating that the transgene was effective against both genes.

Discussion
SCMV is an agronomically important Potyvirus infecting a variety of monocotyledonous species including important crop plants such as maize and sugarcane. It is transmitted in nature by five different insect vectors: Aphis craccivora, Aphis gossypii, Macrosiphum euphorbiae, Rhopalosiphum maidis, and Rhopalosiphum padi [37]. Outbreak of this virus often causes severe damage to crops leading to high cost to crop yields in more than 70 countries [38]. According to the survey conducted between 2008 and 2010 in Pakistan, the highest incidence of virus infection in sugarcane fields reached up to 38% in the Punjab province of Pakistan during the year 2010 [39].
Because of the potential damage SCMV could do to the agronomically important monocotyledonous crops, many efforts have been made to control the spread and multiplication of this virus. It was reported that protection against severe SCMV strains was achieved in a number of different crops by inoculating plants with mild virus strain through the mechanism of viral cross protection [40,41]. Attempts have also been made in the selection and breeding for naturally occurring resistant crop varieties for controlling SCMV. For example, a European maize inbred line was found to be completely resistant to SCMV and Maize dwarf mosaic virus in green house and field tests [42]. As a number of plant  viruses such closteroviruses, comoviruses, tymoviruses, and nepoviruses utilize cysteine proteinases in their polyprotein processing, a cysteine proteinase inhibitor, cystatin, has been transgenically expressed in tobacco to control the multiplication of these viruses [43], which could be potentially used to control SCMV. However, an effective method for controlling SCMV and SCMV-caused diseases has so far not been found. RNA silencing-based approaches, especially the long hpRNA technology, has been successfully used to engineer plant resistance against viruses, including agronomically important potyviruses, such as Wheat streak mosaic virus, Potato virus Y, and Plum pox virus [44,45]. Various regions of the potyviral genome, such as those encoding the NIa protease and Hc-Pro, have been selected as the target sequences for the hpRNA transgenes in such experiments. For the SCMV, attempts were also made to transform maize with an hpRNA construct targeting the CP gene, generating transgenic plants with variable levels of SCMV resistance [46]. Similarly, SrMV resistant sugarcane transgenic lines were generated using hpRNA transgene targeting CP region showing different levels of virus resistance [47]. Thus, hpRNA transgenes provide a potential opportunity for developing SCMV resistance in important crops such as sugarcane.
While targeting a single viral gene by hpRNA transgenes has proved to be successful, we chose a strategy of simultaneously targeting two essential SCMV genes, CP and Hc-Pro. While CP is essential for viral encapsidation and transmission, Hc-Pro has multiple functions including the function of countering RNA silencing. It was anticipated that simultaneous silencing of the CP and the Hc-Pro genes would both cause direct downregulation of viral RNAs and enhance host antiviral silencing activity, therefore resulting in robust virus resistance in the transgenic plants. We selected 240 bp segments from each of the CP and Hc-Pro genes that correspond to conserved functional motifs, with the purpose of achieving broad resistance against the various SCMV virus isolates. In addition, the maize ubiquitin promoter was chosen to drive the expression of the Ubi-hpCP:Hc-Pro transgene, which has been shown to be suitable for high levels of transgene expression in rice and other monocot species [48] . Finally, in the Ubi-hpCP:Hc-Pro construct the sense and antisense CP:Hc-Pro sequences are separated by a spliceable intron, which is suggested to enhance the efficacy of hpRNA transgenes [49].
The efficacy of the Ubi-hpCP:Hc-Pro transgene was first demonstrated by showing that it could induce efficient silencing against the 35S-GUS:CP:Hc-Pro fusion transgene when transiently coexpressed in rice callus tissue using particle bombardment. When stably transformed into rice plants, this Ubi-hpCP:Hc-Pro transgene gave rise to siRNAs across the independent transgenic lines, which conferred silencing against the biolistically delivered 35S-GUS:CP:Hc-Pro fusion transgene. The size distribution of the siRNAs is consistent with the typical pattern of hpRNA-derived siRNAs [31], with the 21-22 nt siRNAs being more abundant than the 24 nt species. Interestingly, the abundance of siRNAs showed no positive correlation with the copy number of Ubi-hpCP:Hc-Pro transgene insertion, and it appeared that the higher the copy number is, the less the siRNA accumulates.
Analysis of Ubi-hpCP:Hc-Pro transcript accumulation in the subsequent T 1 population further suggested that the Ubi-hpCP:Hc-Pro transgene is subject to transcriptional silencing when the number of transgene insertions is very high, which is consistent with previous reports [50]. It was previously reported that high copy number or polyploid progenitor line undergoes partial or complete transgene silencing which is caused by transacting silencing of active and inactive epialleles [51]. This suggests that for achieving stable virusresistance, transgenic plants with a simple transgene insertion pattern should be selected.
Virus infection assays of the single-copy transgenic line expressing high levels of the Ubi-hpCP:Hc-Pro showed that the transgenic plants are highly resistant to SCMV, which displayed a more vigorous growth than vector control transgenic plants, with almost no detectable SCMV RNA in contrast to SCMV RNA in the vector control plants.
In conclusion, we demonstrate here that high-level SCMV resistance can be achieved using an hpRNA transgene targeting two essential viral genes at the same time. While this strategy has not been tested in sugarcane it is anticipated that the Ubi-hpCP:Hc-Pro construct simultaneously targeting the downregulation of both CP and Hc-Pro expression in SCMV will confer SCMV resistance when transformed into sugarcane.