Amplicon-based RNAi construct targeting beta-C1 gene gives enhanced resistance against cotton leaf curl disease

Abstract

Cotton leaf curl disease (CLCuD) is one of the major limiting factors affecting cotton production in Pakistan for the last three decades. The disease is caused by begomoviruses of the family Geminiviridae. RNA interference (RNAi) is a promising tool that has been proved effective against several pathogens. Using RNAi, different genomic regions of geminiviruses have been targeted to attain sustainable resistance. However, the silencing of the transgene upon virus infection is a limiting factor. Here, we have developed for the first time an amplicon-based RNAi construct to target βC1 gene of betasatellite associated with cotton leaf curl begomoviruses. In addition to producing short interfering (si) RNAs, Rep-based activation or looping out of the construct induced upon virus infection produces multiple copies of transgene that results in accumulation of defective molecules of betasatellite. Subsequent transcription gives rise to increased number of siRNAs that gives enhanced resistance. Transgenic Nicotiana benthamiana plants having RCβ (RNAi construct for betasatellite) were challenged against Cotton leaf curl Khokran virus (CLCuKV) and Cotton leaf curl Multan betasatellite (CLCuMB). Reduced titer of the virus and betasatellite were detected through Southern blot hybridization. Significance of the study has been discussed.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-021-02816-6.

Introduction

CLCuD is one of the devastating diseases of cotton in the Indian sub-continent (Sattar et al. 2017). The disease is caused by begomoviruses of the family Geminiviridae, which is characterized by the plant viruses of single-stranded (ss) DNA genome encapsidated in twinned, quasi-isometric particles (Brown et al. 2012). Family Geminiviridae is divided into nine genera Becurtovirus, Begomovirus, Capulavirus, Curtovirus, Mastrevirus, Eragrovirus, Grablovirus, Topocuvirus and Turncurtovirus (Zerbini et al. 2017; Varsani et al. 2017). Begomoviruses constitute the largest genus of the family having viruses of small, circular genome of up to 2.8 kb which are transmitted by whitefly (Bemisia tabaci). On the basis of genomic components, there are two types of begomoviruses, i.e. monopartite and bipartite. Bipartite begomoviruses comprise two genomic components (DNA A and DNA B), while monopartite begomoviruses have a single genomic component equivalent to DNA A of bipartite begomoviruses. The begomoviruses causing CLCuD are monopartite with associated satellite molecules, i.e. betasatellite and alphasatellite (Zhou 2013a, b).

Betasatellites (genus Betasatellite, family Tolecusatellitidae) are small, circular DNA molecules having genome size half of that of begomovirus i.e. 1.4 kb (Virus Taxonomy: 2019 Release [https://talk.ictvonline.org/taxonomy/). They always require a helper virus for their replication, movement in and transmission between plants (Mansoor et al. 2003; Briddon et al. 2003). More than 1300 full-length β-satellite sequences belonging to 61 species of the genus Betasatellite have been characterized from various host plants in 37 countries (Silva et al. 2017; Yang et al. 2019). Betasatellites encode a single conserved (in both position and sequence) open reading frame (ORF) referred to as βC1 that encodes a protein of 118 amino acids performing all functions of betasatellite including pathogenicity determinant (Qazi et al. 2007), suppression of gene silencing (Amin et al. 2011; Zhou, 2013a, b), interacts with PsbP to subvert antiviral defense (Gnanasekaran et al. 2019), and interacts with snRK1 (Kamal et al. 2019). Other genomic regions include satellite conserved region (SCR) and a region rich in adenine nucleotide (known as A-rich). The SCR contains a predicted stem-loop structure containing the nonanucleotide sequence TAATATTAC which, for geminiviruses, is the origin of virion-strand DNA replication (Hanley-Bowdoin et al. 1999). During first and second epidemic of CLCuD in Punjab, Sindh and northwestern India, several begomoviruses have been found associated with CLCuD including Cotton leaf curl Khokran virus (CLCuKV), Cotton leaf curl Multan virus (CLCuMuV), Cotton leaf curl Alabad virus (CLCuAV), Papaya leaf curl virus (PaLCuV) (Zhou et al. 1998; Kirthi et al. 2004), Cotton leaf curl Burewala virus (CLCuBuV; now known as Cotton leaf curl Khokran virus Burewala strain [CLCuKV-Bur]), Cotton leaf curl Gezira virus (CLCuGV) and Cotton leaf curl Shahdadpur virus (CLCuShV) (Amrao et al. 2010a, b; Tahir et al. 2011; Rajagopalan et al. 2017). Though with small changes in SCR, a single betasatellite is associated with all these begomoviruses, that is Cotton leaf curl Multan betasatellite (CLCuMuB) (Mansoor et al. 2003; Akhtar et al. 2014).

There are several approaches to counter the effect of these viruses on crop plants. Most of these approaches use either protein-based or RNA-based resistance mechanisms. Most recently, CRISPR/Cas9-based genome editing is being widely used against various pathogens (Cao et al. 2021). RNAi is a homology-dependent mechanism of gene silencing in which small, double-stranded RNA molecules, known as siRNAs, are produced from a long dsRNA precursor. It is regarded as a powerful reverse genetics approach for analyzing gene functions. There are various applications of the technology creating novel traits in plants (Mansoor et al. 2006). The expression of any gene involved in a particular trait can be up or down-regulated without affecting other genes expression, thus developing a selective modulation of biochemical pathways underlying crop traits. RNAi makes its use in metabolic engineering by silencing potentially deleterious genes involved in different metabolic pathways. In addition, RNAi could be used to engineer food plants rendering them rich in dietary protein; for example, lowering the levels of natural plant toxins in cotton seeds could make this abundant plant appropriate for human consumption. In short, RNAi has a growing appreciation in the diverse fields of biology, medicine and agriculture (Hebert et al. 2008; Kusaba 2004; Tang et al. 2003).

A number of successful applications of RNAi have already emerged in the field of conferring resistance to common plant viruses (Mansoor et al. 2006; Zadeh and Foster 2004). First evidence of virus resistance through RNAi was reported against PVY by transforming potato plants with the sense and anti-sense transcripts of viral helper component proteinase (HC-Pro) gene (Waterhouse et al. 1998). The technology has been used for the development of resistance against multiple RNA viruses (Hameed et al. 2017). Several approaches have been exploited successfully for the induction of RNAi like the expression of sense gene (S-PTGS), anti-sense gene (AS-PTGS) and inverted repeat (IR-PTGS), aiming to develop resistance against plant viruses including geminiviruses. To counter host defense, geminiviruses, like RNA viruses, encode specific proteins that suppress PTGS, suggesting that defense and counter-defense have co-evolved in plants and viruses (Baulcombe 2002; Voinnet 2001). Voinnet et al. 1999 tested diverse virus types for their abilities to suppress systemic PTGS and conclusively suggested that viruses may have variety of mechanisms by which they overcome silencing. And all types of aforementioned approaches are compromised upon virus infection (Voinnet et al. 1999). According to Bian et al. (2006), tomato plants having transgene of hairpin RNAi construct targeting TLCV-C2 showed a strong RNAi response initially, but the response could not be maintained after virus inoculation because of delayed infection and symptoms appeared after some days. They also noted high level of viral DNA methylation, suggesting that the virus escaped RNAi possibly due to a differential methylation of the de novo synthesized viral and host plant DNA.

In this context, an approach was used based on amplicon-mediated RNA silencing. This approach was first time used by Angell and Baulcombe (Angell and Baulcombe 1997) for PVX and later on for PLRV (Barker et al. 2001). An amplicon is a transgene based on the full-length copy of viral genome or modified viral genome and a constitutive promoter that will induce expression of RNA transcripts of the virus. As the current study is based on developing resistance against CLCuD complex, amplicon-based approach has been used to silence the expression of βC1 gene of betasatellite associated with the begomovirus. For this purpose, modified genome of betasatellite was used for the preparation of amplicon-based RNAi construct targeting βC1. The construct consists of SCR, βC1 promoter and βC1 in both sense and anti-sense orientations separated by the intron of Mungbean yellow mosaic India virus (MYMIV) (Supplementary Figure 1). βC1 promoter, being a strong promoter, should potentially induce the expression of both sense and anti-sense transcripts of βC1. As both the transcripts are separated by intron, its splicing would result in the formation of a double-stranded RNA that must be diced into siRNAs.

Although the transgenic plants would contain siRNAs in every cell of the plant, theoretically the production of siRNAs should be increased upon virus infection. Actually this is the uniqueness of this amplicon-based construct that upon virus infection, in contrary to previous RNAi approaches, the construct is actually activated to form a circular molecule of the same size as of the begomovirus-associated betasatellite. The repeated cycles of replication and transcription would increase the amount of siRNAs and produce defective molecules of betasatellites, which would ultimately come up with enhanced resistance in transgenic plants.

Materials and methods

Agrobacterium-mediated transient analysis of RNAi construct

The synthetic amplicon-based RNAi construct was obtained in cloning vector pTZ57R/T. A schematic diagram of the construct is given in Supplementary Figure 1. The construct was synthesized and the synthetic construct was digested with restriction endonucleases EcoRI/HindIII, cloned in binary vector pGreen0029 at the same restriction sites and confirmed through restriction analysis (Supplementary Figure 2). The construct in binary vector and infectious clones of Cotton leaf curl Khokran virus (CLCuKV) and Cotton leaf curl Multan betasatellite (CLCuMB) were transformed in electro-competent cells of Agrobacterium tumefaciens strain GV-3101. For inoculum preparation, the transformed cells of Agrobacterium were cultured in LB media containing rifampicin (25 µg/ml), kanamycin (50 µg/ml) and tetracycline (10 µg/ml) antibiotics and pelleted in a 50 ml centrifuge tube. The pellet was resuspended in 10 mM MgCl₂ and the final OD₆₀₀ of inoculum was set at 0.6–1. To activate Agrobacterium cells, acetosyringone (final concentration 100 µM) was added in the culture and placed at room temperature for 3–4 h. Inoculum was infiltrated underside of the leaves of plants using 5 ml injection syringe. N. benthamiana plants used for transient analysis were 5 to 6 weeks old.

Stable transformation of RNAi construct in N. benthamiana and analysis of infectivity

The promising results of transient analysis of RNAi construct led to its stable transformation in model plant N. benthamiana. Transgenic N. benthamiana plants were prepared using tissue culture techniques in vitro. The plants were grown throughout on selection media (LB media containing kanamycin and cefotaxime antibiotics) from leaflets to complete plant, till they become able to be shifted to pots. The presence of transgene was confirmed through PCR using SHLP2F (TCTGAGCTCCTGAAAAGGAATACACAC) and SHLP2R (AGAAAGCTTAATTCCAAACACAAACCA) primers (Supplementary Figure 2). Starting from co-cultivation of leaflets with the inoculum containing transgene to the onset of seeds of T_o generation of transgenic N. benthamiana plants, it took nearly seven months. Six lines were developed, each line having fifteen plants. Seeds of T_o generation were again germinated on kanamycin selection media for the production of T₁ generation. This was also a selection assay for transgenics. After successive steps of shifting plants from plates and jars (containing media) to pots, T₁ generation was obtained. Transgene was again confirmed through PCR. These plants were challenged with the infectious clones of both monopartite (CLCuKV/CLCuMB) and bipartite begomoviruses (ToLCNDV) and their responses were noted.

Results

Agrobacterium-mediated transient analysis of hairpin RNAi construct targeting βC1

Before stable transformation of hairpin RNAi construct in tobacco plants, it was transiently expressed in Nicotiana benthamiana just to note whether it replicates and moves systemically or not, and to look for the anticipated resistance response induced through RNA silencing. Though CLCuMV DNA A (isolate PK3) can induce leaf curl symptoms in Nicotiana benthamiana but its co-inoculation with CLCuMB satellite (isolate U89) produces typical and severe disease symptoms characterized by downward leaf curling, yellowing and vein thickening. Also, infectious clones of CLCuMV DNA A alone are not able to induce typical leaf curl symptoms in cotton plant, rather they need associated betasatellite for a successful infection and typical symptoms determination (Briddon et al. 2003). As already mentioned, target region in the construct is βC1, however, it was checked for resistance against CLCuMV alone and CLCuMV plus CLCuMB as well. The positive control plants were Nicotiana benthamiana plants infected with CLCuMV and CLCuMV + CLCuMB while negative controls were healthy N. benthamiana plants without inoculation. There were five treatments and six plants were used for each treatment. The treatments included N. benthamiana plants inoculated with CLCuMV, the plants inoculated with CLCuMV + CLCuMB, the plants inoculated with CLCuMV + RCβ (RNAi construct targeting βC1), the plants inoculated with CLCuMV + CLCuMB + RCβ and non-inoculated healthy plants.

The construct RCβ, infectious clones of virus and its satellite were transformed in GV-3101 strain of Agrobacterium tumefaciens to prepare the inoculum. In treatments III and IV, plants were inoculated with RCβ four days before the inoculation of virus. Idea was to give enough time to plants to produce short RNAs against the target sequence. 14 dpi (days post inoculation), inoculated control plants showed initial disease symptoms and the symptoms turned severe 25 dpi. While the plants with RNAi construct (RCβ) did not impart symptoms 14 dpi and even some plants were comparable to the non-inoculated control plants and very mild symptoms were observed 25 dpi (Fig. 1). As RCβ was designed to target βC1 gene, so theoretically it should block the replication and movement of CLCuMB and plants must impart the symptoms caused by DNA A alone. But interestingly, in plants of both treatments III and IV, symptoms were very mild, indicating that the construct has some influence on DNA A as well. To check virus replication and movement in systemic leaves, Southern hybridization analysis was performed. High titre of virus was noted in control plants DNA and virus levels were remarkably low in plants having RCβ in either combination (Fig. 2). A 1.1 kb PCR fragment amplified using primers CLCV1 (CCGTGCTGCTGCCCCCATTGTCCGCGTCAC) and CLCV2 (CTGCCACAACCATGGATTCACGCACAGGG) and labeled with DIG using a PCR DIG Probe Synthesis Kit (Roche, Germany) was used as probe to detect CLCuMV. Circular nature of RCβ and its systemic movement was confirmed through PCR using genomic DNA from systemic leaves as template. β01/β02 primers were used for this amplification.

Fig. 1.

Photographs of Nicotiana benthamiana plants after inoculation with infectious clones of virus and RNAi construct in different combinations. Symptoms exhibited by N. benthamiana plants infected with CLCuMV alone (a and b), CLCuMV and CLCuMB (c and d), CLCuMV and RCβ (e and f), CLCuMV,CLCuMB and RCβ (g and h). Non-inoculated N. benthamiana plants are shown in i and j as negative controls

Fig. 2.

Southern hybridization of a blot hybridized with CLCuMV DNA-A probe. DNA samples loaded in the gel were extracted from; non-inoculated N. benthamiana plant (lane 1), plants inoculated with CLCuMV (lanes 2, 3 & 4), plants inoculated with CLCuMV + CLCuMB (lanes 5 & 6), plants inoculated with CLCuMV + RCβ (lanes 7, 8 & 9) and plants inoculated with CLCuMV + CLCuMB + RCβ (lanes 10, 11 & 12). DNA was extracted 25 dpi and approximately 10 µg genomic DNA was loaded in each well

Infectivity of RCβ against CLCuKV/CLCuMuB

An infectious clone of Cotton leaf curl Khokran virus (CLCuKV) and Cotton leaf curl Multan betasatellite (CLCuMB) was transformed into electro-competent cells of Agrobacterium tumefaciens strain GV3101and inoculum was prepared for its agro-infiltration with an OD₆₀₀ of 0.5. As T₁ is a segregating generation, presence of transgene in these plants was confirmed through PCR (Fig. 3). Only the confirmed transgenic plants were used for infectivity analysis. 10 non-transgenic control plants were also grown along transgenics which were of the same age. The plants of two lines were not fully grown and were unable to be inoculated, so four lines were used for infectivity analysis. 10 plants from each line were infiltrated at the rate of 3 leaves per plant. Five plants of healthy N. benthamiana were also used as a negative control which were not inoculated. Severe leaf curling and vein yellowing were observed in non-transgenic control plants 25 dpi. Transgenic plants were divided into three categories depending upon their phenotypic responses to virus infection (Fig. 4).

• I.13 plants remained symptomless just like non-inoculated control plants of N. benthamiana. Even 50 dpi, there were no symptoms at all (as shown in panels A, B, C and D in Fig. 4).
• II.15 plants out of 40 showed very mild symptoms of leaf curling and vein yellowing (as shown in panels E, F, G and H in Fig. 4). These plants were kept under observation and they maintained the same level of symptoms expression 50 dpi.
• III.12 plants out of 40 showed delayed expression of symptoms. They also imparted milder symptoms 25 dpi just like the plants of second category, which turned severe in the next two weeks. After 50 days, the plants were completely susceptible showing severe downward leaf curling and vein yellowing just like non-inoculated control plants.

Fig. 3.

PCR for the confirmation of transgene using genomic DNA of T_o and T₁generation plants as template. The ethidium bromide-stained agarose gel was photographed under UV illumination. SHLP2F and SHLP2R primers were used for an amplification of 834 bp. Well no. 1 shows a negative control for PCR in which template DNA belonged to a non-transgenic healthy N. benthamiana plant; well no. 2 and 3 represent positive controls having plasmid containing transgene and inoculum used for co-cultivation as template, respectively; while the wells from 4 to 7 show transgenic lines of T₀ generation and 8 to 13 show transgenic lines of T₁ generation. A 1 kbDNA size marker was electrophoresed in lanes M

Fig. 4.

Photographs of N. benthamiana plants taken at 25 days after inoculation. Following inoculation with CLCuKV and CLCuMB, transgenic plants which remained asymptomatic are shown in a, b, c and d; the plants which imparted mild symptoms are presented in e, f, g and h; non-transgenic control plants expressing typical disease symptoms are shown in i and j. k, l Represent non-inoculated, healthy N. benthamiana plants

All asymptomatic and symptomatic plants (both of mild and severe symptoms) tested were positive for the presence of CLCuKV by PCR; wherease, 90% of plants were negative for the presence of CLCuMB. All inoculated, non-transgenic plants were positive for the presence of both CLCuKV and CLCuMB by PCR (results not shown). Southern blot hybridization was performed to determine the level of vrial DNA in leaves developing at the time of, or developed after, inoculation. Probe used was the same 1.1 kb PCR fragment amplified using primers CLCV1 and CLCV2 and labeled with DIG using a PCR DIG Probe Synthesis Kit (Roche, Germany). Thick bands represent a high titre of begomovirus ‘CLCuKV’ in non-transgenic plants used as control (lanes 2, 3 in panels A and B in Fig. 5). While, the virus titre was remarkably low in all the transgenic plants (with a few exceptions). The results of Southern blot hybridization were also consistent for the levels of CLCuMB. Probe used for the detection of betasatellite was an amplification product of 357 bps using primers betaC1 F (TACATATGATGACAACGAGCGG) and betaC1 R (TTAAGC TTAAACGGTGAACTTTTT) and labeled with DIG using a PCR DIG Probe Synthesis Kit (Roche, Germany). In non-transgenic plants, the levels of betasatellite were high (as shown in lanes 2, 3 in panels C and D in Fig. 5). On the other hand, no signal of betasatellite was detected in transgenic plants (transgenic lines L3 and L4 in panel D in Fig. 5). A comparatively lower titre of replicative form of betasatellite was also detected in some samples belonging to lines L1 and L2 (panel C in Fig. 5). It was interesting to note that these were the same samples in which the levels of begomovirus were also comparatively high (as noted in lanes 4, 7, 8 and 11 in panels A and C in Fig. 5).

Fig. 5.

Southern hybridization of blots probed with CLCuMV DNA A (a, b) and βC1 of CLCuMB (c, d). DNA samples run on gel were extracted from a non-inoculated, healthy N. benthamiana plant (lane 1), non-transgenic N. benthamiana plants inoculated with CLCuKV and CLCuMB (lanes 2 & 3). The samples run in lanes 4 to 7 belong to N. benthamiana plants of transgenic line ‘L1’ and from lanes 8 to 11 were extracted from N. benthamiana plants of transgenic line ‘L2’ after inoculation with CLCuKV and CLCuMB (a and c). Similarly, the samples loaded in lanes 12 to 15 were extracted from transgenic plants belonging to transgenic line ‘L3’ of N. benthamiana following inoculation with infectious clones of CLCuKV and CLCuMB; and the DNA sample in lane 16 to 19 were extracted from N. benthamiana plants of transgenic line ‘L4’ after inoculation with CLCuKV and CLCuMB. Total genomic DNA was extracted from the leaves developing at the time of, or developed after, inoculation and was sampled at 25 dpi. An approximately equal amount of DNA (10 µg) was loaded in each well

Since the construct was designed in a way that upon virus infection, due to Rep-mediated activation, the transgene will loop out in a circular molecule and start replicating. Thus, its circular nature was also confirmed through PCR giving an amplification product of approximately 1.4 kb, using universal primers for betasatellites amplification (β01/β02) designed on SCR (Fig. 6). In this case, the template DNA used was extracted from the leaves of transgenic plants infected with CLCuKV alone. Infectivity of CLCuKV also resulted in mild symptoms and reduced level of viral DNA.

Fig. 6.

PCR for the confirmation of looping out of transgene. The ethidium bromide-stained agarose gel was photographed under UV illumination. The samples loaded on gel resulted from the PCR reaction with primer pair β01/β02 and DNA extracted from the leaves of transgenic plants inoculated with CLCuKV. The sample in lane C resulted from PCR amplification with DNA extracted from a healthy N. benthamiana plant, while the samples in all other lanes were extracted from the leaves of transgenic plants of four lines inoculated with CLCuKV. Leaves were sampled 25 dpi. A 1 kbDNA size marker was electrophoresed in lane M. The sizes (bp) of selected marker bands are indicated on the right

Discussion

RNA interference (RNAi) is a homology-dependent gene silencing mechanism which is triggered by the formation of double-stranded (ds) RNA and proceeds with the production of short interfering (si) RNAs. Inverted repeat/hairpin RNA constructs have been proved to be more efficient in resistance development than S/AS-PTGS constructs (Waterhouse et al. 1998; Chuang and Meyerowitz 2000). But the drawback of these mechanisms is their suppression upon virus infection attributed mainly to the geminivirus proteins acting as suppressors of gene silencing (Bisaro 2006). The novelty of amplicon-based RNAi construct is its enhanced production of siRNAs upon virus infection. This is because both arms of the hairpin are ligated with the SCR of cotton leaf curl betasatellites, thus providing two origins of replication (ori) on the same construct (Supplementary figure 1). Upon infection, geminivirus Rep can recognize the ori at one end, nick it and religate it at the second ori position (Laufs et al. 1995). This will make a circular molecule, capable of producing more short RNAs upon repeated replication and transcription. Looping out of the circular molecule of construct has been confirmed by PCR using universal beta primers (Fig. 6). Actually the construct RCβ has been designed to serve a dual purpose function; first, the transgenic plants would produce siRNAs targeting βC1, and second, the looped out circular molecule will not only increase siRNAs production but it will make defective molecules (having no betaC1) with a size less than that of betasatellite as well. The circular molecule of construct is of 1.4 kb but the splitting of intron may result in the production of RNA with no βC1 due to the complementarity of its sense and anti-sense sequences.

Only a few amino acid changes have been observed previously in the amino acid sequence of old CLCuMB and the recombinant betasatellite (CLCuMB^Bur) associated with resistance breakdown in Punjab in 2001 (Akhtar et al. 2014). Comparison of other recombinant betasatellites, cloned recently and included in study, also revealed the conserved nature of βC1 (Zubair et al. 2017). This consistency of the conserved sequence of βC1 makes it a good target for RNAi. Furthermore, the size of βC1 lies in the size specificity of hpRNA (98–853 bases) of the target sequence as determined by Wesley et al. 2001 for efficient silencing in transgenic plants (Wesley et al. 2001).

Though inverted repeat gene silencing is highly target-specific because of homology-dependent nature of the mechanism, and the RCβ should, in principle, target only βC1 lowering the titre of replicating betasatellite. But the low titre of begomovirus in infectivity analysis suggests that the construct also inhibited or interfered in the replication of begomovirus (Fig. 5). There might be two possible reasons for such interference. First, the defective molecules produced upon virus infection due to the looping out of construct in circular form, may obstruct the replication of begomovirus by competing with the virus for host resources, thereby reducing its level (Patil and Dasgupta 2006; Stanley and Townsend 1985). Second, the infectivity results are supported by the work of Yang et al. (2011) in which they characterized siRNAs from a monopartite begomovirus (TYLCCNV) and its satellite (TYLCCNB) using Solexa-based deep sequencing. They concluded that the presence or absence of betasatellite/βC1 affects the amount of viral sRNAs(V-sRNAs) by altering V-sRNAs generating hotspots. In N. benthamiana, not a natural host of TYLCCNV/TYLCCNB, the presence of intact betasatellite promoted the production of V-sRNAs from 1.81 to 3.37%. More sRNAs means more hindrances in virus replication and movement (Yang et al. 2011).

Most of the plants of transgenic lines L3 and L4 remained symptomless and their southern analysis was also in line with the phenotypic results. Southern blot having the genomic DNA of these lines could not show even a single band of replicating betasatellite, representing that betasatellites are being efficiently targeted. Begomovirus level was also reduced in these lines (Fig. 5). While some plants in lines L1 and L2 showed mild symptoms, and there were also some other plants in the same lines whose symptoms turned severe 8 weeks post inoculation. In southern hybridization experiments, these were the same plants which showed a detectable level of betasatellite and a comparatively increased titre of begomovirus. The plant sample numbered as 11 in line L2 showed increased titre of begomovirus and the betasatellite. Regarding symptoms expression, this was amongst the most susceptible plants of all lines. Thus, the results of southern hybridization correlate with the phenotypic expression of plants.

Another important thing is that upon virus infection, differences in responses were observed by different plants of the same line. Possible explanation is that infectivity analyses were performed on T₁ generation and being tetraploid, all N. benthamiana plants were not in fully homozygous form. Determination of correlation of short RNAs of the transgene with the level of DNA of begomovirus/betasatellite would be an interesting study to carry out in future.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contribution

SA performed experiments and prepared the draft of the manuscript. MNT helped in the conduct of different experiments. IA and SM conceived the idea, designed the experiments, supervised the work and helped in technical writing.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest in the publication.