The Internal Extra Sequence Regions in Satellite RNA TA-Tb Are Important for Suppressing RNA Accumulations of Cucumber Mosaic Virus to Attenuate the Virulence of the Helper Virus

Abstract

Cucumber mosaic virus (CMV) infection is often associated with satellite RNA (satRNA), which can sometimes interfere with the replication and symptom expression of CMV. However, the mechanism underlying symptom attenuation has remained unclear. We previously discovered a larger type (than the usual type) of satellite RNA (satRNA TA-Tb) of CMV that reduced the symptom severity of CMV. Herein, we show that satRNA TA-Tb is associated with a reduction in CMV RNA accumulation, and particularly, a strong reduction of RNA4 accumulation at later stages of infection. Deletion analysis showed that the deletion of ten nucleotides of 5′ and 3′ termini, but not the internal sequence regions proximal to the 5′- and 3′-terminal regions, abolished satRNA TA-Tb replication. The alignment of satRNA TA-Tb with usual satRNA isolates showed four internal extra sequence regions (exR1–4) in satRNA TA-Tb. A satRNA TA-Tb mutant with deletion in the exR1 region retained the ability to attenuate CMV symptoms, whereas deletion of the exR2–4 regions abolished the attenuating effect of satRNA TA-Tb, but did not affect its replication. Overall, these results suggest that some short, internal extra sequence regions are dispensable for satRNA TA-Tb replication, but important for symptom attenuation function, supporting the possibility that the RNA structure of satRNA TA-Tb is important for its function in symptom attenuation.

1. Introduction

Many plant viruses are associated with linear or circular single-stranded satellite RNAs (satRNAs), subviral agents that depend on associated (helper) viruses for their replication, encapsidation, and transmission, but share very little or no sequence homology with the genomes of the helper viruses and, generally, are not essential in the infection cycles of the helper viruses [1,2,3]. Linear satRNAs can be categorized into two groups: nonstructural protein-encoding large satRNAs (800–1500 nt) and noncoding small satRNAs (<700 nt) [4,5]. Apart from their dependence on helper viruses, satRNAs may also modulate the symptom expression and genome accumulation of helper viruses. For these reasons, satRNAs have been widely studied in order to further understand virus replication, pathogenicity, and host antiviral responses, and have also been used as biocontrol agents against plant diseases caused by helper viruses [3,4,6,7].

Cucumber mosaic virus (CMV) is a positive-sense, single-stranded RNA virus belonging to the genus Cucumovirus in the family Bromoviridae [8]. CMV has a tripartite genome, designated as RNA1, RNA2, and RNA3, and each of which is encapsidated in icosahedral particles 29 nm in diameter [9]. RNA1 and RNA2 encode two large proteins (1a and 2a, respectively) that are essential for the replication of the viral genome [10]. RNA2 also encodes an additional small protein (2b) expressed through subgenomic RNA (RNA4a), and whose function involves the suppression of host antiviral RNA silencing [11,12]. RNA3 is bicistronic, encoding for the 3a movement protein (MP) and the coat protein (CP), which is translated through CP-coding subgenomic RNA, commonly called RNA4 [13,14]. CMV is considered to be one of the most important plant viruses due to its economic impacts and intensive scientific studies [15,16]. CMV is known as a widespread plant virus that has the broadest host range, infecting more than 1000 plant species including various important plant crops [17], and is even found to naturally infect or associate with fungi [18,19].

CMV infection is often associated with small, linear single-stranded satRNAs with highly conserved secondary structures ranging from 330 to 400 nt long, which are usually encapsidated in the virions [5,20]. The majority of CMV satRNAs are either asymptomatic or attenuate the symptom expressions of CMV [21,22]. However, there are instances in certain plant species where CMV satRNA isolates intensified viral symptom expressions, such as through the induction of necrosis (programmed cell death) in tomatoes by necrogenic satRNAs [23], or a bright yellow symptom in Solanaceae species occurring through downregulation of the mRNA of ChlI, a key gene for tobacco chlorophyll synthesis that is mediated by Y satRNA-derived small RNA [21,24,25]. In particular, it has recently been demonstrated that the bright yellow color of the leaves attracts aphid vectors of CMV, while the small RNAs of Y satRNA enrich the mRNA of the ATP-binding cassette subfamily G member 4 gene in aphids, thereby promoting aphid wing formation [26]. SatRNA-mediated viral symptom attenuation is commonly associated with the reduction of CMV accumulation [27,28,29]. The molecular mechanism underlying this reducing effect of satRNAs on CMV titer is not fully understood, although some studies suggest varying mechanisms involving competition between satRNA and the CMV genome for viral replicase [30], reduced expression of the 2b silencing suppressor [31], or interference with the function of viral silencing suppressors [32]. Due to their ameliorative effect on CMV-induced disease, satRNAs have been applied as biocontrol agents to protect crop plants in fields [33].

We have previously identified a CMV satRNA named satRNA TA-Tb (accession no. MF142365) from tobacco plants. In Nicotiana benthamiana plants, infection with satRNA TA-Tb caused a significant reduction in symptom severity caused by the CMV Fny strain [34]. Moreover, we also showed that the inoculation of satRNA TA-Tb after 35 days pre-infection with CMV was also able to attenuate viral symptom severity, demonstrating a new potential application of satRNAs for the protection of plants against CMV-induced disease [7]. The RNA structure of satRNA TA-Tb was analyzed using selective 2′-hydroxyl acylation by the primer extension (SHAPE) method in combination with structural prediction using software. Previous analyses have suggested that the RNA structure of satRNA TA-Tb consists of five 5′- and 3′-proximal stem-loops and one internal, large, multi-branched stem-loop, along with two possible pairs of kissing loops [35]. However, the effect of satRNA TA-Tb on CMV accumulation and the mechanism of its symptom attenuation remained unclear.

To further characterize satRNA TA-Tb and gain insight into its mechanism of CMV symptom attenuation, in this study, we carried out deletion analysis to investigate the sequence regions in satRNA TA-Tb that are required for its replication and its attenuating effect on CMV symptom expression. The results of our study could provide a scientific basis for the further development of satRNA TA-Tb as a biocontrol agent for controlling plant diseases caused by CMV.

2. Materials and Methods

2.1. Plant Growth and Virus Inoculation

N. benthamiana plants were grown in a plant growth room at 25 °C with 16 h light and 8 h dark cycles. The infectious clone plasmids containing CMV Fny RNA1 (acc. no. D00356), RNA2 (D00355), and RNA3 (D10538) [36] were kindly provided by Professor Tao Xiaorong (Nanjing Agricultural University) and used for the agroinoculation of N. benthamiana plants, as described previously [37]. At least three plants were used for the inoculation of each satRNA TA-Tb wild type and mutant derivative.

2.2. Construction of satRNA TA-Tb Infectious cDNA Clone and Its Mutant Derivatives

An infectious clone of satRNA TA-Tb was prepared by PCR amplification of a pUC19-satCMVTA-Tb plasmid [35] using satRNA TA-Tb-specific primers. All primers used in this study are listed in Supplementary Table S1. The PCR product was inserted between StuI and SmaI sites of the expression cassette (2X35S-HDV ribozyme-NOS) in the mini binary vector pCB301 [36] through homologous recombination (ClonExpress Ultra One Step Cloning Kit, Vazyme, Nanjing, China). The resulting plasmid construct was transformed into the Agrobacterium strain GV3101 and used for the agroinoculation of N. benthamiana plants, together with Agrobacterium-mediated infectious cDNA clones of the CMV Fny strain. The satRNA Ta-Tb deletion mutants were generated using an overlapping PCR-based method using primers described in Supplementary Table S1. The PCR-generated mutant fragments were cloned to binary vector pCB301 and transformed into the Agrobacterium strain GV3101, as described above.

2.3. RNA Extraction and Northern Blot Analysis

Total RNA was extracted from non-inoculated upper leaves, as described previously [38]. Northern blot analysis was carried out using a Digoxigenin-labelled DNA probe specific to the CMV genome (conserved 234–238 nt 3′-terminal regions of RNA1, RNA2, and RNA3) and satRNA TA-Tb prepared with the DIG-High Prime DNA Labeling and Detection Starter Kit II (Roche Diagnostics, Mannheim, Germany). The hybridization conditions and the detection of RNAs were as described in the DIG Application Manual supplied by Roche.

2.4. Western Blot Analysis

Western blot analysis using an antibody specific to CMV CP was carried out as described previously [39]. For the preparation of plant total protein samples, leaves of N. benthamiana plants were ground using liquid nitrogen and 0.3~0.5 g powdered leaves were placed in 1.5 mL centrifuge tubes. Water (200 μL) was added and mixed vigorously. Homogenate (50 μL) was taken and mixed with an equal volume of 2× sample buffer (100 mM Tris-HCl pH 6.8, 20% glycerol, 4% sodium dodecyl sulphate (SDS), 0.2% bromophenol blue, and 5% β-mercaptoethanol), followed by boiling for 5 min. After cooling on ice for 5 min, the sample was centrifuged at 12,000 rpm for 10 min at 4 °C and the supernatant was collected. Protein samples underwent SDS-polyacrylamide gel electrophoresis using 5% stacking gel (80 V) and 12.5% separating gel (120 V) until bromophenol blue migrated out of the gel. The separated proteins from the gel were transferred onto PVDF membrane that had been treated with methanol for 20 min using semi-dry electroblotting (Bio-Rad Laboratories, Hercules, CA, USA) at 18 V for 40 min. After rinsing the membrane with deionized water, the membrane was soaked in 10 mL of blocking solution (1% skim milk) for 1~1.5 h or overnight at 4 °C. Next, 4 μL of primary polyclonal antibody serum (1:5000 dilution) was added to the blocking solution, which was incubated on a shaker at room temperature for 1~1.5 h. After washing the membrane 6 times with 10~15 mL TBST buffer (50 mM Tris pH 7.5, 150 mM NaCl, and 0.1% Tween 20) for 5 min each time, the secondary antibody goat anti-rabbit IgG labelled with horseradish peroxidase (1:2500 dilution, Biodragon, Chengdu, China) was used to incubate the membrane at room temperature for 1~1.5 h. The membrane was washed 6 times with TBST buffer, for 10 min each time. Protein signals were developed using SuperSignal™ West Pico PLUS Chemiluminescent (Thermo Fisher Scientific, Waltham, MA, USA) and detected using a Storm 820 Molecular Imager (GE HealthCare, Chicago, IL, USA).

2.5. Sequence Analysis

The sequence alignment of satRNA isolates was performed using CLC Genomics Workbench 3.6.5 (CLC bio, Aarhus, Denmark).

3. Results

3.1. Replication of satRNA TA-Tb Is Associated with Reduced Accumulation of CMV RNA Genome

To facilitate further investigation of satRNA TA-Tb’s characteristics, an infectious cDNA clone of satRNA TA-Tb was generated. The resulting plasmid construct was used for the agroinoculation of N. benthamiana plants, together with a CMV Fny strain. Yellow mosaics and curly leaves, which are the typical CMV Fny symptoms in N. benthamiana, began to appear at seven days post-inoculation (dpi), but up to 12 dpi, no obvious difference in viral symptom severity was observed between plants that did and did not undergo satRNA TA-Tb inoculation (Figure 1A). However, starting from 14–16 dpi, the plants with satRNA TA-Tb inoculation showed less severe symptoms than the plants without satRNA TA-Tb inoculation; satRNA TA-Tb-inoculated plants showed extended height and leaf growth, while satRNA TA-Tb-free plants remained severely stunted, with strongly curled leaves (Figure 1A). Notably, at a later stage of virus infection (28–40 dpi), the attenuating effect of satRNA TA-Tb inoculation on CMV symptom severity was much more pronounced (Figure 1A).

To investigate the relationship between CMV symptoms and virus accumulation, northern blot analysis was carried out. As expected, the blot analysis confirmed the accumulation of satRNA TA-Tb in inoculated plants throughout the period of 3–40 dpi (Figure 1B), indicating that agroinoculation methods are reliable for the successful inoculation of satRNA. At 3–7 dpi, CMV RNA1-3 accumulated to a comparable level in satRNA TA-Tb-infected and -free plants; however, from 14 dpi onward, the presence of satRNA TA-Tb was associated with a reduction in CMV RNA accumulation. In particular, at 28 dpi a strong reduction in RNA4 accumulation was observed, while at 40 dpi, CMV RNA accumulations were below a detectable level (Figure 1B,C). These observations indicate that CMV symptom attenuation by satRNA TA-Tb is associated with reductions in CMV genome accumulation and particularly strong reductions in RNA4 accumulation at later stages of virus infection.

3.2. Deletions of Ten Nucleotides of 5′ and 3′ Termini Abolish Replication of satRNA TA-Tb

SatRNA TA-Tb is a larger type of CMV satRNA, with a size ranging from 368–405 nt, as compared to the usual (shorter) satRNAs with lengths of 332–342 nt [20,35,40]. Sequence alignment of satRNA TA-Tb with other usual types of satRNA—namely satRNA D4 (acc. no. M30586), which induces a mild amelioration of symptoms in tobacco and lethal necrosis in tomatoes [41,42]; satRNA T1 (acc. no. DQ785472), which attenuates CMV symptoms in N. benthamiana [27]; and satRNA TA-ra (acc. no. MF142364), which enhances CMV symptoms in N. benthamiana [34]—indicated four internal extra sequence regions (9–17 nt) in satRNA TA-Tb, designated as exR1, exR2, exR3, and exR4, respectively (Figure 2). It was also noted that the 5′- and 3′-proximal regions and other internal sequence regions were relatively conserved among these satellite RNAs (Figure 2).

Our previous study predicted the formation of small distinct stem-loops in the 5′- and 3′-proximal regions [35]. To investigate the roles of the 5′- and 3′-proximal regions of satRNA TA-Tb in its biology, a series of mutants containing deletions at nucleotide positions 1–10, 11–20, and 21–30 relative from the 5′- and 3′-terminal nucleotides—referred to as TA-Tb ∆5R1-10, ∆5R11-20, ∆5R21-30, ∆3R1-10, ∆3R11-20, and ∆3R21-30 mutants (see Figure 2 and Figure 3A)—were generated and used for the agroinoculation of N. benthamiana along with CMV Fny, as described above. At 16 dpi, none of the plants inoculated with the deletion mutants showed CMV symptom attenuation as those inoculated with the TA-Tb wild type (WT, Figure 3B) did. Northern blot analysis detected TA-Tb ∆5R11-20, ∆5R21-30, ∆3R11-20, and ∆3R21-30, but not TA-Tb ∆5R1-10 or ∆3R1-10, as having similar accumulation levels to that of TA-Tb WT (Figure 3C). However, the accumulation of these deletion mutants was not associated with any reduction in CMV RNA accumulation (Figure 3C,D), which is consistent with the observation of CMV symptoms (Figure 3B). Thus, these observations show that the deletions of ten nucleotides of the 5′ and 3′ termini abrogate the replication of satRNA TA-Tb, whereas the internal sequences proximal to the 5′- and 3′-terminal regions are dispensable for the replication, but critical for CMV symptom attenuation.

3.3. Internal Extra Sequence Regions of satRNA TA-Tb Are Important for Attenuation of CMV Symptoms

To investigate the possible roles of four internal extra sequence regions in the biological effects of satRNA TA-Tb, mutants containing deletions of each of these extra sequences (Figure 2 and Figure 4A) were generated and used for the agroinoculation of N. benthamiana along with CMV Fny. At 28 dpi, the plants inoculated with TA-Tb WT and ∆exR1 showed similar attenuations of CMV symptom severity, whereas plants inoculated with TA-Tb ∆exR2–4 showed severe symptoms similar to those of satRNA TA-Tb-free plants (Figure 4B). Northern blot analysis detected accumulation levels similar to those of WT in all satRNA TA-Tb deletion mutants (Figure 4C). This suggests that deletion of the internal extra sequences did not affect the replication of satRNA TA-Tb. Consistent with the symptom attenuation, in plants infected with satRNA TA-Tb ∆exR1 but not ∆exR2–4, accumulation levels of RNA1–3—and, to a larger extent, RNA4—were reduced, similarly to those of plants infected with TA-Tb WT (Figure 4C,D). Interestingly, in plants infected with satRNA TA-Tb ∆exR4, accumulation levels of RNA4 were elevated (Figure 4C,D). Western blot analysis using an antibody specific to CMV CP was carried out. Consistent with the RNA4 accumulation levels, CMV CP showed reduced CP levels in plants infected with SatRNA TA-Tb WT and ∆exR1, but increased CP levels in plants infected with satRNA TA-Tb ∆exR4 (Figure 4C). These results indicate that the internal extra sequences exR2, exR3, and exR4, but not exR1, are important for satRNA TA-Tb activity in the suppression of RNA4 accumulation. These observations further reinforce the association of satRNA TA-Tb-mediated symptom attenuation with the strong suppression of CMV RNA accumulation.

4. Discussion

In this study we showed that CMV symptom attenuation by satRNA TA-Tb is associated with the overall reduction in CMV RNA accumulation. As CMV satRNAs do not encode proteins, their ameliorating effects on the symptom expression and accumulation of helper viruses are generally attributed to their replication and RNA sequence/structure. In vitro RNA synthesis of the viral genome and satRNA of CMV using purified CMV RdRP, as well as the trans CMV replication system in plants, showed that the rate of viral genome synthesis is lower in the presence of satRNA [27,30]. This observation supports the view that the replication competitiveness of satRNA contributes to its suppression of helper virus accumulation. As the suppression of CMV RNA accumulation by satRNA TA-Tb was more prominent at later stages of infection (from 14 dpi onward, Figure 1B,C), it is necessary to examine whether satRNA TA-Tb confers replication competitiveness onto CMV by experimenting using in vitro or single-cell systems (for example, protoplast infection systems). CMV satRNAs, particularly the secondary structures, are recognized by plant Dicer-like proteins to produce small interfering (si)RNAs [32,43]. It has been suggested that these highly abundant satRNA-derived siRNAs could bind and saturate the 2b silencing suppressor, compromising its functions in the suppression of antiviral RNA silencing and the induction of developmental abnormalities and symptoms through the interference of endogenous siRNA and micro RNA pathways [32]. It is interesting to investigate the abundancy and characteristics of siRNAs derived from satRNA TA-Tb and examine whether they originate from some of the specific RNA structures determined by our previous study [35].

The molecular studies on larger types of CMV satRNAs have been limited so far. To gain deeper insight into the molecular and biological characteristics of satRNA TA-Tb, a deletion analysis of satRNA TA-Tb was carried out in this study. The analysis showed that ten nucleotides of 5′ and 3′ termini are essential for satellite replication. This result is quite expected, because the 5′- and 3′- terminal regions of RNA viruses are known to be important for the synthesis of viral RNA genomes [44]. However, the deletion of internal sequences proximal to the 5′- and 3′-terminal regions (5R11-20, 5R21-30, 3R11-20, and 3R21-30, Figure 2 and Figure 3A) did not affect the viability of satRNA TA-Tb (Figure 3C). Structural prediction by software, supported by SHAPE analysis, suggested the formation of 5′- and 3′-proximal small stem-loops (nucleotide positions 122 and 114, relative from the 5′- and 3′-terminal nucleotides, respectively) in satRNA TA-Tb [35]. Thus, it might be that the formation of these RNA stem-loop structures is not important for the replication of satRNA TA-Tb. Overall, the deletions of internal sequence regions (including extra internal sequences) carried out in this study did not affect the infectivity of satRNA TA-Tb, but the deletions (except the deletion in exR1) abolished its ability to attenuate CMV symptoms. Interestingly, here, the infection of satRNA TA-Tb ∆exR4 enhanced RNA4 accumulation. Thus, satRNA TA-Tb seems to regulate the synthesis of CP subgenomic RNA. Based on our RNA structural analysis of satRNA TA-Tb [35], exR2-4 regions are located in the multi-branched stem-loop involved in the formation of a kissing loop. This observation suggests that the symptom attenuation function of satRNA TA-Tb is regulated by complex intramolecular interactions related to specific RNA structures. In contrast, small internal deletions abolished Y satRNA viability [45]. This may reflect the different flexibility of sequence requirements for viability between the larger and usual (shorter) satellite RNAs.

Our RNA blot analysis showed the strong suppression of RNA4 accumulation at later stages of infection. A previous study identified a larger CMV satRNA (XJs1, acc. no. DQ070748) from sugar beet. Notably, this satRNA has high sequence similarity (95.83%) with satRNA TA-Tb, and causes strong viral symptom alleviation in addition to markedly reducing RNA4 accumulation, as indicated by quantitative RT-PCR [46]. It would be interesting to further investigate whether the larger types of CMV satRNAs have similar characteristics in terms of the suppression of RNA4 accumulation and interaction with the RNA3 intercistronic region. As CMV CP is not required for virus replication, but is important for long-distance virus movement [47,48], the reduction of RNA4 accumulation may partially explain why satRNA TA-Tb infection prominently affects CMV accumulation at later stages, but not at the initial stage of infection.

5. Conclusions

In this study, deletion analysis of satRNA TA-Tb was carried out. Our results show that the deletion of ten nucleotides of 5′ and 3′ ends abolishes satellite replication, whereas ten nucleotides’ internal sequence regions—proximal to the 5′- and 3′-terminal regions and some short, extra sequences in the central region—are dispensable for replication, although most of them are essential for the attenuation of CMV symptom expression and the reduction of CMV genome accumulation. These observations support the view that the RNA structure of satRNA TA-Tb is important for its function in symptom attenuation.