Background
Viruses are obligate intracellular pathogens. They hijack host functions, divert host resources and suppress host defense responses to achieve successful infection [
1]. These involve an array of interactions with cellular factors, which, inevitably or coincidentally, often lead to host physiological disorders manifested by a variety of disease symptoms [
2,
3]. Understanding molecular details from infection of a virus to symptom development of the host is one major mission of plant virologists. Transcriptome profiling has been used extensively in the past decade to understand mechanisms underlying plant-virus interaction [
4,
5]. Transcriptional response of plants to virus infection is shown to vary depending on virus species, virus strains and the genetic backgrounds of host plants [
6‐
8]. However, it shows a tight link with phenotypes and thus is useful to reveal how a virus colonizes a host, how a host mounts a defense response against a virus, and how a compatible virus-host interaction results in disease symptoms [
6‐
8]. Also, these studies find that some genes may be commonly regulated by different viruses in different host plants [
9]. For example, a set of ribosomal genes have been shown to be up-regulated in Arabidopsis,
Nicotiana benthamiana and rice infected with
Turnip mosaic virus (TuMV), Plum pox potyvirus (PPV) and
Rice stripe virus (RSV), respectively [
10‐
12].
Rice, one of the main crop plants as well as a model for monocot plant research [
13], is host to many viruses. Among them,
Rice dwarf virus (RDV), a member of the genus
Phytoreovirus in the family
Reoviridae, is one of the most widespread and disastrous rice-infecting viruses causing great yield reduction in south East Asia [
14‐
16]. RDV is transmitted in a propagative and circulative manner by leafhoppers (
Nephotettix spp.) [
17]. Typical symptoms associated with RDV infection include severe dwarfism, increased tilling and white chlorotic specks on the infected leaves [
18].
RDV are icosahedral double-shelled particles of approximately 70 nm in diameter. The genome of RDV is composed of 12 segments of double stranded RNAs, which are named S1-S12, respectively, according to their migration during sodium dodecyl sulfate–polyacrylamide gel electrophoresis. S1, S2, S3, S5, S7, S8, and S9 encode seven structural proteins, namely, P1, P2, P3, P5, P7, P8, and P9, respectively. P1, a putative RNA polymerase; P5, a putative guanylyltransferase; and P7, a nonspecific nucleic acid binding protein form the core of RDV together with viral dsRNAs [
19]. P3 and P8 are major components of the inner and outer protein shells that encapsidate the core, respectively [
20,
21]. P2 and P9 are minor components of the outer capsid [
22,
23]. The structural features and the process of assembly of RDV virions have been well studied [
24,
25]. Besides structural proteins, RDV encodes at least five non-structural proteins, namely Pns4, Pns6, Pns10, Pns11, and Pns12, respectively. Pns6, Pns11 and Pns12 are matrix proteins of viroplasm, which is the putative site of viral replication [
26]. Pns4 is a phosphoprotein and is localized around the viroplasm matrix in insect cells [
27]. Several proteins of RDV have been shown to play specific roles in RDV-rice interaction. For example, Pns6 was identified as a viral movement protein and Pns10 as a RNA silencing suppressor of RDV [
28,
29]. P2 interacts with ent-kaurene oxidases of rice, which leads to reduced biosynthesis of gibberellins and rice dwarf symptoms [
30].
In this study, the transcriptome of the
indica subspecies of rice, namely
Oryza sativa L. ssp.
indica cv Yixiang2292, in response to RDV infection was profiled using Affymetrix GeneChips, which contains probes representing the entire genome of rice [
13] (
http://www.affymetrix.com). Our results further confirm the notion that induction of defense related genes is common for rice infected with RDV and there are correlations between transcriptional changes and symptom development in RDV-infected rice.
Discussion
The transcriptome of RDV infected rice plants was profiled in this study. A number of genes are differentially expressed in RDV infected rice. Changes of most of these genes are consistent with previous studies carried out using
N. benthamiana or
Arabidopsis thaliana [
4,
10,
11,
35‐
39]. Also, we find induction of a set of defense related genes including PR genes, WRKY transcription factors and several genes functioning in RNA silencing. This is consistent with reports of Shimizu
et al. [
9] and Satoh
et al. [
7] showing that increased expression of defense related genes may be a common response of rice infected with RDV [
7,
40]. However, our results indicate that RDV induced the expression of far more genes than it suppressed. This is in sharp contrast to the report of Shimizu
et al.[
40]. Multiple resons may be responsible for the inconsistency. But the most plausible one is that transcriptome change in response to RDV infection is host-specific. In the study of Shimizu
et al. [
40], the
Japonica subspecies of rice, namely
Oryza sativa L. cv. Nipponbare, was used, whereas in this study, the
indica subspecies of rice, namely
Oryza sativa L. ssp.
indica cv Yixiang2292, was used. Yixiang 2292, the rice variety used in this study, shows moderate resistance to RDV infection. It can develop typical symptoms of RDV infection, but the symptoms are not as severe as those of more susceptible varieties. A number of recent studies have demonstrated that there is a co-relation between symptom severity and transcriptional alteration in different virus-host combinations [
6‐
8,
41].
Many genes related to protein synthesis (Figure
1) were found and the GO term Ribosome was significantly enriched in the DEGs (Table
1). This is consistent with several studies showing that up-regulation of ribosomal genes and a set of other genes involved in protein synthesis could be a general response of plants to many viruses [
10‐
12]. It has been suggested that this may be a strategy used by the virus to enhance the capacity of the cell to synthesize proteins [
10‐
12].
As a two-subunit ribonucleoprotein complex comprising tens of ribosomal proteins and four species of ribosomal RNAs, the biogenesis of ribosome is one of the most energy consuming cellular processes [
42‐
44]. So it is anticipated that synthesis of ribosomal components should be downregulated in response to environmental cues, as it has been shown in yeast and in Arabidopsis [
45,
46]. Therefore, increased expression of ribosomal genes in virus infected plants may be a result of specific virus-host interaction.
Here, we show that RDV infection also causes a significant alteration of many nucleolar genes (Table
1). The fact that RDV Pns11 has a nuclear localization signal and induces the expression of nucleolus-related genes in tobacco (Figure
4) and the observation that nucleoli seems to be affected in RDV infected rice (Figure
3) support the notion that this alteration is specific and may be useful for RDV. Nucleolus is the site of ribosomal RNA synthesis and processing and ribosome maturation [
47]. Therefore, it is possible that, besides ribosome, RDV may also target nucleolus to manipulate the translation machinery of rice. Interestingly, there is evidence that certain nucleolar components or its overall state play a crucial role in controlling ribosomal gene expression and biogenesis [
46,
48]. So it is intriguing to speculate that RDV may specifically target nucleolus to enhance expression of ribosomal genes. In this regard, it is worth noting that a number of viruses, including many RNA viruses whose primary site of replication is the cytoplasm, encode special proteins to target nucleolus [
49]. It would be very interesting to test the link between nucleolar targeting of these viruses and ribosome biogenesis of their hosts.
Besides ribosomal genes, malfunction of nucleolus may be responsible for altered expression of many other genes detected in this study. For example, emerging evidence suggests that nucleolus might play a role in the small interfering RNA (siRNA) pathway [
47,
50]. Therefore, many genes controlled by siRNAs may be altered because of malfunction of nucleolus in RDV infected rice. Consistent with this, we found a large number of genes encoding transposon/retrotransposon-related proteins in the DEGs (Figure
1). It is well known that tansposon or transposon-related genes are transcriptionally controlled by epigenetic modifications, in which siRNAs play an important role [
51]. To our knowledge, altered expression of transposon/retrotransposon-related genes has never been reported in virus infected plants. However, we do not favor the possibility that this is specific to RDV. Instead, DEGs were classified automatically using web-based tools in most previous studies. In this way, transposon/retrotransposon-related genes tend to be classified into “unknown” genes and be excluded for further analysis.
Materials and methods
Sources of virus and insects
RDV Fujian isolate, China, was maintained in “Taizhong-1” rice plants grown in greenhouses at 25 ± 3°C, 55 ± 5% RH and under natural sunlight. Insects (Nephotettix cincticeps) source: high infectious green rice leafhoppers cultured in our lab with five generations of artificial rearing on rice seedlings.
Plant growth and inoculation
Seeds (Oryza sativa L. ssp. indica cv Yixiang2292) were sowed and germinated on a pot (60 mm in diameter and 50 mm in height) that had been filled with commercial soil mixture (FAFARD SOILS, Southern Agricultural Insecticides Inc Palmetto, FL, 34221). Rice seedlings were subjected to a two-day inoculation using high infectious green rice leafhoppers or virus-free insects (for mock inoculation) by the one test tube-one-seedling method. Inoculated seedlings were transplanted to an iron dish filled with cultivation layer soil of experimental farmland. They were kept in a south-facing greenhouse at 25 ± 3°C with 55 ± 5% RH and under natural sunlight. The aerial parts of 8 entire rice plants were sampled randomly and pooled at 22 dpi, i.e., 7d after appearance of the symptom (the earliest symptoms, i.e. white chlorotic specks in newly developed leaves, appeared at approximately 15 dpi). The samples were flash-frozen in liquid nitrogen, and stored at -80°C for until use.
RNA preparation and microarray hybridization and scanning
Total RNA was extracted from the virus- or mock-inoculated leaves with TRIzol reagent (Invitrogen). RNA was further purified using RNeasy columns (Qiagen, Valencia, CA, USA). An aliquot of 2 μg of total RNA was used to synthesize double-stranded cDNA, then produced biotin-tagged cRNA using MessageAmp™ II aRNA Amplification Kit. The resulting bio-tagged cRNA were fragmented to strands of 35 to 200 bases in length according to Affymetrix's protocols. The fragmented cRNA was hybridized to Affymetrix Rice Genome Array containing 51,279 transcripts which includes approximately 48,564
japonica transcripts and 1,260 transcripts representing the
indica cultivar (
http://www.affymetrix.com). Hybridization was performed at 45°C with rotation for 16 h (Affymetrix GeneChip Hybridization Oven 640). The GeneChip arrays were washed and then stained (streptavidin-phycoerythrin) on an Affymetrix Fluidics Station 450 followed by scanning on GeneChip Scanner 3000. We altogether used 6 chips to perform the analysis of 6 RNA samples.
Microarray data analysis
Hybridization data were analyzed using GeneChip Operating software (GCOS 1.4). The scanned images were firstly assessed by visual inspection then analyzed to generate raw data files saved as CEL files using the default setting of GCOS 1.4. A global scaling procedure was performed to normalize the arrays using dChip software. In a comparison analysis, two class unpaired method was applied in the Significant Analysis of Microarray (SAM) software to identify significantly differentially expressed genes between Test group and Control group. All differentially expressed genes were analyzed using the web-based Molecular Annotation System 3.0 (MAS 3.0,
http://bioinfo.capitalbio.com/mas/). MAS 2.0 integrate three different open source pathway resources-KEGG, BioCarta and GenMAPP. In the MAS 3.0 tool, the pathways and GO were ranked with statistical significance by calculating their
P-values based on hypergeometric distribution. The GeneChip hybridization and scanning were performed at the Microarray Resource Laboratory at Beijing CapitalBio Corporation, Beijing, China, in which GeneChip microarray service was certificated by Affymetrix.
Transient expression in leaves of N. benthamiana
Agro-infiltration for transient expression in leaves of
Nicotiana benthamiana, Leuzinger was carried out as described [
52]. Briefly, individual
Agrobacterium GV3101 strains with different expression constructs (or empty vector as control) were co-infiltrated into
N. benthamiana leaves using a syringe without needle. After 3 day of transient expression, leaves were harvested for RNA extraction.
Real-time PCR assay
Total RNA used for verification of microarray data was prepared from plants that had been grown independently of those used for isolation of RNA for microarray analysis. One Step RNA PCR Kit (AMV) (TaKaRa, Japan) was used. Gene-specific primers were designed by Primer 5 (for a list of the primers used in this study, see Additional file
3: Table S3) and synthesized by Boya Company (Shanghai, China). Relative quantitation method was used. Rice UBQ11 gene and tobacco EF-1α were used as the control to normalize all data [
53] (for a list of these genes primer sequence, see Additional file
3: Table S3).
Electron microscopy
For electron microscopy experiments, RDV infected and health rice samples were fixed with 2.5% glutaraldehyde at 4°C overnight, washed in 0.1 M phosphate buffer (pH 7.0) for 3 times (15 min per time), and post-fixed in phosphate-buffered 1.0% OsO4 for 2 h. Then the tissues were buffer-washed, dehydrated with ethanol (50%, 70%, 80%, 90%, 95% and 100%) and embedded in Epon-Araldite. Ultrathin sections (70–90 nm) were cut with a Reichert ultra-microtome, stained with aqueous uranyl acetate and lead citrate, and examined with a Jeol JEM-1230 transmission electron microscope (Jeol, Tokyo, Japan).
Acknowledgements
We thank Guangpu Li for his critical reading of the manuscript. This work was supported by grants from the National Basic Research Program 973 (2014CB138402, 2013CBA01403 and 2010CB126203), Natural Science Foundation of China (31272018, 31201491 and 31171821), Key Project of the National Research Program of China (2012BAD19B03), Doctoral Fund of Ministry of Education of China (20113515110001, 20113515120004 and 20123515120005), Education department of Fujian Province Office Program (JA11078 and JA11080) and Natural Science Foundation of Fujian Province of China (2011 J05051 and 2013 J01089).
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
Conceived and designed the experiments: JGW, ZJW and LHX. Performed the experiments and analyzed the data: JGW, LY, ZGD and KCW. Wrote the paper: JGW, ZGD, LY. All authors read and approved the final manuscript.