Differential proteomics profiling of the ova between healthy and Rice stripe virus-infected female insects of Laodelphax striatellus

Rice stripe virus-infected females of the small brown planthopper (SBPH, Laodelphax striatellus) usually lay fewer eggs with a longer hatch period, low hatchability, malformation and retarded or defective development compared with healthy females. To explore the molecular mechanism of those phenomena, we analyzed the differential proteomics profiling of the ova between viruliferous and healthy female insects using an isobaric tag for relative and absolute quantitation (iTRAQ) approach. We obtained 147 differentially accumulated proteins: 98 (66.7%) proteins increased, but 49 (33.3%) decreased in the ova of the viruliferous females. RT-qPCR was used to verify the 12 differential expressed proteins from iTRAQ, finding that trends in the transcriptional change for the 12 genes were consistent with those at the proteomic level. Differentially expressed proteins that were associated with meiosis (serine/threonine-protein phosphatase 2B and cyclin B3) and mitosis (cyclin B3 and dynein heavy chain) in viruliferous ova may contribute to low hatchability and defective or retarded development. Alterations in the abundance of proteins involved in the respiratory chain and nutrition metabolism may affect embryonic development. Our study begins to explain macroscopical developmental phenomena and explore the mechanisms by which Rice stripe virus impacts the development of SBPH.

Identification of differentially expressed proteins between viruliferous and healthy ova by iTRAQ. Differentially expressed proteins between RSV-infected and healthy ova were identified and quantified by 2-plex iTRAQ labeling and LC-MS/MS analysis, respectively (Fig. 2). Based on the LC-MS/MS analysis, 334 proteins were identified from the viruliferous and healthy ova. Among those proteins, 147 were differentially accumulated between the two samples (false discovery rate [FDR] < 0.01, fold changes > 1.2 or < 0.83): 98 (66.7%) increased and 49 (33.3%) decreased under the RSV-infection condition. Detailed information on the differentially expressed proteins, accession numbers and ratios are showed in the Table 1. Bioinformatics analysis. To understand the differentially accumulated proteins, the proteins were disposed by bioinformatic tools. All differentially expressed proteins were submitted to Uniprot (http://www.uniprot.org) for gene annotation, including molecular function, biological process and cellular component. For both upregulated and downregulated proteins, the main molecular functions were binding and ATP binding. According to biological process, upregulated proteins were mainly classified as development/growth, translation and response, while downregulated proteins were mainly in development/growth, metabolic process, and mitotic process. The cellular component of upregulated and downregulated proteins was mainly cytoplasm, nucleus and membrane. Detailed information can be found in Figs 3 and 4. We analyzed pathways of the differential proteins through the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.kegg.jp/kegg/). In the upregulated proteins, the main pathways were related to ribosome, signaling pathway and metabolism process. Similarly, Figure 2. Experimental workflow. The viruliferous and healthy female SBPHs were dissected when they reached the 4 th peak hatching period when ova were mature, and viruliferous and healthy ova samples, respectively, were collected and lysed. Differentially expressed proteins were quantified relatively using iTRAQ labeling (tags 113 and 114, respectively) and analyzed by LC-MS/MS. At the end of the study, we conducted a general bioinformatics analysis to provide a complete list of RSV-responsive proteins in the ova and verified some proteins by RT-qPCR. Continued among the downregulated proteins, pathways were the dominant grouping sector for proteins involved with metabolism process, signaling pathway and biosynthesis (Fig. 5).
Validation of the proteomics data at the RNA level by RT-qPCR. From a proteomics perspective, we found many proteins that differentially accumulated in ova infected with RSV compared with uninfected ova.
To evaluate the proteomic data and correlation between mRNA transcription level and protein abundance, we performed RT-qPCR to quantify the mRNA transcript level for 12 proteins that were selected according to the proportion of their up-and downregulation and availability of the mRNA sequence from the SBPH transcriptome 20 ( Table 2). The biological processes of those proteins are mainly metabolic process (Ls-ACC, Ls-Vha68, Ls-OGT  and Ls-Idh), cell cycle (Ls-Pp2B-14D and Ls-Dhc64C), response (Ls-Rpt6), and transport (Ls-Arf102F and Ls-Dhc64C). The trend in transcriptional variation for all the selected proteins was consistent with the proteomic changes determined in the iTRAQ analysis, suggesting that iTRAQ is a reliable way to identify and quantity the expressed differentially proteins of the SBPH (Fig. 6).

Discussion
For nonparthenogenetic insects, such as Drosophila melanogaster and SBPH, the development of the mature egg is arrested during meiosis prophase I and suspended in its metabolism and cell cycle until the insect ovulates 21 . The eggs are reactivated by mechanical stimulation and hydration resulting from passage of the egg within the narrow oviducts. First, they complete meiosis, and the male and female pronuclei become integrated to form the zygote, which then undergoes rapid cleavage cycles in which the nucleus divides without cytokinesis. These processes, from the end of prophase I to the rapid cleavage cycle 13, still use the maternal mRNA and proteins; zygotic transcription does not occur until the mid-blastula transition (MBT), when the divided nucleus becomes separated by a membrane and independent cells form 22 . Considering these processes in the context of the results of our study, we deduced that meiosis in viruliferous ova is likely to be disturbed by the decreased level of G2/mitotic-specific cyclin B3 and the increased level of serine/threonine-protein phosphatase 2B catalytic subunit 2 (Pp2B-14D). Cyclin-B3, a positive regulatory subunit of the cyclin-dependent kinase, is correlated with female fertility. In the crab ovary, increased expression of cyclin B during late vitellogenesis and final maturation of ova is considered to associated with meiotic maturation of the oocyte 23 . In addition, a mutation in cyclin B3 in Drosophila females leads to abnormal oogenesis, fewer eggs laid and malformed embryos 24 . In Caenorhabditis elegans, the loss of cyclin B3 in embryos through RNA interference (RNAi) blocks meiosis II and causes meiotic defects 25 . Pp2B-14D is a subunit of calcineurin that is necessary for meiotic progression beyond metaphase I 26 . Overexpression of a persistently active form of Pp2B-14D in a Drosophila female germline also causes meiotic defects 27 . Therefore, the repression of maternal cyclin-B3 and the accumulation of Pp2B-14D induced by RSV may delay the completion of meiosis and lead to defective eggs, thus explaining the low hatchability and defective development.
Our study also showed that the rapid cleavage cycles of the embryonic nucleus may be disturbed by lowered level of cell cycle-related proteins such as cyclin B3 in viruliferous ova. Cyclin B3 controls the transition from metaphase to anaphase and participates in many cell cycle events such as the timely progression of mitotis and the onset of anaphase. Knockdown of cyclin B3 in Caenorhabditis elegans embryo results in a longer prophase and prometaphase, and a prolonged delay in metaphase 25 . Loss of mitotic cyclins during cleavage cycles 8 and 9 in Drosophila as a result of RNAi causes various mitotic defects and even nuclear arrest 28 . Dynein heavy chain, cytoplasmic (Dhc64C), which was also downregulated in the viruliferous ova of SBPH, is related to spindle formation, movement of chromosomes in prometaphase and anaphase A and control of the timing of the onset of anaphase 29 . The absence of cytoplasmic dynein by RNAi leads to metaphase arrest and mitotic defects such as anaphase delay and chromosome misalignment in the Drosophila S2 cell line 30 . Therefore, reduced levels of cyclin B3 and Dhc64C in the viruliferous ova suggest that RSV infection may impair and arrest mitosis, which may also contribute to delayed or defective development of eggs from viruliferous females.
Previous studies have suggested that the mitochondria genes are still silenced before MBT, which means that the embryo utilizes the mitochondrial transcripts and proteins of the ova to perform respiratory-chain function 31 . But viral infection disturbs the morphology, location and respiratory chain of the mitochondria 32 . Similarly, in our study, various proteins were upregulated, e.g., many electron-transport-chain related proteins such as two components of NADH dehydrogenase (NADH dehydrogenase iron-sulfur protein 7 and 2) and cytochrome c oxidase subunit II in viruliferous ova, indicating a disorder in respiratory chain and oxidative phosphorylation compared with the healthy sample, which may impact the synthesis of ATP and influence subsequent fertilization, cleavage and embryonic development 33 . Changes in the respiratory chain caused by RSV infection may also induce the production of reactive oxygen species (ROS) beyond the level of the antioxidant system 34 . Because of the lack of mitochondrial DNA-protecting proteins and poor restoration mechanism, mitochondrial DNA (mtDNA) adjacent to the respiration chain becomes the preferred target of ROS 35 . The impairment of mtDNA may also induce the release of an apoptotic factor and result in cellular apoptosis 36 .
The yolk of the ovum contains nutrients including protein, lipid and glycogen that are transported from the nurse cells and fat body to support eventual embryonic development 37,38 . RSV infection induced changes in 26 proteins (17.7%) participating in metabolic processes, including protein, lipid and glycogen metabolism, which may affect nutrient utilization for embryonic development (Table 1). This alteration to nutrition may contribute to the upregulation of forkhead box protein O (FoxO), which functions to handle nutrient changes during development by regulating the insulin signaling pathway and cooperating with the cAMP pathway 39 . In Drosophila melanogaster, 28% of the nutrition-related genes are regulated by FoxO. Moreover, FoxO may induce a decrease in cell number by mediating insulin signaling and inhibit development through the gene d4E-BP 40 , which may prolong the hatch period.
Heat shock 70 kDa proteins (Hsp70) regulate translocation, the assembly and folding of proteins and inhibit the caspase-dependent apoptosis 41 . Our result showed that RSV induced the expression of Hsp70. The high level of Hsp70 is considered to improve the survival rate during hyperthermia, but may lead to the over-stimulation or -inhibition many signaling pathways related to cell multiplication, maturation and apoptosis, which will eventually influence development, growth and survival 42,43 .
In our study, the change in expression of some proteins may be due to aiding in replication of RSV. Without any system for protein synthesis, RSV needs to take over the ribosomal proteins (RPs) and enzymes of the host for translation of viral proteins and replication 44 . Guanine nucleotide-binding protein (RACK1), a protein that was higher in viral sample, was demonstrated to be a cellular factor that aids virus infection through an internal ribosome entry site and contributes to virus translation and replication in Drosophila melanogaster 45 . Through   a comparative analysis of the transcriptome of SBPH, the level of two RPs was shown to increase in viruliferous insects 20 . Among the differentially expressed proteins in our study, 20 RPs were identified, and 18 (90%) accumulated to a high level in the viruliferous samples, suggesting that RPs might play key roles in viral protein synthesis for RSV duplication, indicating that RSV may proliferate and accumulate uninterruptedly in ova. Some of the enriched RPs, such as RpS23, RpL11, RPL21, RpL8, and other non-RP protein (hsp83, cyclin B3 and tubulin) also participate in the duplication of the centrosome or in the centrosome cycle 28,46,47 . The number of centrosomes has a significant impact on the number of spindle poles and accurate chromosome segregation 48 . Therefore, superabundant RPs and centrosome-related proteins for SBPH may lead to an alteration in the centrosome number, which will lead to mono-or multi-polar spindles and failure of chromosome segregation and errors in cell division 49,50 . In our study, some proteins were upregulated probably to assist virus transmission in the ovum or early embryo. Transport protein Sec61 and transitional endoplasmic reticulum ATPase TER94, both upregulated in our study, participate in virus entry and infection in insects and mammals 51 . Three subunits of vacuolar ATPase (subunit E and D and catalytic subunit A) also accumulated in the viral sample; vacuolar ATPase is involved in viral entry, in releasing nucleic acid, replication and proper folding of viral proteins 52 . Meanwhile some enriched antiviral proteins were found in the differentially expressed proteins, which may inhibit the influence of RSV before or after spawning. Putative ATP-dependent RNA helicase me31b was identified as performing an antiviral function in both insect cells and adult flies 53 . Ras-related protein Rab6 regulates phagocytosis to inhibit virus infection through actin reorganization in Drosophila melanogaster and shrimp 54 .
In conclusion, we obtained 147 differentially expressed proteins, 98 were upregulated and 49 downregulated, in the ova between viruliferous and healthy female insects of L. striatellus through the iTRAQ method.
Determining the variations in proteins should help us to understand the effect of RSV on SBPH ova at the proteomic level and explain phenomena such as low hatchability, developmental retardation and defects. But some protein changes induced by RSV cannot be explained well and need further experimental study. Our analysis of changes in proteins in the RSV-infected SBPH ova provides insights into the molecular mechanisms underlying RSV-induced phenomena.

Methods
Insect rearing and determining the infection rate of ova. Viruliferous and healthy SBPH were raised in separate glass beakers that each contained about 15 rice plants. The glass beakers were placed in an incubator at 28 °C with 16 h light/8 h dark. Rice plants are replaced with new ones every week to supply adequately nutrients. To verify whether the ova laid by viruliferous females were infected, an individual viruliferous female insect was allowed to feed in a glass beaker, and 16 offspring (3 rd instar nymphs) were collected and checked for RSV through a dot blot immunobinding assay using a monoclonal antibody against RSV and the method of Wang 55 .
Protein extraction and proteinase digestion. After the female SBPH reached the 4 th peak hatching period 56 , we excised the ova from the ovary of the females. The samples were dissolved in moderate lysis buffer (7 M carbamide, 2 M thiocarbamide, 0.1% CHAPS) and suspended for several seconds, then broken by ultrasonication (1.2 s on, 2 s off) and then incubated at room temperature for 30 min before being centrifuged at 15,000 × g for 20 min at 4 °C. The supernatant was then transferred into a new tube. The Bradford method was used to measure the protein concentration 57 . After overnight digestion in 50 μ L trypsin solution at 37 °C, viruliferous and healthy samples were labeled with iTRAQ reagents 113 and 114 (AB Sciex, Foster City, USA), respectively.

2DLC-MS/MS analysis.
The labeled peptide fragments from each sample were reconstituted with mobile phase A (98% ddH 2 O, 2% acetonitrile, pH 10) and pre-separated with mobile phase B (98% acetonitrile, 2% ddH 2 O, pH 10) using RIGOL L-3000 High performance Liquid Chromatography system with an RP analytical column (Durashell-C18, 4.6 mm × 250 mm, 5 μ m, 100 Å) at 0.7 mL min −1 .Then the peptides were redissolved in 2% methyl alcohol and 0.1% formic acid and subsequently separated using a ABI-5600 system (Applied Biosystems) with an EASY-Spray column (12 cm × 75 μ m, C18, 3 μ m) at 350 nL min −1 . Mobile phase A and mobile phase B were 100% H 2 O with 0.1% formic acid and 100% acetonitrile with 0.1% formic acid, respectively. Verification by real-time PCR. We used Bioedit software (version: 7.2.5.0) to create a local transcriptome of SBPH based on sequencing data assembled from SRX016333 and SRX016334 of NCBI by RunAssembly in the program Newbler (version 2.6) 60 . The 12 selected protein sequences, based on iTRAQ data, were downloaded through the accession number from the NCBI and subjected to a tBlastn similarity search against the local transcriptome. Retrieved gene sequences with an expectation value (E) less than 10 −10 were considered Scientific RepoRts | 6:27216 | DOI: 10.1038/srep27216 to be credible and were used to design the RT-qPCR primers using the program Primer Premier Version 5.0. Viruliferous and healthy samples were triturated in TRIzol (Invitrogen) to extract total RNA. With the FastQuant RT Kit (TIANGEN), 1000 ng RNA was reverse-transcribed to synthesize cDNA. RT-qPCR was performed using the SYBR Green SuperReal PreMix (TIANGEN) with the ABI 7500 Real Time PCR thermal cycler (Applied Biosystems) and the following cycle program: 15 min at 95 °C, followed by 40 cycles of 10 s at 95 °C, 32 s at 60 °C and 72 °C for 32 s. β-actin was selected as a reference gene to normalize the expression level of target genes. Relative gene expression was computed using the 2 −ΔΔCT method 61 . The experiments were repeated 3 times independently. Healthy samples were used as a negative control. The primers used for the RT-qPCR to verify the iTRAQ result are shown in Supplementary Table S1.