Transcriptomic and proteomic responses to herbivory in cultivated, Bt-transgenic and wild rice

Background: Strategies are still employed to decrease insect damage in crop production, including conventional breeding with wild germplasm resources and transgenic technology with the insertion of foreign genes, while the insect-resistant mechanism of these strategies remains unclear. Results: Under the feeding of brown planthopper ( Nilaparvata lugens ), cultivated rice (WT) showed less DEGs (568) and DAPs (4) than transgenic rice (2098 and 11) and wild rice CL (1990 and 39) and DX (1932 and 25). Hierarchical cluster of DEGs showed gene expression of CL and DX were similar, slightly distinct to GT, and clearly different from WT. DEGs assigned to the GO terms were less in WT rice than GT, CL and DX, and “Metabolic process”, “cellular process”, “response to stimulus” were dominant. Wild rice CL significantly enriched in KEGG pathways of “Metabolic pathways”, “biosynthesis of secondary metabolites”, “plant-pathogen interaction” and “plant hormone signal transduction”. The iTRAQ analysis confirmed the results of RNA-seq, which showing the least GO terms and KEGG pathways responding to herbivory in the cultivated rice. Synthesize conclusions: This study demonstrated that similarity in the transcriptomic and proteomic response to herbivory for the wild rice and Bt-transgenic rice, while cultivated rice lack of enough pathways in response to herbivory. Our results highlighted the importance of conservation of crop wild species.


Background
Insect damage is one of main factors in reducing agricultural production [1], and the methods of controlling insect pests in agricultural systems include the application of chemical pesticides [2], biological control with releasing the natural enemies of insects [3], high crop diversity and good farm practices [4], breeding and growing resistant varieties [5]. The use of chemical pesticides contaminated food and the environment [2,6]. Thus, germplasm and genomic resources from wild related species and have provided an opportunity for crop genetic improvement and are developed in controlling insect pests [1,7].
In addition, since the genetically modified (GM) crops have experienced to a rapid development, insect-resistant Bt-transgenic crops from bacterium Bacillus thuringiensis have been cultivated worldwide [8]. However, the adoption of GM crops receives concerns on the risk to human health and ecological environment. There are no consensus results of risk assessment with the development of GM plants, although numerous studies on risks of GM plants have been done. Environmental risk assessments include gene flow from GM crops to their wild relatives [9,10], potential risks on nontarget organisms [11,12], resistance evolution of target organisms [13][14][15]. Health risk assessments include substantial equivalence [16,17], non-targeted organisms on food/feed consumption [18,19], traceabiligy of transgenic proteins [20,21]. However, cisgenic and intragenic plants, modified with genetic material derived from the species itself or sexually compatible species, are expected to be out of regulation [22]. Thus, detecting potential insect-resistant genes or related pathways in wild species are key in developing insect-resistance varieties. "Omic" methods, transcriptomic, proteomic and metabolomic analysis, are usually used to investigate functional genes in plants. Transcriptomic profiling by RNA sequencing is becoming an attractive method as it facilitates rapid generation, identification and quantification of large number of transcripts. Global analysis of transcriptome in a species under different stress conditions will facilitate identification of specific pathways and genes responsible for tolerance against a particular stress.
Therefore, the objectives of this study are 1) to explore the physiological and molecular basis of insect herbivory responsiveness in cultivated rice (Oryza sativa), insect-resistant transgenic rice expressing Bacillus thuringiensis (Bt) and wild rice (O. rufipogon), and 2) to detect the variation of genes related plant response to herbivory in rice domestication and the insertion of a foreign gene.
Wild rice possesses a number of elite genes in resistance to insects, diseases, drought, and other stresses [23,24]. Insect-resistant Bt-transgenic rice has been developed and granted biosafety certificates in China [25]. Comparing cultivated and Bt-transgenic rice can understand a transcriptomic variation of the Bt transgene insertion, and comparative physiological studies can prove the superiority of wild rice genotypes over cultivated and Bt-transgenic rice in terms of insect resistance.

Results
Transcriptome profiling herbivory induced changes at transcriptome level and responsive metabolic pathways in rice seedlings, collected from control and herbivore stressed plants. After removing low quality reads, a total of > 1.17 × 10 7 reads were obtained (Table 1). About 81.44-88.18% of the reads were mapped to the indica rice reference genome, and 67.57-73.55% of the reads were assigned to the indica rice reference genes (Table S2). A total of 18104-18907 genes were identified and quantified (Table 1).   (Fig. 3). The high number of DEGs involved in "metabolic process", "cellular process" and "response to stimulus" in the biological process category indicated that rice leaves undergone exquisite metabolic activities and activated some protective mechanisms during the 72 h herbivory. "Catalytic activity", "binding" and "transporter activity" in the category of molecular function were prominent classes, and "cell", "cell part" and "organelle" dominant in the cellular component (Fig. 3).
Based on the criteria of Q-values < 0.05, a total of 21 KEGG pathways were significantly enriched in the four rice types ( Fig. 4; Table S3). CL significantly enriched the most KEGG pathways (18 pathways), with the dominant pathway "metabolic pathways (536 DEGs)" followed by "biosynthesis of  4). Two pathways "plant-pathogen interaction" and "amino sugar and nucleotide sugar metabolism" were enriched in both CL, DX and GT. The "biosynthesis of secondary metabolites" pathway was enriched in CL, DX and WT but not in GT (Table S3).

Validation Of Differential Expression Genes Through Qrt-pcr
Ten DEGs, enriched in the KEGG pathways of "plant hormone signal transduction" and "plantpathogen interaction", were confirmed by qRT-PCR, including three DEGs in ethylene (ET), two gibberellin (GA), one salicylic acid (SA), one auxin (AUX), one mitogen-activated protein kinases (MAPK), one brassinosteroid, one calmodulin (CALM). These regulatory elements involve in the networks of plants responding to herbivory [26]. The expression pattern of these DEGs in the qRT-PCR analysis showed the same trend as the RNA-Seq analysis did (Table 3). In RNA-Seq, differentially expressed genes (DEGs) were defined with the RPKM absolute value of log2Ratio ≥ 1 fold and false discovery rate (FDR) ≤ 0.001; "-" indicated down-regulated.
Venn diagram showed no common DAP for both the four genotypes, and one common DAP (gi|125544232) for CL, DX and GT (Fig. S3). Wild rice CL had four common DAPs with DX and two with WT respectively, and transgenic rice GT had one common DAP (gi|374277679) with DX (Fig. S3).
Photosynthesis and tryptophan metabolism KEGG pathways were significantly involved in DX plants responding to herbivory (Fig. 6).
DAPs in transgenic rice GT were significantly enriched in 74 GO terms, including 58 biological process, 4 cell component and 10 molecular function terms (Table S6) (Table S7). No significant KEGG pathway was found.

Comparison Of Rna-seq And Itraq
RNA-seq and iTRAQ analysis showed 33 common items in wild rice CL, three in wild rice DX, three in Bt-transgenic rice GT and four in cultivated rice WT. Among these items, based on DAPs (> 1.5 fold change) and DEGS (log2Ratio ≥ 1 fold and FDR ≤ 0.001), two common items (gi|115489014 and gi|53791994) in CL, one item in DX (gi|115450521), GT (gi|75248671) and WT (gi|115489014), were found (Table S8). The same pattern was found between RNA-seq and iTRAQ for gi|115489014 (upregulated) in CL, gi|115450521 (up-regulated) in DX and gi|75248671 (down-regulated) in GT. The other two items were opposite between RNA-seq and iTRAQ for gi|53791994 (up-regulated in DAP vs. down-regulated in DEG) in CL and gi|115489014 (down-regulated in DAP vs. up-regulated in DEG) in WT. These items involved in GO terms, including response to biotic stimulus, defense response, oxidation-reduction process, hydrogen peroxide catabolic process, chlorophyll biosynthetic process, photosynthesis and response to ethylene stimulus.

Discussion
Our results showed genes and proteins involved in plant response to brown planthopper (BPH, N. lugens) in four rice genotypes. A total of 19 genes resistance to BPH have been reported in cultivated and wild rice species and been used for rice breeding and production [1]. Although we here did not  Table S9, S10).
We found that the DEGs and DAPs in response to herbivory were mostly distinct from each other, which indicates that RNA-seq and proteomic methods need to be considered together in studying gene regulatory networks during the process of plants responding to biotic stresses. In fact, the expression of genes does not mean the related proteins will express [27]. An alternative reason is the low number of DAPs in this study, which resulted from the lack of multiple fractionation or separation methods that were usually used to improve signal-to-noise and proteome coverage and to reduce interference between peptides in quantitative proteomics [28].
The insertion of Bt gene may promote the response of plants to herbivory because GT showed more DEGs and DAPs than cultivated rice. Foreign genes can induce plants to motivate the expression of many other genes. Li et al. (2010) found genes related with defense response upregulated after induction of a foreign protein in tobacco [29]. However, it is difficult to understand the induction mechanism of foreign gene to other genes.
Our results support crop domestication weakened the response of plants to herbivory. We found that cultivated rice showed less DEGs and DAPs than wild rice. In addition, the plant-pathogen interaction process was specially enriched in the DEGs and DAPs of wild rice CL and DX, but not the cultivated rice (Figs. 4, 6). Number of those genes were much less in cultivated rice than in wild rice (Table S9, S10). Furthermore, wild rice showed strong change in the expression of genes involved in plant hormone signaling, but cultivated rice did not (Fig. 4). JA signaling plays a central role in the herbivory response. We detected that several JAZ genes (OsJAZ5, OsJAZ6, OsJAZ8, OsJAZ9 and OsJAZ12) were differentially expressed in wild rice during herbivory, but not in cultivated rice (Supplemental Table   S9). Ethylene signaling appears to be also particularly important for the herbivory response of wild rice as many genes involved in this pathway were specially differentially expressed in wild rice (Supplemental Table S9) It is widely known that cultivated rice (O. sativa) was domesticated from common wild rice [30], during which lots of functional genes were lost [31]. Crop domestication is considered to cause a genetic bottleneck and reduced genetic diversity throughout the genome, and genes that influence desirable phenotypes experiences a drastic loss of diversity [32]. The genetic diversity of cultivated rice is less than that of wild rice [33,34]. Contrast to wild rice, cultivated rice suffered less herbivory pressure because of huge pesticide application in fields [2]. The target genes of selection (e.g. herbivory) may decrease in nucleotide diversity. In addition, genomic recombination affects the influence of selection on genomic regions via genomic hitchhiking.
Within wild rice, the CL ecotype showed more DEGs, GO terms and KEGG pathways than the DX ecotype, which means the latter was less sensitive to the herbivory of N. lugens than the former. This could be explained by the difference in habitat conditions. The CL population locates in a remote hill with high plant diversity while the DX population was normally grown together with crops including the cultivated rice [35]. Plant-insect interaction depends on environmental abiotic and biotic factors [36]. Herbivore history, a strong driver of plants growth, affects not only plant response to herbivore attack but plant community parameters [37]. Our results indicate growing environment plays an important role in regulating gene networks of plant response to herbivory. Thus, the strategy of in situ conservation is key in the biological conservation of wild germplasm resources.

Conclusions
In conclusion, our results support that rice domestication lost some genes related plant response to The resulting clean reads were mapped to the reference genomes of indica rice 9311, using normalized by calculating the read per kilobase per million mapped reads (RPKM = total exon reads/mapped reads in million X exon length in kb) for each gene. The differentially expressed genes (DEGs) were defined with the RPKM absolute value of log2Ratio ≥ 1 fold and false discovery rate (FDR) ≤ 0.001 (Fig. S1). We performed cluster analysis of gene expression patterns with cluster software and Java Treeview software.
The DEGs were annotated according to molecular function, biological process, and cellular component

LC-MS/MS analysis
After desalted, the labeled peptides were subjected to the MS analysis using a commercial 5600 TripleTOFTM coupled with an Eksigent Nano LC-Ultra 1D plus HPLC system (Eksigent; Dublin, CA, USA). Nano LC was performed via a "trap and elute" configuration and the mobile phase included A

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Availability of data and material
All data generated during this study are included in this published article and its supplementary information files.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was financially supported by the National Environmental Protection Public Welfare Science and Technology Research Program of China (grant 201309038).

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.