Comparative proteomic investigation of drought responses in foxtail millet

Pan et al. BMC Plant Biology (2018) 18:315<br /> https://doi.org/10.1186/s12870-018-1533-9<br /> <br /> <br /> <br /> <br /> RESEARCH ARTICLE Open Access<br /> <br /> Comparative proteomic investigation of<br /> drought responses in foxtail millet<br /> Jiaowen Pan1, Zhen Li1, Qingguo Wang1, Anna K. Garrell5, Min Liu4, Yanan Guan2,3, Wenqing Zhou5 and<br /> Wei Liu1,3*<br /> <br /> <br /> Abstract<br /> Background: Foxtail millet (Setaria italica L. P. Beauv) has been considered as a tractable model crop in recent years<br /> due to its short growing cycle, lower amount of repetitive DNA, inbreeding nature, small diploid genome, and<br /> outstanding abiotic stress-tolerance characteristics. With modern agriculture facing various adversities, it’s urgent to<br /> dissect the mechanisms of how foxtail millet responds and adapts to drought and stress on the proteomic-level.<br /> Results: In this research, a total of 2474 differentially expressed proteins were identified by quantitative proteomic<br /> analysis after subjecting foxtail millet seedlings to drought conditions. 321 of these 2474 proteins exhibited significant<br /> expression changes, including 252 up-regulated proteins and 69 down-regulated proteins. The resulting proteins could<br /> then be divided into different categories, such as stress and defense responses, photosynthesis, carbon metabolism,<br /> ROS scavenging, protein synthesis, etc., according to Gene Ontology annotation. Proteins implicated in fatty acid and<br /> amino acid metabolism, polyamine biosynthesis, hormone metabolism, and cell wall modifications were also identified.<br /> These obtained differential proteins and their possible biological functions under drought stress all suggested that<br /> various physiological and metabolic processes might function cooperatively to configure a new dynamic homeostasis<br /> in organisms. The expression patterns of five drought-responsive proteins were further validated using western blot<br /> analysis. The qRT-PCR was also carried out to analyze the transcription levels of 21 differentially expressed proteins. The<br /> results showed large inconsistency in the variation between proteins and the corresponding mRNAs, which showed<br /> once again that post-transcriptional modification performs crucial roles in regulating gene expression.<br /> Conclusion: The results offered a valuable inventory of proteins that may be involved in drought response<br /> and adaption, and provided a regulatory network of different metabolic pathways under stress stimulation.<br /> This study will illuminate the stress tolerance mechanisms of foxtail millet, and shed some light on crop<br /> germplasm breeding and innovation.<br /> Keywords: Foxtail millet (Setaria italica L.), Drought stress, Comparative proteomics, Expression pattern,<br /> Western blot, qRT-PCR<br /> <br /> <br /> Background minerals, especially when compared to those of rice,<br /> Foxtail millet (Setaria italica L.) is an ancient crop in wheat, and maize, it was named first among the “Five<br /> the subfamily of Panicoideae, and is distributed world- Grains of China” due to its high nutritional values [1].<br /> wide in arid and semi-arid regions. It originated in Foxtail millet possess most noticeable morphological<br /> North China, and was domesticated more than 8,700 and anatomical attributes, such as dense root distribu-<br /> years ago. As foxtail millet grains are rich in protein and tion, thick cell wall, small leaf area and epidermal cell ar-<br /> rangement, which endow it with strong drought<br /> tolerance and high water use efficiency, and further<br /> * Correspondence: wheiliu@163.com<br /> 1<br /> Biotechnology Research Center, Shandong Academy of Agricultural<br /> allow it to be primarily cultivated in arid, semi-arid, and<br /> Sciences; Key Laboratory of Genetic Improvement, Ecology and Physiology of barren regions. Today, foxtail millet is attracting more<br /> Crops, Jinan 250100, Shandong, China attention in agricultural production, especially as global<br /> 3<br /> College of Life Sciences, Shandong Normal University, Jinan 250014,<br /> Shandong, China<br /> warming and lacking water resources become increas-<br /> Full list of author information is available at the end of the article ingly severe in the world [2].<br /> © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0<br /> International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and<br /> reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to<br /> the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver<br /> (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.<br /> Pan et al. BMC Plant Biology (2018) 18:315 Page 2 of 19<br /> <br /> <br /> <br /> <br /> Additionally, foxtail millet carries attractive qualities responsive proteins were identified. Bioinformatic analysis<br /> such as a small diploid genome (~490 Mb), inbreed- revealed that these differential proteins may take part in<br /> ing nature, less repetitive DNA, short growing cycle various biological processes. These biological processes<br /> and abiotic stress-tolerance [3]. As opposed to other may function synergistically by initiating different re-<br /> proximal plants, such as pearl millet, switchgrass, and sponse mechanisms on the protein level to reconfigure<br /> napiergrass, these features highlight it as a model and achieve new homeostasis in drought conditions. Our<br /> crop for exploring the mechanisms of drought toler- results begin filling the gap in our knowledge regarding<br /> ance, evolutionary genomics, architectural traits, C4 the proteomic activity and regulated response mechanisms<br /> photosynthesis and the physiology of bioenergy crops under drought conditions in foxtail millet, which will fur-<br /> [4]. Recently, the genome sequences of foxtail millet ther deepen the understanding of the physiological and<br /> cultivars“Yugu1” and “Zhanggu” have been sequenced molecular basis of stress tolerance in crops.<br /> and submitted by the US Department of Energy Joint<br /> Genomic Institute and Beijing Genomics Institute Materials and methods<br /> (BGI) of China, respectively [3, 5]. Plant materials and growth conditions<br /> After the genomic sequence of foxtail millet was re- The foxtail millet variety, Yugu1, which is known to be a<br /> leased, the vital stress-related gene families, such as the drought resistant variety and whose genome has been<br /> SiNAC, SiWD40 and SiALDH gene families, were sys- sequenced, was used for all experiments [15]. Plastic<br /> tematically analyzed and identified [6–8]. The SiPLDa1, pots (21 cm in diameter × 21 cm in height) were used as<br /> SiDREB2, SiNAC, SiOPR1, and SiAGO1b genes were all experimental units. Each pot was filled with 3-kg soil<br /> reported to mediate various stress responses and devel- consisting of a mixture of nutrient soil and loamy sand<br /> opmental processes during dehydration stress [6, 9–12]. in a ratio of 1:1. Plants were grown in greenhouse with<br /> Deep sequencing technology was also used to investigate well-watered conditions under 30/25 °C day/night cycle<br /> the genome-wide transcriptome reconfiguration of fox- with a 14-h photoperiod for three weeks. The drought<br /> tail millet under drought stress, and a great number of treatments were performed as previously described [16].<br /> differentially expressed genes (2,824), long noncoding Soil moisture of well-watered and drought-treated ex-<br /> RNAs (lncRNAs) and small interfering RNAs (siRNAs) perimental units was controlled at 60–70% and 20-30%<br /> were identified [13]. Under dehydration stress, 105 and of field capacity respectively, and the treatments lasted<br /> 84 differentially expressed genes were identified in fox- for 7 days. The pots were randomized in four replicates<br /> tail millet roots and shoots, respectively, and the re- between the two treatments. After drought treatments,<br /> sponses of genes involved in gluconeogenesis and seedlings were immediately harvested and frozen in li-<br /> glycolysis pathways took place earlier in roots compared quid nitrogen and stored at −80 °C for protein and RNA<br /> to shoots. Furthermore, the protein degradation pathway extraction, and then to perform proteomic, western blot<br /> may also perform a key role in drought tolerance of fox- and gene expression analysis.<br /> tail millet [11].<br /> Although drought-responsive genes and noncoding Protein extraction and trypsin digestion<br /> RNAs (ncRNAs) were identified, there have been hardly Protein extraction and trypsin digestion were performed<br /> any systematic investigation summaries of protein profil- according to a previously established method [17] with<br /> ing for drought stressed foxtail millet. Protein profiling slight modification. Samples were first grinded in liquid<br /> will contribute to the systematic scrutiny of changes in nitrogen, then the tissue powder in lysis buffer was<br /> protein levels and activities, and provide information ultra-sonicated three times. The remaining debris was<br /> about which proteins may participate in certain biological removed after centrifugation, and the protein was pre-<br /> processes. Recently, tandem mass tags (TMT), combined cipitated with 15% TCA and washed with cold acetone.<br /> with liquid chromatography−quadruple mass spectrom- The precipitates were re-suspended in buffer made of<br /> etry (LC−MS/MS) analysis, has been utilized as an useful 100 mM TEAB and 8 M urea at pH 8.0. The protein<br /> quantitative proteomic technique, which facilitates simul- concentration was determined using 2-D Quant kit (GE<br /> taneous identification and relative quantification of pro- Healthcare B80648356 Piscataway USA).<br /> teins with great efficiency and accuracy. This method is For protein digestion, the protein solution was reduced<br /> also widely used for quantitative comparative analysis of with 10 mM DTT and alkylated with 20 mM Iodoacetic<br /> plant proteomes [14]. In this study, the TMT combined acid. After that, the protein samples were diluted by<br /> with LC-MS/MS-based proteomic approach was used, adding 100 mM TEAB with urea concentration less than<br /> and the differentially expressed proteins in foxtail millet 2M. Finally, the protein was digested at 1:50 and 1:100<br /> seedlings after drought treatment were quantitatively trypsin-to-protein mass ratios with two replicates of<br /> identified. There were 2474 differential proteins that were each. For each sample, approximately 100 μg of protein<br /> quantitatively identified, among which, 321 drought were digested for the following experiments.<br /> Pan et al. BMC Plant Biology (2018) 18:315 Page 3 of 19<br /> <br /> <br /> <br /> <br /> TMT labeling and HPLC fractionation content, were performed as described previously [19]. The<br /> After trypsin digestion, the peptide was desalted by Glycine betaine (GB) content was measured according to<br /> Strata X C18 SPE column (Phenomenex) and vacuum- [20]. The detections of spermine and spermidine were car-<br /> dried. The peptide was reconstituted using 6-plex ried out according to the method of [21].<br /> TMT kit according to the manufacturer’s protocol<br /> and the sample was then fractionated according to Western blot analysis<br /> protocol of [17]. Total proteins were extracted from foxtail millet seed-<br /> lings in extraction buffer containing 100 mM Tris pH<br /> LC-MS/MS analysis 8.0, 5 mM EDTA, 1 mM PMSF and 0.2%<br /> The peptides were dissolved in 0.1% formic acid (FA), β-mercaptoethanol. Protein extracts were separated on<br /> and then directly loaded onto a reversed-phase pre-col- 12% SDS−PAGE gels and then transferred to a PVDF<br /> umn (Acclaim PepMap 100, Thermo Scientific). The membrane using a Mini Trans-Blot cell (Beijin Jun Yi<br /> peptide separation was performed using a reversed- electrophoresis equipment, China). The membranes<br /> phase analytical column (Acclaim PepMap RSLC, were first blocked with 10% bovine serum albumin in<br /> Thermo Scientific). LC-MS/MS analysis was performed TBST buffer (100mM NaCl2, 20mM Tris pH 8.0 and<br /> according to protocol of [17]. 0.5% Tween-20) for 2h, and then incubated with poly-<br /> clonal antibodies at a 1:1000 dilution for another 2h at<br /> Database searching room temperature. After washing three times with TBST<br /> The resulting MS/MS data were processed using Mascot buffer, the membranes were incubated with secondary<br /> search engine (v.2.3.0). Tandem mass spectra were antibody of horseradish peroxidase (HRP)-conjugated<br /> blasted against Uniprot_foxtail_4555 (http://www.uni- goat anti-rabbit IgG at a 1:1000 dilution for 2h. After<br /> prot.org/taxonomy/4555) database concatenated with re- three times further washes in TBST buffer,the mem-<br /> verse decoy database. Trypsin/P was specified as the branes were visualized with a 3,3′-diaminobenzidine<br /> cleavage enzyme allowing up to two missing cleavages. (DAB) detection system (Sangon Biotech, China). The<br /> Mass error was set to 10 ppm for precursor ions and western blot analysis was repeated twice. The SiActin<br /> 0.02 Da for fragment ions. Carbamidomethyl on Cys, (XM_004978702) was used as the internal control to<br /> TMT6plex (N-term) and TMT6plex (K) were specified quantify protein loading of different samples. Antibody<br /> as the fixed modifications, and oxidation on Met was production was conducted as described above. The pep-<br /> specified as the variable modification. FDR was adjusted tides of SiPPR (XM_004978236) (residues 186-508)<br /> to 20. and SiRLK (XP_004956304.1) (residues 1-386) were<br /> Gene Ontology (GO) annotation proteome was de- expressed and purified as described [22]. The purified<br /> rived from the UniProt-GOA database (www. http:// peptide was injected into rabbits for polyclonal anti-<br /> www.ebi.ac.uk/GOA/). The proteins were classified by body preparation. Using these polyclonal antibodies,<br /> Gene Ontology annotation based on three categories: only one band was detected at the corresponding pos-<br /> biological process, cellular component and molecular ition in the Western Blot. The antibodies for SiCAT<br /> function. The functional description of identified protein (XM_004985783), SiHSP70 (XM_004981194), SiTuBu-<br /> domains was annotated by InterProScan (a sequence lin (XM_004981865) and Actin were purchased from<br /> analysis application) based on protein sequence align- Agrisera, Sweden(product numbers were AS09 501,<br /> ment method, using the InterPro domain database. AS08 371, AS10 680, and AS13 2640, respectively.<br /> Kyoto Encyclopedia of Genes and Genomes (KEGG)<br /> database (http://www.genome.ad.jp/kegg/) was used to RNA isolation and quantitative real-time PCR<br /> annotate protein pathway: first, by using the KEGG on- Total RNA from foxtail millet was extracted with Trizol re-<br /> line service tool, KAAS, to annotate e the protein’s agent (TaKaRa) according to the manufacturer’s instruc-<br /> KEGG database description, then by mapping the anno- tions. The RNA electropherogram is shown in Additional<br /> tation results on the KEGG pathway database using file 1: Figure S1. First strand cDNAs were synthesized using<br /> KEGG online service tool, KEGG mapper. Wolf-psort the First Strand cDNA Synthesis kit (TaKaRa). Quantitative<br /> (http://www.genscript.com/wolf-psort.html) was used to Real-time PCR was performed in 7500 real-time PCR ma-<br /> predict the protein’s subcellular localization. The chine (Applied Biosystems) using the FastStart Universal<br /> Protein-Protein Interaction analysis was performed ac- SYBR Green (Roche) Master. The FastStart Universal SYBR<br /> cording to [18]. Green (Roche) Master is supplemented with ROX reference<br /> dye for background noise correction. Each PCR reaction<br /> Physiological parameters measurements was carried out with gene-specific primers in a total volume<br /> The measurements of physiological parameters, such as of 20μL containing 10μL SYBR Green Master mix, 0.5μM<br /> antioxidant enzyme activity and proline and soluble sugar gene-specific primers, and appropriately diluted cDNA.<br /> Pan et al. BMC Plant Biology (2018) 18:315 Page 4 of 19<br /> <br /> <br /> <br /> <br /> The foxtail millet actin gene SiActin was used as the in-<br /> ternal reference [7]. All primers were annealed at 56 °C.<br /> Each PCR reaction was repeated three times independently.<br /> Relative gene expression was calculated according to the<br /> delta-delta Ct method [7]. All primers are listed in Add-<br /> itional file 2: Table S1.<br /> <br /> Results<br /> Identification and quantification of drought-responsive<br /> proteins of foxtail millet<br /> After natural drought treatment for 7 days, the foxtail<br /> millet seedlings showed stunted growth, and yellowish,<br /> wilting and curled leaves compared to those of the un-<br /> treated control (Fig. 1). These seedlings were then used<br /> for quantitative proteomic analysis. A total of 4074 an-<br /> notated proteins were identified in two biological repli-<br /> cates (Additional file 3: Table S2). Among the identified Fig. 2 Pie chart showing the distribution of differentially expressed<br /> 4074 proteins, 2474 proteins were found in all four repli- proteins. T/C Ratio: relative fold change abundance of proteins in Foxtail<br /> millet under drought stress compared with control (cutoff of over 1.5<br /> cates and could be quantified (Additional file 4: Table<br /> for increased expression and less than 1/1.5 (0.67) for decreased<br /> S3). In published reports, the cutoff values of 1.2- to expression). a: Up-regulated proteins, b: Down-regulated proteins<br /> 1.5-fold change threshold and p-value (p < 0.05) was<br /> adopted [17, 23–25]. We adopted more stringently<br /> threshold of fold changes (cutoff of over 1.5 for in- increase of less than 2 folds. The variations of 60 de-<br /> creased expression and less than 1/1.5 (0.67) for de- creased proteins ranged from 0.67-0.5 fold compared to<br /> creased expression) and p-value