Research Article Received: 17 November 2014
Revised: 2 February 2015
Accepted article published: 11 March 2015
Published online in Wiley Online Library: 1 April 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7166
Leaf proteome comparison of two GM common bean varieties and their non-GM counterparts by principal component analysis Pedro A Valentim-Neto, Gabriela B Rossi, Kelly B Anacleto, Carla S de Mello, Geisi M Balsamo and Ana Carolina M Arisi* Abstract BACKGROUND: A genetically modified (GM) common bean event, namely Embrapa 5.1, was approved for commercialization in Brazil. The present work aimed to use principal component analysis (PCA) to compare the proteomic profile of this GM common bean and its non-GM counterpart. RESULTS: Seedlings from four Brazilian common bean varieties were grown under controlled environmental conditions. Leaf proteomic profiles were analyzed by two-dimensional gel electrophoresis (2DE). First, a comparison among 12 gels from four common bean varieties was performed by PCA using volume percentage of 198 matched spots, presented in all gels. The first two principal components (PC) accounted for 46.8% of total variation. Two groups were clearly separated by the first component: Pérola and GM Pérola from Pontal and GM Pontal. Secondly, another comparison among six gels from the same variety GM and its non-GM counterpart was performed by PCA; in this case it was possible to distinguish GM and non-GM. CONCLUSION: Separation between leaf proteomic profile of GM common bean variety and its counterpart was observed only when they were compared in pairs. These results showed higher similarity between GM variety and its counterpart than between two common bean varieties. PCA is a useful tool to compare proteomes of GM and non-GM plant varieties. © 2015 Society of Chemical Industry Supporting information may be found in the online version of this article. Keywords: proteome; proteomics; Phaseolus vulgaris; GMO; Embrapa 5.1
INTRODUCTION
J Sci Food Agric 2016; 96: 927–932
∗
Correspondence to: Ana Carolina M Arisi, Food Science and Technology Department, Federal University of Santa Catarina, Rod. Admar Gonzaga 1346, 88034-001, Florianópolis – SC, Brazil. E-mail: [email protected] Food Science and Technology Department, Federal University of Santa Catarina, 88034-001, Florianópolis, SC, Brazil
www.soci.org
© 2015 Society of Chemical Industry
927
Common bean (Phaseolus vulgaris) is a source of protein in the diet of over a billion people worldwide. Golden mosaic of common bean is a disease caused by bean golden mosaic virus and it is one of the greatest constraints on common bean production in Latin America, with significant yield losses. RNA interference (RNAi) was used to silence the AC1 viral gene expression in transgenic common bean.1 In these genetically modified (GM) plants, the sequence-specific degradation of target mRNA interferes with viral replication, reducing or preventing viral DNA accumulation and, consequently, appearance of symptoms.1 Twenty-two GM lines were obtained with an intron-hairpin construct designed to induce post-transcriptional gene silencing of the AC1 gene. These lines were first evaluated under greenhouse conditions; two of these GM lines (named Embrapa 2.3 and Embrapa 5.1) showed high resistance upon inoculation with viruliferous whiteflies.2 In 2011 Brazilian National Technical Commission on Biosafety (CTNBio) approved Embrapa 5.1 event for cultivation and consumption in Brazil. This GM common bean was the first commercial GM plant developed in Latin America by the Brazilian Agricultural Research Corporation (Embrapa). A detailed molecular characterization of Embrapa 5.1 event was performed.3 Identification of the transgene insert confirmed the presence of a single locus
corresponding to two intact copies of the RNAi cassette in opposite orientation and three intact copies of the AtAhas gene. It is flanked by Phaseolus genomic sequences and interspersed by one nuclear and three chloroplastic genomic sequences.3 Embryo chemical composition of three Embrapa 5.1 common bean varieties and their non-GM counterparts (Pérola, Pontal and Olathe varieties) were compared by high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR) and this methodology was proposed to discriminate GM and non-GM common bean.4 Authors interested in the genetic content itself have analyzed embryo composition by HR-MAS NMR using principal component analysis (PCA), which is a useful tool for handling multivariate data. NMR and PCA confirmed the similarities in chemical profiles among the genotypes Olathe Pinto, Pérola and Pontal, as well as the three Embrapa 5.1 GM varieties, suggesting a natural grouping according to GM or conventional origin. The authors
www.soci.org stated that it was possible to evaluate genetic modification by comparing metabolic profiles of GM and non-GM common beans.4 Profiling techniques such as metabolomics, transcriptomics and proteomics have been suggested as non-targeted approaches for genetic modification analysis.5 – 10 Proteins are key players in gene function and are directly involved in metabolism and cellular development; thus proteome forms the central bridge between transcriptome and metabolome.11 Proteomic studies would provide important information for understanding changes in biological processes after genetic modification.12 Among other methods established in proteomics,13 – 15 two-dimensional gel electrophoresis (2DE) is a simple and frequently used technique.16 – 23 Proteomic analysis using 2DE has demonstrated the capacity to differentiate varieties on their protein content in several plants, such as common bean, soybean, maize, wheat and rice.6 – 18,24 – 33 In the present study, we used proteomic analysis and PCA to compare four Brazilian common bean varieties (Phaseolus vulgaris L. Embrapa 5.1 derived from the varieties Pérola and Pontal and their non-GM counterpart varieties). It is the first leaf proteomic analysis of this GM common bean.
MATERIAL AND METHODS Plant material Genetically modified common bean Phaseolus vulgaris L. Embrapa 5.1 derived from the varieties Pérola and Pontal and their non-GM counterpart varieties were provided by Embrapa Arroz e Feijão (Santo Antônio de Goiás, Goiás State, Brazil). This GM event contains an intron-hairpin construct designed to induce post-transcriptional gene silencing of the AC1 viral gene.1 Embrapa 5.1 shows high resistance to bean golden mosaic virus. Leaf material was obtained as follows: Nine seeds of each variety were placed on moistened germination paper and germinated in a growth chamber at 25 ∘ C in the dark for 3 days in May 2013. After germination, all 36 seedlings were grown side by side in pots containing soil in a controlled-environment growth chamber, adjusted to 12 h photoperiod, photosynthetic active radiation of 150 μmol m−2 s−1 , at 25 ∘ C, watered daily. After 7 days, leaves were collected and stored at −80 ∘ C. DNA extraction and PCR for GM common bean detection Genomic DNA was isolated from leaves, previously powdered in liquid nitrogen, using DNeasy Plant mini kit. Samples were subject to PCR using specific primers for Embrapa 5.1 GM common bean detection34 and for Phaseolus vulgaris detection.35 MON 810 GM maize and Roundup Ready GM soybean were used as negative control samples.
928
Protein extraction Total protein was extracted from a pool of six leaves from three plants for each variety. Leaves were ground to a fine powder in liquid nitrogen using a mortar and pestle, and 300 mg of each sample were extracted with 10 mL extraction buffer (0.5 mol L−1 Tris–HCl, pH 8.0, 100 mmol L−1 EDTA, 0.7 mol L−1 sucrose, 0.1 g L−1 CHAPS, 14 mmol L−1 DTT and 1 mmol L−1 PMSF). Proteins were precipitated overnight with 0.1 mmol L−1 ammonium acetate in cold methanol. After centrifugation, dried protein pellets were suspended in 300 μL of a solution containing 7 mol L−1 urea, 2 mol L−1 thiourea, 0.3 g L−1 CHAPS, 0.2 mL L−1 IPG buffer solution, pH 4–7, and 0.15 g L−1 DTT. Protein concentrations were determined using a 2D Quant kit (GE Healthcare). Three protein extracts were prepared from each variety.
wileyonlinelibrary.com/jsfa
PA Valentim-Neto et al. 2DE gel electrophoresis Leaf proteomic profiles were analyzed by two-dimensional gel electrophoresis (2DE). Each protein extract was used for each gel, so three 2DE gels were used per variety. Isoelectric focusing (IEF) was carried out using 13 cm pH 4–7 IPG strips (GE Healthcare). Strips were rehydrated for 18 h with a solution containing 250 μg total soluble protein diluted in rehydration buffer containing 0.2 mL L−1 IPG buffer pH 4–7 (GE Healthcare) in a total volume of 250 μL. After rehydration, strips were focused using an Ettan IPGphor 3 isoelectric focusing system (GE Healthcare) under the following conditions: step of 500 V until 500 V h, voltage gradients of 1000 V and 8000 V until 14500 V h, and a final step of 8000 V until 17 800 V h, up to a total of 34 000 V h, at a limit of 50 mA per strip. After focusing, strips were kept at −80 ∘ C for at least 18 h. The proteins in the IPG strips were subjected to reduction with 10 mg mL−1 DTT in 5 mL equilibration buffer (6 mol L−1 urea, 50 mmol L−1 Tris–HCl (pH 8.8), 3 mL L−1 glycerol, 0.2 g L−1 SDS and 2.5 mg L−1 bromophenol blue), followed by alkylation with 25 mg mL−1 iodoacetamide in the same buffer. The strips were then loaded on top of 12.5% polyacrylamide gel; a molecular weight marker 10–250 kDa (Precision Plus Protein Standards, BioRad) was used. Second-dimension SDS-PAGE gels were run in an SE 600 Ruby System (GE Healthcare) under 15 mA per gel for 30 min and 30 mA per gel for 4 h. The temperature was kept at 10 ∘ C using a MultiTemp III thermostatic circulator (GE Healthcare). Protein spots were visualized by staining the gels with 0.01 g L−1 Coomassie Brilliant Blue G-250 (Bio-Rad, Hercules, CA, USA) as described previously.36 Image and data analysis Gels were scanned in an Image Scanner System II and analyzed with ImageMaster Platinum software v. 7.0 (both from GE Healthcare) using automatic matching. Number of total spots was detected according to the following parameters: smooth ≥ 5, saliency ≥ 100 and area ≥ 4. Relative spot volumes (%Vol) were compared and analyzed by two approaches. In first approach, 12 gels of four varieties were compared all together. In the second approach, six gels of two varieties, each GM variety and its counterpart, were compared. All spots which matched in 12 gels or in six gels were selected and their volume percentage was log2 transformed. All data was median centered before analysis. PCA was performed using R language.37
RESULTS In this study, four common bean varieties were subjected to comparative leaf proteomic profiling. Seedlings from common bean varieties, Embrapa 5.1 GM Pérola and GM Pontal varieties (PerGM and PoGM) and non-GM Pérola and Pontal varieties (Per and Po), were grown side by side. In order to verify the GM identity of seedlings, specific PCR assays for Embrapa 5.1 detection and common bean detection were conducted. The two GM varieties were positive and the two non-GM varieties were negative for Embrapa 5.1 detection PCR assay, and the four varieties were positive for common bean detection PCR assay (supporting information, Fig. S1). Three protein extractions were carried out from leaves of each variety; therefore 12 extracts were prepared and 2DE gel was performed for each extract (Fig. 1). Proteomic analysis was performed in two approaches: first, all 12 gels were analyzed using Image Master software. An average of 513 ± 80 protein spots was detected in all 2DE maps, ranging from 422 to 689 total spots (Table 1). Average value of total protein spots for three gels from the same variety
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
Leaf proteome comparison of GM common bean by PCA
www.soci.org
Figure 1. Representative maps of common bean leaves protein profile from non-GM Pontal and GM Pontal varieties. For the first dimension, 250 μg of proteins were loaded on 13 cm IPG strips with pH 4–7 linear gradient; 12.5% SDS-PAGE gels were used for the second dimension. Gels were stained with Coomassie Brilliant Blue G-250.
Table 1. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM)
Table 3. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from two common bean varieties, Pérola (Per) and GM Pérola (PerGM). Gel
Gel Po1 Po2 (ref ) Po3 PoGM1 PoGM2 PoGM3 Per1 Per2 Per3 PerGM1 PerGM2 PerGM3
Total spots
Matched spots (%)
544 689 609 556 520 532 422 455 446 442 510 436
83 98 93 83 93 90 91 87 89 92 98 92
slope 1.26 – 1.04 1.13 0.79 0.92 0.66 0.62 0.71 0.61 0.71 0.67
0.83 – 0.9 0.77 0.84 0.84 0.7 0.74 0.76 0.72 0.73 0.78
Table 2. Average spot number, standard deviation and coefficient of variation of 2DE maps of leaves from four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM) Variety Po PoGM Per PerGM
Average total spots 614 536 441 463
SD 72.6 18.3 17.1 41.1
Coefficient of variation (%) 11.8 3.4 3.9 8.9
Matched spots (%)
slope
r
Per1 Per2 Per3 PerGM1 PerGM2 (ref.) PerGM3
617 583 574 616 715 594
95 90 91 91 96 90
0.99 0.9 0.91 0.91 0.94
0.88 0.93 0.94 0.91 0.91
Table 4. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from two common bean varieties: Pontal (Po), GM Pontal (PoGM) Gel
Total spots
Po1 Po2 (ref.) Po3 PoGM1 PoGM2 PoGM3
644 775 678 669 559 588
Matched spots (%) 89 97 91 92 89 87
slope 1.14 – 1.01 1.01 0.79 0.8
r 0.81 – 0.91 0.8 0.79 0.83
and the correlation coefficients of scatter plots ranged from 0.7 to 0.9 (Table 1). In the second approach, comparing matched spots of 2DE maps of two common bean varieties (Pérola and GM Pérola), the slopes of the best-fit line varied from 0.9 to 0.99 and the correlation coefficients of scatter plots were in the range 0.88–0.94 (Table 3). Comparing matched spots of 2DE maps of the others two common bean varieties (Pontal and GM Pontal), the slopes varied from 0.79 to 1.14 and the correlation coefficients of scatter plots ranged from 0.79 to 0.91 (Table 4).
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
929
ranged from 441 to 614 spots, presenting coefficients of variation lower than 12% (Table 2). Variability among the gels was analyzed by scatter plots of matched spots. In the first approach, comparing matched spots of all 2DE maps of all four common bean varieties in relation to one reference gel, the slopes of the best-fit line varied from 0.61 to 1.26 J Sci Food Agric 2016; 96: 927–932
Total spots
r
www.soci.org
PA Valentim-Neto et al.
Figure 2. PCA plots of 198 matched spots volume (%Vol) of 2DE gels from leaves of four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM).
In the first approach, comparing all 12 2DE maps, 198 matched spots were detected. PCA was performed using the percentage volume of these matched spots as dataset. These data were log2 transformed to obtain normalized data from volume percentage of matched spots (supporting information, Fig. S2) for subsequent analysis using R language. When all 12 gels were analyzed by PCA (198 matched spots), the first 11 components explained 100% of dataset variation and 33.6% of the variation being allocated to the first component. The first two principal components (PC1 and PC2) accounted for 46.8% of total data variation (Fig. 2). Proteomic profiles from Pérola and Pontal varieties were clearly separated by the first component. However, a clear separation between GM common bean Pontal and non-GM Pontal was not observed. In the same manner, profiles from GM Pérola and non-GM Pérola were very close together in PCA plot (Fig. 2). In the second approach, leaf proteomic profiles were analyzed in pairs: GM variety and its counterpart. Three gels of GM Pontal and three gels of Pontal varieties were compared and they presented 364 matched spots. PC1 and PC2 explained 56.2% of total data variability and, in this case, PCA separated GM and non-GM gels (Fig. 3). When six gels of GM Pérola and Pérola were analyzed, there were 425 matched spots. PC1 and PC2 accounted for 49.4% of total variability and likewise PCA separated GM and non-GM gels (Fig. 4).
Figure 3. PCA plots of 364 matched spots volume (%Vol) of 2DE gels from leaves of two common bean varieties: Pontal (Po) and GM Pontal (PoGM).
Figure 4. PCA plots of 425 matched spots volume (%Vol) of 2DE gels from leaves of two common bean varieties: Pérola (Per) and GM Pérola (PerGM).
DISCUSSION
930
Common bean (Phaseolus vulgaris L.) is part of the staple diet in many countries of Latin America, and is cultivated by small and large farmers in all regions of Brazil. Embrapa 5.1 was the first GM common bean approved for cultivation and consumption in Brazil.2 This first leaf proteomic analysis to compare by PCA two GM common bean varieties and two non-GM counterpart varieties revealed that there are strong similarities among the GM common bean varieties and counterparts. Three 2DE gels of leaf protein extracts from the same variety presented high similarity (Table 2). The first approach of the analysis consisted of scatter plots of matched spots among four common bean varieties; slopes of the best-fit line and the correlation coefficients (r) presented high variation (Table 1). However, in the second approach, slopes and r
wileyonlinelibrary.com/jsfa
were close to 1 when scatter plots of matched spots were analyzed among six gels of two common bean varieties, GM variety and its counterpart (Tables 3 and 4), presenting high homogeneity. In the first approach, two-dimensional gels were separated into two groups by PCA (Fig. 2); one group contained three gels of GM Pérola and three gels of non-GM Pérola samples, and the other group contained three gels of GM Pontal and three gels of non-GM Pontal samples. This result is in accordance to previous works which compared GM and non-GM varieties using proteomic analysis.18,27,38,39 In general, the Omics comparisons revealed that genetic modification has less impact on plant gene expression and composition than that of conventional plant breeding.40 On the other hand, it was possible to distinguish three Embrapa 5.1 GM varieties from three non-GM common bean varieties
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
Leaf proteome comparison of GM common bean by PCA (Pérola, Pontal and Olathe Pinto) by comparing their embryo metabolic profiles,4 different from our results obtained for two varieties of Embrapa 5.1 common bean by comparing leaf proteomic profiles. A clear distinction between GM and non-GM varieties was observed on the PCA of leaf proteome profile only in the second approach, which analysis was performed in pairs: GM and its counterpart (Figs 3 and 4). Comparison between GM Pontal and non-GM Pontal leaf proteome gels shows clear separation on PC1 (Fig. 3). The same pattern was also observed comparing GM Pérola with non-GM Pérola samples (Fig. 4). MON810 maize leaf proteome was compared among four GM Brazilian varieties and their four non-GM counterparts; GM maize proteomic profiles were similar to non-GM counterparts proteomes.27 Previous study of MON810 maize grain proteome concluded that GM and non-GM maize Spanish varieties presented virtually identical grain proteomic profiles.18 In a previous work, a set of potato transcriptome profiles were compared by PCA.10 An untargeted comparative analysis based on transcriptomic profiles was proposed as a complementary tool for food evaluation, for which the necessity of transcript identification was not required. In this study conducted on virus-resistant GM common bean varieties, it was demonstrated that proteomic profiles from GM and non-GM common bean leaves were grouped close to each other in the PCA plots when compared to proteomic profiles of two breeding varieties. It was also shown that PCA is a useful tool to compare proteomes of GM and non-GM plant varieties, without the necessity for protein identification.
ACKNOWLEDGEMENTS We would like to express our gratitude to Josias C Faria for providing the seeds of GM common bean Embrapa 5.1. The work was financially supported by CNPq grant 470683/2012-0. PAVN is the recipient of a CAPES PNPD postdoctoral fellowship; GBR and KBA, of a CNPq IC fellowship; CSM and GMB, a CAPES PhD fellowship; and ACMA, a CNPq PQ-2 fellowship.
SUPPORTING INFORMATION Supporting information may be found in the online version of this article.
REFERENCES
J Sci Food Agric 2016; 96: 927–932
8 Metzdorff SB, Kok EJ, Knuthsen P and Pedersen J, Evaluation of a non-targeted ‘Omic’ approach in the safety assessment of genetically modified plants. Plant Biol 8:662–672 (2006). 9 van Dijk JP, Leifert C, Barros E and Kok EJ, Gene expression profiling for food safety assessment: examples in potato and maize. Regul Toxicol Pharmacol 58:S21–S25 (2010). 10 van Dijk JP, de Mello CS, Voorhilijzen MM, Hutten RCB, Arisi ACM, Jansen JJ et al., Safety assessment of plant varieties using transcriptomics profiling and a one-class classifier. Regul Toxicol Pharmacol 70:297–303 (2014). 11 Salekdeh GH and Komatsu S, Crop proteomics: aim at sustainable agriculture of tomorrow. Proteomics 7:2976–2996 (2007). 12 Gong CY and Wang T, Proteomic evaluation of genetically modified crops: current status and challenges. Front Plant Sci 4:41 (2013). 13 Bindschedler LV and Cramer R, Quantitative plant proteomics. Proteomics 11:756–775 (2011). 14 van Wijk KJ, Challenges and prospects of plant proteomics. Plant Physiol 126:501–508 (2001). 15 Valdes A, Simo C, Ibanez C and Garcia-Canas V, Foodomics strategies for the analysis of transgenic foods. TrAC, Trends Anal Chem 52:2–15 (2013). 16 Ruebelt MC, Leimgruber NK, Lipp M, Reynolds TL, Nemeth MA, Astwood JD et al., Application of two-dimensional gel electrophoresis to interrogate alterations in the proteome of genetically modified crops. 1. Assessing analytical validation. J Agric Food Chem 54:2154–2161 (2006). 17 Zolla L, Rinalducci S, Antonioli P and Righetti PG, Proteomics as a complementary tool for identifying unintended side effects occurring in transgenic maize seeds as a result of genetic modifications. J Proteome Res 7:1850–1861 (2008). 18 Coll A, Nadal A, Rossignol M, Puigdomenech P and Pla M, Proteomic analysis of MON810 and comparable non-GM maize varieties grown in agricultural fields. Transgenic Res 20:939–949 (2011). 19 Hakeem KR, Chandna R, Ahmad A, Qureshi MI and Iqbal M, Proteomic analysis for low and high nitrogen-responsive proteins in the leaves of rice genotypes grown at three nitrogen levels. Appl Biochem Biotechnol 168:834–850 (2012). 20 Otani M, Taniguchi T, Sakai A, Seta J, Kadoyama K, Nakamura-Hirota T et al., Phosphoproteome profiling using a fluorescent phosphosensor dye in two-dimensional polyacrylamide gel electrophoresis. Appl Biochem Biotechnol 164:804–818 (2011). 21 Hew C-S and Gam L-H, Proteome analysis of abundant proteins extracted from the leaf of Gynura procumbens (Lour.) Merr. Appl Biochem Biotechnol 165:1577–1586 (2011). 22 Gutierrez-Sanchez G, Atwood J, Kolli VSK, Roussos S and Augur C, Initial proteome analysis of caffeine-induced proteins in Aspergillus tamarii using two-dimensional fluorescence difference gel electrophoresis. Appl Biochem Biotechnol 166:2064–2077 (2012). 23 Campos A, Puerto M, Prieto A, Camean A, Almeida AM, Coelho AV et al., Protein extraction and two-dimensional gel electrophoresis of proteins in the marine mussel Mytilus galloprovincialis: an important tool for protein expression studies, food quality and safety assessment. J Sci Food Agric 93:1779–1787 (2013). 24 Brandao AR, Barbosa HS and Arruda MAZ, Image analysis of two-dimensional gel electrophoresis for comparative proteomics of transgenic and non-transgenic soybean seeds. J Proteomics 73:1433–1440 (2010). 25 De La Fuente M, Borrajo A, Bermudez J, Lores M, Alonso J, Lopez M et al., 2-DE-based proteomic analysis of common bean (Phaseolus vulgaris L.) seeds. J Proteomics 74:262–267 (2011). 26 Tian L, Zhang L, Zhang J, Song Y and Guo Y, Differential proteomic analysis of soluble extracellular proteins reveals the cysteine protease and cystatin involved in suspension-cultured cell proliferation in rice. Biochim Biophys Acta – Proteins Proteomics 1794:459–467 (2009). 27 Balsamo GM, Cangahuala-Inocente GC, Bertoldo JB, Terenzi H and Arisi ACM, Proteomic analysis of four Brazilian MON810 maize varieties and their four non-genetically-modified isogenic varieties. J Agric Food Chem 59:11553–11559 (2011). 28 Hakeem KR, Khan F, Chandna R, Siddiqui TO and Iqbal M, Genotypic variability among soybean genotypes under NaCl stress and proteome analysis of salt-tolerant genotype. Appl Biochem Biotechnol 168:2309–2329 (2012). 29 Sahnoun-Abid I, Recorbet G, Zuber H, Sommerer N, Centeno D, Dumas-Gaudot E et al., Proteomic characterisation of subclover seed
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
931
1 Bonfim K, Faria JC, Nogueira EOPL, Mendes EA and Aragao FJL, RNAi-mediated resistance to Bean golden mosaic virus in genetically engineered common bean (Phaseolus vulgaris). Mol Plant–Microbe Interact 20:717–726 (2007). 2 Aragao FJL and Faria JC, First transgenic geminivirus-resistant plant in the field. Nat Biotechnol 27:1086–1088 (2009). 3 Aragao FJL, Nogueira EOPL, Tinoco MLP and Faria JC, Molecular characterization of the first commercial transgenic common bean immune to the Bean golden mosaic virus. J Biotechnol 166:42–50 (2013). 4 Choze R, Alcantara GB, Alves ED, Silva L, Faria JC and Liao LM, Distinction between a transgenic and a conventional common bean genotype by H-1 HR-MAS NMR. Food Chem 141:2841–2847 (2013). 5 Kok EJ, Keijer J, Kleter GA and Kuiper HA, Comparative safety assessment of plant-derived foods. Regul Toxicol Pharmacol 50:98–113 (2008). 6 Anttonen MJ, Lehesranta S, Auriola S, Rohlig RM, Engel KH and Karenlampi SO, Genetic and environmental influence on maize kernel proteome. J Proteome Res 9:6160–6168 (2010). 7 Cellini F, Chesson A, Colquhoun I, Constable A, Davies HV, Engel KH et al., Unintended effects and their detection in genetically modified crops. Food Chem Toxicol 42:1089–1125 (2004).
www.soci.org
www.soci.org
30
31 32
33
34
storage proteins during germination. J Sci Food Agric 89:1787–1801 (2009). Liu W, Zhang YZ, Gao X, Wang K, Wang SL, Zhang Y et al., Comparative proteome analysis of glutenin synthesis and accumulation in developing grains between superior and poor quality bread wheat cultivars. J Sci Food Agric 92:106–115 (2012). Badowiec A and Weidner S, Proteomic changes in the roots of germinating Phaseolus vulgaris seeds in response to chilling stress and post-stress recovery. J Plant Physiol 171:389–398 (2014). Lopez-Pedrouso M, Alonso J and Zapata C, Evidence for phosphorylation of the major seed storage protein of the common bean and its phosphorylation-dependent degradation during germination. Plant Mol Biol 84:415–428 (2014). Altenbach SB, Tanaka CK and Allen PV, Quantitative proteomic analysis of wheat grain proteins reveals differential effects of silencing of omega-5 gliadin genes in transgenic lines. J Cereal Sci 59:118–125 (2014). Dinon AZ, Brod FCA, Mello CS, Oliveira EMM, Faria JC and Arisi ACM, Primers and probes development for specific PCR detection of genetically modified common bean (Phaseolus vulgaris) Embrapa 5.1. J Agric Food Chem 60:4672–4677 (2012).
PA Valentim-Neto et al. 35 Venturelli GL, Brod FCA, Rossi GB, Zimmermann NF, Oliveira JP, Faria JC et al., A specific endogenous reference for genetically modified common bean (Phaseolus vulgaris L.) DNA quantification by real-time PCR targeting lectin gene. Mol Biotechnol 56:1060–1068 (2014). 36 Ferrari C, Amaral F, Bueno J, Scariot M, Valentim-Neto P and Arisi A, Expressed proteins of Herbaspirillum seropedicae in maize (DKB240) roots–bacteria interaction revealed using proteomics. Appl Biochem Biotechnol 174:2267–2277 (2014). 37 R Core Team, R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna (2013). 38 Barros E, Lezar S, Anttonen MJ, van Dijk JP, Rohlig RM, Kok EJ et al., Comparison of two GM maize varieties with a near-isogenic non-GM variety using transcriptomics, proteomics and metabolomics. Plant Biotechnol J 8:436–451 (2010). 39 Ricroch AE, Berge JB and Kuntz M, Evaluation of genetically engineered crops using transcriptomic, proteomic, and metabolomic profiling techniques. Plant Physiol 155:1752–1761 (2011). 40 Ricroch AE, Assessment of GE food safety using ‘-omics’ techniques and long-term animal feeding studies. New Biotechnol 30:349–354 (2013).
932 wileyonlinelibrary.com/jsfa
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
Revised: 2 February 2015
Accepted article published: 11 March 2015
Published online in Wiley Online Library: 1 April 2015
(wileyonlinelibrary.com) DOI 10.1002/jsfa.7166
Leaf proteome comparison of two GM common bean varieties and their non-GM counterparts by principal component analysis Pedro A Valentim-Neto, Gabriela B Rossi, Kelly B Anacleto, Carla S de Mello, Geisi M Balsamo and Ana Carolina M Arisi* Abstract BACKGROUND: A genetically modified (GM) common bean event, namely Embrapa 5.1, was approved for commercialization in Brazil. The present work aimed to use principal component analysis (PCA) to compare the proteomic profile of this GM common bean and its non-GM counterpart. RESULTS: Seedlings from four Brazilian common bean varieties were grown under controlled environmental conditions. Leaf proteomic profiles were analyzed by two-dimensional gel electrophoresis (2DE). First, a comparison among 12 gels from four common bean varieties was performed by PCA using volume percentage of 198 matched spots, presented in all gels. The first two principal components (PC) accounted for 46.8% of total variation. Two groups were clearly separated by the first component: Pérola and GM Pérola from Pontal and GM Pontal. Secondly, another comparison among six gels from the same variety GM and its non-GM counterpart was performed by PCA; in this case it was possible to distinguish GM and non-GM. CONCLUSION: Separation between leaf proteomic profile of GM common bean variety and its counterpart was observed only when they were compared in pairs. These results showed higher similarity between GM variety and its counterpart than between two common bean varieties. PCA is a useful tool to compare proteomes of GM and non-GM plant varieties. © 2015 Society of Chemical Industry Supporting information may be found in the online version of this article. Keywords: proteome; proteomics; Phaseolus vulgaris; GMO; Embrapa 5.1
INTRODUCTION
J Sci Food Agric 2016; 96: 927–932
∗
Correspondence to: Ana Carolina M Arisi, Food Science and Technology Department, Federal University of Santa Catarina, Rod. Admar Gonzaga 1346, 88034-001, Florianópolis – SC, Brazil. E-mail: [email protected] Food Science and Technology Department, Federal University of Santa Catarina, 88034-001, Florianópolis, SC, Brazil
www.soci.org
© 2015 Society of Chemical Industry
927
Common bean (Phaseolus vulgaris) is a source of protein in the diet of over a billion people worldwide. Golden mosaic of common bean is a disease caused by bean golden mosaic virus and it is one of the greatest constraints on common bean production in Latin America, with significant yield losses. RNA interference (RNAi) was used to silence the AC1 viral gene expression in transgenic common bean.1 In these genetically modified (GM) plants, the sequence-specific degradation of target mRNA interferes with viral replication, reducing or preventing viral DNA accumulation and, consequently, appearance of symptoms.1 Twenty-two GM lines were obtained with an intron-hairpin construct designed to induce post-transcriptional gene silencing of the AC1 gene. These lines were first evaluated under greenhouse conditions; two of these GM lines (named Embrapa 2.3 and Embrapa 5.1) showed high resistance upon inoculation with viruliferous whiteflies.2 In 2011 Brazilian National Technical Commission on Biosafety (CTNBio) approved Embrapa 5.1 event for cultivation and consumption in Brazil. This GM common bean was the first commercial GM plant developed in Latin America by the Brazilian Agricultural Research Corporation (Embrapa). A detailed molecular characterization of Embrapa 5.1 event was performed.3 Identification of the transgene insert confirmed the presence of a single locus
corresponding to two intact copies of the RNAi cassette in opposite orientation and three intact copies of the AtAhas gene. It is flanked by Phaseolus genomic sequences and interspersed by one nuclear and three chloroplastic genomic sequences.3 Embryo chemical composition of three Embrapa 5.1 common bean varieties and their non-GM counterparts (Pérola, Pontal and Olathe varieties) were compared by high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR) and this methodology was proposed to discriminate GM and non-GM common bean.4 Authors interested in the genetic content itself have analyzed embryo composition by HR-MAS NMR using principal component analysis (PCA), which is a useful tool for handling multivariate data. NMR and PCA confirmed the similarities in chemical profiles among the genotypes Olathe Pinto, Pérola and Pontal, as well as the three Embrapa 5.1 GM varieties, suggesting a natural grouping according to GM or conventional origin. The authors
www.soci.org stated that it was possible to evaluate genetic modification by comparing metabolic profiles of GM and non-GM common beans.4 Profiling techniques such as metabolomics, transcriptomics and proteomics have been suggested as non-targeted approaches for genetic modification analysis.5 – 10 Proteins are key players in gene function and are directly involved in metabolism and cellular development; thus proteome forms the central bridge between transcriptome and metabolome.11 Proteomic studies would provide important information for understanding changes in biological processes after genetic modification.12 Among other methods established in proteomics,13 – 15 two-dimensional gel electrophoresis (2DE) is a simple and frequently used technique.16 – 23 Proteomic analysis using 2DE has demonstrated the capacity to differentiate varieties on their protein content in several plants, such as common bean, soybean, maize, wheat and rice.6 – 18,24 – 33 In the present study, we used proteomic analysis and PCA to compare four Brazilian common bean varieties (Phaseolus vulgaris L. Embrapa 5.1 derived from the varieties Pérola and Pontal and their non-GM counterpart varieties). It is the first leaf proteomic analysis of this GM common bean.
MATERIAL AND METHODS Plant material Genetically modified common bean Phaseolus vulgaris L. Embrapa 5.1 derived from the varieties Pérola and Pontal and their non-GM counterpart varieties were provided by Embrapa Arroz e Feijão (Santo Antônio de Goiás, Goiás State, Brazil). This GM event contains an intron-hairpin construct designed to induce post-transcriptional gene silencing of the AC1 viral gene.1 Embrapa 5.1 shows high resistance to bean golden mosaic virus. Leaf material was obtained as follows: Nine seeds of each variety were placed on moistened germination paper and germinated in a growth chamber at 25 ∘ C in the dark for 3 days in May 2013. After germination, all 36 seedlings were grown side by side in pots containing soil in a controlled-environment growth chamber, adjusted to 12 h photoperiod, photosynthetic active radiation of 150 μmol m−2 s−1 , at 25 ∘ C, watered daily. After 7 days, leaves were collected and stored at −80 ∘ C. DNA extraction and PCR for GM common bean detection Genomic DNA was isolated from leaves, previously powdered in liquid nitrogen, using DNeasy Plant mini kit. Samples were subject to PCR using specific primers for Embrapa 5.1 GM common bean detection34 and for Phaseolus vulgaris detection.35 MON 810 GM maize and Roundup Ready GM soybean were used as negative control samples.
928
Protein extraction Total protein was extracted from a pool of six leaves from three plants for each variety. Leaves were ground to a fine powder in liquid nitrogen using a mortar and pestle, and 300 mg of each sample were extracted with 10 mL extraction buffer (0.5 mol L−1 Tris–HCl, pH 8.0, 100 mmol L−1 EDTA, 0.7 mol L−1 sucrose, 0.1 g L−1 CHAPS, 14 mmol L−1 DTT and 1 mmol L−1 PMSF). Proteins were precipitated overnight with 0.1 mmol L−1 ammonium acetate in cold methanol. After centrifugation, dried protein pellets were suspended in 300 μL of a solution containing 7 mol L−1 urea, 2 mol L−1 thiourea, 0.3 g L−1 CHAPS, 0.2 mL L−1 IPG buffer solution, pH 4–7, and 0.15 g L−1 DTT. Protein concentrations were determined using a 2D Quant kit (GE Healthcare). Three protein extracts were prepared from each variety.
wileyonlinelibrary.com/jsfa
PA Valentim-Neto et al. 2DE gel electrophoresis Leaf proteomic profiles were analyzed by two-dimensional gel electrophoresis (2DE). Each protein extract was used for each gel, so three 2DE gels were used per variety. Isoelectric focusing (IEF) was carried out using 13 cm pH 4–7 IPG strips (GE Healthcare). Strips were rehydrated for 18 h with a solution containing 250 μg total soluble protein diluted in rehydration buffer containing 0.2 mL L−1 IPG buffer pH 4–7 (GE Healthcare) in a total volume of 250 μL. After rehydration, strips were focused using an Ettan IPGphor 3 isoelectric focusing system (GE Healthcare) under the following conditions: step of 500 V until 500 V h, voltage gradients of 1000 V and 8000 V until 14500 V h, and a final step of 8000 V until 17 800 V h, up to a total of 34 000 V h, at a limit of 50 mA per strip. After focusing, strips were kept at −80 ∘ C for at least 18 h. The proteins in the IPG strips were subjected to reduction with 10 mg mL−1 DTT in 5 mL equilibration buffer (6 mol L−1 urea, 50 mmol L−1 Tris–HCl (pH 8.8), 3 mL L−1 glycerol, 0.2 g L−1 SDS and 2.5 mg L−1 bromophenol blue), followed by alkylation with 25 mg mL−1 iodoacetamide in the same buffer. The strips were then loaded on top of 12.5% polyacrylamide gel; a molecular weight marker 10–250 kDa (Precision Plus Protein Standards, BioRad) was used. Second-dimension SDS-PAGE gels were run in an SE 600 Ruby System (GE Healthcare) under 15 mA per gel for 30 min and 30 mA per gel for 4 h. The temperature was kept at 10 ∘ C using a MultiTemp III thermostatic circulator (GE Healthcare). Protein spots were visualized by staining the gels with 0.01 g L−1 Coomassie Brilliant Blue G-250 (Bio-Rad, Hercules, CA, USA) as described previously.36 Image and data analysis Gels were scanned in an Image Scanner System II and analyzed with ImageMaster Platinum software v. 7.0 (both from GE Healthcare) using automatic matching. Number of total spots was detected according to the following parameters: smooth ≥ 5, saliency ≥ 100 and area ≥ 4. Relative spot volumes (%Vol) were compared and analyzed by two approaches. In first approach, 12 gels of four varieties were compared all together. In the second approach, six gels of two varieties, each GM variety and its counterpart, were compared. All spots which matched in 12 gels or in six gels were selected and their volume percentage was log2 transformed. All data was median centered before analysis. PCA was performed using R language.37
RESULTS In this study, four common bean varieties were subjected to comparative leaf proteomic profiling. Seedlings from common bean varieties, Embrapa 5.1 GM Pérola and GM Pontal varieties (PerGM and PoGM) and non-GM Pérola and Pontal varieties (Per and Po), were grown side by side. In order to verify the GM identity of seedlings, specific PCR assays for Embrapa 5.1 detection and common bean detection were conducted. The two GM varieties were positive and the two non-GM varieties were negative for Embrapa 5.1 detection PCR assay, and the four varieties were positive for common bean detection PCR assay (supporting information, Fig. S1). Three protein extractions were carried out from leaves of each variety; therefore 12 extracts were prepared and 2DE gel was performed for each extract (Fig. 1). Proteomic analysis was performed in two approaches: first, all 12 gels were analyzed using Image Master software. An average of 513 ± 80 protein spots was detected in all 2DE maps, ranging from 422 to 689 total spots (Table 1). Average value of total protein spots for three gels from the same variety
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
Leaf proteome comparison of GM common bean by PCA
www.soci.org
Figure 1. Representative maps of common bean leaves protein profile from non-GM Pontal and GM Pontal varieties. For the first dimension, 250 μg of proteins were loaded on 13 cm IPG strips with pH 4–7 linear gradient; 12.5% SDS-PAGE gels were used for the second dimension. Gels were stained with Coomassie Brilliant Blue G-250.
Table 1. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM)
Table 3. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from two common bean varieties, Pérola (Per) and GM Pérola (PerGM). Gel
Gel Po1 Po2 (ref ) Po3 PoGM1 PoGM2 PoGM3 Per1 Per2 Per3 PerGM1 PerGM2 PerGM3
Total spots
Matched spots (%)
544 689 609 556 520 532 422 455 446 442 510 436
83 98 93 83 93 90 91 87 89 92 98 92
slope 1.26 – 1.04 1.13 0.79 0.92 0.66 0.62 0.71 0.61 0.71 0.67
0.83 – 0.9 0.77 0.84 0.84 0.7 0.74 0.76 0.72 0.73 0.78
Table 2. Average spot number, standard deviation and coefficient of variation of 2DE maps of leaves from four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM) Variety Po PoGM Per PerGM
Average total spots 614 536 441 463
SD 72.6 18.3 17.1 41.1
Coefficient of variation (%) 11.8 3.4 3.9 8.9
Matched spots (%)
slope
r
Per1 Per2 Per3 PerGM1 PerGM2 (ref.) PerGM3
617 583 574 616 715 594
95 90 91 91 96 90
0.99 0.9 0.91 0.91 0.94
0.88 0.93 0.94 0.91 0.91
Table 4. Total spot number, matched spots (%), slope and correlation coefficients (r) of scatter plots of 2DE maps of leaves from two common bean varieties: Pontal (Po), GM Pontal (PoGM) Gel
Total spots
Po1 Po2 (ref.) Po3 PoGM1 PoGM2 PoGM3
644 775 678 669 559 588
Matched spots (%) 89 97 91 92 89 87
slope 1.14 – 1.01 1.01 0.79 0.8
r 0.81 – 0.91 0.8 0.79 0.83
and the correlation coefficients of scatter plots ranged from 0.7 to 0.9 (Table 1). In the second approach, comparing matched spots of 2DE maps of two common bean varieties (Pérola and GM Pérola), the slopes of the best-fit line varied from 0.9 to 0.99 and the correlation coefficients of scatter plots were in the range 0.88–0.94 (Table 3). Comparing matched spots of 2DE maps of the others two common bean varieties (Pontal and GM Pontal), the slopes varied from 0.79 to 1.14 and the correlation coefficients of scatter plots ranged from 0.79 to 0.91 (Table 4).
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
929
ranged from 441 to 614 spots, presenting coefficients of variation lower than 12% (Table 2). Variability among the gels was analyzed by scatter plots of matched spots. In the first approach, comparing matched spots of all 2DE maps of all four common bean varieties in relation to one reference gel, the slopes of the best-fit line varied from 0.61 to 1.26 J Sci Food Agric 2016; 96: 927–932
Total spots
r
www.soci.org
PA Valentim-Neto et al.
Figure 2. PCA plots of 198 matched spots volume (%Vol) of 2DE gels from leaves of four common bean varieties: Pontal (Po), GM Pontal (PoGM), Pérola (Per) and GM Pérola (PerGM).
In the first approach, comparing all 12 2DE maps, 198 matched spots were detected. PCA was performed using the percentage volume of these matched spots as dataset. These data were log2 transformed to obtain normalized data from volume percentage of matched spots (supporting information, Fig. S2) for subsequent analysis using R language. When all 12 gels were analyzed by PCA (198 matched spots), the first 11 components explained 100% of dataset variation and 33.6% of the variation being allocated to the first component. The first two principal components (PC1 and PC2) accounted for 46.8% of total data variation (Fig. 2). Proteomic profiles from Pérola and Pontal varieties were clearly separated by the first component. However, a clear separation between GM common bean Pontal and non-GM Pontal was not observed. In the same manner, profiles from GM Pérola and non-GM Pérola were very close together in PCA plot (Fig. 2). In the second approach, leaf proteomic profiles were analyzed in pairs: GM variety and its counterpart. Three gels of GM Pontal and three gels of Pontal varieties were compared and they presented 364 matched spots. PC1 and PC2 explained 56.2% of total data variability and, in this case, PCA separated GM and non-GM gels (Fig. 3). When six gels of GM Pérola and Pérola were analyzed, there were 425 matched spots. PC1 and PC2 accounted for 49.4% of total variability and likewise PCA separated GM and non-GM gels (Fig. 4).
Figure 3. PCA plots of 364 matched spots volume (%Vol) of 2DE gels from leaves of two common bean varieties: Pontal (Po) and GM Pontal (PoGM).
Figure 4. PCA plots of 425 matched spots volume (%Vol) of 2DE gels from leaves of two common bean varieties: Pérola (Per) and GM Pérola (PerGM).
DISCUSSION
930
Common bean (Phaseolus vulgaris L.) is part of the staple diet in many countries of Latin America, and is cultivated by small and large farmers in all regions of Brazil. Embrapa 5.1 was the first GM common bean approved for cultivation and consumption in Brazil.2 This first leaf proteomic analysis to compare by PCA two GM common bean varieties and two non-GM counterpart varieties revealed that there are strong similarities among the GM common bean varieties and counterparts. Three 2DE gels of leaf protein extracts from the same variety presented high similarity (Table 2). The first approach of the analysis consisted of scatter plots of matched spots among four common bean varieties; slopes of the best-fit line and the correlation coefficients (r) presented high variation (Table 1). However, in the second approach, slopes and r
wileyonlinelibrary.com/jsfa
were close to 1 when scatter plots of matched spots were analyzed among six gels of two common bean varieties, GM variety and its counterpart (Tables 3 and 4), presenting high homogeneity. In the first approach, two-dimensional gels were separated into two groups by PCA (Fig. 2); one group contained three gels of GM Pérola and three gels of non-GM Pérola samples, and the other group contained three gels of GM Pontal and three gels of non-GM Pontal samples. This result is in accordance to previous works which compared GM and non-GM varieties using proteomic analysis.18,27,38,39 In general, the Omics comparisons revealed that genetic modification has less impact on plant gene expression and composition than that of conventional plant breeding.40 On the other hand, it was possible to distinguish three Embrapa 5.1 GM varieties from three non-GM common bean varieties
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
Leaf proteome comparison of GM common bean by PCA (Pérola, Pontal and Olathe Pinto) by comparing their embryo metabolic profiles,4 different from our results obtained for two varieties of Embrapa 5.1 common bean by comparing leaf proteomic profiles. A clear distinction between GM and non-GM varieties was observed on the PCA of leaf proteome profile only in the second approach, which analysis was performed in pairs: GM and its counterpart (Figs 3 and 4). Comparison between GM Pontal and non-GM Pontal leaf proteome gels shows clear separation on PC1 (Fig. 3). The same pattern was also observed comparing GM Pérola with non-GM Pérola samples (Fig. 4). MON810 maize leaf proteome was compared among four GM Brazilian varieties and their four non-GM counterparts; GM maize proteomic profiles were similar to non-GM counterparts proteomes.27 Previous study of MON810 maize grain proteome concluded that GM and non-GM maize Spanish varieties presented virtually identical grain proteomic profiles.18 In a previous work, a set of potato transcriptome profiles were compared by PCA.10 An untargeted comparative analysis based on transcriptomic profiles was proposed as a complementary tool for food evaluation, for which the necessity of transcript identification was not required. In this study conducted on virus-resistant GM common bean varieties, it was demonstrated that proteomic profiles from GM and non-GM common bean leaves were grouped close to each other in the PCA plots when compared to proteomic profiles of two breeding varieties. It was also shown that PCA is a useful tool to compare proteomes of GM and non-GM plant varieties, without the necessity for protein identification.
ACKNOWLEDGEMENTS We would like to express our gratitude to Josias C Faria for providing the seeds of GM common bean Embrapa 5.1. The work was financially supported by CNPq grant 470683/2012-0. PAVN is the recipient of a CAPES PNPD postdoctoral fellowship; GBR and KBA, of a CNPq IC fellowship; CSM and GMB, a CAPES PhD fellowship; and ACMA, a CNPq PQ-2 fellowship.
SUPPORTING INFORMATION Supporting information may be found in the online version of this article.
REFERENCES
J Sci Food Agric 2016; 96: 927–932
8 Metzdorff SB, Kok EJ, Knuthsen P and Pedersen J, Evaluation of a non-targeted ‘Omic’ approach in the safety assessment of genetically modified plants. Plant Biol 8:662–672 (2006). 9 van Dijk JP, Leifert C, Barros E and Kok EJ, Gene expression profiling for food safety assessment: examples in potato and maize. Regul Toxicol Pharmacol 58:S21–S25 (2010). 10 van Dijk JP, de Mello CS, Voorhilijzen MM, Hutten RCB, Arisi ACM, Jansen JJ et al., Safety assessment of plant varieties using transcriptomics profiling and a one-class classifier. Regul Toxicol Pharmacol 70:297–303 (2014). 11 Salekdeh GH and Komatsu S, Crop proteomics: aim at sustainable agriculture of tomorrow. Proteomics 7:2976–2996 (2007). 12 Gong CY and Wang T, Proteomic evaluation of genetically modified crops: current status and challenges. Front Plant Sci 4:41 (2013). 13 Bindschedler LV and Cramer R, Quantitative plant proteomics. Proteomics 11:756–775 (2011). 14 van Wijk KJ, Challenges and prospects of plant proteomics. Plant Physiol 126:501–508 (2001). 15 Valdes A, Simo C, Ibanez C and Garcia-Canas V, Foodomics strategies for the analysis of transgenic foods. TrAC, Trends Anal Chem 52:2–15 (2013). 16 Ruebelt MC, Leimgruber NK, Lipp M, Reynolds TL, Nemeth MA, Astwood JD et al., Application of two-dimensional gel electrophoresis to interrogate alterations in the proteome of genetically modified crops. 1. Assessing analytical validation. J Agric Food Chem 54:2154–2161 (2006). 17 Zolla L, Rinalducci S, Antonioli P and Righetti PG, Proteomics as a complementary tool for identifying unintended side effects occurring in transgenic maize seeds as a result of genetic modifications. J Proteome Res 7:1850–1861 (2008). 18 Coll A, Nadal A, Rossignol M, Puigdomenech P and Pla M, Proteomic analysis of MON810 and comparable non-GM maize varieties grown in agricultural fields. Transgenic Res 20:939–949 (2011). 19 Hakeem KR, Chandna R, Ahmad A, Qureshi MI and Iqbal M, Proteomic analysis for low and high nitrogen-responsive proteins in the leaves of rice genotypes grown at three nitrogen levels. Appl Biochem Biotechnol 168:834–850 (2012). 20 Otani M, Taniguchi T, Sakai A, Seta J, Kadoyama K, Nakamura-Hirota T et al., Phosphoproteome profiling using a fluorescent phosphosensor dye in two-dimensional polyacrylamide gel electrophoresis. Appl Biochem Biotechnol 164:804–818 (2011). 21 Hew C-S and Gam L-H, Proteome analysis of abundant proteins extracted from the leaf of Gynura procumbens (Lour.) Merr. Appl Biochem Biotechnol 165:1577–1586 (2011). 22 Gutierrez-Sanchez G, Atwood J, Kolli VSK, Roussos S and Augur C, Initial proteome analysis of caffeine-induced proteins in Aspergillus tamarii using two-dimensional fluorescence difference gel electrophoresis. Appl Biochem Biotechnol 166:2064–2077 (2012). 23 Campos A, Puerto M, Prieto A, Camean A, Almeida AM, Coelho AV et al., Protein extraction and two-dimensional gel electrophoresis of proteins in the marine mussel Mytilus galloprovincialis: an important tool for protein expression studies, food quality and safety assessment. J Sci Food Agric 93:1779–1787 (2013). 24 Brandao AR, Barbosa HS and Arruda MAZ, Image analysis of two-dimensional gel electrophoresis for comparative proteomics of transgenic and non-transgenic soybean seeds. J Proteomics 73:1433–1440 (2010). 25 De La Fuente M, Borrajo A, Bermudez J, Lores M, Alonso J, Lopez M et al., 2-DE-based proteomic analysis of common bean (Phaseolus vulgaris L.) seeds. J Proteomics 74:262–267 (2011). 26 Tian L, Zhang L, Zhang J, Song Y and Guo Y, Differential proteomic analysis of soluble extracellular proteins reveals the cysteine protease and cystatin involved in suspension-cultured cell proliferation in rice. Biochim Biophys Acta – Proteins Proteomics 1794:459–467 (2009). 27 Balsamo GM, Cangahuala-Inocente GC, Bertoldo JB, Terenzi H and Arisi ACM, Proteomic analysis of four Brazilian MON810 maize varieties and their four non-genetically-modified isogenic varieties. J Agric Food Chem 59:11553–11559 (2011). 28 Hakeem KR, Khan F, Chandna R, Siddiqui TO and Iqbal M, Genotypic variability among soybean genotypes under NaCl stress and proteome analysis of salt-tolerant genotype. Appl Biochem Biotechnol 168:2309–2329 (2012). 29 Sahnoun-Abid I, Recorbet G, Zuber H, Sommerer N, Centeno D, Dumas-Gaudot E et al., Proteomic characterisation of subclover seed
© 2015 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
931
1 Bonfim K, Faria JC, Nogueira EOPL, Mendes EA and Aragao FJL, RNAi-mediated resistance to Bean golden mosaic virus in genetically engineered common bean (Phaseolus vulgaris). Mol Plant–Microbe Interact 20:717–726 (2007). 2 Aragao FJL and Faria JC, First transgenic geminivirus-resistant plant in the field. Nat Biotechnol 27:1086–1088 (2009). 3 Aragao FJL, Nogueira EOPL, Tinoco MLP and Faria JC, Molecular characterization of the first commercial transgenic common bean immune to the Bean golden mosaic virus. J Biotechnol 166:42–50 (2013). 4 Choze R, Alcantara GB, Alves ED, Silva L, Faria JC and Liao LM, Distinction between a transgenic and a conventional common bean genotype by H-1 HR-MAS NMR. Food Chem 141:2841–2847 (2013). 5 Kok EJ, Keijer J, Kleter GA and Kuiper HA, Comparative safety assessment of plant-derived foods. Regul Toxicol Pharmacol 50:98–113 (2008). 6 Anttonen MJ, Lehesranta S, Auriola S, Rohlig RM, Engel KH and Karenlampi SO, Genetic and environmental influence on maize kernel proteome. J Proteome Res 9:6160–6168 (2010). 7 Cellini F, Chesson A, Colquhoun I, Constable A, Davies HV, Engel KH et al., Unintended effects and their detection in genetically modified crops. Food Chem Toxicol 42:1089–1125 (2004).
www.soci.org
www.soci.org
30
31 32
33
34
storage proteins during germination. J Sci Food Agric 89:1787–1801 (2009). Liu W, Zhang YZ, Gao X, Wang K, Wang SL, Zhang Y et al., Comparative proteome analysis of glutenin synthesis and accumulation in developing grains between superior and poor quality bread wheat cultivars. J Sci Food Agric 92:106–115 (2012). Badowiec A and Weidner S, Proteomic changes in the roots of germinating Phaseolus vulgaris seeds in response to chilling stress and post-stress recovery. J Plant Physiol 171:389–398 (2014). Lopez-Pedrouso M, Alonso J and Zapata C, Evidence for phosphorylation of the major seed storage protein of the common bean and its phosphorylation-dependent degradation during germination. Plant Mol Biol 84:415–428 (2014). Altenbach SB, Tanaka CK and Allen PV, Quantitative proteomic analysis of wheat grain proteins reveals differential effects of silencing of omega-5 gliadin genes in transgenic lines. J Cereal Sci 59:118–125 (2014). Dinon AZ, Brod FCA, Mello CS, Oliveira EMM, Faria JC and Arisi ACM, Primers and probes development for specific PCR detection of genetically modified common bean (Phaseolus vulgaris) Embrapa 5.1. J Agric Food Chem 60:4672–4677 (2012).
PA Valentim-Neto et al. 35 Venturelli GL, Brod FCA, Rossi GB, Zimmermann NF, Oliveira JP, Faria JC et al., A specific endogenous reference for genetically modified common bean (Phaseolus vulgaris L.) DNA quantification by real-time PCR targeting lectin gene. Mol Biotechnol 56:1060–1068 (2014). 36 Ferrari C, Amaral F, Bueno J, Scariot M, Valentim-Neto P and Arisi A, Expressed proteins of Herbaspirillum seropedicae in maize (DKB240) roots–bacteria interaction revealed using proteomics. Appl Biochem Biotechnol 174:2267–2277 (2014). 37 R Core Team, R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna (2013). 38 Barros E, Lezar S, Anttonen MJ, van Dijk JP, Rohlig RM, Kok EJ et al., Comparison of two GM maize varieties with a near-isogenic non-GM variety using transcriptomics, proteomics and metabolomics. Plant Biotechnol J 8:436–451 (2010). 39 Ricroch AE, Berge JB and Kuntz M, Evaluation of genetically engineered crops using transcriptomic, proteomic, and metabolomic profiling techniques. Plant Physiol 155:1752–1761 (2011). 40 Ricroch AE, Assessment of GE food safety using ‘-omics’ techniques and long-term animal feeding studies. New Biotechnol 30:349–354 (2013).
932 wileyonlinelibrary.com/jsfa
© 2015 Society of Chemical Industry
J Sci Food Agric 2016; 96: 927–932
