Article pubs.acs.org/jpr
iTRAQ Protein Profile Analysis of Tomato Green-ripe Mutant Reveals New Aspects Critical for Fruit Ripening Xiaoqi Pan, Benzhong Zhu, Hongliang Zhu, Yuexi Chen, Huiqin Tian, Yunbo Luo, and Daqi Fu* The College of Food Science and Nutritional Engineering, China Agricultural University, No. 17 Tsinghua East Road, Beijing 100083, PR China S Supporting Information *
ABSTRACT: Green-ripe (Gr) tomato carries a dominant mutation and yields a nonripening fruit phenotype. The mutation results from a 334 bp deletion in a gene of unknown function at the Gr locus. The putative influence of Gr gene-deletion mutation on biochemical changes underlying the nonripening phenotype remains largely unknown. Respiration of Gr fruit was found to be reduced at mature green and breaker stage of ripening, while the fruit softening was dramatically prolonged. We studied the proteome of Gr mutant fruit using high-throughput iTRAQ and high-resolution mass spectrometry and identified 43 proteins representing 43 individual genes as potential influence-targets of Gr mutated fruit. The identified proteins are involved in several ripening-related pathways including cell-wall metabolism, photosynthesis, oxidative phosphorylation, carbohydrate and fatty acid metabolism, protein synthesis, and processing. Affected protein levels are correlated with the corresponding gene transcript levels. The modulation in the accumulation levels of PI(U1)P, PGIP, and PG2 supported the delayed softening phenotype of the Gr fruit. Further investigation in GR gene-silencing fruit ascertained the doubtless modulation of these targets by the deletion mutation of GR gene. KEYWORDS: Green-ripe, tomato fruit, proteomics, iTRAQ, mass spectrometry, VIGS
■
gene, which in turn leads to ripening inhibition.8 The fruit of Gr mutant undergoes an increase in ripening-related ethylene production but fails to fully ripen. Also, the inhibition of ripening is not restored by exogenous ethylene, indicating that ethylene signaling is depressed in the Gr fruit.1 Overexpression of SlGR, the full-length cDNA of GR gene under the control of the cauliflower mosaic virus, recreates the Gr mutant phenotype but fails to obtain a whole-plant reduction in ethylene responsiveness.8 Arabidopsis REVERSION TO ETHYLENE SENSITIVITY1 (AtRTE1), which is the homologue of tomato GR gene, functions at or upstream of the ethylene receptors (ETRs) and facilitates the conformational change in the receptor following ethylene binding.9 Previous studies showed that loss of AtRTE1 function in Arabidopsis lead to enhanced ethylene responsiveness, whereas overexpression resulted in reduced sensitivity,9b and it seems that AtRTE1 acts predominantly through the ETR1 receptor, although the exact mechanisms of this interaction are not fully understood.10 Similarly, little information about the function of tomato GR has been uncovered. To date, collective results provide evidence that GR may modulate a subset of ripening-related pathways at the posttranscriptional level.1,8,10 Here we utilized high-throughput iTRAQ (Isobaric Tag for Relative and Absolute Quantification) combined with
INTRODUCTION The ripening of a fleshy fruit represents the culmination of a series of biochemical processes that are regulated by interactions between developmental programs and environmental inputs.1 The changes that occur during ripening process impart desirable characteristics to the fruit such as attractive color, softening, and accumulation of sugars and volatiles.2 Ripe fruits are dietary sources of nutrients such as vitamins, minerals, antioxidants, and fiber. Therefore, understanding the process of fruit ripening is of significance to basic plant biology, horticulture, and the nutritional engineering of food. Ethylene, the chemically simplest plant hormone, plays a vital role in the ripening program of climacteric fruits including tomato (Solanum lycopersicum).3 In most cases, ethylene regulates the developmental processes of fruits through initiating a variety of signal transduction pathways,4 and this modulation is fulfilled by a combination of ethylene and ethylene receptors as well as the downstream signal transductions.5 However, in some cases, ethylene alone is not sufficient,1 suggesting that ethyleneindependent (or developmental) pathways play a role in the fruit ripening program. Several ripening mutants of tomato are known and have been studied over the years to provide an understanding of the ripening in tomato.1,6−8 Gr is a typical monogenic tomato mutant derived from a deletion of 334-bp in the 5′-flanking region of an unknown gene at the Gr locus. This deletion results an ectopic expression of GR © 2014 American Chemical Society
Received: November 8, 2013 Published: March 4, 2014 1979
dx.doi.org/10.1021/pr401091n | J. Proteome Res. 2014, 13, 1979−1993
Journal of Proteome Research
Article
iTRAQ Labeling and SCX Fractionation
high-efficient VIGS (Virus-induced Gene Silencing), and other complementary molecular and physiological techniques to identify changed protein levels and gene expression caused by the ectopic expression of GR. The proteomic data was integrated into a network containing several metabolic or regulatory pathways to gain a systematic and broader elucidation of processes in the Gr mutant fruit. The findings show a novel relation among the ectopic expression of GR, changes in metabolic characteristics, and the nonripening phenotype of Gr fruit. The deductive information provides avenues to further dissect how the ripening program of GR fruit is modulated.
■
The trypsin-digested proteins were acidified with 10% trifluoroacetic acid and desalted on a C18 solid-phase extraction cartridge. Desalted peptides were then labeled with iTRAQ-8plex reagents (Applied Biosystems) according to the manufacturer’s instructions. Tissue samples from AC fruit were labeled with reagents 114 and 118, and samples from Gr mutant fruit were labeled with reagents 116 and 121. Two independent biological experiments with two technical repeats were performed. The incubation was allowed to proceed at room temperature for 2 h and then stopped by the addition of 10 mM KH2PO4, 25% ACN, pH 2.6, followed by centrifugation at 14 000g for 10 min to remove the aggregated proteins. Subsequently, all four labeled samples were pooled, vacuum-dried, and further fractionated offline using strong cation exchange (SCX) chromatography (UltremexSCX, 4.6 mm diameter, 250 mm length, 200 Å size, 5 μm particle size, Shimadzu). In total, 40 fractions were collected and combined into 13 final fractions, followed by a pooling to obtain a total of eight fractions for LC−MS/MS analysis.
MATERIALS AND METHODS
Plant Material
Tomato seeds were planted in a mixture of potting soil/ vermiculite (2:1 [v/v]), and seedlings with three to four true leaves were transplanted into commercial tomato-cultivated soil. All plants were grown in the greenhouse at 22 °C with 75% relative humidity under a 16 h light/8 h dark regime. Flowers were tagged one day after anthesis (DPA). Fruit of AC (Solanum lycopersicum cv. Ailsa Craig) were harvested at MG (mature green, 34 DPA), BK (breaker, 38 DPA), PK (pink, 42 DPA), and RR (red ripe, 48 DPA) stages of fruit ripening. Gr mutant fruit were harvested at the corresponding ripening stages, with the OR (orange ripe) stages around 60 DPA. Fruit of Micro-Tom was harvested at RR (red ripe, average 35 DPA) ripening stage. Ripening stages were also confirmed by visual analysis of color, size, and shape.11
LC−MS/MS Analysis
Mass spectrometric analysis of the iTRAQ-labeled samples was performed as previously described14 with a slight modification. Nanoflow electrospray ionization tandem mass spectrometric analysis was carried out using an LTQ Orbitrap Velos (Thermo Scientific, Bremen, Germany) interfaced with nanoACQuity (Waters) system. Peptides from SCX fractions were enriched using a C18 Trap column and separated on an analytical column (1.7 μm, 100 μm ×100 mm) at a flow rate of 300 nL/min using a linear gradient of 5−35% acetonitrile (ACN) over 40 min. Mass spectrometric analysis was carried out in a data-dependent manner with full scans acquired using the Orbitrap mass analyzer. The resolution was set to ≥30 000 at m/z 400, and the automatic gain control was set to 500 000 ions. From each MS scan, the 20 most intense precursor ions were selected for MS/MS fragmentation and detected at a mass resolution of 15 000 at m/z 400. The fragmentation was carried out using higher energy collision dissociation (HCD) with 40% normalized collision energy. The ions selected for fragmentation were excluded for 30 s. The automatic gain control for full Fourier transform mass spectrometry and MS/MS was set to 1and 0.1 million ions, respectively, with a maximum time of accumulation of 500 ms. For each experiment, two technical repeats were performed. For accurate mass measurements, the lock mass option was enabled.
Protein Extraction and Digestion
Total proteins were extracted from tomato fruit at BK stage as previously described.12 Five grams of frozen fruit tissue was finely powdered in liquid nitrogen and resuspended in 15 mL of extraction buffer (1% polyvinylpolypyrrolidone (PVPP), 0.7 M sucrose, 0.1 M KCl, 0.5 M Tris-HCl pH 7.5, 500 mM EDTA, 1 mM (PMSF), 2% β-mercaptoethanol). Then, an equal volume of Tris-HCl pH 7.5-saturated phenol was added, and the mixture was rehomogenized for 10 min on ice and finally centrifuged at 12 000g for 15 min. The upper aqueous phase was removed and reextracted with extraction buffer as previously described. Proteins were precipitated from the final aqueous phase with five volumes of saturated ammonium acetate in methanol overnight at −20 °C and pelleted by centrifugation at 12 000g for 30 min. Protein pellets were lyophilized and stored at −80 °C or immediately dissolved in protein extraction buffer composed of 6 M urea, 50 mM triethylammonium bicarbonate, pH 8.5, and 2% CHAPS for 1 h at 6 °C under constant shaking. Protein extracts were centrifuged at 14 000g for 10 min at 10 °C. The supernatant was collected, and protein concentration was determined by Bradford assay (1976)13 using bovine serum albumin as a standard. Total protein (100 μg) was reduced by adding dithiothreitol to a final concentration of 10 mM and incubated at 37 °C for 1 h. Iodoacetamide was then added to a final concentration of 55 mM, and the mixture was incubated for 1 h at room temperature in the dark. Subsequently, 10 mM of dithiothreitol was added and proteins were diluted in 50 mM triethylammonium bicarbonate and 1 mM CaCl2 to reduce the urea concentration to less than 0.6 M, followed by a digestion with 40 μg of sequencing grade porcine trypsin (Promega, Madison, WI) at 37 °C overnight. Protein extractions cover the two biological replicates, with each biological replicate consisting of an independent pool of samples from nine fruits.
Protein Identification and Functional Annotation
ProteinPilot (v2.0.1 Applied Biosystems/MDS Sciex) was used to simultaneously identify and quantify proteins. Tandem mass spectra were extracted and searched using MS/MS data interpretation algorithms within Protein Pilot (Paragon algorithm). Searches were against the “Solanaceae” and “Tomato” databases of NCBI nonredundant protein database containing an appended randomized database for false-positive assessment. Proteins were quantified on the basis of at least three confident MS/MS spectra (allowing generation of a p value). A p value smaller than 0.05 (with 95% confidence) and fold-change thresholds (1.2) were the used parameters. For protein quantification, the filters were set as follows: (1) “weighted” was chosen for protein ratio type; (2) minimum precursor charge was set to 1 and minimum peptides was set to 2; and (3) only unique peptides were used to quantify proteins. All identified proteins were classified according to annotations using the UniProt knowledgebase (Swiss-Prot/ TrEMBL, http://www.uniprot.org/) and the GO database 1980
dx.doi.org/10.1021/pr401091n | J. Proteome Res. 2014, 13, 1979−1993
Journal of Proteome Research
Article
(http://www.geneontology.org/). The biological process and molecular function were elucidated according to David database (http://david.abcc.ncifcrf.gov/), GOTERM (http://go. princeton.edu/cgi-bin/GOTermFinder), and PANTHER (http://www.pantherdb.org), respectively.15 Cellular components were elucidated according to GOTERM. The sequences of differentially displayed proteins were extracted to conduct alignments against the Cluster of Orthologous Groups (COG) database16 using the local alignment tool BLASTP. The KEGG database (http://www.genome.jp/kegg/) and PANTHER_PATHWAY (http://www.pantherdb.org/pathway/) were used to complement elucidation of the pathways of differentially displayed proteins. Plasmid Construction and VIGS Treatment
For the VIGS of tomato Micro-Tom fruit genes, the pTRV1 and pTRV2 VIGS vectors were adopted. A 347 bp fragment of the SlGR gene corresponding to bases 336−682 of the SlGR gene (NCBI accession NM_001246966) and a 304-bp fragment of the SlPSY gene corresponding to bases 129−432 of the SlPSY gene (NCBI accession KC767847) were PCR-amplified from tomato cDNA using primers shown in Supplementary Table 3 in the Supporting Information (with EcoR I + BamH I restriction sites and Nco I + Sac I restriction sites, respectively). The resulting products were cloned into pTRV2 to form pTRV2SlGR-SlPSY or pTRV2-SlPSY. Plant infiltration for VIGS treatment was performed as previously described (Fu et al. 2005).17 Agrobacterium strain GV3101 containing TRV-VIGS vectors was grown at 28 °C in LB medium containing 10 mM MES and 20 μM acetosyringone with appropriate antibiotics. After shaking for 18 h, Agrobacterium cells were harvested and resuspended in the Agrobacterium infiltration (AI) buffer (10 mM MgCl2, 10 mM MES, pH 5.6, 150 μM acetosyringone) to a final OD600 of 1.0. Resuspensions of pTRV1 and pTRV2 or its derivative vectors were mixed at a ratio of 1:1 and infiltrated into the carpopodium of a tomato fruit attached to the plant with a 1 mL syringe (Figure 1). Tomato fruit infiltrated with pTRV2 or pTRV2-SlPSY alone was used as control.
Figure 1. Schematic depiction of virus-induced gene silencing (VIGS) treatment in tomato plant. Target cDNA clones were placed in between the duplicated CaMV 35S promoter (2 × 35S) and the nopaline synthase terminator (NOSt) in the T-DNA vector of qTRV2. The plasmids of pTRV1, pTRV2 were transferred into Agrobacterium (GV3101) separately. Agrobacterium cultures containing pTRV1 and pTRV2 were mixed in 1:1 ratio and infiltrated into the carpopodium of tomato fruit attached to the plant. M 2000 refers to DNA Marker 2000. 1−SlGR, 2SlPSY.LB, and RB refer to left and right borders of T-DNA. Rz, self-cleaving ribozyme. MCS, multiple cloning sites. AI buffer refers to Agrobacterium infiltration buffer.
RNA Extraction and qRT-PCR
Fresh pericarp tissue of tomato fruit was used to extract total RNA as previously described.18 The RNA concentration and purity were measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific). The RNA integrity was checked by agarose gel electrophoresis. cDNA was synthesized using reversetranscription system (Promega) as described.19 The quantitative real-time PCR (qRT-PCR) was performed using SYBR Green PCR Master Mix with a BIO-RAD real-time PCR System (BIO-RAD CFX96, USA). qRT-PCR conditions were as follows: 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Fluorescence changes of SYBR Green were monitored automatically in each cycle, and the threshold cycle (Ct) over the background was calculated for each reaction. Samples were normalized using 18S rRNA (SGN-U581385) and the relative expression levels were measured using the 2(−ΔCt) analysis method.20 Oligonucleotide primers used in this study are listed in Supplementary Table 4 in the Supporting Information. RNA was extracted from a pool of samples from six different fruit, followed by cDNA synthesis. The qRT-PCR data presented are representative of three independent experiments.
overnight at room temperature, the hardness of pericarp was assayed using a penetrometer (mod. FT 327) according to the manufacturer’s instructions. All measurements were performed in triplicate, with each consisting of nine fruits. Respiration Measurement
Respiration was assayed from fruit at different ripening stages by enclosing them in airtight glass containers for 1 h at 20 ± 0.5 °C and 50% relative humidity (RH). Then, 1 mL gas samples were withdrawn and injected into a gas chromatograph (Shimadzu GC-14C, Japan) and interfaced to a 1.5 m long GDX-502 packed column (Tianjin, China) with a thermal conductivity detector (TCD) for CO2 production analysis. All measurements were replicated at least six times. Samples were normalized based on a standard curve and expressed as nL FW g−1 h−1. Statistical Analysis
Pericarp Hardness Analysis
Microsoft Excel 2010 and SPSS 13.0 (SPSS, Chicago, IL) were used for the statistical analyses. Data were subjected to an analysis of variance (ANOVA), and a comparison of means was
Fresh fruits were detached from different lines at MG, BK, and RR/OR ripening stages. After the pericarp was cleaned and aged 1981
dx.doi.org/10.1021/pr401091n | J. Proteome Res. 2014, 13, 1979−1993
Journal of Proteome Research
Article
Figure 2. Overview of differentially displayed proteins in Gr mutant fruit. (A) Phenotype of AC and Gr tomato fruit of BK stage used in iTRAQ study. (B) Area-proportional Venn diagrams depict the overlap of identified tomato proteins between Gr mutant and the wild-type AC fruit by iTRAQ measurements covering two biological replicates. (C) Abundance-distribution of differentially displayed proteins in Gr mutants versus AC fruit across two biological replicates. The red spots refer to the significantly accumulated proteins, and green spots refer to those decreased displayed ones.
carried out by Student’s t test. Differences were considered significant with p < 0.05.
■
cellular component, and molecular functions, which can be classified into 18, 9, and 6 categories, respectively. In terms of the number of proteins, the largest group within biological process is metabolic process (20.63%), followed by cellular process (19.05%) and response to stimulus (12.70%). The molecular function of these proteins mainly belongs to: catalytic activity (41.18%), binding (35.29%), electron carrier activity (8.82%), and structural molecule activity (8.82%), with occurrences at cell (23.53%), cell part (23.53%), organelle (19.61%), and membrane (10.78%) (Figure 3). To predict the biological functions of the differentially displayed 43 proteins underlying the nonripening phenotype of Gr fruit, we analyzed COG enrichment according to the sequence of proteins. Figure 4A shows a significantly functional enrichment of these proteins, indicating a differential enrichment for energy production and conversion, posttranslational modification, protein turnover, and chaperones. This prediction complemented the GO category and helped to dissect the nonripening processes of Gr mutant. Among the 43 proteins, putative ripening-related proteins according to their biological function were associated with protein modification (18.6%), photosynthesis (11.63%), cellular metabolic process (9.30%), cellwall metabolism (6.98%), oxidation−reduction process (4.65%), catabolic process (4.65%), ethylene biosynthesis (4.65%), and transport (4.65%) (Figure 4B).
RESULTS AND DISCUSSION
Protein Identification and Quantification of Gr Mutant Fruit using iTRAQ
To gain an expression profile of GR gene, we used qRT-PCR to determine the transcript level of GR gene in fruit of MG, BK, and RR/OR stages. Fruit of BK stage showed the highest expression level of GR gene, compared with MG and RR/OR fruits (Supplementary Figure 1 in the Supporting Information). To discover the mechanism underlying the nonripening phenotype of Gr mutant, total proteins in Gr fruit and changes in the protein profile compared with that of AC fruit were explored using the iTRAQ technique, and proteins were extracted from pericarp tissue of BK stage Gr mutant and AC (WT) fruits. A total of 2366 and 2385 proteins were identified in the two biological replicates. Among which, 2203 protein pairs composed from the mutant and WT fruit are common in the two biological replicates, with an overlap of 92.74%. The Venn diagram (Figure 2B) depicts detected numbers of common and differentially detected proteins from the two biological replicates. Each of these proteins was unambiguously identified using the criteria described in the Materials and Methods section. The detailed information of these proteins covering two biological replications is shown in Supplementary Table 1 in the Supporting Information. Among the identified proteins, 43 proteins exhibited a differential expression pattern in Gr mutant fruit compared with that in the AC WT (Supplementary Table 2 in the Supporting Information; Table 1). Seventeen proteins had increased and 26 proteins decreased in abundance in Gr mutant compared with the AC control fruit (with a fold change 1.2, p value
