Identification of cold-stress responsive proteins in Anabasis aphylla seedlings via the iTRAQ proteomics technique

ABSTRACT Cold stress is one of the major environmental factors that affects plant growth, development, and species distributions. In order to study the molecular mechanism of cold-stress responses in Anabasis aphylla seedlings, we used the isobaric tags for relative and absolute quantification (iTRAQ) technique to identify differential protein expression under cold stress. In total, 211 differentially expressed proteins were identified, including 109 up-accumulated proteins and 102 down-accumulated proteins, these cold-stress response proteins were mainly identified as proteins involved in carbohydrate and energy metabolism, protein metabolism and translation, stress response, transcription-related, amino acid metabolism, signal transduction, and membrane and transport. GO and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses indicated that most of the proteins related to cold-stress response were involved in carbohydrate and energy metabolism, indicating A. aphylla seedlings adapted to cold conditions mainly through changes in energy metabolism pathways. Additionally, the correspondence between mRNA transcript levels and protein abundance levels for nine accumulated proteins were also tested by qRT-PCR, only one protein did not show consistent trends in mRNA and protein levels. These data clarify our current understandings of stress responses in seedlings of species such as A. aphylla and identify the molecular mechanisms of the cold-stress responses.


Introduction
Cold stress is a major abiotic factor that has noticeable effects on plant physiology and growth. It limits the geographical distribution and productivity of many wild plants and crops. Cold stress broadly influences the physiological and biochemical processes of plants, including photosynthesis (Hu et al. 2000), respiration (Xu 2007), oxygen and antioxidant enzymatic activity (Kang 2008), and water circulation (Sun 2011). In order to tolerate and adapt to adverse growth conditions, plants improve their cold resistance through the regulation of metabolic pathways. Plant responses to cold stress usually occur through changes in membrane composition, antioxidant components, and gene transcription rates under cold stress (Lukatkin 2002). Many genes that respond to cold stress have been documented in the model species Arabidopsis thaliana (Nordin et al. 1993;Welin et al. 1995;Medina et al. 1999). In recent years, proteomics combined with cold-resistant mutation analysis has significantly increased our understanding of signaling pathways and other important molecular mechanisms involved in cold acclimation (Chen et al. 2014;Janda et al. 2014). To date, proteomics has been used to analyze the cold resistance of a diverse array of plant species, including Chorispora bungeana, Oryza sativa (rice), Musa nana (dwarf banana), Buchloe dactyloides, and the desert woody plants Ammopiptanthus mongolicus and Populus euphratica (Yue et al. 2009;Tan 2012;Zhou 2013;Chen 2015;Pang 2015;Zhou et al. 2016), through this body of research, many proteins associated with cold-stress responses have been identified.
The subshrub Anabasis aphylla belongs to the family Chenopodiaceae and is one of the main species found in the soil desert area of the Junggar Basin, where it has an important role in water and soil conservation throughout the diluvial fan and plain. Furthermore, it is a plant species with great value . In this area, snowmelt is a vital water source for A. aphylla seed germination, and it can regulate plant phenology and ecosystem nutrient cycling. Through a combination of field investigation and observations with a simulation experiment, we found that a large number of A. aphylla seeds germinated in snowmelt in the field, which is consistent with Anabasis elatior germination characteristics (Han et al. 2011). Shimono and Kudo (2005) also found Peucedanum multivittatum seeds can germinate at 0°C and observed seeds that were buried in snow beds can emerge in the midst of freezing-thawing cycles. Thus, A. aphylla seeds clearly have the ability to germinate during snowmelt periods, and seedlings must therefore physiologically adapt to freezing conditions. This property has notable ecological significance for A. aphylla seeds and seedlings in coping with climatic fluctuations and unusually rapid warming in early spring.
Isobaric tags for relative and absolute quantification (iTRAQ) is a quantitative technique for analyzing cell metabolic changes. Many reports have used iTRAQ technology to identify and quantitatively analyze proteins. For example, Xie et al. (2016) successfully analyzed protein expression in tobacco leaves under drought stress. Wang (2013) identified differentially expressed proteins (DEPs) in Lycium ruthenicum through the iTRAQ technique and subjected them to gene ontology (GO) analysis, which revealed significant overrepresentations of DEPs among some biological functions. Additionally, Wang et al. (2016) applied the quantitative analysis of DEPs in maize leaves under cold stress using the iTRAQ technique. However, there have been no published reports focused on cold adaptation in A. aphylla, which highlights the need for research on not only the cold resistance of A. aphylla, but also the expression of genes and proteins that respond to cold stress. This study therefore used the quantitative protein technology iTRAQ to identify and characterize the molecular mechanism of cold-stress responses in A. aphylla seedlings.

Plant materials and stress treatments
A. aphylla seeds were obtained from the southern edge of the Gurbantünggüt Desert in Xinjiang, China. All collected A. aphylla seeds were sterilized by soaking them in 10% H 2 O 2 for 30 min, after which they were rinsed with distilled water. For the control treatment, the sterilized seeds were germinated on filter paper discs soaked with distilled water in Petri dishes at room temperature and a 16-h light/8-h dark photoperiod until seedlings were approximately 3 cm in length. For the cold treatment, 3-cm-long seedlings were kept in a −3°C growth chamber for 12 h each day for 1 week. For each treatment, 3-g seedlings were selected as samples and quickly frozen in liquid nitrogen individually and stored at −80°C for protein extraction for iTRAQbased comparative proteomic analyses. Both treatments had two biological replicates.

Protein extraction
Protein extraction was conducted according to the procedure described by Isaacson et al. (2006) with some modifications. Each A. aphylla sample was individually grinded into powder with liquid nitrogen. Then, 10 mL of cooled acetone containing 10% tricarboxylic acid (TCA) was added to 1 g of sample powder at −20°C for a 1-h incubation. Samples were then centrifuged at 15,000 × g for 15 min at 4°C. The resulting pellet was collected, cooled acetone was again added at −20°C for another 1-h incubation, and the centrifugation and cooled acetone addition steps were then repeated again. After centrifugation at 15,000 × g for 15 min at 4°C again, the resulting pellet was collected, dried with a vacuum freeze dryer, and suspended in cold phenol extraction buffer, to which an equal volume of phenol saturated with Tris-HCl (pH 7.5) was added. The mixture was maintained at 4°C for 30 min and then centrifuged at 5000 × g for 30 min, after which the upper phenolic phase was collected and an equal volume of phenol extraction buffer was added. Thereafter, five volumes of cold 0.1 M ammonium acetate in methanol was added to the collected phenol phase, and the mixture was store at −20°C for a 1-h incubation and then centrifuged for 30 min at 5000 × g at 4°C. Then two volumes (based on the volume of the last collected phenolic phase) of ice-cold methanol were used to wash the pellet, which was then mixed gently and centrifuged for 30 min at 5000 × g at 4°C; this step was then repeated two more times using acetone instead of methanol, and the sample was centrifuged for 30 min at 5000 × g at 4°C. The pellet was then dried, dissolved in lysis solution at 30°C for a 1h incubation, and centrifuged at 15,000 × g for 15 min at room temperature. The supernatant was collected and centrifuged again. The resulting supernatant was the extracted protein solution. The concentrations of the protein extracts were determined using the bicinchoninic acid (BCA) method (Smith et al. 1985) and stored at −80°C before the subsequent iTRAQ analysis was conducted.

SDS-PAGE electrophoresis
For each replicate and treatment, 10-μg samples were subjected to electrophoresis in a 12% SDS-PAGE gel. The gel was visualized using Coomassie brilliant blue (CBB) stain according to Candiano's protocol (Giovanni et al. 2004). The gels were fixed for 2 h, stained for 12 h, and washed prior to the spots of interest being excised. The stained gel was scanned using an Image Scanner (GE Healthcare, USA) at a resolution of 300 dots per inch.

Protein reduction, cysteine blocks, and digests
For each spot of interest identified by SDS-PAGE, five volumes of cold acetone were added to 100 μg of each protein sample, incubated at −20°C for 1 h, and centrifuged at 12,000 rpm for 15 min at 4°C, and the resulting pellet was collected and dried with a vacuum freeze dryer. To each pellet, 50 μL of dissolution buffer and 4 μL of reducing reagent were added. The solution was incubated at 60°C for 1 h, and 2 μL of cysteine-blocking reagents was added, followed by a 10-min incubation at room temperature. The protein solution was then cleaned using a 10 kDa ultrafiltration tube by centrifugation at 12,000 rpm for 20 min, and then 100 μL of dissolution buffer was added, followed by centrifugation at 12,000 rpm for 15 min; this step consisting of adding dissolution buffer and centrifugation was repeated three times. The column was then placed in a new tube, to which 50 μL of sequencing-grade trypsin (50 ng/μL) was added, and incubated at 37°C for 12 h. The samples were then centrifuged at 12,000 rpm for 20 min, and the peptide was collected. The filter columns were then transferred to new collection tubes and 50 μL of dissolution buffer was added, after which the two filter solutions were combined.

Protein labeling and MS analysis
Each vial of iTRAQ reagent was brought to room temperature, with centrifugal iTRAQ reagent at the bottom of each reagent tube. Then, 150 μL of ethanol was added to each room-temperature iTRAQ reagent vial, and 50 μL of each sample (containing 100 μg of peptide) was added to new individual tubes, to which iTRAQ reagent was added, followed by incubation at room temperature for 2 h, after which 100 μL of water was added to halt the labeling reaction. The cold-treated sample replicates were labeled with iTRAQ tags 115 and 116, while the untreated samples were labeled with tags 113 and 114. After labeling, the samples were mixed and centrifuged, and the supernatant was collected. Finally, the sample was dried in a vacuum freeze dryer for iTRAQ analysis.

2D-LC-MS/MS analysis
Reverse-phase high-performance liquid chromatography Dried samples were re-suspended with 100 μL of buffer A. Reverse-phase high-performance liquid chromatography (RPLC) was conducted on the Agilent 1200 HPLC System (Agilent, Santa Clara, CA, USA) using an Agilent HPLC column (narrow-bore 2.1 × 150 mm, 5 μm) with detection at 215 and 280 nm. Separation was performed at 0.3 mL/min. The first segment was collected between 0 and 5 min, and each subsequent segment was collected within 4.5 min intervals between 6 and 45 min, with the last segment collected between 46 and 50 min, yielding a total of 10 segments. Each segment was dried in a vacuum freeze dryer before LC-MS/MS analysis.

RPLC-MS/MS analysis
A. aphylla samples were re-suspended with Nano-RPLC buffer A. Online Nano-RPLC was conducted on the Eksigent nanoLC-Ultra ™ 2D System (AB SCIEX, Framingham, MA, USA). The samples were loaded onto a C 18 nanoLC trap column (100 µm × 3 cm, C 18 , 3 µm, 150 Å) and washed with Nano-RPLC Buffer A (0.1% formic acid, 2% acetonitrile) at 2 μL/min for 10 min. An elution gradient of 5-35% acetonitrile (0.1% formic acid) over the course of a 70 min gradient was used on an analytical ChromXP C 18 column (75 μm × 15 cm, C 18 , 3 μm 120 Å) with a spray tip. Data acquisition was performed with a Triple TOF 5600 System (AB SCIEX) fitted with a Nanospray III source (AB SCIEX) and a pulled quartz tip as the emitter (New Objectives, Woburn, MA, USA). Data were acquired using an ion spray voltage of 2.5 kV, curtain gas at a pressure of 30 PSI, nebulizer gas at 5 PSI, and an interface heater temperature of 150°C. For information-dependent acquisition, survey scans were acquired in 250 ms, and as many as 35 product ion scans were collected if they exceeded a threshold of 150 counts per second with a 2 + to 5 + charge-state. The total cycle time was fixed at 2.5 s. A rolling collision energy setting was applied to all precursor ions for collision-induced dissociation. Dynamic exclusion was set for half of the peak width (18 s), and the precursor was then refreshed off the exclusion list.

Protein identification and quantification
Data were processed with Protein Pilot Software v. 5.0 (AB SCIEX) against a Haloxylon ammodendron database using the Paragon algorithm (Shilov et al. 2007). The experimental data from tandem mass spectrometry (MS) were used to match the theoretical data to identify proteins. The analysis was conducted using the National Center for Biotechnology Information (NCBI) FASTA database and UniProt protein database. Differentially accumulated proteins were classified according to GO (http://www.geneontology.org) and Kyoto Encyclopedia of Genes and Genomes (KEGG; http://www. genome.jp/kegg/ or http://www.kegg.jp/) categories to infer molecular functions, cellular components, biological processes, and significant pathways involved in response to cold stress.

Enzyme activity assays
The activities of superoxide dismutase (SOD) and peroxidase (POD) were assayed according to the procedure described by Wang and Huang (2015). Malate dehydrogenase (MDH) and glutathione peroxidase (GSH-Px) were measured according to the methods detailed by Zhu (1990) and Huang and Wu (1999), respectively.

qRT-PCR analysis
Total RNA was extracted from cold-treated and control A. aphylla seedlings using Trizol reagent (Takara, Shiga, Japan), and cDNA was reverse-transcribed using the Trans-Start Tip Green qPCR SuperMix (TransGen Biotech, Beijing, China). Gene-specific primers used for qRT-PCR, which were designed using primer3 according to the nucleic acid sequences obtained from the Haloxylon ammodendron database. The PCR reaction was carried out in a 10-µL reaction volume containing 5 µL of TransStart Tip Green qPCR SuperMix (TransGen Biotech), 1 µL of template, 0.25 µL of forward primer, 0.25 µL of reverse primer, and 3.5 µL of ddH 2 O with thermal cycling conditions of 94°C for 30 s, followed by 40 cycles of 94°C for 5 s, 60°C for 30 s, and 20°C for 10 s.

Primary data analysis and protein identification
Mascot identified a total of 525,160 spectra generated by the iTRAQ experiment. Of these, 138,117 spectra matched known spectra, with 28,341 unique peptides, and 4145 proteins on the basis of the applied >0.05 (10%) cutoff ( Figure 1). The distribution of the number of peptides defining each protein is shown in Figure 2. More than 75% of the proteins included at least two peptides. GO annotation and KEGG analyses were carried out to identify the roles of the differentially accumulated proteins, and the results are shown in Figure 3.

Identification of differentially accumulated protein species by iTRAQ
Protein species were selected according to two criteria: ≥1.5 × fold change and p-value <0.05. Based on these criteria, 211 differentially accumulated proteins were identified in A. aphylla seedlings, 109 of which were up-accumulated and 102 were down-accumulated under cold-stress conditions (Table 1). The main biological functions for the 211 differentially proteins were carbohydrate and energy metabolism, protein metabolic process, stress response, transcription-related, amino acid metabolism, signal transduction, and membrane and transport. Among them, some proteins, such as fructose-bisphosphate protein, glyceraldehyde-3phosphate dehydrogenase, pyruvate kinase, citrate synthase, MDH, glucose dehydrogenase (GDH), photosynthesis protein, adenylate kinase (AK), ribosomal protein, elongation factor, heat shock protein, enzyme scavenging system proteins, TOPLESS protein, acetolactate synthase, glycine decarboxylase, membrane protein, channel protein, and transporter protein, were determined to be involved in these biological processes (these proteins are further highlighted in the section 'Discussion'). In addition, we found 26 proteins involved in unknown biological processes. Detailed information is shown in Figure 4.

Analysis of differentially expressed enzymes
Reactive oxygen species (ROS)-scavenging antioxidant mechanisms play an important role in maintaining suitable concentrations of ROS (Hasanuzzaman et al. 2012). To validate some of the DEPs, four enzymes involved in ROS scavenging and carbohydrate metabolism were selected for enzymatic activity analysis. The activities of SOD, MDH, and GSH-Px enzymes were increased by cold stress compared with the control treatment, though POD activity was decreased in the cold-stress treatment ( Figure 5). These data corresponded with consistent trends in the iTRAQ results.

Transcriptional expression analysis by qRT-PCR
In order to elucidate the correlation in expression level between the mRNA and protein species, the expression levels of the transcripts corresponding to nine proteins were validated by qRT-PCR ( Figure 6). The expression of eight transcripts was consistent with the abundance of the corresponding protein species including photosystem proteins, peroxiredoxin Q, and some ribosomal proteins (RPs). This indicated that the accumulation of these proteins is associated with mRNA transcript levels. However, the transcript expression assay for one of the proteins showed a trend that contrasted with the increased protein abundance, accordingly, this change in protein levels may have been caused by various posttranslational modifications such as glycosylation, protein phosphorylation, and ubiquitination under cold stress, rather than increased transcription.

Carbohydrate and energy metabolism
Cold stress has a substantial effect on the abundance of proteins involved in carbohydrate and energy metabolism processes such as glycolysis, the TCA cycle, and the pentose phosphate pathway (PPP) in A. aphylla seedlings. In our study, the abundance of fructose-bisphosphate proteins (FBPs) was increased under cold stress. FBP aldolase is a key enzyme in glycolytic metabolism, and it catalyzes 1,6fructose diphosphate into dihydroxyacetone phosphate and glyceraldehyde 3-phosphate. Excessive expression of FBP has been reported as a defense against abiotic stresses in Arabidopsis (Lu et al. 2012 Citrate synthase is an enzyme that controls the TCA cycle, in which it is the rate-limiting factor. In our results, the abundance of citrate synthase was down-accumulated under cold stress, which was consistent with a previous report on Capsicum annuum (Sánchez-Be et al. 2012). MDH catalyzes malic acid to oxaloacetate, after which oxaloacetate enters the TCA cycle, the abundance of this enzyme increased under cold stress in our data. Measured levels of MDH activity were also consistent with this result, indicating that the TCA cycle was inhibited in A. aphylla seedlings after cold-stress treatment.
Three proteins related to PPP were identified by the iTRAQ technique in our results: GDH, UTP-glucose-1-phosphate uridylyltransferase, and 6-phosphogluconolactonase. GDH is a key regulatory enzyme of the PPP, and its accumulation was enhanced under cold stress. However, both UTPglucose-1-phosphate uridylyltransferase and 6-phosphogluconolactonase, which are involved in the PPP, were downaccumulated in our study. Differential protein accumulation of proteins involved in this metabolic pathway suggested that complicated regulation mechanisms respond to cold stress in A. aphylla seedlings.
Cold can also affect different aspects of photosynthesis significantly. Our proteomics results indicated that most of the proteins involved in photosynthesis were down-accumulated under cold stress. The down-accumulated proteins included the chloroplast chlorophyll A-B binding protein, photosystem II CP43 chlorophyll apoprotein, photosystem II protein D1, photosystem I P700 apoprotein A2 chain f, and the crystal structure of plant photosystem protein. Among these proteins, photosystem I P700 apoprotein A2 and photosystem II CP43 chlorophyll apoprotein reflected the overall trend observed in the qRT-PCR experiment. Chloroplast chlorophyll A-B binding protein was also differentially expressed in Populus euphratica (Yue et al. 2009). Similarly, an up-accumulated protein was identified in our data, namely oxygenevolving enhancer protein 1. This result disagreed with a previously conducted expression profile of this protein in maize (Wang et al. 2016). Accordingly, A. aphylla seedlings may physiologically adapt to cold stress, in part, through changes in photosynthesis.   AK plays a key role in maintaining the balance of cell energy. As a protein kinase, AK catalyzes AMP to ADP while an organism is at a low energy level. ADP is oxidized into ATP, which is the direct supplier of energy in organisms through oxidative phosphorylation, thereby ensuring the normal vital activities of organisms (Lai et al. 2001). As such, the accumulation of AK may affect the level of ADP or ATP. Two AK proteins in our results were up-accumulated, suggesting that A. aphylla seedlings resist cold stress by storing additional energy.
Most proteins related to catabolism were also up-accumulated, whereas most of the photosynthesis proteins were down-accumulated under cold stress, which is consistent with a previous report on Chorispora bungeana (Sun 2011). This may be a consequence of cold affecting the activity of various photosynthetic enzymes in the chloroplast, thus resulting in the degradation of many important photosynthetic proteins. In order to survive, plants need more energy to sustain their life activities through processes such as catabolism and in this way to adapt cold stress.

Protein metabolism
Proteins are involved in the regulation of various metabolic activities in organisms, including the processes of translation, digestion, folding, and degradation in plants (Reinbothe et al. 2010). We identified several proteins that are involved in protein metabolism using the iTRAQ data. We found eight DEPs related to protein synthesis, including six RPs and two elongation factors. RPs are essential for protein synthesis and have been revealed to play important roles in metabolism, cell division, and growth (Wang 2013). In our study, the abundance of five RPs decreased, while one increased. Among the RPs, the qRT-PCR validation results of ribosomal protein S2, ribosomal protein S3, and ribosomal protein S4 were consistent with the protein accumulation results. Several relevant studies have linked abundances of RPs to abiotic stress level. One previous study reported that some ribosomal protein expression levels decreased while other specific ribosomal components increased (Laura et al. 2011). Cheng et al. (2016 also showed some RPs related to protein processing were up-or down-accumulated under abiotic stress in two hemp varieties. Additionally, we found that the abundance of translation elongation factor 1A-6 was up-accumulated under cold stress. This elongation factor is an organelle protein that plays a central role in the elongation phase of protein synthesis. Conversely, the abundance of elongation factor 2 was down-accumulated under cold stress, indicating that the expression levels of different elongation factors respond differently to cold stress. Heat shock proteins (HSPs) are ubiquitous protective proteins in plants. HSPs mainly act as a molecular chaperone to prevent protein denaturation and repair denatured proteins through binding with target proteins, which maintains the stability of plant internal environments, HSPs play a key role in plant development and stress-resistance processes . In recent years, HSPs have been shown to increase dramatically in expression level under high temperatures, drought, peroxidation, and heavy metal stress (Wang et al. 2004). Under cold stress, the abundance of HSP70 was up-accumulated. Other proteins related to protein folding were also identified as differentially accumulated in our study, including 10 kDa chaperonin-like, chaperonin CPN60-like, cytosolic oligopeptidase A, and dnaJ protein P58IPK homolog isoform X2. Among these proteins, dnaJ is an accessory protein that enables ATP enzyme activity of HSP70, thereby regulating its activity, and it has a variety of functions, including folding new proteins, endocytosis, polypeptide transport via cell organelles, adjustments to stress responses, and the pre-degradation of protein targeting (Cyr et al. 1994;Krzysztof et al. 2008). Oligopeptidase may be involved in the degradation of proteasome-generated peptides with salicylic acid (Moreau et al. 2013). These two proteins were up-accumulated under cold stress, suggesting that the process of protein degradation increases with protein metabolism.

Stress response
Cold stress can cause excess production of ROS, which are important signal molecules in plants; however, ROS can also cause irreversible damage in plant cells (Asada 1999).
In order to relieve cellular damage caused by ROS, plants have developed ROS-scavenging systems that include enzymatic scavenging systems, such as SOD, ascorbate peroxidase (APX), GSH-Px, SAM, and POD. SOD is an enzyme that eliminates the harmful substances produced by metabolic processes. APX is considered to be a key enzyme in the detoxification of hydrogen peroxide in  chloroplasts (Asada 1999). Under cold stress, the abundance of Cu/Zn SOD was increased, and this observation is consistent with the response of maize seedlings to cold stress in a previous study (Wang et al. 2016). APX showed overexpression in stressful conditions such as drought, cold, high salt, and heavy metal stress, enabling plant chloroplasts and mitochondria to resist stress caused by oxidative damage (Neilson et al. 2011). However, our results showed that the abundance of APX decreased under cold stress. Therefore, A. aphylla appears to resist cold through the accumulation of other antioxidant enzymes other than APX.
GSH-Px catalyzes glutathione into oxidized glutathione, enables the toxic peroxide reduction of nontoxic hydroxyl compounds, and promotes the decomposition of H 2 O 2 , thus protecting the structure and function of the cell membrane from interference and damage caused by peroxide (Nathalie et al. 1998). The increased abundance of GSH-Px under cold stress suggested that the protein helps cope with cold stress in A. aphylla seedlings. The GSH-Px enzyme activity assay also confirmed this finding.
POD is involved in respiration, photosynthesis, and the oxidation of auxin, its activity changes continuously in the process of plant growth and development (Zhao et al. 2013). In the present study, the POD isozyme levels decreased in response to cold stress, but the abundance of 3-ketoacyl-CoA thiolase 2, peroxisomal was increased. In previous studies, the abundance of POD increased under abiotic stress in wheat, barley, and cucumber, but decreased in creeping bentgrass (Peng et al. 2009;Witzel et al. 2009;Du et al. 2010;Xu et al. 2010). This suggested that the expression mechanism of peroxidase in plants under cold stress is complex.
S-adenosylmethionine synthase catalyzes methionine and ATP into SAM, and S-adenosylmethionine synthase 2 is also a methyl donor. It participates in the synthesis of ethylene, polyamine, and plant hormones, substances that are closely linked to plant stress responses (Cui et al. 2005). The abundance of S-adenosylmethionine synthase 2 was increased under cold stress. Consistent with this observation, a previous study reported that S-adenosylmethionine synthase 2 was upaccumulated by cold stress in hybrid rice seedlings (Yan et al. 2006). Accordingly, the accumulation of antioxidant proteins may dissipate excess excitation energy and thus protect plant against cold stress.

Transcription and signal transduction
The regulation of transcription has great significance in the responses of plants to biotic and abiotic stresses (Jiang et al. 2007). The protein TOPLESS is a transcriptional corepressor that can repress the expression of root-promoting genes in the top half of the embryo to allow proper differentiation of the shoot pole during the transition stage of embryogenesis, it is also involved in the cellular synthesis of RNA on a template of DNA (Wei et al. 2015). A previous study has also shown that TOPLESS protein expression is related to transcription and stress response (Ke et al. 2015). In our data, TOPLESS was down-accumulated under cold stress. In Arabidopsis, the nucleolar protein is relevant to rRNA modification, which is vital in the response to abiotic stress  Figure 6. Analysis of transcript levels of the differentially accumulated A. aphylla proteins under the control and cold-stress treatments by qRT-PCR. (Pendle et al. 2004). In our study, nucleolar protein decreased in abundance, suggesting that the down-accumulation of this protein in A. aphylla seedlings plays an important role in cold-stress response.
Mitogen-activated protein kinase homolog NTF6 is an enzyme that converts ATP to ADP and a phosphoprotein and is involved in the H 2 O 2 -mediated mitogen-activated protein kinase-signaling pathway. In our present study, mitogen-activated protein kinase homolog NTF6 showed higher accumulation levels, but in contrast, this protein showed downregulation in tobacco cells (Sasabe et al. 2006). This suggests that cold-resistant plants may physiologically adapt by altering signal transduction pathways.

Amino acid metabolism and membrane transport
As amino acids are the basic units of proteins and precursors for all protein production, their metabolism and availability have significant effects on plant growth, photosynthesis, and stress resistance (Yang et al. 2013). The protein 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase catalyzes the transfer of a methyl group from 5methyltetrahydrofolate to homocysteine resulting in methionine formation in L-methionine biosynthesis (Ravanel et al. 2004). Glycine dehydrogenase is involved in the glycine catabolic process and is also one of the key components of the glycine cleavage system (Palmieri et al. 2010). An abundance of glycine decarboxylase may generate more glutathione, a major antioxidant and free-radical scavenger in plants (Noctor et al. 1999;Foyer and Noctor 2000 ). Acetolactate synthase catalyzes the formation of acetolactate from pyruvate, and it is involved in the first step in valine and isoleucine biosynthesis (Mazur et al. 1987). This protein is also involved in the first steps of both the subpathway that synthesizes L-isoleucine from 2-oxobutanoate and the subpathway that synthesizes L-valine from pyruvate, which is part of amino acid biosynthesis. Both L-isoleucine and L-valine function in the repair of plant tissues and provide energy for plant growth (Wang et al. 2011). Under cold stress, the abundance of acetolactate synthase was increased, which suggested that the biosynthesis of L-isoleucine and L-valine had increased.
Membrane steroid-binding protein 2-like is an integral component of the membrane. Porin/voltage-dependent anion-selective channel protein is involved in transmembrane transport, an essential role in the maintenance of ion homeostasis in plant cells. Porin/voltage-dependent anion-selective channel protein was down-accumulated in our study as well as in a similar study conducted with cucumber (Huang et al. 2009). Manganese-transporting ATPase PDR2 mediates manganese transport into the endoplasmic reticulum, which is vital in pollen and root development through its impact on protein secretion and transport processes (Wang et al. 2008). Mitochondrial dicarboxylate transporter is an important transport protein in the inner membrane of chloroplasts. It catalyzes the exchange of dicarboxylic acids like malate and succinate for inorganic phosphate (Picault et al. 2002). These two transport proteins showed higher abundance in response to cold stress. They were induced in Arabidopsis thaliana and Ananas comosus (Kinoshita et al. 2011;Mirvat et al. 2016). As such, A. aphylla seedlings appear to adapt to the cold environments, in part, through alterations in their amino acid metabolism and membrane transport proteins.

Conclusion
A proteomics analysis using iTRAQ technology was performed to analyze differential protein accumulation under cold stress in A. aphylla seedlings. A total of 211 differentially accumulated proteins were identified between control and cold stress conditions seedlings, among which, 109 proteins were up-accumulated and 102 were down-accumulated. On the basis of GO and KEGG pathway enrichment analysis, we concluded that most of the proteins related to cold-stress response were involved in carbohydrate and energy metabolism, suggesting that A. aphylla seedlings primarily adapted to cold conditions through accumulations of proteins involved in energy metabolism pathways. In addition, proteins in some biological processes related to cold stress also had an effect on cold resistance of A. aphylla seedlings, including protein translation, transcriptional regulation, amino acid metabolism, signal transduction, and membrane transport. We also used qRT-PCR technique to confirm positive relationships in expression level between the mRNA transcript and protein species. These results showed that accumulation of a diversity of proteins overall, i.e. in eight out of nine assayed proteins, was associated with mRNA expression levels; however, posttranslational modifications were the likely explanation for decreased accumulation of one of the assayed proteins. Generally speaking, our analysis of differentially accumulated proteins in seedlings under cold stress elucidates the molecular mechanisms involved in cold tolerance in A. aphylla and other species.

Availability of data and material
The datasets supporting the conclusions of this article are available in the Gene Ontology (http://www.geneontology. org), Kyoto Encyclopedia of Genes and Genomes (http:// www.genome.jp/kegg/ or http://www.kegg.jp/) and Uniprot (http://www.uniprot.org/).