Using 7 cm immobilized pH gradient strips to determine levels of clinically relevant proteins in wheat grain extracts

The aim of the work was to test a relatively simple proteomics approach based on phenol extraction and two-dimensional gel electrophoresis (2-DE) with 7 cm immobilized pH gradient strips for the determination of clinically relevant proteins in wheat grain. Using this approach, 157 2-DE spots were quantified in biological triplicate, out of which 55 were identified by matrix-assisted laser desorption/ionization – time of flight tandem mass spectrometry. Clinically relevant proteins associated with celiac disease, wheat dependent exercise induced anaphylaxis, baker’s asthma, and food allergy, were detected in 24 2-DE spots. However, alcohol-soluble gliadins were not detected with this approach. The comparison with a recent quantitative study suggested that gel-based and gel-free proteomics approaches are complementary for the detection and quantification of clinically relevant proteins in wheat grain.


Introduction
The main component of the wheat grain are storage proteins with gluten as the major part representing as much as 80% of total protein content (Ferranti et al., 2007). Gluten is a mixture of gliadins and glutenins that differ in their electrophoretic mobility (Payne et al., 1985;Jacobsen et al., 2007). Gluten is also the main allergen in the wheat grain and is responsible for nutritive intolerances such as celiac disease or gluten-sensitive enteropathy (Rubio-Tapia et al., 2009), and various allergies (Battais et al., 2008;Sapone et al., 2012;Mauro Martin et al., 2014). In addition to storage proteins, wheat grain allergens include enzymatic and structural proteins such as prolamins, cupins, and Bet v1 protein family (Breiteneder and Mills, 2005). Out of these, prolamins are dominant and include α-amylase and protease inhibitors, 2S albumins, and non-specific lipid transfer proteins (nsLTPs; Breiteneder and Radauer, 2004;Mills et al., 2004).
Protein two-dimensional gel electrophoresis (2-DE) has been extensively used to characterize wheat grain proteins. For instance, 2-DE followed with immunoblotting and tandem mass spectrometry (MS/MS) resulted into the identification of nine subunits of low molecular weight (LMW) glutenins, serpin, α-amylase inhibitor, and α-gliadin in wheat flour (Akagawa et al., 2007). The combination of 2-DE and MS/MS identified several allergenic proteins, such as serpins, in dough liquor of four wheat cultivars under abiotic stress (Sancho et al., 2008). Similar approaches detected heat responsible allergenic proteins, such as α-amylase inhibitors or serpins, in the endosperm of developing wheat grains under heat stress (Hurkman et al., 2009). Additionally, several allergenic proteins were detected in Korean sprouting wheat cultivars using 2-DE and matrix assisted laser desorption/ionization-tandem Time of Flight (MALDI-TOF) MS/MS (Kamal et al., 2009). Importantly, 20 allergenic proteins in wheat grains were detected using proteomics approach based on 2-DE in combination with 17 cm immobilized pH gradient (IPG) strips and MS/MS (Yang et al., 2011). Similarly, 2-DE in combination with isoelectric focusing (IEF) capillary tube gels, three different proteases, and MS/MS resulted in the detection of 476 2-DE spots out of which 233 were identified, including well-known allergens (Dupont et al., 2011). The 2-DE was also used to analyze wheat with genetically altered omega-5 gliadin content (Altenbach et al., 2014). Interestingly, this study showed that unique genetic transformation events with the same RNA interference construct may have differential effects on the wheat grain proteome (Altenbach et al., 2014). This study highlights the importance of proteomic analyses in the study of genetic transformations (Altenbach et al., 2014).
The above studies showed that a 2-DE approach is effective in the characterization of wheat grain proteins. However, 2-DE can be labor, resources, and time consuming, especially when long IPG strips are used. The aim of this study was to test a relatively simple 2-DE approach based on 7 cm IPG strips for the detection of clinically relevant proteins of wheat grain.

Plant Material and Protein Extraction
Seeds of winter wheat cultivar Viginta were obtained from SELEKT LtD, Bučany, Slovak Republic. Proteins were extracted in biological triplicate from 500 mg of dry seeds. Seeds were ground in liquid nitrogen and proteins were extracted with phenol-based extraction media [50% (v/v) phenol, 0.45 M sucrose, 5 mM EDTA, 0.2% (v/v) 2-mercaptoethanol, 50 mM Tris-HCl, pH 8.8]. Sample was stirred and homogenized for 30 min at 4 • C. The phenol phase was removed after centrifugation at 5000 × g for 10 min at 4 • C. Proteins were precipitated from the phenol phase by the addition of five volumes of ice-cold 0.1 M ammonium acetate in 100% methanol, and incubated at −20 • C overnight. The protein pellet was extensively washed twice using 0.1 M ice cold ammonium acetate in 100% methanol, followed by 80% ice cold acetone, and finally with 70% ice cold ethanol and precipitates were collected by centrifugation for 15 min., 5000 × g at 4 • C. Total protein concentration was determined using the Bradford (1976) assay with Bovine Serum Albumin as the standard.

Two-Dimensional Gel Electrophoresis
Samples (50 μg protein) were diluted in 100 μl of IEF buffer [8 M urea, 2 M thiourea, 2% (w/v) CHAPS, 2% (v/v) Triton X-100, 50 mM dithiothreitol], 3 μl of ampholytes were added, and loaded onto 7 cm IPG strips of pH 3-10 (ReadyStrip TM IPG Strips BioRad) for IEF. Isoelectric focusing was carried out using Protean IEF Cell (Bio-Rad) with the following conditions: 150 V for 150 VH, 500 V for 500 VH, and 4000 V for 15,000 VH including initial active rehydration for 12 h at 50 V. For the second dimension (SDS-PAGE), IPG strips were incubated in SDS equilibration buffer [1.5 M Tris-HCl pH 6.8, 6 M urea, 30% (v/v) glycerol, 5% (w/v) SDS) for 15 min with 2% (w/v) dithiothreitol] followed by a second equilibration step of 15 min with the equilibration buffer containing 2.5% (w/v) iodoacetamide. The equilibrated strips were loaded on the top of 10% polyacrylamide gel and the electrophoresis was run at 80 V until the dye reached the bottom of the gel. Gels were stained for 16 h with Coomassie Brilliant Blue G-250 at room temperature. The 2-DE gels were digitalized using a GS-800 Calibrated Densitometer (Bio-Rad) at 300 dpi and 16 bit grayscale. Digitalized gels were analyzed with PDQuest 8.0 software (Bio-Rad).
The mass spectrometer was operated by flexControl 3.3 software (Bruker). For every position 4000 shots were summed in positive reflector mode in the range of 700-3500 mass to charge (m/z). Following that, up to 25 of the most intense precursor peaks per sample were selected for the MS/MS analysis with the minimal signal to noise (S/N) ratio set to 15. Abundant trypsin and keratin peaks were specified in the exclusion list. Fragmentation spectra were acquired by accumulation of 3000 laser shots in positive reflector LIFT mode. Fragmentation was achieved by laser induced dissociation (LID) mechanism by 50% increase in laser power, without the introduction of a collision gas. Simultaneously detector voltage was boosted by 80%.

Processing of MS/MS Data
Acquired spectra were processed by flexAnalysis 3.3 software (Bruker). A sophisticated numerical annotation procedure (SNAP) algorithm was used for peak picking to calculate exact monoisotopic masses. For the precursor spectra the S/N threshold was set to 10 and the resulting spectra were externally recalibrated against data from an adjacent spot containing nine peptides of the Peptide calibration standard 2 (Bruker). For the fragment spectra S/N threshold was set to 5, also baseline subtraction (TopHat algorithm), and smoothing (Savitzky-Golay algorithm 3 cycles with 0.15 m/z width) were applied.
The MS/MS peak lists were imported into the ProteinScape 2.1 proteomic data management software (Bruker). Peptide identification was performed by an in-house Mascot 2.3 server (Matrix Science), querying against the non-redundant Triticeae plant protein UniProt database downloaded on April, 2014 (100 981 entries). Additionally, protein assignments were verified by searches against the SwissProt database from June 2014 (545 388 sequences) that included major contaminants such as trypsin or keratin. Search parameters were the following: fixed cysteine carbamidomethylation, variable methionine oxidation, one missed trypsin cleavage site, 40 ppm precursor mass tolerance, 0.5 Da fragment mass tolerance. Protein identifications were accepted if at least two different matched peptides had ion score higher than 30, meaning p < 0.05.
For allergenicity assessments, identified proteins were queried against the Allergome database 1 containing 2994 allergen entries, using Allergome Aligner module with an embedded NCBI blastp v.2.2.18 algorithm. Only hits with 100% sequence identity were accepted as clinically relevant allergens. The MS/MS proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD002067.

Results
The 2-DE-based proteomics approach in combination with 7 cm IPG strips (Figure 1) quantified 157 2-DE spots in biological triplicate (Supplementary Figure S1; Supplementary Table S1) out of which 55 were identified ( Table 1). Identified proteins were classified according to previous apporaches (Bevan et al., 1998) into five functional classes (Figure 2). The most abundant class was 37 proteins associated with destination and storage, followed by nine proteins associated with metabolism and five energy proteins (Figure 2). This study also detected two signaling proteins and two proteins associated with disease/defense ( Table 1). All identified proteins were assigned on the 2-DE gel (Supplementary Figure S2). The most abundant protein on this reference map is the high molecular weight (HMW) glutenin subunit (GS) with a relative volume (%V) of 8.8 (spot number 2909) followed by 11-S seed storage domain containing protein (3302) with %V of 5.6 ( Table 1).

As Much as 45% of Identified Proteins were Associated with Various Allergies or Food Intolerances
To determine clinical relevance of the identified proteins, sequences were queried against the Allergome database 1 which contains 2994 allergen entries (Supplementary Table S2). This approach detected clinically relevant proteins in 24 2-DE spots which represented 13 non-redundant accession numbers ( Table 1). Out of these, nine 2-DE spots were identified as HMW GS, seven as serpins, three as α-amylase inhibitors, two as LMW GS, two as glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and one as Cys peroxiredoxin (PER1; Table 1). All 1 http://www.allergome.org/ FIGURE 1 | Experimental workflow. Proteins were isolated from the winter wheat variety Viginta using a phenol-based method and resolved by 7 cm immobilized pH gradient (IPG) strip. After analysis with PDQuest, two-dimensional gel electrophoresis (2-DE) spots were excised from the gels, and analyzed by MALDI-TOF/TOF. From 55 identified 2-DE spots, 24 contained clinically relevant proteins. These proteins were assigned onto a 2-DE reference map.
these proteins were assigned on the 2-DE gel in order to establish the reference map of clinically relevant proteins of wheat grain (Figure 3). All 24 2-DE spots presented on this reference map were color-coded based on protein identification to visualize regions of the 2-DE gel with a prevalence of clinically relevant proteins (Figure 3). The most abundant protein is HMW GS (2-DE spot number 2909) followed LMW GS (7211; Table 1). The 2-DE spots identified as serpin (2202), GAPDH (3216), GAPDH, and α-amylase inhibitor (4009) (3203) showed the lowest abundance on this reference map (Figure 3; Table 1). To reveal overall abundances of the detected clinically relevant proteins, relative volumes of each 2-DE spot were summed based on protein identifications. Using this approach, the summed (total) relative volumes for each of eight detected clinically relevant proteins was established (Figure 4). It was revealed that HMW GS, LMW GS, and serpins are highly abundant in wheat grain (Figure 4).

Discussion
The aim of this work was to test a 2-DE proteomics approach with 7 cm IPG strips and phenol-based protein extraction for the detection of clinically relevant proteins in wheat grain. Classical methods for protein isolation from wheat grain are based on isopropanol extraction (van den Broeck et al., 2009). We successfully  implemented this method and have previously determined quantities of wheat grain proteins using gel-free proteomics approach (Uvackova et al., 2013a,b).
The majority of wheat grain allergenic proteins detected in the present study were not quantified in our recent MS-based study (Uvackova et al., 2013b). This finding is in agreement with a recent investigation of soybean under flooding stress, where only 9 out of 115 proteins were detected by both gel-based and gel-free proteomics approaches in root tips (Yin et al., 2014). Similar results have been shown in the analysis of the honey bees hemolymph proteome, where only 27% of proteins were detected with both approaches (Bogaerts et al., 2009).
Based on this, it is tempting to speculate that gel-based and gel-free approaches are complementary for the detection and quantification of wheat grain allergenic proteins. However, the complementarity of gel-based and gel-free proteomics approaches was suggested previously (Luque-Garcia et al., 2011;Abdallah et al., 2012). The combination of gel-based and gel-free proteomics was shown to be effective for the analyses of soybean under flooding (Yin et al., 2014), phytopathogenic fungus Botrytis cinerea (Gonzalez-Fernandez et al., 2013), Nicotiana tabacum trichomes (Van Cutsem et al., 2011), the honeybee hemolymph proteome (Bogaerts et al., 2009), or during soybean seed filling (Agrawal et al., 2008).

Conclusion
This study has demonstrated that phenol-based protein extraction in combination with 2-DE and 7 cm IPG strips is capable of determining clinically relevant proteins in wheat grain extracts. However, important clinically relevant proteins, such as alcohol-soluble gliadins were not detected with this approach. The comparison of these data with previous work suggests that gel-based and gelfree proteomics are complementary approaches for the determination of clinically relevant proteins in wheat grain extracts.