Multivariate analysis of 2-DE protein patterns--practical approaches

Practical approaches to the use of multivariate data analysis of 2-DE protein patterns are demonstrated by three independent strategies for the image analysis and the multivariate analysis on the same set of 2-DE data. Four wheat varieties were selected on the basis of their baking quality. Two of the varieties were of strong baking quality and hard wheat kernel and two were of weak baking quality and soft kernel. Gliadins at different stages of grain development were analyzed by the application of multivariate data analysis on images of 2-DEs. Patterns related to the wheat varieties, harvest times and quality were detected on images of 2-DE protein patterns for all the three strategies. The use of the multivariate methods was evaluated in the alignment and matching procedures of 2-DE gels. All the three strategies were able to discriminate the samples according to quality, harvest time and variety, although different subsets of protein spots were selected. The explorative approach of using multivariate data analysis and variable selection in the analyses of 2-DEs seems to be promising as a fast, reliable and convenient way of screening and transforming many gel images into spot quantities.     

Comparison of ethanol-soluble proteins from different rye (Secale cereale) varieties by two-dimensional electrophoresis

The major storage proteins from six rye varieties, grown under the same conditions in 1997 and 1998 in Rønhave, Denmark, were analyzed by two‐dimensional (2‐D) polyacrylamide gel electrophoresis. The proteins were extracted from ground rye kernels with 70% ethanol and separated by 2‐D electrophoresis. The gels were scanned, compared using ImageMaster® software and the data sets were analyzed by principal component analysis (PCA) using THE UNSCRAMBLER software. Afterwards MATLAB was used to make a cluster analysis of the varieties based on PCA. The analysis of the gels showed, that the protein patterns (number of different proteins and their isoelectric points and molecular weights) from the six rye varieties were different. Based on the presence of unique cultivar‐specific spots it was possible to differentiate between all six varieties if the two harvest years were investigated separately. When the results were combined from the two years five varieties could be differentiated. The results from the PCA confirmed the finding of the unique spots and cluster analysis was made in order to illustrate the results. The combination of the results from 2‐D electrophoresis and other grain characteristics showed that one protein spot was located close to the parameters bread volume and bread height.     

An improved pixel-based approach for analyzing images in two-dimensional gel electrophoresis

An improved pixel-based approach for analyzing 2-DE images is presented. The key feature of the method is to create a mask based on all gels in the experiment using image morphology, followed by multivariate analysis on the pixel level. The method reduces the impact of noise and background by identifying regions in the image where protein spots are present, but make no assumption on individual spot boundaries for isolated spots. This makes it possible to detect significant changes in complex regions, and visualize these changes over multiple gels in an easy way. False missing values and spot volumes caused by imposing erroneous spot boundaries are thus circumvented. The approach presented gives improved pixel-based information from the gels, and is also an alternative to existing methods for data-reduction, significance testing and visualization of 2-DE data. Results are compared with software using a common spot boundary approach on an experiment consisting of 35 full size gel images. Gel alignment is required before analysis.     

Determination of wheat quality by mass spectrometry and multivariate data analysis

Multivariate analysis has been applied as support to proteome analysis in order to implement an easier and faster way of data handling based on separation by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry. The characterisation phase in proteome analysis by means of simple visual inspection is a demanding process and also insecure because subjectivity is the controlling element. Multivariate analysis offers, to a considerable extent, objectivity and must therefore be regarded as a neutral way to evaluate results obtained by proteome analysis.Proteome analysis of storage proteins from the wheat gluten complex based on two-dimensional electrophoresis and analysis of the N-terminal sequence has revealed a protein homologous to gamma-gliadins, tentatively associated with quality and within the molecular weight range 27-35 kDa. Further examinations of gliadin data based on mass spectrometry revealed that quality among wheat varieties could be determined by means of principal component analysis. Further examinations by interval partial least squares made it possible to encircle an overall optimal molecular weight interval from 31.5 to 33.7 kDa. The use of multivariate analysis on data from mass spectrometry has thus shown to be a promising technique to minimize the number of two-dimensional gels within the field of proteome analysis.     

Variety identification of wheat using mass spectrometry with neural networks and the influence of mass spectra processing prior to neural network analysis

The performance of matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry with neural networks in wheat variety classification is further evaluated.1 Two principal issues were studied: (a) the number of varieties that could be classified correctly; and (b) various means of pre-processing mass spectrometric data. The number of wheat varieties tested was increased from 10 to 30. The main pre-processing method investigated was based on Gaussian smoothing of the spectra, but other methods based on normalisation procedures and multiplicative scatter correction of data were also used. With the final method, it was possible to classify 30 wheat varieties with 87% correctly classified mass spectra and a correlation coefficient of 0.90.     

Comparison of the Results of SDS PAGE and Chip Electrophoresis of Wheat Storage Proteins

Chip electrophoresis was used for the study of the high-molecular-weight glutenin subunit (HMW-GS) composition of wheat. A model set of samples of Czech wheat varieties was used and HMW-GSs were detected by SDS-PAGE. A part of the common HMW-GSs present in wheat were identified by chip electrophoresis. Chip electrophoresis provides a very good reproducibility of HMW-GS patterns. Moreover this approach is much faster than the conventional SDS PAGE methods requiring several hours for an analysis. Another advantage over traditional gel electrophoresis is lower sample and reagent volume requirements, as well as specialized protein standards for accurate reproducibility and quantification.     

Fast wheat variety classification by capillary gel electrophoresis-on-a-chip after single-step one-grain high molecular weight glutenin extraction

Wheat variety identification based on one-step single-grain wheat extraction and fast capillary gel electrophoresis-on-a-chip (CGE-on-a-chip) analyses was evaluated for 15 different wheat varieties grown in Austria. The results of the capillary-based separation system were compared to the internationally accepted method from the International Union for the Protection of New Varieties of Plants which is based on time-consuming sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) analysis. Comparable protein patterns were observed making the CGE-on-a-chip system a promising tool for high-throughput analysis in food control. For the development of a robust method protein extraction, shelf life of wheat extracts and the instrument’s variability were evaluated. It turned out that a one-step single-grain wheat extraction allowed the sample to be stored at 4 °C for up to 4 weeks without losing any valuable protein information. Furthermore, the technical variation of the whole method is very low making the biological variation of the selected wheat grains the only uncertain factor. Additionally, two unsupervised statistical methods (hierarchical cluster analysis and principal component analysis) were used for variety identification. Identification was successful for a reduced data set of 14 samples from five different wheat varieties making the combination of CGE-on-a-chip analysis of one-step single-grain extraction in combination with automatic data evaluation a promising tool for fast wheat differentiation (within a day).     

A new method for assigning common spot boundaries for multiple gels in two-dimensional gel electrophoresis

The benefits of defining common spot boundaries when several gels from 2-DE are compared and analyzed have lately been stressed by both commercial software producers and users of this software. Though the importance of common spot boundaries is clearly stated, few reports exist that target this issue explicitly. In this study a method for defining common spots boundaries is developed, called the spot density method. The method consists of the following steps: segmentation and spot identification on each individual gel, transferring the spot-center coordinates for all gels onto a single new gel, collecting spot centers clustered together in the new gel and finally assigning pixels and new spot boundaries based on the spots in each cluster. The method is compared to a synthetic gel approach, and validated by visual inspection of three representative areas in the gels. The gel images need to be aligned prior to segmentation and spot identification, but the method can be used regardless of the choice of segmentation procedure. This makes the method an easy extension to existing methods for spot identification and matching. Conclusions based on the visual inspection are that the spot density method identifies partly overlapping spots and low-intensity spots better than the synthetic gel approach.     

Matrix-assisted laser desorption/ionization time-of-flight analysis of low-concentration oligonucleotides and mini-sequencing products

A new sample preparation procedure has been developed to improve the matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) analysis of low-concentration oligonucleotides. In this method, a few microL of the oligonucleotide solutions are first dispensed and allow drying and shrinking to a small spot on an anchoring target. Thereafter, a small volume (0.1 microL) of the saturated 3-hydroxypicolinic acid (3-HPA) solution is added to the top of the shrunk oligonucleotide spot. It is found that the sample prepared by this procedure is more homogenous and substantially reduces the need to search for 'sweet' spots. Importantly, the increased shot-to-shot and sample-to-sample reproducibility makes it possible to perform high-throughput and quantitative analysis. We demonstrated that this procedure allowed the robust detection of oligonucleotides at a level of 0.01 microM and mini-sequencing products produced using only 50 fmol of the extension primer.     

Independent component analysis of 2-D electrophoresis gels

We present a novel application of independent component analysis (ICA), an exploratory data analysis technique, to two-dimensional electrophoresis (2-DE) gels, which have been used to analyze differentially expressed proteins across groups. Unlike currently used pixel-wise statistical tests, ICA is a data-driven approach that utilizes the information contained in the entire gel data. We also apply ICA on wavelet-transformed 2-DE gels to address the high dimensionality and noise problems typically found in 2-DE gels. Also, we use an analysis-of-variance (ANOVA) approach as a benchmark for comparison. Using simulated data, we show that ICA detects the group differences accurately in both the spatial and wavelet domains. We also apply these techniques to real 2-DE gels. ICA proves to be much faster than ANOVA, and unlike ANOVA it does not depend on the selection of a threshold. Application of principal component analysis reduces the dimensionality and tends to improve the performance by reducing the noise.     

Mass spectrometry and partial least-squares regression: a tool for identification of wheat variety and end-use quality

Rapid methods for the identification of wheat varieties and their end-use quality have been developed. The methods combine the analysis of wheat protein extracts by mass spectrometry with partial least-squares regression in order to predict the variety or end-use quality of unknown wheat samples. The whole process takes approximately 30 min. Extracts of alcohol-soluble storage proteins (gliadins) from wheat were analysed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Partial least-squares regression was subsequently applied using these mass spectra for making models that could predict the wheat variety or end-use quality. Previously, an artificial neural network was used to identify wheat varieties based on their protein mass spectra profiles. The present study showed that partial least-squares regression is at least as useful as neural networks for this identification. Furthermore, it was demonstrated that partial least-squares regression could be used to predict wheat end-use quality, which has not been possible using neural networks.     

A two-dimensional electrophoresis-related laboratory information processing system: spot matching.

An approach for the computer-assisted analysis of two-dimensional gels has been developed as a part of our laboratory information processing system (LIPS). This approach relies in part on an algorithm for the pairwise matching of protein spots. The matching process initially matches spots based on a cross-correlational measure of how well neighboring spots align. While this first pass correctly determines most spot correspondences and noncorrespondences, it can make errors. Higher accuracy is obtained by monitoring the consistency of spot match decisions in a second pass, which demands that neighboring spot pairs that align spatially must also have been found to match in the first pass. Pairwise comparisons of gels are combined into n-way comparisons by matching spot lists of gels to "master" gel spot lists, which in turn are matched to higher level masters, resulting in a hierarchy of matched spots. After each pairwise match the results are reviewed and corrected with the assistance of a graphical match-editor. Results are given for 19 single-cell-derived lymphoid clones in which the presence of a mutation had previously been established, each processed in duplicate. Only one of 46 spot changes failed to be detected, which demonstrates that the strategy is sensitive and efficient for detecting qualitative spot differences.     

Target list building for volatile metabolite profiling of fruit

Fruit flavour is the combination of numerous biochemicals: sugars for sweetness, acids for sourness and volatile metabolites for aroma. The objective of this study was to establish a method to develop a target list of statistically relevant compounds for the characterization of melon from non-targeted data, while preserving the profile information. Five different varieties were sampled (sampling 12 biological replicates from 12 plants) using dynamic headspace extraction, then analysed by gas chromatography–mass spectrometry in full scan mode. Using Metalign and SIMCA-P software the raw data was spectrally aligned and then subjected to principal component analysis (PCA). The principal component analysis plot showed good separation of the five varieties based on their full scan GC–MS profile. Mass spectral data points responsible for the differences between varieties were highlighted by further statistical analysis. The mass spectra were then reconstructed and the corresponding chemicals identified using library search or reference standards were available to create a new target component list. To validate the new target list, the initial data set was re-processed using the targeted approach and the results subjected again to principal component analysis. The two representations showed excellent agreement on the separation of the five varieties. The new target list obtained from this study can be applied to differentiate and characterize the volatile profile of melon varieties using a list of statistically significant compounds.     

Use of principal components analysis for mutation detection with two-dimensional electrophoresis protein separations.

The application of two-dimensional electrophoresis (2-DE) to mutation detection requires the capability to monitor each protein in a 2-DE pattern for significant changes in abundance indicative of a mutation event. Previously, mutation searches were done using a univariate outlier detection method in which each protein spot was considered independently in a classical outlier search. An alternative approach to analysis of 2-DE patterns for quantitative changes is a multivariate procedure which takes advantage of the observation that protein spots in a 2-DE pattern often represent correlated rather than independent measurements. We have compared the efficiency of univariate and multivariate procedures for mutation detection using data from the Argonne National Laboratory 2-DE database of mouse liver proteins. Analyses involving a total of over 1500 gels were performed to compare the performance of a multivariate method based on principal components analysis (PCA) with the univariate method. Up to 279 spots from each pattern were used for PCA. First, a simulation was performed to assess the detection efficiency of PCA for single protein spots decreased in abundance by 50%. Then, the ability to detect actual mutations was tested using eight confirmed mutations. Results show that, compared to a univariate approach to analysis of data from the mouse model system, the multivariate method increases the number of protein spots on each 2-DE pattern that can be monitored for quantitative changes indicative of mutations by compensating for variables that contribute to the background quantitative variability of protein spots.     

Rapid imaging, using a cooled charge-coupled-device, of fluorescent two-dimensional polyacrylamide gels produced by labelling proteins in the first-dimensional isoelectric focusing gel with the fluorophore 2-methoxy-2,4-diphenyl-3(2H)furanone

A new method for visualising proteins in two-dimensional polyacrylamide gels was developed. Proteins were labelled with the fluorophore 2-methoxy-2,4-diphenyl-3(2H)furanone (MDPF) while present in the first-dimensional gel after isoelectric focusing and subsequently electrophoresed into the second-dimensional gel. High resolution spot patterns were produced and compared with other methods of visualisation. A new rapid imaging system based on a cooled charge-coupled-device was used to view the two-dimensional fluorescent protein spot patterns. The method allows the immediate and rapid imaging of two-dimensional gels at the end of electrophoresis with no further processing.     

Electrophoresis of gliadins in long acrylamide gels: method and nomenclature

Gliadins, defined as wheat kernel storage proteins soluble in 70% ethanol, possess an electrophoretic diversity permitting us to identify different varieties. Because of genetic proximity, however, it sometimes proves impossible to distinguish between different varieties using standard methods. An electrophoresis method, utilizing a discontinuous buffer system with aluminum lactate and potassium lactate in the gel and sodium lactate in the electrode vessels, is described for two types of gels 18 cm and 32 cm in length. The number of bands is higher than with standard methods, and is significantly increased with long gels, facilitating the distinction between varieties. A nomenclature for the new bands is presented for several widely different cultivars.     

Analysis of protein/polypeptide interactions in human plasma using nondenaturing micro-2-DE followed by 3-D SDS-PAGE and MS

Previously, we have reported on the analysis of human plasma proteins on a nondenaturing micro-2-DE (mu2-DE) gel, using in-gel digestion followed by MALDI-MS and PMF [1]. Many of the spots on the mu2-DE gel showed apparent masses much larger than the calculated masses of their assigned polypeptides, suggesting noncovalent or covalent interactions between the polypeptides. In the present study, we aimed to further analyze the plasma protein spots on a nondenaturing mu2-DE gel, on which protein/polypeptide interactions have been suggested. The proteins in the spots were extracted under alkaline conditions and subjected to 3-D separation using SDS-PAGE in microslab gel format (muSDS gel) with or without the sample treatment of reduction-alkylation. The clear bands in each lane of the muSDS gels demonstrated the successful extraction of proteins from the relevant gel spot and visualized the relative contents of the polypeptides in the spot. Most of the bands were assigned by in-gel digestion followed by MALDI-MS and PMF (MASCOT/Swiss-Prot). The large discrepancy between the apparent mass value of a protein spot and the estimated mass values of the polypeptide bands on a nonreducing muSDS gel strongly suggested noncovalent polypeptide interactions. The differences in the polypeptide separation patterns on the muSDS gels, between with and without the treatment of reduction-alkylation, confirmed polypeptide disulfide bonding. The method employed here, aiming to integrate information on the proteins separated on nondenaturing 2-DE gels with that on the interactions between polypeptides, would help the comprehensive understanding of complex protein systems.