Sources of Variability among Replicate Samples Separated by Two-Dimensional Gel Electrophoresis

Two-dimensional gel electrophoresis (2DE) offers high-resolution separation for intact proteins. However, variability in the appearance of spots can limit the ability to identify true differences between conditions. Variability can occur at a number of levels. Individual samples can differ because of biological variability. Technical variability can occur during protein extraction, processing, or storage. Another potential source of variability occurs during analysis of the gels and is not a result of any of the causes of variability named above. We performed a study designed to focus only on the variability caused by analysis. We separated three aliquots of rat left ventricle and analyzed differences in protein abundance on the replicate 2D gels. As the samples loaded on each gel were identical, differences in protein abundance are caused by variability in separation or interpretation of the gels. Protein spots were compared across gels by quantile values to determine differences. Fourteen percent of spots had a maximum difference in intensity of 0.4 quantile values or more between replicates. We then looked individually at the spots to determine the cause of differences between the measured intensities. Reasons for differences were: failure to identify a spot (59%), differences in spot boundaries (13%), difference in the peak height (6%), and a combination of these factors (21). This study demonstrates that spot identification and characterization make major contributions to variability seen with 2DE. Methods to highlight why measured protein spot abundance is different could reduce these errors.     

The QUEST system for quantitative analysis of two-dimensional gels

The strategies and methods used by the QUEST system for two-dimensional gel analysis are described, and the performance of the system is evaluated. Radiolabeled proteins, resolved on two-dimensional gels and detected using calibrated exposures to film, are quantified in units of disintegrations per minute or as a fraction of the total protein radioactivity applied to the gel. Spot quantitation and resolution of overlapping spots is performed by two-dimensional gaussian fitting. Pattern matching is carried out for groups of gels called matchsets, and within each matchset every gel is matched to every other gel. During the matching process, spots are automatically added to each pattern at positions where unmatched spots were detected in other patterns. This results in enhanced accuracy for both spot detection and for matching. The spot fitting procedure is repeated after matching. Tests show that up to 97% of spots in each pattern can be matched and that fewer than 1% of the spots are matched inconsistently. Approximately 2000 proteins are detected from typical gels. Of these 1600 are high quality spots. Tests to measure the coefficient of variation of spot quantitation versus spot quality show that the average coefficient of variation for high quality spots is 21%. The intensities of the detected proteins range from 4 to 20,000 ppm of total protein synthesis. The QUEST analysis system has been used to build a quantitative database for the proteins of normal and transformed REF52 cells, as presented in the accompanying reports (Garrels, J., and Franza, B. R., Jr. (1989) J. Biol. Chem. 264, 5283-5298, 5299-5312).     

Evaluation of two-dimensional difference gel electrophoresis for protein profiling. Soluble proteins of the marine bacterium Pirellula sp. strain 1

Two-dimensional gel electrophoresis (2DE) is a central tool of proteome research, since it allows separation of complex protein mixtures at highest resolution. Quantification of gene expression at the protein level requires sensitive visualization of protein spots over a wide linear range. Two-dimensional difference gel electrophoresis (2D DIGE) is a new fluorescent technique for protein labeling in 2DE gels. Proteins are labeled prior to electrophoresis with fluorescent CyDyes trade mark and differently labeled samples are then co-separated on the same 2DE gel. We evaluated 2D DIGE for detection and quantification of proteins specific for glucose or N-acetylglucosamine metabolism in the marine bacterium Pirellula sp. strain 1. The experiment was based on 10 parallel 2DE gels. Detection and comparison of the protein spots were performed with the DeCyder trade mark software that uses an internal standard to quantify differences in protein abundance with high statistical confidence; 24 proteins differing in abundance by a factor of at least 1.5 (t test value <10(-9)) were identified. For comparison, another experiment was carried out with four SYPRO-Ruby-stained 2DE gels for each of the two growth conditions; image analysis was done with the ImageMaster trade mark 2D Elite software. Sensitivity of the CyDye fluors was evaluated by comparing Cy2, Cy3, Cy5, SYPRO Ruby, silver, and colloidal Coomassie staining. Three replicate gels, each loaded with 50 microg of protein, were run for each stain and the gels were analyzed with the ImageMaster software. Labeling with CyDyes allowed detection of almost as many protein spots as staining with silver or SYPRO Ruby.     

A probabilistic treatment of the missing spot problem in 2D gel electrophoresis experiments

Two-dimensional SDS-PAGE gel electrophoresis using post-run staining is widely used to measure the abundances of thousands of protein spots simultaneously. Usually, the protein abundances of two or more biological groups are compared using biological and technical replicates. After gel separation and staining, the spots are detected, spot volumes are quantified, and spots are matched across gels. There are almost always many missing values in the resulting data set. The missing values arise either because the corresponding proteins have very low abundances (or are absent) or because of experimental errors such as incomplete/over focusing in the first dimension or varying run times in the second dimension as well as faulty spot detection and matching. In this study, we show that the probability for a spot to be missing can be modeled by a logistic regression function of the logarithm of the volume. Furthermore, we present an algorithm that takes a set of gels with technical and biological replicates as input and estimates the average protein abundances in the biological groups from the number of missing spots and measured volumes of the present spots using a maximum likelihood approach. Confidence intervals for abundances and p-values for differential expression between two groups are calculated using bootstrap sampling. The algorithm is compared to two standard approaches, one that discards missing values and one that sets all missing values to zero. We have evaluated this approach in two different gel data sets of different biological origin. An R-program, implementing the algorithm, is freely available at http://bioinfo.thep .lu.se/MissingValues2Dgels.html.     

Detection of human somatic cell structural gene mutations by two-dimensional electrophoresis

The feasibility of detecting human somatic structural gene mutations by two dimensional electrophoresis has been investigated. A lymphoblastoid cell line was grown as a mass culture in the presence of ethylnitrosourea, after which cells were regrown as single cell clones. A total of 257 polypeptide spots were analyzed in gels derived from 186 clones. Four structural mutations were detected by visual analysis of the gels. Computer analysis of gels corresponding to the mutant clones was also undertaken. At a spot size threshold of 200 spots to be matched using a computer algorithm, all four mutant polypeptides were detected. These results indicate the usefulness of the two-dimensional approach for mutagenesis studies at the protein level.     

Assessment of pleiotropic effects of a gene substitution in pea by two-dimensional polyacrylamide gel electrophoresis

We examined, by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE), near-isogenic lines of the r-gene in pea (Pisum sativum) which determines round (RR) vs. wrinkled (rr) seed. The study was undertaken to assess the number of protein changes resulting from a single gene substitution as a means of quantifying pleiotropic effects. A total of 636 to 770 resolvable polypeptides were identical in all respects between RR and rr for roots, shoots, leaflets, stipules, young ovaries, and young embryos. A single difference between the lines became evident about 21-23 days after anthesis in the embryos. Mature seeds of the two lines showed 62 spot differences in addition to differences in four clusters of spots, representing about 10% of the total number of spots visible on the gels. The protein differences are presumably involved in the many known physiological differences of the two seed types. 2-D PAGE analyses of near-isogenic lines are likely to be valuable in a number of quantitative and developmental genetic contexts.     

Technical strategies to reduce the amount of "false significant" results in quantitative proteomics

When the p-value is set at <0.05 in statistical group comparisons, a 5% rate of "false significant" results is expected. In order to test the reliability of our 2-DE method, we loaded each of 24 gels with equal-sized samples (200 mug protein from pooled rat brain, pH 4-7, stained with ruthenium fluorescent stain for visualization) and statistically compared the first 12 gels with the last 12. In numerous experiments the rate of significant differences found far exceeded 5%. Several factors were identified as causing the following rates of false significant differences in spot intensities: (i) running samples in two different 2-DE runs (42%), (ii) running second dimension gels produced in two different gel casters (16%), (iii) normalizing the entire gel instead of separately normalizing several different gel zones (11%), (iv) using IPG strips from different packages (19%), (v) dividing the whole sample into subgroups during software analysis (9%). After controlling for all these factors, the rates of "false positive" results in our experiments were regularly reduced to approximately 5%. This is an indispensable prerequisite for avoiding too high a rate of false positive results in experiments in which different subgroups are compared statistically.     

Proteome approaches to characterize seed storage proteins related to ditelocentric chromosomes in common wheat (Triticum aestivum L.)

Changes in protein composition of wheat endosperm proteome were investigated in 39 ditelocentric chromosome lines of common wheat (Triticum aestivum L.) cv. Chinese Spring. Two-dimensional gel electrophoresis followed by Coomassie Brilliant Blue staining has resolved a total of 105 protein spots in a gel. Quantitative image analysis of protein spots was performed by PDQuest. Variations in protein spots between the euploid and the 39 ditelocentric lines were evaluated by spot number, appearance, disappearance and intensity. A specific spot present in all gels was taken as an internal standard, and the intensity of all other spots was calculated as the ratio of the internal standard. Out of the 1755 major spots detected in 39 ditelocentric lines, 1372 (78%) spots were found variable in different spot parameters: 147 (11%) disappeared, 978 (71%) up-regulated and 247 (18%) down-regulated. Correlation studies in changes in protein intensities among 24 protein spots across the ditelocentric lines were performed. High correlations in changes of protein intensities were observed among the proteins encoded by genes located in the homoeologous arms. Locations of structural genes controlling 26 spots were identified in 10 chromosomal arms. Multiple regulators of the same protein located at various chromosomal arms were also noticed. Identification of structural genes for most of the proteins was found difficult due to multiple regulators encoding the same protein. Two novel subunits (1B(Z,) 1BDz), the structure of which are very similar to the high molecular weight glutenin subunit 12, were identified, and the chromosome arm locations of these subunits were assigned.     

Differential radioactive proteomic analysis of microdissected renal cell carcinoma tissue by 54 cm isoelectric focusing in serial immobilized pH gradient gels

We present a proof of principle study, using laser microdissection and pressure catapulting (LMPC) of two clinical tissue samples, each containing approximately 3.8 microg renal cell carcinoma protein and 3.8 microg normal kidney protein respectively from one patient. The study involved separate radio-iodination of each sample with both (125)I and (131)I, dual inverse replicate sample loading to high resolution 54 cm "daisy chain" serial immobilized pH gradient isoelectric focusing (IPG-IEF) 2D-PAGE gels, co-electrophoretic separation of cross-labeled proteins from different samples, and precision multiplex differential radioactive imaging to obtain signals specific for each sample coelectrophoresed within single gels but labeled with different isotopes of iodine, providing extremely precise intra-gel estimates of the abundance ratio for protein spots from both samples. Twelve multiplexed analytical radioactive SDS-gels from 4 serial IPG-IEF gels provided 24 individual radioactive images for a comprehensive analytical protein multiplex quantification study. A further 12 SDS gels containing (125)I-labeled sample were coelectrophoresed with preparative protein amounts obtained from whole tissue sections for the mass spectrometric identification of comigrating proteins. This consumed <40% of the (125)I-labeled sample, and <20% of the (131)I-labeled sample from the respective original 3.8 microg samples. Twenty-nine proteins were identified by mass spectrometry with PMF scores >70 that were >2-fold differentially abundant between the samples and t-test probabilities <0.05. We conclude that this combination of technologies provides excellent quality protein multiplex data for the differential abundance analysis of large numbers of proteins from extremely small samples, and is applicable to a broad range of clinical and related applications.     

Proteome analysis of rat hippocampal neurons by multiple large gel two-dimensional electrophoresis

The goal of the present study was to detect as many protein spots as possible in mammalian cells using two-dimensional gel electrophoresis (2-DE). For proteome analysis, it is of importance to reveal as many proteins as possible. A single standard 2-DE gel (pH 3-10, 18 cm x 20 cm, 13.5% gel) could detect 853 spots from proteins of cultured rat hippocampal neurons when visualized by silver staining. To increase the resolution of the separation and the number of detectable proteins by 2-DE, we utilized seven different narrow pH range immobilized pH gradients in the first dimension. In the second dimension, fourteen long SDS polyacrylamide gels were used: seven 7.5% gels for the separation of high molecular mass proteins (> or = 40 kDa) and seven 13.5% gels for the separation of low molecular mass proteins (< or = 40 kDa). Three hundred and sixty microg of proteins from cultured hippocampal neurons were loaded on to individual gels and visualized by silver staining. All 14 gel images were assembled into a 70 cm x 67 cm cybergel that contained 6677 protein spots, thereby indicating that the utilization of the present strategy led to a 783% increase in the number of detected spots in comparison to the standard procedure. Loading double the amount (720 microg) of proteins on to a 13.5% gel led to a 184% increase in the number of detected spots, thereby indicating that the present strategy has a potential to display more protein spots in the cybergels.     

Evaluation of individual protein errors in silver-stained two-dimensional gels

The relationship between spot volume and variation for all protein spots observed on large format 2D gels when utilising silver stain technology and a model system based on mammalian NSO cell extracts is reported. By running multiple gels we have shown that the reproducibility of data generated in this way is dependent on individual protein spot volumes, which in turn are directly correlated with the coefficient of variation. The coefficients of variation across all observed protein spots were highest for low abundant proteins which are the primary contributors to process error, and lowest for more abundant proteins. Using the relationship between spot volume and coefficient of variation we show it is necessary to calculate variation for individual protein spot volumes. The inherent limitations of silver staining therefore mean that errors in individual protein spot volumes must be considered when assessing significant changes in protein spot volume and not global error.     

A simple pattern classification method for alcohol-responsive proteins that are differentially expressed in mouse brain

Proteomic analysis of brain tissues obtained from two inbred mice, C57BL/6J (B6, an alcohol-preferring strain) and DBA/2J (D2, an alcohol-avoiding strain), that were orally administered 1.5 g/kg ethanol, was performed to investigate alcohol-responsive proteins. To analyze relationships of alcohol-responsive protein spots between B6 and D2 mice, we have developed a simple spot classification method (SCM) for the fully matched spot data sets produced by the Melanie 4 analysis software using the paired two-dimensional (2-D) gels of two strains over time. By applying SCM, 55 protein spots that were differentially expressed in brain tissue were classified into 16 patterns as mirror images (2x8 patterns), and additionally in an ordered fashion such as 'fast turn over' and 'slow turn over' forms, depending on the frequency of repetition and rate of changed expression profile in 2-D gels over time. Searching for any interaction proteins through databases of interacting proteins using the classified data set has led to the construction of a linkage map, which reveals the interrelationship of the alcohol-responsive proteins between different species. Thus, it is suggested that the different responses for alcohol between B6 and D2 may come from differences of the response rates and interactions of different variants of the alcohol-responsive protein family.     

Modeling biological variability in 2-D gel proteomic carcinogenesis experiments

We propose a statistical method to model the underlying distribution of protein spot volumes in 2-D gels using a generalized model (GM). We apply this approach to discover mechanisms of chemical carcinogenesis in a rodent model. We generated 247 protein spots that were common to all gels (n = 18). Traditional statistical methods found 6.5% (13 out of 247) significant protein spots, our GM approach yielded a total of 53 (22.5%) differentially expressed protein spots.     

Identification of PSME3 as a novel serum tumor marker for colorectal cancer by combining two-dimensional polyacrylamide gel electrophoresis with a strictly mass spectrometry-based approach for data analysis

The purpose of this study was to identify and validate novel serological protein biomarkers of human colorectal cancer (CRC). Proteins from matched CRC and adjacent normal tissue samples were resolved by two-dimensional gel electrophoresis. From each gel all spots were excised, and enveloped proteins were identified by MS. By comparison of the resulting protein profiles, dysregulated proteins can be identified. A list of all identified proteins and validation of five exemplarily selected proteins, elevated in CRC was reported previously (Roessler, M., Rollinger, W., Palme, S., Hagmann, M. L., Berndt, P., Engel, A. M., Schneidinger, B., Pfeffer, M., Andres, H., Karl, J., Bodenmuller, H., Ruschoff, J., Henkel, T., Rohr, G., Rossol, S., Rosch, W., Langen, H., Zolg, W., and Tacke, M. (2005) Identification of nicotinamide N-methyltransferase as a novel serum tumor marker for colorectal cancer. Clin. Cancer Res. 11, 6550-6557). Here we describe identification and initial validation of another potential marker protein for CRC. Comparison of tissue protein profiles revealed strong elevation of proteasome activator complex subunit 3 (PSME3) expression in CRC tissue. This dysregulation was not detectable based on the spot pattern. The PSME3-containing spot on tumor gels showed no visible difference to the corresponding spot on matched control gels. MS analysis revealed the presence of two proteins, PSME3 and annexin 4 (ANXA4) in one and the same spot on tumor gels, whereas the matched spot contained only one protein, ANXA4, on control gels. Therefore, dysregulation of PSME3 was masked by ANXA4 and could only be recognized by MS-based analysis but not by image analysis. To validate this finding, antibody to PSME3 was developed, and up-regulation in CRC was confirmed by Western blot analysis and immunohistochemistry. Finally by developing a highly sensitive immunoassay, PSME3 could be detected in human sera and was significantly elevated in CRC patients compared with healthy donors and patients with benign bowel disease. We propose that PSME3 be considered a novel serum tumor marker for CRC that may have significance in the detection and in the management of patients with this disease. Further studies are needed to fully assess the potential clinical value of this marker candidate.     

Biological variation of the platelet proteome in the elderly population and its implication for biomarker research

Knowledge about the extent of total variation experienced between samples from different individuals is of great importance for the design of not only proteomics but every clinical study. This variation defines the smallest statistically significant detectable signal difference when comparing two groups of individuals. We isolated platelets from 20 healthy human volunteers aged 56-100 years because this age group is most commonly encountered in the clinics. We determined the technical and total variation experienced in a proteome analysis using two-dimensional DIGE with IPGs in the pI ranges 4-7 and 6-9. Only spots that were reproducibly detectable in at least 90% of all gels (n = 908) were included in the study. All spots had a similar technical variation with a median coefficient of variation (cv) of about 7%. In contrast, spots showed a more diverse total variation between individuals with a surprisingly low median cv of only 18%. Because most known biomarkers show an effect size in a 1-2-fold range of their cv, any future clinical proteomics study with platelets will require an analytical method that is able to detect such small quantitative differences. In addition, we calculated the minimal number of samples (sample size) needed to detect given protein expression differences with statistical significance.     

失血性休克大鼠肝脏蛋白质的差异分析

 应用蛋白质双向凝胶电泳 (two-dimensional polyacrylamide gel electrophoresis, 2-DE) 技术,分析在急性重度失血性休克 (refractory hemorrhagic shock, RHS) 条件下,大鼠肝脏蛋白质组表达的差异.16 只雄性 Wistar 大鼠随机分成正常对照组 (sham hemorrhage shock, SHS) 和 RHS 模型组,每组 8 只.采用股动脉放血的方法制备模型,在规定时间内处死大鼠并分离肝脏,提取肝脏总蛋白质后进行 2-DE.运用 Image Master 2D Platinum v 5.0 凝胶图像分析软件对 2-DE 凝胶图像进行差异表达分析.有意义的差异蛋白质点用基质辅助激光解析电离飞行时间质谱进行肽质量指纹图谱分析,借助 Swiss-prot 数据库进行蛋白质搜索和鉴定.SHS 组和 RHS 组肝脏的 2-DE 图谱,分别平均识别到 698±11 和 700±13 个蛋白质点,SHS 组和 RHS 组肝脏间平均匹配率达88%～92 %.共发现 10 个差异有意义的蛋白质点,鉴定出了肿瘤抑制性抗原gp96、葡萄糖调节蛋白58、过氧还蛋白Ⅰ、细胞色素b5、谷胱甘肽转移酶、ATP合酶β亚单位、二磷酸果糖酶 B、三磷酸甘油醛脱氢酶等8种蛋白质.结果表明,以 2-DE 技术得到重复性和分辨率都较好的 2-DE图谱,并初步鉴定急性重度失血性休克后大鼠肝脏的差异表达蛋白质,为深入研究失血性休克的生理病理机制及寻找失血性休克预防和治疗的生物标志物提供了依据.     

Multivariate data analysis of two-dimensional gel electrophoresis protein patterns from few samples

One application of 2D gel electrophoresis is to reveal differences in protein pattern between two or more groups of individuals, attributable to their group membership. Multivariate data analytical methods are useful in pinpointing the spots relevant for discrimination by focusing not only on single spot differences, but on the covariance structure between proteins. However, their outcome is dependent on data scaling, and they may fail in producing valid multivariate models due to the much higher number of "irrelevant" spots present in the gels. The case where only few gels are available and where the aim is to find as many as possible of the group-dependent proteins seems particularly difficult to handle. The present paper investigates such a case regarding the effect of scaling and of prefiltering by univariate nonparametric statistics on the selection of spots. Besides, a modified 'autoscaling' of the full data set based on within-group standard deviations is introduced and shown to be advantageous in revealing potential group-dependent proteins additional to those found by prefiltering.     

High-throughput array production using precision glass syringes

The advantages of using 1, 96, or 384 precision glass syringes in automated high-throughput microdispensers in creating highly uniform and reproducible DNA, protein, and organic compound array filters and slides are described. Using the Hydra Microdispenser and Tango Liquid Handling system, 0.1-5 ng (in 50-300 nL) PCR-amplified, human cancer-related genes and housekeeping genes were spotted onto nylon membranes and coated slides. Protein solutions of 50 microg/mL to 1 mg/mL were spotted onto coated slides or onto MaxiSorp 96-well plates. Up to 6144 spots/membrane and up to 1000 spots/slide were printed. The size of the spots created by glass syringes was uniform and reproducible (precision variation of less than 5%) from spot to spot and membrane to membrane. Using a Tango 384 system, a total of ten 6144-spot filters can be produced in approximately 25 min, translating into a spotting speed of 2.5 min/membrane.     

温敏性品种雌雄蚁蚕蛋白质双向电泳图谱差异分析

为探讨温敏性品种限伴雌雄蚁体蛋白组分的差异, 对刚孵出的雌（黑色）、雄（赤色）蚁体蛋白用双向电泳的技术和方法进行蛋白组分的比较分析。结果显示:在易溶蛋白中, 雄蚁和雌蚁分别检测到269和250个蛋白点, 能相互匹配的蛋白点占86.71％, 特异蛋白雄蚁有43个, 雌蚁有25个。在难溶蛋白中, 雄蚁和雌蚁分别检测到370和427个蛋白点, 能互相匹配的占80.10％, 雄、雌蚁的特异蛋白分别为52个和106个。结果表明雌雄蚁体的蛋白组成存在差异。     

Elsie 4: quantitative computer analysis of sets of two-dimensional gel electrophoretograms

We have developed and refined a system for quantitative computer analysis of two-dimensional polyacrylamide gel electrophoretograms. The system, named Elsie 4, is based on one described by Vo et al. (Anal. Biochem. 112, 258 (1981]. It is highly automated. Elsie 4 can find, and measure the intensity of, almost any spot resolvable on two-dimensional gels, including spots visible only as shoulders off larger spots and spots so close together that there is no "valley" between them. It can automatically match the spot patterns of different gels, potentially without the need for a user to provide landmark matches. The matches between paired gels let us follow the synthesis of any spot through a set of gels. Information about a group of matched spots can be obtained by referring to any spot in the group. There is generally no need for a standard or reference gel. Data for two experiments can be combined and compared by matching any gel in one experiment with any gel in the other. There are ways to automatically find possible mismatches in sets of gels. Scans and the results of the analysis can be shown on an image displayer. The programs use function libraries; this helps ensure consistency and increase portability. The programs and functions can be linked together in many ways; this lets users build custom programs for analysis of specific experiments.