A novel strategy for quantitative proteomics using isotope-coded protein labels

Stable isotope labelling in combination with mass spectrometry has emerged as a powerful tool to identify and relatively quantify thousands of proteins within complex protein mixtures. Here we describe a novel method, termed isotope-coded protein label (ICPL), which is capable of high-throughput quantitative proteome profiling on a global scale. Since ICPL is based on stable isotope tagging at the frequent free amino groups of isolated intact proteins, it is applicable to any protein sample, including extracts from tissues or body fluids, and compatible to all separation methods currently employed in proteome studies. The method showed highly accurate and reproducible quantification of proteins and yielded high sequence coverage, indispensable for the detection of post-translational modifications and protein isoforms. The efficiency (e.g. accuracy, dynamic range, sensitivity, speed) of the approach is demonstrated by comparative analysis of two differentially spiked proteomes.     

Use of a fluorescent internal protein standard to achieve quantitative two-dimensional gel electrophoresis

2-DE is a powerful separation method for complex protein mixtures. However, large intergel variations in spot intensity limit its use for quantitative proteomics studies. To address this issue, we developed a fluorescent internal protein standard for use in 2-DE analysis. Protein samples are spiked with an Alexa-labeled internal standard (ALIS) prior to separation with 2-DE. Due to the high extinction coefficient of the Alexa-fluor, incorporation of 0.1% of total protein is sufficient to allow visualization of the internal standard yet low enough to avoid interference in subsequent quantification and identification steps. Following 2-DE, total proteins are visualized with fluorescent postelectrophoretic stains spectrally separated from ALIS. Four protein stains, Deep Purple, Sulforhodamine G, ruthenium II-tris(bathophenanthroline disulfonate) (RuTBS), and SYPRO Ruby, including improved purification and staining protocols for RuTBS and ten-fold dilutions of SYPRO Ruby were evaluated. All staining protocols were compatible with the ALIS method and had similar LODs (1-4 ng) and dynamic ranges (10(3)). ALIS is a powerful normalization method for quantitative 2-DE which avoids potential problems associated with dual spot migration patterns observed in the DIGE method. Furthermore, ALIS provides significantly improved normality in the distribution of spot abundance-variance compared to normalization through division by the total spot volume.     

A quantitative investigation into the losses of proteins at different stages of a two-dimensional gel electrophoresis procedure

We report the results of a systematic investigation to quantify the losses of protein during a well-established two-dimensional polyacrylamide gel electrophoresis (2-DE) procedure. Radioactively labelled proteins ([(14)C]bovine serum albumin and a homogenate prepared from the liver of a rat that had been injected with [(35)S]methionine) were used, and recovery was quantified by digesting pieces of gel in H(2)O(2) and subjecting the digests to liquid scintillation counting. When samples were loaded onto the first dimension immobilised pH gradient strips by in-gel rehydration, recovery of protein from the strips was 44-80% of the amount of protein loaded, depending on the amount of protein in the sample. Most of the unrecovered protein appeared to have adhered to the reswelling tray. Losses during isoelectric focusing (IEF) were much smaller (7-14%), although approximately 2% of the protein appeared to migrate from sample strips to adjacent blank strips in the focussing apparatus. A further 17-24% of the proteins were lost into the buffers during equilibration prior to running in the second dimension. Losses during the second dimension run and subsequent staining with SYPRO Ruby amounted to less than 10%. The overall loss during 2-DE was reduced by approximately 25% when proteins were loaded onto the IEF strips using sample cups instead of by in-gel rehydration. These extensive and variable losses during the 2-DE procedure mean that spot intensities on 2-DE gels cannot be used to derive reliable, quantitative information on the amounts of proteins present in the original sample.     

Identification of membrane-associated proteins from Campylobacter jejuni strains using complementary proteomics technologies

Campylobacter jejuni is the leading cause of food- and water-borne illness world-wide. The membrane-associated proteome of a recent C. jejuni gastrointestinal isolate (JHH1) was generated by sodium carbonate precipitation and ultracentrifugation followed by 2-DE and MALDI-TOF MS as well as 2-DLC (strong cation exchange followed by RP chromatography) of trypsin digests coupled to MS/MS (2-DLC/MS/MS). 2-DE/MS identified 77 proteins, 44 of which were predicted membrane proteins, while 2-DLC/MS/MS identified 432 proteins, of which 206 were predicted to be membrane associated. A total of 453 unique proteins (27.4% of the C. jejuni theoretical proteome), including 187 bona fide membrane proteins were identified in this study. Membrane proteins were also compared between C. jejuni JHH1 and ATCC 700297 to identify factors potentially associated with increased gastrointestinal virulence. We identified 28 proteins that were significantly (>two-fold) more abundant in, or unique to, JHH1, including eight proteins involved in chemotaxis signal transduction and flagellar motility, the amino acid-binding surface antigens CjaA and CjaC, and four outer membrane proteins (OMPs) of unknown function (Cj0129c, Cj1031, Cj1279c, and Cj1721c). Immunoblotting using convalescent patient sera generated post-gastrointestinal infection revealed 13 (JHH1) and 12 (ATCC 700297) immunoreactive proteins. These included flagellin (FlaA) and CadF as well as Omp18, Omp50, Cj1721c, PEB1A, PEB2, and PEB4A. This study provides a comprehensive analysis of membrane-associated proteins from C. jejuni.     

Potato tuber proteomics: comparison of two complementary extraction methods designed for 2-DE of acidic proteins

Two protein extraction procedures were tested in order to remove interfering compounds prior to 2-DE of potato tubers. These methods using SDS lysis buffer and phenol-phase extraction were compared regarding the quality of the resulting 2-D gel. While the resolution of SDS extracts on semipreparative gels seems better, both methods lead to similar extraction yields and total number of spots. The procedures are complementary regarding the Mr range of preferentially extracted proteins.     

Subcellular proteomics of cell differentiation: quantitative analysis of the plasma membrane proteome of Caco-2 cells

Human colorectal carcinoma (Caco-2) cells undergo in culture spontaneous enterocytic differentiation, characterized by polarization and appearance of the functional apical brush border membrane. To provide insights into the biology of differentiation, we have performed a comparative proteomic analysis of the plasma membranes from proliferating cells (PCs) and the apical membranes from differentiated cells (DCs). Proteins were resolved by SDS-PAGE, in-gel digested and analyzed by RP-LC and MS/MS. Alternatively, proteins were digested in solution, and tryptic peptides were labeled with isotopic tags and analyzed by 2-D LC followed by MS/MS. Among the 1125 proteins identified in both proteomes, 76 were found to be significantly increased in the membranes of DCs and 61 were increased in PCs. Majority of the proteins increased in the apical membranes were metabolic enzymes, proteins involved in the maintenance of cellular structure, transmembrane transporters, and proteins regulating vesicular transport. In contrast, majority of the proteins increased in the membranes of PCs were involved in gene expression, protein synthesis, and folding. Both groups contained many novel proteins with yet to be identified functions, which could provide potential new markers of the intestinal cells or of colorectal cancer.     

Plasma proteomics of lung cancer by a linkage of multi-dimensional liquid chromatography and two-dimensional difference gel electrophoresis

To investigate aberrant plasma proteins in lung cancer, we compared the proteomic profiles of serum from five lung cancer patients and from four healthy volunteers. Immuno-affinity chromatography was used to deplete highly abundant plasma proteins, and the resulting plasma samples were separated into eight fractions by anion-exchange chromatography. Quantitative protein profiles of the fractionated samples were generated by two-dimensional difference gel electrophoresis, in which the experimental samples and the internal control samples were labeled with different dyes and co-separated by two-dimensional polyacrylamide gel electrophoresis. This approach succeeded in resolving 3890 protein spots. For 364 of the protein spots, the expression level in lung cancer was more than twofold different from that in the healthy volunteers. These differences were statistically significant (Student's t-test, p-value less than 0.05). Mass spectrometric protein identification revealed that the 364 protein spots corresponded to 58 gene products, including the classical plasma proteins and the tissue-leakage proteins catalase, clusterin, ficolin, gelsolin, lumican, tetranectin, triosephosphate isomerase and vitronectin. The combination of multi-dimensional liquid chromatography and two-dimensional difference gel electrophoresis provides a valuable tool for serum proteomics in lung cancer.     

Partial characterization of the proteome of the mouse striatum

Many diseases of the mammalian CNS, including Parkinson's (PD) and Lesch Nyhan disease (LND), are associated with programmatic neurodegeneration or dysfunction of dopaminergic neurons in the mesencephalon, the nigrostriatal pathway, and its projections in the striatum [1-4]. Proteomic studies on brain tissue of both animal models and human PD patients have provided evidence for dysfunction and damage of many pathways, including oxidative stress-related damage, ubiquitin-proteasome dysfunction, mitochondrial energy metabolism deficiencies, and synaptic function [5-11]. To date no such proteomic studies have been reported in the related and rare basal ganglia disorder LND, a developmental rather than a neurodegenerative neurological disorder caused by deficiency of the enzyme hypoxanthine-guanine phosphoribosyltransferase (HPRT) that regulates a major step in the purine salvage pathway [12]. Many studies have demonstrated that the both human LND patients and a mouse knockout model of HPRT deficiency have significantly reduced levels and uptake of dopamine in the striatum [4, 13-16] that is likely to be the principal cause of the CNS disorder. The precise molecular and cellular mechanisms that underlie this neurotransmitter defect are unknown.     

Investigating sample pooling strategies for DIGE experiments to address biological variability

If biological questions are to be answered using quantitative proteomics, it is essential to design experiments which have sufficient power to be able to detect changes in expression. Sample subpooling is a strategy that can be used to reduce the variance but still allow studies to encompass biological variation. Underlying sample pooling strategies is the biological averaging assumption that the measurements taken on the pool are equal to the average of the measurements taken on the individuals. This study finds no evidence of a systematic bias triggered by sample pooling for DIGE and that pooling can be useful in reducing biological variation. For the first time in quantitative proteomics, the two sources of variance were decoupled and it was found that technical variance predominates for mouse brain, while biological variance predominates for human brain. A power analysis found that as the number of individuals pooled increased, then the number of replicates needed declined but the number of biological samples increased. Repeat measures of biological samples decreased the numbers of samples required but increased the number of gels needed. An example cost benefit analysis demonstrates how researchers can optimise their experiments while taking into account the available resources.     

Simulated hyperglycemia in rat cardiomyocytes: a proteomics approach for improved analysis of cellular alterations

Diabetic hyperglycemia can lead to stress-related cellular apoptosis of cardiac tissue. However, the mechanism by which hyperglycemia inflicts this damage on the structure and function of the heart is unclear. In this study, we examined the relationship between proteome alterations, mitochondrial function, and major biochemical and electrophysiological changes affecting cardiac performance during simulated short-term hyperglycemia. Two-dimensional comparative proteomics analysis of rat hearts perfused with glucose at high (30 mM) or control (5.5 mM) levels revealed that glucose loading alters cardiomyocyte proteomes. It increased expression levels of initial enzymes of the tricarboxylic acid cycle, and of enzymes of fatty acid beta-oxidation, with consequent up-regulation of enzymes of mitochondrial electron transport. It also markedly decreased expression of enzymes of glycolysis and the final steps of the tricarboxylic acid cycle. Glucose loading increased the rate of Bax-independent apoptosis. High glucose increased the duration of the action potential and elevated level of intracellular cytoplasmic calcium. Surprisingly, glucose loading did not influence levels of nitric oxide or mitochondrial superoxide in isolated cardiomyocytes. In summary, short-term simulated hyperglycemia attenuated expression of many anti-apoptotic proteins. This effect was apparently mediated via alterations in multiple biochemical pathways that collectively increased apoptotic susceptibility.     

A multifaceted analysis of viperid snake venoms by two-dimensional gel electrophoresis: an approach to understanding venom proteomics

The complexity of Viperid venoms has long been appreciated by investigators in the fields of toxinology and medicine. However, it is only recently that the depth of that complexity has become somewhat quantitatively and qualitatively appreciated. With the resurgence of two-dimensional gel electrophoresis (2-DE) and the advances in mass spectrometry virtually all venom components can be visualized and identified given sufficient effort and resources. Here we present the use of 2-DE for examining venom complexity as well as demonstrating interesting approaches to selectively delineate subpopulations of venom proteins based on particular characteristics of the proteins such as antibody cross-reactivity or enzymatic activities. 2-DE comparisons between venoms from different species of the same genus (Bothrops) of snake clearly demonstrated both the similarity as well as the apparent diversity among these venoms. Using liquid chromatography/tandem mass spectrometry we were able to identify regions of the two-dimensional gels from each venom in which certain classes of proteins were found. 2-DE was also used to compare venoms from Crotalus atrox and Bothrops jararaca. For these venoms a variety of staining/detection protocols was utilized to compare and contrast the venoms. Specifically, we used various stains to visualize subpopulations of the venom proteomes of these snakes, including Coomassie, Silver, Sypro Ruby and Pro-Q-Emerald. Using specific antibodies in Western blot analyses of 2-DE of the venoms we have examined subpopulations of proteins in these venoms including the serine proteinase proteome, the metalloproteinase proteome, and the phospholipases A2 proteome. A functional assessment of the gelatinolytic activity of these venoms was also performed by zymography. These approaches have given rise to a more thorough understanding of venom complexity and the toxins comprising these venoms and provide insights to investigators who wish to focus on these venom subpopulations of proteins in future studies.     

Comparison of extraction procedures for proteome analysis of Streptococcus pneumoniae and a basic reference map

Streptococcus pneumoniae is an important human pathogen causing life-threatening invasive diseases such as pneumonia, meningitis and bacteraemia. Despite major advances in our understanding of pneumococcal mechanisms of pathogenicity obtained through genomic studies very little has been achieved on the characterisation of the proteome of this pathogen. The highly complex structure of its cell envelope particularly amongst the various capsular forms enables the cell to resist lysis by conventional mechanical methods. It is therefore highly desirable to develop a cellular lysis and protein solubilisation procedure that minimises protein losses and allows for maximum possible coverage of the proteome of S. pneumoniae. Here we have utilised various combinations of mechanical or enzymatic cell lysis with two protein solubilisation mixtures urea/CHAPS-based mixture or SDS/DTT-based mixture in order to achieve best quality protein profiles using two proteomic technologies surface-enhanced laser desorption ionisation (SELDI) TOF MS and 2-DE. While urea/CHAPS-based mixture combined with freeze/thawing provided enough material for good-quality SELDI TOF MS fingerprints, a combination of mechanical, enzymatic and chemical lysis was needed to be used to successfully extract the desired protein content for 2-DE analysis. The methods chosen were also assessed for reproducibility and tested on various capsular types of S. pneumoniae. As a result, good-quality and reproducible profiles were created using various ProteinChip arrays and more than 800 protein spots were separated on a single 2-D gel of S. pneumoniae. Twenty-five of the most abundant protein spots were identified using LC/MS/MS to create a reference map of S. pneumoniae. The proteins identified included glycolytic enzymes such as glyceraldehyde 3-phosphate dehydrogenase, phosphoglycerate kinase, enolase etc. Several fermentation enzymes were also present including two of the components of the arginine deiminase system. Proteins involved in protein synthesis, such as translation factors and ribosomal proteins, as well as several chaperone proteins were also identified.     

Beta-amyloid treatment of two complementary P301L tau-expressing Alzheimer's disease models reveals similar deregulated cellular processes

Alzheimer's disease (AD) is characterized by Abeta peptide-containing plaques and tau-containing neurofibrillary tangles (NFTs). Both pathologies have been combined by crossing Abeta plaque-forming APP mutant mice with NFT-forming P301L tau mutant mice or by stereotaxically injecting beta-amyloid peptide 1-42 (Abeta42) into brains of P301L tau mutant mice. In cell culture, Abeta42 induces filamentous tau aggregates. To understand which processes are disrupted by Abeta42 in the presence of tau aggregates, we applied comparative proteomics to Abeta42-treated P301L tau-expressing neuroblastoma cells and the amygdala of P301L tau transgenic mice stereotaxically injected with Abeta42. Remarkably, a significant fraction of proteins altered in both systems belonged to the same functional categories, i.e. stress response and metabolism. We also identified model-specific effects of Abeta42 treatment such as differences in cell signaling proteins in the cellular model and of cytoskeletal and synapse associated proteins in the amygdala. By Western blotting (WB) and immunohistochemistry (IHC), we were able to show that 72% of the tested candidates were altered in human AD brain with a major emphasis on stress-related unfolded protein responsive candidates. These data highlight these processes as potentially important initiators in the Abeta42-mediated pathogenic cascade in AD and further support the role of unfolded proteins in the course of AD.     

Comparative proteomic study of acute pulmonary embolism in a rat model

Pulmonary embolism (PE) is a common, potentially fatal disease, whose blood clots originate from the deep venous system of the lower extremities. PE is of clinical importance because of the considerable mortality and morbidity. In this study, at first we established a rat PE model by injecting 3-4 emboli into the left jugular vein. Before collecting the lung tissues, we perfused them with saline through the right jugular vein and at the same time cut off the right carotid to remove the blood. Then we separated and identified differentially expressed proteins in lung tissues at different time points using the techniques of 2-DE and MS. After image analysis of 2-DE gels, 46 protein spots of interest were excised from the gels and identified by MALDI-TOF-MS. Thirty-two protein spots of them found their corresponding protein candidates in the database. These proteins are associated with distinct aspects of PE such as the contractive function of smooth muscles, metabolism of energy, collagen and toxicant, cellular differentiation, apoptosis and injury, blood pressure adjustment, maintaining of acid-base balance, and so on. Ten of the identified proteins were validated by semiquantitative RT-PCR, and three of them were further validated by Western blot analysis. The differential expression patterns of these proteins suggest the distinct roles they may play in different stages of the rat PE model, and information from this study may be helpful to uncover the pathophysiologic molecular mechanisms involved in PE.     

Optimization of paper bridge loading for 2-DE analysis in the basic pH region: application to the mitochondrial subproteome

Separation of basic proteins with 2-DE presents technical challenges involving protein precipitation, load limitations, and streaking. Cardiac mitochondria are enriched in basic proteins and difficult to resolve by 2-DE. We investigated two methods, cup and paper bridge, for sample loading of this subproteome into the basic range (pH 6-11) gels. Paper bridge loading consistently produced improved resolution of both analytical and preparative protein loads. A unique benefit of this technique is that proteins retained in the paper bridge after loading basic gels can be reloaded onto lower pH gradients (pH 4-7), allowing valued samples to be analyzed on multiple pH ranges.     

Potential protein markers for nutritional health effects on colorectal cancer in the mouse as revealed by proteomics analysis

It is suggested that colorectal cancer might be prevented by changes in diet, and vegetable consumption has been demonstrated to have a protective effect. Until now, little is known about the effects of vegetable consumption at the proteome level. Therefore, the effect of increased vegetable intake on the protein expression in the colonic mucosa of healthy mice was studied. Aim was to identify the proteins that are differentially expressed by increased vegetable consumption and to discriminate their possible role in the protection against colorectal cancer. Mice were fed four different vegetable diets, which was followed by analysis of total cellular protein from colonic mucosal cells by a combination of 2-DE and MS. We found 30 proteins that were differentially expressed in one or more diets as compared to the control diet. Six could be identified by MALDI-TOF MS: myosin regulatory light chain 2, carbonic anhydrase I, high-mobility group protein 1, pancreatitis-associated protein 3, glyceraldehyde-3-phosphate dehydrogenase and ATP synthase oligomycin sensitivity conferral protein. Alterations in the levels of these proteins agree with a role in the protection against colon cancer. We conclude that these proteins are suitable markers for the health effect of food on cancer. The observed altered protein levels therefore provide support for the protective effects of vegetables against colorectal cancer.     

Changes in liver protein abundance in inbred alcohol-preferring rats due to chronic alcohol exposure, as measured through a proteomics approach

This study compares the total liver proteome of inbred alcohol‐preferring line (iP) rats exposed to alcohol with iP rats without alcohol experience. Rat liver proteins were extracted using a three‐step procedure. Each of the three solutions solubilizes a different set of proteins. The extracted proteins were separated by 2‐DE. Scanned gels of two sample groups, alcohol‐exposed iP and alcohol‐naïve iP, were compared, revealing many protein spots with significantly higher or lower densities. These spots were cut from the gel, destained, and subjected to trypsin digestion and subsequent identification by LC‐MS/MS. Twenty‐four individual rats, 12 alcohol‐naïve, and 12 alcohol‐exposed, were used in this study. Two groups, each containing six naïve and six exposed animals, were created for statistical comparison. For the first group, 64 spots were observed to have statistically significant intensity differences upon alcohol exposure across all three extracts while 118 such spots were found in the second group. There were 113 unique proteins in both groups together. The majority of these proteins were enzymes. Significant changes are observed for three major metabolic pathways: glycolysis, gluconeogenesis, and fatty acid β‐oxidation. In addition, enzymes involved in protein synthesis and antioxidant activity show significant changes in abundance in response to alcohol exposure.     

A proteomic approach to studying the differentiation of neural stem cells

The mechanisms that regulate the maintenance of stem cell self-renewal versus differentiation are complex and remain mostly unknown. Understanding neurogenesis and neural cell differentiation presents a unique challenge for the treatment of nervous system disorders. To gain more insight into molecular mechanisms of the differentiation of neural cells, we combined the advantage of porcine fetal neural stem cells (NSCs) in vitro differentiation model and proteomic analysis. Using 2-DE followed by MS, we profiled constituent proteins of NSCs and their differentiated progenies at first and then indicated protein species that were significantly up- or down-regulated during the differentiation. The largest identified group of constituent proteins was related to RNA and protein metabolism and processing, including chaperones, and the second largest consisted of proteins involved in cell organization (cytoskeleton and annexins). Differentiation of neural cells was found to be accompanied by changes in the expression of proteins involved in DNA and RNA binding, mRNA processing and transport, stress responses, iron storage, and redox regulation. Additional immunoblot analysis verified the induction of alpha-B crystallin and heterogeneous nuclear ribonucleoproteins (hnRNPs) A1 and A2/B1. Furthermore, immunocytochemistry demonstrated specific localization of alpha-B crystallin in the cytoplasm or nucleus of glial cells and confirmed cellular expression patterns of hnRNPs A1 and A2/B1. These findings represent a significant step towards understanding neural cell differentiation and identification of the regulatory proteins associated with this process.     

Proteomics of rice and Cochliobolus miyabeanus fungal interaction: Insight into proteins at intracellular and extracellular spaces

Necrotrophic fungal pathogen Cochliobolus miyabeanus causes brown spot disease in rice leaves upon infection, resulting in critical rice yield loss. To better understand the rice–C. miyabeanus interaction, we employed proteomic approaches to establish differential proteomes of total and secreted proteins from the inoculated leaves. The 2DE approach after PEG‐fractionation of total proteins coupled with MS (MALDI‐TOF/TOF and nESI‐LC‐MS/MS) analyses led to identification of 49 unique proteins out of 63 differential spots. SDS‐PAGE in combination with nESI‐LC‐MS/MS shotgun approach was applied to identify secreted proteins in the leaf apoplast upon infection and resulted in cataloging of 501 unique proteins, of which 470 and 31 proteins were secreted from rice and C. miyabeanus, respectively. Proteins mapped onto metabolic pathways implied their reprogramming upon infection. The enzymes involved in Calvin cycle and glycolysis decreased in their protein abundance, whereas enzymes in the TCA cycle, amino acids, and ethylene biosynthesis increased. Differential proteomes also generated distribution of identified proteins in the intracellular and extracellular spaces, providing a better insight into defense responses of proteins in rice against C. miyabeanus. Established proteome of the rice–C. miyabeanus interaction serves not only as a good resource for the scientific community but also highlights its significance from biological aspects.