Multi-component immunoaffinity subtraction chromatography: an innovative step towards a comprehensive survey of the human plasma proteome

In order to discover novel protein markers indicative of disease processes or drug effects, the proteomics technology platform most commonly used consists of high resolution protein separation by two-dimensional electrophoresis (2-DE), mass spectrometric identification of proteins from stained gel spots and a bioinformatic data analysis process supported by statistics. This approach has been more successful in profiling proteins and their disease- or treatment-related quantitative changes in tissue homogenates than in plasma samples. Plasma protein display and quantitation suffer from several disadvantages: very high abundance of a few proteins; high heterogeneity of many proteins resulting in long charge trains; crowding of 2-DE separated protein spots in the molecular mass range between 45-80 kD and in the isoelectric point range between 4.5 and 6. Therefore, proteomic technologies are needed that address these problems and particularly allow accurate quantitation of a larger number of less abundant proteins in plasma and other body fluids. The immunoaffinity-based protein subtraction chromatography (IASC) described here removes multiple proteins present in plasma and serum in high concentrations effectively and reproducibly. Applying IASC as an upfront plasma sample preparation process for 2-DE, the protein spot pattern observed in gels changes dramatically and at least 350 additional lower abundance proteins are visualized. Affinity-purified polyclonal antibodies (pAbs) are the immunoaffinity reagents used to specifically remove the abundant proteins such as albumin, immunoglobulin G, immunoglobulin A, transferrin, haptoglobin, alpha-1-antitrypsin, hemopexin, transthyretin, alpha-2-HS glycoprotein, alpha-1-acid glycoprotein, alpha-2-macroglobulin and fibrinogen from human plasma samples. To render the immunoaffinity subtraction procedure recyclable, the pAbs are immobilized and cross-linked on chromatographic matrices. Antibody-coupled matrices specific for one protein each can be pooled to form mixed-bed IASC columns. We show that up to ten affinity-bound plasma proteins with similar solubility characteristics are eluted from a mixed-bed column in one step. This facilitates automated chromatographic processing of plasma samples in high throughput, which is desirable in proteomic disease marker discovery projects.     

Distinct proteome features of plasma microparticles

Plasma microparticles (MPs) are spherical cell membrane fragments derived from either apoptotic or activated cells. Characterized by a rich phospholipid moiety and many protein constituents, MPs normally circulate in the blood and contribute to numerous physiological processes. In disease states, MPs derived from the injured organ likely contain valuable markers for determining the site, type, and extent of disease pathology. However, the basic protein characteristics of plasma MPs have yet to be described. In this study, MPs from a pooled plasma sample derived from 16 healthy donors, all of group A blood type, were prepared by ultracentrifugation. Flow cytometry confirmed that a majority of these MPs are smaller than 1 microm. Factor Xa generation assay revealed the presence of tissue factor activity in these MPs, confirming MPs' role in initiating blood coagulation. The MP proteome was analyzed by two-dimensional (2-D) gel electrophoresis performed in triplicate, and compared with a 2-D gel of pooled whole plasma and blood platelets. Overall, plasma MPs displayed distinct protein features and a greater number of protein spots (1021-1055) than that detected in whole plasma (331-370). Protein spots expressed in high abundance in the MP proteome were then excised and submitted for protein identity determination. This process provided protein identification for 169 protein spots and reported their relative protein quantities within the MP proteome. These 169 protein spots represented 83 different proteins and their respective isoforms. Thirty of these proteins have never before been reported in previous proteome analyses of human plasma. These results provide unprecedented information on the MP proteome and create a basis for future studies to understand MP biology and pathophysiology.     

BiomarkerDigger: A versatile disease proteome database and analysis platform for the identification of plasma cancer biomarkers

We have developed a proteome database (DB), BiomarkerDigger ( http://biomarkerdigger.org ) that automates data analysis, searching, and metadata‐gathering function. The metadata‐gathering function searches proteome DBs for protein–protein interaction, Gene Ontology, protein domain, Online Mendelian Inheritance in Man, and tissue expression profile information and integrates it into protein data sets that are accessed through a search function in BiomarkerDigger. This DB also facilitates cross‐proteome comparisons by classifying proteins based on their annotation. BiomarkerDigger highlights relationships between a given protein in a proteomic data set and any known biomarkers or biomarker candidates. The newly developed BiomarkerDigger system is useful for multi‐level synthesis, comparison, and analyses of data sets obtained from currently available web sources. We demonstrate the application of this resource to the identification of a serological biomarker for hepatocellular carcinoma by comparison of plasma and tissue proteomic data sets from healthy volunteers and cancer patients.     

Peptide enrichment and protein fractionation using selective electrophoresis

In recent times, the analysis of the peptidome has become increasingly valuable to gain a better understanding of the critical roles native peptides play in biological processes. Here, we show a technique using a novel electrophoretic device named MF10, for the fractionation of proteins and peptides based on size and also pH in low volume liquid phase under an electric field. A 1 microM, 7-protein and peptide standard mix ranging from 1 to 25 kDa has been used to show peptide migration into a fraction contained by 1-5 kDa membranes. Simultaneous fractionation of the higher mass protein standards to the correct fraction also occurred. To assess the MF10's ability to fractionate more complex samples, human plasma was used to enrich for the peptidome below 5 kDa in the presence of the proteome. Peptide enrichment was achieved while simultaneously fractionating higher mass proteins to three other mass restricted fractions. The utility of this approach is demonstrated with the identification (with at least 2 ppm mass accuracy) of 76 unique peptides, equating to 22 proteins enriched to the 1-5 kDa fraction of the MF10.     

Systematical evaluation of the effects of sample collection procedures on low-molecular-weight serum/plasma proteome profiling

Blood is an ideal source for biomarker discovery. However, little has been done to address the effects of sampling, handling and storage procedures on serum/plasma proteomes. We used magnetic bead-based MALDI-TOF MS to systematically evaluate the influence of each procedure on low-molecular-weight serum/plasma proteome profiling on the basis of the whole spectra. We found that sampling procedures, including the selection of blood collection tubes and anticoagulants, variations in clotting time or time lag before centrifugation, and hemolysis, displayed significant effects on the proteomes. Moreover, serum and plasma were mutually incompatible for proteome comparison. By contrast, overnight fasting, handling procedures, including centrifugation speeds (1500 x g vs. 3000 x g) or time (15 min vs. 30 min), and storage conditions, such as at 4 degrees C or 25 degrees C for up to 24 h or at -80 degrees C for up to 3 months, and repeated freeze/thaw of up to ten cycles, had relatively minor effects on the proteomes based upon our analysis of about 100 peaks. We concluded that low-molecular-weight serum/plasma proteomes were diversely affected by sampling, handling and storage with most change from variations of sampling procedures. We therefore suggest the necessity of standardizing sampling procedure for proteome comparison and biomarker discovery.     

Gender differences in rat plasma proteome in response to high-fat diet

Knowledge of gender differences is important because nutritional recommendations on the basis of data collected using predominantly male subjects may not be valid for women. In the present study, we performed proteomic analysis in plasma of rats fed a high-fat diet (HFD) using 2-DE combined with MALDI-TOF-MS for analysis of differential regulation patterns between male and female plasma proteins. Male rats gained more body weight with increased values of biochemical parameters than female rats. Image analysis and further statistical analysis allowed detection and identification of 31 proteins that were significantly modulated in a gender-dependent manner in response to HFD. Those differential expressed proteins were classified into three groups based on their regulation patterns in response to diet and gender. Consequently, we found 13 proteins showing gender-different regulation in both normal diet (ND) and HFD, where 9 proteins showed identical regulation patterns (Group I) and 4 proteins exhibited opposite regulation mode (Group II) between the genders. Eighteen proteins showed no gender-difference but HFD-responsive regulation (Group III). Of these, Apo A-IV, CRP precursor, Hp precursor, and FGG showed a clear gender difference in both ND and HFD, with the same regulation patterns. Present proteomic research into gender-dimorphic protein modulation in plasma would aid in improvement of gender awareness in the health care system and in implementation of evidence-based gender-specific clinical recommendations.     

Multidimensional fractionation of the bovine skeletal muscle proteome.

The ultimate goal of proteomics is to understand complex biological systems. The first step toward this end is the discovery of protein differences by profiling a given proteome. One approach to proteome profiling is to fractionate it into intact proteins, with subsequent identification and quantitation. In this work, lysates of bovine skeletal muscle were prepared. Reproducible proteome profiles were generated by an automatic two-dimensional protein fractionation system. Proteins were separated by isoelectric point and then by hydrophobicity. The data collected from both separations were used to generate proteome profiles. A high protein content fraction with pl above 8.5 was digested with trypsin, and its main protein component was identified as lysozyme C by matrix assisted laser desorption/ionization-time of flight mass spectrometry.     

Dissecting the human plasma proteome and inflammatory response biomarkers

A central focus of clinical proteomics is to search for biomarkers in plasma for diagnostic and therapeutic use. We studied a set of plasma proteins accessed from the Healthy Human Individual's Integrated Plasma Proteome (HIP²) database, a larger set of curated human proteins, and a subset of inflammatory proteins, for overlap with sets of known protein biomarkers, drug targets, and secreted proteins. Most inflammatory proteins were found to occur in plasma, and over three times the level of biomarkers were found in inflammatory plasma proteins and their interacting protein neighbors compared to the sets of plasma and curated human proteins. Percentage overlaps with Gene Ontology terms were similar between the curated human set and plasma protein set, yet the set of inflammatory plasma proteins had a distinct ontology‐based profile. Most of the major hub proteins within protein‐protein interaction networks of tissue‐specific sets of inflammatory proteins were found to occur in disease pathways. The present study presents a systematic approach for profiling a plasma subproteome's relationship to both its potential range of clinical application and its overlap with complex disease.     

Depletion efficiency and recovery of trace markers from a multiparameter immunodepletion column

The selective removal of high-abundance proteins is considered to be an important prerequisite for a sensitive proteome analysis in plasma. In this study, we examined the "multiaffinity removal system", an immunoaffinity depletion column targeted against six plasma proteins. As determined by sandwich ELISA, the depletion rate for each target protein is >99% over 200 cycles of regeneration. Our data give evidence that two column antibodies are slowly inactivated during the repeated use of the column; however, the individual depletion rate meets the specification of the manufacturer. To estimate a potential loss of analytes after the immunodepletion, we performed spiking/recovery experiments with a selection of tumor markers at concentrations in the lower to medium ng/mL range. The average recovery of 9 out of 11 markers is 78%. A significant proportion of two other markers binds to the column. Based on the average marker recovery and a depletion of ;85% of the total protein we estimate a five-fold enrichment of a potential biomarker by the use of this depletion column. We conclude that the selective depletion of plasma proteins by immunoaffinity chromatography is a valid strategy for the enrichment of potential biomarkers sought by proteomics methodologies.     

Effect of fungal polysaccharides on the modulation of plasma proteins in streptozotocin-induced diabetic rats

To search out novel biomarkers for monitoring diabetes prognosis, we examined the effect of hypoglycemic fungal exopolysaccharides (EPS) on the differential levels of plasma proteins in streptozotocin-induced diabetic rats. The orally administrated EPS exhibited an excellent hypoglycemic effect, lowering the average plasma glucose level, and increasing insulin secretion in diabetic rats. The 2-DE analysis of rat plasma demonstrated that about 500 visualized spots were differentially regulated, of which 20 spots were identified as principal diabetes-associated proteins. The distinct effect of diabetes induction on the pattern of rat plasma proteins includes the down-regulation of albumin, apolipoprotein E (Apo E), alpha1-inhibitor-3, fetuin beta, Gc-globulin, hemopexin, vitronectin, and transthyretin (TTR) monomer, and the up-regulation of Apo A-I, Apo A-IV, ceruloplasmin, alpha1-antitrypsin, serine protease inhibitor III, and transferrin. Those protein levels were interestingly restored to those of healthy rats by EPS treatment, although the order of magnitude of the changes differed widely. Two proteins of interest showed distinct differential expression with opposite trends: TTR tetramer was significantly down-regulated and immunoglobulin (Ig) kappa light chain was significantly up-regulated upon diabetes induction, both of which were also normalized to those of healthy groups after EPS treatment.     

Plasma proteome analysis in diet-induced obesity-prone and obesity-resistant rats

One of the major issues in the field of obesity is why some humans become obese and others resist development of obesity when exposed to high-calorie diets. Despite the same genetic background, namely obesity-prone (OP) and -resistant (OR) rats, differing responses have been demonstrated in a high fat diet-induced rodent model. The aim of the present study was to discover novel obesity-related biomarkers for susceptibility and/or resistance to obesity by proteomic analysis of OP and OR rat plasma. After feeding of high fat diet, OP rats gained approximately 25% more body weight than OR rats and were used for proteomic analysis using 2-DE combined with MALDI-TOF-MS. We categorized identified proteins into three groups by analysis of both average spot density in each group and individual spot density of six rats as a function of body weight. Consequently, category (1) included inter-α-inhibitor H4 heavy chain and fetuin B precursor, which can be used as novel plasma biomarkers for risk of obesity. Nine proteins of category (2) and (3) can also be plausible plasma markers in the study of obesity. This proteomic study is an important advancement over the previous steps needed for identification of OP and OR rats.     

Heat-shock protein 27: a potential biomarker for hepatocellular carcinoma identified by serum proteome analysis

Hepatocellular carcinoma (HCC) is the third leading cause of cancer mortality worldwide and ranks second in China. The prognosis of HCC remains dismal mainly because of its late diagnosis, especially in patients with coexisting chronic liver diseases. To identify serum biomarkers for HCC, sera from 20 healthy volunteers, 20 hepatitis B virus (HBV) infected patients and 20 HCC patients were selected for screening study and same number of sera into the same three groups were used for validation study. A strategy including sonication, albumin and immunoglobulin G (IgG) depletion and desalting was optimized for screening differentially expressed proteins of low abundance in serum. By 2-DE image analysis and MALDI-TOF-MS/MS identification, eight proteins including heat-shock protein 27 (HSP27), alpha-fetoprotein (AFP), alpha-1 antitrypsin, clusterin, caeruloplasmin, haptoglobin alpha2 chain, tranferrin and transthyretin were found significantly changed among the healthy, HBV and HCC groups. Further validation study by Western blot showed the detection of HSP27 in 90% HCC sera and two HBV sera, but in none of normal sera. Thus, 2-DE based serum proteome analysis can be useful in the screening of serum biomarkers for HCC and HSP27 could aid in the diagnosis of HCC though further validation is needed.     

Characterization of a high field Orbitrap mass spectrometer for proteome analysis

The field of proteomics continues to be driven by improvements in analytical technology, notably in peptide separation, quantitative MS, and informatics. In this study, we have characterized a hybrid linear ion trap high field Orbitrap mass spectrometer (Orbitrap Elite) for proteomic applications. The very high resolution available on this instrument allows 95% of all peptide masses to be measured with sub‐ppm accuracy that in turn improves protein identification by database searching. We further confirm again that mass accuracy in tandem mass spectra is a valuable parameter for improving the success of protein identification. The new CID rapid scan type of the Orbitrap Elite achieves similar performance as higher energy collision induced dissociation fragmentation and both allow the identification of hundreds of proteins from as little as 0.1 ng of protein digest on column. The new instrument outperforms its predecessor the Orbitrap Velos by a considerable margin on each metric assessed that makes it a valuable and versatile tool for MS‐based proteomics.     

The dynamics of the proteome: strategies for measuring protein turnover on a proteome-wide scale.

Quantitative proteomics captures the steady-state amount of a protein in a cell but does not explain how a change in protein amount is manifest -- whether through a change in synthesis or a change in degradation. If we are to understand the changes in the proteome, we will need to define such processes. In this brief review, strategies for the determination of intracellular protein dynamics on a proteome-wide scale are discussed.     

Establishment of a proteome profile and identification of molecular markers for mouse spermatogonial stem cells

Spermatogonial stem cells (SSCs) are undifferentiated cells that are required to maintain spermatogenesis throughout the reproductive life of mammals. Although SSC transplantation and culture provide a powerful tool to identify the mechanisms regulating SSC function, the precise signalling mechanisms governing SSC self‐renewal and specific surface markers for purifying SSCs remain to be clearly determined. In the present study, we established a steady SSC culture according to the method described by Shinohara's lab. Fertile progeny was produced after transplantation of cultured SSCs into infertile mouse testis, and the red fluorescence exhibited by the culture cell membranes was stably and continuously transmitted to the offspring. Next, via advanced mass spectrometry and an optimized proteomics platform, we constructed the proteome profile, with 682 proteins expressed in SSCs. Furthermore bioinformatics analysis showed that the list contained several known molecules that are regulated in SSCs. Several nucleoproteins and membrane proteins were chosen for further exploration using immunofluorescence and RT‐PCR. The results showed that SALL1, EZH2, and RCOR2 are possibly involved in the self‐renewal mechanism of SSCs. Furthermore, the results of tissue‐specific expression analysis showed that Gpat2 and Pld6 were uniquely and highly expressed in mouse testes and cultured SSCs. The cellular localization of PLD6 was further explored and the results showed it was primarily expressed in the spermatogonial membrane of mouse testes and cultured SSCs. The proteins identified in this study form the basis for further exploring the molecular mechanism of self‐renewal in SSCs and for identifying specific surface markers of SSCs.     

Evaluation of reproducibility of protein identification results after multidimensional human serum protein separation

We developed a gel- and label-free proteomics platform for comparative studies of human serum. The method involves the depletion of the six most abundant proteins, protein fractionation by Off-Gel IEF and RP-HPLC, followed by tryptic digestion, LC-MS/MS, protein identification, and relative quantification using probabilistic peptide match score summation (PMSS). We evaluated performance and reproducibility of the complete platform and the individual dimensions, by using chromatograms of the RP-HPLC runs, PMSS based abundance scores and abundance distributions as objective endpoints. We were interested if a relationship exists between the quantity ratio and the PMSS score ratio. The complete analysis was performed four times with two sets of serum samples containing different concentrations of spiked bovine beta-lactoglobulin (0.1 and 0.3%, w/w). The two concentrations resulted in significantly differing PMSS scores when compared to the variability in PMSS scores of all other protein identifications. We identified 196 proteins, of which 116 were identified four times in corresponding fractions whereof 73 qualified for relative quantification. Finally, we characterized the PMSS based protein abundance distributions with respect to the two dimensions of fractionation and discussed some interesting patterns representing discrete isoforms. We conclude that combination of Off-Gel electrophoresis (OGE) and HPLC is a reproducible protein fractionation technique, that PMSS is applicable for relative quantification, that the number of quantifiable proteins is always smaller than the number of identified proteins and that reproducibility of protein identifications should supplement probabilistic acceptance criteria.     

In-depth analysis of the human CSF proteome using protein prefractionation

The identification of disease markers in human body fluids requires an extensive and thorough analysis of its protein constituents. In the present study, we have extended our analysis of the human cerebrospinal fluid (CSF) proteome using protein prefractional followed by shotgun mass spectrometry. After the removal of abundant protein components from the mixture with the help of immunodepletion affinity chromatography, we used either anion exchange chromatography or sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to further subfractionate the proteins present in CSFs. Each protein subfraction was enzyme digested and analyzed by tandem mass spectrometry and the resulting data evaluated using the Spectrum Mill software. Different subfractionation methods resulted in the identification of a grant total of 259 proteins in CSF from a patient with normal pressure hydrocephalus. The greatest number of protein, 240 in total, were identified after prefractionating the CSF proteins by immunodepletion and SDS-PAGE. Immuno-depletion combined with anion exchange fractionation resulted in 112 proteins and 74 proteins were found when only immunodepletion of the CSF samples was carried out. All methods used showed a significant increase in the number of identified proteins as compared with nondepleted and unfractionated CSF sample analysis, which yielded only 38 protein identifications. The present work establishes a platform for future studies aimed at a detailed comparative proteome analysis of CSFs from different groups of patients suffering from various psychiatric and neurological disorders.     

Benefits of heat treatment to the protease packed neutrophil for proteome analysis: halting protein degradation

Neutrophils, cells of the innate immune system, contain an array of proteases and reactive oxygen species-generating enzymes that assist in controlling the invasion of bacteria and pathogens. The high content of intracellular proteolytic enzymes makes them difficult cells to work with as they can degrade proteins of potential interest. Here, we describe the benefits of heat treatment of neutrophils in reducing protein degradation for subsequent proteome analysis. Neutrophils isolated from four healthy volunteers were each divided into three aliquots and subjected to different preparation methods for 2-DE: (i) Heat treatment, (ii) resuspension in NP40 lysis buffer and (iii) resuspension in standard 2-DE lysis buffer. Representative spots found to be statistically significant between groups (p<0.01) were excised and identified by LC-MS/MS, three of which were validated by immunoblotting. Heat-treated samples contained proteins in the high-molecular-weight range that were absent from NP40-treated samples. Moreover, NP40-treated samples showed an increase in spot number and volume at lower molecular weights suggestive of protein degradation. Incorporating heat treatment into sample preparation resulted in the identification of proteins that may not have previously been detected due to sample degradation, thus leading to a more comprehensive 2-DE map of the human neutrophil proteome.