Progress with proteome projects: why all proteins expressed by a genome should be identified and how to do it

Liquid chromatography and electrospray mass spectrometric mapping of peptides from human plasma filtrate

We present a multidimensional approach to map the composition of complex peptide mixtures obtained as crude extract from biological liquids by (1) cation exchange chromatography and (2) subsequent microbore reversed-phase liquid chromatography and electrospray mass spectrometry coupling (LC-MS). Human hemofiltrate is an equivalent to blood and is used to obtain peptide material in large quantities from patients with chronic renal failure. The upper exclusion limit of the filtration membranes used results in a protein-free filtrate containing peptides in a range up to 20 ku. Using this unique peptide source, several thousand peptides were detected and an LC-MS data base of circulating human peptides was created. The search for known peptides by their molecular mass is a reliable method to guide peptide purification.

Nanoliter chemistry combined with mass spectrometry for peptide mapping of proteins from single mammalian cell lysates

A nanoliter-chemistry station combined with matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry was developed to characterize proteins at the attomole level. Chemical reactions including protein digestion were carried out in nanoliter or subnanoliter volumes, followed by microspot sample deposition of the digest to a MALDI-TOF mass spectrometer. Accurate mass determination of the peptides from the enzyme digest, in conjunction with protein database searching, allowed the identification of the proteins in the protein database. This method is particularly useful for handling small-volume samples such as in single-cell analysis. The high sensitivity and specificity of this method were demonstrated by peptide mapping and identifying hemoglobin variants of sickle cell disease from a single red blood cell. The approach of combining nanoliter chemistry with highly sensitive mass spectrometric analysis should find general use in characterizing proteins from biological systems where only a limited amount of material is available for interrogation.

Multiple parameter cross-species protein identification using MultiIdent--a world-wide web accessible tool

Recent increases in the number of genome sequencing projects means that the amount of protein sequence in databases is increasing at an astonishing pace. In proteome studies, this is facilitating the identification of proteins from molecularly well-defined organisms. However, in studies of proteins from the majority of organisms, proteins must be identified by comparing analytical data to sequences in databases from other species. This process is known as cross-species protein identification. Here we present a new program, MultiIdent, which uses multiple protein parameters such as amino acid composition, peptide masses, sequence tags, estimated protein pI and mass, to achieve cross-species protein identification. The program is structured so that protein amino acid composition, which is highly conserved across species boundaries, first generates a set of candidate proteins. These proteins are then queried with other protein parameters such as sequence tags and peptide masses. A final list of database entries which considers all analytical parameters is presented, ranked by an integrated score. We illustrate the power of the approach with the identification of a set of standard proteins, and the identification of proteins from dog heart separated by two-dimensional gel electrophoresis. The MultiIdent program is available on the world-wide web at: http://www.expasy.ch/sprot/multiident.h tml.

Hydrolysis and amino acid composition of proteins

Amino acid composition analysis is a classical protein analysis method, which finds a wide application in medical and food science research and is indispensable for protein quantification. It is a complex technique, comprising two steps, hydrolysis of the substrate and chromatographic separation and detection of the residues. A properly performed hydrolysis is a prerequisite of a successful analysis. The most significant developments of the technology in the last decade consist in the (i) reduction of the hydrolysis time by the use of microwave radiation energy; (ii) improvement in the sensitivity of the residue detection, the quantification of the sensitive residues and separation of the enantiomeric forms of the amino acids; (iii) application of amino acid analysis in the large-scale protein identification by database search; and (iv) gradual replacement of the original ion exchange residue separation by reversed-phase high-performance liquid chromatography. Amino acid analysis is currently facing an enormous competition in the determination of the identity of proteins and amino acid homologs by the essentially faster mass spectrometry techniques. The amino acid analysis technology needs further simplification and automation of the hydrolysis, chromatography and detection steps to withstand the pressure exerted by the other technologies.

Proteome in perspective

This review describes briefly proteome science. It explains why proteome science or proteomics emerged only recently and why a shift from genomics to proteomics is occurring. This review further illustrates that proteomics can unravel new domains in nature's complexity. Finally, it demonstrates that proteomics is offering new tools for the study of complex biological or medical problems.

Towards an automated approach for protein identification in proteome projects

The development of automated, high throughput technologies for the rapid identification of proteins is essential for large-scale proteome projects. While a degree of automation already exists in some stages of the protein identification process, such as automated acquisition of matrix assisted laser desorption ionisation-time of flight (MALDI-TOF) mass spectra, efficient interfaces between different stages are still lacking. We report the development of a highly automated, integrated system for large scale identification of proteins separated by two-dimensional gel electrophoresis (2-DE), based on peptide mass fingerprinting. A prototype robotic system was used to image and excise 288 protein spots from an amido black stained polyvinylidene difluoride (PVDF) blot. Protein samples were enzymatically digested with a commercial automated liquid handling system. MALDI-TOF mass spectrometry was used to acquire mass spectra automatically, and the data analysed with novel automated peptide mass fingerprinting database interrogation software. Using this highly automated system, we were able to identify 95 proteins on the basis of peptide mass fingerprinting, isoelectric point and molecular weight, in a period of less than ten working days. Advantages, problems, and future developments in robotic excision systems, liquid handling, and automated database interrogation software are discussed.

The Australian Proteome Analysis Facility (APAF): assembling large scale proteomics through integration and automation

The field of proteomics opens new possibilities for the mass screening of proteins from many different sources. While genomics is well understood to be a big science field, proteomics is just emerging as such. This paper describes the setting up of the first national proteomics facility. The facility has been funded by the Australian government and this funding has allowed the design of purpose built, integrated laboratories with state of the art equipment for large scale proteome research.

Proteome and proteomics: new technologies, new concepts, and new words

The goal of proteomics is a comprehensive, quantitative description of protein expression and its changes under the influence of biological perturbations such as disease or drug treatment. Quantitative analysis of protein expression data obtained by high-throughput methods has led us to define the concept of "regulatory homology" and use it to begin to elucidate the basic structure of gene expression control in vivo. Such investigations lay the groundwork for construction of comprehensive databases of mechanisms (cataloguing possible biological outcomes), the next logical step after the soon to be completed cataloguing of genes and gene products. Mechanism databases provide a roadmap towards effective therapeutic intervention that is more direct than that offered by conventional genomics approaches.

An algorithm for the identification of proteins using peptides with ragged N- or C-termini generated by sequential endo- and exopeptidase digestions

We have developed an algorithm (MassDynSearch) for identifying proteins using a combination of peptide masses with small associated sequences (tags). Unlike the approach developed by Matthias Mann, 'Tag searching', in which the sequence tags are generated by gas phase fragmentation of peptides in a mass spectrometer, 'Rag Tag' searching uses peptide tags which are generated enzymatically or chemically. The protein is digested either chemically or with an endopeptidase and the resultant mixture is then subjected to partial exopeptidase degradation. The mixture is analyzed by matrix assisted laser desorption and ionization time of flight mass spectrometry and a list of intact peptide masses is generated, each associated with a set of degradation product masses which serve as unique tags. These 'tagged masses' are used as the input to an algorithm we have written, MassDynSearch, which searches protein and DNA databases for proteins which contain similar tagged motifs. The method is simple, rapid and can be fully automated. The main advantage of this approach is that the specificity of the initial digestion is unimportant since multiple peptides with tags are used to search the database. This is especially useful for proteins like membrane, cytoskeletal, and other proteins where specific endopeptidases are less efficient and lower specificity proteases such as chymotrypsin, pepsin, and elastase must be used.

In vitro model system for the identification and characterization of proteins involved in inflammatory processes

An in vitro model featuring important inflammatory cellular states was established, based on the murine monocyte/macrophage cell line RAW 264.7. Macrophages are key players in chronic inflammation, and major parts of the biochemical reactions taking place in vivo, e.g., the production of proinflammatory cytokines, can be triggered in vitro by stimulation of the cells with bacterial lipopolysaccharide (LPS). A mastergel, representing a synthetic image of the expressed basic set of cellular proteins, was designed by a computer-assisted overlay of a statistically significant number of two-dimensional electrophoresis (2-DE) gels of unstimulated RAW 264.7 cells. This image served as a reference for qualitative and quantitative changes in the protein pattern induced by stimulation of the macrophages with LPS. The optimal conditions for LPS stimulation were evaluated by monitoring the expression and secretion of the proinflammatory cytokine tumor necrosis factor-alpha(TNF-alpha). The comparison of the mastergel with the 2-DE gels of LPS-stimulated cells revealed several changes in the protein pattern. In order to prove the relevance of the presented model system, we focused on two low molecular weight proteins, which showed significant changes in the apparent concentration in a 2-DE pattern. These proteins were further characterized by microsequencing of internal peptides. A comparison of the obtained sequences with protein databases identified them as cofilin and keratinocyte lipid-binding protein.

Two-dimensional gel electrophoresis for proteome projects: the effects of protein hydrophobicity and copy number

Two-dimensional (2-D) gel electrophoresis is often used in proteome projects to provide a global view of the proteins expressed in any cell or tissue type. Here we have investigated the effects of protein hydrophobicity and cellular protein copy number on a protein's presence or absence on a two-dimensional gel. The average hydropathy values of all known proteins from Bacillus subtilis, Escherichia coli and Saccharomyces cerevisiae were calculated, thus defining the range of protein hydrophobicity and hydrophilicity in these organisms. The average hydropathy values were then calculated for a total of 427 proteins from these species, which had been identified elsewhere on 2-D gels. Strikingly, it was seen that no highly hydrophobic proteins, as defined by average hydrophobicity values, have been found to date on 2-D gel separations of whole cell lysates. A clear hydrophobicity cutoff point was seen, above which current 2-D electrophoresis methods appear not to be useful for protein separation. The effect of cellular protein copy number on a protein's presence on a 2-D gel was investigated by means of a graphical model. This model showed how variations in protein loading and copy number per cell interact to determine the quantity of a protein that will be present on a 2-D gel. Considering the current maximum in 2-D gel loading capacity, it was found that 2-D probably can not visualize or produce analytical quantities of proteins present at less than 1000 copies per cell. We conclude that further developments of 2-D electrophoresis techniques are desirable to enable the visualization and analysis of all proteins expressed by a cell or tissue.

Protein phosphorylation: technologies for the identification of phosphoamino acids

Protein phosphorylation plays a central role in many biological and biomedical phenomena. In this review, while a brief overview of the occurrence and function of protein phosphorylation is given, the primary focus is on studies related to the detection and analysis of phosphorylation both in vivo and in vitro. We focus on phosphorylation of serine, threonine and tyrosine, the most commonly phosphorylated amino acids in eukaryotes. Technologies such as radiolabelling, antibody recognition, chromatographic methods (HPLC, TLC), electrophoresis, Edman sequencing and mass spectrometry are reviewed. We consider the speed, simplicity and sensitivity of tools for detection and identification of protein phosphorylation, as well as quantitation and site characterisation. The limitations of currently available methods are summarised.

Protein identification with N and C-terminal sequence tags in proteome projects

Genome sequences are available for increasing numbers of organisms. The proteomes (protein complement expressed by the genome) of many such organisms are being studied with two-dimensional (2D) gel electrophoresis. Here we have investigated the application of short N-terminal and C-terminal sequence tags to the identification of proteins separated on 2D gels. The theoretical N and C termini of 15, 519 proteins, representing all SWISS-PROT entries for the organisms Mycoplasma genitalium, Bacillus subtilis, Escherichia coli, Saccharomyces cerevisiae and human, were analysed. Sequence tags were found to be surprisingly specific, with N-terminal tags of four amino acid residues found to be unique for between 43% and 83% of proteins, and C-terminal tags of four amino acid residues unique for between 74% and 97% of proteins, depending on the species studied. Sequence tags of five amino acid residues were found to be even more specific. To utilise this specificity of sequence tags for protein identification, we created a world-wide web-accessible protein identification program, TagIdent (http://www.expasy.ch/www/tools.html), which matches sequence tags of up to six amino acid residues as well as estimated protein pI and mass against proteins in the SWISS-PROT database. We demonstrate the utility of this identification approach with sequence tags generated from 91 different E. coli proteins purified by 2D gel electrophoresis. Fifty-one proteins were unambiguously identified by virtue of their sequence tags and estimated pI and mass, and a further 11 proteins identified when sequence tags were combined with protein amino acid composition data. We conlcude that the TagIdent identification approach is best suited to the identification of proteins from prokaryotes whose complete genome sequences are available. The approach is less well suited to proteins from eukaryotes, as many eukaryotic proteins are not amenable to sequencing via Edman degradation, and tag protein identification cannot be unambiguous unless an organism's complete sequence is available.

Extraction of membrane proteins by differential solubilization for separation using two-dimensional gel electrophoresis

We describe the extraction and enrichment of membrane proteins for separation by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) after differential solubilization of an Escherichia coli cell lysate. In a simple three-step sequential solubilization protocol applicable for whole cell lysates, membrane proteins are partitioned from other cellular proteins by their insolubility in solutions conventionally used for isoelectric focusing (IEF). As the first step, Tris-base was used to solubilize many cytosolic proteins. The resultant pellet was then subjected to conventional solubilizing solutions (urea, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, dithiothreitol, Tris, carrier ampholytes). Following the completion of this step, 89% of the initial E. coli sample mass was solubilized. Finally, the membrane protein rich pellet was partially solubilized using a combination of urea, thiourea, tributyl phosphine and multiple zwitterionic surfactants. Using N-terminal sequence tagging and peptide mass fingerprinting we have identified 11 membrane proteins from this pellet. Two of these outer membrane proteins (Omp), OmpW and OmpX, have previously been known only as an open reading frame in E. coli, while OmpC, OmpT and OmpTOLC have not previously been identified on a 2-D gel. The prefractionation of an entire cell lysate into multiple fractions, based on solubility, results in simplified protein patterns following 2-D PAGE using broad-range pH 3.5-10 immobilized pH gradients (IPGs). Additional advantages of sample prefractionation are that protein identification and gel matching, for database construction, is a more manageable task, the procedure requires no specialized apparatus, and the sequential extraction is conducted in a single centrifuge tube, minimizing protein loss.

Comprehensive two-dimensional high-performance liquid chromatography for the isolation of overexpressed proteins and proteome mapping

A two-dimensional liquid chromatographic system is described here which uses size-exclusion liquid chromatography (SEC) followed by reversed-phase liquid chromatography (RPLC) to separate the mixture of proteins resulting from the lysis of Escherichia coli cells and to isolate the proteins that they produce. The size-exclusion chromatography can be conducted under either denaturing or nondenaturing conditions. Peaks eluting from the first dimension are automatically subjected to reversed-phase chromatography to separate similarly sized proteins on the basis of their various hydrophobicities. The RPLC also serves to desalt the analytes so that they can be detected in the deep ultraviolet region at 215 nm regardless of the SEC mobile phase used. The two-dimensional (2D) chromatograms produced in this manner then strongly resemble the format of stained 2D gels, in that spots are displayed on a X-Y axis and intensity represents quantity of analyte. Following chromatographic separation, the analytes are deposited into six 96-well (576 total) polypropylene microtiter plates via a fraction collector. Interesting fractions are analyzed by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF/MS) or electrospray mass spectrometry (ESI/MS) depending on sample concentration, which both yield accurate (2 to 0.02%) molecular weight information on intact proteins without any additional sample preparation, electroblotting, destaining, etc. The remaining 97% of a fraction can then be used for other analyses, such Edman sequencing, amino acid analysis, or proteolytic digestion and sequencing by tandem mass spectrometry. This 2D HPLC protein purification and identification system was used to isolate the src homology (SH2) domain of the nonreceptor tyrosine kinase pp60c-src and beta-lactamase, both inserted into E. coli, as well as a number of native proteins comprising a small portion of the E. coli proteome.

An integrated approach to proteome analysis: identification of proteins associated with cardiac hypertrophy

Hypertrophy of cardiac myocytes is a primary response of the heart to overload, and is an independent predictor of heart failure and death. Distinct cellular phenotypes are associated with hypertrophy resulting from different causes. These phenotypes have been described by others at the molecular level by analysis of gene transcription patterns. An alternative approach is the analysis of large-scale protein expression patterns (the proteome) by two-dimensional polyacrylamide gel electrophoresis. Realization of this goal requires the ability to rigorously analyze complex 2D gel images, efficiently digest individual gel isolated proteins (especially those expressed at low levels), and analyze the resulting peptides with high sensitivity for rapid database searches. We have undertaken to improve the technology and experimental approaches to these challenges in order to effectively study a cell culture model for cardiac hypertrophy. The 2D gel patterns for cell lysates from multiple samples of cardiac myocytes with or without phenylephrine-induced hypertrophy were analyzed and spots which changed in abundance with statistical significance were located. Eleven such spots were identified using improved procedures for in-gel digestion of silver-stained proteins and high-sensitivity mass spectrometry. The incorporation of low levels of sodium dodecyl sulfate into the digestion buffer improved peptide recovery. The combination of matrix-assisted laser desorption mass spectrometry for initial measurements and capillary liquid chromatography-ion trap mass spectrometry for peptide sequence determination yielded efficient protein identification. The integration of 2D gel image analysis and routine identification of proteins present in gels at the subpicomole level represents a general model for proteome studies relating genomic sequence with protein expression patterns.

High throughput protein characterization by automated reverse-phase chromatography/electrospray tandem mass spectrometry

We describe an integrated workstation for the automated, high-throughput, and conclusive identification of proteins by reverse-phase chromatography electrospray ionization tandem mass spectrometry. The instrumentation consists of a refrigerated autosampler, a submicrobore reverse-phase liquid chromatograph, and an electrospray triple quadrupole mass spectrometer. For protein identification, enzymatic digests of either homogeneous polypeptides or simple protein mixtures were generated and loaded into the autosampler. Samples were sequentially injected every 32 min. Ions of eluting peptides were automatically selected by the mass spectrometer and subjected to collision-induced dissociation. Following each run, the resulting tandem mass spectra were automatically analyzed by SEQUEST, a program that correlates uninterpreted peptide fragmentation patterns with amino acid sequences contained in databases. Protein identification was established by SEQUEST_SUMMARY a program that combines the SEQUEST scores of peptides originating from the same protein and ranks the cumulative results in a short summary. The workstation's performance was demonstrated by the unattended identification of 90 proteins from the yeast Saccharomyces cerevisiae, which were separated by high-resolution two-dimensional PAGE. The system was found to be very robust and identification was reliably and conclusively established for proteins if quantities exceeding 1-5 pmol were applied to the gel. The level of automation, the throughput, and the reliability of the results suggest that this system will be useful for the many projects that require the characterization of large numbers of proteins.

Functional genomics: going forwards from the databases

Extrapolating systematically from gene sequence to function is undoubtedly the major challenge facing industry and academia alike as we approach the end of the millennium. Many electronic and laboratory approaches are being developed to meet this challenge but the rate of evolution of these is not keeping pace with the speed of sequence generation.

Large-scale identification of proteins of Haemophilus influenzae by amino acid composition analysis

Two-dimensional protein maps of microorganisms are useful tools for elucidation and detection of target proteins, a process essential in the development of new pharmaceutical products. We applied amino acid composition analysis, following separation by two-dimensional gel electrophoresis, for large-scale identification of proteins of Haemophilus influenzae. H. influenzae is a bacterium of pharmaceutical interest of which the entire genome, comprising approximately 1700 open reading frames, has been sequenced. For amino acid analysis, we used both precolumn derivatization of amino acids followed by reversed-phase chromatography of the derivatized residues and post-column derivatization of the residues previously separated on an ion exchanger. The composition analyses derived from both methods allowed the identification of 110 protein spots. The proteins were identified using the AACompldent software on the ExPASy server accessible via the World Wide Web with a success rate of 52%. In some cases, introduction of the analysis data of 12 residues was sufficient for a correct identification. Proteins which contained an unusually high percentage of one residue could be unambiguously identified. Amino acid composition analysis proved to be an error-robust, efficient method for protein identification. The method can be practically established in every biochemical laboratory and, complementary to mass spectrometry, represents an important analytical tool for the mapping of the proteomes of organisms of interest.

Effect of protein application mode and acrylamide concentration on the resolution of protein spots separated by two-dimensional gel electrophoresis

Two-dimensional gel electrophoresis separates large numbers of proteins in two steps on the basis of differences in their pIs and molecular masses. The separation is usually performed on immobilized pH gradient strips, followed by gradient polyacrylamide gels separating proteins with molecular masses between 5-200 kDa. For the first-dimensional separation the protein samples are usually applied near one end of the strip. Using total soluble protein extracts of the bacterium Haemophilus influenzae, we found that simultaneous sample application at both the basic and the acidic ends of the strip resulted in detection of more and stronger protein spots in comparison with sample application at one end only. Because many proteins of an organism have similar pI and Mr values, an overlapping of protein spots is frequently observed in the second-dimensional separation. The soluble protein fraction of H. influenzae was further separated on gels of constant acrylamide concentration between 7.5% and 15.0%. We found that for proteins of molecular mass within certain ranges, the gels of homogeneous acrylamide concentration provided more efficient spot separation than the gradient gels. The observed improvements in spot resolution may be useful in the characterization of proteins from other organisms or cell lines.

Identification of proteins by matrix-assisted laser desorption ionization-mass spectrometry following in-gel digestion in low-salt, nonvolatile buffer and simplified peptide recovery

Matrix-assisted laser desorption ionization-mass spectrometry is an efficient analytical method for large-scale identification of proteins separated by two-dimensional polyacrylamide gel electrophoresis. Following in-gel digestion, the salt present in the peptide extracts is usually removed by chromatography prior to analysis. Desalting is a labor-intensive and time-consuming step, limiting the total number of samples that can be processed daily. We improved the daily sample output by performing the in-gel protein digestion in low-salt, nonvolatile buffer and simplifying the recovery of the generated peptides, collecting them in a small volume by sonication. This technique is routinely used for identification of proteins of Haemophilus influenzae and human brain. The methodology described facilitates the analytical process and allows the analysis of hundreds of proteins per day. Furthermore, it represents an essential step toward process automation.

Proteome research: complementarity and limitations with respect to the RNA and DNA worlds

A methodological overview of proteome analysis is provided along with details of efforts to achieve high-throughput screening (HTS) of protein samples derived from two-dimensional electrophoresis gels. For both previously sequenced organisms and those lacking significant DNA sequence information, mass spectrometry has a key role to play in achieving HTS. Prototype robotics designed to conduct appropriate chemistries and deliver 700-1000 protein (genes) per day to batteries of mass spectrometers or liquid chromatography (LC)-based analyses are well advanced, as are efforts to produce high density gridded arrays containing > 1000 proteins on a single matrix assisted laser desorption ionisation/time-of-flight (MALDI-TOF) sample stage. High sensitivity HTS of proteins is proposed by employing principally mass spectrometry in an hierarchical manner: (i) MALDI-TOF-mass spectrometry (MS) on at least 1000 proteins per day; (ii) electrospray ionisation (ESI)/MS/MS for analysis of peptides with respect to predicted fragmentation patterns or by sequence tagging; and (iii) ESI/MS/MS for peptide sequencing. Genomic sequences when complemented with information derived from hybridisation assays and proteome analysis may herald in a new era of holistic cellular biology. The current preoccupation with the absolute quantity of gene-product (RNA and/or protein) should move backstage with respect to more molecularly relevant parameters, such as: molecular half-life; synthesis rate; functional competence (presence or absence of mutations); reaction kinetics; the influence of individual gene-products on biochemical flux; the influence of the environment, cell-cycle, stress and disease on gene-products; and the collective roles of multigenic and epigenetic phenomena governing cellular processes. Proteome analysis is demonstrated as being capable of proceeding independently of DNA sequence information and aiding in genomic annotation. Its ability to confirm the existence of gene-products predicted from DNA sequence is a major contribution to genomic science. The workings of software engines necessary to achieve large-scale proteome analysis are outlined, along with trends towards miniaturisation, analyte concentration and protein detection independent of staining technologies. A challenge for proteome analysis into the future will be to reduce its dependence on two-dimensional (2-D) gel electrophoresis as the preferred method of separating complex mixtures of cellular proteins. Nonetheless, proteome analysis already represents a means of efficiently complementing differential display, high density expression arrays, expressed sequence tags, direct or subtractive hybridisation, chromosomal linkage studies and nucleic acid sequencing as a problem solving tool in molecular biology.

Purification by reflux electrophoresis of whey proteins and of a recombinant protein expressed in Dictyostelium discoideum

Protein purification that combines the use of molecular mass exclusion membranes with electrophoresis is particularly powerful as it uses properties inherent to both techniques. The use of membranes allows efficient processing and is easily scaled up, while electrophoresis permits high resolution separation under mild conditions. The Gradiflow apparatus combines these two technologies as it uses polyacrylamide membranes to influence electrokinetic separations. The reflux electrophoresis process consists of a series of cycles incorporating a forward phase and a reverse phase. The forward phase involves collection of a target protein that passes through a separation membrane before trailing proteins in the same solution. The forward phase is repeated following clearance of the membrane in the reverse phase by reversing the current. We have devised a strategy to establish optimal reflux separation parameters, where membranes are chosen for a particular operating range and protein transfer is monitored at different pH values. In addition, forward and reverse phase times are determined during this process. Two examples of the reflux method are described. In the first case, we described the purification strategy for proteins from a complex mixture which contains proteins of higher electrophoretic mobility than the target protein. This is a two-step procedure, where first proteins of higher mobility than the target protein are removed from the solution by a series of reflux cycles, so that the target protein remains as the leading fraction. In the second step the target protein is collected, as it has become the leading fraction of the remaining proteins. In the second example we report the development of a reflux strategy which allowed a rapid one-step preparative purification of a recombinant protein, expressed in Dictyostelium discoideum. These strategies demonstrate that the Gradiflow is amenable to a wide range of applications, as the protein of interest is not necessarily required to be the leading fraction in solution.

Two-dimensional map of Haemophilus influenzae following protein enrichment by heparin chromatography

Two-dimensional gel electrophoresis separates several hundred protein molecules in one single experiment and is efficiently used to study the products expressed by different genomes. Low-copy-number gene products are invisible on a stained two-dimensional map and must be enriched such that sufficient amounts are present for visualization and identification. We investigated the enrichment of proteins of the bacterium Haemophilus influenzae by chromatography on immobilized heparin which has affinity for growth and protein biosynthesis factors. Total soluble proteins of the microorganism were fractionated on Heparin-Actigel which resulted in enrichment of approximately 160 proteins. The eluates, representing about 40% of the applied proteins, were analyzed by two-dimensional gel electrophoresis and the protein spots were characterized by amino acid composition analysis and matrix-assisted laser desorption ionization mass spectrometry. The proteins enriched by chromatography on the heparin gel were not exclusively low-copy-number gene products and they did not exclusively belong to one single class of proteins. The proteins that bound to the heparin gel are indicated in a two-dimensional protein map which includes more than 110 newly identified proteins.

A role for Edman degradation in proteome studies

Advances in protein database design and the software used to access the sequence data has led to progress in using protein attributes such as amino acid composition and peptide masses to identify proteins separated by two-dimensional electrophoresis. However, Edman degradation remains the principal technique for protein identification and it presents a significant bottleneck in the progress towards rapid protein identification. Simple modifications to the sequencing hardware, which automate the delivery of protein spots into the sequencer, and parallel sequencing of the protein spots represent a significant advance in the use of Edman degradation to rapidly generate the powerful protein attribute, an N-terminal sequence tag.

Identification of wallaby milk whey proteins separated by two-dimensional electrophoresis, using amino acid analysis and sequence tagging

Micropreparative two-dimensional polyacrylamide gel electrophoresis has been used to separate milk whey proteins from the Tammar wallaby (Macropus eugenii). We have used a combination of amino acid analysis and N-terminal sequence tagging as a rapid and sensitive method to identify the major whey proteins. Using these techniques, we confidently identified alpha-lactalbumin and late lactation protein. While these are the only two M. eugenii whey proteins with a corresponding SWISS-PROT entry, we demonstrate that by using amino acid analysis and matching across species boundaries, we can identify previously unsequenced conserved wallaby whey proteins including beta-lactoglobulin and serum albumin.

Large-scale protein modelling and integration with the SWISS-PROT and SWISS-2DPAGE databases: the example of Escherichia coli

Knowledge-based molecular modelling of proteins has proven useful in many instances, including the rational design of mutagenesis experiments, but it has generally been limited by the availability of expensive computer hardware and software. To overcome these limitations, we developed the SWISS-MODEL server for automated knowledge-based protein modelling. The SWISS-MODEL server uses the Brookhaven Protein Data Bank as a source of structural information and automatically generates protein models for sequences which share significant similarities with at least one protein of known three-dimensional structure. We have now used the software framework of the server to generate large collections of protein models, and established the SWISS-MODEL Repository, a new database for automatically generated and theoretical protein models. This repository is directly integrated with the SWISS-PROT and SWISS-2DPAGE databases through the ExPASy World Wide Web server (URL is http://expasy.hcuge.ch). Here we present an illustration of this process by an application to the Escherichia coli sequences.

Prefractionation of protein samples prior to two-dimensional electrophoresis

Thousands of proteins may be visualised on a two-dimensional (2-D) gel, but only hundreds are present at levels sufficient for chemical analysis. Therefore, prefractionation of protein samples prior to 2-D polyacrylamide gel electrophoresis (PAGE) will be important for the investigation of proteins that are present at sub-picogram levels in physiological samples. We describe an approach to prefractionate protein samples prior to 2-D PAGE using the Gradiflow, which is a new (preparative) electrokinetic membrane apparatus designed to fractionate proteins in a number of different ways. We have fractionated human serum under nonreducing conditions using the 'reflux' mode, in which proteins are fractionated according to their relative mobility under controlled electrophoretic conditions, where the current is periodically reversed. We describe how fractionation occurs and present examples of enrichment of specific proteins.

The Dictyostelium discoideum proteome--the SWISS-2DPAGE database of the multicellular aggregate (slug)

The cellular slime mold Dictyostelium discoideum is a eukaryotic microorganism which has developmental life stages attractive to the cell and molecular biologist. By displaying the two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) protein map of different developmental stages, the key molecules can be identified and characterised, allowing a detailed understanding of the D. discoideum proteome. Here we describe the preparation of reference gel of the D. discoideum multicellular aggregate, the slug. Proteins were separated by 2-D PAGE with immobilised pH gradients (pH 3.5-10) in the first dimension and sodium dodecyl sulfate (SDS)-PAGE in the second dimension. Micropreparative gels were electroblotted onto polyvinylidene difluoride (PVDF) membranes and 150 spots were visualised by amido black staining. Protein spots were excised and 31 were putatively identified by matching their amino acid composition, estimated isoelectric point (pI) and molecular weight (M(r)) against the SWISS-PROT database with the ExPASy AAcompID tool (http:// expasy.hcuge.ch/ch2d/aacompi.html). A total of 25 proteins were identified by matching against database entries for D. discoideum, and another six by cross-species matching against database entries for Saccharomyces cerevisiae proteins. This map will be available in the SWISS-2DPAGE database.

Detailed peptide characterization using PEPTIDEMASS--a World-Wide-Web-accessible tool

In peptide mass fingerprinting, there are frequently peptides whose masses cannot be explained. These are usually attributed to either a missed cleavage site during the chemical or enzymatic cutting process, the lack of reduction and alkylation of a protein, protein modifications like the oxidation of methionine, or the presence of protein post-translational modifications. However, they could equally be due to database errors, unusual splicing events, variants of a protein in a population, or artifactual protein modifications. Unfortunately the verification of each of these possibilities can be tedious and time-consuming. To better utilize annotated protein databases for the understanding of peptide mass fingerprinting data, we have written the program "PEPTIDEMASS". This program generates the theoretical peptide masses of any protein in the SWISS-PROT database, or of any sequence specified by the user. If the sequence is derived from the SWISS-PROT database, the program takes into account any annotations for that protein in order to generate the peptide masses. In this manner, the user can obtain the predicted masses of peptides from proteins which are known to have signal sequences, propeptides, transit peptides, simple post-translational modifications, and disulfide bonds. Users are also warned if any peptide masses are subject to change from protein isoforms, database conflicts, or an mRNA splicing variation. The program is freely accessible to the scientific community via the ExPASy World Wide Web server, at the URL address: http://www.expasy.ch/www/tools.html.

An information-intensive approach to the molecular pharmacology of cancer

Since 1990, the National Cancer Institute (NCI) has screened more than 60,000 compounds against a panel of 60 human cancer cell lines. The 50-percent growth-inhibitory concentration (GI50) for any single cell line is simply an index of cytotoxicity or cytostasis, but the patterns of 60 such GI50 values encode unexpectedly rich, detailed information on mechanisms of drug action and drug resistance. Each compound's pattern is like a fingerprint, essentially unique among the many billions of distinguishable possibilities. These activity patterns are being used in conjunction with molecular structural features of the tested agents to explore the NCI's database of more than 460,000 compounds, and they are providing insight into potential target molecules and modulators of activity in the 60 cell lines. For example, the information is being used to search for candidate anticancer drugs that are not dependent on intact p53 suppressor gene function for their activity. It remains to be seen how effective this information-intensive strategy will be at generating new clinically active agents.

Human 2-D PAGE databases for proteome analysis in health and disease: http://biobase.dk/cgi-bin/celis

Human 2-D PAGE Databases established at the Danish Centre for Human Genome Research are now available on the World Wide Web (http://biobase.dk/cgi-bin/celis). The databanks, which offer a comprehensive approach to the analysis of the human proteome both in health and disease, contain data on known and unknown proteins recorded in various IEF and NEPHGE 2-D PAGE reference maps (non-cultured keratinocytes, non-cultured transitional cell carcinomas, MRC-5 fibroblasts and urine). One can display names and information on specific protein spots by clicking on the image of the gel representing the 2-D gel map in which one is interested. In addition, the database can be searched by protein name, keywords or organelle or cellular component. The entry files contain links to other databases such as Medline, Swiss-Prot, PIR, PDB, CySPID, OMIM, Methabolic pathways, etc. The on-line information is updated regularly.

Characterisation of proteins from two-dimensional electrophoresis gels by matrix-assisted laser desorption mass spectrometry and amino acid compositional analysis

Amino acid compositional analysis and peptide mass fingerprinting by matrix assisted laser desorption mass spectrometry have been used to characterise proteins obtained from two-dimensional electrophoresis (2-DE) separations of human cardiac proteins. A group of twelve protein spots was selected for analysis. The identities of eight of the proteins had been determined by conventional protein characterisation methods, two were unknown proteins and two had putative identities from protein database spot comparison. Amino acid analysis and peptide mass fingerprinting gave corresponding identities for seven of the twelve proteins, which also agreed with our initial identifications. Three proteins which had been identified previously were not confirmed in this study and putative identities were obtained for the two unknown proteins. The advantages, problems and use of amino acid analysis and peptide mass fingerprinting for the analysis of proteins from 2-DE are discussed. The data highlight the need to use orthogonal techniques for the unequivocal identification of proteins from 2-DE gels.