Application of Two-Dimensional Gel Electrophoresis to Microbial Systems

The term proteomics was coined by Marc R. Wilkins in 1994, whilst he was developing the concept as PhD student in Macquarie University (Wilkins et al., 1996a). Proteomics was mainly derived from the field of two-dimensional gel electrophoresis (2-DE). With the arrival of the concept of proteomics or proteome analysis, the method of 2-DE has gained greater significance. Although alternative methods of protein separation for proteomics have been developed, 2-DE has remained the core technology of choice for protein separation.


Introduction
Proteome analysis represents large-scale analysis of the proteins in an organism, simultaneously (Pandey and Mann, 2000), thus facilitating with the elucidation of gene function (Bro and Nielsen, 2004).
The term proteomics was coined by Marc R. Wilkins in 1994, whilst he was developing the concept as PhD student in Macquarie University (Wilkins et al., 1996a). Proteomics was mainly derived from the field of two-dimensional gel electrophoresis ( Two-dimensional gel electrophoresis is derived from 1-D SDS-PAGE, and expands the number of proteins resolved on an electrophoresis gel by separating the proteins based on their native charge and molecular mass. Smithies and Poulik (1956) separated serum proteins using a 2-D combination of paper and starch gel electrophoresis for the first time. The basic method of 2-DE was introduced over 30 years ago (O'Farrell, 1975;Klose, 1975) and represents a technology crucial to the field of proteomics. The coupling of isoelectric focusing (IEF) for the first dimension separation with SDS-PAGE in the second dimension resulted in a 2-D method in which proteins were being distributed across the twodimensional gel profile (Dunn & Görg, 2001). The Proteomic approach was created by combining 2-DE technology with mass spectrometry and bioinformatics. Any improvements, both in mass spectrometers to allow higher imaging quality and more sensitivity and resolution as well as in software for more efficient data mining provides more and detailed information of the proteome. This state-of-the-art technology involves separation, identification and quantification of proteins. It offers many advantages which include the identification of target proteins amongst a pool of gene products in cell and tissue extracts. Having high resolving power, the method provides a greater level of the purified protein for subsequent characterization using mass spectrometry; and the proteins that are separated on the gel are not destroyed and can thus be used for further analysis (Westermeier, 2005).
338 Table 1. The list of protein modifications in the signaling pathway (Seo & Lee, 2004).
These modifications, such as deamidation or oxidation of old cellular proteins may represent aging of the protein (Hipkiss, 2006) or they can occur in an enzymatically regulated fashion after the proteins are translated (Zhang, 2009).
In fact the primary sequence which contains motifs that allow different PTMs (Gupta et al., 2007) are actually found on a protein at any given time in a specific tissue, and cannot be predicted.
Posttranslational modification can bring about change in parameters such as molecular weight and the isoelectric point of proteins (Halligan, 2009). There are in fact more than 360 known chemical modifications of proteins (http://www.abrf. org), which include natural PTMs such as phosphorylation, glycosylation and acylation , as well as artifacts such as oxidation or deamination that might occur naturally inside cells but can also act as artifacts during protein preparation (Benson et al., 2006). It should be noted that shifting of the protein spot (pI or M r ) is usually more likely to be a kind of protein modification.
Oxidation is one of the most common modifications that affect mainly methionine and cysteine residues, but also lysine, histdine, tryptophan and proline, which are oxidized to carbonyl derivatives. Furthermore, the carbonyl content of protein has been used to estimate oxidation of the protein during aging (Stadtman, 1992;Madian & Regnier, 2010). We could end up missing all peptides containing either of the amino acid, because their masses will increase from 2 to 48 Da.
The pulse -chase experiment allows the estimation of the half life of protein degradation. Phosphorylation of proteins can be observed very conveniently by in vivo metabolic labelling using 32 p-or 33 p -ortho phosphate (Cohen, 2000). The sensitivity of metabolic labeling with 35 s amino acids is about three to five times higher than protein detection by the silver staining method (Gygi et al., 2002).
Films or phosphorimagers can be used to detect radioactivity incorporated into the proteins. It is also possible to use two films, and covering the gel with aluminum foil can deliver the 35 S and 32 P signals differentially.
Posttranslational modification enhances the structural diversity and functionality of proteins directly by providing a larger repertoire of chemical properties than is possible using the 20 standard amino acids specific for the genetic code. Therefore, it is regarded as a dynamic phenomenon with a central role in many biological processes.
The applications of two dimensional gels are extensive, particularly in regard to detecting and quantifying modification in genome expression during development under different environmental and stress conditions (Anderson & Anderson, 1998;Wilkins et al., 1996b). If the modified group can be removed by chemical or enzymatic treatment, then 2D gels can be used to identify the position of the modified proteins (Nyman, 2001).
Phosphorylation is the most commonly occurring form of posttranslational modification and the most significant form of regulatory modification in both prokaryotic and eukaryotic cells. The quantitative aspect of this process is important because the phosphoproteome is not only complex but also extremely dynamic (Reinders & Sickmann, 2005). Several reviews have reported a variety of techniques and methodologies for the analysis of phosphoproteins (Thingholm et al., 2009;Paradela & Alber, 2008;Parker et al., 2010).
The primary difference in the detection of the phosphorylated protein is the staining method. This method must be specific for phosphorylated proteins and must not interact with the nonphosphorylated species. Since many phosphorylation-regulated proteins are present in small quantities, the staining or labeling procedures must be sensitive. These criteria are met when using both 32 p labeling and immunostaining procedures.
The radioactive proteins are used during fractionation procedures, such as 2D-PAGE or HPLC, so as to identify the amino acid types that are modified; the phosphoproteins are completely hydrolyzed and the phosphoamino acid content determined. The specific sites of phosphorylation can be determined by proteolytic digestion of the radiolabeled protein, separation and detection of phosphorylated peptides (e.g. by two-dimensional peptide mapping), followed by Edman sequencing. To measure differences in relative abundances of phosphorylation, 32 p-labeled proteomes can be separated by 2D-PAGE and the relative spot intensities are subsequently compared (Cohen, 2000).
One of the most common forms of posttranslational modification that occurs within the cell involves the attachment of carbohydrates to proteins. In fact 2D-GE is the method of choice for visualizing glycoproteins (Schäffer et al., 2001).
Standard techniques for in-gel or on-membrane protein staining appear to work poorly with glycoproteins and special methods are required. The periodic acid-Schiff method, biotinhydrazide, fluorescein, semicarbazide and the glycoprotein specific staining reagent (pro-Q Emearald, which is 50 times more sensitive than other methods) are used for detection and staining of glycoproteins (Steinberg et al., 2001).
For identification of glycosylated proteins, identification of the sites of glycosylation, quantitation of the extent of glycosylation of each site, identification of the number of different glycoforms and structural characterization of the glycolytic side chain are necessary. Detection of glycoproteins can be achieved by 2DGE and polyvinylidene difluoride membranes (Harvey, 2001).
Glycosylation is the most common and complex PTM, (Sharon & Lis, 1997) and has many different biological roles (Varki, 1993;Varki et al., 1999). These roles vary from those which are related primarily to general effects of the size and shape of the glycan, such as protein folding and assembly of protein complexes (Helenius & Aebi, 2001), to those which depend upon the specific configuration of the branched glycan structures, such as cell recognition, cell -cell interaction and immune responses (Rudd et al., 2001).
The main methods that have been used for preparation of glycosylated samples can be divided into two main categories: -sample preparation for detailed characterization of a relatively pure protein. -sample preparation for analysis of low level, complex proteomic samples.
It is possible to prepare proteomic samples for glycosylation-specific analysis by combining 2 D-GE separation with glycoprotein-specific staining, (Hart et al., 2003) but this is subject to the limitation of 2D-GE procedures including limited throughput and difficulties in membrane glycoproteins.The most established affinity techniques (lectin affinity chromatography) can be used to enrich samples for subsequent analysis (Kuster et al., 2001).

Protein -Protein interactions
The majority of proteins of a cell usually function as part of much larger complexes that are often in the static or transitory form, and it is the latter form which takes part in signaling and metabolic pathways.
Proteins within the cells often interact with small molecules, nucleic acids and/or other proteins (Guidi et al., 2010). In order to study interactions between proteins, methods have been devised that are based on the detection of binary interactions which involve interactions between pairs of proteins , and those that are based on the detection of complex interactions that consist of interactions between multiple proteins that form complexes (Phizicky et al., 2003). Such interactions can be confirmed or rejected by the many available genetic and bioinformatics methods. In fact, by applying classical genetics that combines different mutations in small cells or organisms and observing the resulting phenotype, a shortcut to functionally significant interactions that have a recognizable effect on the overall phenotype can be provided (Giot et al., 2003).
Another genetic approach involves a strategy that screens for enhancer mutations. In this situation the individual mutations in the genes that code for proteins X and Y do not prevent interaction and are therefore considered as viable, however the presence of simultaneous mutations in both genes prevents the interaction and leads to what is called a lethal phenotype, thus confirming that these proteins are part of the same complex in the same biochemical or signaling pathway (Tong et al., 2001).
In the bioinformatics approach, three methods have been developed for the confirmation of protein interaction directly from genomic data (Halligan et al., 2004). One of the methods is referred to as the domain fusion or Rosetta stone method (Marcotte 2000). It is based on the principle that protein domains are structurally and functionally independent units that can operate either as discrete polypeptides or as part of the same polypeptide chain. Hence, multidomain proteins in one species may be represented by two or more interacting subunits.
Another of the three methods is the comparative genomic method, which is based on the fact that bacterial genes are usually arranged into operons and although their sequences are diverse, such genes are often functionally related. Thus, if two genes are located next to each other in a series of bacterial genomes, it may be that they are functionally related even if their products are found to interact (Marcotte, 2000). However, there is also evidence that genes whose functions are apparently unrelated may be organized into operons (Marcotte, 2000).
The third bioinformatics-based method is associated with phylogentic profiling, which makes use of the evolutionary conservation of genes with the same function. For example the presence of three or four conserved genes of unknown function in 20 aerobic bacteria and their absence in 20 anaerobic ones indicate that the products of such genes are highly likely to be involved in aerobic metabolism (Marcotte, 2000).
In addition to the above approaches, biochemical and physical methods are required to verify the genetic and bioinformatic methods. However biochemical and physical procedures are often faced with limitations with regard to the number of interactions thus making such approaches unsuitable for the purpose of analyzing protein interactions on a global scale.
In contrast to the above approaches, the gel-based proteomics approach, otherwise known as two-dimensional electrophoresis (2-DE), has been found to be especially useful in the study of protein-protein interactions, because it allows for an improved separation of proteins as well as the detection of specific interacting protein isoforms of a protein that arise from posttranslational modification (Mann & Jensen, 2003). In fact the detection of protein interactions by the proteomics methodology has been considered as one of the most challenging and rewarding approaches. The objective of single studies is often to identify all interacting partners of a single protein, but when several studies are taken together, they can be used to identify all interactions within a single signaling module .
Interactions of a proteomic nature have been analyzed only in some exceptional studies (Ho et al., 2002;Krogen et al., 2006). The results obtained from such studies are often incomplete because of the temporal and unstable nature of protein-protein interactions, and the acquisition of different results from different methods and their complexities. Proteomic parameters can change, from seconds or minutes (e.g. signaling) to hours, days and ever longer time periods, such as in degenerative diseases.
It is important to analyze all these interactions on a proteomic scale. Several proteomic studies have in fact significantly enhanced our knowledge regarding the interacting partners and functions of single proteins, or whole protein complexes. Nevertheless, based on the methods used, it may be difficult to understand whether, for example a protein shows a weak but specific interaction or a strong but nonspecific interaction (Jensen, 2004).
Hence, comparing and combining data from different studies must be carried out with a great deal of care and precision, because of the application of different technologies during such studies.

Metabolic engineering
It is obvious that all proteins of a living cell are not expressed simultaneously. This makes it possible to compare the proteome maps under different conditions. The up-regulated and down-regulated proteins which are stimulated by withdrawal of nutrients, or any other physical and chemical changes, can be determined by 2-DE. In this way proteins of high significance are found. Proteomics has become an important tool in understanding gene function with regard to many metabolic engineering strategies. In this section, the potential use of proteomic analysis in metabolic engineering will be illustrated.
Metabolic engineering involves the optimization of genetic and regulatory processes within cells to increase production of certain substances of human interest by the cells. It involves the alteration of the cells genetic makeup, in order to obtain a specific phenotype (Vemuri et al., 2005). This strategy often tries to minimize cellular energy associated with cell reproduction and proliferation and also attempts to reduce the production of cellular wastes.
One of the main features of metabolic engineering involves metabolic pathway manipulation, which has been classified by Cameron & Tong (1993) into five groups. These are: Improvement of the yield and productivity of products made by microorganisms; expanding the spectrum of substrates that can be metabolized by an organism; forming new and unique products; improving cellular properties, and degradation of xenobiotics.
The genetic and regulatory changes can have significant effects on the complex cellular machinery and hence the cells' ability to survive. In metabolic engineering strategies, a great deal of focus is currently being placed on regulatory networks in the cell to engineer an efficient metabolism besides the direct deletion of or overexpression of the genes that code for metabolic enzymes (Vemuri et al., 2005). Because of the significance of regulatory control in metabolic processes several investigations are currently being carried out to understand regulation at various levels of the metabolic hierarchy (both local and global). The availability of biological data has helped with the identification of the individual components (genes, proteins, and metabolites) of a biological system, thus making it easier to unravel interactions between such components that culminate in what is known as the phenotype. Hence, in addition to an integrated understanding of physiology, but also for practical applications of using biological systems as cell factories, it is very important to reveal and identify such components. Recent "-omics (genome, trancriptome, interactome, proteome, metabolome, fluxome)" approaches have extended knowledge regarding regulation at the gene, protein, and metabolite levels, and thus have had a great influence on the progress associated with metabolic engineering. A combination of global information derived from various levels of metabolic hierarchy is absolutely essential in comprehending and assessing the relationship between changes in gene expression and the resulting phenotype (Vemuri et al., 2005).
One of the "omics" that has allowed a better comprehension of regulation is proteomics. The awareness of protein abundance helps with understanding the extent to which regulatory proteins and transcription binding factors take part in the subsequent change that occurs in the gene expression profile. The initiation of translation and the following regulation is mainly dependant on the ribosome-binding site. After the detection of a signal, the regulatory proteins bind to the promoters and recruit RNA polymerase enzymes to the transcription start site (Vemuri et al., 2005). Two-dimensional (2D) gel electrophoresis has so far been the method of choice in analyzing proteomes, with a good turnover of information. However, one of the drawbacks of this method is that it mostly detects proteins expressed at high concentrations. In order to counteract this problem, staining methods have been developed to improve the accuracy and the sensitivity of protein detection and quantification. Nevertheless, several regulatory proteins are present in the cell at very low concentrations, hence, alongside 2D-gel electrophoresis, sensitive high-throughput methods for accurate protein detection and quantification is widely acknowledged (Vemuri et al., 2005) and is currently being implemented (Bernhardt et al., 2003).
A wide array of microbial systems has been and is currently being used for production of many products that are of biotechnological and metabolic significance. Here presented are a few examples.

Lactic acid bacteria
There have been only a few studies of the regulation of carbon flow in Lactic acid bacteria (LAB) with regard to the "omic" technologies. In fact little is known regarding the regulation of glycolytic enzymes in the emerging pathogen Enterococcus faecalis. In a study by Mehmeti and colleagues (2011) regulation of central carbon metabolism at the level of biosynthesis of the participating proteins was demonstrated in E. faecalis V583 . Such research is of vital interest because of the pathogenic aspects of E. faecalis, in addition to significance of lactic acid production for LAB, and the fact that LAB are widely used for production of lactic acid in fermented food. The data derived from such research will inevitably help with metabolic engineering strategies associated with product development in LAB. During homolactic fermentation in LAB, pyruvate is converted to lactate in addition to a number of minor metabolites, such as acetic acid, acetaldehyde, ethanol, acetoin, and acetate, but, under certain environmental conditions, metabolism switches to heterolactic fermentation (mixedacid producing formate, acetate, acetoin, ethanol, and CO 2 as the final products (Mehmeti et al., 2011). An example of such a situation was observed in Lactococcus lactis, where mixedacid fermentation has been shown to occur at low growth rates under microaerobic conditions (Jensen et al., 2001;Mehmeti et al., 2011), under true carbon-limited conditions, and while growing at low pH on carbon sources other than glucose (Melchiorsen et al., 2002;Mehmeti et al., 2011)). Mixed-acid fermentation has also been observed in E. faecalis V583, but only after removal of the lactate dehydrogenase (LDH) activity (Jonsson, et al., 2009)). This emerging pathogen has two ldh genes, but it is the ldh-1 gene which is the main contributor to lactate production (Mehmeti et al., 2011).
The "omics" approach that involves metabolic, transcriptomic and proteomic technologies will no doubt assist in elucidating the shift from homolactic to heterolactic fermentation in LAB. This has already been demonstrated in L. lactis, where enzyme levels have been found to be regulated in response to growth conditions, and correlations between metabolic and transcriptomic or proteomic data have been established (Dressaire et al., 2009;Mehmeti et al., 2011). In this latest research, Mehmeti and colleagues (2011) compared a constructed lactate dehydrogenase (LDH)-negative mutant of E. faecalis V583 (ldh1.2) with its wild type via metabolic, transcriptomic, and proteomic analyses. The mutant was found to grow at the same rate as the wild type but converted glucose to ethanol, formate, and acetoin. Microarray analysis was also performed showing that LDH deficiency had overwhelming transcriptional effects, where 43 genes in the mutant were found to be upregulated, and 45 were found to be downregulated. Most of the upregulated genes were identified as those involved in energy metabolism and transport. Subsequently, 45 differentially expressed proteins were identified using the technique of two-dimensional (2D) gel analysis in conjuction with MALDI-TOF/MS analysis. A comparison of transcriptomic and proteomic data suggested that for several proteins the level of expression is regulated beyond the level of transcription. Comparative proteomic and transcriptomic and metabolic analyses showed that the pyruvate catabolic genes, including the truncated ldh gene, showed highly increased transcription in the mutant., and that protein expression was regulated beyond the level of transcription, as suggested by Mehmeti and colleagues (2011), who also demonstrated that these genes, along with a number of other differentially expressed genes, are preceded by sequences with homology to binding sites for the global redox-sensing repressor, Rex, of Staphylococcus aureus. The results of this research demonstrates that the NADH/NAD ratio is involved in the transcriptional regulation of these genes, thus having a critical role in the regulatory network controlling energy metabolism in E. faecalis (Mehmeti et al., 2011).

Vitreoscilla stercoraria
The obligate aerobic bacterium, Vitreoscilla stercoraria, produces the oxygen binding protein, Vitreoscilla hemoglobin (VHb), the expression of which under hypoxic conditions in various organisms, such as bacteria, yeasts, fungi, and plant cells, , improves growth, increases protein secretion and metabolic productivity and stress resistance, mediates ATP synthesis and detoxifies the adverse effects of nitric oxide. However, the mechanism of VHb action has still not been clearly understood. With such positive effects, VHb is currently being applied to various cell-based biotechnological processes including metabolic engineering, production of valuable metabolites, and fermentation (Isarankura-Na- Ayudhya et al., 2008).
In the study by Isarankura-Na-Ayudhya and colleagues (2008), the effect of VHb production on the protein expression profile of E. coli was investigated using 2-DE and peptide mass fingerprinting. Following fusion of VHb with green fluorescent protein (GFP), GFPuv was selected as a reporter molecule to demonstrate vgb gene expression, which is under the control of lac promoter. In fact, the location of the fusion protein (VHbGFP) on the 2D gels can easily be distinguished from other high abundant proteins. The resulting protein spots were then identified by MALDI-TOF mass spectrometry.
Data showed that in addition to the loss of tryptophanase that is responsible for tryptophan, cysteine and serine catabolism, VHbGFPuv expression also down-regulated proteins involved in various metabolic pathways, such as glycerol kinase, isocitrate dehydrogenase, aldehyde dehydrogenase, and D-glucose-D-galactose binding protein. The data obtained from this proteomic approach point to the critical roles that VHb plays at the level of cellular carbon and nitrogen consumptions. It may also be involved in the regulation of other metabolic pathway intermediates, via autoregulation of the catabolite repressor regulons ( Isarankura-Na-Ayudhya et al., 2008) .

Amycolatopsis balhimycina
The identification and unraveling of biochemical pathways and enzymes by the proteomics approach has helped greatly with targeting proteins of interest for eventual genetic manipulations, so as to optimize the reactions that culminate in the production of beneficial products in various microbial systems. A recent example of this technology has involved the use of the differential proteomic approach which has shown that the production of the antibiotic balhimycin in batch culture is associated with the up-regulation of enzymes involved in the biosynthesis of antibiotic precursors. These proteomic data point to the possible sustainable balhimycin production through availability of increased levels of precursors, such as tyrosine, that has also proved to be effective in balhimycin production (Gallo et al., 2010).
Chemostat cultures are one of the techniques that help with understanding the relationship between the physiology of a microorganism and its metabolism. In a study by Gallo and colleagues (2010), chemostat cultures of Amycolatopsis balhimycina were used to reach steadystate conditions for the purpose of biomass accumulation under the same growth rate conditions, with or without balhimycin production. The minimal defined media consisted of low Pi concentrations and proficient glucose, or high Pi levels and limiting glucose concentrations. The biomass obtained from these cultivations were then analyzed by a comparative proteomic study in order to reveal the expression of genes that are involved in A. balhimycina primary and secondary metabolism which are associated with biomass production and antibiotic synthesis (Gallo et al., 2010). The medium containing low Pi levels produced balhimycin. Accordingly, quantitative RT-PCR revealed up-regulation of the bal genes, responsible for balhimycin biosynthesis, and of phoP, phoR, pstS and phoD, which are involved in the Pi limitation stress response (Gallo et al., 2010).
The analysis of the biomass by 2-D Differential Gel Electrophoresis (DIGE) and the protein identification carried out by mass spectrometry and computer-assisted 2-D reference-map (http://www.unipa.it/ampuglia/Abal-proteome-maps), showed a differential expression for proteins involved in many metabolic pathways such as the central carbon and phosphate metabolism. Furthermore, the DIGE technique demonstrated that in the presence of low Pi levels, proteins involved in the production of primary metabolic intermediates and cofactors required for balhimycin biosynthesis were up-regulated. The bioinformatic approach was subsequently used to show the presence of PHO box-like regulatory elements in the upstream regions of nine differentially expressed genes (Gallo et al., 2010). The proteomic approach illustrated a relationship between primary metabolism and antibiotic production, which can be used in metabolic engineering strategies to increase antibiotic yield (Gallo et al., 2010).

Escherichia coli
According to Renzone and colleagues (2005), identification of regulators and regulatory networks is essential to control, predict or engineer bacterial behavior. In fact, the identification of crucial stress-related genes in particular will uncover targets for a) specific manipulation to promote or limit cellular growth, b) development of useful tools to screen for tolerant or sensitive strains and c) evaluation of the culture fitness (an indication whether bacteria are fully adapted or stressed), which can help to optimize parameters for growth in culture (Renzone et al., 2005).
In a study by Han et al. (2001), recombinant E. coli strains harboring heterologous polyhydroxyalkanoate (PHA) biosynthesic genes that were shown to accumulate unusually large amounts of PHA were analyzed. In this study, integrated cellular responses of metabolically engineered E. coli to the accumulation of poly (3-hydroxybutyrate) (PHB) in the early stationary phase were analyzed at the protein level by two-dimensional gel electrophoresis. In the presence of accumulated levels of PHB, 20 proteins showed altered expression levels, 13 of which were identified using the mass spectrometry technique. Three heat shock proteins, GroEL, GroES, and DnaK, were significantly up-regulated in PHBaccumulating cells, but, proteins involved in the protein biosynthetic pathways were adversely affected. The proteomics approach illustrated that the increased synthesis of two glycolytic enzymes and one enzyme belonging to the Entner-Doudoroff pathway was likely to have arisen from the cellular demand for large amounts of acetyl coenzyme A and NADPH that are required for PHB biosynthesis. Furthermore, in the presence of accumulated levels of PHB, the expression of the yfiD gene was highly induced; this gene codes for a 14.3-kDa protein that is usually produced at low pH levels. This analysis suggests that the accumulation of PHB in E. coli represents a situation of stress leading to a reduction in the cells' ability to synthesize proteins and induction of the expression of various protective proteins (Han et al., 2001(Han et al., , 2011.

Xanthophyllomyces dendrorhous
The bacidiomycete, Xanthophyllomyces dendrorhous, is one of the best sources of the antioxidant carotenoid astaxanthin, and is used for the microbiological production of this antioxidant. Being of biotechnological significance, many investigations have recently been carried out, including metabolic engineering strategies to improve astaxanthin yield. The first proteomic analysis of the strain ATCC 24230 has been carried out recently by Matinez-Moya and colleagues (2011). The technique of 2-DE in conjunction with matrix-assistedlaser-desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) were used to produce protein profiles before and during the induction of carotenogenesis . Approximately 600 protein spots were observed, 131 of which were found to be nonredundant proteins. In addition, 50 differentially expressed proteins that were classified as a result of distinct expression patterns were also identified. Proteomic analyses revealed that enzymes involved in acetyl-CoA synthesis were present at higher concentrations prior to the process of carotenogenesis (Martinez-Moya et al., 2011). However, redox-and stress-related proteins were up-regulated during the induction of carotenogenesis. Higher levels of the carotenoid biosynthetic enzymes, mevalonate kinase and phytoene/squalene synthase, were observed during induction and accumulation of carotenoids. Nonetheless, the usual antioxidant enzymes, such as catalase, glutathione peroxidase and the cytosolic superoxide dismutases, were not detected. The gel-based proteomic approach succeeded in identifying the potential carotenogenesis-related proteins, which also included those involved in carbohydrate and lipid biosynthetic pathways, and the several redox-and stress related 347 proteins. Martinez-Moya and colleagues (2011) suggest that these proteomic-derived data also indicate that X. dendrorhous accumulates astaxanthin under aerobic conditions, so as to scavenge the reactive oxygen species (ROS) generated during metabolism (Martinez-Moya et al., 2011). Nevertheless, for the sake of both basic research and metabolic engineering, detailed transcriptomic, proteomic and metabolomic studies are required to produce an integrated understanding of the biochemical, physiological and biological processes of X. dendrorhous, (Martinez-Moya et al., 2011).

Characterization of mutant proteins
The elucidation of many regulatory pathways in various microorganisms has involved the creation of mutations in genes that code for enzymes involved in the regulatory pathways and networks. In this regard the presence of mutant proteins has been investigated using several techniques. However, one technology which allows us to identify mutant proteins in the cell proteome and understand the effects of such mutants on other proteins, with insight into the interconnections between the regulatory networks in the proteome under different environmental conditions, is proteomic analysis involving the use of 2-DE. This section will summarize the applications of 2-DE in this regard.
The characterization of mutant proteins in response to elucidation of regulatory proteins and pathways has been greatly aided by the use of 2D-gel electrophoresis. The application of this technique to this area of research in microbial physiology has been extensive and has managed to unravel many regulatory components associated with bacterial metabolism (van Vliet et al., 1998;Coppee et al., 2001;Cheng et al., 2009;Kint et al., 2009;Egan et al., 2002;Panmanee et al., 2008;Cox et al., 2009;Shin et al., 2007;Friedman et al., 2006). Some applications of this technique are described below.

Bacillus subtilis
Recent advances in the application of 2D-gel electrophoresis have been applied to the metabolic system of the Gram positive bacterium, B. subtilis (Bernhardt et al., 2003;Hecker et al., 2009).
The introduction of the highly sensitive technique of 2D-gel electrophoresis into bacterial physiology by Neidhardt and van Bogelen more than 20 years ago, otherwise known as "proteomics" in the mid-90s (Wilkins et al., 1996b), opened a new era in this field of research (Bernhardt et al., 2003;Hecker et al., 2009).
The complete genome sequence and protein annotation of B. subtilis have been made available, which has provided advantages in many proteomic studies. (Kunst et al., 1997;Seul et al., 2011). Extensive information can simply be derived from the genome sequence. However, it is the proteome that signifies the phenotype, representing the cell at the molecular level (Bernhardt et al., 2003). Up to 10,000 proteins (Klose & Kobalz, 1995;Bernhardt et al., 2011) can be separated on one single 2D-gel, with the majority of all proteins synthesized in bacterial cells being visualized simultaneously (including alkaline and extracellular cells). Identification of proteins by mass spectrometry (MALDI TOF MS, ESI MS, etc.) supported by N-terminal sequencing will no longer be problematic once the genome sequence of the organism is known (Bernhardt et al., 2003).
According to Bernhardt et al., 2003, physiological proteomics provides information at the protein level regarding the physiological state of a cell , however, the next step in proteomics involves analyzing the kinetics of the protein pattern under different environmental conditions that represent nature. The growth of B. subtilis in the upper layers of soil, involves short periods of growth followed by long non-growth periods brought about by stress and starvation. To survive such situations, B. subtilis has developed cellular adaptation strategies derived from complex genetic and regulatory networks (Bernhardt et al., 2003). Such networks are comprised of many regulons, that have a unique adaptive function and are controlled by one global regulator. In order to comprehend bacterial physiology, the analysis of this network, its dissection into single regulons, and a definition of the adaptive function of all proteins within the regulons is essential (Msadek, 1999;Bernhardt et al., 2003). Usually more than one regulon is induced by extracellular stimuli; this group when expressed in the protein form is referred to as the stimulon and contributes to stress adaptaion (VanBogelen & Neidhardt, 1990). In order to elucidate adaptaional networks, stimulons and their respective regulons must be defined and distinguished. This is where the role of identifying and characterizing mutant proteins for comparative purposes takes on a significant role. Proteins/ genes belonging to regulons can be identified if mutants in global regulators are available (Bernhardt et al., 2003). The structure of presently known as well as still unknown regulons can be analyzed by comparing wild-type protein expression patterns to deregulated mutant strains carrying inactivated global regulatory genes as demonstrated by the σ B -dependent general stress regulon of B. subtilis as a model (Hecker & Völker, 1998;Hecker et al., 2009).
In order to understand global gene regulation, the study of single regulons derived from the genome is not appropriate because single regulons are part of a tight adaptational network. However, in order to understand and unravel global gene regulation, the use of a comprehensive computer-aided inspection and matching of various 2D gels loaded with radioactively labeled proteins becomes the method of choice (Antelmann et al., 2000;Bernhardt et al., 2003). Although DNA array techniques can be comprehensive and generate a huge quantity of data, but they are often very difficult to interpret. However, a fast gelbased proteomics method that can assist with this kind of analysis, involves the use of the dual channel imaging of 2D protein gels (Bernhardt et al., 1999;Hecker et al., 2009), which has made the search for proteins belonging to stimulons or regulons more practical. In this technique, proteins are assigned to stimulons or regulons simply by detecting red (newly induced) or green (repressed) proteins. In fact, the green proteins which comprise a substantial part of the protein pool in the non-growing cell, cannot be detected when using DNA arrays. According to Bernhardt and colleagues (2003), the dual channel imaging technique combines staining techniques to visualize accumulated proteins and autoradiography to uncover proteins that are synthesized at defined states. Two digitized images of 2D gels have to be generated and combined in using alternate additive color channels. One of them-the densitogram-showing proteins accumulated in the cell that had been visualized by (silver) staining techniques is pseudocolored green. The second image of an autoradiograph showing the proteins synthesized during a 5-min L-[ 35 S]methionine pulse label is pseudocolored red. As a result, accumulated as well as newly synthesized vegetative proteins take on a yellow color. After imposition of a glucosestarvation stimulus, however, proteins newly synthesized in response to that stimulus are red because they have not yet accumulated in the cell. Proteins, the synthesis of which, have been switched off by the stimulus, change their color from yellow to green (Bernhardt et al., 2003). The major advantage of this technique is that it allows for the matching of protein patterns on a single gel as opposed to many different gels, which is one of the bottlenecks in data analysis (Bernhardt et al., 1999;Hecker et al., 2009)). Hence, dual channel imaging helps with providing a large quantity of information regarding the relative amount and rate of synthesis of each single protein (Bernhardt et al., 2003). Already identified spots synthesized at a higher rate in the wild type are indicated by text labels, and nonidentified spots by white circles (taken from the article by Bernhardt et al., 2003). For a list of protein spots that are induced in a σ B -dependent manner refer to Bernhardt et al. (2003).
2D protein gel electrophoresis when used in conjunction with MALDI TOF mass spectrometry and dual channel imaging technique can provide a comprehensive outline of the physiology of a bacterial cell population entering the stationary growth phase, as has been demonstrated by Bernhardt and colleagues (2003) to evaluate and visualize the overall regulation of protein synthesis during glucose starvation. A sequential series of overlays obtained at specific points of the growth curve helps with visualization of the developmental processes at a proteomics scale. A substantial reprogramming of the protein synthesis pattern was found during glucose starvation, where 150 proteins were synthesized de novo and the synthesis of almost 400 proteins was terminated. These 150 proteins were identified as belonging to different general and specific regulation groups, which were demonstrated by the analysis of mutants in genes encoding global regulators. Following glucose starvation, a series of proteins are induced that belong to two main regulatory groups: general stress/starvation responses induced by different stresses or starvation stimuli (σ B -dependent general stress regulon, stringent response, sporulation), and glucosestarvation-specific responses (decline of glycolysis, utilization of alternative carbon sources, and gluconeogenesis). Hence, in addition to being able to identify stimulons and regulons by the dual channel approach, the fate of each single protein could also be followed by the three-color code system in this technique (Fig. 2) (Bernhardt et al., 2003;Hecker et al., 2009).
Mutant protein characterization using the gel-based proteomics approach was also recently used to understand the global regulation by the bacterial PyrR regulatory protein (Seul et al., 2011). The enzymes involved in the denovo biosynthesis of uridine monophosphate (UMP) and cistrones encoding a uracil permease , together with the regulatory protein PyrR are encoded by the B. subtilis pyrimidine biosynthetic (pyr) operon. The PyrR is a bifunctional protein that possesses both pyr mRNA-binding regulatory function and uracil phosphoribosyltransferase activity (Seul et al., 2011).
In a study by Seul and colleagues (2011), proteomic analysis of PyrR-associated protein expression in B. subtilis cells was carried out, where the cellular proteome of B. subtilis DB104 was compared with that of a pyrR knockout mutant (PyrR Mut) using a gel-based differential proteomic strategy. In order to have an insight into global regulation by the pyrR deletion, the cellular proteomes of both B. subtilis DB104 and B. subtilis DB104 ΔpyrR were compared following growth in minimal medium without pyrimidines. The proteins were analyzed by MALDI-TOF mass spectrometry and matched with database search. Variations in levels of expression of the cytosolic proteins in both strains were demonstrated by 2D-gel electrophoresis. Proteomic analysis revealed, approximately 1,300 spots, 172 of which showed quantitative variations; amongst the latter 42 high quantitatively variant proteins were identified on the silver stained 2D-gel with an isoelectric point (pI) between 4 and 10. These results showed that production of the pyrimidine biosynthetic enzymes (PyrAA, PyrAB, PyrB, PyrC, PyrD, and PyrF) were significantly increased in B. subtilis DB104 ΔpyrR (Seul et al., 2011). Proteins associated with carbohydrate metabolism, elongation protein synthesis, metabolism of cofactors and vitamins, motility, tRNA synthetase, catalase, ATPbinding protein, and cell division protein FtsZ were also overproduced in the PyrRdeficient mutant. Analytical data suggests in addition to pyrimidine biosynthesis, the PyrR protein might be involved in a number of other metabolisms in the bacterial cell (Seul et al., 2011).
In conclusion, this gel-based differential proteomics study enabled to elucidate an example of global regulation in the model bacterial strain B. subtilis affected by a gene deletion.

Microbial systematics and epidemiology
Classification and identification of microorganisms are traditionally carried out on the basis of cell morphology, cellular metabolism, cell structure and cell components such as fatty acids, carbohydrates and quinines. The uncertainty of conventional methods to identify the microbial strains has arisen from the lack of distinctive structures in most strains and the lateral gene transfer between some species. Modern microbial classification puts emphasis on molecular systematics, using genetic techniques like guanine-cytosine ratio determination, DNA hybridization, and sequencing of conserved genes such as the ribosomal RNA gene. Molecular phylogenetics uses DNA sequences as information to build a relationship tree showing the possible evolution of various organisms. Application of molecular phylogenetics for organism's evolutionary relationship was pioneered in the 1970s by Charles G. Sibley. Recently new molecular technologies based on genomics and proteomics methods have been developed to identify and characterize bacteria. Although these new molecular technologies can be high throughput and rapid, they are counted as complement techniques for traditional methods (Emerson et al., 2008).
Molecular techniques capable of typing bacteria on the basis of their proteomes are used for pathogenesis and epidemiological investigations (Cash, 2009). In 1990, Costas presented a new approach to bacterial typing by 1-D SDS PAGE. Costas and colleagues paid particular attention to analyze a whole cell protein using a high resolution SDS PAGE. Besides, they considered a detailed analysis of gel data by using high-resolution densitometers and software. Three groups of bacterial isolates belonging to Acromobacter, Neisseriaceae and Providencia were analyzed by this method, and subsequently the strategy of 1-D protein electrophoresis was widely used for typing of a number of bacterial isolates such as Streptococcus porcinus, Bacillus thuringiensis, Helicobacter pylori, etc (Duarte et al., 2005;Konecka et al 2007;Costas et al., 1990). The Haemophilus influenzae proteome was analyzed by 2-DE and extended to three other members of the Haemophilus genus in 1997. Data analysis was carried out by the Nonlinear Dynamics Phoretix 2D (version 3.1) analytical software (Cash et al., 1997). The E. coli isolates collected from clinical specimens were also discriminated by 2-DE. This technique provides a high resolution large-scale screening for differentiation of bacterial isolates. Despite the advantages of 2-DE for taxonomic and epidemiological studies, 1-D SDS PAGE is mostly preferred because it is a simple and straightforward method (Cash, 2009).
Differential proteomics is very helpful for rapid detection of bacteria. Surface enhanced laser desorption ionization (SELDI)-time of flight which is a rapid MALDI MS based technology compatible with the ProteinChip Array has been developed. This is a valuable technique to determine the microbial phylogeny and discriminate different bacterial strains. The protein expression profiles of different strains of Streptococcus pneumoniae grown under different conditions were evaluated by SELDI analysis. This protocol was also successfully applied to a wide range of Gram positive and negative bacteria (Barzaghi et al., 2004). This technology has a potential to be used as a complement to 2-DE gel electrophoresis for the purpose of microbial proteome analysis.
An MS-based proteomics approach for bacterial evolutionary studies has been recently presented. In this method, a bacterial proteome database was prepared from protein coding ORFs found in 170 fully sequenced bacterial genomes. By analysis of amino acid sequences of tryptic peptides obtained by LC-ESI MS/MS, phylogenetic profiles of these peptides were obtained. It has been shown that the application of this proteomic method is possible to classify the bacterial isolate at the strain level using sufficient amounts of sequence information derived from the MS/MS experiments (Dwornzanski et al., 2006).
In a similar approach, differential proteomics can be used to compare single residues and oligopeptide compositions of the organisms' proteomes. The oligopeptides which are either universally over-or under-abundant, constituting overall properties of the proteomic landscape, as well as oligopeptides whose over-or under-abundance is phyla-or speciesspecific have been proposed. Principal component analysis (PCA) provides a new method to study the landscape of compositional motifs among different species and deduce their phylogenetic relationships (Pe'er et al., 2004).
The differential proteomics approach has been known as a powerful method for evaluating evolutionary relationships amongst prokaryotes or eukaryotes at different evolutionary units such as strains, species, genera and even kingdoms (Enard et al., 2002, Roth et al., 2009, Smithies & Poulik, 1956.
The proteome of different strains of bacteria can be analyzed to find the biomarkers related to various diseases. For this reason, a microbial proteomics database system was set up at the Max Planck Institute for Infection Biology based on 2DE/MS. Many studies have been undertaken to detect biomarkers for various conditions using differential proteomics. Distinguishing pathological from harmless bacteria and identification of a biomarker for a pathogen by LC MS and LC MS/MS analyses have been reported (Mini et al., 2006, Mamone et al., 2009. Certain related examples are presented as follows.

Helicobacter pylori
H. pylori eventually causes gastric/duodenal ulcers or even gastric cancer. Helicobacter infections are the reason for approximately one million people dying annually. Chemotherapy alone is not sufficient for Helicobacter eradication. Therefore, effective vaccines are considered as promising strategies to control this important pathogen. To develop a protective vaccine, a comprehensive list of all possible protein antigens has been provided by the H. pylori genome database (Alm et al., 1999). Since the clinical isolates of H. pylori differ remarkably in their genome sequences, antigens for a subunit vaccine should be selected from a core set of 1281 genes (Salama et al., 2000). It needs large-scale screening and clinical trials to find a protective Helicobacter antigen amongst all possible candidates (Ferrero & Labigne, 2001). To overcome this problem, global techniques such as DNA microarray and proteomics have been employed to identify the promising vaccine antigen subsets rapidly. The proteome of H. pylori has been analyzed by 2-DE and LC-MS (Bumann et al., 2001, Govorun et al., 2003. Immunoblotting of 2-DE gels using human sera has provided global information on the immunoproteome of H. pylori (Jungblut et al., 2000). A comparative proteomic and immunoproteomic analysis has been carried out to identify the antigenic patterns of different H. pylori strains. The results have shown that immunoblotting is suitable as a diagnostic test (Mini et al., 2006). Recently, 2-DE protein maps of H. pylori strain 10K, probed against single sera from H. pylori-positive patients and immunoreactive spots were identified by MALDI-TOF-MS (Lahner et al., 2011). The H. pylori proteome, subproteomes including immunoproteomes, serotome and surface exposed proteins data are stored in a proteomics database (http://www.mpiib-berlin.mpg.de/2D-PAGE/) which is necessary for vaccine development. It has been shown that an antioxidant protein, alkylhydroperoxide reductase (AhpC), is an abundant and important antioxidant present in H. pylori. Oxidative stress-induced AhpC with chaperone activity in vivo was investigated by co-immunoprecipitation, 2-DE followed by nano-liquid chromatography coupled with tandem mass spectrometry (nanoLC-MS/MS). Consequently a significant correlation between the AhpC magnitude of inflammatory damage was detected by immunoblotting assays and endoscopic examinations. AphC was thus suggested as a biomarker for gastric patients (Huang et al., 2011).

Mycobacterium
The Mycobacterium tuberculosis H37Rv genome encoding approximately 4000 proteins has been completely sequenced (Camus et al., 2002). This suggests that Mycobacterium is an ideal model organism for proteomics. Proteome analysis of the virulent and attenuated mycobacterial strain has been carried out in order to identify those proteins having a significant role in its pathogenicity and persistence in the host . Culture supernatant proteins of M. tuberculosis were analyzed by combination of high resolution 2-DE, MS-based techniques consisting of MALDI-MS peptide mass finger printing (PMF), ESI-MS/MS, MALDI-MS PMF and N-terminal sequencing by Edman degradation, and 137 different proteins were finally identified. A small set of protein-specific signature peptide masses was designated as the minimal protein identifier (MPI) upon database comparisons of MALDI spectra. The MPI approach takes into consideration the proposition of protein identity of two sample mass spectra. It is a powerful approach for both a reliable identification of low molecular mass proteins and protein fragments as well as tracking proteins in 2-DE gels. The MPI approach was successfully employed to identify the low molecular mass fragments of mycobacterial elongation factor EF-Tu (Tuf; Rv0685). The 14 kDa antigen (HspX; Rv2031c), the 10 kDa chaperon (GroEs; Rv3418c) and the conserved hypothetical protein Rv0569 of M. tuberculosis H37Rv were tracked in 2-DE gels by this approach .
The genome of M. leprae, an obligate intracellular pathogen causing the disease leprosy, has completely been sequenced mainly in order to identify those genes that are expressed during mycobacterial pathogenesis. Proteins from the cytosol and membrane subcellular fractions were separated by 2-DE and identified by mass spectrometry. The proteins identified in the membrane fraction were analyzed by ESI-MS/MS. They were mostly associated with protein synthesis, secretion and heat shock. Proteins present in the cytosol fraction were separated by 2-DE and 172 spots were analyzed by ESI-MS/MS. The complete list of proteins in both fractions has been described and those involved in virulence, adaptation, detoxification and intermediary metabolism have been identified (Marques et al., 2004).

Vibrio cholerae
V. cholera, the causative agent of severe diarrheal disease, is a Gram-negative bacterium with two different physiological states, in the aquatic environment and in the human small intestine. The whole cell proteome of the V. cholera strain N16961 under anaerobic conditions, approximating the in vivo microenvironment, was separated by 2-DE and the protein spots compared with those in the aerobic environment. Under aerobic conditions, some proteins involved in substrate transport, amino acid metabolism and aerobic respiration were found to be abundant. The increased abundance of some proteins related to motility was observed when the bacterium was grown under anaerobic conditions, thus suggesting a meticulous correlation between V. cholera motility and pathogenesis (Kan et al., 2004). This proteome analysis provides useful information for detection of the antigens by immunoproteomics for the purpose of vaccine development.

Listeria monocytogenes
The pathogen L. monocytogenes causes a severe food-borne infection leading to meningitis, encephalitis and spontaneous abortion in pregnant women. Surface proteins of pathogenic bacteria mediate the main interactions between the bacterial cell and the host. The cell wall subproteome of L. monocytogenes was detected by 2-DE and then identified by N-terminal sequencing and MALDI mass fingerprinting after tryptic in-gel digestion and purification of the resulting peptides. Three proteins were f o u n d t o h a v e n o o r t h o l o g u e i n t h e nonpathogenic L. innocua and might be involved in virulence. Some cytoplasmic proteins such as enolase, glyceraldehyde -3-phosphate dehydrogenase, heat-shock factor DnaK and elongation factor TU were observed in the cell wall proteome unexpectedly. They have neither a secretion signal nor a known surface binding domain. Immunoelectron microscopy demonstrated that they are able to bind human plasminogen specifically. This may suggests that proteomic investigations are necessary to confirm the theoretical predictions of protein localization and function from genome sequence (Schaumburg et al., 2004).

Evaluation of proteins involved in the toxic response
Proteomics can be employed to analyze the microbial or cell responses to an environmental change, such as different culture conditions. The regular stresses in nature are temperature, nutrients, oxygen and toxins. In differential proteomics, two or more sets of proteins from similar but distinct samples that are exposed to different conditions are compared. This is the main application of proteomics which can be used to identify biological markers (Lovrić, 2011). The Differential proteomics approach screens and analyzes proteins qualitatively and quantitatively in order to detect the differential proteins and identify them by mass spectrometric data.
Stimulon is a set of proteins whose amount or rate of synthesis changes in response to a single stimulus . Stimulons are directly identified using protein expression profiles as a quantitative list of proteins which are produced by cells under a given condition. A 2-DE gel run from a sample reveals most of a particular protein expression profile. The protein expression profile of a control condition can be compared with that seen in a test condition to identify the stimulon for that circumstance. Indeed, one of the prime objectives in proteomics is to define up-or down-regulated proteins when a cell is exposed to a certain stimulus. Regulon is a set of proteins whose synthesis is regulated by the same regulatory protein (VanBogelen et al., 1999). The protein expression profiles of mutant strains can be compared to those of wild type strains in order to define proteins with the expression characteristics of a regulon member. Most stimulons consist of multiple regulons. For example, heat shock response of E. coli is controlled by at least two regulons of σ S and σ E (O' Connor et al., 2000).
In medical microbiology, proteomics has a great role in defining the proteins synthesized by pathogenic bacteria following their exposure to eukaryotic cells. The synthesis of certain proteins has been found to be induced during cocultivation of bacteria with host cells. For example, the synthesis of the bacterial heat shock proteins is induced during Brucella abortus infection of bovine and murine macrophages. The changes in the patterns of in vivo gene expression of pathogens have been revealed using 2-DE combined with metabolic radiolabelling or immunoblotting (Cash, 2000). The expression pattern of proteins at higher levels in intracellular bacteria is generally similar to those induced in bacteria in response to stress conditions including extreme acidity, oxygen and high temperature. On the contrary, proteomic studies can be used to investigate the host response to microbial infections. It is possible to compare uninfected and infected whole protein patterns using 2-DE. This can be used to find biological biomarkers. The identification of immunogenic proteins by using 2-DE, immunoblotting and polyclonal sera will significantly aid in vaccine development. One of the major problems for the control of infection is bacterial resistance to antibiotics. Bacteria recognize drugs and antibiotics as toxic materials and produce several proteins in response to these stresses. High resolution 2-DE has been employed to investigate the betalactam antibiotics-resistance in Pseudomonas aeruginosa and S. pneumoniae (Cash, 2000). Identification of those proteins involved in drug resistance will lead to improvements in future antimicrobials. In a comparative study, proteome of H. influenza in response to Ro-64-1874, a 2,4-diaminopyrimidine derivative like trimethoprim as dihydrofulate reductase inhibitor, and to standard antibiotics, such as trimethoprim and sulfamethoxazol were analyzed. The expression profile resulting from exposure to Ro-64-1874 showed a good correlation with those in the database in response to standard antibiotics. Thus, the effectiveness of this new antimicrobial drug was confirmed (Grag & Keck, 1999). In this case, the identification of up-or down-regulated proteins is unimportant. Having a database of responses derived from known compounds is generally sufficient in the study of structurefunction relationship. But more detailed analysis is required for compounds with novel modes of action.
Proteomic/immunoproteomic analyses of H. pylori, Neisseria menangitidis, Streptococcus pyogenes, Bacillus anthracis and M. tuberculosis have been carried out to find new potential vaccine candidates. A large data set has been provided by proteomic studies of different strains and a lot of potentially useful antigens have been introduced as vaccine candidates. However, the data derived from proteome analysis are not consistent. It may be due to the pan-genome, defined as a species gene pool, of some bacterial species being open (Adamczyk-Poplawska et al., 2011). Recent vaccine developments using proteomic studies are summarized in Figure 3.

Concluding remarks
The relative simplicity of microbial cells has made them an attractive target for numerous extensive experiments directed toward understanding the physiology and function of the smallest part of life, in a quest for improved disease prevention and treatment and production of valuable metabolites used in various industries. Following decades of study on microbial systems that are considered as separate compartments, this is the new era ofomics to examine the holistic behavior of a microbial cell.
In this chapter, we focused on proteomics and its applications to microbial systems. Proteome (protein complement of a genome) should help to unravel biochemical and physiological mechanisms at the functional molecular level. We explained how the basic technology of 2-DE can be employed for the separation and characterization of proteins, identification of PTMs and detection of protein-protein interactions. This technique followed by powerful methods of data analysis has a remarkable role in manipulation of metabolic pathways, especially for the improvement of the yield and productivity of microbial products. Furthermore, microbial systematics and epidemiology can be studied by molecular technologies based on genomics and proteomics. The approach of differential proteomics, which compares the distinct proteomes of cells exposed to different conditions, such as normal versus treated ones, could be very helpful in the detection of any environmental change or stress. This review represents the extensive applications of 2-DE in the various areas of the basic science of biology, e.g., cell physiology and molecular biology as well as the applied science of biology, viz. medical and industrial microbiology.

Future perspectives
The fundamental goal of molecular biological research is to determine the function of genes, the role of proteins in metabolic pathways and networks and finally provide a detailed understanding of how these molecules interact and collaborate to work a biological system under different conditions. This is achieved by considering biological systems as a whole and not individual part, which is referred to as systems biology. The new biology is rapidly growing through the concept of -omics. There are strict relations among the major -omics consisting of genomics, metabonomics and proteomics. The sub-groups of these major fields of -omics study the specific interactions/processes/molecules quantitatively and qualitatively. The entire data derived from the specific groups are necessary to achieve an understanding of the properties of a whole cell or system.
Obviously, proteomics provides more applicable data than genomics, although genomics is a pre-requisite for proteomics. The new -omics derived from proteomics are coming into being as peptidomics, glycomics, phosphoproteomics, interactomics etc. that try to supply more detailed information in the fields of function, regulation and interaction of peptides/proteins.
Two dimensional gel electrophoresis is still widely accepted as the powerful method capable of separating proteins from highly complex samples. However, sophisticated devices, authoritative techniques and dominant mathematical methods have been developed to analyze proteomics data. For example, the dual channel imaging technique for 2-DE analysis accompanied with MALDI-TOF mass spectrometry can provide comprehensive information of a proteome. Because of the remarkable capabilities of proteomics in the enhancement of our knowledge regarding the qualitative and qualitative properties of a whole cell, it has been employed as a beneficial tool in microbiological research. Those applications of proteomics in this field of study, as mentioned in this chapter, pave the way for improvement of new products by these micofactories through deep information on microbial physiology, responses and interactions obtained from proteomic data. Therefore, there is no doubt that holistic information contributed by the major -omics is essentially required for the better understanding of a microbial cell.