Multidimensional liquid chromatography tandem mass spectrometry for biological discovery (Proteomics)

1. Introduction

The goal of biological studies is to connect genes and proteins to specific processes. In the past, dissection of biological processes entailed detailed and focused studies of a specific gene or protein. The availability of complete genome sequences has made possible broader studies to dissect the functions of proteins and genes in entire pathways, thus extending the knowledge of the process. Proteomics in an organism is facilitated by the existence of a completed genome sequence, which allows more facile and higher-throughput correlations between gene and protein sequences than has been possible with more traditional protein-sequencing methods. Correlations are derived through the use of mass spectrometry data that reveals the identity of a protein through the molecular weights of peptides in a tryptic map or by fragment ions of individual peptides. Fragmentation of individual peptides is accomplished through the use of tandem mass spectrometers that can select individual ions, induce dissociation, and record the resulting product ions (Hunt et al., 1986). The resulting fragment ions represent the amino acid sequence of the peptide. Fragmentation patterns can be matched to sequences in a database, thus identifying the protein from which the peptide was derived. An analytical advantage to this method is the ability to measure peptide mixtures derived from proteolytically digested protein mixtures (Figure 1) (Yates, 1998). This approach represents an efficient and sensitive method to identify the components of protein complexes, protein localization (e.g., proteins located in organelles or subcellular spaces, cellular membranes), and protein modifications (Link et al., 1997; Wu et al., 2003).


To accommodate complex protein mixtures, separation systems of high resolution are desirable. Multidimensional liquid chromatography (MudPIT) provides high-resolution separations of peptides in an automated, convenient manner. Link et al. (1999) developed an integrated form of MudPIT that combines strong cation exchange (SCX) resin with a reversed-phase (RP) to create a biphasic column (Figure 2). The combination of SCX and RP packing materials creates a bimodal, orthogonal separation by charge and hydrophobicity. Modification to this process was introduced by McDonald et al. (2002) to add a third layer of chromatography material to effect an on-column removal of buffer salts that may be present in the sample and to improve retention of peptides on the SCX phase. Washburn et al. (2001) demonstrated that this approach was effective for an unbiased analysis of the components of yeast cells by identifying large and small proteins, basic and acidic proteins, and membrane proteins.

The process of shotgun proteomics involves digesting a protein mixture and then separating the peptides into a tandem mass spectrometer. Peptide ions are individually selected=

Figure 1 The process of shotgun proteomics involves digesting a protein mixture and then separating the peptides into a tandem mass spectrometer. Peptide ions are individually selected for collision-induced dissociation and then each resulting spectrum is searched through a sequence database. The results are filtered, assembled, and organized to view the results

An integrated multidimensional liquid chromatography column to perform high-resolution separations of peptides. The tip of the column is packed with reversed-phase (RP) packing material followed by a layer of strong cation exchange (SCX) resin and then a short layer of RP material. Peptides are electrosprayed out the end of the column directly into a mass spectrometer

Figure 2 An integrated multidimensional liquid chromatography column to perform high-resolution separations of peptides. The tip of the column is packed with reversed-phase (RP) packing material followed by a layer of strong cation exchange (SCX) resin and then a short layer of RP material. Peptides are electrosprayed out the end of the column directly into a mass spectrometer

2. Analysis of protein complexes

This strategy has been used to improve the analysis of protein complexes and to localize proteins (Figure 3). A recent study used biochemical purification of protein complexes combined with identification using the MudPIT approach (Hazbun et al., 2003). The essential hypothetical open reading frames were all epitope-tagged with the Tandem Affinity Purification (TAP) tag that combines a protein A IgG binding domain with the calmodulin-binding peptide (Rigaut et al., 1999). The two binding domains are linked together with a short amino acid sequencing containing the tobacco etch virus protease cleavage site. Inclusion of the cleavage site allows protein complexes to be removed from the initial purification step on an IgG column by proteolytic cleavage. The complex is recovered by passing the eluate through a calmodulin column and then released by elution with EDTA (ethylenediamine tetraacetic acid) (Gavin et al., 2002). The complex is then digested using trypsin to produce a collection of peptides. Peptides are loaded onto a two-dimensional LC column and separated directly into a tandem mass spectrometer. Tandem mass spectra were automatically collected and then searched against the yeast sequence database to identify the proteins in the mixture. Of the 100 tagged proteins analyzed, complexes were identified for 29 of them. On the basis of the identities of the proteins interacting with the tagged proteins in combination with data from colocalization and yeast two-hybrid experiments, gene ontology annotations could be assigned to the proteins. This study and others have shown MudPIT to be a very effective method to identify proteins in complexes (Graumann et al., 2003) as well as to identify sites of modifications in proteins of complexes (Cheeseman et al., 2002; MacCoss et al., 2002).

Protein complexes can be analyzed by incorporating an epitope into a protein sequence that contains protein A, the tobacco etch virus protease cleavage site, and the calmodulin-binding peptide. After elution of the complex from the columns, the proteins are digested and analyzed by MudPIT

Figure 3 Protein complexes can be analyzed by incorporating an epitope into a protein sequence that contains protein A, the tobacco etch virus protease cleavage site, and the calmodulin-binding peptide. After elution of the complex from the columns, the proteins are digested and analyzed by MudPIT

3. Analysis of subcellular compartments

A particular challenge for proteomics studies is the identification of proteins in membranes and organelles. These studies are complicated because it is difficult to enrich for membrane or organelle proteins without contamination and because these proteins have limited solubility in buffers compatible with mass spectrometry. By using MudPIT, subtractive analysis approaches can be utilized as well as strategies to digest only the soluble segments of membrane proteins.

Schirmer et al. (2003) used a subtractive approach to identify proteins of the nuclear envelope (NE) (Figure 4). Enrichment of the NE is particularly challenging because it is contiguous with the endoplasmic reticulum (ER) and intermixed with mitochondria. An enriched fraction of the microsomal membranes (MM) that includes both ER and mitochondria can be generated. Schirmer et al. generated an exhaustive proteome analysis of the microsomal fraction and an enriched NE fraction using MudPIT. Proteins identified in the MM fraction were then subtracted from those identified in the NE fraction. Sixty-seven proteins were identified as hypothetical, integral membrane proteins. This example illustrates a powerful aspect of the shotgun proteomics method enabled by MudPIT. Membrane proteins were readily identified using this approach, since it is not necessary that the proteins be solubilized for digestion only that digestion of the soluble portions of the membrane proteins is achieved. A possible limitation of the approach is potential loss of information from those membrane proteins with minimal regions exposed.

Subtractive proteomics was applied to the nuclear envelope of mammalian cells by enrichment of the microsomal membrane fraction for exhaustive proteome analysis. An enriched nuclear envelope fraction was then analyzed and proteins in common between the two were subtracted from the list of identified proteins

Figure 4 Subtractive proteomics was applied to the nuclear envelope of mammalian cells by enrichment of the microsomal membrane fraction for exhaustive proteome analysis. An enriched nuclear envelope fraction was then analyzed and proteins in common between the two were subtracted from the list of identified proteins

Next post:

Previous post: