Comparative Proteomics 2D Gel Technology

The principles of 2D electrophoresis as applied to proteomics are as follows: solubilized, denatured whole proteins are separated in a first dimension by their isolectric properties, then in a second dimension by their molecular weight. Thus, at the end of the process each protein would be expected to occupy a point within the 2D space, the position of which depends on its isolectric point (the combination of ionic charges on all of its side chains and modifications) and its molecular weight. In most cases this combination is unique, theoretically allowing the separation and concentration of all proteins if sample and gel size are unlimited. Even without knowing the identity of a single protein, the potential of this technology to profile the effects of diseases and therapeutics at the level of active proteins is obvious. Because fully translated and modified proteins may be analyzed directly from the patient, information is gained on which is additional to that and which is accessible through genomic microarrays.

In practice, a number of challenges are presented by 2D proteomics. Not least is the acquisition of samples. In contrast to genomic microarrays, relatively large samples are required in order to produce a detectable "spot" of protein, and although it may be routine to remove such tissues as tumors or such fluids as CSF in sufficient bulk from patients, willing "normal" donors are naturally very scarce. Once samples have been obtained, it is vital to preserve the proteins they contain as quickly as possible and in a state suitable for both long-term storage and future 2D analysis. For the analysis itself, the major hurdles are the conversion from one chromatographic medium to another within the same apparatus and reproducibility of the final protein display.

Sample Preparation

Proteins destined for isolectric focusing and molecular weight fractionation must be fully denatured in order to obtain accurate and reproducible data. Thus, lysis buffers are used, which combine strong reducing agents to break cysteine sulfhydryl bonds, and chaotropic mixtures, which disrupt the ionic and hydrogen bonds and hydrophobic partitioning of folded proteins. Other covalent modifications of proteins are preserved by these buffers, enabling observation of their effects on protein mobility in both the elec-trophoretic dimensions. Protein concentration needs to be high when the sample is applied to the system, necessitating maximal solute concentrations in lysis buffers. Many protein degrading and modifying enzymes continue to act even under harsh conditions, and so chemical inhibitors of these are included in the buffer in order to "freeze" the exact state of proteins at the point of extraction. The amounts of different proteins in biological samples varies to an extraordinary degree, to the extent that up to 90% of the protein content may be dominated by the products of just a few genes. Serum in particular contains large quantities of such proteins as albumins and immunoglobulins, and other proteins can be obscured even after 2D separation, or their relative concentration is too low to be detected in a standard protein loading. Depletion of these major components prior to analysis can be highly advantageous in revealing new serum markers, although further reproducibility problems can be introduced by this step.

2D Electrophoresis

The development of industrialized proteomics has led to some standardization in 2D gel technology, so that apparatus and reagents are now supplied commercially (e.g., Biorad, Hercules, California, U.S.A. GE Healthcare, Little Chalfont, U.K.). Batch production and quality control are essential for reproducibility, but it is mostly in sample quality and loading and in downstream data processing that the greatest gains are to be made. Proteins are resolved in solution in the first dimension, migrating within an immobilized pH gradient (IPG) until their individual isoelectric point is reached. The fractionated proteins are then coated with anionic detergent; when the current is switched by 90°, the proteins move into a thin polyacrylamide gel and are separated according to their molecular weight. After electrophoresis the proteins are stained and the images of the gels are recorded. Both methods represented here have a robust history in protein analysis, in the form of isolectric focusing (IEF) gels and SDS-poly-acrylamide gel electrophoresis (SDS-PAGE). The integration of the two provides a very powerful analytical tool, but there are caveats and technical challenges involved at each stage. First, there is the problem of hydrophobic proteins. Traditionally, these are fractionated in gels by solubilization in ionic detergent, usually SDS, which produces an overall negative charge on every protein. This abrogates the natural isoelectric point (pi) of the protein, however, and so cannot be used for IEF. As a result, 2D gels tend to underrepresent hydrophobic species, such as integral membrane proteins. Second, the process of dimension switching can only be successful if the proteins are fully ionized by detergent infusion and with effective transfer from IPG to PAGE, enabling each protein to move in the second dimension as a discrete spot within the gel. Next, imaging of the fractionated proteins must be optimum. After electrophoresis, proteins are prevented from diffusing by fixing in weak acid to partially precipitate and attach them to the gel matrix. However, the gel itself is a very weak structure, which is not self-supporting. Torsions induced by polymerization cause the gel to change shape on release from the glass plates that are used to support it during electrophoresis. This problem can be overcome by binding the gel to one of the plates, preventing any movement of the electrophoretic matrix. Parallel to this advance in enhancing reproducibility has been an increase in the sensitivity of protein detection using fluorescent dyes with extensive linear dynamic ranges and detection limits extending into the femtomole range. Despite the caveats of 2D gel analysis, the result of these developments is that an excess of 2000 proteins can now be detected in a single sample using standard commercially available apparatus. By measuring the intensity of fluorescence, each protein spot (or "feature") can be accurately quantified relative to the total amount of protein in the gel. Finally, there is the challenge of mining the information provided by the gels. Slight fluctuations inevitably occur in gel composition and running conditions, with the result that for almost every protein there is always a slight positional variation between gels. This factor alone would have doomed 2D electrophoresis as a comparative tool had it not been for the development of software to manipulate gel images postacquisition, moving protein features relative to a set of trigonometric markers in order to produce a virtual, composite image. The accumulation of protein features by running sample replicates and adding to sample (patient) numbers is used to generate an electronic "master gel" comprising many thousands of protein features, all identifiable and quantifiable on the original, stored gels. This kind of image manipulation was pioneered by such proteomics companies as Oxford GlycoSciences and are now produced by leading proteomics suppliers, for example, GE Healthcare, U.K. (Ettan Progen-esisTM). The generation of such a large number of data points has itself necessitated additional software to mine and interpret the information. For example, the Swiss Institute of Bioinformatics (Geneva, Switzerland) developed the Melanie series of software programs; proteomics companies have developed sophisticated image analysis platform, such as Rosetta™ (OGS), Kepler™ (LSB), and ImageMaster™ (GE Healthcare), which aim to compare feature presence and intensity from different gels and samples using a variety of statistical parameters. Although these programs are essentially designed to manage large volumes of data, they also provide an important link with the next stage of the proteomic process—protein annotation—by identifying features that are interesting by virtue of statistically validated alteration in disease. Another approach has been to reduce sample requirements by employing direct comparison using differential staining of proteins prior to gel chromatography. GE Healthcare's two-color system, Ettan DIGE, aims to overcome gel-gel variation by running comparative samples on the same gel after prestaining each with a dye that fluoresces at a different emission wavelength. This system also employs an internal pooled standard so that different gels can be compared. A major source of error can be overcome using this approach, but the problem of consistent protein loading remains due to the viscosity of high concentration protein lysates required for detection of whole proteomes. Protein subfractions are much easier to standardize for gel loading, but the fractionation process itself can introduce broad variability.

Protein Identification

The application of the mass spectrometer to the study of proteins is fundamentally in the generation of precise mass measurements of ionized peptides. Indeed, molecular weight measurement of whole proteins as a screening process has been utilized in some proteomic platforms (e.g., SELDI, see section "Other Proteomic Technologies"). But it is in the area of peptide fragmentation, both enzymatic and via ionization, that MS has revolutionized protein identification. Precise measurements of tryptic peptides can be used to identify a protein from within a mixture, but the sequence of amino acids within these peptides can also be elucidated by secondary fragmentation of selected peptide ions in a tandem mass spectrometer (MS/MS). Measurements of ionized fragments produced by MS/MS can be reconstructed into subsets of possible amino acid sequences, which are then compared with theoretical tryptic fragments of electronic translations of real and predicted messenger RNA sequences found in genomic databases. From the set of candidate "hits" obtained in this way, the real peptide is identified by the comparison of the theoretical full fragmentation spectrum of each candidate with the original, real spectrum. Superposition of spectra identifies the peptide (and, therefore, the gene encoding it).

Automation of peptide ion sequencing process using such computer algorithms as Sequest (28) enabled extensive comparison of spectral analyses with genomic databases to elucidate protein sequence. This opened the possibility of high throughput protein annotation. Prior to this, researchers were dependent upon Edman degradation sequencing of proteins, requiring large quantities of purified protein. In contrast, MS/MS sequencing can accurately identify proteins in quantities close to the limit of detection of 2D gels, that is, in the femtomole range, from a starting material of less than 1 mg of the total protein. Rather than devoting an entire project to the purification of a protein of interest, this process can be completed for large numbers of different proteins from sample acquisition to protein sequence in a matter of days.

Cure Your Yeast Infection For Good

Cure Your Yeast Infection For Good

The term vaginitis is one that is applied to any inflammation or infection of the vagina, and there are many different conditions that are categorized together under this ‘broad’ heading, including bacterial vaginosis, trichomoniasis and non-infectious vaginitis.

Get My Free Ebook


Post a comment