When the term "genome" was coined in the 1960s, it was applied specifically to collections of genes, the physical unit of heredity, and the DNA that comprised them. Because RNA's major function is to translate genes into proteins and because RNA is named for its corresponding DNA sequence, it is convenient to speak of RNA as a component of the genome and expression genomics. The same logic might have extended to the polypeptide sequences dictated by RNA. Recognizing that expressed protein measurement offered challenges beyond those previously encountered in gene expression, a new term was coined in 1994, "the proteome" (72). A brief discussion of the additional challenges posed in the study of protein systems helps contextualize the pace of functional proteomics relative to functional genomics.
In Section 1, we discussed the paramount importance of sequencing the human genome for genomics analysis. DNA sequencing has direct relevance for proteomic analysis, in that polypeptide can often be traced to originating DNA sequence. However, proteins, unlike DNA and RNA, undergo extensive posttranslational modifications. Modifications in many cases dictate the biologic properties of interest, independent of polypeptide sequence. It has been estimated that there, on average, 10 different forms of any given protein, each with different properties. It is a combination of the relative and absolute abundance of the protein forms that dictates biologic function. Although there are protein databases similar to those for genes, there is nothing akin to the Human Genome Project that would allow for automation in the way that has been done for DNA and RNA.
In addition to the existence of multiple posttranslational forms of a given protein, protein structure is more complex than that of DNA or RNA. Nucleic acids are composed of only 4 bases sequenced in linear fashion. Proteins are composed of almost two dozen amino acids in which secondary, tertiary, and quaternary structure often dictate function. In summary, whereas genomic analysis focuses essentially on the quantitative expression of genes, the field of proteomics has a broader set of challenges (73). Current proteomics research can be divided into three distinct categories to address these challenges (74-76). The first is abundance proteomics: a description of the relative and absolute abundance of different protein species in a given biologic system. The second is cellular pro-teomics: the study of protein-protein interactions and the function of protein networks. Finally, structural proteomics refer to the work of identifying novel proteins and their constituents. Although initial progress in all three areas is promising, no area of proteomics currently approaches the clinical usefulness of expression genomics, as we will see.
The most commonly used tool in the field of abundance proteomics is the technique termed "two-dimensional poly-acrylamide gel electrophoresis (2D PAGE). First described in the 1970s, 2D PAGE separates proteins along two axes—first by pH gradient to sort by charge, and then by electrophoresis to sort by molecular weight. A heterogeneous mixture of proteins from eukaryotic cell lysates can reproducibly resolve up to 10,000 proteins, as seen in Fig. 8. The gel can be digitized and analyzed by software to compare similar protein spots across experimental conditions in a manner similar to other array-type experiments. This technique has been particularly promising in discriminating proteins that might serve as new tumor-specific markers that can be characterized by the approaches presented below.
In the area of cellular proteomics, investigators have employed a microarray experimental design similar to the cDNA array to describe novel protein-protein interactions (77). On a protein microarray, polypeptides are produced in a high-throughput system such as a yeast vector. Individually identified proteins are spotted onto a glass medium by a robot in a
Fig. 8. 2D PAGE. (Courtesy of Liza Makowski, Ph.D., Division of Metabolic and Complex Diseases. Harvard School of Public Health, Boston, MA.)
manner similar to the production of a cDNA array described in Section 3.2. Once the chip is produced, a single labeled protein can be placed in solution, hybridized against the chip, and scanned just as with the DNA microarray. In this way, areas of fluorescence suggest protein interaction networks that can be explored by other methodologies.
Perhaps most exciting in the broader field of proteomics is its potential to identify proteins through structural proteomics. The traditional approach to protein identification has been to clone the gene encoding a protein of interest, a famously laborious task. In the genomics era, however, evolving techniques such as mass spectroscopy have linked physical properties of proteins directly to known DNA sequence information. In the most powerful examples, these tools can be automated for large-scale protein identification. In mass spectroscopy, a sample, such as a protein, is converted into a charged gas. The charged-gas particles are accelerated by an electric field into a linear detector. The time of flight in that linear detector can be measured, a value proportional to the mass of the particle (78). To use mass spectroscopy to identify an individual protein, the protein is isolated by standard methods and digested by an enzyme such as trypsin to generate a reproducible set of pep-tides of varying lengths. This set of peptides will produce a highly reproducible pattern when analyzed by time-of-flight mass spectroscopy. If the protein has previously been described, the pattern can be matched. For a novel protein, a specific pattern can suggest a theoretic peptide sequence by matching particle mass to likely peptides with that mass. The theoretic peptide sequence can be related to a corresponding DNA sequence from the genome (79).
This brief introduction to proteomics is mostly to suggest a range of potential uses for proteomics techniques in biomedical research and, ultimately, for patient care. In terms of clinical uses, however, we have not offered any specific examples. The lack of clinical examples is not an accident, as proteomics remains exclusively a research tool at this time.
Was this article helpful?
Are You Prepared For Your First Baby? Endlessly Searching For Advice and Tips On What To Expect? Then You've Landed At The Right Place With All The Answers! Are you expecting? Is the time getting closer to giving birth to your first baby? So many mothers to be are completely unprepared for motherhood and the arrival of a little one, but stress not, we have all the answers you need!