| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Biology Behind |
Department of Molecular Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas 77030
With the progress of the human genome project, a large-scale analysis of proteins within a single experiment, called proteomics, has gained much attention. Mass spectrometry and related techniques have rapidly developed with genome database availability. Using mass spectrometry techniques, one can identify proteins in the database without going through the tedious and time-consuming traditional techniques of HPLC2 peptide mapping and then Edman degradation, oligonucleotides synthesis, PCR, and gene cloning. Thus, mass spectrometry techniques are gaining popularity as a versatile method for protein identification. Compared with traditional protein sequencing techniques, mass spectrometry analysis of protein is much more sensitive and faster. Now that these techniques are available, the question is, "How we can apply them to biomedical science?" In this issue, Le Naour et al. (1) apply proteomics-based approaches to identify a new breast cancer-specific antigenic marker.
The first step in proteomics strategy is the identification and isolation of a proteome or protein set of interest. The number of proteins in a proteome can vary from several to several thousand. These may be bound together in a complex, or they may be related by other criteria. For example, we can study the set of proteins that are induced upon stimulation by growth factors or drugs, the proteins induced upon overexpression of a certain gene, or those precipitated by a particular antibody. Once such a protein set is isolated, the individual proteins in the set can be separated using methods such as 1D or 2D gel electrophoresis or column chromatography. These techniques allowed us to determine the number of proteins in the set and to isolate each protein individually. When proteins are separated by gel electrophoresis, individual protein bands or spots can be manually excised from the gel. To identify what protein is in an excised band or spot, gel fragments containing a protein can be digested using site-specific proteases to yield peptide fragments. In the past, peptide fragments would be extracted from gel slices, separated by HPLC, and sequenced by Edman degradation. Edman degradation is a chemical sequencing method that sequentially removes one amino acid at a time from the NH2 terminus of a peptide. From peptide sequences, oligonucleotide primers were designed to PCR-amplify DNA sequences from the gene encoding the protein. Because of the degeneracy of the genetic code, devising such primers may not be straightforward. Once DNA sequences were available, it was possible to clone the gene. This was a common strategy for gene discovery and could be accomplished without the use of databases, but it was a tedious and time consuming process.
In the past few years, dramatic improvements of DNA sequencing databases significantly influenced the methodology of protein identification. Also, the development of improved mass spectrometry techniques and database analysis software increased the speed and sensitivity of protein identification (2, 3, 4, 5, 6, 7) . The improved methods remain the same as the old one up to the protease digestion step mentioned above. However, in the new procedures, the peptide fragments generated by the protease do not need to be separated or purified but can be directly analyzed as a mixture using a mass spectrometer. This is usually done by a matrix-assisted laser desorption ionization time-of-flight mass spectrometer (MALDI-TOF MS; Ref. 8 ). Mass spectrometer analysis will yield the molecular mass of each of the peptides in the mixture (mass spectrum). Because of the current availability of DNA and protein databases, the theoretical mass spectrum of known proteins and putative proteins can be calculated. By comparing obtained mass data with calculated theoretical mass data for all of the proteins in the databases, we can identify our unknown protein.
Another type of mass spectrometry frequently used for protein identification is LC-MS/MS. The mixture of peptides after site-specific protease treatment is separated using a capillary HPLC column and introduced into an ESI-MS. In the ESI-MS, each individual peptide is ionized and then collided into an inert gas to break down the peptide into many random fragments. The masses of these fragments can then be determined and compared with the masses of the theoretical breakdown products of peptides in the database. Finding a match will identify the peptide and protein from which it is generated. Both MALDI-TOF MS and LC-MS/MS techniques have the advantage of requiring subpicomol amounts of starting protein. Also both approaches are high-throughput. Although the classical approach of sequencing peptides by Edman degradation produces extremely reliable data which does not rely on databases, it is much slower than the proteomics approach. The proteomics approach uses comparisons between experimental mass measurements and theoretical masses calculated from databases and thus involves probabilities of matching. This approach seems at first to be less reliable than direct sequencing, but by using two different kinds of mass spectrometry techniques, i.e., MALDI and ESI, the probability of correct identification becomes certain. Alternatively, if one mass spectrometry technique is used, the unknown protein can be cleaved by several different site-specific proteases. Any proteins from the database will have to match the mass spectrum of the unknown protein for each specific protease. This removes the possibility of choosing a false match.
In this issue, Le Naour et al. (1) used a proteomics approach to identify a novel circulating marker for breast cancer. They first separated all of the proteins from a breast cancer cell line using 2D gel electrophoresis. They next analyzed the proteins by Western blot using sera from breast cancer patients as the primary antibody source. For control antibodies, they used sera from healthy individuals or from patients with cancers other than breast cancer. Using this approach, they could identify proteins that specifically elicited immune responses in breast cancer patients.
Sera from 4 of 30 breast cancer patients immunoreacted with three isoforms of a 25 kDa protein, whereas none of the control sera recognized this protein. The authors were then able to extract the protein of interest from 2D gels subjected to silver staining. They identified the protein by cleaving it with three proteases, using mass spectrometry to identify the molecular masses of the resulting peptides, and cross-referencing the mass spectrometry data to a protein database.
In this manner, they identified RS/DJ-1, a protein known to regulate RNA-protein interaction and which is expressed in a variety of tissues (9, 10, 11, 12) . In breast cancer tissue, there is a decrease in RS/DJ-1 protein levels in the nucleus and cytoplasm of cells with a concomitant increase of secreted RS/DJ-1. This result helps to explain the presence of anti-RS/DJ-1 antibody in breast cancer patients compared with healthy individuals. The RS/DJ-1 protein is detected in high levels in the sera of 37% of newly diagnosed breast cancer patients.
The approach used in this study is very interesting (Fig. 1)
. There is currently an existing method to identify markers called SEREX (13
, 14)
, or serological identification of antigens by recombinant expression cloning. Briefly, proteins from cancer cells are expressed in Escherichia coli using cDNA expression libraries. Sera from cancer patients are used to identify proteins that exhibit tumor-specific expression. In this current paper, an established tumor cell line was used instead of expression libraries generated in bacteria. A tumor cell line was used to obtain sufficient amounts of protein, which were separated by 2D gel electrophoresis. Sera from breast cancer patients were used to detect proteins that are specifically antigenic in breast cancers. It seems very likely that using tumor cells as a source of protein is more straightforward and relevant to cancer studies than using proteins expressed in E. coli. However, until recently, it would have been much more difficult to identify the antigenic proteins if the starting material came from a cell line rather than from E. coli expressing a cDNA library. With the use of proteomic techniques, this difficulty now can be largely overcome. This study found not only a potentially good protein marker for the diagnosis of breast cancer, but it also determined that abnormal secretion of this protein into circulation might be responsible for the lower abundance of the protein in the cancerous cells.
|
FOOTNOTES
1 To whom requests for reprints should be addressed, at Department of Molecular Pathology, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. 
2 The abbreviations used are: HPLC, high-performance liquid chromatography; ESI-MS, electrospray ionization mass spectrometer; LC, liquid chromatography.
Received 7/31/01; accepted 8/20/01.
REFERENCES
Key Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
