With the growing amount of data from high-throughput technologies like phage display peptide library, computational identification of novel CAMs that are responsible for organ-specific homing of tumor cells becomes necessary. The MCAM database can facilitate this process. For example, local BLAST search can be performed using any short oligonucleotides or peptides as queries against the CAM sequences available from the Download page as an input database (also see the future plan described later). Once the CAMs are identified, the information including expression and functional profile of the protein can be searched using the online MCAM database.
For example, we recently have identified 25 CAMs that were putatively accountable for the aggressive and metastatic behavior of various tumor cells. These 25 CAMs were identified by BLAST search utilizing the sequence data available from the MCAM database and seven amino acid peptides as queries. Among the candidate CAMs, Semaphorin 5A and Plexin B3 were shown to be expressed specifically in pancreatic and prostate cancer cell lines derived from metastatic tumors (Sadanandam et al. 2007). This is one application of the MCAM database in identifying metastasis specific CAMs. In addition, we have used the database to putatively identify Plexin B3 as a functional Semaphorin 5A receptor (unpublished data) and this finding was simulataneously confirmed.
The MCAM database can also provide gene list for designing CAM specific oligonucleotides or cDNAs for microarray experiments to examine the expression profiles of CAMs in various disease processes. Since the database is classified based on function, the up- or down-regulation of genes for specific functions like calcium dependent cell adhesion molecules can be studied using the custom spotted microarrays. Finally, the database can serve as a test or training dataset for identifying TM proteins, especially CAMs. Therefore, this database helps in nucleotide and protein sequence analysis of CAMs to assist in CAM-specific genomics and proteomics experiments.
What if MCAM database did not exist?
Consider, for instance, an organ-specific metastatic marker discovery where the starting point is a seven amino acid peptide generated from a phage display peptide library screening (figure below). The goal is to identify tumor-specific CAMs with these peptide sequences in their ligand binding regions. As shown in figure below, without the MCAM database, screening for tumor-specific CAMs is a very time consuming and error-prone process. The initial step is to conduct several sequence database searches such as Basic Local Alignment Search Tool (BLAST) and to utilize pattern matching searches. These searches can be performed only using general databases such as the non-redundant database from the National Center for Biotechnology Information (NCBI). The search will result in non-specific proteins, due to the presence of cytoplasmic, nuclear and non-cell adhesive transmembrane proteins along with CAMs. In order to narrow down the search, the resulting proteins need to be verified for their cell adhesive properties by visiting different database sources such as Gene Ontology (GO) and UniProt. Though TM databases could serve the purpose, many TM proteins are not cell adhesion molecules and searches using those databases do not necessarily produce results that will identify only CAMs. Some CAMs can be also predicted incorrectly as non-TM proteins. Finally, the expression of the identified CAMs needs to be examined by performing searches using virtual expression databases like CGAP Gene tool - Gene Finder. Considering the size of mammalian genome sequences available in repositories and with the growing number of peptides available from phage-display peptide library screening, these searches are tedious. Therefore, a consolidated database for CAMs that provides sequences and information including gene expression profiles is clearly needed for more efficient research on mammalian CAMs. However, to our best knowledge, there is no such specific database available for adhesion molecules with cross-references to many other sources including virtual gene expression databases. This motivated us to curate a consolidated record of available CAMs and their information.
MCAM Application Flow Chart (under construction).
Sadanandam, A., Varney, M., Kinarsky, L., Ali, H., Mosley, R., & Singh, R. Identification of Functional Cell Adhesion Molecules with Potential Role in Metastasis by a Combination of in vivo Phage Display and in silico Analysis. OMICS. 2007 Spring;11(1):41-57. PMID: 17411395
Sadanandam, A, Pal SN, Ziskovsky J, Hegde P, and Singh RK: MCAM: A database to accelerate the identification of functional cell adhesion molecules. Cancer Informatics, 4:1-4, 2008. PubMed PMID: 19259402; PubMed Central PMCID: PMC2623291.