Biomedical Engineering Reference
In-Depth Information
A few selected CAMs and their phage-display peptides are shown in Table 2.
Among these, Semaphorin 5A (SEMA5A) and Plexin B3 were predicted as tumor
markers for pancreatic and prostate cancers, respectively. To test whether these
i ndings might be applicable to humans, we verii ed the expression of SEMA5A
and Plexin B3 in human pancreatic and prostate cancer cell lines, respectively,
using RT-PCR analysis (Sadanandam
et al
. 2007).
Table 2
Selected CAMs identified by phage display peptide library and bioinformatics
approaches
Phage display peptide
GenBank accession no.
Protein
NAFTPDY
NP_033180
SEMA5A
NP_062533.2
Plexin B3
NP_035069.1
NRP2
YQDSANT
NP_034039.1
CXCR2
TPLQPTA
NP_001034239.1
CD44
KSWKVYV
XP_903659.2
SEMA4C
h e peptides were identii ed using
in vivo
phage display peptide library. h ese peptides were used as queries for
bioinformatics approach to identify CAMs that have tumor-specii c expression leading to organ-specii c metastasis
(Sadanandam
et al
. 2007).
Identification of CAMs by Protein-Protein Interactions
Protein-protein interactions between CAMs expressed on the surface of tumor
cells and the vasculature of secondary organs play a major role during metastasis.
Computational methods can be used to predict interactions between CAMs
that are expressed on the surface of primary tumors and secondary organs. We
used a unique and integrated approach to identify interacting partner(s) of a
CAM, SEMA5A, by beginning with seven CAMs as putative binding partners of
SEMA5A (Sadanandam
et al
. 2008b). We chose SEMA5A because we identii ed
it as one of the candidates from the phage display analysis and observed its
expression in aggressive pancreatic cancer cell lines (Sadanandam
et al
. 2007). h e
putative ligand interacting residues are seven (NAFTPDY) amino acids that were
screened to bind specii cally to endothelia of liver, brain and bone marrow using a
phage display peptide library. In keeping with Dwyer and Root-Bernstein/Dillon
theories of protein evolution, we chose eight proteins, including SEMA5A, for
protein interaction analysis based on similarities in their putative ligand-binding
residues. To achieve the goal of identifying SEMA5A interacting partners, we
used integrated bioinformatics approaches such as hydrophobic complementarity
of protein structure, functional patterns, information on domain-domain
interactions, co-expression of genes and protein evolution. Among the set of
seven proteins selected as putative SEMA5A interacting partners, we found the
