Effectiveness of gene expression profiling for response prediction of rectal cancer to preoperative radiotherapy

BACKGROUND

Our aim was to determine whether the expression levels of specific genes could predict clinical radiosensitivity in human colorectal cancer.

METHODS

Radioresistant colorectal cancer cell lines were established by repeated X-ray exposure (total, 100 Gy), and the gene expressions of the parent and radioresistant cell lines were compared in a microarray analysis. To verify the microarray data, we carried out a reverse transcriptase-polymerase chain reaction analysis of identified genes in clinical samples from 30 irradiated rectal cancer patients.

RESULTS

A comparison of the intensity data for the parent and three radioresistant cell lines revealed 17 upregulated and 142 downregulated genes in all radioresistant cell lines. Next, we focused on two upregulated genes, PTMA (prothymosin alpha) and EIF5a2 (eukaryotic translation initiation factor 5A), in the radioresistant cell lines. In clinical samples, the expression of PTMA was significantly higher in the minor effect group than in the major effect group (P = 0.004), but there were no significant differences in EIF5a2 expression between the two groups.

CONCLUSIONS

We identified radiation-related genes in colorectal cancer and demonstrated that PTMA may play an important role in radiosensitivity. Our findings suggest that PTMA may be a novel marker for predicting the effectiveness of radiotherapy in clinical cases.

Further biochemical characterization of rabbit reticulocyte eIF-4B

In this report, it is shown that eIF-4B exhibits ATP hydrolysis activity in a ribosome-dependent fashion. This activity is significantly enhanced by the addition of eIF-4A and eIF-4F, although neither of these factors, alone or together, display a significant ribosome-dependent ATPase activity. Sepharose-6B gel filtration experiments with combinations of nonlabeled and 14C reductively methylated initiation factors (eIF-4A, eIF-4B, eIF-4F, and eIF-3) provide evidence that both eIF-4B and eIF-4F, but not eIF-4A, interact with ribosomes in the presence of specific factors and ATP. It is not known with which ribosomal subunit(s) these factors interact. In ultraviolet radiation-induced crosslinking studies, an eIF-4A and/or eIF-4F and ATP-dependent crosslinking of eIF-4B to radiolabeled oligo(A)12-18 was observed. Past studies with oxidized-cap mRNA achieved cross-linking of all three of these factors, driven by the specific interaction of eIF-4F p24 with the m7GTP cap structure; this is the first reported instance of only eIF-4B cross-linking to RNA.

Photochemical cross-linking of translation initiation factor 3 to Escherichia coli 50S ribosomal subunits

Translation initiation factor 3 (IF-3) was bound noncovalently to Escherichia coli 50S ribosomal subunits. Irradiation of such complexes with near-ultraviolet light (greater than 285 nm) resulted in covalent attachment of initiation factor 3 to the 50S subunit. Photo-cross-linking attained its maximum level of 40% of that which was noncovalently bound after 90 min of irradiation. Cross-linking was abolished in the presence of either 0.5 M NH4C1 or 0.25 mM aurintricarboxylic acid, indicating that specific binding of initiation factor 3 to the ribosome was a prerequisite for subsequent covalent attachment. Further analysis showed that all the IF-3 was covalently bound to a small number of 50S subunit proteins. The major cross-linked proteins were identified as L2, L7/L12, L11, and L27 by immunochemical techniques. These results are discussed in light of the proposed mechanism for IF-3 function.

Photochemical cross-linking of initiation factor-3 to Escherichia coli 30 S ribosomal subunits

Initiation factor-3 has been photochemically cross-linked to Escherichia coli 30 S ribosomal subunits by means of near-ultraviolet (> 285 nm) irradiation. The cross-linking was judged to be specific since noncovalent binding was required for subsequent cross-linking and was prevented by the presence of 0.5 M NH4Cl or 0.5 mM aurintricarboxylic acid. Covalent attachment reached its maximum level of 8.5 to 11% of the noncovalently bound IF-3 after 60 min of irradation. Cross-linking was unaffected by the presence of the photosensitizer, acetone (5 or 10%), but was reduced to 30 to 40% of its maximum level by addition of 5 mM dithiothreitol, a free radical scavenger. Analysis of the 14C-labeled IF-3 x 30 S subunit covalent complex revealed that the IF-3 was distributed between the protein and RNA of the subunits in a 3:1 ratio. The target proteins have been identified as S7, S11, S12, S18, and S21 by immunochemical techniques.