Activating amino-acid substitutions impair the intrinsic GTPase activity of the ras protein and generate constitutively activated signal complexes with transforming activity. Point mutations in ras
genes have been found in a wide variety of human and murine cancers, especially in human pancreatic cancers (90%) and human colorectal cancers (45%) (Bos et al, 1987
; Almoguera et al, 1988
; Bos, 1989
A large fraction of human cancers harbour point mutations in the ras
gene at codon 12, in which the normal residue is substituted by a Val, Asp or Cys residue. From an immunological perspective, these determinants may represent highly specific epitopes for T-cell (CD4+ and/or CD8+) recognition in cancer immunotherapy (Tsang et al, 1994
; Fossum et al, 1995
). Evaluation of point-mutated ras
as a T-cell epitope could be determined biologically using short synthetic peptides that precisely correspond to the altered sites.
Several laboratories established approaches in both murine and human systems to evaluate point-mutated ras p21 oncogene products as potential tumour-specific targets and to characterise the resulting cellular immune responses.
Studies using those system have shown that mutant ras protein is able to serve as a tumour-specific antigen (Peace et al, 1991a
; Cheever et al, 1995
). T-cells from animals immunised by ras peptides can lyse cells transformed to express the ras products with the same mutation in animal models (Peace et al, 1991b
The observation that ras products can be immunogenic in mice suggested that similar T-cell responses might be present in humans (Peace et al, 1991a
). In vitro
stimulation of human T-cells from some normal individuals or cancer patients with mutant ras peptides results in the expansion of CD4+ and CD8+ precursors, which may exhibit cytotoxicity against autologous or MHC-matched, antigen-bearing target cells (Fossum et al, 1994b
; Juretic et al, 1996
). In addition, humoral responses specific for ras products have been observed in colorectal cancer patients (Takahashi et al, 1995
Other studies showed that the human T-cells can recognise peptides that span the mutated segment of the ras protein and that the ras
peptide-specific T-cells can respond to ras protein containing the same substitution (Jung and Schluesener, 1991
; Gedde-Dahl et al, 1992
; Fossum et al, 1994a
; Qin et al, 1995
). Previous studies have identified that ras
mutation-specific memory T-cells in only two human cancer patients with follicular thyroid cancer and colorectal cancer (Gedde-Dahl et al, 1992
; Fossum et al, 1994a
). In both of these cases, the corresponding ras
mutation could not be detected in the available tumour biopsy samples.
However, it remains unclear whether T-cells can recognise the same mutated peptides that are expressed in tumour tissues from the same individual tissue. We first analysed the correlation between the peptides that induce T-cell response against Ki-ras peptides and the Ki-ras mutations in pancreatic and colorectal cancer tissues.
We are not able to show that T-cells can recognise the mutated ras peptide that is expressed in the tumour cells from the same individuals.
To assess the immune response against Ki-ras peptides in patients with pancreatic and colorectal cancer patients, primary proliferative response assays and IFN-γproduction assays were done. No effect of these peptides could be detected using the primary proliferative response assay, because of the low immunogenicity of the Ki-ras products. On the other hand, it was possible to detect a Ki-ras response by twice stimulation with Ki-ras peptides and stimulation by IL-2 in an IFN-γ production assay. The stimulation of such a low concentration of IL-2 is intended to suppress the secretion of IFN-γ from other cell sources, including the activation of natural killer cells. The purpose of IFN-γ production assays is to determine which peptide is significant for the T-cell activation, and actual concentration is not calculated. We used the 18-mer peptides that can bind to class II molecule, but not to class I molecule. Therefore, the CD4+ cells might be related in this study. However, the primary culture in the IFN-γ production assay contains antigen-presenting cells, so CD8+ cells may be contributed in this assay. T-cell immunity against Ki-ras peptides was detected at lower frequency in colorectal cancer patients than in pancreatic cancer patients. This suggested that the immunogenicity to ras products was recognised more strongly in pancreatic cancer patients who had a high frequency of ras mutations in their tumour tissue, but the T-cells did not recognise the ‘correct’ mutation. T-cells in the patients with pancreatic cancer may have been previously exposed to similar antigens in vivo.
We propose two possible reasons why T-cells from a given individual cannot recognise the same mutated ras peptide expressed in the tumour tissues of that individual.
First, the length of the peptides bound by MHC class II molecules are not strongly constrained. Therefore, the binding of peptides to MHC class II molecules is more promiscuous than the binding of peptides to MHC class I molecules. Peptides that bind to MHC class II molecules are variable in length and their anchor residues lie at various distances from the end of the peptide (Rudensky et al, 1991
; Rammensee et al, 1995
; Hammer et al, 1994
). In patients who respond to wild-type peptide, their anchor motif of class II molecule are considered to be similar to wild-type peptide.
Second, we speculate that tumour cells harbouring a mutation have been eliminated by the immune system in cancer patients. Ki-ras
mutations are thought to occur as a relatively early event in the developmental sequence of colorectal adenocarcinoma (Vogelstein et al, 1988
), and may therefore be expressed early on by most of the tumour cells. The cancer present at the time of biopsy may have progressed further, and may not harbour the same mutation because the cancer cells harbouring the earlier mutation were eliminated by the immune system in an early event of tumour development (Nakagawa et al, 1991
It is suggested that T-cells responding to synthetic 18-mer Ki-ras peptides are restricted by HLA-DR or -DQ class II molecules. Ras p21 is an internally localised biosynthetic protein and is therefore potentially susceptible to the endogenous pathway of antigen processing and subsequent loading with MHC class I or II molecules. It is possible that point-mutated ras p21 proteins may be processed through an exogenous mechanism. In this regard, Harmer et al
reported that point-mutated ras p21 proteins can be found in the external tumour microenvironment as well as in the plasma of tumour-bearing mice, and if so, may be available as exogenous antigens for endosomal processing by antigen-presenting cells and presentation to CD4+ T-cells (Harmer et al, 1991
CD4+ T-cells are thought to play an important and central role in immunoregulation through the production and action of lymphokines. Accordingly, several peptide- or protein-based immunotherapy has great therapeutic potential for cancer patients whose tumours harbour point mutations in codon 12 of the ras p21 proto-oncogene.