Constitutive presence of cytochrome c in the cytosol of a chemoresistant leukemic cell line.

The release of holocytochrome c (cyt c) from mitochondria into the cytosol is reportedly a landmark of the execution phase of apoptosis. As shown here, the P-glycoprotein- (P-gp) expressing K562/ADR cell line (but not the parental K562 cell line) exhibits both cytosolic and mitochondrial cyt c in the absence of any signs of apoptosis. K562/ADR cells were found to be relatively resistant to a variety of different inducers of apoptosis, and blocking the P-gp did not reverse this resistance. The release of cyt c in non-apoptotic K562/ADR cells was not accompanied by that of any other mitochondrial apoptogenic protein, such as AIF or Smac/DIABLO, and was inhibited by Bcl-2 over expression. In addition, using a cell-free system, we show that mitochondria isolated from K562/ADR cells spontaneously released cyt c. These data suggest that cyt c release may be compatible with the preservation of mitochondrial integrity and function, as well as cell proliferation.

Resistance to apoptosis is increased during metastatic dissemination of colon cancer.

Apoptosis dysfunction in metastases has been suggested to participate in their poor response to conventional anticancer treatments. To address this question, we have analyzed the sensitivity to cell death induced by non-steroid anti-inflammatory drug, Sulindac, the most common drug used in colon cancer chemotherapy, 5-fluorouracil (5-FU) and the short chain fatty acid, butyrate (Bu) in cell lines derived from a primary colorectal tumor (ALT-I) as well as the liver (ALT-F) and the lymph-node (ALT-G) metastases. We have previously shown both in vitro by analyzing anchorage-independent cell proliferation and in vivo by subcutaneous injection into athymic nude mice that the ALT-F and ALT-G cells were more tumorigenic than the primary ALT-I cells. All these cell lines, derived from an untreated patient, were highly resistant to apoptosis induced by 5-FU and Sulindac but were sensitive to Bu-induced apoptosis. The resistance to apoptosis was, as quantified by the induction of caspase activity and the relative percentage of apoptotic cells, higher in the metastatic cell lines, than in the ALT cell line. When compared to the primary tumor, more anti-apoptotic bcl-2 and less pro-apoptotic bax were expressed in the liver and lymph node metastatic cell lines. Quite remarkably, the expression of bax was up-regulated during Bu-treatment, a feature that could explain its powerful pro-apoptotic activity.

Potentiation of apoptosis by mitochondria in a cell-free system.

Using a cell-free system, we show that rat liver mitochondria, but not mitochondrial extracts, potentiated apoptosis triggered by cytosols derived from apoptotic cells. Apoptosis potentiated by mitochondria appeared to be inhibited by caspase 3 but not by caspase 1 inhibitors. A cytosolic caspase-3-like activity was increased by the addition of mitochondria to apoptotic cytosols; the latter activation was inhibited by the addition of bcl-2. Chelation of calcium by EGTA significantly and specifically inhibited the apoptosis potentiated by mitochondria as well as the increase of caspase-3-like activity. The incubation of mitochondria with apoptotic cytosols led to the release of cytochrome c, this latter phenomenon being inhibited by EGTA. Calcium or cytochrome c and dATP, however, did not reproduce the mitochondrial potentiation in the absence of the organelle. Thus, mitochondria can initiate and potentiate apoptosis through similar but not identical mechanisms.

Over-expression of the S13 ribosomal protein in actively growing cells.

In order to define biological markers of aggressiveness, 2 rat colon-carcinoma cell lines differing by their tumorigenicity were used to clone genes over-expressed in colon carcinoma as compared with normal epithelial cells. A progressive rat colon-carcinoma clone (PROb) cDNA library was hybridized with 32P-cDNA synthesized from mRNA prepared from these PROb cells, or from regressive cells (REGb) derived from the same tumor. Several clones were isolated after the initial screening. The specificity of each clone was confirmed by RNA blotting. One of these (B9) was found to hybridize to an mRNA 30-fold more abundant in PROb cells than in normal adult rat colon, and was therefore selected for further study. No gene amplification was detected by Southern blot analysis, indicating that the difference in mRNA content was most likely due to an increased transcription of this gene. Sequencing of the cDNA revealed approximately 98% homology with the rat S13 ribosomal protein. The expression level of this gene was determined in a series of rat cell lines with different growth rates. A good correlation was found between these 2 parameters. Our data suggest that the S13 ribosomal-protein gene can be used to evaluate the growth rate of tumor cells, which might be correlated with their aggressiveness. In an initial trial experiment, S13 ribosomal-protein mRNA was detected in a series of human colorectal tumors by in situ hybridization. A strong signal was seen in the 4 tumors analyzed.