Search

Rat fibroblasts transformed by a temperature-sensitive mutant of murine p53 undergo a reversible growth arrest in G1 at 32.5 degrees C, the temperature at which p53 adopts a wild-type conformation. The arrested cells contain inactive cyclin-dependent kinase 2 (cdk2) despite the presence of high levels of cyclin E and cdk-activating kinase activity. This is due in part to p53-dependent expression of the p2l cdk inhibitor. Upon shift to 39 degrees C, wild-type p53 is lost and cdk2 activation and pRb phosphorylation occur concomitantly with loss of p2l. This p53-mediated growth arrest can be abrogated by overexpression of cdk4 and cdk6 but not cdk2 or cyclins, leading to continuous proliferation of transfected cells in the presence of wild-type p53 and p2l. Kinase-inactive counterparts of cdk4 and cdk6 also rescue these cells from growth arrest, implicating a noncatalytic role for cdk4 and cdk6 in this resistance to p53-mediated growth arrest. Aberrant expression of these cell cycle kinases may thus result in an oncogenic interference with inhibitors of cell cycle progression.

An exclusive correlation exists between the ectopic expression of the cell-surface marker, intestinal alkaline phosphatase (IAP), and the tumorigenic phenotype of segregants derived from suppressed, nontumorigenic HeLa x fibroblast cell hybrids. This specific association suggests that loss of tumor suppressor function is closely linked to the re-expression of IAP and, therefore, that IAP may be a critical oncogenic factor in these cells. To address this directly, we have used a HeLa IAP cDNA expression vector (pHIAP) to introduce constitutive IAP expression into a nontumorigenic HeLa x fibroblast cell hybrid. Sequence analysis of the HeLa IAP cDNA revealed 5 separate nucleotide alterations when compared to the native IAP cDNA sequence; however, none of these resulted in amino acid substitutions. Four pHIAP transfectants were analyzed for the presence of the intact integrated IAP cDNA and their relative expression levels of exogenous IAP mRNA and protein. The functional integrity of the cDNA-derived IAP product was confirmed by demonstrating proper enzymatic activity and localization to the extracellular membrane. The tumorigenic potentials of the pHIAP transfectants were assayed by s.c. injection into athymic nude mice. No tumors were observed, even after an 11-week incubation in animals. Therefore, expression of IAP in the nontumorigenic HeLa x fibroblast cell hybrid is not sufficient to confer the tumorigenic phenotype. Although ectopic IAP expression is unlikely to be functionally relevant to tumorigenicity in these hybrids, the significance of IAP as a tumor marker is still evident from its apparent strong association with a tumor suppressor locus.

Fusion of tumorigenic HeLa cells with human skin fibroblasts results in genetically stable hybrids which are nontumorigenic and no longer express the HeLa tumor-associated antigen, intestinal alkaline phosphatase (IAP). Previous analysis of spontaneous segregants of the nontumorigenic hybrid have implicated the loss of one copy of human fibroblast chromosome 11 with reexpression of IAP and tumorigenicity. This observation suggests that a putative HeLa tumor suppressor gene(s) is located on chromosome 11 and that this gene may be a negative regulator of the IAP gene. We have isolated several gamma-ray-induced mutants (GIMs) of the nontumorigenic HeLa x skin fibroblast hybrid CGL1 that were specifically selected for reexpression of IAP to further investigate the potential linkage between IAP regulation and the putative tumor suppressor locus. The GIMs have a wide range of cell morphology and level of IAP expression (nearly a factor of 40). The tumorigenicity of the GIMs was examined by s.c. injection into nude mice and all were found to be tumorigenic. The tumor volume-doubling time is in the range of 4 to 8 days for all the cell lines; however, the lag time to reach 500 mm3 tumor volume was significantly longer when the GIM IAP activity was low (less than 20% relative activity), suggesting perhaps that there is a threshold level of IAP expression required for tumor formation and selection for high IAP expression in vivo. However, studies with tumor reconstitutes of the GIMs and transfection studies with an IAP complementary DNA expression vector indicate that high IAP expression alone is not sufficient to confer rapid tumor growth. Therefore, while the data lend strong support to the continued tight correlation between IAP reexpression and tumorigenicity and to our proposal that the tumor suppressor may negatively regulate the IAP gene, it suggests that selection for other gene activities may be responsible for aggressive tumor growth in this cell hybrid system.