Transformation nonresponsive cells owe their resistance to lack of p65/nuclear factor-kappaB activation

Clonal variants of mouse epidermal JB6 cells that are genetically susceptible (P+) or resistant (P-) to tumor promoter-induced neoplastic transformation exhibit differential activator protein-1 (AP-1) response. Transactivation of AP-1 appears to be necessary but not sufficient to promote transformation in JB6 cells. Inhibition of AP-1 is invariably accompanied by inhibition of nuclear factor-kappaB (NF-kappaB) when transformation is suppressed, suggesting that NF-kappaB may also play a role in neoplastic transformation. We report here that transactivation of NF-kappaB is inducible by tumor promoters in P+ but not in P- JB6 cells. Inhibition of NF-kappaB using a nondegradable mutant of IkappaBalpha suppressed inducible anchorage-independent transformation of P+ JB6 cells, suggesting that NF-kappaB activation is required for tumor promotion. Induced degradation of IkappaBalpha occurred in both P+ and P- JB6 cells, indicating that failure to activate NF-kappaB in P- JB6 cells cannot be attributed to failure to degrade IkappaBalpha. Slightly higher levels of nuclear p65 were seen in P+ than in P- JB6 cells. The p65-specific DNA binding activity was also higher in P+ cells upon induction by tumor necrosis factor-alpha, suggesting that differential NF-kappaB activation may be attributable to changes in p65 activity. Transactivation of p65 protein was substantially higher in P+ than in P- JB6 cells, as determined by assay of Gal4-p65 fusion constructs. Thus activated, p65 may be a limiting factor for NF-kappaB activation and transformation responses. Stable expression of p65 in P- JB6 cells conferred not only inducible NF-kappaB and AP-1 activation but also transformation response to tumor promoters. Therefore, p65/NF-kappaB appears to be not only necessary for but also sufficient to confer tumor promotion response. Although stable expression of p65 in P- cells produced p65 increases in whole cell extracts, only the transfectants exhibiting increased nuclear p65 showed transformation response. Thus, elevation of nuclear p65 appears to be a necessary step for a transformation response. The P-/p65 transfectants showing acquired transformation response also showed elevated p65-specific transactivation response, thus recapitulating the NF-kappaB phenotypes seen in P+ cells. Expression of a transactivation-deficient mutant of Jun or dominant-negative extracellular signal-regulated kinase suppressed both AP-1 activation and p65-specific transactivation in JB6 cells, suggesting that AP-1 activity is needed for p65 transactivation and consequently for NF-kappaB activation. Thus, the transformation nonresponsive P- JB6 cells owe their resistance to lack of NF-kappaB activation and p65 transactivation that appears in turn to be attributable to insufficient AP-1 activation.     

Expression of tyrosine kinase Etk/Bmx and its relationship with AP-1- and NF-kappaB-associated proteins in hepatocellular carcinoma

Etk/Bmx is a cytoplasmic tyrosine kinase, which was first identified in human bone marrow cells. It has been found to play an important role in the regulation of differentiation and tumorigenicity in some cancers. The aim of this study was to investigate the significance of Etk/Bmx expression in hepatocellular carcinoma (HCC) and the relationship between Etk/Bmx and activated protein-1 (AP-1)- and nuclear factor-kappaB (NF-kappaB)-associated proteins. We used immunohistochemisty to examine 40 cases of human HCC along with corresponding nontumor tissues to assess Etk/Bmx, Jun family (c-Jun, JunB, JunD), Fos family (c-Fos, FosB, Fra-1) and NF-kappaB p65 expression in these samples. Etk/Bmx expression was present in 12 of 40 (30%) HCC specimens, 4 of which among the 25 well-differentiated tumors and 8 among the 15 poorly differentiated tumors, respectively. In contrast, 6 of 40 (15%) cases expressed Etk/Bmx in adjacent nontumor tissues. Expression level and cellular localization of Etk/Bmx were different in cancer cells and nontumor cells. Etk/Bmx expression was correlated with histological differentiation, but not with clinicopathological features including tumor size, HBV infection, cirrhosis, and metastasis. There was a close relationship between Etk/Bmx and c-Fos expression in HCC. Etk/Bmx immunopositivity was independent of c-Jun, JunD, FosB, Fra-1 and NF-kappaB p65. Our results indicated that Etk/Bmx may have different biological roles in tumor and nontumor cells, and may be involved in regulating hepatocyte differentiation by c-Fos activation in HCC.     

AP-1 complexes containing cJun and JunB cause cellular transformation of Rat1a fibroblasts and share transcriptional targets

To investigate the role of individual Jun proteins in cell growth and transformation, we have used a doxycycline-inducible retroviral vector to regulate their expression in rat fibroblasts. AP-1 complexes enriched with cJun and JunB result in morphological alterations and anchorage-independent cell growth consistent with a transformation-like phenotype, whereas complexes enriched with JunD had an antiproliferative effect. These results suggest that genes regulated by both cJun and JunB are potentially involved in transformation and that they can be distinguished from those regulated by AP-1 complexes containing JunD. To identify genes regulated by cJun and JunB that may have a role in anchorage-independent growth, we investigated differential gene expression by each of the Jun family members using the Affymetrix Rat oligonucleotide microarray, RG_U34A containing approximately 8000 genes. Differentially regulated genes were identified and grouped for correlation with regulation by the different Jun proteins. A total of 33 candidate genes were found to be differentially regulated by both cJun and JunB and not by JunD. These genes have roles in cell metabolism, growth, signal transduction, migration and adhesion. We validated the differential regulation by cJun and JunB of 10 candidate genes by Northern blot analysis. Of these, eight were further characterized as potential direct targets of AP-1 regulation based on Northern blot results showing differential regulation that correlate with cJun expression. Our results show that inducible cJun and JunB expression result in anchorage-independent growth of Rat1a cells, distinct from JunD-expressing cells. This model system and a functional genomic approach enabled us to differentiate AP-1-regulated genes involved in transformation from AP-1-regulated genes known as bystander genes. This approach significantly reduces the number of bystanders and allows for the targeting of genes specifically involved in transformation.