Thesis Subject



The subject of my thesis is the study of transcriptional mechanisms responsible for the differential expression of the keratin 18 gene between tumorigenic and non-tumorigenic cells of the SW613-S human colon carcinoma cell line.
Two types of clones have been isolated from the SW613-S human colon carcinoma cell line: clones of cells with a high level of amplification of the c-myc gene which are tumorigenic in nude mice (such as clones SW613-3), and clones of non-tumorigenic cells with a low level of amplification (such as clone SW613-B3) (Lavialle et al., Oncogene. 1988, 3, 335-339). Around 20 % of the genes are overexpressed in tumorigenic cells compared to non-tumorigenic. The difference of the expression level results from an increase in transcriptional rate. The keratin 18 (K18) gene is one of these genes and was chosen as a model for this study. To unravel the regulatory DNA sequence elements involved in the differential behavior of the K18 gene between tumorigenic and nontumorigenic SW613-S cells, a study of its promoter was undertaken.
When I started my work in the lab, they have shown by progressive deletions of the promoter that the minimal promoter (TATA box and initiation site) of the K18 gene has by itself a differential behavior between tumorigenic and non-tumorigenic cells (Gunther et al., Moll. Cell. Biol. 1995, 15, 2490-2499).Ý
My first study was to determine whether sequence elements of the minimal K18 promoter were directly implicated in the control of its differential behavior. I started a functional analysis of the minimal promoter by sequence mutations. The results showed that none of the mutations affected the differential behavior of the promoter (Prochasson et al., Exp. Cell. Res. 1999, 248, 243-259). DNase I footprinting experiments were performed with nuclear extracts from tumorigenic and nontumorigenic cells. The same large footprint (from TATA box to position +3) was visible with extracts of both cell types. Altogether, these results led us to the conclusion that the mechanism responsible for the differential behavior of the minimal K18 promoter does not involve the binding of a factor to a specific sequence. During the course of these experiments, I have shown that the initiation site of the K18 promoter is actually 11 bp upstream of the +1 position previously reported and that the TATA box is the only essential element of the minimal promoter.
I investigate the effect of different activators on the activity of the minimal K18 promoter and on its differential behavior between tumorigenic and non-tumorigenic cells. I have shown that the activity of the minimal promoter can be stimulated in different way by two classes of activators: (i) activators like Sp1, AP1 and Ets that stimulate equally the activity of the promoter in both cell types and maintain its differential behavior; (ii) activators like GAL4, TEF-1 (GTIIC) and VP16 that stimulate the activity of the promoter more efficiently in non-tumorigenic than in tumorigenic cells, resulting in the loss of its differential behavior. A likely explanation of this situation is that these two classes of activators stimulate the promoter by different mechanisms. It has been shown recently that the acidic activators VP16 and GCN4 have the ability to recruit factors that exhibit acetyltransferase activity. This raises the possibility that the differential activity of the K18 promoter may be linked to a difference in acetylation regulatory mechanisms between tumorigenic and non-tumorigenic SW613-S cells.
A possible role of the acetylation in the differential behavior of the K18 gene was then considered. I investigated the effect of deacetylase inhibitors treatment on the activity of the K18 promoter. In transient expression assays, sodium butyrate or trichostatin A treatment of the cells abolished the differential behavior of the promoter between the two cell types by specifically stimulating its activity in non-tumorigenic cells. I have shown that the effect of the histone deacetylase inhibitors was not due to a difference in the sensitivity of the two cell types. I also compared the state of acetylation of the histones extracted from untreated cells and no difference in the electrophoretic patterns was detectable, indicating that no widespread and constitutive hyperacetylation of histones occurs in tumorigenic cells, as compared to non-tumorigenic. From all the results reported above we conclude that differences in the level of acetylation of histones and/or of some other proteins controlling the activity of the transcription complex between tumorigenic and non-tumorigenic cells are likely to play a role in the differential activity of the K18 promoter but that these differences probably occur locally, at the promoter level.
Several factors involved in the control of transcription harbor an acetyltransferase activity. This is especially the case for the p300/CBP proteins and their associated factor PCAF. The p300/CBP-PCAF complex is a coactivator for several specific transcription factors, but it could also be more widely involved in the formation of the transcription complex since p300/CBP and PCAF have been found associated with the RNA polymerase II holoenzyme. The adenovirus E1A protein binds to p300/CBP and PCAF and prevents their interaction. In addition it has been shown that the HAT activity of p300 and PCAF is inhibited by E1A. The CR1 domain of E1A is responsible for its interaction with p300/CBP and PCAF.
The effect of the E1A protein on the activity of the K18 promoter was determined in tumorigenic and non-tumorigenic cells. The adenovirus E1A protein inhibits the activity of the K18 promoter in tumorigenic cells but has no effect in non-tumorigenic cells. This inhibitory effect cannot be reversed by treatment of the tumorigenic cells with a histone deacetylase inhibitor and the viral protein prevents the stimulation of the promoter by this inhibitor in non-tumorigenic cells. The CR1 domain of E1A, which is involved in the interaction with the p300/CBP coactivators, is necessary to the inhibitory capacity of E1A.Ý
These results strengthen the idea that the p300/CBP proteins are specifically involved in the differential regulation of the K18 promoter between tumorigenic and non-tumorigenic cells. Therefore, I set out to establish whether the HAT activity of CBP is able to directly stimulate the activity of the K18 promoter. Forced recruitment of GAL4-CBP fusion proteins to the K18 promoter results in a greater stimulation of its activity in non-tumorigenic than in tumorigenic cells, reducing the differential behavior of the promoter. The histone acetyltransferase activity of CBP is essential for this differential stimulation and the presence of the CBP2 domain greatly augments the activation capacity of the fusion protein. From all these results we conclude that alterations of acetylation regulatory mechanisms are responsible for the higher activity of the K18 promoter in tumorigenic cells and that the CBP (or p300) protein is involved in these mechanisms. I have shown by Northern blot and Western blot analysis, that there was no difference of accumulation level of the mRNA and the protein between the two cell types for p300 and CBP. Thus a quantitative difference in the level of one of these factors is not likely to be the cause of the differential activity of the K18 promoter.
I am now looking at a difference between the two cell types that could be qualitative, possibly affecting the HAT activity of p300 and CBP proteins. Some preliminary results on the HAT activity of p300 and CBP, as determined by an immuno-precipitation HAT assay, tend to indicate that there is no difference between the two cell types. Another possibility, that I am investigating, would be that a factor interacting with p300/CBP and modulating the HAT activity of these proteins would differ between the two cell types. (Work is in progress).

Back to the Top

Back to main page