Epigenetic events, such as DNA methylation, are crucial in establishing the correct pattern of gene expression. Disruption of this program leads to an aberrant mRNA transcription and potential loss of anti-cancer checkpoints.
The role of seladin-1 in cancer is still unclear, probably due to the multiple roles of this gene in regulating cell functions. However, a thorough analysis of seladin-1 role in oncogenesis and oxidative stress indicated that its expression is involved in the regulation of Ras-induced transformation and senescence in human and rodent cells [4
]. Apparently, seladin-1 exerts its protective effects against oxidative stress following two independent ways. Seladin-1 is up-regulated as a response to acute oxidative stress, with a cholesterol-dependent mechanism, but is down-regulated upon chronic exposure to oxidative stress. However, also the reduction of seladin-1 expression appears involved in a prosurvival strategy due to its interaction with p53 status and function, as recently demonstrated [25
]. The initial consideration that adrenal glands show very high mRNA expression levels of seladin-1 [1
] and that adrenal carcinomas show a significant reduction of seladin-1 mRNA [20
], has prompted a deeper analysis of this gene expression in adrenal cancer. On this basis, the definition of the regulation of seladin-1 expression seems to be crucial for the comprehension of the mechanisms underlying its downregulation in tumoral tissues.
In the present study we demonstrated for the first time the presence of a functionally active CpG island in the regulatory sequence of Seladin-1 gene. Qualitative and quantitative methylation specific PCR clearly indicated that in the adrenal cancer cell lines H295R and SW13 the CpG island is densely methylated and that the treatment with the 5-Aza was able to decrease DNA methylation. In the same experiments we demonstrated that the expression of seladin-1 mRNA could be directly related to the altered pattern of promoter methylation since exposure of adrenal cell lines to 5-Aza was associated to a significant induction of Seladin-1 mRNA expression in SW13 and H295R lines, even if the effects of 5-AZA on H295R methylation is apparently less evident than in SW13. This difference could refer to the different proliferation rate of the two cell lines. The higher proliferative rate of SW13 can explain the major effect of 5-AZA in reducing DNA methylation, since the incorporation of citosine analogue is much more elevated in cells with a shorter duplication time. After 6 days of treatment, methylation in SW13 is close to 0% and the effect of Seladin-1 mRNA expression is maximal. Conversely, H295R tend to proliferate more slowly and after 6 days 30% of their DNA is still methylated and the gain of mRNA expression, even if significantly increased, is less intense than in SW13.
In addition, we experienced also a different effect of the demethylating agent on the respective proliferation rate. 5-Aza treatment induced a reduction of cell growth in H295R that reaches a 50% inhibition after a 6-days treatment, as previously reported [26
], whereas the reduction of cell line growth in SW13 was only 15% in comparison to controls. Therefore, the discrepancy of the effects in the two cell lines is probably connected to their ability to grow in normal cultural conditions and under 5-AZA treatment.
After these preliminary indications obtained 'in vitro', we tried to confirm the presence of an epigenetic control of seladin-1 expression also in 'ex vivo' samples. Real time analysis performed on the same promoter region in DNA of adrenal carcinomas, adenomas and normal glands showed an inverse relationship between methylation of seladin-1 and its expression. In particular hypermethylation was associated to reduced seladin-1 expression levels in adrenal cancer compared to normal adrenal gland and adenomas. No significant difference between adenomas and normal adrenal glands was evidenced. Thus, at least in adrenal carcinomas, hypermethylation could account for the reduction of mRNA expression [20
To our knowledge, this is the first evidence that the pattern of expression of seladin-1 may be regulated by a differential methylation status of the promoter region of this gene, even if the transcription activation after 5-Aza is not a final confirmation of a direct epigenetic regulation on the gene itself. In fact an indirect effect resulting from the demethylation of other genes regulating seladin-1 expression cannot be excluded. Additional studies should be performed in order to determine whether the degree of methylation may account for the different levels of expression between normal and pathologic tissues detected in other human organs