E2F is a collective term for a family of transcription factors (E2F-1 through 6 and DP 1 and 2) which activate genes required for DNA synthesis, cell cycle progression and cell proliferation. The activity of the E2F family members is regulated by binding of the pocket proteins pRb, pi07 and pi30, which are in turn regulated by cyclin dependent kinases. Deregulation of this pathway, and presumably deregulation of growth, has been noted in many tumour types. However, it has not been established whether deregulation of this pathway has an impact on E2F activity or tumour formation. Therefore, the aim of this study was to examine the regulation of E2F in normal keratinocytes and to try and determine if E2F is deregulated in skin carcinogenesis.
Skin carcinogenesis involves changes in both proliferation and differentiation pathways. Little is known about the normal signals that induce differentiation in keratinocytes, nor what has caused the deregulation of these pathways in skin tumours. Normal skin differentiation is characterised by growth arrest of cells and an associated decrease in expression of proliferation-associated genes such as E2F-1 and cdk-1. This is followed by the induction of epithelial-specific and differentiation-specific genes such as transglutaminse type I and cornifin. These markers of the induction of epithelial differentiation were used to study the role of E2F in skin carcinogenesis.
To explore the role of E2F in keratinocyte differentiation and skin carcinogenesis, four lines of investigation were pursued. 1) What is the role of different E2F family members in normal keratinocyte differentiation? 2) Which E2F family members are deregulated in squamous cell carcinomas? 3) At what stage of skin carcinogenesis does E2F deregulation occur? 4) What is the effect of deliberate manipulation of E2F in normal keratinocytes and skin carcinomas?
1. What is the role of the different E2F family members in normal keratinocyte differentiation?
Using RT-PCR, the expression of E2F family members was examined in both proliferative and differentiated keratinocytes. E2F DNA binding activity and E2F transcriptional activity were also examined. A decrease in E2F-1 mRNA, E2F DNA binding ability and E2F fimction all correspond to the induction of differentiation in normal keratinocytes. This demonstrated a strong correlation between E2F-1 mRNA expression and keratinocyte differentiation, whilst no similar association was found for any other E2F family member.
2. Which E2F family members or pocket proteins are deregulated in skin carcinogenesis?
In contrast to normal keratinocytes, the SCC 25 squamous carcinoma cell line is resistant to differentiation-inducing stimuli. In these cells, E2F-1 mRNA expression, E2F DNA binding activity and E2F transcriptional activity were not decreased by differentiation-inducing agents. No correlation was found between the deregulation of differentiation in the SCC 25 cell line and mRNA expression of any other E2F family members. These data led to the hypothesis that deregulation of E2F activity may contribute to squamous carcinoma formation.
3. At what stage of skin carcinogenesis does E2F deregulation occur?
A cell line from a preneoplastic intraepidermal carcinoma was established and its response to differentiation-inducing stimuli characterised. These cells undergo a growth arrest and express differentiation markers when treated with differentiation-inducing stimuli. However, the differentiation process is incomplete, as demonstrated by hyperplastic growth in organotypic raft cultures. Furthermore, the cells have an extended lifespan in-vitro and can undergo a reversible growth arrest. When induced to differentiate, these cells demonstrated a marked decrease in E2F-1 mRNA expression, similar to that seen in normal keratinocytes. These data suggest that deregulation of E2F-1 may be a later step in skin carcinogenesis. A second cell line derived from a basal cell carcinoma was characterised. These cells overexpress the Ptch gene, suggesting that they originated from the tumour, yet they express an intermediate filament profile more typical of fibroblasts. These cells did not express epithelial differentiation markers when treated with differentiation-inducing stimuli and were stimulated to grow in the presence of IFNy. Expression of E2F-1 mRNA did not correlate with the proliferation rate of these cells. A cell line that overexpresses the Ptch gene has not been previously reported.
4. What is the effect of deliberate manipulation of E2F in normal keratinocytes and skin carcinomas?
To address this question, transgenic mice overexpressing E2F-1 from keratinocyte specific promoters were developed. Despite normal pupping rates a transgenic line could not be established. The effect of E2F-1 overexpression was then examined in cultured keratinocytes. Overexpression of E2F-1 in cells induced to differentiate prevented the induction of differentiation markers and permitted the continued expression from proliferation-associated promoter constructs. Conversely, the inhibition of E2F activity in keratinocytes inhibited proliferation-associated genes but did not induce differentiation. Furthermore, inhibiting E2F activity inhibited proliferationassociated genes in 3 of 5 epithelial cell lines. In conclusion, these data show a sfrong association between E2F deregulation and the latter stages of carcinoma development. Therefore, targeting E2F activity in squamous carcinoma may be of therapeutic benefit.