IKK[alpha] is a 85KD serine/threonine protein kinase and a subunit of the IKK complex, which contains IKK[alpha], IKK[beta], and IKK[gamma]. IKK[alpha] and IKK[beta] are highly conserved and they contain three functional domains of kinase domain, leucine zipper (LZ), and helix-loop-helix (HLH). Although IKK[alpha] and IKK[beta] can phosphorylate IκB proteins in vitro, IKK[alpha] and IKK[beta] have distinct physiological functions during mouse development. Genetic studies showed that IKK[alpha] is essential for embryonic skin development in mice. Mice deficient in IKK[alpha] display a hyperplastic epidermis that lacks terminal differentiation, resulting a death soon after birth because of the severely impaired skin. Recently, we reported a reduction in IKK[alpha] expression and identified somatic Ikk[alpha] mutations in a high proportion of poorly differentiated human squamous cell carcinomas (SCCs) (Liu et al., 2006). The aim of this study is to investigate the novel role of IKK[alpha] in skin carcinogenesis. We firstly examined IKK[alpha] expression and Ikk[alpha] mutations in human SCCs and found a reduction of IKK[alpha] in poorly differentiated human SCCs and identified somatic Ikk[alpha] mutations in exon 15 of Ikk[alpha] in human SCCs. We then examined the susceptibility of Ikk[alpha] hemizygotes to chemical carcinogeninduced skin carcinogenesis. In this chemical carcinogen-induced skin carcinogenesis setting, 7,12-dimethylbenz[a]anthracene (DMBA) induces Ras mutations and 12-Otetradecanoyl-phorbol-13-acetate (TPA) promotes Ras-initiated cell proliferation. We found two times more papillomas and eleven times more carcinomas in Ikk[alpha superscript +/-] mice than in Ikk [alpha] superscript +/+] mice induced by DMBA/TPA. Ikk[alpha superscript +/-] mice developed larger and earlier tumors than did Ikk[alpha superscript +/+] mice. Poorly differentiated carcinomas expressed low levels of IKK[alpha]. Ninety five percent of the Ikk[alpha superscript +/-] carcinomas and 44% of the Ikk[alpha superscript +/-] papillomas lost the remaining wild type Ikk[alpha] allele. This result indicates that the remaining one wild type Ikk[alpha] allele is important for preventing malignant carcinoma conversion. Also Ikk[alpha] mutations were detected in these skin tumors. Reduced IKK[alpha] was found to enhance TPA-induced mitogenic and angiogenic activities in mouse skin. Taken together, these results suggest that reduction of IKK[alpha] expression provides a selective growth advantage, which cooperates with DMBA-initiated Ras activity to promote skin carcinogenesis. In addition, we observed a small group of FVB female Ikk [alpha superscript +/-] mice for 1.5 years and found that 12/ 24 mice developed various spontaneous tumors including mammary gland carcinomas, uterine and ovary tumors, and dermal fibrosacomas. Somatic Ikk[alpha] mutations, elevated IKK/ NF[subscript -k]B and extracellular signal-regulated kinases (ERK) activities and elevated cyclin D1 levels were detected in these spontaneous tumors. These results suggest that these molecular alterations may contribute to the development of these tumors although the precise role of the down-regulation of IKK in the development of the tumors remains to be determined. Overall, our data and other published results suggest that IKK[alpha] is a new tumor suppressor in men and mice. / text

Mutation or inactivation of RB occurs in most human tumors and results in the deregulation of several E2F family transcription factors. Among the E2F family, E2F3 has been implicated as a key regulator of cell proliferation and E2f3 gene amplification and overexpression is detected in some human tumors. To study the role of E2F3a in tumor development, we established a transgenic mouse model expressing E2F3a in a number of epithelial tissues via a keratin 5 (K5) promoter. Transgenic expression of E2F3a leads to hyperproliferation, hyperplasia and increased levels of p53-independent apoptosis in transgenic epidermis. Consistent with data from human cancers, the E2f3a transgene is found to have a weak oncogenic activity on its own and to enhance the response to a skin carcinogenesis protocol. While E2F3a induces apoptosis in the absence of p53, the inactivation of both p53 and p73, but not p73 alone, significantly impairs apoptosis induced by E2F3a. This suggests that both p53 and p73 contribute to E2F3a induced apoptosis but that their function is compensatory. Even though data suggest that E2F3a carries out its unique apoptotic activity in part through another E2F family member E2F1, unlike E2F1, the ARF tumor suppressor is required for E2F3a-induced apoptosis. While both E2F3a and E2F1 require ATM for apoptosis, E2F3a activates ATM through a distinct mechanism from E2F1. The overexpression of E2F3a results in the accumulation of DNA damage in K5 transgenic keratinocytes and normal human fibroblasts (NHFs). In response to this, the DNA damage checkpoint kinase ATM is activated, and phosphorylation of the downstream targets p53 and the histone variant H2AX are significantly increased. Additional studies show that increased Cdk activity and aberrant DNA replication contributes to DNA damage, ATM activation and apoptosis in response to deregulated E2F3a, which suggest that aberrant replication imposed by deregulated E2F3a plays an important role in the activation of the ATM DNA damage response pathway. Activation of ATM by E2F3a is not affected by loss of ARF or E2F1. Meanwhile, E2F3a-induced ARF upregulation is not affected by E2F1 loss. The above results indicate that E2F3a engages several parallel pathways involving E2F1, ARF and the ATM kinase, and these pathways cooperate to promote apoptosis.