NICD1 Cooperates With P53 And Lkb1 Loss In A Mouse Model Of Squamous Cell Lung Cancer

Background and objective:Lung squamous cell carcinoma (LSCC) is one of the most lethal types of cancer and progresses in a stepwise fashion to invasive carcinoma. With limited treatment options, the 5-year survival rate of LSCC is only 15%. In the process of LSCC, it underwent normal epithelium, hyperplasi, squamous metaplasia, dysplasia, carcinoma in situ and invasive carcinoma. Given that we believe that stepwise progression is important in LSCC. There are four mouse models of squamous cell lung cancer. (1) LKB1/K-Ras:tumor suppressor LKB1 inactivation and K-Ras activation. (2) IKKa kinase-dead knock-in:decreased IKKa. (3) Sox2/LKB1:tumor suppressor deletions. (4) LKB1/Pten:tumor suppressor deletions. In addition, cytokeratin 5 (K5) cells are regarded as potential stem cells in lung and may be associated with lung cancers. We want to use Cre-loxP system to generate a premalignancy model since understanding the progression of LSCC is essential for early diagnosis, treatment, and prevention. Furthermore, we want to study whether K5 specific cells have an ability of cancer stem cells and contribute to the tumor progression. At last, we want to compare the differences between the K5 cell-derived lung lesion and the non-specific cell-derived lung lesion.Methods:Cancer Genome Atlas Research Network (TCGA,2012) recently sequenced 178 human SCCs, identifying numerous genetic alterations that may serve as valuable therapeutic targets. We used classical transgenic mice to manipulate genes including NICD1, P53 and Lkbl, which are commonly found in LSCC based on previous data from the TCGA as well as recent data from in vitro studies on mouse airway basal stem cells (ABSCs) in air liquid interface (ALI) culture in our lab (unpublished data). We use Ru486 to induce K5 specific cells or Adenovirus Cre to induce all non-specific cells to activate NICD1 and delete the tumor suppressor P53 and Lkbl.Results:Based on similar manipulations in vivo, we obtained LSCC tumors at various stages of premalignancy. These phenotypes were visualized in lung sections by light and confocal microscopy of H&E staining as well as immunofluorescent staining for markers of LSCC, including p63, K5 and K14 and confirmed by a pathologist.(1) Lung squamous cell carcinomas were observed in Ru486 treated mice, suggesting that K5-derived cells might be lung cancer stem cell. In all mice: ①The phenotypes of NICD1 wild-type mice were normal. ②All NICD1 homozygous mutant mice died early after birth or before they were born. ③For NICD1 heterozygous mutant mice, the phenotypes of mice varied significantly according to the different P53 and Lkb1 genotypes:The phenotypes of NICD1+fl; P53fl/fl, NICD1+/fl; Lkb1fl/fl and NICD1+/fl; P53+/fl; Lkb1fl/fl are normal. Lung squamous cell carcinomas were observed in NICD1+/fl; P53+/fl; LKB1/fl、NICD1+/fl; P53fl/fl; LKB1+/fl and NICD1+/fl; P53fl/fl; LKB1fl/fl mice. There was 1,2 and 2 tumor mice in each group. The incidence was 25.0%,28.6% and 33.3%, respectively. The latency was 5 to 12 months. All lung tumors found in H&E staining were further immunofluorescence stained, expressing high lung squamous cell marker K5, whereas lung squamous cell markers TTF1 and cytokeratin 8 (K8) were lost. All these confirmed the spontaneous lung tumors were squamous cell lung carcinomas.(2) Lung squamous cell carcinomas were observed in AdenoCre treated mice, suggesting that K5-derived cells might be lung cancer stem cell. In all mice: ①The phenotypes of NICD1 wild-type mice were normal. ②All NICD1 homozygous mutant mice died early after birth or before they were born. ③For NICD1 heterozygous mutant mice, the phenotypes of mice varied significantly according to the different P53 and Lkbl genotypes:The phenotypes of NICD1+/fl; P53fl/fl、NICD1+/fl; Lkblfl/fl and NICD1+/fl; P53+/fl; Lkblfl/fl are normal. Lung squamous cell carcinomas were observed in NICD1+/fl; P53+/fl; LKB1+/fl, NICD1+/fl; P53fl/fl; LKB1+/fl and NICDl+/fl; PSfl/fl; LKB1fl/fl mice. There was 1 mouse in each group. The incidence was 100%. The latency was 5 to 9 months. All lung tumors found in H&E staining were further immunofluorescence stained, expressing high lung squamous cell marker K5, whereas lung squamous cell markers TTF1, ProSPC and cytokeratin 8 (K8) were lost. All these confirmed the spontaneous lung tumors were squamous cell lung carcinomas.Conclusion:Although some LSCC in vivo mouse models exist, these are histologically not very relevant to the human disease and take a long time to form tumors. Moreover, no model currently shows the formation of premalignant lesions with stepwise progression to LSCC, hence our study is applicable to human LSCC and will further our understanding of the biology of LSCC.