Transgenic cyclin E triggers dysplasia and multiple pulmonary adenocarcinomas

Abstract

Cyclin E is a critical G(1)-S cell cycle regulator aberrantly expressed in bronchial premalignancy and lung cancer. Cyclin E expression negatively affects lung cancer prognosis. Its role in lung carcinogenesis was explored. Retroviral cyclin E transduction promoted pulmonary epithelial cell growth, and small interfering RNA targeting of cyclin E repressed this growth. Murine transgenic lines were engineered to mimic aberrant cyclin E expression in the lung. Wild-type and proteasome degradation-resistant human cyclin E transgenic lines were independently driven by the human surfactant C (SP-C) promoter. Chromosome instability (CIN), pulmonary dysplasia, sonic hedgehog (Shh) pathway activation, adenocarcinomas, and metastases occurred. Notably, high expression of degradation-resistant cyclin E frequently caused dysplasia and multiple lung adenocarcinomas. Thus, recapitulation of aberrant cyclin E expression as seen in human premalignant and malignant lung lesions reproduces in the mouse frequent features of lung carcinogenesis, including CIN, Shh pathway activation, dysplasia, single or multiple lung cancers, or presence of metastases. This article reports unique mouse lung cancer models that replicate many carcinogenic changes found in patients. These models provide insights into the carcinogenesis process and implicate cyclin E as a therapeutic target in the lung.

Fig. 1.

Cyclin E effects on growth of pulmonary epithelial cells. (A) BEAS-2B cells were independently transiently transfected with cyclin E targeting or control siRNAs. (Left) Cyclin E protein expression was examined by immunoblot analyses 96 h after siRNA transfection. Actin expression served as a loading control. (Right) Viable cell numbers 96 h after transfection. Mean values and SDs from three independent experiments are shown. Significant difference was observed between control and cyclin E targeting siRNAs (P < 0.0001). (B) C10 cells were independently transduced with WT cyclin E, degradation-resistant cyclin E or an insertless control vector. (Left) Exogenous cyclin E expression was examined by immunoblot analysis. The degradation-resistant cyclin E species was stabilized. Actin expression served as a loading control. (Right) Anchorage-independent colonies ≥75 μm in diameter were scored at 2 weeks.

Fig. 2.

Generation of human SP-C (hSP-C) driven WT cyclin E or degradation-resistant (T62A/T380A) hcyclin E transgenic lines. (A) The cyclin E transgenic constructs. Restriction endonuclease sites used for cloning these species are shown. (B) Exogenous hcyclin E immunoblot expression profiles in normal lung tissues of transgenic mice. Actin expression served as a loading control. (C) Immunohistochemical detection of hcyclin E in lung tissues of a representative WT (line 2) transgenic cyclin E mouse and a nontransgenic (Tg⁻) control mouse that did not stain. The left and right arrows in Right indicate hcyclin E nuclear staining in representative pneumocyte and bronchiole epithelial cells, respectively.

Fig. 3.

Pathology and immunohistochemical expression of hcyclin E in transgenic lines. (A) Lungs from representative nontransgenic (Tg⁻) control (Left), WT cyclin E (line 2, Center), and degradation-resistant cyclin E (line 4, Right) transgenic mice are shown. Independent lung tumors are indicated by arrows. (B) Hematoxylin/eosin staining of lung tissue sections. Histopathologically normal lung tissue (Left) from a representative Tg⁻ control mouse and dysplasia (Center) and adenocarcinoma (Right) from a representative degradation-resistant transgenic cyclin E line 4 mouse are shown. (C) Immunohistochemical detection of hcyclin E in dysplastic and malignant lung lesions from the same degradation-resistant cyclin E transgenic mouse in B. Arrow indicates hcyclin E nuclear staining in a cell from a dysplastic lesion. (D) Immunohistochemical detection of Ki-67 in a representative Tg⁻ control mouse and the same lung lesions from the cyclin E line in B. (E) A representative metastasis (arrow depicts thymic metastasis) present in WT cyclin E transgenic line 2 is shown.

Fig. 4.

Activation of the Shh pathway in cyclin E transgenic mice and comparison with human adenomatous hyperplasia. (A) Shh, Gli1, and actin immunoblot expression profiles in lung tissues from Tg⁻ (control), WT transgenic cyclin E (data not shown), and degradation-resistant transgenic cyclin E mice. Two representative age- and sex-matched groups are shown. Lanes 1 and 7, normal (N) lung tissues from Tg⁻ control mice; lanes 2 and 8, normal (N) lung tissues from transgenic cyclin E line 4 mice; lanes 3 and 4, a pair (depicted by line) of normal (N) lung and malignant (T) lung tissues of a representative transgenic cyclin E line 4 mouse. Comparisons appear for paired specimens from other transgenic cyclin E line 4 mice in lanes 5 and 6 and in lanes 9 and 10. Shh and Gli1 immunoblot expression was observed in malignant lung tissues from WT transgenic cyclin E lines (data not shown). Actin expression served as a loading control. (B) Gli1 findings in B were independently confirmed by a Gli1 immunohistochemical assay on tissues from WT and degradation-resistant (data not shown) cyclin E transgenic lines. Increased Gli1 expression (relative to normal lung) appears in dysplasia and lung adenocarcinoma from the depicted cyclin E transgenic line 2. (C) Immunohistochemical detection of cyclin E (Left) and Gli1 (Right) in a human adenomatous hyperplasia lesion.