Cop1 constitutively regulates c-Jun protein stability and functions as a tumor suppressor in mice

Abstract

Biochemical studies have suggested conflicting roles for the E3 ubiquitin ligase constitutive photomorphogenesis protein 1 (Cop1; also known as Rfwd2) in tumorigenesis, providing evidence for both the oncoprotein c-Jun and the tumor suppressor p53 as its targets. Here we present what we believe to be the first in vivo investigation of the role of Cop1 in cancer etiology. Using an innovative genetic approach to generate an allelic series of Cop1, we found that Cop1 hypomorphic mice spontaneously developed malignancy at a high frequency in the first year of life and were highly susceptible to radiation-induced lymphomagenesis. Further analysis revealed that c-Jun was a key physiological target for Cop1 and that Cop1 constitutively kept c-Jun at low levels in vivo and thereby modulated c-Jun/AP-1 transcriptional activity. Importantly, Cop1 deficiency stimulated cell proliferation in a c-Jun-dependent manner. Focal deletions of COP1 were observed at significant frequency across several cancer types, and COP1 loss was determined to be one of the mechanisms leading to c-Jun upregulation in human cancer. We therefore conclude that Cop1 is a tumor suppressor that functions, at least in part, by antagonizing c-Jun oncogenic activity. In the absence of evidence for a genetic interaction between Cop1 and p53, our data strongly argue against the use of Cop1-inhibitory drugs for cancer therapy.

Figure 1. The gene trap approach developed for this study allows the generation of an allelic series of Cop1.

(A) The structure of the pGT01xf gene trap vector along with the structures of the wild-type, trapped (assimilated to a null allele), and hypomorphic alleles. The hypomorphic allele is generated by Cre-mediated conversion of the trapped allele. Exons are indicated by black boxes. En2 intr1, intron 1 of Engrailed2; βgeo, lacz/neomycin reporter/selection cassette. (B) A representative PCR analysis of E9.5 embryos with the indicated genotypes (left panel). The 3 primers used are described in Methods. Discrimination of the wild-type, null, and hypomorphic alleles by PCR using 2 different primer sets (right panel). (C) Wild-type (+/+ or +/–), homozygous trapped Cop1^–/– (–/–), and heterozygous embryos harboring one hypomorphic and one trapped allele (h/–) at E10.5 and E14.5, respectively. Control (+/–) mice and mice homozygous for the hypomorphic mutation at P21. Scale bars (left to right): 1 mm, 5 mm, and 3 cm. (D) Total lysates were prepared from E9.5 embryos with the indicated genotypes. Cop1 and vinculin were detected by Western blotting. The relative amount of Cop1 in the various protein lysates is indicated as percentage relative to the amount in wild-type mice (+/+; 100%).

Figure 2. Cop1 regulates c-Jun protein stability in vivo.

(A–C) Cop1 disruption causes a dramatic increase in c-Jun protein levels in vivo. (A) Total lysates were prepared from E9.5 embryos with the indicated Cop1 genotypes. c-Jun and vinculin (Vinc) were detected by Western blotting. (B) Immunohistochemistry for c-Jun on parasagittal sections from E9.5 Cop1^+/– and Cop1^–/– embryos. Scale bar (left panels): 1 mm. Original magnification (right panels), ×10. (C) Total lysates were prepared from MEFs, fetal liver cells, and heart of mice with the indicated Cop1 genotypes. c-Jun, Cop1, and vinculin were detected by Western blotting. (D) Determination of c-Jun half-life in early-passage MEFs. MEFs were metabolically pulse labeled using [³⁵S]methionine for 2 hours, and protein extracts were subsequently prepared at the indicated time points. Endogenous c-Jun was immunoprecipitated from these extracts and analyzed by Western blotting. The c-Jun signal was quantified and normalized to a stable and aspecific protein (*). (E) Cells with the indicated Cop1 genotypes were transduced with empty lentivirus (E) or lentivirus encoding full-length (FL) Cop1, a Cop1 deletion mutant lacking its WD40 domain (WD), and a Cop1 RING finger mutant (R). Samples were analyzed by Western blotting for protein expression as indicated. Vinculin served as loading control.

Figure 3. Cop1 modulates AP-1 activity in vivo.

(A–C) MEFs were exposed to 30 J/m² irradiation. (A) Comparison of AP-1–dependent luciferase activity in Cop1^+/+ and Cop1^hypo/– MEFs. The data represent the mean ± SD of 3 independent experiments. (B) Samples were analyzed by Western blotting. (C) Transcriptional analysis of c-Jun target genes by Q-RT-PCR analyses. The data were normalized to the level of expression in nontreated control, which is set to 1. The data represent the mean ± SD of 3 independent experiments. (D) Western blot analysis of early-passage MEFs. (E) Transcriptional analysis of c-Jun target genes by Q-RT-PCR analyses in p53-deficient MEFs. The data were normalized to the level of expression in control MEFs, which is set to 1. The data represent the mean ± SD of 2 independent experiments. (F) Comparison of AP-1–dependent luciferase activity in different organs from P6 Cop1^hypo/+ and Cop1^hypo/hypo mice. The data were first normalized to the total protein levels and then to the luciferase activity observed in Cop1^hypo/+ mice, which is set to 1. The data represent the mean ± SD of 4 independent biological replicates. (G) Comparison of AP-1–dependent luciferase activity in skin from P6 Cop1^hypo/+ and Cop1^hypo/hypo mice exposed to UV-B irradiation (750 mJ/cm²). The data were normalized first to the total protein levels and then to the luciferase activity observed in Cop1^hypo/+ mice, which is set to 1. The data represent the mean ± SD of 4 independent biological replicates. (H) Immunohistochemistry for phosphorylated (Ser63) c-Jun on skin sections from P6 Cop1^+/+ and Cop1^hypo/hypo mice. Mice were either nontreated (NT) or exposed to UV-B irradiation. Original magnification, ×20.

Figure 4. Cop1 deficiency stimulates cell proliferation in a c-Jun–dependent manner.

(A–H) All experiments described were performed with cells lacking p53. (A) Growth curve of Cop1^+/+ and Cop1^hypo/– MEFs; cells were seeded in triplicate at 2 × 10⁵ cells/60-mm dish. The numbers refer to the mean values ± SD of 2 independent MEF cultures. (B) Growth curves of MEFs infected with recombinant retroviruses (pBABE) encoding c-Jun or puromycin resistance only. Cells were seeded in triplicate at 2 × 10⁵ cells/60-mm dish. The numbers refer to the mean values ± SD of 2 independently infected MEF cultures. (C) Samples from B were analyzed by Western blotting. (D) Growth curves of cells expressing shRNAmir–c-Jun (shJun) or shRNAmir-scramble (Ctr). Cells were seeded in triplicate at 2 × 10⁵ cells/60-mm dish. The numbers refer to the mean values ± SD of 2 independently infected MEF cultures. (E) Colony assays were performed with cells expressing shRNAmir–c-Jun or shRNAmir-scramble (Ctr). Data are presented relative to the number of colonies obtained in the Cop1^hypo/– cultures infected with the control shRNA, which is set to 1. The numbers refer to the mean values ± SD of 2 independently infected MEF cultures. (F) Samples from D and E were analyzed by Western blotting. (G) Growth curve of Cop1^+/+ and Cop1^hypo/– MEFs exposed to 10 μM SP600125; cells were seeded in triplicate at 10⁵ cells/100-mm dish. The numbers refer to the mean values ± SD of 2 independent MEF cultures. (H) Colony assays using cells from A, B, and G.

Figure 5. Cop1^hypo/hypo mice are tumor prone.

(A–C) Spontaneous tumor formation in Cop1^hypo/hypo mice. (A) 50% of Cop1^hypo/hypo mice analyzed (n = 22) by histopathological examination spontaneously developed tumors in their first year of life. None of the Cop1^hypo/+ control littermates analyzed (n = 10) showed hyperplasia or tumor development during the course of the experiment (P = 0.036). (B) Most Cop1^hypo/hypo mice developed T cell lymphoma. Thymi from Cop1^hypo/+ and Cop1^hypo/hypo littermates at 20 weeks; the thymus of the Cop1^hypo/hypo mouse is tumorous. (C) Left panels: H&E staining of sections of organs from a Cop1^hypo/hypo mouse that developed lymphoma (lymphoid infiltrates are indicated by the arrows). Right panels: Immunohistochemistry for CD3 on selected organs from Cop1^hypo/+ and Cop1^hypo/hypo mice; a dramatic increase in CD3⁺ cells in Cop1^hypo/hypo tissues supports a T cell origin of the infiltrates (arrows). Original magnification, ×14. (D and E) Cop1 deficiency leads to increased infiltration of CD3⁺ cells and tumorigenesis following exposure to a single sublethal dose (4 Gy) of ionizing radiation. (D) Survival following 4 Gy of whole body irradiation decreased in the Cop1^hypo/hypo group (n = 28) at 14–32 weeks following irradiation in comparison with the group of control Cop1^hypo/+ littermates (n = 26) (Kaplan-Meier log-rank analysis, P = 0.0002). (E) H&E staining of sections of several organs from a Cop1^hypo/+ control and a Cop1^hypo/hypo mouse that had developed lymphoma. Immunohistochemistry on several organs from Cop1^hypo/hypo mouse shows a dramatic increase in the number of CD3⁺ cells, suggesting a T cell origin of the infiltrates (see arrows). Original magnification, ×14. (F) Immunohistochemistry on several organs shows high levels of phosphorylated (Ser63) c-Jun in the infiltrates of Cop1^hypo/hypo mouse. Original magnification, ×40.

Figure 6. Inverse correlation between COP1 and c-JUN protein levels in human prostate cancers.

(A) COP1 gene-centric expression analysis as revealed by Oncomine. COP1 is significantly underexpressed in a subset of prostate cancers relative to normal prostate tissue (P = 0.001). The y axis represents normalized expression units. Box-and-whisker plots indicate median values of the data and 25th and 75th percentiles of the data sets. Whiskers indicate minimum and maximum data values that are not outliers. P value was calculated by using the Student’s t test. The number of samples (n) in each class is indicated. (B) Western blot analysis of COP1 and c-JUN in several prostate cancer cell lines. COP1 expression is lost in PC3 cells. γ-Tubulin (γ-TUB) served as loading control. (C) Detection by Western blotting of c-JUN and COP1 in PC3 and LNCaP cells treated with the proteasome inhibitor MG-132 for 2, 4, or 6 hours before lysis. γ-Tubulin served as loading control. (D) PC3 cells were transfected with FLAG-tagged expression vectors for COP1. COP1 expression in LNCaP cells was knocked down by siRNA. Western blot analysis confirmed ectopic and knockdown COP1 expression. An inverse correlation between COP1 and c-JUN protein levels was observed. γ-Tubulin served as loading control. (E) Growth curve of PC3 cells transfected with FLAG-tagged empty and COP1 expression vectors; cells were seeded in triplicate at 1.5 × 10⁵ cells/60-mm dish. The numbers refer to the mean values ± SD of 3 independent transfection experiments.