Skip Navigation


Carcinogenesis Advance Access originally published online on February 6, 2008
Carcinogenesis 2008 29(4):858-865; doi:10.1093/carcin/bgn021
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/4/858    most recent
bgn021v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by He, H.
Right arrow Articles by Sun, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by He, H.
Right arrow Articles by Sun, Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

SAG/ROC2/RBX2 E3 ligase promotes UVB-induced skin hyperplasia, but not skin tumors, by simultaneously targeting c-Jun/AP-1 and p27

Hongbin He{dagger}, Qingyang Gu{dagger}, Min Zheng, Daniel Normolle and Yi Sun*

Department of Radiation Oncology, University of Michigan Comprehensive Cancer Center, 1500 East Medical Center Drive, Ann Arbor, MI 48109, USA

* To whom correspondence should be addressed. Tel: +1 734 615 1989; Fax: +1 734 647 9654; Email: sunyi{at}umich.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 
Sensitive to apoptosis gene (SAG)/regulator of cullins-2/RING box protein 2 is a stress-responsive RING component of Skp-1/Cullins/F-box protein E3 ubiquitin ligase. When overexpressed, SAG inhibits apoptosis induced by reactive oxygen species or hypoxia. Here, we report that SAG overexpression inhibits ultraviolet (UV) B-induced apoptosis in mouse JB6 epidermal cells. Using a transgenic mouse model, in which SAG expression was targeted primarily to epidermis by a K14 promoter, we showed that, at the early stage of UVB skin carcinogenesis (10 weeks post-UVB exposure), c-Jun, p27, p53, c-Fos and cyclin D1 were strongly induced. While having no effect on UVB-induced p53, c-Fos and cyclin D1, SAG-transgenic expression reduced the levels of c-Jun and p27 and inhibited AP-1 activity. The net outcome of SAG-mediated inhibition of c-Jun/AP-1 (pro-tumor promotion) and of p27 (antiproliferation) increased skin hyperplasia, with no apparent effect on apoptosis, as evidenced by increased skin thickness, and increased rate of DNA synthesis, but hardly any apoptosis. Although skin hyperplasia was promoted, SAG-transgenic expression had no significant effect on tumor formation in the later stage of UVB carcinogenesis. Thus, by simultaneously targeting c-Jun and p27, SAG accelerates UVB-induced skin hyperplasia, but not carcinogenesis.

Abbreviations: BrdU, 5-bromo-2'-deoxy-uridine; PFA, paraformaldehyde; ROC/RBX, Regulator of Cullins/RING box protein; SAG, sensitive to apoptosis gene; SCF, Skp1-Cullins-F-box protein; SiRNA, small interfering RNA; TUNEL, terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling; UV, ultraviolet


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Funding
 References
 
Carcinogenesis, caused by physical (e.g. radiation), chemical or viral (e.g. oncogenic) mechanisms, is a multistage process of coordinated acquisition of favorable genetic lesions and complex interactions between tumor and host tissues, which ultimately leads to an aggressive metastatic phenotype (1). Ultraviolet (UV) light is considered a complete carcinogen, with characteristics of both tumor initiators and promoters. It acts as an initiator by producing irreversible mutagenic damage such as formation of pyrimidine dimers in DNA and acts as a promoter by inducing epigenetic changes that cause expansion of the initiated cell population. Indeed, chronic exposure to UV radiation from sunlight is the primary cause of non-melanoma and melanoma skin cancer in humans (2), which are associated with >1 million cases diagnosed yearly in USA alone (3).

Sensitive to apoptosis gene (SAG) was initially cloned as a redox-inducible gene that encodes an evolutionarily conserved RING finger protein (4), and was later found to be the second family member of regulator of cullins-1/RING box protein 1 (4–8), the RING component of Skp1, cullin1 and F-box protein (SCF) E3 ubiquitin ligases, which promote ubiquitination and degradation of a variety of protein substrates (9). SAG has been shown previously to regulate apoptosis and cell proliferation. When overexpressed, SAG protects cancer cells from apoptosis induced by redox agents by inhibiting cytochrome c release and caspase activation in a RING domain-dependent manner (4,10) and attenuates injury or apoptosis induced by ischemia and reperfusion in in vivo mouse brain (11) and in rat cardomyocytes (12). Importantly, SAG is overexpressed in a subset of colon carcinoma tissues and in non-small-cell lung carcinomas (13,14). SAG overexpression promotes cell growth under serum starved conditions by inhibiting p27 accumulation, whereas SAG downregulation inhibits cancer cell growth (13,15). Thus, SAG may regulate carcinogenesis via modulating both cell proliferation and apoptosis.

In this report, we focus on the role of SAG in apoptosis protection against UVB exposure and in UVB-induced skin hyperplasia and carcinogenesis. We show that SAG expression inhibits UVB-induced apoptosis in a mouse JB6 epidermal cell model. In an in vivo transgenic model, where SAG is targeted to express primarily in epidermis, UVB-induced levels of c-Jun and activity of AP-1 is reduced along with reduction of p27 levels. The net outcome of these SAG-mediated changes is increased skin hyperplasia, but not tumor formation. Thus, SAG increases UVB-induced skin hyperplasia, but not carcinogenesis, by simultaneously targeting c-Jun/AP-1 and p27, which have opposite effects on tumor promotion and cell proliferation.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Funding
 References
 
Cell culture, epidermal primary culture, UV exposure and apoptosis assays
Mouse JB6-Cl.41 epidermal cells were cultured in 5% fetal bovine serum-containing Dulbecco’s modified Eagle’s medium. Cells were infected with recombinant SAG-expressing lentivirus construct (LT-SAG), along with the empty vector control (LT-vec), as described (16). Primary murine skin keratinocytes were isolated from the epidermis of 1-day-old mice and cultured as described (17). The UVB exposure was conducted under Ultralite Panelite lamps consisting of six FS36T12 UVB-HO bulbs (54% UVB, 31% UVA2, 15% UVA1) with Kodacel membrane filter to remove wavelengths <290 nm (18). The exposure periods were calculated using a UVX radiometer (Ultra-Violet Products, San Gabriel, CA). Assays for UVB-induced apoptosis in JB6-Cl.41 cells include trypan blue exclusion, fluorescent activated cell sorting (FACS) analysis and caspase-3 activation, as described (19).

Western blotting analyses
Cell lysates were prepared after UVB exposure and subjected to western blotting analysis (4) using antibodies against c-Jun, histone H2A, cyclin D1 (Santa Cruz Biotechnology, Santa Cruz, CA), p27 (BD Transduction Laboratories, San Jose, CA), p53 (Novocastra, Bannockburn, IL), β-actin (Sigma, St Louis, MO) and SAG.

SAG transgenic and SAG-AP-1-luc transgenic mice and in vivo AP-1 activity assay
FVB/N mice transgenic for human SAG under the control of the human keratin 14 promoter (SAG-Tg mice) and SAG-AP-1-luc mice, derived from crossing between SAG-Tg and AP-1-luc SKH mice (20) and their genotyping, have been described previously (24). A total of six AP-1-Luc(+)/SAG-Tg(–) mice and thirteen AP-1-Luc(+)/SAG-Tg(+) mice were anesthetized and 3 x 1.5 mm ear punches were collected from the right ears before UVB exposure and from the left ears after receiving a 10 kJ/m2 dose of UVB at 24 h post-UVB exposure. Tissues were ground with a microtube pestle and lysed in 1x passive lysis buffer for 1 h and followed by luciferase activity measurement (24).

UV exposure and gel retardation assay for AP-1 binding
SAG-Tg(–) and SAG-Tg(+) mice, at an age of 8 weeks, were exposed to UVB three times per week for 10 weeks, at a dose of 2.5 kJ/m2 for the first 4 weeks, 5 kJ/m2 for weeks 5–8 and 10 kJ/m2 for the remaining 2 weeks (21). The age-paired control mice received no UVB. Twenty-four hours after the last UVB exposure, the mouse skin tissues were subjected to analyses of immunohistochemistry, skin thickness measurement, 5-bromo-2'-deoxy-uridine (BrdU) and terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL), whereas the epidermal cells were scraped off and subjected to nuclear extract preparation and a gel retardation assay (16,24). The oligonucleotide sequence for AP-1-binding site is AP-1-01: 5'-CGCTTGATGAGTCAGCCGGAA-3' and for p53 is p21-p01: 5'-GAACATGTCCCAACATGTTG-3'.

In vivo ubiquitination
Human lung H1299 cells were transiently transfected with His-Ub, p27 alone or in combination with psiSAG or psiCont. Cells were harvested 48 h after transfection and split into two aliquots with one for direct western blotting analysis and the other for in vivo ubiqutination assay, as described (16,22). Briefly, cell pellets were lysed in buffer A (6 M guanidinium–HCl, 0.1 M Na2HPO4/NaH2PO4, 10 mM Tris–HCl, pH 8.0, 10 mM β-mercaptoethanol) and incubated with Ni-NTA beads (Qiagen, Valencia, CA) at room temperature for 4 h. Beads were washed once with each of buffer A, buffer B (8 M urea, 0.1 M Na2HPO4/NaH2PO4, 10 mM Tris–HCl, pH 8.0, 10 mM β-mercaptoethanol) and buffer C (8 M urea, 0.1 M Na2HPO4/NaH2PO4, 10 mM Tris–HCl, pH 6.3, 10 mM β-mercaptoethanol). Proteins were eluted from beads with buffer D (200 mM imidazole, 0.15 M Tris–HCl, pH 6.7, 30% glycerol, 0.72 M β-mercaptoethanol and 5% sodium dodecyl sulfate). The eluted proteins were analyzed by western blotting for polyubiquitination of p27 with anti-p27 antibody.

Immunohistochemistry
After 10 weeks of UVB exposure, the skin tissue was excised and fixed in 4% paraformaldehyde (PFA)–phosphate-buffered saline for 3 days at 4°C, and embedded in paraffin. Four micrometer thick sections were cut for immunostaining with an ABC kit (Vector Labs, Burlingame, CA). The primary antibodies used were c-Jun, p27, c-Fos, p53 and cyclin D1. Negative controls were included for each study by using normal goat serum to replace primary antibodies. The papilloma to carcinoma conversion was determined through histopathological observation after hematoxylin and eosin staining.

Measurement of skin thickness
Skin hyperplasia after 10 weeks of UVB exposure was determined by measurement of the epidermal thickness after hematoxylin and eosin staining. A total of 16 measurements were made for each mouse in a systematic manner under a microscope, starting from the top of the basement membrane to the bottom of the strateum corneum. The skin thickness from each mouse was then averaged within the group, expressed as millimeter after x400 magnification (mm/400) and subjected to Student’s t-test to determine the statistical significance of differences between the two lines.

BrdU incorporation assay
SAG-Tg(+) and SAG-Tg(–) mice were injected intraperitoneal with undiluted BrdU-labeling reagent, 2 ml/100 g (body weight), 2 h prior to killing, but 22 h after last UVB exposure of 10 weeks. The skin tissue was collected and fixed in 4% PFA–phosphate-buffered saline. The assay was performed with BrdU Labeling and Detection Kit II (Roche, Indianapolis, IN). Slides were developed with nitroblue tetrazolium/5-bromo-4-chloro-3'-indolyphosphate- p-toluidine and counterstained with Eosin (Richard-Allan Scientific, Kalamazoo, MI). For quantification of BrdU-positive cells, two random areas were selected from each mouse and the number of positive cells out of a total of 1000 epidermal cells was counted. Student’s t-test was performed to determine the statistical significance of differences between the two groups.

TUNEL assay
Skin tissue post-UVB exposure was excised and fixed in 4% PFA–phosphate-buffered saline and embedded in paraffin. Four micrometer thick sections were cut and deparaffinized slides were used for the TUNEL assay using the In Situ Cell Death Detection Kit (Roche). The slides were stained with TUNEL reaction mixture, along with label solution for the negative control, and counterstained with propidium iodide counterstaining solution. The slides were mounted and analyzed under a fluorescence microscope. For quantification of TUNEL-positive cells, two random areas were selected from each mouse skin. The number of positive labeling cells (cells in green), out of a total of 1000 cells, were counted in each area. Student’s t-test was performed to determine the statistical significance of differences between the two groups.

UVB exposure and skin carcinogenesis
Eight-week-old female K14-SAG mice (n = 17) and their negative littermates (n = 16) were irradiated on their shaved backs with a bank of Q-Panel Ultraviolet B-313 EL/12 sunlamps (Q-Panel Lab Products, Cleveland, OH) with doses of UVB increasing from 2.5 to 10 kJ/m2 for a total of 29 weeks (three times per week with 2.5 kJ/m2 for 4 week, with 5 kJ/m2 for 4 week and then with 10 kJ/m2 for the remaining weeks) (21). UVB exposure periods were calculated using a UVB radiometer (Ultra-Violet Products). Tumor development (at both ears and the back of trunk) was recorded weekly throughout 29 weeks. Mice were killed once their tumors reached 7 x 7 mm in size, and tumor samples were fixed in PFA, embedded in parablast and analyzed (24).

Statistical analysis of tumor count and growth experiments:
Data were analyzed using SAS v9.1 (SAS Institute). All hypotheses were tested at the 0.05 significance level. Tumor counts were compared using Poisson regression. The product-limit (Kaplan–Meier) estimates of the time to tumor formation functions of the SAG-Tg(+) and SAG-Tg(–) mice were compared by the log-rank test. Tumor growth rates were compared assuming an exponential growth model; the log-transformed volumes of each tumor were regressed on week of measurement using a linear random effects model that accounts for correlated within-animal and within-tumor measurements. The regression coefficients for week of measurement estimate the exponential growth rate parameters, which were compared between SAG-Tg(+) and SAG-Tg(–) mice by an F test.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Funding
 References
 
SAG overexpression inhibits UVB-induced apoptosis
It has been shown previously that UVB induced apoptosis in JB6-Cl.41 epidermal cells (23) and SAG protected cells from apoptosis induced by reactive oxygen species and hypoxia both in vitro and in vivo (4,11,12). We investigated whether SAG overexpression would inhibit apoptosis induced by UVB exposure. A lentivirus-based SAG-expressing construct was made (16) to infect JB6-Cl.41 cells, and a 2- to 3-fold increase of SAG levels was achieved (Figure 1A). This level of SAG overexpression, which is within physiological range (16), rendered epidermal cells significantly more resistant to UVB-induced apoptosis, as demonstrated by reduced dead cells by trypan blue exclusion assay (Figure 1B), reduced sub-G1 apoptotic population in fluorescent activated cell sorting analysis (Figure 1C), and reduced activation of caspase-3 activity (Figure 1D). Thus, SAG overexpression caused a partial protection of JB6-Cl.41 cells from UVB-induced apoptosis.


Figure 1
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. SAG overexpression protects cells from UVB-induced apoptosis: mouse JB6-Cl.41 cells were infected with the empty lentivirus vector (LT-vec) or lentivirus expressing Hemagglutinin A tagged SAG (LT-SAG). Cell lysates were prepared and subjected to immunoblotting analysis using anti-SAG antibody and β-actin as a loading control (A) Both control and SAG overexpressed cells were subjected to UV exposure (400 mJ/cm2) and apoptosis assayed 24 h after exposure by trypan blue exclusion (B) fluorescent activated cell sorting analysis (C) and caspase-3 activity (D).

 
SAG-transgenic expression inhibited UVB-induced AP-1 binding and AP-1 transactivation in mouse skin tissues
We next determined the potential role of SAG in UVB-induced skin hyperplasia, apoptosis and carcinogenesis, using an in vivo SAG-transgenic model, in which the SAG transgene was targeted to express primarily in skin epidermis by the K14 promoter in inbred FVB/N mice, a strain shown previously to be sensitive to UVB carcinogenesis (21). Two independent lines, 345 and 710, were characterized, with similar results in a 9,10-dimethylbenz(a)anthracene-12-0-tetradecanoylphorbol-13-acetate skin carcinogenesis protocol (24). The 710 line was, therefore, used in this study. We have recently found that SAG E3 ubiquitin ligase could, by promoting c-Jun ubiquitination and degradation, inhibit AP-1 activity in cultured cells (16). We determined if SAG transgene expression inhibited AP-1 activity, which is known to be induced by UVB and to play a promoting role in UVB-induced skin carcinogenesis (20). As shown in a gel retardation assay in Figure 2A, control mice, without exposure to UVB, did not have any AP-1 DNA-binding activity, regardless of SAG-transgenic expression (lanes 1–4, samples 1 and 2). However, 10 weeks of UVB exposure, at a frequency of three time a week, produced significant induction of AP-1 binding in three out of three SAG-Tg(–) lines tested (lanes 5–11, samples 3–5). Significantly, SAG-transgenic expression caused a remarkable reduction of UVB-induced AP-1 binding in three out of three SAG-Tg(+) mice (lanes 12–18, samples 6–8). Since AP-1 binding can be supershifted by c-Jun antibody (lanes 6, 8, 10, 13, 15 and 17), and be completely blocked by 50x cold oligonucleotide (lanes 11 and 18), it was concluded that the AP-1 complex contained c-Jun and that the binding was specific. Thus, SAG-transgenic expression inhibits AP-1 binding induced by UVB exposure.


Figure 2
View larger version (58K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Reduction of UVB-induced AP-1 activity by SAG-transgenic expression: nuclear extracts were prepared 24 h after last UVB exposure of a total of 10 weeks, three times a week from mouse skin tissues of three SAG-Tg(–) and three SAG-Tg(+) mice, and subjected to gel retardation assay for AP-1 and p53 specific binding (A and B) as well as western blotting of histone H2A for equal loading of nuclear proteins (C). The brackets in the figure (A) indicate that the nuclear extracts in these lanes were from the same sample without or with the addition of the antibody or cold competitive oligo, as indicated. (D) SAG-transgenic expression reduces UVB-induced AP-1 activity in vivo: SAG-Tg(+) mice were crossed with AP-1-Luc reporter mice and offsprings were genotyped. The mice that were heterozygous with AP-1-Luc-reporter and SAG-Tg(+), n = 13 or SAG-Tg(–), n = 6 were used. The samples were collected from right ears prior to UVB exposure and from left ears 24 h after single UVB exposure at 10 kJ/m2 and subjected to luciferase assay. Fold induction was calculated by setting the luciferase light unit from untreated ear as 1; **P < 0.01.

 
We next studied if UVB exposure would cause any differential induction of p53, which is an UV-inducible gene, but is not subjected to degradation by SAG E3 ubiquitin ligase. The same eight samples of nuclear extracts were subjected to a gel retardation assay using the p53 consensus binding sequence. It has been established previously that in the in vitro gel retardation assay, p53 hardly binds to its consensus sequence, unless a p53 antibody against its C-terminal region is present (25,26). We therefore included p53 antibody (pAb421) in some reactions to visualize the p53 binding. As shown in Figure 2B, little or no binding was found in non-exposed controls, regardless of SAG-transgenic expression (lanes 1–4), but UVB exposure caused a significant induction of p53 activity, which was not affected by SAG-transgenic expression (lanes 5–10 versus 12–17). Due to tight p53-DNA binding in the presence of p53 antibody, an excess of 50x cold oligonucleotide only partially blocked the binding (lanes 11 and 18). The similar levels of histone H2A among all eight samples, as shown in Figure 2C, indicated equal amounts of the nuclear protein were loaded from each sample. Thus, a short-term UVB exposure caused significant activation of AP-1 and p53, which is, remarkably, inhibited or not affected by SAG-transgenic expression, respectively.

To further determine whether reduced AP-1 binding can be reflected by reduced AP-1 transactivation in vivo, we crossed AP-1 luciferase reporter mice (20) with SAG-Tg(+) mice to generate mice that are heterozygous for the AP-1-Luc reporter in SAG-Tg(+) or SAG-Tg(–) background, respectively. The AP-1 activity, as quantified by luciferase light intensity, was measured in ear-punched tissues 24 h post-UVB exposure. AP-1 transactivation activity was induced up to 100-fold 24 h after exposure, which was significantly inhibited (up to 2.5-fold, P < 0.01) by SAG-transgenic expression (Figure 2D). Taken together, it is clearly demonstrated that SAG, upon transgenic expression, significantly inhibited UVB-induced AP-1 activities without affecting p53 activity.

SAG-transgenic expression reduced the levels of c-Jun and p27
We next determined the effect of SAG-transgenic expression on growth-regulatory proteins, c-Jun, p27 and cyclin D1, in response to UVB exposure, using primary keratinocytes isolated from 1-day-old postnatal pups from SAG-transgenic and non-transgenic littermates. The primary cells were prepared from two independent pups from each line. First, we confirmed SAG-transgenic expression, which was readily detectable in cells from transgenic lines, but not from controls (Figure 3A, top panel, labeled as FLAG-SAG). Second, we determined the effect of SAG-transgenic expression on c-Jun at both basal and UV-induced levels. As shown in Figure 3A (2nd panel), SAG-transgenic expression caused a 2- to 3-fold reduction of basal c-Jun levels in keratinocytes, probably due to targeted degradation, as described previously (16). UVB exposure induced a ~4-fold increase of c-Jun levels in both lines, but the total c-Jun levels after UV exposure were 2- to 3-fold higher in non-transgenic than in transgenic cells, apparently due to their lower basal c-Jun level. This result strongly suggests that the lower c-Jun levels are responsible for the observed decrease of AP-1 activity in SAG-transgenic skin tissues (Figure 2A and D).


Figure 3
View larger version (90K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Effect of SAG-transgenic expression on UVB-induced protein expression: (A) SAG expression in primary keratinocytes inhibits UVB-induced accumulation of c-Jun and p27: primary culture were prepared from SAG-Tg(+) and SAG-Tg(–) mice and were subjected to a single UVB exposure at indicated doses. Cells were harvested 24 h after exposure and subjected to western blotting analysis for SAG, c-Jun, p27, cyclin D1 and β-actin. The fold change was calculated after densitometry quantification with β-actin normalization, setting the unexposed control of SAG-Tg(–) mouse as 1. (B) SAG silencing inhibits p27 polyubiquitination: H1299 cells were transiently transfected with His-Ub (lane 2) or p27 (lane 1) alone or His-Ub and p27 in combination with psiSAG (lane 3) or psiCont (lane 4) and subjected to in vivo ubiquitination assay by Ni-bead purification of ubiquitinated p27, followed by detection of polyubiquitinated p27 by p27 antibody. (C) SAG-transgenic expression reduces staining of c-Jun and p27 in epidermal cells: mouse skin tissues after 10 weeks UVB exposure from SAG-Tg(–) and SAG-Tg(+) mice were subjected to immunohistochemistry staining using antibodies against c-Jun, p27, cyclin D1, p53 and c-Fos. Magnification x400.

 
Third, we determined the effect of SAG-transgenic expression on p27, a well-known, naturally occurring inhibitor of cyclin-dependent kinases, and a known substrate of SAG–SCF E3 ubiquitin ligase (9,15,27). Interestingly, p27 was also subjected to SAG-mediated degradation and UVB induction, with a pattern similar to c-Jun: lower basal and UVB-induced levels in the SAG-transgenic line and a similar magnitude of UVB induction between the two lines (Figure 3A, 3rd panel). On the other hand, cyclin D1, which is degraded by SCF-Fbx4/{alpha}B-crystallin (28), was not induced by single exposure of UVB nor significantly affected by SAG-transgenic expression (4th panel), suggesting a certain degree of SAG targeting specificity. Our results demonstrated that SAG-transgenic expression causes simultaneous reduction of c-Jun/AP-1, a tumor promoting transcription factor and of p27, a tumor suppressor.

We have shown previously that SAG small interfering RNA (siRNA) silencing inhibited c-Jun polyubiquitination (16). We determined whether this is the case for p27 as well. H1299 cells were transiently cotransfected with His-Ub and p27, alone or in combination of plasmid expressing control siRNA or SAG siRNA. Ubiquitinated p27 was purified by Ni-NTA beads and detected by antibody against p27. As shown in Figure 3B, p27 polyubiquitination can be readily detected in cells transfected with His-Ub, p27 and control siRNA (lane 4), indicating that endogenous SCF components are sufficient to promote p27 polyubiquitination upon p27 overexpression via transfection. When cotransfected with plasmid that silenced SAG expression, p27 polyubiquitination was remarkably inhibited (lane 3). The lack of complete inhibition of p27 polyubiquitination upon SAG silencing is due to the presence of ROC1/RBX1, a SAG family member known to promote p27 degradation (9). Thus, SAG contributes to p27 polyubiquitination.

We followed up this interesting observation with in situ immunohistochemical analysis. Expression of transgenic SAG in mouse skin in SAG-Tg mice was demonstrated previously (24). As shown in Figure 3C, neither c-Jun nor p27 was detectable in control, non-UVB-exposed epidermal layer of mouse skin (top panel, left two columns). UVB exposure (3 times per week for 10 weeks) induced strong nuclear staining of both c-Jun and p27 in the majority of epidermal cells in non-transgenic SAG-Tg(–) mice (middle). In SAG-Tg(+) mice, both the number of cells with nuclear staining of c-Jun and p27 and the staining intensity in c-Jun- and p27-positive cells were reduced (bottom panel). On the other hand, 10 weeks of UVB exposure induced cyclin D1 expression mainly at the basal cells, but SAG-transgenic expression had no effect (middle panel), consistent with observation made in primary kerotinocytes. Finally, SAG-transgenic expression had no effect on UVB-induced p53 (4th panel) and c-Fos (last panel), known to be degraded by other E3 ligases, Mdm2 (29) and Ubr1 (30), respectively. Thus, SAG, upon ectopic expression, reduces UVB-induced levels of c-Jun and p27, both in primary cultures and in mouse skin in situ. It is worth noting that unlike c-Jun and p27, a single UVB exposure of primary keratinocytes failed to induce cyclin D1 expression (Figure 3A), whereas multiple UVB exposures of mouse skin caused a significantly induction of cyclin D1, regardless of SAG-transgenic expression (Figure 3C).

SAG-transgenic expression promoted UVB-induced skin hyperplasia
During immunohistochemistry analysis of c-Jun and p27 expression, we noticed differences in the thickness of the epidermal layer between SAG-Tg(–) and SAG-Tg(+) mice, suggesting an effect of SAG-transgenic expression in skin hyperplasia. We followed up this observation more precisely using two independent assays. The first was the hematoxylin and eosin staining, followed by the measurement of epidermal layer thickness. Representative areas, shown in Figure 4A (top panels), clearly demonstrated that SAG-transgenic expression remarkably increases the thickness of the epidermis (P < 0.05). The second was BrdU incorporation assay to measure cells with active DNA synthesis. BrdU-labeled cells (stained in blue) were mainly localized in the basal cell layer. The number of positive cells was significantly increased upon SAG-transgenic expression (panel c, P < 0.01) (Figure 4B). Finally, we excluded the possibility that increased skin thickness in SAG-transgenic mice was due to a reduced rate of apoptosis by the TUNEL assay. Apoptotic cells (green dots) were hardly seen after 10 week UVB exposure, and no difference was observed between the two lines (P > 0.05, Figure 4C). ICH using antibody against active caspase-3 also failed to detect apoptotic cells (data not shown). Thus, it appears that SAG-transgenic expression increases the thickness of the epidermis induced by UVB exposure by promoting proliferation, rather than inhibiting apoptosis, which was hardly seen in the epidermis after 10 weeks of UVB exposure.


Figure 4
View larger version (50K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Increased hyperplasia, but not apoptosis, in UVB-exposed skin epidermis by SAG-transgenic expression: SAG-Tg(–) and SAG-Tg(+) mice after 10 weeks of UVB exposure were given intraperitoneal injections of the BrdU labeling reagent 2 h prior to killing. The skin tissue was collected and subjected to hematoxylin and eosin staining and measurement of thickness (A), BrdU incorporation assay (B) or TUNEL apoptotic assay (C). A representative area was shown. The thickness of epidermal layer (brackets) was measured under a microscope and expressed as millimeter after x400 magnification (mm/400). The BrdU-positive cells in a total of 1000 epidermal cells were counted. The propidium iodide staining for nucleus was shown in red, whereas TUNEL-positive nucleus was shown in green. The number of positive cells out of a total of 1000 epidermal cells was counted and data were summarized and analyzed by Student’s t-test; *P < 0.05; **P < 0.01.

 
SAG-transgenic expression had no significant effects on the formation of UVB-induced skin tumors or on papilloma-to-carcinoma conversion
We next determined the effect of SAG-transgenic expression on UVB-induced skin carcinogenesis. The tumor appearance and growth rates in ear and dorsal trunk were recorded for 29 weeks and summarized in Table I and Figure 5. Tumor incidence in ear (A) or dorsum (B), tumor growth rate in both locations (C) and mean number of tumors formed in both locations (D and E) were not significantly different between SAG-Tg(+) and SAG(–) mice. We also measured the rate of papilloma to carcinoma conversion in UVB-induced tumors in ear and dorsum harvested at the end of experiment. A conversion rate of 66.6% was observed among a total of 15 and 9 ear tumors analyzed from SAG-Tg(–) and SAG-Tg(+) mice, respectively. The conversion rate in dorsal tumors was much lower: 9.1% (one tumor) versus 25% (two tumors) in SAG-Tg(–) and SAG-Tg(+) lines, respectively. However, no statistical difference was observed due to the limited number of dorsal tumors. Thus, SAG-transgenic expression had no effect on promoting tumor formation and papilloma to carcinoma conversion during UVB carcinogenesis.


View this table:
[in this window]
[in a new window]

  		Table I. Effect of SAG in UVB-induced skin tumors

 


Figure 5
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. UBV-induced skin carcinogenesis is not affected by SAG-transgenic expression: SAG-Tg(+) (n = 17) and SAG-Tg(–) (n = 16) lines were subjected to UVB exposure three times a week for 29 weeks. The number of papillomas detected by palpation and size of tumor detected by a caliper in each group were recorded weekly after its appearance and plotted. Shown are the tumor incidence in ear (A) and dorsum (B), tumor growth rates in both locations (C, the boxes present the 25th and 75th percentiles. The centerline is the median, the plus is the mean and the cross is an outlier) and mean number of tumor in both locations (D and E).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Funding
 References
 
SCF E3 ubiquitin ligases constitute the largest family of E3 ubiquitin ligases that attach ubiquitin to a variety of cell-regulatory proteins for targeted degradation by the 26S proteasome (9). The core SCF E3 ubiquitin ligase is a complex of regulator of cullins-1/RING box protein 1-cullins or SAG-cullins that recruit E2 (31), while the substrate specificity of the SCF complex is determined by the F-box proteins that recognize the substrates through their WD40 or leucine-rich repeat (LRR) domains (32). For example, F-box protein Fbw7 recognizes c-Jun, whereas Skp2 recognizes p27 for targeted degradation (for review, see ref. 9). We have recently found that, as a RING component of SCF E3 ubiquitin ligases, SAG promotes ubiquitination and degradation of c-Jun, an essential member of the AP-1 transcription factor. As a consequence, TPA-induced and AP-1-mediated neoplastic transformation in the JB6 epidermal cell model was inhibited upon SAG expression or enhanced upon SAG siRNA silencing (16). This implies that SAG inhibits neoplastic transformation by targeting c-Jun/AP-1. On the other hand, our earlier study had shown that SAG could promote cell proliferation under serum starvation through inhibition of p27 accumulation (15). It appears that the net outcome of SAG expression for tumor promotion and cell proliferation would most probably be determined by the degree of targeted degradation of the SAG substrate proteins in a tissue-specific and/or cell-content-dependent manner.

Here, we directly addressed this important issue in a SAG-transgenic mouse model during UVB skin carcinogenesis. We found that targeted SAG expression in primary kerationcytes led to lower basal levels of both c-Jun and p27. Although both proteins were induced by UVB to a similar extent, regardless of SAG-transgenic expression, the total levels of induced c-Jun and p27 were lower in SAG-transgenic cells because of the lower initial basal levels (Figure 3A). The difference was more striking in in situ mouse skin tissue, where targeted SAG expression blocked UVB induction of both c-Jun and p27 dramatically (Figure 3C). Interestingly, this SAG-mediated reduction of c-Jun level/AP-1 activity (pro-tumor promotion) and reduction of p27 level (anti-proliferation) led to a net outcome of increased skin hyperplasia. This is clearly demonstrated by increased skin thickness and increased DNA synthesis, but not apoptosis (Figure 4). The results suggest that p27 plays a more important role in controlling epidermal hyperplasia under this context of UVB skin proliferation. On the other hand, SAG-transgenic expression had no significant effect on promoting tumor formation, suggesting a neutralizing effect of p27 reduction and AP-1 inactivation. However, we cannot exclude the potential contributions of other growth-regulatory substrates of SAG–SCF E3 ubiquitin ligases in the process.

We have recently determined the role of SAG in skin carcinogenesis induced by DMBA/TPA using the same SAG-transgenic mouse model (24). We found that SAG-transgenic expression regulated skin carcinogenesis by a stage-dependent targeting of c-Jun/AP-1 and Inhibitor {kappa}B/Nuclear factor-{kappa}B. At the early stage, SAG-transgenic expression inhibited AP-1 activity by targeting c-Jun, leading to a reduced skin hyperplasia (reduced skin thickness) and reduced rate of tumor formation, with a longer latent period. At the later stage, however, SAG-transgenic expression activated Nuclear factor-{kappa}B by targeting Inhibitor {kappa}B, leading to apoptosis inhibition and larger tumor sizes. This apparent difference was determined by the availability of F-box proteins, Fbx7 (expressed in normal skin to target c-Jun degradation at early stage) and β-TrCP (overexpressed in tumor to target Inhibitor {kappa}B degradation at later stage), with SAG playing a rate-limiting factor (24). Thus, SAG-transgenic expression either promotes or inhibits skin hyperplasia induced by UVB (this study) or DMBA/TPA (24), respectively. The opposite effect appears to be carcinogen/tumor promoter dependent with a net outcome largely determined by the levels of SAG substrates that control cell proliferation and tumor promotion.

We showed here that SAG overexpression inhibits UVB-induced apoptosis in JB6 epidermal cells, but not in mouse skin tissues. The major reason for this apparent discrepancy is that in JB6 cells, UVB, with a dose of 400 mJ/cm2, induced significant levels of apoptosis (>35% population, Figure 1C), whereas in mouse skin tissues, 10 weeks of exposure of UVB (2.5 kJ/m2 for the first 4 weeks, 5 kJ/m2 for weeks 5–8 and 10 kJ/m2 for remaining 2 weeks) induced very minimal levels of apoptosis (~1% population, Figure 4C). Although we observed some SAG protection, it is not statistically significant. Thus, it is very difficult to detect a difference when apoptosis is at background levels.

In summary, when overexpressed in mouse JB6 epidermal culture cells, SAG protects UVB-induced apoptosis. When targeted to express in mouse skin epidermal cells, SAG promotes UVB-induced hyperplasia, but not carcinogenesis. This is through enhanced proliferation, not reduced apoptosis. SAG-mediated degradation of c-Jun and p27 appears to contribute to this process, implying a significant role of p27 in inhibition of UVB skin hyperplasia.


   Funding
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
References
 
National Cancer Institute (1R01CA111554, 1R01CA118762) to Y.S.


   Footnotes
 
{dagger} These authors contributed equally to this work. Back


   Acknowledgments
 
We would like to thank Drs G.Fisher and J.Elder for their help in setting up the UVB exposure experiments and for the use of their equipments. We would also like to thank Dr T.Bowden for AP-1-Luc mice and Dr A.Dlugosz for discussion.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Funding
 References
 

  1. Hanahan D, et al. The hallmarks of cancer. Cell (2000) 100:57–70.[CrossRef][Web of Science][Medline]
  2. Epstein JH. Photocarcinogenesis, skin cancer, and aging. J. Am. Acad. Dermatol. (1983) 9:487–502.[Web of Science][Medline]
  3. Einspahr JG, et al. Skin cancer chemoprevention: strategies to save our skin. Recent Results Cancer Res. (2003) 163:151–164. discussion 264–266.[Medline]
  4. Duan H, et al. SAG, a novel zinc RING finger protein that protects cells from apoptosis induced by redox agents. Mol. Cell. Biol. (1999) 19:3145–3155.[Abstract/Free Full Text]
  5. Kamura T, et al. The Rbx1 subunit of SCF and VHL E3 ubiquitin ligase activates Rub1 modification of cullins Cdc53 and Cul2. Genes Dev. (1999) 13:2928–2933.[Abstract/Free Full Text]
  6. Ohta T, et al. ROC1, a homolog of APC11, represents a family of cullin partners with an associated ubiquitin ligase activity. Mol. Cell (1999) 3:535–541.[CrossRef][Web of Science][Medline]
  7. Swaroop M, et al. Yeast homolog of human SAG/ROC2/Rbx2/Hrt2 is essential for cell growth, but not for germination: chip profiling implicates its role in cell cycle regulation. Oncogene (2000) 19:2855–2866.[CrossRef][Web of Science][Medline]
  8. Tan P, et al. Recruitment of a ROC1-CUL1 ubiquitin ligase by Skp1 and HOS to catalyze the ubiquitination of IkBa. Mol. Cell (1999) 3:527–533.[CrossRef][Web of Science][Medline]
  9. Nakayama KI, et al. Ubiquitin ligases: cell-cycle control and cancer. Nat. Rev. Cancer (2006) 6:369–381.[CrossRef][Web of Science][Medline]
  10. Sun Y. Alteration of SAG mRNA in human cancer cell lines: requirement for the RING finger domain for apoptosis protection. Carcinogenesis (1999) 20:1899–1903.[Abstract/Free Full Text]
  11. Yang GY, et al. Attenuation of ischemia-induced mouse brain injury by SAG, a redox- inducible antioxidant protein. J. Cereb. Blood Flow Metab. (2001) 21:722–733.[CrossRef][Web of Science][Medline]
  12. Chanalaris A, et al. SAG attenuates apoptotic cell death caused by simulated ischaemia/reoxygenation in rat cardiomyocytes. J. Mol. Cell. Cardiol. (2003) 35:257–264.[CrossRef][Web of Science][Medline]
  13. Huang Y, et al. Elevated expression of SAG/ROC2/Rbx2/Hrt2 in human colon carcinomas: SAG does not induce neoplastic transformation, but its antisense transfection inhibits tumor cell growth. Mol. Carcinog. (2001) 30:62–70.[CrossRef][Web of Science][Medline]
  14. Sasaki H, et al. Expression of the sensitive to apoptosis gene, SAG, as a prognostic marker in nonsmall cell lung cancer. Int. J. Cancer (2001) 95:375–377.[CrossRef][Web of Science][Medline]
  15. Duan H, et al. Promotion of S-phase entry and cell growth under serum starvation by SAG/ROC2/Rbx2/Hrt2, an E3 ubiquitin ligase component: association with inhibition of p27 accumulation. Mol. Carcinog. (2001) 30:37–46.[CrossRef][Web of Science][Medline]
  16. Gu Q, et al. SAG/ROC2/Rbx2 is a novel activator protein-1 target that promotes c-Jun degradation and inhibits 12-O-tetradecanoylphorbol-13-acetate-induced neoplastic transformation. Cancer Res. (2007) 67:3616–3625.[Abstract/Free Full Text]
  17. Dlugosz AA, et al. Isolation and utilization of epidermal keratinocytes for oncogene research. Meth. Enzymol. (1995) 254:3–20.[Web of Science][Medline]
  18. Quan T, et al. Ultraviolet irradiation alters transforming growth factor beta/smad pathway in human skin in vivo. J. Invest. Dermatol. (2002) 119:499–506.[CrossRef][Web of Science][Medline]
  19. Bockbrader KM, et al. A small molecule Smac-mimic compound induces apoptosis and sensitizes TRAIL- and etoposide-induced apoptosis in breast cancer cells. Oncogene (2005) 24:7381–7388.[CrossRef][Web of Science][Medline]
  20. Cooper SJ, et al. Expression of dominant negative c-jun inhibits ultraviolet B-induced squamous cell carcinoma number and size in an SKH-1 hairless mouse model. Mol. Cancer Res. (2003) 1:848–854.[Abstract/Free Full Text]
  21. Rossiter H, et al. Targeted expression of bcl-2 to murine basal epidermal keratinocytes results in paradoxical retardation of ultraviolet- and chemical-induced tumorigenesis. Cancer Res. (2001) 61:3619–3626.[Abstract/Free Full Text]
  22. Xirodimas D, et al. Different effects of p14ARF on the levels of ubiquitinated p53 and Mdm2 in vivo. Oncogene (2001) 20:4972–4983.[CrossRef][Web of Science][Medline]
  23. Dhanalakshmi S, et al. Silibinin up-regulates DNA-protein kinase-dependent p53 activation to enhance UVB-induced apoptosis in mouse epithelial JB6 cells. J. Biol. Chem. (2005) 280:20375–20383.[Abstract/Free Full Text]
  24. Gu Q, et al. SAG/ROC2 E3 ligase regulates skin carcinogenesis by stage dependent targeting of c-Jun/AP1 and IkB/NF-kB. J. Cell Biol. (2007) 178:1009–1023.[Abstract/Free Full Text]
  25. Bian J, et al. p53CP, a putative p53 competing protein that specifically binds to the consensus p53 DNA binding sites: a third member of the p53 family? Proc. Natl Acad. Sci. USA (1997) 94:14753–14758.[Abstract/Free Full Text]
  26. Jayaraman J, et al. Activation of p53 sequence-specific DNA binding by short single strands of DNA requires the p53 C-terminus. Cell (1995) 81:1021–1029.[CrossRef][Web of Science][Medline]
  27. Kim YS, et al. Phosphorylation of threonine-10 on CKBBP1/SAG/ROC2/Rbx2 by protein kinase CKII promotes the degradation of I{kappa}B{alpha} and p27kip1. J. Biol. Chem. (2003) 278:28462–28469.[Abstract/Free Full Text]
  28. Lin DI, et al. Phosphorylation-dependent ubiquitination of cyclin D1 by the SCF(FBX4-alphaB crystallin) complex. Mol. Cell (2006) 24:355–366.[CrossRef][Web of Science][Medline]
  29. Haupt Y, et al. Mdm2 promotes the rapid degradation of p53. Nature (1997) 387:296–299.[CrossRef][Medline]
  30. Sasaki T, et al. Spatiotemporal regulation of c-Fos by ERK5 and the E3 ubiquitin ligase UBR1, and its biological role. Mol. Cell (2006) 24:63–75.[CrossRef][Web of Science][Medline]
  31. Wu K, et al. The SCF(HOS/beta-TRCP)-ROC1 E3 ubiquitin ligase utilizes two distinct domains within CUL1 for substrate targeting and ubiquitin ligation. Mol. Cell. Biol. (2000) 20:1382–1393.[Abstract/Free Full Text]
  32. Zheng N, et al. Structure of the Cul1-Rbx1-Skp1-F boxSkp2 SCF ubiquitin ligase complex. Nature (2002) 416:703–709.[CrossRef][Medline]
Received October 7, 2007; revised December 20, 2007; accepted January 12, 2008.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
29/4/858    most recent
bgn021v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by He, H.
Right arrow Articles by Sun, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by He, H.
Right arrow Articles by Sun, Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?