Here, we assessed the protective effect of silibinin on UVB-induced skin carcinogenesis in SKH-1 hairless mice. Topical application of silibinin before or immediately after UVB exposure or its dietary feeding resulted in a strong protection against photocarcinogenesis, in terms of tumor multiplicity (60–66%; P < 0.001), tumor volume per mouse (93–97%; P < 0.001) and tumor volume per tumor (80–91%; P < 0.001). Silibinin also moderately inhibited tumor incidence (5–15%; P < 0.01) and delayed tumor latency period (up to 4 weeks; P < 0.01–0.001). To investigate in vivo molecular mechanisms of silibinin efficacy, tumors and uninvolved skin from tumor-bearing mice were examined immunohistochemically for proliferation, p53, apoptosis, and activated caspase-3. Silibinin treatment showed a strong decrease (P < 0.001) in proliferating cell nuclear antigen-positive cells and an increase in p53-positive (P < 0.005–0.001), terminal deoxynucleotidyltransferase-mediated nick end labeling-positive (P < 0.005–0.001), and cleaved caspase-3–positive cells (P < 0.001). Western blot analysis of normal skin and tumor lysates showed that silibinin decreases the levels of cyclin-dependent kinase 2 and cyclin-dependent kinase 4 and associated cyclins A, E, and D1, together with an up-regulation of Cip1/p21, Kip1/p27, and p53. Silibinin also showed a strong phosphorylation of extracellular signal-regulated protein kinase 1/2, stress-activated protein kinase/c-JUN NH2-terminal kinase 1/2, and p38 mitogen-activated protein kinases but inhibited Akt phosphorylation and decreased survivin levels with an increase in cleaved caspase-3. Together, these results show a strong preventive efficacy of silibinin against photocarcinogenesis, which involves the inhibition of DNA synthesis, cell proliferation, and cell cycle progression and an induction of apoptosis. Furthermore, these results also identify in vivo molecular mechanisms of silibinin efficacy against photocarcinogenesis.

More than one million new cases of nonmelanoma skin cancers including basal cell carcinoma and squamous cell carcinoma are diagnosed annually in the United States, resulting mostly from repeated sunlight exposure (1), in which UVB radiation is the causal etiologic factor (2). The first step in UVB-induced carcinogenesis involves induction of DNA damage via cyclobutane pyrimidine dimers and pyrimidine [6-4] pyrimidone photoproducts (2). DNA damage triggers a rapid increase in p53 that enhances Cip1/p21 synthesis and shuts off cell replication and DNA synthesis, allowing more time for DNA repair and/or apoptotic death of cells carrying damaged DNA (3). Unrepaired DNA damage, however, disrupts cellular processes finally introducing wrong bases into DNA, which results in mutations leading to loss or inappropriate expression of affected genes (4). UVB-induced initiated cells follow multiple additional hits by repeated UVB exposure leading to benign followed by malignant skin tumors in which clonal expansion of initiated cells involves oxidant–antioxidant imbalance, mitogenic and anti-apoptotic signaling cascades, and altered cell cycle and apoptosis regulation (5).

Together, these observations suggest that inhibiting UVB-caused alterations in skin might be useful in preventing nonmelanoma skin cancers. Several studies have shown that various phytochemicals protect against UVB-induced skin damages and cancer (5, 6, 7). Silymarin (flavonoid from milk thistle) and silibinin (bioactive component in silymarin) also cause strong prevention against UVB skin damages and photocarcinogenesis (8, 9). Furthermore, silibinin exerts dual efficacy in protecting or enhancing apoptosis in HaCaT cells depending on UVB damage, suggesting that it works as a UVB-damage sensor in deciding cell fate in its efficacy (10). Here, we report detailed mechanism-based efficacy studies with silibinin in UVB-induced skin tumorigenesis model. Photoprotective effects of silibinin revealed potential antitumor mechanisms involving inhibition of DNA synthesis and cell proliferation and apoptosis induction accompanied by modulation of p53 and cyclin-dependent kinase (CDK)-cyclin–CDK inhibitor levels. Additionally, alterations in mitogen-activated protein kinases (MAPKs) and Akt signaling, survivin expression, and caspase-3 activation were associated with the preventive efficacy of silibinin against photocarcinogenesis.

Experimental Design.

Silibinin diet (1%, w/w) was prepared by Dyets, Inc. (Bethlehem, PA) Female SKH-1 hairless mice (5 weeks old) were from Charles River Laboratories (Wilmington, MA). The UVB light source was four FS-40-T-12-UVB sunlamps with UVB Spectra 305 Dosimeter (Daavlin Co., Bryan, OH) emitting 80% radiation within 280 to 340 nm with a peak at 314 nm monitored with SEL 240 photodetector, 103 filter, and 1008 diffuser attached to IL1400A Research Radiometer (8, 9).

A long-term study was conducted assessing silibinin efficacy on UVB (180 mJ/cm2 5 d/week for 25 weeks)-caused skin tumorigenesis and associated molecular events in female SKH-1 hairless mice. With or without UVB groups had 20 or 10 mice, respectively. Treatments were (a) unexposed/untreated control, (b) 9 mg of topical silibinin per 200 μL of acetone per mouse, (c) 180 mJ/cm2 UVB five times per week for 25 weeks, (d) 9 mg of topical silibinin per 200 μL of acetone per mouse 30 minutes before UVB, (e) 9 mg of topical silibinin per 200 μL of acetone per mouse immediately after UVB, (f) 1% silibinin in diet and UVB for 25 weeks, and (g) 1% silibinin in diet for 25 weeks. Body weight and diet consumption were recorded throughout experiment, and for tumor study, data were analyzed for (a) percent tumor incidence, (b) tumor multiplicity (number per group), and (c) tumor volume per mouse (8). At experiment termination, tumors and uninvolved skin (from tumor-bearing mice) and normal skin (controls) were collected and used for immunohistochemical and/or Western analyses.

Bromodeoxyuridine Incorporation.

Animals received injections i.p. of bromodeoxyuridine (BrdUrd; 50 mg/kg) and sacrificed 1 hour later; tissues were collected and processed; and sections were made. Endogenous peroxidase was blocked by 5% hydrogen peroxide in methanol for 10 minutes followed by incubation with 1 n HCl for 30 minutes at 37°C. Sections were incubated with mouse monoclonal anti-BrdUrd antibody (Sigma) for 2 h, rinsed with PBS, and incubated with conjugated horseradish peroxidase–streptavidin for 30 minutes at room temperature. Sections were then incubated with 3,3′-diaminobenzidine working solution for 10 minutes at room temperature, counterstained with diluted Harris hematoxylin, dehydrated, and mounted.

Immunostaining.

Tissues were fixed in 10% formalin for 8 to 10 hours at 4°C, dehydrated in ethanol, cleared in xylene, and embedded in PolyFin. Four-μm serial sections were cut, processed, and immunostained using monoclonal proliferating cell nuclear antigen (PCNA; Dako, Carpinteria, CA), p53 (Novocastra Laboratories, Newcastle Upon Tyne, UK), and cleaved caspase-3 (Cell Signaling, Beverly, MA) antibodies (9). For negative staining controls, N-Universal Negative Control-mouse or rabbit antibody (Dako) was used under identical conditions. Terminal deoxynucleotidyltransferase-mediated nick end labeling (TUNEL) staining for apoptotic cells was done as published recently (9). All immunohistochemical analyses were done using Zeiss Axioscop 2 microscope (Carl Zeiss, Inc., Jena, Germany).

Lysate Preparation and Immunoblotting.

Tissues were homogenized in lysis buffer, and lysates were prepared as published (10). Western immunoblotting was performed to probe cyclin A (H-432); cyclin E (C-19); cyclin D1 (M-20); CDK2 (M-2); CDK4 (H-303; Santa Cruz Biotechnology, Santa Cruz, CA); p53 (Novocastra Laboratories); survivin (Novus, Littleton, CO); phospho- and total extracellular signal-regulated protein kinase (ERK) 1/2, c-JUN NH2-terminal kinase (JNK) 1/2, p38, and Akt; cleaved caspase-3 (Cell Signaling); and Cip1/p21 (Upstate), Kip1/p27 (Ab-1, Clone DCS-72.F6), and tubulin-α (Ab-2; Lab Vision, Fremont, CA) as published recently (10).

Statistical Analysis.

Statistical significance of difference between UVB alone versus all other groups was determined by one-way ANOVA followed by Bonferroni t test using SigmaStat 2.03. P < 0.05 was considered statistically significant.

Protective Effects of Topical Application/Dietary Feeding of Silibinin on Ultraviolet B-Induced Tumorigenesis.

To evaluate silibinin efficacy against UVB-induced tumorigenesis, three different protocols were selected in which it was applied topically (a) before or (b) immediately after UVB or (c) fed in diet. UVB exposure of mice at 180 mJ/cm2 dose 5 d/week caused 100% tumor incidence at 25 weeks; however, silibinin treatments before or immediately after UVB or its dietary feeding resulted in 85 to 95% incidence accounting for moderate (5–15%) protection (Fig. 1,A). In terms of latency period (appearance of first tumor), compared with UVB group showing tumor appearance at 12 weeks, silibinin pre- or post-topical application or its dietary feeding showed first tumor appearance at 16, 14, and 15 weeks (P < 0.01–0.001), accounting for a delay in latency period by 4, 2, and 3 weeks, respectively (Fig. 1 A).

Whereas silibinin efficacy toward tumor incidence and latency was moderate, it showed very strong effects on tumor multiplicity and volume. Compared with UVB alone, silibinin treatments in three different protocols resulted in a strong reduction in total number of tumors per group throughout experiment (Fig. 1,B). At 25 weeks, compared with UVB group with a total of 273 tumors, number of tumors per group in silibinin pre- or post-topically treated or dietary-fed group were 92, 106, and 108, accounting for 66, 61, and 60% inhibition (P < 0.001), respectively (Fig. 1,B). Similarly, in terms of number of tumors per mouse, compared with UVB group with 13.65 ± 1.48 tumors, silibinin topical treatment in pre- or post-UVB or dietary feeding showed 4.6 ± 0.64, 5.3 ± 0.55, and 5.4 ± 0.83 tumors/mouse accounting for 60 to 66% protection (P < 0.001), respectively (data not shown). In case of tumor volume, compared with UVB group with 2765 ± 759 mm3 tumor volume/mouse, total tumor volume per mouse in pre- or post-silibinin topical treatment or dietary feeding group was only 92 ± 19, 98 ± 17, and 200 ± 53 mm3 accounting for 93 to 97% inhibition (P < 0.001), respectively (Fig. 1 C). Similarly, when data were analyzed for tumor volume per tumor, compared with UVB group with 188 ± 50 mm3 volume/tumor, silibinin pre- or post-topical treatment or dietary feeding resulted in only 16 ± 3, 17 ± 2, and 40 ± 15 mm3 volume/tumor, respectively, accounting for 80 to 91% (P < 0.001) inhibition (data not shown). Overall, our completed tumorigenesis experiment clearly and convincingly shows a strong protective effect of topically applied (before or immediately after UVB) and dietary-fed silibinin against UVB-induced tumorigenesis in mouse skin.

To assess whether silibinin supplementation affects diet consumption, diet consumption profiles were followed for control groups of mice on regular or 1% silibinin supplemented (w/w) diet for 25 weeks, a time-period identical to experimental groups. Compared with regular diet, animals in silibinin dietary supplement group consumed diet at a much higher level throughout 25 weeks of study (data not shown). Body weight gain profile is one of the most extensively used measures for an overall systemic toxicity and untoward side effects in cancer chemoprevention (11). Accordingly, we also recorded body weight for each mouse in all control and experimental groups. Neither topical nor dietary silibinin showed any noticeable change in body weight gain profile throughout study when compared with controls. Conversely, animals in UVB alone group showed a strong decrease in body weight with increasing weeks of UVB when compared with those in (a) control, (b) silibinin treatments in two different protocols, and (c) silibinin treatments in three different protocols and UVB (Fig. 1 D). Together, these observations suggest that dietary silibinin is well accepted by mice in terms of food consumption and that topical and dietary silibinin treatments do not exert toxicity or adverse health effects in mice.

Silibinin Inhibits Bromodeoxyuridine Incorporation in Epidermal Cells.

To evaluate proliferating cells predominantly in S phase, we measured BrdUrd incorporation into the DNA of epidermal cells (Fig. 2,A). UVB irradiation for 25 weeks showed strong immunoreactivity for BrdUrd in uninvolved skin (data not shown) accounting for 23.2 ± 2.2% BrdUrd-positive cells compared with 5.3 ± 1.5, 4.8 ± 2.5, and 7.1 ± 1.5% in untreated, dietary, and topical silibinin–treated but unirradiated groups (4.4-fold increase in UVB alone versus unirradiated group, P < 0.001), respectively (Fig. 2,A). Topical application of silibinin before or immediately after UVB or its dietary feeding, however, resulted in 11.8 ± 1.1% (P < 0.001), 14.9 ± 1.2% (P < 0.001), and 16.9 ± 1.1% (P < 0.005) BrdUrd-positive cells in uninvolved skin, accounting for 49, 35, and 27% decrease, respectively (Fig. 2 A), suggesting that silibinin inhibits UVB-induced DNA synthesis in mouse skin epidermal cells.

Silibinin Inhibits Cell Proliferation in Ultraviolet B-Induced Tumorigenesis.

To further determine whether topical or dietary silibinin treatment affects UVB-induced cell proliferation, uninvolved skin and tumor samples and respective unirradiated skin samples were evaluated for PCNA immunostaining (data not shown). In unirradiated mice receiving normal diet, 4.2 ± 1.5% PCNA-positive cells were observed in skin that were comparable with those in topically (4.3 ± 0.6) or dietary (4.8 ± 1.5) silibinin alone group (Fig. 2,B). A strong PCNA immunostaining, however, was observed in tumors and epidermis from uninvolved skin in UVB alone group (data not shown), accounting for 38.8 ± 1.6% and 21.6 ± 1.4% PCNA-positive cells (P < 0.001 versus unirradiated controls), respectively (Fig. 2,B and C). Silibinin treatment before or immediately after UVB or its dietary feeding decreased PCNA-positive cells to 16.3 ± 1.4% (P < 0.001), 18.0 ± 1.1% (P < 0.001), and 15.7 ± 1.2% (P < 0.005) in tumor samples, accounting for 54 to 60% decrease (Fig. 2,C). Silibinin treatment also decreased (12–36%), although to a lesser degree, PCNA-positive cells to 16.2 ± 1.8% (P < 0.001), 18.9 ± 1.7% (P < 0.001), and 13.8 ± 1.1% (P < 0.001) in uninvolved skin from tumor-bearing mice, respectively (Fig. 2 B).

Silibinin Up-regulates p53 Protein Level in Ultraviolet BInduced Tumorigenesis.

p53 is known to play a key role in cell cycle regulation and apoptosis (12), and therefore, next, we analyzed its expression by immunostaining (data not shown) in tumor and uninvolved skin samples. Exposure of mice to multiple UVB irradiation in 25 weeks tumorigenesis protocol resulted in 14.7 ± 1.7% (P < 0.001) and 11.2 ± 1.1% (P < 0.001) p53-positive cells in tumor and uninvolved skin sections, respectively, compared with 1.8 ± 1.1% in unirradiated controls (Fig. 3,A and B). Silibinin treatment before or immediately after UVB or its dietary feeding, however, increased p53-positive cells to 31.8 ± 1.2% (P < 0.001), 24.2 ± 1.5% (P < 0.005), and 32.0 ± 2.2% (P < 0.005) in the tumors, respectively, accounting for 1.7- to 2.3-fold increase over UVB alone tumors (Fig. 3,A). Similarly, silibinin showed an increase in p53-positive cells to 19.8 ± 1.7% (P < 0.001), 17.5 ± 1.8% (P < 0.005), and 17.1 ± 1.9% (P < 0.005) in uninvolved skin, respectively, accounting for 1.5- to 1.8-fold increase over UVB alone group (Fig. 3,B). In other control groups, dietary or topical silibinin alone, without UVB, did not show any change in p53 positive cells in the skin when compared with unirradiated controls (Fig. 3 B).

Silibinin Increases Apoptosis in Ultraviolet B-Induced Tumorigenesis.

DNA is the primary target of UVB damage, causing either apoptotic death of damaged cells or fixation of mutation in p53 leading to tumor initiation (4). p53 accumulation after DNA damage by UVB plays an important role in DNA repair and apoptotic death of damaged cells (4). Based on our results showing that tumors and uninvolved skin from tumor-bearing mice in three different silibinin treatment protocols have significantly higher levels of p53 compared with UVB alone, we next assessed percentage apoptotic cells by TUNEL staining (data not shown). Quantitative analyses of TUNEL-positive cells showed 15.4 ± 1.7% (11-fold, P < 0.001) and 12.7 ± 1.4% (9-fold, P < 0.001) apoptotic cells in tumor and uninvolved skin samples, respectively, from UVB alone group compared with 1.4 ± 0.3% in unirradiated controls (Fig. 4,A and B). However, silibinin treatment before or immediately after UVB or its dietary feeding further increased the TUNEL-positive cells to 26.1 ± 1.8% (P < 0.001), 30.3 ± 2.6% (P < 0.001), and 24.9 ± 1.5% (P < 0.001) in tumor samples, respectively, accounting for 1.6- to 2.0-fold increase over UVB alone tumors (Fig. 4,A). Similar silibinin treatments also resulted in an increase in TUNEL-positive cells to 22.2 ± 2.2% (P < 0.001), 20.9 ± 2.2% (P < 0.001), and 18.4 ± 2.1% (P < 0.001) in uninvolved skin, respectively, which accounted for 1.4- to 1.7-fold increase compared with UVB alone (Fig. 4 B). Together, these results suggested the apoptotic potential of silibinin in chronic UVB-exposed skin as well as tumors.

Silibinin Up-regulates Cleaved Caspase-3 Level in Ultraviolet B-Induced Tumorigenesis.

To assess possible involvement of caspase activation in silibinin-caused increase in apoptosis, next we performed cleaved caspase-3 immunostaining in tumors and uninvolved skin samples from tumor-bearing mice and respective unirradiated control skin samples (data not shown). Quantitative analyses of immunostaining revealed that topical application of silibinin before or immediately after UVB, or its dietary feeding for 25 weeks increases percent of cleaved caspase-3–positive cells by 2.9- to 3.6-fold (P < 0.001) in tumor and 1.7- to 2.2-fold (P < 0.001) in uninvolved skin samples compared with UVB alone group (Fig. 4,C and D). Exposure of mice to UVB (180 mJ/cm2 5 d/week for 25 weeks) resulted in 2.6 ± 1.4% and 1.8 ± 0.2% cleaved caspase-3–positive cells in tumor and uninvolved skin sections, respectively (Fig. 4,C and D), with no detectable cleaved caspase-3 staining in unirradiated controls without or with silibinin treatment (Fig. 4,D). However, silibinin treatment in three different protocols increased caspase-3–positive cells to 7.7 ± 0.7% (P < 0.001), 7.5 ± 0.9% (P < 0.01), and 9.4 ± 1.4% (P < 0.005) in tumor sections (Fig. 4,C) and to 3.9 ± 0.4% (P < 0.001), 3.2 ± 0.3% (P < 0.001), and 3.0 ± 0.4% (P < 0.001) in uninvolved skin sections, respectively (Fig. 4 D).

Silibinin Modulates Cell Cycle Regulators in Ultraviolet B-Induced Tumorigenesis.

Enhanced expression of cell cycle regulatory proteins such as CDKs and cyclins or decreased expression of CDK inhibitors have been implicated in photocarcinogenesis (13). Based on our results showing an inhibition of tumor growth and cell proliferation together with an increase in p53 expression by silibinin in UVB-induced tumorigenesis protocols, we rationalized that silibinin might have modulatory effects on cell cycle regulatory proteins. As shown in Fig. 5,A, compared with normal skin from age-matched unirradiated controls, tumor tissues from UVB alone group showed a strong increase in cyclins (cyclin A, E, and D1) and CDK2 levels; protein expression of CDK4 was reduced considerably in tumors compared with normal skin. However, we observed a strong decrease in cyclin A, cyclin E, cyclin D1, CDK2, and CDK4 levels in tumors harvested from the groups of mice treated with silibinin before or immediately after UVB, or its dietary feeding compared with UVB alone tumors (Fig. 5,A). Next, we examined levels of Cip1/p21, which is a universal inhibitor of cell cycle progression and is transcriptionally activated by p53 after UVB (12, 14). Consistent with silibinin effect in up-regulating p53 protein levels, we observed an increase in Cip1/p21 in silibinin-UVB–treated groups of tumors compared with UVB alone (Fig. 5,A). Kip1/p27 is another important CDK inhibitor that regulates CDK–cyclin activity at G1–S transition (14). Protein levels of Kip1/p27 were also strongly up-regulated in silibinin-UVB–treated groups of tumors compared with those in UVB alone group (Fig. 5,A). In the studies assessing p53 levels by immunoblotting, consistent with immunostaining data, compared with unirradiated (normal) skin, tumor samples from UVB alone group showed strong levels of p53 protein; however, tumors from silibinin-UVB–treated groups of mice showed much stronger levels of p53 (Fig. 5,A). Reprobing membranes with α-tubulin antibody confirmed equal protein loading in each case (Fig. 5 A). Together, these results suggest that silibinin treatment induces expression of Cip1/p21 and Kip1/p27, accompanied by a decrease in CDK2, CDK4, cyclin A, cyclin E, and cyclin D1 that would overall reduce CDK–cyclin kinase activity followed by other events causing cell cycle arrest thereby inhibition of tumor cell growth.

Silibinin Up-regulates Mitogen-Activated Protein Kinase Activation Cascades in Ultraviolet B-Induced Tumorigenesis, but Inhibits in Uninvolved Skin.

UV-caused activation of MAPK family protein kinases, i.e., ERK, JNK, and p38, and their roles in cell cycle and apoptosis regulation as well as anti-apoptotic cell survival, signaling that influence overall UVB-induced tumorigenesis has been well studied in recent years by several laboratories (5, 15, 16). Based on these studies and our observations showing that tumors from silibinin treatment groups have decreased PCNA staining, together with an increased apoptotic cell population as well as strong levels of cleaved caspase-3, we next investigated the levels of phospho- and total ERK1/2, JNK1/2, and p38 kinase in tumors from UVB alone and silibinin-UVB–treated groups of mice. Phospho-ERK1/2, JNK1/2, and p38 kinase levels were strongly higher without any change in their total protein levels in silibinin-UVB–treated tumors in three different protocols compared with those from UVB alone group or normal skin (Fig. 5,B). Reprobing of membranes with α-tubulin antibody confirmed equal protein loading. Because ERK1/2 activation is also reported to inhibit cell cycle progression via CDK inhibitor up-regulation (17) and because activation of JNK1/2 and p38/MAPK is widely reported with apoptosis (18), our results suggest the roles of MAPK pathways in silibinin-caused alteration of cell-cycle regulatory proteins and/or apoptotic cell death, which might be an important component of silibinin efficacy against UVB-induced tumorigenesis. Conversely to results in tumors, silibinin treatment in three different protocols resulted in almost complete decrease in all three phospho-MAPK levels without any changes in their total protein levels in uninvolved skin from tumor-bearing mice, when compared with UVB alone samples (Fig. 5 D). This observation clearly suggests a dual efficacy of silibinin in (a) protecting nontumorigenic skin cells against UVB-caused MAPK activation that associates with proliferation and transformation, and (b) enhancing apoptotic death of tumorigenic skin cells by further activating MAPKs and other cascades.

Silibinin Inhibits Akt Activation, Decreases Survivin Expression, and Strongly Activates Caspase-3 Cleavage in Ultraviolet B-Induced Tumorigenesis.

Involvement of Akt in diverse tumorigenic activities suggests that Akt activation alone might be sufficient to induce cancer (19), and accordingly identification of inhibitors or modulators of PI3K/Akt pathway could be critical for developing novel therapeutic strategies directed at neoplasms exhibiting Akt activation (19). Consistent with this notion, we observed an activation of Akt in terms of strong phospho-Akt levels in tumors from UVB alone group compared with unirradiated skin (Fig. 5,C). However, tumor samples from silibinin-UVB–treated groups showed only marginal levels of Akt phosphorylation without any noticeable changes in total Akt protein (Fig. 5,C). Next, we analyzed the survivin level that is a potential modulator of keratinocytic apoptosis and is distinguished from other apoptotic regulators by its absence in most normal tissues and highly selective expression in cancers, making it a particularly promising target for cancer therapy (20). The level of survivin was strongly elevated in tumors from UVB alone group compared with normal skin; however, these strong levels of survivin in UVB alone tumors totally diminished to undetectable levels in tumor samples from silibinin-UVB–treated groups in three different protocols (Fig. 5,C). Because survivin is an inhibitor of caspase-3 activation (20), our immunoblot data showing a modest reactivity for cleaved caspase-3 in UVB alone tumors versus very strong levels in tumors from silibinin-UVB–treated groups (Fig. 5 C) are in accord with our survivin results and apoptotic effect of silibinin observed in tumors. Together, it could be reasoned that down-regulation of Akt activation by silibinin renders cells more prone not only to direct apoptotic effects of p53, but also to all other concurrent apoptotic signals that are normally blocked by activated Akt.

UVB exposure of skin results in various biological responses either through direct DNA damage (2), or via formation of free radicals and reactive oxygen species that damage DNA and non-DNA cellular targets (4, 21, 22). UVB-induced biological responses include cell cycle arrest, apoptosis, cell proliferation, and reparative hyperplasia, which are involved in pathophysiology of erythema, premature skin aging, and skin cancer (4, 21, 22). These studies suggest that protecting skin against UVB-induced biological responses mentioned above would also be effective in reducing nonmelanoma skin cancer incidence. With this rationale, a major aim of present study was to evaluate silibinin efficacy against UVB-induced tumorigenesis and define molecular mechanisms of observed efficacy. Our results, for the first time, show a strong suppression of photocarcinogenesis by silibinin, which involves its effect on inhibition of DNA synthesis and cell proliferation and induction of apoptosis. These antitumor effects of silibinin were accompanied by an increased accumulation of p53 protein and an induction of CDK inhibitors with a decrease in CDK–cyclin levels. Furthermore, silibinin activated MAPK but inhibited Akt pathway and decreased survivin level with a concomitant increase in caspase-3 activation.

In normal human and mouse epidermis, cells are constantly turning over where stem cells divide and generate into keratinocytes that differentiate and desquamate on the surface of skin; differentiated cells, thereby, are constantly replaced by proliferating cells from basal layer (22). PCNA, a subunit of DNA polymerase, plays a crucial role in DNA replication and repair and serves as a biomarker of proliferation (22). Earlier studies have reported that inhibition of DNA synthesis and cell proliferation by black tea, green tea, caffeine, or epigallocatechin is associated with their cancer chemopreventive efficacy (7, 23). Silibinin also inhibited UVB-induced DNA synthesis and cell proliferation, as observed by a decrease in BrdUrd incorporation and PCNA-positive cells in uninvolved skin and tumors, suggesting a plausible role of in vivo antiproliferative effect of silibinin in its overall efficacy against UVB-induced tumorigenesis.

Chemopreventive agents enhancing the levels of p53 and other tumor suppressor proteins is relatively an unexplored approach (24, 25); however, agents such as green tea and caffeine have been shown to selectively induce apoptosis of UVB-damaged cells by up-regulating wild-type p53 protein (26, 27). Interestingly, p53 simultaneously induces cascade of events for G1 arrest and apoptosis in same cell type (28); for example high doses of UV irradiation induce apoptosis in human keratinocytes, and low doses activate repair of UV-induced DNA damage (29). Ability of p53 to promote growth arrest/repair and thus survival on one hand, but apoptosis on other, raises the question of what determines the choice between these two opposite events? It is proposed that p53-dependent apoptosis is an important response to DNA damage, allowing selective removal of precancerous cells (30). Furthermore, Tron et al.(31) proposed “dual-role” model for p53-regulated responses to UV damage, indicating that after UV exposure, basal keratinocytes repair damaged DNA but differentiating keratinocytes undergo apoptosis, and both processes are regulated by p53. These studies suggest that if DNA repair is not feasible, damaged cells could undergo apoptosis, which can be considered as a last escape mechanism from mutagenesis/carcinogenesis. Consistent with these reports, we observed a strong association between an increased levels of p53 and apoptosis induction in silibinin-treated chronic UVB exposed skin and tumors compared with UVB alone samples. Interestingly, these results are in contrast with another recent study by us showing that in an acute single UVB exposure protocol, silibinin treatments under identical conditions cause an up-regulation of p53 protein but protect cells from apoptotic death (9). Together, the results of present study and those published recently by us (9), clearly demonstrate “dual efficacy” of silibinin in protecting cells against UVB-induced apoptosis if damage is acute (9) and enhancing apoptosis if UVB-induced damage is chronic; however, both situations involve p53 up-regulation by silibinin as reported by Tron et al.(31).

Regulation of cyclin–CDK complexes plays a key role in cell cycle progression at different phases in which CDKs are negatively regulated by a group of functionally related proteins known as CDK inhibitors such as Kip/Cip family members (32, 33). Cip1/p21 is a universal CDK inhibitor and is mainly regulated by p53 (14). Cip1/p21 binds PCNA and inhibits PCNA function in DNA replication in a p53-dependent manner (34); Kip1/p27 is up-regulated in response to antiproliferative signals (35). Consistent with these reports, silibinin treatments resulted in an up-regulation of Kip1/p27 and Cip1/p21 and decreased CDK2, CDK4, cyclin E, cyclin A and cyclin D1 protein levels in tumors, which could possibly be associated with a cell cycle arrest in tumor cells causing a decrease in proliferation as observed by a decreased PCNA immunostaining. Overall, it could be suggested that inhibition in cell proliferation could be one of the major mechanisms of silibinin efficacy against UVB-induced tumorigenesis.

MAPKs are important upstream regulators of transcription factor activities, and their signaling is critical to transduction of a wide variety of extracellular stimuli into intracellular cascades, thereby controlling the cellular events such as proliferation, differentiation, and apoptosis (36). MAPK signaling cascades are important in regulating magnitude of UV-induced cellular responses (15), in which ERK1/2 activation causes cell proliferation and survival, but JNK1/2 and p38 kinase seem to exert anti- and pro-apoptotic functions (18). ERK activation also exerts dual role where its very high intensity halts cell cycle by inducing CDK inhibitor Cip1/p21 and Kip1/27 expression (37). Ceramide, Taxol, and etoposide have very different intracellular targets yet induce apoptosis through ERK1/2 activation cascade in A431 cells; they also activate JNK1/2 and p38 kinases in their apoptotic response in cancer cells (18). JNK1/2 and p38 kinase also mediate various forms of cellular stress, such as damage repair mechanisms, cell growth arrest, and apoptotic cell death (38). P38 mediates UVB apoptosis by inducing cytochrome c release into cytosol followed by caspase-3 activation in HaCaT cells (39). Conversely, UVB-caused induction of p38 activity followed by activation of transcription factor AP-1 is also reported in HaCaT human keratinocytes, which could lead to pro-survival/anti-apoptotic response (40). Consistent with these reports, silibinin treatments in UVB tumorigenesis studies resulted in a strong increase in ERK1/2, JNK1/2, and p38 phosphorylation in tumor samples, suggesting their possible involvement in an overall silibinin efficacy in causing apoptosis induction in tumor cells.

Akt signaling plays a prominent role in several processes considered hallmarks of cancer (19). Akt makes a tumor cell overly responsive to ambient levels of growth factors for survival that normally would not provoke proliferation. Furthermore, Akt activation affects cell cycle progression, through regulation of cyclin D stability (41) and a decrease in Kip1/p27 via its phosphorylation and degradation (42). In apoptosis-prone cells, p53-dependent signaling also enables Akt down-regulation, predisposing them to rapid apoptosis in response to a combination of cellular stress and decreased survival signals (43). An activated p53 causes a rapid decrease in steady-state levels of Akt by its caspase-mediated degradation (44). Consistent with these reports, we observed an almost complete decrease in phospho-Akt in tumor samples from silibinin-UVB–treated groups compared with its strong levels in tumors from UVB alone group, suggesting possible involvement of cross-talk between Akt and p53 in apoptosis regulation in tumor cells from silibinin treatment groups. Based on this association, it could be reasoned that silibinin-caused pronounced accumulation of p53 might also result in a down-regulation of Akt activation rendering cells more prone not only to direct apoptotic effects of p53, but also to all other concurrent apoptotic signals, outcomes of which are normally blocked by activated Akt.

Recent studies have explained and established molecular mechanism of apoptosis regulation by survivin via its interaction with caspases pathway (45). Molecular antagonists of survivin including antisense, ribozymes, short interferring RNA sequences, and dominant-negative mutants have been shown to cause caspase-dependent cell death, enhancement of apoptotic stimuli, and anticancer activity, in vivo(46, 47). Furthermore, a recent study shows that survivin expression in mouse skin prevents papilloma regression and promotes tumor progression (48). Consistent with these reports, silibinin treatment decreased survivin level in tumors that might have a direct role in caspase-3 activation causing apoptotic death of tumor cells in silibinin treatment groups as observed in present study. In addition, cell death induced through p53 pathway is also executed by caspase cascade, which is the central event in effector phase of apoptosis (49). In this regard, silibinin treatment strongly induced activated caspase-3 levels and the number of TUNEL-positive cells in tumor and uninvolved skin samples.

In summary, the findings in present study show that topical application of silibinin before or immediately after UVB irradiation or its dietary feeding to SKH-1 mice exposed to UVB, strongly inhibits UVB-induced tumorigenesis. Our data also suggest that observed preventive efficacy of silibinin is most likely mediated via an inhibition in DNA synthesis, cell proliferation, and cell cycle arrest and an induction of apoptosis. As summarized in Fig. 6, the molecular events identified to be associated with silibinin efficacy include (a) an increased accumulation of p53 protein, (b) an induction of CDK inhibitors together with a decrease in CDK–cyclin levels, (c) activation of MAPK pathways, (d) inhibition of Akt signaling, and (e) decreased level of survivin concomitant with caspase-3 activation.

Grant support: United States Public Health Service grant CA64514.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Rajesh Agarwal, Department of Pharmaceutical Sciences, School of Pharmacy, University of Colorado Health Sciences Center, 4200 East Ninth Street, Box C238, Denver, CO 80262. Phone: 303-315-1381; Fax: 303-315-6281; E-mail: Rajesh.Agarwal@UCHSC.edu

1
Jemal A, Tiwari RC, Murray T, et al Cancer statistics, 2004.
CA Cancer J Clin
,
54
:
8
-29,  
2004
.
2
Mitchell DL, Greinert R, de Gruijl FR, et al Effects of chronic low-dose ultraviolet B radiation on DNA damage and repair in mouse skin.
Cancer Res
,
59
:
2875
-84,  
1999
.
3
Zambetti GP, Levine AJ. A comparison of the biological activities of wild-type and m-p53.
FASEB J
,
7
:
855
-65,  
1993
.
4
Matsumura Y, Ananthaswamy HN. Toxic effects of ultraviolet radiation on the skin.
Toxicol Appl Pharmacol
,
195
:
298
-308,  
2004
.
5
Bowden GT. Prevention of non-melanoma skin cancer by targeting ultraviolet-B-light signaling.
Nat Rev Cancer
,
4
:
23
-35,  
2004
.
6
Harris RB, Alberts DS. Strategies for skin cancer prevention.
Int J Dermatol
,
43
:
243
-51,  
2004
.
7
Conney AH. Tailoring cancer chemoprevention regimens to the individual.
J Cell Biochem
,
91
:
277
-86,  
2004
.
8
Katiyar SK, Korman NJ, Mukhtar H, Agarwal R. Protective effects of silymarin against photocarcinogenesis in a mouse skin model.
J Natl Cancer Inst (Bethesda)
,
89
:
56
-66,  
1997
.
9
Dhanalakshmi S, Mallikarjuna GU, Singh RP, Agarwal R. Silibinin prevents ultraviolet radiation-caused skin damages in SKH-1 hairless mice via a decrease in thymine dimer positive cells and an up-regulation of p53–p21/Cip1 in epidermis. Carcinogenesis. Epub 2004 Mar 19.
10
Dhanalakshmi S, Mallikarjuna GU, Singh RP, Agarwal R. Dual-efficacy of silibinin in protecting or enhancing ultraviolet B radiation-caused apoptosis in human immortalized keratinocyte HaCaT cells.
Carcinogenesis
,
25
:
99
-106,  
2004
.
11
Singh RP, Dhanalakshmi S, Tyagi AK, Chan DC, Agarwal C, Agarwal R. Dietary feeding of silibinin inhibits advance human prostate carcinoma growth in athymic nude mice and increases plasma insulin-like growth factor-binding protein-3 levels.
Cancer Res
,
62
:
3063
-9,  
2002
.
12
Vogelstein B, Lane D, Levine AJ. Surfing the p53 network.
Nature
,
408
:
307
-10,  
2000
.
13
Hong JT, Kim EJ, Ahn KS, et al Inhibitory effect of glycolic acid on ultraviolet-induced skin tumorigenesis in SKH-1 hairless mice and its mechanism of action.
Mol Carcinog
,
31
:
152
-60,  
2001
.
14
Grana X, Reddy EP. Cell cycle control in mammalian cells: role of cyclins, cyclin dependent kinases (CDKs), growth suppressor genes and cyclin-dependent kinase inhibitors (CKIs).
Oncogene
,
11
:
211
-9,  
1995
.
15
Bode AM, Dong Z. Mitogen-activated protein kinase activation in UV-induced signal transduction.
Sci STKE
,
167
:
1
-15,  
2003
.
16
Chen W, Bowden GT. Activation of p38 MAP kinase and ERK are required for ultraviolet-B induced c-fos gene expression in human keratinocytes.
Oncogene
,
18
:
7469
-76,  
1999
.
17
Cheng M, Sexl V, Sherr CJ, Roussel MF. Assembly of cyclin D-dependent kinase and titration of p27Kip1 regulated by mitogenactivated protein kinase kinase (MEK1).
Proc Natl Acad Sci USA
,
95
:
1091
-6,  
1998
.
18
Boldt S, Weidle UH, Kolch W. The role of MAPK pathways in the action of chemotherapeutic drugs.
Carcinogenesis
,
23
:
1831
-8,  
2002
.
19
Testa JR, Bellacosa A. AKT plays a central role in tumorigenesis.
Proc Natl Acad Sci USA
,
98
:
10983
-5,  
2001
.
20
Altieri DC. The molecular basis and potential role of survivin in cancer diagnosis and therapy.
Trends Mol Med
,
7
:
542
-7,  
2001
.
21
Matsumura Y, Ananthaswamy HN. Short-term and long-term cellular and molecular events following UV irradiation of skin: implications for molecular medicine.
Expert Rev Mol Med
,
2002
:
1
-22,  
2002
.
22
Ouhtit A, Muller HK, Davis DW, Ullrich SE, McConkey D, Ananthaswamy HN. Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin.
Am J Pathol
,
156
:
201
-7,  
2000
.
23
Lu YP, Lou YR, Xie JG, Yen P, Huang MT, Conney AH. Inhibitory effect of black tea on the growth of established skin tumors in mice: effects on tumor size, apoptosis, mitosis and bromodeoxyuridine incorporation into DNA.
Carcinogenesis
,
18
:
2163
-9,  
1997
.
24
Smith ML, Fornace AJ, Jr. p53-mediated protective responses to UV irradiation.
Proc Natl Acad Sci USA
,
94
:
12255
-7,  
1997
.
25
Liu M, Pelling JC, Ju J, Chu E, Brash DE. Antioxidant action via p53-mediated apoptosis.
Cancer Res
,
58
:
1723
-9,  
1998
.
26
Lu YP, Lou YR, Li XH, et al Stimulatory effect of oral administration of green tea or caffeine on ultraviolet light-induced increases in epidermal wild-type p53, p21 (WAF1/CIP1), and apoptotic sunburn cells in SKH-1 mice.
Cancer Res
,
60
:
4785
-91,  
2000
.
27
Lu YP, Lou YR, Xie JG, et al Topical applications of caffeine or (−)-epigallocatechin gallate (EGCG) inhibit carcinogenesis and selectively increase apoptosis in UVB-induced skin tumors in mice.
Proc Natl Acad Sci USA
,
99
:
12455
-60,  
2002
.
28
Liebermann DA, Hoffman B, Steinman RA. Molecular controls of growth arrest and apoptosis: p53-dependent and independent pathways.
Oncogene
,
11
:
199
-210,  
1995
.
29
Cotton J, Spandau DF. Ultraviolet B-radiation dose influences the induction of apoptosis, and p53 in human keratinocytes.
Radiat Res
,
147
:
148
-55,  
1997
.
30
Ziegler A, Jonason AS, Leffell DJ, et al Sunburn and p53 in the onset of skin cancer.
Nature
,
372
:
773
-6,  
1994
.
31
Tron VA, Trotter MJ, Tang L, et al p53-regulated apoptosis is differentiation dependent in ultraviolet B-irradiated mouse keratinocytes.
Am J Pathol
,
153
:
579
-85,  
1998
.
32
Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression.
Genes Dev
,
13
:
1501
-12,  
1999
.
33
Morgan DO. Principles of CDK regulation.
Nature
,
374
:
131
-4,  
1995
.
34
Fotedar R, Bendjennat M, Fotedar A. Role of p21WAF1 in the cellular response to UV.
Cell Cycle
,
3
:
134
-7,  
2004
.
35
Polyak K, Kato JY, Solomon MJ, et al p27Kip1, a cyclin–CDK inhibitor, links transforming growth factor-β and contact inhibition to cell cycle arrest.
Genes Dev
,
8
:
9
-22,  
1994
.
36
Seger R, Krebs EG. The MAPK signaling cascade.
FASEB J
,
9
:
726
-35,  
1995
.
37
Marshall C. How do small GTPase signal transduction pathways regulate cell cycle entry?.
Curr Opin Cell Biol
,
11
:
732
-6,  
1999
.
38
Ichijo H. From receptors to stress-activated MAP kinases.
Oncogene
,
18
:
6087
-93,  
1999
.
39
Assefa Z, Vantieghem A, Garmyn M, et al p38 mitogen-activated protein kinase regulates a novel, caspase-independent pathway for the mitochondrial cytochrome c release in ultraviolet B radiation-induced apoptosis.
J Biol Chem
,
275
:
21416
-21,  
2000
.
40
Chen W, Bowden GT. Role of p38 mitogen-activated protein kinases in ultraviolet-B irradiation-induced activator protein 1 activation in human keratinocytes.
Mol Carcinog
,
28
:
196
-202,  
2000
.
41
Muise-Helmericks RC, Grimes HL, Bellacosa A, Malstrom SE, Tsichlis PN, Rosen NJ. Cyclin D expression is controlled post-transcriptionally via phosphatidylinositol 3-kinase/Akt-dependent pathway.
J Biol Chem
,
273
:
29864
-72,  
1998
.
42
Collado M, Medema RH, Garcia-Cao I, et al Inhibition of the phosphoinositide 3-kinase pathway induces a senescence-like arrest mediated by p27Kip1.
J Biol Chem
,
275
:
21960
-8,  
2000
.
43
Gottlieb TM, Leal JF, Seger R, Taya Y, Oren M. Cross-talk between Akt, p53 and Mdm2: possible implications for the regulation of apoptosis.
Oncogene
,
21
:
1299
-303,  
2002
.
44
Bachelder RE, Ribick MJ, Marchetti A, et al p53 inhibits α 6 β 4 integrin survival signaling by promoting the caspase 3-dependent cleavage of AKT/PKB.
J Cell Biol
,
147
:
1063
-72,  
1999
.
45
Altieri DC. Survivin, versatile modulation of cell division and apoptosis in cancer.
Oncogene
,
22
:
8581
-9,  
2003
.
46
Choi KS, Lee TH, Jung MH. Ribozyme-mediated cleavage of the human survivin mRNA and inhibition of antiapoptotic function of survivin in MCF-7 cells.
Cancer Gene Ther
,
10
:
87
-95,  
2003
.
47
Pennati M, Binda M, Colella G, et al Radiosensitization of human melanoma cells by ribozyme-mediated inhibition of survivin expression.
J Investig Dermatol
,
120
:
648
-54,  
2003
.
48
Allen SM, Florell SR, Hanks AN, et al Survivin expression in mouse skin prevents papilloma regression and promotes chemical-induced tumor progression.
Cancer Res
,
63
:
567
-72,  
2003
.
49
Green D, Kroemer G. The central executioners of apoptosis: caspases or mitochondria?.
Trends Cell Biol
,
8
:
267
-71,  
1998
.