Background
Skin cancer is one of the most common types of cancer accounting for at least 40% of cases worldwide and is even more common among Caucasians [
1]. Carcinogenesis is a multistage process and it is characterized by three stages including initiation, promotion and progression. Synthetic or natural agents, which may hinder, halt or reverse the process of carcinogenesis at different stages, would be considered an effective approach in cancer prevention and treatment. Medicinal plants have been reported to inhibit tumor formation in vivo by blocking the initiation phase or slowing down the promotion and progression phases of carcinogenesis [
2‐
4]. Conventional chemotherapy or radiotherapy are often accompanied with adverse side-effects. Therefore, there is an increasing interest in pharmacological evaluation of various natural products that possess anticancer and chemopreventive properties with limited side effects [
5].
Daucus carota (Linnaeus) ssp. carota, known as wild carrot, is a member of the family Umbelliferae (Apiacae). The plant grows in moderate regions throughout the world [
6] and is commonly consumed as a salad in the Mediterranean diet or used as an additive in some recipes [
7]. In Lebanon, it is traditionally used for the treatment of gastric ulcer, diabetes, muscle pain and cancer. The plant has been also reported to exhibit antilithic, diuretic [
6,
8] antibacterial, and antifungal activities [
9,
10]. Previous studies in our laboratory revealed that
Daucus carota oil extract (DCOE) possesses antioxidant [
11], anti-inflammatory and anti-ulcer activities [
12]. It also exhibited potent cytotoxicity against colon (Caco-2, HT-29), breast (MCF-7, MDA-MB-23) [
11] and human acute myeloid leukemia cells [
13], as well as significant anti-tumor promoting effect against DMBA/TPA skin carcinogenesis in mice [
4]. Recently, DCOE was subjected to chromatographic separation and fractions were shown to possess antioxidant, hepatoprotective [
14] as well as anticancer activity in vitro [
14,
15]. The pentane/diethyl ether fraction (F2) containing a major compound identified as 2-himachalen-6-ol (61.4%), was found to suppress proliferation of HT-29 cells by inducing apoptosis [
14]. It also inhibited motility of MDA-MB231 and SF-268 cells and reduced invasion of B16F-10 cells [
15].
Keratinocytes are the most abundant cell type making more than 95% of the skin cells. There are three main types of skin cancer: melanoma, Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). The most common types of cancer in humans are the ones that directly target keratinocytes and include basal cell carcinomas (BCC) and squamous cell carcinomas (SCC) [
16]. In contrast, melanoma is the least common, but the most aggressive type of skin cancer and it is responsible for the majority of skin cancer mortalities. Previous work in our laboratory revealed that DCOE exhibited remarkable antitumor-promoting activity against DMBA/TPA induced squamous cell carcinoma (papilloma) in mice. The first aim of this study, therefore, was to evaluate the in vitro anti-proliferative effect of F2 fraction in human skin carcinogenesis HaCaT model with its three different stages of tumorigenicity (non-tumorigenic, non-invasive and invasive). These cells are considered the counterparts of the chemically induced squamous cell carcinoma in the DMBA/TPA skin carcinogenesis model in mice. The second aim of this study was to further assess the in vivo antitumor efficacy of F2 fraction in DMBA/TPA skin carcinogenesis mouse model.
Methods
Chemicals and reagents
Dimethyl sulphoxide (DMSO), Dulbecco’s modified Eagle’s medium (DMEM), Trypsin, 7,12-Dimethylbenz (a) anthracene (DMBA), 12-O-tetradecanoylphorbol-13-acetate (TPA), were purchased from Sigma-Aldrich, St. Louis, USA. WST-1 reagent was purchased from Roche, Mannheim, Germany. Silica gel 60 was purchased from ACROS organics, New Jersey, USA and Silica gel 40 (35–70) mesh were purchased from Sigma Aldrich, St. Louis, USA. All other chemicals used in the experiments were of analytical grade.
Sample collection and oil extraction
Daucus carota (Linnaeus) ssp. carota mature umbels (closed, yellow-brownish) were collected at the post flowering season between July and August from Byblos, Lebanon. The plant was identified according to the characteristics described in the “Handbook of Medicinal Herbs” [
6] and confirmed by Dr. A. Houri, a Lebanese plant expert at the Lebanese American University. A voucher specimen of the plant material used in this study has been deposited in the Department of Natural Sciences herbarium. The extraction procedure was carried out according to the method described by Zeinab et al. [
4]. Briefly,
Daucus carota umbels were air dried in the shade and then cut into small pieces for oil extraction in methanol/acetone (1:1) for 72 h. The extract was then filtered and evaporated to dryness under reduced pressure. The residue was centrifuged and the oil was dried over anhydrous sodium sulfate. The final yield (3.47%) was stored in a closed amber bottle at 4 °C until use.
DCOE fractionation
DCOE fractionation was carried out as described previously [
14]. Briefly, thirty grams of DCOE were chromatographed on a silica gel column (size 35–70 mesh), eluted with pentane (100%), pentane: diethyl ether (50:50), diethyl ether (100%) and chloroform: methanol (93:7) to yield F1, F2, F3 and F4 fractions respectively. Fractions were analyzed by TLC using hexane:ethyl acetate (70:30) as mobile phase and plates were stained with 2% anisaldehyde. The chemical components of various fractions were determined as previously described [
17].
Animals
All in vivo experiments were performed on male adult BALB/c mice (6 weeks old) bred in the animal facility of the Biology Department at the Lebanese American University. Mice, weighing 18–21 g were housed under optimum conditions of temperature set at 22 ± 2 °C, humidity at 50 ± 5% and an alternating cycle of light and dark. The animals were supplied with standard laboratory chow diet and water. All experimental protocols were approved by the Department of Natural Sciences Animal Ethical Committee, which complies with the Guide for the Care and Use of Laboratory Animals (Committee for the Update of the Guide for the Care and Use of Laboratory Animals, 2010).
Cell lines and culture
The human HaCaT keratinocytes, HaCaT (non-tumorigenic), HaCaT-ras A5 (benign tumorigenic) and HaCaT-ras II4 (malignant) [
18,
19] were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 μg/mL streptomycin and 100 U/mL penicillin. Cells were maintained in a humidified atmosphere containing 5% CO
2 at 37 °C.
Cell proliferation assay
Proliferation of the HaCaT and HaCaT-ras variants was tested using the WST-1 assay. Cells were plated in 96-welled plates at a concentration of 5×104 cell/mL for 24 h. Afterwards, the three cell lines were treated with increasing concentrations (10, 25, 50 and 100 μg/mL) of DCOE fractions dissolved in DMSO for 48 h. Cell survival was assessed using WST-1 reagent (Roche Applied Science, Penzberg, Germany). The color intensity was quantified at 450 nm using Multiskan FC microplate ELISA reader (Thermo fisher Scientific, Rockford, IL, USA).
Cell cycle analysis
The effect of F2 fraction on cell cycle distribution was assessed by flow cytometry after staining of HaCaT-ras A5, and HaCaT-ras II4 cells with propidium iodide (PI). Briefly, both cells lines (1x105cells/mL) were treated with different concentrations (25 and 50 μg/mL) of F2 fraction and cultured in 6-well plates for 48 h. The treated cells were harvested, washed with PBS and fixed with 70% ethanol on ice. The cells were then washed with cold PBS, suspended in 200 μL 1× Propidium Iodide + RNase staining solution and incubate at 37 °C in the dark for 30 min. Propidium Iodide Flow Cytometry Kit (Abcam, Cambridge, UK) was used for cell cycle analysis. The DNA content of the cells was measured by C6 flow cytometer (BD Accuri Cytometers, Ann Arbor, MI USA) and the population of each phase was determined using CFlow Plus analysis software (BD Accuri Cytometers, Ann Arbor, USA).
Western blot
HaCaT-ras II4 cells were treated with two different concentrations of F2 (25 and 50 μg/mL) for 48 h. Adherent and non-adherent cells were collected on ice, washed twice with PBS, lysed with lysis buffer and centrifuged at 12,000 g for 10 min at 4 °C. The cell lysate was heated at 100 °C for 5 min, and the protein content was determined by the Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA).
Equal protein concentrations were subjected to Western blot analysis as described previously [
20]. The PVDF membranes were blocked with blocking buffer (1× TBS, 0.1% Tween-20, 5% skim milk) for 2 h and then probed with primary rabbit polyclonal antibodies to Caspase-3, AKT, p-AKT, ERK and p-ERK (Abcam, Cambridge, UK) and mouse monoclonal antibodies to actin, p53, p21, Bcl-2, BAX (Santa Cruz, CA) at dilution ranging from 1/1000 – 1/5000 at 4 °C overnight. Later, the primary antibodies were washed away with TBST for 2 h and the membranes were treated with horseradish peroxidase (HRP)-coupled secondary antibodies in blocking buffer at a dilution of 1/5000 (Abcam, Cambridge, UK) for 1 h, and washed with TBST afterwards. Finally, detection of each protein was performed using the chemiluminescence ECL kit (Abcam plc, Cambridge, UK). Blot images were obtained with the image lab Software (BioRad, Chemidoc imaging instrument).
The dorsal surface of BALB/c mice was shaved using electric clippers. Papillomas were initiated by a single topical application of 190 nmol of 7,12-Dimethylbenz (a) anthracene (DMBA) in 0.2 mL acetone. Three weeks after initiation, the shaved area of the back of the mice were treated by topically applying 8 nmol TPA in 0.2 mL acetone twice weekly till the end of the experiment [
21]. Before TPA promotion (30 min), three groups of 10 mice each received intraperitoneal injections of F2- fraction (10 mg/kg, 50 mg/kg, 200 mg/kg) dissolved in DMSO. Control group (
n = 10) received an intraperitoneal injections of DMSO. Papilloma development was followed weekly for 21 weeks. Tumor incidence was assessed as the percentage of mice in each group having tumors on their back. The tumor volume was measured at week 15, 17, 20. Tumor volume was estimated in comparison to specific reference volume models. Briefly, papillomas length and width were measured using a caliper and matched with similar clay tumor models of known standardized volumes [
16]. Papilloma smaller than 1 mm in diameter were not counted. Mice in each group were weighed at week 5, 13 and 21.
Statistical analysis
In vitro data were analyzed for statistical significance using one way analysis of variance (ANOVA). Values of the different tested parameters within each group are presented as mean ± SEM. Significant main effect differences were tested using Bonferroni post hoc test for multiple comparisons. All data were analyzed with the statistical package SPSS 18, and differences between groups were considered statistically significant if
p < 0.05. The IC
50 was calculated by a nonlinear regression curve with the use of Prism Graph Pad Prism version 5.0 software for Windows. In the in vivo study, statistical analysis was done using Linear Mixed Effect Models in R (3.2.2) with the function lie from the name package. Two models were built, one for the number of papilloma and another for the papilloma volume. In both models, the time and dosage were considered as fixed effects while the mice id was considered as a random effect. Raw data are available in a supporting Excel (Additional file
1) file presented along with the manuscript.
Discussion
DCOE have been proven lately to possess promising chemopreventive activity against DMBA/TPA-induced skin carcinogenesis [
4]. The present study was carried out to evaluate the effects of DCOE fractions in vitro on different tumorigenic HaCaT cells. Additionally, the fraction (F2) containing the major compound 2-himachalen-6-ol (61.4%) was tested in vivo by using the DMBA/TPA chemical carcinogenesis model. Treatment with the different DCOE fractions caused a significant dose-dependent decrease in viability in all treated cells, with F1 and F2 fractions being the most potent. Interestingly, the obtained results revealed a cytotoxic selectivity of the fractions (2.4 – 3 folds) against the tumorigenic variants, as compared to the non-tumorigenic HaCaT cells. Similar to previous studies with human breast [
4,
20] and colon cancer (HT-29 Caco-2) cells [
14], the F1 and F2 fractions were found to be the most potent ones. The major sesquiterpenes present in F1, the α-humulene and β-caryophyllene, have been reported to be active against various types of human cancer cells [
22‐
24]. On the other hand, the cytotoxicity of the F2 fraction could be primarily due to the presence of the major sesquiterpene 2-himachalene-6-ol which constitutes 61.4%. Flow cytometry analysis on HaCaT-ras II4 and HaCaT-ras A5 cells clearly indicated that F2 treatment induced accumulation of cells in the sub-G1 apoptotic phase and decreases the fraction of cells being in S and G2/M phases. Together this suggests induction of apoptosis and cell cycle arrest by F2. Similar results were obtained on human colon and breast cancer cells after treatment with both F1 and F2 fractions [
14,
20]. To explore the possible mechanisms leading to apoptosis induced by F2 treatment, several components of the apoptotic pathways were examined. The present data reveal that treatment of HaCaT-ras II4 cells with F2 caused an up-regulation of the pro-apoptotic Bax and a decrease in the anti-apoptotic protein Bcl-2, resulting in a lower Bcl2/Bax ratio which favors apoptotic cell death. The anti-apoptotic protein, Bcl-2, has been shown to form a complex with Bax protein, thereby preventing its pro-apoptotic activity [
25]. Bcl2 protein is known to inhibit apoptosis by preventing Bax translocation to mitochondrial membrane and subsequent release of cytochrome c [
26,
27]. The present results also showed that treatment of HaCaT-ras II4 cells with F2 fraction induced a decrease in the level of the inactive caspase-3 indicating its potential cleavage into the activated form. Once activated, the executioner caspase-3 causes the proteolytic cleavage and activation of a wide variety of downstream protein targets [
28]. Accordingly, these results suggest that the induction of apoptosis might be mediated via the mitochondrial pathway.
The induction of cell apoptosis by p53 and p21 (WAF1, Cip-1) is well documented [
29]. Although p21 is a direct p53 target, it is also known to be regulated by several other tumor suppressors [
30]. The current results show that F2 causes an increase in the expression of p21 with a decrease in p53 levels in HaCaT-ras II4 cells. Such observation suggests that the cell cycle arrest and apoptosis in HaCaT-ras II4 cells is mediated through the up-regulation of the expression of p21 in a p53-independent manner. The altered response in p53 expression may be attributed to the presence of mutant p53 in these cells since previous studies conducted in our laboratory have shown that F2 fraction caused p53 up-regulation in HT-29 colon cancer cells [
14]. The present findings are in accordance with previous reports indicating that polyphenols such as curcumin [
31], and other wine polyphenols including caffeic acid, resveratrol and quercetin [
32] down-regulate p53 in cancer cells harboring mutant p53 gene.
Cellular signaling pathways such as the MAPK pathway and PI3K pathway which are involved in controlling cell cycle, cell proliferation and apoptosis are routinely found to be upregulated in cancer cells [
33,
34]. ERK1/2 (MAPK) is a transcription factor that leads to the expression of several gene products which positively regulate cell survival [
35]. It was reported that constitutively activated ERK is responsible for the increased cell growth in various cancers [
35,
36]. Furthermore, activation of ERK 1/2 has also been shown to prevent apoptosis in response to various types of stimuli such as tumor necrosis factor and radiation [
37]. Whereas, down-regulation of ERK activity is known to be associated with apoptosis [
38]. In the current study, treatment of HaCaT-ras II4 cells with F2 fraction significantly inhibited the phosphorylation of ERK. This down regulation of activation was associated with a decrease in cell survival and proliferation. We speculate that the induction of apoptosis by F2 may be mediated partly through the inhibition of the MAPK/ERK pathway. The AKT protein, main downstream effector of PI3K, is a key mediator of cell survival [
39]. AKT activity has been reported previously to be increased in various cancers, including melanoma, colon, brain, breast and prostate [
33,
40,
41]. Additionally, the PI3K/AKT survival pathway inhibits apoptosis and promotes tumor cell survival through different types of stimuli such as growth factor withdrawal and loss of cell adhesion [
42]. The present data revealed that F2 fraction caused a considerable reduction in AKT phosphorylation in the HaCaT-ras II4 cells, reflecting an inhibition of the PI3K/AKT pathway.
The chemopreventive effect of the F2 fraction was evaluated in DMBA/TPA-induced mouse skin carcinogenesis. The results demonstrate that intraperitoneal injection of F2 fraction at the promotion stage caused a significant reduction in papilloma incidence, yield and volume. These results are consistent with previous work conducted in our laboratory using unfractionated DCOE [
4]. In addition, F2 fraction was shown to markedly increase the level of glutathione S-transferase (GST) in CCl4-treated mice [
17], an enzyme involved in the detoxification of carcinogens and in turn protect cellular components against toxic compounds and chemical carcinogenesis [
43]. Therefore, the chemopreventive action of F2 against DMBA/TPA-induced skin carcinogenesis may be partly attributed to this enhanced activity of GST. Earlier studies in our labs have shown that DCOE percentage yield was 3.47% and contained a major compound constituting around 30% of the crude oil content [
4]. In the present study, F2 fraction was shown to be the most active of DCOE fractions and contained around 60% of that major compound. Assuming that it is the major biologically active compound of DCOE, then a 70 kg adult consuming 67 g of the edible umbel would achieve the dose of 10 mg/kg body weight of this compound. The consumption of the whole umbel would provide the consumer with additional anticancer compounds present in F1 fraction and to a lesser extent in F3 and F4 fractions. Since F1 and F2 fraction are the most potent fractions and not water soluble, it is recommended for maximal benefit to consume the whole umbel rather than by infusion form. In the present study, the animals were treated for 21 weeks with doses of 10–200 mg/kg body weight without incidence of death, thus indicating high safety of the F2 fraction. In comparison with a commonly used chemotherapeutic drug, cisplatin, the lowest dose used in the current study is about three times higher. Knowing that it is still a fraction, chromatographic purification of the F2 would most likely lead to the use of a lower and comparable dose to cisplatin.
Acknowledgements
We would like to than Dr. Georges Khazen for his assistance in statistical analysis and Mr. Jean Karam for his help in animal handling.