Sesquiterpene lactones isolated from indigenous Middle Eastern plants inhibit tumor promoter-induced transformation of JB6 cells

Primary mouse keratinocytes (PMKs) were freshly prepared from one- to-two day-old neonatal BALB-c mice as described by Yuspa et al.[31]. The SP-1 benign tumor cell lines were produced in SENCAR mice [31]. The neoplastic PAM212 cell line is a differentiated squamous cell carcinoma (SCC) that spontaneously transformed in vitro[32]. I7 is a spindle cell line derived from a skin carcinoma formed from PMKs infected with the v-rasHa and c-fos oncogenes and grafted to nude mice [32]. PAM212, SP1, and I7 cell lines were generously provided by Dr. Stuart H. Yuspa (NIH, Bethesda, MD). The JB6P + cell line is a tumor promoter-sensitive clonal variant (clone 41, subclone 5a), derived from the JB6 model for tumor promotion, and originally derived from primary mouse epidermal cells [33]. The JB6P + cell line was generously provided by Dr. Nancy Colburn (NCI, Frederick, MD).

SP1, PAM212, and PMK cells were cultured in fresh Eagle Minimum Essential Medium (EMEM) (Bio Whittaker, Cambrex Co., MD) containing 10% chelated fetal bovine serum (FBS) with no more than 0.05 mM Ca⁺⁺ to maintain a basal proliferating cell phenotype [34], 1% L-glutamine, and 1% penicillin-streptomysin antibiotics (Gibco-BRL Life Technologies, Carlsbad, CA). I7 cells were cultured in complete EMEM medium with 10% FBS, 2 mM L-glutamine, and 1% penicillin-streptomysin. JB6P + cells were cultured in EMEM (SIGMA, M2279) containing 4% heat inactivated FBS (Gibco BRL Life), 2 mM L-glutamine, 25 μ g/mL of gentamicin (SIGMA SG 1397 M10) and 1% non-essential amino acids (NEAA) (Gibco). JB6P + cells were used up to ten passages in culture to avoid spontaneous transformation in vitro. All cells were grown in a humidified incubator which was set at 95% air and 5% CO₂ except for PMKs which were grown in 93% air and 7% CO₂.

Extraction, purification, and identification of the SL β-tan and Sal A from Achillea falcata and Centaurea ainetensis, respectively, were performed as previously described [11, 14]. Briefly, the plant material was soaked in methanol and then subjected to filtration and several fractionation steps where the different fractions were subjected to bio-guided fractionation. The sub-fractions with the most potent anti-proliferative activities were further purified, and the pure bioactive compounds, Sal A from Centaurea ainetensis and β-tan from Achillea falcata were identified using ¹ H and ¹³ C NMR identified using several spectroscopic techniques including 1D and 2D NMR as well as mass spectrometry, UV, and IR. β-tan and Sal A were prepared from a stock of 20 mg/ml diluted in absolute ethanol. Cells were treated with the indicated concentrations of β-tan and Sal A. For the control conditions, concentrations of ethanol in culture medium did not exceed 0.1% which had no effect on the growth of cells (data not shown).

Cell growth was assayed at indicated time points using the MTT Cell Proliferation Kit according to manufacturer’s instructions (Roche Diagnostics). The proliferation assay is an MTT-based method which measures the ability of metabolically active cells to convert tetrazolium salt into a blue formazan product, the absorbance of which is recorded at 595 nm using an ELISA microplate reader. Cell growth results were expressed as percentage of control and were derived from the mean of triplicate wells.

Cells were seeded in 96-well plates, at a density of 1 x 10⁵ cells/ml in 100 μl media, and incubated until confluency reached 50%. After which the media was removed and 100 μl of fresh media containing different concentrations of β-tan or Sal A were placed for treatment conditions, or a maximum of 0.1% ethanol in media for control conditions. For MTT assays using the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) (Enzo Life Sciences, USA), JB6P + cells were treated with either 5 nM TPA [35] in media only, or with the indicated concentrations of β-tan or Sal A with or without 5 nM TPA co-treatment.

Colony growth in soft agar is a well-established index of cell transformation [24]. Anchorage-independent growth was studied using the CytoSelect^TM 96-Well Cell Transformation Assay kit (Cell Biolabs) according to manufacturer’s instructions. The base agar layer (0.6% agar, 10% FBS, 2 mM L-glutamine and 25 μg/ml gentamicin) was layered into wells of a 96-well plate and allowed to solidify. Once solidified, the cell agar layer containing 0.4% agar with JB6P + cells treated with the indicated concentrations of β-tan and Sal A, with 5 nM TPA [23, 35, 36], in complete EMEM (10% FBS), was layered on top of the base agar layer. The indicated concentrations of β-tan and Sal A were then prepared in complete EMEM (10% FBS), with 5 nM TPA and placed over the solidified cell agar layer. The cells were incubated for 9 ± 1 day at 37°C and 5% CO₂, replenished with the indicated concentrations of β-tan and Sal A with 5 nM TPA every 3 days. Colonies were photographed and then quantified using the CyQuant GR Dye where the fluorescence was measured using a 96-well fluorometer set at a 485/520 nm filter set.

JB6P + cells were seeded in 24-well plates (1 x 10⁵ cells/ml), and at 60–80% confluency, cells were co-transfected with the AP-1 (pXP2-35alb-Luc, 0.8 μg) or NF-κB (pGL2-IL-6–Luc, 0.8 μg) firefly luciferase reporter plasmids with the renilla luciferase reporter plasmid (pRL-SV40, 0.04 μg). The pXP2-35alb-Luc harbors the albumin promoter upstream from the luciferase gene. Within this promoter, the GCN4 oligo sequence, which harbors the AP-1 binding site, was ligated. The pGL2-IL-6–Luc uses the IL-6 promoter region containing four putative NF-κB binding sites. These reporter plasmids were kindly provided by Dr. Nancy Colburn (NCI). Co-transfection was done using Lipofectamine^TM 2000 with PLUS^TM reagent (Invitrogen), without antibiotics for 3 h at 37°C, 5% CO₂, then replenished with complete EMEM (4% heat-inactivated FBS, with antibiotics) for at least 12 h. Cells were then treated with the indicated concentrations of β-tan and Sal A, with or without 16 nM TPA for 24 h as described [35]. Cell lysates were then prepared and luminescence measured using the Dual Luciferase Reporter Assay Kit (Promega) as per manufacturer’s instructions. The firefly reporter transfection efficiencies were normalized relative to the renilla luciferase activity generated by this vector and plotted as percentage of control.

JB6P + cells were plated in 100 mm dishes at a density of 50,000 cells/ml. At 80–90% confluency, cells were starved with 0.1% FBS for 24 h, then were pre-treated with either 10 μg/ml β-tan or 15 μg/ml Sal A for 1 hr followed by 15 min or 6 h 32 nM TPA [35, 37]. Whole cell protein extracts (30 μg) were prepared as described [10] and probed overnight at +4 °C with primary antibodies against MMP-9 (Chemicon, Millipore) MMP-2 (Chemicon, Millipore), GAPDH, IκBα, cyclin D₁, p16, Bax and Bcl-2 (Santa Cruz Biotechnology, Inc.) followed by secondary antibodies conjugated with horseradish peroxidase. Equal protein loading and quality were verified through GAPDH reprobing and Ponceau staining of membranes. The immunocomplexes were visualized using enhanced chemiluminescent kits obtained from Santa Cruz (ECL system). Bands were quantified using ImageQuant software and the Molecular Dynamics 860 System (Molecular Dynamics, Sunnyvale, CA). In some western blots, adjustments of brightness and contrast were applied to all bands of the same membrane image.

We have previously shown, in a murine in vitro model of epidermal carcinogenesis, that Sal A selectively inhibits the cell growth of papilloma and SCC cell lines without significantly affecting the growth of normal cells [10]. Here, we characterized the growth-inhibitory effects of β-tan in vitro using an MTT-based assay. In this model, the primary mouse keratinocytes (PMKs) are representatives of normal cells, the SP-1 cell line as benign tumor cells, PAM-212 cell line as SCC, and the spindle I7 cells as aggressive and metastasizing tumor cells. Treatment with β-tan caused a dose-dependent growth inhibition at 24 h, where a concentration of 10 μg/ml decreased cell growth significantly by 49 ± 7% (p < 0.01) in PAM 212 cells compared to a 6 ± 1% decrease in PMKs cell growth (Figure 2A). The benign SP-1 cells and spindle I7 cells appeared to be less sensitive at this concentration, showing a 26 ± 10% and 30 ± 4% decrease, respectively, that were not significantly different than the normal PMKs (Figure 2A). We have previously performed similar experiments on Sal A and found that 10 μg/ml is selective for tumor cells [10]. In this study, we used this same concentration (10 μg/ml) to study the effect of both β-tan and Sal A on JB6P + cell growth and transformation. β-tan and Sal A produced a dose-dependent growth inhibition in JB6P + cells (Figure 2B). Treatment with 10 μg/ml β-tan and Sal A inhibited JB6P + cell growth by a significant 74 ± 7% and 51 ± 4% (p < 0.01), respectively (Figure 2B). These results show that at low concentrations, both molecules preferentially inhibited the growth of JB6P + cells versus normal keratinocytes, eliminating the possibility that the anti-tumor promoting effects of β-tan and Sal A is due to drug cytotoxicity.

We investigated the anti-tumor promoting properties of β-tan and Sal A in JB6P + cells. Tumor promoters, such as the phorbol ester 12-O- tetradecanoylphorbol-13-acetate (TPA), increase JB6P + cell growth and transformation. Treatment of JB6P + cells with TPA alone significantly increased their growth at 48 h by approximately 160 ± 7% relative to control (p < 0.001) (Figure 3A). However, co-treatment with β-tan or Sal A with TPA for 48 h inhibited tumor promoter-induced proliferation of JB6P + cells (Figure 3A). β-tan treatment for 48 h at 1 or 2.5 μg/ml did not cause a significant growth inhibition of JB6P + cell proliferation compared to control treated cells (p > 0.05) (Figure 3A). However, co-treatment of 2.5 μg/ml β-tan with TPA showed a significant (p < 0.001) inhibition of TPA-induced proliferation, by 28 ± 10%, relative to the TPA-treated cells; whereas, co-treatment of 1 μg/ml β-tan with TPA showed no significant inhibition on TPA-induced proliferation (p > 0.05) (Figure 3A). β-tan concentrations of 5 and 10 μg/ml had a significant growth inhibitory effect after 48 h on JB6P + cells (by 70 ± 3% and 80 ± 3% respectively) relative to control (p < 0.001), and when co-treated with TPA, cell proliferation was significantly decreased (p < 0.001) (Figure 3A).

Treatment with Sal A at 5 μg/ml had no growth inhibitory effect in JB6P + cells while this concentration caused a significant inhibition of TPA-induced proliferation by 33 ± 20% relative to the TPA-treated cells (p < 0.001) (Figure 3A). Higher concentrations of Sal A at 10 or 15 μg/ml caused a significant 63 ± 3% and 65 ± 1% decrease in cell proliferation, respectively, with or without the presence of TPA (p < 0.001) (Figure 3A). These results indicate that both SL molecules reduced tumor promoter-induced proliferation of JB6P + cells at concentrations that did not affect the growth of normal cells.

To test whether these two SL molecules inhibit tumor promoter-induced cell transformation, we determined their effects on anchorage-independent cell growth in soft agar, which is a hallmark of malignant transformation. In the presence of tumor promoters, the immortalized but non-tumorigenic JB6P + cells become tumorigenic, forming colonies in an anchorage-independent manner [23]. JB6P + cells treated with only TPA, but not solvent control, exhibit colony growth in soft agar (Figure 3B, C). Importantly, upon co-treatment of β-tan or Sal A with TPA, colony formation was inhibited in a concentration-dependent manner in JB6P + cells (Figure 3B, C). At 0.25 μg/ml, neither β-tan nor Sal A decreased JB6P + colony growth 9 ± 1 day after seeding; however, at 2.5 μg/ml concentrations, which were non-cytotoxic to normal and JB6P + cells by MTT (Figure 2), β-tan and Sal A significantly inhibited tumor promoter-induced colony formation by 66 ± 8% and 51 ± 8%, respectively (p < 0.001) (Figure 3B, C). Both SL molecules completely abrogated colony growth 9 ± 1 day post-seeding at 5 μg/ml concentrations. These results show that β-tan and Sal A inhibit tumor promoter-induced JB6P + cell transformation.

Elevated levels of AP-1 and NF-κB activities are hallmarks of malignant transformation [19, 27, 37]. Since β-tan and Sal A both inhibited tumor promoter-induced cell transformation, we hypothesized that these SL molecules mediate their anti-tumor promoting activities by repressing AP-1, NF-κB, or both transcriptional activities.

The application of TPA alone dramatically increased AP-1 and NF-κB luciferase activities in JB6P + cells by four- and approximately two-fold, respectively, compared to control (Figure 4). We tested the effects of β-tan and Sal A on TPA-induced AP-1 and NF-κB transcriptional activities for 24 hours, using 5 μg/ml concentrations as these completely abrogated colony formation with minimal effects on primary keratinocyte cell growth. Unexpectedly, at this concentration, β-tan showed a significant 2.5-fold increase in basal AP-1 activity, relative to control (p < 0.01) and did not decrease TPA-induced AP-1 activity (Figure 4A). Importantly, 5 μg/ml β-tan showed a significant inhibition of basal and TPA-induced NF-κB activity by 50 ± 4% and 64 ± 4%, respectively, at 24 h (p < 0.001) (Figure 4A).

Sal A (5 μg/ml) did not modulate basal AP-1 activity, but caused a non-statistically significant decrease in TPA-induced AP-1 activity. Interestingly, Sal A significantly decreased basal and TPA-induced NF-κB transcriptional activities at 24 h by 37 ± 6% and 54 ± 5%, respectively (p < 0.01) (Figure 4B). Our experiments show that both β-tan and Sal A decreased basal and tumor promoter-induced NF-κB activities, which in fact is a characteristic property of SL [6].

In JB6 cells, both AP-1 and NF-κB activities are essential for the transformation response, which can be attributed to their roles in the transcriptional activation of genes controlling cellular proliferation, metastasis, angiogenesis, tumor invasion, and apoptosis [38, 39]. We next investigated the effect of β-tan and Sal A on the protein levels of key downstream targets of the AP-1 and NF-κB pathways known to be induced by tumor promoters in cell transformation and tumor progression. These target genes are modulated by tumor promoters at early time points; therefore, we pretreated JB6P + cells for one hour with high concentrations of β-tan (10 μg/ml) and Sal A (15 μg/ml), followed by TPA for 15 minutes or 6 hours. We chose these high concentrations that kill approximately 70% of cells by 24 h to be able to detect early protein changes of key AP-1 and NF-κB target genes. Protein levels of metalloproteinase 9 (MMP-9) were induced by approximately 11-fold in TPA-treated JB6P + cells as early as 15 minutes and were reduced to basal levels and by approximately 50% by pre-treatment with β-tan and Sal A, respectively (Figure 5). On the other hand, MMP-2 protein levels were induced by three-fold in TPA-treated JB6P + cells at 15 minutes but were not reduced by β-tan or Sal A pretreatment. As early as 15 minutes post-TPA treatment, cyclin D₁ protein levels were increased by four-fold, and were slightly decreased upon pretreatment with β-tan (Figure 5). The cyclin-dependent kinase inhibitor (CDKI) p16 was reduced by TPA at 15 minutes and 6 hours, and pretreatment with β-tan or Sal A increased p16 protein levels to control or higher levels by 6 hours (Figure 5). Furthermore, we investigated the changes in pro-apoptotic Bax and anti-apoptotic Bcl-2 proteins upon treatment with β-tan or Sal A in the presence of TPA. These apoptotic regulators are also key target genes for mediating the AP-1 and NF-κB transformation response. An increase in the ratio of pro-apoptotic over anti-apoptotic Bcl-2 proteins leads to an increase in mitochondrial permeability and subsequent release of cytochrome c, an event central to apoptotic activation [40]. Treatment with TPA alone reduced the pro-apoptotic Bax/Bcl-2 protein ratio to 0.3 folds of control as early as 15 minutes (Figure 5). Pre-treatment with β-tan or Sal A restored the Bax/Bcl-2 protein ratio to almost control values at 15 minutes and to more than two- and four-fold of control values at 6 hours post-TPA treatment (Figure 5).

Since both SL molecules inhibited TPA-induced NF-κB transactivation, we next studied their effects on the NF-κB inhibitor, IκBα. Treatment with TPA alone abrogated IκBα protein levels as early as 15 minutes (Figure 5). Interestingly, only pre-treatment with β-tan restored IκBα protein levels after 15 minutes of TPA-treatment. These results indicate that pretreatment with β-tan or Sal A regulate TPA-induced AP-1 and NF-кB target genes that are involved in the regulation of cell growth, cell migration, and metastasis.