Introduction

Thyroid carcinoma is the most common malignancy of the endocrine system, which accounts for the majority of deaths from endocrine cancers.^1, 2 Three types of thyroid carcinomas are derived from thyroid follicular cells: papillary, follicular and anaplastic carcinomas. Differentiated (papillary and follicular) carcinomas have relatively good prognosis, whereas anaplastic (undifferentiated) carcinoma is highly aggressive with a mean survival time of less than 8 months.³ Although most differentiated thyroid carcinomas treated by surgery and ¹³¹I therapy can be cured, 10–20% of patients die from recurrence or progression of advanced differentiated tumors due to lack of effective treatments. Anaplastic carcinoma carries a far graver prognosis. The primary tumor may be debulked by surgery, but most are inoperable when first seen because they are too large or too fixed and may have spread beyond the thyroid bed. Anaplastic tumor cells are not responsive to radioiodine and most modalities of chemotherapy and radiotherapy. Thus, development of novel therapeutic approaches to anaplastic carcinoma becomes necessary and gene therapy may be a viable alternative approach to thyroid carcinoma refractory to conventional treatment.^4, 5

Cytokine gene therapy has been extensively used for the treatment of cancer in experimental animal models and clinical trials.^6, 7 Among many cytokine-based anticancer therapies, interleukin-12 (IL-12) provides one of the most significant antitumor activities in many experimental animal models.⁸ We have previously demonstrated that IL-12 is effective against anaplastic thyroid carcinoma in a mouse model.⁹ The mechanisms leading to tumor regression were mainly dependent on natural killer (NK) cells. However, NK cell-mediated cytotoxicity against IL-12-transfected anaplastic thyroid carcinoma cells ARO (ARO/IL-12) could not be completely neutralized by anti-NK cell antibody, suggesting that other mechanisms of IL-12-mediated antitumor activity may be involved or tumor target structure in ARO/IL-12 cells might be changed as a result of IL-12 expression and such a change may make them more susceptible to be killed.

To further explore the mechanisms of IL-12-mediated antitumor activity, in the present study, we investigated the gene expression profile of ARO and ARO/IL-12 cells by microarray analysis of more than 3000 genes, and found that cannabinoid receptor 2 (CB2) was expressed eightfold higher in ARO/IL-12 cells as compared to ARO cells. We further investigated the effect of CB2 expression in ARO cells on the agonist-induced apoptosis and sensitivity of ARO cells to chemotherapy drug paclitaxel.

Materials and methods

Animals

Female BALB/c (nu/nu) nude mice 6 to 8 weeks of age were used as model hosts for anaplastic thyroid tumors. To establish subcutaneous thyroid tumor, mice were injected subcutaneously with 2 × 10⁶ ARO cells. Palpable tumors developed within 10–14 days. Tumor volume was calculated by the formula: width² × length × 0.5.

Cell line and cell culture

ARO, a human anaplastic thyroid carcinoma cell line, and IL-12- or CB2-transfected ARO were propagated in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum, 2 mm l-glutamine, 100 U ml⁻¹ penicillin and 100 μg ml⁻¹ streptomycin at 37 °C in humidified atmosphere containing 5% CO₂.

RNA extraction, RT-PCR, northern hybridization and microarray analysis

Total RNA was extracted by quanidinium thiocyanate–phenol–chloroform method. The integrity of RNA was verified by denaturing gel electrophoresis, and northern hybridization was carried out using 1.2 kb full-length CB2 cDNA probe as previously described.¹⁰ Clontech’s Altas human cDNA expression array (Atlas Glass Human 3.8 II with 3757 genes) was used to compare gene expression profile between ARO and ARO/IL-12 cell lines according to the manufacturer’s procedure. Expression of IL-12 receptor B1 (IL-12RB1) and B2 (IL-12RB2) subunits were determined by reverse transcription (RT)-PCR analysis. Primer sequences and PCR amplification were as follows: IL-12RB1 sense: 5′-TCTTCCTCTTCCTGCTGTCC-3′, antisense: 5′-CCTCATACTGCCAGGAGCACTC-3′, 94 °C for 1 min, 48 °C for 1 min, 72 °C for 1 min, 35 cycles; IL-12RB2 sense: 5′-CATGGAACAGCATTCCAGTCCA-3′, antisense: 5′-AATGCTTGGTGCCACAAACGCC-3′, 94 °C for 1 min, 48 °C for 1 min, 72 °C for 1 min, 35 cycles.

Quantitative real-time RT-PCR analysis

Total RNAs from ARO, ARO/IL-12 and ARO stimulated with 10 ng ml⁻¹ IL-12 for 16 h were isolated as described above. Total RNA (2 μg each) was reverse-transcribed using Promega reverse transcription system (Promega, Madison, WI). LightCycler DNA Master SYBR Green 1kit was used for quantitative real-time PCR analysis according to the manufacturer’s protocols (Roche, Mannheim, Germany). The cDNA mix was diluted 10-fold and 2 μl of the dilution was used for real-time PCR analysis. PCR primers for 290 bp CB2 cDNA fragment were as follows: 5′-TCAGTGGAATCTGAAGGGCC-3′ (sense) and 5′-ATGCAAAGACCACACTGGCCA-3′ (antisense). The mRNA level of housekeeping gene glyceraldehyde-3-phosphate dehydrogenase was used as an internal control and a 300-bp PCR product was amplified using the following two primers: 5′-ACAGTCAGCCGCATCTTCTT-3′ (sense) and 5′-TTGATTTTGGAGGGATCTCG-3′ (antisense). The PCR conditions were as follows: 95 °C for 30 s followed by 35 cycles of amplification (95 °C for 5 s, 48 °C for 5 s and 72 °C for 10 s). The resulting concentration of CB2 PCR products was normalized by comparison with glyceraldehyde-3-phosphate dehydrogenase and was used to determine the CB2 mRNA level in both ARO, ARO/CB2 or ARO/IL-12 cell lines.

Confocal microscopy

Both ARO and ARO/IL-12 cells grown in coverslips were fixed in 3.7% (vol/vol) formaldehyde. CB2 was visualized by indirect immunofluorescence using polyclonal rabbit anti-CB2 antibody (Cayman, MI) and TRITC-conjugated goat anti-rabbit secondary antibody (Pierce, Rockford, IL). Confocal microscopy images were quantified using the Image Analyzing program provided by Carl Zeiss, Germany.

Cloning and expression of CB2 in ARO cells

The full-length CB2 cDNA was amplified by RT-PCR from ARO/IL-12 cells using the following two primers: 5′-TCAGTGGAATCTGAAGGGCC-3′ (sense) and 5′-TTAAATTGGGAAGAGGCCTC-3′ (antisense). The resulting 1.1-kb cDNA fragment was cloned into pcDNA 3.1 expression vector under the control of cytomegalovirus promoter (Invitrogen, Carlsbad, CA). ARO cells were transfected with either CB2/CMVpDNA or vector alone as previously described.⁹ Cells were selected in the presence of 400 μg ml⁻¹ G418 for 3 weeks (Sigma, St Louis, MO).

Cell proliferation and viability assay

ARO, ARO/IL-12 and ARO/CB2 cells were seeded onto 96-well plate with a concentration of 1 × 10⁴ cells per well in triplicate. Cell proliferation was determined after 24 h by bromodeoxyuridine incorporation assay using Promega’ BrdU cell proliferation assay kit (Promega, Madison, WI). For cell viability assay, 1 × 10⁴ cells per well in triplicate were treated with different concentrations of JWH133, WIN-55,212-2 or paclitaxel for 24 h at 37 °C. Cell viability was determined using Promega’s CellTiter-Glo Luminescent cell viability assay kit.

Flow cytometry analysis for apoptosis

Cells (1 × 10⁵) were double stained with fluorescein isothiocyanate-conjugated annexin V and propidium iodide for 15 min at room temperature using Molecular Probes’s apoptosis detection kit (Invitrogen, CA), and then analyzed on a FACScan flow cytometer (BD Biosciences, San Jose, CA). Annexin V and propidium iodide emissions were detected in the FL-1 and FL-2 channels, respectively.¹¹

Antitumoral action of cannabinoid in vivo

Tumors were induced in nude mice by subcutaneous flank injection of 2 × 10⁶ ARO or ARO/CB2 cells. When tumors reached an average size of 250 mm³, animals were assigned randomly to various groups and received daily intratumoral injection of vehicle or 50 μg ml⁻¹ of CB2 ligand JWH133 in 50 μl of saline supplemented with 5 mg ml⁻¹ bovine serum albumin for 14 days as previously described.¹²

Statistical analyses

Tumor burden and apoptosis among different treatment groups were analyzed by the Student’s t-test. Differences were considered statistically significant when the P-value was <0.05.

Results

CB2 overexpression in ARO/IL-12 cells

We examined gene expression profile of ARO and ARO/IL-12 by microarray analysis of 3757 genes using Atlas Glass Human 3.8 II microarray. The most highly expressed gene was CB2, which was expressed eightfold higher in ARO/IL-12 cells than in ARO cells. To validate the microarray data, we performed northern blot, quantitative PCR and confocal microscopy analysis of CB2 expression in both ARO and ARO/IL-12 cells. As shown in Figure 1, CB2 was overexpressed in ARO/IL-12 cells. Furthermore, CB2 expression could be induced in ARO cells by IL-12 (threefold increase as compared to control), and IL-12 receptor B1 and B2 subunits were expressed in both ARO and ARO/IL-12 cells (Figure 1). Interestingly, IL-12 receptor B2 was expressed higher in ARO/IL-12 cells than ARO cells (Figure 1D). IL-12 receptor B2 gene has been reported to function as a tumor suppressor in human B-cell malignancies.¹³

Effect of CB2 overexpression on the induction of apoptosis by CB2 agonists

To study the effect of CB2 expression on the induction of apoptosis by CB2 agonist in ARO cells, and eliminate the effect of IL-12 expression on apoptosis in ARO/IL-12 cells, we cloned the entire coding region of CB2 cDNA into pcDNA 3.1 expression vector and transfected into ARO cells (ARO/CB2). The transfected cells were selected and stable clones were pooled for apoptosis study. Quantitative real-time RT-PCR analysis showed that the level of CB2 expression in ARO/CB2 cells was 6.7±0.42-fold higher than that in ARO cells transfected with vector alone (data not shown). Two CB2 agonists JWH133 and WIN-55,212-2 were used for induction of apoptosis in ARO, ARO/vector, ARO/CB2 and ARO/IL-12 cells. JWH133 was a potent CB2-selective agonist with approximately 200-fold selective over CB1 receptors. WIN-55,212–2 acted on both CB1 and CB2 with approximately 20-fold selective over CB1 receptors. As shown in Figure 2, both JWH133 and WIN-55,212–2 treatments resulted in a dose-dependent increase of apoptosis in ARO/CB2 and ARO/IL-12 cells as compared to ARO or ARO/vector cells (P<0.01). There was also a twofold increase in spontaneous apoptosis in ARO/IL-12 cells.

Regression of ARO/CB2 tumor following JWH133 treatment

We next investigated whether JW133 can reduce tumor growth in nude mice. ARO, ARO/vector and ARO/CB2 tumors received daily intratumoral injection of 50 μg ml⁻¹of JWH133 or vehicle for 3 weeks. Tumor load was measured 60 days after inoculation. As shown in Figure 3, there was no significant difference in tumor load among mice bearing ARO, ARO/vector or ARO/CB2 tumors treated with vehicle only or ARO tumors treated with JWH133. However, the ARO/CB2 tumor load was significantly reduced: 2.34±0.57 vs 4.47±0.94 g in the mice bearing ARO tumors. The significant reduction in tumor load in ARO/CB2 group was statistically significant (P<0.01). The effect of CB2 expression on ARO cell proliferation was also examined. We did not find any significant difference between ARO and ARO/CB2 cells (data not shown).

Enhancement of paclitaxel-induced apoptosis in ARO/IL-12 and ARO/CB2 cells

Paclitaxel is an anticancer agent and inhibits cell cycle progression by accumulating cells in M phase. We were interested in knowing whether CB2 or IL-12 expression can enhance the effect of paclitaxel. Therefore, the apoptosis of ARO, ARO/vector, ARO/CB2 and ARO/IL-12 cells was analyzed before and after paclitaxel treatment by annexin V and PI staining. As shown in Figure 4, higher level of apoptosis (twofold) was found in ARO/CB2 and ARO/IL-12 cells after 15 nm paclitaxel treatment as compared to ARO cells. These results indicated that induction of CB2 or IL-12 expression could enhance cytotoxicity of paclitaxel.

Discussion

We have demonstrated for the first time that CB2 expression is induced following IL-12 expression in ARO cell line. The overexpression of CB2 renders the cells more susceptible to CB2 agonist-mediated apoptosis and regression of the tumor. Furthermore, we have shown significant increase in apoptosis in ARO/IL-12 and ARO/CB2 cells following paclitaxel treatment.

Tumor formation and growth depend mainly on the inability of the organism to elicit a potent immune response, and on the formation of new blood vessels that enable tumor nutrition. IL-12 gene therapy can target both processes and result in tumor regression. Indeed, we have found significant thyroid tumor regression following IL-12 gene transfer in nude mice.⁹ The IL-12-mediated CB2 induction may also contribute to thyroid tumor regression. Since there is no significant reduction of tumor growth in nude mice inoculated with ARO/CB2 cells as compared to ARO cells, the endogenous CB2 agonists may not play a significant role in the tumor regression, even though endogenous mixed CB1/CB2 agonist anandamide was reported to inhibit the proliferation of human breast cancer cell lines MCF-7 and EFM-19 in vitro. The most plausible explanation may be the change of tumor target structure in ARO cells following IL-12 or CB2 expression, which makes it more easily to be killed by IL-12 or other agent-mediated cytotoxicity. This hypothesis is supported by the increased sensitivity to paclitaxel-induced apoptosis in ARO/CB2 and ARO/IL-12 cells. Cannabinoids exert pro-apoptotic actions in tumor cells via the cannabinoid receptor. It has been reported that CB2 receptor can induce apoptosis in human leukemia cells via p38 mitogen-activated protein kinase (MAPK) activation and a ceramide-dependent stimulation of the mitochondrial intrinsic pathway.^14, 15 The apoptosis induced by paclitaxel also depends on the activation of MAPK pathway. Taxol treatment strongly activated ERK and p38 MAPKs in MCF7 breast cancer cells.¹⁶ Therefore, the increase in sensitivity to paclitaxel-induced apoptosis in ARO/CB2 and ARO/IL-12 cells is likely due to increased CB2 expression and activation of MAPK pathway.

Numerous reports have indicated that cannabinoids, the active components of Cannabis sativa (marijuana) and their derivatives, can inhibit cancer cell growth, angiogenesis, invasion and metastasis by modulating key cell-signalling pathways, such as MAPKs and phosphatidylinositol 3-kinase pathways,^{17, 18, 19, 20} both of which have been involved in thyroid carcinoma.²¹ For example, multiple genetic defects were found in the RET/PTC(TRK)–RAS–BRAF–MEK–MAPK kinase pathway. The most frequent one was BRAF mutation: 44% found in papillary thyroid carcinoma and 24% in anaplastic thyroid carcinoma.²² BRAF mutation would lead to increased phosphorylated pERK protein expression. Recent study has shown that Δ9-tetrahydrocannabinol, the major psychoactive ingredient of marijuana, can downregulate genes involved in this pathway, resulting in decreased phosphorylated pERK protein expression.²³ Although the evidence for its medical use is compelling, its legal or licensed use in medicine is still a controversial issue in most countries due to widespread illegal use of cannabis as a recreational drug. There are two cannabinoid-specific receptors: CB1 and CB2. CB1 receptors are predominantly expressed in the brain, whereas CB2 receptors are primarily found in the immune system. To overcome drug addiction and abuse, several groups have developed new synthetic and non-habit-forming cannabinoid receptor agonists and have demonstrated significant antitumor effect in many types of cancer.^{12, 24, 25, 26, 27, 28, 29} More recently, Herrera et al.³⁰ have shown that the apoptosis induced by CB2 activation is through ceramide-dependent activation of the mitochondrial intrinsic pathway.¹⁴ Interestingly, overexpression of CB1 and CB2 has been reported to correlate with improved prognosis of patients with hepatocellular carcinoma.

The discovery of IL-12-induced CB2 overexpression in thyroid cancer cells may offer a new target for anaplastic thyroid cancer treatment: activation of CB2 would trigger apoptosis of thyroid cancer cells and tumor regression. Because CB2 agonists lack psychotropic effects, they may serve as novel anticancer agents to target and kill thyroid cancer cells.

References

1. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A National Cancer Data Base report on 53 856 cases of thyroid carcinoma treated in the US 1985–1995 [see comments]. Cancer 1998; 83: 2638–2648. | Article | PubMed | ISI | ChemPort |
2. Farid NR, Shi Y, Zou M. Molecular basis of thyroid cancer. Endocr Rev 1994; 15: 202–232. | Article | PubMed | ISI | ChemPort |
3. Are C, Shaha AR. Anaplastic thyroid carcinoma: biology, pathogenesis, prognostic factors, and treatment approaches. Ann Surg Oncol 2006; 13: 453–464. | Article | PubMed |
4. Schott M, Scherbaum WA. Immunotherapy and gene therapy of thyroid cancer. Minerva Endocrinol 2004; 29: 175–187. | PubMed | ChemPort |
5. Barzon L, Pacenti M, Boscaro M, Palu G. Gene therapy for thyroid cancer. Expert Opin Biol Ther 2004; 4: 1225–1239. | Article | PubMed | ChemPort |
6. Nanni P, Forni G, Lollini PL. Cytokine gene therapy: hopes and pitfalls. Ann Oncol 1999; 10: 261–266. | Article | PubMed | ChemPort |
7. Wysocki PJ, Karczewska-Dzionk A, Mackiewicz-Wysocka M, Mackiewicz A. Human cancer gene therapy with cytokine gene-modified cells. Expert Opin Biol Ther 2004; 4: 1595–1607. | Article | PubMed | ChemPort |
8. Sangro B, Melero I, Qian C, Prieto J. Gene therapy of cancer based on interleukin 12. Curr Gene Ther 2005; 5: 573–581. | Article | PubMed | ChemPort |
9. Shi Y, Parhar RS, Zou M, Baitei E, Kessie G, Farid NR et al. Gene therapy of anaplastic thyroid carcinoma with a single-chain interleukin-12 fusion protein. Hum Gene Ther 2003; 14: 1741–1751. | Article | PubMed | ChemPort |
10. Zou M, Famulski KS, Parhar RS, Baitei E, Al-Mohanna FA, Farid NR et al. Microarray analysis of metastasis-associated gene expression profiling in a murine model of thyroid carcinoma pulmonary metastasis: identification of S100A4 (Mts1) gene overexpression as a poor prognostic marker for thyroid carcinoma. J Clin Endocrinol Metab 2004; 89: 6146–6154. | Article | PubMed | ISI | ChemPort |
11. Shi Y, Zou M, Collison K, Baitei EY, Al-Makhalafi Z, Farid NR et al. Ribonucleic acid interference targeting S100A4 (Mts1) suppresses tumor growth and metastasis of anaplastic thyroid carcinoma in a mouse model. J Clin Endocrinol Metab 2006; 91: 2373–2379. | Article | PubMed | ChemPort |
12. Sanchez C, de Ceballos ML, del Pulgar TG, Rueda D, Corbacho C, Velasco G et al. Inhibition of glioma growth in vivo by selective activation of the CB(2) cannabinoid receptor. Cancer Res 2001; 61: 5784–5789. | PubMed | ISI | ChemPort |
13. Airoldi I, Di Carlo E, Banelli B, Moserle L, Cocco C, Pezzolo A et al. The IL-12Rbeta2 gene functions as a tumor suppressor in human B cell malignancies. J Clin Invest 2004; 113: 1651–1659. | Article | PubMed | ChemPort |
14. Herrera B, Carracedo A, Diez-Zaera M, Gomez del Pulgar T, Guzman M, Velasco G. The CB2 cannabinoid receptor signals apoptosis via ceramide-dependent activation of the mitochondrial intrinsic pathway. Exp Cell Res2006; 312: 2121–2131. | Article | PubMed | ISI | ChemPort |
15. Herrera B, Carracedo A, Diez-Zaera M, Guzman M, Velasco G. p38 MAPK is involved in CB2 receptor-induced apoptosis of human leukaemia cells. FEBS Lett 2005; 579: 5084–5088. | Article | PubMed | ISI | ChemPort |
16. Bacus SS, Gudkov AV, Lowe M, Lyass L, Yung Y, Komarov AP et al. Taxol-induced apoptosis depends on MAP kinase pathways (ERK and p38) and is independent of p53. Oncogene 2001; 20: 147–155. | Article | PubMed | ISI | ChemPort |
17. Guzman M. Cannabinoids: potential anticancer agents. Nat Rev Cancer 2003; 3: 745–755. | Article | PubMed | ISI | ChemPort |
18. Bifulco M, Laezza C, Pisanti S, Gazzerro P. Cannabinoids and cancer: pros and cons of an antitumour strategy. Br J Pharmacol 2006; 148: 123–135. | Article | PubMed | ChemPort |
19. Bifulco M, Di Marzo V. Targeting the endocannabinoid system in cancer therapy: a call for further research. Nat Med 2002; 8: 547–550. | Article | PubMed | ISI | ChemPort |
20. Blazquez C, Gonzalez-Feria L, Alvarez L, Haro A, Casanova ML, Guzman M. Cannabinoids inhibit the vascular endothelial growth factor pathway in gliomas. Cancer Res 2004; 64: 5617–5623. | Article | PubMed | ISI | ChemPort |
21. Vasko VV, Saji M. Molecular mechanisms involved in differentiated thyroid cancer invasion and metastasis. Curr Opin Oncol 2007; 19: 11–17. | Article | PubMed | ChemPort |
22. Xing M. BRAF mutation in thyroid cancer. Endocr Relat Cancer 2005; 12: 245–262. | Article | PubMed | ISI | ChemPort |
23. Powles T, te Poele R, Shamash J, Chaplin T, Propper D, Joel S et al. Cannabis-induced cytotoxicity in leukemic cell lines: the role of the cannabinoid receptors and the MAPK pathway. Blood 2005; 105: 1214–1221. | Article | PubMed | ChemPort |
24. Casanova ML, Blazquez C, Martinez-Palacio J, Villanueva C, Fernandez-Acenero MJ, Huffman JW et al. Inhibition of skin tumor growth and angiogenesis in vivo by activation of cannabinoid receptors. J Clin Invest2003; 111: 43–50. | Article | PubMed | ISI | ChemPort |
25. Sarfaraz S, Afaq F, Adhami VM, Mukhtar H. Cannabinoid receptor as a novel target for the treatment of prostate cancer. Cancer Res 2005; 65: 1635–1641. | Article | PubMed | ISI | ChemPort |
26. Blazquez C, Carracedo A, Barrado L, Real PJ, Fernandez-Luna JL, Velasco G et al. Cannabinoid receptors as novel targets for the treatment of melanoma. FASEB J 2006; 20: 2633–2635. | Article | PubMed | ChemPort |
27. Carracedo A, Gironella M, Lorente M, Garcia S, Guzman M, Velasco G et al. Cannabinoids induce apoptosis of pancreatic tumor cells via endoplasmic reticulum stress-related genes. Cancer Res 2006; 66: 6748–6755. | Article | PubMed | ISI | ChemPort |
28. McKallip RJ, Lombard C, Fisher M, Martin BR, Ryu S, Grant S et al. Targeting CB2 cannabinoid receptors as a novel therapy to treat malignant lymphoblastic disease. Blood 2002; 100: 627–634. | Article | PubMed | ISI | ChemPort |
29. Aguado T, Carracedo A, Julien B, Velasco G, Milman G, Mechoulam R et al. Cannabinoids induce glioma stem-like cell differentiation and inhibit gliomagenesis. J Biol Chem 2007; 282: 6854–6862. | Article | PubMed | ISI | ChemPort |
30. Xu X, Liu Y, Huang S, Liu G, Xie C, Zhou J et al. Overexpression of cannabinoid receptors CB1 and CB2 correlates with improved prognosis of patients with hepatocellular carcinoma. Cancer Genet Cytogenet 2006; 171: 31–38. | Article | PubMed | ISI | ChemPort |

Acknowledgements

This research project was supported by a grant from King Abdulaziz City for Science and Technology (ARP-24-11). We thank Dr Raafat M El-Sayed from Animal Facility for his excellent support.