TRAIL-induced apoptosis of thyroid cancer cells: potential for therapeutic intervention Manzoor Ahmad* ,1 and Yufei Shi 1 1 Department of Biological and Medical Research, MBC-03, King Faisal Specialist Hospital and Research Centre, Riyadh 11211, Saudi Arabia To determine whether the apoptotic machinery of thyroid cancer cells is functional and could be activated for tumoricidal purposes, we examined the apoptosis induced by the cytokines TNF-a, Fas and TRAIL in thyroid cancer cell lines, NPA and SW579. Interestingly, out of these cytokines, only TRAIL was able to trigger signi®cant apoptosis. The tumoricidal eect of TRAIL was further enhanced by CHX, suggesting the presence of CHX-sensitive inhibitor(s) of apoptosis in these thyroid cancer cell lines. The anti-apoptotic proteins like FLAME-1, Bcl-2 and Bcl-xL are believed to be such CHX-sensitive inhibitors in various types of cancer cells. We, however, provide the evidence using NPA and SW579 cell lines that these proteins were not aected by the CHX treatment in thyroid cancer cells. The apoptosis of thyroid cancer cells was mediated by the classical activation of caspases that in turn activated the DNA Fragmentation Factor (DFF-45). To elucidate the role of individual caspases in TRAIL-mediated apoptosis, the inhibitory eects of several general and speci®c tetra- peptide caspase inhibitors were studied. The inhibitors of caspase-1, -6, -8, and -9 as well as general upstream inhibitors of apoptosis could dramatically inhibit TRAIL-induced apoptosis in thyroid cancer cells. Caspase-2 and -3 inhibitors, on the other hand, had no signi®cant eect. When the cells were treated with either agonistic Fas antibody (CH11) or TNF-a, no apoptotic changes were observed. The apoptosis induced by agonistic Fas Ab could be seen only after a prolonged exposure (24 h) to CHX, whereas TNF-a had no eect even in the presence of CHX. The ecacy of TRAIL was also tested on other types of thyroid cancer cells like ARO, FRO (anaplastic carcinoma) and TPC-1 (papil- lary carcinoma) and compared to that triggered by other death inducing cytokines FasL and TNF-a. Again TRAIL was more potent in triggering apoptosis than Fas and TNF-a. Since TRAIL is eective in selectively killing thyroid tumor cells without aecting normal thyrocytes and also does not cause organ toxicity and in¯ammation in vivo, its potential for the treatment of thyroid cancer seems very promising. Oncogene (2000) 19, 3363 ± 3371. Keywords: apoptosis; thyroid cancer; Fas; TNF-a; TRAIL Introduction Thyroid carcinoma has a relatively high incidence in the Middle East where its crude relative frequency has been reported as high as 10.5% in Kuwaiti females while in the US it accounts for only 1.6% of all cancers in women (Farid et al., 1994). Papillary carcinoma accounts for 85% of dierentiated thyroid carcinomas while the remaining epithelial thyroid cell tumors are predominantly follicular carcinomas. As in most solid tumors, the malignant transformation of thyroid epithelial cells is thought to arise through multiple genetic changes involving the activation of proto- oncogenes and the loss of tumor suppressor genes by point mutation, gene rearrangement and deletion (Farid et al., 1994; Cerutti et al., 1996; Sherr, 1996). Some progress has been made to elucidate the molecular mechanisms that determine thyroid tumor development and progression. The genetic alterations of several oncogenes (ras, ret, trk) and tumor suppressor genes (p53, Rb, p16/CDKN2, p21/Waf1) have been identi®ed to be involved in thyroid tumorigenesis (Shi et al., 1996; Tung et al., 1996). It is believed that ras oncogene activation is an early speci®c event for follicular cancers whereas the rearrangements of ret and trk proto-oncogenes are speci®c for papillary carcinoma (Farid et al., 1994). Apoptosis is an intrinsic and fundamental biological process that plays an important role in the develop- ment of multicellular organisms and in maintaining tissue homeostasis (Steller, 1995; Jacobson et al., 1997; Ashkenazi and Dixit, 1998). It is becoming increasingly clear that the deregulation of apoptosis is a critical component of a multi-step tumorigenesis process (Thompson, 1995; Rudin and Thompson, 1997). Three major apoptosis-inducing cytokines have been identi- ®ed so far. These are tumor necrosis factor alpha (TNF-a), Fas ligand (Fas L) and tumor necrosis factor- related apoptosis-inducing ligand (TRAIL), which induce apoptosis through activation of death machin- ery by binding to their cognate death receptors (Ashkenazi and Dixit, 1998; Nagata, 1997; MacFarlane et al., 1997). These receptors share a common domain of 80 amino acids known as the `death domain' and this domain is indispensable for the initiation of signaling cascade that results in activation of aspar- tate-speci®c cysteine proteases now called caspases (Alnemri et al., 1996; Ashkenazi and Dixit, 1998). Activated caspases then bring about the signature changes of apoptosis like membrane blebbing, chro- matin condensation, DNA fragmentation and cyto- plasmic shrinkage by cleaving important cellular proteins (Thornberry and Lazebnik, 1998). Little is known however, whether the apoptotic machinery of Oncogene (2000) 19, 3363 ± 3371 ã 2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00 www.nature.com/onc *Correspondence: M Ahmad, Laboratory of Molecular Apoptosis and Cancer Therapy, Department of Biological and Medical Research, #226, King Faisal Specialist Hospital and Research Center, MBC-03, P.O. Box 3354, Riyadh 11211, Saudi Arabia Received 10 November 1999; revised 2 May 2000; accepted 10 May 2000