(2002) 21, 4702 – 4705 ª 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc

Frequent mutations of Fas in nasal NK/T cell lymphoma

Tetsuya Takakuwa1, Zhiming Dong1, Shinichi Nakatsuka1, Shizuo Kojya2, Yasuaki Harabuchi3, Woo-Ick Yang5, Shigekazu Nagata4 and Katsuyuki Aozasa*,1

1Department of Pathology, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan; 2Department of Otorhinolaryngology, University of Ryukyus, Faculty of Medicine, Okinawa 903-0215, Japan; 3Department of Otorhinolaryngology, Asahikawa Medical College, Asahikawa 078-8510, Japan; 4Department of Genetics, Osaka University, Graduate School of Medicine, Osaka 565-0871, Japan; 5Department of Pathology, Yonsei University College of Medicine, Seoul, Korea

Fas (Apo-1/CD95) is a cell-surface involved in Fas (Apo-1/CD95) is a 45 kDa membrane cell death signaling through binding of Fas . belonging to the receptor family, Mutation of Fas gene in lymphoid cells results in and mediates () accumulation of these cells, which might thus contribute through binding of (FasL) (Suda et al., to lymphomagenesis. We examined the open reading 1993). Fas consists of 325 amino acids with a single frame of Fas cDNA in 14 cases of nasal NK/T-cell transmembrane domain, including signal peptide. The lymphoma. Mutations of Fas gene were detected in seven 80-amino acid portion in the cytoplasm, designated as a (50%) of 14 cases which comprised four frameshift, two death-signaling domain, is essential for the apoptotic missense, and one silent mutations. Frameshift mutations (Itoh and Nagata, 1993). Fas gene were caused by insertion of 1 bp (A) at nucleotide 1095 mutations have been reported both in the lymphoid in two cases and by deletion of 1 bp at nucleotide 597 lineage and epithelial malignancies; about 10% of cases and at 704, respectively, in one each. Mouse T-cell with multiple myeloma (Landowski et al., 1997), 11% lymphoma cells transfected with two missense mutated with sporadic non-Hodgkin’s lymphoma (Gronbaek et and frameshift mutations caused by insertion of al., 1998), 20% with Hodgkin and Reed-Sternberg cells 1 bp (A) at nucleotide 1095 were resistant to apoptosis in classical Hodgkin’s disease (Muschen et al., 2000), induced by the anti-Fas antibody. These findings 11.6% with gastric cancer (Park et al., 2001), 7.7 – 20.3% suggested that accumulation of lymphoid cells with Fas with (Lee et al., 1999b; Boldrini et al., 2001); mutations provides a basis for the development of nasal and 28% with urinary bladder cancer (Lee et al., 1999a). NK/T-cell lymphoma. Recent study showed that the nasal NKTCL Oncogene (2002) 21, 4702 – 4705. doi:10.1038/sj.onc. frequently express FasL together with Fas (Ng et al., 1205571 1999), which might be responsible for local tissue damage resulting in necrosis of tissue around tumors Keywords: nasal NK/T cell lymphoma; Fas; apoptosis; and contribute to the clinical aggressiveness of these mutation tumors. Alternatively, some mechanisms for resistance to Fas/FasL-induced apoptosis might work in the development of nasal NKTCL, thus Fas gene muta- tions could be one of the mechanisms for development Nasal NK/T-cell lymphoma (NKTCL) is a clinical of nasal NKTCL. Therefore, we examined the open condition of lethal midline granuloma that shows reading frame of Fas cDNA in 14 cases of nasal necrotic, granulomatous lesions in the upper respiratory NKTCL. Patients were admitted to the hospitals tract, especially in the nasal cavity (Jaffe et al., 1996). during the period from 1996 to 2000. Age of patients The disease is more frequent in Asian countries than in ranged from 36 to 82 years (median, 63) with a male to Western countries (Aozasa et al., 1992), and was female ratio of 11 : 3. All patients presented with the ONCOGENOMICS demonstrated to be highly associated with Epstein-Barr necrotic and granulomatous lesions in the upper virus (EBV) infection (Harabuchi et al., 1990). Recently, respiratory tract, mostly nasal cavity, which were mutations of c- proto-oncogene (Hongyo et al., 2000) biopsied for the histologic diagnosis before treatment. and tumor suppressor p53 gene (Li et al., 2000) were Histologic sections were reviewed by two of the reported. However, understanding of the molecular authors (T Takakuwa, K Aozasa) for diagnosis. mechanism of nasal NKTCL remained inadequate. Varying degrees of necrosis were found in the upper respiratory tract lesions. Diffuse proliferation of large atypical cells with mono- or multinucleus was observed accompanying with varying numbers of lymphocytes, *Correspondence: K Aozasa, Department of Pathology (C3), Osaka plasma cells, and macrophages, giving a polymorphous University Medical school, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan; E-mail: [email protected] appearance. An angiocentric pattern of proliferation Received 15 October 2001; revised 15 April 2002; accepted 15 April was occasionally observed. Immunohistochemically, 7 + + + 2002 tumor cells were CD20 , TIA-1 , CD16 , CD56 , Fas gene mutations in nasal lymphoma T Takakuwa et al 4703 CD3+,orCD43+. These histological and immunohis- nucleotide 1094. Two cases showed insertion of 1 bp tochemical findings were in agreement with those in the (A) at nucleotide 1095, which resulted in generation of nasal NKTCL (Jaffe et al., 1996). Denaturing gradient a frameshift. As a result, stop codon was introduced at gel electrophoresis analysis was employed to examine residue 304. Another case showed deletion of 1 bp (T) the rearrangement of the T-cell receptor g chain genes in a 7-(T) tract from nucleotide 591 to 597 of Fas in six of the present cases with proven clonal coding sequence. The remaining case had deletion of proliferation of EBV genome-containing cells, as 1 bp (C) at nucleotide 704. These two mutations previously described (Ohsawa et al., 1999). All six generated frameshifts, and resulted in stop codon at cases did not show clonally rearranged bands, residue 170. Because these mutants affect transmem- indicating the absence of T-cell clones in these cases. brane domain, the abnormal protein might not be RNA in situ hybridization using EBER-1 probe, which expressed on the cell surface. was performed as previously described (Weiss et al., Three point mutations were detected in 9, and 1992), revealed the positive signals in the nucleus of the two of these were missense ones, which caused proliferating cells in all cases, indicating the presence of substitutions of nonconserved amino acid. These EBV. mutations within the death domain affected codons Mutations of Fas gene were detected in seven (50%) 246 (ATC?GTC; Ile?Val) and 287 (CTT?CCT; of 14 cases of nasal NKTCL. Whether the detected Fas Leu?Pro) (GeneBank accession number M67454) mutations were heterozygous or homozygous could not (Itoh et al., 1991). be determined because nasal lesions showed a poly- Anti-human Fas antibody kills mouse T-cell lympho- morphous pattern of proliferation containing the ma cells WR19L which express human Fas cDNA. tumor cells and inflammatory reactive cells. The Fas This indicates that the Fas receptor mediates an mutations comprised of four frameshift, two missense, apoptotic signal into cells. To confirm whether the and one silent mutation (Figure 1). All but one of the mutations detected in the present cases abolish the silent mutations affected the death domain which is apoptotic signal transduction of the Fas receptor, these essential for apoptotic signal transduction. Previous Fas gene mutants were prepared and introduced into studies on the lymphoid lineage (Landowski et al., WR19L cells. Northern blot analysis demonstrated that 1997; Gronbaek et al., 1998; Muschen et al., 2000; Fas gene were expressed successfully (Figure 2a). Takakuwa et al., 2001) and epithelial malignancies (Lee However, fluorescence-activated cell sorting (FACS) et al., 1999a,b; Boldrini et al., 2001; Park et al., 2001) analysis with use of anti-Fas showed that the majority of Fas mutations observd in detected the expression of the human Fas receptor on these diseases were also located in the death domain. the cell surface only when Fas gene mutants containing Fas gene has a 6-(A) tract from nucleotide 1088 to a conserved transmembrane domain (A978G, 1095 ins A, T1102C) were introduced (Figure 2b). The ability of the mutated Fas on the cell surface to transduce the apoptotic signal was then examined in the established transformants. To exclude the clonal variations of Fas receptor expression, two independent clones were chosen for each mutant. Clones expressing the wild type Fas receptor were killed by Fas antibody. On the other hand, clones expressing the Fas receptor with any mutations (A978G, 1095 ins A, T1102C) were resistant to apoptosis induced by the anti-FAS anti- body (Figure 3). The NKTCL cells were reported to express Fas and FasL (Ng et al., 1999). Because cross-linking of FasL with Fas induces apoptosis, interaction of NKTCL cells from each other should result in apoptosis of the nasal NKTCL cells, thus prohibiting development of the disease. Therefore, some mechanism for resistance Figure 1 Summary of the Fas gene mutations found in patients of nasal NKTCL cells to Fas/FasL-induced apoptosis with nasal NKTCL. Total RNA was extracted using TRIzol re- might work for tumor development. Mutations of Fas agent (Gibco BRL, NY, USA) and was reverse-transcribed by gene could be one of the mechanisms. Mutations of random hexamer priming. The Fas open reading frame was am- Fas gene were detected in seven of 14 (50%) present plified using primer pairs of 5’-CACTTCGGAGGATTGCTCAA- cases of nasal NKTCL. Relatively high frequency of CA-3’ and 5’-TATGTTGGCTCTTC-AGCGCTA-3’ and then cloned in the pCR2.1-TOPO (Invitrogen, CA, USA). For the con- Fas gene mutations was reported in cases with thyroid trol of potential PCR error, eight to 12 clones derived from two lymphoma (65.4%) (Takakuwa et al., 2001) and other different PCR reactions were sequenced, respectively. When com- kinds of malignant lymphoma of mucosa-associated mon mutations were found in two PCR, we regarded them as lymphoid tissue (MALT) type (Seeberger et al., 2001), definite mutations. The shaded rectangles at the COOH-terminus of the protein represents the small peptide added because of the although this remains a subject of controversy (Bertoni frameshift in the gene encoding FAS. TM: transmembrane et al., 2000). It is well-recognized that follicular domain, Mr: molecular weight gastritis caused by the infection of Helicobacter pylori

Oncogene Fas gene mutations in nasal lymphoma T Takakuwa et al 4704

Figure 3 Cytolytic effect of anti-Fas antibody on WR19L trans- formant clones was carried out as described previously (Itoh et al., 1991). The mouse WR19L cell line transfected with plasmid pEF-BOS-EX vector alone and its transformant clones expressing recombinant human Fas protein with (T1102C, A978G, 1095 ins A) or without (wild type) mutations were incubated with various concentrations of anti-Fas antibody at 378C for 16 h. Viable cell counts were determined by the trypan blue exclusion method, and the percentage of viable cells was plotted

provide a basis for development of gastric lymphoma of MALT type (Wotherspoon et al., 1993; Enno et al., 1995), and chronic lymphocytic thyroiditis for the thyroid lymphoma (Kato et al., 1985; Holm et al., 1985). Coexistence of inflammatory cells are common findings in these lymphomas, which is also constantly found in nasal NKTCL (Jaffe et al., 1996). Taken together, resistance to apoptosis induced by mutations Figure 2 The 1.1 kb EcoRI-XbaI fragment containing the full of Fas gene might be involved in the development of length of human Fas cDNA with and without mutations were transferred into a mammalian expression plasmid pEF-BOS-EX lymphomas accompanying with inflammatory cells. (Mizushima and Nagata, 1990). All mutant constructions were One base insertion within a 6-(A) tract from verified by DNA sequencing using dideoxy chain termination. nucleotide 1088 to 1094 was found in two cases and Mouse T cell lymphoma cell lines WR19L (ATCC TIB52) were one base deletion within a 7-(T) tract from nucleotide transfected with the plasmid by electroporation as described pre- 591 to 597 in one case. The insertion of 1 bp (A) at viously (Itoh et al., 1991). After selection with 0.9 mg/ml of G- 418, transformants expressing the mutated Fas were identified nucleotide 1095 is also found in lymphomas arising in by Northern blotting and FACS analysis. (a) Total RNA was pre- thyroid, stomach, pleura, breast, and brain but not in pared from WR19L transfected with plasmid pEF-BOS-EX alone epithelial malignancies such as testicular and ovarian (i), transformants expressing the wild type human Fas receptor cancer (unpublished data). The mononucleotide tract in (ii), 1095 ins A (iii), A978G (iv), T1102C (v), 597 del T (Vi), and 704 del C (Vii). 15 mg of total RNA was electrophoresed the coding sequence is a mutational hotspot in many on a 1.3% agarose gel containing 6.6% formaldehyde and ana- kinds of genes, such as transforming growth factor b lysed by Northern hybridization. As a probe, the 1.1 kb EcoRI- receptor type II gene (Markowitz et al., 1995; Parsons et XbaI fragment containing the full length of human Fas cDNA al., 1995), Insulin-like growth factor receptor type II was labeled with digoxigenin-11 dUTP using a DIG High prime gene (Souza et al., 1996), hMSH3, hMSH6 (Malkhos- according to the manufacturer’s instructions (Roche Molecular Biochemicals, Mannheim, Germany). 28S ribosome RNA served yan et al., 1996), and BAX gene (Rampino et al., 1997). as control. (b) FACS analysis was performed using mouse anti- The insertion of 1 bp (A) at nucleotide 1095 might be a human Fas monoclonal antibody (CH-11, MBL, Nagoya, Japan) mutational hotspot in lymphomas of extranodal origin. and fluorescein isothiocyanate-conjugated anti-mouse Ig antibody In conclusion, the present study provides evidence (Capel, Ohio, USA). WR19L transfected with plasmid pEF-BOS- EX alone (i), transformants expressing the wild type human Fas that the Fas-mediated apoptotic pathway is abrogated receptor (ii), 1095 ins A (iii), A978G (iv), T1102C (v), 597 del in cases of nasal NKTCL. These findings suggest that T (Vi), and 704 del C (Vii). The dashed line represents parental accumulation of lymphoid cells with Fas mutations WR19L provides a basis for development of nasal NKTCL.

References

Aozasa K, Yang WJ, Lee YB, Pan WS, Wy YF and Horiuchi BertoniF,ConconiA,LuminariS,RealiniC,RoggeroE, KHyun BHTajima K. (1992). Int. J. Cancer, 52, 673 – 674. Baldini L, Carobbio S, Cavalli F, Neri A and Zucca E. (2000). Leukemia, 14, 446 – 448.

Oncogene Fas gene mutations in nasal lymphoma T Takakuwa et al 4705 Boldrini L, Faviana P, Pistolesi F, Gisfredi S, Di QD, Lucchi Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, M, Mussi A, Angeletti CA, Baldinotti F, Fogli A, Simi P, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW and Basolo F and Fontanini G. (2001). Oncogene, 20, 6632 – Vogelstein B. (1995). Science, 268, 1336 – 1338. 6637. Mizushima S and Nagata S. (1990). Nucleic Acids Res., 18, Enno A, O’Rourke JL, Howlett CR, Jack A, Dixon MF and 5322. Lee A. (1995). Am. J. Pathol., 147, 217 – 222. MuschenM,ReD,BrauningerA,WolfJ,HansmannML, Gronbaek K, Straten PT, Ralfkiaer E, Ahrenkiel V, Diehl V, Kuppers R and Rajewsky K. (2000). Cancer Res., Andersen MK, Hansen NE, Zeuthen J, Hou JK and 60, 5640 – 5643. Guldberg P. (1998). Blood, 92, 3018 – 3024. Ng CS, Lo ST and Chan JK. (1999). Hum. Pathol., 30, 48 – Harabuchi Y, Yamanaka N, Kataura A, Imai S, Kinoshita 53. T, Mizuno F and Osato T. (1990). Lancet, 335, 128 – 130. Ohsawa M, Nakatsuka S, Kanno H, Miwa H, Kojya S, Holm LE, Blomgren H and Lowhagen T. (1985). N. Engl. J. Harabuchi Y, Yang WI and Aozasa K. (1999). Int. J. Med., 312, 601 – 604. Cancer, 81, 865 – 870. Hongyo T, Li T, Syaifudin M, Baskar R, Ikeda H, Kanakura Park WS, Oh RR, Kim YS, Park JY, Lee SH, Shin MS, Kim Y, Aozasa K and Nomura T. (2000). Cancer Res., 60, SY, Kim PJ, Lee HK, Yoo NY and Lee JY. (2001). J. 2345 – 2347. Pathol., 193, 162 – 168. Itoh N and Nagata S. (1993). J. Biol. Chem., 268, 10932 – Parsons R, Myeroff LL, Liu B, Willson JK, Markowitz SD, 10937. Kinzler KW and Vogelstein B. (1995). Cancer Res., 55, Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, 5548 – 5550. Sameshima M, Hase A, Seto Y and Nagata S. (1991). Cell, RampinoN,YamamotoH,IonovY,LiY,SawaiH,ReedJC 66, 233 – 243. and Perucho M. (1997). Science, 275, 967 – 969. Jaffe ES, Chan JK, Su IJ, Frizzera G, Mori S, Feller AC and Seeberger H, Starostik P, Schwarz S, Knorr C, Kalla J, Ott Ho FC. (1996). Am. J. Surg. Pathol., 20, 103 – 111. G, Muller HH and Greiner A. (2001). Lab. Invest., 81, Kato I, Tajima K, Suchi T, Aozasa K, Matsuzuka F, Kuma 977 – 986. K and Tominaga S. (1985). Jpn. J. Cancer Res., 76, 1085 – Souza RF, Appel R, Yin J, Wang S, Smolinski KN, 1090. AbrahamJM,ZouTT,ShiYQ,LeiJ,CottrellJ,Cymes Landowski TH, Qu N, Buyuksal I, Painter JS and Dalton K, Biden K, Simms L, Leggett B, Lynch PM, Frazier M, WS. (1997). Blood, 90, 4266 – 4270. Powell SM, Harpaz N, Sugimura H, Young J and Meltzer Lee SH, Shin MS, Park WS, Kim SY, Dong SM, Pi JH, Lee SJ. (1996). Nat. Genet., 14, 255 – 257. HK, Kim HS, Jang JJ, Kim CS, Kim SH, Lee JY and Yoo Suda T, Takahashi T, Golstein P and Nagata S. (1993). Cell, NJ. (1999a). Cancer Res., 59, 3068 – 3072. 75, 1169 – 1178. Lee SH, Shin MS, Park WS, Kim SY, Kim HS, Han JY, Park Takakuwa T, Dong Z, Takayama H, Matsuzuka F, Nagata S GS,DongSM,PiJH,KimCS,KimSH,LeeJYandYoo and Aozasa K. (2001). Cancer Res., 61, 1382 – 1385. NJ. (1999b). Oncogene, 18, 3754 – 3760. Weiss LM, Jaffe ES, Liu XF, Chen YY, Shibata D and Li T, Hongyo T, Syaifudin M, Nomura T, Dong Z, Shingu Medeiros LJ. (1992). Blood, 79, 1789 – 1795. N, Kojya S, Nakatsuka S and Aozasa K. (2000). Lab. Wotherspoon AC, Doglioni C, Diss TC, Pan L, Moschini A, Invest., 80, 493 – 499. de Boni M and Isaacson PG. (1993). Lancet, 342, 575 – 577. Malkhosyan S, Rampino N, Yamamoto H and Perucho M. (1996). Nature, 382, 499 – 500.

Oncogene