Simian virus 40-like DNA sequences in human papillary thyroid ...

37 downloads 59 Views 330KB Size Report
with thyroid carcinomas. We studied 69 patients with papillary thyroid carcinoma (age: 12±76 years) and found that SV40-like sequences were present in three ...
Oncogene (1998) 16, 665 ± 669  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00

SHORT REPORT

Simian virus 40-like DNA sequences in human papillary thyroid carcinomas F Pacini1, A Vivaldi1, M Santoro3, M Fedele4, A Fusco5, C Romei1, F Basolo2 and A Pinchera1 1

Istituto di Endocrinologia, Metodologia Clinica e Medicina del Lavoro, via Paradisa 2, and 2Istituto di Anatomia Patologica, via Roma 57, Universita' di Pisa, 56124 Pisa; 3Dipartimento di Biologia e Patologia Cellulare e Molecolare c/o Centro di Endocrinologia ed Oncologia Sperimentale del C.N.R., Universita' di Napoli, via Pansini 5, 80131 Napoli; 4Istituto Nazionale dei Tumori di Napoli, via M. Semmola, 80131 Napoli; 5Dipartimento di Medicina Sperimentale e Clinica, Facolta' di Medicina e Chirurgia di Catanzaro, Universita' di Reggio Calabria, via T. Campanella 5, 88100 Catanzaro, Italia

Sequences of the SV40 virus, a virus of Asian macaques, have been found in human tumors, such as pleural mesotheliomas, ependimomas and choroid plexus tumors. Transgenic mice carrying the SV40 large T gene under the transcriptional control of the thyroglobulin gene promoter, develop thyroid dedi€erentiation and follicular thyroid cell proliferation, leading to thyroid hyperplasia and adenocarcinomas. On these bases we investigated the presence of SV40 DNA sequences in 69 samples of papillary thyroid carcinomas (PTC) and in other thyroid and non-thyroid carcinomas, as well as in benign thyroid diseases. By Southern blot and PCR ampli®cation followed by sequence analysis, we found the presence of SV40-related sequences integrated in the tumoral DNA of three cases of PTC. At least the 203 bp fragment of the aminoterminus of large T antigen, the 294 bp fragment of the VP1 gene and the 483 bp entire regulatory region were present in the tumoral DNA of these patients. SV40 sequences were not found in tissues other than PTC. Our results demonstrate that, in addition to previous ®ndings in mesotheliomas and brain tumors, SV40 is somehow linked to papillary thyroid carcinoma. Although our data do not demonstrate a causative role in the development of PTC, this possibility must be considered and requires further studies. Keywords: SV40; thyroid cancer; autoimmune thyroiditis; oncogene

Malignant tumors of the thyroid gland vary considerably in aggressiveness, ranging from a well-differentiated clinically indolent to an undi€erentiated often lethal phenotype (Williams, 1979). Papillary and follicular carcinomas represent the most frequent di€erentiated forms of thyroid malignancies. Etiologically, papillary histotype is associated with exposure to ionizing radiations (Williams et al., 1995; Miller, 1995; Ron et al., 1995), and follicular histotype with iodine de®ciency (Galanti et al., 1995; Pettersson et al., 1996). These di€erentiated tumors show di€erent genetic lesions, since ras mutations are the most frequent genetic alterations in follicular carcinomas, whereas they are rarely detected in papillary carcinomas (Santoro et al., 1995). In contrast, in papillary tumors activation of RET/PTC and TRK oncogenes represents

Correspondence: F Pacini Received 23 May 1997; revised 8 September 1997; accepted 9 September 1997

the most important genetic lesions, being present in about one third of cases (Santoro et al., 1992, 1994). The SV40 virus is a virus of Asian macaques that is tumorigenic for rodents and can transform human cells in vitro. SV40 sequences have been found in a large proportion of pleural mesotheliomas, ependimomas and choroid plexus tumors and other brain tumors (Bergsagel et al., 1992; Carbone et al., 1994; Lednicky et al., 1995; Martini et al., 1996). Some evidences have been accumulated about the possibility that SV40 large T antigen interferes with thyroid cell growth and di€erentiation. In fact transgenic mice generated by using a construct in which JC virus large T (JCV is a SV40-related virus which infects humans) was under the transcriptional control of SV40 promoter (Feigenbaum et al., 1992), developed thyroid hyperplasia. Conversely, targeted thyroid speci®c expression achieved by generating transgenic mice carrying the SV40 large T gene under the transcriptional control of the thyroglobulin gene promoter, which is thyroid speci®c, induced thyroid dedi€erentiation and proliferation of the follicular thyroid cells leading to thyroid hyperplasia and adenocarcinomas (Ledent et al., 1991). Moreover, the introduction in normal thyroid follicular cells of the SV40 large T gene, allows them to escape from early mortality and to loose all thyroid di€erentiated functions (Bond et al., 1996). These evidences prompted us to investigate the possibility that the SV40 virus might be associated with thyroid carcinomas. We studied 69 patients with papillary thyroid carcinoma (age: 12 ± 76 years) and found that SV40-like sequences were present in three of them. Seven normal peritumoral thyroid tissues, one Hashimoto's thyroiditis, ®ve toxic di€use goiters, three medullary carcinomas and nine breast carcinomas scored negative for SV40 sequences. DNAs from 69 papillary thyroid carcinomas were extracted with a method which provides both the integrated and not integrated DNA. Samples were then digested by the restriction enzyme BamHI, which makes only one cut in the SV40 genome, thus allowing to distinguish the episomal and the integrated state of the virus DNA. In the ®rst case, a single band corresponding to the size of the linearized virus is obtained, and in the second case, one or more bands of di€erent molecular weight are expected. Subsequently, the samples were analysed by Southern blot for genomic integration of SV40 related sequences, by hybridization with SV40 probes. As shown in Figure 1a, three out of 69 PTC samples (4.3%) showed the presence of the SV40 sequence,

SV40-like DNA in papillary thyroid carcinoma F Pacini et al

666

while seven normal thyroid tissues and four blood DNAs from patients with PTC were negative for SV40. No positive case was found in samples from patients with medullary thyroid cancer (n=3) or breast carcinoma (n=9), nor from patients with benign thyroid diseases. Since no multiple bands were detected by Southern blot in the three positive cases, it can be assumed that the neoplastic tissue takes origin from the clonal proliferation of a single cell carrying the integrated SV40. Southern blot of the three positive cases was also performed after digesting the tumor DNA with the restriction enzyme PstI which is able to cut the SV40 genome at two sites, generating a fragment of 4.0 Kbp, including the coding sequences for VP2, part of VP1 (the viral capsid proteins), and the regulatory region, small t and large T antigen, except for the carboxyterminus of the last protein. As shown in Figure 1b, a fragment of the expected size was observed in all the three positive cases, hybridizing in high stringency conditions with both the complete SV40 genome probe and with the PvuII fragment probe, covering the coding sequence for VP2. To con®rm the presence of SV40 sequences in genomic DNA of thyroid tumors, two primers corresponding to the amino-terminal region of large

T antigen were used to amplify SV40 sequences in the thyroid tumor samples. These primers are expected to amplify a 105 bp size fragment, using as template the SV40 genome. The PCR ampli®cation was followed by hybridization with a 32P-labelled internal oligonucleotide; as shown in Figure 2a, the three samples, which resulted positive by Southern blot analysis, showed ampli®cation of an SV40 fragment of the expected size. We then used primers LA1 and LA2, which amplify a 294 bp region of the VP1 gene, the SV40 major capsid protein: the PCR product was hybridized with the entire SV40 DNA and again all the three samples resulted positive, showing a fragment of the expected size (Figure 2b). To assess if a cross-hybridization with SV40 closely related viruses (BK, JC, etc.) which often infects humans (Arthur and Shah, 1989) had occurred, we next sequenced a region of sample 2 encompassing a 9 bp insert present in SV40 related papovaviruses but not in SV40 DNA. The PCR ampli®ed DNA using primers Pyv.for and Pyv.rev, which are speci®c for a 172 bp sequence in the aminoterminus region of large T gene, and are partially superimposed to the region previously ampli®ed with primers SV.for and SV.rev, was directly sequenced: as shown in Figure 3a, the sequence perfectly matches with SV40 DNA and the

b

a C1 C2

1

2

3

4

5

— 23.1 — 9.4 23.1 — — 6.5 9.4 — 6.5 —

4.3 —

C

— 4.3

1

2

3

4

5.2 Kb

— 2.3 — 2.0

4.0 Kb

2.3 — 2.0 — Figure 1 Southern blot autoradiogram of DNAs extracted from papillary thyroid cancers hybridized with a 32P-labeled SV40 probe. DNA extraction and Southern blot analysis were performed following standard procedures (Maniatis et al., 1982). Each gel contained HindIII digested lambda DNA as molecular weight marker, and, as positive control, 0.1 ± 1 genome equivalent of the SV40 DNA used as probe. The entire SV40 DNA (Sigma) was used as probe or the 1446 bp PvuII restriction fragment covering the VP2 region, puri®ed from low melting agarose gel by PCR puri®cation kit (Promega); the probes were labelled by random priming (Stratagene) with [32P]dCTP to speci®c activities416109 d.p.m./mg. (a) Lanes 1 and 3: normal thyroid tissue; lanes 2, 4 and 5: PTC samples. In these experiments BamHI was used as cutter enzyme and the SV40 complete genome as probe. Lanes C1 and C2: positive controls, respectively 1 and 0.1 copy equivalent per cell of SV40 genome linearized by BamHI digestion. (b) Lanes 1, 2 and 3, PTC samples, lane 4 normal thyroid tissue, all digested with PstI which cuts SV40 DNA in two fragments of 4.0 Kbp and 1.2 Kbp; here the 4.0 Kbp restriction fragment is evident in all the three papillary samples. The SV40 complete genome was used as probe. Similar results were obtained when using as probe the 1446 bp PvuII restriction fragment, covering the VP2 region. Lane C, reconstruction track of 1 copy equivalent per cell of SV40 genome linearized by BamHI digestion

SV40-like DNA in papillary thyroid carcinoma F Pacini et al

a

b 1

2

3

4

5

6

1

2

3

4

5

6

294 bp 105 bp Figure 2 Southern blot analysis of PCR products. All possible precautions were taken to avoid tissue samples and PCR assays contamination; besides, in any PCR experiment, we always had a negative control. The regions chosen for PCR ampli®cation were the SV40 early and late region; for the early region (a) we used primers designed by Bergsagel et al. (1992): SV.for 3 (TGAGGCTACTGCTGACTCTCAACA, from nucleotide 4476 ± 4453) and SV.rev (GCATGACTCAAAAAACTTAGCAATTCTG, from nucleotide 4399 ± 4372); for the late region (b) we used primers designed by Lednicky et al. (1995) to detect a segment encoding the C terminus of the major capsid protein: LA1 (GGGTGTTGGGCCCTTGTGCAAAGC, from nucleotide 2251 ± 2274) and LA2 (CATGTCTGGATCCCCAGGAAGCTC, from 2545 ± 2522). The PCR cycling reactions were performed in a Thermolyne Termptronic DNA thermal cycler (PBI); each reaction contained 50 picomoles of primers, 200 mM dNTPs, 1.5 mM MgCl2 and 2.5 units of Taq DNA polymerase in 50 ml. The ®rst cycle was 948C 5 min, 528C 1 min, 728C 1 min, followed by 30 cycles of 948C 1 min, 528C 1 min, 728C 1 min. Upon completion, another 2.5 units of Taq polymerase were added and 30 additional cycles were performed with a ®nal extension of 6 min at 728C. One tenth of the PCR assay mixture was subjected to electrophoresis on 4% agarose gels. Primers Sv.for3 and Sv.rev gave an ampli®cation fragment of 105 bp, primers LA1 and LA2 one of 294 bp. The gels were blotted on Gene Screen Plus membranes (NEN), as recommended by the manufacturer, and these hybridized with a labelled oligonucleotide probe representing an internal region of the ampli®ed PCR products (GGAAAGTCCTTGGGGTCTTCTACC) when the SV.for3 and SV.rev primers were used, and the complete SV40 genome when LA1 and LA2 primers were used. Filters were prehybridized, hybridized and washed according to standard procedures, and exposed to Kodak XAR-5 ®lms for 4 h. Lanes 1, 2 and 3, PTC samples; lane 4, normal thyroid tissue, lane 5 blood lymphocytes from a positive case, lane 6, negative control (no template DNA)

9 bp insert is absent. Besides, we sequenced the SV40 regulatory region of sample 2, because this sequence is speci®c among papovaviruses; using primers RA3 and RA4, speci®c for the SV40 regulatory region, sample 2 was PCR ampli®ed and directly sequenced and the Figure 3b shows that the region matches the SV40 sequence. Finally, we performed immunohistochemistry for the large T antigen using a monoclonal antibody directed against the large T and the small t antigens (SV40 TAg Ab-1, Oncogene Science Inc.). Positive staining was found in the three SV40 positive papillary tumors, limited to the cytoplasms, but not in negative controls (data not shown). Recently, SV40 has been detected in human brain tumors, such as choroid plexus and ependymoma tumors (Bergsagel et al., 1992; Lednicky et al., 1995; Martini et al., 1996), human mesotheliomas (Carbone et al., 1994; Cristaudo et al., 1995), and metastases of a human melanoma (Soriano et al., 1974). In the present study we show the presence of SV40-related sequences, integrated in the tumoral tissue of three papillary thyroid carcinomas. Our results do not demonstrate that the entire SV40 genome is integrated in the positive samples; however the PCR and sequence analysis demonstrate that at least the 203 bp

fragment of the gene coding for the amino terminus of large T antigen, the 294 bp fragment of the VP1, the viral major capsid protein, and the 483 bp entire regulatory region are present. Although the presence of the SV40 sequences in these tumoral tissues do not demonstrate that it has a causative role in the development of papillary thyroid carcinomas, this possibility must be taken into account for several reasons. Firstly, the observed integration of SV40 sequences in the tumoral genomic DNA is consistent with a role played by these sequences in tumorigenesis: indeed in this way viral genes can be propagated to the tumoral cell progeny, as observed in the case of other DNA tumor viruses (Das et al., 1992). Furthermore, thyroid expression of the SV40 large T gene is able to induce poorly di€erentiated thyroid carcinomas in mice (Ledent et al., 1991), and immortalization of human thyroid cultures have been achieved by transfection with the SV40 large T gene (Bond et al., 1996). The possibility that SV40 virus could infect humans has been denied for several years, assuming that SV40 infections in humans are rare and harmless. However, low-level antibody response and replication of the SV40 virus have been documented in subjects accidentally infected with SV40 by the intranasal route (Morris et al., 1961). Moreover, SV40 was shown to be capable of establishing low-grade infections in children fed contaminated poliovaccine over 30 years ago (from 1955 ± 1963), with virus being excreted in the stool for as long as 5 weeks (Melnick et al., 1962; Shah et al., 1976). At least two of our patients with SV40-positive tumor, those born in 1954 and 1961, could have been exposed to SV40contaminated vaccine. Several mechanisms may be envisaged to explain the involvement of the SV40 in the thyroid carcinogenesis process. SV40 large T antigen is able to bind and neutralize the protein products of Rb and p53 tumorsuppressors. Experimental models have demonstrated that, not only p53 inactivation, but also Rb blockage by the thyroid targeted expression of the Human Papillomavirus E7 oncogene is able to cause the development of thyroid neoplasms (Ledent et al., 1995). In addition, it is known that SV40 large T induces IGF-1 production in transfected cell lines, and a functional IGF-1 receptor is necessary for the growth stimulating e€ects of SV40 (Porcu et al., 1992; Sell et al., 1993). Both events, i.e. presence of IGF-1 receptors and stimulation of thyroid cell replication by IGF-1, have been demonstrated in normal and neoplastic follicular thyroid cells (Takahashi et al., 1995; Balkany et al., 1995). Histological examination of thyroid tumors harbouring SV40 DNA sequences, revealed the typical features of papillary thyroid carcinoma. However, two of them were also characterized by histological features of classical Hashimoto's thyroiditis, with lymphocytic in®ltration of the thyroid and high titers of circulating anti-thyroid autoantibodies (data not shown). Although it is possible that this represents only a fortuitous association (nearly 2% of our patients with papillary carcinoma have associated Hashimoto's thyroiditis), it is interesting the possibility that SV40 may play some role in the development in the in¯ammatory reaction. Indeed, a viral etiology has

667

SV40-like DNA in papillary thyroid carcinoma F Pacini et al

668 a

b ACGT

ACGT 4546 G G G A G C A G T G G T G G A A T G g g g a g c a gt g g t g ga at g

316 A A A G A G G A A C T T G G T T A G G T aaagaggaacttggttaggt

CCTTTAATGAGGAAAACCTGTT cctttaatgaggaaaacctgtt

ACCTTCTGAGGCGGAAAGAACC a c c tt c t g a g g cg g a a agaac c

TTGCTCAGAAGAAATGCCATCTA t t g c t c a g a a g a a a t g c c a ta

G T G A T G A T T C T A C T C C T CC A A A A g t g a t g a t t c t a c t c c t cc a a a a

C T C A A C A T T C T A C C T CC A A A A c t c a a c a t t c t a c c t cc a a a a

A G C T G T G G A A T G T G T G T C AG T agctgtggaatgtgtgtcagt

TAGGTGTGGAAAGTCCCCAG t a g g g t g t g g a a a g t cc c c a g

GCTCCCCAGCAGGCAGAAGTA g c t c c c c a g ca gg cag a a g t a

TGCAAAGCATGCATCTCAATTA tgcaaagcatgcatctcaatta A A G A A G A G A A A G G T 4423 aagaagagaaaggt

GTCAGCAACCAGGTGTGGA gtcagcaa ccaggtgtgga A A G T 166 a a gt

Figure 3 Direct DNA sequencing (Sanger et al., 1977) of the PCR products of PTC sample of patient 2. (a) DNA was ampli®ed with primers speci®c for a 172 bp region of the aminoterminal of large T antigen common to several papovavirus, designed by Bergsagel et al. (1992): Pyv.for (TAGGTGCCAACCTATGGAACAGA, from nucleotide 4574 ± 4552) and Pyv.rev (GGAAAGTCTTTAGGGTCTTCTACC, from 4425 ± 4402; as SV40 virus, the tumoral sample lacks a 9 bp insert (arrow) characterizing this sequence as SV40-like. (b) DNA was ampli®ed using primers for the complete SV40 regulatory region designed by Lednicky et al. (1995): RA3 (GCGTGACAGCCGGCGCAGCACCA, from nucleotide 358 ± 336) and RA4 (GTCCATTAGCTGCAAAGATTCCTC, from 5119 ± 5142). The entire sequence was identical to the SV40 regulatory region (only the ®rst 150 bp are shown in the ®gure)

been suggested for Hashimoto thyroiditis (Ciampolillo et al., 1989). Further studies will be necessary to ascertain whether natural or mutated strains of SV40 are present in thyroid carcinomas and whether the integrated viral sequences have transforming ability.

Acknowledgements We thank Dr C Giani (Pisa) for providing the sample of breast carcinomas. This work was supported by grants from: Associazione Italiana per la Ricerca sul Cancro (AIRC); European Communities: COSUCT 94-0090 and NUFISA F14C CT-960002.

References Arthur RR and Shah KV. (1989). Prog. Med. Virol. Basel Karger, 36, 42 ± 61. Balkany V and Cushing GW. (1995). Thyroid, 5, 47 ± 50. Bergsagel DJ, Finegold MJ, Butel JS, Kupsky WJ and Garcea RL. (1992). N. Engl. J. Med., 326, 988 ± 993. Bond AJ, Ness GO, Rowson J, Ivan M, White D and Wynford-Thomas D. (1996). Int. J. Cancer, 67, 563 ± 672. Carbone M, Pass HI, Rizzo P, Marinetti MR, Di Muzio M, Mew DJY, Levine AS and Procopio A. (1994). Oncogene, 9, 1781 ± 1790. Ciampolillo A, Marini V, Mirakian R, Buscema M, Schlz T, Pujol-Borrel R and Bottazzo GF. (1989). Lancet, 20 May (8647), 1096 ± 1100. Cristaudo A, Vivaldi A, Sensales G, Guglielmi G, Ciancia E, Elisei R and Ottenga F. (1995). JEPTO., 14, 29 ± 34. Das BC, Sharma JK, Gopalkrishna V, Das DK, Singh V, Gissmann L, Hausen HZ, Luthra UK. (1992). J. Med. Virol., 36, 239 ± 245. Feigenbaum L, Hinrichs SH and Jay G. (1992). J. Virol., 66, 1176 ± 1182. Galanti MR, Sparen P, Karlsson A, Grimelius L and Ekbom A. (1995). Int. J. Cancer, 61, 615 ± 621. Ledent C, Dumont J, Vassart G and Parmentier M. (1991). Endocrinology, 129, 1391 ± 1401.

Ledent C, Marcotte A, Dumont JE, Vassart G and Parmentier M. (1995). Oncogene, 10, 1789 ± 1797. Lednicky JA, Garcea RL, Bergsagel DJ and Butel JS. (1995). Virology, 212, 710 ± 717. Maniatis T, Fritsch EF and Sambrook J. (1989). Molecular cloning: a laboratory manual., 2nd Edition, Cold Spring Harbor Laboratory Press: New York. Martini F, Iaccheri L, Lazzarin L, Carinci P, Corallini A, Gerosa M, Iuzzolino P, Barbanti-Brodano G and Tognon M. (1996). Cancer Res., 56, 4820 ± 4825. Melnick JL and Stinebaugh S. (1962). Proc. Soc. Exp. Biol. Med., 109, 965 ± 968. Miller RW. (1995). Environ. Health Perspect., 103 (Suppl 6), 41 ± 44. Morris JA, Johnson KM, Aulisio CG, Chanock RM and Knight V. (1961). Proc. Soc. Exp. Biol. Med., 108, 56 ± 59. Pettersson B, Coleman MP, Ron E and Adami HO. (1996). Int. J. Cancer, 65, 13 ± 19. Porcu P, Ferber A, Pietrzkowski Z, Roberts CT, Adamo M, LeRoith D and Baserga R. (1992). Mol. Cell. Biol., 12, 5069 ± 5077. Ron E, Lubin JH, Shore RE, Mabuchi K, Modan B, Pottern LM, Schneider AB, Tucker MA and Boice Jr, JD. (1995). Radiat. Res., 141, 259 ± 277.

SV40-like DNA in papillary thyroid carcinoma F Pacini et al

Sanger F, Niklen S and Coulson AR. (1977). Proc. Nat. Acad. Sci. USA, 74, 5463 ± 5467. Santoro M, Carlomagno F, Hay ID, Herrmann MA, Grieco M, Melillo R, Pierotti MA, Bongarzone I, Della Porta G, Berger N, Peix JL, Paulin C, Fabien N, Vecchio G, Jenkins RB and Fusco A. (1992). J. Clin. Invest., 89, 1517 ± 1522. Santoro M, Dathan NA, Berlingieri MT, Bongarzoni I, Paulin C, Grieco M, Pierotti MA, Vecchio G and Fusco A. (1994). Oncogene, 9, 509 ± 516. Santoro M, Grieco M, Melillo RM, Fusco A and Vecchio G. (1995). Eur. J. Endocr., 133, 513 ± 522. Sell C, Rubini M, Rubin R, Liu JP, Efstradiatis A and Baserga R. (1993). Proc. Natl. Acad. Sci. USA, 9, 11217 ± 11221.

Shah K and Nathanson N. (1976). Am. J. Epidemiol., 103, 1 ± 12. Soriano F, Shelburne CE and Gokcen M. (1974). Nature, 249, 421 ± 424. Takahashi MH, Thomas GA and Williams ED. (1995). Br. J. Cancer, 72, 813 ± 817. Williams ED. (1979). Clin. Endocrine Metab., 8, 193 ± 207. Williams ED, Pacini F and Pinchera A. (1995). J. Endocrinol. Invest., 18, 144 ± 146.

669