ORIGINAL ARTICLE
‘Hit and run’ oncogenesis by human papillomavirus type 18 DNA T. IWASAKA, Y. HAYASHI, M. YOKOYAMA, K. HARA,N. MATSUO AND H. SUGIMORI From the Department of Obstetrics and Gynecology, S a g a Medical School, S a g a , J a p a n
Actu Ohstet Gynecol Scand 1992; 71: 219-223 Transfection of an immortalized cell line (AE), derived from Syrian hamster embryo cells, with human papillomavirus type 18 (HPV 18) DNA induced morphological transformation and these transformed cells were tumorigenic in nude mice. Southern blot analysis revealed that the transfected viral DNA was retained in all the cell lines tested, however, all these transformed cells contained only less than one copy per cell of viral genome. Eleven cloned cell lines were established from a tumor cell line obtained after explantation of a tumor into a nude mouse. Two lines revealed no viral sequences by both Southern blot hybridization and polymerase chain reaction, whereas the nine others contained the remaining viral sequences. These results are highly suggestive of a ‘hit and run’ oncogenesis by this virus.
Key words: ‘Hit and run’ oncogenesis; human papillomavirus type 18; in vitro transformation; hamster embryo cells Submitted October 28, 1991 Accepted December 3 1 , 1991
Human papillomavirus (HPV) sequences are present in up to 90% of cervical cancer tissues, with the majority of these tumors containing either HPV 16 or HPV 18 DNA (1-5). This strongly suggests a causative rBle for these viruses in the development of malignancy. The transforming potentials of these two viruses have been well studied in vitro. The oncogenic potential of total genome of HPV 16 or HPV 18 has been shown in mouse 3T3 cells, rat 3Y-1 cells, and hamster AE cells, immortalized lines (6-9), but not in primary culture cells. It has also been shown that the E6 and E7 open reading frames (ORF) of these viruses have major transforming activity both in rodent and human cells (10-!4). From these studies it is clear that the E6 and E7 genes encode for the major transforming activity of HPV 16 and HPV 18. In all HPV-containing transformed cells or cervical cancer tissues analysed, E6 and E7 ORFs are always maintained. It is therefore
crucial to determine whether there is a continued requirement for these genes in transformed cells or whether these genes behave in a ‘hit and run’ fashion and, after the initial transforming event, are no longer required. In our previous study, we noted the oncogenic potential of HPV 16 and H P V 18 DNA on hamster embryo cells already immortalized with transfection of specific fragment of herpes simplex virus type 2(HSV-2) DNA. All the cell lines obtained by HPV 16 DNA transfection contained more than one copy per cell of viral DNA, while no viral DNA was detected in two cell lines obtained after transfection of HPV 18 DNA[9]. We proposed the ‘hit and run’ mechanism in the oncogenesis by H P V 18 DNA. In the present study, we asked whether HPV DNA is required for the transformation and for the maintenance of a cancer state.
Actci Ohstrt Gynecol Scand 71 (1YY2)
220
T. lwasuku et ul.
Table 1 . Sequences of oligonucleotide primer pairs and their locations in the HPV 18 genome
HPV 18 primer I HPV I X primer 2
Sequence
Location
5' TCACGAGCAATTAAGCGACT 3' 5' CTG AGCTTTCTACTACTAGC 3'
67(k6XY X(WX27
HPV 18 DNA was originally isolated from cervical carcinoma cells and cloned at its unique EcoRI site into pBR322 (2). This cloned form of DNA was obtained from the Japanese Cancer Research Resources Bank, Tokyo.
grams of each Hind 111 linearized plasmid D N A was mixed with 5 pg of salmon sperm D N A and precipitated with calcium phosphate. Aliquots of 0.5 ml were applied to subconfluent monolayers of A E cells (passage 62: 2.5 x 10' cells per 60 mm dish). Cultures were refed fresh medium at 5 hours after transfection and were then fed every 2 to 3 days. Foci were isolated with glass cylinders and then expanded for further analysis.
Cells and D N A trunsfection
nrmorigenicity
A E cells, an immortalized cell line, established by transfcction of Syrian hamster embryo cells with immortalizing sequences of HSV-2 DNA ( l S ) , were maintained in modified Dulbccco's Eagle reinforced medium with 10% fetal calf serum. Three micro-
Transformed cells were collected and suspended in serum-free culture medium. A total number of 1 X lo7 cells in 0.1 ml of the medium was injected subcutaneously into Balb/c nude mice (4 weeks old) and the mice were then examined weekly.
Methods Recombinant HPV D N A
Analysis of H P V D N A sequences High molecular weight cellular D N A was extracted from the transformed cells and the cell lines explanted from the tumors. Ten micrograms of cellular DNA from each cell line was digested with restriction endonuclease EcoRl or Pst- 1, electrophoresed. transferred to a nylon membrane and hybridization was carried out with the probe of HPV 18 DNA, as previously described (9).
Polymeruse chain reaction
Fig. 1 . Southern blot hybridization of DNA samples from transformed cells. Ten microgram of each cellular DNA was digested with both EcoRl and Pst-I, subjected to electrophoresis on 0.7"/, agarose gel, transferrcd to nylon filters and hybridized to a '?P-labeled EcoKI fragment of HPV 18. The DNA analysed was obtained from three HPV IH-transformed cell lines (18-1, 18-2 and 18-3: lanes 4, 6, and 8. respectively). four tumor cell lines (lS-lT, 18-2T, lX-3T, and 18-4T: lanes S, 7, 9, and 10. respectively), and A E cells transfected with pBR322 D N A (lane 3). Samples of HPV 1 X DNA corresponding to 1 and 3 copies per cell of HPV-1X DNA sequences were used for a reconstruction experiment (lanes 2 and I , respectively).
To detect a small amount of HPV DNA, polymerase chain reactin (PCR) was adopted. The region chosen for HPV 18 specific primers was within E 7 O R F (Table I). Amplification reaction was performed on one microgram of each cellular D N A using a DNA Amplification Reagent Kit (Perkin Elmer Cetus, Norwalk, U.S.A.). The samples were overlaid with mineral oil and subjected to 35 cycles of amplification. A cycle represents (a) denaturation for 1 minute at Y4"C, (b) annealing for 2 minutes at 55°C and (c) primer extension for 2 minutes at 72°C. To detect the amplified D N A fragment, 15 pI of the reaction mixture was analysed by electrophoresis on 3% (w/v) NuSieve gels, and the amplified products were visualized by staining with ethidium bromide.
'Hit und run' oncogenesis by H P V type I8
221
Polymerase chain reaction Analysis of D N A extracted from A E cells # (a negative control) and clones # I , #3, #4, and #8 was performed by thirty-five cycles of PCR with HPV 18 specific primers. HPV 18 DNA as well as clones #3 and #4 yielded clear bands between 154bp and 220bp corresponding in size to that expected for HPV 18 D N A (158bp), while no corresponding band was visualized in cases o f AE cells as well as clones # 1 and #8 (Fig. 3).
Tumorigenicity Fig, 2. Southcrn blot hybridization of DNA samples from cloncd cell lines. Cellular DNA was extracted from each cloned cell line obtained from ii t u m o r cell line (18-T4). digested with Pst-1, and subjected to Southern blot hybridization. Lanes 1-11: clones # I to # I t , respectively. Lane 12: 18-4T. Lanes 13 and 14: 1 and 3 copies per cell of HPV 18 DNA sequences as a positive control.
TOassess the tumorigenic activity of the four cloned Cell lines, each clone was further propagated and transplanted into nude mice. HPV DNA-positive clones formed tumors in nude mice about four weeks after transplantation (#3 and #4), About 5 weeks were required for definite tumors to form in of HPV DNA negative clones ( # I and #8).
Results Southern blot hybridizution High molecular weight DNA was extracted from four cell lines. designated as 18-1, 18-2, 18-3, and 18-4, obtained from separated foci, four cell lines, designated as 18-1T, 18-2T, 18-3T, and 18-4T, obtained by explant culture o f the tumors derived from the respective cell lines. and A E cells transfected with pBR322. To examine the retention of HPV DNA sequences, total DNAs extracted from these cell lines were digested with EcoRI and Pst-1 and subjected to Southern blot analysis with the "Plabeled HPV 18 D N A . The results showed that less than one copy per cell of HPV 18 DNA was detected in all the cell lines tested (Fig. I ) . However, these apparent lower copy numbers in these cell lines might be the reflection of lower hybridization efficiency in high molecular DNA. compared with that in purified HPV DNA. To examine whether the HPV D N A was maintained in all the transformed cells, hybridization analysis was performed in the cloncd cell lines. Eleven clones were obtained from the 18-4T cell line which contains the least HPV D N A and cellular D N A was extracted from each clone. Hybridization was carried out in the same fashion after digestion with Pst-1. Of the eleven clones, eight contained approximately one copy per cell of viral D N A and one contained 0.5 copy per cell (#4), while the remaining two revealed no viral D N A (#I and #8) (Fig. 2).
Fig. 3 . PCR prcducts revealed by ethidium bromide staining of electrophoretic gel. The thermophilic Taq polymerase was used in 35 cycles of amplification before electrophoresis. Lane I : PCR products form 1 pg of extracted DNA from A E cells. Lanes 2-5: PCR products from I pg of extracted D N A from cloned cell lines #1, #3,#4, and #8. respectively. Lane 6: PCR products from I pg of recombinant HPV 18 DNA as a positive control. Size markers (pBR322lHinf 1 digest) are given on the left and the band represcnting specific PCR products from HPV 18 DNA is shown by the arrow on the right.
222
T. lwasuka et a1
Discussion I n the oncogenesis by HPV DNA, it has been generally accepted that HPV DNA or its expression is essential for the maintenance of a transformed state. Our previous observation of the oncogencsis by HPV 16 DNA is consistent with this idea (9). In contrast, the present study showed that HPV 18 sequences were apparently not present in some cloned cell lines derived from a tumor cell line obtained after transfection of this virus DNA. This is in accord with our previous observation that viral sequences were absent in the HPV 18-induced transformed cells (9). Thereafter repeated Southern blot hybridization and PCR revealed viral sequences in these transformed cells, albeit the amount being scanty. On the basis of the results presented here, it is suggested that viral sequences are not always required for the maintenance of a transformed state in the oncogenesis by HPV 18, in other words HPV 18 behaves in a ‘hit and run’ fashion. The apparent contradictions in the r61e of HPV 16 and HPV 18 in transforming AE cells might be due to differences in affinity to cellular DNA or to differences in the facility to delete from the transformed celis. Consistent with this idea is the observation that HPV 16 DNA sequences are more frequently present in cervical cancer tissues than are those of HPV 18 DNA, and moreover, in a fair percentage, viral sequences are absent or are present in a very small amount. This different potential to transform cells in vitro between HPV 16 and HPV 18 seems to be analogous to that between bovine papillomavirus (BPV)-1 and BPV-4. BPV-1-transformed cells contain multiple copies of viral DNA and the expression of the BPV-1 genome appears to be required for the maintenance of the transformed state (16, 17). While, in BPV-4-induced transformation, only 9 of 60 ceH lines harbour BPV-4 DNA and there is no evident relationship between the presence of the viral DNA and the transformed phenotype, which is in accord with the ‘hit and run’ oncogenesis (18). The contribution of cellular oncogenes or tumor suppression genes must be considered for the ‘hit and run’ oncogenesis. Expression or insertion of HPV sequences might directly induce amplification or overexpression of cellular oncogenes or involve deletion or a point mutation of tumor suppression genes, consequently the cell line is converted to a tumorigenic one. Once oncogenic transformation has been established, viral sequences are not always necessary for the maintenance of a transformed state. This apparent mechanism in the HPV oncogenesis is not unexpected since the same mechanism functions for transformation by DNA tumor viruses, herAcia Obstei Gynecol &and 71 (1YY2)
pes simplex virus, human cytomegalovirus and adenovirus, where no consistent viral sequences are maintained on expression in tumor cells (19). We cannot completely rule out the possibility that this transformation is spontaneous. It is, however, unlikely because the AE cell line is stable and no spontaneous transformation has been observed during more than 60 passages and in any experimental group set up for control transfection. We have still to rule out the possibility that the transformed cells have retained a small fragment of HPV 18 DNA that activates cellular oncogenes by insertional mutagenesis, if this fragment is too small to be detected by Southern blot hybridization or PCR. Such a mechanism has been postulated for HPV 18 in some cervical carcinomas where the viral genome is integrated upstream of the c-myc oncogene (20). At the present time, it is difficult to exclude this possibility and conversely it is also difficult to prove that the retained small fragment is essential for the maintenance of a transformed state. Furthermore, deletion of viral DNA in tumor cells is not always advantageous for tumorigenicity, as shown in our in vivo experiment in which the viral DNA positive-clones formed tumors in nude mice one week earlier than those lacking the viral DNA. More detailed studies, including cellular oncogenes and tumor suppression genes, on the cell lines, with or without viral DNA, are expected to provide further information on the r61e of the virus in the development of papillomavirus-associated cancers in humans.
Acknowledgment We would like to thank Mariko Ohara for careful review of the manuscript.
References Durst M, Gissmann L, lkenberg H, zur Hausen H. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions. Proc Natl Acad Sci USA 1983; 80. 3812-5. 2 . Boshart M, Gissmann L, Ikenberg H, Kleinheinz A , Scheurlen W, zur Hausen H. A new type of papillomavirus DNA, its presence in genital cancer biopsies and in cell lines derived from cervical cancer. EMBO J 1.
1984; 3: 1151-7. 3. zur Hausen H. Genital papillomavirus infection. Prog Med Virol 1985; 32: 15-21. 4. McCance DJ. Human papillomavirus and cancer. Biochem Biophys Acta 1986; 823: 195-205. 5. Macnab JC, Walkinshow SA, Cordiner JW, Clements
‘Hit and run’ oncogenesis by HPV type 18 JB. Human papillomavirus in clinically and histologically normal tissue o f patients with genital cancer. N Engl J Med 19x6; 315: 1052-8. 6. Yasumoto S, Burkhardt AL, Doniger J , DiPaolo JA. Human papillomavirus typc 16 DNA-induced malignant transformation of NIH 3T3 cells. J Virol 1986; 57: 572-7. 7. Tsunokawa Y , Takehc N , Kasamatsu T, Terada M, Sugimura T. Transforming activity o f human papillomavirus type 16 DNA sequences in a cervical cancer. Proc Natl Acad Sci USA 1986; 83: 220(&3. 8. Matlashcwski G , Osborn K , Murray A. Banks L. Crawford L. Cancer cells. Vol 5: Papillomaviruses. NY: Cold Spring Harbor Laboratory Press, 1987: 195-9. 9. lwasaka T, Yokoyama M, Hayashi Y , Sugimori H. Combined herpes simplex virus type 2 and human papillomavirus type 16 o r 18 deoxyribonucleic acid leads to oncogenic transformation. Am J Obstet Gynecol 1988; 159: 1251-5. 10. Bedell MA, Jones KH, Laimins LA. The E6-E7 region o f human papillomavirus type 18 is sufficient for transformation of NIH3T3 and Rat-I cells. J Virol 1987; 61: 363540. 11. Pirisi L, Yasumoto S, Feller M, Doniger J, DiPaolo JA. Transformation of human fibroblasts and keratinocytes with human papillomavirus type 16 DNA. J Virol 1987; 61: 1061-6. 12. Kanda T, Furuno A , Yoshiike K. Human papillomavirus type I6 open reading frame E7 encodes a transforming gene for rat 3Y1 cells. J Virol 1988; 62: 61G3. 13. Schlegel R , Phelps WC, Zhang YL, Barbosa M. Quantitative keratinocyte assay detects two biological activities of human papillomavirus DNA and identifies viral types associated with cervical carcinoma. EMBO J 1988; 7: 3181-7. 14. Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R . The E6 and E7 genes of the human papillomavirus
15.
16.
17.
18.
19.
20.
223
type 16 together are necessary and sufficient for transformation of primary human keratinocytes. J Virol 1989; 63: 4417-21. Hayashi Y . lwasaka T, Smith CC, Aurelian L, Lewis GK, Ts’o POP. Multistep transformation by defined fragments of herpes simplex virus type 2 DNA: oncogenic rcgion and its gcne product. Proc Natl Acad Sci USA 1985; 82: 8493-7. Law M-F, Lowy DR, Dvoretzky J , Howley P. Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. Proc Natl Acad Sci USA 1981; 78: 2727-31. Amtmann E, Muller K, Knapp A, Sauer G. Reversion of bovine papillomavirus-induced transformation and immortalization by a xanthate compound. Exp Cell Res 1985; 161: 541-50. Smith KT, Campo MS. ‘Hit and run’ transformation of mouse C127 cells by bovine papillomavirus type 4: The viral DNA is required for the initiation but not for maintenance of the transformed phenotype. Virology 1988; 164: 3 9 4 7 . Macnab JC. Herpes simplex virus and human cytomegalovirus: Their rBle in morphological transformation and genital cancers. J Gen Virol 1987; 68: 2525-50. Durst M, Croce CM, Gissmann L, Schwarz E, Huebner K. Papillomavirus sequences integrate near cellular oncogenes in some cervical carcinomas. Proc Natl Acad Sci USA 1987; 84: 107C-4.
Address for correspondence:
Tsuyoshi Iwasaka, M.D. Department of Obstetrics and Gynecology Saga Medical School 1-1, Nabeshima 5-chome Saga 849 Japan
Acta Obstet Gynecol Scand 71 (1992)
‘Hit and run’ oncogenesis by human papillomavirus type 18 DNA T. IWASAKA, Y. HAYASHI, M. YOKOYAMA, K. HARA,N. MATSUO AND H. SUGIMORI From the Department of Obstetrics and Gynecology, S a g a Medical School, S a g a , J a p a n
Actu Ohstet Gynecol Scand 1992; 71: 219-223 Transfection of an immortalized cell line (AE), derived from Syrian hamster embryo cells, with human papillomavirus type 18 (HPV 18) DNA induced morphological transformation and these transformed cells were tumorigenic in nude mice. Southern blot analysis revealed that the transfected viral DNA was retained in all the cell lines tested, however, all these transformed cells contained only less than one copy per cell of viral genome. Eleven cloned cell lines were established from a tumor cell line obtained after explantation of a tumor into a nude mouse. Two lines revealed no viral sequences by both Southern blot hybridization and polymerase chain reaction, whereas the nine others contained the remaining viral sequences. These results are highly suggestive of a ‘hit and run’ oncogenesis by this virus.
Key words: ‘Hit and run’ oncogenesis; human papillomavirus type 18; in vitro transformation; hamster embryo cells Submitted October 28, 1991 Accepted December 3 1 , 1991
Human papillomavirus (HPV) sequences are present in up to 90% of cervical cancer tissues, with the majority of these tumors containing either HPV 16 or HPV 18 DNA (1-5). This strongly suggests a causative rBle for these viruses in the development of malignancy. The transforming potentials of these two viruses have been well studied in vitro. The oncogenic potential of total genome of HPV 16 or HPV 18 has been shown in mouse 3T3 cells, rat 3Y-1 cells, and hamster AE cells, immortalized lines (6-9), but not in primary culture cells. It has also been shown that the E6 and E7 open reading frames (ORF) of these viruses have major transforming activity both in rodent and human cells (10-!4). From these studies it is clear that the E6 and E7 genes encode for the major transforming activity of HPV 16 and HPV 18. In all HPV-containing transformed cells or cervical cancer tissues analysed, E6 and E7 ORFs are always maintained. It is therefore
crucial to determine whether there is a continued requirement for these genes in transformed cells or whether these genes behave in a ‘hit and run’ fashion and, after the initial transforming event, are no longer required. In our previous study, we noted the oncogenic potential of HPV 16 and H P V 18 DNA on hamster embryo cells already immortalized with transfection of specific fragment of herpes simplex virus type 2(HSV-2) DNA. All the cell lines obtained by HPV 16 DNA transfection contained more than one copy per cell of viral DNA, while no viral DNA was detected in two cell lines obtained after transfection of HPV 18 DNA[9]. We proposed the ‘hit and run’ mechanism in the oncogenesis by H P V 18 DNA. In the present study, we asked whether HPV DNA is required for the transformation and for the maintenance of a cancer state.
Actci Ohstrt Gynecol Scand 71 (1YY2)
220
T. lwasuku et ul.
Table 1 . Sequences of oligonucleotide primer pairs and their locations in the HPV 18 genome
HPV 18 primer I HPV I X primer 2
Sequence
Location
5' TCACGAGCAATTAAGCGACT 3' 5' CTG AGCTTTCTACTACTAGC 3'
67(k6XY X(WX27
HPV 18 DNA was originally isolated from cervical carcinoma cells and cloned at its unique EcoRI site into pBR322 (2). This cloned form of DNA was obtained from the Japanese Cancer Research Resources Bank, Tokyo.
grams of each Hind 111 linearized plasmid D N A was mixed with 5 pg of salmon sperm D N A and precipitated with calcium phosphate. Aliquots of 0.5 ml were applied to subconfluent monolayers of A E cells (passage 62: 2.5 x 10' cells per 60 mm dish). Cultures were refed fresh medium at 5 hours after transfection and were then fed every 2 to 3 days. Foci were isolated with glass cylinders and then expanded for further analysis.
Cells and D N A trunsfection
nrmorigenicity
A E cells, an immortalized cell line, established by transfcction of Syrian hamster embryo cells with immortalizing sequences of HSV-2 DNA ( l S ) , were maintained in modified Dulbccco's Eagle reinforced medium with 10% fetal calf serum. Three micro-
Transformed cells were collected and suspended in serum-free culture medium. A total number of 1 X lo7 cells in 0.1 ml of the medium was injected subcutaneously into Balb/c nude mice (4 weeks old) and the mice were then examined weekly.
Methods Recombinant HPV D N A
Analysis of H P V D N A sequences High molecular weight cellular D N A was extracted from the transformed cells and the cell lines explanted from the tumors. Ten micrograms of cellular DNA from each cell line was digested with restriction endonuclease EcoRl or Pst- 1, electrophoresed. transferred to a nylon membrane and hybridization was carried out with the probe of HPV 18 DNA, as previously described (9).
Polymeruse chain reaction
Fig. 1 . Southern blot hybridization of DNA samples from transformed cells. Ten microgram of each cellular DNA was digested with both EcoRl and Pst-I, subjected to electrophoresis on 0.7"/, agarose gel, transferrcd to nylon filters and hybridized to a '?P-labeled EcoKI fragment of HPV 18. The DNA analysed was obtained from three HPV IH-transformed cell lines (18-1, 18-2 and 18-3: lanes 4, 6, and 8. respectively). four tumor cell lines (lS-lT, 18-2T, lX-3T, and 18-4T: lanes S, 7, 9, and 10. respectively), and A E cells transfected with pBR322 D N A (lane 3). Samples of HPV 1 X DNA corresponding to 1 and 3 copies per cell of HPV-1X DNA sequences were used for a reconstruction experiment (lanes 2 and I , respectively).
To detect a small amount of HPV DNA, polymerase chain reactin (PCR) was adopted. The region chosen for HPV 18 specific primers was within E 7 O R F (Table I). Amplification reaction was performed on one microgram of each cellular D N A using a DNA Amplification Reagent Kit (Perkin Elmer Cetus, Norwalk, U.S.A.). The samples were overlaid with mineral oil and subjected to 35 cycles of amplification. A cycle represents (a) denaturation for 1 minute at Y4"C, (b) annealing for 2 minutes at 55°C and (c) primer extension for 2 minutes at 72°C. To detect the amplified D N A fragment, 15 pI of the reaction mixture was analysed by electrophoresis on 3% (w/v) NuSieve gels, and the amplified products were visualized by staining with ethidium bromide.
'Hit und run' oncogenesis by H P V type I8
221
Polymerase chain reaction Analysis of D N A extracted from A E cells # (a negative control) and clones # I , #3, #4, and #8 was performed by thirty-five cycles of PCR with HPV 18 specific primers. HPV 18 DNA as well as clones #3 and #4 yielded clear bands between 154bp and 220bp corresponding in size to that expected for HPV 18 D N A (158bp), while no corresponding band was visualized in cases o f AE cells as well as clones # 1 and #8 (Fig. 3).
Tumorigenicity Fig, 2. Southcrn blot hybridization of DNA samples from cloncd cell lines. Cellular DNA was extracted from each cloned cell line obtained from ii t u m o r cell line (18-T4). digested with Pst-1, and subjected to Southern blot hybridization. Lanes 1-11: clones # I to # I t , respectively. Lane 12: 18-4T. Lanes 13 and 14: 1 and 3 copies per cell of HPV 18 DNA sequences as a positive control.
TOassess the tumorigenic activity of the four cloned Cell lines, each clone was further propagated and transplanted into nude mice. HPV DNA-positive clones formed tumors in nude mice about four weeks after transplantation (#3 and #4), About 5 weeks were required for definite tumors to form in of HPV DNA negative clones ( # I and #8).
Results Southern blot hybridizution High molecular weight DNA was extracted from four cell lines. designated as 18-1, 18-2, 18-3, and 18-4, obtained from separated foci, four cell lines, designated as 18-1T, 18-2T, 18-3T, and 18-4T, obtained by explant culture o f the tumors derived from the respective cell lines. and A E cells transfected with pBR322. To examine the retention of HPV DNA sequences, total DNAs extracted from these cell lines were digested with EcoRI and Pst-1 and subjected to Southern blot analysis with the "Plabeled HPV 18 D N A . The results showed that less than one copy per cell of HPV 18 DNA was detected in all the cell lines tested (Fig. I ) . However, these apparent lower copy numbers in these cell lines might be the reflection of lower hybridization efficiency in high molecular DNA. compared with that in purified HPV DNA. To examine whether the HPV D N A was maintained in all the transformed cells, hybridization analysis was performed in the cloncd cell lines. Eleven clones were obtained from the 18-4T cell line which contains the least HPV D N A and cellular D N A was extracted from each clone. Hybridization was carried out in the same fashion after digestion with Pst-1. Of the eleven clones, eight contained approximately one copy per cell of viral D N A and one contained 0.5 copy per cell (#4), while the remaining two revealed no viral D N A (#I and #8) (Fig. 2).
Fig. 3 . PCR prcducts revealed by ethidium bromide staining of electrophoretic gel. The thermophilic Taq polymerase was used in 35 cycles of amplification before electrophoresis. Lane I : PCR products form 1 pg of extracted DNA from A E cells. Lanes 2-5: PCR products from I pg of extracted D N A from cloned cell lines #1, #3,#4, and #8. respectively. Lane 6: PCR products from I pg of recombinant HPV 18 DNA as a positive control. Size markers (pBR322lHinf 1 digest) are given on the left and the band represcnting specific PCR products from HPV 18 DNA is shown by the arrow on the right.
222
T. lwasuka et a1
Discussion I n the oncogenesis by HPV DNA, it has been generally accepted that HPV DNA or its expression is essential for the maintenance of a transformed state. Our previous observation of the oncogencsis by HPV 16 DNA is consistent with this idea (9). In contrast, the present study showed that HPV 18 sequences were apparently not present in some cloned cell lines derived from a tumor cell line obtained after transfection of this virus DNA. This is in accord with our previous observation that viral sequences were absent in the HPV 18-induced transformed cells (9). Thereafter repeated Southern blot hybridization and PCR revealed viral sequences in these transformed cells, albeit the amount being scanty. On the basis of the results presented here, it is suggested that viral sequences are not always required for the maintenance of a transformed state in the oncogenesis by HPV 18, in other words HPV 18 behaves in a ‘hit and run’ fashion. The apparent contradictions in the r61e of HPV 16 and HPV 18 in transforming AE cells might be due to differences in affinity to cellular DNA or to differences in the facility to delete from the transformed celis. Consistent with this idea is the observation that HPV 16 DNA sequences are more frequently present in cervical cancer tissues than are those of HPV 18 DNA, and moreover, in a fair percentage, viral sequences are absent or are present in a very small amount. This different potential to transform cells in vitro between HPV 16 and HPV 18 seems to be analogous to that between bovine papillomavirus (BPV)-1 and BPV-4. BPV-1-transformed cells contain multiple copies of viral DNA and the expression of the BPV-1 genome appears to be required for the maintenance of the transformed state (16, 17). While, in BPV-4-induced transformation, only 9 of 60 ceH lines harbour BPV-4 DNA and there is no evident relationship between the presence of the viral DNA and the transformed phenotype, which is in accord with the ‘hit and run’ oncogenesis (18). The contribution of cellular oncogenes or tumor suppression genes must be considered for the ‘hit and run’ oncogenesis. Expression or insertion of HPV sequences might directly induce amplification or overexpression of cellular oncogenes or involve deletion or a point mutation of tumor suppression genes, consequently the cell line is converted to a tumorigenic one. Once oncogenic transformation has been established, viral sequences are not always necessary for the maintenance of a transformed state. This apparent mechanism in the HPV oncogenesis is not unexpected since the same mechanism functions for transformation by DNA tumor viruses, herAcia Obstei Gynecol &and 71 (1YY2)
pes simplex virus, human cytomegalovirus and adenovirus, where no consistent viral sequences are maintained on expression in tumor cells (19). We cannot completely rule out the possibility that this transformation is spontaneous. It is, however, unlikely because the AE cell line is stable and no spontaneous transformation has been observed during more than 60 passages and in any experimental group set up for control transfection. We have still to rule out the possibility that the transformed cells have retained a small fragment of HPV 18 DNA that activates cellular oncogenes by insertional mutagenesis, if this fragment is too small to be detected by Southern blot hybridization or PCR. Such a mechanism has been postulated for HPV 18 in some cervical carcinomas where the viral genome is integrated upstream of the c-myc oncogene (20). At the present time, it is difficult to exclude this possibility and conversely it is also difficult to prove that the retained small fragment is essential for the maintenance of a transformed state. Furthermore, deletion of viral DNA in tumor cells is not always advantageous for tumorigenicity, as shown in our in vivo experiment in which the viral DNA positive-clones formed tumors in nude mice one week earlier than those lacking the viral DNA. More detailed studies, including cellular oncogenes and tumor suppression genes, on the cell lines, with or without viral DNA, are expected to provide further information on the r61e of the virus in the development of papillomavirus-associated cancers in humans.
Acknowledgment We would like to thank Mariko Ohara for careful review of the manuscript.
References Durst M, Gissmann L, lkenberg H, zur Hausen H. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions. Proc Natl Acad Sci USA 1983; 80. 3812-5. 2 . Boshart M, Gissmann L, Ikenberg H, Kleinheinz A , Scheurlen W, zur Hausen H. A new type of papillomavirus DNA, its presence in genital cancer biopsies and in cell lines derived from cervical cancer. EMBO J 1.
1984; 3: 1151-7. 3. zur Hausen H. Genital papillomavirus infection. Prog Med Virol 1985; 32: 15-21. 4. McCance DJ. Human papillomavirus and cancer. Biochem Biophys Acta 1986; 823: 195-205. 5. Macnab JC, Walkinshow SA, Cordiner JW, Clements
‘Hit and run’ oncogenesis by HPV type 18 JB. Human papillomavirus in clinically and histologically normal tissue o f patients with genital cancer. N Engl J Med 19x6; 315: 1052-8. 6. Yasumoto S, Burkhardt AL, Doniger J , DiPaolo JA. Human papillomavirus typc 16 DNA-induced malignant transformation of NIH 3T3 cells. J Virol 1986; 57: 572-7. 7. Tsunokawa Y , Takehc N , Kasamatsu T, Terada M, Sugimura T. Transforming activity o f human papillomavirus type 16 DNA sequences in a cervical cancer. Proc Natl Acad Sci USA 1986; 83: 220(&3. 8. Matlashcwski G , Osborn K , Murray A. Banks L. Crawford L. Cancer cells. Vol 5: Papillomaviruses. NY: Cold Spring Harbor Laboratory Press, 1987: 195-9. 9. lwasaka T, Yokoyama M, Hayashi Y , Sugimori H. Combined herpes simplex virus type 2 and human papillomavirus type 16 o r 18 deoxyribonucleic acid leads to oncogenic transformation. Am J Obstet Gynecol 1988; 159: 1251-5. 10. Bedell MA, Jones KH, Laimins LA. The E6-E7 region o f human papillomavirus type 18 is sufficient for transformation of NIH3T3 and Rat-I cells. J Virol 1987; 61: 363540. 11. Pirisi L, Yasumoto S, Feller M, Doniger J, DiPaolo JA. Transformation of human fibroblasts and keratinocytes with human papillomavirus type 16 DNA. J Virol 1987; 61: 1061-6. 12. Kanda T, Furuno A , Yoshiike K. Human papillomavirus type I6 open reading frame E7 encodes a transforming gene for rat 3Y1 cells. J Virol 1988; 62: 61G3. 13. Schlegel R , Phelps WC, Zhang YL, Barbosa M. Quantitative keratinocyte assay detects two biological activities of human papillomavirus DNA and identifies viral types associated with cervical carcinoma. EMBO J 1988; 7: 3181-7. 14. Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R . The E6 and E7 genes of the human papillomavirus
15.
16.
17.
18.
19.
20.
223
type 16 together are necessary and sufficient for transformation of primary human keratinocytes. J Virol 1989; 63: 4417-21. Hayashi Y . lwasaka T, Smith CC, Aurelian L, Lewis GK, Ts’o POP. Multistep transformation by defined fragments of herpes simplex virus type 2 DNA: oncogenic rcgion and its gcne product. Proc Natl Acad Sci USA 1985; 82: 8493-7. Law M-F, Lowy DR, Dvoretzky J , Howley P. Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. Proc Natl Acad Sci USA 1981; 78: 2727-31. Amtmann E, Muller K, Knapp A, Sauer G. Reversion of bovine papillomavirus-induced transformation and immortalization by a xanthate compound. Exp Cell Res 1985; 161: 541-50. Smith KT, Campo MS. ‘Hit and run’ transformation of mouse C127 cells by bovine papillomavirus type 4: The viral DNA is required for the initiation but not for maintenance of the transformed phenotype. Virology 1988; 164: 3 9 4 7 . Macnab JC. Herpes simplex virus and human cytomegalovirus: Their rBle in morphological transformation and genital cancers. J Gen Virol 1987; 68: 2525-50. Durst M, Croce CM, Gissmann L, Schwarz E, Huebner K. Papillomavirus sequences integrate near cellular oncogenes in some cervical carcinomas. Proc Natl Acad Sci USA 1987; 84: 107C-4.
Address for correspondence:
Tsuyoshi Iwasaka, M.D. Department of Obstetrics and Gynecology Saga Medical School 1-1, Nabeshima 5-chome Saga 849 Japan
Acta Obstet Gynecol Scand 71 (1992)
