VIROLOGY
181, 374~377 (1991)
Differences
in Transformation
Activity
between
LUISA LINA
HPV-18 and HPV-16 Map to the Viral LCR-EG-E7
Region
VILLA AND RICHARD SCHLEGEL’
Ludwig institute for Cancer Research, Sao Paula, Brazil; Laboratory of Tumor Virus B/o/ogy, Nat/anal Cancer Institute, National Institutes of Health, Bethesda, Maryland,. and Department of Pathology, Georgetown Un/vers/ty, Washmgton D. C. Received
October 8, 1990; accepted
December
4, 1990
Homologous, subgenomic fragments of the viral LCR and E6/E7 transforming genes of HPV-18 and HPV-16 were amplified from several primary cervical, penile, and vulvar tumors and cloned into a pUC-18-derived vector. When assayed by a quantitative transformation assay using primary human keratinocytes. the subgenomic regions of HPV-16 and HPV-18 exhibited transforming activities similar to that of the full-length, prototype HPV genomes. More importantly, the HPV-18 LCR-EG-E7 region was approximately lo- to 50-fold more active than that of HPV-16. These studies demonstrate (1) that the transforming activity differences previously observed between prototype HPV-16 and HPV-18 map to the LCR-EG-E7 region, and (2) that individual and independent isolates of HPV-16 and HPV-18 exhibit consisQ 1991 Academic press, IW. tent differences in transforming potential, even when isolated from different anatomic sites.
Human papillomaviruses (HPVs) are associated with the development of malignant squamous neoplasms of the cervix, vulva, penis, and anus and also induce the formation of benign squamous epidermal tumors (warts) ( 1). Although there are greater than 60 different genotypes of HPV, only a limited number are associated with malignant anogenital tumors. HPV-16 and HPV-18 account for approximately 70% of HPV-positive cervical carcinomas and the remaining cervical tumors either contain additional HPV types or lack HPV (approximately 5%). HPV-16 and HPV-18 exhibit very different biological and biochemical activities from other HPVs which infect the genital tract (such as HPV-6 and HPV-11). For example, HPV-16 and HPV18 (but not HPV-6 or HPV-11) can transform primary keratinocytes in vitro (Z-5). Genetic studies have shown that both the E6 and E7 genes are sufficient and necessary for the efficient immortalization of human keratinocytes (6-8). In addition, only the E6 and E7 genes of these malignant-associated HPVs associate efficiently with the cellular tumor suppressor proteins p53 and pRb, respectively (9, 10). Presumably this biochemical difference is responsible for their unique in vitro activity as well as their association with human malignancy. Although both HPV-16 and HPV-18 transform primary keratinocytes in vitro, they do so with markedly different efficiencies. HPV-18 is approximately 1O-fold more active for keratinocyte transformation than HPV16 (8). The molecular basis for this difference has not been defined but could result from properties unique to
’ To whom reprint requests
the prototype HPV isolates (such as the interrupted El ORF in the prototype HPV-16 genome), from differences in the E6/E7 genes which induce the in vitro transformation of keratinocytes, or from differences in the transcriptional regulation of the E6/E7 genes mediated by the LCR or E2 protein. This study was designed to determine whether the higher transforming activity of the prototype HPV-18 was a general property of independent isolates, and whether its higher biological activity mapped to the LCR/EG/E7 region of these oncogenic HPVs. The strategy for the isolation of the LCR-EG-E7 region from human tumors is illustrated in Fig. 1. In brief, the procedure consisted of purifying cellular DNA from carcinomas of the uterine cervix, vulva, and penis from patients in Brazil using standard techniques ( 17). The HPV DNA present in each sample was typed by Southern hybridization and only those samples containing HPV-16 or HPV-18 were further studied. One hundred nanograms of purified tumor DNA was then amplified by polymerase chain reaction (PCR) ( 72) using the designated oligonucleotides corresponding to the 5’ end of the LCR (defined as the terminus of the Ll ORF) and the 3’ end of the E7 gene (immediately following the termination codon). In the case of HPV-18, an additional oligonucleotide was used to amplify the LCR from an internal position (bp 7201 ). For each of the oligonucleotides, a nonhomologous 5’ terminus encoding either a HindIll or Xhol restriction site was included to facilitate the unidirectional cloning of the amplified fragment into a pUC18 plasmid containing the SV40 early polyadenylation signals ( 13). The orientation of the amplified HPV LCR-EG-E7 region was verified by direct sequencing.
should be addressed.
0042-6822/g 1 $3.00 Copyright 0 1991 by Academic Press. Inc All rights of reproduction I” any form resewed.
374
SHORT COMMUNICATIONS
375
Hind III CGTAllAAGCllMCGTAAGCTGTAAGTATTG nb fq
GTATCAMGCTT
-T
GCGTGTACGTGCCAOOAAG 7114
CGT All AA0 Cl7 TOT AT0 Al-l GCA llG TAT GG
XhOl AGA Tffi TAG CGA CTA GOA CXjG AGC TCC GAT TA
I=/
878
COT TGT TAC COA CTA GGT fl,G AGC TCA GTA GC 927
FIG. 1. Amplification and cloning of the LCR-EG-E7 region of HPV-16 and HPV-18 from human tumors. DNA was extracted and purified from primary human cervical, penile, and vulvar carcinomas. Oligonucleotides corresponding to the 5’end of the LCR (defined as the terminus of the Ll ORF in both HPV-16, bp 7141, and HPV-18, bp 7114)and the 3’endof the E7 ORF(bp 878for HPV-16 and bp927forHPV-18)wereusedto amplify the LCR-EG-E7 region (- 1.6 kb) with inclusive HindIll and Xhol cloning sites as indicated. 100 ng of tumor DNA or 0.1 ng of plasmid DNA were added to a 100-~1 reaction mixture containing 10 mn/r Tris-HCI, pH 8.3, 50 mM KCI, 1.5 mM MgCI,, 0.01% gelatin, 200 pM each dATP, dlTP, dCTP, and dGTP, 1 pM each primer, and 2.5 units of Taq polymerase (Perkin-Elmer Cetus), and submitted to 30 cycles of 1 min at 94”, 2 min at 55”, 5 min at 70” in a thermal cycler (Perkin-Elmer Cetus). DNA fragments of the expected size were gel purified and cloned unidirectionally into a pUCl8-derived vector containing the SV40 polyadenylation signals ( 13).
For purposes of comparison with HPV DNA amplified from tumors, we also amplified the same LCR-EGE7 regions from the prototype clones of HPV-16 ( 14) and HPV-18 ( 15) (kindly provided by Dr. H. zur Hausen, German Cancer Research Center, Heidelberg, FRG). Each of the above constructed plasmids was then evaluated in a quantitative transformation assay and compared with full-length genomes of HPV-16 and HPV-18. Similar to previous studies (4, 8), the complete HPV-18 genome (~18) induced approximately 90 transformed keratinocyte colonies whereas the HPV16 genome (~16) induced only 8 colonies (Fig. 2). Nasseri, Temple, and Lorincz have recently confirmed this biological difference between HPV-18 and HPV-16 as well as demonstrated that other HPV types associated with cervical carcinoma also exhibit characteristic differences in transformation efficiency (personal communication). This marked difference between HPV-18 and HPV-16 was observed for the isolated LCR-EG-E7 regions of HPV-18 (pf18) and HPV-16 (pf 16). That is, pf 18 induced 100 transformed colonies
in contrast to pf16, which induced 4. The relative transforming activity of HPV-18 to HPV-16 is therefore maintained in a subgenomic fragment containing only the LCR-EG-E7 region, indicating that other viral genes such as El and E2 are not responsible for this difference in activity. In addition, most of the transforming activity of HPV-18 can be accounted for by this subgenomic fragment, although differences in plasmid molarity might contribute slightly to this observation. Transfections were performed with equivalent amounts of DNA (5 pg) and the plasmid size for pl8 is approximately 11 kb whereas that for pf18 is 5 kb. Since saturation for lipofection-mediated transfection occurs between 2.5 and 5 pg DNA, it is possible that some of the activity of pf18 is attributable to a relatively higher concentration of DNA. At a minimum, however, the LCR-EG-E7 fragment of HPV-18 exhibits 50% of the activity of the full-length genome. Figure 3 demonstrates a dramatic difference in biological activity between new isolates of the HPV-16 and HPV-18 LCR-EG-E7 regions. The five new iso-
SHORT COMMUNICATIONS
376
x c $ *O -0 ' B
80
i f
40
20
0 pl8
~118
~16
pfl6
2. Comparison of the transforming activities of full-length HPV-18 (~18) and HPV-16 (~16) with their respective subgenomic fragments, pf18 and pf16. Full-length genomes of HPV-18 and HPV16 cloned into pUC18 were compared with the subgenomic fragments described in Fig. 1 for their ability to transform human keratinocytes in vitro. 10’ primary keratinocytes were transfected with 5 rg DNA of each of the listed plasmid DNAs using cationic liposomes (20) as previously described (8). In brief, the HPV-16 and HPV-18 plasmids (~16 and ~18) were digested with BarnHI or Ncol, respectively, to remove the viral genome from the vector. In the case of the subgenomic fragments, the plasmids were cleaved with HindIll to linearize the DNA. Following transfection, the keratinocytes were passaged (1:3 split) in serum-free medium (GIBCO) and after 3-5 days (at cell confluence) changed to medium containing 2.0 rnM calcium and 10% serum. Proliferating, differentiation-resistant colonies were counted after 3 weeks (8). The insert exhibits the reproducibility of the difference observed’ between pf16 and pf18. Each bar represents an independent transformation assay for each construct. FIG.
lates of HPV-16 induced 3-7 transformed colonies of keratinocytes whereas the two new HPV-18 isolates induced 40-60 colonies. This 1O-fold difference in activity parallels the difference observed previously between the prototype isolates of HPV-16 and HPV-18 and indicates that variations in the LCR-EG-E7 region account for the enhanced transforming activity of HPV18. It is also noteworthy that despite the variations in the anatomic site of origin of the five HPV-16 isolates, they exhibited very similar biological activities. Two of the amplified HPVs (ptl6vl and ptl6v2) were derived from bowenoid papulosis of the vulva. The amplified LCR-EG-E7 regions of several Brazilian HPV-16 isolates have been sequenced and contain several base changes from the prototype HPV-16 genome (unpublished data). These differences were present in several independent PCR amplifications and
were identical in several independent tumors, suggesting a variation in the HPV-16 genome which is present in the Brazilian population. PCR amplifications of the prototype HPV-16 and HPV-18 were also sequenced and were shown to be identical to the starting plasmid DNA (data not shown). Further mapping of the LCR-EG-E7 fragment should define the specific region responsible for the increased activity of HPV-18. It is possible that variations in the E6 and E7 proteins of HPV-18 and HPV-16 might account for the variant transforming activity of these viruses. However, the biochemical data currently available suggest that the E6 and E7 proteins of both of these viruses are very similar in their ability to associate with the cellular proteins of p53 and pRb, respectively (9, 10). Differences in viral promoter or enhancer strengths might also explain the variation observed between HPV-16 and HPV-18. Indeed, a previous report has documented that the HPV-18. promoter is more active in both primary human keratinocytes and cells derived from a cervical carcinoma than the HPV-16 promoter ( 16). Interestingly, however, there is no
80
FIG. 3. Comparison of the transforming activity of the LCR-EG-E7 region amplified from HPV-16 and HPV-18 genomes present in tumor cells. Subgenomic fragments of HPV-16 and HPV-18 amplified from genital tumors were cloned and transfected into keratinocytes as described in Figs. 1 and 2. Specimens consisted of HPV-16.containing squamous cell carcinomas of the uterine cervix (ptl6c), penis (ptl6p1, ptl p2), and bowenoid papulosis of the vulva (ptl6v1, ptl6v2). HPV- 18 DNA was amplified from HeLa cell DNA using oligonucleotides positioned at either bp 71 14 (pHeLa) or bp 7201 (pHeLa*). There was no detectable difference in the transforming activity of the HPV-18 DNA amplified from these two adjacent sites. ptl8*c represents amplified HPV-18 DNA from a cervical carcinoma using the oligonucleotide positioned at bp 7201. Attempts to amplify the HPV-18 genome with the oligonucleotide at bp 71 14 were unsuccessful, suggesting that this region was either rearranged or deleted In the tumor cells, Each bar represents the results of an independent transfection assay for each construct.
SHORT COMMUNICATIONS
quantitative difference in the stable expression of HPV RNA in the cell lines generated by HPV-16 or HPV-18, suggesting that the viral promoters are functioning at a similar level following cellular transformation (data not shown). However, it is possible that these promoters might function differently during the transient periods that precede keratinocyte immortalization (8). These in vitro studies showing a difference in the transforming activity of HPV-18 and HPV-16 also have a correlate in the ability of HPV-18 to effect a more rapid alteration of keratinocyte differentiation as assayed on collagen rafts in vitro ( 17), and in nude mouse grafts in vim (78). In addition, there is preliminary information that HPV-18 may be associated clinically with a more aggressive tumor phenotype ( 19). Despite this apparent difference in in vivo and in vitro activity, HPV-18 accounts for only 1O-20% of cervical carcinomas whereas HPV-16 accounts for 50-60%. It is very possible that this is attributable to a very low incidence rate of HPV-18 infection in the general population compared to HPV-16. If these different, malignancy-associated HPVs do indeed induce in viva transformation with different frequencies and produce different cellular phenotypes, it will further add to the importance of genotyping the HPVs present in clinical specimens. ACKNOWLEDGMENTS We thank Janet Byrne for oligonucleotide synthesis; Jesse Quintero, M. Cecilia Costa, and A. Novello-Netto for excellent technical assistance; and Carl C. Baker for providing the plasniid CCB 119. We also acknowledge Helen Romanczuk and Peter Howley for critical reading of the manuscript.
REFERENCES 1. ZUR HAUSEN, H., and SCHNEIDER,A., ln “The Papovaviridae” Saltman and P. M. Howley, Eds.), Vol. 2, pp. 246-263. num, New York, 1987.
(N. Ple-
377
2. PIRISI, L., YASUMOTO. S., FELLER, M., DONIGER,J., and DIPAOLO, J. A., J. viral. 61, 1061-1066 (1987). 3. DURST, M., DZARLIEVA-PETRUSEVSKA, R. T., BOUKAMP, P., FUSENIG, N. E., and GISSMANN, L., Oncogene 1,251-256 (1987). 4. SCHLEGEL, R., PHELPS, W. C. ZHANG, Y-L., and BARB&A, M., EMf3OJ. 7(10), 3181-3187(1988). 5. KAUR, P., and MCDOUGALL, J. K., Virology 173,302-310 ( 1989). 6. MUNGER, K., PHELPS,W. C., BUSS, V., HOWLEY,P. M., and SCHLEGEL, R., J. Virol. 63, 4417-4421 (1989). HAWLEY-NELSON, P., VOUSDEN, K. H., HUBBERT, N. L., Lowv, D. R., and SCHILLER,J., EMBO 1. 6, 3905-3910 (1989). BARBOSA, M. S., and SCHLEGEL, R., Oncogene 4, 1529-1523 (1989). WERNESS, B. A., LEVINE, A. J., and HOWLEY, P. M., Science (1990). 10. DYSON, N., HOWLEY, P. M., MUNGER, K., and HARLOW, E., Science 243,934-937 ( 1989). 11. SAMBROOK,J., FRITSCH,E. F.,and MANIATIS, T., “Molecular Cloning: A Laboratory Manual,” 2nd edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. 12. SAIKI, R., SCHARF, S., FALOONA, F., MULLIS, K. B., HORN, G. T., ERLICH, H. A., and ARNHEIM, N., Science 230, 1350-1354 (1985). 13. BAKER, C. C., In “Recombinant Systems in Protein Expression” (K. K. Alitalo, M. L. Huhtala, 1. Knowles, and A. Vaheri, Eds.), pp. 75-86. Elsevier, Amsterdam/New York, 1990. 14. DURST, M., GISSMANN, L., IKENBERG,H., and ZUR HAUSEN. H., froc. Nat/. Acad. Sci. USA 60, 3812-3815 (1983). 15. EOSHART, M., GISSMANN, L., IKENBERG, L., KLEINHEINZ, A., SCHEURLEN,W., and ZUR HAUSEN, H., EMBOJ. 3, 1151-l 157 (1984). 16. ROMANCZUK, H., THIERRY, F., and HOWLEY, P. M., 1. vi/o/. 64, 2849-2859 (1990). 17. MCCANCE, D. J., KOPAN, R., FUCHS, E., and LAIMINS, L., Proc. Natl. Acad. Sci. USA 65,7169-7 173 ( 1988). 18. WOODWORTH,C. D.. WAGGONER,S., BARNES,W., STOLER, M. H., and DIPAOLO, J. A., Cancer Res. 50, 3709-3715 (1990). 19. BARNES, W., DELGADO, G., KURMAN, R. J., PETRILLI,E. S., SMITH, D. M.. AHMED, S., LORINCZ,A. T., TEMPLE, G. F., BENNY JENSON, A.. and LANCASTER,W. D., Gynecol. Oncol. 29,267-273 (1988). 20. FELGNER,P. L., GADEK, T. R., HOLM, M., ROMAN, R., CHAN, H. W., WENZ, M., NORTHROP,J. P., RINGOLD, G. M., and DANIELSEN, M., Proc. Natl. Aced. Sci. USA 84, 7413-7417 (1987).
181, 374~377 (1991)
Differences
in Transformation
Activity
between
LUISA LINA
HPV-18 and HPV-16 Map to the Viral LCR-EG-E7
Region
VILLA AND RICHARD SCHLEGEL’
Ludwig institute for Cancer Research, Sao Paula, Brazil; Laboratory of Tumor Virus B/o/ogy, Nat/anal Cancer Institute, National Institutes of Health, Bethesda, Maryland,. and Department of Pathology, Georgetown Un/vers/ty, Washmgton D. C. Received
October 8, 1990; accepted
December
4, 1990
Homologous, subgenomic fragments of the viral LCR and E6/E7 transforming genes of HPV-18 and HPV-16 were amplified from several primary cervical, penile, and vulvar tumors and cloned into a pUC-18-derived vector. When assayed by a quantitative transformation assay using primary human keratinocytes. the subgenomic regions of HPV-16 and HPV-18 exhibited transforming activities similar to that of the full-length, prototype HPV genomes. More importantly, the HPV-18 LCR-EG-E7 region was approximately lo- to 50-fold more active than that of HPV-16. These studies demonstrate (1) that the transforming activity differences previously observed between prototype HPV-16 and HPV-18 map to the LCR-EG-E7 region, and (2) that individual and independent isolates of HPV-16 and HPV-18 exhibit consisQ 1991 Academic press, IW. tent differences in transforming potential, even when isolated from different anatomic sites.
Human papillomaviruses (HPVs) are associated with the development of malignant squamous neoplasms of the cervix, vulva, penis, and anus and also induce the formation of benign squamous epidermal tumors (warts) ( 1). Although there are greater than 60 different genotypes of HPV, only a limited number are associated with malignant anogenital tumors. HPV-16 and HPV-18 account for approximately 70% of HPV-positive cervical carcinomas and the remaining cervical tumors either contain additional HPV types or lack HPV (approximately 5%). HPV-16 and HPV-18 exhibit very different biological and biochemical activities from other HPVs which infect the genital tract (such as HPV-6 and HPV-11). For example, HPV-16 and HPV18 (but not HPV-6 or HPV-11) can transform primary keratinocytes in vitro (Z-5). Genetic studies have shown that both the E6 and E7 genes are sufficient and necessary for the efficient immortalization of human keratinocytes (6-8). In addition, only the E6 and E7 genes of these malignant-associated HPVs associate efficiently with the cellular tumor suppressor proteins p53 and pRb, respectively (9, 10). Presumably this biochemical difference is responsible for their unique in vitro activity as well as their association with human malignancy. Although both HPV-16 and HPV-18 transform primary keratinocytes in vitro, they do so with markedly different efficiencies. HPV-18 is approximately 1O-fold more active for keratinocyte transformation than HPV16 (8). The molecular basis for this difference has not been defined but could result from properties unique to
’ To whom reprint requests
the prototype HPV isolates (such as the interrupted El ORF in the prototype HPV-16 genome), from differences in the E6/E7 genes which induce the in vitro transformation of keratinocytes, or from differences in the transcriptional regulation of the E6/E7 genes mediated by the LCR or E2 protein. This study was designed to determine whether the higher transforming activity of the prototype HPV-18 was a general property of independent isolates, and whether its higher biological activity mapped to the LCR/EG/E7 region of these oncogenic HPVs. The strategy for the isolation of the LCR-EG-E7 region from human tumors is illustrated in Fig. 1. In brief, the procedure consisted of purifying cellular DNA from carcinomas of the uterine cervix, vulva, and penis from patients in Brazil using standard techniques ( 17). The HPV DNA present in each sample was typed by Southern hybridization and only those samples containing HPV-16 or HPV-18 were further studied. One hundred nanograms of purified tumor DNA was then amplified by polymerase chain reaction (PCR) ( 72) using the designated oligonucleotides corresponding to the 5’ end of the LCR (defined as the terminus of the Ll ORF) and the 3’ end of the E7 gene (immediately following the termination codon). In the case of HPV-18, an additional oligonucleotide was used to amplify the LCR from an internal position (bp 7201 ). For each of the oligonucleotides, a nonhomologous 5’ terminus encoding either a HindIll or Xhol restriction site was included to facilitate the unidirectional cloning of the amplified fragment into a pUC18 plasmid containing the SV40 early polyadenylation signals ( 13). The orientation of the amplified HPV LCR-EG-E7 region was verified by direct sequencing.
should be addressed.
0042-6822/g 1 $3.00 Copyright 0 1991 by Academic Press. Inc All rights of reproduction I” any form resewed.
374
SHORT COMMUNICATIONS
375
Hind III CGTAllAAGCllMCGTAAGCTGTAAGTATTG nb fq
GTATCAMGCTT
-T
GCGTGTACGTGCCAOOAAG 7114
CGT All AA0 Cl7 TOT AT0 Al-l GCA llG TAT GG
XhOl AGA Tffi TAG CGA CTA GOA CXjG AGC TCC GAT TA
I=/
878
COT TGT TAC COA CTA GGT fl,G AGC TCA GTA GC 927
FIG. 1. Amplification and cloning of the LCR-EG-E7 region of HPV-16 and HPV-18 from human tumors. DNA was extracted and purified from primary human cervical, penile, and vulvar carcinomas. Oligonucleotides corresponding to the 5’end of the LCR (defined as the terminus of the Ll ORF in both HPV-16, bp 7141, and HPV-18, bp 7114)and the 3’endof the E7 ORF(bp 878for HPV-16 and bp927forHPV-18)wereusedto amplify the LCR-EG-E7 region (- 1.6 kb) with inclusive HindIll and Xhol cloning sites as indicated. 100 ng of tumor DNA or 0.1 ng of plasmid DNA were added to a 100-~1 reaction mixture containing 10 mn/r Tris-HCI, pH 8.3, 50 mM KCI, 1.5 mM MgCI,, 0.01% gelatin, 200 pM each dATP, dlTP, dCTP, and dGTP, 1 pM each primer, and 2.5 units of Taq polymerase (Perkin-Elmer Cetus), and submitted to 30 cycles of 1 min at 94”, 2 min at 55”, 5 min at 70” in a thermal cycler (Perkin-Elmer Cetus). DNA fragments of the expected size were gel purified and cloned unidirectionally into a pUCl8-derived vector containing the SV40 polyadenylation signals ( 13).
For purposes of comparison with HPV DNA amplified from tumors, we also amplified the same LCR-EGE7 regions from the prototype clones of HPV-16 ( 14) and HPV-18 ( 15) (kindly provided by Dr. H. zur Hausen, German Cancer Research Center, Heidelberg, FRG). Each of the above constructed plasmids was then evaluated in a quantitative transformation assay and compared with full-length genomes of HPV-16 and HPV-18. Similar to previous studies (4, 8), the complete HPV-18 genome (~18) induced approximately 90 transformed keratinocyte colonies whereas the HPV16 genome (~16) induced only 8 colonies (Fig. 2). Nasseri, Temple, and Lorincz have recently confirmed this biological difference between HPV-18 and HPV-16 as well as demonstrated that other HPV types associated with cervical carcinoma also exhibit characteristic differences in transformation efficiency (personal communication). This marked difference between HPV-18 and HPV-16 was observed for the isolated LCR-EG-E7 regions of HPV-18 (pf18) and HPV-16 (pf 16). That is, pf 18 induced 100 transformed colonies
in contrast to pf16, which induced 4. The relative transforming activity of HPV-18 to HPV-16 is therefore maintained in a subgenomic fragment containing only the LCR-EG-E7 region, indicating that other viral genes such as El and E2 are not responsible for this difference in activity. In addition, most of the transforming activity of HPV-18 can be accounted for by this subgenomic fragment, although differences in plasmid molarity might contribute slightly to this observation. Transfections were performed with equivalent amounts of DNA (5 pg) and the plasmid size for pl8 is approximately 11 kb whereas that for pf18 is 5 kb. Since saturation for lipofection-mediated transfection occurs between 2.5 and 5 pg DNA, it is possible that some of the activity of pf18 is attributable to a relatively higher concentration of DNA. At a minimum, however, the LCR-EG-E7 fragment of HPV-18 exhibits 50% of the activity of the full-length genome. Figure 3 demonstrates a dramatic difference in biological activity between new isolates of the HPV-16 and HPV-18 LCR-EG-E7 regions. The five new iso-
SHORT COMMUNICATIONS
376
x c $ *O -0 ' B
80
i f
40
20
0 pl8
~118
~16
pfl6
2. Comparison of the transforming activities of full-length HPV-18 (~18) and HPV-16 (~16) with their respective subgenomic fragments, pf18 and pf16. Full-length genomes of HPV-18 and HPV16 cloned into pUC18 were compared with the subgenomic fragments described in Fig. 1 for their ability to transform human keratinocytes in vitro. 10’ primary keratinocytes were transfected with 5 rg DNA of each of the listed plasmid DNAs using cationic liposomes (20) as previously described (8). In brief, the HPV-16 and HPV-18 plasmids (~16 and ~18) were digested with BarnHI or Ncol, respectively, to remove the viral genome from the vector. In the case of the subgenomic fragments, the plasmids were cleaved with HindIll to linearize the DNA. Following transfection, the keratinocytes were passaged (1:3 split) in serum-free medium (GIBCO) and after 3-5 days (at cell confluence) changed to medium containing 2.0 rnM calcium and 10% serum. Proliferating, differentiation-resistant colonies were counted after 3 weeks (8). The insert exhibits the reproducibility of the difference observed’ between pf16 and pf18. Each bar represents an independent transformation assay for each construct. FIG.
lates of HPV-16 induced 3-7 transformed colonies of keratinocytes whereas the two new HPV-18 isolates induced 40-60 colonies. This 1O-fold difference in activity parallels the difference observed previously between the prototype isolates of HPV-16 and HPV-18 and indicates that variations in the LCR-EG-E7 region account for the enhanced transforming activity of HPV18. It is also noteworthy that despite the variations in the anatomic site of origin of the five HPV-16 isolates, they exhibited very similar biological activities. Two of the amplified HPVs (ptl6vl and ptl6v2) were derived from bowenoid papulosis of the vulva. The amplified LCR-EG-E7 regions of several Brazilian HPV-16 isolates have been sequenced and contain several base changes from the prototype HPV-16 genome (unpublished data). These differences were present in several independent PCR amplifications and
were identical in several independent tumors, suggesting a variation in the HPV-16 genome which is present in the Brazilian population. PCR amplifications of the prototype HPV-16 and HPV-18 were also sequenced and were shown to be identical to the starting plasmid DNA (data not shown). Further mapping of the LCR-EG-E7 fragment should define the specific region responsible for the increased activity of HPV-18. It is possible that variations in the E6 and E7 proteins of HPV-18 and HPV-16 might account for the variant transforming activity of these viruses. However, the biochemical data currently available suggest that the E6 and E7 proteins of both of these viruses are very similar in their ability to associate with the cellular proteins of p53 and pRb, respectively (9, 10). Differences in viral promoter or enhancer strengths might also explain the variation observed between HPV-16 and HPV-18. Indeed, a previous report has documented that the HPV-18. promoter is more active in both primary human keratinocytes and cells derived from a cervical carcinoma than the HPV-16 promoter ( 16). Interestingly, however, there is no
80
FIG. 3. Comparison of the transforming activity of the LCR-EG-E7 region amplified from HPV-16 and HPV-18 genomes present in tumor cells. Subgenomic fragments of HPV-16 and HPV-18 amplified from genital tumors were cloned and transfected into keratinocytes as described in Figs. 1 and 2. Specimens consisted of HPV-16.containing squamous cell carcinomas of the uterine cervix (ptl6c), penis (ptl6p1, ptl p2), and bowenoid papulosis of the vulva (ptl6v1, ptl6v2). HPV- 18 DNA was amplified from HeLa cell DNA using oligonucleotides positioned at either bp 71 14 (pHeLa) or bp 7201 (pHeLa*). There was no detectable difference in the transforming activity of the HPV-18 DNA amplified from these two adjacent sites. ptl8*c represents amplified HPV-18 DNA from a cervical carcinoma using the oligonucleotide positioned at bp 7201. Attempts to amplify the HPV-18 genome with the oligonucleotide at bp 71 14 were unsuccessful, suggesting that this region was either rearranged or deleted In the tumor cells, Each bar represents the results of an independent transfection assay for each construct.
SHORT COMMUNICATIONS
quantitative difference in the stable expression of HPV RNA in the cell lines generated by HPV-16 or HPV-18, suggesting that the viral promoters are functioning at a similar level following cellular transformation (data not shown). However, it is possible that these promoters might function differently during the transient periods that precede keratinocyte immortalization (8). These in vitro studies showing a difference in the transforming activity of HPV-18 and HPV-16 also have a correlate in the ability of HPV-18 to effect a more rapid alteration of keratinocyte differentiation as assayed on collagen rafts in vitro ( 17), and in nude mouse grafts in vim (78). In addition, there is preliminary information that HPV-18 may be associated clinically with a more aggressive tumor phenotype ( 19). Despite this apparent difference in in vivo and in vitro activity, HPV-18 accounts for only 1O-20% of cervical carcinomas whereas HPV-16 accounts for 50-60%. It is very possible that this is attributable to a very low incidence rate of HPV-18 infection in the general population compared to HPV-16. If these different, malignancy-associated HPVs do indeed induce in viva transformation with different frequencies and produce different cellular phenotypes, it will further add to the importance of genotyping the HPVs present in clinical specimens. ACKNOWLEDGMENTS We thank Janet Byrne for oligonucleotide synthesis; Jesse Quintero, M. Cecilia Costa, and A. Novello-Netto for excellent technical assistance; and Carl C. Baker for providing the plasniid CCB 119. We also acknowledge Helen Romanczuk and Peter Howley for critical reading of the manuscript.
REFERENCES 1. ZUR HAUSEN, H., and SCHNEIDER,A., ln “The Papovaviridae” Saltman and P. M. Howley, Eds.), Vol. 2, pp. 246-263. num, New York, 1987.
(N. Ple-
377
2. PIRISI, L., YASUMOTO. S., FELLER, M., DONIGER,J., and DIPAOLO, J. A., J. viral. 61, 1061-1066 (1987). 3. DURST, M., DZARLIEVA-PETRUSEVSKA, R. T., BOUKAMP, P., FUSENIG, N. E., and GISSMANN, L., Oncogene 1,251-256 (1987). 4. SCHLEGEL, R., PHELPS, W. C. ZHANG, Y-L., and BARB&A, M., EMf3OJ. 7(10), 3181-3187(1988). 5. KAUR, P., and MCDOUGALL, J. K., Virology 173,302-310 ( 1989). 6. MUNGER, K., PHELPS,W. C., BUSS, V., HOWLEY,P. M., and SCHLEGEL, R., J. Virol. 63, 4417-4421 (1989). HAWLEY-NELSON, P., VOUSDEN, K. H., HUBBERT, N. L., Lowv, D. R., and SCHILLER,J., EMBO 1. 6, 3905-3910 (1989). BARBOSA, M. S., and SCHLEGEL, R., Oncogene 4, 1529-1523 (1989). WERNESS, B. A., LEVINE, A. J., and HOWLEY, P. M., Science (1990). 10. DYSON, N., HOWLEY, P. M., MUNGER, K., and HARLOW, E., Science 243,934-937 ( 1989). 11. SAMBROOK,J., FRITSCH,E. F.,and MANIATIS, T., “Molecular Cloning: A Laboratory Manual,” 2nd edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. 12. SAIKI, R., SCHARF, S., FALOONA, F., MULLIS, K. B., HORN, G. T., ERLICH, H. A., and ARNHEIM, N., Science 230, 1350-1354 (1985). 13. BAKER, C. C., In “Recombinant Systems in Protein Expression” (K. K. Alitalo, M. L. Huhtala, 1. Knowles, and A. Vaheri, Eds.), pp. 75-86. Elsevier, Amsterdam/New York, 1990. 14. DURST, M., GISSMANN, L., IKENBERG,H., and ZUR HAUSEN. H., froc. Nat/. Acad. Sci. USA 60, 3812-3815 (1983). 15. EOSHART, M., GISSMANN, L., IKENBERG, L., KLEINHEINZ, A., SCHEURLEN,W., and ZUR HAUSEN, H., EMBOJ. 3, 1151-l 157 (1984). 16. ROMANCZUK, H., THIERRY, F., and HOWLEY, P. M., 1. vi/o/. 64, 2849-2859 (1990). 17. MCCANCE, D. J., KOPAN, R., FUCHS, E., and LAIMINS, L., Proc. Natl. Acad. Sci. USA 65,7169-7 173 ( 1988). 18. WOODWORTH,C. D.. WAGGONER,S., BARNES,W., STOLER, M. H., and DIPAOLO, J. A., Cancer Res. 50, 3709-3715 (1990). 19. BARNES, W., DELGADO, G., KURMAN, R. J., PETRILLI,E. S., SMITH, D. M.. AHMED, S., LORINCZ,A. T., TEMPLE, G. F., BENNY JENSON, A.. and LANCASTER,W. D., Gynecol. Oncol. 29,267-273 (1988). 20. FELGNER,P. L., GADEK, T. R., HOLM, M., ROMAN, R., CHAN, H. W., WENZ, M., NORTHROP,J. P., RINGOLD, G. M., and DANIELSEN, M., Proc. Natl. Aced. Sci. USA 84, 7413-7417 (1987).
