Int. J. Cancer: 51,587-592 (1992) 0 1992 Wiley-Liss, Inc.

miia

PJbllcat on 01 tne Internat onal Union Aga nsr Cancer Phblication de I Jnion nternationa e Contre le Cancer

HUMAN PAPILLOMAVIRUS (HPV) DNA IN PENILE CARCINOMAS AND IN TWO CELL LINES FROM HIGH-INCIDENCE AREAS FOR GENITAL CANCERS IN AFRICA Maria L. TORNESELLO', Franco M. BUONAGURO', Elke BETH-GIRALDO',S.K. KYALWAZI~ and Gaetano GIRALDO'.~ 'Istituto Nazionale Tumori, "Fondazione Pascale", Cappella Cangiani, 1-80131,Naples, Italy; 2New Mulago Hospital, Makerere University, Kampala, Uganda. Biopsies of 13 penile cancers (PC), from patients living in regions of Uganda with a high incidence of genital cancers, were studied for the presence, molecular characteristicsand physical state of DNA related to that of human papillomavirus (HPV) types 6, I I, 16, 18, 31 and 33. HPV DNA sequences were detected in all PC specimens by dot/Southern blot analyses and by gene amplification of DNA sequences highly conserved among several HPVs. HPV 16 DNA sequences were found in one PC; DNA sequences with low homology to HPV16 or HPVl8 were present in all other samples. Viral DNA is primarily integrated in the cellular DNA. To isolate and characterize a possible highly oncogenic HPV, a genomic library of the DNA extracted from the PC-8 biopsy has been constructed in the EcoRl arms of the EMBL4 phage. A single phage containing 8.30-kb HPVl6-related sequences has been identified and the 3 segments of 0.45, 0.65 and 7.2 kb. released by EcoRl digestion, have been independently subcloned in pUC I8 for further analysis.

o 1992 Wiley-Liss,Inc. Human genital cancer reveals epidemiologic characteristics of infectious events (Rotkin, 1973) and several types of genital papillomavirus (HPV) have been found associated with premalignant and malignant lesions (Schwarz et al., 1983; Durst et al., 1983; Boshart et al., 1984) as well as cell cultures derived from them (Yee et al., 1985). While penile cancer (PC) in general is a rare tumor, accounting for less than 1% of male cancers (Young et al., 19Sl), in certain regions, e.g. Uganda, it represented 12% of all male cancers diagnosed between 1964 and 1968 with an average incidence of 2.2 cases/100,000 men per year (Dodge et al., 1973). Moreover in some ethnic groups such as the Nyoro living in the Bunyoro district of Uganda, this type of malignancy accounted for 41.4% of all male cancers. In the search for factors associated with the high incidence of PC, studies revealed that geographical locations rather than ethnic background could be associated with such malignancies, supporting the concept that environmental agents, e.g. oncogenic viruses, are probably involved in the development of this type of tumor (Schmauz and Jain, 1971). Furthermore, a strong correlation of high incidence rates for cancer of the penis and cervix has been observed in central China (Li et al., 1982) and in several countries (Sorahan and Crombie, 1981) as well as an increased risk of cervical cancer in spouses of patients with penile cancer (Cartwright and Sinson, 1980; Graham et al., 1980) suggesting that the 2 diseases may be etiologically associated and that sexually transmissible agents should be considered. An etiological role for HPVs has been proposed for benign and malignant genital tumors (zur Hausen, 1977; zur Hausen and Schneider, 1987). HPV types 6, 11,41,42,43 and 44 have been associated with benign anogenital lesions and therefore considered to convey low oncogenic risk (Gissman et al., 1983; Beaudenon et al., 1987; Lorincz et al., 1989). In contrast, HPV 16, 18 and, less frequently, HPV 31, 33, 35, 39, 45, 51 and 52, have been associated with malignant changes, intraepithelial neoplasia and invasive carcinomas, and designated high-risk HPV types (Durst et al., 1983; Beaudenon et al., 1986; Lorincz et al., 1987; Shimoda et al., 1988).

The HPV-transforming activity seems to be mainly associated with the E6/E7 regions which are preferentially transcribed in cervical cancers and cell lines derived from them (Schwarz et al., 1985; Smotkin and Wettstein, 1986). Both E6/E7 genes of high-risk HPVs (HPV 16 and 18) are necessary for in vitro transformation of human keratinocytes (Kaur et al., 1989); the E7 gene, on the other hand, is sufficient to transform established murine cells (Voudsen et al., 1988). The greater biologic activity observed for high-risk HPV types has been attributed to differences in E6 and/or E7 gene expression as well as to differences in the intrinsic potency of the E6 and/or E7 protein products (Barbosa et al., 1991). The high incidence rate of genital cancers in some geographical regions and/or in some clinical conditions could be due to the presence of unknown, highly oncogenic HPV strains or variants of known HPVs (16, 18 etc.) with further different biological activities of the E6/E7-transforming regions and/or differences in the regulation of their expression. Environmental co-factors, chemical or biological, such as the human immunodeficiency virus (HIV), could further modulate or act synergistically with the HPV-transforming activity. The purpose of our study was to search for and identify high-risk HPV type(s)/strain(s) associated with penile carcinomas deriving from a high-incidence area for this type of malignancy, namely Uganda. MATERIAL AND METHODS

Tumor biopsies and cell cultures Tumor biopsies from Ugandan patients with penile carcinoma, histologically confirmed to be squamous-cell carcinomas (New Mulago Hospital, Makerere University of Kampala, Uganda) were frozen 1"Cimin and preserved in liquid nitrogen at -192°C until further use for establishment of long-term tumor-cell lines and/or molecular hybridization analyses. Two tumor-cell lines (PCA-5 and PCA-23) have been established (Gentile et al., 1987). Moreover the following human cervical carcinoma cell lines were obtained from ATCC (Rockville, MD): CaSki, SiHa and HeLa. DNA hybridization experiments DNA from frozen biopsy samples, 2 cell lines derived from them and control cell lines (CaSki, SiHa and HeLa) was extracted by proteinase-K treatment followed by several phenol and chloroform-isoamyl alcohol extractions (Gissman et al., 1983). Up to 10 pg of cellular DNAs were spotted in parallel on 2 nitrocellulose filters and tested at high (Tm -40°C) and low (Tm -20°C) stringency with the DNA of several HPV strains. HPV DNA samples positive by dot blot were cleaved with EcoRI restriction endonuclease, separated in 1% agarose gels and blotted onto nitrocellulose. The filters were pre-treated with 0.1 ml/cm2 of a solution containing 6~ 3To whom correspondence and reprint requests should be addressed. Received: January 14,1992.

588

TORNESELLO E T A L .

SSC, l o x Denhardt's, 0.1% SDS at 56°C for at least 2 hr. Hybridization solutions contained 4~ SSC, 10x Denhardt's, 0.1% SDS, 0.01 M Tris and 1 X lo6 cpm/ml of nick-translated HPVs DNA-labeled at specific activity > 1 X los cpm/pg DNA with SO pCi of 32P-dCTP, 3000 Ci/mmol. After incubation for 1 day at 56°C (low stringency) or 68°C (high stringency), the filters were washed with 2 x SSC, 0.1% SDS at 50°C and exposed at -70°C to X-ray films (Amersham Hyperfilm MP, Aylesbury, UK) using intensifying screens.

Tris pH 7.4) and proteinase K 100 pg/ml at 37°C for 24 hr. NaCl was added to a final concentration of 1 M and the cell lysate was incubated at 4°C overnight. After centrifugation at 25,000 g for 40 min at 0"C, the supernatant was collected and incubated with RNase at a final concentration of 10 &ml for 30 min at 37°C. Episomal DNA was then extracted by phenol/chloroform-isoamylalcohol and pelleted with 0.2 M KAc and 2.5 volume of EtOH by centrifugation at 18,000 g for 20 min at 0°C. DNA pellet was dissolved in TE.

Selective extraction method of HPV DNA One gram of tumor biopsy was minced and treated with 1-2 ml of Hirt lysing solution (0.6% SDS, 10 mM EDTA, 10 mM

Molecular cloning of HPV DNA sequences The DNA obtained from PC8 biopsy was partially cleaved with EcoRZ in 4 aliquots for different time intervals (5, 10, 20 and 30 min) and run on agarose gel to determine digestion conditions required to achieve an average length distribution of 20 kb. The DNA fragments were ligated to the EcoRI arms of the lambda E M B U bacteriophage, packaged in vitro and plated on Eschen'chia coli NMS39 strain. Phages (1 x lo6) were screened by the plaque hybridization assay (Benton and Davis, 1977) using a 32P-labeled HPV 16 DNA probe, at Tm -40°C. DNA of a single recombinant phage positive for HPV DNA was cleaved with several restriction endonucleases and tested by Southern blot hybridization analysis, following standard procedures. The 0.45-,0.65- and 7.2-kb EcoRI fragments, that cover the whole HPV sequence, were subcloned in pUC18 plasmid. Polymerase chain reaction (PCR) analysis Oligonucleotide primers were used to amplify a segment within the E6 region of HPV6, 11, 16, 18 (Shibata et al., 1988), a region within the HPV LI open reading frame, highly conserved among HPVs (Manos et al., 1989), and the HIV-1 gag region (Ou et al., 1988). Positive and negative controls were

F~GURE1-Southern blot analyses of PC samples. Results obtained from hybridizations performed at high stringency are presented in the top 3 panels. Several human cervical cell lines, HPV-Dositive. are used as internal control: CaSki (Cab SiHa (Si). HeLar(He) and C4-1 (C4). Calf thymus DNA'(CT) and'the lambda DNAiHindIZZ marker (A) are included. Southern-blot analysis at low stringency of PC samples digested with EcoRZ and the corresponding undigested Hirt extracts (lanes labeled with H), are shown in the bottom panel.

FIGURE 2 - Southern blot analysis of PCA.5 and PCA23 DNA as in Figure 1. Hybridizations are performed at low stringency. HPV 16, 18, 6, 11 and HPV 16a, 18a lanes represent HPV reconstructions of 20 ng and 30 pg respectively.

HPV DNA IN GENITAL CANCER IN AFRICA

used to exclude cross-contamination. D N A samples were amplified for up to 40 cycles (annealing at 37°C for 2 min, extension at 70°C for 3 min and denaturation at 94°C for 1min) in a SO-pl sample volume with 1U of Perkin-Elmer (Norwalk, CT) Cetus Taq polymerase. Amplification products were analysed by liquid hybridization using as probes 40-mer oligos from the central portion of the amplified viral sequences, followed by gel electrophoresis and autoradiography. RESULTS

DNAs extracted from 13 PC biopsies and the 2 carcinoma cell lines PCA-S and PCA-23 were tested for the presence of HPVs by dot-blot analysis using as probes HPV 6,11,16,18,31 and 33 DNA. D N A sequences homologous to HPV 16 were detected at low stringency (Tm -40°C) in PC12, PC15 and PC22 and at high stringency (Tm -20°C) in PC 8. Sequences homologous to HPV 18 were detected at low stringency in PC5, PC8, PC12, PC1S and PC22 samples; at high stringency a weak hybridization signal was present only in PC8 and PC22. The HPV 6 and HPV 11 hybridization pattern was similar to that obtained with HPV 18. Neither HPV-31- nor HPV-33related DNAs were found in these samples. PC DNA samples, HPV-positive by dot blot, were digested with EcoRl and serially tested by Southern blot analysis. A t high Stringency, weak hybridization signals are generally obtained, as shown in Figure 1, using HPV 18, 16 and 31 as probes. In PC8, for example, a 7.2-kb sequence homologous to

589

HPV18 and 16 was detected. All PC D N A samples shown in the lower panel of Figure 1 are positive at low stringency when probed with HPV 16, suggesting the presence of sequences with low homology to such HPV strains. Sequences homologous to HPV 16 are not detected in human DNA tested under the same hybridization conditions (Fig. 2). Re-hybridization of the nitrocellulose with human repetitive sequences of the Alu family confirms that equivalent amounts of D N A were loaded for each sample. The absence of pBR sequences was proved by hybridization with a PUC vector (data not shown). PCA-5 and PCA-23 cell cultures, established from tumor biopsy PCS, were tested for the retention of HPV sequences. Bands larger than 8 kb were detected only at low stringency in both of them with the HPV 16 and HPV 16/18 probes under conditions standardized for the detection of 1.5 copies of HPV reconstructions (Fig. 2). HPV6 and 11 D N A sequences were not detected under the same conditions. The presence of HPV D N A sequences in the PC samples was also tested by PCR. For each sample, 1 pg of D N A was sequentially incubated with (1) 7 2.5-mer primer sets specific to the E6-coding region of each HPV strain (Shibata et al., 1988); (2) a primer set that amplifies a region within the HPV L1 open reading frame highly conserved among the HPVs (Manos et al., 1989); and (3) an oligonucleotide set that amplifies the HIV-1 gag region 1551-1665 (Ou et al., 1988). In the upper panel of Figure 3, the 400-bp segment amplified with HPV16specific primer and probe sets are shown in PC22, 17, 15, 11, 10, 7, 5 and 8 samples. Sequences homologous to HIV-1 have

FIGURE3 - HPV and HIV detection by PCR in PC biopsies. The molecular size marker (M) is given by theAluZ-cleaved pBR322 DNA. In the upper panel 16/6 lane represents the HPV 16 and HPV 6 1-pg reconstruction, 16a and 6a the 0.01-pg reconstruction of the corresponding HPV strains. The 400-bp amplified band is present in PC 17,15,11,10,7and 8. In the lower panel HIVa and HIVb lanes represent HIV-1 reconstructions of 1 and 0.01 pg, respectively.

5 90

TORNESELLO ET A L TABLE I -ANALYSIS FOR HPV AND HIV IN PENILE CARCINOMAS FROM UGANDA

HPV 6 DiSPCR',2

P

C 7 8 1

1 2 PCA 5

23

4

+

16 DiSPCR

11 DiS

-

+

+ + + + + + + 2 + +

5 2 -

+ +

+ +

+ + + -

+ + -

-

+ + + -

+ -

+ +

+

+ -

+

-

-

+ -

+ +

PUC DIS

+

-

+ +

+

+ +

HPV consensus sequences PCR

HIV PCR

+

-

+ + + +

+ - -

+

+

+

-

-

-

-

+

-

-

-

-

+

-

-

-

-

+

-

-

+ -

33 DISPCR

31 DIS

18 DISPCR

-

-

rfr

-

-

-

-

-

-

f rfr

-

-

-

IDIS = Dot-blot and Southern-blot analysis; all hybridizations were obtained at low stringency (Tm-40°C); only the PC 8 was positive for the HPV 16 at high stringency.-2PCR = Polymerase chain reaction using sets of primers specific for each virus, followed by liquid hybridization with 'ZP-labeled probes, electrophoresisin 3% agarose and autoradiography.

Bgl I1

EcoRI

Csti

EcoRI

I I

Piti

PStI

I

I

EcoRI

HindIII

I

EcoRI EcoRIHindIII

FIGURE4 - Partial restriction map of the HPV clone isolated from the PC8 biopsy. Restriction map and location of HPV-related sequences were obtained by multiple restriction endonucleases and Southern-blot analyses.

not been detected by PCR. Table I summarizes the results obtained in PC 7, 8, 12, 15, 22 and PCA-5 and PCA-23, with dot/Southern blot and PCR analysis. Double positives could be due to double infections or simply to the low stringency used in the hybridization step of dot and Southern blot analysis and in the PCR annealing step (45°C). Finally, all PC samples tested by the reverse blot assay against all published HPV types (1-60) are negative for the presence of HPV sequences. These results suggest that an average of

Human papillomavirus (HPV) DNA in penile carcinomas and in two cell lines from high-incidence areas for genital cancers in Africa.

Biopsies of 13 penile cancers (PC), from patients living in regions of Uganda with a high incidence of genital cancers, were studied for the presence,...
1MB Sizes 0 Downloads 0 Views