Biology and biochemistry of papillomaviruses

Detection of papillomaviral-like DNA sequences in premalignant and malignant perineal lesions of sheep

Small hyperkeratotic and ulcerated lesions and clinical cancers were isolated from the perineal region of sheep and examined for evidence of papillomavirus infection by various criteria including gross morphology, histology, immunohistochemistry and DNA hybridisation. No specific diagnostic features of papillomaviral infection by immunohistochemistry were found, although some lesions showed gross morphological and histological features similar to papillomaviral effect in other species. DNA hybridisation analysis, using human papillomaviral type 11, 13, 16 and 18 DNA probes under conditions of reduced stringency (Tm-40 degrees C), detected homologous sequences in two thirds of the biopsies examined. These homologous sequences occurred in benign hyperkeratosis as well as invasive squamous cell carcinomas but were much more frequently isolated from carcinomas. This finding suggests that a papillomavirus is associated with the development of squamous cell carcinomas of the perineum of sheep.

Late promoter of human papillomavirus type 8 and its regulation

Human papillomavirus type 8 (HPV8) belongs to the HPV types associated with skin carcinomas of patients with epidermodysplasia verruciformis (EV). Its noncoding regulatory sequences (NCR) were shown to drive the expression of the reporter gene chloramphenicol acetyltransferase (cat) in transient assays with human epithelial cells (HT3 cells). This constitutive activity could be enhanced by coexpression of the HPV8 transactivator protein E2. The analysis of 5' deletions of the NCR showed that the EV-specific sequence motif M33 and the neighboring AP1 site are essential for the promoter activity, whereas 44 nucleotides located immediately upstream of M33 are strongly inhibitory. The same effects were observed in simian virus 40-immortalized fetal keratinocytes (SV61 cells) and spontaneously immortalized skin keratinocytes (HaCaT cells). By using primer extension and RNase protection analyses two promoters could be identified within the HPV8 NCR. A nested set of weak signals, corresponding to start sites between positions 175 to 179, represented the previously described E6 promoter. The vast majority of transcripts was initiated at position 7535 and shown to undergo processing at an NCR-internal splice donor (positions 1 to 8). The promoter P7535 is similar to late promoters of other skin-associated papillomaviruses as far as localization, transcript structure, and sequence characteristics are concerned. To confirm that P7535-initiated transcripts proceed indeed to the L1 gene for the major capsid protein, viral mRNAs from an HPV8-induced lesion of a patient with EV were characterized by RNase protection and sequence analysis of polymerase chain reaction-amplified cDNAs. The NCR leader (positions 7535 to 4) appeared in two messages with three exons each. The third exon started with the second ATG codon of L1 in both cases; the short central exons from the 3' part of the early coding region were defined by a common splice acceptor site (position 3303) and different splice donor sites (positions 3443 and 3704).

Prognostic importance of human papillomavirus type 16 DNA in cervical cancer

Human papillomavirus (HPV) type 16 DNA is frequent in invasive cervical cancers. Among 43 patients with invasive cervical cancer, HPV-16-positive tumors spread to the parametrial and pelvic lymph nodes significantly more often than did HPV-16-negative tumors (P less than 0.05). Demonstration of HPV-16 DNA in invasive cervical cancers may be an additional prognostic factor for this disease.

Comparison of a one-step and a two-step polymerase chain reaction with degenerate general primers in a population-based study of human papillomavirus infection in young Swedish women

The prevalence of human papillomavirus (HPV) infection in cervical cell scrapes from young women was determined by polymerase chain reaction (PCR) by using general primer pairs localized within the L1 region. With a one-step general PCR, 5.9% (35 of 590) of young women in a population-based study were found to contain HPV DNA. The proportion of HPV-positive women increased with age, from 1.4% (1 of 69) among women aged 19 years to 9.2% (13 of 142) among women aged 25 years. Among the cervical scrapes from women with normal cytology, 5.6% (30 of 539) harbored HPV DNA. A total of 5 of 19 (26.3%) of the women with pathological signs were positive for HPV DNA. By a two-step PCR, using nested general primers, 20.3% (118 of 581) of all women were shown to contain HPV DNA. The proportion of HPV-positive women also increased with age, from 17.4% (12 of 69) among women aged 19 years to 31.9% (43 of 135) among women aged 25 years, when the two-step PCR was used. Some 19.2% (102 of 530) of cervical scrapes from women with normal cytology contained HPV DNA. Among the women with pathological signs, 16 of 19 (84.2%) were positive for HPV DNA. The HPV DNA-positive specimens were demonstrated to contain HPV type 6, 11, 16, 18, 31, 33, 35, 39, 40, 45, 55, or 56. The most prevalent HPV types were 6 (2.0%) and 16 (2.7%). More than one type was found in 16 specimens. Sixty HPV-positive samples could not be typed.

Effective vaccination against papilloma development by immunization with L1 or L2 structural protein of cottontail rabbit papillomavirus

Immunization of rabbits with either L1, the major structural protein, or L2, a minor structural protein of cottontail rabbit papillomavirus (CRPV), protected against challenge with the virus. Neutralizing antibodies were elicited by both the L1 and L2 trpE fusion proteins. Neutralization with anti-L1 serum, however, was more efficient than with anti-L2 serum. In contrast, when tested on Western blots the immune response to L2 was stronger than to L1. Rabbits were also protected against CRPV infection by immunization with L1 expressing recombinant vaccinia virus. Sera from two of three rabbits immunized with recombinant vaccinia virus were negative on Western blots but all three were positive in ELISA's with nondenatured fusion protein or in immunoprecipitations. The results suggest that both the viral structural proteins, L1 and L2, merit consideration in the development of a vaccine against papillomavirus.

Cervical cancer--what role for human papillomavirus?

OBJECTIVE

To review the role of human papillomavirus (HPV) as a causative agent of cancer of the cervix. DATA SOURCES, data synthesis, study selection: Medical journals, recently published text books related to cancer of the cervix and HPV and Papillomavirus Reports were examined to review the pathology of cervical cancer and its precursor lesions, its epidemiology in Australia and overseas, methods of detection of HPV (in particular molecular biology techniques used to diagnose HPV) and evidence linking HPV with genital neoplasia.

CONCLUSION

While there is compelling evidence strongly linking certain HPV types with genital cancer, a causative role is yet to be proven and the aetiology is most likely multifactorial. Detection and typing of high risk genotypes of HPV in the genital tract as a diagnostic exercise to identify those women most at risk of developing genital neoplasia is not currently recommended.

Tumorigenic transformation of murine keratinocytes by the E5 genes of bovine papillomavirus type 1 and human papillomavirus type 16

To examine the biological properties of the bovine papillomavirus type 1 (BPV) and human papillomavirus type 16 (HPV16) E5 genes, each was cloned separately into a retroviral expression vector and helper-free recombinant viruses were generated in packaging cell lines. The BPV E5 retroviruses efficiently caused morphologic and tumorigenic transformation of cultured lines of murine fibroblasts, whereas the HPV16 E5 viruses were inactive in these assays. In contrast, infection of the p117 established line of murine epidermal keratinocytes with either the BPV or the HPV16 E5 retrovirus resulted in the generation of tumorigenic cells. Pam212 murine keratinocytes were also transformed to tumorigenicity by the HPV16 E5 gene but not by the gene carrying a frameshift mutation. These results establish that the HPV16 E5 gene is a transforming gene in cells related to its normal host epithelial cells.

Biological properties of the deer papillomavirus E5 gene in mouse C127 cells: growth transformation, induction of DNA synthesis, and activation of the platelet-derived growth factor receptor

We determined the biological activities of the 44-amino-acid deer papillomavirus (DPV) E5 protein in mouse C127 cells. The DPV E5 gene can induce focus formation, stable and acute morphologic transformation, and DNA synthesis, and it activates the platelet-derived growth factor (PDGF) beta receptor as assessed by increased constitutive tyrosine phosphorylation of mature and precursor receptor forms. Thus, the DPV E5 protein has biological activities similar to those of the closely related E5 protein from bovine papillomavirus type 1. Moreover, like the bovine papillomavirus type 1 E5 protein, the DPV E5 protein shares a short region of sequence similarity with the B chain of PDGF. These results show that activation of the PDGF receptor is a general property of fibropapillomavirus E5 proteins, lending support to our previous proposal (L. Petti, L. Nilson, and D. DiMaio, EMBO J. 10:845-855, 1991) that activation of the PDGF receptor may play a central role in transformation of fibroblasts by E5 proteins.

Identification of T- and B-cell epitopes of the E7 protein of human papillomavirus type 16

There is strong evidence implicating human papillomavirus type 16 (HPV16) in the genesis of human genital cancer. Viral DNA has been identified in invasive carcinoma of the uterine cervix and in cell lines derived from cervical carcinomas. These sequences are actively transcribed, and translation products corresponding to the early (E)-region genes have been identified. The most abundant viral protein is the E7 protein, which has been shown to possess transforming activity for both established and primary cells. In addition, it has been shown to bind to a cellular tumor suppressor, the retinoblastoma gene product (pRb-105). In view of these properties, we have undertaken the immunological analysis of this protein and have identified four T-cell epitopes and three B-cell epitopes by using a series of overlapping peptides spanning the entire HPV16 E7 sequence. Two of the B-cell epitopes were recognized by antisera from mice with three different murine (H-2) haplotypes (k, d, and s) immunized with two different E7 fusion proteins and from Fischer rats seeded with baby rat kidney cells transformed by HPV16 E7 and ras. A third B-cell epitope was recognized by antisera from CBA mice seeded with HPV16 E7-expressing L cells. Two regions of the protein contain common B- and T-cell epitopes, one of which appears to be particularly immunodominant.

Use of the polymerase chain reaction to study the relationship between human papillomavirus infections and cervical cancer

Although it is now evident that human papillomaviruses (HPV) are strongly associated with cervical cancer, their etiological role in the oncogenesis of this disease is still unknown. However, HPV screening may identify women at risk of acquiring this disease. With the recent development of the polymerase chain reaction (PCR), it has become possible to detect small numbers of human papillomavirus genomes in clinical samples. The sensitivity and specificity of this technique, together with the possibility of performing the test on crude cervical scrapes, makes PCR the method of choice for screening. In this paper, data on the detection of human papillomavirus by PCR are presented and the applicability of this technique for the screening of human papillomavirus genotypes is evaluated. The question arises whether screening for diagnostic purposes must include all the human papillomavirus types associated with infections of the genital tract or only those which are strongly associated with cervical cancer (HPV 16 and HPV 18). It is proposed that an international council must be created that is responsible for standardised epidemiological screening strategies and follow-up programmes.

The open reading frame L2 of cottontail rabbit papillomavirus contains antibody-inducing neutralizing epitopes

Polyclonal antisera were generated against bacterially derived fusion proteins of the open reading frames (ORFs) of the capsid proteins of cottontail rabbit papillomavirus (CRPV). The carboxy-terminal two-thirds of CRPV L1 and the carboxy-terminal half of CRPV L2 were cloned into a bacterial expression vector and induced proteins were used as antigen and immunogen. The polyclonal antisera were tested in a series of immunological assays, including ELISA, Western blot, and neutralization of CRPV. ELISA demonstrated that the polyclonal antisera raised against expressed L1 proteins reacted strongly to disrupted CRPV virion antigen and weakly both to intact CRPV virion and disrupted BPV-1 virion. Anti-CRPV L2 antisera reacted strongly only to intact and disrupted CRPV virion antigen. Viral capsid proteins of CRPV were detected in Western blots of HPV-11, BPV-1, and CRPV virus particles by these polyclonal antisera. The anti-L1 sera recognized the major capsid protein (60 kDa) and the anti-L2 sera identified a 76-kDa viral protein of CRPV. Only the antisera generated against expressed L2 neutralized CRPV. The neutralizing titer of the anti-L2 sera, however, was several orders of magnitude lower than the titer of a neutralizing polyclonal antiserum that was generated by immunizations with intact CRPV virions.

Characterization of BPV-like DNA in equine sarcoids

The DNA from equine sarcoid samples from New York State and Switzerland was isolated and probed with bovine papillomavirus type 1 (BPV-1) to determine if BPV genomes were present. Twelve of 13 sarcoids from New York State and 17/20 sarcoids from Switzerland contained DNA that hybridized to the BPV-1 probe. Restriction enzyme analysis of the positive samples demonstrated restriction fragment profiles characteristic of BPV-1 in 22 sarcoids and restriction fragment profiles characteristic of bovine papillomavirus type 2 (BPV-2) in 7 sarcoids. In addition, three tissues histologically diagnosed as pyogranulomatous dermatitis, fibropapilloma, and fibrosarcoma contained BPV-like DNA. Tissues with BPV-1-like and BPV-2-like DNA contained an average of 285.7 (21 to 808) and 125.8 (2 to 762) BPV-like genomes per cell, respectively. Minor differences in the restriction fragment profiles of the BPV-like DNA and evidence for partial BPV-like genomes were found in some sarcoids. BPV-like DNA was not detected in lymphocyte DNA from sarcoid-affected horses. These results confirm previous observations and support the hypothesis that bovine papillomavirus, or a very similar virus, is linked to the cause of equine sarcoid.

Risk of cervical intraepithelial neoplasia in women with glomerulonephritis

OBJECTIVE

To investigate the occurrence of cervical intraepithelial neoplasia in women with glomerulonephritis and its possible association with immunosuppressive treatment.

DESIGN

Retrospective study of cytological or histological specimens from women presenting with glomerulonephritis and a group of case and age matched controls.

SETTING

University department of pathology, Norway.

PATIENTS

81 women presenting with glomerulonephritis from 1981 to 1988, from whom gynaecological cytological or histological specimens were available. A group of 162 case and age matched controls.

MAIN OUTCOME MEASURES

Age when glomerulonephritis of cervical intraepithelial neoplasia was diagnosed, type and characteristics of kidney lesion, stage of cervical intraepithelial neoplasia and presence of human papillomavirus, use of immunosuppressive treatment.

RESULTS

Cervical intraepithelial neoplasia was more common in women with glomerulonephritis than in their controls (16/81 (20%) v 7/162 (4%), p less than 0.001) and was more advanced in those with glomerulonephritis than in the controls (9/81 (11%) of the study group had grade III cervical intraepithelial neoplasia compared with 1/162 (1%) of the controls). The increased occurrence of cervical lesions was independent of the use of immunosuppressive treatment, but the individual lesions tended to be more advanced when it was used (four of the seven cervical lesions in women with glomerulonephritis who had received immunosuppressive treatment were carcinoma in situ). Of the nine cervical lesions tested, seven were virus associated.

CONCLUSION

Women with glomerulonephritis should have regular cervical smears, irrespective of their use of immunosuppressive treatment.

Detection of papillomavirus DNA in precancerous lesions of the ears of sheep

Sixty-seven benign precancerous cutaneous lesions from the ears of 51 sheep were examined for papillomavirus DNA by hybridisation to radioactively labelled or biotinylated probes of bovine papillomavirus type 1 (BPV 1) DNA under varying conditions of stringency. An additional 16 precancerous lesions from other cutaneous sites on 15 sheep and 15 samples of lesion-free skin from nine sheep were similarly examined. Both total genomic and subgenomic probes were used. DNA from 10 aural lesions and one vulval lesion reacted with the probe in a manner indicative of the presence of episomal papillomavirus DNA. Papillomavirus DNA was detected at low stringency in eight of the 10 aural lesions and the vulval lesions, and at high stringency in two aural lesions. Three out of the 8 aural lesions that were positive at low stringency reacted when re-tested at high stringency. Hybridisation with one of the samples of lesion-free ovine skin produced occasional equivocal signals. One particular positive lesion, an ovine aural cutaneous horn, was studied in more detail. When treated with restriction endonucleases, its restriction enzyme pattern was the same as that for BPV 2 DNA with eight of twelve enzymes and the same as that for BPV 1 DNA with two of the twelve enzymes. It was concluded that this ovine papillomavirus was more closely related to BPV 2 than to BPV 1. The possibility that it could be a subtype of BPV 2 is discussed.

Mapping of linear epitopes of human papillomavirus type 16: the E1, E2, E4, E5, E6 and E7 open reading frames

Certain types of human papillomavirus (HPV), especially HPV type 16, are associated with proliferative lesions of the cervix uteri that can progress to malignancy. In order to map the linear epitopes of the HPV-encoded proteins, we have synthesized the predicted amino acid sequences of the open reading frames (ORFs) in the early region of HPV 16, as a set of 94 synthetic 20-residue peptides overlapping each other with 5 amino acids. The peptides were tested for reactivity with IgA, IgG and IgM antibodies in the sera of 30 patients with HPV 16-carrying cervical neoplasia. The EI ORF had only low immunoreactivity, but several relatively minor epitopes were identified in the carboxyterminal part. The E2 ORF was found to contain several epitopes that were highly immunoreactive with a majority (up to 87%) of the cervical cancer patients' sera. The E4 ORF had one major, regularly IgA- and IgG-reactive epitope, whereas the E5 and E6 ORFs had only a few minor epitopes. The E7 ORF had several epitopes that were highly immunoreactive, but only with a minority of patients' sera. The 10 most immunoreactive peptides were also analyzed for immunoreactivity with 60 control sera, of which 22 were derived from patients with parotid gland tumors and 38 were derived from healthy volunteers. Most of the peptides were also immunoreactive with the control sera. However, the IgA antibodies, and to some extent the IgG antibodies, were found at much lower levels among the controls.

Incidence of antibodies to human papillomavirus type 1 in patients with cutaneous and mucosal papillomas

A study of antibodies to human papillomavirus (HPV) 1 is reported in a series of sera from 126 patients with papilloma lesions of different clinical types (verruca plantaris, vulgaris, plana, condyloma acuminata, laryngeal papilloma) using a sensitive solid-phase enzyme-linked immunosorbent assay (ELISA). Our findings showed a significant prevalence of specific HPV 1 IgG antibodies in patients with plantar warts (65%). This incidence was considerably lower in patients with other papilloma lesions, in which positive sera may reflect a past infection with HPV 1. The ELISA test was evaluated as nearly 100 times more sensitive than the indirect immunofluorescence test, and IgG titers to HPV 1 in patients' sera were usually low. Serum samples taken at different times during the course of HPV infection in a few patients with refractory lesions did not show any significant change in their antibody titer.

Lymphotoxin production by regional lymph node lymphocytes in patients with uterine cervical cancer

The cytotoxin production by regional lymph node cells was examined in 25 patients with uterine cervical cancer and 10 patients with uterine myoma. The patients in stage I had significantly increased spontaneous release of cytotoxins compared with that in stages II, III, and IV. The spontaneous release in stages III and IV was markedly reduced. There was no difference in the release of cytotoxins from peripheral blood lymphocytes between cancer patients and patients with myoma or healthy controls. The cytotoxin production by lymph node cells was increased in stage III by stimulating with formalin-fixed QG-K cells derived from uterine cervical cancer, but not in stages I and II. Almost all of the cytotoxic activity of cytotoxin was abrogated by antilymphotoxin antibody. However, the cytotoxin activity was partially inhibited by anti-tumor necrosis factor antibody. These results suggest that cytotoxins released from the regional lymph node cells of uterine cancer patients are derived from, most of all, lymphotoxin.

Mapping of linear epitopes of human papillomavirus type 16: the L1 and L2 open reading frames

Certain types of human papillomavirus (HPV), notably HPV type 16, are associated with flat or inverted proliferative lesions of the cervix uteri that can progress to malignancy. As a first step towards the serological study of the epidemiology of HPV, we have synthesized the entire amino acid sequences of the 2 major viral capsid proteins of HPV type 16, L1 and L2, as a set of 66 synthetic 20-residue peptides with an overlap of 5 amino acids. The peptides were tested for reactivity with IgA, IgG and IgM antibodies in the sera of 30 patients with HPV-16-carrying cervical neoplasms. Both IgG and IgM antibody responses were detected, but most of the reactivity found was of the IgA class. The most immunoreactive peptides were further analyzed for reactivity with sera from 22 patients with parotid gland tumors and with sera from 38 healthy individuals. The L2-encoded protein contained only one major linear epitope, which was not specific for HPV-16-carrying neoplasms. In contrast, the L1-encoded protein contained several epitopes that were regularly immunoreactive with antibodies present in the sera of patients with HPV-16-carrying cervical neoplasms, but only rarely so in the sera of patients with other tumors or of healthy individuals.

Immunological cross-reactivity to laboratory-produced HPV-11 virions of polysera raised against bacterially derived fusion proteins and synthetic peptides of HPV-6b and HPV-16 capsid proteins

Polysera raised in rabbits to bacterially derived fusion proteins and synthetic peptides of the L1 and L2 ORFs of HPV-6b and -16 were tested for cross-reactivity to laboratory-produced infectious HPV-11 virions. The polysera were analyzed in a series of five different immunological assays including immunoperoxidase staining of the koilocytotic nuclei in sections of formalin-fixed, paraffin-embedded as well as fresh frozen sections of HPV-11 experimental condylomas generated in the athymic nude mouse xenograft system, ELISA, Western blots, and neutralization of infectious HPV-11 virions. ELISA and Western blot assays were used to determine whether the polysera identified external or internal epitopes on HPV-11 virions, and whether there was cross-reactivity to BPV-1 or laboratory-produced CRPV virions. Seven of a total of 12 sera were positive for reactivity to HPV-11 in one or more assays, but none of the reactivity was directed to external epitopes on the intact virions as determined by ELISA. None of the L1 products generated group-specific antigen (GSA) antisera including a synthetic peptide spanning the GSA site. The combination of assays clearly demonstrated that apparent false positive and false negative reactivities of different antisera were obtained for each assay system tested. Thus, no single assay could be used reliably to determine the true antiviral reactivity of a given polysera.

Serological responses to papillomavirus group-specific antigens in women with neoplasia of the cervix uteri

Certain types of human papillomaviruses have been linked to the development of carcinoma of the cervix uteri. We have analyzed 114 serum specimens from women with cervical intraepithelial neoplasia (CIN) or carcinoma of the cervix uteri for the presence of serum antibodies against purified, disrupted bovine papillomavirus (BPV). The titers of immunoglobulin A (IgA) antibodies against BPV were slightly elevated (P less than 0.025) in the sera from CIN or cervical carcinoma patients compared with the titers of 139 serum specimens from sex- and age-matched healthy controls. In contrast, both the IgG and IgM serum antibody titers against BPV were significantly decreased for CIN and cervical carcinoma patients compared with those of healthy controls (P less than 0.001 and P less than 0.005, respectively). These results suggest that the difference between IgA and IgG or IgM antibodies to papillomavirus group-specific antigens may represent interesting serological parameters that could possibly be used in the epidemiologic study of women at risk for CIN.

Modulation of HPV18 and BPV1 transcription in human keratinocytes by simian virus 40 large T antigen and adenovirus type 5 E1A antigen

Transcription of early open reading frames initiated from the long control region (LCR) of HPV18 and BPV1 is known to be modulated by homologous and heterologous papillomarvirus E2 gene products. Using CAT constructs transfected into normal human keratinocytes, we show that SV40 large T antigen activates transcription from the LCR of both viruses, whereas Ad5-E1a antigen represses transcription from the HPV18-LCR but activates transcription from BPV1-LCR. Experiments using constructs containing subfragments of the HPV18-LCR cloned in enhancer configuration ahead of the SV40 early promoter or the HSV1-Tk promoter suggest that the effect of Ad5-E1a antigen on HPV18 transcription is probably due to a repression of the enhancer function of the LCR. The mechanism of transcription stimulation by SV40 large T antigen is less clear. The 230 bp Rsa1-Rsa1 central domain of the HPV18-LCR seems involved both in transcriptional stimulation by SV40 large T antigen and transcriptional inhibition by adenovirus E1a antigen.

Humoral assays of human sera to disrupted and nondisrupted epitopes of human papillomavirus type 1

The use of different assay systems and the disparity in results obtained has meant that we have little understanding about the role played by the humoral response during human papillomavirus (HPV) infection. Human antibody responses have so far appeared to be largely directed against the major capsid protein, L1. This protein possesses both type-specific and type-common antigenic determinants but it is not known which of these is important in vivo during the natural course of infection. In this study humoral responses of 83 individuals to purified HPV 1 virions were tested in three types of antibody assay. Western blot analysis detected antibodies in only eight of the serum samples, whereas an enzyme-linked immunosorbent assay (ELISA) and immunoprecipitation assay using nondisrupted HPV 1 virions showed positive antibody reactivities for 71 and 64 individuals, respectively. We suggest from these results that the humoral response to L1 is mainly directed against native conformational epitopes present on the whole HPV 1 particle and that type-common epitopes are not largely involved. This was further demonstrated by the fact that when samples were tested in the same ELISA system using disrupted HPV 1 virions as the antigen instead of whole virus particles, the number of positive sera was reduced to 9 out of 83. We further conclude that humoral assays using antigenic material pertaining to disrupted HPV epitopes are of limited use, at least in the case of HPV type 1. There were no obvious correlations between the antibody assay results and clinical histories of wart infection except that a lower number of positive serum reactivities were found among the group of individuals claiming to have no past history of HPV infection.

Cytogenetic analysis of eight human papillomavirus immortalized human keratinocyte cell lines

Cytogenetic analysis was performed on human papillomavirus (HPV)-immortalized cell lines. Lines were established by co-transfection of primary human keratinocyte cells with HPV type 16 or 18 DNA and pSV2neo. The resulting clonal lines contained integrated HPV DNA and exhibited extended life spans in culture but were non-tumorigenic in nude mice. Two HPV16-immortalized lines (FEPE1L8 and FEPE1L9) and 4 HPV18-immortalized lines (FEA, FEH18L, FEP18-5 and FEP18-11) were established. Two additional lines were derived by subsequent treatment of the FEA line with TPA and by further transfection with HSVII DNA. Cytogenetic analysis revealed that all lines were abnormal, containing a variety of numerical and structural aberrations. Six of the 8 lines were hyper-triploid and 2 were near-diploid. Examination of lines FEA, FEH18L and FEP18-11 at multiple passages in culture revealed that the lines were clonal and chromosomally stable over extended passage in culture. Structural rearrangements were most common in chromosomes 1 and 3 but also occurred in chromosomes 5, 7, 8, 12, 16 and 22. Marker chromosomes were present in all cell lines. A small metacentric marker, possibly an isochromosome for the short arm of chromosome 5, was consistently present in the FEA line and its derivatives (FEAB10 and FEAT) as well as the FEH18L line. A loss or reduction in copy number of chromosome 13 was seen in 5 of the 8 cell lines.

Human papillomavirus associated with adenocarcinoma and adenosquamous carcinoma of the cervix: analysis by in situ hybridization

The incidence of adenocarcinoma of the cervix appears to be increasing. Recent reports have demonstrated an association between adenocarcinoma of the cervix and human papillomavirus (HPV) by Southern blot hybridizations. In situ deoxyribonucleic acid (DNA) hybridization was performed on paraffin-embedded specimens to localize the source of HPV DNA. In pure adenocarcinoma five of six specimens were positive for HPV DNA. Four specimens contained HPV type 18 and one HPV type 16. Only one of three adenosquamous lesions was positive and it contained both HPV types 16 and 31. These findings suggest an association between HPV and adenocarcinoma of the cervix.

A rapid DNA probe test for detecting human papilloma virus types 6/11 and 16 in biopsy specimens

A simple, rapid and sensitive non-radioactive in situ hybridization assay for human papillomavirus (HPV) has been developed and used to detect HPV 6/11 and HPV16 DNA in ano-genital biopsy specimens. A comparative study to determine the sensitivity and the specificity of this assay relative to Southern blot hybridization assays was performed using one hundred biopsy specimens. The sensitivity of the in situ hybridization assay was 88% for HPV type 6/11 and 89% for HPV type 16. The specificity of the test was 99% for both virus types. In addition to its high sensitivity and specificity, this in situ hybridization assay for HPV uses a non-radioactive detection system. The assay is faster and easier to perform than the Southern blot method. Also, in situ hybridization assays permit the simultaneous evaluation of the histology, as well as the DNA content of biopsy tissues because they do not result in the destruction of tissue or cell morphology. This prototype HPV DNA assay was developed using DNA probes for HPV 6/11 and HPV 16 DNA in anogenital biopsy specimens. However, the protocol developed in these studies can easily be extended to include the use of probes for detecting other HPV types in a variety of tissues.

Sequence divergence yet conserved physical characteristics among the E4 proteins of cutaneous human papillomaviruses

Human papillomavirus (HPV) types 1, 2, and 4 together comprise the major cause of cutaneous papillomas in the general population. We have aligned the genomes of these three viruses by partial sequence analysis, and have sequenced the E4 open reading frames (ORFs) of HPV 2 and HPV 4. After expression as beta-gal fusion proteins in bacteria, antibodies raised to the putative E4 gene-products of both virus types were used to identify the native E4 proteins in naturally occurring tumors. At the primary amino acid sequence level, the E4 protein of HPV 2 was found to be most homologous with those of HPV 6 and 11 and was not closely related to those of HPV 1 or 4. Although the E4 ORF represents a region of weak homology amongst papillomaviruses, the E4 encoded proteins showed significant conservation in their physical characteristics. Like those of HPV 1, the E4 proteins of both HPV 2 and HPV 4 were found to be composed of a major low-molecular-weight doublet (16.5/18K for HPV 2, 20/21K for HPV 4, c.f. 16/17K for HPV 1) along with minor high-molecular-weight species, which probably represent dimers of the smaller proteins, (33K for HPV 2, 40K for HPV 4, c.f. 32/34K for HPV 1). The E4 products of all three virus types were multiply charged, and exhibited a characteristic migration pattern following alkaline urea gel electrophoresis. Although the levels of E4 expression in tumors induced by the different virus types was very different, this was found to correlate closely with the level of virus production characteristic of each virus type. In all three cases, E4 proteins were found to be primarily cytoplasmic, and to be associated with the distinctive cytoplasmic inclusion granules characteristic of each virus type. The poor sequence conservation between the E4 protein of HPVs 1, 2, and 4, taken alongside the ability of these viruses to infect similar histological sites, suggests that E4 may not be involved in determining tissue specificity. Our results suggest conserved physical characteristics (acidic, multiply charged, ability to form dimers) and similar site of expression may be the important factors for E4 function.

Cutaneous fibropapillomas of green turtles (Chelonia mydas)

Six juvenile green turtles (Chelonia mydas) from the Indian River Lagoon System, Florida, U.S.A., with multiple cutaneous fibropapillomas, were kept in isolation and examined over a 6-month period. Histologically, the fibropapillomas consisted of a slightly to moderately hyperplastic epidermis overlying a thickened hypercellular dermis. In the earliest lesions, ballooning degeneration was present predominantly in the stratum basale where rete ridges extended into the dermis; aggregates of mixed inflammatory cells were present around dermal vessels. As the lesions matured, they developed an arborizing, papillary pattern. More mature lesions had a less verrucous, often ulcerated surface, with the dermis composed primarily of large collagenous fascicles and relatively few fibroblasts. While numerous trematode eggs were present within dermal capillaries of a histologically similar biopsy specimen from an Hawaiian green turtle, no trematode eggs were observed in any of 28 biopsies examined from the six Florida green turtles in this study. Low stringency Southern blot hybridization and a reverse Southern blot failed to demonstrate papillomavirus DNA in any of the samples extracted. Ultrastructural evaluation of the earliest lesions demonstrated membrane-bound intracytoplasmic vacuoles within epidermal cells in the stratum basale. Similar vacuoles were also observed in the epidermal intercellular spaces and within the dermis. Occasionally, particles with electron-dense centres and measuring 155 to 190 nm were observed in these vacuoles.

Sensitive detection of nucleic acids and protein of human papillomavirus type 6 in respiratory and genital tract papillomata

We have developed a sensitive method to detect and localize HPV-6 viral DNA, mRNA and protein in biopsy specimens of genital and respiratory tract lesions by using in situ hybridization and immunoperoxidase assays on sections of plastic-embedded tissue. This modified in situ hybridization technique, using ultrathin sections and strand-specific 3H-labelled riboprobes, offers the advantages of superior morphological preservation and detection of viral genomes at low copy number with good resolution. This modified immunocytochemistry provides better sensitivity when compared to previous methods using paraffin-embedded materials. In respiratory tract lesions, immunoperoxidase assay detected only a few capsid antigen positive cells, while in the genital tract lesions, there were more capsid antigen positive cells. Southern transfer analyses and in situ hybridizations demonstrated the presence of more viral nucleic acids in genital tract papillomata than respiratory tract papillomata. Epithelial cells throughout the papillomata were infected by HPV-6 as evidenced by positive hybridization, with more viral DNA present in superficial cells. Our results suggest that genital tract epithelium is more permissive for HPV-6 replication than respiratory tract epithelium. Using stand-specific probes synthesized from subgenomic fragments of the HPV-6 genome in conjunction with nuclease digestions, we were able to demonstrate that HPV-6 transcripts specific to open reading frames (ORFs) E6, E7, E1, L1, and L2 occur in maturing superficial cells. In contrast, transcripts specific to ORFs E1, E2, E4, E5a, and E5b could be detected throughout the whole of the epithelium with more signals noted at the basal cell areas. In addition, the distribution of HPV-6 nucleic acids and protein in a carcinoma in situ of the larynx was analyzed. In comparison to benign respiratory tract papillomata, more viral DNA was found in the malignant lesion, but the pattern and distribution of transcription and capsid antigen was similar.

Identification of DNA-protein interactions and enhancer activity at the 5' end of the upstream regulatory region in human papillomavirus type 11

We have examined the 5' end of the noncoding region of the genome of a human papillomavirus, HPV-11, for regulatory elements using permissive host cells. This region of unknown function in the upstream regulatory region (URR) is known to have unusual DNA structure and frequently contains rearrangements which are associated with some more virulent isolates. This 5' 269-bp fragment was found to exhibit both specific DNA-protein binding using laryngeal papilloma protein extracts and enhancer activity in normal and papillomatous primary laryngeal cells. The viral DNA flanks the L1 open reading frame and does not contain the viral E2 binding site. Three distinct protein binding sites are contained in a 50-bp region of the fragment. This fragment, as a whole, functions as an enhancer in primary laryngeal and papilloma cells when ligated to the SV40 promoter and SV40 T-antigen gene. We conclude that this part of the noncoding region of the papillomaviruses has elements characteristic of regulatory elements in cells permissive for infection by these viruses.

Phosphorylation of the human papillomavirus type 1 E4 proteins in vivo and in vitro

Following incubation of HPV 1-induced warts in the presence of [32P] phosphate several of the E4-encoded proteins were found to be radiolabeled. Two-dimensional isoelectric focusing sodium dodecyl sulfate-polyacrylamide gel electrophoresis indicated that the 17K E4 polypeptides had incorporated [32P]phosphate whereas those of 16K were unlabeled. Purified E4 gene products were separated by ion exchange chromatography into a large number of different species, which were of similar size but of different charge due to varying extents of phosphorylated peptides have been isolated and identified. Phosphoserine and phosphothreonine were identified in all 16/17K E4 fractions but not phosphotyrosine. Both HPV 1 E4 16K and 17K fractions were phosphorylated in vitro by cAMP-dependent protein kinase but not by myosin light chain kinase or by phosphorylase kinase. Incubation with cAMP PK gave incorporation of approx. 0.5 mole phosphate/mol of protein indicating that the cAMP-dependent protein kinase site(s) was partially phosphorylated in vivo. This view was supported by the fact that species which were more heavily phosphorylated in vivo incorporated less phosphate after cAMP-dependent protein kinase phosphorylation. HPV 1 E4 was also phosphorylated at serine and threonine residues by a crude cytoplasmic extract prepared from cultured human keratinocytes and cultured human retinoblasts. These results are discussed in the light of the known effects of phosphorylation on the interactions of other keratinocyte-specific proteins.

Comparison of human papillomavirus type 1 serotyping by monoclonal antibodies with genotyping by in situ hybridization of plantar warts

Thirty plantar warts were analyzed for the presence of HPV-1 type-specific and PV genus-specific capsid antigens by immunofluorescence (IF) using monoclonal and polyclonal antibodies and type-specific HPV-1 DNA employing in situ hybridization methods. Fifteen of 30 plantar warts were positive by IF for PV genus-specific structural viral antigens. Thirteen of the 15 productively infected plantar warts expressed intranuclear HPV-1 type-specific capsid antigens and viral DNA, which were detected in the same distribution in each individual wart. The 2 productively infected plantar warts that did not react with HPV-1 type-specific MoAbs did not react with HPV-1 type-specific DNA by in situ hybridization. Thus, serotyping of HPV-1 capsid antigens by monoclonal antibodies is concordant with genotyping of HPV-1 viral DNA by in situ hybridization in productively infected plantar warts.

Tissue site-specific enhancer function of the upstream regulatory region of human papillomavirus type 11 in cultured keratinocytes

The human papillomavirus type 11 regulatory region was cloned upstream of a reporter complex and microinjected into nuclei of individual primary human keratinocytes. Genital and laryngeal keratinocytes, normal host cells for this virus, exhibited higher levels of expression than cutaneous keratinocytes. We conclude that a papillomavirus enhancer(s) shows preferences among epithelial cells from different tissue sites.

Detection of IgA antibodies against human papillomavirus in cervical secretions from patients with cervical intraepithelial neoplasia

Although human papillomavirus (HPV) is involved in the etiology of cervical carcinoma, there are no cervical carcinoma-specific HPV serologic tests available. We investigated the presence of broadly cross-reactive IgA antibodies to papillomavirus (PV) in cervico-vaginal secretions from patients with condylomata and cervical intraepithelial neoplasia (CIN). Purified bovine PV (BPV) virions were used as antigen in an enzyme-linked immunosorbent assay (ELISA). Forty-two women whose ages ranged from 20 to 50 participated in the study. Eight of 9 patients with CIN had IgA antibodies against PV in their cervical secretions. Three of 9 patients with koilocytosis and condylomas but no CIN had IgA antibodies to PV. Six of 24 women with normal Pap-smear and colposcopy also had IgA antibodies against PV in their cervical secretions. The proportion of IgA-positive cervical secretions was significantly higher in the CIN group than in the normal group (p less than 0.005). Our data suggest that IgA antibodies to PV may be a useful marker for CIN.

Nucleic acid hybridization as a diagnostic tool for the detection of human papillomaviruses

At this time nucleic acid hybridization tests are the most sensitive and reproduceable methods for the detection and differentiation of HPV types in clinical samples. The hybridization method of choice depends on the information desired and the availability of the proper diagnostic nucleic acid probes. Assuming most of the HPV nucleic acid probes become readily available in the near future, then the most sensitive test for screening clinical specimens--albeit the most laborious--will be the Southern blot procedure. As more information covering the involvement of HPV infections with the progression of lesions from benign to malignant is compiled, the need to know the particular subtype or status of HPV integration may become more or less important in the screening of clinical samples. If this information becomes less important, or unnecessary for a simple screening procedure, then dot blot hybridization may prove to be a much easier method for obtaining the information desired. If the sensitivity of in-situ hybridization using non-radioactive probes increases, then this method would become the fastest, easiest, and cleanest technique for the screening of a large number of clinical samples where limited information is desired. The ideal test for the future would be automated. In order to automate a test for HPV infections, the test design must become much simpler, and in order to design a simpler test, more information will be needed concerning the biology of HPV infections, how they cause benign or malignant cellular proliferation, and how the host immune system responds to HPV infections.

Immortalization of human foreskin keratinocytes by various human papillomavirus DNAs corresponds to their association with cervical carcinoma

Normal human foreskin keratinocytes cotransfected with the neomycin resistance gene and recombinant human papillomavirus (HPV) DNAs (types 16, 18, 31, and 33) that have a high or moderate association with cervical malignancy acquired immortality and contained integrated and transcriptionally active viral genomes. Only transcripts from the intact E6 and E7 genes were detected in at least one cell line, suggesting that one or both of these genes are responsible for immortalization. Recombinant HPV DNAs with low or no oncogenic potential for cervical cancer (HPV1a, -5, -6b, and -11) induced small G418-resistant colonies that senesced as did the nontransfected cells. These colonies contained only episomal virus DNA; therefore, integration of HPV sequences is important for immortalization of keratinocytes. This study suggests that the virus-encoded immortalization function contributes to the pathogenesis of cervical carcinoma.

Human papillomaviruses and carcinomas

The recognition of multiple types of human papillomaviruses has resulted in remarkable progress in the detection of persisting viral nucleic acid sequences in carcinomas. The consistent transcription in tumors of two early open reading frames, E6 and E7, with few exceptions (Lehn et al., 1985), indicates a role for the products of these genes in the induction and/or maintenance of the transformed state. A number of studies have shown that in vitro transformation can be achieved by transfection of E6/E7 DNA, and proteins encoded by these DNA sequences can be demonstrated in primary human keratinocytes immortalized by this DNA (Kaur et al., 1989). Mutagenesis experiments are needed to determine the absolute requirement for and function of these genes in transformation. A preferential association of some types with benign lesions while others may be frequently found in malignant tumors has been observed. HPV types 5 and 8 in epidermodysplasia verruciformis patients and types 16, 18, 31, 33, etc. in genital lesions are most frequently associated with progression to malignancy, whereas other types, such as HPV-6,-10, -11, and -20, are regularly identified in benign warts. Such distinctions are not absolute but provide the initial steps toward establishing a causal role for some human papillomaviruses in carcinomas. The need for well-designed epidemiological studies in concert with optimum molecular and serologic evaluations is evident (Armstrong et al., 1988). The data from human and animal studies indicate that papillomaviruses contribute significantly to the development of many, if not all, carcinomas, but we do not yet have a clear understanding of the importance of other interacting viral, chemical, or cellular factors. The application of gene cloning and non-stringent hybridization (Law et al., 1979) has provided us with an apparently ever-increasing catalog of human papillomaviruses. More effort is now required to establish their prevalence, the natural history of infection, and the mechanism of neoplastic transformation.

Comparison of filter in situ deoxyribonucleic acid hybridization with cytologic, colposcopic, and histopathologic examination for detection of human papillomavirus infection in women with cervical intraepithelial neoplasia

Material from uterine cervical scrapings of 98 women with cervical intraepithelial neoplasia were analyzed by filter in situ hybridization for human papillomavirus infection. Concurrently obtained Papanicolaou smears and colposcopically directed biopsy specimens were also examined for papillomavirus infection. Hybridization was superior to cytologic and colposcopic examinations and was equivalent to histopathologic study for papillomavirus detection. Infection with virus types 6 and/or 11 was associated with milder disease, whereas virus types 16 and/or 18 infection alone or in association with types 6 and/or 11 was associated with more severe disease. Because papillomavirus infection may not be detected by cytologic or colposcopic examination and specific virus types have been documented to be associated with invasive disease, deoxyribonucleic acid hybridization analysis for at least these virus types should be carried out in conjunction with conventional procedures when evaluating women with cervical disease. Filter in situ hybridization is a simple, rapid, noninvasive procedure and has enhanced diagnostic value over conventional procedures by defining infecting virus types.

Nucleic acid analysis by sandwich hybridization

One of the most significant achievements of the biochemist during the past two decades is the use to which immunologically based assays have been put in clinical diagnosis (Hood et al.: Immunology, 1984). The problem faced and surmounted by immunologists in effecting the transition from research tool to routine clinical assay bears a remarkable similarity to that confronting the molecular biologist today; i.e., how can nucleic acid hybridization, a technique of obvious potential (Meinkoth and Wahl: Anal Biochem 138:267-284, 1984; Syvanen: Med Biol 64:313-324, 1986; Matthews and Kricka: Anal Biochem 169:1-25, 1988), be modified in order to fulfill all necessary parameters of a routine diagnostic assay? There are several such requirements, and the importance placed on each depends on the objectives of the assay: the technique must be sensitive, specific, and reproducible. Other advantages would be cost-effectiveness, ease of manipulation, and amenability to automation. Ideally, the signal detection should be based on a non-radioactive system, because of the instability of probes labelled with isotopes like 32p, and the potential hazards involved in their handling and disposal. The sandwich hybridization for the analysis of nucleic acid sequences was first used in 1977 (Dunn and Hassell: Cell 12:23-36, 1977), but its potential as a diagnostic assay was not realized until 1983, when it was applied to the detection of adenovirus DNA in nasopharyngeal aspirates from children with acute respiratory infection (Ranki et al: Gene 21:77-85, 1983). It has since been modified and used not only for the detection of microbial infection (Virtanen et al.: Lancet i:381-383, 1983; Ranki et al.: Cur Top Microbiol Immunol 104:307-318, 1983; Lehtomaki et al.: J Clin Microbiol 24:108-111, 1986; Virtanen et al.: J Clin Microbiol 20:1083-1088, 1984; Palva and Ranki: Clin Lab Med 5:475-490, 1985; Polsky-Cynkin et al.: Clin Chem 31:1438-1443, 1985; Parkkinen et al.: J Med Virol 20:279-288, 1986; Palva: FEMS Microbiol Lett 28:85-91, 1985; Palva et al: FEMS Microbiol Lett 23:83-89, 1984; Zolg et al.: Mol Biochem Parasitol 22:145-151, 1987; Palva: J Clin Microbiol 18:92-100, 1983), but also for the analysis of nucleotide sequence variations (Langdale and Malcolm: Gene 36:201-210, 1985). We will discuss the development of the sandwich technique and the advantages it conveys over the more conventional nucleic acid hybridization formats, together with new developments which will ensure that it earns a place alongside immunoassay in the diagnostic laboratory.

Detection of EBV and HPV DNA sequences in oral "hairy" leukoplakia by in situ hybridization

Oral "hairy" leukoplakia (OHL) is a white lesion of the oral mucosa, usually located on the lateral tongue among human immunodeficiency virus (HIV) positive acquired immunodeficiency syndrome (AIDS)-risk patients. The lesion has been reported to be associated with Epstein-Barr virus (EBV) and human papillomavirus (HPV). Twenty surgical biopsy specimens were evaluated for the presence of HPV genus-specific antigen, HPV 2/4, 6/11, and 16/18 DNA and EBV DNA by in situ hybridization employing formalin-fixed paraffin-embedded sections. Three cases exhibited immunoreactivity for HPV genus-specific antigen, with localization in cytopathically altered upper spinous layer keratinocytes. HPV 16, 18, or related DNA sequences were identifiable in a single case. Alternatively, employing an EBV long internal repeat subgenomic probe, 19 cases were found to harbor EBV DNA. In all positive cases, the hybrids were localized to upper spinous layer keratinocytes exhibiting nuclear/cytoplasmic vesiculation. It is concluded that OHL is consistently associated with EBV; furthermore, viral replication, as evidenced by DNA localization, corresponds to ultrastructural evidence of capsid and envelope assembly in the more differentiated layers of oral epithelium.

Involvement of human papillomavirus type 8 genes E6 and E7 in transformation and replication

We investigated the transforming activity of human papillomavirus type 8 (HPV8) by expressing all early open reading frames from a heterologous promoter in different rodent fibroblast lines. Morphological transformation was observed only in G418-selected mouse C127 and Rat 1 cells containing an intact E6-coding region. E6 of HPV8 did not transform NIH 3T3 cells as did E6 of bovine papillomavirus type 1. C127 cells transformed by E6 were anchorage independent and had a reduced serum requirement but did not form tumors in nude mice. E7 of HPV8 showed no transforming potential, in contrast to E7 of HPV18 and HPV16. It was, however, able to complement an E7 mutant of bovine papillomavirus type 1 with a defect in high-copy-number DNA maintenance. The data indicate that the expression of the HPV8 E6 open reading frame is sufficient to induce morphological transformation in rodent fibroblasts, whereas E7 is involved in the replication of the viral DNA.