Analysis of human papillomavirus types in exophytic condylomata acuminata by hybrid capture and Southern blot techniques.

Exophytic condylomata acuminata of the external genitalia of 40 patients were analyzed for human papillomavirus (HPV) DNA by the Southern blot and hybrid capture methods. All lesions were initially analyzed by the Southern blot method by using a mixture of HPV type 6, 11, 16, and 18 whole genomic probes. Southern blots demonstrated characteristic PstI restriction patterns of HPV type 6, 11, or 16 in all but one lesion. HPV 6 subtypes accounted for 28 of 39 HPV-positive lesions. Twenty-seven of these 28 lesions contained HPV type 6a, and 1 lesion contained HPV type 6c. Eight lesions contained HPV type 11 and three contained HPV type 16. Two of the three condylomata acuminata containing HPV type 16 were obtained from solid-organ transplant recipients receiving immunosuppressive medications. The third lesion containing HPV type 16 was a typical exophytic condyloma acuminatum from a woman with previously resected vulvar carcinoma. The hybrid capture assay detected HPV DNAs in all lesions except the Southern blot-negative lesion. Twenty-five lesions were positive for the A probe only (HPV types 6 and 11 and related types). All of these lesions were found to contain HPV type 6 or 11 sequences in the Southern blot assay. The remaining 14 lesions were positive for both the A probe and the B probe (HPV types 16 and 18 and related types). The strongest signal in these 14 lesions by the hybrid capture assay was consistent with the result of the Southern blot assay in all but one case. We conclude that (i) HPV type 6a is the most common type found in these lesions, (ii) HPV type 16 may be present more often in exophytic condylomata acuminata from immunosuppressed individuals, (iii) hybrid capture is a useful tool for documenting the presence of HPV sequences in DNAs from exophytic condylomata acuminata, and (iv) in samples containing multiple HPV types, hybrid capture allows detection of minority HPV types.     

Detection of human papillomavirus DNA in urinary bladder carcinoma by in situ hybridisation.

AIMS: To investigate the sensitivity of an in situ hybridisation system to detect human papillomavirus (HPV) infection in transitional cell bladder cancer and to evaluate the advantages of analysing multiple biopsies; to examine the correlation between HPV tumour infection detected by in situ hybridisation and the presence of serum anti-HPV antibodies detected by enzyme linked immunosorbent assay (ELISA); and to relate the presence of viral infection to grade, stage, and follow up in cases of bladder cancer. METHODS: The in situ hybridisation technique was used with broad spectrum and type specific (6/11, 16/18, 31/33/35) probes against HPV DNA in formalin fixed, paraffin embedded tissues from 43 cases of bladder cancer. The results were analysed for the presence and type of papillomavirus and correlated with clinicopathological variables. RESULTS: The presence of HPV DNA was identified by the in situ hybridisation technique in 17 of 43 cases of bladder cancer; 12 of these were serum antibody positive and 10 had had multiple biopsies. Fifteen of the cases that were negative for HPV DNA by in situ hybridisation had positive serum serology when tested by ELISA. In 14 cases, the HPV was either types 16/18 or types 31/33/35, both of which carry high oncogenic risk. The stage (p < 0.05) and grade (NS) of the tumour and the outcome on follow up (p < 0.05) were correlated with the presence of HPV infection. CONCLUSIONS: ELISA is not useful in identifying patients with HPV positive bladder cancer, but the use of several probes and multiple biopsies increases the detection rate of HPV in neoplastic tissues. The association between tumour virus infection and high grade/high stage tumours and worse outcome suggests that HPV infection of neoplastic tissue has a negative effect on the behaviour and evolution of transitional cell bladder carcinoma.     

Influence of fixation on human papillomavirus DNA detection in frozen and embedded paraffin lesions by in situ hybridization with biotinylated probes.

Series of frozen or paraffin-embedded tissues from various body sites, taken from non-immunosuppressed or immunosuppressed patients with persistent papilloma lesions were examined for the presence of group specific antigen from human papillomavirus (HPV) by indirect immunofluorescence or HPV DNA by in situ hybridization with biotinylated probes. We have shown that it is possible to detect HPV DNA after fixation of tissues in neutral formalin, Bouin's or Baker's solution. However, the sensitivity was reduced as compared to frozen tissues. The HPV DNA was detected in nuclei of heavily infected epithelial cells such as plantar or hand warts or in dispersed cells containing high copy numbers of HPV DNA from lesions such as squamous cell carcinomas or keratoacanthomas. In premalignant or malignant lesions of both immunosuppressed or non-immunosuppressed patients, HPV DNA was rarely detected after fixation. HPV types commonly described for skin and genital samples were identified in non-immunosuppressed patients, whereas in transplant recipients oncogenic HPV type 16 was identified in benign warts as well as in premalignant or malignant lesions. Positive reactions with several HPV types were more frequent in lesions from grafted patients than from the normal population. Virus antigen was detectable more frequently in frozen sections than in fixed tissues. Our findings indicate that in situ hybridization is an appropriate and rapid technique to study the presence of HPV infection. However, numerous controls are needed to avoid misinterpretations.     

Absence of human papilloma virus in cervical adenocarcinoma determined by in situ hybridisation.

A few studies using DNA technology have suggested that human papillomavirus (HPV) may be an aetiological factor for adenocarcinoma of the uterine cervix. Twenty one cases of cervical adenocarcinoma were studied by in situ hybridisation using biotinylated DNA probes for HPV types 6, 11, 16 and 18 and a streptavidin, biotinylated alkaline phosphatase detection system. Intranuclear HPV DNA was detected in none of the adenocarcinomas, while positive controls gave a clear intranuclear signal. Adjacent areas of normal, koilocytic, and dysplastic squamous epithelium also gave positive results. It may be that squamous epithelium contaminates adenocarcinomas reported as positive by Southern blotting. Our results showing absence of detectable HPV DNA within adenocarcinomas suggest that HPV infection may not have a major role in the aetiology of adenocarcinoma of the uterine cervix.     

Detection of human papillomavirus in large cell neuroendocrine carcinoma of the uterine cervix: a study of 12 cases

AIM: To investigate the role of human papillomavirus (HPV) in large cell neuroendocrine carcinoma (LCNEC) of the uterine cervix. METHODS: Twelve archival, immunohistochemically and/or electron microscopically confirmed cases of cervical LCNEC were studied. Non-isotopic in situ hybridisation (NISH) was performed on the formalin fixed, paraffin wax embedded biopsies using digoxigenin labelled probes to HPV types 6, 11, 16, 18, 31, and 33. The tumours were then subjected to polymerase chain reaction (PCR) analysis using GP5+/GP6+ consensus primers to the HPV L1 gene, in addition to type specific primers to the E6 and E6/E7 genes. RESULTS: HPV-16 was detected by NISH and/or PCR in seven of the 12 carcinomas. Two additional tumours were HPV-18 positive by NISH and/or PCR. HPV DNA was not detected in the three remaining cases. CONCLUSION: Integration of high risk HPV, in particular type 16 and to a lesser extent type 18, is associated with this uncommon variant of cervical carcinoma.     

Prevalence of HPV cervical infection in a family planning clinic determined by polymerase chain reaction and dot blot hybridisation

The overall prevalence of human papillomavirus (HPV) cervical infection in 131 women attending a family planning clinic was 7% (HPV 6/11, 16, 18, 31) by dot blot hybridisation, 53% (HPV 11, 16, 31) by polymerase chain reaction (PCR), and 56% by the two methods combined. HPV 16 and 18 were the commonest types (4% each) by dot blot, HPV 16 (39%) by PCR. Fifteen percent of subjects had mildly abnormal cervical cytology (grades 1A, 2A, or 3). There was no significant correlation between cytological abnormality and HPV positivity, or between cytological or HPV status and other postulated risk factors for cervical neoplasia. It is concluded that PCR is considerably more sensitive than dot blot DNA hybridisation in detecting HPV cervical infection in such a "low risk" setting, where HPV copy number may be low. Firm conclusions cannot be drawn from our results regarding a causal role for HPV or other factors in the development of cervical neoplasia.     

Evaluation of immunoassays for the detection and typing of PCR amplified human papillomavirus DNA.

AIMS: To evaluate different hybridisation techniques to detect and type human papillomavirus (HPV) DNAs amplified by consensus primer polymerase chain reaction (PCR) in biopsy and cytological specimens. METHODS: A hybrid capture-immunoassay in microtitre wells was performed to detect HPV sequences amplified by PCR and typed by specific oligoprobes. Consensus primers were used to amplify a sequence within the L1 open reading frame, and direct digoxigenin labelling of amplified products was performed during the amplification reaction. The amplified product was separately hybridised with six biotinylated type specific probes (HPV6, 11, 16, 18, 31, and 33); hybrids were then captured into streptavidin coated microtitre wells and detected by a spectrophotometer as an ELISA using antidigoxigenin Fab fragment labelled with peroxidase and a colorimetric substrate. The results were compared with the dot-blot immunoassay used to detect and type PCR amplified HPV DNA sequences. Consensus primers were used to generate the same unlabelled PCR product; digoxigenin labelled type specific probes for HPV6, 11, 16, 18, 31, and 33 were used and hybrids visualised by colorimetric immunoenzymatic reaction. Thirty nine biopsy specimens and 31 cytological samples were tested by the PCR-ELISA and by standard PCR followed by dot-blot hybridisation. RESULTS: The PCR-ELISA proved to be more sensitive than standard PCR with dot-blot hybridisation typing. All samples positive for HPV-DNA in standard PCR with dot-blot hybridisation method were confirmed positive by the PCR-ELISA assay; however, seven samples were positive only by PCR-ELISA. CONCLUSIONS: The PCR-ELISA assay, which can be performed in one day, is easily standardised and therefore seems to be a practical, sensitive, and reliable diagnostic tool for the detection and typing of HPV genomes in biopsy and in cytological specimens in the routine diagnostic laboratory.     

Detection of human papillomavirus DNA in cell scrapes and formalin-fixed, paraffin-embedded tissue of the uterine cervix by filter in situ hybridisation

Filter in situ hybridisation (FISH) was used to detect the presence of DNA of human papillomavirus (HPV) types 6/11 or 16/18 in cell scrapes (CYTOFISH) and formalin-fixed, paraffin-embedded biopsies (HISTOFISH) taken from the uterine cervices of 19 women. Paraffin tissue sections collected for HISTOFISH were either digested with pepsin or lysed with alkali/Triton X-100. The digest or lysate of the tissue sections and cell scrapes were applied to nylon or nitrocellulose membranes for nucleic acid hybridisation using 32P-labeled HPV-DNA probes. CYTOFISH and HISTOFISH were compared directly by taking samples for each method from the cervices of the same women. Of 19 women examined by colposcopy, cytology, and histology, eight were assessed as normal and 11 had evidence of a cervical disorder and/or the presence of HPV infection. Whereas no HPV-DNA was detected in the normal cases, the presence of HPV-DNAs was detected by both CYTOFISH and HISTOFISH in 11 cases with histological evidence of HPV infection and/or dysplasia. In these HPV positive cases, eight contained HPV 16/18, two HPV 6/11, and one a mixed infection of HPV 6/11/16/18. The high correlation between the results of CYTOFISH and HISTOFISH shows that formalin-fixed, paraffin-embedded cervical biopsies are suitable specimens for the detection and typing of HPV-DNA by FISH. Both CYTOFISH and HISTOFISH should facilitate studies on the prevalence and distribution of HPVs and their association with neoplasia.     

Comparative in situ hybridisation study of juvenile laryngeal papillomatosis in Papua New Guinea and Australia.

A comparative study of cases of juvenile laryngeal papillomatosis from Papua New Guinea (n = 3) and Brisbane, Australia (n = 9) was carried out. In situ hybridisation reactions for human papillomavirus (HPV) types 6 and 11 occurred in 11 cases. All three cases from Papua New Guinea and eight from Australia gave positive signals. A negative reaction was observed in one Australian case. The intensity of the reaction was strong in seven cases, moderate in one, and weak in three. An equivocal reaction was also noted with probes for types 16 and 18, and types 31, 33, and 35 in two cases from Australia and one from Papua New Guinea. It is concluded that as similar staining patterns and intensities occurred in cases from both areas, the aetiology is the same. The equivocal reactions noted in three cases were probably due to cross hybridisation rather than multiple infection.     

Prevalence of human papillomavirus in sinonasal papillomas: a study using polymerase chain reaction and in situ hybridization.

Inverted and fungiform papillomas of the sinonasal cavity share a common origin from the Schneiderian membrane, but they differ widely in their rates of recurrence and progression to carcinoma. To determine the role of human papillomavirus in the etiology of these lesions, 15 inverted papillomas, five fungiform papillomas, and two squamous cell carcinomas associated with inverted papilloma were examined for the presence of HPV by in situ hybridization (ISH) and polymerase chain reaction (PCR). ISH was carried out on formalin-fixed, paraffin-embedded material using HPV types 6/11, 16/18, and 31/33/35 DNA probes. Tissue DNA was amplified by PCR with HPV L1 consensus primers, and the product was detected by gel electrophoresis, Southern blotting, and hybridization with type specific probes (HPV types 6/11, 16, 18). Three of 15 inverted papillomas and two of five fungiform papillomas were positive for HPV 6/11 by ISH, whereas PCR detected HPV 6/11 sequences in two of 15 inverted and three of five fungiform papillomas. Biopsies from two patients who had serial resections contained HPV 6/11 in the original lesions and all recurrences. No HPV was detected in the carcinomas by ISH, whereas PCR detected HPV 16 in one carcinoma. These findings confirm the presence of HPV DNA sequences in both inverted and fungiform sinonasal papillomas as well as in an associated squamous carcinoma. This would suggest a role for HPV in the pathogenesis of Schneiderian membrane lesions. Furthermore, our data indicate that ISH and PCR are equally sensitive in detecting HPV in sinonasal papillomas.     

Demonstration of multiple HPV types in normal cervix and in cervical squamous cell carcinoma using the polymerase chain reaction on paraffin wax embedded material.

The prevalence of human papilloma virus (HPV) types 6, 11, 16 and 18 was investigated using the polymerase chain reaction on formalin fixed, paraffin wax embedded material in 19 cases of cervical squamous cell carcinoma and in 10 normal cervices. HPV DNA was detected in 16 of 19 carcinomas, with multiple types present in 11 of these. HPV 16 or 18, or both, were present in all cases in which HPV was shown. Six of 10 cases of normal cervix contained HPV; five of these contained two or more HPV types, including HPV 16 or 18, or both. This study shows the feasibility of using the PCR on paraffin wax embedded material and indicates a high rate of carriage of multiple HPV types in both normal and neoplastic cervix.     

Detection of HPV DNA in archival specimens of cervical cancer using in situ hybridisation and the polymerase chain reaction.

An archival survey of 98 cervical cancer specimens dating from the 1920s to the 1980s was undertaken to determine whether changes had occurred in the prevalence of human papilloma-virus (HPV) DNA. HPV DNA was detected in paraffin sections of cancers fixed in 10% formalin by in situ hybridisation (ISH) using HPV 6, 11, 16, and 18 32P-labelled DNA probes under conditions of high stringency; and by the polymerase chain reaction (PCR) using 20-mer oligonucleotide primers to amplify 109 bases of the E6 region of HPV 16. In 30 instances results obtained from Southern blot hybridisations which had been carried out on specimens of fresh tissue from the same cancers collected during the 1980s were available for comparison. The rates of HPV DNA detection in cervical cancers ranged from 83% (by Southern or PCR) and 70% (by ISH) on specimens from the 1980s, to 50% and 63% (by ISH and PCR, respectively) on specimens from the 1920s. HPV 16 was by far the most common type, being identified by Southern or ISH in approximately 92% of HPV DNA-positive specimens. No significant change in the prevalence of HPV DNA, or of HPV types, in cervical cancers was found over the 65 year period examined.     

Etiology of breast carcinoma: no apparent role for papillomavirus types 6/11/16/18.

A recent study has shown that human papillomavirus (HPV) types 16 and 18 can immortalize normal breast epithelium, and raised the possibility that HPV may be etiologically related to some cases of breast cancer. In order to investigate this possibility, we performed polymerase chain reaction (PCR) assays for HPV types 6, 11, 16 and 18 in 15 papillomas, 15 papillary carcinomas, and 13 infiltrating ductal carcinomas of the breast. No HPV-related DNA sequences were identified by Southern blotting of the PCR products. It therefore seems unlikely that a significant percentage of human breast carcinomas is etiologically related to infection with one of these HPV types.     

Diagnostic sensitivity of polymerase chain reaction and Southern blot hybridization for the detection of human papillomavirus DNA in biopsy specimens from cervical lesions.

Human papillomaviruses (HPVs) are associated with benign and malignant neoplasms of the cervix. One of the criteria for their etiologic role requires an assessment of whether virtually all or only a small fraction of lesions contain viral genomes. DNA preparations from colposcopically directed punch biopsies of cervical lesions were analyzed by Southern blot hybridization and the polymerase chain reaction (PCR) for the presence of HPV DNA. The biopsy specimens represented different pathologic entities (koilocytosis, condyloma, cervical intraepithelial neoplasia, and invasive carcinoma). In Southern blot hybridization with radioactive probes for HPV 11, 16, 18, 31, and 33, HPV DNA was detected in 74% of the biopsy specimens (42 of 57 cases), with the predominant types being HPV 16 and HPV 18. In contrast, after PCR amplification with primers yielding fragments of characteristic size for HPV 11, 16, and 18, the analysis of the same 57 biopsy specimens revealed that all samples were positive for at least one HPV type. To exclude false-positive PCR results, controls without HPV DNA were interspersed at regular intervals, and results were evaluated only if these controls remained HPV negative. To exclude false-negative results due to failure of the reaction, a target sequence within the c-Ha-ras-1 gene was used as an internal control. All HPV typing results obtained by Southern blot hybridization were in agreement with HPV typing by PCR. The higher number of positive samples in the latter analysis stems from the increased sensitivity of PCR, which was which was effective in identifying as few as 10-100 HPV DNA molecules; in contrast, the sensitivity of Southern blot hybridization was 1 pg, or approximately 10(5) molecules of HPV DNA. The authors conclude that, with sufficiently sensitive diagnostic methods, HPV DNA can be detected in most, if not all, neoplastic cervical lesions.     

A non-radioactive PCR enzyme-immunoassay enables a rapid identification of HPV 16 and 18 in cervical scrapes after GP5+/6+ PCR

In previous studies, general primer mediated PCR (GP5+/6+ PCR) was applied successfully to detect a broad spectrum of human papillomaviruses (HPV) in cervical scrapes. In order to facilitate PCR based HPV detection and typing, a colourimetric microtitre plate based hybridisation assay was developed. The method utilised one biotinylated primer (bio-GP6+) in the GP-PCR. Biotinylated PCR products were captured on streptavidin coated microtitre plates, denaturated and hybridised to digoxigenin (DIG) labelled HPV specific internal oligo probes. The DIG labelled hybrids were detected using an enzyme immunoassay (EIA). Since HPV 16 and 18 are the most common HPV types found in cervical carcinomas, this approach was initiated for these two types. Cross-hybridisation reactions were not detected when the specificity of this PCR-EIA for HPV 16 and 18 was tested on a panel of 20 different HPV genotypes. The sensitivity of the assay was found to be between 10 and 100 HPV 16 and 18 viral genomes in a background of 100 ng cellular DNA. This was similar to the detection limit of Southern blot analysis of PCR products with radioactively labelled oligonucleotides. A group of cytomorphologically normal (n = 89) and abnormal (n = 96) cervical scrapes were composed of HPV 16 and HPV 18 positive and HPV negative scrapes. All HPV 16 and 18 positive smears were detected by PCR-EIA. These results indicate that PCR-EIA has the potential for a rapid and sensitive HPV DNA test for day-to-day routine examination of cervical scrapes. © 1996 Wiley-Liss, Inc.     

Detection of human papillomavirus types 6 and 11 E4 gene products in condylomata acuminatum.

Polyclonal antiserum to an Escherichia coli-produced beta-galactosidase/E4 fusion protein of human papillomavirus type 6b (antiserum 256), and affinity purified HPV 11 anti-E4 antibodies were tested for reactivity in Western blots with bacterially expressed trpE/E4 fusion proteins of HPV types 6b, 11, 16, and 18. To further characterize the affinity purified anti-E4 antibodies, a dot-immunobinding assay was performed using overlapping synthetic HPV 11 E1E4 peptides as antigens. Protein extracts of condylomata acuminatum from 18 patients containing HPV type 6 or 11 DNA sequences were tested in Western blots using antiserum 256 or affinity purified HPV 11 anti-E4 antibodies. In the Western blots of the trpE proteins, antiserum 256 identified the HPV types 6b and 11 fusion proteins; the affinity purified HPV 11 anti-E4 antibodies identified only the HPV 11 fusion protein. In the dot-immunobinding assay, three HPV 11 peptides were recognized, each containing a shared 8 amino acid sequence that differs significantly from the corresponding sequences of HPV types 6b, 16, or 18. In the Western blots of protein extracts from 18 condylomata acuminatum samples shown to contain HPV types 6 or 11 DNA, putative E4 gene products were identified in six samples by antiserum 256. The affinity purified HPV 11 anti-E4 antibodies identified putative E4 gene products in one of these same six lesions, which was shown to contain HPV 11 sequences by the Southern blot method. All six samples containing E4 gene products were from women. Three of these women were pregnant, one had serum antibodies to the human immunodeficiency virus, and one was a renal transplant recipient receiving glucocorticoids.(ABSTRACT TRUNCATED AT 400 WORDS)     

Detection of human papillomavirus types 6, 11, 16 and 18 in mucosal and cutaneous lesions by the multiplex polymerase chain reaction.

A multiplex polymerase chain reaction (PCR) based on the simultaneous amplification of human papillomavirus (HPV) types 6/11, 16 and 18 in a single-step procedure was developed, using primers chosen in the E6-E7 region. The specificity and sensitivity of this technique have been proved by amplifying mixtures or various amounts of plasmid-containing HPV DNA; it allowed the detection of as few as 5-25 HPV DNA copies. Application of the multiplex PCR to 71 clinical samples showed that HPV DNA was detected in 80% (45/57 cases) of mucosal biopsies and 35% (5/14 cases) of cutaneous specimens. HPV 16 was predominant in high-grade CIN whereas HPV 6 and 11 were detected more frequently in genital condylomas and laryngeal papillomas. In cutaneous Bowen's disease HPV 16, 18 or 6/11 + 16 were detected and in squamous cell carcinomas HPV 6/11 or 16 were found. After sequence amplification with primers of one HPV type, the clinical samples displayed the same HPV types but the frequency of positive and coinfected lesions increased. Thus, multiplex PCR is a valuable technique for typing HPV DNA but coinfections may be underestimated.     

Use of Probemix and OmniProbe biotinylated cDNA probes for detecting HPV infection in biopsy specimens from the genital tract.

AIMS--To compare two commercially available pan probes for the identification of human papillomavirus (HPV) DNA expression in histological sections and to type the HPV positive cases. METHODS--97 formalin fixed, paraffin wax embedded biopsy specimens from the genital tract were tested for HPV positivity with in situ hybridisation using biotinylated cDNA pan probes--Probemix (Enzo) and OmniProbe (Digene). The HPV positive cases were further tested with HPV types 6/11, 16/18, and 31/33/35/51, and the HPV type was related to the histological diagnosis. Formalin fixed, HeLa cells (10-50 HPV 18 copies per cell) and SiHa cells (1-2 HPV 16 copies per cell) were used as reference cell lines. RESULTS--32% of the specimens gave positive nucleic signals with both Probemix and OmniProbe. Of these, 84% could be further characterised with regard to HPV types 6/11, 16/18, and 31/33/35/51; 4% of all cases were positive with either Probemix or OmniProbe. The concordance of these probes was high, 96% altogether. HeLa cells stained positive but SiHa cells did not. CONCLUSION--There is no difference between Probemix and OmniProbe for the general detection of HPV. The mean detection limit of these probes is about 20 copies a cell.     

Detection and typing of human papillomavirus infection of the uterine cervix by dot blot hybridisation: comparison of scrapes and biopsies

Paired scrapes and biopsies from 100 women attending a routine colposcopy clinic were examined by dot blot DNA hybridisation for infection with human papillomavirus (HPV) types 6/11, 16, 18, and 31; 51% of the scrapes and 50% of the biopsies were positive for HPV infection. Scrapes detected more HPV 18 (10% vs. 2%, P = less than 0.05) and HPV 31 (7% vs. 3%, not significant) than did the biopsies, but biopsies detected more HPV 16 (42% vs. 33%, not significant). Comparison of the results for each patient revealed that the correlation between scrapes and biopsies was not very close: only 34 patients were HPV-positive by both sampling methods, whereas 67 were positive if the results were combined. This report discusses the implications of these findings.     

Human papillomaviruses and cervical cancer: analysis of histopathologic features associated with different viral types

The histopathologic features of 41 cervical carcinomas were correlated with the presence of human papillomavirus (HPV). Southern blots of DNA extracted from the tumors were hybridized with 32P-labeled type specific probes for HPV 6, 11, 16, 18, and 31. HPV was found in 26/41 (63%) of the tumors. The HPV types were: HPV 16 in 17 tumors (41%), HPV 18 in six tumors (15%) and HPV 31 in two tumors (5%). No tumor hybridized to either HPV 6 or HPV 11. HPV was identified in all histologic subtypes of cervical carcinoma; however, different HPV types were associated with specific histologic features. HPV 18 was identified in four of eight adenocarcinomas, while HPV 16 was found in only one. HPV 16 was most strongly associated with the keratinizing tumors. It was found in 10/13 (77%) of the large cell keratinizing (LCK) and in only 4/16 (25%) of the large cell nonkeratinizing cervical carcinomas (LCNK). A mucoepidermoid with extensive keratinization and pearl formation also contained HPV 16. One of three additional adenosquamous carcinomas had HPV 31, as did one LCNK tumor. In one LCK tumor, a HPV was identified that hybridized to both HPV 16 and 18. The LCNK group contained the highest percentage of tumors in which no papillomavirus DNA was identified (9/16 lacked HPV DNA). No papillomavirus was detected in six tumors from other sites or in five cervical specimens with no histologic evidence of HPV infection. These data indicate that HPV is involved in all major histologic types of cervical carcinoma, and suggest that the different HPV types transform slightly different cell populations, or that transformation by HPV 18 tends to induce adeno-differentiation while HPV 16 leads to squamous maturation.