Physical state and expression of HPV DNA in benign and dysplastic cervical tissue: different levels of viral integration are correlated with lesion grade.

OBJECTIVE: Human papillomavirus (HPV) infection is the most important event in the malignant transformation of human cervical epithelium. Several high-risk (HR-)HPV subtypes have been identified, which lead to CIN and subsequently to invasive carcinoma. The reason for this phenomenon is still unknown, but it seems to be related to the physical state of HPV DNA.
METHODS: Digene HC II test was used to identify HR- and/or low-risk (LR-)HPV infections in cervical swabs of 275 women attending our clinic for routine cytological screening and/or colposcopy because of an abnormal Pap smear comprising low-grade squamous intraepithelial lesions (LGSIL) and high-grade SIL (HGSIL). Specific HR (16, 18, 31, 33, 52b, 58) and LR (6, 11) subtypes were characterized in cervical biopsies of 10 women with benign cellular changes and of 68 women with CIN I-III by the PCR-restriction enzyme method. The physical state of HPV DNA (episomal, mixed and integrated form) was analyzed by bi-dimensional (2D)-gel electrophoresis. In addition, mRNA expression of E6/E7 genes was analyzed by RT-PCR. Furthermore, the relative virus load was determined in nine selected cases. The physical state and transcriptional activity of HPV DNA were then correlated to histopathological results.
RESULTS: LR-HPV infection [27 cases (9.8%)] and HR-HPV infection [121 cases (44%)] of cervical swabs were clearly correlated to the degree of SIL. Further HPV typing in cervical biopsies of 78 women showed that HPV6 and 11 were restricted to benign cellular changes, CIN I and II, whereas HPV16 and 18 were observed predominantly in CIN III/CIS (P=0.01). No clear distribution pattern was observed for HPV31, 33, 52b and 58. Expression of HPV E6 and E7 transcripts was uniformly correlated with the different physical state of HPV DNA. Analyzing the physical state of these HPV subtypes, HPV6 and 11 could only be detected as an episomal form, independent of SIL grade. In normal epithelium and in CIN I and II, HPV16 and 18 were exclusively found in the episomal form. In CIN III/CIS, 15 of 30 cases of HPV16 (50%) and 16 of 17 cases of HPV18 (94%) were exclusively integrated into the host genome. Like HPV16/18, HPV31, 33, 52b and 58 were also present in the episomal form in normal epithelium and in CIN I and II, but were integrated in 80% of the CIN III/CIS (4/5) cases.
CONCLUSION: Absent integration of HPV16 DNA in some CIN III/CIS suggests that integration is not always required for progression early dysplastic lesions. In contrast, integration of HPV type 18 and others appears to be of major importance for the transforming efficacy of cervical dysplasia. The applied method represents a sensitive instrument to assess the physical state of HPV and is useful to predict the progression of disease.