Study of stringency conditions for human papillomavirus DNA detection on cell lines, frozen and paraffin-embedded tissue sections by in situ hybridization with biotinylated probes

Abstract

In situ hybridization was mainly used for typing human papillomavirus (HPV) in paraffin-embedded or frozen sections under stringent conditions (SC). We tested 5 different conditions of stringency with biotinylated HPV 1, 2, 16 and 18 probes on 3 cell lines (Siha and CaSki with HPV16, HeLa with HPV18) by varying the concentration of formamide in the hybridization mixture and washings in order to determine the stringency conditions to be used to assess the presence of HPV and its typing: A-low stringency, hybridization at 35° C below the melting temperature of DNA (Tm-35° C) and washings without formamide; B-low stringency, hybridization and washings at Tm-35° C; C-medium stringency, hybridization at Tm-35° C and washings at Tm-12° C; D-high stringency, hybridization at Tm-12° C and washing without formamide; E-very high stringency, hybridization and washings at −12° C. This study showed that HPV typing required a high stringency. On the contrary, under non stringent conditions (NSC), each cell line was positive with the heterologous probes. When 3 to 5 stringency conditions were assayed on 4 frozen samples, similar results were obtained. Typing required high stringency conditions whereas NSC allowed HPV detection. Furthermore, this study demonstrated the specificity of the reaction in lesions positive with more than one type. Stringent (Tm-12° C) and non stringent (Tm-35° C) conditions of hybridization were further applied to 57 biopsy sections (17 frozen and 40 paraffin-embedded specimens) from typical wart lesions and lesions suspected of HPV. Nineteen samples were totally negative under both NSC and SC, and considered as non-infected by HPV. In 22 specimens positive, under both NSC and stringent conditions (SC), the HPV type was identified. Ten specimens reacted with 1, 2 or 3 HPV types under NSC but the HPV DNA was not typed with the probes used. Six lesions were negative under NSC but were typed under SC. Most paraffin sections were labeled only with one HPV probe under NSC, whereas frozen sections were often labeled with 2 or 3 HPV probes. The HPV probe positive under SC was usually positive under NSC in both frozen and paraffin sections. HPV type 1 probe was more frequently positive under NSC in paraffin- embedded sections than the others and the 4 probes tested were equally positive in frozen sections. These findings show the interest of in situ hybridization in low stringency conditions since 17% of our lesions (10/57) were positive only under NSC: HPV DNA was detected but not typed with the probes used. Frozen sections were more frequently positive than paraffin sections, suggesting a loss of DNA accessibility in the latter, due to the fixation or processing before hybridization.