Figure 1.
Schematic outline of the recruitment of lesions included into the study.
34 HPV 16 infected cervical lesions were recruited from a gynecopathology archive of cervical punch and cone biopsies. Sections of these lesions were stained with H&E, p16INK4a and L1. 11 of these displayed koilocytes and L1 positive superficial squamous epithelial cells. Of these lesions, three showed a continuum of p16INK4a-positive epithelial areas indicating transforming HPV infections. 2 additional biopsies displaying p16INK4a-positive epithelial areas were recruited from other biopsies not directly showing this continuum of intra-lesional progression for the in depth methylation analysis as shown in Figure 5.
Figure 2.
Microdissection of cervical tissues representing latent, permissive, and transforming modes of HPV infection.
Laser microdissection of paraffin-embedded section containing areas with: HPV 16 infected epithelium without any morphological changes (normal), permissive infections (i.e. cervical intraepithelial low grade lesions lacking diffuse p16INK4a immuno-reactivity but displaying koilocytes), and adjacent transforming infections (CIN 3 lesions with diffuse p16INK4a immune reactivity). This tissue sections were micro-dissected and DNA was isolated from cell fractions representing the basal, intermediate, and superficial cell layers, as well as from transformed p16INK4a-positive cells.
Figure 3.
Differentiation dependent HPV16 URR methylation in the latent infections.
The methylation status of the HPV16 URR was determined by bisulfite treatment followed by direct DNA sequencing of amplified DNA fragments using 3 sets of primers covering the complete HPV16 URR as indicated by the blue arrows in the upper part of the Figure. DNA derived from laser-microdissected basal, intermediated and superficial layers of the cervical normal HPV 16 infected epithelium was amplified and cloned. 12 individual clones were sequenced to identify the presence and patterns of the methylated CpG dinucleotides. The Figure visualizes schematically the methylation of 16 CpG dinucleotides (positions from 7198 to 161) in the URR of HPV-16 genomes. The analyzed region encompasses 3 functionally distinct segments that have been referred to as the 5′ LCR, the enhancer, and the promoter. Open circles represent an unmethylated CpGs, filled circles represent methylated CpGs.
Figure 4.
Differentiation dependent HPV16 URR methylation in the permissive infections.
The methylation status of the HPV16 URR determined by bisulfite treatment followed by direct DNA sequencing of amplified DNA fragments using 3 sets of primers covering the complete HPV16 URR. DNA derived from laser-microdissected basal, intermediated and superficial layers of the cervical epithelium with permissive HPV16 infection was amplified and cloned. 12 individual clones were sequenced to identify the presence and patterns of the methylated CpG dinucleotides. The Figure visualizes schematically the methylation of 16 CpG dinucleotides (positions from 7198 to 161) in the URR of HPV 16 genomes. Open circles represent unmethylated CpGs, filled circles represent methylated CpGs.
Figure 5.
HPV16 URR methylation in the transforming infections.
The methylation status of the HPV16 URR determined by bisulfite treatment followed by direct DNA sequencing of amplified DNA fragments using 3 sets of primers covering the complete HPV16 URR. DNA derived from laser-microdissected p16INK4a-positive transformed epithelium was amplified and cloned. 12 individual clones were sequenced to identify the presence and patterns of the methylated CpG dinucleotides. The Figure visualizes schematically the methylation of 16 CpG dinucleotides (positions from 7198 to 161) in the URR of HPV 16 genomes. Open circles represent unmethylated CpGs, filled circles represent methylated CpGs.
Figure 6.
Schematic representation of the relative methylation frequencies of 16 CpG nucleotides within the HPV16 URR.
The degree of methylation of 12 independent clones each isolated from basal, intermediate and superficial cell layers from the three latent, permissive and five transforming parts of the cervical epithelium investigated by genomic sequencing as shown in Figure 3, 4 and 5. The y-axis shows the mean percentage of methylated HPV16 genomes at each of the 16 CpG dinucleotides ± standard deviation (SD). Mean values and SDs (three independent lesions per each CpG) were calculated from Sigma Plot 10.0. Nucleotide numbers refer to those of the HPV16 reference clone (GenBank NC_001526).
Figure 7.
Methylation of the E2BS1 increases the p97 promoter activity.
A. To measure the impact of methylation of the E2BS1 on the p97 promoter activity site specific modifications within the E2BS1 of LCR HPV16 of pGLuc reporter vector construct were generated using modified oligonucleotides. Three constructs containing either an unmethylated (wt E2BS1), a mutated (mutE2BS1) or a methylated (methE2BS1) E2BS1 were used. B and C. These plasmid constructs were co-transfected into C33A and NHK cells and increasing amount of E2 expression vector. The secreted Gaussia luciferase activities were normalized using the corresponding internal ß-galactosidase activities. Each value represents the mean ±standard deviation of at least three independent transfection experiments, each performed in triplicate. D. To analyze the impact of methylation of E2BS1 on the early promoter activity in the natural context of the HPV 16 genome where E2 expression is under control of the p97 early promoter, NHK were transiently transfected with full length HPV16 genome (wt or methylated E2BS1). E. The effect of selective E2BS1 methylation in full-length HPV16 on E6 gene expression was measured by real-time PCR. Wild type and HPV16 full-length genome and the full length HPV16 genome with 2 selectively methylated CpGs within E2BS1 were transfected into normal human foreskin keratinocytes (NHK.f.). Total RNA from cells 24, 48 and 72 hours after transfection was assayed for the expression of E6 mRNA. Relative luciferase activities were calculated with respect to the values of the wt construct, which was set to 1, for each time point. The data represent the mean of four independent experiments performed with each sample in triplicate with error bars indicating ± S.D. Mean values and SDs were calculated from Sigma Plot 10.0. F. To test whether methylation recruits other binding of transcription factors to the region of E2BS1, we performed EMSA analysis using nuclear cell extracts isolated from NHK. Methylated and unmethylated DNA probes spanning the regions containing the E2BS1 were used. The methylated probes showed three shifted bands by adding nuclear cell extracts (line 4, 6). Unmethylated probes were not shifted by the same nuclear extract (lane 3, 5). Competition experiments were performed with non-labeled probes (lines 7 and 8).
Table 1.
Primers used for bisulfite genomic sequencing of the HPV16 URR.