A splice donor in E6 influences keratinocyte immortalization by beta-HPV49

Abstract

Human papillomaviruses (HPV) from the genus beta have been implicated in the development of cutaneous squamous cell cancer in epidermodysplasia verruciformis and organ transplant patients. In contrast to alpha-high-risk HPV, which cause ano-genital and oropharyngeal cancers, beta-HPV replication is not well understood. The beta-HPV49 transcriptome was analyzed by RNA sequencing using stable keratinocyte cell lines maintaining high levels of extrachromosomally replicating E8- genomes, which can be established due to a lack of the viral E8^E2 repressor protein. This analysis indicated the presence of four transcription start sites, two polyadenylation signals, and splice donor (SD) and acceptor sites consistent with the conserved gene expression patterns of animal and human PV. Surprisingly, a novel SD in the E6 oncogene (SD217) was identified resembling the SD in E6 of carcinogenic alpha-HPV. Mutation of SD217 enhanced E6 protein expression but had no influence on the growth of keratinocytes transduced with retroviral HPV49 E6 and E7 expression vectors. Inactivation of SD217 in the context of the HPV49 wild-type genome did not enable immortalization and prevented immortalization in the context of the E8- genome. The analysis of SD217 mutant genomes revealed a strong down-regulation of SD217 usage, but only weak effects on other viral transcripts. This suggests that SD217 does not contribute to immortalization by modulating viral gene expression. Usage of SD217 is increased in immortalized E8- cell lines compared with transiently transfected cells, which may indicate that long-term extrachromosomal maintenance requires reduced E6 protein levels.IMPORTANCEHigh-risk (hr) human papillomaviruses (HPV) from the genus alpha cause ano-genital and oropharyngeal cancers, whereas beta-HPV have been implicated to cause skin cancer in epidermodysplasia verruciformis and organ transplant patients. In contrast to alpha hr-HPV, the replication cycle of beta-HPV is not very well understood. Transcriptional profiling of beta-HPV49 by RNA sequencing reveals transcription start sites and splice sites conserved among HPV. Surprisingly, a splice donor site in the E6 oncogene (SD217), previously only described for hr-HPV, was identified that controls E6 oncoprotein levels and is required for immortalization of keratinocytes by the HPV49 genome.

Keywords: E6; RNA splicing; beta-HPV; viral transcriptome.

Fig 1

Visualization of read coverage of RNA sequencing data mapped to the HPV49 reference genome linearized at nt. 1. HPV49 open reading frames (ORFs), the URR, and spliced genes are indicated below. Splice signals (SD; SA) followed by the nucleotide number are indicated by arrows and are shown in black for previously identified sites and in green for major novel sites. The positions of consensus poly adenylation (polyA) sites are indicated by arrows. Data were obtained from three different cell lines derived from different donors. A log10 scale is used on the y-axis.

Fig 2

Exon junctions from RNA sequencing data of three different HPV49 E8- genome containing cell lines were identified (HISAT2 aligner, enable spliced alignment), displayed and counted with the integrated genome viewer, and are depicted as a Sashimi plot. Read abundances on the y-axis are shown on a linear scale. Only splice junctions with >100 supporting reads are shown. Only splice junctions with more than 100 reads and present in all three cell lines are listed in the table. The linearized HPV49 reference genome with ORFs is shown below. The positions of consensus poly adenylation sites are indicated by pins.

Fig 3

(A) Agarose gel analysis of the RT-qPCR product (P) to detect the SD217/SA3281 splice junction in total RNA isolated from HPV49 E8- cells. A size marker (M) is shown on the left. A partial sequence of the RT-qPCR product obtained is shown below. qPCR analysis of unspliced E6 and spliced SD217/SA3281 transcripts in stable HPV49 E8- cell lines using total RNA (B) or in NHK transiently transfected with HPV49 wt or E8- genomes harvested 3, 6, or 9 days post transfection using polyA⁺-enriched RNA (C). (B) n = 9, paired t-test. (***P < 0.001); (C) n = 4. Error bars indicate the SEM.

Fig 4

(A) Sequence alignment of the SD consensus (25), HPV49 SD217 wt, SD217 mt, and codon-optimized (co) E6 sequences. (B) qPCR analysis of unspliced E6 transcripts using polyA⁺-enriched RNA isolated from C33A cells transfected with pSG HPV49 3xHA E6 (E6), pSG HPV49 3xHA E6 SD217 mt (E6 SD217 mt), or pSG HPV49 3xHA E6co (E6 co) plasmids. n = 5, statistical significance was determined by a ratio-paired t-test (***P < 0.01). (C) Immunoblot analysis of C33A (left panel) or NHK (right panel) transfected with the empty vector or expression vectors for 3xHA-tagged wt E6, E6 co, or E6 SD217 mt is shown below. E6 was detected with an anti-HA antibody, and HSP90 was used as a loading control. In the right panel, a pre-stained protein marker (M) is shown.

Fig 5

(A) Overview of the immortalization capabilities of HPV49 E6 and E7 in NHK. NHK from three different donors were transduced with combinations of recombinant retroviruses as indicated in the table. + indicates immortalization, +/- indicates prolonged life span, and - indicates no prolonged life span. (B) Growth curves of cell lines immortalized with HPV49 E7 and with or without different E6 (wt, co, and SD217 mt). Data are derived from five to nine independent experiments for E6/E7 cell lines. The growth curve of HPV49 E7 is derived from one experiment. Error bars indicate the SEM.

Fig 6

Viral gene expression analysis of NHK transiently transfected with different HPV49 genomes as indicated by qPCR using PGK1 as a reference gene. (A) PolyA⁺-enriched RNA was isolated 6 days post transfection and analyzed for SD217/SA3281, E6, E1, and E7 transcripts. (B) Total RNA was isolated 6 days post transfection and analyzed for URR^E4, E8^E2, E1^E2, and E1^E4 transcripts. Values were calculated from plasmid standard curves. Data are derived from five (A) or nine (B) independent transfection experiments. Statistical significance was determined using a ratio-paired t-test using wt/ E8- as reference for the respective SD217 mt (*P < 0.05). Error bars indicate the SEM.

Fig 7

Transcript map of HPV49. The HPV49 genome was linearized at nt. 7,500 to enable the depiction of transcripts initiated at the putative promoter in the 5′URR. ORFs are shown above the linearized genome. Potential TSS are depicted by arrows and polyadenylation signals by pins. Spliced transcripts are shown below the linearized genome. Solid lines represent introns removed from spliced transcripts and dotted lines indicate unknown 5′- or 3′-extensions. Numbers indicate the last or first nt. of the exon.