Background
Papillomaviruses are small double-stranded DNA viruses that are classified into 16 genera according to their genotypes [
1,
2]. The best-studied PVs belong to the group of alpha-subtype viruses (α-PVs) that infect the oral and anogenital mucosa of humans and that can cause condylomas, genital warts (low-risk α-HPVs, such as HPV6 and 11) or anogenital cancers (high-risk α-HPVs, such as HPV16, 18 and 31). The organization of the HPV genome, which is conserved among PVs, contains early (E) and late (L) coding regions and a non-coding long control region (LCR) also known as the upstream regulatory region (URR). The early region genes are expressed at different levels throughout the whole infection period and are responsible for virus DNA replication (E1 and E2), transcriptional regulation (E2) and cellular regulation/transformation (E6, E7, E4, and E5). The late genes, which encode the viral capsid proteins L1 and L2, are expressed in terminally differentiated keratinocytes during the productive phase of infection. The non-coding upstream regulatory region contains
cis-elements for cellular factors, as well as for viral DNA replication, transcription and genome maintenance. The DNA replication cycle of PVs can be divided into 3 phases: initial genome amplification, during which the copy number increases rapidly in infected cultures to up to 50-100 copies per cell; the latent replication phase, during which the viral genome is replicated in synchrony with the cellular DNA; and the vegetative phase, which includes the second genome amplification, viral capsid proteins expression and virion assembly.
Two vaccines against α-papillomaviruses have been developed and marketed as Cervarix (GlaxoSmithKline, UK) and Gardasil (Merck Research Laboratories, US). For review see Malik et al. [
3]. These vaccines induce protective neutralizing antibodies against L1 protein epitopes and target the high-risk viruses HPV16 and HPV18, which are the etiological agents responsible for the induction of cervical carcinoma. Gardasil also provides protection against genital warts and cancers of the anus, vagina and vulva by targeting the low-risk viruses HPV6B and HPV11. These vaccines are highly effective in raising specific immune responses and provide protection against HPV6B, HPV11, HPV16 and HPV18 viruses when vaccinations are performed in young girls before sexual exposure [
4,
5]. The factors preventing the wide and effective use of the current HPV vaccines include their cost and their partial protection against many other papillomavirus types. Although, cross-protection for closely related HPV types like HPV31 and 45 has been observed [
6] there are at least 13 HPV types which are potentially oncogenic [
7]. Also, current vaccines do not generate therapeutic effect against pre-existing HPV infection. Therefore, a real unmet medical need for drugs with a wider spectrum of specificity against most papillomaviruses exists.
One of the more specific targets for therapeutic intervention against HPV infection could be the replication machinery of HPVs. The wide variety of papillomaviruses can be specifically targeted for drug intervention by focusing on the expression of the viral E1 and E2 proteins and their interaction with cellular proteins and specific features of the replication machinery. Such drugs may show wide specificity for HPVs because of the great similarities in the interactions of the E1 and E2 proteins of different HPV subtypes with their specific cellular counterparts and DNA sequences. Thus, drugs developed against specific HPV replication stages may have a specific effect on a wide variety of HPVs. To screen for such HPV replication inhibitors, a cell-based assay must be developed in which HPV genome replication can be effectively quantitated in a high-throughput screening (HTS) format.
The initial phases of PV infection until the generation of viral particles can be studied in raft cultures using transfected PV genomes in human primary keratinocytes or squamous carcinoma cell lines (e.g.
, SCC-4) [
8]. Alternatively, naturally derived cell lines like W12 (HPV16) or CIN612 (HPV31) already harboring replicating HPV DNA episomes allow the latent and vegetative phases of the PV life cycle to be studied [
9-
11]. Although investigating the molecular mechanisms of HPV replication in raft cultures is important for a complete understanding of viral genome replication in differentiating cells of specific tissues, this method is difficult to use for screening potential drug candidates in HTS assays. This issue also applies to the use of primary keratinocyte cultures for HPV replication because of the need for donors of primary cells, in addition to issues concerning the genetic uniformity of the assay. Alternatively, NIKS cells which are non-HPV-containing immortalized keratinocytes could be used to develop an HTS assay, although the robustness of this methodology must be improved for the effective use of this system [
11,
12].
Previously, we have successfully employed the human osteosarcoma U2OS cell line to analyze genome replication and gene expression in α- and β-HPVs [
13-
16]. The initial amplification and latent phases of stable PV replication can be monitored effectively and the subcloning of stable HPV cell lines can be performed in this cell line. Additionally, cloned HPV cell lines can be cultured under dense conditions, thereby triggering the second amplification phase in the case of α-HPVs, which is reminiscent of the vegetative amplification that occurs during the HPV life cycle before late genes expression [
13]. However, virus particle assembly has not been detected in these cells because sufficient expression of the capsid proteins L1 and L2 cannot be induced [
15,
17]. Transcription maps of HPV18 and HPV5 have been compiled in the U2OS cell line [
15,
17] and compared with previous studies [
18,
19]. This comparison concluded that transcription maps of these viruses in U2OS cells and in the keratinocytes
in vivo are very similar, if not identical. Therefore the construction of a high-throughput screening system for inhibitors of the gene transcription and genome replication processes of these viruses in U2OS cells could be possible [
15,
17,
18].
The primary aims of this work were to elucidate the molecular mechanisms of viral gene expression and genome replication further and to confirm whether U2OS cells can be used as a convenient system for molecular studies of HPV11 and as a platform for screening antiviral compounds. We found that the gene expression profile of the HPV11 genome in U2OS cells is very similar to the previously described gene expression in keratinocytes [
20-
27]. Thus, our data suggest that the HPV11 transcription map obtained herein reflects the situation in vivo and confirm that a U2OS-based system is potentially suitable for screening anti-HPV11 drugs that suppress viral DNA replication.
Discussion
Previously, we have used the human osteosarcoma cell line U2OS to study α- and β-HPV DNA replication and gene expression patterns [
13-
17]. Our analyses employing the HPV11wt and E8- genomes in U2OS cells showed that the events of the papillomavirus three-step replication cycle could be mimicked, as in native keratinocyte cell lines, including the first amplification of the virus genome; the latent phase, where the genome copy number is stable; and the second genome amplification phase. The initiation of the replication of the HPV11wt genome, and particularly the E8- genome, was efficient in U2OS cells. Some differences in maintenance were observed when HPV11-positive cells were grown as actively proliferating cultures versus slow dividing dense cultures; intensive passaging of the cells led to some reduction of the viral genome copies for both, the HPV11wt and E8- genomes. The same tendency was also observed for the subconfluently cultivated low-risk HPV11 and HPV6b subclones when we introduced re-ligated HPV genomes into U2OS cells [
13]. The high-risk HPV18 and HPV16 maintain stably their genomes in the cells under subconfluent conditions with active passaging [
13]. Subclones of the β-papillomavirus HPV5 and HPV8 also maintain their copy numbers under subconfluent conditions [
13,
15]. In contrast, under confluent culture conditions, the HPV11wt genomes are effectively maintained in the U2OS cells, while in the case of HPV11E8- mutant 5 to 6 times amplification of the genome occurred. These data indicate that the amplification of the HPV11 E8- mutant genomes under confluent culture conditions generally exhibits the features of α-HPV replication, as previously described for the α-papillomaviruses HPV18 and HPV16 [
13,
17]. Notably, the β-papillomavirus HPV5 is not capable of amplifying the replication signal in confluent cultures of U2OS cells [
15]. One of the reasons may be the difference in the structural organization of the URR regions and regulation of gene expression between the α- and β-papillomaviruses or the response to certain positive cellular factors capable of activating viral replication, which are expressed at considerably higher levels compared with those of subconfluent cultures. Alternatively, cell division in confluent cultures is slowed down, which allows the viral genome to replicate for a longer period because HPV genomes are not restricted to once-per-cell cycle replication, and genome replication occurs not only in S phase but also in G2 phase [
11,
14,
40].
The transcriptional data obtained from the transiently transfected HPV11-positive U2OS cell line are identical to the transcripts found in condylomata acuminata [
20], experimental condylomata [
23] and the human squamous carcinoma cell line SCC-4 [
26]. Indications of transcription events from late region genes could be observed from both promoter analyses and the mapping of the polyA signal. However, the isolation of intact mRNAs from the L1 and L2 genes was unsuccessful, indicating that these transcripts are not stable and that the L1 and L2 structural proteins are not expressed in U2OS cells. Although the second amplification of the viral genome could be observed, the expression of the late protein mRNA was not detected.
The E8 ORF is highly conserved among α-HPVs. During HPV11 transcripts analyses, we identified two mRNAs with an E8 ORF within their sequences: one possibly encoding the E8
^E1 fusion protein and a second possibly encoding the E8
^E2 fusion protein. The fusion protein E8
^E2 is responsible for the repression of viral protein expression, including that of the viral replication proteins E1 and E2 [
15,
17,
31,
34]. Mutation of the E8 start codon in the HPV11 genome increase DNA replication by at least 6-fold. However, in the case of the HPV5, HPV18 and HPV31, the effect on DNA replication initiation is even greater, ranging from 30- to over 100-fold [
15,
17,
34]. Analyzing the expression of the E8
^E1 and E8
^E2 fusion proteins and their effect on HPV11E8- genome replication in the present study revealed that the E8
^E1 protein did not affect DNA replication and may be important in other aspects of the HPV11 life cycle. In contrast, the E8
^E2 protein repressed genome replication in a concentration-dependent manner. Analyzing the HPV11E8- genome mRNA transcripts revealed that the E8
^E2 protein negatively regulates its own mRNA transcription, in addition to repressing the transcription of E1 and E2 mRNAs, suggesting that this protein is part of the negative regulatory loop for the viral gene expression.
Analysis of the transcripts obtained from HPV11-positive U2OS cells indicated that the only transcript encoding the E1 replication protein is the polycistronic mRNA consisting of all of the early genes in the HPV11 genome: E6, E7, E1, E2, E4, E5a and E5b. Previous studies of HPV11 transcripts and E1 protein expression have suggested that E1 can also be translated from mRNA in which the TSS is at the end of the E7 ORF; therefore, the E6 and E7 genes are absent [
20]. To study this difference, we designed two E1 expression vectors: one containing the E6 and E7 leader sequence and a second in which the leader sequence was absent. The E1 protein from the L+ plasmid was functional and supported DNA replication, whereas the E1 protein from the L- plasmid was not expressed and did not support viral DNA replication as effectively as that from the L+ plasmid. This difference might be explained by mRNA stability or by translational effectiveness. In the case of high-risk α-HPVs (e.g., HPV16, HPV18, and HPV31), an intron within the E6 ORF that is spliced in E1-containing mRNAs, which might be the key event in the expression of biologically active E1 protein [
17,
36,
38,
39]. To study this possibility in the HPV11 system, we generated E1 expression vectors in which the leader sequence was substituted with one or two heterologous artificial introns. The splicing event between the promoter and E1 gene restored the effect of the leader sequence. The E1 protein was expressed effectively and supported viral DNA replication. Two introns have an even greater effect on E1 protein expression and activity. In contrast, the analysis of the transcripts obtained from the L+ sequence did not reveal any introns within the leader sequence, indicating that other regulatory factors in the HPV11 E6/E7 region assure effective and biologically active E1 protein expression.
Materials and methods
Cell lines and transfection
U2OS cells obtained from the American Type Culture Collection (ATCC) (number HTB-96) were grown in Iscove’s modified Dulbecco’s medium (IMDM) supplemented with 10% fetal calf serum (FCS). The cells were transfected via electroporation (220 V, 975 μF) using a Bio-Rad Gene Pulser II apparatus supplied with a capacitance extender (Bio-Rad Laboratories). Episomal HPV11 genome amplification was achieved by growing transfected cells under confluent conditions for 10 days without passaging. At 48 h post-transfection, the cells were subjected to geneticin sulfate G418 selection (final concentration, 400 μg/ml, BioTop) to eliminate untransfected cells. The medium was changed every second day. Total DNA was extracted at the time points indicated below.
Plasmids
The HPV11 genome sequence and its numbering are based on the sequence of GenBank accession no. M14119. Minicircle producing plasmid backbone pMC.BESBX [
41] was inserted into BamHI site in the HPV11 genome L1 region (between nt 7072 and 7073 in HPV11 genome). Almost all of bacterial backbone is removed during minicircle plasmid preparation [
14]. The final product is circular HPV11 genome DNA with 91 bacterial backbone derived nucleotides between positions 7072 and 7073. For the HPV11E8- mutant genome, the ATG start codon from pMC.BESPXHPV11wt (at the nt 1242 position in HPV11 genome) was mutated to ACG. To produce HPV11 genome for electroporation: HPV11wt and HPV11E8- minicircle genomes were produced from the parental plasmids in
E. coli and purified as described by Reinson et al. [
14]. To achieve complete removal of the unrecombined parental plasmids and the pMC.BESPX vector, the plasmids extracted from the bacterial cells were subjected to an additional gel purification step.
The E8^E1 expression vector pQMCFH11E8E1 was generated by amplification of the HPV11 E1 C-terminus from the plasmid pMC.BESPXHPV11wt with the primers 11E8-E1_F_NheI (5′-TCTGCTAGCGCCACCATGGCTATTCTGAAGTGGAAGCTGCAACGCAGAAATGGGAATGCAGTATATGAACTATC-3′) and 11E1_R_Bsp (5′-CCCTTCGAATCATAAAGTTCTAACAACTGATCCTG-3′). The E8^E2 expression vector pQMCFH11E8E2 was generated by amplification of the HPV11 E2 C-terminus from the pMC.BESPXHPV11wt plasmid with the primers 11E8-E2_F_NheI (5′-TCTGCTAGCGCCACCATGGCTATTCTGAAGTGGAAGCTGCAACGCAGCACTGTACGAGAAGTATCCATTG-3′) and 11E2_R_Bsp119I (5′-CCCTTCGAATTACAATAAATGTAATGACATAAACCC-3′). The PCR products were cleaved with the enzymes NheI and Bsp119I (Thermo Scientific) and inserted into the vector pQMCF_MCS_9 (Icosagen Cell Factory Ltd.), which was opened through NheI/Bsp119I digestion.
The cloning of the HPV11 URR plasmid and the L+ E1 and E2 expression vectors was described previously by Kadaja et al. (designated pUCURR-11, pMHE1-11 and pQMNE2-11, respectively) [
37]. The HPV11 L- E1 expression vector was obtained by amplification of the E1 ORF from the HPV11 L+ E1 expression vector (nt 823 to 2781 in the HPV11 genome) using the primers E1.Xba-11 (5′-CCTCTAGATCCATGGCGGACGATTCA-3′) and E1.3.11 (5′-GGAAGCTTTTCTTCATAAAGTTCTAACAACTGATCC-3′). The PCR product was cleaved with XbaI and HindIII and inserted into the XbaI/HindIII site of the eukaryotic expression vector pQM-Ntag/Ai + (Icosagen Cell Factory Ltd.). The HPV11 Int1 and Int2 E1 expression vectors were obtained by cleaving the vector pRL-TK (Promega) with the enzymes HindIII and NheI and by cloning the pRL-TK intron-containing region into the XbaI site of the HPV11 L- vector. By blunt-end cloning, we obtained E1 expression plasmids with one (Int1) or two (Int2) introns between the CMV promoter and the E1 gene. All of the modifications of the HPV11 L+ E1 expression vector described in Kadaja et al. were also performed for the L-, Int1 and Int2 E1 expression plasmids [
37].
RNA extraction and rapid amplification of cDNA ends (RACE)
U2OS cells were transfected with 500 ng of the HPV11wt or HPV11E8- genome, and samples used for establish the stable expression were transfected with 2 μg of the linearized pBabe-Neo construct. PolyA
+ RNA templates were extracted from cells using a Micro-FastTrack™ 2.0 Kit (Invitrogen) at 4 and 10 days post-transfection. Then, 275 ng or 350 ng of polyA
+ RNA was used as a template for 5′ or 3′ RACE analysis, respectively. Both the 5′ and 3’ RACE assays were performed using a SMARTer™ RACE cDNA Amplification Kit (Clontech) according to the manufacturer’s instructions. The positions and the sequences of the HPV11-specific primers used for the amplification of the RACE products are shown in Figure
2. The amplified RACE products were purified from agarose gels, cloned and fully sequenced as single clones.
RT-PCR
Samples of 200 ng of polyA
+ RNA extracted from cells using a Micro-FastTrack™ 2.0 kit (Invitrogen) were subjected to DNase treatment with a Turbo DNA-free Kit (Ambion), and cDNA was synthesized using a Superscript III First Strand cDNA synthesis Kit (Invitrogen) according to the manufacturer’s instructions. Then, 4 μl of cDNA was used in a single PCR reaction, along with 4 μl of 5× HOT FIREPol® Blend Master Mix with BSA, 12.5 mM MgCl
2 (Solis Biodyne) and the HPV11-specific primers pr1241 and pr3692c (Figure
2) in a 20 μl of total reaction volume. Amplification of the newly synthesized cDNA was performed using the following program: 95°C 15′, 94°C 30″, 65°C 30″, and 72°C 2′, for a maximum of 25 cycles. The RT-PCR products were purified from agarose gels, cloned and fully sequenced as single clones.
Replication analysis
Extrachromosomal DNA was extracted from the transfected cells at 1 to 5 days post-transfection via the modified Hirt method [
42], or total genomic DNA was extracted at 10 to 21 days post-transfection from U2OS cells as described previously [
13]. Before Southern blot analysis, half of each extrachromosomal DNA sample or 3 μg of each total DNA sample was treated with the linearizing enzyme HindIII, and non-replicated, bacterially produced dam-methylated input DNA was fragmented with DpnI (Thermo Scientific). The DNA samples were resolved on a 0.8% (extrachromosomal DNA) or 0.6% (total genomic DNA) agarose gel using 1× Tris-acetate-EDTA (TAE) buffer. Electrophoresis was performed at 0.8 V/cm for 20 h. The separated DNA fragments were treated with 0.1 M hydrochloric acid, denatured with Sol A (0.5 M NaOH, 1.5 M NaCl), neutralized with Sol B (1 M Tris pH 7.4, 1.5 M NaCl), transferred to a Hybond-N+ membrane (Amersham Pharmacia Biotech) via the Southern blot method and hybridized with a radiolabelled HPV11-specific probe. HPV11 replication signals were detected via exposure to X-ray film.
Western blotting
Cells in a 60 mm culture dish were washed twice with 1× PBS, detached from plates with PBS-3 mM EDTA (pH 7.5) and collected via centrifugation. The cells were lysed with 1× Laemmli buffer (4% SDS, 20% glycerol, 120 mM Tris-Cl (pH 6.8), and 200 mM DTT), boiled at 100°C for 5 min and centrifuged for 10 min at 14000×g. The proteins were separated in 8 to 10% polyacrylamide-SDS gels and transferred to Immobilon-P membranes (Millipore, USA). The HPV11 E1 protein with a hemagglutinin (HA) epitope tag was detected using the rat polyclonal antibody 3F10-HRP (12013819001; Roche), which was diluted 1:500. An antibody against the cellular marker α-tubulin (B-512, Sigma-Aldrich) was used at a dilution of 1:5000. For visualization, a secondary α-mouse-HRP antibody (LabAS) was used at a dilution of 1:10000.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declared that they have no competing interests.
Authors’ contributions
Conceived and designed the experiments: HIP AM MU. Performed the experiments: HIP. Analyzed the data: HIP, AM, EU, MU. Contributed to the writing of the manuscript: HIP, AM, EU, MU. All authors read and approved the final manuscript.