The Human Papillomavirus (HPV) E1 protein regulates the expression of cellular genes involved in immune response

The Human Papillomavirus (HPV) E1 protein is the only viral protein with enzymatic activity. The main known function of this protein is the regulation of the viral DNA replication. Nevertheless, it has been demonstrated that the ablation of HPV18 E1 mRNA in HeLa cells promotes a deregulation of several genes, particularly those involved in host defense mechanisms against viral infections; however, the specific contribution of E1 protein in HPV-independent context has not been studied. The aim of this work was to determine the effect of the HPV E1 protein in the regulation of cellular gene expression profiles evaluated through RNA-seq. We found that E1 proteins from HPV16 and 18 induced an overexpression of different set of genes associated with proliferation and differentiation processes, as well as downregulation of immune response genes, including IFNβ1 and IFNλ1 and Interferon-stimulated gene (ISG), which are important components involved in the antiviral immune response. Together, our results indicate that HR-(High-Risk) and LR-(Low-Risk) HPV E1 proteins play an important role in inhibiting the anti-viral immune response.


Results
HPV E1 proteins regulate cellular gene expression. In order to analyze the effect of HPV E1 from low and high-risk HPVs on cellular gene expression profiles, an RNA-Seq analysis was performed from HaCaT cells expressing the HPV E1 proteins from three viral types. E1 expression from HPV tested types was evaluated 24 h post-transfection by RT-PCR and western blot as shown in Fig. 1A,B, respectively. Western blot analysis detected the E1 protein at approximately 72 KDa as well as other bands produced by the E1 cleavage according to the reported by Moody et al. 16 . Complete blot of the membrane is displayed in Supplementary Fig. 1.
Once E1 expression was confirmed, RNA-Seq was carried out using RNA coming from HaCaT cells transfected with the different HPV E1 expressing plasmids, as well as the pCA vector. Data analysis was performed considering a p-value of ≤0.05 as significant. Based on the log 2-fold change values, we found that HPV16 E1 up-regulates 199 genes and down-regulates 103 genes while HPV18 E1 up-regulates 45 genes and down-regulates 41, and HPV11 E1 up-regulates 34 genes and down-regulates 38 (Additional files 1, 2 and 3). When analyzing the cellular process affected by the E1 deregulated genes, we found that HPV E1 affects genes involved mainly in immune response as well as viral genome regulation (Table 1). These results indicate that the E1 protein from both, high and low risk HPVs have an effect on different cellular processes but mainly associated with immune response.
To further dissect the effect of E1 on the expression of cellular genes, a second analysis was made to considering only the genes that were significant based on the adjusted p-value (p-adj) of ≤0.05. The results indicated that HPV16 E1 modulates 36 genes; while 15 and 14 genes were regulated by E1 HPV18 and HPV 11, respectively ( Table 2). Based on the log 2-fold change values, we found that HPV16 E1 up-regulates 22 genes and down-regulates 14 genes, HPV18 E1 up-regulates 9 genes and down-regulates 6 genes; and HPV11 E1 up-regulates 4 genes and down-regulates 10 genes (Fig. 2). To analyze the cellular processes in which genes affected by E1 are involved, a Gene Ontology analysis (ClueGO) was performed. Supporting the previous results, our data revealed that HPV16 E1 up-regulated genes that participate in immune response, cell-substrate junction assembly and cell proliferation (Fig. 3A,C). Moreover, HPV16 E1 down-regulated genes involved in immune response processes such as response to type I interferon and regulation of type 2 immune response (Fig. 3B,D). Meanwhile, Gene Ontology analysis revealed that genes up-regulated by HPV18 E1 are involved in response to mineralocorticoid and corticosterone (Fig. 4A,C), whilst down-regulated genes are related to processes associated with regulation of type 2 immune response (Fig. 4B,D). Finally, when analyzing genes de-regulated by the HPV11 E1 protein, no cellular processes were associated to the up-regulated genes, probably because of the small number of genes (four genes) compared to those up-regulated by the high-risk HPVs. In contrast, the ClueGO program showed that down-regulated genes by HPV11 E1 are involved in processes such as the negative regulation of viral genome replication, tyrosine phosphorylation of the STAT protein, type I interferon signaling pathway and regulation of type 2 immune response (Fig. 5A,B).
The E1 proteins from low-and high-risk HPVs regulate sets of genes in common. We determined whether genes altered by E1 proteins are shared among the different HPV types evaluated. Venn diagram, from the obtained data shows that HPV16 and HPV18 E1 proteins deregulated four shared genes: FOSB, DUSP6, NR4A and CCA1 (Fig. 6A). The expression of FOSB was verified by qRT-PCR (Fig. 6B), finding increased mRNA levels in the presence of E1 from HR-HPVs compared to the pCA expressing cells (p < 0.001).  www.nature.com/scientificreports www.nature.com/scientificreports/ HPV16 and HPV11 types shared only one gene (IFIT2). It was also found that E1 from HPV18 and HPV11 do not share de-regulated genes. Interestingly, four immune response genes are deregulated by the three HPV types (IFNB1, IFNL1, CCL5 or RANTES, and RSAD2 or Viperin) (Fig. 6A). This suggests that the HPV E1 proteins might have a high impact on the expression of genes associated with the immune response. In addition, it shows that this effect is shared between high and low risk HPVs. E1 proteins from HPV16, 18 and 11 decrease IFNβ1 and IFNλ1 expression. As it was shown, HPV E1 proteins induced a down-regulation in the expression of four genes involved in immune response (CCL5, RSAD2, IFNB1 and IFNL1). Due to their important role in the viral innate immune response 17 , both IFNβ1 and IFNλ1, were selected to perform further analyses. IFNβ1 and IFNλ1 promote the expression of ISG genes that help to control infectious processes allowing an efficient antiviral immune response 18 . Expression of IFNβ1 and IFNλ1 was validated through qPCR assays in HaCaT cells expressing the different E1 proteins. Interestingly, E1 viral proteins decreased IFNβ1 (Fig. 7A) and IFNλ1 (Fig. 7B) mRNA levels compared to the control vector (p < 0.001). Since the all three HPV types affected the expression of ISGs, we next tested the mRNA levels of CCL5 and RSAD2 (viperin) finding a significant decrease in presence of the three E1 proteins compared to pCA expressing cells (p < 0.001) (Fig. 8). Additionally, the mRNA levels of IFIT2, the only gene shared between HPV16 and HPV11 types, were significantly diminished (p < 0.001) by both E1 proteins (Fig. 8).
Since HPV E1 proteins inhibit the expression of IFN, we wonder if E1 could interfere with the production of IFN after stimulating with an inductor of IFNβ. Poly I:C is a synthetic analog of viral dsRNA and it is a well-established inductor of the IFNβ production 19,20  A previous report showed that 0.1 μg of Poly I:C was effective for stimulation and those amounts did not affect the viability of HaCaT cells at 24 hours 21 ; therefore, 0.1 μg was selected to perform subsequent experiments. To determine whether E1 interferes with IFNβ1 expression during Poly I:C treatment, HaCaT cells were co-transfected with 0.1 μg of Poly I:C and 4 μg of each HPV E1-expressing plasmid or the control plasmid. After www.nature.com/scientificreports www.nature.com/scientificreports/ 24 h post-transfection, expression of IFNβ1 was evaluated by qPCR. Figure 9B shows that cells expressing HPV16, 18 and 11 E1 proteins exhibit a decrease in the levels of IFNβ1 compared with cells transfected only with poly I:C plus the pCA vector (p < 0.0001). Additionally, it was demonstrated that 0.1 μg of Poly I:C promoted the expression of IFNλ1 mRNA in the presence of the control vector; while such expression decreased in E1-expressing cells (p < 0.0001) (Fig. 9C). Taken together, these results demonstrate that HPV E1 proteins from low-and high-risk decrease the expression of IFN even after stimulating IFN gene transcription, showing that HPV16, 18 and 11 E1 proteins downregulate the expression of two key components of the antiviral immune response.
In order to verify the possible mechanism implicated in the downregulation of IFNs induced by the HPV E1 proteins, we evaluated the levels of NFκB, a master regulator of immune related genes 22 , as well as the levels of IkBα, which is a negative regulator of NFκB. As shown in Fig. 9D, HPV E1 proteins increased p52 (NFκB) protein levels, which is mainly a positive regulator of the NFκB non-canonical pathway. Additionally, in the presence of E1 proteins, IκBα (NFκB canonical pathway negative regulator) protein levels decreased. These results indicate that HPV E1 activates not only the NFκB canonical pathway, but also the non-canonical pathway.

Discussion
The function so far attributed to the HPV E1 protein has been its participation in the viral replication process, along with E2 11 . However, it has been shown that E1 also associates with other cellular proteins that regulate epigenetic processes 12,13,23 . Thus, the ability of E1 to interact with chromatin modifiers to regulate viral replication, suggests that this protein could also affect cellular processes in the host genome such as gene expression.
In the present study, we determined the effect of E1 from high and low risk HPVs on cellular gene expression analyzed by RNA-seq observing that E1 protein affects the expression of different cellular processes, however, a greater effect on cellular processes associated with immune response (p-value < 0.05) was observed. More restricted analysis in which only genes with an adjusted p-value (p-adj < 0.05) were selected, identified a smaller number of genes regulated by the presence of E1, which are involved in immune response more specifically in antiviral immune response. Our results are in agreement with those of Castillo et al., (2014) who reported a change in the expression of genes associated with the innate immune response after the ablation of E1 expression in HeLa cells 15 . Furthermore, when comparing the three viral types tested, our results demonstrated that not only E1 from HPV18 promotes a change in gene expression but also E1 from HPV16 and from low-risk HPV 11, suggesting a probable similar mechanism in the regulation of gene expression performed by the E1 proteins. Particularly, 4 genes were downregulated in common by the three HPV E1 proteins (IFNβ1, IFNλ1, Viperin and CCL5), which participate in the antiviral immune response 17,24-26 . www.nature.com/scientificreports www.nature.com/scientificreports/ IFNβ1 and IFNλ1, are key mediators of the antiviral immune response, which ultimately promote the expression of interferon-stimulated (ISG) genes, helping to control the infectious processes, through multiple mechanisms such as protein translation inhibition, viral RNA degradation as well as the activation and survival of innate and adaptive immune cells such as Dendritic cells (DCs), macrophages, NK cells, and T cells, activating both innate and adaptive immune response to inhibit viral infection 18,27,28 . Due to the relevance of such elements, IFNβ1 and IFNλ1 were selected for further validation through qPCR, that demonstrated that mRNA levels decreased in the presence of the three viral types. Importantly, the E1 proteins blocked the IFNβ expression even after stimulating with Poly I:C which produces a similar effect than a viral infection, indicating that the downregulation induced by the E1 proteins at a transcriptional level has an impact during a normal challenge in immortalized keratinocytes. Similarly, when analyzing IFNλ1, we found that E1 also decreased its expression after Poly I:C treatment, suggesting that E1 could be acting as an immunosuppressive element in HPV infections. The fact that E1 decreases IFNβ1 and IFNλ1 mRNA levels, lead us to evaluate the expression of some ISGs, finding a decrease in the expression of CCL5, RSAD2 (viperin) and IFIT2 in the presence of E1 of HPV16, 18 and 11. Such results support the data obtained by RNA-seq, in which these ISGs were also diminished, demonstrating that E1 affects the expression of IFNs, which in turn affects the expression of ISGs. Thus, suggesting that E1 may play an important role in the antiviral immune response.
We found a reduction of the negative regulator IκBα in the presence of the three HPV E1 proteins, indicating that the NFκB canonical pathway is active. These results are in agreement with those reported by Nakahara et al., (2015) where HPV16 E1 induced the activation of NFκB in primary cervical keratinocytes 29 . To test whether E1 could activate the NFκB non-canonical pathway, we analyzed p52 levels, a product of NF-κB2/p100 processing. It is known that p52 associates with RelB in the cytoplasm and the complex RelB:p52 is required for the induction of several pro-inflammatory genes 30 . Interestingly, we found an increase in p52 levels in the presence of the three HPV E1 proteins. These results suggest that HPV E1 protein is somehow activating the two NFκB pathways promoting the expression of genes such as cytokines, chemokines, adhesion molecules, enzymes that produce secondary inflammatory mediators, and inhibitors of apoptosis 31 . Nevertheless, while the activation of NFκB should promote the expression of IFNβ we found that E1 decreases its expression. This contradictory effect could be due to the fact that NFκB is not essential for IFNβ expression 32 , although E1 could favor the expression of other NFκB genes. So it would be interesting to determine whether E1 affects other transcriptional factors or cofactors, that could regulate the expression of IFNβ, such as IRFs, since it has been shown that the Interferon Regulatory Factor 3 (IRF-3) promotes the expression of IFNβ 33 as has been shown for E6 and E7 oncoproteins [34][35][36] . www.nature.com/scientificreports www.nature.com/scientificreports/ That E1 blocks the IFNs production indicates that in addition to its participation in HPV infection, E1 could act on HPV persistency; nevertheless, the particular role of E1 in such process needs to be addressed. Therefore, it would be interesting to analyze the mechanism by which E1 interferes with the expression and production of IFN, as well as its effect on the expression of ISG. Some studies have demonstrated that the HPV produces an  www.nature.com/scientificreports www.nature.com/scientificreports/ alteration of key immunological elements either in primary keratinocytes in a context of whole genome as well as in cell lines. Several of the altered genes found in this study were confirmed by those studies including the analysis of IFNβ and RANTES 37,38 , and ISGs 15 . It is worth mentioning that the inhibition of INF expression has been observed in several viral infections such as Hepatitis, Herpesvirus and HTLV-1, where viruses use this strategy to evade the innate immune response favoring the infectious process [39][40][41] . Our results propose a novel mechanism in the modulation of the immune response induced by the HPV E1 protein, which could be associated to the replicative cycle and the viral persistence by evading the immune response.  HaCaT cells transfected with pCA vector control and HPV E1 expressing plasmids were collected and RNA extracted 24 h after transfection. Then, cDNA was synthesized and the mRNA levels of CCL5, RSAD2 (Viperin) and IFIT2 genes were measured using specific primers (Supplementary Table 1). Relative mRNA levels of CCL5, RSAD2 (Viperin) and IFIT2 genes showing a decrease in presence of the HPV-16, 18 and 11 E1. Values are expressed as the difference in double delta-Ct compared with pCA control transfected cells. The expression of the housekeeping gene 18 S was used for normalizing. Bars represent the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 vs control (pCA). www.nature.com/scientificreports www.nature.com/scientificreports/ Among the processes associated with the development of cancer induced by HPV, viral persistence represents a leading key event at the beginning of this process 42 . Recently it was determined that E1 is expressed throughout the different stages of cervical cancer 43 . The authors analyzed the levels of HPV16 E1 mRNA in biopsies with different lesion degrees (CIN) as well as in cervical cancer, observing an increased expression of E1 as long as the lesion advances. In this regard, the analysis of genes differentially regulated by the E1 proteins, specifically those from the high-risk HPVs, revealed that E1 from both HPV16 and 18 types promote the expression of genes associated with cell proliferation, differentiation, cell viability and metabolism regulation (FOSB, DUSP6, NR4A and CCNA1) [44][45][46] , processes that could favor the replicative cycle. However, it has been demonstrated that overexpression of some of these genes play an important role in carcinogenesis associated with poor prognosis, resistance to treatment, recurrence and poor survival in some cancers (non-small-cell lung carcinoma, ovarian cancer) [46][47][48] . Moreover, E1 from HPV16 regulated the expression of genes such as MMP1, EREG, ANGEPTL4 which regulate cellular processes such as keratinocyte migration and re-epithelialization, proliferation and angiogenesis 49,50 , interestingly, the overexpression of those genes has been associated with the carcinogenesis process 26,[51][52][53][54] . Taken together, these data suggest that E1 could play a role in the development of HPV-related cancers through the overexpression of certain cellular genes. During the carcinogenic process there exists a deregulation of gene expression and epigenetic mechanisms, which play an important role at the different stages of cancer; these mechanisms are capable of altering the organization of chromatin, through chromatin modifiers, favoring or inhibiting the expression of genes 55 . It is known that the HPV genome organizes itself into a chromatin like structure 56 , regulating viral replication and transcription, which requires the participation of chromatin modifiers. Also, it has been shown that E1 interacts with the histone H1 and with the Ini1/hSNF5 12,13 , which are involved in the regulation of gene expression. Therefore, the ability of E1 to interact with those proteins, as well as with other chromatin modifiers as acetyltransferases (HATs) and deacetylases (HDACs), could be one of the mechanisms implicated in the regulation of gene expression performed by the E1 protein. In addition, this suggests that E1 could not only regulate the HPV genome, but also the host genome, promoting gene expression that could be favoring in part the infectious process and probably carcinogenesis.
It is also possible that E1 regulates gene expression by directly binding to the cellular genome on regulatory elements (promoters or enhancers), through its DNA binding domain 57 . This has been demonstrated for the Epstein-Barr virus (EBV) EBNA1 protein, which directly bind to cellular promoters, which was related to changes in gene expression profiles 58 . However, it is likely that an indirect interaction of E1 with DNA through its interaction with transcription factors or with chromatin modifiers could be happening. Therefore, it would be interesting to study new E1 interacting partners that allow the modulation of gene expression, including the modulation of other cellular processes, such as its participation in cell transformation.

conclusions
Our results demonstrate that HPV E1 proteins induces changes in cellular gene expression, which is a process that is conserved among high-and low-risk HPVs. E1 of HPV16 and 18 are capable of regulating the expression of genes involved in processes of proliferation, migration and metastasis which are associated with carcinogenesis. Otherwise, E1 of the three viral types were able to modulate the expression of genes associated with the antiviral immune response, and also avoided the IFNβ1 and IFNλ1 expression, as well as of ISGs. Although the mechanism for such regulation remains unclear, our results suggest that both, the canonical, as well as the non-canonical NFκB pathways are affected by the HPV E1 proteins. Nevertheless, the participation of NFκB pathways in the regulation of IFN needs further investigation. Therefore, in addition to its well-known role in viral replication, the E1 protein may also play an important role in the evasion of the immune response to favor the replicative cycle, persistence and development of cancer.

Materials and Methods
Cell culture and transfections. The  Plasmids. HPV16 and HPV11 HA tagged E1 expressing plasmids were kindly provided from Dr. Mart Ustav (University of Tartu, Estonia). HPV18 HA tagged E1 expressing plasmid was previously described by our group 59 . The pCA vector was used as transfection control. All plasmids were verified by sequencing.

RNA Isolation and cDNA synthesis.
HaCaT cells were seeded in a 60 mm culture dish and transfected with 4.5 µg of each of the three viral types E1 plasmid in triplicate. After 24 h of transfection, total RNA was isolated from each condition using the RNeasy Mini kit (Qiagen, Hilden, Germany), according to the manufacturer's protocol. RNA was resuspended in 30 μL of RNAse-free H 2 O, the was treated with the DNAse Free DNA removal kit (Thermo Fisher Scientific, Waltham, MA, USA) and quantified. 400 µg of RNA was reverse-transcribed with Oligo dT utilizing the GeneAmp RNA PCR Core Kit (Applied Biosystems, Foster City, CA, USA) The expression of E1 from the different HPVs were verified by RT-PCR using specific primers listed in Supplementary Table 1. RNA-Seq: Illumina library preparation and sequencing. Total RNA was isolated from three independent transfections per condition as indicated. Total RNA was quantified spectrophotometrically using a NanoDrop ND-1000 spectrometer (Thermo Scientific, Waltham, MA, USA); integrity was analyzed by capillary electrophoresis using the Bioanalyzer 2100 (Agilent Technologies Inc., Santa Clara, CA, USA), all samples exhibited a high-quality RNA integrity number (RIN) > 9.0. RNA-seq libraries from all replicas and conditions (n = 12) were constructed using the TruSeq Stranded mRNA Library Preparation kit (Ilumina, San Diego, CA) as 76-bp paired-end reads length. Samples were sequenced using an Illumina NextSeq 500 platform, generating at least 40 million reads per sample. Paired-end reads were assigned quality scores and aligned to the reference genome RefSeq (hg19) using bowtie and then count files were generate with Htseq-count, and diferential expression performed with DESeq2 where p-value of < 0.05 and Adjusted p-value < 0.05 were consider as statistically significant. The pipepline was carried out using the tools provided in Illumina´s BaseSpace. Biological processes analysis (GO category) was performed using differentially expressed genes and visualized with ClueGo, a Cytoscape plug-in. Ingenuity pathway analysis was performed to identify canonical signaling pathways and functional pathways affected by the differentially expressed genes. Fisher's exact test was done automatically by the software.
Gene expression assay. Total RNA was isolated using the RNeasy mini kit (Qiagen, Hilden, Germany).
The obtained RNA was treated with the DNAse Free DNA removal kit (Thermo Fisher Scientific, Waltham, MA,