Hepatitis C Core Protein Induces a Genotype-Specific Susceptibility of Hepatocytes to TNF-Induced Death In Vitro and In Vivo

Hepatitis C virus (HCV) core protein is a multifunctional protein that is involved in the proliferation, inflammation, and apoptosis mechanism of hepatocytes. HCV core protein genetic variability has been implicated in various outcomes of HCV pathology and treatment. In the present study, we aimed to analyze the role of the HCV core protein in tumor necrosis factor α (TNFα)-induced death under the viewpoint of HCV genetic variability. Immortalized hepatocytes (IHH), and not the Huh 7.5 hepatoma cell line, stably expressing HCV subtype 4a and HCV subtype 4f core proteins showed that only the HCV 4a core protein could increase sensitivity to TNFα-induced death. Development of two transgenic mice expressing the two different core proteins under the liver-specific promoter of transthyretin (TTR) allowed for the in vivo assessment of the role of the core in TNFα-induced death. Using the TNFα-dependent model of lipopolysaccharide/D-galactosamine (LPS/Dgal), we were able to recapitulate the in vitro results in IHH cells in vivo. Transgenic mice expressing the HCV 4a core protein were more susceptible to the LPS/Dgal model, while mice expressing the HCV 4f core protein had the same susceptibility as their littermate controls. Transcriptome analysis in liver biopsies from these transgenic mice gave insights into HCV core molecular pathogenesis while linking HCV core protein genetic variability to differential pathology in vivo.


Introduction
Hepatitis C virus (HCV) imposes a significant health burden, infecting an estimated 71.1 million individuals worldwide [1]. HCV is responsible for acute and chronic liver disease that often leads to fibrosis/cirrhosis, liver failure, and hepatocellular carcinoma (HCC). The development of potent interferon-free treatments has reduced the incidence of HCV-related liver disease and mortality [2,3]. However, a significant percentage of patients with advanced liver disease will develop HCC, in spite of treatment-induced viral clearance [4].

Lentivirus Expression Vectors and Stable Cell Lines
Core4a and core4f genes ( Figure S1) were amplified by PCR and cloned into the lentiviral vector pWPI-BLR [30], a derivative of the bicistronic lentiviral vector pWPI. In this vector, the expression of the gene is directed by the human elongation factor 1 alpha (EF1-α) promoter. Blasticidin resistance allows for selection in mammalian cells. Lentivirus particles were produced in 293T cells using packaging constructs pCMVR8.91 and pMD.G, as previously described [31]. Infection of Huh7.5 and IHH cells with lentivirus particles and selection with blasticidin yielded the respective Huh7.5-core4a, Huh7.5-core4f, IHH-core4a, and IHH-core4f stable cell lines. Lentivirus particles produced with empty pWPI-BLR vectors were used for the production of the control stable cell lines Huh7.5-C and IHH-C.

TNF-Cycloheximide Cytotoxicity Assay
TNF-cycloheximide (CHX) cytotoxicity assays were carried out on both Huh and IHH derived stable cell lines. Cells were plated at a 10 5 density in a 96-well plate and cultured for 24 h in Dulbecco's modified Eagle medium (Thermo Fisher Scientific, Waltham, MA, USA) and supplemented with non-essential amino acids, 2 mM L-glutamine (Thermo Fisher Scientific, Waltham, MA, USA), 100 µg/mL of penicillin/streptomycin (Thermo Fisher Scientific, USA), and 10% fetal bovine serum at 37 • C and 5% CO 2 . Cells were treated with 2 µg/mL CHX (Cell Signaling Technology, Danvers, MA, USA) and a range of human TNFα (Peprotech, Rocky Hill, NJ, USA) concentrations and were incubated at 37 • C and 5% CO 2 for 18 h. MTT assay was carried out, as previously described [31].

UV Irradiation Assays
UV irradiation assays were carried out on IHH derived stable cell lines. Cells were plated at a 10 5 density in a 96-well plate and irradiated using UV-C at 1 J/m 2 s for a range of time periods in a UV crosslinker (Stratagene, Santa Clara, CA, USA). Cells were subsequently cultured for 24 h in Dulbecco's modified Eagle medium (Thermo Fisher Scientific, USA) and supplemented with non-essential amino acids, 2 mM L-glutamine (Thermo Fisher Scientific), 100 µg/mL of penicillin/streptomycin (Thermo Fisher Scientific, USA), and 10% fetal bovine serum at 37 • C and 5% CO 2 . MTT assay was carried out as previously described [31].

Transgenic Mice
HCV core 4a and core 4f genes were amplified by PCR from pCI/core-4aR, pCI/core-4fC plasmid (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6060129/) (accessed on 12 April 2022) and placed downstream of a transthyretin (TTR) liver-specific promoter [32]. Core 4a and core 4f genes was inserted into the StuI site of the pTTR1-ExV3 plasmid, and the transgenes were prepared by purifying the HindIII fragment containing TTR promoter and the core 4a/core 4f coding sequence. The TTR promoter was kindly provided by Dr. Iannis Talianidis, Institute of Molecular Biology and Biotechnology of FORTH in Crete, Heraklion, Greece. Transgenic CBA-C57BL/6 mice, which express liver-specifically core 4a (TTRcore4a) and core 4f (TTRcore4f) proteins, were generated in the Transgenesis Facility of the Biomedical Sciences Research Center "Alexander Fleming", Vari, Greece. Mice were maintained under specific pathogen-free (SPF) conditions. Western blot using anti-GAPDH (6C5, Abcam, Cambridge, MA, USA) and anti-HCV core (C7-50, Abcam, Cambridge, MA, USA) antibodies was conducted according to the manufacturer's general guidelines.

Murine Model of Acute Liver Failure
Eight-week-old male mice were administered Lipopolysaccharide/D-galactosamine (LPS/Dgal), and their acute liver failure was plotted in a Kaplan-Meier survival curve, as described previously [33]. Statistical comparison was conducted by the Log-rank (Mantel-Cox) test using default parameters in GraphPad Prism 8.02 (Dotmatics, Boston, MA, USA).

RNA Sequencing and Bioinformatics Analysis
Total RNA was extracted using Trizol (Thermo Fisher Scientific, Waltham, MA, USA) from biopsies of the large liver lobe of TTR-core4a, TTR-core4f transgenic mice and their littermate wild-type (wt) controls. Total RNA was sequenced using the 3 RNA sequencing (3 RNAseq) protocol (Lexogen, Vienna, Wien, Austria) on an Ion Proton sequencer at the Genomics Facility of the Biomedical Sciences Research Center "Alexander Fleming", Vari, Greece. Raw 3 RNAseq reads were aligned on the Mus_musculus.GRCm38.96 reference murine transcriptome using salmon v1.6.0 [34] and its quasi-mapping capabilities. Read counts were imported into a custom R script using tximport v 1.20.0 [35] and provided as input for differential gene expression analysis (DGEA) via DESeq2 v1.32.0 [36]. DESeq2's plotPCA function was also used to create the PCA plot. Gene annotation mappings from Ensembl mus musculus gene identifiers to murine and human entrez gene identifiers were performed using the org.Mm.eg.db [37], org.Hs.eg.db [38], and R packages. Overrepresentation pathway analysis and visualizations were performed via the Reactome [39] database using the clusterProfiler v4.0.5 [40] R package on the genes which were found differentially expressed with a fold regulation of ±2 and adjusted -p < 0.01. clusterProfiler was also used for the creation of the gene-concept networks linking Reactome pathways to their respective genes (cnetplots). Finally, EnhancedVolcano v1.10.0 [41] was used in the creation of the volcano plot for the differentially expressed genes from 3 RNAseq. Datasets are available at BSRC "Alexander Fleming" genomic repository (https://genomics-lab.fleming. gr/fleming/external/Karakasiliotis/run342/metaseqr_quantseq_run342a/index.html) (accessed on 20 November 2020).

Real-Time RT-PCR
Quantitative real-time reverse transcription (RT) polymerase chain reaction (PCR) was conducted using cDNA from RNA extracted from liver biopsies. Two µg of RNA were used as a template for reverse transcription by murine leukemia virus reverse transcriptase  (Table S1). The conditions used for cycling were the following: 40 cycles of 95 • C for 10 s, 60 • C for 20 s, and 72 • C for 15 s. Relative mRNA expression was calculated using the ∆∆Ct method [42] and GAPDH mRNA as a normalizer. The two-tailed t-test was used for p-value calculation using default parameters in GraphPad Prism 8.02 (Dotmatics, Boston, MA, USA).

HCV Core Presents Differential Susseptibility to TNFα-Induced Death In Vitro
HCV core proteins from two clinical isolates, corresponding to subtypes 4a and 4f, previously isolated in Romania and Cameroon, were stably expressed in Huh7.5 and IHH cell lines [43]. The pWI-BLR vector was used for cloning the core 4a and core 4f genes and the respective lentivirus particles production ( Figure 1a). Core 4a and core 4f stably expressing Huh7.5 hepatoma cell lines (Figure 1b), treated with TNFα and cycloheximide, showed similar death responses when compared to control cells ( Figure 1c). As Huh7.5 cells may have deregulated death-related pathways, we used a non-cancerous cell line; the immortalized human hepatocytes (IHH). IHH cells were tested for their hepatic origin using the AT1A HNF4A mRNA markers as compared to Huh7.5 and 293T cells (Figure 1d). In IHH cells, the core 4a protein resulted in a significant reduction of cell viability at various concentrations of TNFα, while core 4f showed minimal effect on cell viability ( Figure 1e). Thus, core 4a and core 4f proteins not only presented differential modulation of TNFα pathway, but also this effect was cell line specific. To assess the specificity of this result, we tested the IHH stable cell lines in a different model of apoptosis: the UV-induced apoptosis. UV irradiation of IHH 4a and IHH 4f cells resulted in similar levels of cell death between cell lines and as compared to IHH control cells (Figure 1f). Similar levels of UV-induced death may signify that the apoptosis pathway where TNFα-induced apoptosis and UV-induced apoptosis converge is not differentially affected by the two types of core protein. induced apoptosis. UV irradiation of IHH 4a and IHH 4f cells resulted in similar leve cell death between cell lines and as compared to IHH control cells (Figure 1f). Sim levels of UV-induced death may signify that the apoptosis pathway where TNFα-ind apoptosis and UV-induced apoptosis converge is not differentially affected by the types of core protein.

HCV Core Expression Results in Differential Susseptibility to the LPS/Dgal Hepatic Fa Model
As previously described, the role of the HCV core protein in the TNFα pathwa various cell lines is controversial. Thus, we aimed for an in vivo animal model using tr genic mice that express HCV core proteins. A TTR-promoter-driven liver-specific ex sion of HCV core 4a and HCV core 4f proteins ( Figure 2a) allowed us to assess the in role of the two subtypes of core protein. TTRcore4a and TTRcore4f transgenic mice generated using pronuclear injection, and core protein expression was assessed u Western blot (Figure 2b), while HCV core mRNA expression was measured in var tissues to verify tissue specificity ( Figure S2). The transgene for both lines was inhe at the expected 1:1 ratio ( Figure S3).
A well studied in vivo mouse model for liver pathology is LPS/Dgal acute liver ure. The model is based on TNF-induced death of hepatocytes after LPS administra and D-galactosamine hepatocyte death sensitization. TTRcore4a mice showed incre susceptibility to LPS/Dgal treatment as compared to littermate control (Figure 2c). result was verified using a second founder line bearing the same transgene. On the o hand, TTRcore4f mice showed similar susceptibility to LPS/Dgal treatment as comp

HCV Core Expression Results in Differential Susseptibility to the LPS/Dgal Hepatic Failure Model
As previously described, the role of the HCV core protein in the TNFα pathway in various cell lines is controversial. Thus, we aimed for an in vivo animal model using transgenic mice that express HCV core proteins. A TTR-promoter-driven liver-specific expression of HCV core 4a and HCV core 4f proteins (Figure 2a) allowed us to assess the in vivo role of the two subtypes of core protein. TTRcore4a and TTRcore4f transgenic mice were generated using pronuclear injection, and core protein expression was assessed using Western blot (Figure 2b), while HCV core mRNA expression was measured in various tissues to verify tissue specificity ( Figure S2). The transgene for both lines was inherited at the expected 1:1 ratio ( Figure S3).
A well studied in vivo mouse model for liver pathology is LPS/Dgal acute liver failure. The model is based on TNF-induced death of hepatocytes after LPS administration and D-galactosamine hepatocyte death sensitization. TTRcore4a mice showed increased susceptibility to LPS/Dgal treatment as compared to littermate control (Figure 2c). The result was verified using a second founder line bearing the same transgene. On the other hand, TTRcore4f mice showed similar susceptibility to LPS/Dgal treatment as compared to littermate control. Although, at later timepoints, a non-significant increase in death was observed (Figure 2c). The result was verified using a second founder line bearing the same transgene. to littermate control. Although, at later timepoints, a non-significant increase in death observed (Figure 2c). The result was verified using a second founder line bearing the transgene.

Subtype-Specific Modulation of Molecular Pathways by HCV Core In Vivo
The difference in the response to LPS/Dgal between TTRcore4a and TTRcore4f t genic mice was investigated through differential transcriptome analysis using 3′RNA Differential transcriptome analysis of TTRcore4a (Table 1) and TTRcore4f (Table 2), pared to their littermate controls, yielded a small number of upregulated and down lated genes using a cut-off of FDR < 0.05 and fold change > 2 or < −2. The resulted d entially expressed genes did not fall into a statistically significant gene set after gen enrichment analysis (www.gsea-msigdb.org, accessed on 12 April 2022). However, vidual genes that have been in the past implicated in liver pathology were identified bles 1 and 2). These mRNAs were quantified using real-time RT-PCR validating the informatics analysis (Figure 3). SAA1 and SAA2 mRNAs were significantly upregu in TTRcore4a mice compared to their littermate wt controls, while ADAMDEC1, MT2, HAMP2, and GSN mRNAs were downregulated. SAA2 mRNA was signific upregulated in TTRcore4f mice compared to their littermate wt controls, while no mR significant to liver pathology were found downregulated.
To further explore the pathogenetic mechanism, we further explored the differe expression dataset for genes regulated downstream of SAA1, SAA2, ADAMDEC1, MT2, and GSN or related to cell death and liver proliferation. Because SAA1 overex sion has shown to increase various cytokines [44], we assessed the expression of C CCL3, CCL4, CCL7, and CCL11. The mRNA expression for 5 cytokines was found s icantly upregulated in TTRcore4a mice ( Figure 3). Accordingly, the relative abundan a specific T-cell marker IL2RG was significantly increased in TTRcore4a mouse livers ure 3). While proinflammatory cytokines were upregulated, we assessed the expre of TNFα related mRNAs, such as TNFα, TNFR1, TNFR2, and cFLIP. TNFα, TNFR1

Subtype-Specific Modulation of Molecular Pathways by HCV Core In Vivo
The difference in the response to LPS/Dgal between TTRcore4a and TTRcore4f transgenic mice was investigated through differential transcriptome analysis using 3 RNAseq. Differential transcriptome analysis of TTRcore4a (Table 1) and TTRcore4f  (Table 2), compared to their littermate controls, yielded a small number of upregulated and downregulated genes using a cut-off of FDR < 0.05 and fold change >2 or <−2. The resulted differentially expressed genes did not fall into a statistically significant gene set after gene set enrichment analysis (www.gsea-msigdb.org, accessed on 12 April 2022). However, individual genes that have been in the past implicated in liver pathology were identified (Tables 1 and 2). These mRNAs were quantified using real-time RT-PCR validating the bioinformatics analysis ( Figure 3). SAA1 and SAA2 mRNAs were significantly upregulated in TTRcore4a mice compared to their littermate wt controls, while ADAMDEC1, MT1, MT2, HAMP2, and GSN mRNAs were downregulated. SAA2 mRNA was significantly upregulated in TTRcore4f mice compared to their littermate wt controls, while no mRNAs significant to liver pathology were found downregulated.
To further explore the pathogenetic mechanism, we further explored the differential expression dataset for genes regulated downstream of SAA1, SAA2, ADAMDEC1, MT1, MT2, and GSN or related to cell death and liver proliferation. Because SAA1 overexpression has shown to increase various cytokines [44], we assessed the expression of CCL2, CCL3, CCL4, CCL7, and CCL11. The mRNA expression for 5 cytokines was found significantly upregulated in TTRcore4a mice ( Figure 3). Accordingly, the relative abundance of a specific T-cell marker IL2RG was significantly increased in TTRcore4a mouse livers (Figure 3). While proinflammatory cytokines were upregulated, we assessed the expression of TNFα related mRNAs, such as TNFα, TNFR1, TNFR2, and cFLIP. TNFα, TNFR1, and TNFR2 mRNAs were expressed at similar levels ( Figure 3) in both mouse lines, while cFLIP was overexpressed in TTRcore4f mice, reflecting a possible anti-apoptotic potential.

Discussion
Recent advances in direct acting antiviral development have revolutionized the therapy of HCV infection. However, a sustained virological response in a significant proportion of patients does not reverse liver pathology [45,46]. Pathological mechanisms in liver fibrosis and cirrhosis in HCV patients [47,48], and in vivo models [49], have mainly highlighted the deregulation of inflammatory and metabolic processes. The HCV core protein is central in the host-pathogen interaction network in HCV pathology, involved in inflammatory and metabolic aspects. The HCV core protein interacts with cellular proto-oncogenes and changes their expression patterns, resulting to hepatocarcinogenesis [50,51]. In vitro, HCV core protein activated interleukin 6 (IL-6) and interleukin 8 (IL-8) through TLR2 while activating NF-κB in parallel [52,53]. Stimulation of naïve human macrophages with HCV core induces TNF-α and IL-6 production through the TLR2 pathway [10]. HCV core binding to transcription factor signal transducer and activator of transcription 1 (STAT1) appears to inhibit the activation of innate immunity through the type I interferon pathway [11]. Downstream of p53, HCV core interacts with p21WAF that negatively regulates the kinase/cyclin system, consequently accelerating the cell cycle [12,13]. The development of transgenic mice that expressed the HCV core protein in a liver-specific manner led to steatosis, a clinical feature that is characteristic of patients with chronic HCV infection [54,55]. These mice, at a later stage in their lives, spontaneously presented hepatocellular carcinoma [55][56][57]. Moreover, crossing HCV core transgenic mice with PPARα −/− mice indicated that the HCV core protein is implicated in the activation of PPARα, a nuclear receptor that has an essential role in lipid homeostasis [57]. TNFα has a central role in liver inflammation [58] and metabolic [59] processes, and it constitutes a hallmark in liver fibrosis [60]. However, conflicting published results did not allow for a definitive role of HCV core protein in TNFα signaling. On one hand, the HCV core was found to interact with the TNFR1 receptor (HCV genotype 1b in Huh-7 cells) [22], to enhance apoptotic signaling through FADD (HCV genotype 1b in Huh-7 cells) [23], and to block the survival factor NF-κB (HCV genotype 2a JFH1 clone in Huh-7.5 cells) [20]. Accordingly, another report exemplified HCV core enhanced TRAIL signaling through increased activation of Bid and the mitochondrial branch of apoptosis (HCV genotype 1b in Huh-7 cells) [19]. In contrast, the HCV core was found to inhibit death by TNFα (HCV genotype 1a in MCF7 cells) [24,25] through the enhancement of c-FLIP (HCV genotype 1a in HepG2 cells) [26] and to inhibit Fas-induced apoptosis through enhancement of Bcl-xL (HCV genotype 1b J4L6S clone in HepG2 cells) [21]. Considering the impact of HCV genetic variability on HCV pathogenesis [29], it is likely that the observed discrepancies are actually the effect of polymorphisms within the core protein of various genotypes/strains used in the above studies.
In order to assess the role of HCV genetic variability on TNFα-induced death, we used core proteins from two subtypes of HCV genotype 4, namely, 4a and 4f. These two core type have been recently shown to differentially modulate the Wnt pathway in 293T and Huh7.5 cells [43]. The HCV 4f core protein enhanced Wnt signaling more than the HCV 4a core protein. Interestingly, an amino acid substitution (S71T) in core 4a was able to recapitulate the effect of core 4f, highlighting the role of the specific amino acid in Wnt pathway regulation [43]. The interplay of Wnt and TNFα signaling has been well documented [61][62][63][64]. HCV core 4a and core 4f proteins did not present differential susceptibility to TNFα-induced death in Huh7.5, the cell line that is often used for HCVrelated research. As Huh7.5 cells present a strong epithelial to mesenchymal transition (EMT) phenotype [65], and have accumulated several mutations regarding apoptosis related genes such as BAX, MAP3K1, and TP53 [66], we used an immortalized human hepatocyte cell line (IHH). The IHH core 4a protein induced higher susceptibility to TNFαinduced death, highlighting the importance not only of HCV genetic variability but, also, the importance of the cell line genetic background and phenotype.
Mouse models have been at the forefront of the understanding of the role of the HCV core protein in liver pathology, mainly dissecting the steatosis-hepatocarcinogenesis axis [49]. With the development of two novel core-protein-expressing mouse transgenic lines, TTRcore4a and TTRcore4f, we aimed for the recapitulation of TNFα susceptibility observed in IHH cells. The LPS/Dgal-liver-failure mouse model is well known to be dependent on TNFα. The model is based on TNFα production after LPS treatment and subsequent TNFα-induced apoptosis, as TNFα and tumor necrosis factor receptor 1 (TNFR1) knockout mice showed no liver pathology [67]. TTRcore4a mice showed enhanced susceptibility to the LPS/Dgal model, while TTRcore4f mice showed similar susceptibility to their littermate controls. The effect was similar to the effect presented in IHH cell lines stably expressing core 4a and core 4f. Thus, HCV subtype 4a and 4f core proteins seem to present a differential effect on liver pathology, further supporting the role of HCV core protein variability in a genotype-specific pathology. As the amino acid sequence of the two strains were different only in two positions, and the amino acid 191 is cleaved off in cells by signal peptidase [68], the presented effect may be attributed to position 71 that has been, in the past, implicated in various pathological mechanisms.
Analysis of total mRNA transcriptome in the TTRcore4a and TTRcore4f transgenic resulted in a specific small set of differentially expressed mRNAs. Although pathway analysis did not yield statistically significant gene groups, some key mRNAs in liver pathology were identified and further analyzed using real time RT-PCR. One of the most downregulated mRNAs in TTRcore4a livers was ADAM-like DECysin-1 (ADAMDEC1), which is a secreted metalloprotease [69] that is highly associated with the gastrointestinal tract. Adamdec1 −/− mice were more susceptible to the induction of bacterial and dextran sodium sulphate (DSS) induced colitis, and they presented increased bacteremia [70]. Gelsolin (Gsn) was one of the mRNAs found downregulated only in TTRcore4a mice. Gsn deficient mice presented increased susceptibility to the anti-Fas model of liver failure [71], a model that is based on similar signaling pathways to LPS/Dgal model, as Fas receptor belongs to the TNF receptor superfamily [72]. SAA1 mRNA, in contrast to SAA2 mRNA, was upregulated only in TTRcore4a mice. In previous reports, hepatic overexpression of SAA1 aggravated fatty liver inflammation by promoting intrahepatic platelet aggregation [73], while, recently, SAA1 was shown to exacerbate hepatic steatosis via the TLR4-mediated NF-κB signaling pathway [74]. SAA1 overexpression may lead to the induction of pro-inflammatory genes as acute response mediators during infection [75], and they aggravated T cell-mediated hepatitis in mice [44]. The mRNA of Mt1 and Mt2 metalotheionines were found downregulated in both transgenic mice and, thus, may not account for the differential response to LPS/Dgal, although they are highly associated with the response to HCV infection [76]. Finally, another mRNA associated with liver function was that of hepcidin (HAMP) and hepcidin 2 (HAMP2), which are mouse-specific paralogs of HAMP. HAMP was minimally affected in both transgenic mice, while HAMP2 cannot be linked to HCV infection, as it is mouse-specific. However, combined with the effect on ADAMDEC1 and metallothionines, expression may pinpoint a deregulation in metal metabolism in the core-expressing liver, a hallmark in liver pathology [77]. The differential expression analysis highlighted several mediators of pathogenesis that require validation in human biopsies and may shed light into hepatocyte death during HCV infection, directly associated with liver fibrosis and cirrhosis.
Further analysis of the presented phenotype signified differences in the inflammatory status of the livers of TTRcore4a mice. Thus, we further analyzed inflammation effectors and modulators, such as the cytokines shown before, to aggravate T cell-mediated hepatitis following SAA1 overexpression [44]. Indeed, we observed an increased expression of CCL2, CCL3, CCL7, and CCL11 in TTRcore4a mouse livers, while a statistically significant increase in the presence of IL2RG marker was also observed. Expression of TNFα mRNA and the mRNAs of its receptors TNFR1 and TNFR2 mRNAs remained the same between transgenic mice and their wild-type controls. This pro-inflammatory environment signifies a sensitized background towards liver inflammation that possibly aggravates the phenotype in the LPS/DGAL model, similarly to the T cell-mediated hepatitis model published previously [44]. As cFLIP is a direct link between the inflammatory status, TNF-induced apoptosis [78,79], and the HCV core protein pathogenesis [26], we assessed the role of HCV core in the expression of cFLIP in vivo. Intriguingly, cFLIP was significantly induced in the livers of TTRcore4f mice, possibly contributing to the observed difference between TTRcore4a and TTRcore4f mice in the LPS/DGAL model.
Overall, in this study, we identified an aggravating/sensitizing role of HCV core protein in TNFα-induced apoptosis in both in vitro and in vivo models. This role was determined by the genetic variability of the HCV core protein and the background of the mammalian cells used and, thus, our study offers an alternative point of view to the conflicting reports of current bibliography on TNFα-induced apoptosis. Our findings should be validated in patients and infectious systems. Despite the fact that overexpression models have shed light onto core protein pathogenesis, they still remain artificial systems [49,80].
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/v14112521/s1, Figure S1: Alignment of HCV 4a and HCV 4f core protein; Figure S2: Expression of core mRNA in various tissues; Figure S3: Characterization of mouse lines; Table S1: Oligonucleotide primers for real time RT-PCR used for the verification of the differentially expressed genes in transgenic mice.  Data Availability Statement: Total 3 RNA sequencing datasets are available at the BSRC "Alexander Fleming" genomic repository (https://genomics-lab.fleming.gr/fleming/external/Karakasiliotis/ run342/metaseqr_quantseq_run342a/index.html) (accessed on 20 November 2020).