Rapid reversal of innate immune dysregulation in blood of patients and livers of humanized mice with HCV following DAA therapy

Chronic hepatitis C virus (HCV) infection results in sustained immune activation in both the periphery and hepatic tissue. HCV infection induces innate immune signaling that is responsible for recognition of dsRNA, leading to activation of transcription factors and production of Type I and III IFNs, as well as pro-inflammatory cytokines and chemokines. Continued activation of host-immune mediated inflammation is thought to contribute to pathologic changes that result in progressive hepatic fibrosis. The current standard treatment for chronic HCV infection is directly-acting antivirals (DAAs), which have provided the unique opportunity to determine whether successful, rapid treatment-induced eradication of viral RNA normalizes the dysregulated antiviral innate immune response in patients chronically infected with HCV. Results First, in patients receiving two different combinations of DAAs, we found that DAAs induced not only rapid viral clearance, but also a re-setting of antiviral immune responses in the peripheral blood. Specifically, we see a rapid decline in the expression of genes associated with chronic IFN stimulation (IFIT3, USP18, IFIT1) as well as a rapid decline in genes associated with inflammation (IL1β, CXCL10, CXCL11) in the peripheral blood that precedes the complete removal of virus from the blood. Interestingly, this rapid reversal of innate immune activation was not seen in patients who successfully clear chronic HCV infection using IFN-based therapy. Next, using a novel humanized mouse model (Fah-/-RAG2-/-IL2rgnull—FRG), we assessed the changes that occur in the hepatic tissue following DAA treatment. DAA-mediated rapid HCV clearance resulted in blunting of the expression of proinflammatory responses while functionally restoring the RIG-I/MAVS axis in the liver of humanized mice. Conclusions Collectively, our data demonstrate that the rapid viral clearance following treatment with DAAs results in the rebalancing of innate antiviral response in both the peripheral blood and the liver as well as enhanced antiviral signaling within previously infected hepatocytes.


Results
First, in patients receiving two different combinations of DAAs, we found that DAAs induced not only rapid viral clearance, but also a re-setting of antiviral immune responses in the peripheral blood. Specifically, we see a rapid decline in the expression of genes associated with chronic IFN stimulation (IFIT3, USP18, IFIT1) as well as a rapid decline in genes associated with inflammation (IL1β, CXCL10, CXCL11) in the peripheral blood that precedes the complete removal of virus from the blood. Interestingly, this rapid reversal of innate immune activation was not seen in patients who successfully clear chronic HCV infection using IFNbased therapy. Next, using a novel humanized mouse model (Fah -/-RAG2 -/-IL2rg null -FRG), we assessed the changes that occur in the hepatic tissue following DAA treatment. DAAmediated rapid HCV clearance resulted in blunting of the expression of proinflammatory responses while functionally restoring the RIG-I/MAVS axis in the liver of humanized mice. PLOS  Analysis of NFkB response. Patient PBMCs harvested prior to, and subsequent to DAA treatment were lysed with radioimmunoprecipitation assay buffer (50 mM Tris pH 7.4, 150 mM NaCl, 1% (v/v) Triton X-100, 0.5% (w/v) Na-deoxycholate, 250 nM okadaic acid, 1:100 Phosphatase Inhibitor Cocktail Set II (Calbiochem, USA), 1:100 Protease Inhibitor Cocktail P8340 (Sigma, USA)) and 14 μg total protein was resolved on a 4-20% Criterion™ TGX™ polyacrylamide gel (BioRad, USA). Following electrophoresis, proteins were transferred onto nitrocellulose membranes and blocked overnight at 4˚C in Odyssey Blocking Buffer (TBS) (LI-COR, USA). Antibodies used for analysis were mouse anti-NF-κB p65 (L8F6) (Cell Signaling Technology, USA), rabbit anti-NF-kB p65 (phospho S536) (EP2294Y) (Abcam, USA), and mouse anti-Actin (clone C4) (Millipore, USA), with secondary antibodies peroxidase-conjugated goat anti-mouse IgG, donkey anti-rabbit IgG, and Alexa Fluor 790-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch, USA). Pixel intensity of imaged immunoblot bands were quantified using NIH ImageJ software, and activation of the NF-kB pathway was inferred by dividing the value of phospho-NF-kB p65 by the value of total NF-kB p65 for each patient sample. To reduce variance between repeat experiments on different membranes, the ratio derived from samples in the first lane of each gel were arbitrarily set to one and all remaining samples from that experiment were scaled to match. Statistical analysis was performed using a two-tailed Wilcoxon matched pairs test on Prism 5 software (Graph Pad).
Mice. FRG mice were obtained from Yecuris (Tualatin, OR) and maintenance of human liver chimerism was performed as described [6,7]. Degree of human hepatocyte engraftment was measured using a human albumin ELISA (Bethyl Laboratories, Montgomery, Tx). All mice utilized in experiments had >90% engraftment of human hepatocytes. For infection, mice were intravenously administered serum from a patient chronically infected with HCV geneotype 1a. The infectious dose of HCV was 1.53x10 6 IU. For DAA treatment mice were given Asunaprevir (0.3 mg daily), Declatasvir (0.5mg BID), and Sofosbuvir (2mg daily) or DMSO as a control via oral gavage. During treatment infectious load was measured in the peripheral blood every 24 hours via craniofacial bleed. Mice were monitored daily for signs of distress and no mice had to be euthanized prior to completion of the study. Following a 14-day course of treatment mice were euthanized with CO 2 followed by cervical dislocation and livers were fixed in formalin for 30min at 25˚C and embedded in paraffin. This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All experiments were approved by the institutional animal care and use committee (IACUC) at the University of Colorado Anschutz medical campus.

Immunohistochemistry and confocal microscopy
For sections of paraffin-embedded liver biopsies, we used the following procedure involving 4 major steps, including deparaffinization of tissue, antigen retrieval, tissue permeabilization, and immunostaining. Slide-mounted paraffin-embedded biopsy tissue was deparaffinized by first placing slides on a 58˚C heat block for 5 minutes in order to preheat them to the working temperature. Preheated slides were then placed in a glass Coplin jar containing 1x EZ Dewax solution (Biogenex) and were incubated for 5 minutes at room temperature. Slides were then placed in xylene and incubated for 5 minutes at room temperature followed by sequential 5 minute incubations in a Coplin jar each containing 100% ethanol, 95% ethanol, 70% ethanol, and distilled ultrapure water. Slides were placed in IFA buffer (PBS containing 0.05% TWEEN-20) for 5 minutes, and were then subjected to antigen retrieval. For antigen retrieval of deparaffinized slide-mounted tissue, slides were placed in a plastic Coplin jar containing 1x AR10 solution (Biogenex), and a vented lid was placed on the jar. The jar was then set inside a microwave oven and incubated with an oven setting on high while constantly monitoring through the oven door to define the first point of boiling of liquid within the jar. The oven power was turned off when the first point of boiling was observed, the liquid removed from the jar and the jar re-filled with 1x AR10 solution maintained at 4˚C. The jar (containing the slides) was then allowed to cool for 10 minutes at room temperature. This cycle of heating and cooling was repeated, after which the jar was allowed to cool at room temperature for 30 minutes. Slides were then removed from the jar and rinsed sequentially with distilled ultrapure water and 1X PBS. Tissues were permeabilized by placing slides in a Coplin jar containing PBS/1.0% TritonX-100 for 5 minutes at 25˚C. Slides were placed in 1X PBS for 5 minutes then air dried. For immunostaining, slides were first incubated in PBS containing 10% normal goat serum for 1 hour. Primary antibody was applied and the slides were incubated for one hour at 25˚C. The slides were rinsed three times 5 minutes each with IFA buffer followed by application of secondary antibody coupled to Alexa488 or Alexa 594 fluorophores (Molecular Probes). After a 1 hour incubation at 25˚C, the slides were rinsed three times with IFA buffer (PBS containing 0.05% TWEEN-20), allowed to dry, and cover slips were mounted using Prolong Gold mounting medium (ThermoFisher Scientific) for 24 hours before imaging. Tissues were imaged using a Nikon Eclipse TE2000 inverted microscope with the Nikon C1 laser scanning confocal module utilizing a 10 mW Argon laser emitting light at 488 nm wavelength, a 1 mW HeNe laser emitting light at 543 nm wavelength, and a 5 mW HeNe laser emitting light at 633 nm wavelength. Digital images were collected as 0.

Introduction
Chronic infection with hepatitis C Virus (HCV) results in chronic upregulation of many genes associated with innate immune activation [8]. This chronic stimulation of the innate immune response and subsequent activation of hepatic stellate cells are the primary driver of liver inflammation and ultimately cirrhosis [9]. HCV stimulates the innate immune response through multiple mechanisms including activation of TLRs 2,3,4,7/8, and 9 [10]. Therefore, HCV infection can induce activation of virtually every human innate immune cell in both the liver and the peripheral blood, which results in the increased expression of many genes and proteins associated with inflammation and consequent liver fibrosis. For example, chronic HCV infection results in the accumulation of active IL1β in the serum of patients [11]. Additionally, IL1β is associated with the activation of HSCs [12,13], ultimately leading to liver fibrosis. This chronic inflammation caused by IL1β as well as other factors has established that chronic activation of the innate immune response by HCV is critical for the progression to end-stage liver disease. Previous standard of care therapy with IFNα/ribavirin utilized the innate immune response to rid the body of HCV infection in part by boosting the innate immune compartment. However, this treatment regimen had limited success [14], immunomodulatory effects on the adaptive immune system which are critical for viral clearance, and well-documented and significant side effects [15][16][17]. The recent advances in therapy for HCV, via DAAs, has allowed for the dissection of how rapid HCV clearance alters innate immune activation and ultimately liver disease progression in the absence of the immunomodulatory effects of IFN.
In this report, we document how rapid clearance of HCV resets innate immune activation via DAA therapy, how treatment with DAA therapy differs from IFNα/ribavirin therapy in its effect on innate immune activation, and finally how DAA therapy alters innate immune activation in the hepatic compartment of humanized mice. Our observations define the molecular changes in the peripheral blood that are associated with successful clearance of HCV with novel IFN-free DAA therapies.

DAA treatment results in rapid downregulation of the innate immune signature in the peripheral blood of chronically infected patients
We performed microarray mRNA expression profiling in PBMCs of patients who received a three-drug regimen of DAAs: daclatasvir (NS5A inhibitor), asunaprevir (NS3 protease inhibitor), and beclabuvir (NS5B inhibitor). The median age was 55 years old; 88.9% of patients were men. The median pre-treatment viral level was 488,859 IU/ML and the median fibrotest score was 0.47; all patients experienced an SVR12 (S1A Table). DAA-mediated cure of chronic HCV infection resulted in significant transcriptional changes; specifically, when analyzing the top genes that are either increased or decreased in the peripheral blood at the end of DAA therapy, we observe a re-setting of innate immune antiviral responses and an increase in adaptive immune responses in the peripheral blood (Table 1). These transcriptional changes were confirmed using RT-PCR (S1 Fig). Using IPA-ingenuity software (Qiagen), we found that in addition to direct transcriptional changes, DAA therapy results in the predicted reduction in activation state of several upstream regulators of innate immune signaling such as Toll-like receptors (TLRs), chemokines, IFN-α2, and interleukin-1 (Table 2). Thus, after examination of innate immune signaling we observed significant dampening of innate immune responses both in gene expression levels (e.g., IL1β) and predicted activation state (e.g., TLR3) ( Table 2 and Fig 1A).
To confirm these transcriptional changes and determine whether these were unique to the DAA regimen used or a common transcriptional signature associated with DAA-mediated rapid HCV clearance, we performed RT-PCR on PBMC from the cohort above (S1A Table) and a cohort treated with ledipasvir-sofosbuvir (S1B Table). We found remarkably similar transcriptional changes regardless of treatment regimen. In both IFN-free treatment regimens, we found statistically significant decreases in C-X-C motif chemokine (CXCL)-10 and CXCL11 and reduced expression of the innate immune signaling molecules RIG-I, EIF2AK2, STAT1 and IRF7 ( Fig 1B). Therefore, the rapid dampening of innate immune activation following rapid viral clearance with IFN-free DAA therapy is independent of the treatment regimen utilized. In addition to dampening of the transcriptional activation of innate immune genes in the peripheral blood of patients treated with DAA therapy, we observed a statistically significant reduction in the phosphorylation of the NFκB p65 subunit protein in the peripheral blood following DAA therapy (Fig 1C and 1D). This result is consistent with the predicted downregulation of NFκBIA activity revealed in Table 2. Thus, rapid viral clearance with DAA therapy results in a rapid reduction in innate immune activation at both the transcriptional and post-translational level in the peripheral blood.
Next, we examined the kinetics of these transcriptional changes relative to viral decline in patients receiving DAAs. Using this longitudinal analysis we found that both CXCL10 and CXCL11 mirrored the rapid decline of HCV in the peripheral blood. Moreover, we found the expression of genes directly associated with innate immune sensing of IFN, such as ISG56 (IFIT1) and USP18, were rapidly decreased (Fig 2). In contrast, changes in RIG-I lagged behind and IRF3 expression did not decrease, suggesting distinct patterns of innate immune gene expression following DAA therapy. Interestingly, the expression of IL1β over the course of DAA therapy in patient cohort 2 was highly variable (perhaps related to different frequencies on monocytes), and thus at a global level, we did not detect a significant decrease in IL1β transcription in this cohort. These findings contrast with the significant decrease in IL1β transcription in the peripheral blood of patient cohort 1 by both microarray (Tables 1 and 2) and RT-PCR (S1 Fig). These data suggest that while both regimens of DAAs used in this study have the same clinical endpoint (cure of HCV), regimens may have slightly differing effects on the transcriptional signature in the peripheral blood. Taken together, these findings confirm and expand other published data [18], demonstrating IFN-free therapy can rapidly reverse the activation of some, but not all aspects of the innate immune response in the peripheral blood.

Differential mRNA expression between DAA and IFN/ribavirin therapy
To determine which transcriptional changes are strictly associated with DAA therapy versus changes that merely are associated with viral cure, we utilized a published cohort of individuals treated with IFNα/ribavirin. DAA treatment directly interferes with the HCV replication cycle without altering the immune responses, in contrast to the use of IFN/ribavirin which has known immunomodulatory effects [10]. Therefore, we hypothesized that DAA therapy in the absence of IFN/ribavirin results in rapid reversal of HCV-induced chronic innate immune activation. To address the effects of these different regimens on innate immune activation, we directly compared our microarray data to the published microarray results of the Virahep-C cohort [19]. To control for potential differences in the rate of viral clearance between DAA therapy and IFN/ribavirin we focused our analysis on the seven patients with a rapid and sustained virologic responses. The Virahep-C time points for which microarray data was available was at pre-treatment and at 10 weeks of therapy (Fig 3A). When comparing the top 1000 genes  that were significantly changed from pretreatment in both data sets, we found that only 32 (3.2%) genes were significantly different in both data sets with only 20 (62.5%) of those genes having a shared transcriptional change (Fig 3A). Comparison of these patients with end of DAA treatment patients (week 12) revealed many striking differences. Specifically, the dampening of the antiviral immune response seen with DAA therapy is not evident at 10 weeks of IFN/ribavirin therapy-even though these patients had cleared the virus ( Fig 3B). As expected, a majority of the transcriptional changes between IFN-free DAA therapy and IFN/Ribavirin therapy are either directly or indirectly related to IFN treatment. To further document how the transcriptional changes following DAA therapy are represented in the peripheral blood of patients treated with IFNα/Ribavirin, we interrogated the 520 genes that were statistically different (p<0.01) from end of treatment to pretreatment in the DAA cohort. From this gene set we found 214 genes that were also measured in the IFNα/Ribavirin data set (S2 Fig). Using this methodology, we were able to validate the similarities and differences in gene expression between DAA and IFNα/Ribavirin therapy (S2A Fig). Additionally, when examining significant differences between these genes, we find that a vast majority of genes that are statistically different following DAA therapy are not statistically changed in the IFNα/Ribavirin data set (S2B Fig). Next, we compared the innate immune gene expression in the peripheral blood of our cohort of IFN-free DAA treated individuals twelve weeks after completion of therapy to the signature of IFN/Ribavirin-treated individuals twenty four weeks after the end of treatment because the Virahep-C patients for whom microarray data were available were still receiving of the ratio of pS-NFκB to NFκB (right) prior to treatment (circles) and at twelve weeks following the end of treatment (squares). Lines represent individual patient samples. *P = 0.03, Wilcoxon matched-pairs signed rank test. N = 11 patients.
https://doi.org/10.1371/journal.pone.0186213.g001  IFN/Ribavirin which affects gene expression. This latter analysis allowed for the interrogation of differences in transcriptional signature between DAA therapy and IFN-containing regimens in the absence of direct effects of IFN therapy on transcriptional signature as well as the durability of these differences. Surprisingly, in contrast to the sustained dampening of innate immune signaling seen with DAA therapy we did not see reduction in genes such as CXCL10, STAT1 and RIG-I long after the end of IFN/Ribavirin treatment and clearance of the virus (Fig 3C). Taken together, these data suggest that treatment with DAAs results in the rapid and prolonged dampening of antiviral innate signaling while treatment with IFN/Ribavirin can still be associated with persistent activation of innate immune responses even long after clearance of virus.

IFN-free DAA therapy results in the rapid reduction of innate immune transcriptional signature in the livers of humanized mice
In order to determine how reflective peripheral blood changes are relative to hepatic gene expression, we utilized the human liver-chimeric FRG mice [20]. Briefly, in this mouse strain, the selection pressure for transplanted hepatocytes is accentuated due to absence of the enzyme fumaryl acetoacetate hydrolase (FAH), which leads to an accumulation of toxic tyrosine catabolites within mouse hepatocytes. This genetically determined toxicity is preventable by oral administration of 2-(2-nitro-4-trifluoro-methylbenzoyl)-1,3-cyclohexanedione (NTBC), which blocks hydroxyphenylpyruvate dioxygenase activity upstream of FAH and therefore prevents the accumulation of hepatotoxic metabolites [21]. After withdrawal of NTBC, engrafted human hepatocytes repopulate the mouse liver; serum albumin is a reliable correlate of human hepatocyte repopulation [20], and in our mice, we saw high circulating levels of human albumin. Following reconstitution of the mouse liver with human hepatocytes we established stable HCV infection in these mice following intravenous injection of serum from a patient with HCV genotype 1a infection (Fig 4A). The mice were then treated with either vehicle control (DMSO) or a cocktail of daclatasvir, asunaprevir, and sofosbuvir. This treatment regimen induced suppression of circulating HCV within 72 hours to levels below 42 IU/mL ( Fig 4B). As noted in the peripheral blood of patients receiving DAAs, there was a significant decrease in hepatic expression of CXCL10 and CXCL11, but no significant difference in RIG-I transcription (S3 Fig). We and others have shown that innate recognition of HCV in hepatocytes through dsRNA sensors (such as RIG-I) results in recruitment of mitochondrial antiviral signaling (MAVS, also known as CARDIF/IPS-1/VISA) and TRAF3 (Tumor Necrosis Factor receptor-associated factor-3) [22]. Once sufficient viral proteins have accumulated in the cytosol, HCV uses its multifunctional NS3/4A protease, essential for HCV replication, to cleave MAVS and ablate RIG-I mediated innate immune signaling [23]. MAVS is actively suppressed by HCV infection [24]. Within 14 days of triple DAA treatment, there was increased protein expression of MAVS and an increase in the protein IFITM1, a tight junction protein only induced by IFN, that inhibits HCV entry [25] (Fig 4C and S4 Fig). These findings are important because it may suggest that DAAs additionally block HCV spread among hepatocytes. A recent study of patients with chronic HCV receiving sofosbuvir-ribavirin demonstrated that SVR was associated with a decrease in hepatic expression of multiple IFN genes [18]. In contrast to that human study we found that within 14 days DAA treatment in this humanized model there are no significant changes in the expression of IFN genes, possibly due to the residual viral protein in the liver after 14 days of DAA therapy (Fig 4C) or due to the fact that these mice lack innate immune cells known to produce IFNs. Despite these differences, the transcriptional changes we document in the peripheral blood of patients receiving DAA therapy partially reflect changes that are present in the hepatic compartment of humanized mice (S3 Fig).

Discussion
Our longitudinal analyses demonstrate that IFN-free DAA regimens in patients with chronic HCV infection result in global resetting of innate immune signaling and inflammatory pathways. Specifically, we observe a significant decrease in the transcription of the cytokine IL1β, a cytokine that is associated with innate immune activation, liver inflammation, and progression to fibrosis. Furthermore, we demonstrate a decrease in the phosphorylation of the NFκB protein which is directly associated with innate immune activation and downstream signaling. These data suggest that unlike treatment with IFNα/ribavirin, DAA therapy rapidly decreases the expression of genes associated with innate immune activation and liver damage in the DAA therapy rapidly reverses HCV-induced innate immune activation peripheral blood. In support of this notion, several early studies have demonstrated that patients treated with DAA therapy have either a halt in the progression of liver fibrosis or a reversal in liver fibrosis score [26]. Our analysis of the kinetics of PBMCs transcriptional signature highlight the rapid downregulation of CXCL10, CXCL11, IFIT1, and USP18 but delayed or no significant changes in other innate immune genes such as RIGI, IFITM1, and IRF3. These differential patterns in gene expression following DAA therapy may be related to disengagement of specific HCV components and downregulation of key innate immune evasion strategies engaged by HCV to establish chronic infection [27]. It has been demonstrated that the expression level of CXCL10 in the peripheral blood during treatment with IFNα/ribavirin is associated with response to therapy [28]. In our study we find that all patients treated with DAA therapy rapidly decrease the expression of CXCL10 in the peripheral blood, suggesting that CXCL10 is not a predictor of response to DAA therapy but, rather, a marker of reestablishment of immune homeostasis following clearance of chronic HCV infection. Further, CXCL10 and CXCL11 are chemokines that are associated with recruitment of innate immune cells to the site of inflammation and liver inflammation, thus the rapid reduction in these chemokines suggest a rapid decrease in hepatic inflammation and possibly a resetting of the cellular composition in the peripheral blood. These findings corroborate our and other findings that during DAA therapy there is a rapid normalization in both the innate and adaptive immune cell phenotypes in the peripheral blood [2,29,30]. Because the expected outcome of current DAA treatment is SVR, we did not include patients who experienced viral relapse, which could be considered a relative limitation of the study and therefore we are unable to determine if there are any molecular characteristics that are associated with DAA non-responsiveness. However, future studies with larger cohorts focused on individuals who fail to respond to front-line DAA therapies will address these shortcomings. Transcriptional changes associated with DAA therapy versus changes that are associated with IFN-based viral cure were identified. Using this approach, we were able to highlight several key differences in the transcriptional profile in the peripheral blood between patients treated with DAAs and those treated with IFNα/ribavirin. We found that despite the rapid clearance of HCV in both cohorts we observed patients treated with IFNα/ribavirin maintained a transcriptional signature associated with innate immune activation, suggesting that DAA therapies uniquely result in the rapid return to innate immune homeostasis. One potential limitation with this analysis is that IFN treatment is well known to modulate the transcriptional levels of several genes identified in our analysis. To mitigate the direct effect of IFN we examined the transcriptional changes associated with IFN-free and IFN-containing regimens well after the end of treatment. In this analysis we found that the transcriptional differences between IFN-free and IFN-containing treatment regimens were maintained long after removal of treatment as demonstrated by persistent, elevated expression of CXCL10 and CXCL11 twenty-four weeks after removal of IFNα/ribavirin therapy. These data suggest that resolution of inflammation in the liver associated with chronic HCV infection is much more rapid in patients treated with DAA therapy than in patients treated with IFNα/ribavirin and this may be ultimately reflected in regression of liver fibrosis following HCV cure. While our transcriptional data only addressed changes in the peripheral blood of human patients, we were able to utilize a humanized mouse model to establish chronic HCV infection and also treat the infection with DAA therapy. Using this strategy, we were able to demonstrate that the transcriptional changes that occurred in the peripheral blood of patients treated with DAAs may partially, but not entirely, reflect intrahepatic transcriptional changes. Specifically, we see that the rapid reduction in the expression of CXCL10 and CXCL11 in the peripheral blood is recapitulated in the liver of humanized mice treated with a regimen of DAAs. A significant limitation with these murine studies is that residual murine hepatocytes (and non-parenchymal cells) are present in the liver, making it a human-mouse chimera. In this study we utilized mice that had >90% of the liver reconstituted with human hepatocytes, but despite this high degree of chimerism, we were unable to completely examine the transcriptional response of human hepatocytes to HCV infection or DAA therapy due to contamination with murine transcripts. Specifically, we performed RNAseq analysis on the liver of HCV-infected mice treated with DAA therapy or DMSO control. However, contamination with murine transcripts precluded accurate analysis of the data. To counteract these problems, we selected transcripts of interest in which the primers did not detect murine transcripts. Thus, our data does not provide the complete picture of transcriptional changes in the liver following DAA therapy, but we believe our findings correlate well with the published findings from the liver of patients treated with DAAs, and the differences observed utilizing our mouse model may also be a result of the incomplete clearance of HCV protein in the liver due because of the immune-incompetent state of these humanized mice. Thus, our findings demonstrate DAA therapy rapidly reduces inflammation in the peripheral blood and liver, a finding that is not associated with IFNα/ribavirin therapy. Taken together, our study highlights not only the important transcriptional pathways associated with HCV cure but also the molecular pathways that are associated with resolution of hepatic inflammation.
Supporting information S1  Paired t-test on both DAA and the GSE11342 IFN dataset using the 70 day timepoint on the former using those individuals in the IFN dataset that completely cleared the virus by day 70. All uncharacterized genes (without Ref Seq #) were eliminated from both gene lists, leaving 38,598 out of 53,617 genes for DAA and 21,174 out of 22,284 for GSE11342 (IFN dataset). A. X-axis is the mean difference (mean log2 expression Post-mean log2 expression Pre) of the DAA treatment, Y axis is the mean difference (mean log2 expression Post-mean log2 expression Pre) of the IFN dataset. The color code is the mean difference of the DAA samples minus the mean difference of the interferon samples (blue-grey would be more decreased in DAA compared to interferon, red is the opposite). The top 10 genes with the largest and smallest positive and negative values respectively for the mean difference of DAA-Interferon (this is the color code values) were labeled. B. Taking the 520 genes that were different between Post vs Pre-treatment with DAA at a P<0.01 (paired t-test) and merged any available data from the interferon Post vs Pre-samples for all common genes. This left 214 genes different in pre vs. post DAA treatment that were also measured in the interferon dataset.