View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Virology 430 (2012) 43–52

Contents lists available at SciVerse ScienceDirect

Virology

journal homepage: www.elsevier.com/locate/yviro

Chronic immune activation is a distinguishing feature of liver and PBMC signatures from HCV/HIV coinfected patients and may contribute to hepatic fibrogenesis

Angela L. Rasmussen a,1, I-Ming Wang b,1, Margaret C. Shuhart c,1, Sean C. Proll a, Yudong He b, Razvan Cristescu b, Chris Roberts b, Victoria S. Carter a, Christopher M. Williams a, Deborah L. Diamond a, Janine T. Bryan a, Roger Ulrich b,2, Marcus J. Korth a, Lisa V. Thomassen d, Michael G. Katze a,n

a Department of Microbiology, University of Washington, Box 358070, Seattle, WA 98195, USA b Informatics and Analysis, Merck Research Laboratories, 770 Sumneytown Pike, West Point, PA 19846, USA c Department of Medicine, University of Washington, Harborview Medical Center, Box 359773, Seattle, WA 98104, USA d Department of Pathology, University of Washington, 325 Ninth Avenue, Box 359791, Seattle, WA 98104-2499, USA

article info abstract

Article history: Hepatitis C virus/human immunodeficiency virus (HCV/HIV) coinfected patients demonstrate acceler- Received 22 February 2012 ated progression to severe liver injury in comparison to HCV monoinfected patients, although the Returned to author for revisions underlying mechanisms are unclear owing to infection of separate tissue compartments with two 6 April 2012 distinct viral pathogens. Microarray analysis of paired liver biopsy and peripheral blood mononuclear Accepted 17 April 2012 cell (PBMC) specimens from HCV/HIV coinfected and HCV monoinfected patients identified a gene Available online 16 May 2012 expression signature associated with increased inflammation and immune activation that was present Keywords: only in liver and PBMC samples from coinfected patients. We also identified in these samples liver- and Hepatitis C virus PBMC-specific signatures enriched with fibrogenic/hepatic stellate activation and proinflammatory Human immunodeficiency virus , respectively. Finally, Bayesian networks were constructed by assimilating these data with HCV/HIV coinfection existing data from liver and PBMC samples from other cohorts, augmenting enrichment of biologically Liver Peripheral blood important pathways and further indicating that chronic immune activation in HCV/HIV coinfection may Hepatic fibrosis exacerbate liver disease progression in coinfected patients. Systems biology & 2012 Elsevier Inc. All rights reserved. Gene expression Pathogenesis

Introduction has been dramatically decreasing. However, HCV/HIV-coinfected patients are increasingly experiencing morbidity and mortality Chronic HCV infection may lead to severe hepatic injury, associated with HCV-induced liver disease. including fibrosis, cirrhosis, and hepatocellular carcinoma (HCC). The mechanisms underlying rapid liver disease progression in Liver disease typically progresses slowly, with severe disease HCV/HIV co-infected patients are poorly understood and are developing over 10–30 years following HCV infection. However, likely multifactorial. Contributing factors which may accelerate human immunodeficiency virus (HIV) coinfection can profoundly liver disease progression include HIV-associated immune dys- impact the natural history of HCV infection. Several studies have function (Gonzalez et al., 2009b; Sandberg et al., 2010), oxidative demonstrated that HCV/HIV coinfected patients exhibit increased stress (Baum et al., 2011; Lin et al., 2011; Wanchu et al., 2009), hepatic inflammation (Di Martino et al., 2001; Martinez-Sierra apoptosis (Jang et al., 2011), enhanced HCV replication (Lin et al., et al., 2003; Pol et al., 1998), and progress more rapidly to severe 2008), viral modulation of cytokine/chemokine responses (Allison liver disease (Graham et al., 2001; Soto et al., 1997) in comparison et al., 2009), and HIV infection or activation of resident liver cells to HCV-monoinfected patients. Due to the effectiveness of highly such as hepatic stellate cells or Kupffer cells (Tuyama et al., 2010). active antiretroviral therapy (HAART), the incidence of HIV Unfortunately, no widely accepted experimental model of HCV/ progression to acquired immune deficiency syndrome (AIDS) HIV coinfection is currently available. Therefore, detailed studies of liver disease pathogenesis have been limited, and are restricted to clinical samples. n Corresponding author. Fax: þ1 206 732 6056. Previously, we analyzed liver biopsies from HCV mono- and E-mail address: [email protected] (M.G. Katze). HCV/HIV coinfected patients and identified gene expression 1 These authors contributed equally to this work. signatures indicative of dysregulation of apoptosis, increased 2 Present address: Gilead Sciences, Inc., 199 East Blaine Street, Seattle, WA 98102, USA. immune cell activation, and decreased innate antiviral responses

0042-6822/$ - see front matter & 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.virol.2012.04.011 44 A.L. Rasmussen et al. / Virology 430 (2012) 43–52 compared to normal liver reference samples (Walters et al., 2006). (92%) and demonstrated significantly lower CD4 T cell counts However, a specific intrahepatic signature distinguishing HCV/ (mean 451 versus 1120, p¼.0014) compared to the HCV mono- HIV coinfected from HCV monoinfected patients was not identi- infected group. Also, the majority of coinfected patients (77%) fied. In this study, we utilized core needle liver biopsies and a received anti-retroviral treatment, which likely contributed to the whole genome microarray with over twice the human transcrip- lack of detectable HIV RNA in 63% of the subjects. Both groups tome coverage of those used previously to identify intrahepatic were naı¨ve to HCV therapy. Seventy percent of the subjects gene signatures specifically associated with HCV/HIV. We also reported abstinence from alcohol use for at least one year prior explored previous observations that HCV mono- and HCV/HIV to enrollment. We also performed a correlation analysis to assess coinfection induce unique immunologic gene expression profiles which clinical parameters might confound our gene expression in peripheral mononuclear blood cells (PBMC) (Kottilil et al., results, and identified HIV status and consequent CD4 cell counts 2009). In addition to liver biopsy specimens, PBMCs from the as the only parameters significantly influencing gene expression same subjects were evaluated by microarray to investigate how (data not shown). It is likely that the correlation with CD4 T cell transcriptional programming of immune cells in the periphery counts reflects the lower counts observed in the HIV-positive might contribute to hepatic inflammation, cellular infiltration, subjects. Due to the limited sample size, a multivariate analysis and disease progression. was not performed. We specifically did not find any association Expression profiling of clinical specimens can prove challen- between gene expression and other HIV-specific covariates such ging due to numerous confounding variables, limitations on as serum HIV quantitation (suppressed or not) and the use of sample sizes, and heterogeneous subject characteristics. In an antiretroviral therapy, though we note that in this study, these attempt to mitigate some of these challenges, and to expand the patient subgroups are too small to perform a statistically robust scope of our functional analysis, we used our transcriptomic data correlation analysis. to construct Bayesian networks in conjunction with published liver (Lamb et al., 2011; Schadt et al., 2008; Zhong et al., 2010) Common transcriptional signatures in liver and PBMC samples from and PBMC gene expression data (Emilsson et al., 2008) obtained HCV/HIV coinfected patients from larger independent cohorts. These networks allowed us to identify additional genes in critical pathways functionally asso- Using a two-way ANOVA approach, a molecular signature ciated with HCV/HIV co-infection, enhancing our understanding common to liver and PBMC samples from coinfected patients was of the complex biology of liver disease progression in HCV/HIV identified. A total of 467 upregulated and 338 downregulated coinfected patients. differentially expressed genes (DEG; po.01) were identified in both samples (Fig. 1A; Table S1). Common differential expression in both liver and PBMC may indicate that these pathways are Results comparably regulated separately in both tissues, and/or that many PBMC have migrated into and substantially contribute to Subject characteristics the overall gene expression profile in liver. Functional analysis was performed on the 805 DEG common to both tissues using Demographic and clinical characteristics of the subjects are Ingenuity Pathway Analysis (IPA). Many of the top functional shown in Table 1. In total, 10 HCV monoinfected and 13 HCV/HIV categories among the upregulated genes were associated with coinfected patients were recruited in this study. The two groups inflammation, indicating that distinct mechanisms may drive were well-matched for age, ethnic background, body mass index progression of hepatic inflammation in coinfected patients. (BMI), duration of HCV infection, alcohol use, serum ALT levels, Among 84 upregulated genes associated with inflammatory and HCV genotype, HCV viral load, and Batts–Ludwig liver disease immunological responses and disease (Table S1), we observed stage. The HCV/HIV coinfected group was predominantly male DEG associated with components of complement, chemokines,

Table 1 Characteristics of HCV-infected subjects.

HIV-negative (n¼10) HIV-positive (n¼13) P value

Age 47.277.5 42.677.8 .18 Gender (% male) 60 92 .06 Race (%) .44 Caucasian 90 69 African American 10 23 Native American 0 7.7 HCV duration (years) 24.577.8 22.0711.0 .57 Alcohol use in lifetime (gm/d) (median, IQR) 24.1, 40.7 17.0, 46.9 .80 Alcohol use in last 6 months (gm/d) (median, IQR) 0, 0 0, .9 .45 ALT (U/L) 92750 68736 .23 HCV genotype (% 1) 80 77 .86 HCV RNA level (log 10) 6.27.5 6.371.0 .60 Antiretroviral therapy (%) NA 77 – HIV RNA level (log 10)a NA 4.57.7 – HIV RNA level, % undetectable (o400 copies/mL) NA 54 – CD4 cell count (cells/mL) 11207 471 4517292 .0014 HCV disease grade (0–4)b 2.57.7 1.87.8 .05 HCV disease stage (0–4)b 2.17.3 2.171.0 .9

Values expressed as mean7SD unless stated otherwise. NA, not applicable; IQR, interquartile range. a In those with detectable virus. b Batts–Ludwig scoring system. A.L. Rasmussen et al. / Virology 430 (2012) 43–52 45

patient id 124 131 133 138 HCV/HIV HCV HCV/HIV HCV 140 144 145 149 151 154 96 107 122 Liver 126 127 130 134 139 141 143 147 152 156 124 131 133 138 140 144 145 151 149 154 96 107 122 PBMC 126 127 130 134 139 141 143 147 152 156 Log (Ratio)

-0.190 0.19

Fig. 1. Common gene signature distinguishing HCV/HIV coinfected patients. (A) Heatmap of 805 significant differentially expressed genes in liver and PBMC from HCV/HIV coinfected patients compared to HCV monoinfected as determined by two-way ANOVA (po.01). (B) IPA network showing connected genes related to immune cell migration and inflammation from the common signature.

and antigen presentation and T cell activation (Fig. 1B). The Identification of hepatic signatures of coinfection presence of gene expression changes associated with immune activation and migration may indicate enhanced trafficking of We also sought to identify significant hepatic DEG character- activated immune effector cells from the periphery to the liver in istic of HCV/HIV coinfected patients. Using one-way ANOVA HCV/HIV coinfected patients. (po.05, fold change 41.15 in at least 7 samples), we identified 46 A.L. Rasmussen et al. / Virology 430 (2012) 43–52 transcriptional signatures in the coinfected patient cohort using from coinfected patients, over 250 genes related to infectious microarray data from liver samples only (Fig. 2A; Table S2). disease and immune responses were upregulated, including 67 Functional analysis by IPA demonstrated that in liver samples specifically related to HIV infection. Many of these genes are

patient id 124 131 133 138 HCV/HIV HCV HCV/HIV HCV 140 144 145 149 151 154 96 107

Liver 122 126 127 130 134 139 141 143 147 152 156 124 131 133 138 140 144 145 151 149 154 96 107

PBMC 122 126 127 130 134 139 141 143 147 152 156 Log (Ratio)

-0.190 0.19

Fig. 2. Compartmental gene signatures distinguishing HCV/HIV coinfected patients. (A) Heatmap of 3039 significantly differentially expressed genes in liver or PBMC identified by one-way ANOVA (po.05, fold change 41.15 in at least 7 patients) and separated by K-means clustering. (B) IPA network of intrahepatic signature showing directly connected molecules associated with inflammation and immunity. A.L. Rasmussen et al. / Virology 430 (2012) 43–52 47 functionally similar to those identified in the common signature, upregulation of the genes encoding costimulatory molecules and are associated with chemotaxis and cellular migration, CD86 and numerous HLA alleles. The molecular signature also including various chemokines, integrins (ITGAD, ITGA5, ITGA7, featured concomitant downregulation of genes known to sup- ITGB2), actin and tubulin cytoskeletal components (ACTB, ACTG, press T cell activation, such as cytotoxic T lymphocyte-associated TUBA1A, TUBA1C, TUBA4A, TUBG1), and multiple Ras-like homo- 4 (CTLA-4). These are consistent with the chronic activa- log members and related genes [RHOA, RHOB, RHOD, RHOG, tion of HIV-infected T cells. RHOQ, Ras-related C3 botulinum toxin 1, rho family small GTP Interestingly, we also identified a subset of transcripts asso- binding protein 1 (RAC1)] involved in the recruitment of circulat- ciated with fibrosis and extracellular matrix deposition upregu- ing immune cells to the liver. Many DEG associated with T cell lated in HCV/HIV coinfected PBMC, including several types of activation processes, including human leukocyte antigens collagen (COL22A1, COL3A1, COL4A1) and galectins (LGALS1, (HLA-DQA1, HLA-DQB1) that mediate antigen presentation, mole- LGALS3, LGALS8). Notably, both galectin-1 and galectin-3 have cules associated with dendritic cell (DC) maturation (CD209), and been implicated in HSC activation and hepatic fibrogenesis T cell receptor signaling machinery (CD3, CD8A) were upregu- (Henderson et al., 2006; Jiang et al., 2011; Maeda et al., 2003), lated. This suggests that the T cells infiltrating the liver in as have Rho GTPases and rearrangement of the actin cytoskeleton coinfected patients may be transcriptionally distinct from T cells (Kato et al., 1999). Circulating PBMC in HCV/HIV coinfected of HCV monoinfected patients, and may have a more activated patients thus display increased fibrogenic gene expression, and phenotype due to chronic HIV infection. The substantial overlap may subsequently enhance HSC activation and collagen deposi- (286 genes) between the hepatic signature and the signature tion upon immune infiltration of the liver. common to both liver and PBMC (identified using two-way ANOVA) supports an important contributing role for infiltrating Overexpression of REG1A and REG3A secreted immune cells within the livers of coinfected patients. We further observed the upregulation of genes functionally In addition to evaluating the global DEG signature associated linked to hepatic stellate cell (HSC) activation including the Rho with HCV/HIV coinfection, we sought to identify molecules highly family GTPases, actins, myosin chain 6 (MYL6), and cell expressed by the liver which may be candidates for evaluation division cycle 42 (CDC42), a major regulator of cytoskeletal as potential therapeutic or diagnostic targets. From DEG in the remodeling previously associated with chronic liver injury and intrahepatic signature, we observed that regenerating islet- tumorigenesis (van Hengel et al., 2008) and HCV-induced hepatic derived 1 alpha and 3 alpha (REG1A and REG3A) were among fibrosis (Smith et al., 2003). This network of molecules (Fig. 2B) the most highly upregulated in a subset of 7 patients. Reg1a and links activation of HSCs, inflammatory signaling, and immune cell Reg3a are both secreted growth factors associated with pancrea- infiltration. We also identified several secreted factors that con- tic islet cell regeneration, which were previously linked to a nect activated macrophages and Kupffer cells to HSC activation, number of gastrointestinal cancers, including hepatocellular car- including platelet-derived growth factor receptor, alpha polypep- cinoma (Cavard et al., 2006; Yuan et al., 2005). Reg1a and Reg3a tide (PDGFRA) and vascular endothelial growth factor (VEGF), and accelerate liver regeneration (Lieu et al., 2005; Wang et al., 2009), the chemokines CCL2, CXCL3, CCL13, and CCL5. Additionally, and Reg3a has been implicated in the activation of pancreatic genes known to be expressed by activated HSCs, such as Bcl-2 stellate cells, as well as deposition of collagen and associated associated X (BAX), which are also related to chemoattraction and extracellular matrix components (Li et al., 2009), suggesting that apoptotic induction were upregulated. This contrasts with the these molecules may be linked to liver disease progression. downregulation of genes associated with inhibition of apoptosis Therefore, we evaluated whether or not these molecules might (such as baculovirus IAP repeat-containing 5 [BIRC5]) and cell have potential as prognostic indicators of HCV/HIV coinfection- cycle progression (such as p21-activated kinase 7 [PAK7] and associated liver disease. kinesin family member 11 [KIF11]) and is consistent with a report By assessing Reg1a and Reg3a levels in patient plasma speci- that HSC apoptosis is anticorrelated with HCV-induced fibrogen- mens by immunoblot (Fig. 3), we observed that HCV/HIV patients esis (Gonzalez et al., 2009a). Taken together, these findings exhibited substantially higher levels of circulating Reg1a species suggest that in HCV/HIV coinfected patients, increased migration compared to monoinfected individuals. In particular, a higher of activated immune effector cells likely contributes to increased molecular weight form of Reg1a corresponding to either a precursor hepatic inflammation. This in turn may lead to increased HSC activation, hastening collagen deposition and fibrogenesis and HCV monoinfected HIV/HCV coinfected possibly contributing to more rapid liver disease progression in HCV/HIV coinfected patients. Patient ID 124 131 133 138 140 154 096 126 127 134 141 147 Reg1α 20 kDa

Identification of peripheral signatures of coinfection Reg1α 14 kDa In the PBMC compartment, functional analysis using IPA revealed that many of the upregulated genes were associated with immune induction and immunological disease. Specifically, Reg3α 14 kDa 271 genes associated with inflammatory responses, antigen pre- sentation, and humoral immune responses, as well as hematolo- gical, immunological, and inflammatory disease were significantly Fig. 3. Expression of Reg1a and Reg3a proteins. Immunoblot analysis of Reg1a and Reg3a protein in patient plasma specimens. Molecular weight of the detected upregulated (Table S3). Genes within this group include chemo- Reg1a and Reg3a proteins are indicated on the left side of the figure. Reg1a is kine receptors and other molecules associated with chemotaxis detected in the 14–22 kDa range, consistent with differential proteolytic proces- and lymphocyte homing including the HIV receptor CCR5, com- sing and/or glycosylation heterogeneity observed as a result of proteolytic plement components, and proinflammatory cytokines associated processing and differential glycosylation than monoinfected patients. Addition- ally, a higher molecular weight band at approximately 20 kDa represents either with T cell activation such as IL-11, IL-17RA, and IL-15. Addition- the precursor form of Reg1a or one of the other glycosylated isoforms of this ally, we observed numerous gene expression changes associated protein reported within the 17–22 kDa size range (De Reggi et al., 1995) present in with antigen presentation and CD4 T cell activation, including all samples. 48 A.L. Rasmussen et al. / Virology 430 (2012) 43–52 form or a heavily glycosylated protein, was consistently expressed at them one edge out to obtain a network with 1117 genes. higher levels. Secreted Reg3a levels were not significantly different Annotation of genes in this PBMC network indicates an associa- between the two patient conditions, although Reg3a was detectable tion with T and B cell signaling, interaction between dendritic and in all tested samples. This suggests that the higher molecular weight natural killer cells, and communication between innate and form of circulating Reg1a may be a candidate for evaluating hepatic adaptive immune cells (Table S4). These results are consistent disease progression by a non-invasive, blood-based study, and with findings shown in Fig. 1B, and suggest that PBMC from warrants further evaluation. coinfected patients exhibit the differential regulation of genes bridging innate and adaptive immunity. Expanding co-infected liver and PBMC signatures by a network approach Discussion The cross-sectional design of this study restricted detection of liver and PBMC signature DEG to one time point. To identify Understanding disease pathogenesis in HCV/HIV coinfected additional genes which may contribute to liver disease progres- patients presents numerous challenges, one of which is discerning sion in the coinfected tissues, we adopted a Bayesian network how distinct tissue compartments infected with two unique modeling approach to predict interactions between significant pathogens contribute separately and in concert to liver disease DEG in our study and those in networks built from previously progression. Our data suggest that the chronic immune activation published large data sets. Bayesian networks represent statistical characteristic of HIV infection and progression to AIDS (Deeks dependencies between genes (‘‘nodes’’) as ‘‘edges,’’ which are et al., 2004; Hazenberg et al., 2003) is a distinguishing feature of calculated by multiple expression measurements. Leveraging the host response to HCV/HIV coinfection. We observed substan- networks generated from much larger datasets can increase the tial overlap (104 DEG, 24 of which are associated with immune robustness of biological findings from a smaller dataset by further activation and differentiation) between both the individual liver exploring the interrelated connectivity of DEG found in lower and PBMC compartments and the signature common to both abundance. tissues. This further supports the notion that increased immune Combined replicates of liver signature sequences were used as and inflammatory activity are present in both the liver and the the input to Bayesian networks built from three human liver periphery of coinfected patients and are likely important drivers cohorts including 400 normal samples (Schadt et al., 2008), 272 of disease pathogenesis. The enhanced expression of inflamma- samples each of hepatocellular carcinoma (HCC) and adjacent tory genes in both the common and individual compartment normal livers (Lamb et al., 2011), and 950 livers from obese signatures suggests that HIV infection causes both circulating subjects (Zhong et al., 2010). Out of the 2442 liver signature leukocytes, and those infiltrating the liver, to be more activated sequences, 924 genes were present in the three liver networks but than those from HCV monoinfected patients. Consequently, the only those subnetworks with at least 3 nodes (290 genes) were pro-inflammatory functions of these cells may enhance patho- maintained. In subnetworks meeting these criteria, we expanded genic processes and progression of fibrosis and HCV-induced liver along one edge to identify additional connected nodes and obtain injury via immune effector functions, cytokine and chemokine the network shown in Fig. 4A with 2544 genes. secretion, increased migration, and activation of HSC and other We performed functional analysis of this liver network by cellular mediators of fibrogenesis. using the Gene Annotator Module of the Merck Target Gene Another effect of HIV-infected infiltrating immune cells on Information (TGI) Network Analysis and Visualization (NAV) tool, HCV-infected hepatic cells may be to stimulate expression of which finds related gene sets based on functional annotation, and genes with as yet undescribed roles in liver disease progression. ranks significance by calculating an expectation (E) value. Our The growth factors Reg1a and Reg3a have not been associated analysis identified highly enriched pathways involved in chemo- with HCV or HIV infection or pathogenesis. However, their kine signaling, dendritic cell maturation, interaction between apparent role in HCC tumorigenesis and metastasis (Cavard innate and adaptive immunity, T cell receptor signaling, pattern et al., 2006; Yuan et al., 2005), as well as our data indicating that recognition receptor-related pathway, NK cell signaling, and these proteins are secreted, merits further evaluation of these hepatic cholestasis/fibrosis (Table S4). This suggests that the proteins in a larger patient cohort as serum biomarkers with genes added by the network analysis are related to cross-talk prognostic or diagnostic utility. between the innate and adaptive immune system, and the inter- Recent systems biology approaches incorporating high-through- face between these different immune pathways may play an put data types, such as genome-wide DNA genotyping and RNA important role in HCV/HIV coinfection. expression profiling, has allowed the construction of various types of We identified similar functional categories and pathways using genetic networks. Characterizing networks that underlie complex IPA. Some of these overlap with those already identified in Fig. 1B phenotypes such as clinical disease can provide more comprehensive whereas others are novel pathways found to be associated with understanding of the disease pathophysiology and in turn identify or this liver network. Using IPA, we generated a subnetwork using validate disease susceptibility genes (Chen et al., 2008; Derry et al., the genes annotated as involved in communication between 2010; Emilsson et al., 2008; Schadt et al., 2008; Yang et al., 2009). In innate and adaptive immune cells (Fig. 4B). This subnetwork thecurrentstudy,thegenesignaturesweobtainedwerefroma includes additional genes related to induction of innate antiviral single time point (at which disease progression could be hetero- responses (multiple Toll-like receptors; TLR) and inflammation geneous) and from a relatively small number of patients. In an effort (tumor necrosis factor alpha; TNFa), as well as induction of to address these constraints and expand potential biological cover- adaptive immunity via antigen presentation (HLA alleles) and T age, we used coinfected liver and PBMC gene signatures to build cell activation (CD3, CD8). Bayesian statistical causal networks based on published large liver We built a PBMC network model using the similar approach of and blood data sets (Emilsson et al., 2008; Lamb et al., 2011; Schadt inputting 2479 combined replicate signature sequences to the et al., 2008; Zhong et al., 2010). From these networks, we obtained deCODE blood Bayesian network built from about 1000 samples greater enrichment based on functional annotation for the resulting (Emilsson et al., 2008). We applied similar criteria as the liver liver network (Fig. 4AandTable 2) and the PBMC network (Fig. 4B network, maintaining only 131 genes (out of 561 present in the and Table 2) compared to our functional analysis using the signifi- deCODE network) in sub-networks with at least 3 nodes and grew cant DEG identified in this study alone. These expanded networks A.L. Rasmussen et al. / Virology 430 (2012) 43–52 49

T Cell Activation

Antigen Presentation

Inflammation TLR Signaling

Chemotaxis

Fig. 4. Construction of HCV/HIV co-infected liver Bayesian network. (A) Liver network was constructed by overlaying 2442 liver signature sequences onto the network formed from 3 existing liver cohorts as described in the Materials and Methods section. Only those sub-networks with at least 3 connected nodes (290 genes) were retained to grow one edge out to obtain the network shown with 2544 genes. (B) Liver subnetwork of genes annotated as those involved in communication between innate and adaptive immune cells. Solid lines show directly interacting molecules, dashed lines show indirect interactions. Upregulated genes identified in the common signature in this study are colored yellow; genes identified in the network model are colored green. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article).

solidified the pathway connections between genes in both the HCV/ NK cells and DCs mediate both effector and regulatory functions HIV coinfected liver and PBMC molecular signatures, and provided in innate immunity and the induction of adaptive responses (Altfeld greater insight into functionally related genes that might also be et al., 2011; Gonzalez et al., 2010; Kim and Chung, 2009); however, significant for liver disease pathogenesis in a larger population of the HCV/HIV coinfected liver signature did not capture these HCV/HIV coinfected patients. functions as significantly associated with coinfection (Fig. 1B). The 50 A.L. Rasmussen et al. / Virology 430 (2012) 43–52

Table 2 Pathways and genes identified for expanded networks.

TGI NAV pathways Expectation Set Genes identified in network

Liver network Chemokine signaling 1.587E16 189 CCL4;CXCR4;GNB5;PRKCB;AKT2;PRKX;CCR1;CXCR3;CCL14;GNG11;PLCB2;CCL11;PIK3R1; pathway CCL4L1;GRK5;FOXO3;STAT3;CCL2;CRK;RAC2;PIK3CB;ELMO1;CXCL1;IL8;CCL19;CCL5; DOCK2;CCL15;CXCR2;ADCY7;ITK;CXCL16;PIK3CG;PIK3CA;CDC42;CXCR6;PRKACB;CCL23; CXCL6;ROCK1;PIK3R5;CCL4L2;HCK;CCR6;GNB1;VAV3;GNB4;MAP2K1;CXCL3;CCL8;CXCL12; GNG2;CCL3;ARRB1;PREX1;GRK4;CCL13;SHC1;CXCR1;ADCY1 Dendritic cell maturation 4.411E14 148 IL6;CD86;REL;CD247;CD83;COL1A2;FCER1G;AKT2;HLA-DRB1;STAT4;TLR2;LTBR;CREB5; TRGV9;PIK3R1;FCGR1A;FCGR2B;HLA-DMB;TLR4;IL18;TYROBP;RAC2;PIK3CB;TNFRSF11B; HLA-DQB1;FCGR3B;HLA-DMA;B2M;HLA-DRB5;PIK3CG;PIK3CA;CD3E;RELB;NFKBIE; HLA-DRA;FCGR3A;TNF;HLA-B;PIK3C2G;PIK3R5;FCGR1B;CD3D;TNFRSF1B; CD58;IL1B;LY75;HLA-DRB3;FCGR2A;IRF8 Communication between 6.31E12 68 IL6;CD86;CCL4;CD247;CD83;FCER1G;HLA-DRB1;TLR2;TRGV9;TLR4;IL18;CD4;IL8;CCL5; innate and adaptive CCL15;HLA-DRB5;CD8A;CD3E;TLR1;HLA-DRA;TNF;CD3D;TLR8;IL1B;CCL3; immune cells HLA-DRB3;TNFSF13B;CD8B;IFNG;CD28 iCOS-iCOSL signaling in T 2.962E11 100 INPP5D;REL;CD247;IL2RB;NFAT5;FCER1G;AKT2;PRKCQ;HLA-DPB1;TRGV9;PIK3R1; helper cells HLA-DMB;CD4;RAC2;LCP2;PIK3CB;IL2RA;HLA-DMA;ITK;HLA-DRB5;PIK3CG;PIK3CA; CD3E;RELB;NFKBIE;PPP3R1;TRAT1;PIK3C2G;PIK3R5;PTPRC;CD3D;ICOS;LCK; HLA-DRB3;SHC1;CD28 Natural killer cell 6.527E08 103 INPP5D;CD247;PRKCB;FCER1G;AKT2;PRKCQ; signaling SYK;PIK3R1;FYN;LILRB1;KLRK1;TYROBP;RAC2; LCP2;PIK3CB;CD244;FCGR3B;PIK3CG;PIK3CA;SH2D1A;FCGR3A;KLRB1;PIK3C2G;PIK3R5; VAV3;LAIR1;MAP2K1;SIGLEC7;LCK;SHC1;CD300A;HCST

PBMC network Altered T Cell and B Cell .002 80 CD86;FCER1G;TNFSF11;TLR2;FAS;CCL21;CD40LG;HLA-DMA;IL23A;LTB;TLR5;IL1B;IFNG;CD28 signaling in rheumatoid arthritis Crosstalk between .002 92 CD86;CCR7;KIR3DL2;FAS;KLRK1;TYROBP;CD40LG;TLN1;LTB;TNFSF10;CD209;TREM2; dendritic cells and PRF1;IFNG;CD28 natural killer cells Communication between .008 68 CD86;CCR7;FCER1G;TLR2;CD40LG;CCL5;CD8A;TLR5;IL1B;CD8B;IFNG;CD28 innate and adaptive immune cells Role of macrophages .014 294 PRSS35;LRP1;HP;PRKCH;FN1;TNFSF11;TLR2;CALM3;CREB5;PDGFD;TCF7;SFRP4;NFATC2; fibroblasts and CCL5;LTB;PRKCA;MYC;IL6ST;KRAS;GNAQ;CAMK4;IL32;TLR5;OSM;IL1B;C5AR1;MAP2K3;LEF1 endothelial cells in rheumatoid arthritis

expanded liver network, however, did identify significant associa- accelerated liver disease progression observed in coinfected tions with NK and DC functions (Table 2). Innate immunity- patients. Future studies employing longitudinal samples from a associated genes present in the network, but not in the co-infected larger patient cohort should be directed at understanding how liver signature, include TLR1, TLR2, TLR4, and TLR8. TLR signaling these genes are temporally regulated throughout the course of liver provides a critical link between innate and adaptive immunity in disease progression. Additionally, follow up studies should deter- numerous viral infections. In chronic HCV infection, reduced TLR mine the clinical utility of potential biomarkers such as Reg1a to responsiveness is associated with liver dysfunction (Chung et al., identify patients at risk of accelerated severe liver injury. 2011) and chronic inflammation consequent to reduced induction of adaptive immune responses (Dental et al., 2012; Yonkers et al., 2011b). Furthermore, aberrant TLR signaling has been implicated Materials and methods in T cell activation defects in HIV infection (Yonkers et al., 2011a), indicating that these molecules may be important mediators of Study subjects and specimen collection chronic immune activation and inflammation in the HCV/HIV coinfection setting. In addition, the liver network (Table 2)includes Approval for this study was obtained from the University of some key fibrogenic genes such as transforming growth factor-b1,2, Washington Human Subjects Division, and experiments were and 3 (TGFB1, TGFB2 and TGFB3), which have been implicated both performed in compliance with legal, institutional, and ethical in Rho GTPase-mediated HSC activation (Shimada et al., 2011)and standards for human subjects research. All samples were obtained enhancement of HCV replication and associated liver injury (Lin under informed consent. et al., 2008).This network allowed us to bridge gaps present in the Ten subjects infected with HCV and 13 subjects coinfected molecular signatures from our dataset, and establish a more with HCV and HIV were included in this study. Exclusion criteria comprehensive expression profile characteristic of liver disease included chronic hepatitis B infection, alcohol use 41 drink per pathogenesis in HCV/HIV coinfection. day in women and 42 drinks per day in men, and acute infection The observed host transcriptional signatures and networks of any type. An experienced hepatopathologist (L.V.T.) blinded to characteristic of liver and/or PBMC in HCV/HIV coinfected patients all clinical data including HIV status reviewed all liver biopsies. indicate that these compartments are distinguished by increased Liver fibrosis stage was assessed using the Batts–Ludwig system: immune activation consistent with HIV infection. This environ- F0, no fibrosis; F1, portal fibrosis; F2, periportal fibrosis with rare ment may in turn contribute to increased hepatic upregulation of bridges; F3, bridging/septal fibrosis; or F4, cirrhosis (Batts and pro-fibrogenic genes, and reveals a possible mechanism for the Ludwig, 1995). All samples included were deemed adequate for A.L. Rasmussen et al. / Virology 430 (2012) 43–52 51 staging. At the time of liver biopsy, 2–3 mm of liver tissue was Liver and PBMC network construction stored in RNALater (Life Technologies, Mountain View, CA) at 4 1C overnight, and then stored at 80 1C until RNA processing. Whole Networks were constructed as described previously (Zhu et al., blood was obtained by venipuncture within 30 day of the liver 2004,2008) and were visualized using the Target Gene Informa- biopsy (typically within one week of the biopsy). Peripheral blood tion (TGI) Network Analysis and Visualization (NAV) desktop mononuclear cells (PBMCs) were isolated by Ficoll density cen- application developed at Rosetta Inpharmatics. We utilized pre- trifugation (Ficoll-Paque Plus; GE Healthcare Biosciences, Piscat- viously published data sets for liver (Lamb et al., 2011; Schadt away, NJ) within 24 h of blood draw, and 2 106 freshly isolated et al., 2008; Zhong et al., 2010) and PBMC (Emilsson et al., 2008) PBMC were pelleted and resuspended in Trizol (Life Technologies, network constructions. To construct liver and PBMC networks Carlsbad, CA) prior to flash freezing in liquid nitrogen. Samples based on HCV/HIV coinfected gene signatures, the two gene were stored at 80 1C until RNA was prepared according to the signatures were loaded into the database as described in the manufacturer’s instructions. Samples were extracted according to results section and the TGI NAV tool was used to extract all edges manufacturer’s protocol and resuspended in RNAse-free water from this network involving the included signature genes of prior to use in microarray experiments. interest. The TGI NAV tool allows rapid, real-time, graphical analysis of pathway network models built from a comprehensive and fully Microarray platform integrated set of public and proprietary interaction databases available through a back-end central database. In addition, it Samples were amplified and labeled using a custom auto- supports experimentally generated systems biology data such as mated version of the RT/IVT protocol and reagents provided by the statistical associations and causal relationships described in Affymetrix. Hybridization, labeling and scanning were completed the manuscripts cited above and applied in the current study. following the manufacturer’s recommendations using Merck Affymetrix Human Custom Arrays 1.0 containing 43,737 unique probe sequences (Affymetrix, Santa Clara, CA). Sample amplifica- Acknowledgments tion, labeling, and microarray processing were performed by the Rosetta Inpharmatics Gene Expression Laboratory in Seattle, The authors gratefully acknowledge David Gretch and Minjun WA (acquired by Merck and Co., Whitehouse Station, NJ). Expres- Chung Apodaca for accessing patient plasma samples for immu- sion data were loaded into the Resolver (Rosetta proprietary noblot analysis. This work was funded in part by National software database) for transformation, normalization, and error Institute on Drug Abuse grant 1P30DA01562501. modeling.

Appendix A. Supporting information Microarray data analysis Supplementary data associated with this article can be found in Microarray data from each individual liver or PBMC sample the online version at http://dx.doi.org/10.1016/j.virol.2012.04.011. was compared back to a reference pool composed of all corre- sponding samples of liver or PBMC, respectively, from HCV monoinfected subjects. HCV/HIV-coinfected gene sets were estab- References lished based on their statistically significant up- or down-regula- tion (absolute fold change Z1.15 and ANOVA or combined p- Allison, R.D., Katsounas, A., Koziol, D.E., Kleiner, D.E., Alter, H.J., Lempicki, R.A., value r01 in at least 7 samples) when compared to the reference Wood, B., Yang, J., Fullmer, B., Cortez, K.J., Polis, M.A., Kottilil, S., 2009. pool. Signature genes from each tissue were identified and Association of interleukin-15-induced peripheral immune activation with hepatic stellate cell activation in persons coinfected with hepatitis C virus compared. Unsupervised clustering methods, including hierarch- and HIV. J. Infect. Dis. 200 (4), 619–623. ical and k-means clustering, were performed to identify genes Altfeld, M., Fadda, L., Frleta, D., Bhardwaj, N., 2011. DCs and NK cells: critical with similar expression patterns. Ingenuity Pathway Analysis effectors in the immune response to HIV-1. Nat. Rev. Immunol. 11 (3), 176–186. (IPA; Ingenuity Systems, Redwood City, CA) software was used Batts, K.P., Ludwig, J., 1995. Chronic hepatitis. an update on terminology and for functional analysis of significant genes identified by ANOVA. reporting. Am. J. Surg. Pathol. 19 (12), 1409–1417. Baum, M.K., Sales, S., Jayaweera, D.T., Lai, S., Bradwin, G., Rafie, C., Page, J.B., Campa, A., 2011. Coinfection with hepatitis C virus, oxidative stress and antioxidant status in HIV-positive drug users in Miami. HIV Med. 12 (2), 78–86. Immunoblotting Cavard, C., Terris, B., Grimber, G., Christa, L., Audard, V., Radenen-Bussiere, B., Simon, M.T., Renard, C.A., Buendia, M.A., Perret, C., 2006. Overexpression of regenerating islet-derived 1 alpha and 3 alpha genes in human primary liver Patient plasma samples were diluted in M-PER (Thermo tumors with beta-catenin mutations. Oncogene 25 (4), 599–608. Scientific, Rockford, IL) and protein concentration was quantified Chen, Y., Zhu, J., Lum, P.Y., Yang, X., Pinto, S., MacNeil, D.J., Zhang, C., Lamb, J., by BCA assay (Thermo Scientific) per the manufacturer’s instruc- Edwards, S., Sieberts, S.K., Leonardson, A., Castellini, L.W., Wang, S., Champy, M.F., Zhang, B., Emilsson, V., Doss, S., Ghazalpour, A., Horvath, S., Drake, T.A., tions. A Criterion TGX Any kD Tris-glycine polyacrylamide gel Lusis, A.J., Schadt, E.E., 2008. Variations in DNA elucidate molecular networks (Bio-Rad, Hercules, CA) was loaded with 20 mg protein per that cause disease. Nature 452 (7186), 429–435. sample, run using standard SDS-PAGE methods, and transferred Chung, H., Watanabe, T., Kudo, M., Chiba, T., 2011. Correlation between hypo- to a PVDF membrane. Membranes were probed with anti-Reg1a responsiveness to Toll-like receptor ligands and liver dysfunction in patients with chronic hepatitis C virus infection. J. Viral Hepat. 18 (10), e561–e567. (Abcam, Cambridge, MA) and anti-Reg3a (Abcam) primary anti- De Reggi, M., Capon, C., Gharib, B., Wieruszeski, J.M., Michel, R., Fournet, B., 1995. bodies, followed by peroxidase-conjugated donkey anti-rabbit The glycan moiety of human pancreatic lithostathine. Structure characteriza- secondary (Jackson ImmunoResearch Labs, West Grove, PA), and tion and possible pathophysiological implications. Eur. J. Biochem. 230 (2), 503–510. detected using ECL Plus reagent (GE Healthcare Biosciences). Deeks, S.G., Kitchen, C.M., Liu, L., Guo, H., Gascon, R., Narvaez, A.B., Hunt, P., Martin, There is no protein reported to be constitutively secreted at J.N., Kahn, J.O., Levy, J., McGrath, M.S., Hecht, F.M., 2004. Immune activation set similar levels in plasma or serum samples which could act as a point during early HIV infection predicts subsequent CD4þT-cell changes independent of viral load. Blood 104 (4), 942–947. loading control, therefore protein loading per well was normal- Dental, C., Florentin, J., Aouar, B., Gondois-Rey, F., Durantel, D., Baumert, T.F., ized by loading equal quantities of protein per lane. Nunes, J.A., Olive, D., Hirsch, I., Stranska, R., 2012. Hepatitis C virus fails to 52 A.L. Rasmussen et al. / Virology 430 (2012) 43–52

activate NF-kappaB signaling in plasmacytoid dendritic cells. J. Virol. 86 (2), hepatic stellate cells by galectin-1 and galectin-3 through different intracel- 1090–1096. lular signaling pathways. J. Biol. Chem. 278 (21), 18938–18944. Derry, J.M., Zhong, H., Molony, C., MacNeil, D., Guhathakurta, D., Zhang, B., Martinez-Sierra, C., Arizcorreta, A., Diaz, F., Roldan, R., Martin-Herrera, L., Perez- Mudgett, J., Small, K., El Fertak, L., Guimond, A., Selloum, M., Zhao, W., Guzman, E., Giron-Gonzalez, J.A., 2003. Progression of chronic hepatitis C to Champy, M.F., Monassier, L., Vogt, T., Cully, D., Kasarskis, A., Schadt, E.E., liver fibrosis and cirrhosis in patients coinfected with hepatitis C virus and 2010. Identification of genes and networks driving cardiovascular and meta- human immunodeficiency virus. Clin. Infect. Dis. 36 (4), 491–498. bolic phenotypes in a mouse F2 intercross. PLoS One 5 (12), e14319. Pol, S., Lamorthe, B., Thi, N.T., Thiers, V., Carnot, F., Zylberberg, H., Berthelot, P., Di Martino, V., Rufat, P., Boyer, N., Renard, P., Degos, F., Martinot-Peignoux, M., Brechot, C., Nalpas, B., 1998. Retrospective analysis of the impact of HIV Matheron, S., Le Moing, V., Vachon, F., Degott, C., Valla, D., Marcellin, P., 2001. infection and alcohol use on chronic hepatitis C in a large cohort of drug users. The influence of human immunodeficiency virus coinfection on chronic J. Hepatol. 28 (6), 945–950. hepatitis C in injection drug users: a long-term retrospective cohort study. Sandberg, J.K., Falconer, K., Gonzalez, V.D., 2010. Chronic immune activation in the Hepatology 34 (6), 1193–1199. T cell compartment of HCV/HIV-1 co-infected patients. Virulence 1 (3), Emilsson, V., Thorleifsson, G., Zhang, B., Leonardson, A.S., Zink, F., Zhu, J., Carlson, 177–179. S., Helgason, A., Walters, G.B., Gunnarsdottir, S., Mouy, M., Steinthorsdottir, V., Schadt, E.E., Molony, C., Chudin, E., Hao, K., Yang, X., Lum, P.Y., Kasarskis, A., Zhang, Eiriksdottir, G.H., Bjornsdottir, G., Reynisdottir, I., Gudbjartsson, D., Helgadottir, A., B., Wang, S., Suver, C., Zhu, J., Millstein, J., Sieberts, S., Lamb, J., GuhaThakurta, Jonasdottir, A., Styrkarsdottir, U., Gretarsdottir, S., Magnusson, K.P., Stefansson, H., D., Derry, J., Storey, J.D., Avila-Campillo, I., Kruger, M.J., Johnson, J.M., Rohl, C.A., Fossdal, R., Kristjansson, K., Gislason, H.G., Stefansson, T., Leifsson, B.G., van Nas, A., Mehrabian, M., Drake, T.A., Lusis, A.J., Smith, R.C., Guengerich, F.P., Thorsteinsdottir, U., Lamb, J.R., Gulcher, J.R., Reitman, M.L., Kong, A., Schadt, E.E., Strom, S.C., Schuetz, E., Rushmore, T.H., Ulrich, R., 2008. Mapping the genetic Stefansson, K., 2008. Genetics of gene expression and its effect on disease. Nature architecture of gene expression in human liver. PLoS Biol. 6 (5), e107. 452 (7186), 423–428. Shimada, H., Staten, N.R., Rajagopalan, L.E., 2011. TGF-beta1 mediated activation of Gonzalez, S.A., Fiel, M.I., Sauk, J., Canchis, P.W., Liu, R.C., Chiriboga, L., Yee, H.T., Rho kinase induces TGF-beta2 and endothelin-1 expression in human hepatic Jacobson, I.M., Talal, A.H., 2009a. Inverse association between hepatic stellate stellate cells. J. Hepatol. 54 (3), 521–528. cell apoptosis and fibrosis in chronic hepatitis C virus infection. J. Viral Hepat. Smith, M.W., Yue, Z.N., Korth, M.J., Do, H.A., Boix, L., Fausto, N., Bruix, J., Carithers 16 (2), 141–148. Jr., R.L., Katze, M.G., 2003. Hepatitis C virus and liver disease: global Gonzalez, V.D., Falconer, K., Blom, K.G., Reichard, O., Morn, B., Laursen, A.L., Weis, transcriptional profiling and identification of potential markers. Hepatology N., Alaeus, A., Sandberg, J.K., 2009b. High levels of chronic immune activation 38 (6), 1458–1467. in the T-cell compartments of patients coinfected with hepatitis C virus and Soto, B., Sanchez-Quijano, A., Rodrigo, L., del Olmo, J.A., Garcia-Bengoechea, M., human immunodeficiency virus type 1 and on highly active antiretroviral Hernandez-Quero, J., Rey, C., Abad, M.A., Rodriguez, M., Sales Gilabert, M., therapy are reverted by alpha interferon and ribavirin treatment. J. Virol. 83 Gonzalez, F., Miron, P., Caruz, A., Relimpio, F., Torronteras, R., Leal, M., Lissen, (21), 11407–11411. E., 1997. Human immunodeficiency virus infection modifies the natural Gonzalez, V.D., Landay, A.L., Sandberg, J.K., 2010. Innate immunity and chronic history of chronic parenterally-acquired hepatitis C with an unusually rapid immune activation in HCV/HIV-1 co-infection. Clin. Immunol. 135 (1), 12–25. progression to cirrhosis. J. Hepatol. 26 (1), 1–5. Graham, C.S., Baden, L.R., Yu, E., Mrus, J.M., Carnie, J., Heeren, T., Koziel, M.J., 2001. Tuyama, A.C., Hong, F., Saiman, Y., Wang, C., Ozkok, D., Mosoian, A., Chen, P., Chen, Influence of human immunodeficiency virus infection on the course of B.K., Klotman, M.E., Bansal, M.B., 2010. Human immunodeficiency virus (HIV)-1 hepatitis C virus infection: a meta-analysis. Clin. Infect. Dis. 33 (4), 562–569. infects human hepatic stellate cells and promotes collagen I and monocyte Hazenberg, M.D., Otto, S.A., van Benthem, B.H., Roos, M.T., Coutinho, R.A., Lange, chemoattractant protein-1 expression: implications for the pathogenesis of J.M., Hamann, D., Prins, M., Miedema, F., 2003. Persistent immune activation in HIV/hepatitis C virus-induced liver fibrosis. Hepatology 52 (2), 612–622. HIV-1 infection is associated with progression to AIDS. AIDS 17 (13), van Hengel, J., D’Hooge, P., Hooghe, B., Wu, X., Libbrecht, L., De Vos, R., 1881–1888. Quondamatteo, F., Klempt, M., Brakebusch, C., van Roy, F., 2008. Continuous Henderson, N.C., Mackinnon, A.C., Farnworth, S.L., Poirier, F., Russo, F.P., Iredale, cell injury promotes hepatic tumorigenesis in cdc42-deficient mouse liver. J.P., Haslett, C., Simpson, K.J., Sethi, T., 2006. Galectin-3 regulates myofibroblast Gastroenterology 134 (3), 781–792. activation and hepatic fibrosis. Proc. Natl. Acad. Sci. USA 103 (13), 5060–5065. Walters,K.A.,Smith,M.W.,Pal,S.,Thompson,J.C.,Thomas,M.J.,Yeh,M.M.,Thomas,D.L., Jang, J.Y., Shao, R.X., Lin, W., Weinberg, E., Chung, W.J., Tsai, W.L., Zhao, H., Goto, K., Fitzgibbon, M., Proll, S., Fausto, N., Gretch, D.R., Carithers Jr., R.L., Shuhart, M.C., Zhang, L., Mendez-Navarro, J., Jilg, N., Peng, L.F., Brockman, M.A., Chung, R.T., Katze, M.G., 2006. Identification of a specific gene expression pattern associated 2011. HIV infection increases HCV-induced hepatocyte apoptosis. J. Hepatol. with HCV-induced pathogenesis in HCV- and HCV/HIV-infected individuals. 54 (4), 612–620. Virology 350 (2), 453–464. Jiang, J.X., Chen, X., Hsu, D.K., Baghy, K., Serizawa, N., Scott, F., Takada, Y., Fukada, Wanchu, A., Rana, S.V., Pallikkuth, S., Sachdeva, R.K., 2009. Short communication: H., Chen, J., Devaraj, S., Adamson, R., Liu, F.T., Torok, N.J., 2011. Galectin-3 oxidative stress in HIV-infected individuals: a cross-sectional study. AIDS Res. modulates phagocytosis-induced stellate cell activation and liver fibrosis Hum. Retroviruses 25 (12), 1307–1311. in vivo. Am. J. Physiol. Gastrointest. Liver Physiol. Wang, J., Koyota, S., Zhou, X., Ueno, Y., Ma, L., Kawagoe, M., Koizumi, Y., Okamoto, Kato, M., Iwamoto, H., Higashi, N., Sugimoto, R., Uchimura, K., Tada, S., Sakai, H., H., Sugiyama, T., 2009. Expression and localization of regenerating gene I in a Nakamuta, M., Nawata, H., 1999. Role of Rho small GTP binding protein in the regulation of actin cytoskeleton in hepatic stellate cells. J. Hepatol. 31 (1), rat liver regeneration model. Biochem. Biophys. Res. Commun. 380 (3), 91–99. 472–477. Kim, A.Y., Chung, R.T., 2009. Coinfection with HIV-1 and HCV—a one-two punch. Yang, X., Deignan, J.L., Qi, H., Zhu, J., Qian, S., Zhong, J., Torosyan, G., Majid, S., Gastroenterology 137 (3), 795–814. Falkard, B., Kleinhanz, R.R., Karlsson, J., Castellani, L.W., Mumick, S., Wang, K., Kottilil, S., Yan, M.Y., Reitano, K.N., Zhang, X., Lempicki, R., Roby, G., Daucher, M., Xie, T., Coon, M., Zhang, C., Estrada-Smith, D., Farber, C.R., Wang, S.S., van Nas, A., Yang, J., Cortez, K.J., Ghany, M., Polis, M.A., Fauci, A.S., 2009. Human immuno- Ghazalpour, A., Zhang, B., Macneil, D.J., Lamb, J.R., Dipple, K.M., Reitman, M.L., deficiency virus and hepatitis C infections induce distinct immunologic Mehrabian, M., Lum, P.Y., Schadt, E.E., Lusis, A.J., Drake, T.A., 2009. Validation of imprints in peripheral mononuclear cells. Hepatology 50 (1), 34–45. candidate causal genes for obesity that affect shared metabolic pathways and Lamb, J.R., Zhang, C., Xie, T., Wang, K., Zhang, B., Hao, K., Chudin, E., Fraser, H.B., networks. Nat. Genet. 41 (4), 415–423. Millstein, J., Ferguson, M., Suver, C., Ivanovska, I., Scott, M., Philippar, U., Yonkers, N.L., Rodriguez, B., Asaad, R., Lederman, M.M., Anthony, D.D., 2011a. Bansal, D., Zhang, Z., Burchard, J., Smith, R., Greenawalt, D., Cleary, M., Derry, J., Systemic immune activation in HIV infection is associated with decreased Loboda, A., Watters, J., Poon, R.T., Fan, S.T., Yeung, C., Lee, N.P., Guinney, J., MDC responsiveness to TLR ligand and inability to activate naive CD4 T-cells. Molony, C., Emilsson, V., Buser-Doepner, C., Zhu, J., Friend, S., Mao, M., Shaw, PLoS One 6 (9), e23884. P.M., Dai, H., Luk, J.M., Schadt, E.E., 2011. Predictive genes in adjacent normal Yonkers, N.L., Sieg, S., Rodriguez, B., Anthony, D.D., 2011b. Reduced naive CD4 T tissue are preferentially altered by sCNV during tumorigenesis in liver cancer cell numbers and impaired induction of CD27 in response to T cell receptor and may rate limiting. PLoS One 6 (7), e20090. stimulation reflect a state of immune activation in chronic hepatitis C virus Li, L., Bachem, M.G., Zhou, S., Sun, Z., Chen, J., Siech, M., Bimmler, D., Graf, R., 2009. infection. J. Infect. Dis. 203 (5), 635–645. Pancreatitis-associated protein inhibits human pancreatic stellate cell MMP-1 Yuan, R.H., Jeng, Y.M., Chen, H.L., Hsieh, F.J., Yang, C.Y., Lee, P.H., Hsu, H.C., 2005. and -2, TIMP-1 and -2 secretion and RECK expression. Pancreatology 9 (1–2), Opposite roles of human pancreatitis-associated protein and REG1A expres- 99–110. sion in hepatocellular carcinoma: association of pancreatitis-associated pro- Lieu, H.T., Batteux, F., Simon, M.T., Cortes, A., Nicco, C., Zavala, F., Pauloin, A., tein expression with low-stage hepatocellular carcinoma, beta-catenin Tralhao, J.G., Soubrane, O., Weill, B., Brechot, C., Christa, L., 2005. HIP/PAP mutation, and favorable prognosis. Clin. Cancer Res. 11 (7), 2568–2575. accelerates liver regeneration and protects against acetaminophen injury in Zhong, H., Beaulaurier, J., Lum, P.Y., Molony, C., Yang, X., Macneil, D.J., Weingarth, mice. Hepatology 42 (3), 618–626. D.T., Zhang, B., Greenawalt, D., Dobrin, R., Hao, K., Woo, S., Fabre-Suver, C., Lin, W., Weinberg, E.M., Tai, A.W., Peng, L.F., Brockman, M.A., Kim, K.A., Kim, S.S., Qian, S., Tota, M.R., Keller, M.P., Kendziorski, C.M., Yandell, B.S., Castro, V., Attie, Borges, C.B., Shao, R.X., Chung, R.T., 2008. HIV increases HCV replication in a A.D., Kaplan, L.M., Schadt, E.E., 2010. Liver and adipose expression associated TGF-beta1-dependent manner. Gastroenterology 134 (3), 803–811. SNPs are enriched for association to type 2 diabetes. PLoS Genet. 6 (5), Lin, W., Wu, G., Li, S., Weinberg, E.M., Kumthip, K., Peng, L.F., Mendez-Navarro, J., e1000932. Chen, W.C., Jilg, N., Zhao, H., Goto, K., Zhang, L., Brockman, M.A., Schuppan, D., Zhu, J., Lum, P.Y., Lamb, J., GuhaThakurta, D., Edwards, S.W., Thieringer, R., Berger, Chung, R.T., 2011. HIV and HCV cooperatively promote hepatic fibrogenesis via J.P., Wu, M.S., Thompson, J., Sachs, A.B., Schadt, E.E., 2004. An integrative induction of reactive oxygen species and NFkappaB. J. Biol. Chem. 286 (4), genomics approach to the reconstruction of gene networks in segregating 2665–2674. populations. Cytogenet. Genome Res. 105 (2–4), 363–374. Maeda, N., Kawada, N., Seki, S., Arakawa, T., Ikeda, K., Iwao, H., Okuyama, H., Zhu, J., Zhang, B., Schadt, E.E., 2008. A systems biology approach to drug discovery. Hirabayashi, J., Kasai, K., Yoshizato, K., 2003. Stimulation of proliferation of rat Adv. Genet. 60, 603–635.