HCV-Induced Immunometabolic Crosstalk in a Triple-Cell Co-Culture Model Capable of Simulating Systemic Iron Homeostasis

Abstract

Iron is crucial to the regulation of the host innate immune system and the outcome of many infections. Hepatitis C virus (HCV), one of the major viral human pathogens that depends on iron to complete its life cycle, is highly skilled in evading the immune system. This study presents the construction and validation of a physiologically relevant triple-cell co-culture model that was used to investigate the input of iron in HCV infection and the interplay between HCV, iron, and determinants of host innate immunity. We recorded the expression patterns of key proteins of iron homeostasis involved in iron import, export and storage and examined their relation to the iron regulatory hormone hepcidin in hepatocytes, enterocytes and macrophages in the presence and absence of HCV. We then assessed the transcriptional profiles of pro-inflammatory cytokines Interleukin-6 (IL-6) and interleukin-15 (IL-15) and anti-inflammatory interleukin-10 (IL-10) under normal or iron-depleted conditions and determined how these were affected by infection. Our data suggest the presence of a link between iron homeostasis and innate immunity unfolding among liver, intestine, and macrophages, which could participate in the deregulation of innate immune responses observed in early HCV infection. Coupled with iron-assisted enhanced viral propagation, such a mechanism may be important for the establishment of viral persistence and the ensuing chronic liver disease.

Keywords: HCV; co-culture; cytokines; enterocytes; ferritin; hepcidin; innate immunity; iron; macrophages.

Figure A1

Basal gene expression of hepcidin and its inhibitor TMPRSS6 in triple-cell co-cultures. HAMP mRNA expression (a) and hepcidin-secreted peptide levels (b) were measured in mock-infected triple-cell co-cultures. RNA was isolated from Huh7.5 hepatoma cells and THP-1 macrophages and subjected to RT-qPCR at 24 h intervals. 18S rRNA was used as an internal control. Secreted peptide concentrations were measured by ELISA in pooled supernatants. The control of 24 h was arbitrarily set as 100%, and all plotted values are the percentage of this. (c) TMPRSS6 was detected in Huh7.5 whole cell extracts from the same triple-cell co-cultures described in (a,b). Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry and plotted in expression histograms. Values were calculated as a percentage of the 24 h mock-infected control. All graphs depict the average expression from at least three individual experiments, with representative accompanying blots.

Figure A2

Basal hepatic protein expression in mock-infected triple-cell co-cultures. Ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) polypeptides were detected in Huh7.5 whole cell extracts from Supplementary Materials Figure S1. Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the 24 h mock-infected control. All graphs depict average expression from at least three individual experiments, with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05.

Figure A3

Basal intestinal protein expression in mock-infected triple-cell co-cultures. Ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) polypeptides were detected in CaCO₂ whole cell extracts from the same triple-cell co-cultures described in Supplementary Materials Figure S1. Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the 24 h mock-infected control. All graphs depict the average expression from at least three individual experiments with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05.

Figure A4

Basal macrophagic protein expression in mock-infected triple-cell co-cultures. Ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) polypeptides were detected in THP-1 whole cell extracts from the same triple-cell co-cultures described in Supplementary Materials Figure S1. Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the 24 h mock-infected control. All graphs depict the average expression from at least three individual experiments with representative accompanying blots.

Figure A5

Basal pro- and anti-inflammatory cytokine expression profiling in mock-infected triple-cell co-cultures. The gene expression of pro-inflammatory IL-6 (a) and IL-15 (b), as well as anti-inflammatory IL-10 (c), was measured in cellular populations of the triple-cell co-cultures that are known to express the respective cytokine, in the presence or absence of DFO. mRNA expression levels were subjected to RT-qPCR with specific oligonucleotide primers. 18S rRNA was used as an internal control. The 24 h mock-infected control was arbitrarily set as 1, and all plotted values are fold- changes of this. All graphs depict the average expression of replicates acquired from at least three individual experiments. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05; **, p-value ≤ 0.005 (black asterisks: statistical significance of the infected over mock-infected controls; grey asterisks: statistical significance of the infected over mock-infected controls treated with DFO; red asterisks: statistical significance between the infected samples −/+ DFO).

Figure 1

Differential regulation of hepcidin antimicrobial peptide (HAMP) and its inhibitor TMPRSS6 during hepatitis C virus (HCV) infection. HAMP and HCV NS3 mRNA expression (a) and hepcidin-secreted peptide levels (b) were measured in triple-cell co-cultures following infection with HCV JFH-1. RNA was isolated from Huh7.5 hepatoma cells and THP-1 macrophages and subjected to RT-qPCR at 24 h intervals post-infection (p.i.). 18S rRNA was used as an internal control. Secreted peptide concentrations were measured by ELISA in pooled supernatants. The mock-infected controls per time interval were arbitrarily set as 100% (represented by the dashed line), and all plotted values are the percentage of these. (c) TMPRSS6 was detected in Huh7.5 whole cell extracts from the same triple-cell co-cultures described in (a,b). Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry and plotted in expression histograms. Values were calculated as a percentage of the respective mock-infected controls (represented by the dashed line). All graphs depict the average expression from at least three individual experiments, with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05; **, p-value ≤ 0.005.

Figure 2

Differential regulation of hepatic iron-related proteins by HCV infection. Protein expression of ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) were detected in Huh7.5 whole cell extracts from the same triple-cell co-cultures described in Figure 1. Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the respective mock-infected controls (represented by the dashed line). All graphs depict the average expression from at least three individual experiments, with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05; **, p-value ≤ 0.005.

Figure 3

Differential regulation of intestinal iron-related proteins by HCV infection. Protein expression of ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) were detected in CaCO₂ whole cell extracts from the same triple-cell co-cultures described in Figure 1. Isolated protein was used in immunoblotting analysis and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the respective mock-infected controls (represented by the dashed line). All graphs depict average expression from at least three individual experiments, with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05.

Figure 4

Differential regulation of macrophagic iron-related proteins by HCV infection. Protein expression of ferritin (a), FPN (b), TfR1 (c) and DMT1 (100 kDa and 65 kDa isoforms) (d) were detected in THP-1 whole cell extracts from the same triple-cell co-cultures described in Figure 1. Isolated protein was used in immunoblotting analysis, and β-actin was the loading control. The blots were subjected to densitometry, and the resulting expression values were plotted in expression histograms. Values were calculated as a percentage of the respective mock-infected controls (represented by the dashed line). All graphs depict average expression from at least three individual experiments, with representative accompanying blots. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05; **, p-value ≤ 0.005.

Figure 5

Pro- and anti-inflammatory cytokine expression profiling during HCV infection. The gene expression of pro-inflammatory interleukin (IL)-6 (a) and IL-15 (b), as well as anti-inflammatory IL-10 (c), was measured in cellular populations of the triple-cell co-cultures that are known to express the respective cytokine, following HCV infection in the presence or absence of desferrioxamine (DFO). mRNA expression levels were subjected to RT-qPCR with specific oligonucleotide primers. 18S rRNA was used as an internal control. The mock-infected controls per time interval were arbitrarily set as 1, and all plotted values are fold changes of these. All graphs depict the average expression of replicates acquired from at least three individual experiments. Statistical significance is denoted by asterisks as follows: *, p-value ≤ 0.05; **, p-value ≤ 0.005 (black asterisks: statistical significance of infected over the mock-infected controls; grey asterisks: statistical significance of the infected over mock-infected controls treated with DFO; red asterisks: statistical significance between the infected samples −/+ DFO).

Figure 6

Iron homeostasis and innate immunity determinants in HCV infection. Schematic diagram of the immunometabolic interplay between HCV, iron and determinants of the innate immune response in a triple-cell co-culture model. The diagram schematically captures major findings reported in this study by depicting transcriptional alterations of cytokine genes and translational expression changes of iron homeostasis proteins elicited by HCV infection.