Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 2014 Nov 10;143(4):578–587. doi: 10.1111/imm.12336

Conjugated bilirubin affects cytokine profiles in hepatitis A virus infection by modulating function of signal transducer and activator of transcription factors

Flor P Castro-García 1,2, Karla F Corral-Jara 1,2, Griselda Escobedo-Melendez 3,*, Monserrat A Sandoval-Hernandez 4, Yvonne Rosenstein 4, Sonia Roman 2,5, Arturo Panduro 2,5, Nora A Fierro 1,2,5,✉
PMCID: PMC4253506  PMID: 24943111

Abstract

Hepatitis A virus (HAV) infection is the major cause of acute liver failure in paediatric patients. The clinical spectrum of infection is variable, and liver injury is determined by altered hepatic enzyme function and bilirubin concentration. We recently reported differences in cytokine profiles between distinct HAV-induced clinical courses, and bilirubin has been recognized as a potential immune-modulator. However, how bilirubin may affect cytokine profiles underlying the variability in the course of infection has not been determined. Herein, we used a transcription factor (TF) binding site identification approach to retrospectively analyse cytokine expression in HAV-infected children and to predict the entire set of TFs associated with the expression of specific cytokine profiles. The results suggested that modulation of the activity of signal transducers and activators of transcription proteins (STATs) may play a central role during HAV infection. This led us to compare the degree of STAT phosphorylation in peripheral blood lymphoid cells (PBLCs) from paediatric patients with distinct levels of conjugated bilirubin (CB). Low CB levels in sera were associated with increased STAT-1 and STAT-5 phosphorylation. A positive correlation was observed between the serum interleukin-6 (IL-6) content and CB values, whereas higher levels of CB correlated with reduced serum IL-8 values and with a reduction in the proportion of PBLCs positive for STAT-5 phosphorylation. When CB was used to stimulate patients’ PBLCs in vitro, the levels of IL-6 and tumour necrosis factor-α were increased. The data showed that bilirubin plays a role in STAT function and affects cytokine profile expression during HAV infection.

Keywords: bilirubin, hepatitis A virus, signal transducer and activator of transcription protein, transcription factors, viral hepatitis

Introduction

Each year, worldwide, hepatitis A virus (HAV) infects approximately 1·5 million people. The virus responsible for the infection replicates in the liver.1 Transmission of HAV occurs via the faecal–oral route through contaminated food or water; such transmission is associated with unsanitary conditions.2 Although improved hygiene and vaccination have reduced the HAV infection rate, the virus remains widespread in developing countries,3 where the infection is generally acquired in early childhood.4

Because hepatitis A is an acute, self-limiting disease, most cases resolve spontaneously without residual damage or sequel. Nonetheless, during infection, the spectrum of clinical manifestations is broad, ranging from mild to intermediate disease to acute liver failure.5 The causes underlying the variability in the clinical course induced by HAV have not been clearly defined. However, given that HAV is a non-cytopathic virus, the damage to the liver resulting from the infection most likely does not stem directly from virus replication; rather, it is produced by the virus-specific, cell-mediated immune response to infected hepatocytes.6,7

Liver damage caused by viral hepatitis has been associated with host cytotoxic T lymphocytes directed against virus-infected hepatocytes8 and with the establishment of a T helper type 1 cytokine profile.6,7,9 During the resolution of acute HAV infection in a chimpanzee model, a clear dominance of CD4+ T helper cells that produce the cytokines interferon-γ (IFN-γ), tumour necrosis factor-α (TNF-α), interleukin-2 (IL-2) and IL-21 has been described,10 suggesting a unique mechanism to prevent the progression of the infection. Additionally, acute HAV infection is characterized by a limited intra-hepatic type I IFN response,11 and the temporary inhibition of the function of T regulatory (Treg) cells, which occurs during the infection, has been explained in terms of a specific interaction between HAV and its cellular receptor (HAVCR1) on the T-cell surface, in a transforming growth factor-β (TGF-β) -dependent mechanism.12,13 We reported recently that distinct HAV-induced clinical courses are associated with different cytokine profiles.14 In particular, in HAV-infected children, we found that over-expression of TNF-α, together with IL-1α, IL-6, IL-13 and monocyte chemoattractant protein-2 (MCP-2), correlates with high serum levels of conjugated bilirubin (CB). In contrast, in patients with low serum levels of CB, cytokines associated with hepatitis-induced inflammation, TGF-β and IL-8 are dominant, which supports the idea that, during viral infection, changes in cytokine activities are associated with different outcomes.14

Changes in hepatic enzymes, including aspartate aminotransferase (AST) and alanine aminotransferase (ALT), as well as changes in the concentration of bilirubin, have been associated with liver injury during hepatic infection. In particular, CB values > 2 mg/dl are linked with cholestasis, a condition in which substances normally excreted into the bile are retained.15,16 Interestingly, bilirubin, a potent endogenous antioxidant, has been shown to be an immunomodulator.17 Models in vitro have shown that bilirubin concentrations > 25 μm modulate apoptosis of CD4+ T cells and neutrophils18,19 and that the induction of tolerance observed after administration of bilirubin to transplant recipients results from de novo generation of Treg cells.20 In addition, bilirubin is able to decrease IL-2 production in human lymphocytes.21 Therefore, we hypothesized that the interplay between CB serum level and transcriptional control of cytokines may modulate the immune response to HAV and influence the severity of disease.

The approach that we used to understand the molecular basis of transcriptional control of cytokines during HAV infection was the identification of the transcription factor binding site (TFBS).22 Hence, using serum samples from paediatric patients with distinct levels of CB – a measure of distinct clinical courses following HAV infection – we characterized the transcriptional factors (TFs) that potentially may be involved in modulating characteristic cytokine profile expression. The data suggested that the CB-mediated modulation of signal transducers and activators of transcription (STATs) plays a central role during HAV infection. These results will help to improve our understanding of the interplay between metabolic and transcriptional components that modulate immune function during type A viral hepatitis and that could contribute to the resolution of infection during the acute phase.

Materials and methods

Study population

A total of 77 paediatric patients (< 15 years old) were included in this study. The patients were admitted to the Servicio de Infecto-pediatria of the Hospital Civil de Guadalajara Fray Antonio Alcalde (HCFAA) between 2011 and 2013. Hepatitis was defined as hepatomegaly, fever (> 38°), and/or jaundice with elevated values of serum AST (> 38 IU/l) and ALT (> 35 IU/l), as previously described.3 Additionally, CB (> 0·3 mg/dl) and albumin values were measured and clinical features were recorded. Excluded from the study were patients with liver disease who were undergoing treatment with a hepatotoxic drug, those with acute hepatitis E virus (HEV) infection or with chronic hepatitis, and those diagnosed with autoimmune hepatitis; none of the patients had been vaccinated against HAV. A total of 30 healthy paediatric donors (< 15 years old) who had been admitted to the Unidad de Vacunación of the HCFAA but who had not yet been vaccinated against HAV were included in this study as controls. After the children’s parents had provided informed consent, blood samples from patients and healthy donors were obtained by venepuncture. This study was approved by the ethical committee of the HCFAA.

Clinical and demographic data

The demographic and clinical history data were collected using a structured questionnaire, as previously reported.3 The recorded data included age and gender and clinical features, including time of onset of the clinical symptoms, nausea, vomiting, abdominal pain, choluria, acholia, acute liver failure and hepatitis A and B vaccination status.

Serological tests

To detect acute hepatitis A infection, serum samples from patients diagnosed with hepatitis were screened for the presence of anti-HAV IgM and the absence of anti-HAV IgG. All samples were negative for antibodies to the hepatitis B virus (HBV), hepatitis C virus (HCV), and HEV. The presence of anti-HAV IgM and absence of anti-HAV IgG, the surface antigen of HBV (HBsAg), and anti-HCV antibodies was tested by using a third-generation microparticle immunoenzymatic assay [AxSYM HAVAB-M 2·0, AxSYM HBsAg (V2), and AxSYM HCV 3·0; Abbott Laboratories, Chicago, IL] with an AxSYM analyser (Abbott Laboratories). Total anti-hepatitis B core antigen anti-HBc (total IgM and IgG) and anti-HEV antibodies were measured by using immunoenzymatic assays (Monolisa Anti-HBc PLUS, Bio-Rad Laboratories, Chicago, IL, and MP Diagnostics, Geneva, Switzerland, respectively) with a PR 3100 TSC analyser (Bio-Rad). The levels of albumin/globulin, ALT, AST, alkaline phosphatase, total protein, total bilirubin and CB were measured in the serum samples, following routine clinical–laboratory procedures.

Liver injury categorization in hepatitis A-infected children

Patients who tested positive for acute HAV infection (anti-HAV IgM+ and anti-HAV IgG−) and negative for antibodies to HBV, HCV and HEV and who exhibited abnormal levels of ALT and AST (> 38 IU/l and/or > 35 IU/l, respectively) were categorized as previously described:14

  1. Minor HAV-induced liver injury (M-HAV-ILI): patients who exhibited CB levels between > 0·3 and < 2 mg/dl (38 patients).

  2. Intermediate HAV-induced liver injury (I-HAV-ILI): patients who exhibited CB levels > 2 mg/dl (39 patients).

  3. Healthy controls (H): children with normal hepatic enzymatic activity in the absence of HAV, HBV and HCV serological markers.

Analysis of IL-6 and IL-8 in sera

Cytokines in the serum samples were detected by ELISA following the manufacturer’s guidelines. The following reagents were used: human IL-6 and human IL-8 ELISA MAX standard set (BioLegend, San Diego, CA).

CB treatment in vitro

Ficoll-Paque PLUS (GE Healthcare, Uppsala, Sweden) gradient centrifugation was used to isolate peripheral blood lymphoid cells (PBLCs) from anti-coagulated blood samples from three healthy donors exhibiting 0·13, 0·16 and 0·10 mg CB/dl and from three HAV-infected patients who exhibited 0·34, 0·40 and 0·45 mg CB/dl. The buffy coat of each sample was washed three times with PBS solution (300 g; 10 min; room temperature) and resuspended in RPMI-1640 supplemented with 5% fetal calf serum and 5% bovine iron-supplemented calf serum (Hyclone, Logan, UT), with 2 mm l-glutamine, 50 μg/ml penicillin, 50 μg/ml streptomycin and 50 μm β-mercaptoethanol (Sigma, St Louis, MO). Before treatment with CB (Merck-Millipore, Darmstadt, Germany), the purified PBLCs were arrested (2 hr) in RPMI-1640 supplemented with 2% fetal calf serum (Sigma). A total of 2 × 106 purified PBLCs per condition were incubated with increasing concentrations of CB (0, 1, 2 and 3 mg/dl). Interleukin-6 and TNF-α present in the tissue culture media at 48 hr following the treatment were detected by using human IL-6 and human TNF-α ELISA MAX standard set (BioLegend). The PBLCs were then washed and labelled with anti-phospho-STAT-5 (Merck-Millipore) as described below.

Prediction of TFBS

The promoter sequences for TGF-β, IL-8, IL-6, IL-13, IL-1α, TNF-α and MCP-2 were obtained from the ensembl database version 67 (http://www.ensembl.org/info/website/archives/index.html), including sequences 1000 bp upstream and 200 bp downstream from the ATG for each of these seven cytokines. TFBS were identified by using position weight matrices from the transfac database.23 The Patch algorithm was used to identify potential TFBS, taking into consideration the following parameters: (i) pattern matrix of 6 bp; (ii) matching score = 100% of identity; (iii) vertebrate (mammals) position weight matrices: human, and (iv) a lower-score boundary of 87·5. For each gene, based on the predicted DNA-binding sites, we generated a matrix of absence/presence (0, 1) for each TF. A hierarchical clustering analysis was performed to identify groups of TFs associated with common gene profiles through the Pearson correlation as a distance metric and average linkage clustering as linkage method by using Cluster 3·0 and was visualized by using the java tree view program (Lawrence Berkeley National Laboratory, University of California, Oakland, CA).

Phospho-STAT-1, -3 and -5 FACS staining

Before the addition of specific antibodies to blood samples, the red blood cells were lysed with Cal-lyse whole blood lysing solution (Invitrogen, Camarillo, CA). Lymphoid cells were subsequently washed by centrifugation (300 g; 10 min) to remove red cell debris. The cells were then washed and resuspended in fixation buffer (Merck-Millipore) and incubated (10 min; room temperature). The cells were then washed by centrifugation (300 g; 10 min) and resuspended in ice-cold permeabilization buffer (Merck-Millipore) and mixed by vortexing at high speed (20 seconds). The cells were then incubated on ice (10 min) and washed by centrifugation. Anti-phospho-STAT-1, -3, -5 and anti-pan STAT staining was conducted according to the manufacturer’s instructions (Merck-Millipore). Briefly, cells (1 × 106) were resuspended in 100 μl of assay buffer (Merck-Millipore) and incubated with anti-STAT-1, -STAT-3, STAT-5 and anti-pan STAT (30 min; room temperature) while protected from light. The cells were then washed by centrifugation (300 g; 5 min) and resuspended in assay buffer and analysed using a Guava EasyCyte 6 with incyte 2·0 software (Merck-Millipore). The percentage of positive cells was obtained from the acquisition of 10 000 events. Triplicate counts from the 1 × 106 cells resuspended in assay buffer were conducted.

Statistical analysis

The data are presented as the mean ± standard deviation (SD). Statistical comparisons were performed by using graphpad prism software version 5·01 (GraphPad Software, Inc, San Diego, CA). A non-parametric Mann–Whitney U-test was used to calculate the statistical significance of the assay results. A P-value < 0·05 was considered statistically significant. Significant P-values were corrected by using the Bonferroni method to ensure that there were differences between the compared groups. To study associations between variables, the Pearson correlation coefficient was calculated by using simple regression analysis.

Results

CB levels were differentially associated with IL-8 and IL-6 secretion during HAV infection

We previously found differences in the relative cytokine levels during distinct clinical courses of HAV infection.14 Herein, when the IL-8 and IL-6 concentrations in serum samples from HAV-infected patients who had distinct clinical courses were examined, significantly higher concentrations of IL-8 (12·81 pg/ml ± 3·89) were found for HAV-infected children with M-HAV-ILI relative to those (2·62 pg/ml ± 4·27) found for children with I-HAV-ILI; no IL-8 was detectable in healthy donors’ sera (Fig. 1a). In agreement with previous work,14 patients with M-HAV-ILI or I-HAV-ILI had higher IL-6 levels than did healthy donors, and I-HAV-ILI patients exhibited higher concentrations of IL-6 (19·77 pg/ml ± 8·07) relative to patients with M-HAV-ILI (9·2 pg/ml ± 5·24) or healthy donors (1·17 pg/ml ± 2·76) (Fig. 1b). We found a wide variability in the concentrations of IL-8 and IL-6 secreted, such that there was overlap between the concentration ranges of the two groups of patients. For IL-8, the values in the lower range of the M-HAV-ILI group were similar to those in the upper range of the I-HAV-ILI group; a corresponding finding was observed for IL-6 (Fig. 1a,b). Classification of our patients was based on the concentration of CB in serum. To determine if those patients with similar concentrations of IL-8 and IL-6 in the different study groups would have similar serum levels of CB, and hence if CB could play a role in the differential secretion of IL-6 and IL-8 during HAV infection, we analysed the possible correlation between IL-8 and IL-6 concentrations with that of CB in serum. No correlation between IL-8 and CB values was found, although data trended towards a reduction in IL-8 levels at > 2 mg CB/dl (Fig. 1c). In contrast, the data analysis between IL-6 and CB values revealed a positive correlation, particularly in those patients with CB values > 1 mg/dl (Fig. 1d). Our data suggest that IL-6 detected in sera from patients could be related to higher levels of CB, whereas high levels of CB could result in a reduction of IL-8 secretion. We found no significant differences between the clinical and demographic characteristics among the groups. However, in patients with I-HAV-ILI a significant increase in AST levels was found relative to patients with M-HAV-ILI (Table 1). Taken together, these data not only support our previous observation that determination of the cytokine profiles can help to distinguish between M-HAV-ILI and I-HAV-ILI, but also suggest that CB levels in sera correlate with specific cytokine profiles, so defining the outcome of the infection.

Figure 1.

Figure 1

Open in a new tab

Interleukin-8 (IL-8) and IL-6 were differentially regulated by conjugated bilirubin in different hepatitis A virus (HAV) -induced clinical courses. ELISAs were performed to determine the concentrations of cytokines in serum samples from patients with minor HAV-induced liver injury (M-HAV-ILI; n = 30), intermediate HAV-induced liver injury (I-HAV-ILI; n = 30), and healthy donors (H; n = 30). Sera concentrations of IL-8 (a) and of IL-6 (b). Values ± the standard deviation (SD) are presented. The Pearson correlation coefficients for IL-8, IL-6, and conjugated bilirubin (CB) were calculated by using simple regression analysis and are shown in (c) and (d), respectively. P < 0·05 value was considered statistically significant. ***P < 0·0001.

Table 1.

Demographic and clinical characteristics of patients and controls

Patients
Characteristic Healthy controls (n = 30) M-HAV-ILI (n = 38) I-HAV-ILI (n = 39) P value
Gender (% female) 60 53 50 NS
Mean age (years ± SD) 6·87 ± 2·83 6·93 ± 3·26 7·1 ± 3·56 NS
Mean ALT (IU/l ± SD) 22·23 ± 17·58 1875 ± 555·46 1591·07 ± 1018·21 NS
Mean AST (IU/l ± SD) 13·77 ± 11·20 439·22 ± 437·11 1318·96 ± 1028·17 < 0·05
Mean CB (mg/dl ± SD) 0·15 ± 0·10 1·11 ± 0·59 4·39 ± 1·58 −
Anti-HAV IgM − + + –
Anti-HAV IgG − − − –
Open in a new tab

HAV, hepatitis A virus; M-HAV-ILI, minor HAV-induced liver injury; I-HAV-ILI, intermediate HAV-induced liver injury; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CB, conjugated bilirubin; IgM anti-HAV, immunoglobulin M anti-HAV antibody; IgG anti-HAV, immunoglobulin G anti-HAV antibody; SD, standard deviation; NS, not significant.

High concentration of CB induced IL-6 and TNF-α secretion in PBLCs from HAV-infected patients in vitro

Our data indicated that within a specific concentration range of CB (> 2 mg/dl) in sera, a characteristic pro-inflammatory cytokine profile was induced during HAV infection. We investigated whether this is a general mechanism regulated through bilirubin, or whether it is specific to HAV infection. Under basal conditions, for healthy donors and patients with low levels of CB, minimal levels of IL-6 (2·2 pg/ml ± 0·40 and 4·4 pg/ml ± 1·21, respectively) and of TNF-α (3·8 pg/ml ± 0·75 and 6·7 pg/ml ± 2·04) were found (Fig. 2 a,b). However, when increasing concentrations of CB were added, the data trended toward increased IL-6 and TNF-α levels in healthy donors, with this effect being more evident at the higher concentrations of CB. Interestingly, significantly higher concentrations of IL-6 and TNF-α (40·5 pg/ml ± 19·68 and 25·3 pg/ml ± 8·40, respectively) were found in patients with M-HAV-ILI relative to those of healthy donors (7·3 pg/ml ± 1·23 and 6·1 pg/ml ± 1·16) once PBLCs were incubated with 2 mg/dl of CB. Additionally, increasing IL-6 and TNF-α levels were found when 3 mg/dl of CB were added to PBLCs of those patients (62·2 pg/ml ± 18·61 and 40 pg/ml ± 3·85) relative to healthy donors (8·7 pg/ml ± 2·15 and 7 pg/ml ± 1·23) (Fig. 2a,b). These results showed that high concentrations of added CB induced IL-6 and TNF-α secretion from human PBLCs and that this effect was potentiated during HAV infection.

Figure 2.

Figure 2

Open in a new tab

High concentration of conjugated bilirubin (CB) resulted in interleukin-6 (IL-6) and tumour necrosis factor-α (TNF-α) secretion in vitro in lymphoid cells from hepatitis A virus (HAV) -infected patients. Peripheral blood lymphoid cells (PBLCs) isolated from three healthy (H) donors and three patients with minor HAV-induced liver injury (P) were treated with increasing concentrations of CB (0, 1, 2 and 3 mg/dl). IL-6 (a) and TNF-α (b) present in the cell culture media for 48 hr following the treatment were detected by ELISA.

Cytokines were differentially co-regulated by nuclear factor-κB and STATs during distinct HAV-induced clinical courses

Through an analysis in silico, we addressed the possibility that the transcription of those cytokines specific to M-HAV-ILI (IL-8 and TGF-β) or the I-HAV-ILI forms (IL-1α, IL-6, IL-13, TNF-α and MCP-2) of HAV infection may rely on the recruitment of different sets of TFs to the promoter region of the genes that encode the cytokines. As shown in Fig. 3, the more frequently predicted TFs were common to cytokines corresponding to both groups. This list included TFs such as GATA-1 (GATA binding protein 1, globin transcription factor1), GATA-3 (GATA binding protein 3), HNF-1 (hepatocyte nuclear factor 1), PPARg (peroxisome-proliferator-activated receptor gamma), AP-1 (activator protein 1), and NFAT (Nuclear factor of activated T-cells). Interestingly, IL-8 and TGF-β (characteristic of M-HAV-ILI) had binding sites for nuclear factor-κB (NF-κB), whereas MCP-2 (characteristic of I-HAV-ILI) did not. Furthermore, members of the STATs family TFs were predicted to be differentially recruited to the promoters of the different groups of cytokines. Potential association of STAT-1 and STAT-6 was predicted for IL-6, IL-13, TNF-α, TGF-β and IL-1α but not for MCP-2 and IL-8. STAT-5 was potentially associated with all promoters, with the exception of that of IL-8, a cytokine associated with low levels of CB content. These findings suggest a fine control of transcriptional activity and a possible correlation between the level of CB and specific TFs, particularly NF-κB and STAT family members in driving the progression of HAV-induced disease.

Figure 3.

Figure 3

Open in a new tab

Hierarchical clustering identified specific transcription factors (TFs) potentially associated with cytokines corresponding to distinct hepatitis A virus (HAV) -induced clinical courses. The cytokines associated either with minor HAV-induced liver injury [transforming growth factor (TGF-β) and interleukin-8 (IL-8)] or with intermediate HAV-induced liver injury [IL-6, IL-13, IL-13, tumour necrosis factor-α (TNF-α), IL-1α and monocyte chemoattractant protein 1 (MCP-2)] are shown in the upper margin. The association between transcription factors predicted in silico for each cytokine analysed through patch-transfac program is shown in the right margin. Hierarchical clustering was obtained with the average linkage algorithm. In the dendrogram, the colour red identifies a positive prediction and black identifies the absence of a prediction for each factor in each gene.

Differential STAT phosphorylation modulated the outcome of HAV infection

The STAT proteins are DNA-binding TFs that regulate many aspects of growth, survival and differentiation in cells.24,25 The activation of STAT proteins following stimulation is mediated by tyrosine phosphorylation, leading to their dimerization and tetramerization, which facilitate nuclear translocation and binding to specific promoter elements.26 To evaluate the participation of the STAT family members in defining HAV-induced clinical courses, we evaluated STAT phosphorylation in PBLCs from HAV-infected patients and from healthy donors. Minimal phosphorylation of STAT-1, STAT-3, and STAT-5 was found in healthy donors (data not shown). Patients with M-HAV-ILI had a fivefold increase in the percentage of PBLCs positive for phosphorylated STAT-1 relative to patients with I-HAV-ILI (Fig. 4a,d). Conversely, we found a fourfold increase in STAT-5 phosphorylation in patients with M-HAV-ILI (Fig. 4c,f). No significant differences were found for STAT-3 phosphorylation between groups, though the patients with M-HAV-ILI tended to have more phospho-STAT-3-positive cells (Fig. 4b,e). An analysis of double phospho-STAT-positive cells did not reveal changes between groups and staining with an anti-pan STAT antibody showed that cells of all groups expressed equivalent amounts of STAT family members (data not shown). These data suggest a role for STATs in integrating and regulating the transcription of cytokines that differentially modulate the outcome of type HAV infection.

Figure 4.

Figure 4

Open in a new tab

Signal transducer and activator of transcription 1 (STAT-1), STAT-3 and STAT-5 were differentially phosphorylated in distinct clinical courses during hepatitis A virus (HAV) infection. After lysis of erythrocytes from blood samples obtained from minor HAV-induced liver injury (M-HAV-ILI; n = 8) and intermediate HAV-induced liver injury (I-HAV-ILI; n = 9) patients, the lymphoid cells were subsequently stained with anti-phospho-STAT-1-PerCP, anti-phospho-STAT-3-Alexa-488, and anti-phospho-STAT-5-PE and then analysed by using flow cytometry. Representative histograms from individual patients are shown in (a–c). The black histograms correspond to isotype controls and grey histograms correspond to phospho-STAT-1, -3, and -5 staining, respectively. The results are displayed as the percentage of cells with phosphorylated STAT-1 (d), STAT-3 (e), and STAT-5 (f). The value shown is representative of three independent counts. The medians ± the standard deviations (SD) of the percentage of cells with phosphorylated STAT are presented. P < 0·05 was considered statistically significant. *P < 0·05, **P < 0·001.

CB levels modified STAT-5 phosphorylation during HAV infection

Our data pointed to a correlation between cytokine profiles and levels of CB in HAV-infected children. Specifically, results from the identification of TFBS suggested that high expression of TGF-β was associated with STAT-5 activity (Figs 3 and 4). Moreover, we found that, at a serum CB concentration < 2 mg/dl, IL-8 was effectively secreted in HAV-infected patients. We reasoned that STATs could be differentially phosphorylated and recruited depending on CB concentration. To test the hypothesis that bilirubin levels were involved in STAT phosphorylation, we evaluated the possible correlation between the CB levels and the percentage of PBLCs with phosphorylated STAT-1, STAT-3 or STAT-5. No correlation between STAT-1 or STAT-3 phosphorylation was found relative to CB values (data not shown), and STAT-5 phosphorylation did not correlate with low CB values either. However, there was a trend towards a reduction in the percentage of positive cells for phospho-STAT-5 at CB values > 2 mg/dl (Fig. 5a). This finding was consistent with results obtained by treating PBLCs from patients with M-HAV-ILI with increasing concentrations of CB in vitro: a minimal phosphorylation of STAT-5 was obtained at the highest concentrations of CB (Fig 5b). These data suggest that, during HAV infection, STAT-5 phosphorylation can be modulated by CB in a dose-dependent manner. However, large-scale studies are necessary to characterize the functional significance of CB levels in STAT-5 phosphorylation and to determine the consequences over several related signal-transduction pathways.

Figure 5.

Figure 5

Open in a new tab

Conjugated bilirubin (CB) levels played a role in signal transducer and activator of transcription-5 (STAT-5) phosphorylation during hepatitis A virus (HAV) infection. (a) The Pearson correlation coefficients between phospho-STAT-5 and CB values relative to intermediate HAV-induced liver injury (I-HAV-ILI; n = 9) was calculated by simple regression analysis. (b) Peripheral blood lymphoid cells (PBLCs) were isolated from three patients with mild HAV-induced liver injury (M-HAV-ILI). PBLCs were treated with increasing concentrations of CB (0, 1, 2 and 3 mg/dl) for 48 hr at 37°. PBLCs were subsequently stained with anti-phospho-STAT-5-PE and analysed using flow cytometry. The results are displayed as the percentage of cells with phosphorylated STAT-5. The count shown is representative of three independent assays.

Discussion

The results of this study support the concept that bilirubin may play a role in modulating specific immune responses through actions that include intracellular signalling and transcriptional control, ultimately affecting cytokine secretion during HAV infection.

Recently, we reported that different cytokine patterns may be associated with different HAV-induced clinical courses.14 Our current data indicate that during HAV infection there is a fine balance between the CB content and cytokine secretion and suggest that high levels of CB may result in a reduction of IL-8 secretion (Fig. 1). Clinically relevant concentrations of bilirubin can induce apoptosis in neutrophils.27 Furthermore, bilirubin can suppress inflammation and increase antioxidant enzyme generation in activated neonatal neutrophils by down-regulating the lipopolysaccharide-induced generation of IL-8.19 Given that neutrophils are a source of IL-8, it is plausible that the changes in the proportion of neutrophils due to high concentrations of bilirubin resulted in the reduced IL-8 secretion that we found in HAV-infected patients with CB levels > 2 mg/dl. Although a retrospective analysis of clinical records of all the patients included in this study did not reveal significant differences in the amount of neutrophils between study groups, the data trended toward an increased proportion of neutrophils in patients with lower serum CB concentrations (data not shown). Hence, it is plausible that, during HAV infection, high concentrations of CB may affect cytokine secretion by immune cells, including neutrophils.

How bilirubin can regulate the immune system is not yet known. Bilirubin is a product of normal haem degradation.28 Haem oxygenase-1 (HO-1) degrades haem to generate biliverdin, carbon monoxide and ferrous iron.29 Biliverdin is rapidly converted into bilirubin by biliverdin reductase. In the liver, bilirubin is conjugated with glucuronic acid by the enzyme glucuronyltransferase, with the resulting CB being soluble in water.30 Given that HO-1 expression is indispensable for the development and function of CD4+ CD25+ Treg cells20 and given that CB is a product of HO enzyme activity, the effect of CB on immune cell function may be similar that of HO-1. Moreover, the extraordinary antioxidant property of bilirubin may contribute to its immunomodulatory action by regulating cytokine production during HAV infection. Consistent with this were our data in vitro showing that CB was able to induce IL-6 and TNF-α secretion in human PBLCs in a dose-dependent manner (Fig. 2).

Our study of the potential transcriptional differences, which are associated with the cytokine profiles related to distinct CB levels during HAV infection, predicted that the list of overlapping TFs related to the set of cytokines defined for each group would include TFs involved in cell proliferation, maturation and differentiation (Fig. 3).31–35 Interestingly, the promoters of most cytokines characteristic of M-HAV-ILI or I-HAV-ILI have binding sites for NF-κB. The absence of binding sites for NF-κB in the MCP-2 promoter suggests that NF-κB may play a role in defining the inflammatory profile during M-HAV-ILI. Furthermore, members of the STAT family of TFs were predicted to be differentially recruited to the promoters of the different groups of cytokines. These findings revealed subtle differences in transcriptional activity during different HAV-induced clinical courses and coincided with the recognized roles of STAT family members in antiviral defences, the acute-phase response, and hepatic injury,36,37 repair and inflammation.38 Herein, analyses ex vivo of STAT phosphorylation did not show changes in the values of STAT-3 phosphorylation in the analysed groups. In contrast, HAV-infected patients with low levels of CB and associated with TGF-β and IL-8 over-expression showed a significant increase in the percentage of PBLCs in which STAT-1 and STAT-5 were phosphorylated. Differential STAT-5 phosphorylation related to distinct levels of CB during infection was corroborated in vitro (Fig. 5); this finding suggests that CB levels may be important in modulating STAT-5 phosphorylation during HAV infection. Moreover, given that STAT-5 was not predicted for IL-8 (Fig. 3), our results suggest that STAT-5 may be relevant to the specific production of TGF-β in patients with M-HAV-ILI.39 However, further studies with a larger number of patients are needed to evaluate this possibility.

A mechanism recognized to be exploited by a number of RNA viruses, including HCV, to escape host surveillance is direct targeting of cytokine-inducible transcription regulators in the STAT family.36 The involvement of IFN-γ/STAT-1 has been suggested in the pathogenesis of chronic HCV hepatitis, whereas STAT-5 is important in controlling the expression of a wide range of hepatic genes that are essential for cellular function.36,37 Moreover, loss of STAT-5 causes liver fibrosis and cancer development through increased TGF-β and STAT-3 activation.39 Hence, during HAV infection, activation of STATs may mutually regulate each other, tightly controlling the development and progression of disease. STAT-5 has a recognized role in FOXP3 expression and Treg cell function, and an excessive release of TGF-β in the serum during acute viral infections has been demonstrated to inhibit antigen-specific T-cell activation and proliferation as a result of the regulation of Treg cell activity.40 Moreover, genetic variants in the TGF-β promoter related to higher plasma levels of TGF-β have recently been associated with a susceptibility to HAV infection among Mexican-American patients.41 Our data, in conjunction with the role of HO-1 on Treg cell development and function,20 raise the possibility that, during HAV infection, low CB levels may result in the secretion of TGF-β and phosphorylation of STAT-5, which, in turn, will permit Treg cell activity and an efficient control of the inflammatory process during HAV infection.

Information regarding the immunopathology associated with HAV infection is limited, particularly as it relates to the paediatric population. This population is the one with the greater risk of contracting infection. Therefore, the information obtained in this study is of great relevance in this area and supports the notion that, during HAV infection, the concentration of CB may modulate specific immune responses to ultimately define the outcome of the disease (Fig. 6). The data obtained in this study will help to elucidate the immunopathology involved in HAV infection, ant they are expected to help raise interest in further research on the topic.

Figure 6.

Figure 6

Open in a new tab

During hepatitis A virus (HAV) infection, conjugated bilirubin (CB) affected the cytokine profiles. The relationships between immune response, inflammatory processes, and disease outcomes during viral infections are complex. Herein, we proposed a model in which CB modulates immune functions during HAV infection in a dose-dependent manner, through actions that include transcriptional modulation by reducing signal transducer and activator of transcription-5 (STAT-5) phosphorylation at high levels of CB with the subsequent over-production of interleukin-6 (IL-6) and a pro-inflammatory cytokine profile and by modulating the number and/or activity of neutrophils to ultimately affect IL-8 secretion. This is consistent with an increased STAT-5 phosphorylation related to high expression of IL-8 and transforming growth factor-β (TGF-β) in HAV-infected individuals with low levels of CB and with a differential participation of nuclear factor-κB (NF-κB) during the process. The role of distinct immune cells including T lymphocytes, monocytes/macrophages, T regulatory and T helper type 17 cells during the process remains undefined.

Acknowledgments

The authors thank Lisbeth de Paz and Jesus Meza for technical assistance. This work was funded by grants from the Consejo Nacional de Ciencia y Tecnologia (CONACYT) #127229 and #188240 and the Consejo Estatal de Ciencia y Tecnologia de Jalisco (COECYTJAL) #849 to NAF. FPC, KC and MAS were supported by PhD scholarships from the CONACYT. The authors thank Veronica Yakoleff for editing the manuscript and for helpful comments.

Glossary

AST

aspartate aminotransferase

ALT

alanine aminotransferase

CB

conjugated bilirubin

HAVCR1

hepatitis A virus cellular receptor

HAV

hepatitis A virus

HBV

hepatitis B virus

HCV

hepatitis C virus

HEV

hepatitis E virus

HO-1

haem oxygenase-1

IFN

interferon

IL-12

interleukin-12

MCP-2

monocyte chemoattractant protein 2

NF-κB

nuclear factor-κB

PBLCs

peripheral blood lymphoid cells

STATs

signal transducers and activators of transcription factors

TFBS

transcription factor binding site

TFs

transcription factors

TGF

transforming growth factor

TNF

tumour necrosis factor

Treg

regulatory T cells

Disclosures

No competing financial interests exist.

References

  1. Martin A, Lemon SM. Hepatitis A virus: from discovery to vaccines. Hepatology. 2006;43:S164–72. doi: 10.1002/hep.21052. [DOI] [PubMed] [Google Scholar]
  2. Kotwal G, Cannon JL. Environmental persistence and transfer of enteric viruses. Curr Opin Virol. 2014;4C:37–43. doi: 10.1016/j.coviro.2013.12.003. [DOI] [PubMed] [Google Scholar]
  3. Escobedo-Melendez G, Fierro NA, Roman S, et al. Prevalence of hepatitis A, B and C serological markers in children from western Mexico. Ann Hepatol. 2012;11:194–201. [PubMed] [Google Scholar]
  4. Jacobsen KH, Wiersma ST. Hepatitis A virus seroprevalence by age and world region, 1990 and 2005. Vaccine. 2010;28:6653–7. doi: 10.1016/j.vaccine.2010.08.037. [DOI] [PubMed] [Google Scholar]
  5. Jeong S, Lee S. Hepatitis A: clinical manifestations and management. Intervirology. 2010;53:15–9. doi: 10.1159/000252779. [DOI] [PubMed] [Google Scholar]
  6. Corazza N, Badmann A, Lauer C. Immune cell-mediated liver injury. Semin Immunopathol. 2009;31:267–77. doi: 10.1007/s00281-009-0168-1. [DOI] [PubMed] [Google Scholar]
  7. Racanelli V, Rehermann B. The liver as an immunological organ. Hepatology. 2006;43:S54–62. doi: 10.1002/hep.21060. [DOI] [PubMed] [Google Scholar]
  8. Protzer U, Maini MK, Knolle PA. Living in the liver: hepatic infections. Nat Rev Immunol. 2012;12:201–13. doi: 10.1038/nri3169. [DOI] [PubMed] [Google Scholar]
  9. Park SH, Veerapu NS, Shin EC, et al. Subinfectious hepatitis C virus exposures suppress T cell responses against subsequent acute infection. Nat Med. 2013;19:1638–42. doi: 10.1038/nm.3408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Zhou Y, Callendret B, Xu D, et al. Dominance of the CD4+ T helper cell response during acute resolving hepatitis A virus infection. J Exp Med. 2012;209:1481–92. doi: 10.1084/jem.20111906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Lanford RE, Feng Z, Chavez D, et al. Acute hepatitis A virus infection is associated with a limited type I interferon response and persistence of intrahepatic viral RNA. Proc Natl Acad Sci U S A. 2011;108:11223–8. doi: 10.1073/pnas.1101939108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fierro N, Castro-Garcia FP, Panduro A. Rethinking cytokine function during hepatitis A and hepatitis C infections. Adv Biosci Biotechnol. 2013;4:13–8. [Google Scholar]
  13. Manangeeswaran M, Jacques J, Tami C, et al. Binding of hepatitis A virus to its cellular receptor 1 inhibits T-regulatory cell functions in humans. Gastroenterology. 2012;142:1516–25.e3. doi: 10.1053/j.gastro.2012.02.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fierro NA, Escobedo-Melendez G, De Paz L, et al. Cytokine expression profiles associated with distinct clinical courses in hepatitis A virus-infected children. Pediatr Infect Dis J. 2012;31:870–1. doi: 10.1097/INF.0b013e318258e808. [DOI] [PubMed] [Google Scholar]
  15. Pashankar D, Schreiber RA. Jaundice in older children and adolescents. Pediatr Rev. 2001;22:219–26. doi: 10.1542/pir.22-7-219. [DOI] [PubMed] [Google Scholar]
  16. Woolbright BL, Jaeschke H. Novel insight into mechanisms of cholestatic liver injury. World J Gastroenterol. 2012;18:4985–93. doi: 10.3748/wjg.v18.i36.4985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Liu Y, Li P, Lu J, et al. Bilirubin possesses powerful immunomodulatory activity and suppresses experimental autoimmune encephalomyelitis. J Immunol. 2008;181:1887–97. doi: 10.4049/jimmunol.181.3.1887. [DOI] [PubMed] [Google Scholar]
  18. Khan NM, Poduval TB. Immunomodulatory and immunotoxic effects of bilirubin: molecular mechanisms. J Leukoc Biol. 2011;90:997–1015. doi: 10.1189/jlb.0211070. [DOI] [PubMed] [Google Scholar]
  19. Weinberger B, Archer FE, Kathiravan S, et al. Effects of bilirubin on neutrophil responses in newborn infants. Neonatology. 2013;103:105–11. doi: 10.1159/000343097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Rocuts F, Zhang X, Yan J, et al. Bilirubin promotes de novo generation of T regulatory cells. Cell Transplant. 2010;19:443–51. doi: 10.3727/096368909X484680. [DOI] [PubMed] [Google Scholar]
  21. Haga Y, Tempero MA, Kay D, et al. Intracellular accumulation of unconjugated bilirubin inhibits phytohemagglutin-induced proliferation and interleukin-2 production of human lymphocytes. Dig Dis Sci. 1996;41:1468–74. doi: 10.1007/BF02088574. [DOI] [PubMed] [Google Scholar]
  22. Dozmorov MG, Kyker KD, Hauser PJ, et al. From microarray to biology: an integrated experimental, statistical and in silico analysis of how the extracellular matrix modulates the phenotype of cancer cells. BMC Bioinformatics. 2008;9(Suppl 9):S4. doi: 10.1186/1471-2105-9-S9-S4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Wingender E, Chen X, Hehl R, et al. TRANSFAC: an integrated system for gene expression regulation. Nucleic Acids Res. 2000;28:316–9. doi: 10.1093/nar/28.1.316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. O’Shea JJ, Holland SM, Staudt LM. JAKs and STATs in immunity, immunodeficiency, and cancer. N Engl J Med. 2013;368:161–70. doi: 10.1056/NEJMra1202117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Reich NC. STATs get their move on. JAKSTAT. 2013;2:e27080. doi: 10.4161/jkst.27080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. O’Shea JJ, Gadina M, Kanno Y. Cytokine signaling: birth of a pathway. J Immunol. 2011;187:5475–8. doi: 10.4049/jimmunol.1102913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Khan NM, Poduval TB. Bilirubin augments radiation injury and leads to increased infection and mortality in mice: molecular mechanisms. Free Radic Biol Med. 2012;53:1152–69. doi: 10.1016/j.freeradbiomed.2012.07.007. [DOI] [PubMed] [Google Scholar]
  28. Rigato I, Ostrow JD, Tiribelli C. Bilirubin and the risk of common non-hepatic diseases. Trends Mol Med. 2005;11:277–83. doi: 10.1016/j.molmed.2005.04.008. [DOI] [PubMed] [Google Scholar]
  29. Ryter SW, Choi AM. Heme oxygenase-1/carbon monoxide: from metabolism to molecular therapy. Am J Respir Cell Mol Biol. 2009;41:251–60. doi: 10.1165/rcmb.2009-0170TR. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Zhou J, Tracy TS, Remmel RP. Bilirubin glucuronidation revisited: proper assay conditions to estimate enzyme kinetics with recombinant UGT1A1. Drug Metab Dispos. 2010;38:1907–11. doi: 10.1124/dmd.110.033829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Chandra V, Huang P, Hamuro Y, et al. Structure of the intact PPAR-γ-RXR-nuclear receptor complex on DNA. Nature. 2008;456:350–6. doi: 10.1038/nature07413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Costa RH, Kalinichenko VV, Holterman AX, et al. Transcription factors in liver development, differentiation, and regeneration. Hepatology. 2003;38:1331–47. doi: 10.1016/j.hep.2003.09.034. [DOI] [PubMed] [Google Scholar]
  33. Hoesel B, Schmid JA. The complexity of NF-κB signaling in inflammation and cancer. Mol Cancer. 2013;12:86. doi: 10.1186/1476-4598-12-86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pan MG, Xiong Y, Chen F. NFAT gene family in inflammation and cancer. Curr Mol Med. 2013;13:543–54. doi: 10.2174/1566524011313040007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tang Y, Wei Y, He W, et al. GATA transcription factors in vertebrates: evolutionary, structural and functional interplay. Mol Genet Genomics. 2014;289:2013–14. doi: 10.1007/s00438-013-0802-4. [DOI] [PubMed] [Google Scholar]
  36. Adamson AS, Collins K, Laurence A, et al. The current STATus of lymphocyte signaling: new roles for old players. Curr Opin Immunol. 2009;21:161–6. doi: 10.1016/j.coi.2009.03.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Gao B, Wang H, Lafdil F, et al. STAT proteins – key regulators of anti-viral responses, inflammation, and tumorigenesis in the liver. J Hepatol. 2012;57:430–41. doi: 10.1016/j.jhep.2012.01.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kang K, Robinson GW, Hennighausen L. Comprehensive meta-analysis of Signal Transducers and Activators of Transcription (STAT) genomic binding patterns discerns cell-specific cis-regulatory modules. BMC Genomics. 2013;14:4. doi: 10.1186/1471-2164-14-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Hosui A, Kimura A, Yamaji D, et al. Loss of STAT5 causes liver fibrosis and cancer development through increased TGF-β and STAT3 activation. J Exp Med. 2009;206:819–31. doi: 10.1084/jem.20080003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Passerini L, Allan SE, Battaglia M, et al. STAT5-signaling cytokines regulate the expression of FOXP3 in CD4+ CD25+ regulatory T cells and CD4+ CD25– effector T cells. Int Immunol. 2008;20:421–31. doi: 10.1093/intimm/dxn002. [DOI] [PubMed] [Google Scholar]
  41. Zhang L, Yesupriya A, Hu DJ, et al. Variants in ABCB1, TGFB1, and XRCC1 genes and susceptibility to viral hepatitis A infection in Mexican Americans. Hepatology. 2012;55:1008–18. doi: 10.1002/hep.25513. [DOI] [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES