fax: (31) 10-4632793
Favorable effect of adefovir on the number and functionality of myeloid dendritic cells of patients with chronic HBV†
Article first published online: 27 SEP 2006
Copyright © 2006 American Association for the Study of Liver Diseases
Volume 44, Issue 4, pages 907–914, October 2006
How to Cite
van der Molen, R. G., Sprengers, D., Biesta, P. J., Kusters, J. G. and Janssen, H. L. A. (2006), Favorable effect of adefovir on the number and functionality of myeloid dendritic cells of patients with chronic HBV. Hepatology, 44: 907–914. doi: 10.1002/hep.21340
Potential conflict of interest: Nothing to report.
- Issue published online: 27 SEP 2006
- Article first published online: 27 SEP 2006
- Manuscript Accepted: 17 JUL 2006
- Manuscript Received: 11 APR 2006
- Gilead Sciences, Milan, Italy
- Netherlands Organization of Scientific Research (NWO)
In patients with chronic hepatitis B virus (HBV), 2 predominant precursor dendritic cell (DC) subtypes, the myeloid dendritic cell (mDC) and the plasmacytoid dendritic cell (pDC), were recently found to be functionally impaired. HBV DNA was found to be present in the DC subtypes, but no viral replication could be detected. The question remains whether simply the presence of the virus and viral proteins causes this dysfunction of DCs. To address this issue, the effect of viral load reduction resulting from treatment with the nucleotide analogue adefovir dipivoxil on the number and functionality of circulating DCs was studied during 6 months of treatment. Treatment resulted in a mean 5 log10 decrease in the viral load and normalization of alanine aminotransferase within 3 months. The number of mDCs, but not of pDCs, increased significantly over 6 months of treatment to a level comparable to that of uninfected healthy controls. The allostimulatory capacity of isolated and in vitro matured mDCs increased significantly after 3 months of treatment. Accordingly, mDCs exhibited an increased capacity to produce tumor necrosis factor alpha and interleukin-12 after 3-6 months of treatment. There was no change in interferon alpha production by pDCs during treatment. In conclusion, adefovir treatment results in an improvement in the number and functionality of mDCs, but not of pDCs. Our findings provide clues for the reasons why current antiviral therapy does not lead to consistently sustained viral eradication. (HEPATOLOGY 2006;44:907–914.)
Today, more than 400 million people are chronically infected with hepatitis B virus (HBV) and are at risk of developing cirrhosis or hepatocellular carcinoma.1, 2 Chronic HBV infection is the result of a complex interaction between a replicating noncytopathic virus and a down-regulated antiviral immune response. To recover from an HBV infection, both strong humoral and cellular immune responses are required. During an acute infection, class I– and class II–restricted T-cell responses to the virus are vigorous, polyclonal, and multispecific.3–5 Such responses are relatively weak and more narrowly focused in chronically infected patients.6, 7
Dendritic cells (DCs) play an important role in the induction of specific T-cell responses. We recently demonstrated functional impairment of the myeloid dendritic cells (mDCs) and plasmacytoid dendritic cells (pDCs) of patients with chronic HBV infection compared to healthy volunteers.8 We showed reduced maturation, T-cell stimulatory capacity, and tumor necrosis factor alpha (TNF-α) production by mDCs and reduced interferon alpha (IFN-α) production by pDCs. In addition, we have recently shown that regulatory T cells contribute to the impaired immune response in patients with chronic HBV.9 These regulatory T cells were found to suppress HBV-specific T-cell responses. Regulatory T cells can be induced by immature DCs.10 Together, these data suggest an important link between impaired DC function and insufficient T-cell response in chronic HBV infection.
The mechanism by which HBV affects the function of DCs is not clear. Although viral particles and HBV replication intermediates were shown to be present in monocyte-derived DCs of chronic HBV patients,11, 12 we could only detect HBV DNA in the precursor DC subsets, not HBV RNA replication intermediates, suggesting HBV is not replicating.13 Indeed, Untergasser et al. have recently shown that circulating DCs may take up HBV antigens but support neither nucleocytoplasmic transport nor replication.14 Therefore, the question remains how HBV causes dysfunction of the DC subtypes. Is it the presence of the virus or viral proteins in circulation or in the DCs themselves? It is well established that antigen dose and distribution critically affect the induction of T-cell responses.15 Reducing the viral load by inhibition of viral replication may provide insight in the effect of high virus titers on the functionality of DCs. Adefovir dipivoxil, a nucleotide analogue, is a potent inhibitor of HBV replication of wild-type and lamivudine-resistant viruses and has recently been approved for the treatment of chronic HBV infection.16, 17 The aim of this study was to determine the number, phenotype, and functionality of mDC and pDC precursor subsets in peripheral blood of patients with chronic HBV before and during antiviral therapy with adefovir.
Patients and Methods
Twelve patients with chronic HBV (10 male and 2 female) were treated with adefovir dipivoxil (10 mg orally per day). Patient parameters at baseline are shown in Table 1. The median age of the patients was 51 (range 28-67). At baseline the median serum HBV DNA level was 1.3 × 108 geq/mL (range 2.3 × 104-2.1 × 109), and median alanine aminotransferase (ALT) level was 105 U/L (range 39-467). Seven patients were hepatitis B e antigen (HBeAg) positive at baseline, and all patients had biopsy-proven chronic hepatitis with various degrees of liver inflammation and fibrosis or cirrhosis. Fibrosis was scored according to the Ishak system.18 Patients were negative for antibodies against human immunodeficiency virus, hepatitis C, and hepatitis D. None of the patients received any antiviral or immunosuppressive medication in the 6 months prior to starting adefovir treatment. At baseline and after 3 and 6 months of treatment, heparinized peripheral blood samples were taken for virological analysis and isolation of peripheral blood mononuclear cells (PBMCs). All patients gave informed consent before blood donation. Data obtained on the numbers of circulating mDCs and pDCs of the patients were compared with those obtained from 12 age- and sex-matched healthy controls (9 male and 3 female) with a median age of 48 (range 27-53).
|Age (Years)/Sex||Race||ALT (U/L)||HBV DNA (geq/mL)||HBeAg||Ishak Fibrosis Score||Previous AVT|
|1||34/m||Asian||74||2.10 × 109||pos||1||entecavir|
|2||28/m||Asian||160||2.29 × 108||pos||1||Peg-IFN + lam|
|3||34/f||Asian||100||1.00 × 108||pos||2||none|
|4||52/m||Caucasian||110||1.77 × 105||neg||2||none|
|5||50/m||Caucasian||204||1.65 × 108||neg||2||Peg-IFN + lam|
|6||58/m||Caucasian||80||1.73 × 108||pos||3||entecavir|
|7||61/m||Caucasian||467||1.62 × 109||pos||4||none|
|8||55/m||Caucasian||62||6.66 × 106||neg||4||none|
|9||67/m||Caucasian||128||2.32 × 104||neg||5||Peg-IFN + lam|
|10||52/f||Caucasian||133||1.62 × 109||pos||5||none|
|11||49/m||Asian||62||2.40 × 104||neg||6||none|
|12||41/m||Asian||39||3.26 × 106||pos||6||Peg-IFN + lam|
Serum hepatitis B surface antigen (HBsAg), HBeAg, and anti-HBe were determined quantitatively using the IMX system (Abbot Laboratories, North Chicago, IL) according to the manufacturer's instructions. Serum HBV DNA was determined using an HBV monitor assay (Roche Applied Science, Penzberg, Germany; detection limit 1 × 103 geq/mL). When serum HBV DNA was below 1 × 103 geq/mL HBV DNA, the assay was repeated using an in-house-developed TaqMan PCR (detection limit 373 geq/mL) (19).
Analysis of Frequencies and Absolute Numbers of Myeloid and Plasmacytoid Dendritic Cells Subsets in Peripheral Blood.
To determine the number of mDCs, 100 μL of whole blood was incubated with CD45-FITC and CD19-PerCP (Becton Dickinson, San Jose, CA) and anti-BDCA1 (CD1c)-PE (Miltenyi Biotec, Bergisch Gladbach, Germany). For pDCs, 100 μL of whole blood was incubated with CD45-FITC, anti-BDCA4-PE (Miltenyi Biotec), and anti-CD123-biotin (Becton Dickinson) and subsequently, in a secondary step, with strepavidin-PerCP (Becton Dickinson). After staining, red blood cells were lysed with FACS lysing solution (Becton Dickinson). Stained cells were analyzed using a 4-color flow cytometer (FACScalibur, Becton Dickinson) and CellQuest software. At least 50,000-100,000 events per run were acquired. CD45+ cells were gated to determine the percentage of mDCs (CD19− and BDCA1+) and pDCs (CD123+ and BDCA4+). The absolute numbers of mDCs and pDCs in circulating blood were calculated using the percentage of cells relative to the mononuclear cell count, as determined by an automated differential blood count.
Isolation of Dendritic Cells.
PBMCs were obtained by Ficoll-Isopaque gradient centrifugation to isolate mDCs and pDCs. mDCs (BDCA1+) and pDCs (BDCA4+) were isolated by positive immunomagnetic selection using the mini-MACS system (Miltenyi Biotec) as previously described.8 The isolated mDCs and pDCs were analyzed for purity by flow cytometry, and fractions were only used if they were more than 90% pure.
Expression of Cell Surface Molecules on Surfaces of Myeloid Dendritic Cells Determined by Flow Cytometry.
Freshly isolated mDCs (1 × 104 cells) were stained with CD80-FITC (Immunotech, Marseille, France), CD86-APC (Becton Dickinson), or CD40-APC (Becton Dickinson). Cells were stained with the corresponding isotype-matched control monoclonal antibodies as controls. Cells were analyzed on a flow cytometer, and mean fluorescence intensity was determined.
Analysis of T-Cell Stimulatory Capacity of Myeloid Dendritic Cells in a Mixed Lymphocyte Reaction.
Purified mDCs were matured for 24 hours in 96-well flat-bottomed culture plates at different concentrations (1.25, 2.5, 5, or 10 × 103 cells/200 μL) in medium consisting of RPMI 1640 (Bio Whittaker, Verviers, Belgium) with 10% fetal calf serum (Hyclone, Logan, UT), 50 ng/mL interleukin 1beta (IL-1β; Strathmann Biotech, Hannover, Germany), 25 ng/mL TNF-α (Strathmann), and 500 U/mL granulocyte/macrophage colony-stimulating factor (GM-CSF; Leucomax, Novartis Pharma, Arnhem, The Netherlands). The next day culture supernatant was removed, and nylon-wool purified T cells from a normal healthy volunteer (1.5 × 105 cells/200 μL) were added to the DCs. After 5 days, cell proliferation was assessed by the incorporation of [3H]thymidine (Radiochemical Centre, Amersham, Little Chalfont, UK). Then 0.5 μCi/well was added, and cultures were harvested 18 hours later. As a positive control, 5 μg/mL phytohemagglutin (Murex, Paris, France) was added to T cells.
Cytokine Production of Stimulated Myeloid and Plasmacytoid Dendritic Cells.
Purified mDCs were stimulated at a concentration of 4 × 104 cells/200 μL in 96 flat-bottomed Costar plates (Costar, Cambridge, MA) in culture medium (RPMI 1640 with 10% fetal calf serum) with synthetic double-stranded (ds)RNA, polyriboinosinic-polyribocytidylic acid (poly[I:C], 20 μg/mL; Sigma-Aldrich, St. Louis, MO) in combination with recombinant human IFN-γ (1,000 U/mL; Strathmann) and in the presence of GM-CSF (500 U/mL). Purified pDCs were stimulated at a concentration of 2 × 104 cells/200 μL with Staphylococcus aureus Cowan strain I (SAC; 75 μg/mL; Calbiochem, San Diego, CA) and in the presence of IL-3 (10 ng/mL; Strathmann). Cells were cultured at 37°C and 5% CO2 in a humidified incubator. Supernatants were harvested after 24 hours. IL-12p70 (Diaclone, Besançon, France) and IFN-α (Biosource International, Nivelles, Belgium) levels were determined by a standard ELISA according to the manufacturers' instructions. TNF-α and IL-10 levels were determined with a specific solid-phase sandwich ELISA, using monoclonal antibodies and recombinant cytokine standards from Biosource International.
In Vitro Effect of Adefovir on T-Cell Stimulatory Capacity of Myeloid Dendritic Cells.
Purified mDCs isolated from untreated HBV patients were matured for 24 hours in the presence or in the absence of the active metabolite of adefovir, 9-(2-phosphonylmethoxyethyl)adenine (PMEA; kindly provided by Gilead Sciences, Foster City, CA). Concentrations of 1, 10, or 100 ng/mL PMEA were added to the maturation cocktail. After 24 hours the mDCs were washed twice and analyzed for T-cell stimulatory capacity, as described before.
Data are expressed as means ± SEMs, unless indicated otherwise. Data were analyzed and compared to baseline results with SPSS 11.5 for Windows (SPSS, Chicago, IL) using a Wilcoxon matched-pairs signed rank sum test. The Mann-Whitney test was used to compare variables between two independent groups of patients. Correlations were determined using Spearman's correlation test or linear regression with multiple variables. In all analyses a P value of < .05 was considered statistically significant.
Serum HBV DNA and ALT Levels Are Reduced Following Treatment With Adefovir.
The HBV DNA and ALT levels of 12 patients who had started adefovir antiviral treatment were followed for 6 months. Figure 1 shows scatter plots of serum HBV DNA and ALT levels during treatment. The median HBV DNA showed a decrease from 1.3 × 108 to 3.1 × 103 geq/mL within the first 3 months of treatment, after which it remained stable for the next 3 months. The median ALT level decreased from 105 to 40 U/L within the first 3 months of treatment. Six months of adefovir treatment resulted in undetectable HBV DNA in 7 patients (58%) and in ALT normalization in 9 patients (75%). Three patients seroconverted from HBeAg to anti-HBeAg during treatment. None of the patients exhibited loss of serum HBsAg.
Circulating Number of Myeloid Dendritic Cells Is Increased Following Adefovir Treatment.
To determine whether treatment of chronic HBV patients with adefovir influenced circulating numbers of mDCs and pDCs in peripheral blood, the percentages and absolute numbers of mDCs (CD19− and BDCA1+) and pDCs (CD123+ and BDCA4+) in whole blood were determined by flow cytometry at different time points. The percentages (Fig. 2A) and absolute numbers (Fig. 2B) of mDCs increased significantly during therapy compared to baseline levels (P < .05). A significant inverse correlation between ALT and absolute number or frequency of mDCs was observed: r = 0.63 (P < .01) and r = 0.45 (P < .05), respectively. Compared to the 12 age-matched healthy controls, the frequency of mDCs was significantly lower at the start of treatment (P < .01) but was not significantly different after 6 months of treatment. Also, after 6 months of adefovir treatment the absolute number of mDCs of patients was almost completely restored to that of the normal healthy control levels. The absolute number of mDCs before treatment was inversely correlated with the Ishak fibrosis score (r = 0.61; P < .05).
Limited increases in the percentage (Fig. 2C) and absolute number (Fig. 2D) of pDCs was observed during treatment with adefovir. Compared to healthy controls, both the frequency and the absolute number of pDCs were significantly lower in patients at the start of treatment (P < .01), a difference that had disappeared after 6 months of treatment. The absolute number of pretreatment pDCs showed a trend toward being inversely correlated with the Ishak fibrosis score (r = 0.50; P = .11).
Allostimulatory Capacity of Myeloid Dendritic Cells of Chronic HBV Patients Is Increased During Adefovir Treatment.
We previously showed that mDCs of patients with chronic HBV have a decreased capacity to stimulate allogeneic T cells compared to the capacity of mDCs in healthy controls.8 To investigate whether antiviral treatment restores the allostimulatory capacity of mDCs, these cells were isolated at baseline and during adefovir therapy and matured in vitro in the presence of IL-1β and TNF-α. Different numbers of mature mDCs were added to T cells of a healthy third party. Figure 3 shows that mDCs isolated from patients treated for 3 months exhibited an increased capacity to stimulate T cells compared to baseline mDCs at all ratios tested (P < .001). After 6 months of treatment the capacity decreased slightly; however, this was still significantly different from baseline (P < .05). A clear trend was observed toward correlation between the increase in allostimulatory capacity and decrease in HBV DNA (r = 0.73; P = .15), but not between the increase in allostimulatory capacity and decrease in ALT. Background proliferation of T cells was <500 cpm. The proliferation of phytohemagglutinin-stimulated T cells was similar in all assays.
Expression of Costimulatory Molecules CD80, CD86, and CD40 on Matured Myeloid Dendritic Cells Does Not Change during Treatment With Adefovir.
To determine the expression of costimulatory molecules on mature mDCs, flow-cytometry staining was performed. Figure 4 shows the mean fluorescence intensity of CD80, CD86, and CD40 of these matured mDCs at baseline and during treatment with adefovir. No significant differences were observed in the mean fluorescence intensity of CD80, CD86, and CD40 on matured mDCs at the time points investigated.
Capacity of Myeloid and Plasmacytoid Dendritic Cells to Produce Cytokines Is Modulated During Treatment With Adefovir.
To investigate the influence of reduced virus titers on the capacity of the 2 DC subtypes to produce cytokines, mDCs were stimulated with poly(I:C) and IFN-γ, and pDCs were stimulated with SAC. We determined the production of IL-12p70, TNF-α, and IL-10 by mDCs and the production of IFN-α, TNF-α and IL-10 by pDCs (Fig. 5). The capacity of mDCs to produce IL-12 following stimulation was significantly increased after 6 months of adefovir treatment, whereas that to produce IL-10 was significantly decreased (P < .05). TNF-α production was significantly increased after 3 months (P < .05; Fig. 5). A trend toward correlation between decreased viral load, but not ALT, and increased TNF-α production was observed (r = 0.57; P = .07). No correlation was observed for IL-10 or IL-12 and HBV DNA or ALT.
For pDCs significantly decreased production of TNF-α was found compared to baseline (P < .05). No significant differences were observed in the capacity of pDCs to produce IFN-α and IL-10 following treatment with adefovir (Fig. 5).
Adefovir Does Not Directly Affect the T-Cell Stimulatory Capacity of Myeloid Dendritic Cells In Vitro.
The observed effect of adefovir treatment on DC function could be the result of direct interaction of adefovir with mDC. To investigate this, the in vitro effect of adefovir on the T-cell stimulatory capacity of mDCs was studied. Purified mDCs isolated from 4 untreated patients with chronic HBV were matured for 24 hours in the presence of different concentrations of PMEA, the active metabolite of adefovir. No effect was seen on the T-cell stimulatory capacity of the mDCs.
DCs are professional antigen-presenting cells, and the most potent cell type in initiating the primary immune response. Previously, we demonstrated significant functional impairment of the mDCs and pDCs of patients with chronic HBV infection compared with healthy volunteers and patients with chronic inflammatory liver disease of nonviral origin.8 It is unknown by which mechanism DC subsets have been functionally impaired. Because patients with chronic inflammatory liver disease of nonviral origin showed normal DC function, most likely the virus itself is responsible for the dysfunction.8 HBV DNA was detected in mDCs and pDCs, but this may also reflect attachment of the virus to the cell surface or uptake of the virus as a biological function of DC. HBV RNA replication intermediates could not be detected in DC subsets of chronic HBV patients, indicating HBV is not replicating in the DC subsets.13 Chronic HBV patients have large amounts of HBV particles and HBV antigens circulating, which could play a role in the induced functional impairment of DCs. It has been shown previously that HBeAg as well as HBsAg can induce T-cell tolerance in a mouse model.20–22 In addition, we recently demonstrated that purified HBsAg and HBV particles, obtained from the HepG2.2.15 cell line, can functionally impair the capacity of mDCs to stimulate allogeneic T cells.13
The data presented here indicate that during antiviral treatment with subsequent reduction of the viral load, the circulating number of mDCs increases to a level similar to that of normal healthy controls. The increased number of mDCs was correlated with a decrease in ALT. In addition, both the T-cell stimulatory capacity and the capacity to produce IL-12 and TNF-α of mDC increased during treatment with adefovir. During therapy this functional recovery of mDCs followed the same pattern as viral load reduction. In addition, we showed with in vitro experiments that PMEA, the working metabolite of adefovir, did not alter the T-cell stimulatory function of mDCs directly. Together, these data suggest the reduction of the viral load in chronic HBV patients positively affects the function of mDCs to stimulate a T-helper 1–type immune response.
In this study we have shown that patients with chronic HBV who have an indication for antiviral therapy have reduced pretreatment mDCs and pDCs as compared to healthy controls. The number of mDCs was inversely correlated with the Ishak fibrosis score. In our previous study we did not find reduced numbers of mDCs and pDCs in chronic HBV patients as compared to healthy controls.8 However, patients studied previously all had minimal to moderate fibrosis, whereas the current patients had more advanced fibrosis. The number of circulating DCs largely depends on the extent of mobilization of unprimed DCs to inflamed tissue. During adefovir therapy, in parallel to decreasing HBV DNA levels and antigen exposure, liver inflammation is reduced as reflected by ALT level. Therefore, it is possible that the increased number of circulating DCs during adefovir therapy primarily reflects reduced migration of DCs to the liver. This has been described for a mouse model of granulomatous liver disease.23 Also, a recent study by Kunitani et al. showed that in patients with chronic liver disease the number of circulating mDCs was inversely correlated with serum ALT level.24
The increase in T-cell stimulatory function that occurs during treatment does not seem to depend on the expression of costimulatory molecules. This suggests other factors such as reduced expression of the inhibitory costimulatory molecules ICOS ligand, PD-L1 (B7-H1), and PD-L2 (B7-DC) could be important in the decreased allostimulatory capacity.25 Furthermore, the capacity of poly(I:C)/IFN-γ-stimulated mDCs to produce TNF-α was significantly increased after 3 months compared to baseline. This increased capacity to produce TNF-α could be involved in the increased allostimulatory capacity of mDCs. TNF-α is known to contribute to allostimulation acting as an autocrine growth factor for DC-induced T-cell proliferation.26 The slight decreases in the allostimulatory capacity and the capacity to produce TNF-α after 6 months of treatment compared to 3 months could be explained by the mDCs being temporarily activated during the first months of treatment because of the release in antigenic pressure as a consequence of the prominent drop in viral load. An increase in the number of mDCs did not play a role in this, as the number of mDCs per well was equal at all time points studied.
The current data suggest that a significant reduction in the viral load by antiviral therapy increases the capacity of mDCs to produce IL-12 and decreases their capacity to produce IL-10. IL-12 is very important for stimulating natural killer cells, T lymphocytes to produce IFN-γ, and it thereby supports the induction of the T-helper 1–type immune response.27, 28 IL-10, on the other hand, is capable of inhibiting these responses.29 The combined effect of the observed increase in IL-12 production and reduction in IL-10 production by mDCs would suggest they cooperate in achieving sustained control over HBV replication and stimulate a T-helper 1–type immune response. Indeed, we have shown that HBV-specific T-cell responses increase after 6 months of adefovir treatment.30 In addition, Boni et al. have shown that both HLA class II–restricted T-cell responses and cytotoxic T-cell responses were recovered in association with lamivudine-induced reduction of viral load.31, 32
Recently, it has been shown that the capacity of the pDCs of chronic HBV patients to produce IFN-α was severely reduced as compared to healthy controls.8 In the present study, we showed that this capacity did not increase during treatment with adefovir. Instead, the cytokine profile of pDCs during treatment showed a trend toward activating a T-helper 2–type immune response, with increased capacity to produce IL-10 and decreased production of TNF-α. This suggests pDC function may counterbalance the capacity of mDCs to stimulate T-helper 1–type immune responses and subsequently may inhibit HBV-specific immune responses. Together these findings may provide some clues about why the current antiviral therapy does not lead to a consistently sustained viral eradication.
In conclusion, our results show that adefovir treatment of chronic HBV patients affects the number and functionality of mDCs, but not of pDCs. New immunomodulatory antiviral therapies that also target pDCs may be the key to a vigorous and sustained antiviral response.
The authors thank B. Hansen for statistics. Furthermore, we thank W. F. Leemans, M. J. ter Borg, A. Keizerwaard, L. A. van Santen-van der Stel, and C. van de Ent-van Rij for their assistance in obtaining peripheral blood samples and Dr. L. J. W. van der Laan for critical reading of the manuscript.
- 13Functional impairment of mDC in chronic HBV patients: the role of HBV proteins. HEPATOLOGY 2005; 42 (suppl 1): 712A., , , , .
- 30Adefovir treatment of chronic hepatitis B decreases regulatory T cells and partially restores the antiviral immune response. HEPATOLOGY 2005; 42(suppl 1): 714A., , , , , .