A Clinical Study of HBsAg-Activated Dendritic Cells and Cytokine-Induced Killer Cells During the Treatment for Chronic Hepatitis B

Authors


Correspondence to: X.-Y. Ji, Henan Provincial Key Medical Laboratory for Cellular and Molecular Immunology, School of Medicine, Henan University, Kaifeng Henan 475004, China. E-mail: xinying_ji@henu.edu.cn

Abstract

We aim to study the therapeutic effects of HBsAg-activated DCs and cytokine-induced killer (CIK) cells as adoptive immunotherapy in patients with Chronic Hepatitis B (CHB). Autologous HBsAg-activated DC–CIK cells were infused into patients with CHB to evaluate their effect on HBV-DNA, HBsAg, ALT, etc. The viral load in the treatment group decreased significantly (P < 0.001), while that in the control group did not decrease (P > 0.05). Twenty-one patients (63.6% efficiency) in the treatment group had a viral response (≥2 log decrease in viral load), while four patients (14.8% efficiency) from the control group had a viral response. There were significant differences in the viral responses of the two groups (the control group 63.6% versus the control group 14.8%, P < 0.001). We concluded that the immunity was enhanced after HBsAg activation in DCs and CIK cells. Reinfusion of autologous HBsAg-activated DC–CIK cells inhibited HBV proliferation in 21 of 33 (63.6%) patients.

Introduction

Chronic hepatitis B (CHB) infection is a serious medical problem in China. Ninety-three million people are infected, and 20 million have CHB. Interferon and nucleoside analogues are first-line antivirus treatment, but the efficacy is poor due to the related immunodeficiency and the existence of stable covalently closed circular DNA (cccDNA) in the nuclei of hepatocytes. In recent years, dendritic cell (DC) and cytokine-induced killer (CIK) cell research in patients with hepatitis B has identified new potential treatments for chronic hepatitis [1]. The T cell response to HBV is weak and is narrowly focused in chronically infected patients [2], suggesting that it may be a cause of persistent infection. HBV-specific helper and cytotoxic T lymphocytes (CTLs) are barely detectable in peripheral blood of patients with CHB [3], and DCs are specialized antigen-presenting cells that orchestrate immune responses. They stimulate innate and acquired immune responses, but also act as tolerogenic cells for immune responses in a variety of situations. In viral hepatitis, dysfunction of DCs from peripheral blood has been reported. The mechanism of impairment of DC function in patients with CHB is unclear. Immunotherapy with DCs which is a promising method for CHB has been applied in both animals and humans. In recent study, injection of activated DCs loaded with HBV peptide or protein achieved a reduction in HBV-DNA level in some patients. HBeAg negativity was achieved in more than half of the treated patients in one study [4, 5].The in vivo activation of DCs could be an alternative way for inducing antiviral immune responses including possible activation of CTLs against HBV.

In this study, DCs isolated from the peripheral blood of patients with CHB and healthy subjects were cultured, activated with HBsAg before maturation and then cocultured with cytokine-induced killer cells. We evaluated the DC and CIK phenotypes, their ability to secrete cytokines and CIK cell-killing ability. We evaluated the therapeutic effects of HBsAg-activated DC and CIK cells in patients with CHB after autologous transfusion, exploring the possibility of using DC and CIK for clinical CHB cellular immune therapy.

Materials and methods

Study subjects

All study subjects were adults. Patients with CHB were diagnosed according to diagnostic criteria specified in ‘chronic hepatitis B prevention and treatment guidelines’ (2010 edition) [6], that is, serum HBsAg positive, HBeAg positive and HBcAb positive. Cirrhosis, liver cancer, autoimmune hepatitis, alcoholic hepatitis, drug-induced hepatitis and other liver diseases were excluded from the study. The study group included 15 patients with CHB. There were nine men and six women, with an average age of 33.6 years (range: 20–45). The control group included 15 healthy subjects with no hepatitis B virus markers. There were eight men and seven women, with an average of 34.2 years (range: 18–48). The clinical cell therapy group included 33 patients. There were 18 men and 15 women, with an average age of 31.2 years (range: 20–45 years). The clinical therapy control group included 27 patients. There were 15 men and 12 women, with an average age of 35.5 years (range: 25–48 years). This control group only received symptomatic treatment. They did not receive antiviral or immunosuppressive therapy before CIK cell therapy. The study protocol was approved by the hospital ethics committee. An informed consent was signed by each study subject.

Reagents

AIM-V culture medium (Gibco BRL, Carlsbad, CA, USA), human lymphocyte isolation medium Ficoll (Beijing Solai Bao Technology Co., Ltd., Beijing, China), CD3 monoclonal antibody (Wuhan Institute of Biological Products, Wuhan, China), IFN-r (Kaicong Mau Biological, Shanghai, China), rhlL-2 (Liaoning satellite Institute of Biological Products LTD., Shenyang, China), rhGM-CSF, rhIL-1, rhIL-4 (Perprotech company, Suzhou, China), CCK-8 kit and IL-12 kit (Wuhan Boster Biological Engineering Co., Ltd., Wuhan, China) were used for these studies. The donated liver cell line HepG2.2.15 was a generous gift from Prof Liyun Zheng (Zhengzhou Institute of Liver Diseases, Zhengzhou, China). HBV-DNA was detected using the Agilent Technologies Stratagene Mx3000P real-time quantitative PCR detector (Santa Clara, CA, USA). Quantitative detection of HBsAg was performed using the LUMO laminator and related kits (Antu Green Branch Biological Engineering Co., Ltd, Zhengzhou, China). An Olympus CKX41 inverted phase contrast microscope was used to view cells. A Becton Dickinson flow cytometer was used to evaluate surface markers.

Cell culture and amplification of DC and CIK cells
Peripheral blood mononuclear cells

Peripheral blood mononuclear cells (PBMCs) were isolated from patients with CHB receiving cell therapy. Hundred millilitre of blood was collected from each patient. Hundred millilitre of peripheral blood was drawn from healthy subjects under sterile conditions. PBMCs were separated using a Ficoll gradient [7]. The resulting cell concentration was adjusted to 4~5 × 106/ml in AIM-V culture medium. The cells were then cultured at 37 °C in a 5% CO2 incubator for 2 h.

Induction and amplification of DCs

After PBMCs and MNC were incubated for 2 h, the non-adherent cells were collected for later use. AIM-V culture medium containing cytokines (with 1000 U/ml GM-CSF and 1000 U/ml IL-4) was added to the culture flasks containing adherent cells. The cells were placed in a 37 °C, 5% CO2 incubator. The medium was changed, and cytokines were added on the 3rd day. Cultured cells were harvested on the fifth day and equally divided into two flasks. One flask was incubated with hepatitis B surface antigen at a concentration of 1000 ng/ml. The second flask was incubated with a non-HBsAg stimulation control. The two cell treatment groups were cultured under the same conditions. On the 6th day, the cells were collected for testing or cocultured with CIK cells.

In vitro induced and amplified CIK cells

Non-adherent PBMCs and MNC collected after 2-h culture were adjusted with AIM-V medium to a cell concentration of 1~2 × 106/ml. Thousand unit per millilitre of interferon (IFN-γ) was added, and the mixture cultured at 37 °C in a 5% CO2 incubator for 24 h. Interleukin IL-2 (2000 U/ml), IL-1 (1000 U/ml) and anti-CD3 monoclonal antibody (1 ug/ml CD3McAb) were added, and the cells placed in the incubator. The medium was changed once every 3 days, supplied with rhIL-2, and the cell concentration adjusted to 1~2 × 106/ml.

Coculture of DCs and CIKs

On the 6th day, HBsAg-sensitized and non-sensitized DCs were harvested and mixed with CIK at a 1:5 ratio. The mixture was cultured for 3 days with AIM-V culture medium containing 2000 U/ml IL-2. These cells were used for studies in Section 1.4 and 1.5.

Phenotypes and the immune effects of DC and CIK cells
Cytotoxic testing

The DC–CIK cell mixture was cultured for 3 days. The cells were used as effector cells for HepG2.2.15 cells growing at the logarithmic growth phase. The concentration of the HepG2.2.15 cells was adjusted with the AIM-V culture solution to 1 × 105/ml. The DC–CIK cell concentration was adjusted to 5 × 105/ml.

Ninety-six-well culture plate studies: Hundred microlitre of target cells and 100 μl of effector cells (HBsAg-sensitized or HBsAg-sensitized DC–CIK cells) were added to each well. Hundred microlitre of effector cell suspension and 100 μl of AIM-V were added to effector wells. Hundred microlitre of target cell suspension and 100 μl of AIM-V medium were added to target cell wells. Each study condition was duplicated six times. The cells were cultured at 37 °C in a 5% CO2 incubator for 24 h. 20 μl of Cell Counting Kit (CCK-8; Boshide Biotech Inc, Wuhan, China) reagents was added to each well. Cells were incubated at 37 °C for 2 h, and an enzyme-linked immunosorbent assay (wavelength 450 nm) was used to detect absorbance (A). The rate of killing was calculated as: killing rate = [1 − (A experimental well − A effector well)/A target cell well] × 100%.

Detection of cell population phenotypes and observation of cell morphology

DC and CIK cell morphology was observed under an inverted phase contrast microscope on the 3rd and 6th culture days. Cell phenotype was evaluated on the 6th culture day using flow cytometry. DC–CIK cell cultures were evaluated after 48 h and 6 days.

Detection of cytokines

Supernatant from DC and CIK cocultures was collected after 48 h and cryopreserved at −70 °C for future use. IL-12 was detected using ELISA according to manufacturer's recommendation.

DC–CIK harvest and reinfusion

The DC–CIK cell suspension was split into two after 48 h. Half was centrifuged at 157 g for 5 min, the supernatant removed, and the cell pellet resuspended in 5× (v/v) normal saline. The cells were centrifuged again at 157 g for 5 min, and the supernatant removed. This procedure was repeated twice. The final cell pellet was resuspended in 100–150 ml normal saline containing 2% serum albumin and then injected into disposable blood collection bags.

The other half of the suspension was cultured in AIM-V medium containing 2000 U/ml IL-2. At least 109 cells were infused over 50–60 m on the 2nd and 3rd days, respectively, through a forearm vein. The transfusion was performed every 2 days for another three times.

Clinical observation

Fasting serum (before treatment and 4 weeks, 8 weeks, 12 weeks and 24 weeks after treatment) was collected from treatment and control groups and screened for HBV-DNA, HBsAg and liver biochemical biomarker levels. The patient's temperature and other vital signs and adverse events were evaluated within 24 h of cell infusion.

Statistical analysis

Measurement data were calculated as mean ± standard deviation. Rates were expressed as a percentage. A paired t-test was used to compare the same patients before and after treatment. The t-test was used to compare measurement data in the treatment and control groups. Fisher's exact test was used to compare rates in the two groups. SPSS 11.5 (Chicago, IL, USA) software was used for statistical testing. P < 0.05 was considered statistically significant.

Results

DC morphology and cell surface markers

DCs were observed as tufted growth, forming adherent colonies with varying sizes after 3 days of culture. On day 6, the majority of cells were detached and had an irregular surface. No significant differences were observed between cells from health subjects and patients with CHB (Fig. 1).

Figure 1.

Different stages of DCs. DC morphology was observed under an inverted phase contrast microscope on the 3rd and 6th culture days. Cell phenotype was evaluated on the 6th culture day using flow cytometry. DCs were observed as tufted growth, forming adherent colonies with varying sizes after 3 days of culture. On day 6, the majority of cells were detached and had an irregular surface. No significant differences were observed between cells from health subjects and CHB patients.

Flow cytometry was used to analyse DC surface markers. Healthy subjects had higher expression of the DC surface markers CD1a, CD80 and CD83 than that of patients with CHB (t = 12.824, 5.462 and 4.87 respectively, P < 0.001). There was no difference in HLA-DR expression of DCs found in the CHB group and healthy group (t = 0.502, P > 0.05). DCs from the HBsAg-sensitized healthy group had significantly greater expression of HLA-DR, CD1a, CD80 and CD83 than those from patients with CHB (t = 20.872, 5.295, 2.395 and 10.002, respectively, P < 0.05). HBsAg-sensitized DC from the health group had significantly higher expression of the four markers (HLA-DR, CD1a, CD80 and CD83) on DCs than that of non-HBsAg-sensitized DCs (t = 8.910, 7.055, 8.247 and 5.789, respectively, P < 0.01). There was no difference in expression of the four markers in HBsAg-sensitized and non-sensitized DCs from patients with CHB (t = 1.730, 1.565, 1.022 and 1.653, respectively, P > 0.05; Table 1).

Table 1. Comparison of cell surface biomarker expression
BiomarkerHealthy subject group (n = 15)CHB group (n = 18)
HBsAg (+)HBsAg (−)HBsAg (+)HBsAg (−)
CD1a31.83 ± 1.5225.10 ± 2.5017.48 ± 2.2715.65 ± 1.72
HLA-DR96.49 ± 8.6175.90 ± 7.3281.49 ± 7.6577.29 ± 8.43
CD8038.13 ± 4.8525.79 ± 3.1822.13 ± 2.3420.57 ± 2.31
CD8357.33 ± 7.4341.79 ± 7.2734.42 ± 5.7331.79 ± 3.54

CIK cell morphology and surface markers

On the 3rd day of culture, CIK cells were found to have a clustered, tufted clone-like growth. On the 6th day of culture, the majority of cells grew into colonies. Individual cell volume increased significantly over time, although there was variation in cell size and shape. Cells from healthy subjects and patients with CHB appeared similar (Fig. 2).

Figure 2.

Different stages of cytokine-induced killers (CIKs). CIK cell morphology was observed under an inverted phase contrast microscope on the 3rd and 6th culture days. Cell phenotype was evaluated on the 6th culture day using flow cytometry. On the 3rd day of culture, CIK cells were found to have a clustered, tufted clone-like growth. On the 6th day of culture, the majority of cells grew into colonies. Individual cell volume increased significantly over time, although there was variation in cell size and shape. Cells from healthy subjects and Chronic Hepatitis B patients appeared similar.

Flow cytometry analysis demonstrated more CD3/CD8 and CD3/CD56 expression in HBsAg-sensitized CIK cells from healthy subjects than non-sensitized CIK cells (t = 2.701, P < 0.01, t = 5.034, P < 0.001). There was no difference in similar cellular expression from patients with CHB (t = 0.822, 0.975, P > 0.05). Healthy subjects had greater expression of these markers on HBsAg-sensitized CIK cells than patients with CHB (t = 3.364, 4.352, P < 0.05; Table 2).

Table 2. Comparison of CD3/CD8 and CD3/CD56 expression in CIK cells
BiomarkerHealthy subject group (n = 15)CHB Group (n = 18)
HBsAg (+)HBsAg (−)HBsAg (+)HBsAg (−)
CD3/CD874.62 ± 14.2061.24 ± 12.960.31 ± 10.257.37 ± 11.25
CD3/CD5636.28 ± 6.4924.11 ± 6.7525.43 ± 7.6223.34 ± 4.98

Morphological findings in DC–CIK cultures

After coculture of DC–CIK for 48 h, clustered colony-like growth was observed (Fig. 3). The proliferation rate was significantly greater in cocultured cells than cells cultured alone. On the 6th day of coculture, both cell volumes and numbers were significantly greater than those in the initial culture.

Figure 3.

Different stages of DCs-CIKs. On the sixth day, HBsAg sensitized and non-sensitized DCs were harvested, and mixed with CIK at a 1:5 ratio. The mixture was cultured for 3 days with AIM-V culture medium containing 2000 U/ml IL-2. DC and CIK cell morphology was observed under an inverted phase contrast microscope. After co-culture of DC-CIK for 48 h, clustered colony-like growth was observed. The proliferation rate was significantly greater in co-cultured cells than cells cultured alone. On the 6th day of co-culture, both cell volumes and numbers were significantly greater than those in the initial culture.

Immune effects of DC and CIK

A CCK-8 colorimetric assay was used to evaluate ClK cytotoxicity when incubated with HBsAg-sensitized DC and HepG2.2.15 cells (Table 3). ClK cells incubated with HBsAg-sensitized DC from both healthy and patients with CHB had higher killing rates than those of non-sensitized CIK cells (t = 10.660, 9.115, P < 0.001). CIK cells incubated with non-sensitized DCs from healthy subjects had a higher killing rate than those from patients with CHB (t = 7.288, P < 0.001). CIK cells incubated with HBsAg-sensitized DCs from healthy subjects had a significantly higher killing rate than those from patients with CHB (t = 10.158, P < 0.001).

Table 3. The immune effects of CIK cells
BiomarkerHealthy subject group (n = 15)CHB group (n = 18)
HBsAg (+)HBsAg (−)HBsAg (+)HBsAg (−)
Killing rate (%)84.50 ± 8.3257.27 ± 5.3560.02 ± 5.4543.76 ± 5.18
IL-12 (pg/ml)55.36 ± 3.8137.56 ± 3.2429.86 ± 2.1327.58 ± 1.95

The supernatant IL-2 levels at 48 h were higher with cocultured CIK cells from healthy subjects incubated with HBsAg-sensitized DC than those from patients with CHB (t = 21.252, P < 0.0001). Supernatant IL-2 levels from non-sensitized CIK cells were higher in healthy controls than those in patients with CHB (t = 10.917, P < 0.0001). There was no difference in supernatant IL-2 levels between HBsAg-sensitized and unsensitized CIK cell cultures (t = 1.07, P > 0.05).

Changes in clinical indicators

There was no significant difference in HBV-DNA load between the treatment and control groups before treatment (t = 0.340, P > 0.05; Table 4). The HBV-DNA load of the DC–CIK cell treatment group was significantly lower than that of the control group (t = 2.775, P < 0.05) at 24 weeks after treatment. The viral load in the treatment group decreased significantly (t = 4.957, P < 0.001), while that in the control group did not decrease (t = 0.533, P > 0.05). Twenty-one patients (63.6% efficiency) in the treatment group had a viral response (≥2 log decrease in viral load). Four of these patients had no detectable HBV-DNA. Four patients (14.8% efficiency) from the control group had a viral response. No control patient had loss of HBV-DNA expression. There were significant differences in the viral responses between the two groups (P = 0.000).

Table 4. Clinical Indicators after HBsAg sensitized DC-CIK treatment
 Control group (n = 27)Therapy group (n = 33)
DNA (log)HBsAg (ng/ml)ALT (IU/l)TBIL (μmol/l)DNA (log)HbsAg (ng/ml)ALT (IU/l)TBIL (μmol/l)
0 weeks6.75 ± 1.12181.2 ± 79.2184.2 ± 106.139.2 ± 9.56.84 ± 1.06176.8 ± 112.7199.1 ± 108.238.7 ± 11.4
4 weeks6.52 ± 1.28168.5 ± 88.743.5 ± 11.426.5 ± 5.86.50 ± 1.15175.2 ± 106.138.8 ± 6.524.0 ± 6.5
8 weeks6.45 ± 0.85171.2 ± 86.237.4 ± 9.522.5 ± 6.16.27 ± 0.96180.5 ± 110.542.4 ± 7.620.2 ± 3.5
12 weeks6.50 ± 1.15162.0 ± 89.540.1 ± 6.923.0 ± 5.35.82 ± 1.12168.3 ± 96.440.8 ± 8.118.2 ± 4.8
24 weeks6.36 ± 1.7171.9 ± 91.339.8 ± 7.520.6 ± 4.55.29 ± 1.477153.0 ± 120.641.2 ± 5.520.5 ± 6.3

There was no significant change in HBsAg expression in patients with CHB at 24 weeks after treatment (t = 1.179, P > 0.05). HBsAg expression levels in the control group did not change after treatment (t = 0.399, P > 0.05). HBsAg expression did not change after DC–CIK cell infusion in the treatment and control groups (t = 0.169, 0.672, respectively, P > 0.05). Biochemical indicators of liver function in patients with CHB were significantly improved at 4 weeks after treatment. The ALT and TBIL levels did not improve after treatment (t = −0.484, 0.121, P > 0.05, respectively).

Adverse effects

No adverse effects were observed in the control group. Two patients receiving DC–CIK cells had a fever (37.9 °C and 38.4 °C, respectively). The fever resolved a few hours after antipyretic treatment.

Discussion

Specific cellular immunity is an important antiviral mechanism in the body. CTLs and DCs target HBV epitopes. DCs play a major role in cytotoxic mechanisms related to HBV clearance [8, 9]. DCs are the most important and powerful multiple functional antigen-presenting cells in vivo. DCs phagocytize hepatitis B virus, process them into antigenic peptides and present them to CD4+ and CD8+ T cells. CD4+ T cells are activated, proliferate, differentiate and secrete cytokines (IFN-γ, TNF-α, IL-12, etc.) that promote the development of Th0 cells to Th1. CD8+ T cells are activated, identify and combine with HBV-infected hepatic cells and promote hepatocyte cytolysis or apoptosis [10]. DCs activate NK cells and antibody-dependent cytotoxicity (ADCC) cells to destroy HBV-infected hepatocytes [11]. Any defects existing in cellular immunity can affect immune function, leading to chronic infection. The relationship of T lymphocyte and DC abnormalities with hepatitis B immunity is not clear [12].

HBsAg in the peripheral circulation does not induce a complete immune response in patients with CHB. Suppression of HBV replication and reduction in serum viral load and HBeAg levels and HbeAg/anti-HBe serum conversion have been reported after infusion of in vitro HBV-sensitized DCs [13]. The majority of virus-specific T cells isolated from the peripheral blood of patients with CHB have lost their ability to proliferate and produce cytokines. In vitro activated cells can play a role in HBV clearance. CIK cells are a group of heterogeneous mononuclear cells that are produced after culture with a variety of cytokines. CIK cells have powerful antitumour and antiviral T cell functions. Su HB et al. [14] used CIK cells to treat 33 HBV-DNA-positive patients with cirrhosis. HBV-DNA became undetectable in 12 patients after treatment. HBeAg became undetectable in 10 of 14 patients who were HBeAg positive pretreatment. The liver function was improved after treatment. No serious adverse effects were found. Marten et al. [15] found that coculture of DC and CIK cells from the peripheral blood markedly increased their proliferation capacity.

Immunotherapy of chronic HBV infection is a promising therapeutic strategy. There is controversy regarding the types of DC defects in chronic HBV patients and whether they can be improved after in vitro culture and stimulation [16, 17]. Similar to previous reports, we found that healthy subjects had higher DC expression of CD1a, CD80 and CD83 than patients with CHB (P < 0.05). We found no significant difference in cellular HLA-DR expression in patients with CHB and healthy subjects. Healthy subjects had greater expression of DC and CIK cell surface markers, IL-12 concentration in the cell culture supernatant, and greater ClK killing capacity to HepG2.2.15 cells than patients with CHB. HBsAg-sensitized DC/CIK cells from healthy subjects had greater expression of cell surface markers, greater IL-12 concentration in the supernatant of the coculture and greater killing capacity to HepG2.2.15 cells than non-sensitized DC/ClK cells. These findings indicate that the immune activities of DC and CIK cells in patients with CHB are lower than those in healthy subjects. These defects may not fully correct after in vitro treatment. Our findings supported Shi M's report [18] that DC's function in patients with CHB can be partially restored.

We identified antiviral effects in HBsAg-sensitized DC–CIK cells. The total efficacy of immune response was 63.6% (21/33) at 24 weeks after treatment. Four patients lost expression of HBV-DNA (negative rate 12.1%). Further studies are needed to determine the sustained response rate. Overall, no advantage was observed over currently used nucleoside analogues.

Acknowledgment

This study was supported by United Fund for Working Station of Academian from Henan Provincial Association for Science and Technology and Zhengzhou Renmin Hospital.

Ancillary