Interleukin-35 modulates the balance between viral specific CD4⁺CD25⁺CD127^dim/- regulatory T cells and T helper 17 cells in chronic hepatitis B virus infection

The study conformed to the ethical guidelines of the 1975 Declaration of Helsinki. The protocol was approved by the Ethics Committee of The Second Hospital of Jilin University on March 2017, and written informed consent was obtained from each participant. Twenty-one of HLA-A2 restricted CHB patients and 23 of HLA-A2 restricted asymptomatic HBV carriers (ASC) were enrolled in the present study. All patients were hospitalized or followed-up in Department of Hepatopancreatobiliary Medicine Affiliated to The Second Hospital of Jilin University between July 2017 and March 2018, and diagnose was made in accordance with diagnostic standard of Chinese Guideline of Prevention and Treatment for Chronic Hepatitis B (2010 Version). All patients were positive for HBV e antigen (HBeAg), and treatment-naïve to nucleos(t)ide analogue and interferon-α. Serum alanine aminotransferase (ALT) presented more than 2-fold elevation in CHB patients, while ALT level was normal in ASC group. No patients received immunomodulatory therapies 1 year before sampling. Patients who were co-infected with other hepatovirus or human immunodeficiency virus (HIV), or afflicted with immune disorder or malignance diseases were excluded from this study. The baseline characteristics of all enrolled patients were shown in Table 1.

HBV DNA was quantified by a commercial real-time polymerase chain reaction (PCR)-Fluorescence Quantitative Detection Kit for HBV DNA (Da’An Gene, Guangzhou, Guangdong Province, China) with detection limit of 2 log₁₀ IU/mL. HBV surface antigen (HBsAg), anti-HBs, HBeAg, anti-HBe, and anti-HBV core (anti-HBc) was measured by commercial enzyme linked immunosorbent assay (ELISA) kits (Kehua Biotech, Shanghai, China). Serum biochemical assessments were measured by Hitachi 7500 automatic analyzer (Hitachi, Tokyo, Japan).

EDTA anticoagulant peripheral bloods were collected from each patients. PBMCs were isolated by density gradient centrifugation using Ficoll-Hypaque (Sigma-Aldrich, St Louis, MO, USA). CD4⁺ T cells or CD4⁺CD25⁺CD127^dim/- Tregs were purified from PBMC using CD4⁺ Cell Isolation Kit (Miltenyi, Bergisch Gladbach, Germany) or CD4⁺CD25⁺CD127^dim/- Regulatory T Cell Isolation Kit II (Miltenyi) according to manufacturer’s instruction, respectively. The purity of enriched cells was more than 95% by flow cytometry determination.

CD4⁺ T cells or purified CD4⁺CD25⁺CD127^dim/− Tregs were stimulated with recombinant human IL-35 (final concentration: 1 ng/mL; Peprotech, Rocky Hill, NJ, USA) for 12 h. Cells and supernatants were harvested for further experiments.

PBMCs were stimulated with HBV core 18–27 epitope (HBc 18–27, sequence: FLPSDFFPSV; final concentration: 10 μg/mL) in the presence of Brefeldin A (final concentration: 10 μg/mL) for 12 h. Cells were harvested and transferred to FACS tubes, and were stained with anti-CD4-PerCP (BD Bioscience, San Jose, CA, USA), anti-CD25-APC (BD Bioscience), and anti-CD127-FITC (BD Bioscience) for 20 min in the dark at 4 °C. Cells were washed twice with Staining Buffer (BD Bioscience), and were resuspended with 250 μL of Fixation/Permeabilization solution (BD Bioscience) for 20 min at 4 °C. Cells were then washed twice in 1 × BD Perm/Wash buffer (BD Bioscience), and were stained with anti-IL-17-PE (BD Bioscience) for 30 min in the dark at 4 °C. In certain experiments, purified CD4⁺CD25⁺CD127^dim/− Tregs were stained with anti-CCR4-FITC (BD Bioscience) and anti-CCR6-PE (BD Bioscience) for 20 min in the dark at 4 °C. Isotype antibodies were used to separate positive and negative cells in PerCP, APC, FITC, and PE fluorescence channels. Samples were analyzed with FACS Calibur analyzer (BD Bioscience). Acquisitions were performed with CellQuest Pro Software (BD Bioscience), and analyses were performed with FlowJo Version 8.4.2 for Windows (Tree Star, Ashland, OR, USA).

Total RNA was isolated using Trizol reagent (Invitrogen, Thermo Fisher Scientific, Carlsbad, CA, USA) according to manufacturer’s instruction. cDNA was synthesized with oligo(dT) primer and random 6 mers using PrimeScript RT Master Mix (TaKaRa, Beijing, China). Real-time PCR was performed using TB Green Premix Ex Taq II (TaKaRa). The relative gene expression was quantified by 2^-ΔΔCT method using ABI7500 System Sequence Detection Software (Applied Biosystems, Foster, CA, USA). The sequences of primers were as following. FoxP3 sense: 5′-CCT CCC CCA TCA TAT CCT TT-3′; FoxP3 anti-sense: 5′-TTG GGG TTT GTG TTG AGT GA-3′; RORγt sense: 5′-AGT CGG AAG GCA AGA TCA GA-3′; RORγt anti-sense: 5′-CAA GAG AGG TTC TGG GCA AG-3′; IL-12p35 sense: 5′-TTC CCA TGC CTT CAC CAC TC-3′; IL-12p35 anti-sense: 5′-TAA ACA GGC CTC CAC TGT GC-3′; EBI3 sense: 5′-TTA CAA GCG TCA GGG AGC TG-3′; EBI3 anti-sense: 5′-TTC CCC GTA GTC TGT GAG GT-3′; GAPDH sense: 5′-GCA CCG TCA AGG CTG AGA AC-3′; GAPDH anti-sense: 5′-TGG TGA AGA CGC CAG TGG A-3′.

Cytokine production in the supernatants were measured using commercial ELISA kits (CusaBio, Wuhan, Hubei Province, China) according to manufacturer’s instruction.

Western blot was performed as described previously [16]. Briefly, cells were lysed on ice for 5 min in 2 × SDS buffer with β-mercaptoethanol, and were treated in 95 °C for 10 min. Supernatants were harvested by centrifugation for 1 min at 10,000×g. Total proteins were loaded and separated on SDS-PAGE gels, and were electroblotted onto PVDF membrane. The membrane was soaked for 2 h in block solution, and then incubated overnight in the presence of rabbit polyclonal to signal transducers and activators of transcription 3 (STAT3, phospho Y705, ab76315), or STAT3 (ab32500) (Abcam, Cambridge, MA, USA; 1: 1000 dilution). Horseradish peroxidase-conjugated goat anti-rabbit antibody IgG (Abcam; 1: 2000 dilution) was added for additional 2 h incubation. Antigen-antibody complexes were observed by enhanced chemiluminescence (Western Blotting Luminol Reagent, Cell Signaling Technology, Danvers, MA, USA).

All data were analyzed using SPSS Version 21.0 for Windows (Chicago, IL, USA). Shapiro-Wilk test was used for normal distribution assay. Variables following normal distribution were presented as mean ± standard deviation, and statistical significance was determined by Student t test or paired t test. Variables following skewed distribution were presented as median [Q1, Q3], and statistical significance was determined by Mann-Whitney test or Wilcoxon matched pairs test. All tests were two-tailed, and P value of less than 0.05 was considered to indicate significant difference.

PBMCs were isolated from all enrolled patients (21 CHB and 23 ASC), and were stimulated with HBc 18–27 peptide for 12 h. Cells were then stained with anti-CD4, −CD25, −CD127, and -IL-17 for flow cytometry analysis. The gating strategy and representative flow dots for Tregs (CD4⁺CD25⁺CD127^dim/−) and Th17 (CD4⁺IL-17⁺) in both CHB and ASC was shown in Fig. 1a. There was no remarkable different of viral specific Tregs between CHB and ASC (11.99 ± 3.03% vs. 12.44 ± 2.84%, Student t test, P = 0.608, Fig. 1b). However, the percentage of viral specific Th17 cells in CHB was significantly higher than in ASC (0.97 ± 0.20% vs. 0.81 ± 0.13%, Student t test, P = 0.0024, Fig. 1c). Thus, ratio of HBc specific Tregs/Th17 in CHB was notably lower in comparison with ASC (12.62 ± 3.39 vs. 15.68 ± 4.27, Student t test, P = 0.012, Fig. 1d).

CD4⁺ T cells were purified from eighteen CHB and sixteen ASC, and were stimulated with IL-35 for 12 h in the presence of HBc 18–27 peptide. Cells and supernatants were harvested for further experiments. Our previous study has been demonstrated serum IL-35 was elevated in both CHB and ASC [17]. It was also accepted that Tregs and other cell types (including activated myeloid, endothelial cells, regulatory B cells) could secret IL-35 [19]. Thus, IL-12p35 and EBI3 mRNA expression in unstimulated CD4⁺ T cells was firstly screened by real-time PCR. IL-12p35 mRNA was comparable between CHB and ASC (Student t test, P = 0.745, Fig. 2a). However, EBI3 mRNA was significantly elevated in CD4⁺ T cells from CHB when compared with those from ASC (Student t test, P = 0.0009, Fig. 2b). IL-35 stimulation notably increased CD4⁺CD25⁺CD127^dim/− Tregs percentage within CD4⁺ T cells in both CHB (16.68 ± 3.21% vs. 12.73 ± 2.24%, paired t test, P = 0.0004, Fig. 2c) and ASC (15.31 ± 3.20% vs. 13.64 ± 3.32%, paired t test, P = 0.025, Fig. 2c). However, IL-35 treatment only slightly down-regulated Th17 cells frequency in CHB (1.14 ± 0.10% vs. 1.10 ± 0.14%) and ASC (0.87 ± 0.12% vs. 0.82 ± 0.15%), and those differences failed to achieve significances (paired t tests, P = 0.212 and P = 0.127, respectively, Fig. 2d). Although ratio of Tregs/Th17 was remarkable elevated in both CHB (15.28 ± 2.98 vs. 11.35 ± 2.98, paired t test, P = 0.0003, Fig. 2e) and ASC (19.20 ± 4.87 vs. 16.03 ± 4.76, paired t test, P = 0.021, Fig. 2e) in response to IL-35 stimulation, this ratio was still higher in ASC in comparison with in CHB (Student t test, P = 0.0074, Fig. 2e). mRNA relative expression corresponding to key transcriptional factor for Tregs (FoxP3) and Th17 cells (RORγt) was investigated by real-time PCR. FoxP3 mRNA was elevated in both CHB (paired t test, P < 0.0001, Fig. 2f) and ASC (paired t test, P = 0.0076, Fig. 2f) in response IL-35 stimulation. However, IL-35 treatment did not affect RORγt mRNA expression in CD4⁺ T cells from either CHB (paired t test, P = 0.949, Fig. 2g) or ASC (paired t test, P = 0.069, Fig. 2g). Tregs-secreting IL-10 and Th17-secreting IL-17 production in the cultured supernatants were also measured by ELISA. IL-10 expression was elevated in response to IL-35 stimulation in CHB (131.4[55.52, 177.3] vs. 68.61[41.13, 79.40], Wilcoxon matched pairs test, P = 0.0001, Fig. 2h) and ASC (189.3[105.3, 268.6] vs. 109.3[50.35, 130.0], Wilcoxon matched pairs test, P = 0.0002, Fig. 2h). IL-17 production by CD4⁺ T cells was down-regulated with IL-35 stimulation in CHB (33.11[26.09, 37.79] vs. 43.78[29.69, 52.06], Wilcoxon matched pairs test, P = 0.0025, Fig. 2i), however, was comparable in ASC in presence or absence of IL-35 treatment (42.69[24.12, 60.03] vs. 44.56[27.45, 66.61], Wilcoxon matched pairs test, P = 0.124, Fig. 2i). pSTAT3 and STAT3 expression in CD4⁺ T cells was investigated by Western blot (Fig. 2j). STAT3 phosphorylation was significantly down-regulated in CD4⁺ T cells in response to IL-35 treatment, however, total STAT3 expression was comparable in the presence or absence of IL-35 stimulation (Fig. 2j).

Our previous study has been demonstrated that IL-35 stimulation in vitro robustly promoted inhibitory activity of CD4⁺CD25⁺CD127^dim/− Tregs [17]. Th17-like phenotype could also be induced from both human naïve and effector Tregs, leading to reduced suppressive function [20, 21]. Thus, we purified CD4⁺CD25⁺CD127^dim/− Tregs from thirteen CHB patients, and were stimulated with HBc 18–17 peptide in the presence or absence of IL-35 for 12 h. Cells were harvested and tested for CCR4 and CCR6 expression by flow cytometry. Both CCR4⁺ (1.21 ± 0.09% vs. 1.29 ± 0.19%, paired t test, P = 0.038, Fig. 3a) and CCR6⁺ percentage (4.21 ± 0.73% vs. 4.99 ± 1.00%, paired t test, P = 0.006, Fig. 3b) within Tregs was significantly reduced in response to IL-35 stimulation. IL-17 and IL-22 production by Tregs was also measured in cultured supernatants. Both IL-17 (4.12 ± 0.78 pg/mL vs. 5.10 ± 1.39 pg/mL, paired t test, P = 0.027, Fig. 3c) and IL-22 secretion (11.88 ± 2.52 pg/mL vs. 14.74 ± 2.53 pg/mL, paired t test, P = 0.023, Fig. 3d) was remarkably down-regulated with IL-35 stimulation.

Ten of CHB patients received tenofovir disoproxil fumarate (TDF, 300 mg once daily), and all ten patients reached virological response with undetectable HBV DNA in the serum three month post-therapy. CD4⁺ T cells and CD4⁺CD25⁺CD127^dim/− Tregs were purified, and were stimulated with IL-35 in the presence of HBc 18–27 peptide for 12 h. There were no significant difference in viral specific Tregs (11.91 ± 2.49% vs. 12.43 ± 2.89%, paired t test, P = 0.415, Fig. 4a), Th17 cells (0.85 ± 0.15% vs. 0.94 ± 0.24%, paired t test, P = 0.291, Fig. 4b), or IL-35 expression in the serum (76.30 ± 5.27 pg/mL vs. 78.10 ± 6.44 pg/mL, paired t test, P = 0.345, Fig. 4c) in response to TDF therapy. IL-35 stimulation in vitro did not elevated viral specific Tregs (11.75 ± 1.70% vs. 13.34 ± 2.53%, paired t test, P = 0.160, Fig. 4d) or Th17 cells (1.17 ± 0.06% vs. 1.22 ± 0.08%, paired t test, P = 0.193, Fig. 4e) within CD4⁺ T cells from TDF-treated CHB patients. Moreover, CCR4⁺ (1.18 ± 0.13% vs. 1.22 ± 0.27%, paired t test, P = 0.401, Fig. 4f) and CCR6⁺ (4.77 ± 0.91% vs. 5.13 ± 1.23%, paired t test, P = 0.460, Fig. 4g) cells were also did not affected by IL-35 in vitro stimulation in purified CD4⁺CD25⁺CD127^dim/− Tregs from TDF-treated CHB patients.