Autophagy is involved in regulating the immune response of dendritic cells to influenza A (H1N1) pdm09 infection

Summary

Autophagy can mediate antiviral immunity. However, it remains unknown whether autophagy regulates the immune response of dendritic cells (DCs) to influenza A (H1N1) pdm09 infection. In this study, we found that infection with the H1N1 virus induced DC autophagy in an endocytosis‐dependent manner. Compared with autophagy‐deficient Beclin‐1^+/− mice, we found that bone‐marrow‐derived DCs from wild‐type mice (WT BMDCs) presented a more mature phenotype on H1N1 infection. Wild‐type BMDCs secreted higher levels of interleukin‐6 (IL‐6), tumour necrosis factor‐ α (TNF‐α), interferon‐β (IFN‐β), IL‐12p70 and IFN‐γ than did Beclin‐1^+/− BMDCs. In contrast to Beclin‐1^+/− BMDCs, H1N1‐infected WT BMDCs exhibited increased activation of extracellular signal‐regulated kinase, Jun N‐terminal kinase, p38, and nuclear factor‐κB as well as IFN regulatory factor 7 nuclear translocation. Blockade of autophagosomal and lysosomal fusion by bafilomycin A1 decreased the co‐localization of H1N1 viruses, autophagosomes and lysosomes as well as the secretion of IL‐6, TNF‐α and IFN‐β in H1N1‐infected BMDCs. In contrast to Beclin‐1^+/− BMDCs, H1N1‐infected WT BMDCs were more efficient in inducing allogeneic CD4⁺ T‐cell proliferation and driving T helper type 1, 2 and 17 cell differentiation while inhibiting CD4⁺ Foxp3⁺ regulatory T‐cell differentiation. Moreover, WT BMDCs were more efficient at cross‐presenting the ovalbumin antigen to CD8⁺ T cells. We consistently found that Beclin‐1^+/− BMDCs were inferior in their inhibition of H1N1 virus replication and their induction of H1N1‐specific CD4⁺ and CD8⁺ T‐cell responses, which produced lower levels of IL‐6, TNF‐α and IFN‐β in vivo. Our data indicate that autophagy is important in the regulation of the DC immune response to H1N1 infection, thereby extending our understanding of host immune responses to the virus.

Abbreviations

Atg

autophagy related

Beclin‐1^+/− BMDCs

BMDCs from Beclin‐1^+/− mice

DCs

dendritic cells

ERK

extracellular signal‐regulated kinase

H1N1

2009 influenza A pandemic

IFN

interferon

IL

interleukin

IRF7

interferon regulatory factor 7

JNK

Jun N‐terminal kinase

MOI

multiplicity of infection

NF‐κB

nuclear factor‐κB

Th

T helper

TLRs

Toll‐like receptors

TNF‐α

tumour necrosis factor‐α

Treg

regulatory T

WT BMDCs

BMDCs from wild‐type mice

Introduction

Since the onset of the 2009 influenza A pandemic, the H1N1 virus has been the cause of significant morbidity and mortality.1 Although elderly patients and those with co‐morbidities are more likely to experience the worst clinical outcomes, many severe cases have been reported in healthy young adults: approximately 25–50% of all patients hospitalized for H1N1 infection in 2009 died without an obvious co‐existing medical condition.2, 3, 4 The nature of the host immune response to the 2009 H1N1 virus remains largely unknown.5 Increased cytokine and chemokine plasma levels are markers of critical illness and suggest that host immune defence contributes to disease pathogenesis.6 It is therefore important to increase our understanding of the host immune response to H1N1 virus infection and its contribution to viral clearance.

Dendritic cells (DCs) are a crucial cell type in the initiation of robust immunity to influenza A.7 Within the lung, DCs direct both innate and adaptive immune responses to viral pathogens through the secretion of pro‐inflammatory cytokines and type I interferon (IFN) and through their migration and presentation of viral antigens to T cells in the hilar lymph nodes.8 DCs are activated through the detection of viral antigens by pattern recognition receptors, which include Toll‐like receptors (TLRs), retinoic acid‐inducible gene I‐like receptors, nucleotide‐binding oligomerization domain‐like receptors, and C‐type lectin receptors.9 In DC endosomal compartments, TLR7 detects influenza A virus single‐stranded RNA,10 and the activation of TLR signalling is required for DC maturation and increased antigen presentation ability.11

Macroautophagy (hereafter referred to as autophagy) is a highly conserved pathway that enables cells to degrade pieces of themselves in lysosomes to promote their survival under stress, including amino acid starvation, energy deprivation and viral infection.12 The biogenesis of the autophagosome is evolutionarily conserved and requires two ubiquitin‐like conjugation systems. One system generates conjugates of autophagy‐related (Atg) proteins 5 and 12, which associate with Atg16 and are essential for the elongation of the isolation membrane.13, 14 The Atg5–Atg12 complex possesses E3 ubiquitin ligase‐like activity towards the second ubiquitin‐like system, which is the Atg8 (LC3) conjugation system.15 The second system mediates the conjugation of free cytosolic LC3 (LC3‐I) to phosphatidylethanolamine (LC3‐II), which triggers autophagy.16 Autophagy plays a critical role in host immune cells by promoting an effective antiviral immunity response, including the production of IFN‐α by plasmacytoid DCs.17 Autophagy can also facilitate the efficient antigen cross‐priming of virus‐specific CD8⁺ T cells18 and is involved in the production of CXCL10 and IFN‐α by macrophages upon H1N1 virus infection, which suggests that autophagy plays a role in establishing anti‐H1N1 immunity.19 However, whether H1N1 viruses can induce autophagy in DCs as well as the effect of autophagy in DC immunity upon H1N1 virus infection remains to be determined.

In this study, we sought to determine whether H1N1 viruses induce autophagy in DCs. We then explored whether autophagy was implicated in the regulation of the DC immune response to H1N1 virus infection by analysing the response of bone marrow‐derived DCs (BMDCs) from autophagy‐deficient Beclin‐1^+/− mice20 to H1N1 virus infection.

Materials and methods

Mice and virus

For all experiments, only female mice were used. Female C57BL/6J and BALB/c mice (aged 6–8 weeks) were purchased from Joint Ventures Sipper BK Experimental Animal Co. Ltd. (Shanghai, China). Beclin‐1^+/− mice, C57BL/6‐Tg (Tcra Tcrb) 1100Mjb/J (OT‐I) mice and DTR‐CD11c mice were purchased from Jackson Laboratory (Bar Harbor, ME). Mice were kept in specific pathogen‐free facilities in Zhejiang University. All experiments using mice were approved by and performed according to the guidelines of the Animal Ethics Committee of Zhejiang University. The influenza A (H1N1) pdm09 virus strain A/Zhejiang/2/2009 (H1N1) was a gift from Professor Yiyu Lu of the Zhejiang Provincial Centre for Disease Control and Prevention. The viruses were grown in Madin–Darby canine kidney cells and were purified by pre‐adsorption to and elution from turkey red blood cells. Virus infectivity was determined by titration on Madin–Darby canine kidney cells.21, 22

BMDCs generation and H1N1 virus infection

The BMDCs were generated as previously described.23 On day 6, BMDCs were collected and resuspended at a density of 1 × 10⁶/ml. Then the BMDCs were infected with H1N1 viruses at the indicated multiplicity of infection (MOI) for 2 hr and collected for the subsequent experiments.

Immunofluorescence staining

For the detection of autophagosomes, BMDCs from wild‐type mice (WT BMDCs) with or without 2 μg/ml cytochalasin D (Sigma‐Aldrich Chemicals, St Louis, MO) pre‐treatment, WT BMDCs or BMDCs from Beclin‐1^+/− mice (Beclin‐1^+/− BMDCs) were infected with H1N1 viruses for 2 hr. The cells were then fixed with 4% paraformaldehyde, permeabilized with 0·1% Triton‐X‐100 for 5 min and blocked in 5% BSA with 0·1% Tween‐20. The cells were incubated for 5 hr with an antibody against LC3B (D11; Cell Signaling Technology, Danvers, MA). To detect IFN regulatory factor 7 (IRF7) nuclear translocation, WT or Beclin‐1^+/− BMDCs were infected with H1N1 virus at an MOI of 2 for the indicated duration. Then, the BMDCs were fixed, permeabilized and then incubated for 5 hr with antibodies against IRF7 (H‐246; Santa Cruz Biotechnology, Santa Cruz, CA). To detect the H1N1 virus, TLR7, autophagosomes and lysosomes, BMDCs were infected with H1N1 viruses at an MOI of 2 with or without 100 nm bafilomycin A1 (Sigma‐Aldrich Chemicals) for 2 hr. Then, the cells were fixed, permeabilized and incubated for 5 hr with antibodies against H1N1 haemagglutinin (C102; Thermo Fisher Scientific, Rockford, IL), LC3B (D11; Cell Signaling Technology), TLR7 (H‐114, C102; Santa Cruz Biotechnology) and Lamp2 (GL2A7; Abcam, Cambridge, MA). Subsequently, all cells were incubated with a corresponding fluorescence‐labelled secondary antibody for 1 hr. The cells were counterstained with DAPI to label DNA. The stained cells were viewed on a confocal microscope (Leica, SP2, Solms, Germany).

Western blot analysis

To detect autophagy, BMDCs were infected with H1N1 viruses at the indicated MOI for 2 hr. To evaluate the effect of cytochalasin D on autophagy induction, BMDCs were pre‐treated with the indicated concentration of cytochalasin D for 30 min and then treated with 20 pmol/ml rapamycin (Sigma‐Aldrich Chemicals) for 2 hr. To detect the activation of TLR signalling, both WT and Beclin‐1^+/− BMDCs were infected with H1N1 virus at an MOI of 2 for the indicated duration. All cells were washed and lysed. Twenty microgrammes of cell lysate protein was separated by 10% SDS–PAGE and transferred onto a PVDF membrane. After incubation with primary antibodies against LC3B (G‐9), p62 (H‐290), extracellular signal‐regulated kinase (ERK (MK1), phosphorylated ERK (p‐ERK; E‐4), Jun N‐terminal kinase (JNK; D‐2), p‐JNK (9H8), p38 (A‐12), p‐p38 (D‐8), nuclear factor‐κB (NF‐κB C22B4) or p‐NF‐κB (93H1) (all from Santa Cruz Biotechnology) followed by a corresponding horseradish peroxidase‐coupled secondary antibody, specific bands were visualized on the membrane using a chemiluminescence kit (ECL detection Kit, Amersham Bioscience, GE Healthcare Life Sciences, Pittsburgh, PA).

Qualification of H1N1 viruses yield

The yield of H1N1 viruses from H1N1 virus‐infected BMDCs was qualified as previously described with minor modification.24 WT and Beclin‐1^+/− BMDCs were infected with H1N1 viruses at an MOI of 2 for 24 hr. Then the BMDCs were washed and resuspended and cultured in fresh culture medium. After 0, 24, 48 or 72 hr, the BMDCs were lysed three times by freeze–thaw. Then 0·1 ml of the cultural supernatants was serially diluted on the monolayer of Madin–Darby canine kidney cells and 1 × 10⁴ cells were seeded into 96‐well plates 1 day before measurement. The 50% tissue culture infectious dose (TCID₅₀) was measured after 3 days.

Cytokine assays

Wild‐type or Beclin‐1^+/− BMDCs were either infected with H1N1 viruses at an MOI of 2 for 24 hr or cultured in Hanks' balanced salt solution for 2 hr followed by H1N1 virus infection at an MOI of 2 for 24 hr. The levels of interleukin‐6 (IL‐6), tumour necrosis factor‐ Biotechnology (TNF‐α), IFN‐β, IL‐12p70 and IFN‐γ in the culture supernatant (all from Biolegend, San Diego, CA) were measured by ELISA. Wild‐type BMDCs were infected with H1N1 viruses at an MOI of 2 with or without 100 nm bafilomycin A1 for 24 hr. Then, the levels of IL‐6, TNF‐α and IFN‐β in the culture supernatant were detected by ELISA.

Mixed lymphocyte reaction

The BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 viruses at an MOI of 2 for 24 hr, followed by inactivation with 50 μg/ml mitomycin C (Sigma‐Aldrich Chemicals) for 30 min at 37°. Then, BMDCs were co‐cultured with CD4⁺ T cells obtained from splenocytes of BALB/c mice by purification with the CD4⁺ T Cell Isolation Kit II (Miltenyi Biotec, Inc., Auburn, CA) at the indicated ratio for 3 days. The proliferation of CD4⁺ T cells was detected using a CCK‐8 cell proliferation kit (Dojindo Molecular Technologies, Kumamoto, Japan).

Intracellular staining and flow cytometry analysis

Wild‐type or Beclin‐1^+/− BMDCs were infected with H1N1 viruses at an MOI of 2 for 2 hr and were then stained with phycoerythrin‐labelled anti‐MHC‐II (M5/114.15.2), CD80 (16‐10A1), CD86 (GL‐1) and CD40 (3/23) antibodies (all from Biolegend). To confirm the block of H1N1 endocytosis by cytochalasin D, WT BMDCs were pre‐treated with or without 2 μg/ml cytochalasin D and then infected with H1N1 viruses at an MOI of 2 for 2 hr. The cells were fixed and permeabilized using Intracellular Fixation and Permeabilization Buffer (eBioscience, San Diego, CA) and were stained with antibodies against H1N1 haemagglutinin (C102). To determine the differentiation of CD4⁺ T‐cell subsets, BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 virus at an MOI of 2 for 24 hr. The BMDCs were then co‐cultured with naive CD4⁺ T cells (purified with a CD4⁺ naive T‐cell isolation kit; Miltenyi Biotec, Inc.) in the presence of 1 μg/ml anti‐CD3 (17A2) and CD28 (37.51) (both from eBioscience) for 3 days. The cells were stained with anti‐CD4 antibody (GK1.5, eBioscience) followed by fixation and permeabilization with a Foxp3/Transcription Factor Staining Buffer Set (eBioscience). The cells were then stained with antibodies against IFN‐γ (XMG1.2), IL‐4 (11B11), IL‐17 (eBio17B7) or Foxp3 (FJK‐16s) (all from eBioscience) followed by PMA/ionomycin stimulation for 5 hr. All cells were examined by flow cytometry, and the data were analysed using the flowjo software.

Analysis of antigen cross‐presentation

Wild‐type or Beclin‐1^+/− BMDCs were infected with H1N1 virus at an MOI of 2 for 24 hr, either with or without 20 μg/ml ovalbumin (OVA) (Sigma‐Aldrich Chemicals). The infected BMDCs were stained with antibodies against MHC‐I (AF6‐88.5.5.3) or MHC‐I/OVA_257–264 (25‐D1.16) (both from eBioscience) and analysed by flow cytometry or collected and co‐cultured with naive CFSE‐labelled CD8⁺ T cells (purified using CD8 magnetic microbeads from OT‐I mice; Miltenyi Biotec, Inc.) at a ratio of 1 : 10 for 3 days. The proliferation of CD8⁺ T cells was assessed by determining the CFSE dilution by flow cytometry.

Mouse infection

To deplete DCs, CD11c‐DTR mice were given one intraperitoneal injection of diphtheria toxin (Sigma‐Aldrich Chemicals) (16 μg/kg body weight). Twenty‐four hours later, 2 × 10⁶ WT or Beclin‐1^+/− BMDCs were intravenously transferred into these mice, which were then intranasally infected with H1N1 virus (10² plaque‐forming units/mouse) under parenteral anaesthesia.

ELISpot assay

At 7 days post‐H1N1 infection, the mice were killed, and CD4⁺ or CD8⁺ T cells from the spleen were purified using the CD4⁺ T Cell Isolation Kit II or CD8 magnetic microbeads (Miltenyi Biotec, Inc.). CD4⁺ or CD8⁺ T cells were re‐stimulated with H1N1 viruses at an MOI of 2 in the presence of Mitomycin C (Sigma‐Aldrich Chemicals) ‐inactivated BMDCs at T‐cell : BMDC ratio of 10:1 for 24 hr in IL‐2 or IFN‐γ ELISpot plates (eBioscience). Then, the ELISpot assay was performed according to the manufacturer's instructions.

Statistical analysis

Results are expressed as the mean ± SEM. Statistics were analysed by unpaired two‐tailed Student's t‐test, one‐way or two‐way analysis of variance with Newman–Keuls post‐test using a graphpad prism 5·0 software (Graphpad Software, Inc., La Jolla, CA). A P‐value of < 0·05 was considered statistically significant.

Results

Increased induction of autophagy in BMDCs upon H1N1 virus infection

Increased autophagy was detected in macrophages upon H1N1 virus infection19 and in DCs during respiratory syncytial virus infection.25 We hypothesized that autophagy would also be induced in DCs after H1N1 virus infection. As expected, 2 hr after H1N1 virus infection, the conversion of LC3B‐I to LC3B‐II, which is a hallmark of autophagy induction, had markedly increased in BMDCs, as detected by Western blot (Fig. 1a). Along with the higher conversion of LC3B‐I to LC3B‐II, the level of p62 protein in H1N1‐infected BMDCs decreased, suggesting that the H1N1 virus did not inhibit autophagic flux in BMDCs (Fig. 1a). Because an H1N1 MOI > 2 did not increase autophagy, we considered the MOI of 2 to be optimal for autophagy induction in BMDCs (Fig. 1a). We confirmed this result by immunofluorescent staining, which revealed similar levels of autophagy induction as green puncta in BMDCs after H1N1 virus infection (Fig. 1b). It has been reported that viruses can induce autophagy independent of endocytosis.26, 27 To examine whether contact between the H1N1 virus and DCs was sufficient to induce autophagy, we used cytochalasin D to inhibit the endocytosis of H1N1 by DCs. First, we determined whether cytochalasin D affected the induction of autophagy in DCs. After treatment with cytochalasin D at concentrations of 0·5 and 2 μg/ml, BMDCs exhibited no obvious difference in autophagy induction by rapamycin (Fig. 1c). However, at a concentration of 2 μg/ml, cytochalasin D markedly inhibited the endocytosis of H1N1 virus by BMDCs (Fig. 1d). Under these conditions, autophagy in BMDCs was markedly inhibited after H1N1 virus infection (Fig. 1e). These results suggest that H1N1 virus infection can induce autophagy in BMDCs, depending on whether viral endocytosis has occurred.

Figure 1.

Autophagy induction increased in bone‐marrow‐derived dendritic cells (BMDCs) upon H1N1 virus infection. Two hours after infection by the indicated multiplicity of infection (MOI) of H1N1 virus, (a) the levels of LC3B‐I, LC3B‐II and p62 proteins in BMDCs were examined by Western blot and (b) LC3B‐positive puncta in BMDCs were detected by fluorescence staining. (c) BMDCs were pre‐treated with the indicated concentration of cytochalasin D for 30 min and were treated with 20 pmol/ml rapamycin for 2 hr. The levels of LC3B‐I and LC3B‐II proteins were then assessed by Western blot. (d, e) BMDCs were pre‐treated with or without 2 μg/ml cytochalasin D and were infected with H1N1 virus at an MOI of 2 for 2 hr. The endocytosis of H1N1 viruses was detected by flow cytometry via intracellular staining of the H1N1‐specific antigen haemagglutinin (d), and LC3B‐positive puncta in BMDCs were detected by fluorescence staining (e). Data are representative of three or four independent experiments.

Beclin‐1^+/− BMDCs are deficient in maturation and the production of innate and adaptive cytokines upon H1N1 infection

Because autophagy can be induced in BMDCs during H1N1 infection, we sought to determine whether autophagy plays a role in the immune functions of BMDCs upon infection with the virus. Autophagy‐deficient Beclin‐1^+/− mice were used in the following experiments. First, we confirmed whether autophagy was deficient in the BMDCs of Beclin‐1^+/− mice following H1N1 virus infection. As shown in Fig. 2(a), there was no obvious difference in autophagy between the BMDCs of WT and Beclin‐1^+/− mice in the absence of H1N1 infection. However, 2 hr after H1N1 infection, WT BMDCs showed a significant increase in autophagy puncta, whereas these puncta only slightly increased in number in Beclin‐1^+/− BMDCs (Fig. 2a). Western blot also revealed a greater conversion of LC3B‐I to LC3B‐II in WT BMDCs than in Beclin‐1^+/− BMDCs after H1N1 infection. Moreover, p62 protein in WT BMDCs was significantly reduced, but in Beclin‐1^+/− BMDCs, it was slightly reduced after H1N1 infection, indicating normal autophagic flux in both types of BMDC (Fig. 2b). Next, we examined the phenotype of BMDCs obtained from WT and Beclin‐1^+/− mice upon H1N1 virus infection and found that both mice exhibited a more mature phenotype after H1N1 infection. Compared with WT BMDCs, Beclin‐1^+/− BMDCs expressed lower levels of MHC‐II, CD80 and CD86 but comparable CD40 levels upon H1N1 infection (Fig. 2c,d). We next examined the role of autophagy in cytokine production by BMDCs. Similar to the observed phenotypic change; H1N1 infection promoted the secretion of IL‐6, TNF‐α, IFN‐β, IL‐12p70 and IFN‐γ from the BMDCs obtained from both mouse genotypes. Strikingly, the production of these cytokines was significantly reduced in Beclin‐1^+/− BMDCs (Fig. 2e). To exclude the possibility that this effect was caused by the different replication abilities of the H1N1 virus in WT or Beclin‐1^+/− BMDCs, we measured the viral yield in WT and Beclin‐1^+/− BMDCs after H1N1 infection. There was no obvious difference in viral yield between H1N1‐infected BMDCs from WT and Beclin‐1^+/− mice up to72 hr after infection when measured by TCID₅₀ assay (Fig. 2f). To corroborate the autophagy dependence of this effect, BMDCs were subjected to amino acid starvation in Hanks' balanced salt solution culture to induce autophagy before H1N1 infection, and cytokine secretion by BMDCs was assayed. When autophagy was induced in BMDCs, increases in the levels of IL‐6, TNF‐α, IFN‐β, IL‐12p70 and IFN‐γ were detected upon H1N1 infection (Fig. 2g). In addition, H1N1 infection induced transforming growth factor‐β ₁ production by BMDCs in an autophagy‐independent manner (data not shown). These results indicate that autophagy plays a critical role in the maturation and secretion of innate and adaptive cytokines of DCs upon exposure to the H1N1 virus.

Figure 2.

Beclin‐1^+/− bone‐marrow‐derived dendritic cells (BMDCs) are deficient in maturation and the production of innate and adaptive cytokines upon H1N1virus infection. (a, b) BMDCs from wild‐type (WT) or Beclin‐1^+/− mice were infected with H1N1 virus at a multiplicity of infection (MOI) of 2 for 2 hr. The LC3B‐positive puncta in BMDCs were detected by fluorescence staining (a), and the LC3B‐I/II and p62 proteins were detected by Western blot (b). (c) BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 virus at an MOI of 2 for 24 hr. The expression of MHC‐II, CD80, CD86 and CD40 was detected by flow cytometry. (d) Statistical analysis of the mean fluorescent intensity (MFI) of (c). (e) BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 virus at an MOI of 2 for 24 hr. The secretion of interleukin‐6 (IL‐6), tumour necrosis factor‐α (TNF‐α), interferon‐β (IFN‐β), IL‐12p70 and IFN‐γ was measured by ELISA. (f) The H1N1 virus yield by BMDCs from WT or Beclin‐1^+/− mice at the indicated time‐points after H1N1 virus infection were determined by TCID₅₀ assay. (g) BMDCs from WT or Beclin‐1^+/− mice were cultured in Hanks' balanced salt solution medium for 2 hr then were infected by H1N1 virus at an MOI of 2 for 24 hr. The secretion of IL‐6, TNF‐α, IFN‐β, IL‐12p70 and IFN‐γ was examined by ELISA. The data are representative of three independent experiments and are expressed as the means ± SEM. Statistical significance was determined using an unpaired Student's t‐test for (d) and (e), and (g) was analysed by one‐way analysis of variance and Q test. *P < 0·05; **P < 0·01; ***P < 0·001.

TLR signalling activation is compromised in Beclin‐1^+/− BMDCs upon H1N1 virus infection

As mentioned previously, IL‐6, TNF‐α and IFN‐β, which play critical roles in the TLR signalling that mediates innate immunity,28 exhibited a reduction in release in autophagy‐deficient BMDCs upon H1N1 infection. Hence, autophagy probably affects TLR signalling activation in DCs upon H1N1 virus infection. Accordingly, H1N1 virus infection markedly activated the ERK, JNK, p38 and NF‐κB signalling pathways in BMDCs of both genotypes, but the degree of activation was lower in Beclin‐1^+/− BMDCs (Fig. 3a). As a single‐stranded RNA virus, H1N1 has been reported to activate TLR7 signalling.29 Activating TLR7 induces the adaptor protein MyD88, which leads to the phosphorylation of IRF7 and its subsequent nuclear translocation, which in turn induces the production of IFN‐β.30 We therefore examined the nuclear translocation of IRF7 in BMDCs from both mouse strains upon H1N1 infection. Whereas the nuclear translocation of IRF7 increased in WT BMDCs, there was only a slight increase in IRF7 translocation in Beclin‐1^+/− BMDCs after 30 min of H1N1 infection. In both WT and Beclin‐1^+/− BMDCs, the translocation of IRF7 declined after 1 hr and recovered to a basal level after 2 hr of H1N1 infection (Fig. 3b). Hence, these results suggest that autophagy is involved in the activation of TLR signalling in DCs upon H1N1 infection.

Figure 3.

Beclin‐1^+/− bone‐marrow‐derived dendritic cells (BMDCs) exhibit compromised Toll‐like receptor (TLR) signal activation upon H1N1 virus infection. (a, b) BMDCs obtained from wild‐type (WT) or Beclin‐1^+/− mice were infected with H1N1 virus at a multiplicity of infection (MOI) of 2 for the indicated duration. The activation of the extracellular signal‐regulated kinase (ERK), Jun N‐terminal kinase (JNK), p38 and nuclear factor‐κB (NF‐κB) pathways was examined by Western blot (a), and the nuclear translocation of IRF7 was detected by fluorescence staining (b). The data are representative of three independent experiments.

Autophagy promotes TLR signalling activation through the transport of H1N1 viruses to lysosomes

The recognition of viruses by TLR7 requires the transport of cytosolic viral replication intermediates into the lysosome through autophagy.17 However, there is currently no direct evidence to support this notion. We therefore asked whether autophagosomes are the intermediates in the transport of H1N1 viruses to lysosomes. First, we determined whether bafilomycin A1, an inhibitor of autophagosomal/lysosomal fusion,31 inhibited autophagosomal/lysosomal fusion in BMDCs. We found that after treatment with 100 nm bafilomycin A1, almost no co‐localization of autophagosomes and lysosomes could be detected in BMDCs (Fig. 4a). We next examined the co‐localization of autophagosomes, lysosomes and H1N1 viruses with or without bafilomycin A1 treatment. Obvious co‐localization between autophagosomes, H1N1 viruses and lysosomes was detected in BMDCs after H1N1 virus infection for 2 hr (Fig. 4b). However, when treated with bafilomycin A1, the co‐localization of autophagosomes, H1N1 virus and lysosomes was only rarely detected in BMDCs. However, the co‐localization of autophagosomes and H1N1 virus in bafilomycin A1‐treated BMDCs increased (Fig. 4b). In addition, bafilomycin A1 treatment markedly inhibited the co‐localization of TLR7, lysosomes and H1N1 virus, suggesting that autophagosomal/lysosomal fusion is necessary for contact to be made between H1N1 viruses and TLR7 in lysosomes (Fig. 4c). To further characterize the inhibition of autophagosomal/lysosomal fusion after bafilomycin A1 treatment, we evaluated autophagic flux by assessing the levels of LC3B‐II and p62 proteins in BMDCs. We found that bafilomycin A1 treatment obviously increased the level of LC3B‐II protein, along with the apparent accumulation of p62 protein (Fig. 4d). These results suggest that bafilomycin A1 did inhibit autophagic flux in BMDCs, the co‐localization of TLR7, lysosomes and H1N1 viruses and probably TLR signalling activation. As expected, bafilomycin A1 treatment significantly reduced the levels of IL‐6, TNF‐α and IFN‐β produced by H1N1‐infected BMDCs (Fig. 4e). Together, these results provide direct evidence that autophagosomes are the intermediators of H1N1 transport to lysosomes.

Figure 4.

Autophagy promotes Toll‐like receptor (TLR) signalling activation through the transport of H1N1 virus to lysosomes. (a) bone‐marrow‐derived dendritic cells (BMDCs) were treated with or without 100 nm bafilomycin A1 for 2 hr, and then Lamp2 and LC3B proteins were detected by fluorescence staining. (b–d) BMDCs were infected with H1N1 virus at a multiplicity of infection (MOI) of 2, either with or without 100 nm bafilomycin A1, for 2 hr. The H1N1 virus, LC3B and Lamp2 proteins (b) and the H1N1 viruses, TLR7 and Lamp2 proteins (c) were detected by fluorescence staining, and the levels of LC3B‐I, LC3B‐II and p62 proteins were detected by Western blot (d). (e) BMDCs were infected with H1N1 virus at an MOI of 2, either with or without 100 nm bafilomycin A1, for 24 hr. The secretion of interleukin‐6 (IL‐6), tumour necrosis factor‐α (TNF‐α), interferon‐β (IFN‐β) by BMDCs were detected by ELISA. The data are representative of two independent experiments and are expressed as the means ± SEM. Statistical significance was determined using an unpaired Student's t‐test. *P < 0·05; ***P < 0·001.

H1N1‐infected Beclin‐1^+/− BMDCs are impaired in CD4⁺ T antigen presentation and show different potential to drive Th cell differentiation

As mentioned above, autophagy‐deficient BMDCs exhibited an immature phenotype. We therefore wanted to know whether autophagy could regulate the antigen‐presenting ability of BMDCs. The BMDCs obtained from WT and Beclin‐1^+/− mice were co‐cultured with CD4⁺ T cells from BALB/c mice at the indicated ratio for 3 days, and the proliferation of CD4⁺ T cells was then assessed to determine the antigen‐presenting ability of the BMDCs. As shown in Fig. 5(a), before H1N1 virus infection, there was no difference in the antigen‐presenting abilities of WT BMDCs and Beclin‐1^+/− BMDCs. However, the antigen‐presenting ability of Beclin‐1^+/− BMDCs was significantly reduced compared with that of WT BMDCs after H1N1 virus infection (Fig. 5a). Autophagy‐deficient BMDCs also exhibited an altered cytokine secretion profile, which plays a central role in inducing the differentiation of naive CD4⁺ T cells.32, 33, 34 We next compared the effects of BMDCs from WT and Beclin‐1^+/− mice on inducing naive CD4⁺ T‐cell differentiation after H1N1 virus infection. In the absence of H1N1 virus, BMDCs of both genotypes could not induce the differentiation of T helper type 1 (Th1), Th2 and Th17 cells and regulatory T (Treg) cells (Fig. 5b,c). After H1N1 virus infection, WT BMDCs induced more Th1, Th2 and Th17 cell differentiation than did Beclin‐1^+/− BMDCs (Fig. 5b,c). However, Beclin‐1^+/− BMDCs were more efficient at inducing Treg cells compared with WT BMDCs (Fig. 5b,c). These results suggest that autophagy can regulate BMDC antigen presentation to CD4⁺ T cells and the induction of Th cells.

Figure 5.

H1N1‐infected Beclin‐1^+/− bone‐marrow‐derived dendritic cells (BMDCs) are impaired in their ability to present antigens to CD4⁺ T cells, and their ability to drive T helper (Th) cell differentiation is altered. (a) BMDCs from wild‐type (WT) or Beclin‐1^+/− mice were infected with H1N1 virus at a multiplicity of infection (MOI) of 2 for 24 hr, followed by inactivation with 50 μg/ml mitomycin C for 30 min at 37°. Then, BMDCs were co‐cultured with CD4⁺ T cells from BALB/c mice at the indicated ratio for 3 days. The proliferation of CD4⁺ T cells was detected using a CCK‐8 cell proliferation kit. (b) BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 virus at an MOI of 2 for 24 hr, and the BMDCs were co‐cultured with naive CD4⁺ T cells in the presence of 1 μg/ml anti‐CD3/CD28 for 3 days. The differentiation of Th cells was examined by intracellular cytokine staining following stimulation with PMA/ionomycin. Regulatory T cells were detected by nuclear staining for Foxp3. (c) Statistical analysis of (b); the data are representative of three independent experiments and are expressed as the means ± SEM. Statistical significance was determined using two‐way analysis of variance and Q test for (a); **P < 0·01, versus Beclin‐1^+/− + H1N1. Statistical significance was determined using the unpaired Student's t‐test for (c). *P < 0·05; ***P < 0·001.

Autophagy deficiency attenuated the MHC‐I cross‐presenting ability of DCs upon H1N1 infection

We next wanted to determine whether a deficiency in autophagy could affect the ability of DCs to cross‐prime CD8⁺ T cells. After H1N1 virus infection, WT or Beclin‐1^+/− BMDCs were co‐cultured with CD8⁺ T cells from OT‐I mice in the presence of OVA protein, and CD8⁺ T‐cell proliferation was measured. As shown in Fig. 6(a,b), WT BMDCs exhibited a greater potential to induce CD8⁺ T‐cell proliferation than did Beclin‐1^+/− BMDCs, either with or without H1N1 infection, and H1N1 virus infection significantly improved the ability of both BMDCs to stimulate CD8⁺ T‐cell proliferation. We also determined the percentage of IFN‐γ ⁺ CD8⁺ T cells and found similar results for both genotypes (Fig. 6c,d). To determine whether autophagy could affect antigen processing, we examined the levels of MHC‐I molecules expressed by WT and Beclin‐1^+/− BMDCs, either with or without H1N1 infection. We found that H1N1 virus infection increased the levels of MHC‐I molecules overall, but there was no obvious difference between WT and Beclin‐1^+/− BMDCs (Fig. 6e). Moreover, the levels of MHC‐I/OVA_257–264 peptide complexes on these BMDCs underwent similar changes (Fig. 6f). Hence, these results indicate that autophagy‐deficient DCs are compromised in their MHC‐I cross‐presentation ability upon H1N1 infection but that their MHC‐I restricted antigen‐processing ability is unaffected.

Figure 6.

Autophagy deficiency attenuated the MHC‐I cross‐presenting ability of dendritic cells (DCs) upon H1N1 infection. bone‐marrow‐derived DCs (BMDCs) from wild‐type (WT) or Beclin‐1^+/− mice were infected with H1N1 viruses at a multiplicity of infection (MOI) of 2 for 24 hr, either with or without 20 μg/ml ovalbumin (OVA) protein. (a, c) The BMDCs were collected and co‐cultured with CFSE‐labelled CD8⁺ T cells from OT‐I mice for 3 days. The proliferation of CD8⁺ T cells was detected by flow cytometry (a), and the interferon‐γ‐positive (IFN‐γ ⁺) CD8⁺ T cells were detected by intracellular staining (c). (b, d) Statistical analysis of (a) and (c), respectively. (e, f) BMDCs from WT or Beclin‐1^+/− mice were infected with H1N1 virus at an MOI of 2 for 24 hr, either with or without 20 μg/ml OVA. The MHC‐I molecules (e) or MHC‐I/OVA_257–264 complexes (f) on BMDCs were detected by flow cytometry. The data are representative of three independent experiments and are expressed as the means ± SEM. Statistical significance was determined using an unpaired Student's t‐test. *P < 0·05; **P < 0·05; ***P < 0·001.

Autophagy‐deficient BMDCs were impaired in their ability to induce inflammatory responses and H1N1‐specific T‐cell responses in vivo

To clarify whether DCs deficient in autophagy were defective in inducing H1N1‐specific immune responses in vivo we transferred WT and Beclin‐1^+/− BMDCs into DC‐depleted DTR‐CD11c mice and then intranasally infected these mice with H1N1 virus. At 2 days post‐infection, when the viral load in the lungs of the mice showed no obvious discrepancy (data not shown), we detected a pro‐inflammatory response in the bronchoalveolar lavage fluid. We observed lower levels of IL‐6, TNF‐α and IFN‐β in the bronchoalveolar lavage fluid of mice that had received Beclin‐1^+/− BMDCs (Fig. 7a). At 7 days post‐infection, we observed increased H1N1‐specific staining in the lungs of mice treated with Beclin‐1^+/− BMDCs compared with that in mice that had received WT BMDCs (Fig. 7b). Similarly, the viral load was higher in the lungs of mice treated with Beclin‐1^+/− BMDCs (Fig. 7c). To evaluate the H1N1‐specific T‐cell response ex vivo, CD4⁺ or CD8⁺ T cells were purified at 7 days post‐infection and were used in an IL‐2 or IFN‐γ ELISpot assay, respectively. After H1N1 virus was added to mitomycin C‐inactivated BMDCs in vitro for 24 hr, more CD4⁺ or CD8⁺ T cells from mice that received WT BMDCs transfer could produce IL‐2 or IFN‐γ, respectively (Fig. 7d). These results suggest that autophagy‐deficient BMDCs are defective in inducing H1N1‐specific immune responses, which will affect the clearance of the H1N1 virus by the host.

Figure 7.

Autophagy‐deficient bone‐marrow‐derived dendritic cells (BMDCs) exhibited an impaired ability to induce an inflammatory response and an H1N1‐specific T‐cell response in vivo. (a–d) CD11c‐DTR mice received an intraperitoneal injection of diphtheria toxin (16 μg/kg body weight). Twenty‐four hours later, 2 × 10⁶ wild‐type (WT) or Beclin‐1^+/−BMDCs were intravenously transferred into these mice, which were then intranasally infected with H1N1 virus (10² PFU/mouse) under parenteral anaesthesia. At 2 days post‐infection, the mice were killed, and the levels of interleukin‐6 (IL‐6), tumour necrosis factor‐α (TNF‐α), interferon‐β (IFN‐β) in bronchoalveolar lavage fluid were detected by ELISA (a). At 7 days post‐infection, the mice were killed, and the viral loads in the lungs were visualized by H1N1‐specific immunofluorescent staining (b). Alternatively, the virus infectivity in the lungs was quantified by titration on Madin–Darby canine kidney cells (c), or CD4⁺ or CD8⁺ T cells from the spleen were purified and co‐cultured with mitomycin C‐inactivated BMDCs at ratio of 10 : 1, either with or without H1N1 virus at a multiplicity of infection (MOI) of 2 for 24 hr in IL‐2 or IFN‐γ antibody‐coated plates. Then, the IL‐2‐producing CD4⁺ T cells or IFN‐γ‐producing CD8⁺ T cells were quantified by ELISpot assay (d). The data are representative of two independent experiments and are expressed as the means ± SEM. Statistical significance was determined using an unpaired Student's t‐test. *P < 0·05; **P < 0·05.

Discussion

H1N1 virus infection has been reported to induce autophagy in murine macrophages, monkey and canine kidney cells and human alveolar epithelial cells.19, 35, 36 The contribution of autophagy to MHC‐II‐restricted presentation by DCs after H1N1 virus infection has also been evaluated. However, before this report, autophagy in DCs after H1N1 virus infection had not been examined, and whether autophagy could be induced in DCs upon H1N1 infection was unclear. In this study, we found that H1N1 infection markedly increases autophagy, which plays an extensive role in DC immune responses.

Autophagy may be induced in the host cell as a result of different steps in the viral life cycle, including viral tethering and entry; membrane fusion; the exposure of viral components to host sensors; replication‐induced perturbation of cellular homeostasis; and direct stimulation by viral protein.26 In the present study, we found that inhibiting H1N1 endocytosis abolished the induction of autophagy upon H1N1 virus infection. Toll‐like receptor signalling has been reported to regulate autophagy.36, 37, 38 The endocytosis of H1N1 viruses is necessary for TLR7 signalling activation in BMDCs; hence, the exposure of H1N1 virus components (probably TLR7) is likely to be necessary for the induction of autophagy in DCs after H1N1 virus infection. In addition, we confirmed that autophagy plays a critical role in TLR signalling activation in BMDCs, indicating the presence of an autophagy‐dependent positive feedback loop for the activation of innate immunity in DCs after H1N1 infection.

Viruses are thought to be transported into lysosomes during autophagy.17 Before now, no direct evidence had been found to support this notion. In this study, we could not detect co‐localization of the autophagosome, H1N1 virus and lysosome in BMDCs after H1N1 infection. The inhibition of autophagosome fusion to lysosomes by bafilomycin A1, however, abolished the co‐localization of autophagosome, H1N1 virus and lysosome in BMDCs. Moreover, in these BMDCs, increased co‐localization of autophagosomes and H1N1 virus was detected. Therefore, our results provide direct evidence that the transport of H1N1 virus to lysosomes is autophagosome‐dependent.

Compared with WT BMDCs, the production of IFN‐β, which is critical for viral clearance,39 was lower in Beclin‐1^+/− BMDCs. Autophagy, which is deficient in Beclin‐1^+/− BMDCs, has also been reported to enhance viral clearance by transporting the virus into lysosomes for degradation.40 It is therefore reasonable to hypothesize that viral replication in WT BMDCs is stronger than in Beclin‐1^+/− BMDCs. However, according to our results, there was no obvious difference in viral yield between H1N1‐infected WT and Beclin‐1^+/− BMDCs at 72 hr. Hepatitis B virus and influenza A virus have been reported to resist lysosome‐mediated degradation.41, 42 It may also be the case that H1N1 viruses are resistant to lysosome‐mediated degradation. Moreover, lysosomes may shelter H1N1 virus from clearance by the host cell defence system. These putative mechanisms may help to explain the similarity in viral yield between H1N1‐infected WT and Beclin‐1^+/− BMDCs.

After H1N1 virus infection, BMDCs from WT mice induced Th1, Th2 and Th17 cell differentiation to a greater extent than BMDCs obtained from Beclin‐1^+/− mice. Compared with Beclin‐1^+/− BMDCs, BMDCs from WT mice produced more IL‐12 upon H1N1 virus infection, which plays a central role in the Th1 cell differentiation.32 However, neither genotype of BMDC produced detectable levels of IL‐4 upon infection, which could direct Th2 cell differentiation (data not shown). Interleukin‐6 has been determined to control Th2 cell differentiation.43 Higher IL‐6 production in BMDCs from WT mice upon H1N1 infection is probably responsible for their greater potential to induce Th2 cell differentiation. In addition, IL‐6 is important for Th17 cell generation.34 Combined with transforming growth factor‐β ₁ produced by BMDCs, the higher IL‐6 levels in BMDCs obtained from WT mice should generate more Th17 cells upon H1N1 infection than BMDCs from Beclin‐1^+/− mice. Due to the reciprocal modulation of Th17 cells and Treg cells,44 it is reasonable to assume that BMDCs obtained from Beclin‐1^+/− mice induce more Treg cells and fewer Th17 cells. Furthermore, the influenza A H1N1 vaccine is known to up‐regulate Th1/Th2 responses,45 and Th17 has been reported to exert a protective effect during influenza infection.46 Hence, autophagy is likely to control anti‐H1N1 virus immunity by affecting Th cell differentiation.

Limited evidence suggests that autophagy plays a role in promoting MHC‐I cross‐presentation in DCs. Atg5‐deficient DCs are intact in the MHC‐I cross‐presentation.47 However, in that report, the authors did not detect a decline in autophagy in the DCs of Atg5‐deficient mice after peptide stimulation. Moreover, the phenotypes of DCs obtained from Atg5‐deficient or WT were similar. According to our results, Beclin‐1^+/− and WT BMDCs express similar MHC‐I molecules and MHC‐I/OVA_257–264 peptide complexes, which suggests that autophagy does not alter antigen processing via the conventional MHC‐I pathway. However, the lower expression of co‐stimulatory molecules in Beclin‐1^+/− BMDCs may compromise their MHC‐I cross‐presenting ability. In addition, the secretion of IL‐12 in Beclin‐1^+/− BMDCs is inhibited after activation, which is likely to impair their MHC‐I cross‐presenting ability because IL‐12 has been reported to enhance DC MHC‐I cross‐presentation.48

In summary, we found that H1N1 virus infection induces autophagy in DCs. Autophagy‐deficient DCs exhibited a less mature phenotype and reduced production of IL‐6, TNF‐α, IFN‐β, IL‐12p70 and IFN‐γ cytokines upon H1N1 infection. We provide direct evidence that autophagosomes transport H1N1 viruses into lysosomes to activate the TLR signalling pathway and found that autophagic deficiency impaired both the antigen‐presenting ability of DCs and their ability to induce Th cell differentiation after H1N1 infection. Autophagic deficiency also inhibited the MHC‐I cross‐presentation of DCs upon H1N1 infection. Furthermore, autophagy‐deficient BMDCs were compromised in their ability to induce H1N1‐specific innate and adaptive immune responses in vivo. Our results reveal the profile of DC immune response upon H1N1 infection, thereby providing new possible targets for immunotherapy to treat H1N1 influenza.

Authors' contributions

FZ, YC and ZL designed and performed the experiments, analysed the results, drafted the figures and manuscript. ZC, LY FX and JW participated in experiments, analysed the results and drafted the figures. WZ and HL designed the study, interpreted the data and edited the manuscript.

Disclosures

There are no financial conflicts of interest to declare.