Innate Immune Interleukin

昵称37064826 2020-03-22

展开全文

Introduction

Viral myocarditis leading to dilated cardiomyopathy is an important contributor for heart failure.^1–3 Common causal agents for myocarditis are cardiotropic viruses such as the coxsackievirus B3 (CVB3).^2,4 Clinical presentation, however, is highly variable. Some patients experience rapidly progressive heart failure or sudden cardiac death, whereas others develop only mild disease or are even completely asymptomatic.^2,4 It is not known, however, which factors determine the extent of cardiac inflammation and outcome in an individual patient.

Clinical Perspective on p 1554

Host pattern recognition receptors of the innate immune system, such as Toll-like receptors (TLRs) and retinoic acid–inducible gene I (RIG-I)–like receptors (RLRs), identify invading viruses, mediate inflammatory responses, and initiate antiviral pathways such as interferon regulatory factor (IRF)–dependent type I interferon (IFN) release.⁵ TLR-triggered inflammatory responses largely result from MyD88/interleukin-1 receptor–associated kinase 4 (IRAK4)–dependent translocation of nuclear factor-κB (NF-κB).^5,6 In this context, we have demonstrated previously that MyD88 aggravates inflammation in viral myocarditis, and its removal conferred protection through intriguing cross talk with IRF3.⁷ Indeed, increased IFN-β production related to promoted IRF3 dimerization was observed in the absence of MyD88, suggesting an inhibitory function of MyD88 on IRF3.⁷ Besides NF-κB activation, TLR3, TLR4, TLR7, and cytoplasmic RLRs, such as the melanoma differentiation-associated protein 5 (MDA5), which has specificity toward RNA viruses such as the cardiotropic CVB3, induce TANK-binding kinase 1/IKK-ε–dependent activation of IRFs to produce antiviral type I IFNs.^8–10 Several members of the IRF family are known to regulate type I IFN transcription. In particular, IRF7 is considered to be a master regulator for both IFN-α and IFN-β, IRF3 plays a critical role in IFN-β gene activation,^11,12 and IRF5 promotes IFN-α production.^13,14 Type I IFNs, however, are critical to protect the host from viral myocarditis.^4,15

IRAK4, a downstream adaptor molecule to MyD88, is essential for proinflammatory cytokine production on stimulation of interleukin (IL)-1βR, tumor necrosis factor-αR, TLR2, TLR4, TLR7/8, or TLR9 in macrophages and myeloid dendritic cells.^6,16,17 In fact, IRAK4 contributes to TLR2- and TLR4-triggered inflammation in a mouse model of myocardial infarction.^18,19

In the present study, we demonstrate that loss of IRAK4 protects from myocarditis. Despite drastic reduction of TLR-mediated inflammation in the absence of IRAK4, antiviral pathways were markedly activated. This combined effect of enhanced antiviral activity and impaired proinflammatory cytokine release may point to potentially novel therapeutic concepts for virus-triggered inflammatory heart disease.

Methods

Additional details for Methods are provided in the online-only Data Supplement.

Mice, Cells, and Viruses

All mice were of C57Bl/6 background. IRAK4^−/− and MyD88^−/− mice have been described previously.^6,7,18 CCR5^−/−, CCL5^−/−, and CD45.1 wild-type mice were purchased from The Jackson Laboratories. MDA5^−/− mice were from Dr S. Akira of Japan.¹⁰ All animal experiments were conducted in accordance with the Animal Care Committee of the University Health Network of the University of Toronto. The IRAK1/4 kinase inhibitor was from EMD Millipore. The cardiovirulent CVB3 (Gauntt strain) has been described previously.^3,20

Competitive Bone Marrow Chimeric Mice and Macrophage Adoptive Transfer

Mice were lethally irradiated with 2 doses of 6.5 Gy with the use of a GammaCell 40 Exactor (¹³⁷Cs) source and reconstructed intravenously with 10⁷ donor crude bone marrow cells. Reconstitution checked by flow cytometry was 46.6±2.8% for CD45.1 and 53.1±3.2% for CD45.2 donors in both groups. Adoptive transfer was performed intravenously with 10⁶ monocytes/macrophages 2 days after CVB3 infection (10⁵ plaque-forming units [PFU] per mouse).

Isolation of Heart-Infiltrating Cells, Flow Cytometry Analysis, and Cell Sorting

Hearts were digested with the use of Liberase TM Research Grade (Roche Diagnostics) at a working concentration of 0.08 Wuensch U/mL for 45 minutes at 37°C. Heart-infiltrating cells were always triple-stained and first gated on live CD45^high lymphocytes plotted on side-scatter/CD45⁺ cells, acquired with a BD LSR II or a BD FACSCanto II (BD Biosciences), and analyzed with FlowJo (Tree Star). Purity of CD45^high heart-infiltrating cells and CD45⁻ heart cells sorted with the use of a MoFloXDP (Beckman Coulter) was always >97.5%.

Quantitative and Semiquantitative Reverse Transcription Polymerase Chain Reaction

RNA was isolated with Trizol reagent (Invitrogen) according to the manufacturer’s protocol. Reverse transcription was performed with 1 µg RNA. The 2^−ΔΔCt method was used for quantitative reverse transcription polymerase chain reaction gene expression analysis.

Enzyme-Linked Immunosorbent Assay

A commercially available VeriKine ELISA kit (PBL Interferon Source) was used to measure IFN-α, and ELISA Development Kits (PeproTech) were used to measure IFN-γ, CCL4 (macrophage inflammatory protein-1β), and CCL5 (RANTES [regulated on activation, normal T cell expressed and secreted]).

Western Blot Analysis and Coimmunoprecipitation

Tissue or cell lysates were separated with the use of NuPAGE Novex Bis-Tris gel (Invitrogen) or self-made Tris-HCl nondenaturing gel, transferred on polyvinylidene fluoride membrane (Roche Diagnostic), and then immunoblotted with specific antibodies. Bands were visualized with an ECL chemiluminescent substrate (Luminata Crescendo, Millipore). For coimmunoprecipitation, antibodies were first purified with the Antibody Clean-up Kit (Pierce, Thermo Scientific), immobilized, and then covalently coupled to the beads. Proteins were incubated for immunoprecipitation according to the manufacturer’s instructions (IPeX Kit, GeBA).

Statistical Analysis

Survival rates were analyzed by the Kaplan-Meier method, and differences between groups were tested with the log-rank test. Not normally distributed data were analyzed by the nonparametric 2-tailed Mann-Whitney test or by the Kruskal-Wallis test. Data from different treatment groups at different time points were analyzed with 2-way ANOVA and Bonferroni post hoc testing for multiple subgroup comparisons. Data are presented as medians with dot plots or medians with ranges. Statistical analysis was conducted with the use of Prism 5 software (GraphPad Software). Differences were considered statistically significant for P<0.05.

Results

IRAK4 Promotes CVB3 Infection

To assess the relevance of IRAK4 in viral myocarditis, we inoculated C57Bl/6 IRAK4^−/− mice and their IRAK4^+/+ littermate controls with 10⁵ PFU CVB3 per mouse. IRAK4-deficient mice had very low mortality and lower viral proliferation at day 7 after infection compared with their littermate controls (Figure 1A through 1C). Heart function was also better in IRAK4^−/− mice than in IRAK4^+/+ mice (Figure 1D and Figure IA in the online-only Data Supplement). Brain natriuretic peptide significantly increased in IRAK4^+/+ compared with IRAK4^−/− hearts (Figure 1E). Analyzing the levels of infiltrating cells in total hearts, we observed that percentages and absolute numbers per gram of tissue of CD45^high leukocytes were comparable at the baseline level (day 0, before infection) and at day 4, whereas percentages and absolute numbers per gram of tissue of CD45^high cells were markedly enhanced in IRAK4^−/− hearts compared with IRAK4^+/+ hearts 2 days after infection (Figure 1F and 1G). Immunohistochemistry methods were inefficient to detect the small but relevant number of heart-infiltrating cells at day 2 after infection detected by flow cytometry (not shown). Intriguingly, CD11b⁺F480⁺ monocytes/macrophages were predominant among the CD45^high cells reaching the infected heart at day 2 (Figure 1F and 1G).

**Figure 1.** Interleukin-1 receptor–associated kinase 4 (IRAK4)–deficient mice are protected from coxsackievirus B3 (CVB3) infection. A, Survival curve of IRAK4^−/− (n=30) and their IRAK4^+/+ (n=60) littermate control mice (both C57Bl/6 background) intraperitoneally challenged with 10⁵ plaque-forming units (PFU) CVB3 per mouse; log-rank test, P=0.0001. B, Viral titers in the hearts were assessed at days 4, 7, and 10 after infection with the use of plaque-forming assay in HeLa cells; n=5. C, Histological analysis of heart cross sections stained with hematoxylin and eosin at day 7 after infection; magnification ×200 (**left**) and pathological score at indicated days (**right**); n=5. D, Fractional shortening evaluated by echocardiography before CVB3 infection (baseline) and 4 days after infection (CVB3-infected); n=6. E, Brain natriuretic peptide (BNP) was measured by quantitative reverse transcription polymerase chain reaction in total heart tissue collected at indicated days; n=5. F, Flow cytometry analysis of hearts from uninfected (day 0) or CVB3-infected IRAK4^−/− and IRAK4^+/+ mice (10⁵ PFU per mouse) at days 2 and day 4 after infection. Percentages of triple-stained CD45^high-gated heart-infiltrating CD11b⁺F480⁺ macrophages, CD11b⁺Ly6G⁺ granulocytes, CD11c⁺CD11b⁺ myeloid dendritic cells, CD11c⁺B220⁺ plasmacytoid dendritic cells, CD11c⁺CD8⁺ conventional dendritic cells, CD49b⁺CD3⁻ natural killer cells, and CD49b⁺CD3⁺ natural killer T cells are shown; n=4. G, Percentages of heart-infiltrating cell populations (above) and absolute numbers of cells calculated per gram of heart tissue (below); n=4. *P<0.05, **P<0.01, ***P<0.001. One of 3 independent experiments is shown. SSC indicates side scatter.

IRAK4 Reduces Migration of Protective Cells to the Heart During Early Myocarditis

To evaluate the role of heart-infiltrating cells in early myocarditis, we first screened chemokines by microarray in wild-type CVB3-infected hearts 2 days after infection and selected the most overexpressed chemokine ligands and their receptors for further characterization (Figure IIB in the online-only Data Supplement). To this end, heart-infiltrating CD45^high cells were separated from other cardiac cells (hereafter referred to as CD45⁻) by cell sorting 2 days after infection. Quantitative reverse transcription polymerase chain reaction analysis revealed CCL5 as the major chemokine ligand overexpressed in IRAK4^−/− CD45⁻ compared with IRAK4^+/+ CD45⁻ cells (Figure 2A). In addition, CCR5, the major chemokine receptor of CCL5, was significantly more highly expressed on IRAK4^−/− compared with IRAK4^+/+ CD45^high leukocytes (Figure 2B). Importantly, serial quantitative reverse transcription polymerase chain reaction analysis at days 1, 2, and 4 revealed that CCR5 and its ligands (namely, CCL3, CCL4, and CCL5) were exclusively overexpressed 2 days after infection in IRAK4^−/− hearts (Figure 2B). Moreover, CCR5 expression was higher on heart- and pancreas-infiltrating CD45^high cells in IRAK4^−/− mice 2 days after infection but not 4 days after infection (Figure 2C). These findings clearly suggest an early peak of protective CD45^high cells migrating to the heart during the acute phase of viral myocarditis in IRAK4^−/− mice.

**Figure 2.** Interleukin-1 receptor–associated kinase 4 (IRAK4) reduces migration of bone marrow–derived CCR5⁺ macrophages toward the heart. A, CD45⁻ and CD45^high cells were sorted from IRAK4^−/− (white bars) and IRAK4^+/+ (black bars) hearts collected 2 days after coxsackievirus B3 (CVB3) infection (10⁵ plaque-forming units [PFU] per mouse). RNA levels of chemokine ligands (in CD45⁻ cells) and chemokine receptors (in CD45^high cells) were analyzed by quantitative reverse transcription polymerase chain reaction; n=5. B, Total heart tissue from CVB3-injected (10⁵ PFU per mouse) IRAK4^+/+ (white bars) and IRAK4^−/− (black bars) mice collected at days 1, 2, and 4 for RNA extraction analyzed by quantitative reverse transcription polymerase chain reaction; n=5. C, Hearts and pancreata harvested from IRAK4^+/+ (black line) and IRAK4^−/− (blue line) mice infected with CVB3 (10⁵ PFU per mouse) were analyzed by flow cytometry for CD45^highCCR5⁺ infiltrating cells and CD45⁻CCR5⁺ heart or pancreas cells at days 2 and 4 after infection. Gray-tinted histograms represent isotype control; n=5 per group. D, CD45.2 IRAK4^−/− and CD45.2 IRAK4^+/+ bone marrow macrophages (BMM) were stimulated in vitro with TLR7 stimulant and then transferred intravenously into CVB3-infected (10⁵ PFU per mouse) CD45.1 wild-type mice 2 days after infection. Twelve hours after transfer, hearts were harvested, and CD45.2 macrophages were analyzed by flow cytometry. Cells were gated on a CD45.1⁻CD45(.1⁻/.2⁺) population and then on CD45.1⁻CD45(.1⁻/.2⁺) heart-infiltrating cells; n=5 per group. E, Survival scores of wild-type mice infected and transferred with BMM as in D; n=5. F, CD45.1 recipient mice were lethally irradiated and reconstructed with equal amounts of CD45.1 wild-type (WT) and CD45.2 IRAK4^+/+ or of CD45.1 WT and CD45.2 IRAK4^−/− crude bone marrow cells. Heart-infiltrating cells collected from CVB3-infected (10⁵ PFU per mouse) hearts 2 days after infection were analyzed by flow cytometry with antibodies against CD45.1 and CD45.2 cells. CD45.2 IRAK4^+/+ (black line) and CD45.2 IRAK4^−/− (blue line) cells were stained for CCR5⁺; n=5 per group. *P<0.05, **P<0.01, ***P<0.001. One of 3 independent experiments is shown.

To test the effect of IRAK4 on CD11b⁺F480⁺ monocyte/macrophage migration, we adoptively transferred TLR7-stimulated CD45.2⁺ IRAK4^+/+, CD45.2⁺ IRAK4^−/−, or control CD45.2⁺ CCR5^−/− bone marrow macrophages (BMM) into CVB3-infected CD45.1⁺ wild-type mice 2 days after infection. As hypothesized, we observed almost double the amount of heart-infiltrating CD45.2⁺ IRAK4^−/− BMM compared with CD45.2⁺ IRAK4^+/+ BMM (Figure 2D). In addition, CVB3-infected wild-type mice adoptively transferred with IRAK4^−/− BMM, but not IRAK4^+/+ BMM, showed significantly reduced mortality (Figure 2E). Next, we generated competitive mixed bone marrow chimeric mice and observed a remarkably higher ratio of CD45.2⁺ IRAK4^−/− to CD45.1⁺ IRAK4^+/+ cells (5.1/1.0) 2 days after infection compared with the CD45.2⁺ IRAK4^+/+ to CD45.1⁺ IRAK4^+/+ ratio (1.68/1.0) (Figure 2F). Furthermore, CCR5 expression was significantly higher on CD45.2⁺ IRAK4^−/− cells than on CD45.2⁺ IRAK4^+/+ cells in infected hearts, confirming the bone marrow origin of CCR5⁺ heart-infiltrating cells (Figure 2F). Taken together, these data demonstrate that migration of bone marrow–derived CCR5⁺CD11b⁺ monocytes/macrophages to the heart during the first 2 days of viral myocarditis is markedly enhanced in the absence of IRAK4, showing that IRAK4-deficient monocytes/macrophages specifically inhibit viral myocarditis.

CCR5 and CCL5 Protect Against CVB3 Infection

Because CVB3-infected IRAK4^−/− mice showed increased amounts of CD45^high/CCR5⁺ and CCL5-producing CD45⁻ cells in the heart 2 days after CVB3 infection (Figure 2A and 2B), we hypothesized that the CCR5/CCL5 axis may be crucial for protection against early CVB3 myocarditis. Accordingly, we infected CCR5^−/−, CCL5^−/−, and wild-type control mice with 10⁴ PFU CVB3 per mouse. As expected, mortality rates and viral proliferation were dramatically increased in both CCR5^−/− and CCL5^−/− mice (Figure 3A and 3B). Amounts of heart-infiltrating CD45^high leukocytes were comparable in CCL5^+/+ and CCL5^−/− mice 2 days after infection, but we observed remarkable differences in the CD11b⁺F480⁺ monocyte/macrophage subpopulation. Taken together, these findings emphasize the importance of the CCR5/CCL5 axis in viral protection (Figure 3C and 3D).

**Figure 3.** The CCR5/CCL5 axis confers protection against coxsackievirus B3 (CVB3). A, Survival curve of CCR5^−/− (white squares; n=20), CCL5^−/− (gray squares; n=20), and wild-type (WT; black squares; n=20) mice (all C57Bl/6 background) after infection with 10⁴ plaque-forming units (PFU) CVB3 per mouse intraperitoneally; log-rank test, *P=0.0099 for WT vs CCL5^−/−, **P=0.0001 for WT vs CCR5^−/−. B, Viral titers in the hearts of mice described in A were assessed 2 days after CVB3 infection with the use of plaque-forming assay in HeLa cells; n=4. C, Flow cytometry analysis of CCL5^−/− and WT hearts 2 days after CVB3 infection. Triple staining of CD45^high-gated heart-infiltrating CD11c⁺CD11b⁺ myeloid dendritic cells, CD11c⁺PDCA1⁺ plasmacytoid dendritic cells, CD11c⁺CD8α⁺ conventional dendritic cells, CD11b⁺F4/80⁺ macrophages, and CD11b⁺Ly6G⁺ granulocytes is shown; n=5 per group. D, Heart-infiltrating CD45^high cells as in C were stained for CCR5⁺. Black histogram represents WT, blue histogram represents CCL5^−/−, and gray-tinted histogram represents isotype control; n=5. One of 2 independent experiments is shown.

IRAK4 Negatively Regulates CCR5 Expression After CVB3 Infection

To determine the mechanism of IRAK4-mediated inhibition of CCR5 expression, we first measured chemokine receptors on CVB3-infected macrophages in vitro. As expected, only CCR5 transcripts were consistently increased in IRAK4^−/− BMM after CVB3 infection and on stimulation with polyinosine–polycytidynic acid (TLR3) and R848 (TLR7), 2 TLR ligands/adjuvants mimicking CVB3-viral double-strand genome and CVB3-viral single-strand genome, respectively (Figure 4A through 4C and Figure IC in the online-only Data Supplement). Flow cytometry and immunofluorescent staining confirmed that IRAK4 directly inhibited CCR5 (Figure 4D and 4E). Because CCR5 is usually expressed on virus-infected T cells, and regulatory T cells are protective against viral myocarditis,²⁰ we analyzed regulatory T cell levels in the spleens, mediastinal lymph nodes, hearts, or pancreata of CVB3-infected IRAK4^−/−, CCR5^−/−, and their littermate controls, but we did not observe any differences (Figure ID and IE in the online-only Data Supplement). Moreover, CAR and Lck, which are required for viral internalization and for T cell–mediated heart inflammation, were similar in the hearts of CVB3-inoculated IRAK4^+/+ and IRAK4^−/− mice (Figure IF in the online-only Data Supplement). These findings collectively suggest that IRAK4 does not modulate CCR5 expression on T cells but rather on monocytes/macrophages.

**Figure 4.** Interleukin-1 receptor–associated kinase 4 (IRAK4) modulates CCR5 expression. A, IRAK4^+/+ (black bars) and IRAK4^−/− (white bars) bone marrow macrophages were stimulated in vitro with R848. RNA levels of the indicated chemokine receptors were measured by quantitative reverse transcription polymerase chain reaction; n=4. *P<0.05, **P<0.01. B and C, Semiquantitative reverse transcription polymerase chain reaction evaluation of polyinosine–polycytidynic acid (polyIC)–stimulated (B) or coxsackievirus B3 (CVB3)–infected (C) bone marrow macrophages. GAPDH was used as internal control. D, CCR5 expression was evaluated by flow cytometry in IRAK4^+/+, IRAK4^−/−, and CCR5^−/− bone marrow macrophages after 48- and 72-hour in vitro infection with CVB3. Cells were first gated on CD11b⁺, then on F4/80⁺/CCR5⁺; n=5. E, CCR5 immunofluorescent staining on bone marrow macrophages infected for 48 hours with CVB3; magnification, ×200. The graph shows the percentage of counted CCR5⁺ cells (red) divided by the nuclei (blue) ×100; n=5. *P<0.05. F, IRAK4^+/+ and IRAK4^−/− bone marrow macrophages were stimulated with R848 for the indicated times, and phosphorylated Stat5 and Stat1 were measured in total cell lysates. G, Intranuclear flow cytometric analysis of phosphorylated Stat5 4 hours after in vitro infection with CVB3. Black histogram represents IRAK4^+/+ BMM, blue histogram represents IRAK4^−/− bone marrow macrophages, and gray solid histogram represents the staining control with secondary antibody only. Data are representative of 1 of 2 to 3 independent experiments.

According to a search of databases, Stat5 is the major transcription factor with high score and low e-value specific for the promoter region of CCR5.²¹ In fact, in vitro TLR7 stimulation and CVB3 infection of IRAK4^−/− BMM resulted in higher Stat5 phosphorylation, along with Stat1 phosphorylation, compared with IRAK4^+/+ BMM (Figure 4F and 4G). Thus, IRAK4 negatively regulates the major CCR5 transcription factor.

IRAK4 Downregulates Antiviral Protection

IRAK4 promotes proinflammatory cytokine production after NF-κB phosphorylation.^6,16,17 It is not clear, however, whether IRAK4 affects antiviral responses. CD45^high leukocytes isolated from hearts 2 days after CVB3 infection expressed significantly higher RNA levels of IFN-α1 and IFN-γ, whereas CD45⁻ heart-derived cells showed higher levels of IFN-α4 in IRAK4^−/− compared with IRAK4^+/+ mice (Figure 5A). Furthermore, in vitro–stimulated IRAK4^−/− BMM consistently showed increased IFN-α and IFN-γ production compared with IRAK4^+/+ BMM on CVB3 infection and after TLR3 or TLR7 stimulation (Figure 5B and Figures IIA, IIB, IIIA, and III in the online-only Data Supplement). Of note, IFN-α and IFN-γ overexpression was observed only in IRAK4^−/− BMM but not in bone marrow–derived dendritic cells or plasmacytoid dendritic cells (Figure 5B and Figure IIIC in the online-only Data Supplement). These latter findings suggest that IRAK4 specifically inhibits IFN-α and IFN-γ production in monocytes/macrophages on CVB3 infection.

**Figure 5.** Interleukin-1 receptor–associated kinase 4 (IRAK4) downregulates antiviral protection. A, CD45^high cells were sorted from coxsackievirus B3 (CVB3)–injected IRAK4^−/− (white bars) and IRAK4^+/+ (black bars) hearts 2 days after infection and separated from CD45⁻ heart cells. RNA levels of type I and II interferon (IFN) and interleukin-6 (IL-6) were measured by quantitative reverse transcription polymerase chain reaction; n=5. B, IFN-α and IFN-γ concentrations were measured by ELISA at the indicated times after in vitro stimulation with polyinosine–polycytidynic acid (polyIC) in bone marrow macrophages (BMM) and bone marrow–derived dendritic cells (BMDC), R848 in BMM and plasmacytoid dendritic cells (pDC), or infection with CVB3 in BMM (n=4). C, Detection of genomic positive-strand CVB3 in IRAK4^+/+, IRAK4^−/−, MyD88^−/−, and MDA5^−/− BMM. Cells (10⁶) were left untreated (−) or infected with CVB3 (+) for 24 hours. Positive-strand CVB3 sequence and cellular GAPDH were detected in total RNA with the use of reverse transcription polymerase chain reaction. D, Detection of infectious CVB3 in IRAK4^+/+, IRAK4^−/−, MyD88^−/−, and MDA5^−/− BMM infected for 24 hours with CVB3 analyzed with the use of plaque-forming assay in HeLa cells; n=4. E, Wild-type intraperitoneal macrophages were pretreated with dimethyl sulfoxide (vehicle) or IRAK1/4 kinase inhibitor (IRAK1/4 IHN) dissolved in dimethyl sulfoxide for 1 hour, then infected for 24 hours with CVB3 and finally analyzed with the use of plaque-forming assay in HeLa cells; n=4. F, Plaque assay of IRAK4^+/+ (black bars) and IRAK4^−/− (white bars) mouse neonatal cardiomyocytes infected with CVB3 for the indicated times; n=4. G, Levels of CCL4, CCL5, IFN-α, and IFN-γ were measured by ELISA in the supernatants of IRAK4^+/+ (black bars) and IRAK4^−/− (white bars) mouse neonatal cardiomyocytes 24 hours after CVB3 infection; n=4. *P<0.05, **P<0.01. Data are representative of 1 of >3 independent experiments.

To determine whether reduced IFN production impairs CVB3 protection in macrophages, we measured CVB3 propagation in vitro. As expected, reduced CVB3 genomic +RNA replication and lower viral proliferation were observed in IRAK4^−/− and MyD88^−/− BMM compared with IRAK4^+/+ and CVB3 infectivity–prone MDA5^−/− BMM (Figure 5C and 5D). In addition, CVB3 proliferation was reduced in wild-type thioglycollate-elicited intraperitoneal macrophages pretreated with a specific soluble IRAK1/4 kinase inhibitor compared with untreated macrophages (Figure 5E and Figure IIID in the online-only Data Supplement). Moreover, CVB3 proliferation was reduced in IRAK4^−/− mouse neonatal cardiomyocytes (Figure 5F). Interestingly, IRAK4^−/− mouse neonatal cardiomyocytes produced higher amounts of IFN-γ compared with IRAK4^+/+ mouse neonatal cardiomyocytes, although comparable amounts of CCL4 and CCL5, and undetectable levels of IFN-α (Figure 5G). Taken together, these findings show that CVB3 proliferation is impaired in the absence of IRAK4.

IRAK4 Countermodulates IRF3, IRF7, and IRF5 Dimerization

After recognition of viral genomes, cytoplasmic RNA helicase receptors start antiviral responses, which converge to several IRFs.^10,22 Analysis of hearts collected from infected mice at days 0, 1, 2, and 4 after infection showed comparable or lower levels of phosphorylated IRF3 but higher stability of MDA5 in IRAK4^−/− mice compared with the increased MDA5 degradation in IRAK4^+/+ mice (Figure 6A). Because IRF3 did not show significant differences, we evaluated IRF5, which promotes IFN-α transcription as homodimer but inhibits IFN-α transcription when heterodimerized with IRF7.^14,23 We found that IRF5 homodimerization was consistently higher in IRAK4^−/− compared with IRAK4^+/+ hearts (Figure 6B). Using direct in vitro infection with CVB3 and vesicular stomatitis virus (VSV), which is a negative single-strand RNA rhabdovirus that activates another member of the RLRs similar to MDA5 (namely, RIG-I), we found a different pattern of IRF protein upregulation in IRAK4^−/− BMM (Figure IVA through IVC in the online-only Data Supplement). Accordingly, activation of IRF3 and IRF7 was higher in VSV-infected IRAK4^−/− BMM compared with IRAK4^+/+ BMM, whereas CVB3, as already shown in the heart tissue, induced lower phosphorylation of IRF3 and equal nuclear translocation of IRF7 in IRAK4^−/− BMM (Figure 6C and 6D). Furthermore, high MDA5 degradation in IRAK4^+/+ BMM (Figure 6E) confirmed our in vivo observations. In addition, degradation of phosphorylated Stat1 was observed in IRAK4^+/+ BMM only.

Likewise, in vitro CVB3-infected IRAK4^−/− BMM showed increased amounts of homodimerized IRF5 compared with IRAK4^+/+ BMM or MyD88^−/− BMM (Figure 6F). Of note, similar to MDA5, IRF5 was completely degraded in IRAK4^+/+ BMM 9 hours after infection with CVB3 (Figure 6F). Therefore, IRF5 coimmunoprecipitation was performed 6 hours after CVB3 infection. IRF5 coimmunoprecipitation showed high affinity between IRF5 and IRF7 to heterodimerize in IRAK4^+/+ BMM compared with the very low affinity for heterodimerization in IRAK4^−/− BMM (Figure 6G). Taken together, IRAK4 preferentially promotes IRF5/IRF7 heterodimerization while inhibiting IRF5 homodimerization. Thus, IRAK4 modulates antiviral responses on the molecular level via virus-dependent IRF dimerization.

Direct Link Between IRAK4, CCR5/CCL5 Axis, and IFN Production

Evaluating cardiac cytokine levels of CVB3-infected mice 2 days after infection, we observed dramatically reduced IFN-α but an overall increase of inflammatory cytokines in CCR5^−/− and CCL5^−/− mice compared with wild-type controls (Figure 7A). These observations raised the question of whether CCR5 or CCL5 directly regulates IFN-α. To this end, we stimulated BMM with CCL3, CCL4, and CCL5. As hypothesized, only IFN-α was significantly overexpressed in IRAK4^−/− BMM after cultivation with rmCCL3, rmCCL4, or rmCCL5, suggesting that typical CCR5 ligands further enhance induction of IFN-α in the absence of IRAK4 (Figure 7B). In addition, CCR5 acted not only as chemokine receptor for cell mobilization. Indeed, in vitro infection with CVB3 resulted in higher expression of nuclear levels of p65 and lower levels of Stat1 and IRF7 in CCR5^−/− BMM compared with IRAK4^+/+ and IRAK4^−/− BMM, suggesting a yet unknown regulatory function of CCR5 in IFN expression (Figure 7C).

**Figure 7.** Interleukin-1 receptor–associated kinase 4 (IRAK4) exclusively downregulates CCR5 in a nonredundant manner. A, Type I interferon (IFN; **left**) and cytokine/chemokine (**right**) RNA levels were assessed in the hearts of wild-type (WT; black squares), CCR5^−/− (white squares), and CCL5^−/− (gray squares) mice 2 days after coxsackievirus B3 (CVB3) infection by quantitative reverse transcription polymerase chain reaction; n=4. B, IRAK4^+/+ (black bars) and IRAK4^−/− (white bars) bone marrow macrophages were stimulated in vitro with 50 ng/mL rmCCL3, rmCCL4, or rmCCL5. RNA levels of IFN-α, IFN-β, interleukin-6 (IL-6), and CCR5 were assessed by quantitative reverse transcription polymerase chain reaction; n=4. C, Nuclear extraction from IRAK4^+/+, IRAK4^−/−, CCR5^−/−, and CCR5^−/−×IRAK4^−/− double-knockout bone marrow macrophages infected with light dose of CVB3 and probed with Stat1, interferon regulatory factor (IRF)7, and p65 antibodies. GAPDH was used to test the purity of the nuclear extraction, and histone H3 was used as nuclear fraction control. D, Survival curve of IRAK4^−/− (gray triangles; n=6), CCR5^−/− (white squares; n=6), CCR5^−/−×IRAK4^−/− (gray diamonds; n=6), and WT (black squares; n=6) mice after infection with middle dose (10³ plaque-forming units per mouse) of CVB3 intraperitoneally; log-rank test, P=0.8164 for CCR5^−/− vs CCR5^−/−×IRAK4^−/−. *P<0.05, **P<0.01, ***P<0.001. Data are representative of 1 of 2 to 3 independent experiments.

To confirm that IRAK4 exclusively downregulates CCR5, we infected IRAK4^−/−, CCR5^−/−, IRAK4^+/+×CCR5^+/+, and IRAK4^−/−×CCR5^−/− double-knockout mice with 10³ PFU CVB3 per mouse. As expected, CCR5^−/− and IRAK4^−/−×CCR5^−/− mice had a very similar mortality rate, confirming direct and nonredundant IRAK4-dependent regulation of CCR5 (Figure 7D).

Discussion

The present study shows an inhibitory role of IRAK4 in viral myocarditis. For the first time, we reveal IRAK4 as a double-edged sword in viral myocarditis: It acts as a proinflammatory molecule but also blocks protective cell migration and antiviral responses. The ability of the host to limit viral proliferation while minimizing tissue injury attributable to detrimental proinflammatory responses, thus protecting from autoimmune postviral cardiomyopathy, is a prerequisite of favorable outcome.

IRAK4 is a well-known proinflammatory kinase downstream of all known TLRs except TLR3. It complexes with MyD88, IRAK1, IRAK2, and IRAK-M, phosphorylates IRAK1 and IRAK2, and ultimately contributes to NF-κB and mitogen-activated protein kinase activation for inflammatory cytokine transcription.^5,6,16,17 Accordingly, and as reported by our group earlier, IRAK4-deficient mice showed better survival and lower heart inflammation after experimental myocardial infarction.¹⁸ In addition, we recently found that MyD88, which is upstream of IRAK4, directly reduced viral protection by inhibiting IRF3 homodimerization and IFN-β production after CVB3-induced myocarditis.⁷ Another study showing that MyD88 inhibited IRF3-dependent production of IFN-γ on TLR3 stimulation and rhinovirus infection supported our observations.²⁴ However, it is known that murine cytomegalovirus–induced and encephalomyocarditis virus–induced myocarditis are less severe in MyD88-competent than in MyD88-deficient mice.^11,19 The reason for these discrepant findings may be explained by different activations of MyD88-dependent and MyD88-independent pathways. The MyD88-dependent antiviral pathway may exert a more protective function than the MyD88-independent one after murine cytomegalovirus and encephalomyocarditis virus infection. In the present study, MyD88 seems not to counterregulate IRFs. In the case of CVB3 and rhinovirus infection, MyD88, as well as IRAK4, inhibits IRFs-dependent production of type I IFNs on one side while promoting detrimental proapoptotic cytokines on the other side. Future studies should clarify the complex molecular mechanism of IRF interaction with respect to MyD88 and IRAK4 after infection with common cardiotropic viruses.

IRAK4-dependent regulation of IFN-γ production on viral infection or TLR7 stimulation has never been studied in macrophages. Indeed, some studies demonstrated that IRAK4^−/− plasmacytoid dendritic cells were unresponsive to polyinosine–polycytidynic acid and R848 to produce TLR3- and TLR7-dependent type I IFNs, respectively, or after influenza-virus infection, whereas IRAK4 kinase–deficient mouse embryonic fibroblasts and macrophages showed impaired NF-κB and mitogen-activated protein kinase pathways after TLR7 stimulation.^17,25 However, none of these reports clarified how IRAK4 may affect type I IFN production in macrophages. Only 1 report introduced the idea that IRAK4 may be unnecessary to produce IFN-α, showing slightly higher amounts of IFN-α in IRAK4 kinase–deficient bone marrow–derived cells on influenza-virus infection than in wild-type bone marrow–derived cells, indicating that IFN-α production is not reduced in the absence of IRAK4 but rather increased.¹⁷ In the present study, we showed for the first time that IRAK4 downregulated IFN-α production after CVB3 infection and TLR7 stimulation in macrophages.

IRF5 represents another, less studied member of the IRF family, which induces IFN-α when homodimerized but inhibits IFN-α when heterodimerized with IRF7.^13,14 In CVB3-infected IRAK4^−/− mice and macrophages, we observed increased amounts of homodimerized IRF5 but a very low affinity between IRF5 and IRF7 to heterodimerize. In contrast, MyD88-deficient macrophages showed impaired IRF5 homodimerization, which is consistent with the suppressed IRF5 nuclear translocation previously observed in the absence of MyD88 after TLR9 stimulation.²⁶ Collectively, IRAK4 reduces the potential of IRF5 to homodimerize and induce IFN-α, whereas MyD88 preferentially inhibits IRF3 and IFN-β, suggesting that molecules upstream of NF-κB modulate a cell-specific and sophisticated counterregulatory mechanism targeting different members of the IRF family to limit exaggerated and detrimental antiviral responses. In view of severe viral infections, however, this mechanism seems to be a double-edged sword because it drastically reduces antiviral protection.

Pathogen-associated molecular patterns are recognized by TLRs and cytoplasmic RLRs, such as MDA5. MDA5 has high specificity toward positive single-stranded picornaviruses, such as the cardiotropic picornavirus CVB3.¹⁰ MDA5 signals through the adaptor molecule MAVS, thus inducing TANK-binding kinase 1– and IKK-ε–dependent phosphorylation of several IRFs.⁵ Intriguingly, we observed that MDA5 underwent strong degradation after CVB3 infection in IRAK4-competent mice. Picornavirus-induced MDA5 cleavage and degradation, which impair the antiviral machinery and therefore increase viral propagation in the host, have been described previously.²⁷ Thus, our data link enhanced viral propagation and myocarditis susceptibility to IRAK4-mediated cleavage of relevant virus-recognizing receptors, such as MDA5.

Examining cell migration in early myocarditis, we found high percentages and absolute numbers per gram of heart tissue of CCR5⁺CD11b⁺F480⁺ monocytes/macrophages in IRAK4^−/− mice 2 days after infection but not before infection or after 4 days. Accumulation of CCR5⁺CD11b⁺F480⁺ monocytes/macrophages was associated with better survival and improved viral protection in IRAK4-deficient mice, suggesting that IRAK4 blocks the first crucial influx of protective cells of the innate immune system that migrate to the heart to limit the expansion and proliferation of CVB3. Remarkably, our data not only show the well-known importance of CCR5 for cell migration but also define CCR5 for the first time as a component of the innate antiviral machinery. Using double-knockout CCR5^−/−×IRAK4^−/− cells and mice, we conclusively demonstrated the mutual and unique cooperation between CCR5 and IRAK4. Taken together, early cardiac CCR5⁺CD11b⁺F480⁺ monocytes/macrophages represent an efficient cellular substrate of protective antiviral mechanisms in CVB3 myocarditis.

Clinical studies on patients with IRAK4 gene mutation showed severely impaired proinflammatory responses on specific bacterial infection or IL-1β stimulation. In contrast, no severe viral or parasitic infections were reported in the same patients.²⁸ IRAK4-mutated human peripheral blood mononuclear cells and patient-derived fibroblasts were still responsive to TLR3 and TLR4 stimulations and produced IFN-α and IFN-β on in vitro infection with RNA and DNA viruses.²⁸ These clinical observations nicely fit our mechanistic findings in the CVB3 virus myocarditis model. Despite a critical role for IRAK4 in promoting systemic inflammatory responses, IRAK4 also suppresses several antiviral key mechanisms. Thus, IRAK4 appears to be a promising target for new treatments. Thus far, several promising compounds have been developed that precisely reduce IRAK4 kinase–triggered responses in a balanced fashion, reducing IL-1β– and TLR-triggered excessive and detrimental immune responses.²⁹ Thus far, treatments of viral myocarditis with NF-κB inhibitors failed to show any benefit.³⁰ Our results, however, suggest a novel therapeutic concept that combines the use of specific IRAK4 inhibitors together with antiviral compounds. The resiquimod R848 has already been used as antiviral agent. According to our study, polyinosine–polycytidynic acid and R848 promote higher production of IFN-α and IFN-γ and increase migration of protective cells in the absence of IRAK4. From a clinical point of view, our findings therefore strongly support the hypothesis that a combination of IRAK4 inhibitors and TLR3-7/RLR adjuvants might simultaneously improve antiviral responses and reduce proinflammatory signals in patients with acute viral myocarditis (Figure 8).

**Figure 8.** The mechanism orchestrated by interleukin-1 receptor–associated kinase 4 (IRAK4) to induce inflammation and reduce antiviral protection. **Left**, On coxsackievirus B3 (CVB3) infection, under normal conditions IRAK4 promotes melanoma differentiation-associated protein 5 (MDA5) degradation and interferon regulatory factor (IRF)5/IRF7 dimerization, ultimately leading to impaired transcription of interferon (IFN)-α. IRF3 dimerization induces too small amounts of IFN-β, and excessive IRAK4-dependent inflammatory cytokines, such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α), are detrimental, and the host dies. **Right**, In IRAK4 deficiency, MDA5 stability is improved, and IRF5 homodimerization is preferred to IRF5/IRF7 heterodimerization, leading to enhanced transcription and production of IFN-α. Simultaneously, higher levels of IFN-α and IFN-γ support Stat1 and Stat5 phosphorylation. Phosphorylated Stat5 promotes CCR5 transcription and expression. Consequently, CCR5⁺CD11b⁺F480⁺ monocytes/macrophages migrate from the bone marrow to the heart through a CCR5 ligands–dependent chemoattractant gradient, promoting CVB3 cleavage and clearance. NF-κB indicates nuclear factor-κB; and TLR, Toll-like receptor.

Limitations of This Study

We acknowledge that there are several limitations to our study. First, we cannot rule out a direct role of IRAK4 in cardiac myocytes as a contributing factor in the observed myocarditis exacerbation phenotype. Further studies are needed to clearly elucidate the role of IRAK4 in cardiac cells. Another limitation is that we are unable to confirm the differences in CCR5⁺CD11b⁺F480⁺ monocyte/macrophage infiltration in the myocardium at day 2 after infection by immunohistochemistry. This is likely attributable to the small amount of heart-infiltrating cells at this time point below the detection threshold of immunohistochemistry. Flow cytometry had the advantage of analyzing the entire heart tissue and its cellular contents.

Acknowledgments

We are most grateful to Dr Dana Philpott (University of Toronto) for her valuable suggestions and critical reading of the manuscript.

Sources of Funding

This research was supported by the Canadian Institutes of Health Research. Dr Valaperti was supported by the Myocarditis Foundation and the Novartis Foundation. Dr Eriksson received support from the Swiss National Foundation (32002B_130771).

Disclosures

None.

Footnotes

The online-only Data Supplement is available with this article at http://circ./lookup/suppl/doi:10.1161/CIRCULATIONAHA.113.002275/-/DC1.

Correspondence to Peter P. Liu, MD, University of Ottawa Heart Institute, 40 Ruskin St, Room H2263, Ottawa, Ontario, K1Y 4W7, Canada. E-mail peter.liu@utoronto.ca or pliu@ottawaheart.ca

References

1. Maekawa Y, Ouzounian M, Opavsky MA, Liu PP. Connecting the missing link between dilated cardiomyopathy and viral myocarditis: virus, cytoskeleton, and innate immunity.Circulation. 2007; 115:5–8.Link Google Scholar
2. Liu PP, Mason JW. Advances in the understanding of myocarditis.Circulation. 2001; 104:1076–1082.Crossref Medline Google Scholar
3. Liu P, Aitken K, Kong YY, Opavsky MA, Martino T, Dawood F, Wen WH, Kozieradzki I, Bachmaier K, Straus D, Mak TW, Penninger JM. The tyrosine kinase p56lck is essential in coxsackievirus B3-mediated heart disease.Nat Med. 2000; 6:429–434.Crossref Medline Google Scholar
4. Cooper LT. Myocarditis.N Engl J Med. 2009; 360:1526–1538.Crossref Medline Google Scholar
5. Kawai T, Akira S. The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors.Nat Immunol. 2010; 11:373–384.Crossref Medline Google Scholar
6. Suzuki N, Suzuki S, Duncan GS, Millar DG, Wada T, Mirtsos C, Takada H, Wakeham A, Itie A, Li S, Penninger JM, Wesche H, Ohashi PS, Mak TW, Yeh WC. Severe impairment of interleukin-1 and Toll-like receptor signalling in mice lacking IRAK-4.Nature. 2002; 416:750–756.Crossref Medline Google Scholar
7. Fuse K, Chan G, Liu Y, Gudgeon P, Husain M, Chen M, Yeh WC, Akira S, Liu PP. Myeloid differentiation factor-88 plays a crucial role in the pathogenesis of coxsackievirus B3–induced myocarditis and influences type I interferon production.Circulation. 2005; 112:2276–2285.Link Google Scholar
8. Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3.Nature. 2001; 413:732–738.Crossref Medline Google Scholar
9. Diebold SS, Kaisho T, Hemmi H, Akira S, Reis e Sousa C. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA.Science. 2004; 303:1529–1531.Crossref Medline Google Scholar
10. Kato H, Takeuchi O, Sato S, Yoneyama M, Yamamoto M, Matsui K, Uematsu S, Jung A, Kawai T, Ishii KJ, Yamaguchi O, Otsu K, Tsujimura T, Koh CS, Reis e Sousa C, Matsuura Y, Fujita T, Akira S. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses.Nature. 2006; 441:101–105.Crossref Medline Google Scholar
11. Honda K, Yanai H, Negishi H, Asagiri M, Sato M, Mizutani T, Shimada N, Ohba Y, Takaoka A, Yoshida N, Taniguchi T. IRF-7 is the master regulator of type-I interferon-dependent immune responses.Nature. 2005; 434:772–777.Crossref Medline Google Scholar
12. Sato M, Suemori H, Hata N, Asagiri M, Ogasawara K, Nakao K, Nakaya T, Katsuki M, Noguchi S, Tanaka N, Taniguchi T. Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction.Immunity. 2000; 13:539–548.Crossref Medline Google Scholar
13. Ning S, Huye LE, Pagano JS. Interferon regulatory factor 5 represses expression of the Epstein-Barr virus oncoprotein LMP1: braking of the IRF7/LMP1 regulatory circuit.J Virol. 2005; 79:11671–11676.Crossref Medline Google Scholar
14. Barnes BJ, Field AE, Pitha-Rowe PM. Virus-induced heterodimer formation between IRF-5 and IRF-7 modulates assembly of the IFNA enhanceosome in vivo and transcriptional activity of IFNA genes.J Biol Chem. 2003; 278:16630–16641.Crossref Medline Google Scholar
15. Deonarain R, Cerullo D, Fuse K, Liu PP, Fish EN. Protective role for interferon-beta in coxsackievirus B3 infection.Circulation. 2004; 110:3540–3543.Link Google Scholar
16. Kawagoe T, Sato S, Jung A, Yamamoto M, Matsui K, Kato H, Uematsu S, Takeuchi O, Akira S. Essential role of IRAK-4 protein and its kinase activity in Toll-like receptor-mediated immune responses but not in TCR signaling.J Exp Med. 2007; 204:1013–1024.Crossref Medline Google Scholar
17. Kim TW, Staschke K, Bulek K, Yao J, Peters K, Oh KH, Vandenburg Y, Xiao H, Qian W, Hamilton T, Min B, Sen G, Gilmour R, Li X. A critical role for IRAK4 kinase activity in Toll-like receptor-mediated innate immunity.J Exp Med. 2007; 204:1025–1036.Crossref Medline Google Scholar
18. Maekawa Y, Mizue N, Chan A, Shi Y, Liu Y, Dawood S, Chen M, Dawood F, de Couto G, Li GH, Suzuki N, Yeh WC, Gramolini A, Medin JA, Liu PP. Survival and cardiac remodeling after myocardial infarction are critically dependent on the host innate immune interleukin-1 receptor-associated kinase-4 signaling: a regulator of bone marrow-derived dendritic cells.Circulation. 2009; 120:1401–1414.Link Google Scholar
19. Mann DL. The emerging role of innate immunity in the heart and vascular system: for whom the cell tolls.Circ Res. 2011; 108:1133–1145.Link Google Scholar
20. Shi Y, Fukuoka M, Li G, Liu Y, Chen M, Konviser M, Chen X, Opavsky MA, Liu PP. Regulatory T cells protect mice against coxsackievirus-induced myocarditis through the transforming growth factor beta-coxsackie-adenovirus receptor pathway.Circulation. 2010; 121:2624–2634.Link Google Scholar
21. Marinescu VD, Kohane IS, Riva A. The MAPPER database: a multi-genome catalog of putative transcription factor binding sites.Nucleic Acids Res. 2005; 33:D91–D97.Crossref Medline Google Scholar
22. Kato H, Takeuchi O, Mikamo-Satoh E, Hirai R, Kawai T, Matsushita K, Hiiragi A, Dermody TS, Fujita T, Akira S. Length-dependent recognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-I and melanoma differentiation-associated gene 5.J Exp Med. 2008; 205:1601–1610.Crossref Medline Google Scholar
23. Barnes BJ, Richards J, Mancl M, Hanash S, Beretta L, Pitha PM. Global and distinct targets of IRF-5 and IRF-7 during innate response to viral infection.J Biol Chem. 2004; 279:45194–45207.Crossref Medline Google Scholar
24. Siednienko J, Gajanayake T, Fitzgerald KA, Moynagh P, Miggin SM. Absence of MyD88 results in enhanced TLR3-dependent phosphorylation of IRF3 and increased IFN-β and RANTES production.J Immunol. 2011; 186:2514–2522.Crossref Medline Google Scholar
25. Koziczak-Holbro M, Joyce C, Glück A, Kinzel B, Müller M, Tschopp C, Mathison JC, Davis CN, Gram H. IRAK-4 kinase activity is required for interleukin-1 (IL-1) receptor- and toll-like receptor 7-mediated signaling and gene expression.J Biol Chem. 2007; 282:13552–13560.Crossref Medline Google Scholar
26. Takaoka A, Yanai H, Kondo S, Duncan G, Negishi H, Mizutani T, Kano S, Honda K, Ohba Y, Mak TW, Taniguchi T. Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors.Nature. 2005; 434:243–249.Crossref Medline Google Scholar
27. Barral PM, Morrison JM, Drahos J, Gupta P, Sarkar D, Fisher PB, Racaniello VR. MDA-5 is cleaved in poliovirus-infected cells.J Virol. 2007; 81:3677–3684.Crossref Medline Google Scholar
28. Yang K, Puel A, Zhang S, Eidenschenk C, Ku CL, Casrouge A, Picard C, von Bernuth H, Senechal B, Plancoulaine S, Al-Hajjar S, Al-Ghonaium A, Maródi L, Davidson D, Speert D, Roifman C, Garty BZ, Ozinsky A, Barrat FJ, Coffman RL, Miller RL, Li X, Lebon P, Rodriguez-Gallego C, Chapel H, Geissmann F, Jouanguy E, Casanova JL. Human TLR-7-, -8-, and -9-mediated induction of IFN-alpha/beta and -lambda is IRAK-4 dependent and redundant for protective immunity to viruses.Immunity. 2005; 23:465–478.Crossref Medline Google Scholar
29. Wang Z, Wesche H, Stevens T, Walker N, Yeh WC. IRAK-4 inhibitors for inflammation.Curr Top Med Chem. 2009; 9:724–737.Crossref Medline Google Scholar
30. Kindermann I, Barth C, Mahfoud F, Ukena C, Lenski M, Yilmaz A, Klingel K, Kandolf R, Sechtem U, Cooper LT, Böhm M. Update on myocarditis.J Am Coll Cardiol. 2012; 59:779–792.Crossref Medline Google Scholar