Essay Writing Service

The Role of Type I IFN in the Immune System

1.1      The role of type I IFN in the immune system

Type I IFNs consist of many different IFNs with structural homology such as IFN-α (13 different subtypes in human, 14 in mice), IFN-β and other type I IFNs that are less well characterised such as IFN-δ, IFN-ε, IFN-κ, IFN-τ, IFN- and IFN-ω. IFN-δ and IFN-τ are only described in pigs and cattle and do not have a homologue in humans {Pestka, 2004 #172}. Type II IFN (IFN-γ) and type III IFNs (IFN-λ1, IFN-λ2, IFN-λ3, IFN-λ4 {O’Brien, 2014 #174}) also contribute to the broader family of IFNs {Pestka, 2004 #172}.

1.1.1       Cellular sources and targets of type I IFN

A diverse range of cells produces type I IFNs, including most immune cells {Trinchieri, 2010 #181}. All type I IFNs bind to a common heterodimeric transmembrane receptor, the type I IFN receptor that is composed of 2 distinct subunits: the IFNAR1 and IFNAR2. IFNAR1 is constitutively associated with TYK2, while IFNAR2 is associated with JAK1. Upon ligation of the receptor these kinases get auto-phosphorylated. Activated JAK1 and TYK2 subsequently phosphorylate STAT molecules that are present in the cytosol. STATs that are activated in response to type I IFNs are STAT1, 2, 3 and 5 {Platanias, 2005 #157}. After phosphorylation activated STAT1 and STAT2 form heterodimers and translocate to the nucleus where they initiate transcription of IFN-stimulated genes (ISGs) {Platanias, 2005 #157}. The most important transcriptional complex that is induced is the STAT1-STAT2-IRF9 complex, known as IFN-stimulated gene factor 3 (ISGF3). This complex then binds to IFN-stimulated response elements (ISREs) to initiate gene transcription. This leads to the expression of several hundred ISGs, a large number of which function to induce an antiviral state within the cell. Wang et al. {Wang, 2017 #545} recently reported a novel phosphorylation site of STAT2 on T387 which leads to a negative regulation of this response and is constitutively phosphorylated in untreated cells. Upon type I IFN stimulation also other STAT complexes, specifically STAT1 homodimers, can form, translocate to the nucleus and bind to γ-activated sequences within DNA to regulate gene expression {Decker, 2005 #175}. In addition to JAK/STAT signalling, MAPKs are activated downstream of the type I IFN receptor including ERK2 {David, 1995 #176} and p38 {Uddin, 1999 #177}, although the first has been less well documented. p38 can generate IFN-mediated signals independent of STATs. Type I IFNs also activate PI3K signalling downstream of JAKs in an IRS-dependent but STAT-independent manner leading to activation of mTOR {Platanias, 2005 #157}.

This diversity of signalling pathways leads to the transcription of a wide range of genes including genes that encode cytokines and chemokines, antibacterial effectors, pro-apoptotic and anti-apoptotic molecules, and molecules involved in metabolic processes {Rauch, 2013 #775}. Hence, the type IFN response has several intracellular ‘‘check-points’’ and the regulation of its pathway is achieved by targeting signalling pathways leading to its expression, the transcription of ISGs, and the function of IFN-regulated proteins that they encode {Porritt, 2015 #318}. Mechanisms including the induction of the regulatory proteins suppressor of cytokine signalling (SOCS) {Piganis, 2011 #785} and ubiquitin specific peptidase (USP)18 {Francois-Newton, 2011 #787}{Malakhov, 2002 #786}. SOCS proteins are known to target the JAK/STAT pathway and thereby inhibiting STAT receptor binding and subsequent block of downstream signalling {Piganis, 2011 #785}. USP18 is involved in the process of deISGlyation {Malakhov, 2002 #786} and was recently shown to bind to the intracellular domain of the IFNAR2 and therefore inhibiting JAK1 binding and downstream IFN- signalling {Francois-Newton, 2011 #787}.

1.1.2       The role of type I IFN in immune responses and in disease

Type I IFNs induce a program of gene transcription interfering with multiple stages of viral replication but also influence the innate and adaptive immune response to various pathogens. The ability of type I IFNs to restrict viral replication is largely due to ISGs which are either expressed constitutively in cells in response to low levels of type I IFN in the microenvironment or most importantly different ISGs are activated or repressed depending on the pathways triggered upon IFNAR activation. Several ISGs such as Irf1Ifitm3 and Oasl, have been shown to inhibit viral replication {Schoggins, 2011 #178}. It was reported, that Ifnar1–/– mice are susceptible to vesicular stomatitis virus, Semliki forest virus, vaccinia virus and lymphocytic chorio- meningitis virus. However, susceptibility to influenza virus infection is dependent on type IIFN as well as type III IFN {Durbin, 2000 #776}{Garcia-Sastre, 1998 #779}{Koerner, 2007 #777}{Mordstein, 2008 #780}. However, a detrimental role for type I IFN in mouse models of intracellular bacterial infections including Listeria monocytogenes {Auerbuch, 2004 #179} and M.tuberculosis {Manca, 2005 #180} was shown. Type I IFN has been shown to promote lymphocyte apoptosis and subsequent dampening of the innate immune response which is in part due to the induction of IL-10 production by phagocytic cells in L.monocytogenes infection {Carrero, 2006 #781}. Additionally, type I IFN mitigates responsiveness to IFN-γ in L.monocytogenes infected macrophages {Rayamajhi, 2010 #782}. In the context of M.tuberculosis infection, type I IFN mediated the inhibition of the protective cytokine IL-1 {Mayer-Barber, 2011 #610}. In humans, a type I and type II IFN related gene expression signature in the blood has been shown to correlate with the radiological extent of disease {Berry, 2010 #341}. Nevertheless, type I IFNs can also have protective effects in bacterial infections such as Chlamydia trachomatis, Chlamydia pneumoniae, Legionella pneumophila, Shigella flexneri and Salmonella enterica {McNab, 2015 #173}.

Type I IFN also plays a role in autoimmune diseases such as SLE through the over-production of ISGs in active disease {Bennett, 2003 #783}, Sjögren’s syndrome and rheumatoid arthritis {Ronnblom, 2011 #789}. Conversely, type I IFN has been used as a therapy in the treatment of multiple sclerosis {Ann Marrie, 2006 #186}. Patients treated with IFN-β, demonstrate an increase in Il10 mRNA levels {Rudick, 1996 #187} and IL-10 has been linked to the suppression of EAE, the MS mouse model {Bettelli, 1998 #188}{Samoilova, 1998 #189}. IFN- therapy is used to treat some chronic viral infections and cancers {Friedman, 2008 #788}.

1.1.3       The regulation of type I IFN production in macrophages

Innate immune cell types such as macrophages produce type I IFN in response to the stimulation of PRRs by microbial products {Trinchieri, 2010 #181}. In macrophages, TLR3 and 4 induce Ifnb1 mRNA expression via the TRIF-dependent pathway {Yamamoto, 2003 #35}. However, in other cell types, mostly pDCs, the ligation of TLR7, 8 and 9 can activate pathways leading to type I IFN production. However, the induction of a positive feedback loop is necessary to induce the expression of Ifna genes. Therefore, IFN-β signals through the type I IFN receptor to activate IRF7 which subsequently promotes the expression of Ifna genes, with exception of IFN-α4 {Marie, 1998 #182}. Other PRRs such as the RLR RIG-I and melanoma differentiation-associated gene 5 (MDA5), are the main cytosolic receptors and induce robust type I IFN production in response to viral nucleic acids through the activation of IRF3, IRF7, NF-κB and MAPKs {Pichlmair, 2007 #183}{Goubau, 2013 #771}. The cytosolic molecular sensors NLRs NOD1 and NOD2 have also been shown to induce type I IFN in response to Helicobacter pylori {Watanabe, 2010 #184} and Mycobacterium tuberculosis infection {Pandey, 2009 #185}. Furthermore, other DNA motifs in the cytosol can be recognised by the DNA-dependent activator of IFN-regulatory factors (DAI), the DEAD box and DEAH box (DEXD/H box) helicases, and the receptor cytosolic GAMP synthase (cGAS), and lead to the induction of type I IFN production {Goubau, 2013 #771}{Paludan, 2013 #772}. All these signalling pathways converge on a few signalling molecules, such as the IFN-regulatory factor (IRF) family of transcription factors, leading to the transcription of genes encoding type I IFNs. The Ifnb1 gene contains consensus binding sites for IRFs, NF-κB transcription factors as well as AP-1 sites {Honda, 2005 #565}. In line, co-operation between these factors in the induction of Ifnb1 mRNA expression was reported {Thanos, 1995 #784}. ATF3, a member of the AP-1 family was found to regulated IFN- via direct binding distal to the Ifnb1 promoter and act as a transcriptional repressor in macrophages. Yet, ATF3 itself is induced by type I IFN {Labzin, 2015 #259}. However, IRF3 and IRF7, in most cases the fundamental IRFs, are activated by upstream kinases such as IKKε and TBK1. RIG-I and MDA5 use the adaptor mitochondrial antiviral signalling protein (MAVS) to activate TBK1, whereas TLR3 and TLR4 use the adaptor molecule TRIF. TLR7 and TLR9 signals rather via MyD88 than TRIF {Tamura, 2008 #774}{Moynagh, 2005 #773}.

However, a negative regulatory mechanism also exists, which can attenuate the production of type I IFN. In TLR4 and TLR9 stimulated macrophages, the IFN-β production is negatively regulated by the TPL-2/ERK pathway {Yang, 2011 #17}{Kaiser, 2009 #4}, which might be depend on the transcription factor c-Fos {Kaiser, 2009 #4}.

1.2      C57BL/6 and BALB/c as strains of different susceptibilities

Mouse models are often the organism of choice for immunologists. Two of the most commonly used inbred mice strains are C57BL/6 and BALB/c mice. As they differ significantly in their immune responses and they have been used as models to study susceptibility or resistance to various pathogens (Table 2.1). C57BL/6 mice are considered more resistant than BALB/c mice to Burkholderia pseudomallei {Leakey, 1998 #755} and Leishmania major infection {Sacks, 2002 #563} as well as to infection with Listeria spp. {Mainou-Fowler, 1988 #356} and M. avium {Roque, 2007 #753} (Table 2.1). BALB/c mice have been shown to be more resistant than C57BL/6 mice to T. gondii {Schluter, 1999 #564} and Helicobacter spp. infections {Mohammadi, 1996 #754} (Table 2.1). Differences in resistance and susceptibility are also evident in Il10-/- mice in the C57BL/6 and BALB/c background. The deficiency in IL-10 leads to spontaneous onset of enterocolitis dependent on the presence of the gut flora {Sellon, 1998 #756}. However, C57BL/6 mice were less susceptible to the onset of disease than BALB/c mice {Berg, 1996 #757}.

Thus, C57BL/6 and BALB/c mice show different immunological responses in various contexts and these differences have been studied in order to better understand the underlying mechanisms. C57BL/6 mice have been correlated to a dominant TH1 response, while BALB/c mice are associated with a dominant TH2 response {Heinzel, 1989 #758}. Other mechanism leading to their distinct phenotypes have been raised such as differential production of nitric oxide due to increased IFN--depentent iNOS expression in C57BL/6 cells in comparison with BALB/c cells {Oliveira, 2014 #124} or differences in abundance of respiratory chain complexes and lysosomal proteins as well as differential regulation of components belonging to various antioxidant stress systems in C57BL/6 and BALB/c macrophages {Depke, 2014 #351}. In B cells the reduced surface expression of TLR4 in BALB/c cells accounts for the low response of these B cells to LPS {Tsukamoto, 2013 #363}. Although some immunological factors have been found to explain the differences in disease outcome between C57BL/6 and BALB/c macrophages, the mechanisms underlying these traits are complex and only partially understood.

1.3      Overview/Summary

Preliminary data from the O’Garra laboratory showed that macrophages from C57BL/6 mice produce high levels of IL-10 and low level of pro-inflammatory cytokines such as IL-12, IL-1 and TNF- upon stimulation with different TLR ligands, whereas macrophages from BALB/c mice showed a reciprocal cytokine profile (data now published {Howes, 2016 #553}). In addition, it was suggested that type I IFN production may be higher in C57BL/6 macrophages than from BALB/c macrophages and regulate the differential expression of IL-10 and pro-inflammatory cytokines in these cells. Hence, we aimed to expand these findings using microarray analysis of stimulated C57BL/6 and BALB/c macrophages and investigate the mechanism underlying the differential type I IFN production in C57BL/6 and BALB/c macrophages further. Additionally, we intended to dissect the role of type I IFN in the regulation of IL-10 as one important anti-inflammatory cytokine in LPS-stimulated C57BL/6 macrophages investigating the activation or inhibition of kinases as well as transcription factor recruitment. Thus, this study described here in this thesis will shed light on the type I IFN-dependent regulation of IL-10 in LPS stimulated C57BL/6 macrophages and contribute to the understanding of how IL-10 expression is controlled which is critical in the design of immune intervention strategies.

1.4      Mice strains

C57BL/6 wild-type (WT), BALB/c WT, C57BL/6 Ifnar1-/-, C57BL/6 Tlr4-/-, C57BL/6 Nfil3-/- mice were bred and maintained at The Francis Crick Institute, Mill Hill Laboratory and the Francis Crick Institute under specific pathogen-free conditions in accordance with the Home Office, U.K., Animal Scientific Procedures Act, 1986. C57BL/6 Ifnar1-/- breeders were provided by B&K Universal (Hull, U.K.). C57BL/6 Tlr4-/- breeding pairs were kindly given to us by Prof. S. Akira (Osaka University, Osaka, Japan). C57BL/6 Nfil3-/- mice were provided by Andreas Wack. C57BL/6N and BALB/cAnN Myd88-/- mice were purchased from Oriental BioService Inc. and appropriate WT controls were ordered from Taconic (Taconic Farms Inc.). All mice used were female, between 8-16 weeks of age.

1.5      Reagents

1.5.1       Cell culture medium

For all in vitro experiments the following culture medium was used: RPMI 1640; 5% heat-inactivated FCS (Gibco, Thermo Scientific); 0.05 mM 2-Mercaptoethanol (Sigma); 10 mM HEPES buffer; 100 U/ml penicillin; 100 μg/ml streptomycin; 2 mM L-glutamine; 1 mM sodium pyruvate, in the following referred to as cRPMI. Unless otherwise stated, all components were purchased from BioWhittaker.

1.5.2       PRR ligands, heat-killed bacteria and recombinant cytokines

Cells were stimulated, unless otherwise stated, with either cRPMI culture medium (control), Salmonella Minnesota LPS (Alexis) at a final concentration of 10 ng/ml, Pam3CSK4 (InvivoGen) at a final concentration of 200 ng/ml, heat-killed B. pseudomallei 576 (from DSTL, Porton Down) at a final concentration of 500 HkBps : 1 BMDM, CpG1668 (TriLink Biotech) at a final concentration of 0.5 M, Poly I:C (InvivoGen) at a final concentration of 50 g/ml. When indicated, cells were additionally treated with recombinant murine IFN-β or IFN- (PBL) at a final concentration of 2 or 20 ng/ml. Doses have been established in the laboratory beforehand. Duration of stimulations are indicated in figure legends.

1.5.3       Inhibitors

Kinase inhibitor stocks were diluted in DMSO (DMSO vehicle control was included in every experiment) and used at the following final concentrations: 0.1 M BIRB0796 (p38 inhibitor), 0.5 M SB203580 (p38 inhibitor), 0.1 M PD0325901 (MEK1 inhibitor), 1 M U0126 (MEK1/2 inhibitor), 1 M Arry142886 (MEK1/2 inhibitor), 0.125 M rapamycin (mTOR inhibitor, InVitrogen), 0.03 M PI-103 (PI3K inhibitor) and 1 M CT99021 (GSK-3 inhibitor). BIRB0796, SB203580, PD0325901, PI-103, U0126, Arry142886 and CT99021 were kind gifts of Philip Cohen (University of Dundee) if not otherwise stated. Inhibitors were added to BMDM cultures 1h prior or 2h after stimulation with PRR ligands as indicated in the figure legends. The doses were based on recommendations by Cohen et al. for maximal efficiency with minimal off-target effects {Cohen, 2009 #343;Bain, 2007 #114} as well as by the Kinase Profiling Inhibitor Database ( Data from this database have been generated by the International Centre for Kinase profiling within the MRC Protein Phosphorylation Unit at the University of Dundee. Concentrations were confirmed by own titrations experiments using BMDMs (Error! Reference source not found.).

1.6      Differentiation of bone marrow-derived macrophages

Bone marrow was flushed from tibias and femurs and cultured (37C, 5% CO2) in 9 cm Petri dishes (Sterilin Ltd.) at 0.5×106 cells/ml in 8 ml cRPMI supplemented with 10% FCS and 20% L929-cell conditioned medium, containing M-CSF. L929-cell conditioned medium was generated from the L929 cell line with the assistance of Jackie Wilson (Large Scale Laboratory Facility, National Institute for medical research). On day 4, 10 ml cRPMI supplemented with 10% FCS and 20% L929-cell conditioned medium was added to the culture. On day 6, non-adherent cells were removed and adherent cells were harvested by gentle flushing with cold PBS (Gibco, Invitrogen). Cells were seeded in either 48-well tissue culture plates (Corning Inc.) at 0.5×106 cells/well or 6-well tissue culture plates (Corning Inc.) at 4×106 cells/well in cRPMI and rested for 18-24h prior to stimulation.

1.7      Enzyme-linked immunosorbent assay

1.7.1       Quantification of cytokine concentrations

Supernatants were collected from cell cultures and cytokine concentrations quantified by sandwich enzyme-linked immunosorbent assay (ELISA). Maxisorp 96-well plates (Nunc, Thermo Scientific) were used for the assay. IFN-β and IFN- were quantified using a commercially available ELISA kit (PBL) following the manufacturer’s instructions. Matched-pair sandwich ELISAs were used to measure IL-10 and IL-12p40 concentrations. Details are summarised in Table 2.1. ELISA plates were read on a Safiremicroplate reader (TECAN). Standard curve calculation and cytokine concentration were determined using Magellan software.

1.7.2       Quantification of the activity of interferon regulatory factor 3

Nuclear extracts of 2h heat-killed B. pseudomallei–stimulated BMDMs (500:1 B. pseudomallei:BMDM) were prepared using the Nuclear Extract Kit and analysed with the TransAM IFN regulatory factor (IRF) 3 kit for mouse (both from Active Motif) according to the manufacturer’s instructions. Plates were read on a Safiremicroplate reader (TECAN).

Specificity of the assay was tested using wild type consensus oligonucleotides or mutated oligonucleotides as provided by the kit (Error! Reference source not found.).

1.8      Real-time polymerase chain reaction

Supernatants from the cell culture were discarded at the indicated time points and cells washed with pre-warmed PBS. Cells were then lysed with RLT buffer (Qiagen) containing 1% -Mercaptoethanol (Sigma). Lysates were stored at -80C. RNA was harvested and isolated according to the manufacturer’s instructions using RNeasy Mini Kit (QIAGEN) with an on-column DNase digestion step to remove contaminating DNA (RNase-Free DNase kit, Qiagen). Purified RNA concentration was determined with a Nanodrop spectrophotometer (NanoDrop1000, Thermo Scientific).

cDNA was synthesised using High Capacity cDNA Reverse Transcription kit (Applied Biosystems), according to the manufacturer’s instructions followed by RNase H (Promega) treatment for 30 min at 37C to degrade leftover RNA.

Il10, Il12a, Ifnb1, Oas1g, Stat1, Stat3, Irf7 and Irf9 gene expression were quantified by real-time PCR (7900HT, Applied Biosystems and QuantStudio3, ThermoFisher) using the TaqMan assay system and normalised to Hprt1 mRNA. A no-cDNA template control and water only control was always included to ensure that reagents were not contaminated. The primer-probes used are summarised in Table 2.2.

For the quantification of un-spliced Il10 mRNA transcripts, TaqMan primer/probe pairs were designed the way that the forward (sense) primer and TaqMan probe annealed within an exon sequence (in this case exon 3 of the 5 Il10 exons), and the reverse (antisense) primer annealed within an intron sequence (in this case intron 4). Primers were designed by Ashleigh Howes using Primer Express 2.0 software and were custom made by Applied Biosystems. Primer sequences are summarised in Table 2.3.

The gene expression value, expressed in relative units (RU), for each sample was determined and normalised to the house-keeping gene Hprt1 using the delta Ct (ΔCt) calculation: ΔCtgene = 1.8(CtHprt1-Ctgene) x 100,000.

1.9      Determination of mRNA stability

BMDMs were stimulated with LPS or LPS and IFN-β. 1h post stimulation ActinomycinD from Streptomyces sp. (Sigma-Aldrich, 10 μg/ml) was added to the cultures (t = 0) to inhibit DNA synthesis. 30, 60, and 90 min later mRNA was isolated, reverse transcribed into cDNA, and Il10 mRNA quantified as described in 2.5.

1.10 Immunoblotting

In experiments where phosphorylation of proteins was evaluated, BMDMs were rested in 1% FCS for 20 h prior to stimulation in order to reduce the background signal caused by serum response. In all other cases BMDMs were rested as described before (2.3). Cells were stimulated as indicated, after which supernatant was removed and cells washed with PBS and lysed in RIPA buffer composed of 50 mM Tris HCl (Sigma), pH 8; 150 mM NaCl (Sigma); 2 mM EDTA (Sigma); 2 mM sodium pyrophosphate (Sigma); 50 mM sodium fluoride (Sigma); 0.1% SDS (BioRad); 1% NP-40 (Fluka); 0.5% deoxycholate acid (Sigma); 100 mM vanadate (Sigma); complete EDTA-free protease inhibitor cocktail (Roche) or in Triton X-100 lysis buffer (for IRF3 Western blots) composed of 50 mM Tris HCl (Sigma), 1mM EGTA (Sigma), 1mM EDTA (Sigma), 1 mM vanadate (Sigma); 50 mM sodium fluoride (Sigma); 5 mM sodium pyrophosphate (Sigma); 10 mM Sodium glycerophosphate (Sigma), 270 mM Sucrose (Sigma); 1% Triton X-100 (Sigma); 0.1% -Mercaptoethanol (Sigma) and complete EDTA-free protease inhibitor cocktail (Roche). For the separation of cellular and nuclear extracts the Nuclear Extract Kit from Active Motif was used.

Protein concentration was measured using a reducing agent compatible Bicinchoninic Acid (BCA) Protein Assay Kit according to the manufacturer’s instructions (Thermo Scientific).

Proteins were denatured in SDS sample buffer (5 min, 95 C), and resolved on a 10 or 12.5% SDS-polyacrylamide gel and subsequently transferred to polyvinylidene difluoride membranes (Millipore). Membranes were probed with antibodies against phospho-ERK1/2 (T185/Y187), total ERK1/2 (both Invitrogen), phospho-p38 (T180/Y182), total p38, phospho-IRF3 (S396, 4D4G) (Cell Signaling Technology), total IRF3 (Santa Cruz Biotechnology), phospho-STAT1 (Y701), total STAT1, phospho-TBK1 (S172), total TBK1 (both Cell signaling), phospho-GSK-3/ (S21/9) , total GSK-3/ (both from Cell signaling), JUNB (Cell signaling), ATF3 (Atlas antibodies), cJUN (Santa Cruz Biotechnology), BATF (Cell signaling) and GAPDH (FL-335), heat shock protein 90a/b (H-114) (Santa Cruz Biotechnology), actin (Calbiochem) or Tubulin (Abcam) followed by HRP-conjugated rabbit anti-goat IgG, goat anti-rabbit IgG (SouthernBiotech), or goat anti-mouse IgM (Calbiochem). For visualisation, the membrane was incubated with Pierce ECL Western blotting substrate (Thermo Scientific) and exposed to X-ray film (SLS).

1.11 Flow cytometry analysis

BMDMs were stimulated with LPS for 0, 10, 30 and 90 min, washed twice with PBS, and blocked with anti-CD16/CD32 Ab. Subsequently, cells were stained with PE-labelled anti-mouse TLR4 (SA15-21; BioLegend), APC-labelled anti-mouse CD14 (SA14-2, BioLegend) or isotype control (Rat IgG2a, κ) for 30 min at 4 ̊C and acquired using a BD LSR II (BD Biosciences). Data were analysed by FlowJo software.

1.12 Processing of microarray samples and analysis of microarray data set

1.12.1   Microarray processing

BMDMs were stimulated with LPS, heat-killed B. pseudomallei 576 or media (control) at 1×106 cells/ml in a 48 well plate (500 ul cells/well) for 0.5, 1, 3, 4 and 6h. RNA was isolated and purified as described in 2.5. These steps were carried out by Ashleigh Howes. I then confirmed RNA quality (RNA integrity number, range 9–10) using an Agilent 2100 Bioanalyzer (Agilent Technologies) and prepared RNA for microarray using the Illumina®TotalPrep-96 RNA Amplification kit. RNA concentration was determined using a Nanodrop1000 (Thermo Scientific) and quality checked again using an Agilent 2100 Bioanalyzer (Agilent Technologies). 1500 ng cRNA was loaded onto 6 sample Illumina BeadChip Arrays (MouseWG-6 v2). BeadChips were then incubated for 16-20h in hybridisation chambers to allow sample hybridisation. The following day, BeadChips were washed, blocked, treated with streptavidin Cy-3 to ‘stain’, washed, allowed to dry, and stored away from the light until scanned. BeadChips were scanned by an Illumina iSCAN array scanner. Intensity values were generated and background signal subtracted using BeadStudio software (Illumina). Microarray processing was done with the assistance of Dr. Christine Graham (Anne O’Garra laboratory) and Dr. Harsha Jani (Division of Systems Biology, NIMR, London).

1.12.2   Microarray data analysis

Further analysis of microarray data was done using GeneSpring GX version 12.6.1 or 14.5 (Agilent Technologies). All data shown was subjected to the following data processing and normalisation steps: lower threshold of signal intensity of 10. The expression values were log transformed (base 2) and scaled to the 75th percentile of all samples. Following this, the expression value of each gene probe was normalised to the median expression of that gene probe across all samples. After normalisation, all gene probes were quality filtered for those present (p < 0.01) in at least one sample, with 19,191 gene probes having remained. Prior to further analysis, a quality control evaluation was carried out to ensure the integrity of the data. All samples from the experiment were hierarchically clustered according to condition using GeneSpring software. Replicates within experimental groups clustered together demonstrating their similarity. A comparable observation was made using principal component analysis.

If further statistical analysis has been done, it is described in the relevant figure legends. Canonical pathway analysis was conducted using Ingenuity Pathway Analysis (IPA) software (Ingenuity Systems, http://www. Expression data conform to the minimum information about a microarray experiment standards for microarray analysis. Microarray data have been deposited in the Gene Expression Omnibus database ( under accession number GSE79809. Parts of the dataset used in Error! Reference source not found. has not been published and is therefore not deposited.

1.13  Assay for transposase accessible chromatin (ATAC) – library preparation and computational analysis of sequencing data

1.13.1   ATAC library preparation

BMDMs were plated in a 48-well plate at 500 000 cells/well, rested for 20 h and then stimulated as indicated in figure legend. ATAC-library preparation was performed as described in Buenrostro et al. {Buenrostro, 2013 #557}{Buenrostro, 2015 #348} with slight adaptations. At indicated times cells were washed twice with PBS and lysis buffer containing 10 mM Tris-HCl (pH 7.4), 10 mM NaCl, 3 mM MgCl2 and 0.1% IGEPAL CA-630 (Fluka) was added to the plate. 50 000 cells were pelleted and nuclei re-suspended in transposition reaction mix (25 µL 2x TD Buffer, 2.5 µL Tn5 Transposes (Illumina), 22.5 µL nuclease-free H2O) for 2h at 37°C. This transposition time was chosen after a time course analysis from 30 min to 2h and evaluation of the fragment size (Error! Reference source not found.). DNA was purified using the Qiagen PCR purification MinElute Kit and eluted in 10 µL elution buffer included in the kit. Transposed DNA fragments were then amplified via PCR (Table 2.4) using the following cycles: (1) 72°C, 5 min; (2) 98°C, 30 sec; (3) 98°C, 10 sec; (4) 63°C, 30 sec; (5) 72°C, 1 min; (6) repeat steps 3-5, 11x; (7) hold at 4°C. Primers used are summarised in Table 2.5.

After amplification of transposed DNA fragments, the library was purified using the Qiagen PCR purification MiniElut kit followed by an additional clean up step with AMPure beads (Beckman Coulter). Beads were incubated with the library preparation for 5 min at room temperature, followed by two washing steps with 80% ethanol. DNA was eluted with 0.1x Tris-EDTA (Sigma) and stored at -20C. The size of the library fragments was evaluated using the Bioanalyzer (Agilent Technologies). Libraries were then quantified using the Qubit and checked for adapters by ECO real qPCR before being run on the HiSeq2500 (PE50, 1 sample per lane, ~ 400 millionreads per sample). Quality control and sequencing of the library were performed by the High Throughput Sequencing facility at the Francis Crick Institute.

1.13.2   Analysis of ATAC-sequencing data

All data shown were subjected to the following data processing and normalisation steps. Quality and integrity of the sequenced library was determined using FastQC v0.11.5, a quality control tool for high throughput sequence data developed by the Bioinformatics Group at the Babraham Institute, UK. Subsequent, reads were quality trimmed (Trimmomatic 0.36) using the following parameters: ILLUMINACLIP:NexteraPE-PE.fa:2:30:10, SLIDINGWINDOW:3:20 and MINLEN:35, and quality was checked again using FastQC. Reads were mapped to the GRCm38.83(mm10) genome from UCSC using Bowtie2 (2.2.9-foss-2016b) with –sensitive-local parameters (-D 15 -R 2 -N 0 -L 20 -i S,1,0.75) as well as -X 2000 to increase the default of maximum fragment length for valid paired-end alignments from 500 to 2000. Duplications were removed using SAMtools (1.3.1-intel-2016b; rmdup) and artefact regions that tend to show artificially high signal were filtered out using the DAC blacklist for mm10, a comprehensive empirical blacklist identified by the ENCODE and modENCODE consortia {Consortium, 2012 #558}. Before calling peaks with MACS2 ( (; {Feng, 2011 #825}), the read start sites were adjusted to represent the centre of the transposon binding event as Tn5 transposase was shown to bind as a dimer and inserts two adapters separated by 9 bp {Adey, 2010 #345}. Hence, reads aligning to the + strand were off-set by +4 bp, and reads aligning to the – strand were off-set −5 bp. To assess the similarity of biological repeats, DeepTools{Ramirez, 2014 #559} and the R Bioconductor package DiffBind {Ross-Innes, 2012 #603}{Stark, 2011 #828} were used. DiffBind was also used to identify significantly different peaks between groups. For motif finding, significant peaks determined with DiffBind, were input for HOMER package motif finder algorithm{Heinz, 2010 #284}.

1.14 Chromosome immune-precipitation – polymerase chain reaction (ChIP-PCR)

BMDMs were stimulated with LPS and at indicated time points supernatant was removed and cells washed with PBS. Cells were crosslinked in 1% paraformaldehyde for 8 min at room temperature. Glycine was added to a final concentration of 0.125 M and incubated at room temperature for 8 minutes. After three PBS wash steps cells were scraped off and the pellet snap frozen at -80C. This aid the lysis and therefore improves sonication. The defrosted pellet was lysed in 100 l/106 cells of lysis buffer containing 1% SDS (Affymetrix Inc.), 10 mM EDTA (Sigma), 50 mM Tris-HCl pH 8.0 (Sigma) and 1 mM protease inhibitor cocktail (Roche). The lysate was incubated for 10 min on ice in a tube rotator before being sonicated for 40 min (30 sec on, 30 sec off on high setting; Diagenode Bioruptor). Sonication time was established by a previous sonication time course (Error! Reference source not found.). After sonication 30 – 100 l of the samples were frozen at -20C as input. 5×106 cells were used per immunoprecipitation and diluted 1:10 in dilution buffer (0.01% SDS (Sigma), 1.1% Triton-X 100 (Sigma), 1.2 mM EDTA (Sigma), 16.7 mM Tris-HCl pH 8.0 (Sigma),167 mM NaCl (Sigma), 1 mM protease inhibitor cocktail (Roche)). 5 g of antibody (anti-Atf3 Ab, polyclonal, Atlas Antibodies; Rabbit IgG, polyclonal, Isotype Control, Abcam) or a dilution of 1:150 for anti-Junb Ab (monoclonal (C37F9), Cell signalling) was incubated at 4C in a tube rotator overnight. 50 l of beads, pre-treated with 2 mg/ml BSA for at least 2 h, were added per sample and incubated on a rotator at 4C for 4 h. Beads were washed once with low salt buffer (0.1% SDS (Affymetrix Inc.), 1% Triton-X 100 (Sigma), 2 mM EDTA (Sigma), 150 mM NaCl (Sigma), 20 mM Tris-HCl pH 8.0 (Sigma)), once with high salt buffer (0.1% SDS (Affymetrix Inc.), 1% Triton-X 100 (Sigma), 2 mM EDTA (Sigma), 500 mM NaCl (Sigma), 20 mM Tris-HCl pH 8.0 (Sigma)), once with Lithium Chloride Immune wash (0.25 M LiCl (Sigma), 1% IGEPAL CA-630 (Fluka), 1% Sodium deoxycholate (Sigma), 1 mM EDTA (Sigma), 10 mM Tris-HCl pH 8.0 (Sigma)) and twice with Tris-EDTA (Sigma). Samples were eluted into 200 l elution buffer (1% SDS (Affymetrix Inc.), 100 mM NaHCO3 (Sigma)) and incubated at room temperature for 15 min, followed by 15 min at 65 C. The supernatant was collected, 8 l of a 5 M NaCl solution added and incubated overnight at 65 C to reverse the crosslinking. Inputs were defrosted, volume made up to 200 l with Tris-EDTA and equally NaCl is added and sample incubated at 65 C overnight. To digest leftover RNA and proteins within the sample, Rnase A to a final concentration of 100 g/ml (Invitrogen) was added and incubated for 1h at 37 C followed by 1h incubation at 55 C with Proteinase K (Thermo Scientific) (8 l Tris-HCl (pH 6.5 1M, Sigma), 4 l EDTA (0.5M, Sigma), 0.8 l Proteinase K (20mg/ml)). The Zymo Chip Clean and concentrate kit (Zymo Research) was used according to manufacturer’s instructions to concentrate and clean the sample. qRT-PCR was performed using SYBRE Green and primers as listed in Table 2.6.

ChIP-qPCR results were normalised using the Percent Input Method as this includes normalisation for both background levels and input chromatin going into the ChIP. First the dilution factor x of the input was calculated and log transformed (logx,2). This factor was subtracted from the input Ct value and forms the adjusted input Ct value, which gets averaged between the 3 technical replicates (average adjusted input Ct). The percentage of the input for a particular antibody or mock (IgG) was calculated as follows: 100 x 2(average adjusted input Ct – Ct IP).

1.15 Statistics

All data analysis was performed using GraphPad Prism software 7 (GraphPad Software, San Diego, CA). Statistical test and significance values for each experiment are indicated in the figure legends.

1.16 Background

During an infection PRR detect microbial products and prompt cytokine production, which then shapes the immunological response {Takeda, 2015 #233;Kawai, 2010 #110}. IL-12, TNF-, and IL-1 are pro-inflammatory cytokines, which are crucial for resistance against infection. However, when produced at high levels they may contribute to immunopathology {Nathan, 2010 #613}. In contrast, IL-10 is an immunosuppressive cytokine, which dampens pro-inflammatory responses. Yet, it can also lead to defective pathogen clearance {Murray, 2012 #612}{Moore, 2001 #31}. Therefore, the regulation of these cytokines is fundamental in order to allow the generation of an effective but balanced immune response.

C57BL/6 and BALB/c mice show significant differences in their immune responses, which leads to distinct outcomes of infection. Therefore, they are often used for studying susceptibility or resistance to various pathogens {Sacks, 2002 #563}{Mainou-Fowler, 1988 #356}{Schluter, 1999 #564}. We recently showed that C57BL/6 macrophages stimulated with LPS (TLR4), Pam3CSK4 (TLR2) and B. pseudomallei (TLR2 and 4) produce low levels of IL-12, TNF-, and IL-1, but high levels of IL-10 {Howes, 2016 #553}. In contrast, BALB/c macrophages show a reciprocal pattern in the production of these cytokines {Howes, 2016 #553}. PRR such as TLR2 and 4 are localised on the cell surface recognising bacterial motifs such as peptidoglycan (TLR2) or LPS (TLR4) {Takeda, 2015 #233;Kumar, 2009 #34;Akira, 2004 #28}. Upon TLR ligation adaptor molecules containing a TIR domain such as the myeloid differentiation primary-response protein 88 (MyD88) and Toll/IL-1 receptor domain- containing adaptor (TRIF) are recruited {Kumar, 2009 #34;Akira, 2006 #33;Akira, 2004 #28}. The TRIF-dependent pathway is used by TLR4 but not TLR2 signalling and is critical for IFN-β production downstream of this receptor {Akira, 2004 #28}{Yamamoto, 2003 #35}. IL-10 production in macrophages, amongst pro-inflammatory cytokines, is induced by TLRs via the MyD88-dependent and TRIF- dependent pathways {Boonstra, 2006 #27}. Stimulation of TLR4 leads to an activation of both signalling pathways and, importantly, maximal IL-10 production through TLR4 signalling requires the activation of both, the MyD88 and TRIF-dependent pathways {Boonstra, 2006 #27}. Although TLR4 is essential for the signalling of LPS, the response to LPS requires several additional molecules, one of them being CD14. CD14 is a glycosylphosphatidylinositol–anchored molecule preferentially expressed on monocytes/macrophages and neutrophils and is critically involved in the recognition of LPS together with TLR4. CD14 chaperones LPS molecules to the TLR4-MD-2 signalling complex and hence controls the trafficking and signalling functions of TLR4 {da Silva Correia, 2001 #562}{Zanoni, 2011 #276}. Nevertheless, only at low LPS concentrations the signalling via MyD88 is CD14- dependent, whereas signalling through the TRIF pathway requires CD14 at low and high LPS concentrations {Rajaiah, 2015 #347}.

1.17 Aims

Data from our laboratory showed that C57BL/6 macrophages produce more IL-10 compared to BALB/c macrophages upon stimulation of TLR2 and 4 and that this is the opposite for pro-inflammatory cytokines. To untangle more of the differences in the responses between C57BL/6 and BALB/c macrophages, we aimed to answer the following question:

  • Which molecular pathways are associated with the differential expression of IL-10 and pro-inflammatory cytokines in C57BL/6 and BALB/c macrophages?
  • What is the mechanism behind the differential regulation of these pathways?

1.18 Results

1.18.1   Type I IFN signalling as one major difference between C57BL/6 and BALB/c macrophages

To investigate the mechanism underlying the difference in cytokine production by C57BL/6 and BALB/c macrophages upon B. pseudomallei infection, leading mainly to a stimulation of TLR2 and 4, we stimulated macrophages with heat-killed B. pseudomallei for 3 and 6h and undertook an unbiased microarray analysis. At 3h post B. pseudomallei stimulation, 790 genes were found to be differentially regulated in C57BL/6 and BALB/c macrophages (Figure 3.1 A). Of these genes, most of them were upregulated by B. pseudomallei stimulation in macrophages of both strains of mice. The expression of the majority of these up-regulated genes were more strongly up-regulated in C57BL/6 compared to BALB/c macrophages. Genes that were downregulated by B. pseudomallei stimulation in C57BL/6 macrophages were also downregulated in BALB/c macrophages. However, these down-regulated genes were either more highly or more weakly down-regulated in BALB/c macrophages in comparison to C57BL/6 macrophages. At 6h post B. pseudomallei stimulation, 2246 genes were differentially regulated in C57BL/6 and BALB/c macrophages (Figure 3.1 A), suggesting a reinforcement of differential gene expression over time. Additionally, the difference in expression of these 2246 genes between C57BL/6 and BALB/c macrophages is increased compared to the 3h time point (Figure 3.1 A).

To better understand the biological relationships between the differentially regulated genes in C57BL/6 and BALB/c macrophages upon B. pseudomallei stimulation, transcripts that were identified at 3 and 6h post stimulation were subjected to canonical pathway analysis (IPA). The top 10 identified pathways at each time point were found to mainly relate to type I IFN- mediated processes (Figure 3.1 B). These included at the 3h time point IFN signalling, JAK/STAT signalling, activation of IRF by cytosolic PRRs, and role of JAK1, JAK2, TYK2 in IFN signalling; and at the 6h time point the activation of IRF by cytosolic PRRs, role of PKR in IFN induction and antiviral response, and IFN signalling (Figure 3.1 B). Genes related to these pathways together with their fold change expression are listed in Error! Reference source not found..

The induction of most of the differentially regulated genes within these type I IFN–related pathways was higher in stimulated C57BL/6 compared to BALB/c macrophages (Error! Reference source not found.). To validate genes of these pathways which had low log FC values (Error! Reference source not found.) including Oas1g, Stat1, Stat3, Irf7, and Isgf3g (Irf9), macrophages were stimulated with B. pseudomallei for 3 and 6h and expression of these genes was evaluated by qRT-PCR (Figure 3.2). Stat3 and Irf9 expression was significantly increased in C57BL/6 macrophages 3h post B. pseudomallei stimulation compared to BALB/c macrophages (Figure 3.2 A). Oas1g, Stat1 and Irf7 expression was not significantly different between C57BL/6 and BALB/c macrophages 3h post B. pseudomallei stimulation (Figure 3.2 A). Nevertheless, there was a trend of an increased transcription in stimulated C57BL/6 macrophages. Six hours post B. pseudomallei stimulation this trend was enhanced and Oas1g, Stat1 and Irf7 expression was significantly increased in stimulated C57BL/6 compared to BALB/c macrophages as was the expression of Stat3 and Irf9 (Figure 3.2 B).

Our results show that type I IFN signalling pathways as well as type I IFN inducible genes are significantly increased in B. pseudomallei stimulated macrophages from C57BL/6 compared to BALB/c mice. In line with these data, we have shown that mRNA levels of type I IFN are also increased upon B. pseudomallei and LPS stimulation in C57BL/6 macrophages compared to BALB/c macrophages 1h after stimulation (Howes, Taubert et al., Figure3A) {Howes, 2016 #553}. In concordance, type I IFN protein levels are increased in C57BL/6 macrophages compared to BALB/c macrophages, with its peak 3h post stimulation (Howes, Taubert et al., Figure3B) {Howes, 2016 #553}.

These data suggest that type I IFN production and consequently type I IFN signalling are a major difference in the responses of C57BL/6 and BALB/c macrophage to B. pseudomallei and LPS and in part explains their differential expression profile of IL-10 and pro-inflammatory cytokines, as we demonstrated in Howes, Taubert et al. {Howes, 2016 #553}.

1.18.2   Expression and endocytosis of TLR4 and CD14 are similar in C57BL/6 and BALB/c macrophages

The mechanism of why type I IFN expression and signalling is different between these two mouse strains is not understood. Zanoni et al. {Zanoni, 2011 #276} and Rajaiah et al. {Rajaiah, 2015 #347} reported that CD14 is required for LPS-/Escherichia coli-induced TLR4 internalisation into endosomes and that the activation of the TRIF pathway required CD14 at different LPS concentrations. CD14 leads to internalisation of TLR4 complex into endosomes whereupon TRIF-mediated signalling in macrophages is activated, resulting in type I IFN production {Zanoni, 2011 #276}. To determine if CD14 expression on C57BL/6 and BALB/c macrophages is different and might therefore lead to an increased type I IFN expression and signalling in C57BL/6 macrophages, cells were stained with an anti-CD14 antibody and analysed by flow cytometry. The expression of CD14 was higher on C57BL/6 compared to BALB/c macrophages at steady-state (Figure 3.3 A) and 90min post LPS stimulation (Figure 3.3 B), though not significantly increased. CD14 expression over a time course of 90min followed similar kinetics in C57BL/6 and BALB/c macrophages, although CD14 expression was always lower in BALB/c macrophages (Figure 3.3 C).

As CD14 is important in TLR4 endocytosis leading to the activation of the TRIF pathway and the production of type I IFN, we aimed to investigate TLR4 expression and endocytosis on these macrophages upon LPS stimulation. Therefore, we stimulated macrophages with LPS and undertook cell surface staining to evaluate the expression of TLR4 presentation at steady-state and during LPS stimulation on the surface of C57BL/6 and BALB/c macrophages. At steady-state TLR4 surface expression was similar on C57BL/6 and BALB/c macrophages (Figure 3.3 D). Also, at 10, 30 and 90min post LPS stimulation TLR4 expression was similar between C57BL/6 and BALB/c macrophages (Figure 3.3 C, F). The expression of TLR4 decreased over the time course of stimulation on both, C57BL/6 and BALB/c macrophages, indicating the endocytosis of the receptor upon stimulation (Figure 3.3 F). Hence, the rate of TLR4 endocytosis was similar between C57BL/6 and BALB/c macrophages. Additionally, in Howes, Taubert et al. {Howes, 2016 #553} we could demonstrate that there is no difference in Tlr4 mRNA expression between C57BL/6 and BALB/c macrophages at steady-state.

In all experiments LPS was added at a final concentration of 10ng/ml as at this dose IL-10 production was substantially induced in both C57BL/6 and BALB/c macrophages. Additionally, upon stimulation with 10 ng/ml LPS differences in cytokine levels were distinct in C57BL/6 and BALB/c macrophages (titration was carried out by Ashleigh Howes). As Zanoni et al. {Zanoni, 2011 #276} as well as Rajaiah et al. {Rajaiah, 2015 #347} used higher doses of LPS in their experiments, we repeated our experiments with 1g/ml of LPS (Figure 3.4). An increased concentration of LPS reinforced that the expression of CD14 and its endocytosis upon stimulation are not significantly different between C57BL/6 and BALB/c macrophages (Figure 3.4 A). Furthermore, TLR4 expression and its endocytosis were also similar in C57BL/6 and BALB/c macrophages upon stimulation with 1g/ml LPS (Figure 3.4 B). Hence, previous results using LPS at a concentration of 10ng/ml got confirmed.

Altogether, these data suggest that there are no major differences in CD14 and TLR4 expression and endocytosis upon LPS stimulation in C57BL/6 and BALB/c macrophages that could explain the difference in type I IFN production by these cells.

1.18.3   TRIF signalling: TBK1 and IRF3 are more activated in stimulated C57BL/6 macrophages compared to BALB/c macrophages

LPS stimulates TLR4 resulting in the activation of two different downstream adapter molecules, MyD88 and TRIF {Takeda, 2015 #233}. TRIF activates IRF3, a master transcription controller of antiviral responses that leads to the production of type I IFNs {Takeda, 2015 #233}. Therefore, we investigated if any downstream targets of TRIF are differentially activated in C57BL/6 and BALB/c macrophages upon LPS stimulation.

The activation of TRIF leads to the recruitment and phosphorylation of TBK1 on Ser172 {Tu, 2013 #561;Larabi, 2013 #560} via TRAF3, which then phosphorylates the cognate upstream adaptor protein. IRF3 binds to the phosphorylated TRIF, and becomes phosphorylated by TBK1. Phosphorylated IRF3 (S396) subsequently dissociates from the adaptor protein and dimerizes before translocating into the nucleus to induce type I IFNs {Liu, 2015 #223}.

To investigate, if the activation of TBK1 is different in C57BL/6 and BALB/c macrophages, we stimulated the C57BL/6 and BALB/c macrophages with LPS and blotted for TBK1 (pS172). Phosphorylation of TBK1 (S172) was reduced in BALB/c macrophages upon LPS stimulation compared to C57BL/6 macrophages (Figure 3.5 A). However, it has to be pointed out that this experiment was done only once and therefore has to be repeated to be able to draw a strong conclusion. Nevertheless, in accordance with a decrease in TBK1 (S172) phosphorylation we showed a significant difference in IRF3 activity/(S396) phosphorylation using two different methods (ELISA – Figure 3.5 B; western blotting – Figure 3.5 C) in C57BL/6 and BALB/c macrophages upon B. pseudomallei and LPS stimulation.


In agreement with increased TBK1 and IRF3 activation and higher production of type I IFN in C57BL/6 macrophages, we additionally demonstrated an increased phosphorylation of STAT1 (Y701) in LPS-stimulated C57BL/6, BALB/c, and C57BL/6 Ifnar1-/- macrophages (Howes, Taubert et al., Figure3C, 3D) {Howes, 2016 #553}. The phosphorylation of STAT1 (Y701) 2h post stimulation was reduced in BALB/c macrophages compared to C57BL/6. Therefore, the amount of type I IFN produced by BALB/c macrophages is lower compared to C57BL/6 macrophages but is sufficient to activate STAT1. The activation of STAT1 is due to type I IFN signalling as the phosphorylation of STAT1 (Y701) in LPS-stimulated C57BL/6 Ifnar1-/- macrophages is absent (Howes, Taubert et al., Figure3C, 3D) {Howes, 2016 #553}.

1.18.4   MyD88 signalling does not account for differences in type I IFN production of C57BL/6 and BALB/c macrophages stimulated with LPS or B. pseudomallei

Our data demonstrate, that TBK1 and IRF3 are more activated in C57BL/6 compared to BALB/c macrophages as well as downstream type I IFN signalling even though TLR4/CD14 expression and endocytosis are similar. As shown in many co-stimulation studies, MyD88- and TRIF-activating TLR ligands lead to a synergistic cytokine production in innate immune cells {Ting Tan, 2013 #587}{Whitmore, 2004 #588}, indicating cross-talk between the MyD88 and TRIF pathways. Liu et al. showed that TRIF-induced IRF1 can be suppressed by the MyD88 pathway, controlling the magnitude and timing of cytokine expression {Liu, 2015 #586}.

We investigated if signalling through the cytosolic adapter protein MyD88 plays a role in the difference in IFN- production in C57BL/6 and BALB/c macrophages. C57BL/6 and BALB/c WT and MyD88-/- macrophages were stimulated with LPS and B. pseudomallei and the production of IFN- quantified by ELISA (Figure 3.6)The production of IFN- increased significantly in C57BL/6 MyD88-/- compared to WT macrophages with both stimuli (Figure 3.6). Although, the knock-out of MyD88 in BALB/c macrophages did not lead to an increase in IFN- production if stimulated with LPS, the production of IFN- increased significantly upon B. pseudomallei stimulation (Figure 3.6). Nevertheless, IFN- levels did not elevate to the levels produced by WT C57BL/6 macrophages.

Although MyD88 signalling inhibits IFN- production in C57BL/6 macrophages, our data suggest that the differential cytokine production in C57BL/6 versus BALB/c macrophages was not due to immediate signalling downstream of the TLR4-MyD88 axis neither to inhibition of the TRIF pathway by MyD88-dependent signalling.

1.19 Discussion

Balancing pro- and anti-inflammatory immune responses is necessary to guarantee effective but safe pathogen clearance. C57BL/6 and BALB/c mice show significant differences in their immune responses leading to distinct outcomes of infection {Sacks, 2002 #563}{Schluter, 1999 #564} such as in B. pseudomallei infection, where C57BL/6 mice show enhanced resistance compared to BALB/c mice {Tan, 2008 #326}{Titball, 2008 #589}. The higher production of pro-inflammatory cytokines in BALB/c mice has been associated with exacerbated pathology {Tan, 2008 #326}{Ulett, 2000 #590;Ulett, 2000 #591}. However, it is unclear whether this exacerbated pathology is due to the cytokine storm or decreased control of the bacteria. Nevertheless, in vivo infection models show a high degree of complexity. Therefore, it is difficult to completely dissect the mechanisms that underlie the differential production of cytokines in these two strains of mice. Therefore, in Howes, Taubert et al. {Howes, 2016 #553}, we showed in an in vitro setting that C57BL/6 macrophages produce low levels of IL-12, TNF- and IL-1, but high levels of IL-10 in response to TLR4 and TLR2 ligands while BALB/c macrophages show a reciprocal cytokine production {Howes, 2016 #553}.

To study the underlying mechanism of the differential cytokine production by stimulated C57BL/6 and BALB/c macrophages, we undertook a comparative microarray analysis of temporal gene expression in B. pseudomallei–stimulated C57BL/6 and BALB/c macrophages. This analysis revealed major differences of type I IFN–responsive genes and type I IFN pathway genes, including Oas1g, Stat1, Stat3, Irf7, and Irf9 in C57BL/6 compared with BALB/c macrophages. These differences are in accordance with a higher IFN- production in B. pseudomallei-stimulated C57BL/6 macrophages compared with BALB/c macrophages. Together with experiments performed by Ashleigh Howes {Howes, 2016 #553}, these findings demonstrate that fundamental differences in type I IFN induction and signalling in C57BL/6 and BALB/c macrophage responses may contribute to their differential cytokine production. These findings go in hand with findings by Oliveira et al. showing that defective TLR4 signalling for IFN-β expression is responsible for the decreased production of nitric oxide in BALB/c compared to C57BL/6 macrophages in response to LPS {Oliveira, 2014 #124}.

Type I IFNs play a critical role in innate and adaptive immunity during infection with viruses, bacteria, parasites and fungi {McNab, 2015 #173}. As well as for other cytokines, the tight regulation of type I IFN is necessary to keep the right balance between an IFN response that is temporally appropriate and of sufficient magnitude while limiting tissue damage {McNab, 2015 #173}{Porritt, 2015 #318}. In most cell types, type I IFN is induced through the activation of cytosolic receptors that recognize viral or other xenogeneic or autologous nucleic acid {McNab, 2015 #173}. However, in some cell types e.g. macrophages and DCs, type I IFN is produced in response to TLR3 and 4 activation. Type I IFN production via TLR4 activation is dependent on the TRIF pathway {Akira, 2004 #28}{Yamamoto, 2003 #35} and requires CD14 at different LPS concentrations, leading to internalisation of the TLR4 complex into endosomes whereupon TRIF is recruited {Rajaiah, 2015 #347}. However, the differential production of IFN- by C57BL/6 and BALB/c macrophages observed in our study was not due to differential TLR4 or CD14 expression or differences in their endocytosis rate.

Nevertheless, further downstream of the TLR4-TRIF axis TBK1 activation seems to be increased in LPS-stimulated C57BL/6 compared to BALB/c macrophages, which is strengthened by increased IRF3 activation/(S396) phosphorylation post LPS or B. pseudomallei stimulation in C57BL/6 compared to BALB/c macrophages. These data suggest that signalling events affecting the TBK1-IRF3 axis, which are critical for the induction of type I IFN downstream of TLR4 {Honda, 2005 #565}, may be responsible for the enhanced production of type I IFN in C57BL/6 macrophages compared to BALB/c macrophages.

To exclude the possibility of a cross-talk between the MyD88 and TRIF pathway and a possible inhibition of IRF3 activation in BALB/c macrophages as shown by Liu et al. {Liu, 2015 #586} for IRF1, we evaluated IFN- levels in C57BL/6 and BALB/c MyD88-/- macrophages. MyD88 signalling significantly inhibits IFN- production in LPS- and B. pseudomallei-stimulated C57BL/6 macrophages as well as in B. pseudomallei-stimulated BALB/c macrophages. However, the differential production of IFN- by stimulated C57BL/6 and BALB/c macrophages is independent of MyD88 signalling. Sugiyama et al. showed thatthe stimulation of TLR3 and TLR4 pathways in SMAD2/3-/- macrophages lead to enhanced IFN- production and STAT1 phosphorylation compared with WT macrophages. SMAD2 and SMAD3 were found to directly inhibit IRF3, resulting in reduced IFN- production {Sugiyama, 2012 #593}. It needs to be investigated if SMAD2/3 levels are significantly different in C57BL/6 and BALB/c macrophages upon LPS stimulation and whether it might explain the difference in IRF3 activation and the type I IFN induction in these mouse strains. Furthermore, Zanoni et al. reported in DCs, that TLR4 internalisation and IRF3 activation is mediated by PLC-γ2 and Syk {Zanoni, 2011 #276}. Further studies are required to investigate whether this pathway is existent in macrophages and differentially activated in C57BL/6 and BALB/c macrophages.

Recently, Wang et al. reported a new phosphorylation site of STAT2 on T387 that is constitutively phosphorylated in different untreated cancer cell lines and negatively regulates the signal transduction from the type I IFN receptor {Wang, 2017 #545}. We hypothesise, that BALB/c macrophages might show a higher phosphorylation of STAT2 on T387 resulting in reduced type I IFN signalling and a lower production of ISGs. The antibody described and used in the publication by Wang et al. was kindly given to us to test this hypothesis. However, preliminary results were not clear as antibody used was raised against human STAT2 and cross-reactivity might not be given. Hence, further studies are required.

In summary, type I IFN production and type I IFN signalling are a major difference in B. pseudomallei- and LPS-stimulated macrophages from C57BL/6 and BALB/c mice. This difference is due to an increased TBK1 (S172) and IRF3 (S396) phosphorylation in stimulated C57BL/6 macrophages compared to BALB/c macrophages.

With Our Resume Writing Help, You Will Land Your Dream Job
Resume Writing Service, Resume101
Trust your assignments to an essay writing service with the fastest delivery time and fully original content.
Essay Writing Service, EssayPro
Nowadays, the PaperHelp website is a place where you can easily find fast and effective solutions to virtually all academic needs
Universal Writing Solution, PaperHelp
Professional Custom
Professional Custom Essay Writing Services
In need of qualified essay help online or professional assistance with your research paper?
Browsing the web for a reliable custom writing service to give you a hand with college assignment?
Out of time and require quick and moreover effective support with your term paper or dissertation?