CD45 negatively regulates tumour necrosis factor and interleukin-6 production in dendritic cells

Abstract

CD45 is known to regulate signalling through many different surface receptors in diverse haemopoietic cell types. Here we report for the first time that CD45^−/− bone marrow dendritic cells (BMDC) are more activated than CD45^+/+ cells and that tumour necrosis factor (TNF) and interleukin-6 (IL-6) production by BMDC and splenic dendritic cells (sDC), is increased following stimulation via Toll-like receptor (TLR)3 and TLR9. Nuclear factor-κB activation, an important downstream consequence of TLR3 and TLR9 signalling, is also increased in CD45^−/− BMDC. BMDC of CD45^−/− mice also produce more TNF and IL-6 following stimulation with the cytokines TNF and interferon-α. These results show that TLR signalling is increased in CD45^−/− dendritic cells and imply that CD45 is a negative regulator of TLR and cytokine receptor signalling in dendritic cells.

Introduction

The leucocyte common (CD45) antigen is an abundant transmembrane tyrosine phosphatase expressed on all nucleated haemopoietic cells.^1,2 CD45 phosphatase activity is critical for T- and B-cell receptor signal transduction and Sarc (Src) family kinases have been identified as its primary targets. In addition to this well-established role in antigen receptor signal transduction, initial triggering of cytokine production through a variety of other receptors is modulated by CD45. For example, CD45 is required for chemokine and cytokine production by natural killer (NK) cells after stimulation via Fc or major histocompatibility complex (MHC)-binding receptors³ and is important for histamine degranulation after immunoglobulin E receptor crosslinking in mast cells.⁴ Furthermore CD45 can also modulate signalling through cytokine receptors and both positive and negative effects have been reported.^5,6 Janus kinases (Jaks) have been shown to be substrates for CD45 phosphatase and CD45^−/− mice show increased phosphorylation of Jaks and increased responses to cytokines in several haemopoietic cell types.⁵ These data show that alterations in CD45 expression modulate signalling through diverse receptors and thus may affect cytokine production and response in many different cell types.

Eleven distinct mammalian Toll-like receptors (TLRs) have been described, most of which have been shown to function as receptors for pathogen-associated molecular patterns (PAMPS).^7,8 Microbial ligands upon binding to TLRs on antigen presenting cells trigger nuclear factor (NF)κB, leading to up-regulation of cytokines and costimulatory molecules. This in turn activates the innate immune response against the pathogen from which the ligand is derived. Of the TLRs studied TLR3 has been identified as responding to dsRNA, which can be mimicked in vitro using the synthetic analogue dsRNA poly(I:C). Stimulation through TLR3 results in the induction of interferon (IFN)-β, interleukin (IL)-6, IL-12, and tumour necrosis factor (TNF) from macrophages.^9,10 TLR9 recognizes un-methylated CpG DNA motifs, which are found at a much greater frequency in prokaryotic DNA and therefore serve as a molecular marker for bacteria.¹¹ Viral genomes may also harbour CpG motifs, HSV2 is known to interact, via TLR9, with plasmacytoid dendritic cells (DCs) and induce the production of IFN-α.¹² These observations imply that TLR3 and TLR9 play an important role in host defence against virus infections.

No information is available about the role of CD45 in TLR signalling but in view of the data showing effects of CD45 on many other cell surface recptors, we hypothesized that it is very likely that CD45 influences signalling via TLR receptors. As DC and TLR play unique and crucial roles in innate immune mechanisms we set out to determine experimentally if CD45 is required for TLR3 and TLR9 recognition of PAMPs. We analysed directly the effect of CD45 in bone marrow (BM) derived or splenic (s) DC from CD45^−/− mice following TLR3 and TLR9 stimulation. We show that CD45 is a negative regulator for TNF and IL-6 production and also negatively regulates the response to TNF and IFN-α.

Materials and methods

Mice

C57Bl/6 (CD45^+/+) and CD45^−/− mice were obtained from the specific pathogen-free unit of the Institute for Animal Health, Compton, UK. Exon 9 targeted CD45^−/− mice were originally provided by Dr N. Holmes (University of Cambridge). All animal studies were approved by the site ethical review committee and were carried out in accordance with UK Home Office regulations.

Dendritic cell isolation and culture

Bone marrow DCs were prepared from the hind limb bones of CD45^−/− and CD45^+/+ mice. Briefly, bone marrow was flushed from the bones with RPMI-1640 supplemented with GLUTAMAX, 25 mm HEPES (Invitrogen Life Technologies, Paisley, UK), 10% heat-inactivated fetal calf serum (FCS), 10 U/ml penicillin, 10 µg/ml streptomycin, 100 units/ml of polymyxin B (Sigma-Aldrich, Poole, UK) and 5 × 10⁻⁵M β-mercaptoethanol (Invitrogen Life Technologies) (RPMI-FCS). Red blood cells were removed with Red Blood Cell Lysis Buffer (Sigma-Aldrich) according to the manufacturer's instructions. Bone marrow cells were then resuspended at 2 × 10⁵ cells/ml in RPMI-FCS supplemented with 200 ng/ml recombinant murine granulocyte–macrophage colony-stimulating factor (rmuGM-CSF; R & D Biosciences, Abingdon, UK). 2 × 10⁶ cells were cultured in non-tissue-culture treated Petri dishes (BD Biosciences) for 48 hr, after which time a further 10 ml of RPMI-FCS supplemented with 200 ng/ml rmuGM-CSF was added to each dish. BMDCs were cultured for a further 72 hr, 10 ml of medium was removed from each culture plate and replaced with 10 ml of RPMI-FCS supplemented with 100 ng/ml rmuGM-CSF. BMDCs were then cultured for a further 48 hr prior to stimulation.

Splenic DCs were obtained using a variation of the method described by Vremec et al.¹³ Briefly, spleens from eight mice were perfused with RPMI-FCS supplemented with 0·1 m ethylenediaminetetra-acetic acid, 1 mg/ml collagenase (Type III, Worthington Biochemical Corp., Lorne Laboratories, Reading, UK), DNase I (325 K units/ml, Sigma-Aldrich). Spleens were digested for 30 min at 37° and the homogenate passed through a 0·40 µm cell sieve. Low-density cells were isolated on a Nycodenz (1·077 g/ml, Life Technologies Ltd) gradient by centrifugation at 2000 g for 20 min. DCs were further enriched from the low-density fraction using anti-CD11c microbeads (Miltenyi Biotech Ltd, Bisley, UK) according to the manufacturer's instructions. sDCs were further purified based upon staining of cells for CD3, CD19, F4/80 and DX5 and sorting of negative populations on a MoFlo flow cytometer (Cytomation, Fort Collins, CO).

BMDCs and sDCs were stimulated with 1 µm CpG (1826 – tcc atg acg ttc ctg acg tt) or CpG control oligonucleotides (1826K – tcc atg agc ttc ctg agc tt) (Invivogen Technologies, San Diego, CA), 100 µg/ml poly(I:C) RNA (Invivogen), 50 ng/ml rmuTNF (R & D Systems) or 2000 U/ml IFN-α (a kind gift of Dr Agnes Le Bon).

Flow cytometric analysis

Unless otherwise stated all conjugated antibodies were obtained from BD Pharmingen. CD11c-fluoroscein isothiocyanate (FITC; HL3), CD40-Biotin (3/23), CD86-Biotin (GL1), CD49b-FITC (DX5), CD3-FITC (145-2C11), CD19-phycoerythrin (PE; ID3), F4/80-PE (Cl:A3-1) (Serotec) and Streptavidin-APC. Cells (2−5 × 10⁵) were incubated with antibody diluted in phosphate-buffered saline (PBS) supplemented with 2% fetal calf serum, 0·1% NaN₃. Analysis was performed on a FACScalibur using CellQuest software.

Cytokine analysis

Cytokines were quantified in cellular supernatants using mouse inflammation cytometric bead array kits (BD Biosciences) according to the manufacturer's instructions. IFN-α was quantified by enzyme-linked immunosorbent assay (ELISA; PBL, Alexis Corp, Nottingham, UK) according to the manufacturer's instructions.

Immunoblotting

Cells (1 × 10⁶ cells per ml) were collected, washed with ice-cold PBS and resuspended in RIPA lysis buffer (PBS pH 7·4, 1% NP-40, 0·5% sodium deoxycholate, 10 mm NaF and 0·5 mm phenylmethylsulphonyl fluoride), containing protease and phosphatase inhibitors (protease inhibitor cocktail and phosphatase inhibitor cocktail 2; Sigma-Aldrich). Cell lysates were incubated on ice for 30 min and after removing the insoluble materials by centrifugation (16 500 g for 10 min at 4°), the total protein concentration was quantified by BCA protein assay (Pierce Biotechnology, Rockford, IL). Whole cell lysate proteins were separated by 10% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (10 µg of total protein per lane), transferred onto a nitrocellulose membrane and immunoblotted using antibodies specific for IκBα (Santa Cruz Biotechnology, Inc., CA) and β-actin (Sigma-Aldrich) as a loading control. Blots were developed using ECL^TM donkey anti-rabbit horseradish peroxidase (HRP)-linked F(ab′)₂ fragment or sheep antimouse HRP-linked whole antibody and ECL^TM Western Blotting Detection Reagents (Amersham Biosciences, UK Ltd, Amersham, UK).

Statistical analysis

Data from CD45^−/− and CD45^+/+ mice were compared using the two tailed Student's t-test included in Excel software.

Results

TLR3 and TLR9 stimulation of BMDCs in CD45^−/− and CD45^+/+ mice

BMDC from CD45^+/+ and CD45^−/− mice were studied individually (a total of eight experiments were performed). There was no difference in the number of cells recovered after culture of bone marrow from CD45^+/+ and CD45^−/− mice. After excluding dead cells and debris by gating on forward and side scatter, the CD11c staining profiles of CD45^+/+ and CD45^−/− BMDC were very similar so that similar gates for analysis of other cell surface molecules were used (Fig. 1a). BMDCs were stimulated overnight with either the TLR3 ligand poly(I:C) or the TLR9 ligand CpG. Unstimulated CD45^−/− BMDC consistently showed higher CD40, CD86 and IA than CD45^+/+ cells (Fig. 1b and Table 1). Following TLR stimulation CD45^−/− cells showed a greater increase in expression of CD40 which was particularly apparent with poly(I:C), compared to CD45^+/+ cells. Unstimulated CD45^−/− cells contained more CD86 bright cells than CD45^+/+ and after CpG stimulation, and even more after poly(I:C), this CD86 bright population became much more prominent among CD45^−/− BMDC (Fig. 1b). While levels of CD80 and MHC class I were similar in the two cell types and did not alter following stimulation (data not shown), the basal level of MHC class II expression was slightly higher on CD45^−/− cells, which also showed more MHC II bright cells after poly(I:C) stimulation (Table 1). CD45 expression therefore affects both the basal levels of several surface molecules and the extent of up-regulation following stimulation with TLR ligands.

Figure 1.

Characterization of DCs in CD45^−/− and CD45^+/+ mice following TLR3 and TLR9 stimulation. (a) BMDC cultured in GM-CSF were harvested and stained with CD11c and antibodies to other cell surface molecules. Dead cells and debris were gated out on the basis of forward and side scatter. Dot plots of CD11c staining of the scatter gated cells are shown and the gates used for analysis of other cell surface molecules. (b) CD45^+/+ and CD45^−/− BMDCs were stimulated with CpG or poly(I:C) overnight and stained with CD40 and CD86. Filled histograms indicate the isotype control, light solid black line unstimulated, dotted CpG and heavy black line poly(I:C) stimulated BMDCs. (c) Cytokine production was determined in the supernatants of unstimulated (NC), CpG, CpG control or poly(I:C) stimulated BMDC or sDC (d). Data are means and standard deviations of four independent BMDC and three sDC experiments.

Table 1.

Mean fluorescence of DC

		CD40	CD86	IAb
CD45+/+	NC	23.6	39.1	43.73
	CpG	72.64	76.6	49.43
	Poly(I:C)	76.08	383	93.42
CD45−/−	NC	59.9	73.7	86.29
	CpG	246.2	218	121.22
	Poly(I:C)	328.03	591	226.6

DC were stimulated overnight with CpG or polyIC and stained with CD40, CD86 and IA antibodies. The data shown is the average geometric mean fluorescence from two to four experiments.

We next analysed the abilty of BMDCs to produce cytokines following TLR3 and TLR9 stimulation. BMDCs from CD45^−/− and CD45^+/+ mice were stimulated overnight with poly(I:C), CpG or CpG control oligonucleotides and supernatants analysed for the presence of the cytokines IL-6, IL-10, IL-12, monocyte chemoattractant protein-1 (MCP-1), TNF, IFN-γ, and IFN-α. Increased production of the pro-inflammatory cytokines TNF and IL-6 by CD45^−/− BMDCs was detected (Fig. 1c). On average, CD45^−/− BMDC produced twofold more TNF following poly(I:C) and fivefold more TNF following CpG stimulation (P = 0·001) compared to control CD45^+/+ BMDCs. IL-6 production was also increased twofold in CD45^−/− BMDC after poly(I:C) stimulation (P = 0·02). The difference in IL-6 production following CpG stimulation did not reach statistical significance, because IL-6 production was more variable from experiment to experiment, but in all eight experiments it was higher in CD45^−/− BMDC. No differences in the production of MCP-1 or IFN-α were apparent between the CD45^−/− and CD45^+/+ stimulated BMDC preparations (Table 2), while neither IL-10, IL-12, nor IFN-γ could be detected in the supernatants of poly(I:C) or CpG stimulated BMDCs. Taken together these data suggest that lack of CD45 results in increased production of the pro-inflammatory cytokines TNF and IL-6 in response to TLR3 and TLR9 stimulation.

Table 2.

Production of MCP-1 and IFN-α in BMDC

	MCP-1		IFN-α
	CD45+/+	CD45−/−	CD45+/+	CD45−/−
NC	444.4 ± 245.3	372.0 ± 122.9	22.0 ± 7.1	27.0 ± 0
CpG	2067.4 ± 1211.6	2448.9 ± 1054.2	37.0 ± 7.1	42.0 ± 7.1
CpG control	546.0 ± 270.8	673.3 ± 229.7	32 ± 7.1	27 ± 0
poly(I:C)	1767.3 ± 788.5	1818.1 ± 1186.1	1827 ± 141.4	1877 ± 28.3

BMDCs derived from CD45^+/+ or CD45^−/− mice were cultured overnight in the absence of stimulation (NC) or in the presence of CpG (CpG), CpG control oligonucleotide (CpG control) and poly(I:C). Supernatants were analysed for the presence of MCP-1 and IFN-α. Results are the mean in pg/ml and standard deviation of three independent experiments.

TLR3 and TLR9 stimulation of splenic DCs in CD45^−/− and CD45^+/+ mice

Because the relationship of BMDC to ex vivo DCs is unclear, we next studied the response to TLR3 and TLR9 stimulation in splenic DC (sDC). sDCs were isolated from 8 pooled spleens of either CD45^+/+ or CD45^−/− mice and characterized by flow cytometry to assess the presence of contaminating cell populations. Because significant numbers of B and NK cells were present, the sDC were further purified by cell sorting based upon negative selection of CD19, CD3, F4/80 and DX5 expressing cells. Purification in this manner produced sDCs that displayed >98% purity, which were used for the subsequent analysis of cytokines.

Cytokine production in response to TLR9, was assessed in CD45^−/− and CD45^+/+ sDCs following overnight stimulation with CpG, CpG control oligonucleotide, or in unstimulated cells (NC) (Fig. 1d). As with BMDC, increased TNF (P = 0·02) was detected in CD45^−/− sDCs supernatants compared to CD45^+/+ controls. In all three experiments performed, IL-6 production was higher in sDC of CD45^−/− than CD45^+/+ mice, although the difference did not reach statistical significance. No differences in IFN-α production were detected, while IL-10, IL-12, IFN-γ and MCP-1 were below the threshold of detection in supernatants of highly purified sDC though they could be detected in supernatants of sDC not subjected to flow cytometric purification. Similar results were obtained following poly(I:C) stimulation (data not shown). These results again indicate that CD45^−/− sDC like CD45^−/− BMDC have an increased capacity to produce TNF and IL-6 in response to TLR3 and TLR9 stimuli.

NFκB activation following poly(I:C) and CpG stimulation

We next analysed the biochemical mechanisms involved in the increased cytokine production of CD45^−/− cells following poly(I:C) and CpG stimulation. The TLR9 signalling pathway first involves a Toll-interleukin-1 receptor resistance (TIR) domain-containing adaptor molecule, MyD88, and following further downstream interactions activation of NFκB.^12,14 Although the TLR3 pathway uses another adaptor, the TIR domain-containing adaptor inducing IFN-β (TRIF), this again leads to activation of NFκB together with mitogen-activated protein kinase (MAPK) and IFN-regulatory factor 3/7 (IRF3/7).^10,15,16 As both TLR3 and TLR9 result in NFκB activation we assessed its activation in CD45^−/− and CD45^+/+ BMDC.

In unstimulated cells, NFκB is sequestered in the cytoplasm by a family of inhibitory proteins known as IκB, which following TLR activation, are degraded allowing the release and nuclear translocation of NFκB.^17,18 The degradation of IκB family members is an indicator of activation of NFκB and can be easily monitored by Western blotting. We therefore measured the levels of IκBα inhibitory protein in CpG and poly(I:C) treated BMDC (Fig. 2a, b). CD45^−/− BMDC cells show more profound degradation of IκBα (indicating increased activation of NFκB) at 20 and 30 min after CpG stimulation and a less dramatic difference after poly(I:C) compared to CD45^+/+ cells. The degradation of IκBα in CpG-stimulated CD45^−/− BMDC was consistently greater (ratio of IκBα/actin 0·29 and 0·24 in CD45^−/− versus 0·41 and 0·60 in CD45^+/+ cells) than in poly(I:C) treated cells (0·53 and 0·71 in CD45^−/− versus 0·76 and 0·68 in CD45^+/+ cells). However, in both cases, increased activation of NFκB correlates with increased production of IL-6 and TNF in the CD45^−/− cells. These results indicate that NFκB is more active in CD45^−/− cells and that CD45 negatively regulates NFκB activation triggered by CpG or poly(I:C).

Figure 2.

NFκB activation in BMDC. CpG (a) or poly(I:C) (b) induced activation of NFκB in BMDC as assessed by degradation of IκBα. BMDC were stimulated with CpG or poly(I:C) for the indicated times. Whole cell lysates were analysed by Western blotting using antibodies against IκBα and β-actin as a loading control. The ratio of IκBα to actin at each time point is shown. Data are representative of three independent experiments.

Response to cytokines

CD45 can also modulate cytokine receptor signalling and different CD45^−/− haemopoietic cells show increased responses to cytokines.⁵ We therefore studied the effect of TNF and IFN-α on CD45^−/− and CD45^+/+ BMDC.

TNF is considered to be a major pro-inflammatory mediator engaged both in tissue regeneration and apoptosis. Following overnight stimulation with TNF we detected increased IL-6 production in CD45^−/− cells, suggesting that these cells respond more vigorously to TNF (Fig. 3a). The increased response to TNF was further examined by analysing the signalling pathway involved. TNF binds to TNFR1, leading to its trimerization and interaction with the death domain-containing adaptor protein TRADD, although in these experiments we did not see increased cell death in TNF-stimulated cultures. However, TRADD also serves as an assembly platform for binding of adaptor molecules finally leading to NFκB activation. We analysed the activation status of NFκB by measuring the degradation of IκBα. Increased IκBα degradation was detected in CD45^−/− cells (ratio of IκBα/actin 0·17 and 0·23 versus 0·37 and 0·44 in CD45^+/+ cells) (Fig. 3b) indicating increased NFκB activation.

Figure 3.

Response to TNF and IFNα in BMDC.BMDCs were stimulated overnight with TNF (a and b) and IFN-α (c). IL-6 production in response to different doses of TNF is shown in (a), and degradation of IκBα in (b). The production of TNF and IL-6 in response to IFN-α is shown in (c). Data are representative of three experiments.

Stimulation with IFN-α of BMDCs from both CD45^+/+ and CD45^−/− mice also induced production of TNF and IL-6 (Fig. 3c). IFN-α-stimulated CD45^−/− BMDCs produced fourfold more TNF (P = 0·01) and five-fold more IL-6 (P = 0·04) compared to CD45^+/+ BMDCs.

Taken together these results indicate clearly that the lack of CD45 leads to increased responses to TNF and IFN-α.

Discussion

Here we show that CD45 suppresses NFκB activation and negatively regulates TLR3 and TLR9 signalling. Lack of CD45 in CD45^−/− BM and sDC leads to enhanced IL-6 and TNF production and NFκB activation. In three experiments using the TLR7 ligand R837 to stimulate BMDC, there were inconsistent changes in TNF and MCP-1 production but a consistent increase in IL-6. The overall amount of IL-6 produced in the three experiments varied between 7000 and 28 000 pg/ml, making statistical analysis impossible, but these preliminary data, together with the data shown above, indicate that CD45 influences signalling through several TLRs.

The exact mechanism by which CD45 affects IL-6 and TNF production is unknown. Most TLR pathways begin with a common TIR domain-containing adaptor molecule, MyD88, and recruitment of the IL-1R-associated kinase (IRAK) and TNFR-associated factor 6 (TRAF6). Following a sequence of downstream interactions NFκB is activated. One possibility is that CD45 could modulate TLR signalling directly by dephosphorylating an adaptor molecule but this has not been reported. Alternatively CD45 might influence TLR signalling through dephosphorylation of Src kinases. This is a well-established mechanism in T-cell receptor signalling¹ but could also operate for TLR, because macrophages from Bruton's tyrosine-kinase deficient mice show reduced responses to TLR signals, implying that Src kinases are involved in TLR signalling.¹⁹ Signalling through TLRs also leads to activation of yet other signalling pathways. Thus the TLR3 ligand, dsRNA, activates not only NFκB but also MAPK and IRF3/7 pathways. In NK cells CD45 has been shown to regulate the MAPKs, c-JunNH₂-terminal kinase (JNK) and p38 after ligation of Ly49D.³ Although the involvement of these pathways in TLR9 signalling stimulated by CpG oligonucleotides is less clear, CD45 might still regulate a so far ill-defined TLR9 alternative signalling pathway.

In this study we show not only that CD45 negatively regulates cytokine production in DC stimulated via TLRs, but also negatively regulates the response to cytokines. Thus CD45^−/− produce more IL-6 following TNF and more TNF and IL-6 following IFN-α stimulation than do CD45^+/+ cells. Signalling through the TNF receptor is mediated by NFκB activation and we show increased IκBα degradation in TNF-stimulated CD45^−/− BMDC. Although in this study we did not follow these responses beyond 30 min, activated NFκB is known to rapidly induce IκBα in an autoregulatory feedback loop that shuts off the further activation of NFκB, so that IκBα degradation is usually transient.²⁰

Signalling through the type I interferon receptor involves Jak/STAT pathways, which have been shown previously to be modulated by CD45.⁵ Thus it appears that CD45 may influence cytokine production in DCs at two levels: first by modulation of initial signalling through TLRs and second by affecting the well-known autocrine amplification of cytokine production associated with binding of type I IFNs or TNF to their receptors.^21,22

Regulation of DC function by CD45 is clearly extremely complex. In the present study BMDC of both mouse strains showed similar CD11c expression at the end of the initial in vitro culture in GM-CSF, but CD45^−/− cells expressed higher levels of CD40, CD86 and MHC class II and following TLR stimulation up-regulated these molecules to a greater extent. The increased cytokine production observed following TLR stimulation may therefore be because the CD45^−/− BMDC differentiate abnormally in the absence of CD45 and are more activated. Irrespective of this, abnormal function of T and B lymphocytes in CD45^−/− mice, and by extension CD45 transgenic mice or humans with altered CD45 expression, may in part be caused by alterations in function of cells of the innate immune system.

TNF and IL-6 production differed in CD45^−/− and CD45^+/+ DC, MCP-1 and IFN-α production did not. In contrast, in a separate study, we observed that splenic DC of CD45^−/− mice have an impaired capacity to produce type I IFN in response to LCMV infection in vivo, although the exact mechanism underlying this defect has not been determined (Montoya et al., submitted for publication). The apparent discrepancy between these two studies may be because in vivo LCMV activates pattern recognition receptors distinct from TLR9 and TLR3.²³ In the spleen DC are exposed to a complex microenvironment containing many cells that may deliver signals to them. In contrast in the in vitro experiments described here, much more defined signals are delivered to DCs in isolation. Nevertheless these contrasting results indicate that depending on the context CD45 may act as a positive or negative regulator of signalling. This is in line with several reports in the literature.^5,6 Thus CD45^−/− NK cells show dramatically reduced chemokine and cytokine production following Fc or Ly49D activation.³ However CD45^−/− mice produce high levels of several cytokines in the gut even though profoundly lacking in T cells.²⁴ Similar positive and negative roles for CD45 have been documented in antigen receptor signal transduction.²⁵

Although the mechanisms underlying these contrasting effects are not clearly established, it is tempting to speculate that the distribution of CD45 isoforms on cells contributes to differences in signalling. It is well established that the size of the CD45 extracellular domain is important for signalling²⁶ and differences in signalling can be detected in T cells transfected with distinct isoforms.²⁷ Experiments on human lymphocytes from individuals carrying variant alleles of CD45 also indicate that combinations of isoforms can influence TCR signalling and cytokine production.^28,29 Individuals carrying the variant alleles show altered disease susceptibility. The present experiments indicate that altered immune function in variant individuals may be a consequence not only of alterations in T-cell function but altered function of another key cell of the immune system, antigen presenting dendritic cells.

Abbreviations

BMDC

bone marrow dendritic cell

FCS

fetal calf serum

IFN

interferon

IL

interleukin

IRF3/7

IFN-regulatory factor 3/7

Jaks

Janus kinases

MAPK

mitogen-activated protein kinase

MHC

major histocompatibility complex

PAMPS

pathogen-associated molecular patterns

rmuGM-CSF

recombinant murine granulocyte–macrophage colony-stimulating factor

sDC

splenic dendritic cell

Src

Sarc

TIR

Toll-interleukin-1 receptor resistance domain-containing adaptor molecule

TLR

Toll-like receptor

TNF

tumour necrosis factor

TRIF

TIR domain-containing adaptor inducing INF-β