Anti-cd3 Tcr Activation

A Dr. von Hauner Children s Hospital, Ludwig-Maximilians-University, Munich, Germany; b Laboratory of Lymphocyte Activation and Susceptibility to EBV Infection.

SYK expression endows human ZAP70-deficient CD8 T cells with residual TCR signaling

APC/Cy7 Anti-human CD3 Antibody Anti-CD3 - CD3ε is a 20 kD chain of the CD3/T-cell receptor TCR complex which is composed of two CD3ε, one CD3γ, one CD3δ, one.

A Dr. von Hauner Children s Hospital, Ludwig-Maximilians-University, Munich, Germanyb Laboratory of Lymphocyte Activation and Susceptibility to EBV Infection, INSERM UMR1163, Institut IMAGINE, Paris, Francec Paris Descartes University, Sorbonne Paris Cité, Imagine Institut, Paris, Franced BIOSS Centre for Biological Signalling Studies, Faculty of Biology, University of Freiburg, Germanye Center for Chronic Immunodeficiency CCI, University Medical Center, University of Freiburg, Germanyf Faculty of Biology, University of Freiburg, Freiburg, Germanyg Center for Pediatrics and Adolescent Medicine, University Medical Center, University of Freiburg, Germanyh Division of Pediatric Immunology and Rheumatology, Department of Pediatrics, University Hospital Mainz, Germanyi Study Center of Immunodeficiencies, Necker-Enfants Malades Hospital, Assistance Publique-Hôpitaux de Paris AP-HP, Paris, Francej ImmunoDeficiencyCenter Leipzig IDCL, St. Georg Hospital, Leipzig, Germanyk Translational Centre for Regenerative Medicine TRM, University Leipzig, Leipzig, Germanyl Laboratoire d Immunologie, Centre Hospitalier Universitaire de Nantes, Francem Institute for Transfusion Medicine, University Hospital Ulm and Institute for Clinical Transfusion Medicine and Immunogenetics Ulm, German Red Cross Blood Service, Baden-Württemberg-Hessen, Ulm, Germanyn Unité de Régulation Immunitaire et Vaccinologie, Institut Pasteur, Paris, Franceo Unité d Immunologie et Hématologie Pédiatrique, Necker-Enfants Malades Hospital, AP-HP, Paris, Francep Collège de France, Paris, Franceq Laboratory of Human Genetics of Infectious Diseases, Necker Branch, INSERM UMR 1163, Institut IMAGINE, Necker Medical School, Paris, FranceReceived 21 May 2015, Revised 29 June 2015, Accepted 1 July 2015, Available online 14 July 2015

Autosomal recessive human ZAP70 deficiency is a rare cause of combined immunodeficiency CID characterized by defective CD4 T cells and profound CD8 T cell lymphopenia. Herein, we report two novel patients that extend the molecular genetics, the clinical and functional phenotypes associated with the ZAP70 deficiency. The patients presented as infant-onset CID with severe infections caused by varicella zoster virus and live vaccines. Retrospective TCR excision circle newborn screening was normal in both patients. One patient carried a novel non-sense mutation p.A495fsX75 ; the other a previously described misense mutation p.A507V. In contrast to CD4 T cells, the majority of the few CD8 T cells showed expression of the ZAP70-related tyrosine kinase SYK that correlated with residual TCR signaling including calcium flux and degranulation. Our findings highlight the differential requirements of ZAP70 and SYK during thymic development, peripheral homeostasis as well as effector functions of CD4 and CD8 T cells.KeywordsCID; ZAP70; SYK; TCR signaling; TREC newborn screening; Live vaccine adverse event

Copyright 2015 Elsevier Inc. All rights reserved.

The mechanism of T cell avidity maturation has remained elusive. Geisler and co-workers show that induction of the phospholipase PLC- gamma 1 via the alternative p38.

TLR4 deficiency aggravates IL-10–dependent colitis. Previous studies have identified the TLR expression profile of T cells. To investigate the potential role of.

Step II: Addition of Cells: 1. Prepare PBMC and resuspend the cells at 1-2x10 6 /mL of complete RPMI. Note: This density of PBMCs is optimal for TCR-mediated T cell.

TLR4 deficiency aggravates IL-10–dependent colitis. Previous studies have identified the TLR expression profile of T cells 10. To investigate the potential role of these receptors in the development of colitis, we crossed Tlr4–/– or Tlr9–/– mice onto Il10–/– animals. We followed the animals for 2 months and then analyzed their colons for signs of inflammation. While Il10–/– and Il10–/–Tlr9–/– mice did not develop obvious signs of intestinal inflammation at the age of 8 months, the Il10–/–Tlr4–/– mice demonstrated overt colitis at the age of 8 weeks, i.e., thickening of the intestinal wall, diarrhea, enlarged spleen and mesenteric lymph nodes MLNs Supplemental Figure 1; supplemental material available online with this article; doi:

10.1172/JCI40055DS1. Histological analysis of the colon revealed that Il10–/–Tlr4–/– mice developed severe inflammation, with a high degree of epithelial crypt hyperplasia Figure 1A, transverse section and marked infiltration of mononuclear cells in the colonic lamina propria LP Figure 1B, cross section. Figure 1C displays quantitative morphometric analysis of these inflammatory parameters. In addition, we observed goblet cell depletion in the mucosal layers and an increase in epithelial cell proliferation data not shown.

Figure 1

TLR4 deficiency aggravates colitis in Il10–/– mice.

A Histological analysis of 8-week-old B6, Il10–/–, Il10–/–Tlr9–/–, and Il10–/–Tlr4–/– mice maintained under specific pathogen–free conditions showing increased epithelial crypt hyperplasia in Il10–/–Tlr4–/– mice original magnification, 100. B Cross-sectional analysis of colons from the different mice showing greater cellular infiltration in the mice lacking TLR4 original magnification, 100. C Quantitative measurement of crypt length and cellular infiltration in the different groups of mice at 8 weeks of age. D Histological analysis of 4-week-old Il10–/– mice housed with or without 7- to 8-week-old Il10–/–Tlr4–/– mice original magnification, 100. E Quantitative measurements of crypt length and cellular infiltration in co-housed animals. Data represent mean SEM of 3 different experiments. P 0.01.

The colitis observed in Il10–/–Tlr4–/– mice could result from altered microflora in these animals. To evaluate this possibility, we co-housed young Il10–/– with diseased Il10–/–Tlr4–/– mice in the same cage to allow colonization of these 2 groups with the same microflora 11. Age- and sex-matched Il10–/– mice housed separately served as controls. Il10–/– mice, under each housing condition, were monitored weekly for signs of intestinal inflammation for an additional period of 6 weeks, and histological evaluation was performed at the end of this period. Under these conditions, the co-housed Il10–/– mice showed no difference in the development of intestinal inflammation compared with Il10–/– mice housed in separate cages Figure 1D. Quantification of crypt length and cellular infiltration showed no difference between the colons of these 2 different groups Figure 1E. Taken together, these data indicate that the lack of TLR4 accelerates the intestinal inflammation in the Il10–/– host.

Characterization of the inflammatory profile in Il10–/–Tlr4–/– mice.

To determine the inflammatory profile associated with colitis, we measured the mRNA levels of cytokines, chemokines, and cell markers in tissue homogenates obtained from the colons of each group of mice at 8 weeks of age. Consistent with the inflammatory phenotype, we found that Il10–/–Tlr4–/– mice had increased transcript levels of inflammatory cytokines such as IL-6, IFN-γ, IL-1β, and IL-17A, as well as chemokines keratinocyte chemoattractant KC, macrophage inflammatory protein 1α MIP1α, MIP1β, CCL22, and CXCL10. The mRNA levels of the cellular markers F4/80 macrophages, MPO neutrophils, and CD3d/CD4 T cells were also increased in the Il10–/–Tlr4–/– mice as compared with the other tested groups Table 1. To quantify the inflammatory mediators produced by the inflamed colons, we cultured ex vivo colonic explants CEs from Il10–/– and Il10–/–Tlr4–/– mice. Supernatants from these cultures where then analyzed by ELISA. As shown in Figure 2A, the CEs from Il10–/–Tlr4–/– mice released high amounts of IL-17A, IFN-γ, and TNF-α, whereas these cytokines were barely detectable in the supernatants harvested from the Il10–/– CEs. To further evaluate the inflammatory phenotype in the colons of these mice, intraepithelial lymphocytes IELs and LP lymphocytes LPLs from Il10–/– and Il10–/–Tlr4–/– mice were isolated and stimulated with anti-CD3/CD28 Abs. Colonic lymphocytes from Il10–/–Tlr4–/– mice produced significantly higher levels of proinflammatory cytokines such as IL-6, IL-17A, TNF-α, or IFN-γ when compared with colonic lymphocytes from Il10–/– mice Figure 2, B and C.

Figure 2

Il10–/–Tlr4–/– mice develop an increased colonic proinflammatory profile.

A Cytokine levels in CE supernatants after 24 hours of culture ELISA. Represented values are normalized to milligrams of cultured colonic tissue. B and C Cytokine levels obtained from LPLs B and IELs C isolated from the colons of Il10–/– and Il10–/–Tlr4–/– mice and stimulated with anti-CD3/CD28 Abs for 24 hours Data represent mean SEM of 3 different experiments. P P 0.01.

Table 1

qRT-PCR analysis of RNA samples isolated from colon of mice of the different genotypes

TLR4 expression on effector CD4 T cells restrains colitis in an adoptive transfer model.

Colitis in Il10–/– mice depends on CD4 T cells 3. To determine whether the accelerated colitis observed in Il10–/–Tlr4–/– is dependent on effector CD4 T cells, we carried out 2 different experiments. First, we selectively depleted the CD4 population by injecting anti-CD4 depleting Abs into Il10–/–Tlr4–/– mice Supplemental Figure 2A. Depletion of CD4 cells resulted 2 weeks later in significant inhibition of crypt hyperplasia, cellular infiltration, and suppression of the colonic proinflammatory cytokine profile Supplemental Figure 2, B and C. Second, we transferred naive CD4 T cells i.e., CD4 CD45RBhi from Il10–/– or from Il10–/–Tlr4–/– donor mice to Rag1–/– recipients 12 and evaluated the subsequent colitis 8 weeks later. As a control, we cotransferred CD4 CD45RBloCD25 Tregs from wild-type C57BL/6 B6 mice along with the naive CD4 subsets from Il10–/– or Il10–/–Tlr4–/– mice. We observed that naive CD45RBhi T cells from Il10–/–Tlr4–/– induced greater body weight loss than those from Il10–/– mice Figure 3A. As expected, the cotransfer of CD45RBloCD25 T cells from B6 mice inhibited the induction of colitis in the recipients Figure 3B. In accordance with the body weight data, histological analysis of the colonic tissues revealed more severe inflammation in mice that received the naive CD4 T cells from the Il10–/–Tlr4–/– mice. The colitis in these recipients was characterized by depletion of goblet cells and marked abnormalities in the crypt structure, such as crypt hyperplasia, cellular infiltration, and distorted crypt orientation Figure 3C. Morphometric quantification of the crypt length Figure 3D and the number of infiltrating cells Figure 3E revealed a significant increase in these inflammatory parameters in mice transferred with naive TLR4-deficient CD4 cells. In some areas, we detected crypt abscesses that were not detected upon transfer of naive CD4 T cells from Il10–/– mice Figure 3C. Furthermore, increased proinflammatory cytokines were generated by CEs taken from recipients transferred with naive CD4 T cells from Il10–/–Tlr4–/– mice as compared with recipients transferred with naive CD4 T cells from Il10–/– mice Figure 3F. We next analyzed the cytokine production by CD4 T cells harvested from different organs of the recipient mice. CD4 T cells from the spleen, MLNs, and colonic mucosa IELs and LPLs were isolated and stimulated with anti-CD3/CD28 Abs. While there was no significant difference in the cytokine production by splenic CD4 T cells, we found that MLN CD4 T cells isolated from recipients reconstituted with naive Il10–/–Tlr4–/– cells produced higher levels of IFN-γ. Moreover, the CD4 IELs and LPLs from these recipients produced higher levels of IL-6, IFN-γ, and IL-17A Figure 3G. Of note, the adoptive transfer of naive Tlr4–/– CD4 T cells to Rag1–/– recipients also resulted in greater loss of body weight and in more severe colitis than that induced by B6 naive CD4 cells Supplemental Figure 3. To explore whether the increased inflammatory phenotype observed after adoptive transfer of Il10–/–Tlr4–/– T cells was due to decreased Treg development, we checked the frequency of Foxp3 cells by IHC in colonic tissue from Rag1–/– recipients transferred with the indicated CD4 T cell populations. Consistent with previous results 13, the presence of Foxp3 cells was higher in the highly inflamed colons, i.e., the frequency of Foxp3 Tregs arising from the Il10–/–Tlr4–/– CD45RBhi naive population was increased when compared with the number of Foxp3 Tregs observed in the colons of mice receiving Il10–/– CD45RBhi cells Supplemental Figure 4, A and B. Similar numbers of Foxp3 cells were found in the mice receiving Treg cells from B6 mice cotransfer groups. Collectively, these data suggest that TLR4 ligation exerts a tonic inhibitory effect on colitogenic CD4 T cells.

Figure 3

TLR4 expression on CD4 T cells restrains colitis in an adoptive transfer model.

A and B Percentage of initial body weight of Rag1–/– recipients transferred with Il10–/– and Il10–/–Tlr4–/– FACS-sorted naive CD45RBhi T cells A and cotransferred with regulatory CD45RBloCD25 Tregs B6 B. C Microscopic evaluation revealed mild inflammation in the colons of Il10–/– CD45RBhi recipients, while severe inflammation was observed in the colon of Il10–/–Tlr4–/– CD45RBhi recipients original magnification, 100. D and E Quantitative analyses of crypt length and cellular infiltration performed in colonic tissues from the recipient mice. Asterisks represent significant differences compared with control group group receiving Il10–/– naive and B6 Tregs. F Cytokine levels in CE supernatants after 24 hours of culture. G CD4 T cell cytokine response. CD4 cells from colon IELs and LPLs, MLNs, and spleen were isolated from Rag1–/– recipients and stimulated with anti-CD3/CD28 Abs. Cytokine levels were determined 24 hours later. Data represent mean SEM of 2 independent experiments. P P 0.01.

To further verify this effect, we injected anti-TLR4 blocking Ab to Il10–/– mice. The blockade of TLR4 resulted in an increase in the production of proinflammatory cytokines by MLN CD4 T cells after anti-CD3/CD28 Ab stimulation Supplemental Figure 5A. Similarly, CEs from these mice displayed higher levels of IL-6, IFN-γ, IL-17A, and TNF-α Supplemental Figure 5B. Taken together, these data demonstrate that the lack of TLR4 on IL-10–deficient CD4 T cells impacts their inflammatory profile.

Naive CD4 T cells express functional TLR4.

The expression of TLR4 on naive CD4 T cells was recently reported 14, 15. Others groups documented TLR4 expression in Tregs 16 as well as in colitogenic LP CD4 cells 17. Consistent with these reports, we detected TLR4 expression in gated CD3 CD4 cells from spleen of B6 and Il10–/– mice Supplemental Figure 6. As expected, TLR4 expression was absent in the double-deficient Il10–/–Tlr4–/– mice Supplemental Figure 6. Our finding that TLR4-deficient effector CD4 T cells display different inflammatory phenotypes at various sites Figure 3, F and G led us to hypothesize that TLR4 expression on these cells may be the result of a site-specific or an activation-induced regulation. We therefore analyzed the level of TLR4 expression in FACS-sorted CD4 T cells isolated from spleen and MLN of Il10–/– mice by quantitative RT-PCR qPCR. The purity of the analyzed CD4 populations was greater than 99 data not shown. First, we observed different levels of TLR4 expression in CD4 T cells isolated from different organs Figure 4A, and the lowest expression was observed in LPLs and IELs. Second, and consistent with a previous report 15, we observed that the expression of TLR4 in these cells was downregulated by TCR stimulation Figure 4B.

Figure 4

CD4 T cells express functional TLR4.

A qPCR analysis of Tlr4 expression in unstimulated CD4 T cells from spleens Sp, MLNs, LPLs, and IELs of Il10–/– mice. Total splenocytes Total Sp were used as a positive control. B qPCR analysis of Tlr4 expression in spleen CD4 T cells before and 24 hours after anti-CD3/CD28 Abs stimulation. C and D MLN CD4 T cells were treated with LPS for the indicated times. Nuclear lysates were examined for NF-κB activation EMSA C, and cytosolic lysates were tested for ERK1/2, p38, and JNK activation by immunoblotting with phospho-specific Abs D. E and F NF-κB and ERK activation in Tlr4–/– MLN CD4 T cells after LPS stimulation. All CD4 T cells used were FACS sorted, and purity was greater than 98 in all experiments. Stimulation was carried out with 100 ng/ml of LPS in all experiments. Data represent mean SEM and are representative of 3 independent experiments 4 mice per experiment were used in A and B. P P 0.01.

Since TLR4 deficiency in effector CD4 T cells affected their inflammatory phenotype in vivo, we explored TLR4 signaling in CD4 T cells. LPS stimulation led to the activation of the NF-κB signaling pathway Figure 4C, as well as to the phosphorylation of members of the MAPK family, i.e., p38, JNK, and ERK1/2 Figure 4D. As expected, MLN-derived CD4 T cells from Tlr4–/– mice were unresponsive to LPS stimulation Figure 4, E and F.

TLR4 signaling downregulates IFN-γ production through ERK1/2 inhibition. As LPS is a soluble mediator, its interaction with its receptor TLR4 on CD4 T cells is not restricted to lymphoid organs. This is in contrast to the cognate TCR-MHCII complex interaction of T cells with APCs. We therefore reasoned that at intestinal mucosal sites, there is a considerable chance of having TLR4-stimulated CD4 T cells prior to their TCR engagement. To further explore the impact of LPS signaling on CD4 T cell function, splenic CD4 T cells from OVA-transgenic OT-II mice were prestimulated with LPS, or left unstimulated, before being cocultured with OVA-loaded bone marrow dendritic cells BMDCs; wild-type for 5 days. Interestingly, LPS pretreatment significantly reduced the production of IFN-γ by CD4 T cells while increasing the production of IL-17A Figure 5A. The levels of other cytokines, such as TNF-α or IL-2, remained unaffected by LPS pretreatment followed by TCR stimulation data not shown. Similar data were observed in the BMDC-free system, in which CD4 T cells from B6 mice spleen, MLN, and colonic LP were stimulated with LPS followed by anti-CD3/CD28 Abs Supplemental Figure 7, A and B.

Figure 5

TLR4 signaling in CD4 T cells downregulates IFN-γ production through ERK1/2 inhibition.

A Splenic CD4 T cells from OT-II mice were incubated for 2 hours in the presence of 100 ng/ml LPS or medium alone. The cells were then washed and cultured for 5 days with OVA-loaded BMDCs. After coculture, the CD4 cells were restimulated, and the secretion of cytokines was determined 24 hours later. B NF-κB activation was assessed in nuclear extracts by EMSA. C Cytosolic lysates were immunoblotted for MAPK protein activation after anti-CD3/CD28 Ab stimulation with or without LPS 100 ng/ml for 2 hours prior to TCR stimulation. D LPS-dependent regulation of ERK1/2 phosphorylation was confirmed by phospho-protein analysis by flow cytometry as described in Methods. Numbers within histograms denote the percentage of p-ERK1/2–positive cells in stimulated cells when compared with control cells. E Activation of NFAT-1 was measured by immunoblotting of nuclear extracts from CD4 cells stimulated as in C. F RT-PCR analysis of CD4 T cells incubated with the MEK/ERK1/2 inhibitor UO126 or vehicle DMSO before being stimulated with anti-CD3/CD28 Abs for the indicated times. Data in A–F represent mean SEM of 3 independent experiments. P A, were carried out with FACS-sorted MLN-derived CD4 cells from Il10–/– mice purity, 98.

To delineate the signaling mechanism by which LPS exerts its regulatory effect on subsequent TCR activation, we isolated FACS-sorted MLN CD4 T cells from Il10–/– mice and incubated them in the presence or absence of LPS for 2 hours and then stimulated them with anti-CD3/CD28 Abs. LPS stimulation alone induced the translocation of NF-κB Figure 5B. However, LPS pretreatment had no significant effect on the TCR-dependent NF-κB translocation compared with that observed after TCR activation alone Figure 5B. In contrast, phosphorylation of MAPK family members was strongly affected by the LPS pretreatment. Specifically, TLR4 triggering prior to TCR stimulation strongly inhibited p-ERK1/2 levels, whereas p-p38 levels were only slightly affected by LPS pretreatment Figure 5C. Similarly, using intracellular staining FACS, we detected a 3-fold difference in geometric mean fluorescence GMF intensity of p-ERK1/2 after 2 hours of LPS pretreatment Figure 5D. A similar trend of p-ERK1/2 inhibition was also observed for lower concentrations of LPS Supplemental Figure 8, A and B. Activation of nuclear factor of activated T cells–1 NFAT-1 was unaffected by LPS pretreatment Figure 5E. Of note, costimulation of TLR4 and TCR did not have the same inhibitory impact on either the activation of ERK1/2 or the production of IFN-γ Supplemental Figure 9.

Since pretreatment with LPS resulted in a significant reduction in IFN-γ production Figure 5A and inhibition of ERK1/2 phosphorylation Figure 5C in CD4 T cells, we asked whether these findings are linked. To explore this issue, we pretreated the CD4 T cells with the specific MEK1/2 inhibitor UO126 before TCR stimulation with anti-CD3/CDCD28 Abs. Similarly to LPS pretreatment, UO126 pretreatment reduced the mRNA expression level of IFN-γ Figure 5F, while the transcript levels of other proinflammatory cytokines such as IL-6 and IL-17A were not significantly altered by MEK/ERK inhibition data not shown.

TLR4 signaling regulates subsequent TCR signaling events via the induction of MAPK phosphatases. The lack of ERK1/2 activation induced by LPS treatment of CD4 T cells could be explained by either diminished phosphorylation by MEK1/2 or increased dephosphorylation of p-ERK1/2 by specific phosphatases. As shown in Supplemental Figure 10, LPS pretreatment did not inhibit the TCR-dependent phosphorylation of MEK1/2, indicating that inhibition of p-ERK1/2 is not due to an upstream event. MAPK enzymes undergo inactivation by MAPK phosphatases MKPs, also known as dual-specificity protein phosphatases DUSPs, which dephosphorylate both phosphothreonine and phosphotyrosine residues on activated MAPKs 18. Since the expression of several MKPs was documented in T cells 19–21, we tested whether TLR4 triggering regulates p-ERK1/2 through the induction of MKPs. Indeed, we found that LPS stimulation of Il10–/– CD4 T cells led to a marked increase in cytosolic MKP-3 Figure 6A. In accordance with these data, freshly isolated MLN CD4 T cells from Il10–/–Tlr4–/– mice showed reduced expression of MKP-3 Figure 6B. Moreover, LPS pretreatment led to increased levels of both nuclear MKP-1 and cytosolic MKP-3 upon subsequent anti-CD3/CD28 Abs stimulation Figure 6C. In contrast, TCR activation alone did not induce any activation of MKP-1 or MKP-3 in these cells at the tested time period Figure 6C. The phosphorylation level of SHIP-1, a phosphatase that is associated with LPS tolerance in macrophages 22, was largely unaffected by LPS stimulation of Il10–/– CD4 T cells or in Il10–/–Tlr4–/– mice Figure 6, A and B. In contrast, a slight increase in the TCR-dependent phosphorylation of SHIP-1 was observed after LPS pretreatment Figure 6C.

Figure 6

TLR4 modulates TCR-dependent MAPK phosphatase activation. A Immunoblot analysis of the indicated phosphatases after LPS stimulation of CD4 cells at different time points. B Expression of MKPs and p-SHIP1 in freshly isolated, unstimulated CD4 cells from Il10–/– and Il10–/–Tlr4–/– mice. C Immunoblotting of protein extracts from CD4 T cells stimulated with LPS or left untreated before stimulation with anti-CD3/28 Abs for different periods of time. D Analysis of the siRNA knockdown in CD4 T cells 24 hours after transfection with either MKP-3 or control siRNA Ctrl. E TCR-dependent phosphorylation of ERK1/2 24 hours after transfection with MKP-3 or control siRNA. Numbers within histograms denote percentage of p-ERK1/2–positive cells in stimulated cells when compared with control cells. F Cytokine levels measured from 24-hour supernatants of cells treated as in E. Data represent mean SEM. G p-ERK expression in freshly isolated CD4 cells from Il10–/– mice treated with MKP inhibitor in vivo NSC 95397 or vehicle. H Cytokine levels after 24 hours of anti-CD3/28 Ab stimulation of MLN-derived CD4 T cells isolated from Il10–/– mice 3 days after injection of NSC 95397 or vehicle. Data represent mean SEM. Data are representative of at least 2 independent experiments. P cells from Il10–/– or Il10–/–Tlr4–/– mice purity, 98. Stimulation was carried out with 100 ng/ml LPS in all experiments. The lanes shown in B were run on the same gel but were noncontiguous.

As MKP-3 was downregulated in Il10–/–Tlr4–/– CD4 T cells Figure 6B, and since it was shown to specifically regulate ERK1/2 while minimally affecting other MAPKs 23, we next explored its impact on Il10–/– CD4 T cell activation profile by gene silencing Figure 6D. MKP-3 knockdown abrogated the LPS-dependent downregulation of ERK1/2 activation in vitro, as determined by FACS-based intracellular staining Figure 6E. LPS pretreatment resulted in a marked decrease in both the percentage of p-ERK1/2–positive cells and the GMF intensity in the control siRNA transfected cells, whereas such an effect was not observed in MKP-3 siRNA transfected cells Figure 6E. MKP-3 knockdown also abrogated the LPS-dependent downregulation of IFN-γ production by anti-CD3/CD28 Abs stimulation but did not affect IL-17A production Figure 6F.

TLR4 signals through 2 adaptors, MyD88 and TRIF. To further explore the mechanism by which TLR4 exerts its control over MKP-3 activation, we examined whether this effect is dependent on the MyD88- or the TRIF-mediated signaling. We stimulated MLN-derived CD4 T cells from Il10–/–Myd88–/– or Il10–/–/LPS2 mice the LPS2 mouse has a Trif–/– phenotype 24 with LPS and checked the subsequent MKP-3 activation. Interestingly, MKP-3 activation was unaffected in the absence of MyD88 but was strongly reduced in LPS2 CD4 T cells Supplemental Figure 11. These data indicate that TLR4 activates MKP-3 in a TRIF-dependent manner and provide a possible explanation for the different phenotypes observed in Il10–/–Tlr4–/– and Il10–/–Tlr9–/– mice Figure 1, A and B.

To validate that the increased inflammation observed in the absence of TLR4 signaling in the Il10–/– mice is related to the lack of induction of MKPs, we inhibited MKPs in vivo using a cell-permeable, quinone-based inhibitor of dual-specificity phosphatases, NSC 95397, which inhibits both MKP-1 and MKP-3 25. Injection of a single dose of this compound into Il10–/– mice resulted 3 days later in significant body weight loss and diarrhea as compared with vehicle-treated mice Supplemental Figure 12, A and B. Furthermore, the number of CD4 T cells present in MLNs of mice treated with NSC 95397 was also increased when compared with vehicle-treated mice Supplemental Figure 12C. In addition to these signs of intestinal inflammation, the administration of this compound in vivo resulted in the upregulated expression of p-ERK1/2 in freshly isolated MLN-derived CD4 T cells Figure 6G. More importantly, stimulation of these MLN CD4 T cells with anti-CD3/CD28 Abs resulted in increased production of proinflammatory cytokines Figure 6H. Finally, in vivo blockade of TLR4 signaling in Il10–/– mice by administration of anti-TLR4 blocking Abs led to a reduction in the basal expression of MKP-1 and MKP-3 in freshly isolated CD4 T cells from spleen and MLN Supplemental Figure 13. In addition, as discussed above, in vivo blockade of TLR4 led to increased production of proinflammatory cytokines by stimulated MLN CD4 T cells and by CEs Supplemental Figure 5, A and B. However, no changes in body weight or histological score crypt elongation and infiltration were observed in mice treated with TLR4 blocking antibody. Repeated treatment and/or longer follow-up may be required to identify colonic inflammation induced by TLR4 blockade.

Collectively, these data indicate that TLR4 signaling in colitogenic CD4 T cells regulate their activation mainly through the induction of MKP-3 that restrains p-ERK levels upon subsequent TCR stimulation Figure 7.

Figure 7

Proposed mechanism of TLR4-dependent regulation of TCR activation conditional activation. TLR4 triggering of CD4 T cells activates NF-κB and MAPK pathways. However, the induction of MKP1 and MKP3 occurs mainly via the TRIF pathway. The high levels of MKP-1 and MKP-3 restrain a subsequent TCR-dependent activation of p38 and ERK, respectively. This activation of phosphatases in CD4 T cells by TLR4 restrains subsequent TCR-induced phosphorylation events and, hence, modulates CD4 T cell responses and their inflammatory phenotypes. AP-1, activator protein 1; MKK, MAPK kinase.

The activation of a naive T cell requires two signals : ligation of the TCR with the MHC/peptide complex on the APC and cross-linking of costimulatory receptors on.