Costimulatory activation of murine invariant natural killer T cells by toll-like receptor agonists

Cellular Immunology 277 (2012) 33–43

Contents lists available at SciVerse ScienceDirect

Cellular Immunology journal homepage: www.elsevier.com/locate/ycimm

Costimulatory activation of murine invariant natural killer T cells by toll-like receptor agonists Raveendra R. Kulkarni a,1, Alexander Ian Villanueva a, Inas Elawadli a, Preethi Jayanth a,2, Leah R. Read a, S.M. Mansour Haeryfar b, Shayan Sharif a,⇑ a b

Department of Pathobiology, University of Guelph, Guelph, Ontario, Canada N1G 2W1 Department of Microbiology and Immunology, The University of Western Ontario, London, Ontario, Canada

a r t i c l e

i n f o

Article history: Received 17 January 2012 Accepted 4 June 2012 Available online 17 June 2012 Keywords: CD1d Innate immunity T cell receptors a-GalCer CD3 stimulation

a b s t r a c t Invariant NKT (iNKT) cells are glycolipid-reactive lymphocytes with anti-microbial properties. Toll-like receptor (TLR)-primed antigen-presenting cells are known to activate iNKT cells, however, the expression and function of TLRs in iNKT cells remain largely unknown. Here, we show that TCR-activation of murine iNKT cells by a-GalactosylCeramide (a-GalCer) or anti-CD3 antibodies can result in increased expression of TLR genes. TLR3, 5 and 9-mediated costimulation of TCR-preactivated iNKT cells resulted in enhancement of iNKT cell activation, as determined by their cytokine production. Expression of TLR3 and 9 at protein level was also confirmed in TCR-activated iNKT cells. Furthermore, TCR-preactivation followed by TLR9-costimulation of iNKT cells increased their ability to induce maturation of dendritic cells. Thus, our findings show that iNKT cells can up-regulate their TLR expression upon TCR activation and a subsequent TLR-signaling in these cells can lead to their enhanced activation, suggesting a new possible mode of iNKT cell activation. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction Invariant NKT (iNKT) cells are an important subset of ab TCR+ T lymphocytes that express an invariant TCR and recognize glycolipid antigens. Alpha-GalactosylCeramide (a-GalCer), a glycolipid antigen, is a potent activator of iNKT cells. Upon activation, iNKT cells can produce large amounts of T helper (Th)1-type and/or Th2-type cytokines. The importance of iNKT cells in the regulation of immune responses, in preventing autoimmune diseases and in immunity against microbial pathogens has been well documented [1–4]. iNKT cells can be activated either by direct TCR engagement or indirectly through signals received from antigen-presenting cells (APC) that have encountered pathogens [3]. Pattern recognition receptors, such as toll-like receptors (TLR), were earlier thought to be expressed only by APC; however, it has now become evident that TLRs are also expressed by various T cell subsets [5,6]. Several studies have shown direct costimulatory effects of TLR agonists in T cells leading to their proliferation, activation and enhancement of effector functions, in the absence of APC [7–9]. An important ⇑ Corresponding author. Fax: +1 519 824 5930. E-mail address: [email protected] (S. Sharif). Present address: Department of Pediatrics and Emory Vaccine Center, Emory University School of Medicine, Atlanta, GA 30322, USA. 2 Present address: Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada. 1

0008-8749/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cellimm.2012.06.002

observation from these studies is that the TLR costimulation in naïve T cells may require preactivation through their TCR as a prerequisite [10–12]. This notion is supported by findings that suggested memory T cells do not require TCR-stimulation as a prerequisite for responding to direct TLR ligation [7,10,13]. The expression and function of TLRs in iNKT cells are poorly understood. Shimizu et al. (2002) reported the expression of TLR2, 4, 5 and 9 genes in purified (97%) murine liver NK1.1+TCRb+ NKT cells [14,15]. Subsequently, TLR3 and 4 protein expression in hepatic CD1d tetramer positive iNKT cells was reported [16,17]. More recently, Moreno et al. (2009) reported expression of TLR1 to 9, except for TLR8, at the transcript level in purified (93–100%) Va24+Vb11+ human iNKT cells [18]. Previous research has investigated the ‘indirect’ effects of TLR primed APC-derived signals in murine and human iNKT cell activation [18–20]. For example, Nagarajan et al. (2007) showed increased activation of iNKT cells in response to LPS which was dependant on Dendritic Cells (DC)derived IL-12 and IL-18 [19]. Tyznik et al. (2008) showed that coculturing of TLR7/9 primed DCs with splenic iNKT cells can result in enhanced IFN-c production by iNKT cells. These studies along with a few others [18,21,22] collectively indicated the necessity of APC in transmitting TLR-derived signals to iNKT cells leading to enhanced iNKT cell activation. However, despite the expression of certain TLRs in iNKT cells, there is little evidence for the direct TLR-mediated stimulatory or costimulatory effects in the activation of iNKT cells.

34

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

In the present study, we used iNKT cell lines primarily to avoid potential contamination of other cells, although present in few numbers, particularly APC that have abundant TLR expression. We investigated the expression of TLR genes at both transcript and protein levels in murine iNKT cell lines, in the presence or absence of TCR ligation by either a-GalCer or anti-CD3 antibodies. Further, TLR costimulatory effects were investigated in iNKT cells that were stimulated through their TCR. Finally, we examined the role of iNKT cells costimulated using TLR ligands in the maturation of DCs. 2. Materials and methods 2.1. Cell lines Va14-Ja18+ mouse iNKT cell line DN32.D3 [23] was generously provided by Dr. Albert Bendelac (The University of Chicago, Chicago, IL). The rationale for using iNKT cell lines was to exclude the possibility of contamination with other cells, such as APCs, which are known to have abundant expression of a wide range of TLRs and could interfere with our objective of investigating direct effects of TLR ligands on iNKT cells. DN32.D3 cells can be activated with a-GalCer in the absence of APC or FcR-expressing accessory cells and we have recently shown that DN32.D3 cells express their own CD1d, which enables cross-presentation of a-GalCer leading to cellular activation [24]. N38-2C12 mouse iNKT cell line [25,26] was a kind gift from Dr. Kyoko Hayakawa (Fox Chase Cancer Center, Philadelphia, PA). These cells, unlike DN32.D3 cells, express very low CD1d molecules and require exogenous CD1d protein coated to plates for presentation of a-GalCer [26]. DC2.4 cells, an immature dendritic cell line [27] was provided by Dr. Kenneth Rock, University of Massachusetts, referred to, hereafter, as DCs. Cells were grown in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum, nonessential amino acids, 2 mM L-glutamine, 1 mM sodium pyruvate, 100 U/mL penicillin, 100 lg/mL streptomycin, and 50 lM 2-mercaptoethanol, which will, hereafter, be referred to as complete medium. All cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO2. 2.2. In vitro iNKT cell stimulation with TCR and TLR agonists To determine if TCR stimulation of iNKT cells can lead to expression of TLR genes, cells were seeded at 0.5–1.0  105 cells/well of microtiter plates and stimulated with either a-GalCer (KRN7000, Cederlane Laboratories, ON, Canada) or plate-bound anti-CD3 antibodies (eBioscience, San Diego, CA) for 24 h. Two concentrations (high and low) of these TCR agonists were used based on their ability to induce higher and lower levels of activation status in iNKT cells, as determined by production of IL-2 or IL-4 cytokines in a pilot dose titration experiment. The concentrations for a-GalCer were 500 ng/ml (high) and 50 ng/ml (low) while anti-CD3 antibodies were used at 100 ng/ml (high) and 10 ng/ml (low) concentrations. In addition, iNKT cells were also stimulated with TLR agonists alone (without TCR ligation) to examine the TCR-independent activatory effects of TLR agonists. Following stimulation of iNKT cells with either TCR agonists or TLR agonists, cells were collected for RNA extraction at 2, 6 and 24 h post-stimulation. The unstimulated control cells received only the diluent/vehicle used for dissolving the TCR or TLR agonists. RNA samples were further processed as described below. The following TLR agonists were used either alone or in combination or following preactivation by TCR agonists to assess their direct stimulatory or costimulatory effects. Various concentrations for each of the TLR agonists were used in a pilot dose-range finding experiment to choose an optimal concentration which is given in the brackets:

peptidoglycan [TLR2 ligand (5 lg/ml)] and LPS [TLR4 ligand (1 lg/ ml)] were purchased from Sigma, St Louis, MO. Poly I:C [TLR3 ligand (25 lg/ml)], flagellin [TLR5 ligand (1 lg/ml)], Imiquimod [TLR7 ligand (1 lg/ml)], type C CpG ODN [T:R9 ligand]-2395 and control ODN-2395 (10 lg/ml) and type B CpG ODN-1668 and control ODN-1668 (10 lg/ml) were purchased from Invivogen, San Diego, CA. For experiments aimed at evaluating the costimulatory effects of TLR agonists in TCR preactivated iNKT cells, cells were stimulated with TCR agonists for 24 h and washed with complete medium followed by stimulation with TLR agonists. Similar experiments were also carried out to evaluate the costimulatory effects of TLR agonists in iNKT cells pre-treated with murine IL12 (25, 150 and 500 U/ml) or IFN-a (100, 500 and 1000 U/ml) cytokines (eBioscience, San Diego, CA). In some experiments, we also used N38-2C12 iNKT cells to verify if a-GalCer derived signals can provide further TLR-mediated costimulation in these cells. These experiments required CD1d protein pre-coated to plates prior to addition of a-GalCer since these cells, unlike DN32.D3 cells, have very low CD1d expression [26]. Recombinant mouse CD1d1/Fc chimera purchased from R&D Systems was used to coat the plates at 1.5–2 lg/ml followed by addition of a-GalCer (50 ng/ ml) in these experiments. In order to assess the ability of iNKT cells in inducing DC maturation, first iNKT cells were activated with anti-CD3 antibodies (10 or 100 ng/ml) and were subsequently treated with CpG-ODN or non-CpG control ODN, as described above. The TLR-costimulated iNKT cells or their culture supernatants were then added or co-cultured with DCs (5  104/well) for 24 h and the cells were harvested for flow cytometry analysis as well as the culture supernatants were collected for cytokine measurements. The supernatants obtained from the TLR-costimulated iNKT cells were added to the DC cultures in the ratios of 1:1, 1:2, 1:5 and 1:10. 2.3. RNA extraction and cDNA synthesis Total RNA was extracted from iNKT cells stimulated with either

a-GalCer or anti-CD3 antibodies collected at 2, 6 and 24 h poststimulation using TRIzolÒ reagent (Invitrogen, Carlsbad, CA). RNA quality and quantity were estimated spectrophotometrically. Five hundred ng to 1 lg of RNA were reverse-transcribed into cDNA using the SuperScriptÒ First-Strand kit (Invitrogen) using oligo-dT primers and according to the manufacturer’s protocol. Spleen tissue from a C57BL/6 mouse was collected and total RNA was extracted followed by cDNA synthesis for use as positive control for TLR genes expression. 2.4. Preparation of standard curves using plasmid constructs In order to determine the relative gene expression, standard curves were generated for each of the TLR genes, using plasmid constructs containing cloned TLR gene fragments. Briefly, genespecific primers were used to amplify a segment of the target TLR gene with conventional PCR. The PCR amplicon sizes are given in Table 1. The resultant PCR products were then purified and cloned into the pDrive cloning vector (QiagenÒ PCR Cloning Kit, Qiagen Inc., Ontario). Plasmid constructs were then purified from recombinant Escherichia coli DH5a bacterial cells and 10-fold serial dilutions (101–109) were prepared. These plasmid dilutions were analyzed in triplicates using real-time PCR assay to obtain the standard curve. 2.5. Real-time RT- PCR Quantitative real-time PCR (qRT-PCR) was performed using the LightCyclerÒ 480 II (Roche Diagnostics GmbH, Mannheim,

35

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43 Table 1 Primers used in the real-time PCR analysis for quantification of TLR gene expression in iNKT cells. Gene/NCBI accession #

Primer sequence (50 –30 )

Amplicon size (bp)

Annealing temp. (°C)

b-actin/NM_007393.3

F- CCACACCCGCCACCAGTTCG R- ACAGCCCGGGGAGCATCGTC F- GGACTCCTAGGCTCCGGGCA R- TCCGCACCTCCTTGAACGACA F- CGGGGGTCCAACTGGAGAACCT R- GTGGGGGTTCAGTTGGGCGT F- GCCTCCCTGGCTCCTGGCTA R- GGGACTTTGCTGAGTTTCTGATCC F- CGGTCCCGCCAGCCATTTCA R- ATGAGCTGCAGCGGCAAGGG F-GCCTTCAAGAAAGATGTCCTTGGCTCC R- TCCGTGTCCACATCGAAAACACCA F- CTCTTGGCTGTGGCCGTGGG R- GGAGAGTTTGGGCGCTGCGT

111

64

130

60

212

64

138

60

102

64

134

64

136

64

TLR2/NM_011905.3 TLR3/NM_126166.4 TLR4/NM_021297.2 TLR5/NM_016928.2 TLR7/NM_133211.3 TLR9/NM_031178.2

Germany) on 384-well or 96-well plates in a final 20 lL reaction volume. Primers for b-actin, TLR2, 3, 4, 5, 7 and 9 were synthesized by Sigma–Aldrich Canada, Oakville, ON and their sequences are given in Table 1. Between 0.25–0.5 lM of each of the gene-specific primers and 5 ll of 1:10 dilution of cDNA as template and PCR grade water were used in the reaction. The cycling conditions included an initial heat-denaturing step at 95 °C for 10 min, 55 cycles at 95 °C for 10 s, annealing as described in Table 1 for each of the primers, and product elongation and signal acquisition (single mode) at 72 °C for 10 s. Following amplification, the melting curves were determined in a three-segment cycle of 97 °C for 0 s, 60 °C for 15 s, and 95 °C for 0 s in the continuous acquisition mode and the temperature transition rates were set at 0.11 °C/s. 2.6. Cytokine measurement IL-2 production by iNKT cells was considered as the readout assay for iNKT cell activation in the majority of the experiments carried out in the present study as reported before [23,24]. In addition, IL-4 and IFN-c production by iNKT cells was measured, wherever specified. As a means of measuring DC maturation, IL-6, TNF-a and IL-12 production by DCs was also measured. The cytokine concentration in triplicate culture supernatants in each independent experiment was determined using mouse IL-2, IL-4, IFN-c, IL-6, TNF-a or IL-12 sandwich ELISA kits (eBioscience, San Diego, CA) following manufacturer’s instructions. Briefly, 100 ll of capture antibody was coated overnight at 4 °C. The following day, after blocking the wells, the cytokine standard or 100 ll of the culture supernatants (diluted wherever required) were added as antigens and incubated for 2 h at room temperature. A detection antibody (biotinylated anti-mouse cytokine antibody, 100 ll) was added to the wells and incubated at room temperature for 1 h followed by avidin-HRP enzyme for 30 min. One hundred microlitre of 1X TMB solution were added to each well and incubated for 15 min for color development. The reaction was stopped by adding 50 ll of 2 N sulfuric acid to each well and the plates were read at an optical density of 450 nm. 2.7. Flow cytometry analyses To detect intracellular TLR3 and 9 proteins expression in TCRactivated iNKT cells, DN32.D3 cells were stimulated with platebound anti-CD3 antibodies for 24 h and cells were stained with anti-TLR antibodies along with appropriate isotype controls for flow cytometry analysis. For intracellular staining of TLR3 protein, a previously reported protocol was used [28]. Briefly, cells were initially treated with anti-CD16/32 antibodies (e-Bioscience, San Diego, CA) to block FccR and fixed using BD Cytofix/Cytoperm solu-

tion (BD Biosciences). Cells were then permeabilized with permeabilization buffer (e-Bioscience) followed by staining with primary antibody, rat anti-mouse TLR3 antibody (clone 313129 from R&D systems) or rat IgG2a isotype control antibody (e-Bioscience) for 1 h at room temperature (RT). All samples were washed twice in permeabilization buffer and then stained with PE-goat anti-rat IgG (e-Bioscience) for 1 h at RT. For intracellular staining of TLR9, after initial blocking of FccR, cells were fixed using BD Cytofix/Cytoperm solution and incubated with FITC-labeled antimouse TLR9 antibody (clone M9.D6 from e-Bioscience) or isotype control antibody for 30 min at RT. After a final washing with FACS buffer (1% bovine serum albumin in PBS), TLR3, 9 or isotype antibody stained cells were re-suspended in 4% PFA solution and BD FACScan flow cytometer was used for data acquisition. Data analysis was carried out using FlowJo software (Tree Star, Ashland, OR). To determine if the TLR-costimulated DN32.D3 iNKT cells induce DC maturation, the surface expression of the costimulatory maturation markers, CD40, CD80, CD86 and MHC-II was analyzed by Flow Cytometry. Subsequent to treatment with iNKT cell culture supernatants, DCs were harvested and stained with PE-conjugated rat anti-CD40 mAb (clone 1C10), anti-CD86 mAb (clone GL1) or PElabeled rat IgG2a (isotype control) antibodies on ice for 30 min. To confirm the expression of CD80 and MHC-II by DCs, cells were stained with FITC-labeled anti-CD80 mAb (clone 16-10A1), antiMHC-II mAb (clone 14-4-4S) or FITC-labeled Armenian hamster IgG or mouse IgG2a (isotype control) antibodies. The above fluorochrome-labeled mAbs were all purchased from eBioscience. After staining, cells were washed with FACS buffer and fixed with 4% PFA before a BD FACScan flow cytometer was used for data acquisition. Data analysis was carried out using FlowJo software (Tree Star, Ashland, OR).

2.8. Data analysis In experiments that evaluated the relative TLR gene expression in iNKT cells, the PCR efficiency (E) of the plasmid standards was calculated using the formula: E = 10 1/slope. The ratio of the TLR gene expression, as target gene, was based on their relative expression versus the level of b-actin (house keeping gene) expression, as a reference, and was calculated by the formula; Ratio = (ET)DCp target[calibratorsample]/(ER)DC p reference[samplecalibrator]. In this formula, ET, ER, and DCp represented efficiency of target gene in the standard curve, efficiency of reference in the standard curve, and the difference between crossing points, respectively. The relative expression of all the TLR genes was calculated based on the expression b-actin using the Pfaffl’s formula [29] as we have described previously [30,31]. Experimental cell cultures were performed in triplicates and sets of triplicate data were compared

36

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

determined TLR expression in unstimulated as well as in a-GalCer or anti-CD3 antibody stimulated DN32.D3 iNKT cells. TLR3, 4, 5, 7 and 9 genes except TLR2 were found constitutively expressed in DN32.D3 iNKT cells, as detected by real-time PCR (data not shown). Therefore, iNKT cells were first stimulated with the corresponding TLR agonists alone (without TCR ligation) to determine the functional responsiveness of these TLRs compared to unstimulated controls. Two types of CpG ODNs (ODN-1668 and ODN-2395) were used in these experiments as ligands for TLR9. ODN-1668 belongs to type B CpG-ODNs that have been shown to activate mainly B cells, whereas. ODN-2395 belongs to type C CpG-ODNs, which combine features of both types A and B and have been shown to stimulate both DC and B cells [32]. In these experiments, cells treated with CpG ODN-2395 showed significant up-regulation of TLR9 gene transcription at 6 h post-stimulation compared to the cells treated with CpG ODN-2395 control (Fig. 1f). However, no changes in TLR expression in cells treated with CpG ODN-1668 were observed. Hence, CpG ODN-2395 was used in all the experiments described in subsequent sections of this manuscript and referred to hereafter as CpG-ODN. When used alone, no other TLR agonists were able to induce significant transcriptional changes in their corresponding TLR gene expression (data not shown). Quantitative analysis of transcriptional changes in TLR gene expression in iNKT cells stimulated with either a-GalCer or anti-

between the control and the treated groups using the unpaired two-tailed Student’s t-test. Differences in relative TLR gene expression between treatment and control culture groups were considered significant at p 6 0.05. The data were expressed as the mean ± SEM. The abundance of cytokines in the triplicate culture supernatants of iNKT cells was calculated using standard curves generated for each cytokine. The data are expressed as the mean cytokine concentration + SD in triplicate cultures of each independent experiment. These experiments were repeated at least three times independently and the results were consistently reproducible. The results shown represent one independent experiment. Statistical comparisons were performed using the two-tailed t-test and differences were considered statistically significant at p-values <0.05.

3. Results 3.1. Expression of TLR genes in DN32.D3 iNKT cells Although many cell types including various T cell subsets, such as conventional ab, cd, and regulatory T cells are known to express a diverse array of TLRs, either constitutively or post-activation, expression of TLR genes in iNKT cells is poorly understood. We

a

α-GC V

6E-05 relative expression to β-actin

b

TLR3

7E-05

*

c

TLR4

300 250

0.01

200

0.008

150

0.006

100

0.004

1E-05

50

0.002

0E+00

0

α-GC 50 5E-05

α-GC 500

TLR5

0.012

4E-05

*

*

2

6

3E-05 2E-05

2

6

0

2

24

6

24

24

hours post-stimulation

d

relative expression to β-actin

0.1

e

TLR7

0.12

*

f

TLR9

0.6 0.5

*

0.08

0.4

0.06

0.3

0.04

0.2

0.02

0.1

0

*

0

2

6

24

2

6

24

hours post-stimulation Fig. 1. TLR expression in DN32.D3 iNKT cells activated with a-GalCer or CpG-ODN. iNKT cells were stimulated with a-GalCer at 50 ng/ml (a-GC 50), 500 ng/ml (a-GC 500) or with vehicle-only control (a-GC V) for a period of 24 h (panels a–e). In panel f, iNKT cells were stimulated with CpG ODN-2395 as well as the control CpG ODN-2395 at 5 (CpG-5) and 10 lg/ml (CpG-10) for a period of 24 h. Cells were collected at 2, 6 and 24 h post-stimulation and total RNA was extracted followed by cDNA synthesis. Quantitative real-time PCR (qRT–PCR) was performed to determine the expression of target TLR gene in these cells. The results are shown as expression of the TLR gene relative to the housekeeping gene, b-actin expression. Experimental cell cultures were performed in triplicates and sets of triplicate data were compared between the groups using the t-test. Differences in relative TLR gene expression between treatment and control culture groups were considered significant at p 6 0.05. The data are shown as the mean relative expression of target TLR gene ± SEM in triplicate cultures.

37

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

CD3 antibodies was performed and compared to the transcript levels in vehicle-treated cells to determine if TCR signaling can lead to transcriptional up-regulation of TLR genes. As shown in Fig. 1a–e, a-GalCer (50 ng/ml) stimulation induced significant up-regulation of TLR5, 7 and 9 transcripts as early as 2 h post-stimulation compared to controls. In addition, TLR3 and 5 transcripts were significantly up-regulated at 24 and 6 h post-a-GalCer (50 ng/ml) stimulation, respectively. Significant TLR7 transcriptional up-regulation was also found in cells treated with a-GalCer (500 ng/ml) at 24 h post-stimulation. TLR4 gene expression was observed only in the a-GalCer stimulated cells; however, no detectable expression of TLR2 gene was observed in the a-GalCer stimulated cells (data not shown). In addition to a-GalCer, we used anti-CD3 antibodies primarily to bypass the requirement for CD1d-mediated glycolipid presentation and its potential contribution to TLR signaling in DN32.D3 iNKT cells. Analysis of iNKT cells stimulated with anti-CD3 antibodies at both high (100 ng/ml) and low (10 ng/ml) concentrations showed significant up-regulation in TLR3 transcripts at 6 h poststimulation (Fig. 2a), whereas TLR9 gene expression was significantly up-regulated at 24 h post-stimulation (Fig. 2d). In addition, anti-CD3 (100 ng/ml) activation of iNKT cell TCR also induced significantly higher levels of TLR3 transcription at 24 h post-stimulation. However, no changes in the expression of other TLR genes were observed (Fig. 2 and data not shown). Similar to a-GalCer

a

b

TLR3 6E-05

stimulated cells, no detectable expression of TLR2 gene was observed in the anti-CD3 stimulated iNKT cells. 3.2. Expression of TLR proteins in iNKT cells Further, to validate the results of TLR expression at the transcript level, protein expression of two TLRs was examined. Intracellular expression of TLR3 and 9 in DN32.D3 iNKT cells stimulated with plate-bound anti-CD3 antibodies (100 ng/ml) for 24 h was determined by flow cytometry analysis. As shown in Fig. 2e, expression of TLR3 in DN32.D3 cells was found to be constitutive. Importantly, TCR signaling in these cells via the use of anti-CD3 antibodies resulted in a marked increase in their expression of TLR3, as determined by the percentage of positively stained cells and mean fluorescence intensity (MFI). Although TLR9 was not found constitutively expressed in DN32.D3 cells, an increase in TLR9 expression was evident in these cells upon TCR-mediated activation compared to unstimulated controls (Fig. 2f). 3.3. Costimulatory effects of TLR agonists in TCR preactivated iNKT cells The direct stimulatory effect of various TLR agonists on iNKT cell activation in the absence of TCR-mediated preactivation was measured by their production of IL-2. Although, flagellin and CpG-ODN

TLR5

c

5E-06

PBS 4E-06

anti-CD3 10 4E-05

anti-CD3 100

3E-06 2E-06

2E-05

1E-06 0E+00

0E+00

2

d

6

2

24

e

TLR9

6

24

f

5E-04

4E-04

3E-04

2E-04

1E-04

0E+00

2

6

24

Fig. 2. TLR expression in DN32.D3 iNKT cells activated with anti-CD3 antibodies. iNKT cells were stimulated with plate-bound anti-CD3 antibodies at 10 ng/ml (anti-CD3 10), 100 ng/ml (anti-CD3 100) or with PBS control (PBS) for a period of 24 h. Cells were collected at 2, 6 and 24 h post-stimulation and total RNA was extracted followed by cDNA synthesis. Quantitative real-time PCR (qRT–PCR) was performed to determine the expression of target TLR gene in these cells. The results are shown in panels a–d, as expression of the TLR gene relative to the housekeeping gene, b-actin expression. Experimental cell cultures were performed in triplicates and sets of triplicate data were compared between the groups using the t-test. Differences in relative TLR gene expression between treatment and control culture groups were considered significant at p 6 0.05. The data are shown as the mean relative expression of target TLR gene ± SEM in triplicate cultures. Panels e and f show the intracellular protein expression of TLR3 and 9 in the DN32.D3 iNKT cells that were activated with anti-CD3 antibodies, respectively. The numbers in each panel indicate the % of positively stained cells in each group with MFI given in the parenthesis. Data shown here represent one of the three independent experiments.

38

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

stimulation in iNKT cells resulted in a slight increase in IL-2 production at 48 and 72 h post-stimulation compared to control-treated cells, the quantity of IL-2 produced by TLR alone-stimulated iNKT cells, in general, was found to be too low compared to TCR-stimulated iNKT cells (data not shown). Therefore, no definitive conclusions could be made from these experiments. Based on the above observations that indicated TCR signaling in iNKT cells can induce up-regulation of TLR transcripts and proteins led us to evaluate the effects of TLR-costimulation in a-GalCer or anti-CD3 antibody-mediated TCR preactivated iNKT cells in the absence of APC-derived signals. DN32.D3 iNKT cells preactivated with a-GalCer (50 ng/ml) showed significantly enhanced activation of iNKT cells to a subsequent costimulation by poly I:C at 48 h post-TLR stimulation as measured by increased levels of IL-2 and IL-4 compared to controls (Fig. 3a and b). Furthermore, poly I:C-mediated costimulatory activation was also evident in N382C12 iNKT cells preactivated with a-GalCer (50 ng/ml) (Fig. 5a). In sharp contrast, DN32.D3 iNKT cells preactivated with higher concentrations of a-GalCer (500 ng/ml) showed a significant down-regulation in cell activation upon stimulation with the TLR9 agonist, marked by a significant reduction in IL-2 and IL-4 production (Fig. 3c and d). These observations suggest a possible regulatory role of certain TLR agonists such as CpG-ODN in iNKT cells activated through their TCR. TCR-mediated preactivation of DN32.D3 iNKT cells by anti-CD3 antibodies at low or sub-optimal (10 ng/ml) and higher (100 ng/ ml) concentrations led to a subsequent enhancement of activation by some of the TLR agonists at 24 h post-TLR stimulation as measured by production of IL-2 and IL-4. As shown in Fig. 4a, poly

a 500.0

I:C, flagellin and CpG DNA induced a significant increase in IL-2 production by DN32.D3 iNKT cells preactivated with anti-CD3 antibodies at higher concentrations compared to cells that were preactivated with anti-CD3 antibodies with no subsequent TLRcostimulation. Similar TLR5- and TLR9-mediated costimulatory activation was also observed in N38-2C12 iNKT cells preactivated with anti-CD3 antibodies at higher concentrations (Fig. 5c). Furthermore, CpG DNA was found to retain its costimulatory ability in both DN32.D3 as well as N38-2C12 iNKT cells preactivated with low or suboptimal concentrations of anti-CD3 antibodies, as determined by a 2–4-fold increase in IL-2 production compared to controls (Figs. 4b and 5b). Additionally, flagellin and poly I:C treatment of N38-2C12 cells pre-activated with suboptimal concentrations of anti-CD3 antibodies resulted in the costimulatory activation of these cells (Fig. 5b). However, although not statistically significant, poly I:C was also found to show a trend of increase in IL-2 production of DN32.D3 iNKT cells preactivated with anti-CD3 antibodies at these concentrations (Fig. 4b). Peptidoglycan induced a significant reduction in IL-2 and IL-4 production by DN32.D3 iNKT cells preactivated with low concentrations of anti-CD3 antibodies (Fig. 4b and d). In addition, significantly reduced IL-4 and IFN-c production by peptidoglycan was also observed in cells preactivated with the higher concentrations of anti-CD3 antibodies (Fig. 4c and e). Although CD3 ligation in DN32.D3 iNKT cells by anti-CD3 antibodies (100 ng/ml) resulted in production of IFN-c unlike in cells activated with a-GalCer (500 ng/ml), none of the TLR agonists showed any increase in the IFN-c production upon subsequent costimulation (Fig. 4e and data not shown).

b 160.0

*

* 120.0

IL-4 pg/ml

IL-2 pg/ml

400.0

300.0

200.0

40.0

100.0

0.0

c

α-GC 50

PG

pIC

LPS

Fla

Imq

0.0

CpG CpGCx

d

4.0

α-GC 50

PG

pIC

LPS

Fla

Imq

CpG CpGCx

500.0

400.0

2.0

* 1.0

IL-4 pg/ml

3.0

IL-2 ng/ml

80.0

300.0

* 200.0

100.0

0.0

0.0

α-GC 500

PG

pIC

LPS

Fla

Imq

CpG CpGCx

α-GC 500

PG

pIC

LPS

Fla

Imq

CpG CpGCx

Fig. 3. TLR-mediated costimulatory activation of DN32.D3 iNKT cells preactivated with a-GalCer. iNKT cells were activated with a-GalCer (50 ng/ml in panels a and b; 500 ng/ml in panels c and d) for 24 h and the cells were washed three times with RPMI complete medium. Different TLR agonists were added to the culture for 48 h to determine the TLR costimulatory effects on iNKT cell activation. iNKT cell activation was determined by production of IL-2 (panels a and c) and IL-4 (panels b and d) in the culture supernatants by ELISA at 48 h post-TLR stimulation. Statistical significance was determined between the treatment and control groups (⁄p 6 0.05). These experiments were repeated three times independently and the results were found consistently reproducible. The representative data from one independent experiment are shown here as mean IL-2 or IL-4 concentration + SD in triplicate cultures. aGC- a-GalactosylCeremide, PG- peptidoglycan, pIC- poly I:C, LPS- Lipopolysaccharide, Fla- Flagellin and ImqImiquimod.

39

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

a

30

6

*

*

IL-2 ng/mL

20 15 10

4 3 2 1

5

*

0

0

CD3-100

PG

pIC

LPS

Fla

Imq

CpG

CD3-10

CpGCx

d 400

1200

300

IL-4 pg/mL

c 1600 IL-4 pg/mL

*

5

25

IL-2 ng/mL

b

*

800

*

PG

pIC

LPS

Fla

Imq

CpG

CpGCx

pIC

LPS

Fla

Imq

CpG

CpGCx

200

400

100

*

0

0

CD3-100

PG

pIC

LPS

Fla

Imq

CpG

CpGCx

pIC

LPS

Fla

Imq

CpG

CpGCx

CD3-10

PG

e 1200 IFN-γpg/mL

900

*

600

300

0

CD3-100

PG

Fig. 4. TLR-mediated costimulatory activation of DN32.D3 iNKT cells preactivated with anti-CD3 antibodies. iNKT cells were activated with plate bound anti-CD3 antibodies (100 ng/ml in panels a, c and e; 10 ng/ml in panels b and d) for 24 h and the cells were washed three times with RPMI complete medium. Different TLR agonists were added to the culture for 24 h to determine the TLR costimulatory effects on iNKT cell activation. iNKT cell activation was determined by production of IL-2 (panels a and b), IL-4 (panels c and d) and IFN-c (panel e) in the culture supernatants by ELISA at 24 h post-TLR stimulation. Statistical significance was determined between the treatment and control groups (⁄p 6 0.05). These experiments were repeated three times independently and the results were found consistently reproducible. The representative data from one independent experiment are shown here as mean IL-2, IL-4 or IFN-c concentration + SD in triplicate cultures. CD3 refers to anti-CD3 antibodies, PG- peptidoglycan, pIC- poly I:C, LPS- Lipopolysaccharide, Fla- Flagellin, Imq- Imiquimod and CpGCx- CpG-ODN control.

a

b 400

600 500

*

c

*

*

5 *

4

300

300

ng/ml

pg/ml

pg/ml

400

*

200

3

*

2 200 100

1

100

0

0 aGC-50

poly I:C

0 CD3-10

poly I:C

Fla

CpG

CpGCx

CD3-100 poly I:C

Fla

CpG

CpGCx

Fig. 5. TLR-mediated costimulatory activation of N38-2C12 iNKT cells preactivated with a-GalCer or anti-CD3 antibodies. N38-2C12 iNKT cells were activated with a-GalCer (50 ng/ml in panel a) or plate bound anti-CD3 antibodies (10 ng/ml in panel b; 100 ng/ml in panel c) for 24 h and the cells were washed three times with RPMI complete medium. TLR agonists were added to the culture for 24 (in panels b and c) or 48 h (in panel a) to determine the TLR costimulatory effects on iNKT cell activation. iNKT cell activation was determined by production of IL-2 in the culture supernatants by ELISA. Statistical significance was determined between the treatment and control groups (⁄p 6 0.05). Data shown here represent one of the three independent experiments as mean IL-2 concentration + SD in triplicate cultures. aGC refers to a-GalCer, CD3 refers to anti-CD3 antibodies, pIC- poly I:C, Fla- Flagellin and CpGCx- CpG-ODN control.

40

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

We further evaluated if pre-treatment of DN32.D3 iNKT cells with cytokines (IL-12 or IFN-a) can induce subsequent costimulatory activation of these cells by TLR ligands. However, none of the TLR agonists had a stimulatory effect on iNKT cells pre-treated with any of these cytokines (data not shown). 3.4. DC maturation by TLR-costimulated iNKT cells We further determined if the TLR costimulation of iNKT cells preactivated with TCR agonists can influence their ability to mature DC. As measures of DC maturation, we examined surface expression of CD80, CD86, MHC-II, CD40 as well as production of cytokines by DC 2.4 cells. As shown in Fig. 6a–d, the unstimulated DCs constitutively expressed CD40 and CD80 molecules, which is in agreement with a previous report [33]. These cells when cultured in the presence of culture supernatants (at the ratio of 1:5) derived from iNKT cells preactivated with anti-CD3 antibodies (10 or 100 ng/ml) showed a marked increase in their surface expression of CD40 and CD80 molecules, as determined by the per-

centage of positively stained cells and MFI. However, no changes in the DC expression of either CD86 or MHC-II were observed when iNKT cell-derived supernatants were added to DC cultures (data not shown). More importantly, DCs cultured in the presence of supernatants derived from anti-CD3 preactivated plus CpG-costimulated iNKT cells displayed higher surface expression of CD40 and CD80 molecules compared to DC cultured in the presence of anti-CD3 preactivated iNKT cell-derived culture supernatants. This observation indicated that TLR-mediated costimulation in iNKT cells preactivated through their TCR has additive effects on the maturation of DC. In order to determine if the low amounts CpG present in the supernatants added to DCs might have contributed to maturation of DCs, one of the control groups in this experiment consisted of iNKT cells that were not CD3-stimulated but received CpG. However, after addition of these supernatants to DCs, the expression of costimulatory molecules by DCs did not change compared to the control group. This may be either because the final concentration of CpG in the DC culture was too low (1.5–2 ug/

Fig. 6. DC maturation by TLR-costimulated iNKT cells. Dendritic cells were cultured in the presence of culture supernatants derived from iNKT cells stimulated with anti-CD3 or anti-CD3 + CpG ODN. Dendritic Cells were collected at 24 h post-CpG ODN or control CpG ODN stimulation and analyzed by flow cytometry to determine the surface expression of CD40 and CD80 costimulatory molecules. Panel a and b show the expression of CD40 expression on DC cultured with supernatants from iNKT cells preactivated with anti-CD3 antibodies at 10 and 100 ng/ml, respectively. Panel c and d show the expression of CD80 expression in DCs cultured with supernatants from iNKT cells preactivated with anti-CD3 antibodies at 10 and 100 ng/ml, respectively. The numbers in each panel indicate the % of positively stained cells in each group with MFI given in the parenthesis. Production of IL-6 and TNF-a by DCs cultured in presence of CD3 + CpG or control CpG-ODN costimulated iNKT cells is shown in Panel e. Data shown here represent one of the three independent experiments.

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

ml) or the CpG DNA in culture might have been partly degraded after 48 h in culture before being added to DCs. Furthermore, DC cultured in the presence of culture supernatants derived from CD3-preactivated or CD3-preactivated plus TLR9-costimulated iNKT cells produced significantly higher amounts of IL-6 and tumor necrosis factor (TNF)-a (Fig. 6e), although no change was observed in the production of IL-12 by these cells (data not shown). Altogether, these results suggest the possible role of TLR-costimulated iNKT cells in DC maturation.

4. Discussion Growing evidence in recent years has demonstrated that in some species, including rodents, humans and cattle, TLRs are expressed by different T cell subsets [6]. However, to date, there have been only a few reports describing the expression of TLRs in iNKT cells [15–18,34]. Despite the expression of these TLRs by iNKT cells, possible direct costimulatory effects of TLR agonists in these cells have not been adequately addressed. Therefore, using two iNKT cell lines (DN32.D3 and N38-2C12 cells), the present study demonstrated the expression and costimulatory functions of certain TLRs such as that of TLR3, 5 and 9 in iNKT cells following the activation of these cells through their TCR. These observations suggested that some of the TLRs in iNKT cells are functional, thereby proposing a new mode of iNKT cell activation induced by pathogen-associated molecular patterns, in the absence of APC. The iNKT cell lines used in the present study have been extensively used as a useful model to study iNKT cell responses and these are murine iNKT cells with typical Va14-Ja18 rearranged TCRa [23,25,26]. Since these cell lines provide a pure iNKT cell population, we chose to address TLR expression and function in these cells primarily to avoid potential contamination of other cells although present in few numbers, particularly APC that have abundant TLR expression. Our initial experiments showed that treatment of DN32.D3 iNKT cells with CpG-ODN resulted in significant up-regulation of TLR9 gene transcription, thus raising the possibility for existence of a self-amplification loop for TLR expression and function in iNKT cells. This observation can be supported by the recent findings of Gardner et al. (2010) who showed that poly I:C administration in mice can result in increased TLR3 expression in hepatic Va14iNKT cells [16]. Although most TLR genes in iNKT cells were found constitutively expressed, none of the TLR agonists, when used alone, was able to induce optimal iNKT cell activation. This observation is in agreement with a recent study that despite expression of most TLR genes in human iNKT cells, activation of iNKT cells by TLR ligands required DCs to mediate this effect [18]. Therefore, we hypothesized that responsiveness of iNKT cells to direct stimulation by TLR agonists may require certain ‘pre-requisite’ signals. To test this hyopothesis, we used a-GalCer or anti-CD3 antibodies to induce TCR signaling in iNKT cells. The use of anti-CD3 antibody was to by-pass CD1d-mediated TCR signaling in iNKT cells. In addition, these TCR agonists were used at two concentrations (high and low) to evaluate the relative influence of varying intensity of TCRstimulation on TLR expression and function in iNKT cells. Indeed, TCR signaling in DN32.D3 iNKT cells was shown to induce increased TLR expression at both the transcript and protein levels. Notably, a-GalCer or anti-CD3 antibody stimulation at low (or suboptimal) concentration not only induced increased TLR expression in DN32.D3 cells but also enhanced their subsequent costimulatory activation by TLR3 and 9 agonists. Similar TLR3, 5 and 9-mediated costimulatory activation was also observed in N38-2C12 cells preactivated with low concentrations of anti-CD3 antibodies or a-GalCer. These responses indicated that perhaps, weak TCR signals received by iNKT cells may induce protein expression of pre-made

41

receptors to receive microbial signals, leading to an optimal or sufficient activation of these cells. In agreement, Gelman et al. (2006) also found TLR9-mediated costimulatory effects in murine CD4 + T cells preactivated with suboptimal concentrations (10 ng/ml) of anti-CD3 antibody [35]. Collectively, it can be inferred that TLR3, 5 or 9 costimulatory signals can augment iNKT cell responses when combined with a suboptimal dose of a TCR agonist, thus reducing the TCR activation threshold as well as the need for costimulatory signals from APC. Additionally, we observed TLR3, 5 and 9-mediated enhanced costimulatory activation of DN32.D3 and N38-2C12 iNKT cells that were preactivated with a relatively higher concentration (100 ng/ ml) of anti-CD3 antibodies (Figs. 4 and 5). These observations possibly indicate the important role of TLR3, 5 and 9 receptors in the co-activation of TCR-primed iNKT cells. Similar TLR-dependant costimulatory activation of conventional T cells preactivated with anti-CD3 cross-linking has also been previously reported [7,9,13,36]. In contrast, DN32.D3 iNKT cells preactivated with higher concentration (500 ng/ml) of a-GalCer induced significant down-regulation of cell activation following treatment with CpGODN. Interestingly, peptidoglycan treated anti-CD3 preactivated iNKT cells also displayed down-regulated activities. Although we could not detect TLR2 expression in iNKT cells, it is possible that receptors other than TLR2 such as Nod2 proteins may be involved in recognition of peptidoglycan in iNKT cells [37]. Similar findings were previously observed by Caron et al. (2005) who found downregulation of human CD4 + T cell activation by Resiquimod, a TLR7/ 8 agonist [7]. Such contrasting differences in either conventional T cells or iNKT cells may arise from the subtle differences in their TLR signaling as well as costimulation pathways [38,39]. Based on the observations which suggested that TLR expression and function in iNKT cells depended on the choice of TCR agonist and the intensity of TCR cross-linking, it is likely that there are subtle differences in the TCR-TLR signaling cross-talk. Mechanistically, dose-dependant a-GalCer/TCR-mediated preactivation followed by TLR-mediated costimulatory effects in iNKT cells may depend on de novo CD1d biosynthesis and expression in iNKT cells or APC. This is because certain TLR agonists, such as poly I:C and Imiquimod have been shown to up-regulate CD1d expression in myeloid DCs, whereas agonists of TLR2 and 4 can significantly reduce CD1d synthesis and surface expression [21]. Although we did not address this possibility in the present investigation, studies are currently underway to dissect the mechanisms of TCR-TLR cross-talk, which should provide new insight into the regulation of iNKT cell activation in the context of microbial infections. The findings presented here primarily suggest that iNKT cells preactivated with low or suboptimal concentrations of TCR-agonists can lead to subsequent costimulation by poly I:C, flagellin and CpG-ODN. Therefore, it is reasonable to assume that TLR3, 5 and 9 may act as essential co-receptors for iNKT cells in sensing viral/bacterial nucleic acids, in the absence of APC-derived signals. It is noteworthy that certain viruses such as HIV-1 and LCMV can target iNKT cells for their infection and survival [40,41] in which case the virus sensing TLR3 and 9 receptors may provide a vital means of activation for these cells. In support of this notion, recent studies have reported poly I:C/TLR3-mediated activation of human ab T cells as well as murine hepatic iNKT cells [42]. However, despite significant up-regulation of certain TLR transcripts such as TLR5, 7 and 9 in a-GalCer stimulated iNKT cells at suboptimal concentrations, no TLR-mediated costimulatory activation was observed. These results are consistent with previous reports and may partly be due to the lack of protein expression for these TLRs or lack of their surface expression, which suggests that additional accessory molecules may be necessary for TLR signaling in these cells [17,42].

42

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

Another important facet of TLR costimulation in T cells is the functional ability of TLR-costimulated cells in providing signals that help in maturation of DCs. Mature DCs expressing enhanced levels of costimulatory molecules are known to be more effective at priming and augmenting T cell responses [33]. To this end, we observed that culture supernatants from iNKT cells that were preactivated by anti-CD3 antibodies followed by costimulation with CpG-ODN could induce a marked increase in the surface expression of CD40 and CD80 molecules on DCs compared to those that were treated with culture supernatants of iNKT cells that were stimulated via their TCR only (without TLR costimulation). This increase in CD40 and CD80 expression was also associated with a significant increase in production of IL-6 and TNF-a by DCs treated with supernatants of iNKT cells that were preactivated by anti-CD3 and subsequently treated with CpG-ODN. These findings raise the possibility that co-activation of iNKT cells through their TCR and TLR has functional implications for these cells, one of which is enhancement of the capacity of iNKT cells to induce DC maturation. In support of this suggestion, Moreno et al. (2009) have recently reported that human iNKT cells co-cultured with TLRprimed DCs can induce increased expression of costimulatory molecules (maturation markers) on DCs [18]. Furthermore, pulsing of TLR-primed DCs with a-GalCer prior to co-culturing them with iNKT cells resulted in IL-12 production indicating that TCR-reactivity in iNKT cells can effectively enhance DC maturation. However, lack of increased expression of MHC-II and CD86 as well as production of IL-12 by DCs in our study indicate the necessity of additional accessory signals such as engagement of costimulatory molecules like CD40 on DC. In summary, the present study reported augmented expression of certain TLRs in TCR-activated iNKT cells and the direct TLR costimulatory effects using TLR3, 5 and 9 agonists in enhancing activation of TCR preactivated iNKT cells. Our observations also suggested that TLR costimulation in iNKT cells, which had received TCR-derived signals, could provide maturation signals to DCs. Collectively, our findings demonstrate that TCR-dependant recognition of glycolipid antigens in iNKT cells can induce responsiveness for TLR signaling, thus suggesting a new possible mode of iNKT cell activation. Disclosure The authors have no financial conflicts of interest. Acknowledgments We thank Dr. S. Ole Odemuyi for help in the FACS data analysis. Dr. R.R. Kulkarni was a recipient of an Ontario Ministry of Research and Innovation post-doctoral fellowship. References [1] S. Sharif, G.A. Arreaza, P. Zucker, Q.S. Mi, T.L. Delovitch, Regulation of autoimmune disease by natural killer T cells, J. Mol. Med. 5 (2002) 290–300. [2] L. Wu, L. Van Kaer, Natural killer T cells and autoimmune disease, Curr. Mol. Med. 1 (2009) 4–14. [3] E. Tupin, Y. Kinjo, M. Kronenberg, The unique role of natural killer T cells in the response to microorganisms, Nat. Rev. Microbiol. 6 (2007) 405–417. [4] R.R. Kulkarni, S.M. Haeryfar, S. Sharif, The invariant NKT cell subset in anti-viral defenses: a dark horse in anti-influenza immunity?, J Leukoc. Biol. 4 (2010) 635–643. [5] D. Kabelitz, Expression and function of toll-like receptors in T lymphocytes, Curr. Opin. Immunol. 1 (2007) 39–45. [6] R. Kulkarni, S. Behboudi, S. Sharif, Insights into the role of toll-like receptors in modulation of T cell responses, Cell Tissue Res. 343 (2010) 141–152. [7] G. Caron, D. Duluc, I. Fremaux, P. Jeannin, C. David, H. Gascan, Y. Delneste, Direct stimulation of human T cells via TLR5 and TLR7/8: flagellin and R-848 up-regulate proliferation and IFN-gamma production by memory CD4+ T cells, J. Immunol. 3 (2005) 1551–1557.

[8] A. Cottalorda, C. Verschelde, A. Marcais, M. Tomkowiak, P. Musette, S. Uematsu, S. Akira, J. Marvel, N. Bonnefoy-Berard, TLR2 engagement on CD8 T cells lowers the threshold for optimal antigen-induced T cell activation, Eur. J. Immunol. 7 (2006) 1684–1693. [9] A.E. Gelman, J. Zhang, Y. Choi, L.A. Turka, Toll-like receptor ligands directly promote activated CD4+T cell survival, J. Immunol. 10 (2004) 6065–6073. [10] S.M. Lee, Y.D. Joo, S.K. Seo, Expression and function of TLR2 on CD4 versus CD8 T Cells, Immune Netw. 4 (2009) 127–132. [11] B.J. Marsland, C. Nembrini, K. Grun, R. Reissmann, M. Kurrer, C. Leipner, M. Kopf, TLR ligands act directly upon T cells to restore proliferation in the absence of protein kinase C-theta signaling and promote autoimmune myocarditis, J. Immunol. 6 (2007) 3466–3473. [12] R. Simone, A. Floriani, D. Saverino, Stimulation of Human CD4 T Lymphocytes via TLR3, TLR5 and TLR7/8 up-regulates expression of costimulatory and modulates proliferation, Open Microbiol. J. 3 (2009) 1–8. [13] J. Tabiasco, E. Devevre, N. Rufer, B. Salaun, J.C. Cerottini, D. Speiser, P. Romero, Human effector CD8+ T lymphocytes express TLR3 as a functional coreceptor, J. Immunol. 12 (2006) 8708–8713. [14] H. Tsujimoto, S. Ono, A. Matsumoto, T. Kawabata, M. Kinoshita, T. Majima, S. Hiraki, S. Seki, L.L. Moldawer, H. Mochizuki, A critical role of CpG motifs in a murine peritonitis model by their binding to highly expressed toll-like receptor-9 on liver NKT cells, J. Hepatol. 6 (2006) 836–843. [15] H. Shimizu, T. Matsuguchi, Y. Fukuda, I. Nakano, T. Hayakawa, O. Takeuchi, S. Akira, M. Umemura, T. Suda, Y. Yoshikai, Toll-like receptor 2 contributes to liver injury by Salmonella infection through Fas ligand expression on NKT cells in mice, Gastroenterology 4 (2002) 1265–1277. [16] T.R. Gardner, Q. Chen, Y. Jin, M.N. Ajuebor, Toll-like receptor 3 ligand dampens liver inflammation by stimulating Valpha 14 invariant natural killer T cells to negatively regulate gammadeltaT cells, Am. J. Pathol. 4 (2010) 1779–1789. [17] P.W. Askenase, A. Itakura, M.C. Leite-de-Moraes, M. Lisbonne, S. Roongapinun, D.R. Goldstein, M. Szczepanik, TLR-dependent IL-4 production by invariant Valpha14+Jalpha18+ NKT cells to initiate contact sensitivity in vivo, J. Immunol. 10 (2005) 6390–6401. [18] M. Moreno, B.M. Mol, S. von Mensdorff-Pouilly, R.H. Verheijen, E.C. de Jong, B.M. von Blomberg, A.J. van den Eertwegh, R.J. Scheper, H.J. Bontkes, Differential indirect activation of human invariant natural killer T cells by toll-like receptor agonists, Immunotherapy 4 (2009) 557–570. [19] N.A. Nagarajan, M. Kronenberg, Invariant NKT cells amplify the innate immune response to lipopolysaccharide, J. Immunol. 5 (2007) 2706–2713. [20] A.J. Tyznik, E. Tupin, N.A. Nagarajan, M.J. Her, C.A. Benedict, M. Kronenberg, Cutting edge: the mechanism of invariant NKT cell responses to viral danger signals, J. Immunol. 7 (2008) 4452–4456. [21] M.J. Raftery, F. Winau, T. Giese, S.H. Kaufmann, U.E. Schaible, G. Schonrich, Viral danger signals control CD1d de novo synthesis and NKT cell activation, Eur. J. Immunol. 3 (2008) 668–679. [22] Y. Suzuki, D. Wakita, K. Chamoto, Y. Narita, T. Tsuji, T. Takeshima, H. Gyobu, Y. Kawarada, S. Kondo, S. Akira, H. Katoh, H. Ikeda, T. Nishimura, Liposomeencapsulated CpG oligodeoxynucleotides as a potent adjuvant for inducing type 1 innate immunity, Cancer Res. 23 (2004) 8754–8760. [23] O. Lantz, A. Bendelac, An invariant T cell receptor alpha chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4-8- T cells in mice and humans, J. Exp. Med. 3 (1994) 1097–1106. [24] J.K. Stuart, S.P. Bisch, M. Leon-Ponte, J. Hayatsu, D.M. Mazzuca, S.M. Vareki, S.M. Haeryfar, Negative modulation of invariant natural killer T cell responses to glycolipid antigens by p38 MAP kinase, Int. Immunopharmacol. 9 (2010) 1068–1076. [25] M. Gui, J. Li, L.J. Wen, R.R. Hardy, K. Hayakawa, TCR beta chain influences but does not solely control autoreactivity of V alpha 14J281T cells, J. Immunol. 11 (2001) 6239–6246. [26] N. Burdin, L. Brossay, Y. Koezuka, S.T. Smiley, M.J. Grusby, M. Gui, M. Taniguchi, K. Hayakawa, M. Kronenberg, Selective ability of mouse CD1 to present glycolipids: alpha-galactosylceramide specifically stimulates V alpha 14+ NK T lymphocytes, J. Immunol. 7 (1998) 3271–3281. [27] Z. Shen, G. Reznikoff, G. Dranoff, K.L. Rock, Cloned dendritic cells can present exogenous antigens on both MHC class I and class II molecules, J. Immunol. 6 (1997) 2723–2730. [28] W.G. Cho, R.J. Albuquerque, M.E. Kleinman, V. Tarallo, A. Greco, M. Nozaki, M.G. Green, J.Z. Baffi, B.K. Ambati, M. De Falco, J.S. Alexander, A. Brunetti, S. De Falco, J. Ambati, Small interfering RNA-induced TLR3 activation inhibits blood and lymphatic vessel growth, Proc. Natl. Acad. Sci. USA 17 (2009) 7137–7142. [29] M.W. Pfaffl, A new mathematical model for relative quantification in real-time RT–PCR, Nucleic Acids Res. 9 (2001) e45. [30] P. Parvizi, L.R. Read, M.F. Abdul-Careem, A.J. Sarson, C. Lusty, M. Lambourne, N. Thanthrige-Don, S.C. Burgess, S. Sharif, Cytokine gene expression in splenic CD4+ and CD8+ T cell subsets of genetically resistant and susceptible chickens infected with Marek’s disease virus, Vet. Immunol. Immunopathol. 2–4 (2009) 209–217. [31] A.I. Villanueva, R.R. Kulkarni, S. Sharif, Synthetic double-stranded RNA oligonucleotides are immunostimulatory for chicken spleen cells, Dev. Comput. Immunol. 1 (2011) 28–34. [32] J.D. Marshall, K.L. Fearon, D. Higgins, E.M. Hessel, H. Kanzler, C. Abbate, P. Yee, J. Gregorio, T.D. Cruz, J.O. Lizcano, A. Zolotorev, H.M. McClure, K.M. Brasky, K.K. Murthy, R.L. Coffman, G.V. Nest, Superior activity of the type C class of ISS in vitro and in vivo across multiple species, DNA Cell Biol. 2 (2005) 63–72. [33] T. He, C. Tang, S. Xu, T. Moyana, J. Xiang, Interferon gamma stimulates cellular maturation of dendritic cell line DC2.4 leading to induction of efficient

R.R. Kulkarni et al. / Cellular Immunology 277 (2012) 33–43

[34]

[35]

[36]

[37]

cytotoxic T cell responses and antitumor immunity, Cell. Mol. Immunol. 2 (2007) 105–111. T. Hiromatsu, T. Matsuguchi, H. Shimizu, T. Yajima, H. Nishimura, T. Arai, Y. Nimura, Y. Yoshikai, NK T cells stimulated with a ligand for TLR2 at least partly contribute to liver injury caused by Escherichia coli infection in mice, Eur. J. Immunol. 9 (2003) 2511–2519. A.E. Gelman, D.F. LaRosa, J. Zhang, P.T. Walsh, Y. Choi, J.O. Sunyer, L.A. Turka, The adaptor molecule MyD88 activates PI-3 kinase signaling in CD4+ T cells and enables CpG oligodeoxynucleotide-mediated costimulation, Immunity 5 (2006) 783–793. M. McCarron, D.J. Reen, Activated human neonatal CD8+ T cells are subject to immunomodulation by direct TLR2 or TLR5 stimulation, J. Immunol. 1 (2009) 55–62. K.M. Davis, S. Nakamura, J.N. Weiser, Nod2 sensing of lysozyme-digested peptidoglycan promotes macrophage recruitment and clearance of S. pneumoniae colonization in mice, J. Clin. Invest. 121 (2011) 3666–3676.

43

[38] M.J. van den Heuvel, N. Garg, L. Van Kaer, S.M. Haeryfar, NKT cell costimulation: experimental progress and therapeutic promise, Trends Mol. Med. 17 (2010) 65–77. [39] M. Yamamoto, S. Akira, TIR domain-containing adaptors regulate TLR signaling pathways, Adv. Exp. Med. Biol. 560 (2005) 1–9. [40] J.A. Hobbs, S. Cho, T.J. Roberts, V. Sriram, J. Zhang, M. Xu, R.R. Brutkiewicz, Selective loss of natural killer T cells by apoptosis following infection with lymphocytic choriomeningitis virus, J. Virol. 22 (2001) 10746–10754. [41] A. Motsinger, D.W. Haas, A.K. Stanic, L. Van Kaer, S. Joyce, D. Unutmaz, CD1drestricted human natural killer T cells are highly susceptible to human immunodeficiency virus 1 infection, J. Exp. Med. 7 (2002) 869–879. [42] N. Funderburg, A.A. Luciano, W. Jiang, B. Rodriguez, S.F. Sieg, M.M. Lederman, Toll-like receptor ligands induce human T cell activation and death, a model for HIV pathogenesis, PLoS One 4 (2008) e1915.