Journal of Immunological Methods 382 (2012) 150–159

Contents lists available at SciVerse ScienceDirect

Journal of Immunological Methods

j ourna l homepage: www.e lsev ie r .com/ locate / j im

Research paper

Ex-vivo α-Galactosylceramide activation of NKT cells in humansand macaques

Caroline S. Fernandez, Garth Cameron, Dale I. Godfrey, Stephen J. Kent⁎Department of Microbiology and Immunology, University of Melbourne, Parkville, VIC 3010, Australia

a r t i c l e i n f o

⁎ Corresponding author.E-mail address: skent@unimelb.edu.au (S.J. Kent).

0022-1759/$ – see front matter © 2012 Elsevier B.V. Adoi:10.1016/j.jim.2012.05.019

a b s t r a c t

Article history:Received 3 February 2012Received in revised form 18 May 2012Accepted 30 May 2012Available online 7 June 2012

NKT cells are key mediators of antiviral and anticancer immunity. Experiments in mice havedemonstrated that activation of NKT cells in vivo induces the expression of multiple effectormolecules critical to successful immunity. Human clinical trials have shown similar responses,although in vivo activation of NKT cells in humans or primate models are far more limited innumber and scope. Measuring ex vivo activation of NKT cells by the CD1d-restricted glycolipidligand α-Galactosylceramide (α-GalCer) through cytokine expression profiles is a usefulmarker of NKT cell function, but for reasons that are unclear, this approach does not appear towork as well in humans and non-human primate macaque models in comparison to mice. Weperformed a series of experiments on human and macaque (Macaca nemestrina) fresh wholeblood samples to define optimal conditions to detect NKT cell cytokine (TNF, IFNγ, IL-2) anddegranulation marker (CD107a) expression by flow cytometry. We found that conditionspreviously described for mouse splenocyte NKT cell activation were suboptimal on human ormacaque blood NKT cells. In contrast, a 6 h incubation with brefeldin A added for the last 4 h,in a 96-well plate based assay, and using an α-GalCer concentration of 1 μg/ml were optimalmethods to stimulate NKT cells in fresh blood from both humans and macaques. Unexpectedly,we noted that blood NKT cells from macaques infected with SIV were more readily activatedby α-GalCer than NKT cells from uninfected macaques, suggesting that SIV infection may haveprimed the NKT cells. In conclusion, we describe optimized methods for the ex vivo antigen-specific activation of human and macaque blood NKT cells. These assays should be useful inmonitoring NKT cells in disease and in immunotherapy studies.

© 2012 Elsevier B.V. All rights reserved.

Keywords:NKT cellsMacaquesα-GalactosylceramideOptimization of ex-vivo activationSIV infection

1. Introduction

NKT cells are lymphocytes of the innate immune systemthat are important in viral and antitumor immunity throughtheir ability to be rapidly activated and express a wide rangeof effector molecules (Lindqvist et al., 2009; Motohashi et al.,2009; Brigl and Brenner, 2010; Fujii et al., 2010). NKT cellsare activated via an antigen-specific T cell receptor (TCR).Improved methodologies to assess the activation of NKT cellsex vivo should assist dissecting the importance of theseimmune cells in humans and non-human primates.

ll rights reserved.

Type I or semi-invariant NKT cells express an invariant TCR-α chain (Vα14-Jα18 in mice, Vα24-Jα18 in humans) and arerestricted by the MHC-class Ib molecule, CD1d, expressed onAPCs presenting lipid based antigens such as glycolipidantigens (Exley et al., 1997). Lipid based antigens are presentedto NKT cells by direct ligation onto surface CD1d molecules orinternalized into endosomes where loading onto CD1d occursand subsequent membrane surface presentation on a lipid raft(Venkataswamy and Porcelli, 2010). NKT cells respond to abroad variety of lipid-based antigens including self and foreignglycolipid and phospholipid antigens (Venkataswamy andPorcelli, 2010).α-galactosylceramide (α-GalCer) is a syntheticglycosphingolipid derived from the marine sponge, Agelasmauritianus, and is commonly used in mice and human NKTstudies as a potent activator of NKT cells in vivo or in vitro

http://dx.doi.org/10.1016/j.jim.2012.05.019mailto:skent@unimelb.edu.auhttp://dx.doi.org/10.1016/j.jim.2012.05.019http://www.sciencedirect.com/science/journal/00221759

151C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

(Kawano et al., 1997). α-GalCer has immunomodulatoryeffects in cancer immunotherapy (Giaccone et al., 2002;Chang et al., 2005; Uchida et al., 2008; Kunii et al., 2009;Motohashi et al., 2009; Schneiders et al., 2011; Yamasaki et al.,2011; Vivier et al., 2012), autoimmunity (Novak and Lehuen;Hong et al., 2001; Wu and Van Kaer, 2009), bacterial (Brigl andBrenner, 2010; Emoto et al., 2010), and viral infections(Guillonneau et al., 2009; Lindqvist et al., 2009; Schneiderset al., 2011).

Upon activation with α-GalCer, NKT cells produce largeamounts of Th1, Th2 and Th17 cytokines such as IFNγ, TNF,IL-2, IL-4, IL-10, IL-13, IL-17, IL-21 and IL-22 (Godfrey et al.,2010). The cytokines produced by NKT cells trigger theactivation of other cells of the immune system such as NKcells, T and B cells and DCs (Fujii et al., 2003; Cerundolo et al.,2009). The detection of the expression of these cytokines byintracellular cytokine staining and flow cytometry permits ananalysis of NKT cell functional status.

Studies on α-GalCer activation of mice NKT cells ofteninvolve in vivo activation, where α-GalCer induced cytokineproduction by NKT cells, and indirectly downstream by NKcells, is measured in serum (Sullivan et al., 2010), or directlyex-vivo following α-GalCer administration in vivo (Wilson etal., 2003; Uldrich et al., 2005). In the latter case mice areadministered α-GalCer, organs such as spleen, liver or lymphnode are harvested a few hours after challenge and therelevant cells placed in culture with a protein transportinhibitor, without any further activation. Alternatively, theglobal mitogen-induced activation of NKT cells ex vivo isoften assessed by PMA/ionomycin stimulation. Such mito-genic stimulation, while providing information on the totalfunctional potential of NKT cells, may not reflect the in vivocapacity of NKT cells when interacting with antigen pre-sented by CD1d. Furthermore, activation with PMA/ionomy-cin poses difficulties in enumerating intracellular cytokineproduction from NKT cell subsets as the CD4 surface markeris downregulated upon mitogenic stimulation (O'Neil-Andersen and Lawrence, 2002). Many other studies haveassessed activation of mouse NKT cells ex vivo with α-GalCerin combination with protein transport inhibitors such asmonensin to detect intracellular cytokines produced in vitro(Uldrich et al., 2005; Patel et al., 2011).

It is important to translate advances in understandingNKT cell biology in mice towards more directly relevant non-human primate models and humans. For example, there isconsiderable interest in harnessing the antiviral activity ofNKT cells in the setting of chronic viral infections such as HIVinfection of humans or SIV infection of macaques (Fernandezet al., 2009; Snyder-Cappione et al., 2009). However, suchstudies require ex vivo/in vitro analysis of NKT cells, yet atpresent the literature is lacking in clearly defined optimizedmethods for acute ex-vivo activation of NKT cells withα-GalCer either in humans or macaques. The relative abilityof fresh blood NKT cells to respond to acute ex vivo activationwith cognate ligands such as α-GalCer has not been studiedin HIV infection, or SIV infection, compared to naïve SIVuninfected macaques. Such assays would be useful inimmunotherapy or vaccine clinical or pre-clinical trialsdesigned to harness NKT cell based immunity. In this studywe assessed ex vivo NKT function upon stimulation of humanand macaque blood with α-GalCer under a variety of

experimental conditions to optimize methods for NKT cellactivation.

2. Methods

2.1. Healthy human subjects and animals

For all studies on optimized NKT cell activation methods(Figs. 2–5) we studied subjects and animals as detailedbelow. Healthy HIV uninfected human volunteers (n=5aged 23–48 years; four females, one male) with varyingfrequencies of peripheral NKT cells (range of 0.03%-0.23%)were recruited for this study. We also studied a total of 9juvenile pigtail macaques (Macaca nemestrina) aged 2 to6 years. Five macaques were infected with SIVmac251 intra-vaginally and were 21–28 weeks post infection at the time ofthis study. Four SIV-uninfected healthy pigtail macaqueswere also studied. Macaques were sedated with ketamine(10 mg/kg monkey weight) and intravenous peripheralblood was drawn from macaques or humans. Spleens fromtwo naïve C57BL/6 mice were also harvested for mononu-clear cells. All studies were approved by the relevantinstitutional human and animal research ethic committees.

2.2. Activation of macaque or human NKT cells with α-GalCer orPMA/ionomycin

Lyophilised α-Galactosylceramide (Sapphire Biosciences,catalog number SL-232) was resuspended to 200 μg/ml inPBS containing 56 mg/ml sucrose, 7.5 mg/ml L-histidine, 0.5%tween-20 by step-wise addition of the buffer interspersedwith sonication, vortexing and heating at 60 °C. Unlessotherwise stated NKT cells within macaque or humanperipheral blood (200 μl) were activated with 5 μg/ml α-GalCer for 6 h at 37 °C, 5% CO2 in a 96-well U-bottom platefollowed by the addition of brefeldin A (BFA) at 10 μg/ml forthe last 4 h of activation. Mouse anti-human CD107a (cloneH4A3, BD) was added at the beginning of the activationperiod. PMA/ionomycin activation of macaque NKT cells wascarried out with 10 ng/ml phorbol 12-myristate 13-acetate(Sigma) and 3 μM ionomycin calcium salt (Sigma) for 4 h at37 °C, 5% CO2 with the addition of a protein transportinhibitor for the last 2 h of activation. Unstimulated samples(Figs. 2–5) were incubated in PBS buffer containing sucrose,histidine and tween at concentrations given above. Selectedexperiments used mouse IgG1, κ isotype control antibody(APC-H7-clone MOPC-21, BD) as negative staining control forthe CD107a surface stain (Supplementary Fig. 1).

2.3. α-GalCer activation of mouse splenic NKT cells

Spleens were harvested from mice into PBS containing 2%fetal bovine serum, filtered through 70 μm filters andcentrifuged at 350 g for 4 min. Red blood cell lysis wasconducted using 4 ml Red Blood Cell Lysing Buffer (Sigma) atroom temp for 4 min. Live splenocytes were enumeratedwith trypan blue dye exclusion in a heamocytometer. NKTcell activation was performed as previously described in theliterature (Uldrich et al., 2005; Patel et al., 2011). Splenocyteswere incubated with 0.1 μg/ml α-GalCer in 4 wells of a 96-well U-bottom plate at 0.5×106 splenocytes per well for 8 h

152 C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

at 37 °C, 5% C02, with the addition of monensin (GolgiStop,BD) at 2 μM for the last 4 h. The assay was conducted induplicate.

2.4. Reagents and NKT cells analyses

Human or macaque samples: At the end of the activationperiod, macaque or human peripheral blood cells wereincubated at 4 °C overnight and further surface stained witha cocktail containing mouse CD1d tetramer (produced in-house with a baculovirus-based CD1d expression systemoriginally derived from M. Kronenberg, (La Jolla Institute forAllergy and Immunology) loaded with α-GalCer, kindlyprovided by P. Savage (Brigham Young University) (C 24:1PBS-44), mouse anti-human antibodies to CD3 (clone SP34-2,BD), CD4 (Clone L200, BD) and CD8 (clone SK1, BD), with orwithout CD161 (clone DX12, BD for staining human samplesor clone HP-3 G10, eBioscience, for macaque samples) for30 min at room temperature, whereupon red blood cell lysiswas conducted at 10 min at room temperature with 10-foldvolume of BD FACS lysing solution. A fixable live-dead cellstain (Aqua stain, Invitrogen) was added to unstimulated or1 μg/ml α-GalCer activated healthy human or SIV-uninfectedmacaque blood samples in addition to the above reagents inselected experiments to show that the activated NKTs werelive cells (Supplementary Fig. 2). All washes in the followingsteps were carried out with FACS wash buffer containing 0.5%(w/v) bovine serum albumin and 2 mM EDTA in phosphatebuffered saline, at 500 g for 7 min. The reaction wasinactivated by the addition of an equal volume of FACSwash buffer. Cells were centrifuged and then permeabilizedwith 0.5 ml 1× BD FACS Perm for 10 min at room temper-ature. The reaction was stopped with 3.5 ml FACS washbuffer, centrifuged and decanted completely. Intracellularcytokine staining was performed with a cocktail containingmouse anti-human antibodies to TNF ( l MAb11, BD), IFNγ(clone B27, BD), and rat anti-human IL-2 (clone MQ1-17 H12,BD) for 30–60 min at room temperature. In selected exper-iments where a fixable live-dead cell stain was included(described above, Supplementary Fig. 1), mouse IgG1 κisotype control antibodies to TNF (clone PE-Cy7-MOPC-21,BD), and IFNγ (clone Alexa Fluor 700-MOPC-21, BD) and ratIgG2a, κ isotype control antibody to IL-2 (clone APC-R35-95,BD) were added instead. Cells were subsequently washedwith 4 ml FACS wash buffer then resuspended in 30 μlstabilizing fixative (BD) prior to analysis on an LSR II flowcytometer (BD). NKT cells were gated as singlet lymphocytesdouble positive for CD3 and CD1d tetramer loaded withα-GalCer (Fig. 1A shows the gating strategy).

Mouse samples: At the end of the activation period, fourreplicate wells of mouse splenocytes were pooled intoV-bottom FACS tubes, washed twice with Wash buffer (2%fetal bovine serum in phosphate buffered saline) at 350 g for4 min at 4 °C. All washes and centrifugation steps on micesplenic cells were either carried out with Wash buffer asabove or with Perm/Wash (BD). Cells were surface stainedwith hamster anti-mouse TCR beta, rat anti-mouse B220,CD1d tetramer loaded with α-GalCer (as described above),and 7-AAD to discriminate live from dead cells on ice for30 min. Cells were washed twice with Wash buffer, fixed andpermeabilized with 150 μl Cytofix/Cytoperm (BD) solution

for 30 min on ice and washed twice with Perm/Wash (BD).Cells were stained intracellularly with rat anti-mouse TNF,IFNγ, and IL-4 for 30 min on ice. Following two furtherwashes with Perm/Wash (BD) cells were resuspended in150 μl Wash buffer and analyzed immediately on an LSR IIflow cytometer (BD). NKT cells were gated as lymphocytesdouble positive for TCR beta and CD1d tetramer loaded withα-GalCer, and negative for 7-AAD and B220.

2.4.1. Statistical analysesStatistical analyses were performed using SPSS version 20

(IBM, Chicago, IL, USA). Data were analyzed by two-wayANOVA in conjunction with Bonferroni post hoc test.Analyses of individual data sets pertaining to a particularcytokine type was done using one-way ANOVA withBonferroni post hoc test as indicated in Figure legends.Where necessary, (Fig. 2A IL-2 data set from SIV infectedmacaques and Fig. 2B) log10 transformation before ANOVAwas performed for the data to pass or tend towards Levene'stest for equal variances. Where the data had unequalvariances a more stringent significance level of Pb0.01 wasused (Fig. 2B).

3. Results

3.1. Activation of NKT cells in macaques using standardconditions studied in mice

Our goal was to develop reliable conditions to assessantigen-specific NKT cell activation in macaque or humanblood samples. To assess the ability of primate NKT cells torespond to α-GalCer-induced activation ex vivo, we firstassayed the effectiveness of the same experimental condi-tions used in mouse lymphoid cells (Uldrich et al., 2005; Patelet al., 2011) on the ex vivo activation of pigtail macaque orhuman peripheral blood NKT cells. We gated on NKT cells aspreviously described (Fernandez et al., 2009) (Fig. 1A).Lymphocytes that double stained for CD1d:α-GalCer tetra-mer and CD3 were analyzed for the intracellular productionof TNF, IFNγ, and IL-2 and upregulation of a surfacedegranulation marker CD107a. Under the standard condi-tions suitable for activation of mouse splenocyte NKT cells, of0.1 μg/ml α-GalCer in a 6–8 h incubation with the addition ofmonensin for the last 4 h, we observed intracellular expres-sion of multiple cytokines from mouse splenocytes (Fig. 1B).However, negligible amounts of cytokines were detectedfrom NKT cells within either macaque or human peripheralblood (Fig. 1B).

3.2. Comparison of monensin and brefeldin A for intracellularcytokine staining of macaque or human NKT cells

It has previously been reported that the protein transportinhibitor monensin is inhibitory to the production of TNFfrom T cells upon PMA/ionomycin stimulation (O'Neil-Andersen and Lawrence, 2002). We first studied the effectof 2 protein transport inhibitors, monensin (GolgiStop, BD),alone at 2 μM (MN), brefeldin A (BFA) alone at 10 μg/ml, or acombination of both reagents each at a half strengthconcentration (BFA/MN), upon PMA/ionomycin activation ofNKT cells within peripheral blood samples derived from 9

A

B

Fig. 1. Poor stimulation of human and macaque blood NKT cells using conditions reported for mouse splenocyte NKT cell activation. (A) Gating strategy used inflow cytometry of intracellular staining of cytokines expressed by α-GalCer activated NKT cells. Macaque or human NKT cells are defined as lymphocytes doublepositive for α-GalCer analog loaded CD1d tetramer and CD3. (B) Dot plots showing expression of cytokines and a degranulation marker (CD107a) from NKT cellsof mouse splenocytes, macaque peripheral blood and human peripheral blood upon activation with 0.1 μg/ml α-GalCer for 8 h with the addition of monensin forthe last 4 h. Numbers in each rectangular gate indicate the percentage of background subtracted positive cells within the NKT population. IL-4 was substituted forCD107a in the mouse assay. Data representative of 2–5 similar experiments using 2–5 samples per experiment.

153C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

pigtail macaques (Fig. 2A). High levels of TNF or IFNγ (>74%NKT cells expression of either cytokine for each animal) wereobserved in the PMA/ionomycin-activated, BFA-treated pe-ripheral blood NKT cells from SIV-infected monkeys. Incontrast, NKT cells from blood of SIV-uninfected macaquesproduced significantly lower levels of all three cytokines(pb0.0001). In the SIV infected macaques, the effect ofmonensin alone (MN) was significantly (p=0.001) lesseffective on the intracellular accumulation of all threecytokines, particularly TNF and IL-2. In SIV-uninfectedanimals, either BFA/MN or BFA alone were effective reagentsfor intracellular accumulation while BFA alone was better forSIV-infected macaques.

We next assessed the effect of these protein transportinhibitors using macaque and human NKT cells activatedwith α-GalCer (Fig. 2B and C). Similar to the results obtainedin the PMA/ionomycin activation assays above, we found areduction of cytokine accumulation by α-GalCer stimulationof NKT cells in the presence of MN (Fig. 2B; pb0.001). Thiseffect was seen across both macaque and human peripheralblood samples. Overall, cytokine levels were lower withα-GalCer activation compared to mitogenic activation. Inter-estingly, cytokine levels detected from healthy human NKT

cells were significantly higher than in healthy macaques(pb0.001). BFA alone (p=0.001 for SIV-infected), or acombination of BFA/MN (p=0.021 for SIV-uninfected), waseffective in the optimal detection of α-GalCer activation ofNKT cells from SIV-infected or SIV-uninfected animals whileBFA alone was significantly superior to BFA/MN (p=0.020)or MN (pb0.001) for the detection of cytokines from freshhuman NKT cells. Overall, we again observed that healthySIV-uninfected macaques produced less cytokines than SIV-infected macaques (pb0.001).

3.3. Environment of ex vivo NKT cell activation

Cells of the immune system commonly require proximalcontact for formation of immunological synapses and activa-tion (Kuylenstierna et al., 2011). We therefore investigatedthe effect of different culture conditions on ex vivo α-GalCerinduced activation of NKT cells (Fig. 3). NKT cells fromperipheral blood of SIV-infected or -uninfected macaques, orhealthy humans, were incubated in a 96-well U-bottom plate(diameter=5–6 mm) or the larger surface area of a roundbottom FACS tube (diameter=9–10 mm). The plate incuba-tion generally improved the cytokine expression from NKT

A

B

C

Fig. 2. Differential effects of protein transport inhibitors on cytokine production from NKT cells. (A) Peripheral blood from SIV-infected macaques (n=5),SIV-uninfected macaques (n=4) or healthy humans (n=5) were activated with PMA/ionomycin in the presence of 10 μg/ml brefeldin A (BFA), 5 μg/ml BFAtogether with 1 μM Golgi stop containing monensin (BFA/MN) or 2 μM Golgi stop only (MN) in a 4 hour assay with the addition of the relevant protein transportinhibitor for the last 2 h. Bar graphs show the proportion of NKT cells positive by intracellular cytokine staining. Error bars, standard error of mean; one-wayANOVA with Bonferroni post hoc test between BFA alone and MN alone groups (***=b0.001, **=0.007, ns = not significant). (B) Similar to (A) but stimulatedwith 5 μg/ml α-GalCer for 6 h with the addition of a protein transport inhibitor for the last 4 h. (C) Flow cytometric analysis of expression of intracellularcytokines and degranulation marker from NKT cells from one SIV-infected macaque after α-GalCer activation. Unstim, unstimulated samples; naïve macaques,SIV-uninfected macaques. Data representative of 2–6 similar experiments using the number of subjects given above.

154 C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

cells compared to the FACS tube incubation as better NKT cellactivation was observed in humans (p=0.018 in a two-wayANOVA with Bonferroni post hoc test), whilst a trend in thesame direction was seen in samples from SIV-infected or-uninfected macaques. A 96-well plate format was thereforeused for subsequent optimization experiments below.

3.4. Duration of activation of NKT cells

The optimal duration of ex vivo antigen-specific activationof NKT cells for detection of cytokines is unclear, with mostT cell-based activation assays running for 4–6 h with BFAadded for either the whole duration or after the first 2 h ofactivation (Dale et al., 2004). We therefore investigatedthe time required to activate NKT cells optimally withα-GalCer by incubating at 37 °C for 4 or 6 h with the additionof BFA either for the whole duration of the incubation or thelast 4 h of incubation (Fig. 4). In both the SIV infected anduninfected macaques a 6-hour incubation demonstrated atrend towards modestly better NKT activation than a 4-hourincubation, although the differences were not statisticallysignificant. In human blood NKT cells there was significantly

greater NKT cell activation in a 6-hour incubation, regardlessof when BFA was added, compared to a 4-hour incubation(p=0.001).

3.5. Dose responsiveness ofα-GalCer-induced NKT cell activation

Thus far we have developed optimized conditions forα-GalCer activation of NKT such as choice of protein transportinhibitor, and duration and surface area of activation. Mousestudies on purified splenocytes usually use 0.1 μg/ml α-GalCer,but macaque or human NKT cells within whole blood may havealtered sensitivity to α-GalCer ex vivo, which may partly be dueto the abundance of red blood cells present that may non-specifically take up some of the antigen. We therefore sought todetermine an appropriate dose ofα-GalCer suitable formaximalcytokine or degranulation responses using whole blood. NKTcells were activated with 2–5 fold increasing doses of α-GalCerand assayed for intracellular cytokines (Fig. 5A and B). HumanNKT cells showed a significantly elevated cytokine responseto α-GalCer concentrations ≥1 μg/ml (pb0.001). SimilarlyNKT cells from SIV-infected or uninfected macaques exhibitedincreased cytokine production in response to doses >0.1 μg/ml.

A

B

Fig. 3. The effect of the milieu of activation on cytokine production from NKT cells. Peripheral blood cells from SIV-infected macaques, naïve macaques or healthyhumans were activated with (A) 5 μg/ml α-GalCer in a 6 hour assay with the addition of 10 μg/ml brefeldin A (BFA) at the beginning of the assay, in a 96-wellU-bottom plate or a tube. Bar graphs show the proportion of NKT cells positive by intracellular cytokine staining. Error bars, standard error of mean; two-wayANOVA performed on untransformed data. (B) Flow cytometric analysis of expression of intracellular cytokines and degranulation marker from NKT cells fromone SIV-infected macaque upon α-GalCer activation. Unstim, unstimulated samples. Data representative of 2–6 similar experiments.

155C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

For NKT cells from both macaques and humans there wasno difference between 1 μg/ml to 10 μg/mlα-GalCer, suggestingthat saturating levels could be achieved at 1 μg/ml α-GalCerin this assay. If anything, there was a trend towards lowercytokine production at 10 μg/ml α-GalCer in some samplesstudied.

A

Fig. 4. The effect of duration of a-GalCer activation on cytokine production from Nhealthy humans were activated with (A) 5 μg/ml α-GalCer: for 6 h with the additionthe beginning of activation (grey bars), or 4 h with the addition of BFA at the beginpositive by intracellular cytokine staining. Error bars, standard error of mean.degranulation marker from NKT cells from one SIV-infected macaque stimulated witexperiments.

As observed earlier (Fig. 2) we again noticed a markeddifference in levels of cytokine expression between the threegroups studied with human NKT cells being more efficientcytokine producers than healthy, SIV-uninfected macaqueNKT cells (pb0.001 for TNF or IFNγ expression). SIV-uninfected macaque NKT cells produced less TNF or IFNγ

B

KT cells. Peripheral blood from SIV-infected macaques, naïve macaques orof 10 μg/ml BFA for the last 4 h (black bars), 6 h with the addition of BFA at

ning of incubation (white bars). Bar graphs show the proportion of NKT cells(B) Flow cytometric analysis of expression of intracellular cytokines andhα-GalCer. Unstim, unstimulated samples. Data representative of 2–6 similar

A B

Fig. 5. Enhanced cytokine production from NKT cells with >0.1 μg/ml α-GalCer. (A) Expression of intracellular cytokines and degranulation marker from NKTcells following a 6 hour stimulation of peripheral blood from SIV-infected macaques, SIV-uninfected macaques or healthy humans with 10-, 5-, 1-, or 0.1 μg/mlα-GalCer, followed by the addition of 10 μg/ml BFA for the last 4 h of incubation. Error bars, standard error of mean; one-way ANOVA on human TNF or IFNγexpression levels with Bonferroni post hoc test (for TNF samples p=0.015 between 10- and 0.1 μg/ml α-GalCer, p=0.008 between 5- and 0.1 μg/ml α-GalCer,p=0.027 between 1- and 0.1 μg/ml α-GalCer doses; for IFNγ samples p=0.001 between 10- and 0.1 μg/ml α-GalCer, p=0.001 between 5- and 0.1 μg/mlα-GalCer, p=0.005 between 1- and 0.1 μg/ml α-GalCer doses). (B) Flow cytometric analysis of intracellular cytokine expression from NKT cells from oneSIV-infected macaque stimulated with α-GalCer. Unstim, unstimulated samples. Data representative of 2–6 similar experiments.

156 C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

than NKT cells from SIV-infected macaques across allα-GalCer doses studied (pb0.001).

To exclude the possibility of non-specific staining ofintracellular cytokines within NKT cells, particularly fromhealthy macaques where cytokine expression levels arelower, we also stained samples with a fixable live/deadstain and isotype control antibodies. Blood from NKT cellsfrom 5 SIV uninfected macaques or 5 humans, activated with1 μg/ml α-GalCer under optimal conditions defined thus far,were stained with the appropriate isotype control antibodiesand a fixable live-dead cell stain. Isotype control staining didnot reveal any non-specific staining of NKT cells in eitherhuman or macaque blood samples (Supplementary Fig. 1).Further, there was no statistically significant difference inintracellular cytokine accumulation when dead cells weregated or not gated out of the analyses (Supplementary Fig. 2).

3.6. Effect of α-GalCer activation on NKT cell subsets

Phenotypic expression of NKT surface molecules canchange during long term in vitro activation with α-GalCer(Rout et al., 2010), but it is not clear if this also occurs duringthe short 6-hour in vitro activation process we studied. Wetherefore analyzed the surface expression of CD161, CD4, andCD8 on NKT cells of healthy humans or SIV-uninfectedmacaques in blood samples ex vivo, or following a 6 hour invitro culture with or without α-GalCer activation (Fig. 6). Inboth macaque and human blood samples there was nostatistical difference in the expression of the CD161, CD4 orCD8 surface markers in ex vivo blood compared to culturedblood. Further we observed a trend towards lower CD161expression in macaque NKT cells after a 6-hour culture,although this difference was statistically insignificant. Inter-estingly we observed much lower levels of CD161 positiveNKT cells in macaques than in human (pb0.001).

4. Discussion

The ability of NKT cells to rapidly respond with produc-tion of a variety of cytokines and other effector functions iskey to their potent antiviral and anti-tumor properties. Thestudy of antigen-driven NKT cell activation in macaque andhuman whole blood samples is hampered by relatively weakcytokine responses, compared to similar assays using mouseNKT cells. We refined and optimized methods to studyactivation of human or macaque NKT cells with theprototypic ligand, α-Galactosylceramide. We also assessedthe effectiveness of NKT cell activation techniques, as well asstudying the dose responsiveness of α-GalCer itself on NKTcells. Human or macaque ex vivo NKT cell activation studiesusing antigenic stimulation have variously used free α-GalCer at the standard concentration of 0.1 μg/ml with orwithout CD28 co-stimulation, α-GalCer loaded dendriticcells, or CD1d dimer loaded with α-GalCer (Chang et al.,2005; Moll et al., 2009; Snyder-Cappione et al., 2009).However, careful comparative methods for ex vivo activationof primate NKT cells are not clearly defined in the literature.

Our data reveal that ex vivo activation of macaque orhuman NKT cells in fresh peripheral blood with free α-GalCeris best achieved at a dose of 1–5 μg/ml. The addition of BFAfor the intracellular accumulation of expressed cytokinesover a 6-hour incubation period was superior to monensin,consistent with previous literature (O'Neil-Andersen andLawrence, 2002). Further, we suggest the use of a 96-well U-bottom plate and the addition of BFA 2 h after thecommencement of incubation also enhances conditions fordetected NKT cell activation by α-GalCer. It is possible thatNKT cells activated with ‘free’ α-GalCer prefer close contactwith APCs for maximal stimulation. The choice of using platesor tubes has practical advantages and limitations. Assay set-up could be done more efficiently in 96-well plates than intubes, however, subsequent intracellular staining and flow

A

B

Fig. 6. The effect of α-GalCer activation on NKT cell subsets. (A) The expression of CD161, CD4 single positive and CD8 single positive markers on blood of NKTcells from 5 SIV-uninfected macaques or 5 healthy humans was analyzed ex vivo, cultured for 6 h without α-GalCer activation, or activated for 6 h with 1 μg/mlα-GalCer. Error bars, standard error of mean. (B) Flow cytometric analysis of CD1d:αGalCer tetramer staining on lymphocytes, CD1d expression on gated NKTcells, and CD4 or CD8 single positive NKT cell subsets on ex vivo, or α-GalCer stimulated or unstimulated blood from one SIV-uninfected macaque. Intracellularcytokine expression from NKT cells in shown in the last panel. The data shown are representative of 2 similar experiments.

157C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

cytometric analyses usually requires the cells to be trans-ferred to FACS tubes.

Our initial studies on ex vivo activation of macaque NKTcells were carried out on macaque peripheral blood with0.1 μg/ml free α-GalCer stimulation with the addition ofGolgi stop (containing monensin) for intracellular cytokineaccumulation. This resulted in negligible NKT cell activation,despite the fact that macaque NKT cells are capable of robustcytokine production when activated through mitogenicstimuli. In contrast, analysis of mouse splenic NKT cellsunder similar conditions revealed substantially higher intra-cellular cytokine staining, suggesting a greater sensitivity toex vivoα-GalCer activation in these ex vivo assays (Patel et al.,2011). The CD1d molecules on APCs are conserved evolu-tionarily between mice, macaques and humans, as mouseCD1d tetramers loaded with α-GalCer readily detect humanand macaque NKT cells (Matsuda et al.; Benlagha et al.,

2000; Fernandez et al., 2009). It would be interesting to studyα-GalCer-mediated activation of macaque splenic NKT cellscultured in the above conditions to determine any activationdifferential between NKT cells from peripheral blood andspleen. However, given the relative difficulty in obtainingmacaque (or human) splenocyte samples, particularly forlongitudinal studies, optimized methods for studying periph-eral blood NKT cells will be more practical for future researchpurposes.

As previously reported for conventional T cells, we foundmonensin to be less effective for the intracellular accumula-tion of TNF from PMA- or α-GalCer activated macaque orhuman NKT cells (O'Neil-Andersen and Lawrence, 2002). Theintracellular accumulation of IFNγ or IL-2 was also lower inNKT cells treated with monensin alone. Unlike human ormacaque NKT cells, intracellular accumulation of cytokinessuch as TNF, IFNγ or IL-4 in mouse spleen NKT cells activated

158 C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

ex vivo with α-GalCer in the presence of BFA was eithersimilar or slightly elevated compared to that of monensin-only treated cells (data not shown). Nylander et al. (Nylanderand Kalies, 1999) have shown that PMA/ionomycin stimu-lated mouse splenic T cells treated with BFA or monensinproduced similar levels of intracellular IFNγ. It has also beenreported that monensin is more toxic to humanmonocytes inperipheral blood (Schuerwegh et al., 2001) or mouse T cellsin spleen preparations (Nylander and Kalies, 1999) resultingin lower viability after in vitro stimulation. Another factor toconsider in choosing between monensin or BFA treatedperipheral blood is the ease of lysis of red blood cells. Wefind that mitogen-stimulated macaque peripheral bloodpreparations lyse with greater efficiency when treated withmonensin rather than BFA (data not shown).

Our study analyzed fresh blood samples as only a smallvolume of blood is required to detect these relatively rareNKT cells (200 μl/test) and variable loss of viability orintercurrent activation could occur through preparing, stor-ing and thawing PBMC. When we studied a live-dead cellgating strategy we found this was not necessary in the freshblood samples we studied, however this may be much moreimportant when frozen cells are analyzed. For many clinicalstudies, assaying fresh blood samples will not be practicaland in the future, it will be important to optimize NKT cellactivation conditions on thawed PBMC. One study showedthat approximately 30% of healthy human PBMC NKT cellsexpressed IFNγ after an overnight activation with 0.1 μg/mlα-GalCer using both monensin and an anti-CD28 antibody(Chang et al., 2005). We performed overnight cultures forNKT cell activation but found that downregulation of the Tcell receptor occurred, making it difficult to gate accuratelyon NKT cells (data not shown).

This study focused on the use of free α-GalCer to activateNKT cells as this measures the capability of NKT cells torespond to endogenous antigen presenting cells presentingα-GalCer. Recent studies have successfully used a CD1ddimer fused to immunoglobulin (Dimer X, BD) loaded withα-GalCer as the antigen-specific stimulant for human NKTcells (Snyder-Cappione et al., 2009). We also find that humanor macaque NKT cells can be activated with as little as 5 ng/mlDimer X:α-GalCer, with optimal activation at 714 ng/ml (datanot shown). Future studies of the ability of blood NKT cells torespond to either free α-GalCer or CD1d dimer-presentedα-GalCer are likely to be complementary and potentially revealinsights into the primary blood cell presentation ofα-GalCer toNKT cells.

We found that pigtail macaque NKT cells express lowlevels of the NK cell marker, CD161, in contrast to adulthuman NKT cells (Gumperz et al., 2002; Eger et al., 2006).Our finding accords with that of healthy rhesus macaqueperipheral blood lymphocytes where only 0.4% NKT cells areCD161 positive (Gansuvd et al., 2003), and furthermore only9% healthy sooty mangabey NKT lymphocytes express CD161(Rout et al., 2010). We also saw a trend towards lower NKTcell CD161 expression after a 6 hour culture, irrespective ofα-GalCer activation, consistent with rhesus macaque datawhere CD161 expression was lost in both short- and long-term cultures (Gansuvd et al., 2003), but in contrast to sootymangabeys NKT cultured cells where the mean CD161expression of clones was 53% (Rout et al., 2010).

We made the surprising finding that blood NKT cells fromSIV-infected macaques were significantly more capable of exvivo activation and cytokine expression in response to α-GalCer than those of healthy control animals. This wasconsistent at all doses of α-GalCer studied and a variety ofin vitro conditions analyzed, with higher intracellular ex-pression of all three TNF, IFNγ and IL-2 cytokine profiles. SIVinfection of macaques or HIV infection of human conven-tional T or B cells typically leads to dysfunctional responses(McElrath et al., 2008; Klatt et al., 2011). NKT cells of pigtailmacaques or humans can readily be infected and loss of theCD4-expressing subset of NKT cells is observed in chronicinfection (Sandberg et al., 2002; van der Vliet et al., 2002;Fernandez et al., 2009). The remaining NKT cells (primarilyCD8+ NKT cells) in SIV infected macaques clearly maintainthe capacity to undergo high levels of activation ex vivo,potentially reflecting their response to the ongoing infectionand/or inflammation. This raises the possibility that CD8+NKT cells could be important in maintaining partial control ofSIV infection or disease during chronic infection. Furtherstudies on the functional capacity of NKT cell subsets duringSIV infection of macaques and HIV infection of humansshould be illuminating.

In summary, we define optimal conditions to analyzeantigen-specific ex vivo NKT activation in blood samples fromboth humans and macaques. We found that SIV infectedmacaques had higher levels of NKT activation than unin-fected macaques. Our results suggest these assays should beuseful in monitoring NKT cell function during disease andthroughout immunotherapy studies.

Acknowledgments

We thank Wendy Winnall, Konstantinos Kyparissoudis,Angela Chan, Sheilajen Alcantara, Pradeep Sidana, and MeganSchepers for expert assistance and advice. This research wassupported by Australian National Health and Medical Re-search (NHMRC) awards 510448, 629000 and 454569. GC issupported by a Cancer Research Institute postgraduatescholarship. DIG is supported by an NHMRC researchfellowship 454309; SJK is supported by an NHMRC researchfellowship 508937.

Appendix A. Supplementary data

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.jim.2012.05.019.

References

Benlagha, K., Weiss, A., Beavis, A., Teyton, L., Bendelac, A., 2000. In vivoidentification of glycolipid antigen-specific T cells using fluorescentCD1d tetramers. J. Exp. Med. 191, 1895.

Brigl, M., Brenner, M.B., 2010. How invariant natural killer T cells respond toinfection by recognizing microbial or endogenous lipid antigens. Semin.Immunol. 22, 79.

Cerundolo, V., Silk, J.D., Masri, S.H., Salio, M., 2009. Harnessing invariant NKTcells in vaccination strategies. Nat. Rev. Immunol. 9, 28.

Chang, D.H., Osman, K., Connolly, J., Kukreja, A., Krasovsky, J., Pack, M.,Hutchinson, A., Geller, M., Liu, N., Annable, R., Shay, J., Kirchhoff, K., Nishi,N., Ando, Y., Hayashi, K., Hassoun, H., Steinman, R.M., Dhodapkar, M.V.,2005. Sustained expansion of NKT cells and antigen-specific T cells afterinjection of alpha-galactosyl-ceramide loaded mature dendritic cells incancer patients. J. Exp. Med. 201, 1503.

159C.S. Fernandez et al. / Journal of Immunological Methods 382 (2012) 150–159

Dale, C.J., De Rose, R., Stratov, I., Chea, S., Montefiori, D.C., Thomson, S.,Ramshaw, I.A., Coupar, B.E., Boyle, D.B., Law, M., Kent, S.J., 2004. Efficacyof DNA and fowlpox virus priming/boosting vaccines for simian/humanimmunodeficiency virus. J. Virol. 78, 13819.

Eger, K.A., Sundrud, M.S., Motsinger, A.A., Tseng, M., Van Kaer, L., Unutmaz,D., 2006. Human natural killer T cells are heterogeneous in their capacityto reprogram their effector functions. PLoS One 1, e50.

Emoto, M., Emoto, Y., Yoshizawa, I., Kita, E., Shimizu, T., Hurwitz, R.,Brinkmann, V., Kaufmann, S.H.E., 2010. Alpha-GalCer ameliorateslisteriosis by accelerating infiltration of Gr-1+ cells into the liver. Eur.J. Immunol. 40, 1328.

Exley, M., Garcia, J., Balk, S.P., Porcelli, S., 1997. Requirements for CD1drecognition by human invariant Valpha24+ CD4−CD8− T cells. J. Exp.Med. 186, 109.

Fernandez, C.S., Chan, A.C., Kyparissoudis, K., De Rose, R., Godfrey, D.I., Kent,S.J., 2009. Peripheral NKT cells in simian immunodeficiency virus-infected macaques. J. Virol. 83, 1617.

Fujii, S., Shimizu, K., Smith, C., Bonifaz, L., Steinman, R.M., 2003. Activation ofnatural killer T cells by alpha-galactosylceramide rapidly induces the fullmaturation of dendritic cells in vivo and thereby acts as an adjuvant forcombined CD4 and CD8 T cell immunity to a coadministered protein.J. Exp. Med. 198, 267.

Fujii, S., Motohashi, S., Shimizu, K., Nakayama, T., Yoshiga, Y., Taniguchi, M.,2010. Adjuvant activity mediated by iNKT cells. Semin. Immunol. 22, 97.

Gansuvd, B., Hubbard, W.J., Hutchings, A., Thomas, F.T., Goodwin, J., Wilson,S.B., Exley, M.A., Thomas, J.M., 2003. Phenotypic and functionalcharacterization of long-term cultured rhesus macaque spleen-derivedNKT cells. J. Immunol. 171, 2904.

Giaccone, G., Punt, C.J., Ando, Y., Ruijter, R., Nishi, N., Peters, M., vonBlomberg, B.M., Scheper, R.J., van der Vliet, H.J., van den Eertwegh, A.J.,Roelvink, M., Beijnen, J., Zwierzina, H., Pinedo, H.M., 2002. A phase Istudy of the natural killer T-cell ligand alpha-galactosylceramide(KRN7000) in patients with solid tumors. Clin. Cancer Res. 8, 3702.

Godfrey, D.I., Stankovic, S., Baxter, A.G., 2010. Raising the NKT cell family.Nat. Immunol. 11, 197.

Guillonneau, C., Mintern, J.D., Hubert, F.X., Hurt, A.C., Besra, G.S., Porcelli, S.,Barr, I.G., Doherty, P.C., Godfrey, D.I., Turner, S.J., 2009. Combined NKTcell activation and influenza virus vaccination boosts memory CTLgeneration and protective immunity. Proc. Natl. Acad. Sci. U. S. A. 106,3330.

Gumperz, J.E., Miyake, S., Yamamura, T., Brenner, M.B., 2002. Functionallydistinct subsets of CD1d-restricted natural killer T cells revealed byCD1d tetramer staining. J. Exp. Med. 195, 625.

Hong, S., Wilson, M.T., Serizawa, I., Wu, L., Singh, N., Naidenko, O.V., Miura, T.,Haba, T., Scherer, D.C., Wei, J., Kronenberg, M., Koezuka, Y., Van Kaer, L.,2001. The natural killer T-cell ligand alpha-galactosylceramide preventsautoimmune diabetes in non-obese diabetic mice. Nat. Med. 7, 1052.

Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Motoki, K., Ueno, H.,Nakagawa, R., Sato, H., Kondo, E., Koseki, H., Taniguchi, M., 1997. CD1d-restricted and TCR-mediated activation of valpha14 NKT cells byglycosylceramides. Science 278, 1626.

Klatt, N.R., Vinton, C.L., Lynch, R.M., Canary, L.A., Ho, J., Darrah, P.A., Estes, J.D.,Seder, R.A., Moir, S.L., Brenchley, J.M., 2011. SIV infection of rhesusmacaques results in dysfunctional T- and B-cell responses to neo andrecall Leishmania major vaccination. Blood 118, 5803.

Kunii, N., Horiguchi, S., Motohashi, S., Yamamoto, H., Ueno, N., Yamamoto, S.,Sakurai, D., Taniguchi, M., Nakayama, T., Okamoto, Y., 2009. Combinationtherapy of in vitro-expanded natural killer T cells and alpha-galactosylceramide-pulsed antigen-presenting cells in patients withrecurrent head and neck carcinoma. Cancer Sci. 100, 1092.

Kuylenstierna, C., Bjorkstrom, N.K., Andersson, S.K., Sahlstrom, P., Bosnjak, L.,Paquin-Proulx, D., Malmberg, K.J., Ljunggren, H.G., Moll, M., Sandberg,J.K., 2011. NKG2D performs two functions in invariant NKT cells: directTCR-independent activation of NK-like cytolysis and co-stimulation ofactivation by CD1d. Eur. J. Immunol. 41, 1913.

Lindqvist, M., Persson, J., Thorn, K., Harandi, A.M., 2009. The mucosaladjuvant effect of alpha-galactosylceramide for induction of protectiveimmunity to sexually transmitted viral infection. J. Immunol. 182,6435.

Matsuda, J.L., Naidenko, O.V., Gapin, L., Nakayama, T., Taniguchi, M., Wang,C.R., Koezuka, Y., Kronenberg, M., 2000. racking the response of naturalkiller T cells to a glycolipid antigen using CD1d tetramers. JOURNAL OFEXPERIMENTAL MEDICINE; SEP 4 192 (5), 741–753.

McElrath, M.J., De Rosa, S.C., Moodie, Z., Dubey, S., Kierstead, L., Janes, H.,Defawe, O.D., Carter, D.K., Hural, J., Akondy, R., Buchbinder, S.P.,Robertson, M.N., Mehrotra, D.V., Self, S.G., Corey, L., Shiver, J.W.,Casimiro, D.R., 2008. HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis. Lancet 372, 1894.

Moll, M., Kuylenstierna, C., Gonzalez, V.D., Andersson, S.K., Bosnjak, L.,Sonnerborg, A., Quigley, M.F., Sandberg, J.K., 2009. Severe functional

impairment and elevated PD-1 expression in CD1d-restricted NKT cellsretained during chronic HIV-1 infection. Eur. J. Immunol. 39, 902.

Motohashi, S., Nagato, K., Kunii, N., Yamamoto, H., Yamasaki, K., Okita, K.,Hanaoka, H., Shimizu, N., Suzuki, M., Yoshino, I., Taniguchi, M., Fujisawa,T., Nakayama, T., 2009. A phase I–II study of alpha-galactosylceramide-pulsed IL-2/GM-CSF-cultured peripheral blood mononuclear cells inpatients with advanced and recurrent non-small cell lung cancer.J. Immunol. 182, 2492.

Novak, J., Lehuen, A., 2011. Mechanism of regulation of autoimmunity byiNKT cells. CYTOKINE; MAR 53 (3), 263–270.

Nylander, S., Kalies, I., 1999. Brefeldin A, but not monensin, completelyblocks CD69 expression on mouse lymphocytes: efficacy of inhibitors ofprotein secretion in protocols for intracellular cytokine staining by flowcytometry. J. Immunol. Methods 224, 69.

O'Neil-Andersen, N.J., Lawrence, D.A., 2002. Differential modulation ofsurface and intracellular protein expression by T cells after stimulationin the presence of monensin or brefeldin A. Clin. Diagn. Lab. Immunol. 9,243.

Patel, O., Cameron, G., Pellicci, D.G., Liu, Z., Byun, H.S., Beddoe, T., McCluskey,J., Franck, R.W., Castano, A.R., Harrak, Y., Llebaria, A., Bittman, R., Porcelli,S.A., Godfrey, D.I., Rossjohn, J., 2011. NKT TCR recognition of CD1d-alpha-C-galactosylceramide. J. Immunol. 187, 4705.

Rout, N., Else, J.G., Yue, S., Connole, M., Exley, M.A., Kaur, A., 2010. Paucity ofCD4+ natural killer T (NKT) lymphocytes in sooty mangabeys isassociated with lack of NKT cell depletion after SIV infection. PLoS One5, e9787.

Sandberg, J.K., Fast, N.M., Palacios, E.H., Fennelly, G., Dobroszycki, J., Palumbo,P., Wiznia, A., Grant, R.M., Bhardwaj, N., Rosenberg, M.G., Nixon, D.F.,2002. Selective loss of innate CD4(+) V alpha 24 natural killer T cells inhuman immunodeficiency virus infection. J. Virol. 76, 7528.

Schneiders, F.L., Scheper, R.J., von Blomberg, B.M., Woltman, A.M., Janssen,H.L., van den Eertwegh, A.J., Verheul, H.M., de Gruijl, T.D., van der Vliet,H.J., 2011. Clinical experience with alpha-galactosylceramide(KRN7000) in patients with advanced cancer and chronic hepatitis B/Cinfection. Clin. Immunol. 140, 130.

Schuerwegh, A.J., Stevens, W.J., Bridts, C.H., De Clerck, L.S., 2001. Evaluationof monensin and brefeldin A for flow cytometric determination ofinterleukin-1 beta, interleukin-6, and tumor necrosis factor-alpha inmonocytes. Cytometry 46, 172.

Snyder-Cappione, J.E., Loo, C.P., Carvalho, K.I., Kuylenstierna, C., Deeks, S.G.,Hecht, F.M., Rosenberg, M.G., Sandberg, J.K., Kallas, E.G., Nixon, D.F.,2009. Lower cytokine secretion ex vivo by natural killer T cells in HIV-infected individuals is associated with higher CD161 expression. AIDS23, 1965.

Sullivan, B.A., Nagarajan, N.A., Wingender, G., Wang, J., Scott, I., Tsuji, M.,Franck, R.W., Porcelli, S.A., Zajonc, D.M., Kronenberg, M., 2010.Mechanisms for glycolipid antigen-driven cytokine polarization byValpha14i NKT cells. J. Immunol. 184, 141.

Uchida, T., Horiguchi, S., Tanaka, Y., Yamamoto, H., Kunii, N., Motohashi, S.,Taniguchi, M., Nakayama, T., Okamoto, Y., 2008. Phase I study of alpha-galactosylceramide-pulsed antigen presenting cells administration tothe nasal submucosa in unresectable or recurrent head and neck cancer.Cancer Immunol. Immunother. 57, 337.

Uldrich, A.P., Crowe, N.Y., Kyparissoudis, K., Pellicci, D.G., Zhan, Y., Lew, A.M.,Bouillet, P., Strasser, A., Smyth, M.J., Godfrey, D.I., 2005. NKT cellstimulation with glycolipid antigen in vivo: costimulation-dependentexpansion, Bim-dependent contraction, and hyporesponsiveness tofurther antigenic challenge. J. Immunol. 175, 3092.

van der Vliet, H.J., von Blomberg, B.M., Hazenberg, M.D., Nishi, N., Otto, S.A.,van Benthem, B.H., Prins, M., Claessen, F.A., van den Eertwegh, A.J.,Giaccone, G., Miedema, F., Scheper, R.J., Pinedo, H.M., 2002. Selectivedecrease in circulating V alpha 24+V beta 11+ NKT cells during HIVtype 1 infection. J. Immunol. 168, 1490.

Venkataswamy, M.M., Porcelli, S.A., 2010. Lipid and glycolipid antigens ofCD1d-restricted natural killer T cells. Semin. Immunol. 22, 68.

Vivier, E., Ugolini, S., Blaise, D., Chabannon, C., Brossay, L., 2012. Targetingnatural killer cells and natural killer T cells in cancer. Nat. Rev. Immunol.12, 239.

Wilson, M.T., Johansson, C., Olivares-Villagomez, D., Singh, A.K., Stanic, A.K.,Wang, C.R., Joyce, S., Wick, M.J., Van Kaer, L., 2003. The response ofnatural killer T cells to glycolipid antigens is characterized by surfacereceptor down-modulation and expansion. Proc. Natl. Acad. Sci. U. S. A.100, 10913.

Wu, L., Van Kaer, L., 2009. Natural killer T cells and autoimmune disease.Curr. Mol. Med. 9, 4.

Yamasaki, K., Horiguchi, S., Kurosaki, M., Kunii, N., Nagato, K., Hanaoka, H.,Shimizu, N., Ueno, N., Yamamoto, S., Taniguchi, M., Motohashi, S.,Nakayama, T., Okamoto, Y., 2011. Induction of NKT cell-specific immuneresponses in cancer tissues after NKT cell-targeted adoptive immuno-therapy. Clin. Immunol. 138, 255.

Ex-vivo α-Galactosylceramide activation of NKT cells in humans and macaques1. Introduction2. Methods2.1. Healthy human subjects and animals2.2. Activation of macaque or human NKT cells with α-GalCer or PMA/ionomycin2.3. α-GalCer activation of mouse splenic NKT cells2.4. Reagents and NKT cells analyses2.4.1. Statistical analyses

3. Results3.1. Activation of NKT cells in macaques using standard conditions studied in mice3.2. Comparison of monensin and brefeldin A for intracellular cytokine staining of macaque or human NKT cells3.3. Environment of ex vivo NKT cell activation3.4. Duration of activation of NKT cells3.5. Dose responsiveness of α-GalCer-induced NKT cell activation3.6. Effect of α-GalCer activation on NKT cell subsets

4. DiscussionAcknowledgmentsAppendix A. Supplementary dataReferences