Distinct Requirements for Activation of NKT and NK Cells during Viral Infection

Aaron J. Tyznik, Shilpi Verma, Qiao Wang, Mitchell Kronenberg and Chris A. Benedict
J Immunol April 15, 2014, 192 (8) 3676-3685; DOI: https://doi.org/10.4049/jimmunol.1300837

Abstract

NK cells are key regulators of innate defense against mouse CMV (MCMV). Like NK cells, NKT cells also produce high levels of IFN-γ rapidly after MCMV infection. However, whether similar mechanisms govern activation of these two cell types, as well as the significance of NKT cells for host resistance, remain unknown. In this article, we show that, although both NKT and NK cells are activated via cytokines, their particular cytokine requirements differ significantly in vitro and in vivo. IL-12 is required for NKT cell activation in vitro but is not sufficient, whereas NK cells have the capacity to be activated more promiscuously in response to individual cytokines from innate cells. In line with these results, GM-CSF–derived dendritic cells activated only NK cells upon MCMV infection, consistent with their virtual lack of IL-12 production, whereas Flt3 ligand–derived dendritic cells produced IL-12 and activated both NK and NKT cells. In vivo, NKT cell activation was abolished in IL-12−/− mice infected with MCMV, whereas NK cells were still activated. In turn, splenic NK cell activation was more IL-18 dependent. The differential requirements for IL-12 and IL-18 correlated with the levels of cytokine receptor expression by NK and NKT cells. Finally, mice lacking NKT cells showed reduced control of MCMV, and depleting NK cells further enhanced viral replication. Taken together, our results show that NKT and NK cells have differing requirements for cytokine-mediated activation, and both can contribute nonredundantly to MCMV defense, revealing that these two innate lymphocyte subsets function together to fine-tune antiviral responses.

Introduction

Host defense to mouse CMV (MCMV; a β-herpesvirus) involves multiple cell types of both the innate and adaptive immune systems (1). Two innate-like lymphocyte populations, NK cells and NKT cells, are major producers of IFN-γ early during MCMV infection (2, 3). NKT cells are a T lymphocyte subset that is characterized by expression of an invariant TCR α-chain, formed by a Vα14 to Jα18 rearrangement in mice. When paired with several β-chains, prominently Vβ8.2, this α-chain imparts specificity for glycolipids presented by CD1d, a class I–like Ag-presenting molecule. These cells are commonly referred to as type I or invariant NK T (iNKT) cells. The iNKT cell TCR is capable of recognizing several types of glycolipid Ags derived from microbial, environmental, or endogenous sources (4–7). Although NK cells are truly innate lymphocytes, we refer to iNKT cells as innate-like cells, because although they are bona fide T cells that mature in the thymus, they carry out very rapid effector responses.

In addition to TCR/CD1d-dependent activation of iNKT cells, Ag-independent activation of these lymphocytes also can occur: for example, in response to viruses or TLR ligands, which typically results in their exclusive production of IFN-γ, whereas TCR stimulation results in IL-4, as well as other cytokines. This “indirect” iNKT cell activation was shown to involve dendritic cells (DCs), and potentially other APCs, which produce cytokines in response to TLR triggering (3, 8–10). To this point, the key cytokines promoting iNKT cell activation in these experimental systems have been IL-12, IL-18, and IFN-I, with IL-12 playing a dominant role. This also holds true in the case of MCMV infection, with IL-12−/− mice being severely compromised for iNKT activation, whereas IL-18– or IFN-I signaling–deficient mice show only very modest reductions (3, 10). In these cases, in which no foreign lipid Ag is present to contribute to iNKT activation, IL-12 may synergize with TCR recognition of self-Ags presented by CD1d (11). However, iNKT cells are activated normally in MCMV-infected mice where CD1d expression is lacking or blocked (3, 10), providing evidence for a purely cytokine-driven activation process.

Despite being distinct cell lineages, NK and iNKT cells are similar in some respects, including the shared expression of NK cell receptors, dependence upon IL-15 signaling, overlapping tissue localization, and the ability to rapidly release copious amounts of cytokines, particularly IFN-γ, following infection. Consistent with these shared properties, both cell types have been implicated in the immune response to viral, bacterial, and parasitic infections. Further linkage comes from evidence showing that iNKT cells promote the secondary activation of NK cells through iNKT cell expression of CD40L, which mediates the activation of APCs (9, 12). Taken together, these data suggest that iNKT and NK cells could play similar, and potentially redundant, roles in innate defense to infection.

The critical role of NK cell–mediated protection during MCMV infection has been studied extensively (13, 14). In C57BL/6 (B6) mice, recognition of the MCMV m157 protein by NK cells expressing the Ly49H-activating receptor results in enhanced control of early replication (15). In turn, the lack of an Ly49H homolog in BALB/c mice results in significantly enhanced levels of early MCMV replication (16). At very early times of MCMV infection (8–12 h), a burst of IFN-I synthesis emanating from infected stromal cells promotes the activation of NK cell cytolytic activity (1, 17, 18). By ∼36 h, MCMV has completed its first replication cycle, and it subsequently triggers DC populations to produce IL-12 and IL-18, in addition to a second wave of IFN-I (19, 20). Although iNKT cells show markers of activation at 12 h (3), it is only at ∼36 h when a large proportion of NK and iNKT cells produce IFN-γ (2, 3, 10). Notably, plasmacytoid DCs are the primary source of innate cytokines at 36 h, leading to NK and iNKT cell IFN-γ production through TLR9-dependent recognition of MCMV (3, 21, 22).

In this study, we explored the regulation of NK and iNKT cells by APC-produced cytokines and determined whether they have distinct roles in antiviral control. We find that iNKT and NK cells have different cytokine-mediated activation requirements, and they contribute to MCMV innate defense in a nonredundant fashion.

Materials and Methods

Ethics statement

This study was carried out in strict accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocols were approved by the Institutional Animal Care and Use Committee of the La Jolla Institute for Allergy and Immunology (AP112-MK2-0410 and AP087-CB2-0110), which adheres to National Institutes of Health Office of Laboratory Animal Welfare and American Association for the Accreditation of Laboratory Animal Care guidelines. Our institute’s Office of Laboratory Animal Welfare assurance number is A3779-01, and our American Association for the Accreditation of Laboratory Animal Care accreditation number is 000840.

Mice

B6, B6.129S1-Il12btm1Jm/J (IL-12p40−/−), B6.129P2-Il18tm1Aki/J (IL-18−/−), and BALB/c mice were purchased from The Jackson Laboratory. CD1d1−/− mice were a kind gift from Dr. L. Van Kaer (Vanderbilt University, Nashville TN). Tlr9cpg1 mice were a kind gift from Dr. B. Beutler (University of Texas Southwestern Medical Center, Dallas, TX). BALB/c Jα18−/− mice were a kind gift from Dr. M. Taniguchi (Riken Research Center for Allergy and Immunology, Yokohama, Japan) and were maintained as Jα18+/− heterozygotes. Jα18+/+ and Jα18−/− littermates were used for MCMV infection experiments. 4get mice on the B6 background were a kind gift from Dr. R. Locksley (University of California San Francisco, San Francisco CA). All mice were housed in specific pathogen–free conditions.

Reagents and Abs

mAbs to the following mouse Ags were purchased from BD Biosciences, as purified or conjugates to FITC, Alexa Fluor 488, PE, PerCP-cyanin (Cy)5.5, PE-Cy7, allophycocyanin, or Becton Dickinson Horizon V450 or V500: TCR-β, CD11b, CD8α, NK1.1, CD11c, CD44, CD25, CD69, CD4, CD212 (IL-12Rβ1), IL-4, TNF, DX5, and IFN-γ. A mAb to CD218a (IL-18Rα) conjugated to Alexa Fluor 647 was purchased from BioLegend. PE-conjugated αGalCer-CD1d tetramers were generated in our laboratory, as previously described (9), and used to stain cell suspensions. Recombinant mouse IL-12 and IL-18 were purchased from R&D Systems. IFN-β was purchased from PBL Interferon Source.

No-touch iNKT cell isolation

Spleens from 4get mice were isolated at 8–10 wk of age and dissociated into single-cell suspensions, and RBCs were lysed using red cell lysing buffer (Sigma-Aldrich), washed, filtered, and counted. Single-cell preparations were stained with a custom mixture of biotin-labeled Abs containing anti-CD8α, CD11b, CD19, CD24, CD62L, B220, F4/80, Gr-1, and Ter119. Labeled cells were loaded onto a RoboSep cell separation system using a custom enrichment reagent kit and protocol, according to the manufacturer’s instructions (STEMCELL Technologies, Vancouver, BC, Canada). iNKT cells were enriched up to 20%, and eGFP bright cells were sorted using a FACSAria II (BD Biosciences). Sorted cells were 99% GFP+, with the purity of the cells collected being typically >91% iNKT cells, as assessed using CD1d tetramers loaded with αGalCer (Supplemental Fig. 1).

Cell preparation and cell culture

All cell preparations were conducted using complete IMDM (Invitrogen Life Technologies) supplemented with 10% FBS, penicillin-streptomycin-glutamine, and 2-ME. DCs were prepared using two systems. For GM-CSF–derived DCs, bone marrow was harvested and cultured in 10-cm bacterial dishes in complete IMDM supplemented with 20 ng/ml recombinant mouse GM-CSF (provided by Kyowa Hakko Kirin). On day 4, half of the media was replaced with complete IMDM containing GM-CSF (20 ng/ml). GM-CSF–derived bone marrow–derived DCs (BMDCs) were harvested on day 7. For Flt3 ligand (Flt3L)-derived DCs, bone marrow was harvested and cultured with 100 ng/ml recombinant human Flt3L (Amgen) for 8 d in T75 flasks at a concentration of 45 × 106 cells/25 ml. Special attention was paid not to disturb the cultures during the entire incubation period. DCs were treated with either 10 μg/ml type B CpG oligodeoxynucleotide (ODN) 1826 or control ODN 1982 (Alexis Biochemicals), infected with MCMV Smith strain (ATCC VR-1399) at a multiplicity of infection of 3, or mock infected, for 2 h. A total of 2.5 × 105 BMDCs was cultured or not with 5 × 104 sorted iNKT cells or NK cells for 48 h in a volume of 200 μl in a round-bottom 96-well plate at 37°C and 5% CO2 before cytokine detection, as described (23). Cell–cell contact assays were performed by culturing 5 × 104 sorted iNKT cells or NK cells with 200 μl conditioned BMDC culture supernatant. MCMV stocks were prepared from NIH-3T3 fibroblasts. NK cells were purified from splenocytes using NK1.1 PE-labeled mAb, followed by positive selection on a RoboSep cell separation system with a PE Selection kit (STEMCELL Technologies) and sorting for NK1.1+ TCRβ− cells on a FACSAria II. Typically, NK cells were >98% pure.

ELISA

A standard sandwich ELISA was performed to measure mouse IFN-γ, IL-12p70 (R&D Systems), IL-18 (MBL International), and type I IFNs (PBL Interferon Source), following the manufacturer’s instructions.

Viral infections

Salivary gland stocks of MCMV Smith strain were prepared essentially as described (24). BALB/c mice were infected with 1 × 104 PFU virus in 200 μl PBS via i.p. injection. Four days postinfection, mice were killed by CO2 asphyxiation, and organs were harvested and immediately snap-frozen in liquid nitrogen. To determine viral titers, organs were weighed and homogenized, and serial dilutions of homogenates were added to monolayers of NIH-3T3 fibroblasts plated in 24-well plates. Plates were spun at 524 × g for 10 min prior to incubation at 37°C, which increased the sensitivity of the plaque assays ∼6–10 fold. Cells were fixed with formalin, and plaques were visualized with 0.1% crystal violet and quantified.

Cell depletion

Jα18+/+ BALB/c and Jα18−/− BALB/c mice were depleted of NK cells by injecting 50 μl anti-asialo GM1 rabbit polyclonal Ab (Wako) on day −1 of MCMV infection. NK cell depletion was verified in individual mice by flow cytometry after bleeding and just prior to infection, using a DX5-specific mAb; nonspecific cellular depletion was not observed. Additionally, control mice were Ab depleted of CD8+ T cells (clone 2.43). Depletion was verified by flow cytometric staining with anti-CD8 clone 53-6.7 after bleeding and just prior to infection.

Real-time PCR analysis

Total RNA was isolated from sorted splenic iNKT cells and NK cells using an RNeasy Plus kit (QIAGEN). After DNase I treatment (Ambion), RNA was reverse transcribed using an iScript cDNA synthesis kit (Bio-Rad). Real-time PCR was performed using SYBR Green (Bio-Rad) on a LightCycler 480 Real-Time PCR System with PCR primer pairs for il-12rb1, il-12rb2, il-18r1, il-18rab, ifnar1, and ifnar2 from SABiosciences RT2 qPCR Primer Assays (QIAGEN). Target gene expression was assessed by comparative cycling threshold analysis, with L32 mRNA expression as a control. Data are presented in arbitrary units.

Flow cytometry and intracellular cytokine staining

Lymphocytes were stained with αGalCer/CD1d tetramers labeled with streptavidin-allophycocyanin, anti-NK1.1–PerCP PE-cyanin (PECy)5, anti-CD8–PECy7, anti-CD11b–PECy7, anti-TCRβ–allophycocyanin–AF750, and CD25-FITC. All Abs and isotype controls, with the exception of anti-TCRβ–allophycocyanin–AF750 (eBioscience), were purchased from BD Biosciences. Cells were fixed and permeabilized using Cytofix/Cytoperm buffer and stained for intracellular IFN-γ with PE-labeled clone XMG1.2. The data were collected on a LSR II flow cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star).

Statistics

Differences between groups were evaluated using appropriate statistical tests, as recommended by the La Jolla Institute Bioinformatics Core Facility. Results are expressed as mean ± SD, except where indicated. The p values < 0.05 were considered significant.

Results

Infected DC subtypes differ in their activation of iNKT cells

We demonstrated previously in cocultures that MCMV infection of Flt3L-expanded BMDCs promoted purified iNKT cells to secrete IFN-γ, but not other cytokines, such as the IL-4 that these cells secrete when activated with glycolipid Ags. This did not occur when DCs from IL-12 p40–deficient mice were infected. Together with other studies indicating that IL-12 participates in the activation of iNKT cells (9, 23, 25, 26), these data strongly implicated the action of IL-12, rather than IL-23, in promoting iNKT cell effector function.

To test whether other DC types can activate iNKT cells upon MCMV infection, DCs derived from GM-CSF–treated bone marrow cultures also were analyzed. GM-CSF or Flt3L-expanded BMDCs were infected with MCMV, treated with CpG ODN, because the activation of IL-12 secretion was shown to be TLR9 dependent, or mock treated prior to the addition of purified iNKT cells or NK cells, and IFN-γ production in the culture supernatants was measured ∼48 h later. To assess whether GM-CSF–derived BMDCs activated iNKT cells in a CD1d-dependent manner, Cd1d1−/− BMDCs also were tested. In addition, GM-CSF– and Flt3L-derived BMDCs generated from Tlr9cpg/cpg mice were included as negative controls. For Flt3L-expanded BMDCs, MCMV infection and CpG ODN treatment resulted in the production of IFN-γ by both iNKT cells and NK cells (Fig. 1A, 1C). This activation by MCMV and CpG ODN was independent of CD1d expression but required TLR9 (Supplemental Fig. 2), as expected, based on our previous observations (3). CpG ODN treatment of GM-CSF–derived BMDCs also resulted in TLR9-dependent IFN-γ production by iNKT and NK cells and, again, was independent of CD1d (Fig. 1B, Supplemental Fig. 2). However, when GM-CSF–derived BMDCs were infected with MCMV, NK cells were induced to secrete IFN-γ, whereas iNKT cells were unresponsive (Fig. 1D).

FIGURE 1.
FIGURE 1.

IFN-γ production by iNKT and NK cells in DC cocultures. Flt3L-derived BMDCs (A) or GM-CSF–derived BMDCs (B) from the indicated mouse strains were treated with CpG-ODN or control and then cultured with purified iNKT cells or NK cells for 48 h. Flt3L-derived BMDCs (C) or GM-CSF–derived BMDCs (D) from the indicated mouse strains were mock or MCMV infected and cultured with purified iNKT cells or NK cells for 48 h. Supernatants were analyzed for IFN-γ by ELISA. Shown is a representative experiment of six performed. ELISA results represent the mean of one experiment with three replicate cultures measured in triplicate. Error bars represent SEM (n = 9 for each set of BMDCs).

Cell contact is not required for iNKT cell activation by MCMV-infected DCs

MCMV has developed numerous mechanisms to avoid detection by the immune system (1). We (27) previously observed that MCMV-infected GM-CSF–derived BMDCs potently inhibit T cell proliferation through viral modulation of cosignaling pathways, raising the possibility that iNKT cell activation may be similarly suppressed by MCMV. To test this, or any other mechanism dependent on cell contact, supernatants from MCMV-infected Flt3L- or GM-CSF–derived BMDCs were evaluated for their ability to activate iNKT cells. Supernatants from Flt3L-expanded BMDCs activated both NK cells and iNKT cells, indicating that cell contact was not required for the induction of IFN-γ synthesis by these innate lymphocytes. In contrast, the supernatant from GM-CSF–derived BMDCs activated NK cells but not iNKT cells (Fig. 2A). To assess whether a soluble factor released from MCMV-infected GM-CSF–derived BMDCs was inhibiting iNKT cell activation, Flt3L- and GM-CSF–expanded BMDC–derived supernatants were mixed prior to iNKT cell addition. Even at a 10:1 (GM-CSF/Flt3L) ratio of BMDC supernatants, iNKT cells were activated, demonstrating the absence of a potent inhibitor and, therefore, strongly suggesting that an activating factor was absent in the supernatants derived from MCMV-infected GM-CSF–derived BMDCs (Fig. 2B).

FIGURE 2.
FIGURE 2.

Activation of iNKT and NK cells by MCMV-infected DCs does not require cell–cell contact. (A) IFN-γ production by iNKT cells or NK cells cultured for 48 h with supernatants from Flt3L- or GM-CSF–derived BMDCs infected with MCMV. (B) iNKT cells were cultured with individual supernatants, a 1:1 mixture of the two APC types (Flt3L/GM-CSF Cells), or with a 10:1 ratio of supernatants from MCMV-infected DCs (Flt3L/GM-CSF Supes). IFN-γ ELISA results represent the mean of one experiment with two replicate cultures measured in triplicate. Error bars represent SEM (n = 6 for each culture condition). Shown is a representative experiment of three performed.

Infected DC subtypes produce different cytokines

Our results indicated that a secreted factor(s) required for activation of iNKT cells was absent in supernatants from GM-CSF–derived BMDCs infected with MCMV. Our previous work (3, 9), and that of other investigators (10), indicated that IL-12, IL-18, and IFN-I can all contribute to the activation of iNKT cells, suggesting that the absence of one or more of these factors could be responsible. To address this, Flt3L- and GM-CSF–derived BMDCs were treated with CpG ODN and/or infected with MCMV prior to isolation of supernatants for analysis of IL-12p70, IL-18, and IFN-α by ELISA. Following stimulation with CpG ODN, both Flt3L- and GM-CSF–derived DCs produced approximately similar amounts of these cytokines (Fig. 3A). However, when GM-CSF–derived BMDCs were infected with MCMV, virtually no IL-12 was produced. In contrast, IFN-α production was similar for the two infected DC types, and the amount of IL-18 was significantly higher from infected GM-CSF–derived DCs compared with Flt3L-derived DCs (Fig. 3B). Importantly, use of an MCMV-GFP virus indicated that the infection efficiency of these two BMDC cultures was similar (data not shown), a fact also reflected by the high-level production of IL-18 and IFN-α by both BMDC types. In summary, although GM-CSF–derived DCs have the capacity to produce IL-12 upon TLR9 activation with CpG ODN, they produced virtually none when exposed to MCMV, despite producing both IFN-I and IL-18 at high levels in response to viral infection.

FIGURE 3.
FIGURE 3.

Cytokine production by DCs in response to CpG or MCMV. (A) Flt3L- or GM-CSF–derived BMDCs were treated with CpG or control ODN (Non) for 48 h, and supernatants were analyzed for IL-12 p70, IL-18, or IFN-α by ELISA. (B) Same as in (A), but DC subtypes were mock infected or MCMV infected prior to analysis of cytokine levels by ELISA. Data are representative of four independent experiments. Results show the mean of one experiment with three replicate cultures of DCs measured in duplicate for ELISA. Error bars represent SEM (n = 6 for each culture condition).

iNKT cells and NK cells differ in their sensitivity to cytokines

The data obtained from MCMV infection of the two DC subtypes suggested that iNKT cells are more dependent on IL-12 for their activation, whereas, for NK cells, IL-12 may not be essential if other cytokines from innate immune cells are present. To test this, IFN-γ production by purified NK and iNKT cell populations was measured after treatment with various amounts of recombinant cytokines in the absence of DCs. When the concentrations of IL-12 or IL-18 were limiting (<10 pg/ml), these two cytokines functioned synergistically to activate both iNKT cells and NK cells to produce IFN-γ. For example, when iNKT cells and NK cells were cultured with 10 pg/ml of IL-12, both cell types produced large, and roughly equivalent, amounts of IFN-γ across a wide range of IL-18 concentrations (Fig. 4A). These data suggest that this dose of IL-12, with IL-18 at or below the detection limit of the ELISA assay, induced a maximal response of both iNKT and NK cells. In contrast, fixing the IL-18 at 10 pg/ml and titrating in very limiting amounts of IL-12 (∼0.1–10 pg/ml) revealed that iNKT cells were even more sensitive than NK cells to synergistic activation by combinations of these two cytokines (Fig. 4D).

FIGURE 4.
FIGURE 4.

Activation threshold of purified iNKT and NK cells to cytokines. Purified iNKT cells or NK cells were cultured with 10 pg/ml IL-12 and serially diluted amounts of IL-18 (A) or IFN-β (B). Purified iNKT cells or NK cells were cultured with 10 pg/ml IL-18 and serially diluted IFN-β (C) or IL-12 (D). (E) Purified iNKT cells or NK cells were cultured with various doses of IL-12, IL-18, or IFN-β. Forty-eight hours later, supernatants were analyzed for IFN-γ by ELISA. Data are representative of three independent experiments and show the mean of one experiment with three replicate cultures measured in triplicate for ELISA. Error bars represent SEM (n = 9 for each culture condition).

When a dose titration of IFN-β was performed in the presence of 10 pg/ml IL-12, similar to what was done for IL-18, an increase in IFN-γ production by both NK cells and iNKT cells again was observed. However, under these circumstances, NK cells were more sensitive to the synergistic effects of IFN-β and IL-12 than were iNKT cells (Fig. 4B). IFN-β alone could not induce IFN-γ secretion from iNKT cells, regardless of the dose used, whereas it could promote NK cells to secrete IFN-γ when added at higher levels (Fig. 4E, right panel). Surprisingly, a similar observation was made with IL-18 and IL-12 (Fig. 4E). Neither of these cytokines alone, even at 1 ng/ml, could activate purified iNKT cells to secrete significant levels of IFN-γ. In contrast, NK cells activated by IL-18 or IL-12 alone produced high levels of IFN-γ, even at doses as low as 100 and 10 pg/ml, respectively (Fig. 4E). To attempt to exclude a potential contribution of anti-NK1.1 mAb treatment on “priming” NK cell activation during the purification process, several of these experiments were repeated using DX5+CD3− cells isolated from spleens; similar results were obtained (data not shown). Additionally, NK and iNKT cells were purified by cell sorting from BALB/c mice using identical Ab staining panels except TCR-β expression was used to identify iNKT cells. As we observed for C57BL/6 mice, BALB/c NK cells produced IFN-γ when single innate cell–derived cytokines were added to cultures, whereas iNKT cells remained unresponsive, even at high doses of either IL-12 or IL-18 alone (data not shown).

We also determined whether iNKT cells secrete IFN-γ in response to IFN-β and IL-18. Purified iNKT and NK cells were cultured with 10 pg/ml of IL-18, a dose that synergizes with IL-12 to induce their robust activation (Fig. 4D), together with various amounts of IFN-β. Interestingly, iNKT cells were not activated in response to the combination of IL-18 and IFN-β, whereas NK cells were markedly stimulated (Fig. 4C). IFN-γ secretion by NK and iNKT cells has been associated with the ability of IL-12 to induce STAT4 activation (26, 28). It was demonstrated previously that NK cells express a high basal level of STAT4 bound to the IFN-I receptor, and, upon exposure to IFN-I, STAT4 is displaced, leading to IFN-γ secretion (28). In agreement with these findings, when NK cells were exposed to high doses of IFN-β, IL-12, or IL-18, STAT4 phosphorylation was detectable within 1 h (Supplemental Fig. 3A). Meanwhile, STAT4 phosphorylation in iNKT cells was also detectable at high doses of IL-12 (1 ng/ml), giving a bimodal pattern seen in an earlier study (26). However, this was absent at lower IL-12 concentrations (10 pg/ml), which still robustly activated STAT4 phosphorylation in NK cells, and no STAT4 phosphorylation was observed in iNKT cells cultured with high doses of IL-18 or IFN-β alone (Supplemental Fig. 3B). However, STAT4 phosphorylation was detectable in iNKT cells at low IL-12 concentrations when combined with IL-18 or IFN-β. Consequently, the pattern of STAT4 phosphorylation in these two cell types reflects their requirements for the induction of IFN-γ synthesis. Taken together, these results indicate that iNKT cells and NK cells have different cytokine requirements to activate their production of IFN-γ. These results also explain the inability of MCMV-infected GM-CSF–derived BMDCs to activate iNKT cells, despite high levels of virus-induced IFN-β and IL-18 production, because IL-12 production was absent.

Differential sensitivity of NK and iNKT cells to IL-12 in vivo

The amounts of IL-12 and IL-18 induced at ∼36 h after MCMV infection in the sera are in the ng/ml range (29), far above the limiting amounts used to stimulate purified iNKT and NK cells in our in vitro experiments shown in Fig. 4. IFN-β also reaches high levels in the serum at 36 h, but it also exhibits a much earlier peak at 6–8 h in response to MCMV infection of splenic stromal cells (17). Because purified NK cells produced IFN-γ in response to treatment with any one of the single cytokines at ng/ml doses (Fig. 4E), we postulated that in mice genetically deficient for only one of the cytokines, compared with iNKT cells, the NK cells might show enhanced activation upon MCMV infection in vivo. Consistent with previous results (3, 10), iNKT cell IFN-γ synthesis in MCMV-infected mice was strictly dependent on IL-12 when measured by intracellular cytokine staining on cells analyzed directly ex vivo. Although the NK cell response was decreased in the absence of IL-12, these cells still produced measurable levels of IFN-γ in the spleen following MCMV infection (Fig. 5). In contrast, spleen NK cells were unable to produce IFN-γ in the absence of IL-18, whereas iNKT cell IFN-γ synthesis was not diminished (Fig. 5). Therefore, the cytokine requirements for the activation of NK cells to secrete IFN-γ in vivo may be more stringent than when these cells are purified and cultured with cytokine. Interestingly, NK cells isolated from the livers of infected mice retained the ability to secrete IFN-γ in the absence of IL-18, in agreement with previous studies (29), although the response was reduced compared with wild-type (WT) controls (data not shown).

FIGURE 5.
FIGURE 5.

iNKT and NK cell activation following MCMV infection in vivo. (A) IFN-γ expression by NK and iNKT cells from the spleen of indicated mouse strains 36 h postinfection with MCMV (open graphs) compared with isotype controls (shaded graphs). Graphs are representative plots of three independent experiments with three to eight mice/group. (B) IFN-γ mean fluorescence intensity (MFI) of three to eight mice/group 36 h postinfection with MCMV from a minimum of three independent experiments. Error bars represent SEM (n = 3–8/group). *p ≤ 0.01, versus WT NK or iNKT cells, equal-variance Student t test.

Expression of cytokine receptors by NK and iNKT cells

One possible explanation for the differing sensitivity of NK cells and iNKT cells to combinations of IL-12, IL-18, and IFN-I could be varying levels of cytokine receptor expression. Quantitative PCR analysis of cytokine receptor mRNA levels, as well as protein expression for those receptors where Abs are available, was performed in these two cell populations. Compared with NK cells, both Il12rb1 and Il12rb2 expression levels were higher in freshly purified iNKT cells (∼2- and 3-fold, respectively) (Fig. 6A), whereas Il18rb1 and Il18rap expression was ∼3-fold lower (Fig. 6B). Expression of ifnar1 was ∼2-fold higher in NK cells compared with iNKT cells, although the two cell populations expressed similar levels of ifnar2 (Fig. 6C). To verify whether the increase in mRNA expression correlated with protein, IL-12Rβ2 and IL-18Rα expression were analyzed by flow cytometry on iNKT and NK cells (Fig. 6D). Although IL-12Rβ2 was readily detectable on tetramer−/CD3− NK cells, the highest level of expression was on tetramer+/CD3+ iNKT cells. IL-12Rβ1 receptor expression was not detectable on MHC class I–reactive CD8+ αβ or γδ T cells. Expression of the IL-18Rα-chain was detectable on NK cells, but this was not the case for iNKT cells, conventional αβ T cells, or γδ T cells in the spleen (Fig. 6D) (26, 30). Analysis of IL-12R and IL-18R chain expression in iNKT and NK cells isolated from BALB/c mice revealed identical results to those seen in C57BL/6 mice, suggesting that the differential sensitivity of these two cell populations to these innate cytokines may be a general characteristic (data not shown). These trends in the amounts of cytokine receptor expression are consistent with the increased sensitivity of iNKT cells to IL-12 in the presence of IL-18, as well as with the enhanced ability of NK cells to respond to IL-18 + IFN-I. However, the results do not exclude the contribution of additional factors to the differential cytokine sensitivity of the two lymphocyte populations.

FIGURE 6.
FIGURE 6.

Cytokine receptor expression by iNKT and NK cells. Purified splenic iNKT or NK cells were analyzed by quantitative real-time PCR for their expression of IL-12 (A), IL-18 (B), and IFN-I (C) receptor mRNAs. Bar graphs show each receptor chain or receptor-associated protein mRNA level relative to L32 ribosomal protein gene expression. Data are the mean of two independent experiments performed in duplicate. Error bars represent SEM (n = 4). (D) Flow cytometric analysis of CD19− splenocytes for IL-12Rβ2 (left panel) and IL-18Rα (right panel) expression on tetramer+ iNKT cells (solid line), tetramer− NK cells (dotted line), and tetramer− αβ and γδ CD8+ T cells (shaded graph). Graphs are representative plots of three mice/group from two independent experiments. *p < 0.05, unpaired Student t test.

iNKT cells contribute to antiviral innate defense

Previous reports (10, 31) indicated that mice genetically deficient for iNKT cells (Jα18−/−) can control MCMV replication normally. However, this has only been examined in B6 mice, which have very robust NK cell–mediated MCMV defenses due to expression of the Ly49H-activating receptor that interacts with the viral m157 protein (32, 33). In contrast, NK cell–mediated defense against MCMV in BALB/c mice differs significantly from that in B6 mice as the result of a different expression pattern of activating and inhibitory Ly49 family receptors (34), including a lack of Ly49H. Such differences, including the lack of LY49H, also pertain to many other inbred strains and WT mice, as well (35). Consequently, we tested whether iNKT cells contribute to MCMV innate defense in BALB/c mice lacking Ly49H by analyzing viral replication in the spleen and liver of Jα18−/− mice and WT littermate controls. We found that Jα18−/− mice had ∼3–5-fold higher levels of MCMV replication in both of these organs, indicating that iNKT cells play a role in the early control of MCMV replication in some mouse strains (Fig. 7A, 7B). In separate experiments, we compared MCMV replication in Jα18−/− BALB/c mice that were depleted of NK cells using anti-asialo GM1 antisera. Depleting NK cells in mice that lack iNKT cells resulted in a further enhancement of viral replication, which was more dramatic in the liver than the spleen (Fig. 7C, 7D). Because anti-asialo GM1 antisera can potentially deplete activated CD8+ T cells, we verified that MCMV replication was unaltered in BALB/c mice at day 4 following CD8 depletion using CD8-specific Abs (Supplemental Fig. 4). These data establish that iNKT cells contribute significantly, and nonredundantly, to early or innate defense against MCMV in BALB/c mice.

FIGURE 7.
FIGURE 7.

iNKT and NK cells mediate nonredundant, innate defense against MCMV. Groups of BALB/c Jα18+/+ iNKT cell-sufficient (WT) or BALB/c Jα18−/− iNKT cell–deficient littermate control mice were infected with MCMV, and viral PFU levels were determined in the liver (A) and spleen (B) 4 d later. (C and D) BALB/c Jα18−/− iNKT cell–deficient mice were depleted of NK cells (NK dep) or not and subsequently infected with MCMV. MCMV replication levels in the liver (C) and spleen (D) were determined 4 d later. For (A)–(D), results are representative of 12 mice/group from a minimum of two independent experiments. Error bars represent SEM (n = 12/group). Note that the data shown in (C) and (D) are from experiments carried out at separate times from those in (A) and (B); therefore, the absolute values of PFU cannot be compared directly. Differences between groups were evaluated for statistical significance by the two-tailed unpaired Student t test.

Discussion

The importance of NK cells for innate immune responses to viral infections is well established; although iNKT cells also can be activated after exposure to viruses, in the absence of synthetic glycolipid Ag activation, there are relatively few examples in which these responses are protective. However, consistent with an important antiviral role for iNKT cells is the uncontrolled infection in response to the varicella-zoster vaccine in rare individuals lacking iNKT cells or CD1d and the finding that several viruses downregulate CD1d expression, suggesting evolutionary pressure for an immune-evasion mechanism targeting iNKT cells (36–42). In this study, we show that NK cells and iNKT cells exhibit cytokine-induced activation in response to infection with MCMV, but that their cytokine requirements for activation differ, and that both cell types contribute to host protection. These data definitively show that iNKT and NK cells are differentially activated in the context of pathogen infection, perhaps in response to distinct APC subsets or other influences, and indicate they may play distinct roles in the control of infection.

iNKT cells express an invariant TCR that recognizes glycolipids, raising the issue as to how they participate in the immune response to viruses. However, several additional modes of activation allow iNKT cells to be activated by microbes in the absence of a microbial glycolipid presented by CD1d. First, activation can result from the presentation of self-glycolipid Ags by CD1d. The relevant self-Ags may include phospholipids and various types of glycosphingolipids (43), but this response also requires IL-12 production by activated APCs. iNKT cells activated in this way predominantly produce IFN-γ, in contrast to the diverse Th1 and Th2 cytokines that are elicited by strong glycolipid agonists for the TCR (e.g., αGalCer) (6, 26). An exception to the requirement for IL-12 is one study (8) in which APCs were activated with CpG ODN; in this case, IFN-I was required as the cofactor for iNKT cell activation mediated by self-Ag.

Second, an additional pathway for iNKT cell activation is purely cytokine driven and does not require concomitant TCR engagement with any Ag presented by CD1d. In studying MCMV infection, we showed that activation by TLR9-mediated signaling leads to IL-12 release by APCs, resulting in iNKT cell production of IFN-γ by ∼36 h postinfection. CD1d expression was not required, and IL-18 had a minimal role (3). Wesley et al. (10) reported similar results, but they observed a contribution of IFN-I for optimal iNKT cell activation during MCMV infection. In addition, we saw previously that iNKT cells increase expression of the activation marker CD69 as early as 12 h after MCMV infection, which is strictly dependent upon IFN-I produced by MCMV-infected splenic stromal cells (44). However, the activated iNKT cells do not produce IFN-γ at this early time, at least as assessed by intracellular cytokine staining directly ex vivo, although it remains to be determined whether lower amounts of IFN-γ are produced or whether other effector functions are acquired. Interestingly, this “initial” IFN-I produced by stromal cells at 8–12 h of MCMV infection is sufficient to promote IFN-γ production by some NK cells (44), consistent with our results that IFN-β alone can induce IFN-γ production by purified NK cells but not iNKT cells (Fig. 4).

The engagement of activating NK receptors, such as NK1.1, CD94/NKG2C, or NKG2D, provides a third, TCR-independent pathway for iNKT cell activation that is illustrative of their innate- or NK cell–like function (45–47). However, because productive MCMV infection inhibits cell surface expression of ligands for activating NK receptors (48, 49), it is likely that this mechanism of iNKT cell activation would be muted or absent when directly encountering virus-infected cells.

Although the critical importance of IL-12 for iNKT cell activation has been established, in this study we characterized, in detail, the cytokine-activation requirements for iNKT cells compared with NK cells. Interestingly, IL-12 alone did not induce iNKT cells to produce IFN-γ in vitro. This is in agreement with our earlier findings that both IL-12 and IL-18 were required for iNKT cell stimulation in response to LPS in vitro or in vivo (9). Furthermore, activation-induced arrest of iNKT cells in liver sinusoids was most evident when IL-12 and IL-18 were injected together (50). In contrast, our earlier studies (3) of the response to MCMV demonstrated the importance of IL-12, but the absence of IL-18 had only a marginal effect. In this study, we showed that either IL-18 or IFN-I can synergize with IL-12 to activate iNKT cells in vitro. Therefore, in the context of MCMV infection, when high levels of IFN-I are produced, it is likely that IL-18 is largely redundant for iNKT cell activation.

We also now show that the cytokine requirements for NK cell stimulation in vitro to produce IFN-γ are different, and generally less stringent, than are those for iNKT cells. This is most noticeable in the ability of NK cells to respond to single cytokines when added at higher concentrations, as well as their responses to combinations of either IL-18 + IFN-β or IL-12 + IFN-β. The only exception to this trend was that iNKT cells were more responsive to IL-18 in the presence of low IL-12 levels. The differential cytokine sensitivity of these two cell types correlated with the expression of cytokine receptors, suggesting that this could be a contributing factor. Furthermore, our in vivo data are generally consistent with the in vitro results. iNKT cells were unable to produce IFN-γ in IL-12p40−/− mice, whereas NK cells remained partially responsive, presumably as the result of suboptimal activation by IL-18 and IFN-Ι. Somewhat surprisingly, however, production of IFN-γ by NK cells in vivo was found to be highly dependent on IL-18 in the spleen. Based on our in vitro experiments, we would not have predicted this result, because other cytokines from innate cells were present at the time NK cells were analyzed. However, this finding is consistent with previous work showing that IL-18 is required for NK cell production of IFN-γ in the spleen and for expansion of Ly49H+ NK cells in vivo, as well as for IFN-γ production by NK cells cocultured with MCMV-infected GM-CSF–derived BMDCs (29, 51, 52). It should be noted that, although the in vitro work defines the capacity/potential for NK or iNKT cells to respond to minimal cytokine concentrations, negative regulatory influences would be diminished in this set-up. For example, when cytokines are added to purified NK cells, the influence of cell surface molecules expressed by APCs acting on NK cell inhibitory receptors will be minimized. Furthermore, local cytokine concentrations may not reflect serum concentrations, and the existence of such local effects is supported by the organ-specific requirement for IL-18 for MCMV-mediated NK cell activation in the spleen but not the liver (29).

Using two common methods for generating DC populations in vitro, we found that GM-CSF–derived BMDCs were capable of activating NK cells, but not iNKT cells, when infected with MCMV. The defining difference was not in the efficiency of the infection or in their ability to generally detect virus and subsequently produce cytokines. Instead, the defect was in the absence of IL-12 in the supernatant following MCMV infection. We do not know why GM-CSF–derived BMDCs do not produce IL-12 after MCMV infection, because they are competent to do so after exposure to CpG ODN. Regardless, as microbes evolve they are able to evade detection in numerous ways, including the blocking of production of critical cytokines from innate cells, and our results are consistent with previous work showing that productive infection of GM-CSF–derived BMDCs with MCMV inhibits their ability to produce IL-12 in response to secondary TLR activation (53). The results from microarray analysis of gene expression suggest that GM-CSF–derived BMDCs are likely most similar to inflammatory monocytes (54). However, the phenotype of DC subsets changes rapidly postinfection, and it is uncertain whether there is a DC population in vivo after MCMV infection that truly corresponds to GM-CSF–derived BMDCs. Nevertheless, it is certain from many studies that the cytokines produced by innate cells following infection will depend on the infecting agent and the cell subset that is activated. Therefore, extrapolating the MCMV results to other infections, it seems reasonable to propose that circumstances exist where the balance of activated APC types and cytokines produced could favor the preferential activation of NK or iNKT cells.

There are only a few examples where iNKT cells were shown to affect viral clearance in the absence of pharmacologic activation by glycolipid Ag (55); therefore, a key issue is whether iNKT cell responses are important for limiting MCMV. We observed that viral replication was elevated in BALB/c mice that lack iNKT cells, especially in the liver. Depletion of NK cells from Jα18−/− BALB/c mice led to a further increase in replication. These data indicate that NK cell activation postinfection is not completely dependent on iNKT cells, although, in some cases, activation of iNKT cells can contribute to NK cell stimulation; furthermore, considering the effects of NK cell depletion in iNKT cell–deficient mice, NK cell contribution to viral clearance is quantitatively quite similar to iNKT cell–mediated control. Therefore, NK cells and iNKT cells each play unique roles in antiviral defense in mice. Importantly, previous work showed that B6 mice lacking iNKT cells do not show enhanced replication of MCMV, and we saw similar results in Jα18−/− mice generated in this strain (data not shown). This could be due to several factors, including the more robust, Ly49H-dependent NK cell response in B6 mice or perhaps the differential effector cytokine production by iNKT cells in BALB/c mice compared with B6 mice. Because human CMV infection impacts the NK cell repertoire differentially in people of varying genetic backgrounds, perhaps it is not surprising that iNKT cells show a differential importance in controlling MCMV replication in mouse strains, because this is certainly the case for MCMV and NK cells (56). Interestingly, as noted above, a drastic reduction in the number of iNKT cells and decreased CD1d expression have been associated, in a few cases, with disseminated varicella infection following the administration of the vaccine strain (36, 37). Therefore, we must consider the possibility that iNKT cell activity is particularly important in humans for control of infections by Herpesviridae.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank A. Khurana for assistance with the generation of CD1d tetramers, Josh Chong for assistance with plaque assays, Shelby Johnson for assistance with cell isolation and cell sorting, and Petra Krause for scientific and technical assistance for gene quantification.

Footnotes

  • ↵1 M.K. and C.A.B. share senior authorship.

  • This work was supported by National Institutes of Health Grants R01 AI69296 and R37 AI71922 (both to M.K.), R21 AI076864 (to C.A.B. and M.K.), and F32 AI80087 (to A.J.T.).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    B6
    C57BL/6
    BMDC
    bone marrow–derived dendritic cell
    DC
    dendritic cell
    Flt3L
    Flt3 ligand
    iNKT
    invariant NKT
    MCMV
    mouse CMV
    ODN
    oligodeoxynucleotide
    WT
    wild-type.

  • Received March 27, 2013.
  • Accepted February 10, 2014.

References

  1. ↵
    1. Loewendorf A.,
    2. C. A. Benedict
    . 2010. Modulation of host innate and adaptive immune defenses by cytomegalovirus: timing is everything. J. Intern. Med. 267: 483–501.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Orange J. S.,
    2. B. Wang,
    3. C. Terhorst,
    4. C. A. Biron
    . 1995. Requirement for natural killer cell-produced interferon gamma in defense against murine cytomegalovirus infection and enhancement of this defense pathway by interleukin 12 administration. J. Exp. Med. 182: 1045–1056.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Tyznik A. J.,
    2. E. Tupin,
    3. N. A. Nagarajan,
    4. M. J. Her,
    5. C. A. Benedict,
    6. M. Kronenberg
    . 2008. Cutting edge: the mechanism of invariant NKT cell responses to viral danger signals. J. Immunol. 181: 4452–4456.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Brigl M.,
    2. M. B. Brenner
    . 2004. CD1: antigen presentation and T cell function. Annu. Rev. Immunol. 22: 817–890.
    OpenUrlCrossRefPubMed
    1. Kronenberg M.
    2005. Toward an understanding of NKT cell biology: progress and paradoxes. Annu. Rev. Immunol. 23: 877–900.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Bendelac A.,
    2. P. B. Savage,
    3. L. Teyton
    . 2007. The biology of NKT cells. Annu. Rev. Immunol. 25: 297–336.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Wingender G.,
    2. P. Rogers,
    3. G. Batzer,
    4. M. S. Lee,
    5. D. Bai,
    6. B. Pei,
    7. A. Khurana,
    8. M. Kronenberg,
    9. A. A. Horner
    . 2011. Invariant NKT cells are required for airway inflammation induced by environmental antigens. J. Exp. Med. 208: 1151–1162.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Paget C.,
    2. T. Mallevaey,
    3. A. O. Speak,
    4. D. Torres,
    5. J. Fontaine,
    6. K. C. Sheehan,
    7. M. Capron,
    8. B. Ryffel,
    9. C. Faveeuw,
    10. M. Leite de Moraes,
    11. et al
    . 2007. Activation of invariant NKT cells by toll-like receptor 9-stimulated dendritic cells requires type I interferon and charged glycosphingolipids. Immunity 27: 597–609.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Nagarajan N. A.,
    2. M. Kronenberg
    . 2007. Invariant NKT cells amplify the innate immune response to lipopolysaccharide. J. Immunol. 178: 2706–2713.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Wesley J. D.,
    2. M. S. Tessmer,
    3. D. Chaukos,
    4. L. Brossay
    . 2008. NK cell-like behavior of Valpha14i NK T cells during MCMV infection. PLoS Pathog. 4: e1000106.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Cohen N. R.,
    2. R. V. Tatituri,
    3. A. Rivera,
    4. G. F. Watts,
    5. E. Y. Kim,
    6. A. Chiba,
    7. B. B. Fuchs,
    8. E. Mylonakis,
    9. G. S. Besra,
    10. S. M. Levitz,
    11. et al
    . 2011. Innate recognition of cell wall β-glucans drives invariant natural killer T cell responses against fungi. Cell Host Microbe 10: 437–450.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Tyznik A. J.,
    2. E. Farber,
    3. E. Girardi,
    4. A. Birkholz,
    5. Y. Li,
    6. S. Chitale,
    7. R. So,
    8. P. Arora,
    9. A. Khurana,
    10. J. Wang,
    11. et al
    . 2011. Glycolipids that elicit IFN-γ-biased responses from natural killer T cells. Chem. Biol. 18: 1620–1630.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Salazar-Mather T. P.,
    2. K. L. Hokeness
    . 2006. Cytokine and chemokine networks: pathways to antiviral defense. Curr. Top. Microbiol. Immunol. 303: 29–46.
    OpenUrlPubMed
  14. ↵
    1. Pyzik M.,
    2. E. M. Gendron-Pontbriand,
    3. S. M. Vidal
    . 2011. The impact of Ly49-NK cell-dependent recognition of MCMV infection on innate and adaptive immune responses. J. Biomed. Biotechnol. 2011: 641702.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Babić M.,
    2. A. Krmpotić,
    3. S. Jonjić
    . 2011. All is fair in virus-host interactions: NK cells and cytomegalovirus. Trends Mol. Med. 17: 677–685.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Scalzo A. A.,
    2. M. G. Brown,
    3. D. T. Chu,
    4. J. W. Heusel,
    5. W. M. Yokoyama,
    6. C. A. Forbes
    . 1999. Development of intra-natural killer complex (NKC) recombinant and congenic mouse strains for mapping and functional analysis of NK cell regulatory loci. Immunogenetics 49: 238–241.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Schneider K.,
    2. A. Loewendorf,
    3. C. De Trez,
    4. J. Fulton,
    5. A. Rhode,
    6. H. Shumway,
    7. S. Ha,
    8. G. Patterson,
    9. K. Pfeffer,
    10. S. A. Nedospasov,
    11. et al
    . 2008. Lymphotoxin-mediated crosstalk between B cells and splenic stroma promotes the initial type I interferon response to cytomegalovirus. Cell Host Microbe 3: 67–76.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Shellam G. R.,
    2. J. E. Allan,
    3. J. M. Papadimitriou,
    4. G. J. Bancroft
    . 1981. Increased susceptibility to cytomegalovirus infection in beige mutant mice. Proc. Natl. Acad. Sci. USA 78: 5104–5108.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Dalod M.,
    2. T. Hamilton,
    3. R. Salomon,
    4. T. P. Salazar-Mather,
    5. S. C. Henry,
    6. J. D. Hamilton,
    7. C. A. Biron
    . 2003. Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. J. Exp. Med. 197: 885–898.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Zucchini N.,
    2. G. Bessou,
    3. S. H. Robbins,
    4. L. Chasson,
    5. A. Raper,
    6. P. R. Crocker,
    7. M. Dalod
    . 2008. Individual plasmacytoid dendritic cells are major contributors to the production of multiple innate cytokines in an organ-specific manner during viral infection. Int. Immunol. 20: 45–56.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Swiecki M.,
    2. S. Gilfillan,
    3. W. Vermi,
    4. Y. Wang,
    5. M. Colonna
    . 2010. Plasmacytoid dendritic cell ablation impacts early interferon responses and antiviral NK and CD8(+) T cell accrual. Immunity 33: 955–966.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Tabeta K.,
    2. P. Georgel,
    3. E. Janssen,
    4. X. Du,
    5. K. Hoebe,
    6. K. Crozat,
    7. S. Mudd,
    8. L. Shamel,
    9. S. Sovath,
    10. J. Goode,
    11. et al
    . 2004. Toll-like receptors 9 and 3 as essential components of innate immune defense against mouse cytomegalovirus infection. Proc. Natl. Acad. Sci. USA 101: 3516–3521.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Mattner J.,
    2. K. L. Debord,
    3. N. Ismail,
    4. R. D. Goff,
    5. C. Cantu III.,
    6. D. Zhou,
    7. P. Saint-Mezard,
    8. V. Wang,
    9. Y. Gao,
    10. N. Yin,
    11. et al
    . 2005. Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections. Nature 434: 525–529.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Reddehase M. J.,
    2. F. Weiland,
    3. K. Münch,
    4. S. Jonjic,
    5. A. Lüske,
    6. U. H. Koszinowski
    . 1985. Interstitial murine cytomegalovirus pneumonia after irradiation: characterization of cells that limit viral replication during established infection of the lungs. J. Virol. 55: 264–273.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Brigl M.,
    2. L. Bry,
    3. S. C. Kent,
    4. J. E. Gumperz,
    5. M. B. Brenner
    . 2003. Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nat. Immunol. 4: 1230–1237.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Brigl M.,
    2. R. V. Tatituri,
    3. G. F. Watts,
    4. V. Bhowruth,
    5. E. A. Leadbetter,
    6. N. Barton,
    7. N. R. Cohen,
    8. F. F. Hsu,
    9. G. S. Besra,
    10. M. B. Brenner
    . 2011. Innate and cytokine-driven signals, rather than microbial antigens, dominate in natural killer T cell activation during microbial infection. J. Exp. Med. 208: 1163–1177.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Benedict C. A.,
    2. A. Loewendorf,
    3. Z. Garcia,
    4. B. R. Blazar,
    5. E. M. Janssen
    . 2008. Dendritic cell programming by cytomegalovirus stunts naive T cell responses via the PD-L1/PD-1 pathway. J. Immunol. 180: 4836–4847.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Miyagi T.,
    2. M. P. Gil,
    3. X. Wang,
    4. J. Louten,
    5. W. M. Chu,
    6. C. A. Biron
    . 2007. High basal STAT4 balanced by STAT1 induction to control type 1 interferon effects in natural killer cells. J. Exp. Med. 204: 2383–2396.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Pien G. C.,
    2. A. R. Satoskar,
    3. K. Takeda,
    4. S. Akira,
    5. C. A. Biron
    . 2000. Cutting edge: selective IL-18 requirements for induction of compartmental IFN-gamma responses during viral infection. J. Immunol. 165: 4787–4791.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Kastenmüller W.,
    2. P. Torabi-Parizi,
    3. N. Subramanian,
    4. T. Lämmermann,
    5. R. N. Germain
    . 2012. A spatially-organized multicellular innate immune response in lymph nodes limits systemic pathogen spread. Cell 150: 1235–1248.
    OpenUrlCrossRefPubMed
  31. ↵
    1. van Dommelen S. L.,
    2. H. A. Tabarias,
    3. M. J. Smyth,
    4. M. A. Degli-Esposti
    . 2003. Activation of natural killer (NK) T cells during murine cytomegalovirus infection enhances the antiviral response mediated by NK cells. J. Virol. 77: 1877–1884.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Arase H.,
    2. E. S. Mocarski,
    3. A. E. Campbell,
    4. A. B. Hill,
    5. L. L. Lanier
    . 2002. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 296: 1323–1326.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Brown M. G.,
    2. A. O. Dokun,
    3. J. W. Heusel,
    4. H. R. Smith,
    5. D. L. Beckman,
    6. E. A. Blattenberger,
    7. C. E. Dubbelde,
    8. L. R. Stone,
    9. A. A. Scalzo,
    10. W. M. Yokoyama
    . 2001. Vital involvement of a natural killer cell activation receptor in resistance to viral infection. Science 292: 934–937.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Pyzik M.,
    2. B. Charbonneau,
    3. E. M. Gendron-Pontbriand,
    4. M. Babić,
    5. A. Krmpotić,
    6. S. Jonjić,
    7. S. M. Vidal
    . 2011. Distinct MHC class I-dependent NK cell-activating receptors control cytomegalovirus infection in different mouse strains. J. Exp. Med. 208: 1105–1117.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Brown M. G.,
    2. A. A. Scalzo
    . 2008. NK gene complex dynamics and selection for NK cell receptors. Semin. Immunol. 20: 361–368.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Banovic T.,
    2. M. Yanilla,
    3. R. Simmons,
    4. I. Robertson,
    5. W. A. Schroder,
    6. N. C. Raffelt,
    7. Y. A. Wilson,
    8. G. R. Hill,
    9. P. Hogan,
    10. C. B. Nourse
    . 2011. Disseminated varicella infection caused by varicella vaccine strain in a child with low invariant natural killer T cells and diminished CD1d expression. J. Infect. Dis. 204: 1893–1901.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Levy O.,
    2. J. S. Orange,
    3. P. Hibberd,
    4. S. Steinberg,
    5. P. LaRussa,
    6. A. Weinberg,
    7. S. B. Wilson,
    8. A. Shaulov,
    9. G. Fleisher,
    10. R. S. Geha,
    11. et al
    . 2003. Disseminated varicella infection due to the vaccine strain of varicella-zoster virus, in a patient with a novel deficiency in natural killer T cells. J. Infect. Dis. 188: 948–953.
    OpenUrlAbstract/FREE Full Text
    1. Moll M.,
    2. S. K. Andersson,
    3. A. Smed-Sorensen,
    4. J. K. Sandberg
    . 2010. Inhibition of lipid antigen presentation in dendritic cells by HIV-1 Vpu interference with CD1d recycling from endosomal compartments. Blood 116: 1876–1884.
    OpenUrlAbstract/FREE Full Text
    1. Raftery M. J.,
    2. M. Hitzler,
    3. F. Winau,
    4. T. Giese,
    5. B. Plachter,
    6. S. H. E. Kaufmann,
    7. G. Schönrich
    . 2008. Inhibition of CD1 antigen presentation by human cytomegalovirus. J. Virol. 82: 4308–4319.
    OpenUrlAbstract/FREE Full Text
    1. Raftery M. J.,
    2. F. Winau,
    3. T. Giese,
    4. S. H. E. Kaufmann,
    5. U. E. Schaible,
    6. G. Schönrich
    . 2008. Viral danger signals control CD1d de novo synthesis and NKT cell activation. Eur. J. Immunol. 38: 668–679.
    OpenUrlCrossRefPubMed
    1. Webb T. J. R.,
    2. R. A. Litavecz,
    3. M. A. Khan,
    4. W. J. Du,
    5. J. Gervay-Hague,
    6. G. J. Renukaradhya,
    7. R. R. Brutkiewicz
    . 2006. Inhibition of CD1d1-mediated antigen presentation by the vaccinia virus B1R and H5R molecules. Eur. J. Immunol. 36: 2595–2600.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Yuan W. M.,
    2. A. Dasgupta,
    3. P. Cresswell
    . 2006. Herpes simplex virus evades natural killer T cell recognition by suppressing CD1d recycling. Nat. Immunol. 7: 835–842.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Brennan P. J.,
    2. R. V. Tatituri,
    3. M. Brigl,
    4. E. Y. Kim,
    5. A. Tuli,
    6. J. P. Sanderson,
    7. S. D. Gadola,
    8. F. F. Hsu,
    9. G. S. Besra,
    10. M. B. Brenner
    . 2011. Invariant natural killer T cells recognize lipid self antigen induced by microbial danger signals. Nat. Immunol. 12: 1202–1211.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Verma S.,
    2. Q. Wang,
    3. G. Chodaczek,
    4. C. A. Benedict
    . 2013. Lymphoid-tissue stromal cells coordinate innate defense to cytomegalovirus. J. Virol. 87: 6201–6210.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Vilarinho S.,
    2. K. Ogasawara,
    3. S. Nishimura,
    4. L. L. Lanier,
    5. J. L. Baron
    . 2007. Blockade of NKG2D on NKT cells prevents hepatitis and the acute immune response to hepatitis B virus. Proc. Natl. Acad. Sci. USA 104: 18187–18192.
    OpenUrlAbstract/FREE Full Text
    1. Arase H.,
    2. N. Arase,
    3. T. Saito
    . 1996. Interferon gamma production by natural killer (NK) cells and NK1.1+ T cells upon NKR-P1 cross-linking. J. Exp. Med. 183: 2391–2396.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Benlagha K.,
    2. A. Weiss,
    3. A. Beavis,
    4. L. Teyton,
    5. A. Bendelac
    . 2000. In vivo identification of glycolipid antigen-specific T cells using fluorescent CD1d tetramers. J. Exp. Med. 191: 1895–1903.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Krmpotic A.,
    2. M. Hasan,
    3. A. Loewendorf,
    4. T. Saulig,
    5. A. Halenius,
    6. T. Lenac,
    7. B. Polic,
    8. I. Bubic,
    9. A. Kriegeskorte,
    10. E. Pernjak-Pugel,
    11. et al
    . 2005. NK cell activation through the NKG2D ligand MULT-1 is selectively prevented by the glycoprotein encoded by mouse cytomegalovirus gene m145. J. Exp. Med. 201: 211–220.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Lodoen M.,
    2. K. Ogasawara,
    3. J. A. Hamerman,
    4. H. Arase,
    5. J. P. Houchins,
    6. E. S. Mocarski,
    7. L. L. Lanier
    . 2003. NKG2D-mediated natural killer cell protection against cytomegalovirus is impaired by viral gp40 modulation of retinoic acid early inducible 1 gene molecules. J. Exp. Med. 197: 1245–1253.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Velázquez P.,
    2. T. O. Cameron,
    3. Y. Kinjo,
    4. N. Nagarajan,
    5. M. Kronenberg,
    6. M. L. Dustin
    . 2008. Cutting edge: activation by innate cytokines or microbial antigens can cause arrest of natural killer T cell patrolling of liver sinusoids. J. Immunol. 180: 2024–2028.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Andrews D. M.,
    2. A. A. Scalzo,
    3. W. M. Yokoyama,
    4. M. J. Smyth,
    5. M. A. Degli-Esposti
    . 2003. Functional interactions between dendritic cells and NK cells during viral infection. Nat. Immunol. 4: 175–181.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Andoniou C. E.,
    2. S. L. van Dommelen,
    3. V. Voigt,
    4. D. M. Andrews,
    5. G. Brizard,
    6. C. Asselin-Paturel,
    7. T. Delale,
    8. K. J. Stacey,
    9. G. Trinchieri,
    10. M. A. Degli-Esposti
    . 2005. Interaction between conventional dendritic cells and natural killer cells is integral to the activation of effective antiviral immunity. Nat. Immunol. 6: 1011–1019.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Andrews D. M.,
    2. C. E. Andoniou,
    3. F. Granucci,
    4. P. Ricciardi-Castagnoli,
    5. M. A. Degli-Esposti
    . 2001. Infection of dendritic cells by murine cytomegalovirus induces functional paralysis. Nat. Immunol. 2: 1077–1084.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Robbins S. H.,
    2. T. Walzer,
    3. D. Dembélé,
    4. C. Thibault,
    5. A. Defays,
    6. G. Bessou,
    7. H. Xu,
    8. E. Vivier,
    9. M. Sellars,
    10. P. Pierre,
    11. et al
    . 2008. Novel insights into the relationships between dendritic cell subsets in human and mouse revealed by genome-wide expression profiling. Genome Biol. 9: R17.
    OpenUrlCrossRefPubMed
  55. ↵
    1. Tupin E.,
    2. Y. Kinjo,
    3. M. Kronenberg
    . 2007. The unique role of natural killer T cells in the response to microorganisms. Nat. Rev. Microbiol. 5: 405–417.
    OpenUrlCrossRefPubMed
  56. ↵
    1. Xie X. F.,
    2. A. Dighe,
    3. P. Clark,
    4. P. Sabastian,
    5. S. Buss,
    6. M. G. Brown
    . 2007. Deficient major histocompatibility complex-linked innate murine cytomegalovirus immunity in MA/My.L-H2b mice and viral downregulation of H-2k class I proteins. J. Virol. 81: 229–236.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section