Involvement of the 4-1BB/4-1BBL Pathway in Control of Monocyte Numbers by Invariant NKT Cells

Suzanne L. Cole, Kambez H. Benam, Andrew J. McMichael and Ling-Pei Ho
J Immunol April 15, 2014, 192 (8) 3898-3907; DOI: https://doi.org/10.4049/jimmunol.1302385

Abstract

4-1BB is expressed on invariant (i)NKT cells, but its role is unclear. We showed previously that iNKT cells are involved in control of monocyte numbers during influenza A virus (IAV) infection and now question the role of the 4-1BB costimulatory pathway in the cross-talk between these cells. We found that iNKT cells and monocytes interact to promote expression of 4-1BB and 4-1BBL, respectively. Blockade of 4-1BB/L pathway under resting coculture conditions increased apoptosis of iNKT cells and monocytes. However, activation of iNKT cells overrides this survival signal, causing marked apoptosis of monocytes independent of 4-1BB/L. Blocking 4-1BBL in alpha-galactosylceramide-activated iNKT–monocyte cocultures reduced iNKT proliferation and abrogated monocytic IL-12 production. In vivo, expression of 4-1BB and 4-1BBL is increased on iNKT cells and Ly6Chi monocytes, respectively, during IAV infection, and there were lower frequencies of apoptosing Ly6Chi monocytes in the blood of iNKT knockout mice and higher numbers of monocytes in lungs compared with infected wild-type mice. Adoptive transfer of iNKT cells into the lungs of these mice reduced lung Ly6Chi monocytes levels, even when iNKT cells were preincubated with 4-1BB blocking Abs. These findings suggest that under resting conditions, 4-1BB/L engagement during iNKT–monocyte interaction promotes survival of these cells. When iNKT cells are activated, whether by alpha-galactosylceramide or during IAV infection, iNKT cells induced apoptosis of monocytes via a 4-1BB/L–independent mechanism, reducing monocyte numbers. 4-1BB/L costimulation amplified monocyte-mediated proliferation of iNKT cells, indirectly providing a method for monocytes to control their own numbers during infection.

Introduction

The 4-1BB molecule (also known as CD137 and TNFRSF9) is an inducible costimulatory protein from the TNF superfamily best known for enhancing CD8+ T cell proliferation and survival, independent of CD28 (1–3). Its ligand, 4-1BBL (CD137L/TNFSF9), is mainly expressed on myeloid cells (4). As with other members of the TNF superfamily, signaling is bidirectional, and therefore, ligation influences the function of ligand-bearing cells as well as those expressing the receptor (5, 6). 4-1BB signaling preferentially expands CD8+ T cells and induces Th1 responses (7, 8). Its reverse signaling, via 4-1BBL, amplifies the innate immune response by enhancing monocyte proliferation, migration, and maturation into Th1-inducing dendritic cells (DCs) (9–11) and augmenting cytokine production by peritoneal macrophages (12). 4-1BBL reverse signaling also has been reported in nonhematopoietic cells such as epithelial and carcinomatous cells, where it results in injurious inflammation (13, 14). Cross-linking of 4-1BB on activated T cells and NK cells has shown promise in enhancing antitumor immune responses by these cells in murine models (15, 16), and agonistic 4-1BB Abs have been tested in phase I and II clinical trials (17, 18). In addition to T cells and NK cells, 4-1BB is also expressed on regulatory T cells, DCs, monocytes, neutrophils, and iNKT cells (15, 19, 20). The wide expression profile and the capacity for bidirectional signaling make the functional outcome of its stimulation complex and cell dependent (6).

It is well established that iNKT cells express several costimulatory and coinhibitory molecules akin to conventional T cells, but the engagement of these molecules on iNKT cells does not always yield similar functional outcomes to T cells. This is partly due to the preactivated phenotype of iNKT cells (21–23), which sets these cells up for a quicker rate of activation compared with naive T cells. The 4-1BB/L pathway seems to be dispensable in initiating activation of conventional T cells (where other costimulatory signals dominate) but is instead involved in augmenting TCR signals in later phases and in sustaining effector functions. Comparatively, little has been published on the role of 4-1BB/L in iNKT cell activation, although 4-1BB costimulation in conjunction with alpha-galactosylceramide (αGC) has been shown to enhance proliferation and activation of iNKT cells in vivo and exacerbate iNKT-mediated airway hyperresponsiveness (3). To our knowledge, nothing is known of how this pathway is involved in the cross-talk between iNKT cells and monocytes.

In this study, we examined the role of 4-1BB/L in iNKT–monocyte cross-talk, as this cellular interaction may have beneficial consequences in influenza A virus (IAV) infection. We previously showed that iNKT cells caused lysis of monocytes in a CD1d-dependent process in vitro and that iNKT knockout (KO) mice (Jα18−/−) infected with IAV have markedly increased numbers of lung Ly6Chi monocytes, which could have contributed to lung injury in these mice (24). Circulating monocytes are a heterogeneous population of cells, which can differentiate to DCs, macrophages, or TNF-α/iNOS–producing DCs in tissues (25) and can have a suppressive function if exported from the bone marrow in the immature form (myeloid-derived suppressor cells or MDSCs) (26). In IAV infection, Ly6Chi monocytes and TNF-α/iNOS–producing DCs contribute both to host defense and lung injury (27). We had proposed that when the viral load is high, iNKT cells are instrumental in controlling the levels of monocytes and preventing this from spilling into a harmful response (24). We reasoned that if the 4-1BB costimulatory pathway is involved in the cross-talk between these cells, modulating this could potentially influence the functional consequence of the interaction between iNKT cells and monocytes in a disease or infection setting. We hypothesize that at least part of the iNKT–monocyte interaction that regulates monocyte function is mediated or influenced by the 4-1BB/L pathway. The findings from this paper suggest that under steady-state conditions, and without exogenous Ags, iNKT cells interact with monocytes to promote expression of 4-1BB and 4-1BBL, and via this pathway, their own survival. However, iNKT cell activation, whether by αGC or during IAV infection overrides this survival signal, with iNKT cells inducing apoptosis of monocytes via a 4-1BB/L–independent mechanism causing reduction in Ly6Chi monocyte numbers at the target site (infected lungs). Interaction via 4-1BB/L further enhances this effect by increasing monocyte-mediated proliferation and numbers of iNKT cells.

Materials and Methods

iNKT–monocyte coculture assays

Purified human monocytes were cocultured with the well established LH22 iNKT clones (28). PBMCs were isolated using standard density gradient centrifugation on Lymphoprep (Axis Shield, Oslo, Norway) as described previously (29). From these, monocytes were purified by CD14-positive selection using CD14 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany). A total of 2 × 105 monocytes were plated in round-bottom 96-well plates at differing ratios to iNKT cells as described in Results. 4-1BB blocking Ab (4B4-1; BD, Bergen County, NJ) and 4-1BB ligand blocking Ab (C65-485; BD) (30) were added at the beginning of the coculture at 5 μg/ml each. When indicated in text, CD1d blocking Ab (CD1d42; BD Pharmingen, Oxford, U.K.) was added at 10 or 25 μg/ml. Where used, αGC (Enzo Life Sciences, Exeter, U.K.) was added after addition of blocking Abs at 0.4 μg/ml.

Functional assays

Standard 6-h chromium release assays were performed as described previously (28). Where 4-1BB/L blocking was carried out, 5 μg/ml 4-1BB and 4-1BB ligand blocking Abs (clones 4B4-1 and C65-485) were added to effector cells for 1 h prior to addition of target cells.

iNKT clone proliferation was measured using a standard CFSE fluorescence assay (Invitrogen, Paisley, U.K.). iNKT clones were incubated with CFSE according to the manufacturer’s protocol. Cells were washed in culture media and resuspended in PBS supplemented with 1% BSA at a concentration of 1 × 106 cells/ml. A total of 10 mM CFSE was added to the cell suspension, mixed thoroughly, and incubated for 10 min at 37°C in the dark. After incubation, 10 ml ice-cold culture medium was added to the cells, followed by a 10-min incubation on ice. Cells were then washed three times. To induce proliferation, 2 × 105 CFSE-labeled cells were incubated with 5 × 104 monocytes and αGC (0.4μg/ml) in round-bottomed 96-well plates. After 5 d of coculture, fluorescence was measured using a flow cytometer. Reduction in CFSE staining indicates cells that have divided.

Human NK cells were isolated using NK negative selection kit from Miltenyi Biotec (NK cell isolation kit number 130-092-657), according to the manufacturer’s instructions. Anti-NKp46 (9E2; eBioscience) was added to activate NK cells. Nef-specific CD8 T cell clone and clone-specific peptide (FLKEKGGL) was a gift from T. Dong (University of Oxford) (31).

Flow cytometric studies

For surface Ag staining, cells were incubated for 20 min at 4°C in the dark with fluorochrome conjugated mAbs in 96-well plates in FACS buffer (PBS containing 10% FCS and 1% BSA). Cells were then washed twice and resuspended in FACS buffer. Samples were analyzed on three-laser, nine-color, CyAn ADP Flow Cytometers using Summit 4.3 (both Dako, Cambridgeshire, U.K.). For surface staining of mouse cells, cells were blocked with anti-CD16/32 (93; eBioscience, San Diego, CA) at 1 μg/ml for 15 min to block Fc binding prior to washing and applying surface stains as above. Abs from BioLegend (San Diego, CA), Serotec (Raleigh, NC), eBioscience, or Miltenyi Biotec were used: anti-human, 4-1BB (4B4-1), 4-1BBL (5F4), Vα24-Jα18 TCR (6B11) CD3 (UCHT1), CD14 (61D3), and CD107a (eBioH4A3); and anti-mouse, Ly6C (HK1.4), Ly6G (1A8), CD3e (154-2C11), 4-1BBL (TKS1), and 4-1BB (17B5).

For intracellular cytokine staining, cells were incubated in round bottomed 96-well plates for the indicated times, with brefeldin A (Sigma-Aldrich, St. Louis, MO) at 10 μg/ml added for the last 4 h of stimulation. In some cases, 0.4 μg/ml α-GC was added to activate iNKT clones at the start of the culture. For CD107a detection, the Ab was added at the start of the stimulation period. Surface Ags were stained as described above, after which cells were fixed in 1% paraformaldehyde (Sigma-Aldrich) for 15 min at room temperature. Cells were then washed and resuspended in 30 μl 0.5% saponin (Sigma-Aldrich) in FACS buffer with conjugated mAbs. Cells were then washed twice, fixed as described, resuspended in FACS buffer, and analyzed by flow cytometry. Intracellular Abs from eBioscience or BioLegend were used: anti-human TNF-α (MAb11), IFN-γ (4S-B3), IL-4 (8D4-8), and IL-12 (C8.6).

IAV infection of monocytes

Isolated monocytes were infected by culturing 2 × 106 cells with 40 hemagglutinating unit (HAU)/ml A/PR8/34 (H1N1) in 400 μl culture medium without FCS but with trypsin (Sigma-Aldrich) at 2 μg/ml in a 15-ml Falcon tube for 2 h at 37°C and 5% CO2 to allow for virus adsorption. Cells were then washed twice and plated out in 6-well plates for 18 h. Infection was confirmed with hemagglutinin expression as described previously (24).

Mice

C57BL/6 mice were obtained from breeding pairs originally purchased from The Jackson Laboratory (Bar Harbor, ME). Mice lacking the Ja18 TCR gene segment (19) were provided by Prof. M. Taniguchi (RIKEN, Yokohama Japan) and have been backcrossed at least 12 times with the C57BL/6 wild-type (WT) mice. All mice were maintained at the Biomedical Services Unit at the John Radcliffe Hospital (Oxford, U.K.) and were used according to established institutional guidelines. A/PR8/34 was administered intranasally to mice using methods described previously (32). Mice were sacrificed on day 3 after viral challenge, and blood, spleen, and BAL cells were obtained as described previously. All animals were used according to established University of Oxford institutional guidelines (Oxford, U.K.) under the authority of a U.K. Home Office project license.

For iNKT cell transfer, cells were isolated from spleen of IAV-infected C57BL/6 mice using NK1.1+ NKT isolation kit (Miltenyi Biotec) and divided into two aliquots. One was incubated with anti–4-1BB (17B5; eBioscience) at 5 μg/ml diluted in PBS and the other in PBS alone for 20 min at 4°C. Following blocking, iNKT cells were washed and resuspended in single aliquots of 1.5 × 104 purified iNKT cells in 20ul PBS per mouse. The virus was diluted to 0.1 HAU in 20 μl, which was mixed with iNKT cells and immediately used to inoculate mice intranasally so that the final infection used was 0.1 HAU/mouse in a 40-μl volume.

For lung homogenate studies, a standard method was used (33). Briefly, lungs from WT and Jα18−/− mice were first perfused with PBS and then removed, cut into pieces, and incubated with collagenase A (Roche) (1 mg/ml in 10 ml PBS) and DNAse I (Sigma-Aldrich) (200 μg/lung) for 1 h at 37°C on a shaking platform. The resulting suspension was passed through a 23-gauge needle and passed through a 70-μm filter before FACS staining.

Statistics

All statistical analysis was performed using GraphPad Prism 5.01 (GraphPad, La Jolla, CA). Normality of data distribution was first tested using the D’Agostino and Pearson omnibus normality test. Normally distributed data sets were compared using Student t test; where comparisons included data not normally distributed, the Mann–Whitney U test was used to compare data sets. To compare multiple, normally distributed data sets simultaneously, one-way ANOVA with Tukey posttest analysis was used. Where multiple data sets included data that were not normally distributed, Kruskal–Wallis with Dunn’s multiple comparison posttest analysis was performed. A p value < 0.05 was considered significant. Mean values were reported for normally distributed data and median for nonnormal distribution. Error bars on graphs represent SEM unless otherwise stated.

Results

4-1BB and 4-1BBL are induced on iNKT cells and monocytes, respectively, following coculture and activation

We first analyzed expression of 4-1BB and 4-1BBL on human iNKT cells and monocytes. A well-established human iNKT clone (LH22) (28) was cocultured with human monocytes isolated using CD14 MACS beads (95% purity) at 4:1, 1:1, and 1:4 monocyte: iNKT cell ratios for 4–5 h (all ratios showed similar results and trend; data depicted are from 1:1 ratio). iNKT cells and monocytes alone expressed very low levels of 4-1BB or 4-1BBL (mean ± SEM, 1.44 ± 0.39 and 0.78 ± 0.09% of cells, respectively) (Fig. 1A). However, following 4 h of coculture (without exogenous glycolipids), expression of 4-1BB and 4-1BBL were induced on iNKT cells and monocytes respectively (mean ± SEM, 10.4 ± 1.3 and 4.4 ± 1.3%, respectively) (Fig. 1A). Activation with αGC further increased 4-1BB expression on iNKT cells and 4-1BBL on monocytes, respectively (Fig. 1A–C). Because we were interested in the cross-talk between iNKT and monocytes during IAV infection, and IAV infection in vivo caused direct infection of monocytes in the lungs (24), we infected human monocytes in vitro with H1N1 A/PR8/34 and analyzed 4-1BB and 4-1BBL expression on iNKT cells and monocytes. We observed upregulation of 4-1BBL on IAV-infected monocytes 24 h postinfection with a corresponding increase of 4-1BB on iNKT cells (Fig. 1B, 1C). In the coculture, cells that coexpressed iNKT and monocyte markers, and shown as doublet cells on pulse width/SSC analysis on FACS (and therefore two cells physically interacting), showed even higher levels of expression of 4-1BB, although not 4-1BBL (Fig. 1D, 1E), suggesting that physical interaction, in the absence of exogenous glycolipid Ags, caused an increase in 4-1BB on iNKT cells. Because 4-1BB expression also has been reported on activated myeloid cells, we also examined this in the coculture conditions described but found only minimal expression compared with that on iNKT cells (mean of 3–5% with or without coincubation with αGC). There was no expression of 4-1BBL on iNKT cells (data not shown).

FIGURE 1.
FIGURE 1.

4-1BB and 4-1BBL expression on cocultured iNKT cells and monocytes. (A) Representative FACS dot plots showing surface expression of 4-1BBL on monocytes and 4-1BB on iNKT cells either alone (Mono only or iNKT only) or in 1:1 coculture of monocytes/iNKT cells for 4 h without any exogenous stimulation (+0) or with αGC stimulation. (B) 4-1BB expression on iNKT cells cocultured with monocytes after 4 h of αGC stimulation [as in (A)] and 24 h after coculture with IAV-infected monocytes. (C) 4-1BBL expression on monocytes for coculture described in (B). n = 3 experiments. (D) Surface expression of 4-1BB or 4-1BBL on 6B11+CD14+ doublets [gate i in (E)] following a 4-h coculture of monocytes and iNKT cells. (E) Doublets (i) shown as FACS dot plot.

These findings suggest that iNKT cells and monocytes contact each other, and without need for exogenous glycolipid Ag increase expression of 4-1BB and 4-1BBL on iNKT cells and monocytes, respectively. Doublet cells in the coculture ranged from 2 to 5% regardless of the ratio of iNKT to monocytes (data not shown), suggesting that an optimal amount of contacting cells occurred at any one point, and these interacting cells impart signals to increase the 4-1BB and 4-1BBL expression on the two cell groups. Lack of increase in 4-1BBL expression on monocytes in these doublets could be due to these molecules being upregulated later than 4-1BB, so that overall 4-1BBL expression was increased in the monocyte population but not on contacting (doublet) cells. Alternatively, 4-1BBL could be endocytosed and internalized at this time point, as found on CD137L-expressing Hodgkin’s disease cell line KM-H2 (34). Induction of 4-1BB and 4-1BBL with and without direct antigenic stimulation and indirectly via inflammatory signals from IAV infection implicates this pathway in iNKT–monocyte interactions at steady state and during viral infection.

Engagement of 4-1BB/L pathway during iNKT–monocyte interaction maintains monocyte and iNKT survival by preventing apoptosis, but iNKT activation overrides this signal, causing apoptosis of monocytes

Because the 4-1BB pathway is primarily involved in survival and prevention of apoptosis in T cells (35), we first questioned whether it promoted the survival of iNKT cells and monocytes during baseline interaction. Because 4-1BB/4-1BBL signaling is bidirectional (5, 6), iNKT cells and monocytes were cocultured as above with or without a mixture of both 4-1BB and 4-1BBL blocking Abs (both 5 μg/ml) for 24 h. Blocking 4-1BB/L increased monocyte apoptosis (Fig. 2A, 2B), suggesting that the 4-1BB/L pathway prevented apoptosis of monocytes under nonstimulated conditions. Activation of iNKT cells with αGC also increased the proportion of apoptosing monocytes, but this was independent of the 4-1BB/L pathway. To examine whether the apoptotic effect was unique to iNKT cells, we conducted the same experiment but using Ag (nef)-specific CD8 T cell clones (31) or purified NK cells instead of iNKT cells ± relevant activating stimuli (FLKEKGGL peptide for nef-specific CD8 T cell clones and anti-NKp46 for NK cells) at the same ratio and duration of iNKT to monocyte experiments (1:1). These non-iNKT cells did not cause apoptosis of monocytes attributing the apoptotic effect uniquely to iNKT cells (Fig. 2C). Activation of these cells was confirmed by increased TNF-α production by both iNKT, NK, and CD8 T cells/clones after their specific stimulation (Fig. 2D). To examine whether soluble factors were involved in induction of apoptosis on monocytes, Transwell experiments were carried out where primary human monocytes, cocultured with iNKT cells ± αGC, were placed in the top Transwell section, and monocytes alone were placed in the bottom section (ratio of iNKT: top compartment monocyte:bottom compartment monocyte were 1:1:1) for 24 h. This showed that addition of αGC to activate iNKT cells in the top compartment still caused a significant increase in proportion of apoptotic monocytes in the bottom compartment even with separation (Fig. 2E). However, the degree of induction was less (mean of 35% where Transwell inserts were used versus 56% without Transwell separation). This suggests that both soluble factor(s) and physical interaction are required for optimal induction of apoptosis. To examine the level of contribution CD1d makes to the physical part of this interaction, we added 10 or 25 μg/ml CD1d blocking Ab at the start of the coculture and observed a significant decrease in the proportion of cell undergoing apoptosis at 24 h (Fig. 2F).

FIGURE 2.
FIGURE 2.

Changes in proportion of apoptosing iNKT cells and monocytes in response to 4-1BB/L blockade. (A) Representative FACS plots depicting apoptosis of monocytes (gated on 6B11−CD3− cells) following 24-h coculture with iNKT cells at 1:1 ratio. +0 indicates iNKT–monocyte coculture only; +41BB/L block indicates with addition of 4-1BB and 4-1BBL blocking Abs; + αGC indicates in the presence αGC only; +αGC and 4-1BB/L block indicates in presence of both. Cells undergoing apoptosis are defined as Annexin Vhi7AAD− (as compared with isotype staining) and numbers indicate percentage of gated cells. (B) Graphical data from (A); from three independent experiments using monocytes isolated from three to four different individuals each time. All asterisks refer to p < 0.05 (statistical methods described in Materials and Methods). (C) Annexin Vhi monocytes following 24-h coculture at 1:1 ratio with iNKT clones, CD8 T cell clones, or isolated NK cells. (D) Intracellular cytokine measurement of TNF-α produced by iNKT clones ± αGC, CD8 T cell clones ± peptide, or NK cells ± anti-NKp46 following 24-h coculture with monocytes. (E) Frequency of apoptosing monocytes (Annexin Vhi) in lower compartment, following 24-h incubation with monocytes + iNKT clones only (+0) or monocytes + iNKT clones + αGC (+αGC) in upper compartment of Transwell experiment. (F) Effect of CD1d blockade on frequency of apoptosing monocytes. (G) FACS profile of monocytes (CD3−6B11−) and iNKT clones (6B11+) following 5 d of coculture in the presence or absence of αGC and 4-1BB and 4-1BBL blocking Abs. The FACS plot depicts only live (gating out dead cells) cells but numerical values refer to percentage of all cells (including dead cells). Gate (i) live iNKT cells, (ii) dying/apoptosing iNKT cells (measured as Annexin Vhi cells), and (iii) monocytes. (H) Absolute numbers of Ly6Chi monocytes in blood and spleen of naive C57BL/6 or Jα18−/− mice. +0 indicates iNKT + monocytes only; +4-1BB/L block indicates iNKT + monocytes and 4-1BB/L blocking Abs; +αGC indicates iNKT + monocytes + αGC; +αGC+4-1BB/L block indicates iNKT + monocytes + αGC and 4-1BB/L blocking Abs; all added from the start of the coculture.

With regards to apoptosis of iNKT cells, in a 5-d coculture with and without αGC, 4-1BB/L blockade without exogenous stimulation caused an increase in the apoptosing iNKT cell population (6B11midAnnexin V+ cells; marked as ii in Fig. 2G), further supporting the requirement for 4-1BB/L pathway in optimal survival of iNKT cells. Stimulation with αGC increased iNKT cells and reduced monocyte (marked as iii in Fig. 2G) levels. Microscopy of tissue culture plates before staining confirmed that this reduction in monocytes was not due to activated monocytes adhering to the plate. Blocking 4-1BB/L in αGC-activated cocultures caused the percentage of monocytes to increase despite an increase in apoptosing monocytes (see gate iii Fig. 2G). We reasoned that this is secondary to the reduced iNKT numbers in the well, which meant less iNKT-mediated loss of monocytes.

These data suggest that iNKT cells interacting with monocytes contribute to their survival and this is, at least in part, dependent on the 4-1BB/L pathway (Fig. 2B). When 4-1BB/L was blocked during αGC activation, an additional but small enhancement of apoptosis was still observed. Thus, the 4-1BB/L pathway is primarily involved in prevention of apoptosis, and even when this was overwhelmed by apoptotic signals provided by activated iNKT cells, a significant inhibitory effect was still present during this time. For monocytes, the role of 4-1BB/L pathway in their homeostatic survival is supported in vivo by findings of reduced circulating monocyte levels in iNKT KO mice (Fig. 2H). Lack of difference in numbers in the spleen between iNKT KO and WT mice suggests that functional interaction could have occurred in the blood compartment rather than the spleen.

4-1BB is required for optimal proliferation of activated iNKT cells and cytokine production by monocytes

We next turned to functional consequence of 4-1BB/4-1BBL engagement on monocytes and iNKT cells when the latter is activated. Using the same coculture system as Fig. 2H, but with addition of CFSE-stained iNKT cells, we found that blocking 4-1BB/L significantly reduced both the magnitude and the rate of iNKT cell proliferation measured by CD3+ CFSE staining over 5 d (Fig. 3A) but not the levels of early IFN-γ, IL-4, or TNF-α production (fourth hour), although there was a small but highly significant (p < 0.001) increase in IL-4 production with 4-1BB/L blockade at 24 h (Fig. 3B). There was no change in the capacity of iNKT cells to degranulate and kill target cells with or without 4-1BB/L blockade, as shown by CD107a staining and [51Cr] release assay (Fig. 3C, 3D).

FIGURE 3.
FIGURE 3.

Functional consequences of 4-1BB/L blocking on iNKT cells and monocytes. (A) iNKT cell proliferation measured as CFSE dilution for 6B11+ cells. CFSElo cells refer to amount of cells that have CFSE staining below the lowest dividing cells. A total of 2 × 105 iNKT cells were CFSE-labeled and cocultured with 5 × 104 monocytes for 2, 4, or 5 d with or without αGC and 4-1BB and 4-1BBL blocking Abs. Data depicted are from three individuals in one experiment; representative of three independent experiments. (B) Cytokine production from iNKT cells measured by intracellular cytokine staining after 4- and 24-h coculture with monocytes at 1:1 ratio with or without αGC and 4-1BB and 4-1BBL blocking Abs. Data are from two individual experiments, with monocytes from three individuals per experiment. The y-axis refers to percentage of iNKT cells in well–all wells had identical numbers of starting iNKT and monocytes. (C) CD107a expression on 6B11+ iNKT cells following 4-h coculture with monocytes at 5:1 (iNKT/monocyte) ratio with or without αGC and 4-1BB and 4-1BBL blocking Abs. (D) Specific lysis of 51Cr-labeled monocytes by iNKT cells at ratios depicted. Data are from two individual experiments using monocytes from one to three individuals per experiment. (E) Cytokine production from monocytes measured by intracellular cytokine staining after 24-h coculture with iNKT cells at 1:1 ratio with or without αGC and 4-1BB and 4-1BBL blocking Abs. No difference was found at the fourth hour of coculture. (F) MFI of IL-12 and TNF-α expression on monocytes after coculture with iNKT cells ± 4-1BB/L blockade [identical conditions to (E)]. (G) IL-12 levels (by ELISA) in supernatant from cocultures of iNKT cells and monocytes at 1:1 ratio for 24 h. Data are from three individual experiments of two to three individuals per experiment. For (A)–(D) and (F), +0 indicates iNKT + monocytes only; +4-1BB/L block indicates iNKT + monocytes and 4-1BB/L blocking Abs; +αGC indicates iNKT + monocytes + αGC; +αGC+4-1BB/L block indicates iNKT + monocytes + αGC and 4-1BB/L blocking Abs; all added from the start of the coculture. Mono +0 indicates monocytes without iNKT cells; Mono+αGC indicates monocytes without iNKT cells but cocultured with αGC.

Examining the effect of 4-1BB/L blockade on monocytes in the coculture, we found a reduction in IL-12 and TNF-α expression by intracellular cytokine staining after 24 h of stimulation with αGC (Fig. 3E). Notably, iNKT-mediated production of IL-12 by monocytes, at least in vitro, appears to require 4-1BB/L pathway as blockade abrogated IL-12 production (Fig. 3E, 3G). This was not the case with TNF-α. Because there was a possibility that the differences in TNF-α staining was due to the failure to release membrane-bound in TNF-α, we also showed the mean fluorescence index (MFI) of the staining. This showed no significant difference in expression levels for both IL-12 and TNF-α, although the trend is toward lower levels after 4-1BB/L blockade (Fig. 3f). In conjunction with reduced numbers of IL-12– and TNF-α–producing monocytes, we conclude that less monocytes were secreting these cytokine when the 4-1BB/L pathways was blocked. Cytokine levels measured by ELISA in the supernatant collected after 24 h also showed reduced levels of IL-12 (Fig. 3F) but not TNF-α (data not shown). The latter could be due to a large amount of TNF-α production by iNKT cells in this coculture system, which masked smaller changes in TNF-α production by monocytes. For IL-12, intracellular cytokine staining showed that only monocytes produced IL-12 in the coculture (data not shown).

These findings suggest that costimulation via 4-1BB is required for the proliferation but not cytokine production by iNKT cells, and 4-1BBL reverse-signaling contributes to IL-12 and likely TNF-α production by monocytes. The effect on TNF-α production by monocytes may not be significant during iNKT–monocyte interactions because a large amounts of TNF-α is also produced by iNKT cells, which is not affected by the 4-1BB/L pathway.

iNKT cells enhanced apoptosis of monocytes during IAV infection in vitro and in vivo

To examine the relevance of our findings in vivo, we infected C57BL/6 WT and iNKT-deficient (Jα18−/−) mice with H1N1 A/PR8/34 using a well-characterized model (24). On day 3 after inoculation, animals were sacrificed, and samples were obtained for analysis. We found that the frequency of circulating iNKT cells in blood expressing 4-1BB (but not iNKT cells from spleen or liver) (Fig. 4A, 4B), and the proportion of Ly6Chi monocytes expressing 4-1BBL (Fig. 4C) increased after IAV infection. Notably, the proportion of 4-1BB–expressing iNKT cells in the blood is greater than spleen- or liver-derived iNKT cells at baseline (with or without IAV infection) (Fig. 4B). The overall MFI for these cells did not show a significant difference (data not shown). Proportion of 4-1BBL–expressing monocytes and overall MFI of 4-1BBL+ monocytes were increased in blood and spleen after IAV infection regardless of whether they were WT or iNKT KO mice (Fig. 4C, 4D). This suggests that IAV infection increased both the level of 4-1BBL expression on monocytes and the number of 4-1BBL–expressing monocytes. These findings were observed within 3 d of IAV infection in both WT and iNKT KO mice. These data reflect findings in vitro, although the scale of increase in 4-1BBL on monocytes with IAV infection was greater in vivo. Increases in 4-1BBL also were most marked in cells derived from blood.

FIGURE 4.
FIGURE 4.

Expression of 4-1BB/L and apoptotic markers on iNKT cells and monocytes after IAV infection of WT and iNKT KO mice. (A) Representative FACS plot for 4-1BB expression on iNKT cells and 4-1BBL on Ly6Chi monocytes in blood of WT mice on day 3 after IAV infection. (B) Proportion of 4-1BB–expressing iNKT cells in blood, spleen, and liver of WT mice 3 d after IAV infection (n = 6). (C) Proportion of 4-1BBL–expressing Ly6Chi monocytes in blood and spleen of WT and iNKT KO mice on day 3 after IAV infection. (D) MFI of 4-1BBL expression on Ly6Chi monocytes from (C). (E and F) Proportion of apoptosing (Annexin Vhi) Ly6Chi monocytes in blood or spleen of WT or iNKT KO mice on day 3 ± IAV infection, and for circulating Ly6C−Ly6G+ neutrophils (G). All data are from two to three independent experiments using three to four mice per group. In (E), *p < 0.001.

We next questioned whether iNKT cell activation in vivo during IAV infection is required for apoptosis of Ly6Chi monocytes, as was suggested in vitro (Fig. 2A, 2B), by examining the apoptosis of circulating (blood) and spleen-derived Ly6Chi monocytes on day 3 of infection. We found that less of the circulating (but not spleen-derived) monocytes from iNKT KO mice were apoptosing compared with WT after IAV infection (Fig. 4C, 4D). Therefore, more Ly6Chi monocytes were surviving longer in the blood of iNKT KO mice during IAV infection. As shown by others (36, 37), overall, almost all circulating Ly6Chi monocytes were apoptotic (mean ± SEM, 91.02 ± 0.45% in WT and 82.52 ± 1.53% in iNKT KO of Ly6Chi monocytes were Annexinhi), and higher proportions of circulating monocytes were apoptotic compared with spleen-derived monocytes (Fig. 4E, 4F). This is consistent with the concept that monocytes are short-lived cells with a high apoptotic rate, although during inflammatory situations they receive signals to survive longer and home to tissues to contain infection (38). We also note that the reduction in apoptosing cells with IAV infection, and enhancement of this in iNKT KO mice, was not observed for neutrophils (which are not known to interact with iNKT cells) (Fig. 4G).

Thus, these data suggest that iNKT cells are involved in augmenting apoptosis of circulating Ly6Chi monocytes during IAV infection, which could be a method of keeping the numbers of Ly6Chi monocyte under control in the lungs (24). These infection models are well established by our group (24), and although Jα18−/− mice also have other T cell anomalies, the question in this study focuses on whether addition of iNKT cells to the lungs of Jα18−/− mice during IAV infection virus changed the monocyte numbers and whether this is affected by 4-1BB/L blockade. Thus, to investigate whether the 4-1BB/L pathway has a role in iNKT–monocyte interaction and function during IAV infection in vivo, we isolated iNKT cells from spleen of IAV-infected mice and incubated half of the cells with 4-1BB blocking mAb and the other half with PBS. The two aliquots of 1.5 × 104 iNKT cells were then transferred separately, via high-volume intranasal instillation, to iNKT KO mice at the point of IAV inoculation (Fig. 5A). We hypothesized that whether activated iNKT cells directly caused apoptosis of monocytes as suggested by Fig. 2A and 2B, then we would observe a depression in monocyte numbers in the lungs on day 3 (we had previously shown that this was the day of maximal infiltration of Ly6Chi monocytes in this model) (21), and an increase in proportion of apoptosing monocytes in the bronchoalveolar lavage (BAL). If 4-1BB/L had a role in this, the reduction would be abrogated with 4-1BB/L blockade.

FIGURE 5.
FIGURE 5.

Effect of 4-1BB blockade of iNKT cells on lung monocyte numbers. (A) Overview of iNKT cell transfer experiment. (B) Absolute numbers of Ly6Chi monocytes (per ml of BAL fluid) 3 d after IAV infection and transfer of purified iNKT cells (with or without 4-1BB blockade), *p < 0.01. (C) Absolute number of Ly6Chi monocytes in digests of perfused lungs from WT and iNKT KO, on day 3 after IAV infection. (D) Absolute numbers of neutrophils in BAL after 3 d after IAV infection and transfer of purified iNKT cells (with or without 4-1BB blockade) [from same mice as (B)]. No significant difference was found between groups. Data are representative of two independent experiments of three mice per group.

First, we observed an increase in Ly6Chi monocyte numbers in bronchoalveolar lavage (BAL) of iNKT KO mice compared with WT, which we have shown before (Fig. 5B) (24) and in homogenates of perfused lungs (Fig. 5C). Transferring iNKT cells to the lungs of these iNKT KO mice ameliorated this increase. Instillation of 4-1BB/L–blocked iNKT cells did not affect lung Ly6Chi monocyte numbers (Fig. 5B), supporting our in vitro data (Fig. 2B) that 4-1BB/L inhibition of apoptosis is likely to be irrelevant during inflammation or infection. Lung neutrophil numbers were not affected by addition of 4-1BB/L–blocked or untouched iNKT cells to lungs of IAV-infected iNKT KO mice (Fig. 5D).

Because the volume given was the same as the usual intranasal instillation for IAV and we know from previous work that this reached the lower respiratory tract (21), we propose that iNKT cells are able to act both in the lungs and blood (Fig. 4C) to reduce monocyte numbers during IAV infection. This is likely to be via increasing apoptosis of monocytes.

These in vivo studies provide circumstantial evidence that iNKT cells reduce Ly6Chi monocyte numbers in the lungs during IAV infection by inducing their apoptosis in the blood and lungs, thereby causing their death. 4-1BB/L costimulation is unlikely to play a significant role in this process but amplifies monocyte-mediated proliferation of iNKT cells, which in turn cause monocytes to apoptose, leading to a reduction in their numbers.

Discussion

Understanding the role played by costimulatory pathways in iNKT cell activation is important because iNKT cells have potent effector and regulatory capabilities. They are directly involved in destroying malignant cells and clearing microbial pathogens (39–41) and activate a wide range of immune cells including DCs, macrophages, and NK, T, and B cells. We are primarily interested in their interaction with monocytes, which has been relatively understudied. Our previous work suggests that cross-talk between these two groups of cells is important for control of the function and numbers of Ly6Chi monocytes during IAV infection. In this paper, we show that expression of 4-1BB and 4-1BBL on iNKT cells and monocytes, respectively, is inducible upon activation of iNKT cells and IAV infection of monocytes. We also show that 4-1BB/L is involved in the homeostatic maintenance of monocyte and iNKT populations and their proliferation and expansion when these cells are activated. It supports previous data from Vinay et al. (42) who found reduced numbers of iNKT cells in 4-1BB−/− mice. It is likely that during IAV infection or in an inflammatory setting, monocyte expression of CD1d provides a specific pathway for engagement with iNKT cells, causing expansion of iNKT cells, which then reduced the number of monocytes. This is a possible explanation for previous finding of significantly higher number of monocytes in the lungs of iNKT KO mice with severe IAV infection (24). The adoptive transfer studies (Fig. 5) suggest that iNKT cells directly interacted with monocytes in the lungs, causing apoptosis of the monocytes. One caveat to the adoptive transfer studies is that it was not clear how long the blocking Abs would have remained on iNKT cells after the initial coculture. Notwithstanding this, we believe that the most important interaction probably occurred within the first 24 h. 4-1BB and 4-1BBL expression were increased within 4 h after αGC activation and 24 h of IAV infection in vitro, and both 4-1BB/L–dependent cytokine production by monocytes (IL-12 and TNF-α) and monocytic apoptosis were induced by 24 h. Thus, within a 3-d time-frame, the impact of iNKT cells on monocytes and vice versa is likely to have relied on these early signals. Ideally, transfer of iNKT cells from 4-1BB–deficient mice, rather than 4-1BB–blocked iNKT cells would confirm this.

The mechanism by which iNKT cells induced apoptosis in monocytes appears to require both physical interaction and soluble factor(s). CD1d engagement is important as blockade caused loss of nearly all the induction of apoptosis found after activation of iNKT cells. The identity of the soluble factor is unknown. We specifically examined the possibility of TNF-α as a potential candidate but this proved not to be the case as neither blocking of TNF-α (with blocking Abs toward TNFR (TNFR1) as well as a neutralizing TNFα Ab) nor addition of rTNF-α changed the rate or level of apoptosing monocytes in iNKT–monocyte coculture experiments (data not shown), whereas it did increase IL-8 production from the THP-1 cell line. This is perhaps not surprising as high levels of TNF-α were produced by the CD8 T cell clone in the CD8 T cell clone–monocyte coculture experiments (Fig. 2D) but this did not cause apoptosis of monocytes. Apoptosis in this system was also not via Fas/Fas ligand (FasL), as the iNKT clones we used did not express FasL, and in Fas/FasL blocking experiments, no change in apoptosis was observed (data not shown).

However, it did seem that the iNKT cells were unique in their ability to induce apoptosis in monocytes, at least when compared with NK cells and a CD8 T cell clone. This fits with the specific requirement for CD1d in the iNKT–monocyte interaction and presents an interesting scenario where monocytes can seek out iNKT cells via their ability to engage by CD1d–TCR interaction and to maintain a selective connection at rest and during inflammation.

The site of interaction for iNKT cells and monocytes was not determined from this study. However, the finding that 4-1BB expression is highest on circulating iNKT cells (as compared with liver and spleen) and that changes in 4-1BB expression and levels of apoptotic monocytes during IAV infection are most evident in the blood rather than spleen suggest that this may occur in the circulation. This is not entirely implausible because monocytes are produced in the bone marrow and under normal circumstances, then enter the circulation where they reside for a very short time before undergoing spontaneous apoptosis (half life of 1 d in mice, 3 d in humans) (38, 43). Inflammatory stimuli can cause production in the bone marrow to increase as well as induce survival signals that cause monocytes to escape their apoptotic fate and increase their life span, allowing time to home to the site of infection (44). Thus, the opportunity to influence the function of monocytes likely resides in the short period when they have been generated, circulate in blood, and transit through secondary lymphoid organs. The fact that iNKT cell–monocyte interactions seem to require only a small number of interacting cells (Fig. 1D), which remained constant despite changes in the iNKT/monocytes ratio suggests that these cells contact and impart signals rapidly to each other. However, the sheer force generated by circulating cells in blood may make it difficult for these cells to interact in midflow. However, if either the iNKT cell or monocyte is anchored and relatively stationary, this could be possible. iNKT cells are known to crawl along the luminal surface of sinusoidal endothelial cells at very low speed (100-fold lower speed than the rolling phenomenon) (45), thereby providing circulating monocytes with a scaffold of relatively slow-moving iNKT cells to interact with. It is unclear whether this also occurs in nonsinusoidal endothelium. Intriguingly, the same investigators show that these iNKT cells stopped crawling when αGC was injected, suggesting that when activated, they engage with their Ag, presumably presented on CD1d. Other investigators have shown by intravital microscopy that activation of iNKT cells caused production of IL-12 and IL-18, which in turn induced cell arrest within liver sinusoids without a strict requirement for CD1d Ag presentation (46). This cell arrest behavior occurred within 1 h of cytokine exposure and precedes iNKT cell activation. This could again be a stabilizing factor to allow for formation of an immune synapse between iNKT cells and the Ag presentation cell. Our study did show that IL-12 production by monocytes was increased after activation of iNKT cells in a 4-1BB/L–dependent manner (Fig. 3E).

The bone marrow, lymph nodes, spleen, and lungs are other possible sites of interaction for iNKT cells and monocytes. Barral et al. (47) showed that iNKT cells were positioned within the interfollicular zones and the medulla of the lymph node and could actively scan and communicate with macrophages. The same group demonstrated that the majority of iNKT cells patrol (at a very low speed of 6 μm • min−1) around the marginal zone of spleen where they may form a net for receipt of circulating monocytes (48). Finally, in the lungs, it has been postulated that iNKT cells are long-term residents of the lung microvascular network (49). Blood collected from the right and left ventricles of the heart revealed enrichment of NKT cells by ∼20-fold compared with peripheral blood (50), suggesting a continuous accumulation and release of iNKT cells in the microvascular compartments of the lungs. Our in vivo data (Fig. 5B) suggest this is possible.

The significance of the 4-1BB/L pathway in the interaction between iNKT cells and monocytes is suggested by the marked increase in expression when iNKT cells are activated or monocytes are infected. Ligands of the TNF family are upregulated in inflamed tissues (51, 52), but to our knowledge, no previous reports have characterized 4-1BBL expression and function on monocytes during the innate immune response to IAV infection. There was increase in both the level of expression and proportion of iNKT cells and monocytes expressing 4-1BB and 4-1BBL, respectively, during IAV infection. 4-1BB/L costimulation also appears to be required for the full effector function of monocytes because production of two key cytokines, IL-12 and TNF-α, was reduced without this pathway. In particular, it appears that without 4-1BB/L costimulation, IL-12 production from monocytes appears completely abrogated. The reduction in iNKT cell proliferation by ∼50% when 4-1BB/L was blocked suggests a substantial costimulatory contribution. This study is one of only a few to examine the effects of these molecules on specific interacting primary human cells (e.g., iNKT and monocytes) and suggest that functions of this costimulatory pathway differ in steady state compared with inflammatory situations. Our findings reflects those from conventional T cells studies, where 4-1BB was found to have a role in maintaining survival and proliferation of activated T cells, and it has been proposed that one role of 4-1BB/L during IAV infection is to prolong CD8 T cell survival long enough to clear the virus (53). Increase in the percentage of IL-4–producing iNKT cells following 4-1BB/L blocking at 24 h could mean that 4-1BB/L costimulation is a later event (after the first burst of TCR signaling, also observed for conventional T cells) where it suppressed IL-4 production, indirectly polarizing iNKT cells to Th1 bias, in keeping with observations in conventional T cells (8). 4-1BB signals have also been shown to be important in stimulating expression of the antiapoptotic proteins Bcl-2 and Bcl-XL in CD4 T cells (54), and our results suggest that this role may also be relevant for iNKT cells, as 4-1BB was required to maintain optimal proliferation and survival of these cells in vitro.

In summary, our study suggests that monocytes interact with iNKT cells to increase expression of 4-1BBL and 4-1BB, and in conjunction with this pathway, maintain their numbers at baseline. When iNKT cells are activated, 4-1BB costimulation amplifies monocyte-mediated expansion of iNKT cells, which in turn, induced monocytes apoptosis. This provides a potential feedback mechanism for the control of monocyte numbers during inflammation. The study advances the understanding of this costimulatory pathway in iNKT–monocyte cross-talk and also provides information on a potential consequence of in vivo blockade on iNKT cells and monocytes.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Eleni Giannoulatou for statistical input and Yanchun Peng and Tao Dong for generating and providing the HIV nef–specific CD8 T cell clone.

Footnotes

  • This work was supported by the Wellcome Trust.

  • Abbreviations used in this article:

    BAL
    bronchoalveolar lavage
    DC
    dendritic cell
    FasL
    Fas ligand
    αGC
    alpha-galactosylceramide
    HAU
    hemagglutinating unit
    i
    invariant
    IAV
    influenza A virus
    KO
    knockout
    MFI
    mean fluorescence index
    WT
    wild-type.

  • Received September 6, 2013.
  • Accepted February 16, 2014.

References

  1. ↵
    1. Saoulli K.,
    2. S. Y. Lee,
    3. J. L. Cannons,
    4. W. C. Yeh,
    5. A. Santana,
    6. M. D. Goldstein,
    7. N. Bangia,
    8. M. A. DeBenedette,
    9. T. W. Mak,
    10. Y. Choi,
    11. T. H. Watts
    . 1998. CD28-independent, TRAF2-dependent costimulation of resting T cells by 4-1BB ligand. J. Exp. Med. 187: 1849–1862.
    OpenUrlAbstract/FREE Full Text
    1. Kwon B.,
    2. H. W. Lee,
    3. B. S. Kwon
    . 2002. New insights into the role of 4-1BB in immune responses: beyond CD8+ T cells. Trends Immunol. 23: 378–380.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Kim D. H.,
    2. W. S. Chang,
    3. Y. S. Lee,
    4. K. A. Lee,
    5. Y. K. Kim,
    6. B. S. Kwon,
    7. C. Y. Kang
    . 2008. 4-1BB engagement costimulates NKT cell activation and exacerbates NKT cell ligand-induced airway hyperresponsiveness and inflammation. J. Immunol. 180: 2062–2068.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Langstein J.,
    2. J. Michel,
    3. H. Schwarz
    . 1999. CD137 induces proliferation and endomitosis in monocytes. Blood 94: 3161–3168.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Schwarz H.
    2005. Biological activities of reverse signal transduction through CD137 ligand. J. Leukoc. Biol. 77: 281–286.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Langstein J.,
    2. J. Michel,
    3. J. Fritsche,
    4. M. Kreutz,
    5. R. Andreesen,
    6. H. Schwarz
    . 1998. CD137 (ILA/4-1BB), a member of the TNF receptor family, induces monocyte activation via bidirectional signaling. J. Immunol. 160: 2488–2494.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Shuford W. W.,
    2. K. Klussman,
    3. D. D. Tritchler,
    4. D. T. Loo,
    5. J. Chalupny,
    6. A. W. Siadak,
    7. T. J. Brown,
    8. J. Emswiler,
    9. H. Raecho,
    10. C. P. Larsen,
    11. et al
    . 1997. 4-1BB costimulatory signals preferentially induce CD8+ T cell proliferation and lead to the amplification in vivo of cytotoxic T cell responses. J. Exp. Med. 186: 47–55.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Kim Y. J.,
    2. S. H. Kim,
    3. P. Mantel,
    4. B. S. Kwon
    . 1998. Human 4-1BB regulates CD28 co-stimulation to promote Th1 cell responses. Eur. J. Immunol. 28: 881–890.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Ju S. W.,
    2. S. G. Ju,
    3. F. M. Wang,
    4. Z. J. Gu,
    5. Y. H. Qiu,
    6. G. H. Yu,
    7. H. B. Ma,
    8. X. G. Zhang
    . 2003. A functional anti-human 4-1BB ligand monoclonal antibody that enhances proliferation of monocytes by reverse signaling of 4-1BBL. Hybrid. Hybridomics 22: 333–338.
    OpenUrlCrossRefPubMed
    1. Ju S.,
    2. S. Ju,
    3. Y. Ge,
    4. H. Qiu,
    5. B. Lu,
    6. Y. Qiu,
    7. J. Fu,
    8. G. Liu,
    9. Q. Wang,
    10. Y. Hu,
    11. et al
    . 2009. A novel approach to induce human DCs from monocytes by triggering 4-1BBL reverse signaling. Int. Immunol. 21: 1135–1144.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Drenkard D.,
    2. F. M. Becke,
    3. J. Langstein,
    4. T. Spruss,
    5. L. A. Kunz-Schughart,
    6. T. E. Tan,
    7. Y. C. Lim,
    8. H. Schwarz
    . 2007. CD137 is expressed on blood vessel walls at sites of inflammation and enhances monocyte migratory activity. FASEB J. 21: 456–463.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Kim J. D.,
    2. E. A. Lee,
    3. N. N. Quang,
    4. H. R. Cho,
    5. B. Kwon
    . 2011. Recombinant TAT-CD137 ligand cytoplasmic domain fusion protein induces the production of IL-6 and TNF-α in peritoneal macrophages. Immune Netw. 11: 216–222.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Kim H. J.,
    2. J. S. Lee,
    3. J. D. Kim,
    4. H. J. Cha,
    5. A. Kim,
    6. S. K. Lee,
    7. S. C. Lee,
    8. B. S. Kwon,
    9. R. S. Mittler,
    10. H. R. Cho,
    11. B. Kwon
    . 2012. Reverse signaling through the costimulatory ligand CD137L in epithelial cells is essential for natural killer cell-mediated acute tissue inflammation. Proc. Natl. Acad. Sci. USA 109: E13–E22.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Jeon H. J.,
    2. J. H. Choi,
    3. I. H. Jung,
    4. J. G. Park,
    5. M. R. Lee,
    6. M. N. Lee,
    7. B. Kim,
    8. J. Y. Yoo,
    9. S. J. Jeong,
    10. D. Y. Kim,
    11. et al
    . 2010. CD137 (4-1BB) deficiency reduces atherosclerosis in hyperlipidemic mice. Circulation 121: 1124–1133.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Melero I.,
    2. J. V. Johnston,
    3. W. W. Shufford,
    4. R. S. Mittler,
    5. L. Chen
    . 1998. NK1.1 cells express 4-1BB (CDw137) costimulatory molecule and are required for tumor immunity elicited by anti‑4-1BB monoclonal antibodies. Cell. Immunol. 190: 167–172.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Melero I.,
    2. W. W. Shuford,
    3. S. A. Newby,
    4. A. Aruffo,
    5. J. A. Ledbetter,
    6. K. E. Hellström,
    7. R. S. Mittler,
    8. L. Chen
    . 1997. Monoclonal antibodies against the 4-1BB T-cell activation molecule eradicate established tumors. Nat. Med. 3: 682–685.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Houot R.,
    2. M. J. Goldstein,
    3. H. E. Kohrt,
    4. J. H. Myklebust,
    5. A. A. Alizadeh,
    6. J. T. Lin,
    7. J. M. Irish,
    8. J. A. Torchia,
    9. A. Kolstad,
    10. L. Chen,
    11. R. Levy
    . 2009. Therapeutic effect of CD137 immunomodulation in lymphoma and its enhancement by Treg depletion. Blood 114: 3431–3438.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Kwong B.,
    2. S. A. Gai,
    3. J. Elkhader,
    4. K. D. Wittrup,
    5. D. J. Irvine
    . 2013. Localized immunotherapy via liposome-anchored anti-CD137 + IL-2 prevents lethal toxicity and elicits local and systemic antitumor immunity. Cancer Res. 73: 1547–1558.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Wilcox R. A.,
    2. A. I. Chapoval,
    3. K. S. Gorski,
    4. M. Otsuji,
    5. T. Shin,
    6. D. B. Flies,
    7. K. Tamada,
    8. R. S. Mittler,
    9. H. Tsuchiya,
    10. D. M. Pardoll,
    11. L. Chen
    . 2002. Cutting edge: expression of functional CD137 receptor by dendritic cells. J. Immunol. 168: 4262–4267.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Pollok K. E.,
    2. Y. J. Kim,
    3. J. Hurtado,
    4. Z. Zhou,
    5. K. K. Kim,
    6. B. S. Kwon
    . 1994. 4-1BB T-cell antigen binds to mature B cells and macrophages, and costimulates anti‑mu-primed splenic B cells. Eur. J. Immunol. 24: 367–374.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Park S. H.,
    2. K. Benlagha,
    3. D. Lee,
    4. E. Balish,
    5. A. Bendelac
    . 2000. Unaltered phenotype, tissue distribution and function of Vα14+ NKT cells in germ-free mice. Eur. J. Immunol. 30: 620–625.
    OpenUrlCrossRefPubMed
    1. D’Andrea A.,
    2. D. Goux,
    3. C. De Lalla,
    4. Y. Koezuka,
    5. D. Montagna,
    6. A. Moretta,
    7. P. Dellabona,
    8. G. Casorati,
    9. S. Abrignani
    . 2000. Neonatal invariant Vα24+ NKT lymphocytes are activated memory cells. Eur. J. Immunol. 30: 1544–1550.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Stetson D. B.,
    2. M. Mohrs,
    3. R. L. Reinhardt,
    4. J. L. Baron,
    5. Z. E. Wang,
    6. L. Gapin,
    7. M. Kronenberg,
    8. R. M. Locksley
    . 2003. Constitutive cytokine mRNAs mark natural killer (NK) and NK T cells poised for rapid effector function. J. Exp. Med. 198: 1069–1076.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Kok W. L.,
    2. L. Denney,
    3. K. Benam,
    4. S. Cole,
    5. C. Clelland,
    6. A. J. McMichael,
    7. L. P. Ho
    . 2012. Pivotal advance: invariant NKT cells reduce accumulation of inflammatory monocytes in the lungs and decrease immune-pathology during severe influenza A virus infection. J. Leukoc. Biol. 91: 357–368.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Auffray C.,
    2. M. H. Sieweke,
    3. F. Geissmann
    . 2009. Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu. Rev. Immunol. 27: 669–692.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Gabrilovich D. I.,
    2. S. Nagaraj
    . 2009. Myeloid-derived suppressor cells as regulators of the immune system. Nat. Rev. Immunol. 9: 162–174.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Aldridge J. R., Jr..,
    2. C. E. Moseley,
    3. D. A. Boltz,
    4. N. J. Negovetich,
    5. C. Reynolds,
    6. J. Franks,
    7. S. A. Brown,
    8. P. C. Doherty,
    9. R. G. Webster,
    10. P. G. Thomas
    . 2009. TNF/iNOS-producing dendritic cells are the necessary evil of lethal influenza virus infection. Proc. Natl. Acad. Sci. USA 106: 5306–5311.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Ho L. P.,
    2. B. C. Urban,
    3. L. Jones,
    4. G. S. Ogg,
    5. A. J. McMichael
    . 2004. CD4‑CD8αα subset of CD1d-restricted NKT cells controls T cell expansion. J. Immunol. 172: 7350–7358.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Ho L. P.,
    2. B. C. Urban,
    3. D. R. Thickett,
    4. R. J. Davies,
    5. A. J. McMichael
    . 2005. Deficiency of a subset of T-cells with immunoregulatory properties in sarcoidosis. Lancet 365: 1062–1072.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Fernández Do Porto D. A.,
    2. J. O. Jurado,
    3. V. Pasquinelli,
    4. I. B. Alvarez,
    5. R. H. Aspera,
    6. R. M. Musella,
    7. V. E. García
    . 2012. CD137 differentially regulates innate and adaptive immunity against Mycobacterium tuberculosis. Immunol. Cell Biol. 90: 449–456.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Dong T.,
    2. G. Stewart-Jones,
    3. N. Chen,
    4. P. Easterbrook,
    5. X. Xu,
    6. L. Papagno,
    7. V. Appay,
    8. M. Weekes,
    9. C. Conlon,
    10. C. Spina,
    11. et al
    . 2004. HIV-specific cytotoxic T cells from long-term survivors select a unique T cell receptor. J. Exp. Med. 200: 1547–1557.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Ho L. P.,
    2. L. Denney,
    3. K. Luhn,
    4. D. Teoh,
    5. C. Clelland,
    6. A. J. McMichael
    . 2008. Activation of invariant NKT cells enhances the innate immune response and improves the disease course in influenza A virus infection. Eur. J. Immunol. 38: 1913–1922.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Huaux F.,
    2. T. Liu,
    3. B. McGarry,
    4. M. Ullenbruch,
    5. S. H. Phan
    . 2003. Dual roles of IL-4 in lung injury and fibrosis. J. Immunol. 170: 2083–2092.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Ho W. T.,
    2. W. L. Pang,
    3. S. M. Chong,
    4. A. Castella,
    5. S. Al-Salam,
    6. T. E. Tan,
    7. M. C. Moh,
    8. L. K. Koh,
    9. S. U. Gan,
    10. C. K. Cheng,
    11. H. Schwarz
    . 2013. Expression of CD137 on Hodgkin and Reed-Sternberg cells inhibits T-cell activation by eliminating CD137 ligand expression. Cancer Res. 73: 652–661.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Lee H. W.,
    2. S. J. Park,
    3. B. K. Choi,
    4. H. H. Kim,
    5. K. O. Nam,
    6. B. S. Kwon
    . 2002. 4-1BB promotes the survival of CD8+ T lymphocytes by increasing expression of Bcl-xL and Bfl-1. J. Immunol. 169: 4882–4888.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. van Furth R.,
    2. Z. A. Cohn
    . 1968. The origin and kinetics of mononuclear phagocytes. J. Exp. Med. 128: 415–435.
    OpenUrlAbstract
  37. ↵
    1. Zhang Y.,
    2. M. J. Morgan,
    3. K. Chen,
    4. S. Choksi,
    5. Z.-G. Liu
    . 2012. Induction of autophagy is essential for monocyte-macrophage differentiation. Blood 119: 2895–2905.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Parihar A.,
    2. T. D. Eubank,
    3. A. I. Doseff
    . 2010. Monocytes and macrophages regulate immunity through dynamic networks of survival and cell death. J. Innate Immun. 2: 204–215.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Kawano T.,
    2. J. Cui,
    3. Y. Koezuka,
    4. I. Toura,
    5. Y. Kaneko,
    6. H. Sato,
    7. E. Kondo,
    8. M. Harada,
    9. H. Koseki,
    10. T. Nakayama,
    11. et al
    . 1998. Natural killer-like nonspecific tumor cell lysis mediated by specific ligand-activated Vα14 NKT cells. Proc. Natl. Acad. Sci. USA 95: 5690–5693.
    OpenUrlAbstract/FREE Full Text
    1. Takahashi T.,
    2. K. Haraguchi,
    3. S. Chiba,
    4. M. Yasukawa,
    5. Y. Shibata,
    6. H. Hirai
    . 2003. Vα24+ natural killer T-cell responses against T-acute lymphoblastic leukaemia cells: implications for immunotherapy. Br. J. Haematol. 122: 231–239.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Metelitsa L. S.,
    2. K. I. Weinberg,
    3. P. D. Emanuel,
    4. R. C. Seeger
    . 2003. Expression of CD1d by myelomonocytic leukemias provides a target for cytotoxic NKT cells. Leukemia 17: 1068–1077.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Vinay D. S.,
    2. B. K. Choi,
    3. J. S. Bae,
    4. W. Y. Kim,
    5. B. M. Gebhardt,
    6. B. S. Kwon
    . 2004. CD137-deficient mice have reduced NK/NKT cell numbers and function, are resistant to lipopolysaccharide-induced shock syndromes, and have lower IL-4 responses. J. Immunol. 173: 4218–4229.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Fahy R. J.,
    2. A. I. Doseff,
    3. M. D. Wewers
    . 1999. Spontaneous human monocyte apoptosis utilizes a caspase-3‑dependent pathway that is blocked by endotoxin and is independent of caspase-1. J. Immunol. 163: 1755–1762.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Trottier M. D.,
    2. R. Irwin,
    3. Y. Li,
    4. L. R. McCabe,
    5. P. J. Fraker
    . 2012. Enhanced production of early lineages of monocytic and granulocytic cells in mice with colitis. Proc. Natl. Acad. Sci. USA 109: 16594–16599.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Geissmann F.,
    2. T. O. Cameron,
    3. S. Sidobre,
    4. N. Manlongat,
    5. M. Kronenberg,
    6. M. J. Briskin,
    7. M. L. Dustin,
    8. D. R. Littman
    . 2005. Intravascular immune surveillance by CXCR6+ NKT cells patrolling liver sinusoids. PLoS Biol. 3: e113.
    OpenUrlCrossRefPubMed
  46. ↵
    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
  47. ↵
    1. Barral P.,
    2. P. Polzella,
    3. A. Bruckbauer,
    4. N. van Rooijen,
    5. G. S. Besra,
    6. V. Cerundolo,
    7. F. D. Batista
    . 2010. CD169+ macrophages present lipid antigens to mediate early activation of iNKT cells in lymph nodes. Nat. Immunol. 11: 303–312.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Barral P.,
    2. M. D. Sánchez-Niño,
    3. N. van Rooijen,
    4. V. Cerundolo,
    5. F. D. Batista
    . 2012. The location of splenic NKT cells favours their rapid activation by blood-borne antigen. EMBO J. 31: 2378–2390.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Scanlon S. T.,
    2. S. Y. Thomas,
    3. C. M. Ferreira,
    4. L. Bai,
    5. T. Krausz,
    6. P. B. Savage,
    7. A. Bendelac
    . 2011. Airborne lipid antigens mobilize resident intravascular NKT cells to induce allergic airway inflammation. J. Exp. Med. 208: 2113–2124.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Thomas S. Y.,
    2. S. T. Scanlon,
    3. K. G. Griewank,
    4. M. G. Constantinides,
    5. A. K. Savage,
    6. K. A. Barr,
    7. F. Meng,
    8. A. D. Luster,
    9. A. Bendelac
    . 2011. PLZF induces an intravascular surveillance program mediated by long-lived LFA-1‑ICAM-1 interactions. J. Exp. Med. 208: 1179–1188.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Seko Y.,
    2. N. Takahashi,
    3. H. Oshima,
    4. O. Shimozato,
    5. H. Akiba,
    6. K. Takeda,
    7. T. Kobata,
    8. H. Yagita,
    9. K. Okumura,
    10. M. Azuma,
    11. R. Nagai
    . 2001. Expression of tumour necrosis factor (TNF) ligand superfamily co-stimulatory molecules CD30L, CD27L, OX40L, and 4-1BBL in murine hearts with acute myocarditis caused by Coxsackievirus B3. J. Pathol. 195: 593–603.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Malmström V.,
    2. D. Shipton,
    3. B. Singh,
    4. A. Al-Shamkhani,
    5. M. J. Puklavec,
    6. A. N. Barclay,
    7. F. Powrie
    . 2001. CD134L expression on dendritic cells in the mesenteric lymph nodes drives colitis in T cell-restored SCID mice. J. Immunol. 166: 6972–6981.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Lin G. H.,
    2. B. J. Sedgmen,
    3. T. J. Moraes,
    4. L. M. Snell,
    5. D. J. Topham,
    6. T. H. Watts
    . 2009. Endogenous 4-1BB ligand plays a critical role in protection from influenza-induced disease. J. Immunol. 182: 934–947.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    1. Lee H. W.,
    2. K. O. Nam,
    3. S. K. Seo,
    4. Y. H. Kim,
    5. H. Kang,
    6. B. S. Kwon
    . 2003. 4-1BB cross-linking enhances the survival and cell cycle progression of CD4 T lymphocytes. Cell. Immunol. 223: 143–150.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section