Tim-3 regulates pro- and anti-inflammatory cytokine expression in human CD14⁺ monocytes

Regulation of pro- and anti-inflammatory innate immune responses by Tim-3 in macrophages.

Abstract

Tim-3 and PD-1 are powerful immunoinhibitory molecules involved in immune tolerance, autoimmune responses, and antitumor or antiviral immune evasion. A current model for Tim-3 regulation during immune responses suggests a divergent function, such that Tim-3 acts synergistically with TLR signaling pathways in innate immune cells to promote inflammation, yet the same molecule terminates Th1 immunity in adaptive immune cells. To better understand how Tim-3 might be functioning in innate immune responses, we examined the kinetics of Tim-3 expression in human CD14⁺ M/M_Ф in relation to expression of IL-12, a key cytokine in the transition of innate to adaptive immunity. Here, we show that Tim-3 is constitutively expressed on unstimulated peripheral blood CD14⁺ monocytes but decreases rapidly upon TLR stimulation. Conversely, IL-12 expression is low in these cells but increases rapidly in CD14⁺ M/M_Ф in correlation with the decrease in Tim-3. Blocking Tim-3 signaling or silencing Tim-3 expression led to a significant increase in TLR-mediated IL-12 production, as well as a decrease in activation-induced up-regulation of the immunoinhibitor, PD-1; TNF-α production was not altered significantly, but IL-10 production was increased. These results suggest that Tim-3 has a role as a regulator of pro- and anti-inflammatory innate immune responses.

Introduction

Powerful mechanisms are in place to prevent unnecessary activation of host immune cells and autoimmunity. One such mechanism is the intricate balance between positive and negative costimulatory signals delivered to immune cells after antigenic encounter. The recently described Tim-3 and PD-1 immunoinhibitory pathways represent such negative mechanisms at play, with evidence emerging for their involvement in normal immune tolerance, autoimmune responses, and antitumor or antiviral immune evasion [1, 2]. Tim-3 is a type 1 membrane protein with a structurally conserved Ig variable domain and mucin stalk that connects to an intracellular tail [1]. Initially identified as preferentially expressed on activated Th1 cells, but not Th2 cells, Tim-3 has since been found to express and regulate the function of APCs, Tregs, and NK cells [3–10].

Although extensive studies have demonstrated that Tim-3 is an inhibitory molecule on Th1/type-1 cytotoxic T cells, it has been postulated that Tim-3 serves opposing roles in the innate and adaptive immune systems [3]. In the naïve resting immune cells, Tim-3 is primarily expressed on DCs and synergizes with TLRs to promote inflammation; once adaptive Th1 responses are generated, Tim-3 has increased expression on terminally differentiated Th1 cells and ligates with the up-regulated galectin-9 to terminate Th1 immunity [3]. On the one hand, Tim-3 has been shown to negatively regulate macrophage activation, and Tim-3 signaling on cells of the innate immune system critically influences regulation of adaptive immune responses [5, 6, 8, 10]. For example, in vivo administration of Tim-3 antibody enhances a Th1-dependent autoimmune encephalomyelitis by increasing the number and activation of macrophages [8]. Blocking Tim-3 signaling during an innate immune response to viral infection reduces CD80 costimulatory molecule expression on macrophages, leading to a decreased CTLA-4 level in CD4⁺ T cells, decreased Tregs, and increased inflammatory heart disease [6]. On the other hand, galectin-9 has been reported to induce maturation of human monocyte-derived DCs and promote phagocytosis of apoptotic cells and cross-presentation of dying cell-associated antigens to T cells [4, 9]. Therefore, the role of Tim-3 in macrophage functions is clearly complex, controversial, and requires further investigation.

Here, we used kinetic studies to show high expression of Tim-3 on naïve M/M_Ф and rapid Tim-3 down-regulation following TLR stimulation, associated with up-regulation of IL-12 but also IL-10. Proinflammatory IL-12 and anti-inflammatory IL-10 expression, as well as the activation-associated up-regulation of immunoinhibitor PD-1, was dependent on Tim-3 expression and signaling. We thus propose that Tim-3 functions as a brake on TLR-drive IL-12 and IL-10 expression during innate immune responses but that a balance of Tim-3-dependent cytokine signaling may ultimately determine the inflammatory outcome.

MATERIALS AND METHODS

Cell isolation and culture

Human PBMCs were isolated from peripheral blood using Ficoll density gradient centrifugation (Atlanta Biological, Lawrenceville, GA, USA). CD14⁺ M/M_Ф were further purified from PBMCs by positive selection with anti-CD14 magnetic beads or negative selection using a Monocyte Isolation Kit II with column purification per the manufacturer's instruction (Miltenyi Biotec, Auburn, CA, USA). The cells were cultured with RPMI 1640 containing 10% FBS, 100 μg/ml penicillin-streptomycin (Thermo Scientific, Logan, UT, USA), and 2 mM L-glutamine (Thermo Scientific). The protocol was approved by East Tennessee State University/Veteran's Administration IRB (Johnson City, TN, USA), and all patients underwent a written, informed-consent process approved by the IRB.

Flow cytometry

PBMCs from 10 healthy subjects were stimulated with 5 μg/ml LPS (InvivoGen, San Diego, CA, USA) and 5 μg/ml R848 (InvivoGen), which has a synergistic effect on IL-12 production, for various times, as indicated in the presence of Brefeldin A (BioLegend, San Diego, CA, USA) for the last 6 h before harvesting to inhibit cytokine transport. Cell surface staining was carried out using anti-Tim 3-allophycocyanin or anti-Tim 3-PE (R&D Systems, Minneapolis, MN, USA), anti-Tim-3-PE clone F38-2E2 (BioLegend), or anti-PD-1-PE and anti-CD14-FITC (BD Biosciences, San Jose, CA, USA), followed by intracellular IL-12 staining using an Inside Stain Kit (Miltenyi Biotec), per the manufacturer's instruction. Isotype-matched control antibodies (BD Biosciences) were used to determine the levels of background staining. For negatively selected M/M_Ф, cytokines, including IL-10, IL-6, TNF-α, IFN-γ, and the costimulatory molecule CD83, were also detected following LPS/R848 stimulation and Tim-3 blockade. PE-IL-6 and allophycocyanin-TNF-α were purchased from eBioscience (San Diego, CA, USA), and allophycocyanin-CD83, PE-IL-10, and IFN-γ were from Miltenyi Biotec. The cells were analyzed on a FACSCalibur flow cytometry (Becton Dickinson, Franklin Lakes, NJ, USA). Twenty thousand events were collected after gating on the M/M_Ø populations and analyzed by CellQuest software.

TIM-3 blockade

Healthy PBMCs or positively purified CD14⁺ M/M_Ф were incubated with 10 μg/ml anti-TIM-3 or control IgG antibody (R&D Systems) for 48 h; negatively selected CD14⁺ M/M_Ф were incubated with 10 μg/ml LEAF anti-Tim-3 (clone F38-2E2, BioLegend) or LEAF control IgG antibody for 48 h. These groups of purified cells were stimulated with or without 5 μg/ml LPS and 5 μg/ml R848 for 24 h. Flow cytometric analysis of PD-1, IL-12, CD83, and various other cytokine expressions was carried out as described above.

Tim-3 siRNA silencing

THP-1 cells (3×10⁵) were incubated with 60 pmol Tim-3 siRNA duplex or control siRNA in 200 μl siRNA transfection medium in a six-well plate, per the manufacturer's instruction (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Following 6 h of incubation at 37°C, normal growth medium containing 2× concentrations of normal serum and antibiotics was added. The transfected cells were stimulated with 5 μg/ml LPS and R848 for 18 h, followed by detection of Tim-3, IL-12, IL-10, and other cellular markers or cytokines as described above.

RT-PCR and real-time PCR

Total RNAs were isolated using the Qiagen RNeasy mini kit (Qiagen, Valencia, CA, USA) from THP-1 cells following siRNA silencing. After being treated with DNase I (Promega, Madison, WI, USA), 1 μg/ml RNAs were reverse-transcribed (Promega), and the cDNAs were amplified by the PCR at conditions: 94°C 4 min, followed by 94°C 40 s, 58°C 40 s, 72°C 1 min for 33 cycles, and then 72°C 5 min. The primers for gene amplifications include: Tim-3 (910 bp), sense 5′-CAT GTT TTC ACA TCT TCC C-3′, antisense 5′-CTA TGG CAT TGC AAA GCG ACA-3′; and β-actin, sense 5′-ATGGGTCAGAAGGACTCCTATG-3′, antisense 5′-ATCTCCTGCTCGA AGTCTAGAG-3′. The amplified products were analyzed by electrophoresis on 2.0% agarose gels.

Real-time quantitative PCR was performed with a Bio-Rad iQ5 sequence detection system using the SYBR reagents. Primers were designed per GenBank AF450242.1 using Primer Express software (Applied Biosystems, Foster City, CA, USA). Primers for Tim-3 target (96 bp) include sense 5′-TCC AAG GAT GCT TAC CAC CAG-3′, antisense, 5′-GCC AAT GTG GAT ATT TGT GTT AGA T-3′. Primers for internal control β2M were available commercially. The real-time PCR was conducted using the following conditions: 50°C for 2 min, 95°C for 10 min, and 95°C for 15 s, 60°C 15 s, 72°C 30 s for 40 cycles, followed by 72°C 10 min. Quantification of the genes was performed using the comparative threshold method. The expression level of each mRNA was normalized to β2M mRNA and represented as n-fold difference relative to the calibrator control. All PCR assays were performed in triplicate, and results are represented by the mean ± sd.

Statistical analysis

Results are expressed as means ± sd. All of the analyses for determining the level of significance were performed using independent samples′ t test program of SPSS 18 software. Values of *P < 0.05 were considered significant, and values of **P < 0.01 or ***P < 0.001 were considered very significant.

RESULTS AND DISCUSSION

Dynamic expression of Tim-3 and IL-12 in CD14⁺ M/M_Ø following TLR stimulation

As an initial approach to determine the role of Tim-3 in regulation of innate immune cells, we first examined healthy human PBMCs for the cell surface expression of Tim-3 and intracellular expression of IL-12 in resting, “naïve”, and TLR-activated human CD14⁺ M/M_Φ using flow cytometric analysis. As shown in Fig. 1A and B, naïve CD14⁺ M/M_Φ from multiple healthy subjects exhibited a fairly high level of Tim-3 with low if any IL-12 expression; but upon TLR stimulation, Tim-3 expression declined significantly, accompanied by a significant increase in IL-12 production, primarily by CD14⁺/Tim-3^– M/M_Ø. To determine the specific effect of TLR stimulation on M/M_Ø, positively selected, purified CD14⁺ M/M_Ø were stimulated with or without TLR and subjected to the Tim-3/IL-12 detection as described above, and we found similar results (Supplemental Fig. 1). To address the potential issues of monocyte stimulation during positive selection, CD14⁺ monocytes were negatively selected prior to stimulation as above (Fig. 1C) and confirmed these findings; these monocytes were also assayed using a different anti-Tim-3 antibody clone to verify specificity (anti-Tim-3-PE, clone F38-2E2). A time-course of Tim-3 expression (Fig. 1D) revealed a rapid reduction in the first 24 h that appeared to slowly resolve over the ensuing 48 h following TLR stimulation, and this alteration of Tim-3 expression was inversely associated with IL-12 production. Therefore, it appears that a high level of baseline Tim-3 expression in CD14⁺ M/M_Ø declines rapidly upon TLR stimulation, which may allow the cells to elicit IL-12 expression. Other costimulatory molecules/cytokines were also examined in negatively selected monocytes, with increased expression of IL-6, IL-10, and TNF-α observed following TLR stimulation (Fig. 1E).

Figure 1. Dynamic expressions of Tim-3 and IL-12 upon TLR stimulation.

(A) Decreased Tim-3 expression and increased IL-12 production in CD14⁺ M/M_Ø. PBMCs isolated (iso) from healthy subjects were incubated with or without TLR ligand LPS/R848 for 18 h, followed by triple immunostaining and flow cytometric analysis for the expressions of CD14, Tim-3, and IL-12. Cells were first gated on the monocytic population and then analyzed for the percentage of Tim-3 (red panels)- or IL-12 (green panels)-positive cells in the CD14⁺ cell population, which is shown on the top right corner of the dot plot. The relationship of Tim-3 and IL-12 expressions (blue panels) in the gated CD14⁺ M/M_Ø is also shown. SSC/FSC, Side-/forward-scatter. (B) Summary data (mean±sd from 10 healthy subjects) of Tim-3⁺CD14⁺ M/M_Ø and IL-12⁺CD14⁺ M/M_Ø in the resting and activated status are shown. ***P < 0.001. unsti, Unstimulated; sti, stimulated. (C) Tim-3 and IL-12 expressions were detected on negatively purified CD14⁺ M/M_Ø, with or without LPS/R848 stimulation and using anti-Tim-3-PE clone F38-2E2. Summary data (mean±sd from four healthy subjects) of Tim-3⁺CD14⁺ M/M_Ø and IL-12⁺CD14⁺ M/M_Ø in resting and activated status are shown below. *P < 0.05; ***P < 0.001. (D) Kinetics of Tim-3 and IL-12 expressions in CD14⁺ M/M_Ø following TLR stimulation. PBMCs isolated from three healthy subjects were incubated with or without TLR ligand LPS/R848 for various time-points, followed by triple immunostaining and flow cytometric analysis for the expressions of CD14, Tim-3, and IL-12. The percentage of Tim-3⁺CD14⁺ M/M_Ø and IL-12⁺CD14⁺ M/M_Ø at different time-points is calculated, and the mean ± sd of the double-positive cells from these healthy subjects is shown. (E) CD14⁺ M/M_Ø negatively selected from healthy subjects were stimulated with LPS/R848 as above, followed by staining and flow cytometric analysis for the expressions of CD83, IL-6, IL-10, and TNF-α. Summary data of percentages of positive cells are shown. **P < 0.01; ***P < 0.001.

Tim-3 signaling regulates IL-12 expression in human CD14⁺ M/M_Ø

Based on the inverse correlation of Tim-3/IL-12 expression upon TLR stimulation, we proposed that Tim-3 would function as a brake or cap on TLR-mediated IL-12 production. To assess this possibility, we incubated PBMCs with anti-Tim-3 or control antibody followed by TLR stimulation. We observed a significant increase in IL-12 expression by CD14⁺ M/M_Ø upon Tim-3 blockade (Fig. 2A). The data were reproducible in highly purified, positively selected CD14⁺ M/M_Ø and treated in the same way as described (Fig. 2B), suggesting that the improvement of IL-12 expression is an effect of blockade of Tim-3 signaling directly on M/M_Ø instead of secondary on other cells in PBMCs. Blockade of Tim-3 in the absence of TLR stimulation did not alter IL-12 expression, and the improvement of IL-12 expression was only observed in the Tim-3-blocked, TLR-stimulated cells (Fig. 2C). These findings support the possibility that Tim-3 is indeed important in inducing inflammation per se but does so by a decrease in its expression that permits TLR signaling to drive inflammatory responses.

Figure 2. Blockade of Tim-3 improves IL-12 expression.

(A) PBMCs isolated from six healthy subjects were preincubated with Tim-3 antibody or a control antibody for 48 h, followed by TLR stimulation for 24 h. After triple immunostaining, the cells were gated on the monocytic populations, IL-12 expression by CD14⁺ M/M_Ø was analyzed, and the summary data from six subjects are shown below. **P < 0.01. (B) CD14⁺ M/M_Ø were positively purified from PBMCs, treated in the same way as described in A, IL-12 expression by the purified cells was analyzed, and the summary data from three subjects are shown below. **P < 0.01. (C) Tim-3 blockade enhances IL-12 production in CD14⁺ M/M_Ø in the presence, but not in the absence, of TLR stimulation. PBMCs isolated from healthy subjects were preincubated with Tim-3 antibody (R&D Systems) or a control antibody for 48 h and then stimulated without or with LPS/R848 for 18 h, followed by triple immunostaining. The upper panel shows the representative dot plots of IL-12 expression in CD14⁺ M/M_Ø with various treatments. Summary data (mean±sd) of percentages of IL-12⁺CD14⁺ M/M_Ø with various treatments from four healthy subjects are shown below. **P < 0.01.

Tim-3 regulates PD-1 expression on TLR-stimulated M/M_Ø

We have shown recently that PD-1 negatively regulates IL-12 expression by M/M_Ø during chronic hepatitis C virus infection [11]. As PD-1 is an activation-induced negative immunomodulator involved in feedback regulation of immune responses by preventing cell overactivation [10], and Tim-3 appears constitutionally expressed on naïve M/M_Ø and functions as a cap for cell initial activation, we hypothesized that altering Tim-3 function would also affect PD-1 expression. Different from the expression pattern of Tim-3, we observed that naïve CD14⁺ M/M_Ø expressed low levels of PD-1, which increased upon TLR stimulation in Tim-3⁺ and Tim-3^– populations (Fig. 3A). We confirmed these findings in positively (Supplemental Fig. 2) and negatively (Fig. 3B) selected CD14⁺ monocyte populations. Interestingly, we found that PD-1 and Tim-3 could be singly or dually expressed on distinct cell populations; in addition, they exhibited a different expression pattern dynamically. These results suggest that these two inhibitory molecules might function at different stages of cell activation or might help to functionally identify specific cell populations. We then preincubated PBMCs (Fig. 4A) or positively (Supplemental Fig. 3) or negatively (Fig. 4B) purified CD14⁺ M/M_Ø with anti-Tim-3 or isotype control antibody for 48 h, followed by TLR stimulation for 18 h, and analyzed PD-1 and IL-12 expressions in the CD14⁺ M/M_Ø population. We observed a decrease in PD-1 expression and an increase in IL-12 production upon Tim-3 blockade; and costaining of the CD14⁺ M/M_Ø for PD-1 and IL-12 revealed that the increase in IL-12 following Tim-3 blockade and TLR stimulation occurred predominantly in the CD14⁺PD-1^– subpopulation. Thus, the activation-associated PD-1 up-regulation that occurs as a delayed feedback mechanism for preventing cell overactivation appears to be linked and regulated by altering Tim-3 signaling, which appears to function as a cap on initial cell activation.

Figure 3. PD-1 expression is increased following TLR stimulation.

(A) PD-1 expression on CD14⁺ M/M_Ø is increased following TLR stimulation of whole PBMCs. PBMCs isolated from healthy subjects were incubated with or without TLR ligand LPS/R848 for 18 h, followed by triple immunostaining and flow cytometric analysis for the expressions of CD14, Tim-3, and PD-1. The cells were gated on the monocytic population and then analyzed for the percentage of Tim-3 (red panels)- or PD-1 (green panels)-positive cells in the CD14⁺ cells. The relationship of Tim-3 and PD-1 expressions (blue panels) in the gated CD14⁺ M/M_Ø is also shown. Summary data (mean±sd from six healthy subjects) of PD-1⁺CD14⁺ M/M_Ø are shown below. *P < 0.05. (B) Relationship of Tim-3 and PD-1 expressions on negatively selected, purified CD14⁺ M/M_Ø following TLR stimulation and analyzed using anti-Tim-3-PE clone F38-2E2. Summary data (mean±sd from four healthy subjects) of Tim-3⁺ versus PD-1⁺ cells on CD14⁺ M/M_Ø are shown below. *P < 0.05; **P < 0.01.

Figure 4. Tim-3 regulates PD-1 expression in TLR-stimulated M/M_Ø.

(A) Tim-3 blockade inhibits PD-1 expression and enhances IL-12 production in CD14⁺ M/M_Φ following TLR stimulation of PBMCs. PBMCs isolated from healthy subjects were preincubated with Tim-3 antibody (R&D Systems) or a control antibody for 48 h, followed by TLR stimulation for 18 h. After triple immunostaining, the cells were first gated on monocytic populations and then gated further on CD14⁺ M/M_Ø. The representative dot plots show the relationship of PD-1 and IL-12 expression in CD14⁺ M/M_Ø treated with a control IgG or Tim-3 antibody. Summary data (mean±sd from six healthy subjects) of PD-1⁺ versus IL-12⁺ cells in CD14⁺ M/M_Ø treated with a control IgG or Tim-3 antibody are shown below. *P<0.05. (B) Tim-3 blockade (anti-Tim-3 clone F38-2E2) inhibits PD-1 expression and enhances IL-12 production in negatively selected, purified CD14⁺ M/M_Φ following TLR stimulation. Purified CD14⁺ M/M_Φ were treated and analyzed as described in A; a representative dot plot of PD-1⁺ versus IL-12⁺ cells is shown above, and summary data (mean±sd from four healthy subjects)—including PD-1⁺, IL-12⁺ cells, as well as expression of the costimulatory molecule CD83, IL-6, IL-10, and TNF-α, and CD14⁺ M/M_Ø treated with a control IgG or Tim-3 antibody—are shown below; *P < 0.05.

We also examined the expression of other cytokines/costimulatory molecules in negatively selected monocytes following stimulation and Tim-3 blockade using anti-Tim-3 antibody (clone F38-2E2; Fig. 4B). We observed no significant change in expression of CD83 or TNF-α following Tim-3 blockade, whereas IL-10 and IL-6 appeared to be somewhat increased. To confirm that Tim-3 was linked to IL-12 expression, we next silenced Tim-3 expression using THP-1 monocytes and assayed for IL-12 expression following stimulation. As shown in Fig. 5A, Tim-3 expression on THP-1 cells was suppressed significantly by transfecting the cells with Tim-3-specific siRNA compared with control siRNA at 48 h, assessed by flow cytometry, end-point RT-PCR, and real-time PCR. Importantly, we found that Tim-3 silencing enhanced the ability of THP-1 monocytes to express IL-12 (Fig. 5B and C). This was not observed with IL-6 but was for IL-10 expression (Fig. 5D), supporting the possibility that Tim-3 might be a fundamental regulator of monocyte functions.

Figure 5. Tim-3 negatively regulates IL-12 expression in THP-1 cells treated with LPS/R848.

THP-1 cells transfected with Tim-3 siRNA or control siRNA and stimulated with LPS/R848 for 48 h were subjected to flow cytometric analysis for expression of Tim-3, IL-12, and cytokines. (A) THP-1 cells were transfected with Tim-3 siRNA (RNAi) or RNA control. Flow cytometric analysis of Tim-3 expression using anti-Tim-3-PE is shown on the left. RT-PCR as well as real-time PCR analysis was carried out for Tim 3 expression using β2M as an internal control and is shown on the right. Summary data are shown below. *P < 0.05. (B) Tim-3-mediated IL-12 inhibition is partially restored in THP-1 cells transfected with Tim-3 siRNA followed by TLR stimulation. The summary data of three repeated experiments are shown on the right. *P < 0.05. (C) Dual staining of Tim-3 and IL-12 following Tim-3 silencing. FL2/4-H, Fluorescence 2/4-height; APC, allophycocyanin. (D) Flow cytometric detection of IL-6 and IL-10 expressions in THP-1 cells by Tim-3 gene silencing. The summary data of three repeated experiments are shown. *P < 0.05.

As its name implies, the Tim-3 protein was identified originally on the surface of differentiated Th1 cells other than Th2 cells and was thought to be a T cell-specific modulator [1]. We now know that Tim-3 is also expressed on and regulates the functions of other types of immune cells, notably APCs, NK, and Tregs [3–10], and has more complex roles in innate immune regulation. Consistent with our data, increased inflammatory heart disease was observed upon blocking Tim-3 signaling during innate immune responses to viral infection, leading to decreased CTLA-4 levels in CD4⁺ T cells and decreased Tregs [6].

A current model suggests that Tim-3 serves opposing roles in the innate and adaptive immune systems; i.e., in the naïve resting immune system, Tim-3 is highly expressed on human DCs, but not monocytes, and synergizes with TLRs to promote inflammation; once Th1 responses are generated, Tim-3 is highly expressed on terminally differentiated Th1 cells and induces the up-regulation of galectin-9 to terminate Th1 immunity [3]. To explain the divergent roles, investigators have suggested differential effects on downstream signaling pathways, such as ERK. Our data would suggest that Tim-3 serves as a negative regulator of IL-12, dependent on Tim-3 expression levels. In adaptive immune cells, baseline Tim-3 expression on resting CD4⁺ and CD8⁺ T cells is low and is up-regulated upon TCR activation (data not shown). In contrast, baseline Tim-3 expression on resting M/M_Ø is quite high and associated with little if any IL-12 production. These findings support a model in which Tim-3 functions as a brake or cap on certain, but clearly not all, innate immune responses. This brake is at least transiently released, as Tim-3 surface expression declines rapidly following TLR activation, which provides the driving force that promotes inflammatory responses. Releasing Tim-3 (the brake) without providing acceleration (TLR stimulation) would have no consequence, whereas releasing Tim-3 and providing the necessary stimulation would result in enhanced M/M_Ø function. Consistent with this, we have found enhanced IL-12 production in Tim-3-blocked M/M_Ø with TLR stimulation but not in the absence of TLR activation (Fig. 2). The model we propose would be similar to what has been offered for A20, a ubiquitin-editing enzyme involved in TLR-mediated inflammation, which may function as a cap on inflammatory responses that must be released in the setting of an accelerant [12].

Complicating this concept is the finding that Tim-3 blockade or silencing led to increased IL-10 expression (but no significant effect on TNF-α expression), suggesting the possibility that loss of Tim-3 signaling might ultimately dampen Th1 responses. This increase in IL-10 is more consistent with murine data from Anderson et al. [3], where Tim-3 appeared to be proinflammatory in the setting of a monoclonal Tim-3 agonist. The partial opposition of our data to the limited human data presented to date may be a function of many issues. Some data suggest galectin-9 may be proinflammatory, independent of its role in Tim-3 signaling [13]. We are also exploring the possibility that how galectin-9 interacts with Tim-3 (trans- vs. cis-association) might be important to cell activation, as suggested by Hirashima and co-workers [14]. It may also be the case that how the Th1 response is affected is a function of not only Tim-3 expression but also the differentation status of M/M_Ø/DCs and the balance of inhibitory and stimulatory molecules involved. Ultimately, the overall cytokine milieu that would include IL-10, IL-12, and TNF-α may determine whether a pro- or anti-inflammatory response is elicited. This milieu may result from the particular stimulus, its concentration, cell–cell interactions, and other related signals such as those mediated by TLRs.

Collectively, these data provide a firm demonstration that Tim-3 is a regulator for key pro- and anti-inflammatory cytokine production in human CD14⁺ M/M_Ø. How TLR stimulation leads to loss of Tim-3 expression on the surface of CD14⁺ M/M_Ø is not as yet clear, but possibilities include a direct connection via an intracellular intermediate that down-regulates Tim-3 expression or perhaps an effect of TLR stimulation on other immune cells that leads to Tim-3 shedding or ligand-receptor (Gal-9/Tim-3) interactions, which promote receptor internalization. Tim-3 expression appears dynamic and should be evaluated further kinetically in different subsets of cells at various differentiation states in disease models to better understand its role in negative regulation of innate and adaptive immune responses.

AUTHORSHIP

Z.Q.Y. and J.P.M. served as principal investigator and coprincipal investigator of this NIH-sponsored study, conceived of the idea, summarized the data, and wrote the manuscript. Y.Z. designed and performed the experiments and analyzed the data. C.J.M. performed silencing experiments. J.M.W. and X.J.J. performed the experiments. X.Y.W. provided the technical and laboratory support.

DISCLOSURES

The authors declare no competing financial interests. The contents in this publication do not represent the views of the Department of Veterans Affairs or the United States government.