CD1d knockout (KO) mice on a C57BL/6 background and wild-type (WT) C57BL/6 mice were purchased from the Jackson Laboratory and housed in a specific pathogen-free barrier unit. Experiments were conducted at 8–12 weeks of age and gender matched. Animal handling and the experimental procedures were approved by Institutional Animal Care and Use Committee of Henry Ford Health System.

Injection of 5 mL of air into the subcutaneous tissue on the back of mice was followed by injection of an additional 3 mL of air on day 3 and day 5. On day 7, MSU crystal (3 mg in 1 mL) was injected into the air-pouch cavities, and cells were harvested with 2 mL PBS at 3, 6, or 12 h for flow cytometry and enzyme-linked immunosorbent assays (ELISA) analyses (Figure 1A).

Total spleen cells from WT mice were first stained with anti-mouse CD8 biotin and anti-mouse B200 biotin Abs, then CD8⁺ and B220⁺ cells were depleted with anti-biotin beads using auto MACS (Miltenyi Biotec). Negatively selected cells were then stained with anti-mouse TCR-β, anti-mouse CD4 Ab, and PBS57-CD1d tetramers. CD4⁺ iNKT cells, CD4⁻ iNKT cells, and CD4 T cells (>97% purity) were sorted by BD FACS Aria II.

BM cells from WT or CD1d KO mice were incubated in culture media (RPMI with 10% FBS and 30 ng/mL M-CSF) for 7 days to induce BMDMs (M0). BMDMs were pre-loaded with α-galactosylceramide (α-GalCer, 1 µg/mL) for 6 h before they were cocultured with sorted CD4⁺ iNKT cells, CD4⁻ iNKT cells, or CD4⁺ T cells, respectively, at a 2:1 ratio for 36 h, followed by stimulation with MSU crystal (50 µg/mL) for 4 h with or without iNKT cells. After MSU crystal stimulation, BMDMs were collected for flow cytometry analysis or qRT-PCR.

Single-cell suspensions were washed twice with staining buffer and incubated with Fc block (clone 2.4G2). The following conjugated mAbs were used for flow cytometry analysis: TCR-beta (H57-597), PBS57-CD1d tetramers, F4/80 (BM8), Ly6G (1A8), and TNF-α (MP6-XT22). Dead cells were first gated out by propidium iodide (PI) staining. All mAbs were purchased from BD biosciences or eBioscience. Data were analyzed using FlowJo software.

The peritoneal cells were harvested at 6 h after MSU crystal (3 mg in 0.5 mL) treatment through peritoneal cavity. The phagocytosis of MSU crystals was determined by analyzing the side scatter (SSC) change in the flow cytometry.

Total RNA was extracted from BMDMs using a Mammalian Total RNA Miniprep Kit (Sigma). Q-PCR data collected on the QuantStudio 7 were normalized to the β-actin gene in the corresponding sample. Primer sequences are listed in Table S1 in Supplementary Material.

Cytokines in the supernatant of air-pouch cavities were measured using the ELISA Ready-Set-Go kit (eBioscience).

Statistical analysis was performed with Prism 7.0 (GraphPad Software). The two-tailed Student’s t-test was used. Differences were considered statistically significant when P < 0.05.

The air-pouch inflammation model was used for the investigation of synovial inflammation-mediated by MSU. To determine whether iNKT cell recruitment is a general feature in MSU crystal-induced gouty inflammation, cellular components in the washing fluid of the air-pouch cavity were analyzed by flow cytometry at different time points post-MSU crystal challenge. The iNKT cell population was identified by CD1d-tetramer and anti-TCRβ staining (Figure 1B). To our surprise, iNKT cells were recruited into the air-pouch cavities starting at 3 h and reached the peak at 12 h post-MSU crystal challenge (Figure 1B). Thus, it is reasonable to speculate that the iNKT cells recruited to sites of inflammation may have a role in the regulation of gout development.

CD1d is required for iNKT cell selection in the thymus and CD1d KO mice fail to develop iNKT cells. As shown in Figure 2A, there is no iNKT cell migration into air-pouch cavities of CD1d KO mice upon MSU crystal stimulation. To understand the function of iNKT cells in MSU crystal-induced gouty inflammation, we examined inflammatory profiles in WT mice and iNKT-null CD1d KO mice using an air-pouch gouty model. Upon MSU crystal stimulation, the total inflammatory cell number in the washing fluid from air-pouch cavities was significantly increased in CD1d KO mice compared with WT mice at 6 h (P < 0.01) and 12 h (P < 0.01) post-MSU crystal challenge (Figure 2B). We next evaluated neutrophils and macrophages in the washing fluid of air-pouch cavities by flow cytometry (Figure 2C). Even though the frequencies of neutrophils were not significantly increased, the absolute number of neutrophils increased dramatically (P < 0.05 at 6 h, P < 0.01 at 12 h) in CD1d KO mice. Nevertheless, no significant changes in frequency or number of macrophages were observed (Figure 2D). Therefore, the mice without iNKT cells developed more severe MSU crystal-induced inflammation.

Next, the levels of IL-4 and IL-1β in the supernatants from air-pouch cavities were analyzed by ELISA. As expected, both cytokines were elevated at 3 h following MSU crystal treatment in WT mice, then quickly downregulated at 6 and 12 h (Figure 2E). Interestingly, the level of IL-1β was dramatically increased in CD1dKO mice compared with WT mice at 6 h, whereas IL-4 was significantly reduced in CD1dKO mice at 3 h post-MSU crystal challenge. Thus, iNKT cells may regulate cytokine production, reducing IL-1β but increasing IL-4 production by macrophages. Taken together, these results suggest that iNKT cell deficiency significantly enhances the inflammatory response to MSU crystal in vivo.

The uptake of MSU crystal by macrophages is a feature for recognition by NALP3 inflammasome, thereby triggering caspase-1 activation and IL-1β processing (7, 13). Recently, Liu-Bryan et al. found that CD14 mediated ingestion of MSU crystal was able to induce an inflammatory response, leading to IL-1β release, which, however, is independent of modulation of MSU crystal uptake (5, 14). To explore if the increased production of IL-1β in CD1d KO air-pouch cavities was due to the modulation of capacity of MSU crystal uptake in the deletion of iNKT cells, we next preformed macrophage MSU crystal uptake assay in vivo. WT and CD1d KO mice were challenged with MSU crystal peritoneally for 6 h. Subsequently, the macrophages (Ly6G⁻F4/80^hi) were further analyzed for their SSC change, whose increasing would reflect the uptake of MSU crystal. As shown in Figures 3A,B, MSU crystal phagocytosis were comparable between WT and CD1d KO mice, which indicated that MSU crystal-induced severe inflammation, including high production of IL-1β, observed in CD1d KO mice is not due to increased MSU crystal phagocytosis. Thus, CD1d deficiency does not affect phagocytosis. Next, to rule out the possibility that lack of iNKT cells may affect macrophage death during MSU crystals induced gouty inflammation, we did a cell viability assay. As shown in Figure 3C, the frequency of PI⁺ dead macrophages was comparable between WT and CD1d KO mice. Thus, lack of iNKT cells unlikely affect MSU crystal induced macrophage death.

Since CD1d is also expressed in APCs, including macrophages, to exclude the possibility that the exacerbation of MSU crystal-induced gouty inflammation in CD1d KO mice was due to the lack of CD1d in the macrophages, we examined the TNF-α production by CD1dKO BMDMs using flow cytometry. As shown in Figure S1 in Supplementary Material, the frequencies of TNF-α secreting BMDMs from WT and CD1d KO mice were comparable post-MSU crystal stimulation. Thus, CD1d deficiency in macrophages may not affect their cytokine production in response to MSU crystal. Therefore, deficiency of iNKT cells in CD1d KO mice likely contributes to enhanced gout development.

Given that iNKT cells present quite early in sites of inflammation and may interact with CD1d⁺ APCs, macrophages in particular, we raised the hypothesis that the cross talk between iNKT cells and macrophages may interrupt MSU crystal-induced inflammation. There are at least two subsets of iNKT cells, such as CD4⁺ and CD4⁻ iNKT cells, which seem to possess different functions in regulating disease development (15). To investigate the contribution of iNKT cells to the cellular events at the site of gouty inflammation, we performed the in vitro iNKT cell–BMDM coculture experiment. BMDMs from CD1d KO or WT mice were cocultured with either sorted CD4⁺ or CD4⁻ splenic iNKT cells, and treated with α-Galcer, an agent to activate iNKT cells (16). BMDMs were collected after MSU crystal stimulation, with or without the washing off of iNKT cells before MSU crystal stimulation (Figure 4A). As shown in Figure 4B, BMDMs from WT mice cocultured with CD4⁺ iNKT cells produced less TNF-α upon MSU crystal stimulation than macrophages not cocultured with CD4⁺ iNKT cells. Coculture with CD4⁻ iNKT cells did not dramatically change TNF-α production compared with the BMDMs alone. This result suggests that it was CD4⁺ iNKT cells rather than CD4⁻ iNKT cells that played a key role in controlling gouty inflammation. As expected, coculture of iNKT cells with the BMDMs from CD1d KO mice, which are unable to stimulate iNKT cells due to their lack of the CD1d molecule, showed that neither CD4⁻ nor CD4⁺ iNKT cells inhibited TNF-α production (Figure 4C), which suggests that the activation of iNKT cells is required to inhibit TNF-α production in BMDMs. Thus, CD4⁺ iNKT cells regulate gouty inflammation in a CD1d-dependent manner.

Previous studies have suggested that activated iNKT cells modulate macrophage differentiation from M1 to M2 macrophages (17–19). To test if iNKT cells can regulate gouty inflammation through macrophage polarization, we assessed macrophage polarization-related cell markers using qRT-PCR analysis. As shown in Figure 4D, coculture with CD4⁺ iNKT cells significantly upregulated M2-related gene expressions on BMDMs, including Clec7a (P < 0.01), Pdcd1Ig2 (P < 0.01), and IL-4 (P < 0.01), compared with that of BMDMs cultured alone or cocultured with CD4⁺ conventional T cells; whereas expression levels of M1 markers (IL-6 and MIP-1a) were unaffected. Collectively, activation of CD4⁺ iNKT cells promotes M2 polarization, which may contribute to iNKT cell-mediated protection from MSU crystal-induced gouty arthritis.