Soluble Uric Acid Activates the NLRP3 Inflammasome

Abstract

Uric acid is a damage-associated molecular pattern (DAMP), released from ischemic tissues and dying cells which, when crystalized, is able to activate the NLRP3 inflammasome. Soluble uric acid (sUA) is found in high concentrations in the serum of great apes, and even higher in some diseases, before the appearance of crystals. In the present study, we sought to investigate whether uric acid, in the soluble form, could also activate the NLRP3 inflammasome and induce the production of IL-1β. We monitored ROS, mitochondrial area and respiratory parameters from macrophages following sUA stimulus. We observed that sUA is released in a hypoxic environment and is able to induce IL-1β release. This process is followed by production of mitochondrial ROS, ASC speck formation and caspase-1 activation. Nlrp3^-/- macrophages presented a protected redox state, increased maximum and reserve oxygen consumption ratio (OCR) and higher VDAC protein levels when compared to WT and Myd88^-/- cells. Using a disease model characterized by increased sUA levels, we observed a correlation between sUA, inflammasome activation and fibrosis. These findings suggest sUA activates the NLRP3 inflammasome. We propose that future therapeutic strategies for renal fibrosis should include strategies that block sUA or inhibit its recognition by phagocytes.

Figure 1. Soluble uric acid stimulates IL-1β production in a NLRP3- and MyD88-dependent way.

Bone marrow-derived macrophages were stimulated with sUA (180 μΜ) and LPS (10 ηg/mL). Non-stimulated macrophages were used as controls and uricase (1 U/mL) and nigericin (10 μΜ) were used in some experiments. (a) il-1β mRNA was analyzed at different time points. (b) Western blotting of IL-1β and caspase-1 of cell supernatants after 24 hours stimulus. (c) Elisa of IL-1β of supernatants of WT, NLRP3^−/− and MyD88^−/− macrophages after different stimuli for 24 h. (d) Elisa of IL-1β was analyzed in supernatants of WT, Itgax^CreNLRP3^flox and LyZ^CreNLRP3^flox macrophages after sUA+LPS stimulation for 24 h. (e) WT, NLRP3^−/− and MyD88^−/− macrophages were transfected with a luciferase-based inflammasome and protease activity reporter for proteolytic activity of caspase-1 (iGLuc). Luciferase activity was assessed 96 h after the indicated stimulus in the cellular supernatants. (f) Release of lactate dehydrogenase (LDH) into supernatants was measured at 24 hours and shown as percentage of maximum at each stimulus. (g) Immortalized ASC-mCerulean macrophages were cultivated for 24 hours with indicated stimulus and ASC specks are represented as number of specks per ten microscopy fields. (h) Representative confocal photomicrographs of ASC-mCerulean macrophages showing ASC specks (green) produced after the activation of inflammasomes. The plasmatic membrane is indicated in red, the nucleus in blue and crystals are shown by white arrows. In (a), qPCR data were normalized to HPRT expression, and the mean of the control condition was considered 1. In (b), the blots are cropped in order to improve the clarity and conciseness of the membrane presentation. In (e), r.l.u. indicates relative light units. In (h), the white bars represent 50 μm. Data are representative of two independent experiments. n = 2 to 9. *p < 0.05; **p < 0.01; and ***p < 0.001. Nig = Nigericin; UC = uricase.

Figure 2. Soluble uric acid triggers the inflammasome and is responsible for renal damage.

WT, MyD88^−/−, NLRP3^−/−, Casp1^−/− and IL1R^−/− animals were submitted to ureteral obstruction and renal damage was investigated seven days later. (a) Quantitative analysis of the hypoxic area, represented by the proportion of the pimonidazole staining area compared to the total area of the field and (b) representative photomicrographs of pimonidazole uptake. (c) XDH mRNA, (d) sUA and (e) inflammasome-related genes in the kidney tissue of WT mice submitted to UUO and sham ones. (f) Proteinuria in the urine of obstructed pelvises normalized to that in the urine of the animals before surgery. (g) Representative photomicrographs and (h) quantification of collagen deposition, as analyzed by Sirius red staining, in the kidneys of WT, Casp1^−/−, IL1R^−/−, NLRP3^−/− and MyD88^−/− mice. In (a and f), the bars represent 50 μm in all photomicrographs. Sirius red staining and pimonidazole deposition are represented by the percentage of stained area out of the total area in the field. qPCR data were normalized to HPRT expression, and the mean expression in sham mice was considered 1. n = 3 to 13 animals per group in each experiment. *p < 0.05; **p < 0.01; and ***p < 0.001.

Figure 3. Soluble uric acid is responsible for damage in obstructive nephropathy.

Mice were either treated with allopurinol or injected with sUA following ureteral obstruction. (a–f) Allopurinol (Alop) treatment in WT mice. (a) Soluble tissue uric acid levels normalized by tissue protein. (b) Proteinuria in the urine of obstructed pelvises, normalized to that in the urine of the animals before surgery. (c) Type 1 collagen mRNA levels and (d) hydroxyproline levels normalized by tissue protein. (e) Inflammasome-related genes and (f) Elisa of IL-1β in obstructed kidneys. (g–i) SUA injection into WT, NLRP3^−/− and MyD88^−/− mice. (g) Proteinuria in the urine of obstructed pelvises, normalized to that in the urine of the animals before surgery. (h) Type 1 collagen mRNA levels and (i) quantification of collagen deposition, as analyzed by Sirius red staining. Sirius red staining deposition is represented by the percentage of stained area out of the total area in the field. qPCR data were normalized to HPRT expression, and the mean expression in sham mice was considered 1. n = 3 to 15 animals per group in each experiment. *p < 0.05.

Figure 4. Soluble uric acid triggers the NLRP3 inflammasome and mitochondrial ROS production.

WT, NLRP3^−/− and MyD88^−/− bone marrow-derived macrophages were stimulated with sUA in the presence of LPS (sUA+LPS) and non-stimulated macrophages were used as controls. (a) Representative flow cytometry graphs and (b) quantification of MitoSox staining in cells under SUA+LPS stimulus or control conditions for 24 h. (c) Relative reduced over oxidized glutathione (GSH/GSSG), normalized to total glutathione in WT macrophages under SUA+LPS stimulus or control ones for 24 h. (d) Representative flow cytometry graphs and (e) quantification of MitoSox staining in WT macrophages in the presence or absence of N-acetyl-l-cysteine (NAC) under SUA+LPS stimulus for 6 h. (f) Elisa of IL-1β in supernatants of macrophages analyzed as in (d) and (e) after 24 h of stimuli. (g) Transmission electronic photomicrographs of macrophages under different stimuli. Arrows point to mitochondria. The sUA+LPS representative figures are increased by 5000X and the other two figures are increased by 7000X. (h) Quantification of mitochondrial area seen in (g). Data are representative of two independent experiments. n = 2 to 11 *p < 0.05; **p < 0.01; and ***p < 0.001. UC = uricase.

Figure 5. Soluble uric acid promotes mitochondrial modifications in the absence of NLRP3.

Bioenergetic profiles of bone marrow-derived macrophages in the presence or absence of sUA+LPS. (a) Cells (60,000 per well) were treated with respiratory inhibitors and uncoupler at the following concentrations: oligomycin (1 μgmL), CCCP (5 μΜ) and antimycin A (10 μg/mL) plus rotenone (1 μΜ). (a) Representative oxygen consumption rates (OCR) from WT, NLRP3^−/− and MyD88^−/− macrophages. Gray lines represent sUA+LPS stimulated cells and black lines represent control ones. (b, upper graph) Basal OCR (OCR before addition of inhibitors), (b, middle graph) ATP production-dependent OCR (basal OCR minus oligomycin-insensitive-OCR), (b, lower graph) H⁺ leak (oligomycin-insensitive-OCR), (c, upper graph) maximal OCR (highest OCR after CCCP addition), (c, middle graph) reserve respiratory capacity (maximal OCR minus basal OCR), and (c, lower graph) non-mitochondrial respiration (OCR values in the presence of antimycin A plus rotenone) calculated from experiments such as those depicted in Panel a–c. (d) mRNA analysis of mitochondrial membrane component-related genes in different macrophages under sUA+LPS stimulus and non-stimulated conditions. (e) VDAC mRNA expression and (f) western blotting of VDAC. In (a–c), data are representative of three independent experiments and n = 3 to 10. In (d), n = 10 in one experiment. In (e–f), data are representative of two independent experiments and n = 2 to 4. In (f), the blots are cropped in order to improve the clarity and conciseness of the membrane presentation. *p < 0.05; **p < 0.01; and ***p < 0.001.