The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiaotong University School of Medicine (XHEC-2017-104-2). Written informed consent was obtained from the patients before inclusion. Simvastatin ointment at a concentration of 1% was prepared in the laboratory of our hospital pharmacy. Three pediatric patients were treated with simvastatin ointment on the right side of the body twice per day for 6 months and the corresponding placebo ointment without simvastatin on the left side of the body as a control. Efficacy was assessed by comparing clinical photographs taken before and after treatment. Skin samples were obtained from the three patients after treatment.

A human monocyte cell line (THP-1; ATCC, Manassas, VA, USA) was cultured in RPMI 1640 medium (Gibco, Grand Island, NY, USA). Differentiation of THP-1 monocytes into macrophages was stimulated by 24 h of incubation with 50 nM phorbol-12-myristate-13-acetate (Sigma-Aldrich, St. Louis, MO, USA), followed by 24 h of incubation in RPMI 1640 medium without phorbol-12-myristate-13-acetate. Confluent cell cultures were exposed to 20 μg/mL ox-LDL (Yeasen Biotech, Shanghai, China) for 8 h, with or without 2 μM simvastatin (Selleckchem, Houston, TX, USA).

The cells were transfected with adenoviral vectors harboring the desired gene (Ad-RNA) for 24 h. The cells were collected 48 h after transfection or treatment with appropriate reagents. The cells were transfected with lentiviral vectors harboring short hairpin RNA (shRNA), and stable knockdown cells were selected using puromycin (Merck, Kenilworth, NJ, USA). The transcripts of Ad-NLRP3 and 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase (HMGCR) were NM_004895 and NM_000859, respectively. The sequences of the CD36 shRNAs, GSDMD shRNAs and HMGCR shRNAs were 5′-AAGGAATCCCTGTGTATAGAT-3′, 5′-GATGAGGTGCCTCCACAACTT-3′ and 5′-GACAGAATCTACACTCTCA-3′, respectively.

Five-micrometer-thick sections were obtained from each paraffin block and stained with hematoxylin and eosin (H&E; Abcam, Cambridge, UK). The other sections were incubated with the ox-LDL (Abcam, ab14519), NLRP3 (Abcam, ab214185), ASC (Cell Signaling Technology, Danvers, MA, USA, 13833), caspase-1 (Abcam, ab62698), GSDMD (Sigma, WH0079792M1), IL-1β (Cell Signaling Technology, 12242), and IL-18 (Abcam, ab243091) antibodies, and then incubated with fluorescent secondary antibodies (Cell Signaling Technology).

Cytotoxicity was determined by measuring the release of lactate dehydrogenase (LDH) in the cell supernatants using an LDH cytotoxicity assay kit (Beyotime, Jiangsu, China) according to the manufacturer’s instructions.

Pyroptosis was assessed using an apoptosis and necrosis assay kit (Beyotime) according to the manufacturer’s instructions.

IL-1β and IL-18 levels in the cell supernatants were measured using enzyme-linked immunosorbent assay (ELISA) kits (Dakewe, Shenzhen, China) according to the manufacturer’s instructions.

The cells were incubated with the GSDMD-N antibody (Cell Signaling Technology, 36425) and then with the fluorescent secondary antibody (Cell Signaling Technology). Images were captured using a confocal microscope (Olympus, Tokyo, Japan).

Cellular proteins were electrophoresed and transferred onto polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA), and then sequentially detected using primary antibodies, secondary antibodies, and an enhanced chemiluminescence substrate (Millipore). Antibodies against NLRP3 (15101), ASC (13833), caspase-1 (3866), cleaved caspase-1 (4199), GSDMD (93709), GSDMD-N (36425), IL-1β (12703), cleaved IL-1β (83186), and β-actin (4970) were purchased from Cell Signaling Technology. HMGCR (ab174830) antibodies were obtained from Abcam.

Cholesterol content in the macrophages was assessed using a cholesterol quantitation assay (Abcam) according to the manufacturer’s instructions.

The results are expressed as the mean ± standard error. Comparisons between groups were performed using the independent samples t-test. Statistical significance was set at p < 0.05. All experiments were performed independently, with at least three replicates.

All three patients were infants or children presenting yellowish nodules that multiplied throughout the body ( Figure 1A ). Through H&E staining, foam cells and Touton giant cells were observed in all three patients ( Figure 1B ). Immunohistochemical staining (CD68- and CD163-positive as well as S100- and Langerin-negative staining) confirmed the DNPX diagnosis ( Figure 1C ) (2). Blood lipid levels in all three patients were within the normal range and no systemic disease was found. All patients’ characteristics at baseline are presented in Supplementary Table 1 . The severity of the lesions in the three patients significantly improved with 6-month topical simvastatin treatment ( Figure 1D ). No severe adverse events were observed except for slight erythema, which spontaneously alleviated after 3–5 days of simvastatin treatment. H&E staining also revealed fewer foam cells and inflammatory cells after simvastatin treatment. Healthy controls refer to normal peer children having skin without the related disease ( Figure 1D ). To study the mechanism underlying the effects of simvastatin, we performed immunofluorescence staining on the lesion samples. The levels of HMGCR, NLRP3, ASC, caspase-1, GSDMD, IL-1β, and IL-18 in patients with DNPX were higher than those in healthy patients and was attenuated after simvastatin treatment than those on the placebo side at the end of treatment ( Figure 1F ).

The pathogenesis of DNPX is thought to be associated with the abnormal deposition of ox-LDL (6). Therefore, we performed immunofluorescence staining of ox-LDL. Our results showed a significant amount of ox-LDL deposition in the cytoplasm of the upper dermis, such as Touton giant cells. In contrast, a negligible amount was observed in the normal lower dermis and subcutaneous tissue ( Figure 2A ). The results are consistent with those of HE staining. Therefore, we established an in vitro foam cell model for macrophage phagocytosis of Dil-ox-LDL ( Supplementary Figure 1A ). Specifically, we set up a concentration gradient to stimulate macrophage response to ox-LDL. Exposure to ox-LDL enhanced LDH release into the supernatant and increased macrophage proliferation; however, ox-LDL concentrations did not fully positively correlate with the levels of released LDH and negatively correlated with macrophage proliferation, which indicates that ox-LDL not only induces macrophage pyroptosis, but also promotes macrophage proliferation ( Supplementary Figures 1B, C ). A study had reported that ox-LDL causes macrophage proliferation (26), which is consistent with our results. Therefore, we were more concerned about the release of inflammatory cytokines than macrophage proliferation in pyroptosis. Based on the results, 20 µg/mL ox-LDL was used in the subsequent experiments. We also observed pyroptotic bubbles under the microscope ( Supplementary Figure 1D ). Confocal microscopy images showed that GSDMD-N was localized to the cell membrane ( Supplementary Figure 1E ). To further confirm pyroptotic activation, we performed Hoechst/propidium iodide (PI) staining; the number of PI-positive cells increased after ox-LDL treatment ( Supplementary Figure 1F ). Considering that some of the dead cells were discarded from the supernatants, we performed flow cytometry to further confirm the results ( Supplementary Figure 1G ). Our ELISA results showed elevated IL-1β and IL-18 levels in the cell supernatants ( Supplementary Figures 1H, I ). Western blotting showed that the classical pyroptotic pathway was activated, and NLRP3 and ASC expression increased after ox-LDL treatment ( Supplementary Figure 1J ). In addition, the levels of the active forms of these proteins, namely cleaved caspase-1, GSDMD-N, and cleaved IL-1β, significantly increased after ox-LDL exposure ( Supplementary Figure 1J ). GSDMD-N immunofluorescence staining further confirmed this result ( Supplementary Figure 1K ).

To further confirm the effect of cholesterol efflux, we knocked down CD36, a main phagocytic receptor of ox-LDL, to inhibit macrophage phagocytosis, and applied T0901317, an activator of cholesterol excretion receptor, to promote cholesterol excretion. Our results revealed that total cholesterol levels significantly decreased after CD36 knockdown or application of T0901317 ( Supplementary Figure 2A ). In addition, CD36 knockdown or T0901317 application attenuated LDH release into the supernatant, decreased the number of PI-positive cells, and reduced the levels of inflammatory factors such as IL-1β and IL-18 in the cell supernatants ( Supplementary Figures 2B–E ). Western blotting revealed that CD36 knockdown or T0901317 application inhibited the expression of NLRP3 and ASC and prevented the cleavage of caspase-1, GSDMD, and IL-1β ( Supplementary Figure 2F ). These results demonstrated that ox-LDL activated pyroptosis in the macrophages.

To confirm the role of simvastatin in foam cell pyroptosis, we added 2 μM simvastatin to a medium containing ox-LDL. Our results suggested that simvastatin inhibited ox-LDL-induced LDH release ( Figure 2B ) and reduced the number of PI-positive cells ( Figures 2C, D ). Microscopic observation also suggested that the number of pyroptotic bubbles decreased in the group treated with simvastatin ( Figure 2E ). ELISA showed that simvastatin reduced IL-1β and IL-18 levels in the cell supernatants ( Figures 2F, G ). Western blotting revealed that NLRP3, ASC, cleaved caspase-1, GSDMD-N, and cleaved IL-1β levels were significantly decreased after simvastatin treatment ( Figure 2H ). GSDMD-N immunofluorescence staining further verified this result ( Figure 2I ). To verify the mechanism underlying the effect of simvastatin on ox-LDL uptake by macrophages, we used Dil-ox-LDL to quantify ox-LDL phagocytosis. However, no significant differences were observed in the quantity of ox-LDL after simvastatin treatment ( Figure 2J ).

We used 10 μM MCC950, an inhibitor of NLRP3, to verify the role of NLRP3 in foam cell pyroptosis. Our results suggested that MCC950 inhibited LDH release, decreased the number of PI-positive cells, prevented pyroptotic bubble formation, and attenuated IL-1β and IL-18 secretion ( Supplementary Figures 3A–F ). In addition, western blotting revealed that MCC950 inhibited the expression of NLRP3 and ASC and prevented the cleavage of caspase-1, GSDMD, and IL-1β ( Supplementary Figure 3G ). These results were similar to those for simvastatin.

To confirm the involvement of the classical pyroptotic pathway in the effects of simvastatin on foam cells, we first established Ad-NLRP3 macrophages ( Figure 3A ). Our results showed that NLRP3 overexpression enhanced LDH release, which was attenuated by simvastatin in foam cells ( Figure 3B ). The effects of simvastatin on the number of PI-positive cells and pyroptotic bubbles and inflammatory cytokine (IL-1β and IL-18) release were reversed after NLRP3 overexpression ( Figures 3C–G ). Correspondingly, western blotting revealed increased levels of NLRP3, ASC, cleaved caspase-1, GSDMD-N, and cleaved IL-1β ( Figures 3H, I ).

To further confirm pyroptotic involvement, we knocked down GSDMD, a pyroptotic performer protein ( Supplementary Figure 4A ). We found that enhanced LDH release induced by ox-LDL was abrogated by GSDMD knockdown ( Supplementary Figure 4B ). Moreover, the number of PI-positive cells, pyroptotic bubbles, and IL-1β and IL-18 levels in the cell supernatants were not elevated after GSDMD knockdown ( Supplementary Figures 4C–G ). However, NLRP3 and ASC expression still increased, and caspase-1 and IL-1β were activated by ox-LDL even after GSDMD knockdown ( Supplementary Figure 4H ).

To further study the effects of simvastatin on pyroptosis, three types of statins were used to treat the foam cells. No significant differences were observed among the effects of simvastatin, pravastatin, and lovastatin on foam cell pyroptosis ( Supplementary Figures 5A–G ). Statins inhibit HMGCR, a rate-limiting enzyme in cholesterol synthesis (27). In this study, western blotting showed that HMGCR expression was enhanced after ox-LDL treatment, consistent with previous study results (28), and was attenuated after simvastatin treatment ( Figure 2H ). HMGCR activity detection showed that HMGCR activity was enhanced after ox-LDL treatment and attenuated after simvastatin treatment ( Figure 4A ). Therefore, we overexpressed HMGCR, the known target of statins, to verify the target of simvastatin ( Figure 4B ). Our results showed that HMGCR overexpression reversed the effect of simvastatin ( Figures 4C–J ). Through HMGCR knockdown, we found that ox-LDL-induced pyroptosis was attenuated ( Figures 5A–I ).

Immunoprecipitation of either NLRP3 or HMGCR led to the recovery of both proteins, supporting the existence of an interaction between NLRP3 and HMGCR ( Figures 6A, B ). Additionally, simvastatin inhibited the interaction between NLRP3 and HMGCR ( Figures 6C, D ). NLRP3 and HMGCR colocalization in DNPX lesions was determined ( Figure 1F ). Our results indicated that simvastatin inhibited inflammation by causing an interaction between NLRP3 and HMGCR. A schematic model of the effects of simvastatin is shown in Figure 6E .