Alternative autophagy dampens UVB-induced NLRP3 inflammasome activation in human keratinocytes

Abstract

Sunlight exposure results in an inflammatory reaction of the skin commonly known as sunburn, which increases skin cancer risk. In particular, the ultraviolet B (UVB) component of sunlight induces inflammasome activation in keratinocytes to instigate the cutaneous inflammatory responses. Here, we explore the intracellular machinery that maintains skin homeostasis by suppressing UVB-induced inflammasome activation in human keratinocytes. We found that pharmacological inhibition of autophagy promoted UVB-induced NLRP3 inflammasome activation. Unexpectedly, however, gene silencing of Atg5 or Atg7, which are critical for conventional autophagy, had no effect, whereas gene silencing of Beclin1, which is essential not only for conventional autophagy but also for Atg5/Atg7-independent alternative autophagy, promoted UVB-induced inflammasome activation, indicating an involvement of alternative autophagy. We found that damaged mitochondria were highly accumulated in UVB-irradiated keratinocytes when alternative autophagy was inhibited, and they appear to be recognized by NLRP3. Overall, our findings indicate that alternative autophagy, rather than conventional autophagy, suppresses UVB-induced NLRP3 inflammasome activation through the clearance of damaged mitochondria in human keratinocytes and illustrate a previously unknown involvement of alternative autophagy in inflammation. Alternative autophagy may be a new therapeutic target for sunburn and associated cutaneous disorders.

Keywords: alternative autophagy; inflammasome; keratinocyte; skin; sunburn.

Figure 1

Autophagy suppresses UVB-induced NLRP3 inflammasome activation in human keratinocytes.A, quantification of IL-1β production by ELISA in the supernatants of keratinocytes, which were incubated for 24 h in the absence or presence of autophagy inhibitor 3MA, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). B, flow cytometric analysis to detect autophagy using Cyto-ID. Cells were incubated for 24 h in the absence or presence of autophagy inhibitor 3MA, then irradiated or not irradiated with UVB, and further cultured for 6 h (n = 4 per group). C, quantification of IL-1β production by ELISA in the supernatants of keratinocytes incubated for 24 h in the absence or presence of the autophagy inducer rapamycin (Rapa), then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). D, representative example showing the expression level of NLRP3 evaluated by Western blotting in the cell lysates of keratinocytes, which were transfected with either control siRNA or NLRP3 siRNA. E, quantification of IL-1β production by ELISA in the supernatants of keratinocytes, which were transfected with either control siRNA or NLRP3 siRNA and then irradiated or not irradiated with UVB and further cultured for 48 h in the absence or presence of the 3MA (n = 5 per group). F, quantification of IL-1β production by ELISA in the supernatants of keratinocytes, which were transfected with either control siRNA or NLRP3 siRNA (#2) and then irradiated or not irradiated with UVB and further cultured for 48 h in the absence or presence of 3MA (n = 4 per group). G, quantification of IL-1β production by ELISA in the supernatants of keratinocytes incubated for 24 h in the absence or presence of 3MA or NLRP3 inhibitor MCC950, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). H, representative example showing the expression level of NLRP1 evaluated by Western blotting in the cell lysates of keratinocytes, which were transfected with either control siRNA or NLRP1 siRNA. I, quantification of IL-1β production by ELISA in the supernatants of keratinocytes, which were transfected with control siRNA or NLRP1 siRNA, incubated for 24 h in the absence or presence of 3MA, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). All data are expressed as the mean ± SD. ∗ p < 0.05, ∗∗ p < 0.01, and ∗∗∗ p < 0.001, ns: not significant. IL, interleukin; UVB, ultraviolet B.

Figure 2

Atg5/Atg7-independent alternative autophagy participates in the suppression of UVB-induced inflammasome activation in human keratinocytes.A, representative example showing the expression level of Atg5 (left) or Atg7 (right) evaluated by Western blotting in the cell lysates of keratinocytes, which were transfected with either control siRNA or Atg5 siRNA or Atg7 siRNA. B, representative example showing the expression level of LC3B determined by Western blotting in cell lysates of either Atg5- or Atg7-knockdown keratinocytes. C, quantification by ELISA of IL-1β in the supernatants of keratinocytes that were transfected with control siRNA or Atg5 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). D, quantification by ELISA of IL-1β in the supernatants of keratinocytes that were transfected with control siRNA or Atg7 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 3 per group). E, representative example showing the expression level of p-Ulk1⁷⁴⁶ evaluated by immunoprecipitation-Western blotting in the cell lysates of keratinocytes, which were irradiated or not irradiated with UVB and further cultured for 6 h (top). Relative intensity of p-Ulk1⁷⁴⁶ is shown (n = 4 per group) (bottom). F, representative immunocytochemical staining of GS28 (green) and p-Ulk1⁷⁴⁶ (red) in keratinocytes, which were irradiated or not irradiated with UVB and further cultured for 6 h (left). The percentage of p-Ulk1⁷⁴⁶–positive keratinocytes. Cells were counted and averaged across 15 randomly selected images in each experiment. n = 15 in each group (right). Repeated three independent experiments were performed. G, representative example showing the expression level of Beclin1 or Rab9 determined by Western blotting in the cell lysates of keratinocytes transfected with either control siRNA, Beclin1 siRNA, or Rab9 siRNA. H, quantification by ELISA of IL-1β in the supernatants of keratinocytes that were transfected with control siRNA or Beclin1 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 3 per group). I, quantification by ELISA of IL-1β in the supernatants of keratinocytes that were transfected with control siRNA or Rab9 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 48 h (n = 4 per group). Nuclei were stained with Hoechst (blue). All data are expressed as the mean ± SD. ∗ p < 0.05 and ∗∗∗ p < 0.001, ns: not significant. Scale bar represents 10 μm. IL, interleukin; UVB, ultraviolet B.

Figure 3

Accumulation of damaged mitochondria in UVB-irradiated human keratinocytes is promoted under inhibition of alternative autophagy.A, representative staining with mitoSOX (red), an indicator of mitochondrial ROS, in keratinocytes, which were incubated for 24 h in the absence or presence of 3MA, then irradiated or not irradiated with UVB, and further cultured for 18 h. B, the intensity of mitoSOX staining in keratinocytes. Cells were counted and averaged across 15 randomly selected images in each experiment. n = 15 in each group. Repeated three independent experiments were performed. C, representative staining of mitoSOX (red) in keratinocytes, which were transfected with control, Atg5, Atg7, or Beclin1 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 18 h. D, the intensity of mitoSOX staining in keratinocytes. Cells were counted and averaged across 20 randomly selected images in each experiment. n = 20 in each group. Repeated three independent experiments were performed. Nuclei were stained with Hoechst (blue). All data are expressed as the mean ± SD. ∗∗∗ p < 0.001. Scale bar represents 100 μm. UVB, ultraviolet B.

Figure 4

NLRP3 co-localizes with UVB-induced damaged mitochondria in human keratinocytes.A, representative PLA staining (red), which detects protein complexes between NLRP3 and TOM20, in keratinocytes, which were incubated for 24 h in the absence or presence of 3MA, then irradiated or not irradiated with UVB, and further cultured for 18 h. B, the intensity of PLA signals in keratinocytes. Cells were counted and averaged across 15 randomly selected images in each experiment. n = 15 in each group. Repeated three independent experiments were performed. C, representative PLA staining (red) in keratinocytes, which were transfected with control, Atg5, Atg7, or Beclin1 siRNA for 48 h, then irradiated or not irradiated with UVB, and further cultured for 18 h. D, the intensity of PLA signals in keratinocytes. Cells were counted and averaged across 20 randomly selected images in each experiment. n = 20 in each group. Repeated three independent experiments were performed. Nuclei were stained with Hoechst (blue). All data are expressed as the mean ± SD. ∗ p < 0.05 and ∗∗∗ p < 0.001. Scale bar represents 100 μm. UVB, ultraviolet B.

Supporting Figure S1

Supporting Figure S2

Supporting Figure S3

Supporting Figure S4

Supporting Figure S5