Histone acetyltransferase P300 deficiency promotes ferroptosis of vascular smooth muscle cells by activating the HIF-1α/HMOX1 axis

Abstract

Background: E1A-associated 300-kDa protein (P300), an endogenous histone acetyltransferase, contributes to modifications of the chromatin landscape of genes involved in multiple cardiovascular diseases. Ferroptosis of vascular smooth muscle cells (VSMCs) is a novel pathological mechanism of aortic dissection. However, whether P300 regulates VSMC ferroptosis remains unknown.

Methods: Cystine deprivation (CD) and imidazole ketone erastin (IKE) were used to induce VSMC ferroptosis. Two different knockdown plasmids targeting P300 and A-485 (a specific inhibitor of P300) were used to investigate the function of P300 in the ferroptosis of human aortic smooth muscle cells (HASMCs). Cell counting kit-8, lactate dehydrogenase and flow cytometry with propidium iodide staining were performed to assess the cell viability and death under the treatment of CD and IKE. BODIPY-C11 assay, immunofluorescence staining of 4-hydroxynonenal and malondialdehyde assay were conducted to detect the level of lipid peroxidation. Furthermore, co-immunoprecipitation was utilized to explore the interaction between P300 and HIF-1α, HIF-1α and P53.

Results: Compared with normal control, the protein level of P300 was significantly decreased in HASMCs treated with CD and IKE, which was largely nullified by the ferroptosis inhibitor ferrostatin-1 but not by the autophagy inhibitor or apoptosis inhibitor. Knockdown of P300 by short-hairpin RNA or inhibition of P300 activity by A-485 promoted CD- and IKE-induced HASMC ferroptosis, as evidenced by a reduction in cell viability and aggravation of lipid peroxidation of HASMCs. Furthermore, we found that hypoxia-inducible factor-1α (HIF-1α)/heme oxygenase 1 (HMOX1) pathway was responsible for the impacts of P300 on ferroptosis of HASMCs. The results of co-immunoprecipitation demonstrated that P300 and P53 competitively bound HIF-1α to regulate the expression of HMOX1. Under normal conditions, P300 interacted with HIF-1α to inhibit HMOX1 expression, while reduced expression of P300 induced by ferroptosis inducers would favor HIF-1α binding to P53 to trigger HMOX1 overexpression. Furthermore, the aggravated effects of P300 knockdown on HASMC ferroptosis were largely nullified by HIF-1α knockdown or the HIF-1α inhibitor BAY87-2243.

Conclusion: Thus, our results revealed that P300 deficiency or inactivation facilitated CD- and IKE-induced VSMC ferroptosis by activating the HIF-1α/HMOX1 axis, which may contribute to the development of diseases related to VSMC ferroptosis.

Keywords: Ferroptosis; HIF-1α; HMOX1; P300; P53; Vascular smooth muscle cells.

Fig. 1

P300 was downregulated during ferroptosis of HASMCs. (A) Relative cell viability of HASMCs measured with a CCK8 kit after cystine deprivation (CD) treatment for 0, 4, 8, 12, 16, and 24 h (n = 6 per group). (B) Relative cell viability of HASMCs after treatment with different concentrations of imidazole ketone-erastin (IKE) (0, 0.5, 1, 2.5, 5, 10 µM) (n = 6 per group). C-F. The level of lipid ROS (oxidized BODIPY-C11/non-oxidized BODIPY-C11 ratio) detected by using a BODIPY-C11 kit in HASMCs after treatment with CD (C-D) and IKE (E-F) (n = 3 per group). G. Representative images of immunofluorescence staining with 4-HNE in HASMCs treated with DMSO and Fer-1 after CD and IKE stimulation for 16 h. H. Quantitative analysis of immunofluorescence staining of 4-HNE (n = 4 per group). I. The level of MDA measured by using an MDA assay kit in HASMCs treated with DMSO and Fer-1 after CD and IKE stimulation for 16 h. (n = 5 per group). J-M. P300 protein levels in HAMSCs after CD (J-K), IKE (L-M) stimulation for 16 h and treatment with Fer-1, 3-MA and Emricasan were measured by western blot (n = 4 per group). J and L. Representative western blot; K and M. Quantitative results. β-actin served as a loading control. Values are means ± SD; ***p < 0.001, **p < 0.01, NS, no significant

Fig. 2

Knockdown of P300 reduced cell viability and increased cell death of HASMCs with CD- and IKE-induced ferroptosis. A-B. The protein level of P300 were detected by western blot in total and nuclear protein extraction of HASMCs infected with lenti-shRNA or lenti-shP300. (n = 4 per group). (A) Total protein extraction western blot; (B) Nuclear protein extraction. C. Representative images showing cell death induced by CD, IKE, alone or in combination with Fer-1 after HASMCs were infected with lenti-shRNA, lenti-shP300-1 and lenti-shP300-2. D-E. Relative viability of the indicated lentivirus-infected HASMCs evaluated by CCK8 after treatment with CD (D), IKE (E), alone or in combination with Fer-1 (n = 6 per group). F. Relative cell death of HASMCs infected with the indicated lentivirus evaluated by LDH after the treatment described above (n = 5 per group). G-H. Percentage of PI positive indicated lentivirus-infected HASMCs examined by flow cytometry with propidium iodide (PI) staining after treatment as described above (n = 4 per group). Values are means ± SD; ***p < 0.001, NS, no significant

Fig. 3

Downregulation of P300 aggravated the accumulation of lipid peroxidation during HASMC ferroptosis. A-D. The level of lipid ROS (oxidized BODIPY-C11/non-oxidized BODIPY-C11 fluorescence ratio) examined by using BODIPY-C11 kit after treatment with CD (A-B) and IKE (C-D), alone or in combination with Fer-1, in HASMCs infected with lenti-shRNA, lenti-shP300-1 and lenti-shP300-2 (n = 4 per group). E and G. Representative images of immunofluorescence staining of 4-HNE in the indicated lentivirus-infected HASMCs treated with CD (E), IKE (G), alone or in combination with Fer-1. F and H. Quantitative analysis of 4-HNE (n = 4 per group). I-J. The level of MDA was evaluated by an MDA assay kit in indicated lentivirus-infected HASMCs after CD (I), IKE (J), or in combination with Fer-1 stimulation for the indicated time (n = 4 per group). Values are means ± SD; ***p < 0.001, **p < 0.01, *p < 0.05, NS, no significant

Fig. 4

A-485 reduced cell viability and increased cell death of HASMCs treated with CD and IKE. A. A CCK8 assay was performed to show the relative cell viability detected in HASMCs after treatment with different concentrations of A-485 (0, 0.5, 1, 2.5, 5, 10, 15 and 20 µM) (n = 6 per group). B-C. H3K27ac, H3K18ac, and H2BK5ac protein levels in HASMCs treated with A-485 at different concentrations of 0, 0.5, 1, 2.5, 5 and 10 µM were measured by western blot (n = 4 per group). β-actin served as a loading control. D. Representative images showing cell death after treatment with DMSO and A-485 under the ferroptosis models induced by CD and IKE, alone or in combination with Fer-1. E-F. The CCK8 assay showed the relative viability of HASMCs treated with DMSO and A-485 after CD (E), IKE (F), alone or in combination with Fer-1 stimulation for the indicated time (E: n = 4 per group, F: n = 5 per group). G. An LDH assay was performed to examine the cell death of HASMCs after treating as described above (n = 5 per group). H-I. Flow cytometry with PI staining showed the percentage of PI positive cells after treatment with DMSO and A-485 under the ferroptosis models induced by CD (H), IKE (I), alone or in combination with Fer-1 (n = 4 per group). Values are means ± SD; ***p < 0.001, NS, no significant

Fig. 5

A-485 aggravated the accumulation of lipid peroxidation during HASMC ferroptosis. A-D. The ratio of oxidized BODIPY-C11/non-oxidized BODIPY-C11 fluorescence detected by the BODIPY-C11 kit after treatment with DMSO and A-485 under the ferroptosis models induced by CD (A-B), IKE (C-D), alone or in combination with Fer-1 (n = 3 per group). E and G. Representative images of 4-HNE immunofluorescence staining in HASMCs treated with DMSO and A-485 after CD (E), IKE (G), alone or in combination with Fer-1 stimulation for the indicated time. F and H. Quantitative analysis of 4-HNE immunofluorescence staining (n = 4 per group). I-J. The level of MDA in HASMCs treated as described above was examined by an MDA assay kit (n = 3 per group). Values are means ± SD; ***p < 0.001, **p < 0.01, *p < 0.05, NS, no significant

Fig. 6

P300 and P53 competitively bind HIF-1α to regulate HMOX1 expression during ferroptosis of HASMCs. A-B. Western blot analysis and quantification results showing SLC7A11, GPX4, and FSP1 protein levels after CD (A), and IKE (B) stimulation in HASMCs infected with lenti-shRNA, lenti-shP300-1 and lenti-shP300-2 (A-B), β-actin served as a loading control (n = 4 per group). C-D. Western blot analysis and quantification results showing HMOX1 protein levels in HASMCs infected with lenti-shRNA, lenti-shP300-1 and lenti-shP300-2 after CD and IKE stimulation. β-actin served as a loading control (n = 4 per group). E. Co-immunoprecipitation results showed that endogenous P300 interacted with exogeneous HIF-1α in HASMCs after treatment with MG132. F. The interaction between endogenous HIF-1α and exogenous P300 in 293T cells after treatment with MG132 was verified by Co-immunoprecipitation. G. Co-immunoprecipitation showed the interaction between endogenous HIF-1α and endogenous P300 and P53 after treatment with CD and MG132 in nucleus of HASMCs. Values are means ± SD; ***p < 0.001, NS, no significant

Fig. 7

HIF-1α downregulation abolished the pro-ferroptotic effects of P300 deficiency in HASMCs. P300 knockdown HASMCs were infected with lenti-shRNA and lenti-shHIF-1α, and then these HASMCs were used for subsequent experiments. A-B. Relative cell viability was evaluated by CCK8 kit in HASMCs after treatment with CD (A) and IKE (B) (n = 5 per group). C-F. The level of lipid ROS (oxidized BODIPY-C11/non-oxidized BODIPY-C11 fluorescence ratio) examined by the BODIPY-C11 kit in HASMCs treated with CD (C-D) and IKE (E-F) (n = 4 per group). G-H. An MDA assay was used to detect the MDA levels of HASMCs after treatment with CD (G) and IKE (H) (n = 4 per group). I-L. Representative images of immunofluorescence staining and quantitative analysis of 4-HNE in HASMCs among the indicated groups (n = 4 per group). Values are means ± SD; ***p < 0.001, **p < 0.01, *p < 0.05, NS, no significant

Fig. 8

HIF-1α inhibition eliminated the pro-ferroptotic effects of P300 deficiency in HASMCs. A-B. Relative cell viability of HASMCs infected with lenti-shRNA, lenti-shP300-1, and lenti-shP300-2 by using a CCK8 kit after treatment with CD (A) and IKE (B) in the presence or absence of BAY87-2243 (n = 4 per group). C-F. The oxidized BODIPY-C11/non-oxidized BODIPY-C11 fluorescence ratio was evaluated by BODIPY-C11 kit in the indicated lentivirus-infected HASMCs treated with DMSO and BAY87-2243 after CD (C-D) and IKE (E-F) stimulation for 16 h (n = 4 per group). G-H. An MDA assay was used to assess MDA levels in indicated lentivirus-infected HASMCs treated with DMSO and BAY87-2243 after CD (G) and IKE (H) stimulation for the indicated time (n = 4 per group). I-L. Representative images of immunofluorescence staining and quantitative analysis of 4-HNE in HASMCs among the indicated groups in the presence or absence of BAY87-2243 (n = 4 per group). Values are means ± SD; ***p < 0.001, **p < 0.01, *p < 0.05, NS, no significant

Fig. 9

Schematic summary. P300 deficiency and inhibition induced VSMC ferroptosis by activating the HIF-1α/HMOX1 axis. P300 and P53 competitively bound HIF-1α to regulate HMOX1 expression. Under normal conditions, P300 interacted with HIF-1α to inhibit HMOX1 expression. When treatment with ferroptosis inducers, the decrease of P300 contributed to the interaction between P53 and HIF-1α to activate HMOX1 transcription. The activation of HMOX1 decreased cell viability and aggravated lipid peroxidation and further triggered ferroptosis