Targeting the smooth muscle cell Keap1-Nrf2-GSDMD-pyroptosis axis by cryptotanshinone prevents abdominal aortic aneurysm formation

Abstract

Rationale: Abdominal aortic aneurysm (AAA) is an inflammatory, fatal aortic disease that currently lacks any effective drugs. Cryptotanshinone (CTS) is a prominent and inexpensive bioactive substance derived from Salvia miltiorrhiza Bunge, a well-known medicinal herb for treating cardiovascular diseases through its potent anti-inflammatory properties. Nevertheless, the therapeutic effect of CTS on AAA formation remains unknown. Methods: To investigate the therapeutic effect of CTS in AAA, variety of experimental approaches were employed, majorly including AAA mouse model establishment, real-time polymerase chain reaction (PCR), RNA sequencing, western blot, co-immunoprecipitation, scanning/transmission electron microscopy (SEM/TEM), enzyme-linked immunosorbent assay (ELISA), seahorse analysis, immunohistochemistry, and confocal imaging. Results: In this study, we demonstrated that CTS suppressed the formation of AAA in apolipoprotein E knock-out (ApoE^-/-) mice infused with Ang II. A combination of network pharmacology and whole transcriptome sequencing analysis indicated that activation of the Keap1-Nrf2 pathway and regulation of programmed cell death in vascular smooth muscle cells (VSMCs) are closely linked to the anti-AAA effect of CTS. Mechanistically, CTS promoted the transcription of Nrf2 target genes, particularly Hmox-1, which prevented the activation of NLRP3 and GSDMD-initiated pyroptosis in VSMCs, thereby mitigating VSMC inflammation and maintaining the VSMC contractile phenotype. Subsequently, by utilizing molecular docking, together with the cellular thermal shift assay (CETSA) and isothermal titration calorimetry (ITC), a particular binding site was established between CTS and Keap1 at Arg415. To confirm the binding site, site-directed mutagenesis was performed, which intriguingly showed that the Arg415 mutation eliminated the binding between CTS and the Keap1-Nrf2 protein and abrogated the antioxidant and anti-pyroptosis effects of CTS. Furthermore, VSMC-specific Nrf2 knockdown in mice dramatically reversed the protective action of CTS in AAA and the inhibitory effect of CTS on VSMC pyroptosis. Conclusion: Naturally derived CTS exhibits promising efficacy as a treatment drug for AAA through its targeting of the Keap1-Nrf2-GSDMD-pyroptosis axis in VSMCs.

Figure 1

CTS inhibits Ang II-induced AAA formation in ApoE^-/- mice. (A) Schematic illustration of AAA establishment and CTS administration strategy. In the murine AAA model, 12-week-old male ApoE^-/- mice were randomly allocated to four groups and infused with either saline or Ang II (1000 ng/kg/min) for 28 consecutive days. On the day the AAA was started to induce, HP-β-CD (vehicle), low-dose CTS (CTS-L, 15 mg/kg), or high-dose CTS (CTS-H, 50 mg/kg) were intragastrically delivered daily, respectively. (B) Systolic blood pressure was measured at the day before Day 0 and Day 28 (n = 10-15). (C) Survival rate of AAA. (D) Incidence of AAA. (E) Representative images of aortas with AAA. (F) Maximal diameter of the abdominal aorta was measured in four groups: Sham (n = 15), Ang II (n = 10), CTS-L (n = 13), and CTS-H (n = 14). (G-H) Representative images of mouse aorta cross-sections stained with H&E and EVG. In EVG staining, the dark line represents elastin, and blue triangles indicate broken elastin. (I) Evaluation of elastin degradation grade. Elastin degradation was graded on a scale of 1-6: grade 1, no degradation; grade 2: elastin degradation less than< 20% of the entire area; grade 3: elastin degradation between 20 and 40%; grade 4: elastin degradation between 40 and 60%; grade 5: elastin degradation between 60 and 80%; and grade 6: elastin degradation ≥80% (n = 7). (J) Representative images and quantitative analyses of IHC staining for CD68, α-SMA, SM22α, and MMP2, 3 and 9 (n = 5). (K-L) Representative western blots and quantitative analyses of indicated protein expression in mouse aortas (n = 6). (M-N) In-situ zymography immunofluorescence images and quantitative analysis for assessing the MMP activities of the cross-sections from mouse aortas. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: no significant.

Figure 2

CTS orchestrates VSMC phenotypes, inflammation, ROS production, and mitochondrial function in an in vitro AAA model. (A-B) Representative western blots and quantitative analyses of the indicated protein expression in RAVSMCs. RAVSMCs were pre-treated with various concentrations of CTS for 3 h, followed by the addition of TNF-α (10 ng/mL) with an incubation of 24 h (n = 5). (C-D) Representative images and quantitative analysis of DHE staining in murine abdominal aorta (n = 5) (M: tunica media; A: tunica adventitia) (n = 5). (E-H) ROS levels were analyzed through photography (E-F) or flow cytometry (G-H). RAVSMCs were treated as aforementioned in A-B (n = 5). (I-L) SOD and MDA levels were measured from mouse serum Ang II-infused mice with or without CTS (low and high doses) treatment (I-J) or from RVASMCs treated as aforementioned in A-B (K-L) (n = 5). (M) LDH levels were measured from the cell culture supernatants of RVASMCs (n = 5). (N) Representative transmission electron microscope (TEM) images of mitochondria were shown in the RAVSMCs treated with TNF-α and CTS. EM, empty mitochondria; MV, mitochondrial vacuolization; WM, whorl-like inner membrane; DE, dense electron. (O) OCR profile of RAVSMCs. (P) Quantification of mitochondrial respiration function parameters from O (n = 6-8). (Q-R) The quantity of cytosolic DNA and copies of mitochondrial DNA (mtDNA) were measured by qRT-PCR (n = 5) in MAVSMCs. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: no significant.

Figure 3

Network pharmacology combined with RNA-seq analysis indicates that CTS activates the Keap1-Nrf2-HO-1 pathway and suppresses the cell death pathway. (A-B) 2,537 AAA-associated targets and 439 CTS-related targets were collected from free-access databases. Merging these networks revealed 192 common genes, potential targets for CTS in AAA prevention. Cytoscape analysis provided degree values for each target; those exceeding 1.5 times the mean were considered key targets in restraining AAA for CTS. Additionally, the 192 common targets were subjected to DO, KEGG, and GO analyses. (A) The process of screening potential and core targets of CTS in its action on AAA. (B) The DO enrichment analysis revealed that the 192 common genes were primarily associated with inflammatory response and cell death in the biological process ontology. In this analysis, the color and length of the bands were utilized to signify the P-value and the percentage of genes. (C) For RNA-seq analysis, RAVSMCs were pretreated with 10 μM CTS or DMSO for 3 h, followed by TNF-α treatment for another 3 h. The RNA was then collected and extracted for analysis. Comparing TNF-α treatment to co-treatment with CTS and TNF-α, the volcano plot indicated the magnitude and significance of the changes in gene expression in RAVSMCs. (D) The heatmap showed CTS-upregulated genes related to heme and iron metabolism, GSH production and regeneration, NADPH regeneration, the TXN-based antioxidant system, and ROS and xenobiotic detoxification in the RNA-seq analysis. (E-F) The KEGG and GO enrichment analyses revealed the effect of CTS on RAVSMCs at the molecular and cellular levels. All DO, KEGG, and GO analyses were performed using the OmicShare tools (https://www.omicshare.com/tools).

Figure 4

CTS activates the Keap1-Nrf2-HO-1 pathway in VSMCs to prevent AAA. (A) Representative western blots and quantitative analysis of indicated protein expression in mouse aortic tissues from Ang II-infused mice with or without CTS (low and high doses) treatment (n = 6). (B) Representative images and quantitative analyses of IHC staining for Nrf2 and HO-1 from the cross-sections of the mouse abdominal aorta (n = 5). (C) Representative western blots and quantitative analyses of the indicated protein expressions in RAVSMCs treated with different doses of CTS (2.5, 5 and 10 μM). TBHQ is an activator of the Nrf2 pathway used as a positive control (n = 5). (D) Representative western blot analysis and quantitative assessment of Nrf2 in RAVSMCs (n = 5). (E) Immunofluorescence staining showed the distribution and colocalization of Nrf2 (green) in RAVSMCs treated with TNF-α and CTS or DMSO. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: no significant.

Figure 5

CTS suppresses VSMC pyroptosis to prevent AAA. (A) Heatmap showing the gene expression of Gsdmd in the aortas of Ang II/Saline-infused mice from 7 to 28 days. The data are from a reanalysis of the reported microarray dataset (GSE17901). n = 5-7 (upper); Heatmap showing the gene expression of Gsdmd in RAVSMCs treated with TNF-α and CTS (n = 3-4) (lower). (B-E) Representative western blots and quantitative analysis of indicated protein expression in mouse aortic tissues from Ang II-infused mice with or without CTS treatment, as well as RAVSMCs treated with TNF-α and CTS (n = 5-6). (F-G) Representative images and quantitative analyses of IHC staining for NLRP3, Caspase 1, and GSDMD (n = 5). (H) Cytokines in mouse serum from the indicated groups were measured by ELISA (n = 6). (I) Representative phase-contrast imaging assays were performed. Arrowheads indicated balloon-like pyroptotic cells in the RAVSMCs treated with TNF-α and CTS. (J) Representative scanning electronic microscopy (SEM) images of pyroptosis were shown in the RAVSMCs treated with TNF-α and CTS. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ns: no significant.

Figure 6

CTS directly binds to the Arg415 residue of Keap1 to activate Nrf2. (A) The CETSA was performed using intact RAVSMCs with 10 μM of CTS. The stability of Keap1 protein at 37-77 °C was measured by western blot. (B) The ITDRF-CETSA assay was performed using intact RAVSMCs in the presence of different CTS doses (0-102.4 μM) over 4 h. The stability of Keap1 protein at 55 °C was measured by western blot. (C) The 2D image illustrates the interaction of CTS and Keap1 at specific amino acids. Hydrogen bonds are denoted by green and blue dotted lines. (D) The CETSA was performed using purified Keap1 WT/R415K proteins in the presence of CTS (10 μM). The stability of Keap1 proteins under 37-77 °C was measured by western blot. (E-F) Isothermal titration plot of 2 μM CTS (in sample cell) with 40 μM Keap1 WT or Keap1 R415K proteins (in syringe). The inset provided a graphical representation of the thermodynamic parameters, including KD and ΔH. The solid line represents the optimal nonlinear least-squares fit to a single binding site model. (G) Co-immunoprecipitation (Co-IP) of Keap1 in RAVSMCs treated with TNF-α (10 ng/mL) and CTS (10 μM) or DMSO to detect the protein level of Nrf2. (H) Co-IP of Nrf2 in RAVSMCs treated with TNF-α (10 ng/mL) and CTS (10 μM) or DMSO was performed to detect the ubiquitination level of Nrf2. WCL: whole cell lysates. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ns: no significance.

Figure 7

CTS binds to Keap1 at Arg415 to reduce the generation of ROS via activating Nrf2 and inhibiting pyroptosis in TNF-α treated VSMCs. (A-B) ROS levels were analyzed through photography (A) or flow cytometry (B). RAVSMCs were transfected with Keap1-WT and Keap1-R415K plasmids, respectively. After 36 h, the transfected cells were pretreated with CTS for 3 h followed by the addition of TNF-α (10 ng/mL) with an incubation of 24 h. Cells were stained and collected for photography or flow cytometry analysis. (C-D) Representative western blot analysis and quantitative assessment of indicated proteins in RAVSMCs (n = 3). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 8

Nrf2 inhibition reverses the protective effect of CTS in vitro. (A) Representative western blots and quantitative analysis of indicated protein expression in RAVSMCs treated with TNF-α, CTS, and ML385. ML385 was used as an inhibitor of Nrf2 (n = 5). (B) qRT-PCR analysis of the mRNA expression of MMPs and cytokines in RAVSMCs treated with TNF-α, CTS, and ML385 (n = 5). (C) Representative western blots and quantitative analysis of indicated protein expression in RAVSMCs treated with TNF-α, CTS, and ML385 (n = 5). (D-G) ROS levels were analyzed through flow cytometry (D-E) or photography (F-G) in RAVSMCs treated with TNF-α, CTS, and ML385 (n = 5). (H) Representative western blots and quantitative analysis of indicated protein expression in RAVSMCs treated with TNF-α, CTS, and ML385 (n = 5). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ns: no significance.

Figure 9

VSMC-specific Nfe2l2 knockdown abolishes the protective effect of CTS in Ang II-induced mouse AAA model. (A) Mouse organs were collected for GFP-fluorescence detection. Both bright-field and fluorescence images were captured under the same conditions. (B) Representative images of fluorescence imaging of mouse aorta across sections (M, media; A, adventitia). (C) Representative western blots and quantitative analysis of indicated protein expression in mouse brains and aortas. n = 5. (D) Schematic illustration of SMC-specific Nfe2l2 knockout in mice, AAA model establishment, and CTS administration strategy. Ten-week-old male ApoE^-/- mice were tail-vein-injected with AAV2-shCtrl and AAV2-shNfe2l2 viruses, respectively. After two weeks, all the mice were infused with Ang II (1000 ng/kg/min) to establish the AAA model. On the day the AAA was started to induce, HP-β-CD (vehicle) or high-dose CTS (50 mg/kg) were intragastrically delivered daily, respectively. At the endpoint, blood, aorta, and other organs were collected for the following experiments. (E) Percent survival. (F) AAA incidence. (G) Representative images of abdominal aortic aneurysm. (H) Maximal diameter of the abdominal aorta. (I) Representative images and quantitative analysis (for elastin) of vessel cross-sections stained with H&E, EVG, and Masson's trichrome staining. (J) Quantitative analysis of elastin degradation related to EVG staining in panel I. (K) Representative images and quantitative analysis of immunohistochemical staining for Nrf2. (L) Images and quantitative analysis of Nrf2 and α-SMA immunofluorescence in the murine abdominal aorta. (M) Cytokines from mouse serum were measured by ELISA kits (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ns: no significance.

Figure 10

SMC-specific Nfe2l2 knockdown abolishes the effect of CTS on anti-inflammatory, antioxidant, and inhibiting MMP activity and generation in vivo. (A) Representative immunohistochemical staining and quantitative analysis of indicated proteins in mouse aorta cross-sections (n = 5). (B) Representative western blots and quantitative analysis of indicated proteins in murine aorta (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ns: no significance.