3.1 | Inactivation of SERCA2 C674 suppresses PPARγ by activation of calcineurin/NFAT/NF-κB pathways
Our previous study indicates that the inactivation of C674 by causing the accumulation of intracellular Ca2+ to activate calcineurin-mediated NFAT/NF-κB pathways, resulted in SMC phenotypic modulation to accelerate aortic aneurysm. In order to study the mechanism, we first performed RNA sequencing. Bioinformatics analysis showed that inactivation of C674 had a wide range of effects on gene expression, and a variety of signaling pathways were affected, among which PPAR signaling pathway was the most closely related to the SMC phenotypic modulation (Figure 1A). PPAR belongs to the nuclear receptor superfamily, including PPARα, PPARβ and PPARγ. FPKM values showed that inactivation of C674 had no effect on PPARα (WT: 1.49 ± 0.28; SKI: 1.06 ± 0.11) and PPARβ (WT: 25.13 ± 2.22; SKI: 29.74 ± 2.93), but decreased PPARγ2 (WT: 5.65 ± 0.68; SKI: 2.44 ± 0.21). The down-regulation of PPARγ2 was further confirmed at both mRNA level by real-time quantitative PCR (Figure 1B) and protein level by Western blot (Figure 1C) in aorta of SKI mice.
Activation of calcineurin or NF-κB inhibited the expression of PPARγ in adipocytes (Liu et al., 2007) and pulmonary artery SMCs (Xie et al., 2017). To explore whether C674 inactivation regulates the transcription of PPARγ2 by affecting its promoter activity, we constructed a mouse PPARγ2 promoter from -630 to +45 bp, containing NFAT/NF-κB binding sites. Compared with WT SMCs, the activity of PPARγ2 promoter in SKI SMCs was down-regulated (Figure 1D), which could be reversed by either calcineurin inhibitor CsA or NF-κB inhibitor PDTC (Figure 1E). Correspondingly, compared with WT SMCs, the protein expression of PPARγ was down-regulated in SKI SMCs, which could be reversed by either CsA (Figure 1F) or PDTC (Figure 2A), suggesting that PPARγ might be the downstream target of calcineurin/NFAT/NF-κB pathways. We have reported that inactivation of C674 decreased the expression of MYOCD, increased the expression of OPN, MMP2, Col I, Col III, phosphorylated p65NF-κB, VCAM1, and ICAM1, accelerated SMC proliferation, migration and macrophage adhesion to SMCs, and all these characters could be reversed by CsA (Que et al., 2020). CsA inhibits the nuclear translocation of NFAT4 and NF-κB in SKI SMCs (Que et al., 2020). Here we showed that NF-κB inhibitor PDTC had similar effect to CsA in SKI SMCs (Figure 2), further supporting that the activation of NF-κB pathway caused by inactivation of C674 promoted SMC phenotypic modulation. These results indicate that inactivation of C674 promotes the binding of NFAT/NF-κB to PPARγ2 promoter by increasing the nuclear translocation of NFAT/NF-κB, thus negatively regulating the transcription of PPARγ2, resulting in the decrease of PPARγ protein expression.
3.2 | The down-regulated PPARγ2accounts for the phenotypic modulation of SKI SMCs
PPARγ prevents the dedifferentiation of SMCs, restricts SMC proliferation, migration and the activation of inflammatory pathway (Abe et al., 2003; Hamblin et al., 2009). The two isoforms of PPARγ, PPARγ1 and PPARγ2, differ at their N terminal. The protective effect of PPARγ1 on the phenotypic modulation of SMCs was studied (Halabi et al., 2008; Hu et al., 2008; Meredith et al., 2009). However, the role of PPARγ2 in the regulation of SMC phenotype is still unclear. Although PPARγ2 has 30 more amino acids than PPARγ1, it has similar DNA binding area and ligand binding region to PPARγ1 (Werman et al., 1997). In SKI SMCs, overexpression of PPARγ2 increased the expression of MYOCD and decreased the expression of OPN, MMP2, Col I, Col III, phosphorylated p65NF-κB, VCAM1 and ICAM1 (Figure 3A), and inhibited cell proliferation, migration, and macrophage adhesion to SMCs (Figure 3B-D). This provides direct evidence that PPARγ2 inhibits the phenotypic modulation of SMCs, and indicates that the decrease of PPARγ2 is accountable for the phenotypic modulation of SKI SMCs.
3.3 | Activation of PPARγ prevents SMC phenotypic modulation and represses the activities of NFAT/NF-κB in SKI SMCs
Next, we used PPARγ activator pioglitazone to compensate for the down-regulation of PPARγ expression in SKI SMCs. Activation of PPARγ by pioglitazone in SKI SMCs, similar to the overexpression of PPARγ2, also increased the expression of MYOCD and decreased the expression of OPN, MMP2, Col I, Col III, phosphorylated p65NF-κB, VCAM1 and ICAM1 (Figure 4A), and inhibited cell proliferation, migration, and macrophage adhesion to SMCs (Figure 4B-D), which further confirmed the contribution of the down-regulated PPARγ in promoting the phenotypic modulation of SKI SMCs.
Though PPARγ could be a downstream target of calcineurin/NFAT/NF-κB signaling pathways, in return it could also interfere with NFAT and NF-κB to repress their activities (Blanquart et al., 2003). Inactivation of C674 increases the nuclear translocation of NFAT4 and NF-κB to be active (Que et al., 2020). As shown in Figure 5, the activation of PPARγ by pioglitazone could inhibit the nuclear translocation of NFAT4 and NF-κB in SKI SMCs, suggesting the mutual negative regulation between NFAT/NF-κB and PPARγ in SMCs.
3.4 | Activation of PPARγ by pioglitazone ameliorated angiotensin II-induced aortic aneurysm in SKI mice
Activation of PPARγ by pioglitazone protected the formation and rupture of experimental aortic aneurysms in mice (Golledge et al., 2010; Shimada et al., 2015). Next, we used LDLR−/− background mice for angiotensin II-induced aortic aneurysm analysis, and pioglitazone or solvent control administered 7 days after angiotensin II infusion (Figure 6A). After 28 days of angiotensin II infusion, there was no difference in body weight, plasma levels of triglyceride and cholesterol among WT mice treated with solvent control, SKI mice treated with solvent control, and SKI mice treated with pioglitazone (Table 1).
During 28-day angiotensin II infusion, all WT mice treated with solvent control (11 out of 11) and SKI mice treated with pioglitazone (11 out of 11) survived, while 7 out of 11 SKI mice treated with solvent control survived (Figure 6B). As shown in Figure 6C, 8 out of 11 WT mice treated with solvent control developed aortic aneurysms. In contrast, all SKI mice treated with solvent control (11 out of 11) developed aortic aneurysms, in which 4 of them had early death due to the aneurysm rupture. Pioglitazone slightly decreased the incidence of aortic aneurysms in SKI mice (9 out of 11) comparable to WT mice treated with solvent control. Compared with WT mice treated with solvent control, SKI mice treated with solvent control had severer aortic aneurysm, of which 4 out of 11 had ruptured aneurysms. Pioglitazone ameliorated aortic aneurysms in SKI mice with improved composition of aortic aneurysm comparable to WT mice treated with solvent control (Figure 6D). Accordingly, SKI mice treated with solvent control had a higher grade of elastin degradation and a slight increase of collagen deposition than WT mice treated with solvent control, and pioglitazone treatment could reverse the increased elastin degradation and collagen deposition in SKI mice comparable to WT mice treated with solvent control (Figure 6E&F).
Mice (4 out of 11 SKI mice treated with solvent control) that died of aneurysm rupture and mice without aneurysm (3 out of 11 WT mice treated with solvent control and 2 out of 11 SKI mice treated with pioglitazone) during the 28-day angiotensin II infusion were excluded from aneurysm size analysis. The largest aneurysm in each mouse was analyzed. Compared with WT mice treated with solvent control, in SKI mice treated with solvent control, the diameter of aneurysm, the outer area of aneurysm, and the media area with aneurysm, all showed a trend of increase, while the lumen area showed a trend of decrease, and all these characters of SKI mice could be reversed by pioglitazone treatment (Figure 7A). The length and number of aneurysms were significantly larger in SKI mice treated with solvent control compared with WT mice treated with solvent control, while pioglitazone treatment reduced the length and number of aneurysms in SKI mice (Figure 7B&C).
In SKI non-aneurysmal aortas infused with angiotensin II, though pioglitazone had no effect on the protein expression of PPARγ compared with solvent control, it did increase the expression of MYOCD and decrease the expression of OPN, MMP2, Col I, Col III, phosphorylated p65NF-κB, VCAM1, and ICAM1 (Figure 8), further supporting the regulation of PPARγ on these aneurysm-related proteins in vivo . These results indicated that inactivation of C674 by downregulation of PPARγ to promote the phenotypic modulation of SMCs, thus aggravated angiotensin II-induced aortic aneurysms, and pioglitazone by suppressing these aneurysm-related proteins to improve aortic aneurysms.