Aldosterone enhances high phosphate–induced vascular calcification through inhibition of AMPK‐mediated autophagy

Abstract It remains unclear whether the necessity of calcified mellitus induced by high inorganic phosphate (Pi) is required and the roles of autophagy plays in aldosterone (Aldo)‐enhanced vascular calcification (VC) and vascular smooth muscle cell (VSMC) osteogenic differentiation. In the present study, we found that Aldo enhanced VC both in vivo and in vitro only in the presence of high Pi, alongside with increased expression of VSMC osteogenic proteins (BMP2, Runx2 and OCN) and decreased expression of VSMC contractile proteins (α‐SMA, SM22α and smoothelin). However, these effects were blocked by mineralocorticoid receptor inhibitor, spironolactone. In addition, the stimulatory effects of Aldo on VSMC calcification were further accelerated by the autophagy inhibitor, 3‐MA, and were counteracted by the autophagy inducer, rapamycin. Moreover, inhibiting adenosine monophosphate‐activated protein kinase (AMPK) by Compound C attenuated Aldo/MR‐enhanced VC. These results suggested that Aldo facilitates high Pi‐induced VSMC osteogenic phenotypic switch and calcification through MR‐mediated signalling pathways that involve AMPK‐dependent autophagy, which provided new insights into Aldo excess‐associated VC in various settings.

VC, which indicated the involvement of Aldo in VC. [6][7][8][9] Besides Aldo excess, a series of mineral disorders and hormonal imbalance including altered homeostasis of calcium, phosphorus, 1,25-dihydroxyvitamin D and parathyroid hormone, occurs gradually with the decline in kidney function. 10 Therefore, whether Aldo is an initiator or just a promoter of VC remains unclear 11 and warrants further study. Primary aldosteronism (PA) is an endocrine disorder characterized by autonomous Aldo secretion and hypertension. Our recent study showed that PA patients, matched for age, sex and blood pressure, exhibited more severe abdominal aortic calcification than those with essential hypertension. 12 Notably, we noticed that most included PA patients had different degree of reduced estimated glomerular filtration rate, which was tightly linked to increased inorganic phosphate (Pi) concentration. 13 Therefore, whether the calcified mellitus induced by high Pi is required for Aldo-enhanced VC, and the potential mechanisms for Aldo-MR complex to promote VC need to be clarified.
It has been implicated that autophagy acts as a mediator in hyperaldosteronism-induced target-organ damage. [14][15][16] Autophagy is a process highly regulated by adenosine monophosphate-activated protein kinase (AMPK) signalling; impaired autophagy activity in vascular cells is closely associated with vascular disorders. 17 Under various pathophysiological conditions, vascular smooth muscle cells (VSMCs) undergo autophagy and generate double-membrane vesicles called autophagosomes, which fuse with the lysosome leading to lytic degradation of autophagosomal contents. 18 Recent evidence indicated that up-regulation of autophagy counteracted the high Pi-induced phenotypic switch and calcification of VSMCs. 19,20 However, the role of VSMC autophagy in the VC regulated by Aldo-MR complex is unknown.
In the present study, we assessed the necessity of high Pi in the association between Aldo and VC in vivo and in vitro. We then explored the roles of AMPK signalling-dependent autophagy in the Aldo-enhanced VSMC osteogenic differentiation and calcification.

| Animal experiments
After 12 weeks, the animals were killed and mouse serum was collected to determine the blood urea nitrogen (BUN), creatinine (Cr), calcium and phosphorus. The thoracic aortas were excised for further analysis.

| Cell cultures
Primary mouse VSMCs were extracted from 8-week-old male C57BL/6 mice. Briefly, the mice were killed by sodium pentobarbital, and the thoracic aorta was dissected out quickly. After removing the adventitia and intima, the artery segment was cut into 1-2 mm 2 sections and placed in a cell culture flask with DMEM containing 4.5 g/L glucose supplemented with 20% FBS, 100 U/mL penicillin and 100 mg/mL streptomycin (growing medium). After confirming the VSMC phenotype and purity by positive immunostaining for α-SMA ( Figure S1), VSMCs were incubated at 37°C in a humidified atmosphere containing 5% CO 2 and generally switched into DMEM with 5% FBS. VSMCs at passages 4-8 were used for the experiments. The culture medium was replaced every 3 days. As shown in Figure S2, 2.6 mmol/L NaH 2 PO 4 was the minimal concentration to induce VSMC calcification ( Figure S2).

| von Kossa staining and Alizarin red S staining
For von Kossa staining, 3 μm paraffin-embedded artery sections were exposed to a 5% silver nitrate solution and were placed under an ultraviolet light for 2 hours after the standard dewaxing procedure.
For Alizarin Red staining, VSMCs in 6-well plates were washed

times with PBS after indicated treatments and then fixed with 4%
paraformaldehyde for 30 minutes. Then, the cells were washed 3 times with PBS and exposed with 2% Alizarin Red solution (Sigma) for 10 minutes. Positively stained cells display a red-orange colour. 19

| Alkaline phosphatase (ALP) activity and calcium content detection
The ALP Assay Kit (A059-2) and Calcium Assay Kit (C004-2) were from Nanjing Jiancheng Bioengineering Institute. The ALP activity and calcium content were determined according to the manufacturer's instructions. The results were then normalized by the protein content determined by the BCA Protein Assay Kit (P0012) from Beyotime. 23

| Western blotting analysis
The protein samples were extracted from VSMCs with RIPA lysis detected by ECL blotting substrate and then exposed to an X-ray film. The densities of the bands were semi-quantified and normalized against GAPDH by Image J software (Thermo Scientific).  was measured using a microplate reader (Bio-Tek). Cell viability was normalized to control.

| Statistical analysis
All experiments were independently repeated at least 3 times. Data were presented as mean ± SEM. Differences of data among multiple groups were determined by one-way ANOVA followed by a LSD test. Statistical analysis was performed with SPSS version 20 (SPSS, Inc, Chicago, IL), and P < .05 was considered statistically significant.
GraphPad Prism 8.0 was used to draw figures.

| Aldo facilitates calcium deposition in vivo and in vitro
As shown in Table 1, serum phosphorus and BUN levels (both P < .05) were higher in HPD-treated mice than the control and DOCA-treated mice. However, the serum levels of calcium and Cr were similar among the 4 groups.
To assess vascular medial calcification in HPD-induced non-CKD mice, von Kossa staining was performed. As shown in Figure 1A, HPD-

| Aldo promotes oseteogenic phenotypic switch of VSMC in MR-dependent manner in vitro
The switch from contractile VSMCs into osteo-/chondrocytic-like cells has been increasingly accepted as a key mechanism of VC. 26 Thus, we attempted to investigate whether Aldo exerted procalcific effects via triggering VSMC phenotypic transdifferentiation. As shown in Figure 2A,B, the procalcific effect of Aldo was greatly sup-

| Aldo down-regulates high Pi-induced VSMC autophagy
We then explored whether Aldo affected VSMC autophagy in high Pi condition. As shown in Figure 3A,B, high Pi treatment induced a significant increase in LC3-II formation, an indicator of autophagy.
However, Aldo stimulation for 7 days caused a dose-dependent decrease in LC3-II formation and Beclin-1 protein expression and reached a peak inhibition at the dose of 100 nmol/L (P < .05).
As the change of LC3II is possibly due to modulation of autophagosome formation or autophagic degradation, 27 we inhib-

| Aldo-progressed VSMC calcification and osteogenic differentiation are through autophagy inhibition
Next, we aimed to investigate the role of autophagy in VSMC calcification and osteogenic differentiation aggravated by Aldo. As F I G U R E 5 Adenosine monophosphate-activated protein kinase (AMPK) phosphorylation mediated aldosterone (Aldo)-mineralocorticoid receptor-dependent inhibition of vascular smooth muscle cells (VSMC) autophagy. A and B, Representative Western blot bands and semiquantitative analysis of LC3, Beclin-1, phospho-AMPKα (P-AMPKα) and total AMPKα (T-AMPKα) in VSMCs. C, Alizarin red S staining of microscopic (scale bar, 50 μm) images in mice VSMCs. D, Quantitative analysis of the calcium content of VSMCs after induction with different stimuli for 7 d, normalized to the protein levels, and 100 nmol/L Aldo, 1 µmol/L Compound C and 20 μmol/L spironolactone (Sipro) were used. n = 5 per group. *P < .05 vs control group; # P < .05 vs Pi group; & P < .05 vs Aldo + Pi group; % P < .05 vs Aldo + Pi + Spiro group [Colour figure can be viewed at wileyonlinelibrary.com] shown in Figure 4A,B, Aldo augmented high Pi-induced VSMC calcification, and addition of the pharmacological inhibitor, 3-MA (10 μmol/L), further accelerated this effect. However, treatment of the pharmacological inducer, Rapa (10 μmol/L), significantly attenuated Aldo-enhanced VSMC calcification (P < .05). Pre-treating VSMCs with 3-MA further aggravated LC3II down-regulation, whereas Rapa treatment abrogated LC3II reduction induced by Aldo.
As AMPK signalling was reported as an important regulator of autophagy, 15 we next investigated the role of AMPK signalling in the Aldo-MR complex-mediated inhibition of VSMC autophagy. As shown in Figure 5A,B, addition of Aldo reduced the ratio of P-AMPKα

| D ISCUSS I ON
Our present study demonstrates that only in the calcified mellitus induced by high Pi medium, Aldo binds to mineralocorticoid receptor (MR) and induces a switch of VSMCs from contractile to osteogenic phenotype, promoting VSMC calcification. The molecular mechanisms underlying these processes involve AMPK-dependent autophagosome formation.
Previous studies have indicated the association between Aldo excess and VC, but the role of high Pi in this association has been debated. [6][7][8][9]28,29 Some investigators found that only in calcified mellitus induced by high Pi, Aldo promoted VC through a MR-dependent manner. 7,8 In addition, inhibition of type III sodium-dependent Pi cotransporter Pit-1 significantly abrogated the procalcific effects of Aldo, suggesting that it required high Pi mellitus for Aldo-enhanced VC. 9 However, some argued that long-term exposure to Aldo alone could induce VSMC calcification. 6,28,29 In this study, we used DBA/2N mice to build VC model in diet (2%) in standard regimen for building VC in these mice was much higher than that was consumed by humans, 20 Moreover, we revealed that AMPK signalling-dependent VSMC autophagy was a potential molecular mechanism for Aldo/MRmediated calcification in VSMCs.
Multiple mechanisms are involved in VC, 37 such as inflammation, oxidative stress and VSMC apoptosis. Whether these processes are also involved in Aldo-enhanced VC still needs further investigation.
Additionally, although we showed that MR-dependent mechanism played a key role in Aldo procalcific effects, we could not exclude the MR-independent effects in Aldo-increased VC. 38 Therefore, further studies are warranted.
In conclusion, we identified Aldo as a promoter of VC, rather than an initiator. Under calcified conditions, Aldo excess facilitated VSMC osteogenic differentiation and subsequent calcification via inhibition of AMPK-dependent autophagy ( Figure 6). Inhibition of Aldo/MR signalling by MR blockers or pharmacological up-regulation of autophagy may protect against VSMC osteogenesis and VC in the setting of Aldo excess.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets generated and/or analysed during the current study are available from the corresponding author on reasonable request.