Role of smooth muscle cell p53 in pulmonary arterial hypertension

Pulmonary arterial hypertension (PAH) is characterized by remodeling and narrowing of the pulmonary arteries, which lead to elevation of right ventricular pressure, heart failure, and death. Proliferation of pulmonary artery smooth muscle cells (PASMCs) is thought to be central to the pathogenesis of PAH, although the underlying mechanisms are still being explored. The protein p53 is involved in cell cycle coordination, DNA repair, apoptosis, and cellular senescence, but its role in pulmonary hypertension (PH) is not fully known. We developed a mouse model of hypoxia-induced pulmonary hypertension (PH) and found significant reduction of p53 expression in the lungs. Our in vitro experiments with metabolomic analyses and the Seahorse XF extracellular flux analyzer indicated that suppression of p53 expression in PASMCs led to upregulation of glycolysis and downregulation of mitochondrial respiration, suggesting a proliferative phenotype resembling that of cancer cells. It was previously shown that systemic genetic depletion of p53 in a murine PH model led to more severe lung manifestations. Lack of information about the role of cell-specific p53 signaling promoted us to investigate it in our mouse PH model with the inducible Cre-loxP system. We generated a mouse model with SMC-specific gain or loss of p53 function by crossing Myh11-Cre/ERT2 mice with floxed Mdm4 mice or floxed Trp53 mice. After these animals were exposed to hypoxia for 4 weeks, we conducted hemodynamic and echocardiographic studies. Surprisingly, the severity of PH was similar in both groups of mice and there were no differences between the genotypes. Our findings in these mice indicate that activation or suppression of p53 signaling in SMCs has a minor role in the pathogenesis of PH and suggest that p53 signaling in other cells (endothelial cells, immune cells, or fibroblasts) may be involved in the progression of this condition.


Introduction
In patients with pulmonary arterial hypertension (PAH), progressive remodeling and narrowing of the pulmonary arteries eventually result in right ventricular failure and death [1]. Recently, some medications have been shown to improve the clinical outcome [2][3][4], but the overall prognosis remains unacceptably poor. The mechanisms underlying PAH have not been fully characterized and there is a major unmet medical need in this area. It is well accepted that vascular cells in PAH share similar metabolic features with cancer [5]. Inhibition of mitochondrial respiration, suppression of glucose oxidation, and activation of glycolysis provide advantages for proliferating cancer cells [6]. Pulmonary artery smooth muscle cells (PASMCs), endothelial cells, and fibroblasts also develop a proliferative phenotype by such metabolic alterations, which are thought to promote the pathogenesis of PAH [5]. PASMCs from PAH patients and animal models of PAH display elevation of glycolysis that is associated with reduction of glucose oxidation and mitochondrial respiration [7,8], resembling the metabolic profile of cancer cells. Vascular cells involved in the pathology of PAH also become resistant to apoptosis, contributing to uncoordinated cell proliferation that leads to intimal and medial thickening [9]. The protein p53 is a well-characterized transcription factor that is involved in cell cycle coordination, DNA repair, apoptosis, and cellular senescence, as well as in maintenance of genomic stability and suppression of tumorigenesis [10]. Activation of p53 signaling inhibits cell proliferation and suppresses glycolysis. Accordingly, it is highly possible that reduced expression of p53 contributes to the pathogenesis of PAH, while activation of p53 signaling may ameliorate PAH. In support of this concept, it was previously reported that systemic p53 deficiency led to exacerbation of hypoxia-induced pulmonary hypertension (PH) in mice [11]. However, there has been no investigation of the role of cell-specific p53 signaling in PH models. We used an inducible Cre-loxP system targeting smooth muscle cells (SMCs) to investigate the influence of cell-specific p53, and demonstrated that SMC-specific gain or loss of p53 function did not lead to exacerbation of hypoxia-induced PH compared with wild-type mice.

Animal models
All animal experiments were conducted in compliance with the protocol reviewed by the Institutional Animal Care and Use Committee of Niigata University and approved by the President of Niigata University. C57BL/6NCr male mice were purchased from SLC Japan (Shizuoka, Japan). Mice carrying floxed Trp53 alleles (Trp53 fl/fl ) and Cre recombinase in Myh11-positive cells (Myh11-Cre/ERT2) were purchased from Jackson Laboratories. Then Myh11-Cre/ERT2; Trp53 fl/fl mice or Myh11-Cre/ERT2; Mdm4 fl/fl mice were generated by crossing heterozygous Myh11-Cre/ERT2 mice with Trp53 fl/fl or Mdm4 fl/fl mice [12]. To induce Cre recombinase, mice received daily injection with 1 mg of tamoxifen (Sigma Life Science T5648-1G) dissolved in 300 μl of corn oil (Sigma Life Science C267-500ML) for 5 days from 7 weeks old. A mouse model of PH was generated by housing the animals in a hypoxic chamber with an atmosphere of 8% O 2 for 4 weeks from 8 weeks old. Mice were anesthetized with isoflurane for catheter studies. We used a 25G needle with a pressure sensor (FTH-1211B-0018, Primetech) to measure the right ventricle systolic pressure (RVSP), and calculations were done with a LabChart8 (AD Instruments). All studies for pressure measurement are done and analyzed in a blinded way with genotypes. After the animals were euthanized by isoflurane or intraperitoneal barbiturate injection, tissues were quickly corrected for further analyses.

Cell culture
Human pulmonary arterial smooth muscle cells (PASMCs) and culture medium (CC-3182) were purchased from Lonza, and the cells were incubated according to the manufacturer's instructions. In some experiments, hypoxic stress was introduced by maintaining the cells under a 1% O 2 atmosphere for 72 hours in a hypoxic chamber (Stemcell Technologies).

Echocardiography
Echocardiography was performed with a Vevo 2100 High Resolution Imaging System (Visual Sonics Inc.). To minimize variation of the data, cardiac function was only assessed when the heart rate was within the range of 550-650 /min. All studies for echocardiography were performed and analyzed in a blinded way with genotypes.
p53 in pulmonary arterial hypertension

Extracellular flux assay
Cellular oxygen consumption and extracellular acidification were measured with a Seahorse XF extracellular flux analyzer according to the manufacturer's instructions (Agilent). PASMCs were cultured in a Seahorse XF 24-well assay plate in antibiotic-free medium at a density of 20,000 cells/well, and were treated with siRNA for p53 (si-TP53; Invitrogen) and negative control siRNA (Invitrogen) after attachment. The medium was replaced with antibiotic-containing medium after 24 hours. After incubation for a further 24 hours, the plates were washed and the medium was replaced with pre-warmed running medium (XF base medium supplemented with 10% D-glucose, 100 mM pyruvate, and 200 mM glutamine for the mitochondrial stress test, or with 200 mM glutamine alone for the glycolysis stress test). Then incubation was performed in a non-CO 2 incubator for 60 min at 37˚C. The basal oxygen consumption rate and extracellular acidification rate were recorded for 24 min, followed by performance of the mitochondrial stress test (1μM oligomycin, 2 μM FCCP, and 0.5 μM rotenone/antimycin A) and glycolysis stress test (10 mM glucose, 1 μM oligomycin, and 50 mM 2-DG). All reagents were obtained from the Seahorse XF Cell Mito Stress Test Kit (Seahorse Bioscience, #103015-100) and the Seahorse XF Glycolysis Stress Test Kit (Seahorse Bioscience, #103020-100).

Metabolomic analyses
Metabolomic analyses were done according to the method of Soga et al. by using capillary electrophoresis/mass spectrometry (CE/MS), as described previously [13]. Briefly, 1,000,000 PASMCs in a 6-cm dish were treated with si-RNA as described above, washed twice with ice cold 5% mannitol, and let stand for 10 min at room temperature in methanol containing Lmethionine sulfone (25 μM, Alfa Aesar A17027), MES (25 μM, Dojindo 341-01622), and CSA (25 μM, Wako 037-01032). Then the cells were harvested with a cell scraper and 400 μL of supernatant was collected after vortexing for 30 seconds. After addition of CHCl 3 (400 μl) and distilled water (200 μl) with thorough mixing, centrifugation was performed at 10,000 g for 3 minutes at 4˚C. Next, the aqueous layer (400 μl) was transferred to an ultrafiltration tube (UltrafreeMC-PLHCC, Human Metabolome Technologies, UFC3LCCNB-HMT), followed by centrifugation at 9,100 g for 3 hours at 4˚C. Finally, 320 μL of the filtrate was sent to the Institute for Advanced Biosciences at Keio University for further analyses.

Statistical analysis
Results are shown as the mean ± SEM. Outliers and abnormal values were excluded by boxplot analyses. Differences between groups were examined by the two-tailed Student's t-test or twoway ANOVA, which were followed by Tukey's multiple comparison test for comparisons among more than two groups. In all analyses, P<0.05 was considered statistically significant. Analyses were done with SPSS software (version 24).

Generation of the PH model with hypoxic stress
We generated a murine PH model by exposing C57BL6/NCr mice to hypoxia (8% O 2 ) for 4 weeks (Fig 1A). Exposure to hypoxia led to a significant increase of right ventricular systolic pressure (RVSP) (Fig 1B), right ventricular wall thickness (RVWT) (Fig 1C), and right ventricular diameter (RVD) (Fig 1D). In contrast, left ventricular systolic function (S1A Fig), left ventricular end-systolic diameter (S1B Fig), and heart rate (S1C Fig) were comparable between normoxic mice and mice with hypoxia-induced PH. Pathological examination showed pulmonary artery wall thickening in our PH model (Fig 1E), indicating that it displayed features of pulmonary arterial hypertension (PAH). Because p53 null mice were reported to show advanced changes of PH [11], we investigated the overall level of p53 expression in PH lungs and found significant reduction of p53 (Fig 1F), which might contribute to vascular remodeling with progression of PH.

Suppression of p53 expression in PASMCs upregulates glycolysis
Because p53 expression was reduced in PH lungs, we performed in vitro investigation of the role of p53 in PASMCs. We found that exposure to hypoxic stress led to reduction of p53 expression by these cells (Fig 2A). To further characterize the role of p53 in PASMCs, we used siRNA targeting p53 to suppress its expression (Fig 2B). Metabolomic studies showed that suppression of p53 led to a marked increase of glucose-6-phosphate in these cells, as well as a significant increase of fructose-1,6-bisphosphate, pyruvate, and lactate ( Fig 2C). Evaluation of glycolysis by using the Seahorse XF extracellular flux analyzer showed that suppression of p53 expression increased glycolysis, glycolytic capacity, and the glycolytic reserve (Fig 2D). These results indicated that inhibition of p53 led to a metabolic shift toward glycolysis in PASMCs.

Suppression of p53 inhibits the TCA cycle and mitochondrial respiration in PASMCs
Next, we tried to assess the changes of the TCA cycle and mitochondrial respiration with p53 inhibition in PASMCs. Metabolomic studies indicated that acetyl-CoA and citrate were reduced by suppression of p53 expression. There was no change of cis-aconitate, but isocitrate, fumarate, and malate levels were significantly reduced by p53 suppression (Fig 3A). Next, we evaluated mitochondrial respiration by using the Seahorse extracellular flux analyzer, revealing that p53 inhibition reduced basal respiration, maximal respiration, ATP production, and spare capacity (Fig 3B). These results indicated that both the TCA cycle and mitochondrial respiration were suppressed in PASMCs by inhibition of p53. Taken together, the in vitro data indicated that suppression of p53 expression in PASMCs led to metabolic remodeling, which was characterized by activation of glycolysis and inhibition of mitochondrial respiration, mimicking the features of cancer cell metabolism.

Smooth muscle cell-specific gain or loss of p53 function has no impact on hypoxia-induced PH
To further investigate the role of p53 signaling in PASMCs, we generated mouse models with SMC-specific depletion or activation of p53. The SMC-specific p53 depletion model was generated by crossing Myh11-Cre/ERT2 mice with floxed p53 mice (Myh11-Cre/ERT2; Trp53 fl/fl (SMC-p53KO)), while the SMC-specific p53 activation model was generated by crossing Myh11-Cre/ERT2 mice with Mdm4 fl/fl mice (Myh11-Cre/ERT2; Mdm4 fl/fl (SMC-Mdm4KO)). Myh11-Cre/ERT2 mice were analyzed as controls. Myh11-Cre/ERT2 (control), SMC-p53KO mice, and SMC-Mdm4KO mice were housed under hypoxic conditions (8% O 2 ) for 4 weeks ( Fig 4A). First, we investigated p53 expression in aortic tissue with abundant SMCs, revealing that p53 expression was increased in SMC-Mdm4 KO mice and reduced in SMC-p53KO mice (Fig 4B). We also confirmed an increase of p53-positive SMC number in pulmonary arteries of SMC-Mdm4KO mice, while SMC-p53KO mice showed a significant reduction of p53-positive SMC (S2A Fig). Surprisingly, hemodynamic studies showed that RVSP was comparable among the genotypes (Fig 4C). Likewise, echocardiography indicated that RVWT and RVD were similar among the control mice, SMC p53KO mice, and SMC Mdm4KO mice (Fig 4D  and 4E).  (Fig 4F). These results indicated that vascular smooth muscle cell-specific gain or loss of p53 function had no effect on hypoxia-induced PH.

Discussion
Because p53 regulates the cell cycle, DNA repair, apoptosis, cellular senescence, and cellular metabolism, it has an important role in the maintenance of homeostasis [10]. We previously reported that p53-induced cellular senescence is a critical process in cardiovascular and metabolic diseases [14][15][16][17][18][19]. We found that p53 signaling was upregulated in cardiomyocytes, vascular cells, adipocytes, and immune cells in cardiovascular and metabolic diseases, whereas p53 inactivation alleviated these age-related disorders. In cardiovascular and metabolic diseases, senescent cells are characterized by growth arrest and alterations of the genetic profile that contribute to initiation and maintenance of chronic sterile inflammation associated with tissue remodeling [20]. Accordingly, suppression of cellular senescence or elimination of senescent cells may have the potential to become next generation therapy for such age-related disorders [21][22][23]. PAH develops due to remodeling and narrowing of the pulmonary arteries. In this condition, vascular cells such as PASMCs, endothelial cells, and fibroblasts exhibit a proliferative phenotype, and uncoordinated proliferation of these cells contributes to vascular remodeling. In PAH, pulmonary vascular cells develop an anti-apoptotic phenotype and undergo metabolic remodeling that is characterized by increased glycolysis, reduced glucose oxidation, and  p53 in pulmonary arterial hypertension reduced mitochondrial respiration [5]. This is similar to the metabolic profile of cancer cells, which is recognized to provide a survival advantage for proliferating cells. Since p53 signaling inhibits cell proliferation and also suppresses glycolysis, we hypothesized that downregulation of p53 could exacerbate the pathology of PAH, whereas p53 activation may ameliorate PAH. In support of this concept, it was previously reported that systemic p53 deficiency promoted hypoxia-induced PH in mice [11]. Accordingly, we attempted to characterize p53 signaling in specific cell types. PASMCs from PAH patients are reported to show increased glycolysis, reduced glucose oxidation, and reduced mitochondrial respiration, with these metabolic changes having a critical role in vascular remodeling [9,24]. Therefore, we focused on p53 signaling in PASMCs and its potential role in PH. As expected, our in vitro studies showed that genetic suppression of p53 in PASMCs led to upregulation of glycolysis and downregulation of mitochondrial respiration. The metabolic shift was similar to that seen in cancer cells, and we predicted it would promote the progression of PH in an animal model. However, we found that SMC-specific gain or loss of p53 function had no phenotypic impact in a mouse model of hypoxia-induced PH. Hemodynamic studies and echocardiography showed that RVSP, RVWT, and RVD were comparable among control mice, SMC-p53KO mice, and SMC-Mdm4KO mice exposed to hypoxia. These findings indicated that activation or suppression of p53 signaling in SMCs has a minor role in the pathology of PH and suggest that p53 signaling in other cells (including endothelial cells, immune cells, or fibroblasts) may be involved in progression of this condition. PH is recognized to be a multiorgan disease, and tissue remodeling is known to occur in various organs, including the skeletal muscle, spleen, and bone marrow [5,9]. Thus, activation of p53 in SMCs alone may not have been sufficient to influence the course of PH, which would be mediated by coordinated modulation of p53 signaling in other cells.
A potential limitation of this study is the model analyzed. A hypoxia-induced PH model is widely used for PH research; however, it is known that pathological changes are not completely compatible with the pathology of human PAH lung. For example, plexiform lesion does not develop in most cases, and therefore fundamental questions that remain to be answered are whether mouse models are suitable for understanding pathologies in PH patients. Another potential limitation of our study is related to divergent functions of p53. It is well known that p53 is involved in a broad spectrum of cellular processes including cell cycle, apoptosis and cellular senescence in addition to its role in maintaining cell metabolism. Although previous studies suggest that p53 may control pulmonary SMC proliferation via regulation of metabolism, p53 may affect SMC proliferation and/or survival via other mechanisms. Further studies would be required to understand the role of p53 in the pathogenesis of PH.