Early-passage (3–8 passage) human PAVSMC isolated from small (≤1 mm outer diameter) PAs of patients with PAH and non-diseased subjects were provided by UC Davis Lung Center Pulmonary Vascular Disease Program human specimens biobank, University of Pittsburgh Vascular Medicine Institute Cell Processing Core, and the Pulmonary Hypertension Breakthrough Initiative (PHBI) under approved protocols in accordance with Institutional Review Board (IRB) and Committee for Oversight of Research and Clinical Training Involving Decedents (CORID) policies. Cells isolation, characterization and maintenance were performed under PHBI-approved protocols as described previously (10, 32). Cells were maintained in complete PromoCell Smooth Muscle Cell Growth Medium 2 with SupplementPack and Antibiotic-Antimycotic. Before experiments, cells were incubated for 24–48 h in basal media supplemented with 0.1% bovine serum albumin (BSA) if not stated otherwise. For functional experiments (Ki-67, cell counts, TUNEL) a minimum of three technical replicas was performed within one experiment, minimum of 100 cells/subject was analyzed.

To achieve the lipid-deprived cell culture condition, cell culture grade fetal bovine serum (FBS) and BSA were delipidated using Cleanascite™ Lipid Removal Reagent (Biotech Support Group; Monmouth Junction, NJ, United States) according to the manufacturer’s protocol with modification. Briefly, Cleanascite™ was added to the FBS or BSA (1:4 volume ratio). The mixture was incubated for four hours at 4°C with gentle shaking and centrifuged at 16,000 g for 10 min. Supernatant was collected, then a second dose of Cleanascite™ Lipid Removal Reagent was added (1:4 volume ratio), and incubation and centrifugation steps were repeated as described above. The supernatants, consisting of lipid depleted FBS or BSA, were then used as cell culture media supplements to prepare lipid-depleted media.

Intracellular lipid detection was performed using a fluorescent probe for lipid droplets BODIPY 493/503 as previously described in Qiu et al. (33) with modifications. Briefly, cells were washed twice with PBS and then incubated with 2 μM BODIPY 493/503 staining solution (#D3922 Thermo Fisher Scientific, Chicago, IL, United States) for 15 min at 37°C. Then cells were fixed in 4% paraformaldehyde/phosphate-buffered saline (PBS), followed by DAPI staining to detect nuclei. Immunostaining was visualized and images were taken using a Keyence BZ-X800 (Keyence Corporation of America, Itasca, IL, United States) and Zeiss LSM700 confocal microscope (White Plains, NY, United States). A minimum of 100 cells per condition was analyzed.

Cell proliferation was assessed using Ki-67 immunostaining as described previously (10, 32, 34). Briefly, cells were washed with PBS, fixed in 4% paraformaldehyde/PBS, and permeabilized using Triton X-100/PBS solution. Then cells were incubated in blocking solution (2% BSA/PBS), followed by overnight incubation with primary anti-Ki-67 antibody (#9129, Cell Signaling, Danvers, MA, United States) in blocking solution. Next day, the slides were washed with PBS, followed by incubation with secondary chicken anti-rabbit IgG (H + L) Alexa Fluor 594 antibody (#A-21442, Invitrogen, Waltham, MA, United States) and 4′,6-diamidino-2-phenylindole (DAPI) to detect nuclei. Images were taken using a Keyence BZ-X800 microscope.

Cell growth analysis was performed as described previously (10, 32, 34). Briefly, 300,000 cells per well were plated in a six well plate in cell culture media supplemented with 5% FBS. Forty-eight hours later, when cells attached and spread (day 0), the media was changed to 0.1% BSA with or without lipids and incubated for six more days (media was renewed every 48 h). Then, cell counting was performed using the Countess II FL Automated Cell Counter (Thermo Fisher Scientific, Waltham, MA, United States) according to manufacturer’s protocol.

Apoptosis analysis was performed using the In situ Cell Death Detection Kit (Roche, Nutley, NJ, United States) based on terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) technology following the manufacturer’s protocol.

Immunoblot analysis was performed as described previously (10, 32, 34). Antibodies for ACLY (#4332), P-S79-ACC (#11818), ACC (#3676), FASN (#3180), P-S473-Akt (#4060), P-T450-Akt (#12178), Akt (#9272), PThr183/Tyr185-JNK (#4668), JNK (#9252), α/β-Tubulin (#2148), CPT1A (#97361), hexokinase II (HKII) (#2867), phosphofructokinase (PFKP) (#8164) were purchased from Cell Signaling (Danvers, MA, United States). The antibody for P-S455-ACLY (#PA5-97395) was purchased from Thermo Fisher Scientific (Chicago, IL, United States). The antibody for malonyl CoA decarboxylase (MLYCD) (#15265-1-AP) were purchased from Proteintech (Rosemont, IL, United States). Secondary HRP conjugated anti-mouse antibody (ab205719) was purchased from Abcam (Boston, MA, United States), secondary HRP-conjugated anti-rabbit antibody (#7074) was purchased from Cell Signaling (Danvers, MA, United States).

2-Deoxy-D-glucose (2-DG, D6134), IL-6 (I1395), PDGF-BB (GF149), and Akt inhibitor VIII (SIAL-124018) were purchased from Millipore Sigma (St. Louis, MO, United States), 5-tetradecyloxy-2-furoic acid (TOFA, sc-200653) was purchased from Santa Cruz Biotechnology (Dallas, TX, United States), JNK inhibitor (bentamapimod, HY-14761) was purchased from MedChemExpress (Monmouth Junction, NJ, United States).

To measure intracellular lactate amount, the lactate assay kit (#MAK064, Millipore Sigma) was used according to the manufacturer protocol with modifications. Briefly, 2 × 10⁶ cell were homogenized in 200 μL of lactate assay buffer and centrifuged at 21,000 g for 5 min to remove cell debris. The supernatant was deproteinized with a 10 kDa MWCO spin filter (#UFC5003, Millipore Sigma) to remove lactate dehydrogenase, and the lactate assay was performed. Protein concentrations were determined in the supernatants before deproteinization using the BCA protein assay kit (#23227, Thermo Fisher Scientific) and lactate content was normalized to the amount of total protein used for the assay.

Immunoblots, BODIPY 493/503, Ki-67 and apoptosis assays were analyzed using ImageJ (NIH, Bethesda, MD, United States), StatView (SAS Institute, Cary, NC, United States) and GraphPad Prizm 9.02 (GraphPad Software, San Diego, CA, United States). Immunocytochemical analyses were detected and captured using the Keyence BZ-X800 system and software (Keyence Corporation of America, Itasca, IL, United States) and Zeiss LSM700 confocal microscope and software (ZEN, 2009; White Plains, NY, United States). Statistical comparisons between the two groups were performed by Mann–Whitney U test. Statistical comparisons among three or more groups were performed by the Kruskal–Wallis rank test with Dunn pairwise comparison post hoc test. Statistical significance was defined as p ≤ 0.05.

In order to evaluate the status of lipogenesis in PAH PAVSMC, we first tested the expression of key lipogenic enzymes driving the biosynthesis of fatty acids. Immunoblot analysis demonstrated significant increase in phosphorylated ACLY, and elevated protein levels of ACC and FASN in early-passage distal human PAH PAVSMC compared to cells from non-diseased subjects (CTRL) (Figures 1A–D and Supplementary Figure 1A). Interestingly, we detected no significant differences between non-diseased and PAH PAVSMC in protein levels of CPT1A (Figures 1A,E) and MLYCD (Supplementary Figure 1B), the regulatory enzymes in fatty acid β-oxidation and synthesis, in spite of previously reported up-regulation of CPT1A in smooth muscle cells in rats with monocrotaline-induced PH and protective effect of MLYCD deletion against development of hypoxia-induced PH in mice (28, 35). Observed up-regulation of key fatty acid synthesis enzymes suggested that de novo lipid synthesis is altered in PAH PAVSMC. To confirm our observations, we performed immunocytochemical detection of lipid droplets (a major storage depot for neutral lipids, primarily triglycerides) using a fluorescent probe (BODIPY 493/503) to visualize neutral lipid accumulation in PAH PAVSMC. We found that lipid accumulation detected in PAH PAVSMC was preserved in media deprived from exogenous lipids (Figures 1F,G), indicating that PAH PAVSMC have an ability to generate lipids de novo. Importantly, PAH PAVSMC, in contrast to control cells, demonstrated increased growth not only in serum-deprived media, but in lipid-deprived serum-deprived media (Figure 1H). Moreover, 5-tetradecyloxy-2-furoic acid (TOFA), an allosteric inhibitor of ACC, a key enzyme regulating fatty acid synthesis, significantly decreased proliferation and induced apoptosis in PAH PAVSMC (Figures 1I,J), demonstrating that lipogenesis is required for PAH PAVSMC hyper-proliferation and survival. Taken together, these data show that de novo lipid synthesis is up-regulated and required for maintaining the pro-proliferative/pro-survival phenotype of human PAH PAVSMC.

Since fatty acid uptake is not altered in PAH PAVSMC (23) and we found that PAH PAVSMC growth was maintained in lipid-deprived media, we hypothesized that glucose could be a potential source for de novo lipid synthesis in PAH. We found that intracellular lipid content in PAH PAVSMC in lipid-deprived conditions was significantly depleted by the treatment with the non-metabolizable analog of glucose 2-DG. Importantly, co-treatment with glucose metabolite pyruvate significantly attenuated 2-DG-dependent inhibition of intracellular lipid accumulation in human PAH PAVSMC (Figures 2A,B). Furthermore, 2-DG-induced inhibition of PAH PAVSMC proliferation in lipid-deprived conditions was also partially reversed by pyruvate (Figure 2C). These data demonstrate that glucose, metabolized through the glycolysis, serves as the main source for de novo lipid synthesis and supports lipid accumulation and increased proliferation of PAH PAVSMC. Confirming up-regulation of glycolysis, intracellular lactate levels were significantly higher in PAH PAVSMC compared to non-diseased cells (Figure 2D). Moreover, protein levels of key glycolytic enzymes PFKP and HKII were significantly higher in human PAH PAVSMC compared to controls (Figures 2E–G), and HIF1α over accumulation was detected in six out of seven PAVSMC from PAH patients compared to one out of five non-diseased (control) subjects (Supplementary Figure 2). This is in good agreement with previously published data demonstrating the importance of glycolysis for PH (19, 36, 37). Together, our data demonstrate that glycolysis-metabolized glucose serves as a main source for increased de novo lipid synthesis in PAH PAVSMC, supporting a pro-proliferative cell phenotype.

One of the key regulators of glucose and lipid metabolism is a serine/threonine kinase Akt (38–42), which stimulates glucose uptake and glycolysis (43) and promotes de novo lipid synthesis through sterol regulatory binding protein (SREBP)-dependent expression of lipogenic enzymes (43, 44). Accumulating evidence from multiple groups, including ours, demonstrate pathological role of Akt as a regulator of pulmonary vascular cell metabolism, proliferation, pulmonary vascular remodeling, and overall PH (10, 45, 46). This allowed us to hypothesize that Akt might coordinate glucose metabolism and de novo lipid synthesis supporting PAVSMC hyper-proliferation in PAH. In agreement with published studies reporting Akt activation in PAH (10, 32), we found that S473-Akt phosphorylation was significantly increased in human PAH PAVSMC (Figures 3A,B). Importantly, Akt inhibitor VIII suppressed activation of both key lipogenesis enzymes, ACLY and ACC; ACLY deactivation was detected by decrease in S455-ACLY phosphorylation (47), and ACC deactivation was detected by an increase in inhibitory S79-ACC (48) phosphorylation (Figures 3C–F and Supplementary Figures 3A,B), suggesting that Akt supports increased lipogenesis in human PAH PAVSMC. Moreover, Akt inhibitor suppressed phosphorylation of ribosomal protein S6 (Figures 3C,G), the main downstream effector of mTOR complex 1 (mTORC1), a known activator of cell growth, proliferation, and lipid biogenesis (49). In summary, these data demonstrate that de novo lipid synthesis in PAH PAVSMC is regulated by Akt.

Next, we aimed to identify the factor(s) regulating de novo Akt-dependent lipid synthesis in PAH. Several studies demonstrate that lysine deacetylase SIRT7 coordinates cellular metabolic balance by regulating glucose sensing/homeostasis, glycolysis, mitochondria and ribosomal biogenesis, DNA damage response, fatty acid synthesis and overall lipid metabolism (50–56). Moreover, it was recently shown that SIRT7 modulates aortic vascular smooth muscle cell proliferation in wire injury model of the femoral artery (57) and promotes cancer progression by activating Akt (56, 58) and mTORC1 effector S6K1 (59). We found that SIRT7 protein levels were significantly higher in human PAH PAVSMC compared to non-diseased PAVSMC (Figures 4A,B). Importantly, shRNA-induced depletion of SIRT7 in PAH PAVSMC significantly reduced S473 Akt phosphorylation (Figures 4C,D) and activatory phosphorylation of ACLY, and significantly increased inhibitory phosphorylation of ACC in human PAH PAVSMC (Figures 4C,E,F), suggesting that Akt-mediated lipogenesis in human PAH PAVSMC is regulated by SIRT7.

Since SIRT7 does not possess kinase activity, its regulation of Akt phosphorylation status most likely is mediated by intermediate player(s) and one of the potential SIRT7 pro-survival targets is JNK (54). Indeed, we found that shSIRT7 significantly decreased phosphorylation of JNK in PAH PAVSMC (Figures 4C,G), also suggesting that SIRT7 regulates Akt and lipogenesis through JNK. Previous studies have shown that Akt activation via phosphorylation at S473 is achieved through a series of phosphorylation steps, and the initial priming phosphorylation of T450 can be regulated by JNK (60). We found a significant increase in T183/185 JNK phosphorylation in human PAH PAVSMC compared to control PAVSMC (Figures 5A,B), indicating that JNK is activated in PAH PAVSMC. JNK inhibitor HY-14761 significantly reduced both S473 and priming Thr450 Akt phosphorylation in PAH PAVSMC (Figures 5C–E). JNK inhibitor-mediated deactivation of Akt was associated with de-activation of lipogenic enzymes ACLY and ACC in human PAH PAVSMC (Figures 5C,F,G and Supplementary Figures 3C,D). These data indicate that Akt-mediated lipogenesis in human PAH PAVSMC is at least in part regulated by JNK.

In PAH, Akt activation in resident pulmonary vascular cells could be induced by excessive amounts of growth factors and pro-inflammatory mediators. To start determining pro-PH agents modulating SIRT7/JNK axis in PAVSMC, we treated non-diseased human PAVSMC with PDGF and IL-6. Interestingly, while both PGDF and IL-6 significantly increased JNK phosphorylation, only PDGF induced significant increase of SIRT7 protein levels (Supplementary Figures 4A,B), similar to those seen in human PAH PAVSMC. Together, these data suggest that SIRT7/JNK-dependent Akt activation in PAVSMC could be induced by PDGF.

To further clarify the role of Akt in lipid accumulation, hyper-proliferation, and survival of human PAH PAVSMC, we treated cells with Akt inhibitor VIII. Immunocytochemical analysis demonstrated that intracellular lipid accumulation in PAH PAVSMC, maintained in lipid-deprived media, was significantly downregulated by Akt inhibitor VIII (Figures 6A,B). Moreover, treatment with Akt inhibitor VIII significantly reduced proliferation and promoted apoptosis of PAH PAVSMC (Figures 6C,D), demonstrating that Akt supports increased cell proliferation, survival, and lipogenesis in PAH PAVSMC (Figure 6E). Furthermore, inhibition of Akt did not augment 2-DG-dependent inhibition of PAH PAVSMC proliferation (Supplementary Figures 5A,B), suggesting that Akt regulates PAH PAVSMC proliferation at the level of or downstream of glycolytic enzymes. Taken together, our data demonstrate that human PAH PAVSMC have up-regulated lipogenesis supported in an Akt- and glycolysis-dependent manner to sustain increased proliferation and survival, and that Akt signaling is regulated, at least in part, by SIRT7-JNK axis (Figure 6E).