This was a very interesting and exciting paper that I discovered today via Pulmonary Hypertension News: “Vascular stiffness mechanoactivates YAP/TAZ-dependent glutaminolysis to drive pulmonary hypertension”. I’m a large proponent of the hypothesis that metabolic dysregulation and immune dysfunction are key drivers in the development of pulmonary hypertension. This paper adds weight to my case that the former is most likely a key element in PH pathogenesis.

I’ve also always believed that due to the enigmatic nature of the disease, and presence of “cancer”, “autoimmune”, and “metabolic dysregulation” like features, solving PH can potentially solve problems like cancer and autoimmunity. A few examples from the paper as it pertains to cancer:

  • “Here, the identification of glutaminolysis as a mechanoactivated process coregulated with aerobic glycolysis advances our understanding of the regulatory hierarchy seen in the metabolic reprogramming in PH. Such an interface between stiffness and metabolism draws parallels to related reprogramming events proposed in tumors in relation to matrix remodeling.”
  • “Furthermore, it is likely that the metabolic actions of Hippo signaling extend to an even wider sphere of influence than vascular stiffness or PH alone. In the contexts of organ development and tumorigenesis, it is tempting to speculate on the master regulatory role of Hippo signaling in glutaminolysis and glycolysis as a primary mechanism to balance proliferative capacity with efficient energy production.”

Another exciting aspect of the paper is that some of the potential therapies investigated, which show promise based on their results, could be easy and relatively quick to implement: one of the drugs is already approved by the FDA, and the other is currently in clinical trials. This is key because drug approval is almost a decade long process.

Some terms:
PAEC = Pulmonary Artery Endothelial Cell
PASMC = Pulmonary Artery Smooth Muscle Cell
ECM = Extracellular Matrix

Summary, Key Points & Thoughts:

  • Overall goal of this study was to investigate whether an altered metabolism due to an increase in pulmonary vascular stiffness controls cellular proliferation in PH. The primary metabolic pathways studied were glycolysis and anaplerosis (in this case the major anaplerotic pathway examined was glutaminolysis). The aim of these results being “effectively linking vascular stiffness and metabolic dysfunction as 2 integrally related molecular drivers of [PH].”
  • While there is remodeling and proliferation of intima and media in PH, it is becoming increasingly recognized that the ECM plays a key role in driving the disease. According to the paper, a stiff ECM can drive cellular proliferation and survival via activation YAP and TAZ, transcriptional coactivators in the Hippo Signaling pathway, which essentially act as mechanosensors in the pulmonary vasculature. In addition, recently, they “found that pulmonary vascular stiffness activates YAP/TAZ early in PH, thereby inducing the miR-130/301 family to augment further ECM remodeling and cellular proliferation in vivo (1).”
  • It is just now being confirmed that, while hypoxia can induce PH, it is not necessary to induce PH. PH could technically be manifested by a dysregulation in metabolism [my emphasis added]: “Separately, aerobic glycolysis, a chronic shift in energy production from mitochondrial oxidative phosphorylation to glycolysis, has been described as a pathogenic driver of pulmonary arterial endothelial and smooth muscle proliferation and migration in PH (as reviewed by ref. 5). Prior mechanistic studies in PH related to this metabolic shift have historically relied on hypoxic disease modeling (6, 7). Yet, numerous forms of PH — subtypes linked to idiopathic or secondary conditions such as predisposing genetic mutations, congenital heart disease, scleroderma, and HIV infection, to name a few — are also characterized by profound metabolic dysregulation in the absence of obvious hypoxic injury. Data are only just emerging (8) regarding the molecular regulators of metabolic dysfunction operating independent of outright hypoxic stress in PH.”
  • According to the paper in reference to proliferating intimal and medial vascular cells, “increased glycolysis alone is insufficient to meet the total metabolic demands of such proliferating cells. The tricarboxylic acid (TCA) cycle also serves as a source of energy production and provides a critically important reservoir of substrates for the biosynthesis of amino acids, carbohydrates, and lipids (11). Continued functioning of the TCA cycle requires the replenishment of carbon intermediates. This replenishment, or anaplerosis, is accomplished via 2 major pathways: glutaminolysis (deamidation of glutamine via the enzyme glutaminase [GLS1]) and carboxylation of pyruvate to oxaloacetate via ATP-dependent pyruvate carboxylase (PC). Specifically, glutaminolysis via GLS1 activity contributes to anaplerosis by allowing for mobilization of cellular energy, carbon, and nitrogen, particularly in rapidly proliferating cells (12), and serves as a critical process in transformed cells that have switched their metabolism from oxidative phosphorylation to glycolysis in order to maintain cell growth and viability (13). The particular ability of glutaminolysis (and/or pyruvate carboxylation) to support aspartate production for direct induction of proliferation has recently been reported in malignant cells (14, 15).” As a side note, I’ve been considering the pros and cons of oxaloacetate as a supplement for PH. However, this quote is perhaps evidence of oxaloacetate being contraindicated.
  • Their data from cultured PAEC’s and PASMC’s showed that “stiff conditions act as a mechanical stimulus to increase glycolysis and decrease mitochondrial oxidative phosphorylation.” Furthermore, they also observed an increase in lactate/pyruvte ratio, decreased succinate levels and increased lactate production, decreased intracellular glutamine and significant increases in glutamate and aspartate, which is “consistent with a putative anaplerotic process accompanying accelerated glycolysis.” Also, the “[l]evels of 3 key enzymes in PAECs — lactate dehydrogenase A (LDHA), both GLS1 isoforms (KGA and GAC), and PC — implicated in both glycolysis (LDHA) and anaplerosis (GLS1 and PC) were elevated.” In conclusion, “exposure to stiff matrix not only alters glycolysis and oxidative phosphorylation but also controls anaplerotic replenishment of amino acids.”
  • “Taken together, YAP and TAZ are integral to the mechano-triggered, glycolytic and glutaminolytic metabolic reprogramming events initiated by ECM stiffness.” Knockdown (or removal) of YAP and TAZ resulted in a “decreased extracellular lactate and lactate/ pyruvate ratio… YAP/TAZ knockdown also blunted the effects of stiff ECM on intracellular glutamine, glutamate, and aspartate.” Furthermore, “[m]itochondrial membrane potential was also sustained during YAP/TAZ knockdown in stiff matrix.” In a non-stiff ECM, expressing YAP “increased extracellular lactate and lactate/pyruvate ratio… decreased glutamine and increased glutamate and aspartate… and consequently, decreased mitochondrial membrane potential.” Finally, they observed that “only knockdown of YAP and TAZ together, but not YAP or TAZ alone, was sufficient to decrease the target gene expression” of LDHA, GLS1, and PC. This suggests you need the activation of both YAP and TAZ in order to observe changes in metabolism at the cellular level.
  • Next, it was found that “[i]ncreased GLS1 expression and glutaminolysis are critical for sustaining glycolysis and cell proliferation in a stiff environment.” When inhibiting GLS1, they noticed that “PAECs blunted the stiffness induced processes of glutamine consumption, glutamate production, and aspartate production… GLS1 inhibition also decreased glycolysis in stiff matrix, as indicated by decreased extracellular lactate and lactate/pyruvate ratio…”. They noticed similar changes in metabolic activity when inhibiting GLS1 in PASMCs. Another interesting observation they found was that “GLS1 inhibition had a negligible effect on apoptosis and on proliferation in soft matrix but blunted proliferation on stiff matrix in PAEC.”
  • They also found that the products of the actions of the GLS1 enzyme are responsible for sustained cellular proliferation. For example, glutamate or aspartate supplementation was administered to cells lacking GLS1 or YAP/TAZ, and after doing so, they noticed a decrease in proliferation in either PAECs or PASMCs. Importantly, in these cells, “cellular proliferation was at least partially restored by glutamate and more fully restored by aspartate supplementation. Aspartate supplementation similarly reversed the reduced cell migration of GLS1- or YAP/TAZ-deficient PAECs on stiff matrix… Collectively, these results demonstrate that GLS1 and its control of glutamate and aspartate production by glutaminolysis are essential for metabolic reprogramming and consequent vascular cell proliferation and migration specific to stiff matrix exposure.” This also perhaps makes a case for at least avoiding dietary supplementation of these amino acids.
  • Next, moving up from cultured cells, they found that the YAP/TAZ–GLS1 axis activated glycolysis and glutaminolysis in vivo in both animal and human PH PAECs and PASMCs exposed to vascular stiffness: “From these rats, CD31+ endothelial cells were isolated from lungs 3 weeks after exposure to vehicle or monocrotaline… Consistent with our observations of anaplerosis in cultured PAECs grown on stiff matrix… glutamine was decreased…and aspartate was increased… Notably, no significant change in glutamate concentration was observed in these cells… which may suggest an elevated glutamate turnover in pulmonary cells in vivo. They also observed a decrease of succinate which is indicative of a decrease in TCA cycle activity, as well as an increase in the ratio of lactate/pyruvate. A significant increase in the expression of GLS1, LDHA, and PC was also observed.
  • Using microscopy techniques, they also analyzed the monocrotaline-induced PH in rats in situ and “consistent with previous theories of endothelial apoptosis in PH… found an early yet temporary induction of endothelial apoptosis as reflected by cleaved caspase-3 in situ staining and by caspase-3/7 activity (days 0–3 after monocrotaline injection)… This was followed by a subsequent decrease of apoptosis and an increase of smooth muscle and endothelial cell proliferation… correlating with an increase of both splice isoforms (KGA and GAC) of vascular GLS1 expression… Taken together, and consistent with our in vitro findings, these results demonstrated that, following vascular injury and just after an early wave of endothelial apoptosis, the development of pulmonary vascular stiffness and glutaminolysis follows the same kinetics as the increase of proliferation of diseased endothelial and smooth muscle cells in vivo.” Also, in order to “determine whether glutaminolytic reprogramming is an active process in human PAH to sustain pulmonary vascular cell proliferation… a cohort of human patients with PAH (n = 13) stemming from causes ranging from idiopathic and hereditary etiologies to scleroderma… as compared with non-PAH subjects (n = 6) who died from traumatic or unrelated causes” was studied. “Correlating with increased periarteriolar collagen remodeling in PAH cases… a concurrent upregulation of GLS1… PC, and LDHA… was observed in both CD31+ (endothelial) and α-SMA+ (smooth muscle) cells…”. They also found that “in subjects with particularly high pulmonary arterial pressures (mean pulmonary pressure >45 mmHg), lactate/pyruvate ratio was elevated reflective of increased glycolysis, while glutamine/glutamate ratio was decreased and aspartate was increased, indicative of upregulated glutaminolysis and anaplerosis, in comparison with non-PH individuals (mPAP <25 mmHg; Figure 8, E–G). Together, these results support the notion that vascular stiffening activates YAP/TAZ in order to induce a glutaminolytic metabolic switch and vascular proliferation in PAH across both rodent and human instances of disease in vivo.”
  • It was also exciting to read that the “YAP/TAZ–GLS axis induces glycolysis and glutaminolysis in primates with SIV-PAH and in people with HIV-induced PAH.” Exciting because they found these results in a primate model of PH that closely resembles the human form of PH (other animal models of PH do not accurately reflect the pathological phenotype of human PH).
  • They also found that “modulation of pulmonary vascular stiffness and YAP/TAZ dependent mechanotransduction regulates glutaminolysis and PH manifestation in vivo”. In order to test if ECM remodeling and YAP/TAZ modulate cell metabolism in vivo, they tested whether or not the alteration of YAP/TAZ controls both glutaminolysis and PH development in the monocrotaline rat model of PH: “First, using a known pharmacologic inhibitor (β-aminopropionitrile, BAPN) of lysyl oxidase (Lox), the enzyme responsible for collagen cross-linking and consequent matrix stiffening, we determined whether inhibition of ECM stiffening could prevent the metabolic changes and downstream PH manifestations observed in monocrotaline exposed rats… BAPN treatment decreased pulmonary Lox activity and consequent periarteriolar ECM stiffening, as assessed by atomic force microscopy… In line with previous reports (25), a trend toward decreased systemic mean arterial pressure was observed with BAPN treatment… but without adverse effects on left ventricular cardiac function or heart rate… Consistent with our in vitro results, reduction of ECM stiffening by BAPN led to a decrease of YAP and GLS1 expression… YAP-dependent gene expression… and downstream GLS activity, as reflected by direct enzymatic activity measurement… Such metabolic effects further decreased vascular endothelial and smooth muscle proliferation, as reflected by in situ arteriolar staining… and ameliorated hemodynamic and histologic manifestations of PH, as measured by vascular remodeling and muscularization… and right ventricular systolic pressure (RVSP)…”. Also, verteporfin, an inhibitor of YAP, was found to decrease pulmonary arteriolar stiffness. In sum, these data show that in vivo, the “ECM stiffening relies on YAP/TAZ-specific mechanotransduction in order to induce pulmonary vascular glutaminolysis and anaplerosis, proliferation, and PH.”
  • Inhibiting GLS1 induced glutaminolysis “decreases pulmonary vascular cell proliferation in vivo and ameliorates PH.” To test if glutaminolysis itself is essential for the pulmonary vascular proliferation process, they “administered 2 separate pharmacologic inhibitors of GLS1 — C968 (19) and CB-839 (28) — in monocrotaline-exposed rats using either a disease prevention… or a disease reversal… dosing protocol. In both cases, C968 and CB-839 treatments decreased GLS activity in whole rat lung as compared with control… without adverse effects on left ventricular function, heart rate, or systemic blood pressure… Correspondingly, C968 and CB-839 both decreased the presence of proliferation markers (PCNA+ or Ki-67+) in CD31+ or vWF+ (endothelial) and α-SMA+ (smooth muscle) pulmonary arteriolar cells in comparison with control PH rats… without effect on pulmonary arteriolar cell apoptosis…” Both C968 and CB-839 were found to significantly decrease pulmonary arteriolar remodeling, muscularization, RVSP, and right ventricular remodeling. “Taken together, these results directly implicate GLS1 and glutaminolysis, a process dependent on ECM stiffening, as critical metabolic mediators necessary for sustaining pulmonary vascular proliferation in PH.”
  • In their study, they used “multiple independent markers of endothelial lineage (CD31 and vWF) and proliferation (Ki-67 and PCNA)” to obtain results and draw their conclusions. This adds weight to their conclusions, since previously, single markers of proliferation or endothelial cell lineage were used to analyze diseased pulmonary vessels, which yielded results that were challenging to interpret. Using this new method, their results were “consistent with the model of spatiotemporal balance of PAEC apoptosis and proliferation, originally described by Voelkel and colleagues (33) and echoed by others (34).” They found that the “YAP/TAZ–GLS1 activation, glutaminolyis, and proliferation in PAECs closely followed an initiating wave of injury and PAEC apoptosis. Such proliferation may feed into increased PAEC turnover at earlier stages of PH… at which a balance of PAEC apoptosis and proliferation was discerned. However, during later stages of PH when upregulation of YAP and GLS1 was persistent, markers of proliferation were also more evident. As reported by others (35), PAEC apoptosis cannot be ruled out at later disease time points. However, our data indicate that these apoptotic events must be occurring in cells other than the prevalent proliferative component driven by vascular stiffness and glutaminolysis…”. It is known that PAEC apoptosis followed by outgrowth of proliferative PAECs is one of the factors in pathogenesis of PH. However, it could also be that there is a continuous PAEC apoptosis followed by outgrowth of proliferative PAECs. That is, either it is an initial insult causing death to PAEC followed by a proliferation of PAECs, or the cycle of apoptosis and proliferation continues in PH. This is important to differentiate between because therapies that target the induction of apoptosis in the pulmonary vasculature may not be beneficial if it is the case that there is continuous PAEC apoptosis taking place which is the driver of the disease.
  • “Even in settings where the diseased endothelial layer is not overgrown, it is an intriguing possibility that hyperactivated and glutaminolytic PAECs are reprogrammed for pathophenotypes in addition to proliferation, including endothelial-to-mesenchymal transition — a process that has been directly connected to PH (36, 37) and where further proliferation could allow for endothelial cells to feed into medial hyperplasia.”
  • Their findings also revealed a promigratory phenotype that is promoted by matrix stiffness and the YAP-GLS1 axis. Disorders in this migratory process may explain the abnormal angiogenesis seen in PH.
  • The therapeutic options discussed in the study are promising. With Lox inhibitors, “when coupled with targeting of the downstream YAP/TAZ–GLS1 axis, an additive or perhaps synergistic therapeutic benefit may emerge in inhibiting pulmonary vascular proliferation and remodeling. This may be particularly evident with YAP, given the emerging beneficial effects even beyond metabolism of altering Hippo signaling in the pulmonary vasculature (1, 48) and the robust improvement of severe rodent PH using the YAP inhibitor verteporfin alone… Importantly, verteporfin is already approved by the US FDA as an intravenous medication for the treatment of age-related macular degeneration (49). Moreover, an oral form of the GLS1 inhibitor CB-839 is undergoing active clinical development in early human clinical trials for cancer therapy (Clinical Trial NCT02071862). Therefore, a repurposing of verteporfin and CB-839 for treatment of PH, either separately or together with possible optimization for tissue-specific delivery, presents a rare opportunity to offer novel therapeutics in this disease without the delay of decades to develop small-molecular inhibitors de novo.”

Other things learned:

  • Pharmacologic targeting of vascular ECM can improve PH, but in animal models
  • A new model of PH discovered! “Previously, a nonhuman primate model of HIV-induced PAH was described in rhesus macaques infected with SIV (23). Importantly, such a model replicates the hemodynamic and histologic manifestations of PAH. It also displays an incomplete penetrance with 50% to 60% of infected macaques developing PAH, thus consistent with the incomplete penetrance of PAH with HIV infection in humans.”
  • As expected, matrix stiffness corresponds to a decrease in compliance of the pulmonary arteries in response to the administration of vasodilators: “Analysis of invasive hemodynamic data of HIV-PAH subjects revealed a significant decrease of pulmonary arterial compliance (Figure 10A) consistent with an increase of pulmonary artery stiffness in comparison with HIV-infected, non-PAH individuals.”
  • Metabolic remodeling also may affect adventitia: “Separately, in adventitial fibroblasts, we recently described a YAP/TAZ–miR-130/301 feedback loop whereby matrix stiffening spreads through pulmonary vasculature and perhaps even pulmonary parenchyma via mechanoactivation of naive fibroblasts that contact stiffened matrix (1). Given the current findings implicating YAP/TAZ activation with glutaminolysis, it is possible that glutaminolysis and anaplerosis in fibroblasts are also inherently linked to the control of matrix stiffening and remodeling.”
  • AMPK activation is relevant in Hippo signaling: “Moreover, cellular energy status has been implicated as a potent regulator of YAP/TAZ activity either through AMP-kinase activation (9, 10), induction of aerobic glycolysis (40), or mevalonate metabolism.”
  • “Increased glutamine utilization and metabolism have been observed in the right ventricles of rats and humans with PH (16). Cardiac glutaminolysis, however, appeared to be driven by microvascular ischemia rather than through mechanosensation.”


  • Is glutaminolysis pathological in PH? Possible answer: in the lung vasculature, yes, but in other organs such as the heart, maybe the case is different. Or maybe in the heart, it is just a response to ischemia, as mentioned above.
  • Why does mechanical stiffness induced metabolic changes?
  • And one last important question, generously proposed by Dr. Ron Oudiz from the Liu Center for Pulmonary Hypertension in response to this post: “does the abnormal metabolism cause the stiffness or vice-versa?” It still isn’t clear which causes which…

Leave a Reply

Your email address will not be published. Required fields are marked *