The Mechanobiology of Vascular Remodeling in the Aging Lung

Abstract

Aging is accompanied by declining lung function and increasing susceptibility to lung diseases. The role of endothelial dysfunction and vascular remodeling in these changes is supported by growing evidence, but underlying mechanisms remain elusive. In this review we summarize functional, structural, and molecular changes in the aging pulmonary vasculature and explore how interacting aging and mechanobiological cues may drive progressive vascular remodeling in the lungs.

Introduction

The aging lung is susceptible to several of the most common and debilitating pulmonary diseases, including emphysema, idiopathic pulmonary fibrosis, and group 2 pulmonary hypertension. Aging-associated endothelial dysfunction and vascular remodeling have been increasingly found to play roles in organ dysfunction in the systemic vasculature but have been less well studied in the pulmonary circulation. In this review, we discuss what is known about clinical, structural, and molecular changes in the aging pulmonary vasculature, how these interact via mechanobiological feedback, and the downstream consequences for pulmonary disease pathogenesis. We highlight the current knowledge gaps in this field and the promise of further research in this area for modifying and reversing aging-associated pulmonary disease.

Clinical Assessment and Consequences of Aging-Associated Pulmonary Vascular Remodeling

Age-related changes in the systemic arterial circulation in both large elastic arteries and the microcirculation contribute to cardiovascular diseases (1). These changes precede development of overt disease and can occur in the absence of risk factors (2), suggesting that they correlate with age rather than disease. Characteristic changes with normal aging include increasing stiffness of large arteries, progressive endothelial dysfunction [loss of nitric oxide (NO)-dependent dilation] in small and conduit arteries, and impaired angiogenesis (3). These vascular changes correlate with incident systemic hypertension (4–6), left ventricular remodeling (7–9), and microvascular end-organ damage of the brain and kidney (10–14). Given these effects, it is unsurprising that systemic arterial stiffness is a significant independent predictor of cardiovascular mortality (15–19).

Although much less examined than the systemic circulation, the pulmonary vasculature appears to be subject to many of the same aging-related phenotypes. Clinical study of the normal aging pulmonary circulation in humans has been challenging because of the morbidity of invasive measurements of the lung; multiple strategies have been used (FIGURE 1). Invasive pulmonary artery catheterization of nondiseased subjects has demonstrated an increase in systolic and mean pulmonary arterial (PA) pressure with aging (20–23), accompanied by evidence of increased vascular stiffening (24). A large, population-based study by Lam et al. (25) using noninvasive echocardiography found that PA systolic pressure showed a linear correlation with age (r = 0.31), and that the increase in pressure closely mirrored that seen in the systemic circulation. Although the absolute increase in pressure seen in the pulmonary circulation is smaller than that seen in the systemic circulation with age, the relative increase is similar (25) or larger (21–23) and is accompanied by a larger relative increase in vascular resistance (21, 22).

FIGURE 1.

Pathophysiological features and assessment modalities for vascular remodeling in the pulmonary arterial tree This schematic of the pulmonary arterial system illustrates its branching structure, featuring large-diameter elastic arteries, smaller muscular arteries and arterioles, and extensive microcirculation. Unlike in the systemic circulation, the high compliance of the pulmonary arterial tree distributes across the entire vascular bed, contributing to impedance of the system as a whole and subsequent impact on the right ventricle (RV). The impact on the entire circulation is largely assessed invasively with flow and pressure sensors. These measurements can also be modeled based on invasive studies in humans and animals. Large arteries are most susceptible to alteration in elastin/collagen matrix and can be directly imaged with multiple modalities; reliable noninvasive physiological assessment has been technically challenging but is showing increasing feasibility with the latest technological advances. Small arteries are most susceptible to smooth muscle cell (SMC) hypertrophy and changes in tone, which can be assessed ex vivo with atomic force microscopy or examined histologically. The microcirculation suffers rarefaction and impaired angiogenesis with remodeling that can be best studied with micro-computed tomography (CT) or high-throughput, machine-assisted histology. CTA, computed tomography angiography; Dl_CO, diffusion capacity for carbon monoxide; mPAP, mean pulmonary artery pressure; MRA, magnetic resonance angiography; PFTs, pulmonary function tests; PP, pulse pressure; PVR, pulmonary vascular resistance; SV, stroke volume. Figure modified from Ref. 41, with permission from Frontiers in Physiology.

By increasing blood flow and oxygen demand, exercise exaggerates the hemodynamic changes seen with normal aging. Although mean PA pressure increases in all age groups with exercise, the elderly have an increase out of proportion to cardiac output and oxygen consumption compared with younger cohorts (26). Using a vascular distensibility model derived from large-animal studies, Reeves et al. (27) analyzed human exercise data and identified a decrease in vascular distensibility (1.5% vs. 2% change/mmHg) in older (age 63–81 yr) versus younger (age 14–60 yr) subjects. Age-related vascular remodeling therefore appears to require higher driving pressures in order to appropriately accommodate blood flow and recruit lung units during exercise (28).

There is strong evidence that vascular stiffening in the pulmonary circulation is associated with morbidity and mortality. In one community-based sample, elevated PA systolic pressure (PASP) was strongly associated with increased mortality, particularly for those patients with PASP > 30 mmHg or without known cardiopulmonary disease (25). Leibowitz et al. (29) showed similar findings in a separate community-based cohort of patients over the age of 85 yr. Patients with aging-related pulmonary diseases, such as chronic obstructive pulmonary disease (COPD) or interstitial lung disease, have dramatically increased mortality in the setting of pulmonary artery pressures elevated to this extent (30–33). Likewise, in cohorts of patients meeting criteria for pulmonary hypertension, vascular stiffness correlates with functional status and mortality (34–39). Several of these studies showed that stiffness measures were better predictors of mortality than typically measured resistance indexes (35, 38, 39), with particular prognostic utility in patients with normal vascular resistance (34, 38, 39). This evidence strongly links pulmonary vascular remodeling and stiffening to adverse clinical outcomes and has led multiple research groups to assess the therapeutic potential of targeting and reversing this process (40, 41).

Mechanobiological Changes in the Aging Pulmonary Vasculature: Linking Remodeling to Function

Extracellular Matrix Remodeling

Detailed structure-function studies of pulmonary vascular mechanics have been conducted in animal models and ex vivo human tissues (42). These studies have informed our growing understanding of the critical links between vascular structure, composition, stiffness, and function (41). As in the systemic circulation, the pulmonary vasculature transitions from large-diameter conduit arteries through a branching structure that ultimately reaches the alveoli as a delicate capillary network. The changes in pulmonary vascular mechanics that most influence right heart function and health are concentrated in the pulmonary arterial circulation; hence we focus here on these arterial vessels (FIGURE 1). Unlike in the systemic circulation, the compliance of the pulmonary arterial tree is distributed across the entire vascular bed rather than concentrated in the most proximal vessels (43); thus we explore remodeling as it manifests along the entirety of the pulmonary arterial vasculature.

Pulmonary arteries, like those of the systemic circulation, are composed of intimal, medial, and adventitial layers that together generate complex viscoelastic material properties. Mechanical testing studies on systemic and pulmonary arteries in dogs (44), sheep (45), and pigs (46) have emphasized their anisotropic (directional) nature. Much of this anisotropy is associated with the alignment of extracellular matrix (ECM) proteins, chief among them collagen I, which is a highly rigid molecule typically formed in highly aligned bundles to provide strength to mechanically loaded tissues. Interspersed with collagen bundles in the vascular wall are networks of elastin, an intrinsically disordered polymer possessing extraordinary compliance and ability to undergo reversible deformations (elasticity). Together these matrix components contribute to the nonlinear stiffening of vessels as loads initially borne by elastin are gradually taken on by the more rigid collagen fibers (47). Adding to their complexity, vessels have both passive mechanical properties that reflect their structure and composition as well as an active component largely defined by the number of actomyosin cross bridges dynamically engaged in the vascular smooth muscle cells that line the large arteries/arterioles of the pulmonary circulation. Changes in vessel mechanics can thus reflect changes in the proportion and molecular and cellular composition of vessels as well as the cellular activation state (tone) within the muscular layer (FIGURE 2).

FIGURE 2.

Aging and mechanobiological signals intersect to regulate vascular remodeling and pulmonary vascular mechanics The schematic cross section of a pulmonary vessel highlights known aging-related changes in inflammation, senescence, and extracellular matrix remodeling in the vessel wall. On left are shown interacting aging mechanisms of senescence, inflammation, and mitochondrial and endothelial dysfunction that together influence pulmonary vascular biology. These mechanisms engage in reciprocal interactions with the mechanobiological cellular responses to altered shear stress, stiffness, pulsatile flow, and vascular distension shown on right. Within each of these domains are shown key pathways that are increased (green) or decreased (red) in response to aging and mechanobiological cues and that link these interacting mechanisms to each other and to vascular biology. Both sets of pathways are self-reinforcing positive feedback loops containing reciprocal interactions and lacking a single initiating pathogenic mechanism. Together the interaction of aging-related and mechanobiological cues drives structural and function alterations in the pulmonary vasculature, including extracellular matrix (ECM) remodeling, increased vascular tone, and microvascular dysfunction. Over time these changes combine to alter local and systemwide pulmonary vascular mechanics and function. eNOS, nitric oxide synthase 3 or endothelial NOS; MMPs, matrix metalloproteinases; MRTF, myocardin-related transcription factor; NF-κB, nuclear factor-κB; NRF2, nuclear factor-erythroid factor 2-related factor 2; ROS, reactive oxygen species; SMADs, small Mothers Against Decapentaplegic; TAZ, transcriptional coactivator with PDZ-binding motif; TGF-β, transforming growth factor beta; TNF-α, tumor necrosis factor alpha; YAP, yes-associated protein.

Aging gradually remodels components of both the systemic and pulmonary vasculature (40, 48). The vessels of the systemic arteries exhibit gradual thickening of their walls, loss of elastin, and increased diameters of conducting arteries with age. Elastin has a remarkable half-life of ∼40 yr (49) but is mainly synthesized during development and is gradually lost with age (48, 50). Elastin-degrading metalloproteinases increase in the circulation and vessel walls with aging and may contribute to elastin loss (51, 52). Although less studied, similar aging-dependent changes, such as elastin degradation and fiber disorganization, are seen in the aging pulmonary vasculature as well (53). Similar findings are seen along with loss of pulmonary vascular compliance in pulmonary hypertension (54, 55). In this setting, absence or disorganization of elastin leads to earlier loading of collagen in the vessel wall, thereby altering wave propagation down the vascular tree and increasing right ventricular (RV) pressures and hypertrophy (56).

In contrast to elastin, collagen has a shorter half-life but is continuously produced and degraded throughout life (57). Aging increases the formation of nonenzymatic advanced glycation end products (58) that stabilize collagen, leading to loss of collagen’s elasticity and loss of vascular wall compliance (56, 59). Excess collagen deposition and accumulation are key sources of early decreases in vascular distensibility in experimental models of pulmonary hypertension (74), although hilar vessels demonstrate increased stiffness with aging without increased collagen content (53). On a molecular level, decreased nitric oxide (NO) bioavailability with aging (see below) increases the export of transglutaminase 2 to the extracellular matrix (60), where it enhances matrix stiffness by cross-linking collagen, fibronectin, fibrinogen, and laminin complexes. Alterations in matrix metalloproteinases (61), NO bioavailability (61, 62), and oxidative stress (63, 64) associated with aging have been shown to activate vascular transforming growth factor (TGF)-β signaling in both the lung and systemic circulation. TGF-β signaling, in turn, increases collagen I and collagen III deposition (65). In the systemic circulation, this gradual loss of elastin and replacement with collagen contribute directly to the decrease in aortic compliance seen starting in late middle age (50, 66).

Measures of vascular wall mechanics have traditionally emphasized quasi-static measures of deformability. Less emphasized, but likely important, are the viscoelastic (time/rate dependent) aspects of vessel mechanics. Such rate-dependent characteristics are essential to the dissipation or propagation of energy along the vascular wall that occurs with normal pulsatile loading. Changes in viscoelastic tissue mechanics with aging likely reflect altered interactions of collagen and elastin with highly charged proteoglycans distributed throughout the interstitial spaces of the vessel wall (61, 67). However, the contributions of proteoglycans (and other ECM components) to vascular viscoelasticity are less well defined than those for collagen and elastin, requiring further investigation.

Vascular Tone

Smaller muscular arteries and arterioles lack elastin and are composed of mostly smooth muscle by mass; consequently, their vessel wall mechanics are dominated by smooth muscle tone. These smaller vessels still contribute to the compliance of the pulmonary vascular tree because of the high capacitance of the pulmonary circulation. Aging-related increases in resting vascular tone have been well studied in the systemic circulation and arise primarily from impaired endothelium-dependent vasodilation (68–70). At baseline, the vascular endothelium produces NO in response to external stimuli such as acetylcholine and physiological levels of shear stress, leading to smooth muscle relaxation via NO-mediated activation of guanylate cyclase. Aging-associated pathology is felt to arise from NO scavenging by increased reactive oxygen species (ROS) that accumulate with aging (71). In animal models, inhibition of NO production can mimic aging-associated increases in arterial stiffness (72), and prolonged treatment with ROS scavengers in mice can reverse it (73). In human trials, high-dose vitamin C antioxidant infusions have similarly been shown to partially reverse age-associated declines in endothelium-mediated vasodilation (74, 75). Although less well studied, the pulmonary circulation in aged rats also shows impaired endothelium-mediated relaxation and increased oxidative stress. This was associated with high levels of ROS-generating enzymes and could be reversed through inhibition of ROS production (76).

In addition to altering NO-mediated relaxation, there is evidence that ROS can directly impact vascular smooth muscle contractility. In pulmonary arterial smooth muscle cells (PASMCs), increased ROS closely couples to increased free cytosolic Ca²⁺ levels, leading to increased tone and contractility. Moreover, catalase, antioxidants, and pharmacological inhibitors of potential ROS sources reduced or prevented hypoxia-induced elevation in intracellular calcium concentration ([Ca²⁺]_i) in PASMCs (77, 78). ROS are also generated by smooth muscle cells upon exposure to vasoconstrictors such as angiotensin II or endothelin, leading to feedback loops that amplify vascular tone. ROS production and TGF-β signaling likewise mutually cross-activate (63, 64), leading to enhanced vascular stiffening and fibrosis (79). Finally, ex vivo application of cyclical stretch to mimic hypertension from arterial stiffening elevates ROS in both endothelial and smooth muscle cells (80, 81). These findings implicate high-pulsatility flow in mechanobiological remodeling of the pulmonary vasculature through endothelin, TGF-β, and NF-κB signaling (83, 84). NF-κB further promotes recruitment of inflammatory cells that amplify local oxidative stress. These multiple levels of positive feedback involving arterial stiffness, oxidative stress, contractility, and inflammatory signaling are likely critical drivers of aging-associated vascular remodeling (see FIGURE 2).

Impaired Angiogenesis and Microvascular Rarefaction

The aging vascular system tends toward rarefaction and impaired microvascular angiogenesis. Extensive mapping of the vascular and stromal networks of young versus aged organs identified marked vascular attrition in multiple organ beds including the heart, kidney, and spleen (85). These vascular changes preceded markers of accumulating ROS and cellular senescence. Endothelial signatures from organs with rarefaction compared to those without showed a strong association with increased inflammatory cytokine signaling in the endothelium as a cause or consequence of this process. Under in vitro conditions, microvascular endothelial cells derived from aged rodents exhibit impaired angiogenesis in response to VEGF and other growth factors compared with cells derived from young animals (86). Endothelial NO synthesis is involved in angiogenic responses (102), as demonstrated most clearly by significant microvascular rarefaction with long-term endothelial nitric oxide synthase (eNOS) inhibition (103) or absence of eNOS expression (104). Similarly, endothelium-specific sirtuin-1 knockout mice develop capillary rarefaction, reduced exercise-induced neovascularization, impaired endothelium-dependent vasorelaxation, and premature vascular senescence (87). Although Sirtuin1 is involved in several aspects of aging and senescence, it is well known to regulate the transcription and enzymatic activity of eNOS in the endothelium (88, 89), a likely primary driver of this phenotype.

Recent work by Caporarello et al. has extended this concept to the pulmonary vasculature. They documented that young mice recover from bleomycin lung injury with increased capillary density and repair, whereas aged mice exhibit capillary rarefaction and nonresolving lung fibrosis (90). Nos3, which codes for eNOS, was transiently upregulated only in young mice, and young mice deficient in eNOS recapitulated the capillary rarefaction and fibrotic phenotype seen in the aged rodents. This rarefaction in the vasculature may result from impaired blood flow delivery by remodeled upstream vessels, but the temporal sequence is not well understood. Whether the rarefaction itself contributes to vascular resistance or stiffening is also not a resolved question. As a low-resistance system, pulmonary vessels of small size contribute more to vascular resistance and capacitance than in the systemic circulation. Even in the systemic circulation, simulated rarefaction of distal A3 and A4 arterioles (8- to 20-µm-diameter vessels) modestly augments the global resistance of the vascular bed (91). In the lung, the consequences may be more severe given that the organ cannot reduce its perfusion and must accommodate the entirety of the cardiac output.

In summary, age-associated alterations in matrix composition in the large arteries, vascular tone in muscular arteries and arterioles, and rarefaction in the microvasculature may all contribute to vascular stiffening, increased pressures, and high-pulsatility flow in the pulmonary vasculature. These mechanical changes are sensed by pulmonary vascular cells locally through a process known as cellular mechanotransduction. We and others have studied mechanotransduction in response to vascular stiffening in the pulmonary circulation (83, 84, 92, 93, 94, 95, 97, 99–106). Increases in both flow pulsatility and matrix stiffening can activate pathways leading to matrix deposition, likely resulting in further vascular stiffening. The molecular details of these pathological feedback loops are reviewed elsewhere (41, 101, 105, 107, 108). In general, upstream mechanical cues are integrated by the cell via cell membrane- and cytoskeleton-associated signaling into a transcriptional program mediated by cytosolic-nuclear protein shuttles such as MRTF, YAP/TAZ, NF-κB, β-catenin, and Smads. Several of these pathways are also activated by growth factors, cell-cell signaling, and other nonmechanical cues, thereby integrating biological and mechanical signals. Although some mechanotransduction pathways have been found to be activated by age and stiffness in the systemic circulation (61, 109, 110), mechanobiological feedback is presently understudied in the context of lung aging. As key mediators of the cellular machinery driving vascular remodeling, such pathways may be useful targets for intervention.

Prevention and Reversal of Aging-Associated Pulmonary Vascular Remodeling

There has been great interest in reversing aging phenotypes in the vasculature, particularly given that the prominent “vascular-centric theory of aging” posits vascular dysfunction as the primary driver of aging-related pathology in the brain, kidney, cardiovascular, and endocrine systems (3, 111). Approaches to altering vascular remodeling directly through matrix modification have been pursued but have so far failed to translate into clinical practice. Elafin, an endogenously expressed elastase inhibitor, has been found to reduce elastase and metalloproteinase activity, leading to improvements in pulmonary vascular remodeling in multiple pulmonary hypertension models (112, 113). Synthetic elastase inhibitors have faced multiple challenges over several decades, with continued efforts ongoing to translate elastase inhibition into effective therapies for pulmonary diseases (114). Tissue transglutaminase inhibitors, such as cystamine, have been found to reduce vascular stiffness in aged rat aorta (115) and to delay vascular remodeling in monocrotaline-induced pulmonary hypertension (116). Similarly, BAPN, an inhibitor of the collagen cross-linking enzyme lysyl oxidase, reduces vascular remodeling in pulmonary hypertension and accelerated vascular aging models (117–120) but also leads to arterial aneurysms. These structural strategies, although aimed directly at long-term vascular changes, are likely high risk because of the consequences of long-term widespread alterations in structural protein deposition in a human compared with the impact of a few weeks of treatment in small-animal models. Improved techniques for targeting such therapies to the aging vasculature will almost certainly be required for clinical viability of this approach.

Given these challenges, many groups have pursued omics-based approaches to seek druggable targets in the aging process. Evaluation of aging using large data sets across species has yielded several common themes. Cellular and molecular changes involving oxidative metabolism, DNA damage, and telomere shortening are seen in higher mammals, whereas mechanisms like cellular senescence and inflammation are common among all species (121). Comparative single-cell RNA sequencing (RNA-seq) and proteomic analyses of lungs from young (3 mo old) and old (24 mo old) mice highlight many of the same themes (122), including increased transcriptional noise, mitochondrial oxidative phosphorylation, and proinflammatory pathway activation. These changes are seen in pulmonary vascular endothelial, mesothelial, and smooth muscle cells (123). Other groups have studied noninvasive systemic pulse-wave velocities in large human populations to gather cohorts at the extremes of arterial stiffness for age (124). Using these cohorts allows for an examination of genetic and epigenetic factors responsible for divergent age-associated stiffening (125). Data from one cohort revealed hypomethylation of a promoter region in calcium and integrin-binding protein 2, which contributes to stiffening and elevated vascular tone by increasing smooth muscle intracellular calcium concentrations (126). As in the lung, differential regulation of transcriptional programs related to oxidative stress was seen with this approach. It might be fruitful to use high-temporal-resolution magnetic resonance (MR) or other MR surrogates (127) to similarly examine cohorts with abnormal-for-age pulmonary vascular mechanics.

On the basis of these collective observations, multiple groups have explored targeting the redox state of the aging vasculature. In animal models, overexpression of antioxidant-regulating transcription factors, such as sirtuin-1 (128) and NRf2 (129) can rescue aging-associated pathology in the systemic circulation. Antioxidant drugs targeting similar pathways, such as high-dose resveratrol and 4-hydroxy-2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPOL), can improve aortic stiffening and endothelium-dependent vasodilation (73, 130). Targeting of mitochondrial ROS, based on strong evidence that mitochondrial function and number decrease in arteries with age, has a large impact on vascular stiffness when genetically manipulated (131, 132). Pharmacological targeting of mitochondrial ROS using mitoQ improves endothelial dysfunction (73) and decreases pulse-wave velocity, ex vivo aortic stiffness, and aortic elastin degradation (133). In the pulmonary circulation, however, both TEMPOL and mitoQ modestly lower right ventricular pressure but do not result in improvements in pulmonary vascular remodeling in hypoxia or Sugen-hypoxia mouse pulmonary hypertension models (134, 135). This suggests either that these agents only impact age-associated remodeling or that they have decreased efficacy in the pulmonary system.

An alternative pharmacological approach is targeting aging-associated inflammation and cellular dysfunction with drugs called “senolytics.” One end result of accumulated age-associated oxidative, mitochondrial, or replicative stress is cellular senescence (136). Senescent cells exhibit a “senescence-associated secretory phenotype” (SASP) that promotes inflammation, fibrosis, and mechanobiological feedback via NF-κB, TNF-α, and TGF-β activation (137, 138). In the systemic vasculature, senescent cells accumulate with aging (139) and correlate inversely with endothelium-dependent vasodilation in vivo (140); clearance of these cells with senolytics improves vascular endothelial function in aged animals (141). Recent work (142) demonstrated that activation of senescence pathways and SASP was closely associated with irreversibility of pulmonary hypertension (PH) in a rat model, and that senolytic treatment could dramatically improve PH and PH-associated vascular remodeling even at late stages of disease. Such findings are driving interest in identifying safe and effective senolytic agents for use in aging-associated cardiovascular disease in both the systemic and pulmonary circulations (143).

Key Questions and Future Approaches

As emphasized at the outset and throughout this review, remodeling of the vasculature occurs with aging and in parallel with altered pulmonary vascular pressures and mechanics. Ultimately, such changes have been implicated, though in still relatively vague mechanistic terms, in aging-related morbidity and mortality. Key questions that emerge from these studies include: What components of remodeling are most critical to altered vascular mechanics? Can these changes be slowed or reversed? And finally, will interventions targeting pulmonary vascular remodeling ultimately result in improved function and benefit for age-related morbidity and mortality?

As a starting point for the first two questions, the complicating presence of feedback loops in vascular remodeling must be emphasized (41). Identifying fruitful points of intervention in pulmonary vascular remodeling requires a better understanding of where such processes begin and how they propagate. Notably, evidence is present for both proximal-to-distal and distal-to-proximal propagation of vascular remodeling, raising questions about the ultimate origins of aging-related mechanobiological remodeling and therefore the most beneficial target for early diagnosis and intervention (41, 108). A proximal-to-distal cascade of vascular remodeling fits with the clinical evidence that proximal vascular stiffening predicts progression of pulmonary hypertension and that reduction of pressure, flow, and shear stress protects against distal remodeling (105). However, the evidence supporting the priority of proximal stiffening in initiating remodeling is accompanied by limited abilities to measure distal vascular stiffness in living organisms. Relatively recent work employed atomic force microscopy (AFM) to invasively characterize smaller vessels in animal models of pulmonary arterial hypertension (93). This study identified increased stiffening of small vessels that preceded stiffening and thickening of larger vessels and right heart dysfunction. However, AFM was limited to measurements of intrapulmonary vessels and did not address aging-related stiffness changes.

It therefore remains unclear whether aging-related remodeling follows a distal-to-proximal or a proximal-to-distal cascade or can follow both within or across individuals. More recent efforts using AFM analysis have confirmed gradual aging-related stiffening of small intrapulmonary vessels in human lungs (144). Such measurements are only possible postmortem and therefore could not be compared to noninvasive assessments of proximal pulmonary remodeling. Resolving the point of initiation of adverse aging-related pulmonary vascular stiffening will thus require advances in methodologies in order to simultaneously assess large- and small-vessel mechanics efficiently and noninvasively. Although not possible in humans, it is notable that ex vivo perfused lungs can be used to disaggregate distal (steady flow) from proximal (oscillatory flow) vascular stiffening (145). With such methods it should be possible to study the aging vasculature in model organisms to identify the proximal versus distal site of initial pulmonary vascular stiffening and carefully reconstruct the molecular and cellular changes in composition and architecture. Merging such approaches with ongoing efforts to map the multigenomic state of cells in the aging lung (122, 146–149) could allow additional functional and molecular insight into the pathogenesis of vascular remodeling not possible through either approach alone.

Computational modeling may serve an essential function in bridging the gap between work in model organisms, from which extensive remodeling and mechanistic data can be gathered, and extrapolation of these findings to humans, in which much more limited data will be available (150–154). In parallel with modeling, advances in clinical imaging modalities that enable less invasive and more detailed characterization of pulmonary vascular mechanics would broadly benefit efforts to study the natural history of, and interventional studies to modify, pulmonary vascular mechanics. In particular, implementation of cardiac magnetic resonance to noninvasively assess key hemodynamic measures of pulmonary vascular mechanics traditionally assessed by right heart catheterization could enable such efforts (155, 156). Similarly, longitudinal chest computed tomography (CT) imaging has shown value in documenting vascular density and pruning in COPD and bronchiectasis (157–159) and could be used in similar studies of aging. In animal models, microvascular CT has already been used to assess (160) and follow microvascular injury and rarefaction during the development of PH (161). The use of advancing technology allowing noninvasive assessment of this process in aging would greatly advance our understanding of the evolution of aging-related lung diseases in the microvasculature. Advances in dynamic imaging, such as lung MR angiography (162), may also allow noninvasive assessment of pulmonary vascular capacitance at different anatomic levels and reveal the impact of age-related remodeling on blood flow and recruitment during exercise. Combined with thorough and longitudinal clinical data, computational analysis of these imaging studies may yield disease-relevant noninvasive imaging biomarkers for early vascular remodeling.

Ultimately, selection and evaluation of therapeutic interventions will be required to understand whether aging-related vascular mechanobiology can be successfully targeted and whether such interventions translate into improved vascular and pulmonary function and reduced morbidity and mortality. Such interventional studies will benefit from guidance gained from recent successes in targeting vascular function in pulmonary hypertension (163, 164). Careful selection of current and emerging end points will be key to identifying early successes supporting the expansion of pilot studies to large, extended clinical trials likely necessary to determine long-term benefit of vascular aging-targeted therapies. Given the growing weight of evidence implicating vascular mechanobiology in aging-related pulmonary diseases and overall declining cardiopulmonary function, clinical translation of mechanobiological modifications may represent an important advance for protecting pulmonary health in aging populations.