From Development to Regeneration: The Endothelial Interface in Lung Injury and Repair

Abstract

The pulmonary alveolar-capillary niche is a highly specialized interface that balances gas exchange with maintenance functions and repair. Advances in single cell transcriptomics have uncovered endothelial heterogeneity which underlies developmental angiogenesis and plastic responses to injury. Emerging evidence from a neonatal hyperoxia model highlights CAP1 to CAP2 transitions and the role of p53 in maintaining lineage fidelity. Beyond intrinsic lineage plasticity, circulating mediators such as cell-free hemoglobin drive endothelial barrier disruption through oxidative injury and lipid modification. As new signaling pathways and therapeutics targets emerge, complementary strategies are being developed at the cellular level, including adoptive transfer of mesenchymal stromal and immune cells, although mechanisms of endothelial adhesion and homing remain incompletely defined. Finally, biomechanical forces such as shear stress have become critical contextual cues for endothelial signaling, yet remain underrepresented in some experimental models. Together, these insights underscore the central role of endothelial heterogeneity, injury responses, and environmental cues in shaping pulmonary vascular health and repair, with implications for designing targeted therapies in both pediatric and adult lung disease.

INTRODUCTION

The alveolar-capillary niche is a structurally minimal, yet powerfully functional interface, exquisitely adapted for pulmonary gas exchange, immune regulation, and tissue repair. Function follows form as dynamic cell-cell interactions, mechanotransduction, and tightly regulated signaling networks combine to ensure intimate contact and alignment of alveolar epithelial cells and pulmonary microvascular endothelial cells (ECs) to optimize gas exchange. Recent advances in single cell transcriptomics and developmental biology have revealed a previously unknown degree of pulmonary EC heterogeneity and plasticity, as well as EC specific differences in the response to injury across developmental stages.

In humans, alveolarization is predominantly a postnatal process. Alveolar number increases exponentially from approximately 30 million in a term infant to over 300 million in an adult.(1, 2) The process is driven by angiogenesis and remodeling of the pulmonary capillary network.(3, 4) At birth, the pulmonary circulation undergoes an unprecedented, and biologically imperative transition as pulmonary blood flow increases 10-fold and pulmonary arterial pressure decreases by half.(5, 6) Establishment of an air-liquid interface, increased oxygen tension, rhythmic distention of the lung, and an increase in shear stress leads to endothelial nitric oxide synthase (eNOS) activation, nitric oxide production and pulmonary vasodilation.(7)

The pulmonary endothelium comprises multiple transcriptionally distinct populations with spatially and functionally specialized roles.(8, 9) Single cell RNA sequencing (scRNA-seq) studies have identified two major capillary endothelial subtypes: Capillary (CAP) 1 and 2 cells. CAP1 cells exhibit angiogenic, progenitor-like, and reparative gene signatures, while CAP2 cells, which are specified by epithelial-derived VEGFA, adopt a thin, web-like morphology and are in direct contact with the alveolar epithelium to optimize gas exchange.(9, 10) During alveologenesis, the interplay between the epithelium and these EC subtypes ensures proper vascularization, contributes to alveolar septation, and supports formation of a functional air-blood interface. This endothelial heterogeneity likely reflects differences in mechanical stress, metabolic demand, and proximity to epithelial and immune cells across the vascular tree.

Beyond baseline heterogeneity, pulmonary ECs display substantial regenerative capacity following injury. After insults such as influenza infection, lipopolysaccharide (LPS) exposure, or mechanical trauma, ECs activate gene programs associated with proliferation, migration, and vessel remodeling.(11) CAP1 cells can re-enter the cell cycle, migrate to sites of injury, and differentiate into CAP2 cells to restore barrier function and gas exchange capacity.(12) This process involves reactivation of vascular endothelial growth factor (VEGF), Notch, and transcriptional regulators such as activating transcription factor 3 (ATF3).(13, 14)

In adult mammals, this ability of the endothelium to regenerate is also evidenced by compensatory lung growth following pneumonectomy.(15, 16) Here, both sprouting and intussusceptive angiogenesis contribute to alveolar regrowth, coordinated by interactions between ECs, fibroblasts, and pericytes.(15, 17) Intussusceptive angiogenesis allows expansion of the vascular network without disrupting its architecture, highlighting the sophistication of adult lung vascular adaptation.(18, 19)

Despite its regenerative potential, the alveolar-capillary interface is vulnerable to inflammatory and infectious injury. During influenza infection, ECs undergo activation characterized by increased expression of adhesion molecules, cytokines (e.g., IL-6, TNF-α), and matrix-degrading enzymes. This inflammatory cascade leads to endothelial barrier dysfunction, increased permeability, and alveolar edema.(20) Nonetheless, following such injury, CAP1 cells proliferate and repopulate the damaged vasculature, initiating vascular repair.(11, 14)

These injury responses are also strongly shaped by the lung’s developmental stage. Neonatal lungs often demonstrate enhanced resistance to injury, including LPS-induced endothelial dysfunction. Relative to adult ECs, neonatal ECs respond to injury with attenuated inflammatory response, reduced cytokine expression, and more stable tight junction architecture.(21) Conversely, adult lungs are more prone to exaggerated inflammatory responses and compromised barrier function. LPS exposure in adult models leads to increased vascular endothelial (VE)-cadherin internalization, cytoskeletal reorganization, and persistent vascular leak.(22)

The alveolar-capillary niche exemplifies a model of cellular specialization, molecular signaling, and mechanical integration. EC heterogeneity, particularly the CAP1/CAP2 axis, enables the lung to balance gas exchange with dynamic repair and immune function. Importantly, the lung’s ability to respond to injury is profoundly influenced by developmental stage. Understanding these processes is essential not only for developing targeted therapies for acute lung injury (ALI), sepsis, and viral pneumonia, but also for appreciating how lineage plasticity shapes repair.

Endothelial Plasticity and Lineage Maintenance in Alveolar Development and Injury

Building on this framework of heterogeneity and developmental stage, recent work has illuminated how lineage plasticity and transitional endothelial states contribute to regeneration in the injured neonatal lung. Neonatal hyperoxia in mice is a commonly used model of bronchopulmonary dysplasia (BPD), a chronic lung disease of premature infants with compromised vascular development and impaired alveolarization.(23, 24) CAP2 ECs are particularly susceptible to the oxidative injury caused by hyperoxia, undergoing apoptosis and morphological simplification. In response, CAP1 cells initiate a compensatory program, converting into CAP2 ECs to re-establish the alveolar capillary barrier. While this transition preserves function, it also depletes the CAP1 population, which may contribute to long-term vascular rarefaction and defective alveolar structure seen in the disease.(25)

Regulation of CAP1 to CAP2 transition during hyperoxic injury also appears to involve the tumor suppressor gene p53, a transcription factor with central roles in cell-cycle arrest, apoptosis, and senescence. p53 is known to be upregulated in multiple cell lineages in hyperoxia, with oxidative stress widely proposed as a key driver of this activation, and more recently senescence.(12, 26–29) Another mechanism proposed is that the robust activation of the p53 pathway seen across the capillary endothelium in hyperoxia plays a role in maintaining lineage fidelity during vascular repair.(25) In this context, endothelial-specific deletion of p53 results in partial rescue of both alveolar and vascular phenotypes. This rescue is accompanied by the emergence of a “transitional EC” population, transcriptionally intermediate between CAP1 and CAP2 cells based on trajectory analysis, and enriched for genes associated with oxidative stress responses and angiogenesis.(25)

The identification of a transitional endothelial state draws a compelling parallel to epithelial injury-repair mechanisms, such as the emergence of damage-associated transient progenitors (DATPs) which also exhibit p53 upregulation.(30, 31) Although other injury models, such as cell ablation and neonatal deletion of VEGF receptor 2 (Vegfr2 or Kdr) in combination with hyperoxia, have also produced intermediate capillary states, those appear to correspond to earlier CAP2 cell specification also seen during development.(32, 33) By contrast, transitional ECs seem to represent a later point along that specification program, and are specifically seen upon endothelial deletion of p53 in hyperoxia treated mice.(25) Whether they ultimately resolve into fully mature CAP2 cells remains an open question. It is possible that their fate, like transitional epithelial populations, may depend on the nature and severity of the injury. These findings support a broader conceptual framework in which lung regeneration relies on lineage plasticity and stress-responsive transcriptional programs, including those mediated by p53. Importantly, the transitional EC gene signature is conserved in an aberrant capillary population identified in human BPD with pulmonary hypertension (PH), suggesting that this regenerative trajectory is clinically relevant and evolutionarily conserved.(25, 34) By elucidating the mechanisms by which capillary EC subtypes respond to developmental and pathological cues, we can better understand the basis of lung resilience and identify new strategies to enhance vascular repair in pediatric lung disease.

Hemoglobin-Mediated Endothelial Injury in the Alveolar-Capillary Barrier

While lineage plasticity represents an intrinsic mechanism of repair, circulating mediators can directly compromise endothelial integrity and disrupt the alveolar-capillary barrier, leading to pulmonary edema and pathologies such as ALI and acute respiratory distress syndrome (ARDS).(35) The pulmonary endothelial barrier is maintained by cell-cell junction proteins (adherens and tight junctions) connected to the cytoskeleton via catenins and is precisely controlled by a vast interplay of intracellular signaling pathways.(36, 37) Though great progress has been made, a comprehensive understanding of the precise mechanisms regulating pulmonary endothelial injury and repair remains.(38)

The barrier can be disrupted by several circulating mediators, including pathogens, toxins, inflammatory cytokines, and oxidative molecules that directly interact with the endothelium. One such mediator gaining attention as a critical driver of alveolar-capillary barrier disruption is circulating cell-free hemoglobin.(39) This molecule can be released from fragile or damaged red blood cells during inflammatory and hemolytic pathologies, including sickle cell disease, PH, transfusion-related ALI, primary graft dysfunction after lung transplantation, sepsis, and ARDS. The hemoglobin structure (two α- and two β-globins with iron-containing heme groups) allows it to redox cycle under oxidative stress, amplifying injury.(39)

Hemoglobin can also oxidize lipids and lipoproteins, propagating endothelial injury.(40) Indeed, a relationship between circulating hemoglobin or heme and oxidized phospholipids with associated pulmonary and cardiac tissue injury in sickle cell disease has been reported.(41) Meegan and colleagues recently described that oxidation of low-density lipoprotein (LDL) by hemoglobin or heme exacerbated human pulmonary microvascular endothelial barrier dysfunction, partially via the endothelial scavenger receptor lectin-like oxidized LDL receptor 1 (LOX-1).(42)

While these studies reveal exciting novel mechanisms, there is much to learn about the role of hemoglobin-mediated oxidation of lipids and lipoproteins and the subsequent effects on endothelial injury, including the precise interactions between hemoglobin and heme with different species of lipids and lipoproteins (e.g., LDL, high-density lipoprotein, phospholipids, etc.), endothelial scavenger receptors involved in the response (e.g., LOX-1, CD36, SR-BI), and the intracellular signaling events mediating barrier disruption. Nevertheless, these discoveries reveal a novel signaling pathway that has potential to be targeted therapeutically along several points. For example, patients with high levels of circulating hemoglobin might be treated with supplemental haptoglobin,(43–45) the endogenous scavenger of hemoglobin, or with hemoglobin reductants such as acetaminophen.(46–48)

Though there are no specific pharmacological inhibitors of oxidized lipoproteins at this time, targeting their interactions with receptors on the endothelium may have potential; interestingly, therapeutics targeting the LOX-1 receptor, including small molecule inhibitors, monoclonal antibodies, and microRNAs, are already being developed.(49, 50) A combination of therapies may protect or repair the alveolar-capillary barrier in disease or injury. A deeper understanding of the precise mechanisms involved in alveolar-capillary barrier disruption mediated by oxidation of lipoproteins by hemoglobin has great potential to reveal novel therapeutic targets in several inflammatory and hemolytic disorders. Beyond molecular targets, however, complementary strategies are emerging at the cellular level, particularly approaches that directly harness or modify cells to restore endothelial integrity.

Cell-Based Therapeutics and Endothelial Targeting in ALI/ARDS

Severe lung inflammation can result in ALI or, in severe cases, ARDS. Treatment in ALI and ARDS, diseases with a high mortality rate, is supportive and includes low-volume mechanical ventilation to mitigate further lung injury.(51) Although ALI results from lung-endothelial barrier failure, specific barrier-protective therapies are lacking, but are urgently required since ALI may be caused by a range of insults including trauma, pneumonia, sepsis, aspiration, and COVID-19.(51, 52) As part of emerging cell-based therapeutic strategies, several studies have demonstrated protective effects from adoptively transferred mesenchymal stromal (stem) cells (MSCs), alveolar type 2 cells, macrophages and NK cells.(53–58) While some clinical studies have reported survival benefits following MSC therapy,(59–67) others have shown limited efficacy.(68–70) Most of the studies, however, show improvement in inflammatory cytokines, biological markers, and transcriptomic profiles.(59, 61–65, 67, 69–73) These early findings underscore the importance of further investigating the mechanisms of action underlying cell-based therapies to refine treatment protocols and enhance their effectiveness in future clinical applications.

To confer a protective effect adoptively transferred cells, including intravenously (i.v.) administered MSCs, must attach to the vasculature. However, the mechanisms by which i.v. MSCs, the preferred route in clinical trials, attach to the intact pulmonary endothelium remain poorly understood. While systemic capillaries are known to restrict MSC transit, i.v. MSCs preferentially adhere to postcapillary venules, suggesting that MSC can traverse the lung capillaries.(74) This contrasts with recent reports demonstrating substantial MSC attachment to pre-capillary arterioles.(75) Following i.v. administration, MSCs are rapidly retained in the pulmonary circulation,(76) likely due to their size and entrapment in vasoactive regions, as supported by studies showing reduced lung retention in the presence of vasodilators such as nitroprusside.(77) The absence of embolic complications in clinical trials further suggests that lung capillaries are not significantly obstructed by MSCs.(60, 70) Indeed, the high compliance of pulmonary capillaries may facilitate MSC transit more effectively than systemic microvascular beds. Additionally, MSCs are capable of deforming and thinning out to pass through intercellular junctions, potentially enabling capillary transit and even transvascular migration into alveolar spaces, where they may suppress inflammation.(78) This endothelial attachment may also enhance the local delivery of MSC-derived immunosuppressive factors, contributing to their known neutrophil-inhibitory effects and further supporting their therapeutic role in modulating alveolar inflammation.(78–81)

While passive entrapment contributes to retention, multiple studies indicate that receptor-mediated interactions also play a role. MSCs express adhesion molecules such as CD29 (fibronectin receptor) and CD44 (osteopontin receptor),(77, 82) which enable attachment to fibronectin-rich or osteopontin-expressing domains, often upregulated in injured lungs.(83, 84) MSCs exhibit rolling and adhesion on activated endothelium via VLA-4/VCAM-1 and P-selectin interactions, and similar to leukocytes, MSCs rely on surface molecules such as Sialyl Lewis X and α4 integrins for vascular recruitment.(85, 86) Engineering CD44 to HCELL enhances MSC homing and retention, further supporting selectin-based adhesion mechanisms.(87) However, excessive MSC accumulation can be detrimental, as shown by Otsu et al., who demonstrated that high MSC concentrations induced ROS-dependent endothelial apoptosis via connexin-43.(82) Thus, while MSCs appear to home to injured tissue using leukocyte-like mechanisms, refining homing efficiency is essential to maximize therapeutic benefits and avoid harmful effects, particularly in the lung during ALI/ARDS. Therefore, a definitive understanding of whether MSCs and other adoptively transferred cells are retained by specific attachment or by non-specific rheologic mechanisms will critically determine the effectiveness of cell-based therapies against ALI/ARDS.

Shear Stress as a Regulator of Pulmonary Microvascular Homeostasis

In addition to the developmental programs, injury responses, and therapeutic interventions previously described, endothelial behavior is continuously shaped by biomechanical cues. The pulmonary microvasculature is exposed to laminar blood flow, resulting in a homeostatic degree of shear stress that is exerted parallel to flow, and serves as a critical regulator of endothelial signaling. ECs are well-recognized as sensors of both biochemical and mechanical stimuli,(88, 89) but in the alveolar niche their unique proximity to pneumocytes creates a specialized signaling environment. Understanding the baseline signaling pathways of the pulmonary microvasculature is crucial for interpreting aberrations in disease states. Yet, most in vitro studies are performed under static conditions, potentially overlooking key mechanotransductive cues.

Recognizing shear stress as an essential contextual factor highlights an important gap in experimental design and points to future opportunities for refining models of lung endothelial biology. In pulmonary hypertension (PH), for example, pulmonary blood flow is altered and shear stress increased in the distal pulmonary vasculature as a result of vessel narrowing and vascular remodeling.(90) Since changes in shear stress critically influence endothelial signaling, in vitro approaches should be adapted to incorporate flow conditions. ECs grown in static culture can be compared to those exposed to physiologic shear stress to disentangle differences in biochemical signaling, protein expression, and downstream functional outcomes. Despite observations that ECs change their shape in response to shear stress, both in vitro and in vivo, the precise pathways governing this process remain unknown. Similarly, shear stress may be an important determinant of apoptosis susceptibility, a factor crucial to the maintenance of a healthy capillary-alveolar niche in disease states.(91–93)

To address these gaps, experimental systems have been developed utilizing channel slides attached to a rotary pump. This method allows exposure of ECs to constant shear stress,(94) with the advantage of enabling real-time calcium imaging and visualization of acute changes in response to shear stress.(95) In parallel, orbital shaker models with fixed revolutions per minute and media volume, such as the one pioneered by Croglio and colleagues, can be used to study EC morphology, protein abundance, and survival.(93) Shaker systems are particularly useful for generating higher cell yields for protein or RNA assays but are limited by non-linear flow and the inability to capture real-time signaling events. The complementary use of both systems broadens experimental design, combining dynamic imaging with large-scale biochemical readouts. These techniques allow characterization of shear stress-dependent processes such as eNOS production and endothelial-mesenchymal transition, highlighting the importance of modeling biomechanical forces in endothelial research.(94, 96)

Using these in vitro shear stress models, Croglio and colleagues have identified important differences in endothelial signaling and function, including calcium dynamics, cell shape maintenance, and apoptosis susceptibility (unpublished data). Recent work shows that physiological shear stress suppresses apoptosis in human pulmonary microvascular ECs,(93) while Fisher and colleagues demonstrated calcium influx in murine pulmonary ECs in response to acute shear.(97) These findings underscore shear stress as a fundamental regulator of endothelial homeostasis, while also highlighting the limitations of current in vitro systems and the need for in vivo validation. Increased intracellular calcium and altered endothelial proliferation and migration have been described in pulmonary hypertension animal models, supporting some overlap between in vitro and in vivo observations, though systemic influences and flow complexity may account for divergences.(98–100)

Further work integrating physiological flow in animal models of PH or pulmonary artery ligation, will be essential to clarify how shear stress influences capillary-endothelial behavior under both health and disease.(101) In systemic vascular research, carotid casting and aortic coarctation models have long been used to study shear-dependent signaling, and analogous strategies such as pulmonary artery ligation and isolated perfused lung preparations can be applied to the lung.(102) While initially investigated in the context of pulmonary vascular disease, more broadly this work will allow a better picture of the alveolar-capillary niche in other pathological states. In ALI, for instance, hypoxic pulmonary vasoconstriction may increase shear stress and reduce endothelial apoptosis as a protective mechanism, whereas low or oscillatory shear stress (as seen in systemic vessels) has been linked to increased endothelial permeability.(103) Notably, severe BPD often manifests with PH, further highlighting how developmental vascular injury intersects with shear-dependent signaling. As such, shear stress should be viewed not only as a variable to control, but as a lens through which to refine models of lung endothelial biology, setting the stage for future studies aimed at understanding and ultimately repairing the injured pulmonary vasculature.

SUMMARY

From developmental differences in endothelial response to hyperoxic injury, to the biomechanical influence of shear stress on capillary homeostasis, the pulmonary vasculature emerges as a critical determinant of lung health and disease (Figure 1). Novel mechanisms of endothelial dysfunction, such as hemoglobin-mediated lipid oxidation, highlight underappreciated pathways of injury, while cell-based interventions targeting the endothelial interface offer new therapeutic opportunities. Collectively, this review highlights mechanistic insights into lung vascular development, maintenance, injury and repair, underscoring the essential role of the endothelium and its interactions with the alveolar niche. Advancing this knowledge will be key to designing tailored interventions for both pediatric and adult pulmonary diseases.

Figure 1. Mechanisms of alveolar-capillary niche disruption and repair in lung health and injury.

Schematic illustrating cellular dynamics within the alveolar-capillary interface across states of health and injury. The left panel depicts a healthy niche characterized by endothelial cell heterogeneity (CAP1 and CAP2 subtypes), shear stress signaling, and regenerative interactions between alveolar epithelial type 1 and 2 (AT1/2) cells and mesenchymal stromal cells (MSCs). The right panel illustrates pathological features observed in hyperoxia, acute lung injury (ALI), and acute respiratory distress syndrome (ARDS), including immune cell infiltration, hemolysis, reactive oxygen species (ROS)–mediated oxidative injury, barrier dysfunction, and endothelial lineage infidelity. Key players such as red blood cells (RBCs), platelets, and immune cells are shown engaging with the injured endothelium and epithelium, influencing the trajectory of vascular injury and repair. Created in BioRender.