Perspective: pathobiological paradigms in pulmonary hypertension, time for reappraisal

Abstract

For the past 120 years, there has been a progressive evolution of the pathobiological concepts underlying pulmonary hypertension. Conceptual frameworks, build around the paradigms of excessive vasoconstriction (vs. vasodilation) and, more recently, of the cancer-like hypothesis of pulmonary hypertension, have served to consolidate key discoveries; moreover, they have and continue contributing to innovative advances that have been translated into either successful or potential new therapies. However, those frameworks do not fully address the complexity and challenges facing pulmonary hypertension, particularly those involving the marked heterogeneity of disease presentation and the dynamic changes occurring over time in affected tissues and cells. This is particularly relevant in regards to the molecular pathways of pulmonary hypertension; the ever growing understanding of molecular and cellular pathways requires clarification if they drive distinctive pulmonary vascular lesions in a given lung and disease patients with the same group pulmonary hypertension. Novel methodologies and approaches can start dissecting this key challenge in the field as it is critical to address the key angle of heterogeneity of the disease and reappraisal of disease-modifying therapies.

INTRODUCTION

For the past 120 years, the study of pathobiological concepts of pulmonary hypertension (PH) has advanced based on paradigms that reflect investigative advances of a given time, while providing a framework that has shaped future investigations into the disease (71). The impact of the hemodynamic definition of PH after 1948 was dramatic (13): the understanding of PH advanced from the early pathologic descriptors to fundamental physiological insights with the development of pulmonary catheterization (8). This particular stage can be framed as “the age of vasoconstriction” in which functional mechanisms involving control of pulmonary vascular tone would favor vasoconstriction and thus drive the ensuing pulmonary pathological lesions of PH. The vasoconstriction paradigm was heavily anchored on the identification of hypoxic vasoconstriction (HPV) (11); this led to fundamental discoveries that revealed chemical mediators that would enhance or drive vasoconstriction versus decrease vasodilation. The discoveries that stem from the excessive vasoconstriction hypothesis resulted in fundamental therapeutic insights, best illustrated by the current use prostacyclin analogs (65), endothelin receptor blockers (42), and pharmacologic enhancers of endogenously produced nitric oxide (43), all of which remain as mainstays of the modern PH therapeutic armamentarium (25).

Notwithstanding the advances in treatment of PH based on the “excessive vasoconstriction” hypothesis of PH, it became increasingly clear that vascular remodeling contributed significantly to the “fixed” or vasodilation-resistant component of PH (41). In fact, most patients with severe disease have a more prominent component of unreactive pulmonary circulation at presentation. Accordingly, the field underwent a shift with a new focus on the nature of pulmonary vascular remodeling rather than the restrictive view of regulation of pulmonary vascular tone. Many of the initial studies focused on vascular thickening, particularly of the media and adventitia, because these were the most apparent changes in the hypoxic model of disease, which was dominant at the time (34, 51). Early replicative changes in adventitial fibroblasts and medial smooth muscle cells were consistently reported, possibly accounting for the increased numbers of pulmonary vascular cells in these compartments (51). These studies also pointed out important subtleties, including the fact that neither smooth muscle nor adventitial fibroblasts are a phenotypically uniform population, as it was originally thought, but rather each is composed of diverse cells with diverse functions responding in distinct ways to specific stress stimuli, i.e., highly plastic in regards to phenotype and functional properties (9, 14, 66). However, animal models of chronic PH due to environmental hypoxia fall short of capturing fully the type and extent of pulmonary vascular remodeling seen in severe forms of human PAH (50, 64). Most patients with idiopathic pulmonary arterial hypertension (IPAH), the prototypic form of severe PH in humans, are largely unresponsive to vasodilators and their lungs demonstrate a complex variety of vascular lesions, which cannot be reproduced by chronic environmental hypoxia. The relatively short time frame of pulmonary vascular remodeling in hypoxic rodent models (i.e., up to 4 wk) is unlikely to capture the diverse and dynamic temporal changes present in the human disease, in which at presentation, has widespread involvement of the pulmonary vascular bed.

Increasing numbers of investigations into mechanisms involved in vascular remodeling in the 1990s and early 2000, including cutting-edge studies regarding the genetic basis of the disease (5, 10, 68), have led to discoveries that framed a new concept, the so-called “cancer-like paradigm of PH.” This hypothesis, formulated based largely on human studies, addressed mechanistically the pathogenesis of complex vascular lesions in PAH (63). That plexiform lesions in IPAH were found to be monoclonal in nature strongly suggested that stem- or progenitor-like cells could acquire a genetic (or mutational) advantage for growth (either increased cell proliferation or decreased cell death) (31, 62), were genetically unstable (1, 12, 69), had a loss of suppressive cell growth signals (such as the transforming growth factor-β family), and activated signaling pathways to favor their long term survival (33). When compared with the relatively simplified early molecular mechanisms underlying the vasoconstriction hypothesis, the cancer-like paradigm flourished exponentially, with ever more complex signaling pathways, now interfacing with altered cellular metabolism, extracellular matrix remodeling, alterations of the immune system, among others being identified. The complexity can be appreciated in the reviews that followed the initial documentation and elaboration of the cancer-like hypothesis (18, 38, 39, 45, 56). While recognizing the merits of the cancer-like hypothesis, inconsistencies and contradictions have been noted (18, 45). Without a critical reassessment and a proper understanding of the context in which the cancer hypothesis originated, and its strengths and limitations, there is the risk to limit the field by overemphasizing the merits of this “unifying hypothesis” at the expense of critical questions that should shape the investigations in the field. Given that this hypothesis is now “mature” (6), it is imperative for investigators to critically evaluate its merits and weaknesses and decide on how best to move forward by focusing on key next steps in the field.

The fields of biology and medicine have recently experienced an exponential growth of groundbreaking methodologies that can be applied to interrogate fundamental molecular processes. Insights generated by these advances have gone beyond the constraints of simplistic hypotheses, as illustrated by the ever growing understanding of the role of pericancerous stroma in allowing for cancer growth or elements of immunity, many shaping similar cellular actors and molecular mediators involved in PH. As with the pathobiological hypotheses of PH, summaries of accomplishments and perceived needs in the field (20, 24, 35, 59) have emphasized an ever-growing list of molecular and signaling pathways. As described, they appear to act in parallel, simplistically ascribed to be universal to the major pulmonary vascular cells—endothelial cells, smooth muscle cells, and fibroblasts—with no mechanistic insights on how these multiple pathways could be orchestrated at a particular cell level and over time, during the disease course. However, all are reported to have an impact on the key parameters of pulmonary hemodynamics and/or remodeling. The authors’ premise is that there is pressing need to resolve the mechanistic hierarchy of these pathways in relation to disease initiation and progression, taking in consideration the critical nature of a highly heterogeneous disease. This heterogeneity concerns clinical parameters, including initiating factors, such as age at onset, clinical presentation, rate of progression, and response to therapy (53). Importantly, patient-to-patient heterogeneity, involving the types of pulmonary vascular lesions and the corresponding endotypes (i.e., underlying molecular processes driving the specific disease presentation), has been uncovered by our group and underlies the high complexity of disease pathogenesis (46) (Fig. 1). The paucity of investigations of these broader aspects of heterogeneity has resulted in a lack of understanding of specific factors contributing to particular subphenotypes of PAH, negatively impacting the development of more targeted (and individualized) therapies. The authors propose that this is the time to prioritize addressing heterogeneity in PH.

Fig. 1.

Summary of the levels of heterogeneity to be addressed by the Pulmonary Hypertension (PH) field, at the patient, lesional, and cellular composition levels, including the level of the extracellular matrix. The authors consider these goals as central to advancing PH research, allowing for a critical validation of novel paradigms, while building on accomplishments from the past. SMCs, smooth muscle cells.

“CLEAR AND PRESENT DANGER” IN PH: LESIONAL HETEROGENEITY

The pulmonary vascular field has been limited by its lack of understanding regarding the initiation of the disease, notably in relation to group 1 pulmonary arterial hypertension (PAH). The nature of the “triggering injury event” is unclear, often speculatively imputed to be due to excessive vasoconstriction or an undefined infectious agent(s). The incomplete penetrance of mutated genes linked to disease, notably of bone morphogenetic protein receptor II, suggests that a trigger event exists. Moreover, the time span between triggering event(s) and when PH is diagnosed in disparate individuals remains elusive and is a major obstacle to advancing our understanding of PH. The fallback has been to infer this key pathogenetic deficiency in our understanding of PH from animal models; the monocrotaline (4)- and SU5416-chronic hypoxia (54) rat models implicate vascular injury, leading to endothelial cell apoptosis or death (32, 55), which would be required for establishment of the severe chronic stage of PH. However, in large animal models, hypoxia is a critical driver, engaging a multitude of pathogenetic steps that have relevance to the human disease (48). At the end of the day, they represent unproven assumptions of what might happen in patients with PH. The authors propose that these limitations explain the paucity of new therapies developed based on animal studies, rather than resulting from imperfections of study design, such as randomization or masking of experimental groups, or statistical flaws (37). Without knowledge on how the disease starts, different pathogenetic routes cannot be directly/indirectly linked to a single or, alternatively, to multiple trigger events. This pitfall also hampers data from animal models. It is conceivable that, even in the so-called “established” disease stage, each model may at best reflect only elements of the early disease in humans.

When diagnosed, PAH shows established (i.e., potentially irreversible) pulmonary vascular lesions. At the present time, there is no evidence that assigns which critical lesions account for an increase in pulmonary vascular resistance in PH (28, 57). Several lesions are recognized either topographically as located in the intima, media, adventitia, or by the predominance of structural cells that compose each of the specific lesions (61): endothelial cells likely predominate in plexiform and some concentric intima lesions (60); myofibroblasts are predominant in intima obliteration, while smooth muscle cells account for media remodeling; fibroblasts undergo activation in the adventitia. Few mapping studies have localized some obliterative lesions to branching points (7, 67), or supernumerary arteries (27), but how vascular lesions are distributed along the fractal pulmonary artery remains unclear (58). These changes of resident pulmonary vascular cells occur in the context of perivascular inflammation, one of the most consistently reported finding in human and animal models of PH (Fig. 1); importantly, perivascular inflammation is more than a bystander in the disease process, as it impacts on hemodynamic parameters (20, 46, 60). Notwithstanding the discoveries related to gene expression profiling using whole normal and diseased lung tissues (17), the signaling hubs that operate in each of these lesions remain unknown. It should be noted that a thorough investigation regarding molecular profiling of the plexiform lesion did not support the concept of an “angiogenic niche” that harbors cells with quasi neoplastic behavior in vessels neighboring or giving rise to the plexiform lesion, but rather interpreted as an overshot of a regenerative process (not a neoplastic process; see Ref. 28), somewhat akin to abnormal wound healing or abnormal clotting (70). Moreover, it is unclear whether different PAH patients with a similar spectrum of pulmonary vascular lesions share similar lesion-specific pathogenetic pathways or endotypes. None of the individual intimal or medial lesions carry a statistically significant correlation with pulmonary artery pressures or pulmonary vascular resistance (46). However, a trend is noted when hemodynamic is compared with “intima thickening.” Furthermore, the quantification of each type of lesion varies significantly in different lung regions. There is no correlation among lesions in regard to their distribution or frequency, limiting the extent that these data might offer clues to their potential topographical and pathogenetic relationship (46). The aspect of lesional heterogeneity is illustrated by the inability to detect clusters derived from principal component analysis derived from the data of intima or medial remodeling, perivascular inflammation, and plexiform lesions from 62 patients with PAH and 28 controls (Fig. 2) (46).

Fig. 2.

Principal component analyses (PCA) of control (CTL, n = 22), aged controls (CTL2, n = 6), idiopathic pulmonary arterial hypertension (IPAH, n = 48), associated pulmonary arterial hypertension (APAH, n = 12), and venocclusive disease (VOD, n = 2) as described in Ref. 46. The parameters used for the PCA were thickness measurements of intima, media, adventitia, and inflammatory score (methods and data outlined in detail in Ref. 46). Outlined are cluster 1 with an enrichment of controls (CTL and CTL2) and, in cluster 2, with an enrichment of APAH lungs (8 of 12). APAH cases falling outside cluster 2 are outlined by gray arrowheads. VOD cases are highlighted by black arrowheads. Note the IPAH lungs did not conform to specific PCA clusters, underscoring their pathological heterogeneity.

A further level of heterogeneity is that pulmonary vascular and inflammatory cells in PH form phenotypically diverse cell clusters, as noted with pulmonary artery smooth muscle cells and perivascular fibroblasts (49). Several determinants account for this diversity: anatomic location along the pulmonary arterial tree (58), distinct origins of specific cell subpopulations [i.e., bone marrow (2) or transdifferentiation (22, 40)], type and response to injury, and interaction with adjacent structural and recruited inflammatory cells. The time effect on these cell types, e.g., their response to increase in numbers (proliferation) or specific states related to quiescence (progenitor features, cell growth arrest due to senescence), and epigenetic markers is perhaps the most difficult to ascertain. The plasticity accounting for the heterogeneous properties of pulmonary vascular smooth muscle cells has been recently reviewed (49). Time- and location-dependent plasticity also applies to pulmonary endothelial cells (30, 52) and, likely, to adventitia fibroblasts (23); the same limitations apply to inflammatory cells. In fact, the authors are not aware of studies that took into account the topology and time-dependent behaviors of the different cells involved in PH, since only “snapshot” insights have been reported (44, 60). All these elements of plasticity require detailed consideration not only in PH but also in the normal pulmonary circulation. Importantly, age-related phenotypic changes in the apparently normal pulmonary circulation might differ significantly from the pulmonary circulation in younger individuals (46).

It has been increasingly recognized that not only cells (structural and immune cells) participate in the pulmonary vascular lesions but also the extracellular matrix (ECM). The ECM, which is largely understudied in the PH field, is a determinant of location- and time-dependent plasticity of pulmonary vascular cells. Furthermore, the extracellular matrix is in constant flux as the lung matures, ages, and responds to injury (56). ECM remodeling and pulmonary vascular stiffness occur early in the disease process and often precede increases in vascular thickness and even arterial pressure. This suggests changes in ECM and related proteins can be causes and not simply result from changes of vascular cell phenotypes. In established IPAH, there are distinct, compartment-specific changes in expression of genes encoding collagens, tenascin, fibronectin, and osteopontin. Additionally, there is evidence of ongoing breakdown of elastin and collagens with increased expression of elastases and metalloproteinases (21, 24, 38). ECM remodeling and stiffness, through mechanoactivation of multiple signaling pathways, can have profound effects on vascular cell phenotype and function. The fine proteome of pulmonary vascular lesions in human PH (notably in PAH) has not been elucidated thus far. We propose that the intra- and extracellular proteome are key to the thickening of all three vascular compartments and that there are multiple stages of matrix remodeling defined by unique macromolecular components, matrix cross linking, and cellular signatures.

Appropriately fingerprinting the multilevel heterogeneity of pulmonary vascular lesions will lead to better patient stratification, identification of novel therapeutic targets, and better definition of strengths and limitations of animal models. The lesional proteome is predicted to contain disease driver(s), such as elements of the classic and alternative pathways of complement (15). Moreover, this lesional proteome may ultimately manifest increased pulmonary vascular stiffness and drive pathogenic enhanced cell survival/proliferation, resistance to apoptosis, and inflammation (24).

NEW METHODOLOGIES: BREAKING GROUND INTO PH PATHOBIOLOGY

Presently, the cancer field has mastered the application of groundbreaking novel methodologies as opposed to building on traditional concepts (19); these advances have largely revolutionized the diagnosis and treatment of patients with malignancies. The authors propose that, as in cancer, new advances in PH will require the combined use of high-density molecular (genomic and proteomic) in vivo data and the use of three-dimensional models, particularly those described for organoids (3, 36, 73). The cancer field has an advantage in that investigators can obtain and utilize human biopsies in patient-derived xenograft models. Unlike organoids, patient-derived xenograft models have been suggested to retain the architecture and morphology of the original transplanted tumors and may provide better assays for both mechanistic and drug studies (16). Cancer investigators have used patient-derived explant (PDE) cultures (36). In these studies, the tumor itself can be studied ex vivo for assessment of mechanistic and drug responses. PDE can be homogenized, and protein, DNA, RNA, or metabolites can be isolated and measured using a variety of different approaches such as mass spectrometry and transcriptomic, genomic, or metabolic profiling. If applied to the PH field, these approaches should be paralleled by comprehensive analysis of tissues using laser microdissection and advanced molecular studies based on novel genomic platforms optimized to archival control and PH lungs, such as those banked by the Pulmonary Hypertension Breakthrough Initiative (46). Further, new developments such as single-cell RNA sequencing, could allow for the longitudinal characterization of changes within the complex vascular lesions or in the vascular environment in response to ex vivo drug treatment. Methods such as tyramide signal amplification (TSA) can be applied to formalin-fixed paraffin-embedded tissue and is particularly suited to multiplex imaging (47). Moreover, the equivalent to mass cytometry with up to 50 simultaneous markers can be simultaneously visualized and quantified, with the key advantage to map the cellular structures in the lung; tissue-based proteomics further expands these tools while being able to define extracellular proteins at the tissue level.

The authors predict that the evolution of these and other methodologies will converge to allowing studies in the diseased and normal human lung, which key heterogeneities can be determined at the genomic (DNA and RNA), proteomic, and metabolomics levels. The time-dependent parameters may become available based on lung biopsies (with an acceptable safety profile due to technical advances) or by the use of surrogates in the blood, such as progenitor cells (72).

The impact of such an approach will likely extend beyond the group 1 PAH group, but will become transformative to the investigation of largely understudied groups 2–5 of PH, which have lagged significantly behind the insights gained in the PAH group. Moreover, other lung diseases, in the pediatric and adult groups, characterized by significant pulmonary pathology (such as fibrotic diseases, inflammatory/immunity disorders, and COPD-related destruction) will greatly benefit from novel methodologies, leveraging novel human tissue acquisition and studies in archival material from well-characterized cohorts.

In conclusion, we believe that, similar to the emerging idea that cancer is not a single-cell disease but rather the result of complex interactions of tumor cells with surrounding matrix and immune cells, the vascular remodeling that characterizes chronic pulmonary hypertension is also the result of interactions between numerous cell types and their surrounding matrix and inflammatory microenvironment. These interactions are multifaceted, cell type-specific, and dynamic over time. In cancer, there is now strong evidence for extensive inter- and intratumor heterogeneity both within and between human cancer samples and for a critical role of the tumor microenvironment in driving cancer hallmarks and in influencing drug responses (26, 29). In PAH, it may be that our previous lack of appreciation for the diversity and complexity of the vascular lesions that characterize PAH might have accounted for the lack of generalized success of many new therapies. The human PH is silent until more advanced stages, when clinically important symptoms arise. However, much of the data in the field regarding potential new therapies has come from rodent models with very short time courses and/or human cells, which have been studied in two-dimensional cell culture, one cell type at a time. These data have indeed generated large databases and uncovered new signaling pathways but, as in cancer, the caveat is the highly reductionist nature of these approaches, relying on 2D cell lines, devoid of their interaction in vivo or in vitro with their natural microenvironment. A comprehensive approach to overcome the complex heterogeneity of PH will require leveraging groundbreaking methodological advances, which can lead to development of novel insights into the disease and potential new therapies.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants R24 HL123767 (to R.M.T.), P01 HL14985 (to R.M.T. and K.R.S.), and R01 HL08783 and Department of Defense Grant R181125 W81XWH-19-1-0259 (to K.R.S.).

DISCLOSURES

R.M.T. has stock options in Pulmokine. K.R.S. has no conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

R.M.T. and K.R.S. prepared figures; drafted manuscript; edited and revised manuscript; and approved final version of manuscript.