dTAG-13

Occlusive Lung Arterial Lesions in Endothelial-targeted, Fas-Induced Apoptosis (FIA) Transgenic Mice

Abstract

Pulmonary arterial hypertension (PAH) is a deadly condition marked by structural and functional abnormalities of distal pulmonary arterioles, leading to increased vascular resistance and eventual right heart failure. Experimental models suggest that endothelial cell (EC) apoptosis serves as a vital trigger for developing obliterative lung arteriopathy by fostering hyperproliferative and apoptosis-resistant vascular cells. However, whether EC apoptosis alone is sufficient to induce complex lung arteriolar lesions has been unclear.

To investigate, researchers developed a conditional transgenic mouse model expressing the Fas-Induced Apoptosis (FIA) construct under the endothelial-specific Tie2 promoter. In this model (EFIA mice), administration of the small molecule dimerizing agent AP20187 induced modest pulmonary hypertension and obliterative vascular lesions in distal lung arterioles in some mice. These lesions were predominantly composed of proliferating cells, especially CD68 macrophages. Interestingly, while endothelial apoptosis was also observed in the kidney, subsequent arteriopathy was restricted to the lungs.

This study provides compelling evidence that lung endothelial apoptosis can act as a primary trigger for initiating PAH-like phenotypes. It also sheds light on the mechanisms underlying the lung-specific vascular response to endothelial injury, presenting valuable insights for understanding and potentially treating PAH.

Introduction

Pulmonary arterial hypertension (PAH) is a fatal condition characterized by increased pulmonary vascular resistance, ultimately leading to right ventricular hypertrophy and right heart failure. Arteriolar remodeling, rather than increased vasomotor tone, is considered the primary mechanism responsible for the loss of functional pulmonary vascular cross-sectional area. This remodeling often features increased muscularization, concentric intimal hyperplasia, and hallmark plexiform lesions, arising from abnormal vascular cell proliferation and inflammation, which result in occlusive arteriolar changes.

The underlying mechanisms of occlusive arterial remodeling in PAH remain poorly understood. Experimental studies, such as the SU5416 (SU) and chronic hypoxia (CH) rat model, provide insights into potential triggers. SU, an inhibitor of VEGF receptor 2, induces widespread endothelial cell (EC) apoptosis in combination with hypoxia, leading to apoptosis-resistant and hyperproliferative vascular cells, mimicking the plexiform lesions seen in clinical PAH. In this model, EC apoptosis was shown to be necessary for severe pulmonary hypertension (PH) and lesion formation, as caspase inhibition with Z-ASP abrogated both vascular remodeling and hemodynamic changes. However, other experimental models, such as the monocrotaline (MCT) model in rats, demonstrate EC apoptosis without forming complex intimal lesions. Interestingly, caspase inhibition also prevented MCT-induced PH, suggesting that EC apoptosis plays distinct roles in different PAH models, either through triggering dysregulated vascular cell growth or causing degeneration of precapillary arterioles.

Increased EC apoptosis has also been implicated in human PAH. Both hereditary and sporadic cases have been associated with loss-of-function mutations in the bone morphogenetic protein receptor 2 (BMPR2) gene. BMPR2, a member of the transforming growth factor-β (TGF-β) receptor family, regulates EC growth and survival. Mutations in BMPR2 heighten lung EC susceptibility to apoptosis, potentially contributing to PAH development when combined with environmental or genetic “second hits.” However, the lung-specific and region-specific nature of this response remains an enigma, given the widespread expression of BMPR2 across the body.

To test whether EC apoptosis alone is sufficient to induce PAH-like features, a conditional transgenic mouse model was developed using the Fas-Induced Apoptosis (FIA) system driven by the endothelial-specific Tie-2 promoter (EFIA mice). EC apoptosis was triggered by the administration of a small molecule dimerizer, AP20187. This intervention led to modest pulmonary arterial pressure increases and the formation of obliterative vascular lesions localized to distal lung arterioles. These lesions, characterized by cell proliferation, are highly reminiscent of the vascular changes seen in PAH. Importantly, similar arteriopathy was not observed in other organs, highlighting the unique susceptibility of the lung vasculature to EC apoptosis.

These findings establish, for the first time, that EC apoptosis is sufficient to induce complex obliterative arterial remodeling in the lung, a hallmark feature of PAH. This model offers critical insights into the mechanisms driving PAH and paves the way for exploring targeted therapeutic interventions to mitigate vascular remodeling and disease progression.

Materials and Methods

The LNGFR-FKBP-Fas Artificial Gene Construct:

The Fas-Induced Apoptosis (FIA) construct contains the cytoplasmic Fas death domain (FADD) bound to two modified FK506-binding protein (FKBP) domains, which are linked to the membrane-tethering region of the low affinity nerve growth factor receptor (∆LNGFR). FKBPs are a family of highly conserved proteins that are targets of immunosupressent drugs such as FK506 or rapamycin (10). In the FIA construct, AP20187 binds to a mutated FKBP domain, but not endogenous FKBP, and therefore has no known immunosuppressive activity or other biological activity other than dimerization of the FIA construct, activation of the death inducing signaling complex (DISC) and initiation of the caspase cascade (9). The FIA construct was a generous gift from Dr. Donald Cohen, University of Kentucky.

Generation of EFIA Transgenic Mice

The endothelial-specific Tie-2 promoter was cloned upstream of the FIA coding sequence to create the EFIA construct. This construct included a 2.1 kb Tie-2 promoter fragment (1.8kb of the upstream regulatory sequence and 318bp of the 5’ untranslated region of the first exon), combined with 1.7 kb of the enhancer element from the first intron required to produce robust expression in adult mice (11) (12). The EFIA construct was microinjected into ova from super-ovulated CD1 mice to generate viable EFIA transgenic mice. Seven transgenic founder lines were generated and identified from genomic tail DNA by PCR analysis using FIA- specific primers (1446F 5’-ACA TGC CAC TCT CGT CTT-3’; 1696R 5’-TGG CTT CAT TGA CAC CAT-3’), spanning a sequence region encoding the second FKBP and the FADD.

Activation with the AP20187 compound

AP20187 (AP) was a gift from Ariad Pharmaceuticals. Lyophilized AP was dissolved in 100% ethanol at concentration of 12.5 mg/ml and stored at -20°C. For in vivo use, the 12.5 mg/ml stock was diluted to 0.40 mg/ml in an injection solution (consisting of 4% ethanol, 10% polyethylene glycol-400 (PEG-400), and 2% Tween-20 in water). Each mouse was given daily intraperitoneal injections for seven days with 10mg/kg AP20187. Vehicle injections were performed in the same manner with 100% ethanol in the injection solution, in absence of AP20187. All experiments were approved by the University of Ottawa’s Animal Care Ethics Committee and complied with the principles and guidelines of the Canadian Council on Animal Care.

Hemodynamic Assessment

Direct right ventricular systolic pressure (RVSP) was measured following administration of ketamine (100 mg/kg) and xylazine (10 mg/kg). Right ventricular pressure was measured using a pressure catheter transducer and proper readings were determined by observing sequential pressure loops recorded on LabScribe2 software (Transonic Scisense Inc., London, Ontario). Right ventricle hypertrophy was assessed as a weight ratio of the right ventricle (RV) over the left ventricle plus septum (LV+S).

Statistical analysis

All data are presented as mean ± standard error (SEM). Results were analyzed using a one-way analysis of variance (ANOVA) with a post hoc Bonferroni test on GraphPad Prism statistical software. A value of p<0.05 was considered statistically significant. Results In vitro validation of the FIA construct Human pulmonary artery endothelial cells (HPAECs) transfected with the FIA transgene under control of the CMV promoter exhibited expression of NGFR on the membrane surface (Supplemental Figure E1). A dose-dependent increase in apoptosis was seen in response to increasing concentrations of dimerizering agent, AP20187, assessed by flow cytometry using Annexin-V staining (Supplemental Figure E2), up to a maximum of 85% at 3nM of AP. However, even in the absence of the dimerizing agent, there was a moderately basal level of apoptosis (~20%), which is consistent with some leakiness of this conditional system in the context of a strong promoter and relatively high levels of expression in vitro. No apoptosis was detected in untransfected HPAECs that had been treated with AP20187. In vivo validation of the EFIA Construct Three transgenic founder lines were initially selected for assessment based on their EFIA mRNA expression levels, measured through RT-qPCR. After establishing colonies, repeat testing identified two lines with the highest EFIA RNA expression—lines E6780 and E6711. Line E6711 was selected for detailed phenotypic characterization. Apoptosis levels in E6711 and wild-type (WT) mice were evaluated using TUNEL staining under both untreated conditions and following treatment with 10 mg/kg AP20187. Mice were sacrificed at various time points (2, 4, 8, 24, and 48 hours, as well as 1 week post-treatment). Under basal conditions, EFIA mice exhibited no abnormal apoptosis compared to WT vehicle-treated controls. However, in EFIA mice treated with AP20187, apoptosis increased in a dose-dependent manner, peaking at 8 hours post-treatment. Notably, no apoptosis increase was observed in WT mice treated with AP20187. These findings were further validated using immunostaining for activated caspase-3 and caspase-8. Interestingly, while EC apoptosis in EFIA mice was primarily localized to the lungs, it was also detected in other organs, particularly the kidney. Despite this, there was no evidence of renal pathology. No apoptosis was observed in the heart. This lung-specific apoptotic response provides valuable insights into the mechanisms underlying endothelial cell apoptosis and its role in pulmonary vascular remodeling. Hemodynamic Assessment of EFIA Transgenic Mice Under basal conditions, EFIA mice appeared normal and exhibited no increases in systemic (data not shown) or pulmonary arterial pressure. In contrast, administration of 2 or 10mg/kg AP20187 resulted in modest but significant increases in RVSP and RV hypertrophy (RV/LV+S) after only one week of treatment. Administration of AP20187 to WT mice had no effect on RVSP or RV remodeling. As well, there was no increase in systemic arterial pressure in WT or EFIA transgenic mice in response to AP20187 treatment. Lung Histology and Immunohistological Characterization In the EFIA mouse model, approximately 21% (12 out of 56) of the animals developed florid pulmonary vascular lesions in response to AP20187 administration. These lesions appeared as early as one week after endothelial cell (EC) apoptosis was induced. They were almost exclusively localized to small, distal intra-acinar arterioles and were often obliterative, encroaching on or entirely occluding the vascular lumen. The morphology and distribution of these lesions closely resembled the complex plexiform lesions characteristic of human pulmonary arterial hypertension (PAH). The lesions exhibited high levels of cell proliferation, as demonstrated by Ki67 immunostaining, and were further analyzed using immunohistochemistry with various cell surface markers. Staining for CD31 and CD34, markers of microvascular endothelium, revealed significant neovascularization within the lesions, indicative of an "angioproliferative" phenotype. CD68-positive macrophages were the predominant cell type within the lesions, while CD3-positive lymphocytes were present in smaller numbers. Evidence of ongoing apoptosis in the arterial lesions was detected through TUNEL staining. Interestingly, the lesions showed minimal to no expression of α-SMA, a marker for smooth muscle cells, and no evidence of abnormal muscularization was observed in the EFIA mice, as confirmed by further analysis. These findings emphasize the significant role of EC apoptosis in driving the formation of complex obliterative pulmonary lesions and provide a valuable model for studying the underlying mechanisms of vascular remodeling in PAH. Discussion: EC apoptosis is thought to trigger reactive changes in vascular cell growth and survival in the distal pulmonary arterial bed that leads to the development of complex, obliterative arteriopathy and progressive hemodynamic compromise resulting in severe PAH (5). While EC apoptosis may indeed be necessary for the development of PAH, it is unclear whether it is sufficient to produce the complex lung arterial lesions characteristic of this disease. Therefore, the present study sought to establish whether EC apoptosis, induced by endothelial targeted overexpression of a conditional apoptosis-promoting construct (EFIA), by itself could result in a PAH phenotype in transgenic mice. In the present report we show dramatic remodeling of the lung arterial bed in a proportion of EFIA mice, which was associated with hemodynamic changes indicative of PH. To our knowledge, this is the first demonstration that induction of apoptosis using a transgenic strategy targeted to the endothelium can reproduce many of the salient features of this disease, including increased pulmonary arterial pressure, RV hypertrophy and proliferative arterial lesions, which underscores the central role of EC injury and apoptosis in the pathogenesis of PAH. Fas is a member of the tumor necrosis factor (TNF) receptor superfamily. Dimerization of Fas by FasL leads to the recruitment of a Fas-activating death domain (FADD), activation of the caspase cascade and induction of apoptosis (14). The Fas-induced apoptosis (FIA) construct was designed to respond to a synthetic dimerizing agent, AP20187, which selectively binds the re- engineered FK506 binding protein (FKBP) domain, but not for the endogenous FKBP. The FIA construct, introduced into T lymphocytes with a retroviral vector, produced a dose-dependent depletion of T lymphocytes in response to AP20187 in vitro (14). Using macrophage-targeted, Dox-conditional FIA (MAFIA) transgenic mice, Burnet et al demonstrated up to 94% macrophage depletion following the administration of 10mg/kg AP20187 for five consecutive days (9). For the present study, we cloned the FIA construct into a plasmid vector containing an endothelial-selective Tie-2 promoter to develop endothelial-targeted Fas-induced apoptosis- (EFIA) transgenic mice. EFIA transgenic mice demonstrated a pulmonary hemodynamic response after AP20187 treatment, with modest increases in RVSP and RV remodelling (Figure 4). Interestingly, only a subset of these mice (~20%) exhibited exuberant proliferative vascular lesions. This observation suggests that other factors or influences are necessary in order for the occlusive pulmonary arterial phenotype to be manifested in the EFIA model, mirroring the relatively low penetrance of PAH in the context of disease-causing mutations. These additional factors may be epigenetic or environmental, or combination of the two. Arterial lesions in this model were localized exclusively to small arteries and arterioles and showed a number of features reminiscent of characteristic plexiform lesions found in patients with advanced PAH, including a high level of cell proliferation and lumen obliteration. As well, the lesions in the EFIA model appeared to be mainly inflammatory in nature, with a marked predominance of CD68-positive macrophages. It has been well established that inflammation plays an important role in the pathogenesis of PAH, particularly in the earlier stages of lesion formation (15), possibly driven by self-renewing resident macrophages (16). However, unlike human plexiform lesions, there was a paucity of smooth muscle staining in lesions of the EFIA model that might again point to an early stage of development. Despite the EFIA construct being broadly targeted to endothelial cells throughout the body via the Tie2 promoter, the pathological changes were specifically restricted to the lung vasculature. This aligns with the clinical pattern of idiopathic pulmonary arterial hypertension (IPAH), a condition localized exclusively to the lungs. Interestingly, a large proportion of IPAH cases involve mutations in the BMPR2 gene, which, while ubiquitously expressed, predisposes specifically to pulmonary vascular disease. This has been attributed to an increased susceptibility of pulmonary endothelial cells (ECs) to apoptosis, which is a recognized trigger in experimental pulmonary hypertension (PH) models. The unique vulnerability of the lung's fragile microvasculature, especially at the distal, precapillary arterioles, might contribute to this specificity. Supporting this, no EC apoptosis was observed in the myocardial microvasculature in the EFIA model, though apoptosis was detected in the kidney. However, unlike in the lung, the kidney exhibited no associated pathological consequences, highlighting the lung’s distinct response to EC injury. In healthy tissues, apoptosis is typically resolved through efferocytosis, a mechanism that clears cell debris and minimizes inflammation. However, the EFIA model and clinical PAH exhibit a striking inflammatory and proliferative vascular response. The lung’s constant interaction with the external environment may play a role here, as its robust innate immune system could amplify responses to arteriolar EC apoptosis. In earlier studies, targeting lung capillary ECs with peptides inducing apoptosis led to alveolar simplification and emphysema rather than PAH, further emphasizing the importance of the vascular region targeted. Similarly, chronic treatment with the VEGFR2 antagonist SU5416 induced emphysematous changes instead of PAH. While alveolar changes were not observed in the EFIA model, longer treatment periods might reveal overlapping pathology. Spontaneous lesion development in the E6780 transgenic line (without AP20187 administration) suggests spontaneous dimerization of the EFIA construct due to high membrane expression levels. This may reflect a "leakiness" of the system, as indicated by basal EC apoptosis in models overexpressing EFIA with the CMV promoter. Conversely, only a subset of E6711 EFIA mice developed proliferative lung lesions with AP20187 treatment, highlighting the low penetrance of this phenotype. Over time, this limitation, alongside generational exacerbation and challenges in obtaining the dimerizing agent, constrained the ability to further refine and study this model. This work provides insights into the mechanisms of lung-specific vascular remodeling and highlights the complexities of endothelial cell apoptosis in PAH pathogenesis, underscoring the challenges of modeling this condition experimentally. It is perhaps of interest that the marked vascular remodeling in this model appeared to be out of proportion to the more modest hemodynamic changes, even in those animals that exhibited abundant vascular lesions. However, mice appear to be inherently resistant to pulmonary hypertension and do not show the same robust response to monocrotaline pyrrole or SU/CH as are typically seen in rats (23). The modest hemodynamic phenotype in our transgenic model is also consistent with most other transgenic models including BMPR2 deficiency, and may relate to a species difference in susceptibility to PH (23). Moreover, a dissociation between the severity of complex arterial lesions and hemodynamic abnormalities has also been noted anecdotally in patients with PAH (16). This suggests that complex, occlusive arteriopathy may not be the only cause of the increase in pulmonary vascular resistance, and other mechanisms such as arteriolar degeneration and dropout play a more important role in these hemodynamic changes. This is also consistent with the findings of Abe et al. in the rat SU/CH model (24) in which plexiform-like lesions only appeared relatively late, well after the increase in pulmonary arterial pressure was fully established. In summary, this study shows that targeted expression of the FIA construct to the endothelium was sufficient to produce complex obliterative lung vascular lesions with the appearance of early plexiform-like lesions, together with pulmonary hypertension and RV hypertrophy. These data further strengthen the role of EC apoptosis as an initiating mechanism in the development of PAH; however further studies are needed to elucidate the precise molecular mechanisms that link this with the subsequent proliferative and inflammatory changes that result in progressive and irreversible arterial remodelling and PAH. Transgenic models that allow conditional and targeted induction of lung endothelial apoptosis provide a unique opportunity to uncover the downstream pathways that perpetuate arterial inflammation and abnormal vascular cell growth in this disease, and this may provide insights into novel therapeutic targets. Extended Materials and Methods: Generation of EFIA Construct The plasmid pSG5 (Stratagene, California) was strategically modified to express the NGFR-FB-Fas gene under the regulatory control of the mouse Tek2 promoter. The first step involved replacing the original SV40 promoter in pSG5 with the Tek2 promoter. This was accomplished by treating pSG5 with restriction enzymes Nsi I (site 97 in the original plasmid) and Stu I (site 348), which excised the SV40 promoter. Subsequently, a 2 Kbp fragment of the mouse Tek2 promoter, amplified via PCR, was ligated into the now promoter-less pSG5 backbone. To confirm successful incorporation, the modified plasmid—termed pSG5-Tek2—underwent DNA sequencing. The NGFR-FB-Fas gene, sourced from a plasmid kindly provided by Dr. Donald Cohen (University of Kentucky), was excised using BamH I and inserted into the Bgl II site of pSG5-Tek2 through ligation. This generated the final construct, pSG5-Tek2-NGFR-FB-Fas, which integrates the NGFR-FB-Fas gene under the control of the Tek2 promoter for directed expression. The technical precision in this plasmid engineering demonstrates a thoughtful approach to tailoring gene expression systems, ensuring robust promoter activity suitable for experimental needs. Histological Assessment (H&E, TUNEL, IHC, IF): The left lung was inflated with a 1:1 saline:OCT solution through the trachea then fixed (in 4% paraformaldehyde) and processed for paraffin-embedded sections. All hematoxylin & eosin (H&E) and immunohistochemistry (IHC) staining was conducted as per standard protocols. IHC antibodies used were Ki67 (BD Pharmingen), CD68 (Abd Serotec), CD31 (BD Pharmingen) and CD34 (Abcam). Images were acquired using the Aperio Light ScanoScope (Aperio). For the muscularization analysis, 100 vessels were analyzed in each animal lung cross section after staining with α-SMA (Sigma). The vessels were categorized as nonmuscularized (no apparent muscle), partially muscularized (with only a crescent of muscle) or fully muscularized (with a complete medical coat of muscle). Activated caspase antibodies used were against cleaved caspase 3 (CST, Asp175) and cleaved caspase 8 (CST, Asp387) and amplified using VectaFluor Excel Amplified Dylight 594 kit as per manufacturers’ instructions (Vector Labs). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining of lung tissue was following the manufacturer's protocols (Promega). TUNEL and IF slides were imaged using the Confocal Fluoview FV1000 (Olympus). Annexin-V Flow Cytometry Analysis For in vitro validation, the FIA construct was placed downstream of a CMV promoter within the pVAX1 plasmid and transfected into human pulmonary artery endothelial cells. Flow cytometry was performed to assess apoptosis following electroporation with the ΔLNGFR-FKBP-Fas construct and treatment with increasing concentrations of AP20187. Briefly, cells (1 x 105) were incubated in the presence of anti-NGFR-APC antibody (5μg/ml) (Chromoprobe) and annexin V-FITC (5% v/v) (BD Biosciences) to detect successfully transfected and apoptotic cells, respectively. Flow cytometry was performed using a Cytomics™ FC500 flow cytometer (Beckman Coulter) equipped with 488 nm argon and 633 nm helium-neon lasers. Real-Time (RT)-qPCR Lung samples for RNA analysis were placed in RNAlater® stabilizing solution and stored at -80oC. At time of analysis, dTAG-13 lung tissue was disrupted and homogenized using TissueLyser II (Qiagen) and total RNA was extracted using an RNeasy Mini Kit (Qiagen). RT reactions were performed with 1μg total RNA using a QuantiTECT-RT kit (Qiagen) and PCR assays were conducted using a CFX96 Real-Time System (Bio-Rad). RT-qPCR EFIA primers were designed using Intregrated DNA Technology (IDT) software (EFIA#2 For 5’-CTG TGG TTG TGG GTC TTG TG -3’; EFIA#2 Rev 5’-CCT CCC ATC CCC TAA TGA CT -3’), spanning the region encoding LNGFR and first FK-BP. Triplicates were carried out for each sample. All samples were normalized using 18s as an internal control. Relative quantification was performed using the ΔΔCT method.