Abstract Background Marfan syndrome (MFS) is a heritable connective tissue disorder caused by mutations in the Fibrillin‐1 gene, which encodes the extracellular matrix protein fibrillin‐1. Patients with MFS are predisposed to aortic aneurysms and dissections, significantly contributing to mortality. Emerging evidence suggests that endothelial cell (EC) senescence plays a critical role in the pathogenesis of aortic aneurysms in MFS. This study aims to elucidate the role of EC senescence in the development of aortic aneurysms in MFS using a vascular model derived from human induced pluripotent stem cells. Methods and Results We generated human induced pluripotent stem cells lines from 2 patients with MFS carrying specific Fibrillin‐1 mutations and differentiated these into ECs. These MFS–hiPSC‐derived ECs were characterized using immunofluorescence, reverse transcription‐quantitative polymerase chain reaction, and Western blotting. Functional assays including cell proliferation, scratch wound, tube formation, NO content detection, and senescence‐associated β‐galactosidase staining were conducted. RNA sequencing was performed to elucidate underlying signaling pathways, and pharmacological inhibition of the transforming growth factor‐beta pathway was assessed for its therapeutic potential. MFS–hiPSC‐derived ECs recapitulated the pathological features observed in Marfan aortas, particularly pronounced cellular senescence, decreased cell proliferation, and abnormal transforming growth factor‐beta and NF‐κB signaling. These senescent ECs exhibited diminished proliferative and migratory capacities, reduced NO signaling, increased production of inflammatory cytokines, and attenuated responses to inflammatory stimuli. Importantly, senescence and dysfunction in MFS‐hiPSCderived ECs were ameliorated by transforming growth factor‐beta signaling pathway inhibitor, SB‐431542, suggesting a potential therapeutic strategy. Conclusions This study highlights the pivotal role of endothelial cell senescence in the pathogenesis of aortic aneurysms in MFS. Our human induced pluripotent stem cells–based disease model provides new insights into the disease mechanisms and underscores the potential of targeting the transforming growth factor‐beta pathway to mitigate endothelial dysfunction and senescence, offering a promising therapeutic avenue for MFS. Keywords: endothelial cell senescence, fibrillin‐1, induced pluripotent stem cells, Marfan syndrome, TGF‐β Subject Categories: Aortic Dissection, Aneurysm, Vascular Disease __________________________________________________________________ Nonstandard Abbreviations and Acronyms AAD aortic aneurysm and aortic dissection ASMCs aortic smooth muscle cells ECM extracellular matrix EC endothelial cell FBN1 fibrillin‐1 hiPSCs human induced pluripotent stem cells MFS Marfan syndrome MFS‐iECs MFS‐hiPSC‐derived ECs TAAD thoracic aortic aneurysm and dissection TGF‐β transforming growth factor‐beta TJ tight junction VSMCs vascular smooth muscle cells α‐SMA α‐smooth muscle actin Wild‐type iECs wild‐type hiPSC‐derived ECs Clinical Perspective. What Is New? * Marfan syndrome (MFS) patient‐specific induced pluripotent stem cells–derived endothelial cells (ECs) have been used for disease modeling and revealed that EC dysfunction, including EC senescence, plays a pivotal role in MFS aortic aneurysm. * Through the RNA Seq, and relevant experiments, we demonstrated that MFS‐hiPSC‐derived ECs exhibited abnormal transforming growth factor beta signaling activation. * Inhibition of the transforming growth factor beta signaling pathway could attenuate the senescence of ECs from patients with MFS. What Are the Clinical Implications? * This study highlights that EC senescence might be a novel potential target for therapy of aortic aneurysms in patients with MFS. * Cell‐specific target of transforming growth factor beta pathway in ECs might offer a promising therapeutic avenue for aortic aneurysm in patients with MFS. Marfan syndrome (MFS) is a heritable autosomal dominant multisystem disorder of connective tissue, affecting ≈1 in 5000 individuals.[48] ^1 , [49]^2 About 75% of patients with MFS are predisposed to aneurysm/dissection,[50] ^3 , [51]^4 which are the leading cause of death in patients with MFS.[52] ^5 There are currently no effective pharmacological treatments, and only surgical replacement of the aortic root increases life expectancy in patients with MFS. MFS is caused by mutations in FBN1, which encodes fibrillin‐1, a major constituent of microfibrils found in the extracellular matrix. FBN1 gene mutation increased fibrillin protein degradation and elastic fiber fragmentation in the aortic wall, eventually leading to aortic root aneurysmal dilation and aortic dissection.[53] ^6 , [54]^7 Vascular smooth muscle cells (VSMCs) are considered critical in the pathogenesis of thoracic aortic aneurysm and dissection (TAAD), contributing to several dysfunctions, including phenotypic switching, increased senescence and death, production of inflammatory cytokines and matrix metalloproteinases, elevated reactive oxygen species, and defective autophagy.[55] ^8 Recent single‐cell RNA sequencing experiments in Marfan mice and patients have identified disease‐specific VSMC clusters.[56] ^9 , [57]^10 In addition, dysfunction of multiple cell types, including endothelial cells (ECs), may contribute to TAAD.[58] ^11 Recent studies have suggested that EC dysfunction is an early pathologic event in the TAAD and abdominal aortic aneurysm.[59] ^12 , [60]^13 For example, Gould et al. reported that ROBO4 variants in humans predispose individuals to bicuspid aortic valve and thoracic aortic aneurysm, knockout of ROBO4 or mutant ROBO4 expression in mice EC results in impaired barrier function and aortic aneurysm.[61] ^14 Yang et al. demonstrate that EC tight junction (TJ) dysfunction plays a role in the pathogenesis of TAAD.[62] ^15 Additionally, oxidative stress‐driven endothelial dysfunction and dysregulation of endothelial NO signaling has been suggested to be associated with TAAD.[63] ^16 , [64]^17 Interestingly, EC‐specific but not VSMC‐specific At1r inactivation significantly delayed death from ruptured TAAD in Marfan mice.[65] ^18 However, how EC dysfunction leads to TAAD is still largely unknown. It has been suggested that FBN1 defect in Marfan patient or mice, activated VSMC transforming growth factor‐β (TGF‐β) signaling, results in the increased SMAD3 and ERK signaling contribute to aneurysm progression.[66] ^19 However, pathogenic loss‐of‐function variants in the components of TGF‐β signaling, such as the ligands (TGFB2/3), the receptors (TGFBR1/2), and downstream effectors (SMAD2/3), are associated with a syndromic form of TAAD called Loeys‐Dietz syndrome. Many other research results also suggest that the knockout of the classical TGF‐β pathway can impair aortic function and promote aortic rupture.[67] ^19 However, the exact role of TGF‐β in TAAD remains unclear. Some evidence has shown that TGF‐β can regulate VSMC and vascular progenitor cell senescence in patients with MFS.[68] ^20 , [69]^21 Inhibition of TGFβ1 signaling significantly inhibits serum‐free‐induced endothelial cell senescence and thereby improves endothelial function.[70] ^22 Whether EC dysfunction, especially EC senescence, is related to human Marfan TAAD and TGF‐β signaling, would be involved in this process needs to be determined. Currently, most knowledge of MFS pathogenesis has been gained from studies in animal models.[71] ^23 , [72]^24 , [73]^25 , [74]^26 Study of the pathogenesis of TAAD in human MFS is limited, largely due to a lack of appropriate experimental models. On the other hand, due to the diversity of FBN1 gene variants in patients with MFS, the MFS animal model cannot fully replicate this feature. Patient‐specific iPSC modeling has provided us a unique platform to study the pathogenesis of many genetic diseases, such as MFS. In this study, we generated non‐integrative induced pluripotent stem cell (iPSC) lines from 2 patients with MFS. We revealed that MFS‐hiPSC‐derived ECs (MFS‐iECs) exhibited dysfunction, including senescence phenotype. RNAseq analysis indicated the TGF‐β pathway is involved in EC senescence. Pharmacological inhibition of TGF‐β reverses EC senescence, which is providing valuable clues for Marfan pathogenesis and developing a potential therapeutic strategy. METHODS Generation of Induced Pluripotent Stem Cell Lines The data that support the findings of this study are available from the corresponding author upon reasonable request. Sex was not considered a biological variable in our study. The aortic tissue was received from the Department of Cardiothoracic Surgery, The First Affiliated Hospital of Wenzhou Medical University. The first patient with MFS, male, 28 years old, clinical diagnosis: aortic root dilatation and aneurysm formation in the aortic sinus. The second patient with MS, male, 56 years old, clinical diagnosis: dilatation of the aortic sinus and ascending aorta, moderate‐to‐severe aortic regurgitation. Control patient with mitral valve insufficiency, male, 75 years old, clinical diagnosis of: severe mitral valve insufficiency and cardiac insufficiency. The generation of patient‐derived human induced pluripotent stem cells (hiPSCs) was approved by the medical research ethics committee of Wenzhou Medical University, with approval, number 2017‐066, and informed consent was obtained from all the patients. The investigation conformed to the principles outlined in the Declaration of Helsinki. Referring to the method proposed by Jin et al,[75] ^27 we performed the primary aortic smooth muscle cells (ASMCs) isolation and culture. Passage of 3–5 ASMCs were used for iPSC reprogramming. The episomal vector‐mediated integration‐free reprogramming was performed by AMAXA 4D Nucleofector (LONZA) as previous description with minor modification.[76] ^27 Four Episomal plasmids were nucleofected in the ratio 1:1:1:1 (3 μg total; pCXLE‐hUL, pCXLE‐hSK, pCXLE‐hOCT3/4‐shp53‐F and pCXWB‐EBNA1) into 1 × 10^6 ASMCs with Program FF‐120.[77] ^28 Cells were plated into 35 mm dishes with a density of 1.5 × 10^5 cells. After 6 days, 2 × 10^5 cells were replated on a 35 mm matrigel‐coated dish. Twenty‐four hours later, reflashed with the Nuwacell hiPSC/hESC medium‐ncEpic (Nuwacell, RP01001) plus sodium butyrate (0.25 mmol/L) and bFGF (10 ng/mL). Then the medium was changed every other day. About 3 weeks after nucleofection, iPSC colonies appeared. At least 6 colonies with hESC‐like morphology were picked up, and individual colonies were transferred into separate wells in a 12‐well plate coated with Matrigel. Culture medium was changed every day and cells were splited 1:3–1:10 every 4 to 5 days with Versene Solution (Gibco, 5 040 066). Passage of 20–40 iPSCs was used for iPSC relevant experiments. Alkaline Phosphatase Staining BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime, C3206) was used for AP staining as previously described.[78] ^27 EC Differentiation and Purification The EC differentiation protocol follows our previous report.[79] ^29 In short, hiPSCs were seeded at 3×10^5 cells/35 mm dish coated with Matrigel in Nuwacell hiPSC/hESC medium with 5 μmol Y27632 (Selleck). After 24 hours, mesoderm differentiation was initiated by adding N2B27medium containing CP21R7 (1 μmol/L, MCE) and BMP4 (25 ng/mL) for 48 hours. Then EBM‐2 complete medium with VEGF‐A (50 ng/ mL) and bFGF (25 ng/mL) was used to induce EC differentiation (days 4–7), and the medium was changed every day. iPSC‐derived ECs were dissociated into single‐cell suspensions with 0.05% Trypsin–EDTA (Gibco) at 37° C. Anti‐CD144 (VE‐Cadherin) magnetic microbeads (Miltenyi Biotec) were incubated with cell suspension at 4 °C for 20 minutes. Positive cells were then separated using LD columns combined with a Midi MACS Separator (Miltenyi Biotec). The positive cells (ECs) were expanded in ECM Bullet kit medium (Sciencell, #1001) after MACS separation. Passage of 4 to 7 iECs was used for EC‐relevant experiments. Reverse Transcription Polymerase Chain Reaction RNA was extracted using Trizol. Total RNA (1 μg) was reverse transcribed into cDNA using the PrimeScript RT kit (TaKaRa, #RR037A). qPCR was performed and analyzed in the StepOne Plus Real‐Time PCR System (Applied Biosystems) using iTaq Universal SYBR Green SuperMix (Bio‐Rad, #1725124). Primer sequences are listed in Table [80]S1. SDS‐PAGE and Immunoblot Analysis Cells were lysed in RIPA buffer and protein concentrations were determined by BCA protein analysis (Thermo Scientific, #23227). Protein aliquots were electrophoresed on 10% SDS‐PAGE and transferred to nitrocellulose Protran membranes (Whatman, Dassel, Germany). Blots were incubated for 1 hour at RT in blocking buffer that contained 5% milk powder in TBST (150 mmol/L NaCl, 10 mmol/L Tris, pH 8.0, 0.1% Tween‐20) and overnight at 4 °C with primary antibody. Blots were washed in TBST, and then incubated with secondary antibody at RT for 1 hour, washed, and used with SuperSignal West Pico chemiluminescent substrate (Thermo Scientific, #34580). Immunocytochemistry Before being fixed in 4% paraformaldehyde for 20 minutes at RT, cells were washed twice with PBS. Then the cells were permeabilized and blocked using 0.5% Triton X‐100 in PBS with 5% normal goat serum for 1 hour at RT, and incubated overnight with the primary antibodies at 4 °C. After washing, the blots were incubated with secondary antibodies for 1 hour at RT and washed before being visualized. Cell nuclei were stained with 4′,6‐ Diamino‐2‐phenylindole (DAPI). The antibodies used in this study are listed in Table [81]S2. Differentiation Potential Assay In Vitro To test the differentiation potential, the 90% confluence iPSCs were digested with collagenase IV (1 mg/mL) for ≈15 to 20 minutes at 37 °C and cultured in embryoid body medium (DMEM/F12 containing 20% KOSR (Knockout serum replacement), 10 mmol/L NEAA, 2 mmol/L L‐glutamine, 0.1 mmol/L 2‐mercaptoethanol) using a 60 mm ultra‐low attachment dish for 6 days. Then a gelatin‐coated 6‐well plate with DMEM containing 20% FBS is prepared for the embryoid bodies for 6 to 12 days. The culture medium needs to be changed every 2 days. The differentiated cells were fixed and stained with specific markers of the 3 germ layers. Karyotyping Karyotyping of iPSC was performed on G‐band metaphase chromosomes by FeiFan standard technical service Ltd. Briefly, the proliferating passage 3 of iPSCs were blocked by 50 ng/mL of colcemid for 2 hours, digested with trypsin–EDTA into single cell, then treated with hypotonic KCl solution for 20 to 40 minutes at 37 °C. Glass slides were prepared with 3 steps of fixation in methanol/glacial acetic acid (3:1). Then 20 metaphase spreads were counted. Cell Proliferation Curves Cells were seeded in 24‐well plates at 5×10^4/well. After that the cells were digested 2 wells each day and counted. This counting was carried out for 6 days. Finally, the number of cells recorded each day was plotted on a curve known as the cell growth curve. Scratch Wound Model in ECs For the scratch wound assay, cells were seeded onto 12‐well plates. When the culture has reached complete confluence, a thin linear scratch injury was made in the middle of the cell layer with an aseptic pipette tip. Then the wells were washed 3 times with DPBS. ECs migrating from the edge of the injured monolayer were photographed at 0, 6, 12, and 18 hours after scraping. The mobility ratio, which is the migrated cell region/scraped area, was analyzed by ImageJ. Tube‐Formation Assay Matrigel (250 μL) was dispensed into each well of a 24‐well plate (Corning) for 1 hour at 37 °C to allow the gel to solidify. The cells were seeded onto the matrix at a density of 1×10^5 cells per well in 250 μL of EBM‐2 medium and cultured for 12 hours at 37 °C. Total NO Content Detection Total NO content detection of the ECs was performed with a total Nitric Oxide Assay Kit (Beyotime, #S0023) according to the manufacturer's instructions. Senescence‐Associated β‐Galactosidase Staining The senescence of the ECs was performed with a senescence‐associated β‐Galactosidase Staining Kit (Beyotime, #C0602) according to the manufacturer's instructions. RNA‐Seq Data Processing Differential expression analysis was performed using the DESeq2 ([82]http://bioconductor.org/packages/release/bioc/html/DESeq2.html), Differentially expressed genes with |log2FC| >1 and Padj <0.05 were considered to be significantly different expressed genes. Volcano plots and heatmaps were visualized using the ggplot2 package in R. GO function enrichment and KEGG pathway enrichment analysis were performed using the clusterProfiler R software package. When P<0.05, it is considered that the GO or KEGG function is significantly enriched. Gene set enrichment analysis was conducted by clusterProfiler package in R. The genes were ranked according to the degree of differential expression in the 2 sets of samples, and then the predefined gene sets were tested to determine whether they were enriched at the top or bottom of the ranking table. The protein–protein interaction network was constructed from the String database ([83]https://cn.string‐db.org/). Using default settings for K nucleus, degree cut‐off, node score cut‐off, and a maximum depth of 100, the protein–protein interaction network was displayed using the Cytoscape software's molecular complex detection (MCODE) component. Statistical Analysis All data are analyzed using Prism (GraphPad Software, San Diego, CA). All data are expressed as the mean±SD of at least 3 independent experiments. Statistical comparisons were performed using Student t test (95% CI) and 1‐way ANOVA with Tukey test. Values of P<0.05 were considered statistically significant. RESULTS Generation of MFS‐Specific hiPSCs and Differentiation into EC Lineages For MFS disease modeling, we collected an aortic sample from 2 patients with MFS for primary cell culture and iPSC reprogramming. Clinical computed tomography imaging showed both patients presented an enlarged aortic aneurysm (Figure [84]1A). Whole exome sequencing analysis identified different mutations in the FBN1 gene in these patients. The first patient with MFS carried a missense mutation in exon28(c.C3442G, p. P1148A) and the second patient had a stop mutation in exon 37(c.C4567T, p. R1523X). These specific FBN1 mutations were confirmed by Sanger sequencing (Figure [85]1B and [86]1C). Then, we utilized the swissmodel website (https:// [87]swissmodel.expasy.org) to analyze the spatial location of P1148A, R1523X mutation site in the structure of the FBN1 protein (Figure [88]1D). Additionally, to check the effects of FBN1 mutations, we detected the protein expression in both patient‐specific iPSCs and their derived ECs. We observed a significant decrease in FBN1 protein expression in MFS‐iECs compared with the wild‐type one (Figure [89]1E and [90]1F), but no detectable FBN1 in iPSCs. Both mutations are predicted to disrupt the tertiary structure of FBN1, as indicated by the Varsite website ([91]https://www.ebi.ac.uk/thornton‐srv/databases/VarSite), which suggested that both of the mutated locuses may be deleterious (Figure [92]S1A). Using an episomal vector‐based nonintegration iPSC reprograming method, we established 2 Marfan‐specific iPSC lines and 1 wild‐type iPSC line (Figure [93]2A and Figure [94]S1B). These iPSC lines exhibited strong immunofluorescence signals for pluripotency markers Octamer‐Binding Transcription Factor 4 (OCT4), Nanog homeobox, SRY‐box transcription factor 2 (SOX2), TRA‐1‐60, TRA‐1‐81, as well as alkaline phosphatase staining (APS) (Figure [95]2B). RT‐PCR (Figure [96]2C) and Qpcr (Figure [97]2D) confirmed the high expression of pluripotency genes in these iPSC lines, in contrast to primary ASMCs. Karyotyping analysis revealed that all cell lines maintained normal chromosomal structure (Figure [98]2E). An embryoid body–mediated in vitro differentiation assay confirmed that these iPSC lines could differentiate into the 3 germ layers (endoderm, mesoderm, and ectoderm) as indicated by positive staining for α‐fetoprotein, α‐smooth muscle actin, and beta III tubulin, respectively (Figure [99]2F). Figure 1. Clinically relevant information about the patients with MFS. Figure 1 [100]Open in a new tab A, Clinical imaging of 2 patients with MFS. MFS‐1: the first patient with MFS; MFS‐2: the second patient with MFS. B and C, Sanger sequencing of FBN1 mutation (c.C3442G, p. P1148A/c.C4567T, p. R1523X) in the patients with MFS. D, Schematic diagram of the mutation of the structure site of the FBN1 3D protein. E, Immunoblot analysis of FBN1 in Wild‐type iECs and MFS‐iECs. Wild‐type iECs: Wild‐type hiPSC‐derived ECs; MFS‐iEC‐1: the first MFS hiPSC‐derived ECs; MFS‐iEC‐2: the second MFS hiPSC‐derived ECs. F, RT‐qPCR analysis of FBN1 in Wild‐type iECs and MFS‐iECs. Data are presented as the mean±SEM (n=3). *P<0.05, **P<0.01, ***P<0.001. Statistical significance was tested via 1‐way ANOVA with Tukey test in (E) through (F). FBN1 indicates Fibrillin‐1; MFS, Marfan syndrome; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; MFS‐iECs, MFS‐hiPSC‐derived ECs; and Wild‐type iECs, wild‐type hiPSC‐derived ECs. Figure 2. Generation and characterization of hiPSCs from patients with MFS and Wild‐type cells. Figure 2 [101]Open in a new tab A, Schematic diagram of nonintegrative (iPSCs) reprogramming mediated by episomal vectors. B, APS and immunofluorescence staining of pluripotent markers OCT4, NANOG, SOX2, TRA‐1‐60, and TRA‐1‐81. Scale bar=100 μm. Wild‐type iPSC. MFS‐iPSC‐1, MFS‐iPSC‐2. C, Pluripotency markers gene NANOG, SOX2, OCT4 expression analyzed by RT‐PCR. HASMC. H1ESC. D, RT‐qPCR analysis of pluripotency genes OCT4, NANOG, SOX2 expression. E, Karyotyping analysis of both Wild‐type iPSC and MFS‐iPSCs. F, Morphological diagrams of EB differentiation, scale bar=250 μm. EB‐mediated expression analysis of 3 germ layer markers, AFP, α‐SMA, and TUJ1, scale bar=100 μm. AFP indicates α‐fetoprotein; APS, Alkaline phosphatase staining; EB, embryoid body; HASMC, human aortic smooth muscle cell; H1ESC, H1 human embryonic stem cell; hiPSCs, induced pluripotent stem cells; MFS‐iPSC‐1, MFS patient induced pluripotent stem cell line 1; MFS‐iPSC‐2, MFS patient induced pluripotent stem cell line 2; NANOG, Nanog homeobox; OCT4, octamer‐binding transcription factor 4; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; SOX2, SRY‐box transcription factor 2; α‐SMA, α‐smooth muscle actin; TUJ1, beta III tubulin; and Wild‐type iPSC, wild‐type induced pluripotent stem cell. To explore the potential role of ECs in TAAD of patients with MFS, we differentiated these iPSCs into ECs following our previously reported protocol (Figure [102]3A and Figure [103]S1C).[104] ^29 , [105]^30 After EC purification, we observed that MFS‐iECs showed a larger and rounder morphology during culture compared with the wild‐type hiPSC‐derived ECs (wild‐type iECs) (Figure [106]3B). We further characterized both wild‐type iECs and MFS‐iECs by immunofluorescence, reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) and Western blotting (Figure [107]3C through [108]3E). Interestingly, additionally, MFS‐iECs showed lower mRNA and protein expression levels of endothelial markers, including CD31, VE‐Cadherin, and von Willebrand factor, compared with wild‐type iECs (Figure [109]3D and [110]3E). These results suggested that MFS‐iECs exhibit dysfunctional characteristics. Figure 3. Mesoderm‐mediated EC differentiation. Figure 3 [111]Open in a new tab A,, Schematic illustration of the EC differentiation process from iPSCs. B, Morphological diagram of purified ECs derived from wild‐type and MFS patient‐specific iPSCs. Scale bar=250 μm. C, Immunofluorescence staining of EC markers CD31 and VE‐Cad. Scale bar=100 μm. D, RT‐qPCR analysis of EC markers CD31, VE‐Cad, and vWF. E, Immunoblot analysis of EC markers CD31, VE‐Cad, and vWF. Data are presented as the mean±SEM (n=3). *P<0.05, **P<0.01, ***P<0.001. Statistical significance was tested via 1‐way ANOVA with Tukey test in (D), Student t test in (E). EC indicates, endothelial cell; ECM, extracellular matrix; EGM‐2, Endothelial Growth Medium ‐ 2; iPSCs, induced pluripotent stem cell lines; MFS‐iEC‐1, MFS‐hiPSC‐derived EC‐1; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; VE‐Cad, VE‐Cadherin; VEGF, vascular endothelial growth factor; vWF, von Willebrand factor; and Wild‐type iEC, wild‐type hiPSC‐derived EC. MFS‐iECs Exhibited Dysfunction During the maintenance and passage of iPSC‐EC culture, we observed a decreased proliferation ability in MFS‐iECs after several passages. Immunofluorescence staining revealed a lower percentage of Ki67^+ cells in MFS‐iECs compared with wild‐type iECs (Figure [112]4A). RT‐qPCR analysis confirmed a significant decrease in the mRNA levels of proliferation markers Ki67 and PCNA in MFS‐iECs (Figure [113]4B). Cell proliferation curves consistently demonstrated the reduced growth capacity of MFS‐iECs compared with wild‐type iECs (Figure [114]4C). We then evaluated the migration ability of MFS‐iECs using scratch tests and the results showed that the migration ability of both MFS‐iECs was decreased, and the percentage of wound closure for MFS‐iECs reduced to ≈50% in 18 hours (Figure [115]4D). To further detect EC dysfunction in MFS‐iECs, we checked the mRNA expression level of inflammatory cytokines by RT‐qPCR. The results showed that expression of ICAM‐1, IL‐1β, IL‐6, and IL‐8 in MFS‐iECs was all increased compared with that of the wild‐type iECs (Figure [116]4E). In addition, we checked the inflammatory response of iEC to tumor necrosis factor‐α stimulation. We found that tumor necrosis factor‐α treatment increased the inflammatory response of both wild‐type iECs and MFS‐iECs (Figure [117]4E). Interestingly, we noticed that the sensitivity of inflammatory response in MFS‐iEC is decreased, evidenced by the decline of the fold changes of the inflammatory cytokines' expression after tumor necrosis factor‐α treatment (Figure [118]4F). Additionally, tube formation assay experiment shows that the capillary tube formation ability of MFS‐iECs was attenuated (Figure [119]4G). As the NO and the eNOS activity is a key indicator of EC dysfunction, we demonstrated that the expression of eNOS in MFS‐iECs is down‐regulated by Western blotting and RT‐qPCR (Figure [120]4H and [121]4I), and the NO level was also reduced in MFS‐iECs (Figure [122]4J). All these findings suggested that MFS‐iECs exhibit dysfunctional phenotype characterized by reduced proliferation and migration abilities, increased production of inflammatory cytokines, a diminished sensitivity of inflammatory stimuli, and downregulation of NO signaling. These are typical features of senescent ECs. Figure 4. MFS‐iECs exhibited dysfunction. Figure 4 [123]Open in a new tab A, Immunofluorescence staining and quantitation of the proliferation marker Ki67. Scale bar=100 μm. B, RT‐qPCR analysis of the proliferation markers Ki67 and PCNA. C, Proliferation curves of ECs. D, iECs scratch tests and quantitation. Scale bar=250 μm. Wound closure (%)=(Initial scratch area−final scratch area)/(Initial scratch area)×100%. E, RT‐qPCR analysis of inflammatory markers in TNF‐α‐treated and untreated groups. F, Changes in mRNA levels of relevant inflammatory markers in TNF‐α‐treated and untreated ECs. G, Capillary tube formation assay from Wild‐type and MFS‐iECs. Scale bar=250 μm. H, Immunoblot analysis of eNOS in Wild‐type iECs and MFS‐iECs. I, RT‐qPCR analysis of eNOS in Wild‐type iECs and MFS‐iECs. J, NO signal detection in Wild‐type iECs and MFS‐iECs. Data are presented as the mean±SEM (n=3). *P<0.05, **P<0.01, ***P<0.001 for wild‐type iEC versus MFS‐iEC‐1; #P<0.05, ##P<0.01, ###P<0.001 for wild‐type iEC versus MFS‐iEC‐2; Statistical significance was tested via 1‐way ANOVA with Tukey test in (A) through (E), (H) through (I), Student t test in (F), (J). ECs indicates endothelial cells; eNOS, endothelial nitric oxide synthase; MFS‐iECs, MFS‐hiPSC‐derived ECs; PCNA, Proliferating Cell Nuclear Antigen. RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; and TNF‐α, tumor necrosis factor α. MFS‐iECs Showed Obvious Senescence Phenotype To confirm the senescence phenotype in MFS‐iECs, we performed SA‐β‐gal staining assays, which revealed a significant increase in the percentage of SA‐β‐gal positive cells in MFS‐iECs compared with Wild‐type iECs (Figure [124]5A). Furthermore, MFS‐iECs exhibited upregulation of the senescence markers p53 and p21 at both the protein and mRNA levels (Figure [125]5B and [126]5C). In addition, RT‐qPCR results indicated that MFS‐iECs displayed a senescence‐associated secretory phenotype, characterized by the expression of specific inflammatory cytokines, chemokines, and proteases, including matrix metalloproteinases (Figure [127]5E). Western blotting results also showed that phosphorylated p65, a key component of the NF‐κB pathway, was significantly activated in MFS‐iECs (Figure [128]5D). These findings confirm an obvious senescence phenotype in MFS‐iECs. Figure 5. MFS‐iECs showed an obvious senescence phenotype. Figure 5 [129]Open in a new tab A, SA‐β‐gal staining assay in Wild‐type iECs and MFS‐iECs. Scale bar=100 μm. (B) Immunoblot analysis of senescence‐related proteins P53 and P21. C, RT‐qPCR analysis of senescence‐related genes P53 and P21. D, Immunoblot analysis of phosphorylated p65, a marker of the NF‐κB pathway. E, RT‐qPCR analysis of SASP markers. Data are presented as the mean±SEM (n=3). *P<0.05, **P<0.01, ***P<0.001. Statistical significance was tested via 1‐way ANOVA with Tukey test in (A), (C), (E). Student t test in (B), (D) MFS‐iECs indicates MFS‐hiPSC‐derived ECs; MMP, matrix metalloproteinases; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; SASP, senescence‐associated secretory phenotype; and Wild‐type iECs, wild‐type hiPSC‐derived ECs. Changes in MFS‐iEC Transcriptional Profile To further investigate the potential signaling pathways involved in MFS‐iEC dysfunction, particularly EC senescence, we performed RNA sequencing. A total of 4011 differentially expressed genes were identified between MFS‐iECs and Wild‐type iECs, with 2256 genes upregulated and 1755 genes downregulated (|log2(fold change) | > 1, adjusted P value (padj) <0.05) (Figure [130]6A and [131]6B). KEGG enrichment analysis showed that the top 20 upregulated pathways included NF‐κB, TGF‐β, p53, and cellular senescence pathways. Conversely, the top downregulated pathways were mainly related to cell cycle regulation, DNA replication, and other essential cellular processes (Figure [132]6C). Gene set enrichment analysis and protein–protein interaction network consistently revealed that MFS‐iECs were enriched in genes associated with key signaling pathways, including the TGF‐β signaling pathway, p53 signaling pathway, cytokine‐receptor interactions, and the cell cycle (Figure [133]6D and [134]6E). GO analysis indicated that downregulated genes were closely associated with vascular dysfunction, cellular DNA replication, and various functional aspects of cellular senescence (Figure [135]6F). Overall, the transcriptional profiling of MFS‐iECs highlights that their dysfunction and senescence are closely related to key signaling pathways, including TGF‐β, NF‐κB, and cellular inflammation. These pathways play a crucial role in the observed EC senescence and associated dysfunctions in MFS. Figure 6. Changes in MFS‐iEC transcriptional profile. Figure 6 [136]Open in a new tab A, Heatmap illustrating differentiall–y expressed genes in Wild‐type and MFS‐iECs. B, Volcano plot showing the number of upregulated (red) and downregulated (green) genes in MFS‐iECs. C, KEGG analysis (left: KEGG UP right: KEGG down) showing the top 20 pathways with functional changes in MFS‐iECs. D, Gene set enrichment analysis (GSEA) plots showing representative KEGG‐related gene enrichments in MFS‐iECs. (E) PPI network analysis plot of MFS‐iECs. F, GO enrichment analysis BP indicates biological process, CC, cellular component; GSEA, gene set enrichment analysis; MF, molecular function; MFS‐iECs, MFS‐hiPSC‐derived ECs; PPI, protein–protein interaction; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; and SASP, senescence‐associated secretory phenotype. TGF‐β Signaling Pathway Involvement in Regulating MFS‐iEC Senescence RNA sequencing analysis revealed that many components of the TGF‐β signaling pathway were elevated in MFS‐iECs. These components included ligands (TGF‐β2, TGF‐β3), receptors (TGFBR2, ACVR1), and key mediators (SMAD2, SMAD3) (Figure [137]7A). We confirmed the increased expression of these genes using real‐time PCR, which was consistent with the RNAseq results (Figure [138]7B). Additionally, we detected significantly enhanced expression of phosphorylated SMAD2 and SMAD3 (p‐SMAD2, p‐SMAD3) in both MFS‐iEC lines (Figure [139]7C). Interestingly, we also found activation of the noncanonical TGF‐β signaling pathway component p‐AKT, but not p‐ERK1/2 (Figure [140]7D). Figure 7. TGF‐β pathway activation in MFS‐iECs. Figure 7 [141]Open in a new tab A, Heatmap of TGF‐β target genes in MFS‐iECs. B, RT‐qPCR analysis of selected TGF‐β signaling pathway marker genes. C, Immunoblot analysis of phosphorylated SMAD2 and SMAD3, markers of the canonical TGF‐β signaling pathway. D, Immunoblot analysis of phosphorylated AKT, a marker of the noncanonical TGF‐β signaling pathway. Data are presented as the mean±SEM (n=3). *P<0.05, **P <0.01, ***P<0.001. Statistical significance was tested via 1‐way ANOVA with Tukey test in (B), Student t test in (C and D). IPSC‐iECS indicates induced pluripotent stem cells‐derived ECs; MFS‐iECs, MFS‐hiPSC‐derived ECs; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; SMAD, Sma and Mad‐related protein; TGF‐β, transforming growth factor β; and Wild‐type iEC, wild‐type hiPSC‐derived EC. To further investigate whether MFS‐iEC senescence was mediated by the activation of the TGF‐β signaling pathway, we treated MFS‐iECs with SB‐431542, a TGF‐β signaling pathway inhibitor. Exposure to SB‐431542 for 24 hours resulted in a significant decrease in TGF‐β signaling, as evidenced by the reduced expression of p‐SMAD2 and p‐SMAD3 (Figure [142]8A). Moreover, treatment with SB‐431542 reversed the mRNA expression levels of senescence‐associated secretory phenotype components such as IL‐1β, IL‐6, and ICAM‐1, as well as matrix metalloproteinases in MFS‐iECs. The protein expression levels of phosphorylated p65, a central mediator of the inflammatory response, were also inhibited after SB‐431542 treatment (Figure [143]8B and [144]8C). Importantly, the senescence markers p53 and p21 were decreased at both the mRNA and protein levels following treatment (Figure [145]8B and [146]8D). The results of β‐gal staining further demonstrated that SB‐431542 treatment alleviated senescence in MFS‐iECs (Figure [147]8E). These findings suggest that the activation of the TGF‐β signaling pathway plays a crucial role in mediating senescence in MFS‐iECs. Inhibition of this pathway with SB‐431542 effectively mitigates the senescence phenotype, indicating potential therapeutic avenues for treating endothelial dysfunction in MFS. Figure 8. TGF‐β signaling pathway is involved in regulating MFS‐iEC senescence. Figure 8 [148]Open in a new tab A, Immunoblot analysis of phosphorylated SMAD2 and SMAD3 after MFS‐iECs were treated with SB‐431542 for 24 hours. B, RT‐qPCR analysis of senescence‐associated genes after MFS‐iECs were treated with SB‐431542 for 24 hours. C, Immunoblot analysis of phosphorylated p65 after MFS‐iECs were treated with SB‐431542 for 48 hours. D, Immunoblot analysis of senescence‐related proteins P53 and P21 after MFS‐iECs were treated with SB‐431542 for 48 hours. E, SA‐β‐gal staining assay of MFS‐iECs after treatment with SB‐431542 for 48 hours. Scale bar=100 μm. Data are presented as the mean±SEM. (n=3). *P<0.05, **P<0.01, ***P<0.001. Statistical significance was tested via Student t test in (A through E). MFS‐iECs indicates MFS‐hiPSC‐derived ECs; MMP, matrix metalloproteinases; RT‐qPCR, reverse transcription‐quantitative polymerase chain reaction; SB, SB‐431542; SMAD, Sma and Mad‐related protein and TGF‐β, transforming growth factor β. DISCUSSION In this study, we generated patient‐specific iPSCs from 2 patients with MFS and differentiated them into ECs to model disease mechanisms. Our findings demonstrated that MFS‐iECs exhibited significant dysfunction, including cellular senescence. We identified the involvement of TGF‐β and NF‐κB signaling pathways in MFS‐iEC dysfunction. Notably, inhibition of TGF‐β signaling partially reversed the senescence phenotype in MFS‐iECs, suggesting that TGF‐β signaling‐mediated EC senescence may contribute to the development of aortic aneurysm/dissection in patients with MFS. Although numerous studies have indicated that VSMC dysfunction, such as phenotypic switching and apoptosis, plays a critical role in the development of aortic aneurysm/dissection, the underlying molecular and cellular processes remain largely unknown. Recent advancements, including single‐cell RNA sequencing, have revealed that ECs play a pivotal role in the initial stages of aortic aneurysm and dissection pathogenesis.[149] ^11 , [150]^12 , [151]^13 Several studies support this hypothesis. For example, EC‐specific, but not VSMC‐specific, knockout of At1r significantly delayed mortality due to ruptured thoracic aortic aneurysm/dissection in MFS mice and AngII‐induced mouse TAA models.[152] ^18 , [153]^31 Genetic evidence also showed that pathogenic variants in ROBO4, a transmembrane receptor, impaired EC barrier function and predisposed individuals to bicuspid aortic valve and TAA.[154] ^14 Yang et al. used single‐cell RNA sequencing analysis to reveal aberrant endothelial TJ protein expressions in the thoracic aortas of patients with TAAD.[155] ^15 Treatment with AT‐1001, a protease‐activated receptor 2 inhibitor that seals TJs, alleviated impairment of endothelial TJ function and subsequently reduced TAAD incidence, indicating that disruption of endothelial TJ function is an early event before TAAD formation.[156] ^15 Our study, along with others, supports the notion that EC dysfunction is an early key determinant in the initiation and progression of aortic aneurysm/dissection.[157] ^13 , [158]^32 Our study utilized Marfan patient‐specific iPSC disease modeling to demonstrate that ECs derived from 2 Marfan patient iPSCs exhibited dysfunction, including decreased proliferation and migration abilities, higher expression of senescence‐associated secretory phenotype cytokines, and a senescence phenotype. However, the molecular mechanisms underlying EC dysfunction leading to aortic aneurysm/dissection require further exploration. Beyond EC barrier impairment, dysregulation of EC NO signaling has emerged as an important mechanism in aortic aneurysm and aortic dissection (AAD). For instance, Oller et al. reported higher aortic NO and Nos2 levels in Adamts1‐deficient mice and MFS mouse models, while Nos2 inactivation protected both types of mice from aortic pathology.[159] ^17 Recently, Luo et al. linked impairment of the endothelial cystathionine γ‐lyase/hydrogen sulfide (H2S) system to AAD, showing that the endothelial ZEB2‐HDAC1‐NuRD complex transcriptionally represses cystathionine γ‐lyase expression, impairing protein disulfide isomerase S‐sulfhydration and driving AAD.[160] ^33 In our current study, we identified EC senescence as a potential contributor to AAD in patients with MFS. EC senescence is known to promote EC dysfunction and various age‐related vascular diseases,[161] ^34 such as atherosclerosis, stroke, diabetes, and pulmonary hypertension.[162] ^35 Although EC dysfunction has been associated with aortic aneurysm/dissection, there is limited literature specifically linking EC senescence to AAD. To investigate this connection, we performed RNA sequencing and identified that endothelial TGF‐β signaling is a key pathway involved in EC senescence and AAD in MFS. It was previously thought that FBN1 mutations in MFS decreased FBN1 expression in VSMCs, leading to TGF‐β signaling overactivation and causing AAD. However, subsequent studies found that AAD was not mitigated after knockdown of TGF‐β signaling pathway‐related genes, such as TGF‐βR1, TGF‐βR2, SMAD3, and TGF‐β2.[163] ^36 Additionally, blocking TGF‐β systemically accelerated AAD development, indicating a complex role for TGF‐β signaling in preventing thoracic and abdominal atheromatous diseases[164] ^37 , [165]^38 Recent single‐cell RNA sequencing analysis of aortic aneurysm tissues from patients with MFS revealed TGF‐β impairment in VSMCs, fibroblasts, and ECs, although the potential mechanisms remain unknown.[166] ^39 Consistent with these findings, our Marfan iPSC disease model showed aberrant activation of TGF‐β signaling in ECs. Interestingly, Zhu et al. reported that impairment of VSMC TGF‐β signaling by VSMC‐specific disruption of the type II TGF‐β receptor gene (Tgfbr2) in newborn mice resulted in endothelial dysfunction.[167] ^40 These studies suggest that abnormal TGF‐β signaling in both VSMCs and ECs is associated with AAD. Our study indicated that dysregulation of the TGF‐β signaling pathway in ECs might result in EC dysfunction, particularly senescence. Inhibition of TGF‐β signaling with SB‐431542 reversed the senescence phenotype in ECs. TGF‐β signaling has been reported to regulate cellular senescence, including in aortic VSMCs and vascular progenitor cells in MFS.[168] ^20 , [169]^21 However, our study has some limitations. We did not use a Marfan mouse model to further investigate how EC senescence contributes to AAD progression in MFS. Additionally, other signaling pathways or mechanisms, such as metabolic regulation, may also be involved in EC senescence and warrant further investigation.[170] ^41 CONCLUSIONS In summary, our study demonstrated that Marfan iPSC‐derived ECs exhibit senescence and other dysfunctions mediated by TGF‐β signaling. Targeting EC TGF‐β signaling may represent a novel therapeutic strategy for AAD in MFS. Sources of Funding This work was supported by the National Natural Science Foundation of China (32370787, 82070487, 81670454), Major scientific and technological innovation projects of Wenzhou Municipal Science and Technology Bureau (ZY2022009) and the Scientific Research Start‐up Fund of Wenzhou Medical University (QTJ15029). Disclosures None. Supporting information Tables S1–S2 Figure S1 [171]JAH3-14-e037826-s001.pdf^ (815KB, pdf) Acknowledgments