Human pluripotent stem cells (hPSCs), including embryonic stem cells (ESCs) and induced PSCs (iPSCs), are characterized by unlimited proliferation and the ability to differentiate into three germ layers (ectoderm, mesoderm, and endoderm). ESCs are derived from the undifferentiated cell mass of human embryos, whereas iPSCs can be regarded as “artificial ESCs” obtained by reprogramming adult cells such as peripheral blood mononuclear cells, urine cells, skin fibroblasts, or hair keratinocytes with transcription factors[1]. Compared with ESCs, iPSCs overcome ethical issues such as embryo damage during ESC isolation, the limited cell source, and the immunogenicity of allogeneic transplantation. The availability of abundant cell sources has notably increased patients’ psychological acceptance of sampling and transplantation, thereby enhancing promising clinical prospects in regenerative medicine, drug screening, disease modeling, and other related areas.

In 2006, Japanese scientists Takahashi and Yamanaka[2] discovered that when the genes related to stem cell pluripotency and self-renewal, including octamer-binding transcription factor 4 (Oct4), the sex-determining region Y-frame protein 2 (Sox2) gene, the proto-oncogene (C-MYC), and the epidermal zinc finger factor-related (KLF4) gene, were introduced into mouse fibroblasts by retroviruses, adult cells can be de-differentiated into cells with the potential for self-renewal and multidirectional differentiation. Such cells derived from somatic ones were called iPSCs. Subsequently, Takahashi et al[3] introduced these four pluripotent transcription factors into human skin fibroblasts to generate the first human iPSCs (hiPSCs) lines.

However, due to the existence of proto-oncogene C-MYC and the integration of retrovirus, this method has some limitations, such as the possibility of tumorigenesis, low induction efficiency, and poor ability of directed differentiation. To solve these problems, scientists performed further selection and optimization of transcription factors used for reprogramming. For example, in 2007, Yu et al[4] obtained iPSCs using different combinations of transcription factors (Oct4, Sox2, Nanog, and Lin28). In 2008, Kim et al[5] successfully reversed the pluripotency of neural stem cells using only Oct4 and Klf4. Subsequently, only transcription factors Oct3/4 have been considered indispensable, while Sox2, Klf4, and c-Myc have been considered substitutable[6]. Yu et al[7] obtained iPSCs by the non-integrated episomal method, providing an idea to avoid tumors caused by viral vectors. Anokye-Danso et al[8] demonstrated that expression of the miR302/367 cluster rapidly and efficiently reprograms somatic cells to an iPSC state without the need for exogenous transcription factors. Hou et al[9] used small-molecule compounds to reprogram mouse somatic cells into iPSCs. In 2017, Kim et al[10] showed that mechanical stretch stimulation could significantly increase the reprogramming efficiency. In the same year, Blanchard et al[11] discovered a new method of using antibodies to replace transcription factors during reprogramming, avoiding the potential risk of transcription factor introduction. In 2022, Guan et al[12], who created an intermediate plastic state that overcomes the barriers of human somatic cells to chemical stimulation due to a stable epigenome and reduced plasticity, successfully reprogrammed human somatic cells into pluripotent cells using exposure to small molecule chemical substances.

Until now, with research progress, multiple methods to reprogram various types of somatic cells into iPSCs have been developed[13], allowing targeted differentiation of iPSCs into various cells, tissues, and organs. Since then, enormous progress has been made in stem cell biology[14]. For example, iPSCs can generate vascular endothelial cells (VECs), vascular smooth muscle cells (VSMCs), and three-dimensional (3D) vascular organoids through a series of differentiation, providing new hope for disease modeling, drug development, and screening, as well as regenerative medicine research for vascular disorders.

The induced differentiation of iPSCs into blood vessels is basically divided into two types: 2D cell differentiation and 3D vascular organoid differentiation.

Previous studies showed that type IV collagen[31], retinoic acid[32,33], and growth factors platelet-derived growth factor-BB (PDGF-BB)[34,35] and TGF-β1[36-38] were involved in inducing the differentiation of VSMCs. In 2012, Ge et al[39] rapidly induced iPSCs into VSMCs by the method of EB formation according to the previously established human ESCs (hESCs) scheme. The brief process was as follows. First, iPSCs were differentiated into EBs and cultured in suspension for 6 d. EBs were transferred to gelatin-coated dishes and cultured with fresh DM for another 6 d. The cells were digested and placed on Matrigel-coated dishes, followed by culture in VSMCs growth culture medium (SmGM-2) for 1 wk. Then, the cells were transferred to gelatin-coated dishes and cultured in a serum-containing medium for 5 d. This method can induce VSMCs with up to 95% of calponin-positive cells without sorting by flow cytometry, being the simplest EB method to obtain high-purity VSMCs.

Wanjare et al[40] inoculated iPSCs onto plates coated with type IV collagen and cultured them in DM for 6 d. The differentiated cells were then reseeded on type IV collagen-coated plates, and 10 ng/mL PDGF-BB and 1 ng/mL TGF-β1 were added for final VSMCs differentiation. In this method, the derivative of synthetic VSMCs and contractility VSMCs were generated by changing the concentration of serum, PDGF-BB, and TGF-β1 in DM. Some studies also focused on the effects of intrinsic substrate properties (such as the elastic modulus and cross-linking density of hydrogels in a 3D static and dynamic environment) on the phenotypic transformation between synthetic and contractile types of human mural cells derived from hiPSC-derived organoids (ODMCs). They demonstrated that the phenotypic plasticity of ODMCs in response to 2D biological and 3D biological mechanical stimuli is equivalent to that of primary human VSMCs[41].

The main difficulty in studying the differentiation of iPSCs into VSMCs and pericytes is that they have different developmental origins, which are not confined to the mesoderm but can also originate from the neural crest. In 2014, Cheung et al[42] successfully induced iPSCs into neuroectoderm, lateral plate mesoderm, and paraxial mesoderm and then induced differentiation of cells from these different germinal layers to VSMCs by optimizing different combinations of small-molecule compounds. Thus, directional induction of VSMCs from different germinal sources was achieved. Afterward, Patsch et al[43] established a fast and efficient induction scheme using monolayer cell induction. First, cells were differentiated into mesodermal cells using WNT signaling pathway inhibitor BMP4; then, activin-A and PDGF-BB were used for targeted differentiation of mesoderm cells. The entire procedure took 6 d to complete, and the induced cells were of high purity.

Furthermore, some researchers targeted intermediate cell populations such as cardiac progenitor cells and mesenchymal stem cells (MSCs). These intermediate cell populations derived from hPSCs, including hESCs and hiPSCs, play a significant role in the derivation of hPSCs to VSMCs. Park et al[22] demonstrated that CD34⁺ progenitor cells are an important source of VSMCs. Mesodermal cell lines were obtained from hiPSCs treated with PD98059 (a MEK/ERK pathway inhibitor) and BMP4, followed by VEGF165 and bFGF treatment for 6 d to obtain CD34⁺ cells. VSMCs were cultured in endothelial cell growth medium-2 (EGM-2) medium supplemented with PDGF-BB and bFGF for 15-21 d. Bajpai et al[44] generated VSMCs from an intermediate stage of MSCs using soluble signaling and ECM molecules. MSCs were treated with TGF-β1 (10 ng/mL) and heparin (30 mg/mL). After 5 d, MSCs expressed markers of VSMC differentiation, such as myosin heavy chain, α-SMA, calponin, and calmodulin. Zhang et al[45] cultured iPSC-MSCs in EGM-2 medium containing sphingylcholine (5 mmol/L) and TGF-β1 (2 ng/mL). After 3 wk, 50%-60% of the differentiated cells expressed VSMC markers, such as α-SMA and calponin 1. The cells also exhibited a spindle-like morphology and could contract under phenol treatment. Cao et al[46] developed an effective method for generating populations of early cardiovascular progenitor cells (CVPCs) from hiPSCs, key early developmental pathways involved in human cardiovascular norms and CVPCs self-renewal are modulated using a chemo-definition system containing bone BMP4, GSK3 inhibitors CHIR99021, and ascorbic acid. CVPCs were differentiated into VSMCs after 12 d of stimulation with 10 ng/mL PDGF-BB and 2 ng/mL TGF-β1. Additionally, CVPCs can also differentiate into other cardiovascular cells (such as cardiac muscle cells and endothelial cells) with high purity. Further studies found that tensile strain (cyclic uniaxial or circumferential) can increase elastin deposition and cell alignment and improve contractile responses, which further promote the maturation of hiPSC-derived VSMCs[47,48].

Kusuma et al[49] induced the co-differentiation of hPSC to generate early vascular cells (EVCs), which can mature into endothelial cells and pericytes. These cells can self-assemble to form microvascular networks in an engineered matrix. Monolayer cultures were employed without the need for specific differentiation-inducing feeder layers, EB formation, or sorting. Briefly, researchers developed a stepwise vascular lineage differentiation, in which EVCs were cultured in a medium supplemented with TGF-β inhibitor SB431542 and high VEGF concentrations to eliminate the pericyte-mediated inhibition of endothelial cell growth that often occurs when pericytes and endothelial cells are co-cultured[50]. VEcad⁺ cells were induced early during differentiation to ensure endothelial maturation. EVCs obtained in this way were highly purified (> 95%) for CD105 and CD146 expression, which are surface antigens common to ECs and pericytes. This population, composed of CD105⁺CD146⁺VEcad⁺ and CD105⁺CD146⁺ PDGFRβ⁺ subtypes, contained the cells required for microvascular construction and formed a network within collagen and hyaluronic acid (HA)-based hydrogel systems. This integrative approach leverages the inherent self-assembly ability of derived two-cell populations to create microvessels in a deliverable matrix, which has tremendous implications for vessel construction and regenerative medicine.

Orlova et al[51] established 2D co-culture of hPSC-derived ECs and pericytes. This protocol overcomes the shortcomings of previous efforts to standardize the culture and differentiation conditions of hPSCs by not using fully defined reagents, often leading to difficult-to-replicate results or low and difficult-to-scale up differentiation efficiency. In this protocol, mesodermal differentiation of hPSC was induced by BMP4, activin-A, GSK3β inhibitor CHIR99021, and VEGF. Mesenchymal cells were observed on the third day of differentiation. The mesoderm-inducing factor was removed and replaced with a standard medium supplemented with VEGF and TGF-β inhibitor SB431542 to support EC proliferation. On day 10, ECs were isolated using anti-CD31 antibody-coupled magnetic beads. If necessary, pericytes could be obtained from CD31- cells while ECs were isolated. Subsequently, ECs and pericytes could be co-cultured. However, ECs would first adhere to the substrate and form EC islands surrounded by pericytes, which can organize into vascular tube-like structures mimicking the formation of primary vascular plexuses. This co-culture approach allows the study of endothelial network formation and endothelial cell-pericyte interactions. For example, if the disease-affected vascular cell type is unknown, healthy donor or patient-derived ECs can be co-cultured with pericytes from a healthy or affected individual.

The induction of 2D vascular cells, including ECs and SMCs, from patient-specific iPSCs is a crucial step in constructing patient-specific disease models. For instance, VECs induced from iPSCs have been successfully utilized to investigate the pathophysiological mechanisms of familial pulmonary arterial hypertension (FPAH). Researchers discovered that compared to ECs derived from healthy control iPSCs, ECs derived from patients with BMP receptor 2 mutations exhibit reduced adhesion, migration, and angiogenesis potential. This study played a critical role in identifying, evaluating, and developing alternative treatment options for FPAH patients[67]. In another study, researchers utilized SMCs derived from iPSCs to investigate the pathological mechanism of aortic aneurysm (AA) in Marfan syndrome. The model successfully replicated the phenotype observed in the aorta of Marfan syndrome patients, aiding the identification of new therapeutic targets (such as p38 and Kruppel-like factor 4) and offering an innovative human platform for testing new drugs[68].

Using a co-culture model of iPSC-derived ECs and mural cells, researchers investigated the role of the ECM at the junction of blood and brain in the pathological progression of COL4A1/A2 mutation-related cerebral small vessel disease. This study found that the mutation can induce mural cell apoptosis, migration defects, ECM remodeling, and transcriptome changes. Importantly, these mural cell defects have a detrimental effect on endothelial cell tight junctions through a paracrine pathway. The COL4A1/A2 mutant model also expresses high levels of matrix metalloproteinases. Inhibition of matrix metalloproteinase activity can partially prevent ECM abnormalities and mural cell phenotype changes[69].

AA is a potentially fatal disease characterized by aortic dilatation that can only be treated by surgery or endovascular procedures. The underlying mechanism of AA is still unclear due to the heterogeneity of the segmental aorta and the limitations of existing disease models, resulting in insufficient investigation into early preventive treatment. Liu et al[70] established a comprehensive lineage-specific VSMCs-on-a-chip model using hiPSCs to generate cell lineages representing different segments of the aorta. They integrated this organ chip model with high-tech tools, such as transcriptomics, and conducted many experiments, including RNA sequencing, real-time reverse transcriptase-polymerase chain reaction, immunofluorescence, western blotting, and FACS analyses, to study the heterogeneity of the response of the segmental aorta to tensile stress and drug testing. These differences might be associated with the distinct transcription profiles of different lineage-specific VSMCs under tension stress. Furthermore, the organ chip exhibited contractile physiology and perfect fluid coordination, making it conducive to drug testing and demonstrating heterogeneous segmental aortic responses. This model has been evaluated as a novel and appropriate complement to animal models of AA for identifying differential physiological and pharmacological responses in different parts of the aorta. The system has the potential for future disease modeling, drug testing, and personalized treatment for patients with AA.

3D vascular organoids also offer unique advantages over traditional 2D cell culture models in the study of vascular diseases. These vascular organoids, derived from patients, closely resemble natural blood vessels, containing important vascular cell types such as mural cells and ECs and retaining patient-specific metabolic memory. This feature enables blood vessel organoids to better simulate natural pathophysiological responses. In 2019, Wimmer et al[71] constructed a model of diabetic vasculopathy using vascular organoid technology. In another study, researchers constructed a vascular organoid model of CADASIL, offering a new opportunity to explore gene therapy strategies[72]. In a study investigating the mechanisms of endothelial injury and coagulation disorders caused by severe acute respiratory syndrome coronavirus 2[73], human vascular organoids derived from iPSCs were used to model endothelial damage resulting from viral infection. Longitudinal serum proteomics analysis identified abnormal complement patterns driven by the amplification cycle regulated by complement factors B and D (CFB and CFD) in critically ill patients. This aberrant complement pattern initiated endothelial damage, neutrophil activation, and organoid-derived human blood vessel-specific thrombosis, as confirmed by in vivo imaging. These results highlight the role of the alternative complement pathway in exacerbating endothelial injury and inflammation and suggest the potential of targeted therapy for CFD in severe virus-induced inflammatory thrombosis outcomes.

These models not only help us understand the pathological mechanisms of vascular diseases and study the common molecular mechanisms shared between rare and common vascular diseases but also provide potential treatment targets and new avenues for studying vascular diseases.

Although iPSCs have addressed numerous ethical concerns associated with ESCs and offer significant potential for applications, research related to them has still sparked many ethical debates. For example, the transplantation of patient-derived iPSCs and their derived cells or organoid products into non-human animals to observe in vivo phenotypes has raised ethical concerns regarding the creation of chimeric organisms. Chimeric research falls under the category of animal research, and many animal studies are harmful, fatal, and non-therapeutic experiments conducted on vulnerable groups lacking the capacity to provide consent, aiming to benefit less vulnerable groups. Furthermore, many people are concerned that chimeric research raises questions about playing the role of God: When scientists use human cells to create non-human animals, they are essentially creating new creatures and crossing traditional species boundaries. In this process, scientists might be perceived as violating the inherent dignity of human beings by creating and destroying new forms of life for their own purposes[76]. Thus, balancing scientific progress and ethics remains a complex issue that cannot be ignored. It necessitates sustained dialogue and joint efforts from all parties. Furthermore, the establishment of a regulatory framework is necessary to ensure the safe and reliable use of PSCs as a powerful tool in scientific research and clinical applications.

There are still numerous challenges in the study of iPSCs and their derived vascular cells or vascular organoids. The first consideration is the efficiency and safety of the reprogramming techniques used to generate iPSCs. Some studies found that current reprogramming methods might result in genetic abnormalities and epigenetic changes that can impact cell behavior[77]. Furthermore, the tumorigenicity of iPSCs makes their medical safety controversial. Local microenvironment changes resulting from the transplantation of iPSCs and their derivatives might also induce carcinogenic changes in host cells[78]. Moreover, iPSCs and their derivatives have inherent limitations. The differentiation of iPSCs into 2D vascular cells is valuable for creating patient-specific in vitro models to study the pathophysiology of individual blood vessel cell types. However, these approaches cannot reflect the interactions among different cell types or the response of blood vessels to varying blood flow parameters. Tissue-engineered vascular grafts, which are more complex vascular models with a certain integrity, lack the ability to fully represent all developmental stages and might overlook crucial intermediate cell types. Although 3D vascular organoids encompass various cell types, they only partially simulate the function and structure of blood vessels, making it challenging to completely replicate the complexity of real blood vessels. Additionally, they can only maintain activity and function for a limited duration. Maintaining long-term stability and fully reflecting the interaction and overall effect between different organs of the body are difficult. Moreover, the cost of establishing and maintaining iPSC technology is high, and the requirements for experimental conditions and technology are strict. Difficulties in expanding the scale of production also cannot be ignored.

While there are still some challenging problems associated with iPSC technology that need further resolution, recognizing that opportunities and challenges always go hand in hand is important. Personalized medicine is gaining momentum, and the development of NGS technology has significantly enhanced our ability to identify genes with pathogenic changes in rare diseases[79,80]. Once pathogenic gene mutations are identified in patients, it becomes possible to generate iPSCs and their induced 2D vascular cells, vascular organoids, organ chips, and tissue-engineered vascular grafts. These in vitro models can be utilized as patient-specific disease models to facilitate further studies of pathological mechanisms and provide a platform for high-throughput drug screening. The therapeutic potential of iPSCs is also receiving significant attention. Currently, the therapeutic effectiveness of transplanting iPSCs and their derived cells has been validated in several clinical and animal studies. The use of gene editing to correct such cells for the treatment of diseases shows promise as a personalized treatment approach. Combined with the continuous advancement of cutting-edge technologies such as bioprinting and nanotechnology in biomaterial development, their integration with iPSC technology offers new hope for the study of rare vascular diseases.

The study of hiPSCs and their derivatives has consistently been a focal point in medical research. Given the substantial social and economic burden of vascular diseases, researchers have been dedicated to developing more effective models for transformative research. The differentiation of iPSCs into vascular cells represents an undeniable cutting-edge technology that holds great promise for improving the lives of patients with vascular diseases. It paves the way for advancements in vascular regeneration and repair.