Use of iPSC-Derived Smooth Muscle Cells to Model Physiology and Pathology

Abstract

The implementation of human induced pluripotent stem cell (hiPSC) models has introduced an additional tool for identifying molecular mechanisms of disease that complements animal models. Patient-derived or Crispr/Cas9 edited iPSCs differentiated into smooth muscle cells (SMCs) have been leveraged to discover novel mechanisms, screen potential therapeutic strategies, and model in vivo development. The field has evolved over almost fifteen years of research using hiPSC-SMCs and has made significant strides towards overcoming initial challenges such as the lineage specificity of SMC phenotypes. However, challenges both specific (e.g., the lack of specific markers to thoroughly validate hiPSC-SMCs) and general (e.g., a lack of transparency and consensus around methodology in the field) remain. In this review, we highlight the recent successes and remaining challenges of the hiPSC-SMC model.

Graphical Abstract

The discovery of human induced pluripotent stem cells (hiPSCs) by Shinya Yamanaka in 2006 held great promise for researchers throughout biomedical science (1, 2). These cells heralded a new opportunity for personalized medicine: study disease mechanisms, test therapeutics, and even engineer cells or tissue for clinical use using a patient’s own cells. The first publication showing vascular smooth muscle cells (SMCs) derived from human iPSCs was published in 2009 (3). Here, we will discuss some of the inherent difficulties with using hiPSC-derived models for vascular smooth muscle cells, and how the field has worked to overcome these difficulties and fulfill the initial promise of iPSC-derived cells.

Vascular smooth muscle cells are the cellular component of the medial layer of the blood vessel wall. Quiescent SMCs in the vessel wall contract to regulate blood flow and connect with extracellular matrix proteins that provide the tensile strength of the vessel wall, such as elastin and collagens. These cells express smooth muscle-specific contractile proteins, including smooth muscle α-actin (encoded by ACTA2), smooth muscle β-myosin heavy chain (MYH11), calponin (CNN1), SM22α (TAGLN), and smoothelin (SMTN). Two critically important features of vascular smooth muscle cells that make them more difficult to model in vitro are 1) their diverse embryonic origins (4) and 2) their inherent phenotypic plasticity (5).

SMCs can be derived from either ectodermal or mesodermal lineages, including neural crest cells, epicardial lineage, paraxial mesoderm, and the second heart field (Figure 1)(4, 6, 7). Developmental lineage is of critical importance for in vitro modeling of vascular diseases due to regional differences in disease propensity or pathogenesis (8). For example, in the ascending aorta, there are both neural crest-derived and second heart field-derived SMCs (9), and the use of lineage-specific Cre lines in mice has supported that these cells behave differently under pathologic conditions of vascular calcification (10) or aneurysm formation (11, 12). In vitro studies of explanted SMCs from vascular beds of distinct embryonic origins have shown differential responses to specific stimuli (e.g., transforming growth factor β1 (TGFβ1), homocysteine, and others (13-16)) that further emphasize the need for lineage specificity in any disease model.

Figure 1. Diverse embryonic origins of vascular smooth muscle cells.

A schematic of the aorta showing the region-specific origin of SMCs within the aorta and a list of markers of each progenitor cell type. SMCs in the ascending aorta, carotid arteries, and the cerebrovasculature are all derived via neural crest progenitors. Descending aorta SMCs are derived via a somite lineage. Aortic root SMCs are derived from secondary heart field cells, while coronary artery SMCs are derived via epicardium.

SMCs retain significant plasticity, and many vascular diseases including atherosclerosis and aneurysm formation are characterized by phenotypic modulation of SMCs, wherein the cells de-differentiate and take on disease-specific phenotypes such as proliferation and migration, synthesis of extracellular matrix, inflammation, or expression of a heterogeneous range of non-SMC marker genes (5, 17-19). Fate mapping and single cell transcriptomics studies have emphasized how varied this SMC phenotypic modulation can be and how critical it is to disease pathogenesis (20, 21). However, this plasticity raises questions about the validity of in vitro models to recapitulate the “true” SMC phenotype- fundamentally, if cells are proliferating in culture, are they representative of the contractile cells in an intact vessel? How can disease-associated modulated phenotypes be modeled in vitro?

These questions, while not specific to hiPSC models, are still fundamental to the field; it is imperative to be precise in describing how closely hiPSC-derived SMCs represent in vivo phenotypes in terms of both embryonic lineage and quiescence/maturity. One difficulty researchers face is the lack of cell surface markers for fully differentiated SMCs; most publications rely on the contractile genes listed above to define SMC identity, but these all encode intracellular proteins that cannot be easily used to verify the identity of live cells. Although homogeneity of differentiation can be verified by analyses of intracellular markers, there is no mechanism to use this information to increase the purity of the derived cells, e.g. by cell sorting. To date, no published studies have identified embryonic lineage-specific SMC markers, although significant progress has been made in developing differentiation protocols to mimic developmental pathways. Although it was first identified as having an SMC-specific expression pattern nearly 30 years ago (22), the use of cell surface integrin receptor ITGA8 as a marker of SMC identity has not been widespread. A recent inducible Itga8-Cre mouse model (23) confirms the specificity of this marker and suggests the promise of using integrin α8 as a surface marker to ensure the purity of derived SMCs.

Generation of hiPSC-SMCs

The first paper reporting in vitro derivation of SMCs from hiPSCs was published in 2009. This paper utilized a protocol developed for embryonic stem cell (ESC) differentiation: a relatively crude initial differentiation followed by cell sorting for Flk1⁺ VE-cadherin⁻ TRA1-60⁻ cells that were then induced into SMCs by platelet derived growth factor (PDGF)-BB treatment (3). Although this protocol had an extremely low yield (<5%), it was the first proof of principle that directed differentiation of hiPSC-SMCs was possible.

Since that initial report, there have been several advances in the methodology for deriving hiPSC-SMCs, most notably the development of lineage-specific protocols for these differentiations. A critical and promising advance arrived in 2012 from the laboratory of Sanjay Sinha, M.D., Ph.D., who developed parallel protocols for the derivations of SMCs via neuroectodermal, paraxial mesoderm, and lateral plate mesoderm intermediates (24). Lateral plate mesoderm is a precursor of the second heart field (25) and here is used to model the derivation of aortic root SMCs. After a four- or seven-day treatment to induce the intermediate cell type, all three lineages can be induced into SMCs by cotreatment with TGFβ1 and PDGF-BB for twelve days. The same laboratory later published a protocol for derivation of epicardial progenitor-derived SMCs: after derivation of epicardial progenitor cells, epithelial-to-mesenchymal transition and differentiation into SMCs can be induced with TGFβ1 and PDGF-BB similar to the other lineages (26). These protocols (summarized in Figure 2) all led to robust, high efficiency differentiation to SMCs (i.e. >90% purity of cells expressing SMC markers by immunostaining) that recapitulated both the expected differentiated SMC phenotypes and region-specific differences in responses to cytokines based on embryonic origins (27). Specifically, robust expression of both early (ACTA2, TAGLN, CNN1) and late (SMTN, MYH11) markers of SMCs was confirmed by microarray and immunostaining-based analyses. The cells recapitulated origin-specific phenotypes, such as the requirement for the transcriptional co-activator MKL2 in neuroectoderm-to-SMC differentiation but not mesoderm-to-SMC differentiation (27). Further iterations of this protocol indicated improved specificity of the neuroectodermal lineage into a more robust neural crest progenitor lineage (28, 29).

Figure 2. Comparison of lineage-specific differentiation protocols for hiPSC-SMCs.

A) Comparison of published protocols for differentiation of hiPSC-SMCs via neural crest intermediates(24, 30, 70). B) Comparison of published protocols for differentiation of hiPSC-SMCs via mesoderm lineages – the top two lines show paraxial mesoderm or somite intermediate while the bottom two lines show lateral plate mesoderm or second heart field intermediate protocols(24, 30). C) Comparison of published protocols for differentiation of hiPSC-SMCs via epicardium intermediates(24, 30).

To even more closely mimic in vivo differentiation trajectories, a recent paper from the laboratory of Joseph Wu, M.D., Ph.D., outlined more granular in vitro maps to embryonic origin-specific SMCs. Here, a stepwise program ensures the purity of intermediate progenitor cells and robust yields of final SMCs (summarized in Figure 2) (30). Critically, these protocols were designed to mirror sequential signals that promote differentiation during embryonic development. Similar to the earlier Sinha lab protocols, all four of the progenitor lineages described: neural crest, somites (i.e., paraxial mesoderm), secondary heart field (i.e., lateral plate mesoderm), and epicardium, can be differentiated into SMCs by co-treatment with TGFβ1 and PDGF-BB. However, a direct comparison between cells derived using these available lineage-specific protocols has yet to be published. The field would benefit from a comparison evaluating whether the purity of the resulting SMCs (assessed as a percentage of cells expressing SMC markers), the level of expression of SMC markers (assessed by quantitative comparison of mRNA or protein levels), or origin-specific phenotypes (assessed by response to specific stimuli known to provoke the expected response based on lineage identity) are affected by the protocol used to derive the cells.

Despite these dramatic advances in our understanding of hiPSC-SMC derivation, a few difficulties remain. Namely, many reports indicate that derived SMCs are immature and not fully representative of in vivo quiescent SMCs (31, 32). hiPSC-SMCs have been compared directly with explanted primary SMCs from mouse or human aortas and are found to have similar levels of contractile proteins and similar behaviors (27, 33, 34). Explanted SMCs have been used for many years as a model to dissect molecular pathways relevant to disease (35-37). However, proliferation is necessary for the maintenance of cultured cell lines, and for SMCs, increased proliferation often correlates with decreased differentiation or maturity (38). Multiple strategies for improving the maturity of hiPSC-SMCs have been attempted. Extended treatments with TGFβ1 have been shown to induce a more mature and less proliferative phenotype (39). The Sinha protocol includes a provision for long-term culture of 1-3 months in serum-containing media to a more “mature” stage they call S30 (28). The Wu protocol includes an additional treatment step with MEK inhibitor PD0325901 for six days to induce higher expression of MYH11, the most established marker for mature SMCs (30). Other reports utilize an alternative to TGFβ1 or PDGF-BB treatment for the differentiation of hiPSC-SMCs from a mesodermal progenitor: an Alk5 inhibitor called RepSox, and this change is reported to yield higher levels of MYH11 expression as well (40). Finally, an additional strategy for the maturation of hiPSC-SMCs without additional growth factor treatments is to co-culture SMCs with endothelial cells (ECs) with hemodynamic flow (41), mimicking some of the methods used to mature primary SMCs in culture. However, there is again no real comparison available or consensus within the field about best practices for inducing highly mature hiPSC-SMCs.

Applications of hiPSC-SMCs

Disease modeling in monolayer cultures

One of the most prominent, and promising, uses of hiPSC-SMCs is to model specific diseases in culture. These disease models can be used to understand mechanisms, manipulate genetics, or screen potential therapeutic strategies. Importantly, the use of hiPSC-SMCs allows tracking of disease progression through early stages, while patient tissue samples only represent the late-stage or end-stage disease. Additionally, human primary SMC lines are rare and can only be used for a limited number of passages, while SMCs from animals may not completely recapitulate human phenotypes or may be inadequate for studies of genetic factors that are present in the human, but not the mouse, genome. hiPSC-SMCs overcome these barriers by providing a renewable source of human-specific SMCs. Many modeling studies utilize iPSCs reprogrammed from patient-derived cells, either fibroblasts or peripheral blood mononuclear cells. Newer strategies can develop specific disease models using CRISPR/Cas9 editing, which has the significant benefit of creating isogenic control lines to directly compare the effects of a single genetic change. This section will summarize studies and findings from a few of these hiPSC-SMC disease models.

One disease where iPSC-based modeling has successfully contributed to the advancement of the field is Hutchinson-Gilford Progeria syndrome (HGPS), which is caused by autosomal dominant truncation variants in the gene LMNA (42-44). The truncated Lamin A protein is called “progerin” and exerts toxic effects on cells. Dermal fibroblasts from HGPS patients have been reprogrammed into iPSCs and differentiated into multiple cell types, notably including SMCs. Despite significant phenotypic changes in the dermal fibroblasts, HGPS iPSCs are indistinguishable from control cells (45). HGPS iPSCs can differentiate into SMCs with similar efficiency to control cells and have been used to investigate disease mechanisms (46), including how progerin accumulation induces cell death in SMCs. HGPS SMCs exposed to hypoxic or mechanical stress had increased nuclear distortion and cellular senescence compared with control cells derived from unaffected relatives of HGPS patients (47). These findings recapitulate phenotypes observed in HGPS patient tissue, including loss of SMCs and distortion of the nuclear envelope, and provide a cellular model to look more deeply at molecular mechanisms and test potential treatments. One study used lentiviral shRNA infection to knock down the expression of the progerin protein during differentiation to specifically prove that the accelerated cell death phenotype of HGPS SMCs is due to progerin accumulation (45). Further studies have cultured HGPS hiPSC-SMCs in a three-dimensional culture model under flow conditions (these types of culture models will be discussed more in detail below) and have tested specific pharmaceuticals, including batimastat which inhibits MMP13 (48), everolimus which inhibits mTOR signaling (49), and lonafarnib which inhibits farnesyltransferase (50). Each of these drugs improved the phenotype of HGPS hiPSC-SMCs and have also been tested in mouse models and human clinical trials (48, 51, 52). The use of the hiPSC-SMC model accelerates drug discovery potential and allows easier and more direct assessments of its effects specifically on vascular cells, which is of critical importance given that cardiovascular disease is the major cause of mortality in HGPS patients.

A second disease that has been very successfully modeled with hiPSC-SMCs is Marfan Syndrome (MFS), caused by pathogenic variants in the gene FBN1. Reprogrammed fibroblasts from MFS patients differentiate into hiPSC-SMCs at similar efficiency to isogenic controls generated by correction of the mutation using a TALEN genome editing approach (53). MFS hiPSC-SMCs had reduced contractility and calcium flux that are rescued by correction of the FBN1 variant (53). A second group utilized hiPSC-SMCs to perform proteomics analyses comparing SMCs derived via neuroectoderm or lateral plate mesodermal progenitors from both controls and MFS patient-derived cells (54). They identified novel target proteins that are differentially regulated in MFS-derived cells including the collagen-modulating protein MRC2, which has reduced expression in MFS cells compared with controls, along with a set of extracellular matrix proteins upregulated specifically in lateral plate mesoderm-derived MFS SMCs. These ECM proteins may contribute to the site-specific disease pattern in patients with MFS and were validated using immunostaining of human aortic tissue from patients with MFS (54). The same group went on to confirm that a specific inhibitor of one of these targets, integrin αv, blocks pathologic signaling in both the hiPSC-SMC model and a mouse model of MFS, supporting this inhibitor as a potential therapeutic approach (55). Another group used patient-derived MFS hiPSC-SMCs and found that these cells replicate key pathologic phenotypes found in aortic tissue of MFS patients including abnormal deposition of fibrillin-1, the protein encoded by FBN1, increased TGFβ1 canonical signaling, decreased cellular contractility, and increased cell death (28). Prior work on MFS mouse models had identified losartan as a potential therapeutic to reduce MFS-related aortic phenotypes, but clinical trials in humans had not confirmed any positive benefit to patients (56-59). Here, treatment of the MFS hiPSC-SMCs with losartan rescued some phenotypes, including reducing canonical TGFβ1 signaling, but ultimately failed to prevent cell death (28). These results provide a critical explanation for the failed clinical trial, and the authors went on to identify other key molecular signals, including noncanonical TGFβ1 signaling through p38 MAP kinase and the transcription factor KLF4 as novel targets for further investigation. Finally, similar to the previous study, correction of the pathogenic variant via CRISPR/Cas9 editing rescued aberrant phenotypes in the hiPSC-SMCs to the same phenotype as controls (28). This same group later went on to perform a screen for small molecule therapeutics in hiPSC-SMCs from MFS patients with or without CRISPR/Cas9 correction of the FBN1 allele as isogenic test populations. They identified GSK3β inhibition as a critical target for the treatment of MFS (60). Taken together, the use of hiPSC-SMCs to model MFS shows the very strong utility of the hiPSC-SMC system, as it has provided an avenue for a) identifying molecular mechanisms based on the specific embryonic origin of SMCs, b) identification of novel therapeutic targets, c) testing of gene correction strategies, and d) rapid screening of potential therapeutics in a human cell model that better mimics aberrant phenotypes in human disease than mouse models.

The use of hiPSC-SMCs for the study of other heritable aortopathies, including various forms of Loeys-Dietz syndrome, vascular Ehlers-Danlos syndrome, and heritable thoracic aortic disease (HTAD), is less advanced. Although a few studies have reported generating hiPSC lines from patient cells (61-65), there is less published work on differentiated hiPSC-SMCs for these disorders, with a few notable exceptions. The laboratory of Bo Yang, M.D., has published on TGFBR1 variant hiPSCs, with both correction of the variant in patient-derived cells and introduction of the variant in control cells via CRISPR/Cas9 editing (66). The TGFBR1 variant cells had decreased differentiation into SMCs compared with isogenic controls. Combinatorial treatment with activin A and rapamycin during differentiation improved contractile gene levels and mechanical function of cells harboring the mutation (66). The same group also created a SMAD3 loss of function variant using CRISPR/Cas9 editing in hiPSCs. Results showed that loss of SMAD3 similarly resulted in decreased differentiation potential of the cells into SMCs only in cells differentiated via a mesodermal intermediate and not cells differentiated via a neuroectodermal intermediate (67). These data align with clinical information from patients with Loeys-Dietz syndrome, who develop thoracic aneurysms at the aortic root (of mesodermal origin) but much less frequently in the ascending aorta (of neuroectodermal origin) (68), underscoring the potential utility of hiPSC-SMCs to model aortopathies beyond MFS.

The same group separately introduced a NOTCH1 pathogenic variant associated with aortic disease with bicuspid aortic valve using CRISPR/Cas9 editing and found decreased differentiation potential in neural crest-derived hiPSC-SMCs (69). This data recapitulates findings from the same group using patient-derived cells from individuals with bicuspid aortic valves but without known pathogenic variants and with negative testing for NOTCH1 variants. In the previous study, hiPSCs from individuals with bicuspid aortic valve were less able to differentiate into SMCs via neural crest intermediates but not via mesodermal intermediates (70). As above, these data align with previous studies, which have shown that neural crest-derived cells are critical for valve formation (71). Taken together, these data underscore both the critical importance of a lineage-specific differentiation pathway for modeling vascular disease using hiPSCs and the principle that studying rare pathogenic variants can uncover mechanisms of disease that are present in more common or less heritable forms of the same clinical phenotype.

Importantly, primary cells are more often available for aortopathies than for other vascular diseases due to the availability of tissue from surgical samples taken prior to end-stage disease. For other diseases, such as cerebrovascular diseases, the use of hiPSC-SMCs has even greater promise as a more robust source of patient-derived cells than primary explants. Multiple groups have modeled the cerebrovascular disease Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL), caused by pathogenic variants in NOTCH3, using reprogrammed patient-derived iPSCs. Early studies indicated that hiPSC-SMCs from CADASIL patients recapitulate disease associated phenotypes such as increased proliferation, dysregulation of signaling pathways, and cytoskeletal disorganization (72, 73). A second group showed that CADASIL hiPSC-SMCs secreted lower levels of the angiogenic growth factor VEGF, and as a result induced cell death in adjacent endothelium; these defects could be partially rescued by the addition of exogenous VEGF or by knockdown of the mutant NOTCH3 (74). Use of CADASIL hiPSC-SMCs in a co-culture model of an in vitro neurovascular unit showed that the SMCs derived from CADASIL patients were less able to induce barrier function in co-cultured endothelium compared with control cells, suggesting that altered blood-brain barrier function is a potential mechanism of disease (75). An additional study corrected the NOTCH3 variant using an adenine base editor strategy (76). Taken together, these results affirm that hiPSC-SMC modeling can be utilized for diseases of smaller vessels than the aorta. Stem cell-based models have also been used to study amyloid-β metabolism and moyamoya disease (77-79). Pulmonary hypertension due to HIF2A gain of function variants and BMPR2 pathogenic variants have also been modeled with hiPSC-SMCs (80-82).

Although most studies using hiPSC-SMCs as a disease model are attempting to model a monogenic disorder where the pathogenic variant is known, a few studies have used these cells to model more common disease processes. One study used hiPSC-SMCs derived from patients with hypertension with the stated eventual goal being to screen the responsiveness of patient-specific cells to different antihypertensive medications, which would optimize personalized treatments to control blood pressure (83). Another study used healthy control hiPSC-SMCs treated with osteogenic media to study the process of calcification (84). These two studies represent a spectrum of possible uses for hiPSC-SMCs in the study of common disease: hiPSC-SMCs can be a renewable source of vascular cells that function similarly to primary explanted SMCs and are differentiated through the embryonic lineage of the tissue of interest. Also, hiPSC-SMCs from patients with complex, non-heritable diseases can be utilized to assess the impact of genetic variation on disease pathogenesis and treatment response. To date, however, these types of studies have not been extensively pursued.

Published work confirms the promise of using hiPSC-SMCs as a disease model, as most findings either align with known clinical consequences (e.g., disease susceptibility in different vascular beds) or can be validated using in vivo systems. The field is young, and to fulfill that promise, the next stage of discovery using hiPSC-SMCs needs to confirm translational impact: can we actually use an hiPSC-SMC disease model to identify novel therapies that translate to the clinic? Specifically, can hiPSC-SMC models be better predictors of clinical efficacy than mouse models? Disease pathogenesis is complex and may involve more than one cell type: co-culture of hiPSC-SMCs with derived endothelial or immune cells, or stimulation with conditioned media or cytokines produced by other cell types, can overcome these barriers. The two-dimensional culture system described in the above studies allows for the screening of molecules with therapeutic potential. but some phenotypes cannot be effectively modeled in two dimensions – including the biomechanical forces that likely play a role in the pathogenesis of vascular diseases like aortic dissections (85-87). The use of hiPSC-SMCs can also permit the study of these forces using three-dimensional systems that can be subjected to cyclic stretch or flow-based shear stress (88-90).

Three-dimensional culture systems

To overcome some of the limitations present in traditional two-dimensional (2D) monolayer cell culture models, ongoing studies have implemented three-dimensional (3D) cell culture techniques to recreate a more realistic physiological environment. These 3D systems can subsequently be more effectively manipulated for translational research and ultimately implemented in a clinical environment. Traditional 2D cell culture models fail to fully recapitulate cell-cell and cell-matrix interactions in response to stimuli like drugs, growth factors, or biomechanical forces (91, 92). Some of the most common and recently developed 3D cell culture techniques for hiPSC-derived SMCs involve the usage of organ-on-a-chip, vascular cell spheroids, and engineered vascular tissues (eVTs) to name a few (93-95). The ongoing diversification of these techniques has exponentially increased within the last decade, attracting more attention to the potential that 3D models have for the study of vascular function and disease. 3D cultures allow for the proper formation of and interaction with the extracellular matrix (ECM), mimicking cell adhesion behaviors as well as migratory and differentiation cues that are present in vivo (96, 97). Additionally, key interactions with cell adhesion molecules, including integrin-mediated interactions with collagen and fibronectin, allow cells in a 3D environment to mediate gene expression, proliferation, and morphology in response to mechanical cues which are more representative of a true physiological setting (98, 99). Table 1 comprises a list of some of the most recently published works that integrate the usage of tissue engineering methods for hiPSC-SMCs, their application, and their relevance (50, 94, 100-108).

Table 1.

Recent publications detailing three-dimensional culture or tissue engineering applications for hiPSC-SMCs.

Construction of iPSC-derived SMC eVTs
Publication Year	Title	Applications/Techniques	Summary	Citation
2022	HUMAN iPSC-VASCULAR SMOOTH MUSCLE CELL SPHEROIDS DEMONSTRATE SIZE-DEPENDENT ALTERATIONS IN CELLULAR VIABILITY AND SECRETORY FUNCTION	3D Spheroids	Via the production of an inexpensive method to generate derived spheriods, Islam et al. sought to create an optimized vessel to utilize the paracrine secretion ability of hiPSC-SMCs for regenerative wound healing in patients.	[93]
2023	3D PRINTED BIOMIMETIC FLEXIBLE BLOOD VESSELS WITH iPS CELL-LADEN HIERARCHICAL MULTILAYERS	3D-printed Flexible Small Diameter Blood Vessels	Hann et al. successfully developed 3D-printed blood vessels consisting of SMCs and vascularized endothelium. These cells were derived from iPSCs which were embedded in fibrinogen solutions including thrombin to regulate the formation of layers in these 3D-printed blood vessels.	[99]
2023	HUMAN iPSC-DERIVED VASCULAR SMOOTH MUSCLE CELLS IN A FIBRONECTIN FUNCTIONALIZED COLLAGEN HYDROGEL AUGMENT ENDOTHELIAL CELL MORPHOGENESIS	3D Hydrogel	As a way to regulate the detrimental effects of angiogenesis, Duan et al. generated autologous grafts from hiPSC-SMCs in junction with fibronectin to bring about an increase in EC networks, stabilizing engineered vascular constructs.	[100]
2022	ELECTROSPUN BIODEGRADABLE α-AMINO ACID-SUBSTITUTED POLY(ORGANOPHOSPHAZENE) FIBER MATS FOR STEM CELL DIFFERENTIATION TOWARDS VASCULAR SMOOTH MUSCLE CELLS	Biodegradable Organophosphazene Polymers/Electrospun Nano-Fibrous Scaffold	Two PαAPz polymers were synthesized from L-alanine/L-phenylalanine. L-alanine polymers did not degrade and fibrous mats were able to support proper adhesion and proliferation of hiPSC-SMCs. These electrospun fibers were able to support and promote differentiation.	[101]
2020	A DENSE FIBRILLAR COLLAGEN SCAFFOLD DIFFERENTIALLY MODULATES SECRETORY FUNCTION OF iPSC-DERIVED VASCULAR SMOOTH MUSCLE CELLS TO PROMOTE WOUND HEALING	Collagen-Based scaffolds	In order to understand the impact collagen fibrillar density can have on hiPSC-SMC’s paracrine and cytokine secretion ability, Dash et al., laid the foundation to promote regenerative healing by guiding hiPSC-SMCs in their paracrine signaling pathway.	[102]
2020	TISSUE ENGINEERING USING VASCULAR ORGANOIDS FROM HUMAN PLURIPOTENT STEM CELL DERIVED MURAL CELL PHENOTYPES	Methylcellulose-based hydrogel 3D spheroids	To overcome the problems presented by inadequate vascularization in eVTs; Markou et al. developed a coculture model consiting of hiPSC-SMCs & primary human ECs in 3D spheroidal vascular organoids. This method presents an efficient and stable way to generate in vitro neovessels, thus allowing for the proper development and maturation of TEBVs.	[103]
2020	TISSUE-ENGINEERED VASCULAR GRAFTS WITH ADVANCED MECHANICAL STRENGTH FROM HUMAN iPSCs	Biodegredable polyglycolic acid scaffolds	Key limitations, such as mechanical strength, exist in the fabrication of hiPSC-SMC-containing TEBVs. Luo et al. developed a model by which hiPSC-SMCs were embedded in a biodegredable polyglycolic acid scaffold. The TEBVs were subjected to incremental radial distension and performed similarly to native vessels, thus creating a mechanically superior TEBV model.	[104]
iPSC-derived SMC eVTs as a Disease Model
2023	CARDIAC MUSCLE PATCHES CONTAINING FOUR TYPES OF CARDIAC CELLS DERIVED FROM HUMAN PLURIPOTENT STEM CELLS IMPROVE RECOVERY FROM CARDIAC INJURY IN MICE	Human Cardiac Muscle Patches	Lou et al. generated human cardiac muscle patches using hiPSC-SMCs, which helped attenuate the effects of severe acute myocardial infarction. Addition of cardiac fibroblasts allowed for the better organization of major cell behavior patterns and cellular structure.	[105]
2020	INDUCED PLURIPOTENT STEM CELL-DERIVED SMOOTH MUSCLE CELLS INCREASE ANGIOGENESIS AND ACCELERATE DIABETIC WOUND HEALING	3D Collagen Scaffolds	hiPSC-SMCs were embedded in collagen scaffolds, which were then applied to splinted back wounds in diabtetic nude mice to asses thier in vivo healing properties. hiPSC-SMCs increased the concentration of pro-angiogenic cytokines and accelerated diabetic wound healing associated with total and M2 macrophages.	[107]
2023	LONAFARNIB AND EVEROLIMUS REDUCE PATHOLOGY IN iPSC-DERIVED TISSUE ENGINEERED BLOOD VESSEL MODEL OF HUTCHINSON-GILFORD PROGERIA SYNDROME	3D Tissue Engineered Blood Vessels (TEBV)	In their study, Abutaleb et al. overcome previous shortcomings in the generation of TEBVs from hiPSC-SMCs from patients with HGPS. Supplementation with Lonafarnib and everolimus during 3D cell culturing improved endothelial growth and increased SMC marker expression and decreased apoptosis.	[49]
2023	PCSK9 ACTIVATION PROMOTES EARLY ATHEROSCLEROSIS IN A VASCULAR MICROPHYSIOLOGICAL SYSTEM	3D Tissue Engineered Blood Vessels (TEBV)	Proprotein convertase subtilisin/kexin 9 (PCSK9), an enzyme that contributes to vascular inflammation and a possible protagonist for atherosclerosis was studied in hiPSC-SMC-containing TEBVs. Via the production of PCSK9 over-and under-expressing hiPSC-SMCs, PCSK9 activation and represson were assessed in vitro, and the results show that PCSK9 inhibition lowers vascular inflammation.	[106]

Future Clinical Uses of hiPSC-SMCs:

Transplantation of vascular cells to revascularize damaged tissue and improve tissue perfusion is a potential therapeutic strategy for many diseases (109, 110). Co-transplantation of hiPSC-SMCs with other iPSC-derived cells such as endothelial cells (EC) has been shown to efficiently mediate key processes such as neovascularization. For example, hiPSC-SMCs and hiPSC-derived ECs were co-administered to assess their therapeutic potential in a critical limb ischemia (CLI) model: the addition of hiPSC-SMCs in addition to ECs enhanced perfusion due to paracrine effects of the SMCs on EC behavior (111). These results underscore the importance of cell-cell interactions in therapeutic settings. A further improvement on co-transplantation of individual cells could be the transplantation of fully developed vascular units, i.e. eVTs.

Researchers have developed numerous approaches for the generation of eVTs from hiPSC-SMCs for various pre-clinical applications (112, 113). It is important to note that there are limited studies that use hiPSC-SMC for vascular-engineered tissues; although analogous studies have focused on the usage of other hiPSCs-derived cells such as skeletal muscle cells and cardiomyocytes for tissue engineering development (114, 115). Crucially, successful eVTs require the integration of SMC and endothelial cells along with a carefully patterned matrix to fully recapitulate in vivo vessels. Another limiting factor is the presence of key regulators of SMC modulation: one example is elastin, which serves as a direct mediator for SMCs’ maturation and proliferation by directly affecting the polymerization rate of smooth muscle α-alpha actin. Early studies of engineered grafts lacked elastin deposition in the ex vivo vessel walls, but treatment with compounds like epigallocatechin gallate can facilitate the laying down of elastin fibers in hiPSC-SMC populated engineered vascular tissues (116). Further studies are needed to ensure replication of the endogenous structure of elastic arteries: interspersed layers of elastin fibers between each concentric ring of SMCs. Despite the ongoing advances in 3D culturing techniques to more accurately model in vivo conditions, these in vitro vascular tissues are often supplemented with animal-derived reagents for their proper growth and development. To translate the findings of new innovative in vitro models into a clinical setting, hiPSC-SMC-derived tissues must be produced under non-xenogeneic conditions to allow for a wider and safer application in patient therapies (117). However, these changes are not necessary for the research use of hiPSC-derived eVTs for disease modeling.

Use of eVTs for Disease Modeling:

Disease modeling using eVTs has enabled researchers to fine-tune the attributes of hiPSC-SMCs to reflect in vivo conditions more accurately. To develop an alternative for aortic aneurysm mouse models, Liu et al. established a linage-specific SMC-on-a-chip model using hiPSCs derived via different cell lineages that would represent different segments of the aorta (93). These lineages were then tested through various tensile stress conditions, which revealed that all lineages of hiPSC-SMCs within the organ-on-a-chip model upregulated pathways related to cellular contraction and adhesion in response to optimized cyclic stretch as expected, but the test also revealed lineage-based differences in sensitivity to tensile stress (93). This system, once developed, can be used to expand pathological mechanism research in cardiovascular disease by allowing for higher-throughput mechanistic studies on the interactions between different components of the vessel wall. In a similar vein, cardiac spheroids have been developed comprising multiple hiPSC-derived cardiovascular cell lines including SMCs; these spheroids have been shown to provide an accurate system for the modeling of cardiac diseases as well as the testing of potential therapeutics (118). As described in the section above, 3D culture models have begun to be utilized to study genetically triggered vascular diseases, such as HGPS (49).

Although there have been gigantic strides in the development of more efficient techniques within the field of translational cardiovascular research, there still exists a somewhat substantial gap in knowledge when it comes to the effectiveness of engineered vascular tissue. More precise physiological representation is possible using 3D model systems, increasing their applicability as compared with 2D models. New technologies, like 3D-bioprinting, can be leveraged to improve the embedding of cells and matrices still further in physiologically relevant ex vivo tissues, as has been done using hiPSC-derived cardiomyocytes in ventricle-shaped cardiac tissues (119). These approaches show promise for tailored production of patient-specific tissues to address cardiac defects or predisposing conditions. However, significant hurdles must be overcome in terms of testing before these novel approaches can be integrated clinically. A more immediate use for these newly developed engineered tissues are organ-on-a-chip systems, which can produce incredibly specific models for individual patient mutations, such as the case of a heart-on-a-chip model that mimicked cardiomyocytes to better understand dilated cardiomyopathy caused by a point mutation in sodium channel (SCN5A)-related cardiomyopathies (120). These models have not been extensively utilized for vascular and SMC-driven diseases yet, but the technology exists, and this is a promising direction for future research (121).

Current state of hiPSC-SMC research

We have established in this review that major technological strides have been made in terms of the derivation and use of hiPSC-SMCs since the first publication almost 15 years ago. However, there is still a perception that this technology is not in widespread use. To assess this perception, we used the tool dimensions.ai to assess trends in publishing for hiPSC-SMCs compared with hiPSC-derived endothelial cells or cardiomyocytes, which are other cardiovascular cell types with similar timelines for the establishment of hiPSC derivation and similar disease relevance (122). We collected data on publications by year that are associated with the term “induced pluripotent stem cells” and the term “smooth muscle cells” or “endothelial cells” or “cardiomyocytes” and graphed this data either by raw number of publications or the growth in the number of publications since 2009 (Figure 3A,B). There are fewer publications on hiPSC-SMCs than the other two cell types, but the growth of each field is proceeding at a very similar pace (Figure 3A,B). There are far fewer publications on SMCs in general than on endothelial cells (Figure 3C), which suggests that hiPSC-SMC use may lag behind hiPSC-ECs due to underinvestment in SMCs overall. Increased research use of hiPSC-cardiomyocytes, by contrast, may be due to the lack of availability of primary cardiomyocyte cultures, accelerating the use of hiPSC technology for this cell type. However, the growth of hiPSC-SMC research at a similar pace to other hiPSC-derived cell types is encouraging.

Figure 3. Growth of hiPSC-SMC publications over the first 15 years of research.

A) Data mined from dimensions.ai(122) shows the raw number of papers published associated with the terms “induced pluripotent stem cells” and “smooth muscle cells” (SMCs) or “endothelial cells” (ECs) or “cardiomyocytes” (CMs). There are fewer papers on hiPSC-SMCs than on hiPSC-ECs or hiPSC-CMs every year since the first publication of hiPSC-SMCs in 2009. B) By contrast, if we graph the same data by percent increase over time, we can see that the number of publications on hiPSC-SMCs has increased at the same rate as publications on hiPSC-ECs or hiPSC-CMs since 2009. C) SMCs as a whole are underrepresented in the recent literature compared with ECs, but not compared with CMs.

Ongoing challenges do remain that may impede more widespread adoption of this technology (summarized in Table 2). Notably, constantly evolving protocols for both the reprogramming and differentiation of hiPSCs have created difficulties in evaluating reproducibility. Very few studies have been published comparing the differentiation potential of hiPSCs derived via different reprogramming methods, and there are also few studies directly comparing the available SMC differentiation methodologies. Each laboratory seems to utilize their own preferred methodologies, which limits the growth of the field as new researchers may have difficulty assessing the most up-to-date protocols. Improved markers for SMCs, as well as lineage-specific markers, to assess the robustness of origin-specific differentiation protocols are also still needed. A prior review indicated batch-to-batch heterogeneity of hiPSC-SMC differentiations, in particular between hiPSC lines (123). Indeed, our own experience suggests that the SMC differentiation protocol must be optimized slightly for each hiPSC line according to variables such as confluence of the starting cells, time to first passage during the differentiation, and other minor adjustments. Details of the protocol, including specific media composition, plate coating, or cell confluency, are not always included in publications and can have significant effects on hiPSC-SMC quality. These factors yield a high barrier-to-entry for the use of hiPSC-SMCs even beyond the acknowledged high cost of materials and expertise for hiPSC culture in general (124).

Table 2.

Summary of recent advances and ongoing challenges in the field of hiPSC-SMC research.

DIVERSE EMBRYONIC ORIGINS
Major Advances:	Remaining Challenges:
Development of lineage-specific in vitro differentiation protocols	Lack of known lineage-specific SMC marker genes
Recapitulation of vascular-bed specific disease features via lineage specific models	Incomplete characterization of how SMCs from different vascular beds respond to tension/stretch stress, extracellular matrix type, or metabolism
INHERENT PHENOTYPIC PLASTICITY
hiPSC-SMCs recapitulate phenotypes of primary cultured SMCs	Need for increased in vivo validation
Improved maturity of hiPSC-SMCs via extended growth factor treatment, 3D culture, or co-cultured methods	Lack of comparative studies assessing the best methodology for increased maturity
RIGOR & REPRODUCIBILITY
Published detailed protocols for in vitro differentiation	Lack of consensus/constantly evolving protocols
Identification of surface marker Itga8 to improve purity of derived cells	Publications do not always include quantification of differentiation efficiency or specific methodological details
	Batch-to-batch or cell-to-cell line heterogeneity of differentiation

As described at the start of this review, SMCs also present unique challenges, including high levels of plasticity that call into question the ability to truly model in vivo phenotypes in an inherently proliferative culture system. Even assessing whether hiPSC-SMCs mimic quiescent SMC phenotypes is difficult due to the heterogeneity of SMCs across vascular beds, age, and other variables. The proliferation of single cell sequencing datasets from human tissue should accelerate the discovery of markers for specific subtypes of SMCs, whether based on lineage origin or genetic factors, that can be utilized by hiPSC-SMC researchers to further validate their results. However, as of now, these datasets are only just beginning to be publicly available, and it will take time to perform meta-analyses that can identify the most useful phenotypes for validation.

Despite these acknowledged challenges, it is encouraging to see continued innovation within the field. Continued refinement of origin-specific SMC differentiation protocols has led us to a point where faithful recapitulation of in vivo development is attainable. The ease of genomic editing technology allows the rapid production of isogenic control cells for studying genetically-triggered vascular diseases, reducing the impact of human variability on the success of future research. In the future, the use of the developing three-dimensional culture technologies for the modeling of diseases with fundamental mechanical pathology, such as heritable aortopathies, will accelerate the discovery of molecular mechanisms and therapeutics for these disorders. The use of hiPSC-SMCs still has some theoretical benefits over animal models for the discovery and testing of therapeutics: namely the possibility of speed/high throughput approaches that would be impossible in animal models, and the use of human or even patient-specific cells to test therapies in the patient’s specific genetic context. Eventually, these three-dimensional systems can be optimized for clinical use to re-vascularize tissue with damaged vessels or prepare patient-specific vascular grafts. In other words, the potential for this technology remains high, and we predict continued growth of the hiPSC-SMC field over the coming 15 years.

Highlights.

• Human induced pluripotent stem cells (hiPSCs) have emerged as an important source of derived vascular smooth muscle cells (SMCs) over the past fifteen years of research

• hiPSC-SMCs have been successfully used for disease modeling in 2D and 3D cultures, including identification of disease mechanisms and screening of potential therapeutics.

• hiPSC-SMCs can also be used as a source of cells for engineered vascular tissues for translational research and potential clinical applications.

• The field has made significant strides towards overcoming initial challenges such as developing lineage specific protocols for deriving hiPSC-SMCs.

• Widespread use of hiPSC-SMCs is hampered by the lack of specific markers to thoroughly validate hiPSC-SMCs and a lack of transparency and consensus around methodology in the field.

List of Abbreviations

CM

Cardiomyocyte

EC

Endothelial cell

ECM

Extracellular matrix

ESC

Embryonic stem cell

eVT

Engineered vascular tissue

hiPSC

Human induced pluripotent stem cell

hiPSC-SMC

Human induced pluripotent stem cell-derived smooth muscle cell

HGPS

Hutchinson-Gilford progeria syndrome

HTAD

Heritable thoracic aortic disease

MFS

Marfan Syndrome

PDGF-BB

Platelet-derived growth factor-BB

SMC

Smooth muscle cell

TEBV

Tissue engineered blood vessel

TGFβ1

Transforming growth factor β1

2D

Two-dimensional

3D

Three-dimensional