OUP user menu

Derivation and maturation of synthetic and contractile vascular smooth muscle cells from human pluripotent stem cells

Maureen Wanjare, Frederick Kuo, Sharon Gerecht
DOI: http://dx.doi.org/10.1093/cvr/cvs315 First published online: 11 October 2012

Abstract

Aims Embryonic vascular smooth muscle cells (vSMCs) have a synthetic phenotype; in adults, they commit to the mature contractile phenotype. Research shows that human pluripotent stem cells (hPSCs) differentiate into vSMCs, but nobody has yet documented their maturation into the synthetic or contractile phenotypes. This study sought to control the fate decisions of hPSC derivatives to guide their maturation towards a desired phenotype.

Methods and results The long-term differentiation of hPSCs, including the integration-free-induced PSC line, in high serum with platelet-derived growth factor-BB (PDGF-BB) and transforming growth factor-β1, allowed us to induce the synthetic vSMC (Syn-vSMC) phenotype with increased extracellular matrix (ECM) protein expression and reduced expression of contractile proteins. By monitoring the expression of two contractile proteins, smooth muscle myosin heavy chain (SMMHC) and elastin, we show that serum starvation and PDGF-BB deprivation caused maturation towards the contractile vSMC (Con-vSMC) phenotype. Con-vSMCs differ distinctively from Syn-vSMC derivatives in their condensed morphology, prominent filamentous arrangement of cytoskeleton proteins, production and assembly of elastin, low proliferation, numerous and active caveolae, enlarged endoplasmic reticulum, and ample stress fibres and bundles, as well as their high contractility. When transplanted subcutaneously into nude mice, the human Con-vSMCs aligned next to the host's growing functional vasculature, with occasional circumferential wrapping and vascular tube narrowing.

Conclusion We control hPSC differentiation into synthetic or contractile phenotypes by using appropriate concentrations of relevant factors. Deriving Con-vSMCs from an integration-free hiPSC line may prove useful for regenerative therapy involving blood vessel differentiation and stabilization.

  • Smooth muscle
  • Stem cell
  • Cell differentiation

1. Introduction

The stabilization of blood vessels occurs by extracellular matrix (ECM) formation, as well as through the recruitment of mural cells, which include vascular smooth muscle cells (vSMCs) and pericytes. While pericytes are found in the microvasculature, such as in capillaries, vSMCs surround larger vessels such as arteries and veins. During angiogenesis, endothelial cells (ECs) proliferate; connect to pre-existing blood vessels; and, through lumen formation, develop endothelial tubules (a process known as intussusception).1 After the formation of the nascent tubes composed of ECs, surrounding undifferentiated mesenchymal cells get recruited and become differentiated into proliferating vSMCs, which are needed to stabilize the formed tubules.2,3 A platelet-derived growth factor (PDGF-BB)4,5 and a transforming growth factor (TGF-β1)6,7 act as signalling cues for the recruitment and differentiation of vSMCs. Research has suggested that vSMCs become quiescent after birth, taking on the contractile phenotype found in adult vessels.8,9

During neovascularization in the embryo10 or during vessel development, vSMCs have a synthetic phenotype, which is characterized by high proliferation, migration, and ECM protein production.11 In adult blood vessels, vSMCs play an important role in vessel stabilization; therefore, they commit to the mature contractile phenotype, characterized by low proliferation, expression of contractile proteins—namely, smooth muscle myosin heavy chain (SMMHC), and elastin—and low synthetic activity.11

Adult vSMCs wrap around the vessel layer of ECs and contract to regulate and maintain blood vessel diameter in order to counteract the pulsatile blood pressure generated by the heart.12 Remarkably, vSMCs do not stay in a particular terminally differentiated state. Instead, they exhibit plasticity—they can reversibly take on either a contractile or a synthetic phenotype.11

Pluripotent stem cells (PSCs), including human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs), serve as a reliable source for vSMCs because they can self-renew and proliferate. PSCs first differentiate into the mesoderm13 and later into the vascular lineages, including vSMCs.14,15 Collagen IV,16 retinoic acid,1719 and the growth factors PDGF-BB5,14,1921 and TGF-β15 have been implicated in the inducement of vSMC differentiation. Vascular SMCs have previously been derived from human iPSCs from skin fibroblasts22 and human aortic smooth muscle.23 To the best of our knowledge, no study has demonstrated the regulation of both contractile proteins, SMMHC and elastin, in the course of the differentiation and maturation of vSMCs from PSCs.

Our previous studies demonstrated the derivation of vascular smooth muscle-like cells (SMLCs) from hESCs using monolayer cultures supplemented with PDGF-BB and TGF-β1.5,24 This study hypothesizes that hPSC-derived SMLCs can be guided to acquire either a synthetic phenotype or a contractile phenotype.

2. Methods

A detailed Methods section is available in the Supplementary material online.

2.1 Cell culture

2.1.1 Human PSCs

The hESC lines H9 and H13 (passages 15–40; WiCell Research Institute, Madison, WI, USA) and the hiPSC lines MR3125 and BC126,27 (kindly provided by Dr Cheng, JHU School of Medicine) were grown on inactivated mouse embryonic fibroblast feeder layers (GlobalStem, Rockville, MD, USA) in growth medium comprising 80% ES-DMEM/F12 (GlobalStem), 20% KnockOut Serum Replacement (Invitrogen, Carlsbad, CA, USA), and 4 ng/mL basic fibroblast growth factor (bFGF; Invitrogen) or in growth medium composed of KnockOut DMEM (Invitrogen) as basal medium with 20% KnockOut Serum Replacement (Invitrogen), 1% GlutaMAX (Invitrogen), 10 ng/mL FGF2 (PeproTech, Rocky Hill, NJ, USA), 1% MEM Non-Essential Amino Acids (Invitrogen), 0.1% β-mercaptoethanol (BME; Invitrogen), and 1% antibiotic-antimycotic (Invitrogen). All hPSCs were passaged every 4–6 days using 1 mg/mL of type IV collagenase (Invitrogen). Media were changed daily.

2.1.2 Human vSMCs

Human aorta vSMCs (ATCC, Manassas, VA, USA; up to passage 7) were used for the control cell type and were cultured according to the manufacturer's recommended protocol in the complete SMC growth medium specified by ATCC. We also examined primary human aorta vSMCs (Promocell, Heidelberg, Germany; passages 2–5) and cultured the cells following the manufacturer's protocol in their recommended Smooth Muscle Cell Growth Medium 2 (Promocell).

2.2 Vascular SMC differentiation protocol

Human PSCs were collected through digestion with TrypLE (Invitrogen), and a 40 µm mesh strainer (BD Biosciences, San Jose, CA, USA) was used to separate the cells into individual cell suspensions. The cells were seeded at a concentration of 5 × 104 cells/cm2 onto plates previously coated with collagen IV (R&D Systems, Minneapolis, MN, USA). The hPSCs were cultured for 6 days in a differentiation medium composed of alpha-MEM (Invitrogen), 10% FBS (Hyclone), and 0.1 mM β-mercaptoethanol (Invitrogen), with the media changed daily. On Day 6, the differentiated cells were collected through digestion with TrypLE (Invitrogen), separated with a 40 µm mesh strainer, and seeded at a concentration of 1.25 × 104 cells/cm2 on collagen-IV-coated plates. We then cultured the differentiating hPSCs in differentiation medium with the addition of 10 ng/mL PDGF-BB (R&D Systems) and 1 ng/mL TGF-β1 (R&D Systems) for six additional days (a total of 12 days) for SMLCs. We cultured hPSC-derived SMLCs for the time periods and with the media components detailed throughout this paper, changing the media every second day.

2.3 Real-time quantitative RT–PCR

Two-step reverse transcription polymerase chain reaction (RT–PCR) were performed on differentiated hPSCs at various time points as described previously28 for SMA, CALPONIN, SM22, SMMHC, COL1, FN1, ELN, MMP1, MMP2, MT1-MMP, SRF, MYOCD, ERK, YAP1, SMAD3, β-ACTIN, and GAPDH.

2.4 Immunofluorescence

Cells were fixed, permeabilized, washed, and incubated for 1h with anti-human SMA (1:200; Dako, Glostrup, Denmark), anti-human calponin (1:200; Dako), anti-human SM22 (1:200, Abcam, Cambridge, MA, USA), and anti-human SMMHC (3:100; Dako), anti-human fibronectin (1:200; Sigma-Aldrich), anti-human collagen (1:200; Abcam), anti-human elastin (3:100 Abcam), or anti-human Ki67 (1:50, Invitrogen) for 1h. Cells were rinsed twice with PBS and incubated with FITC-conjugated phalloidin (1:40; Molecular Probes, Eugene, OR, USA), anti-mouse IgG Cy3 conjugate (1:50; Sigma-Aldrich), or anti-rabbit IgG Alexa Fluor 488 conjugate (1:1000; Molecular Probes) for 1h. Cells were then rinsed with PBS and incubated with DAPI (1:1000; Roche Diagnostics) for 10min. We examined the immunolabelled cells using fluorescence microscopy (Olympus BX60; Olympus, Center Valley, PA, USA).

2.5 Western blot

We evaluated protein amounts from whole-cell lysates, quantified using the DC assay (BioRad, Hercules, CA, USA), and loaded a concentration of 50 μg of isolated protein from each of the indicated samples per well into a 12.5% SDS–PAGE gel (BioRad). Proteins were transferred to nitrocellulose membranes, blocked for 1 h and incubated overnight with constant shaking and primary antibody (antibodies indicated above). Membranes were washed and incubated for 2 h with either anti-rabbit HRP (1:1000; Cell Signaling Technology, Boston, MA, USA) or anti-mouse HRP (1:3000; Cell Signaling Technology), washed, developed using enhanced chemiluminescence (Pierce), and visualized using the ChemiDoc XRS+ System (BioRad). Images were acquired using BioRad Quantity One software.

2.6 Functional contraction studies

Contraction studies were performed in response to carbachol as previously described.5,19,23 Briefly, hPSC derivatives were cultured (as detailed elsewhere in this paper), washed, and induced for contraction by incubation with 10–5 M carbachol (Calbiochem, Darmstadt, Germany) in DMEM medium (Invitrogen) for 30 min. We visualized the cells using calcein, a cytoplasm-viable fluorescence dye. A series of time-lapse images were taken using a microscope with a 10 × objective lens (Axiovert; Carl Zeiss, Thornwood, NY, USA). We calculated the cell contraction percentage as the difference in the area covered by the cells before (at time zero) and after contraction (at time 30 min).

2.7 Subcutaneous matrigel implantation

PSC-derived vSMCs were trypsinized, collected, and stained with the PKH26 (Sigma-Aldrich) membrane dye. We encapsulated a total of 0.5 × 106 PSC-vSMCs in reduced growth factor Matrigel (BD Biosciences) and 20 μL of EGM-2 media (endothelial growth media). The Matrigel, which contained 250 ng/mL of bFGF (R&D Systems), was injected subcutaneously into each side of the dorsal region of 6- to 8-week-old nude mice. On Day 7, we injected isolectin GS-IB4 from Griffonia simplicifolia and Alexa Fluor 488 conjugate (Invitrogen) through the tail veins of the mice. After 20 min, we euthanized the mice by CO2 asphyxiation and harvested the Matrigel plugs, which were fixed in 3.7% formaldehyde (Sigma-Aldrich) for 1 h. A sequence of z-stack images was obtained using confocal microscopy (LSM 510 Meta, Carl Zeiss, Inc.). Vessel diameters from the short axes of the lumen of the vessel were determined from the three-dimensional 3D confocal images. The lumen diameters of vessels that contained areas with and without PSC-vSMC wrapping were measured using ImageJ (NIH) and known pixel:length ratios. The Johns Hopkins University Institutional Animal Care and Use Committee approved all animal protocols. The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication, 8th Edition, 2011).

2.8 Histology

After confocal analysis, the fixed construct explants were dehydrated in graded ethanol (70–100%), embedded in paraffin, serially sectioned using a microtome (5 μm), and stained with either haematoxylin and eosin or immunohistochemistry for anti-human elastin (Dako, Glostrup, Denmark). Mouse and human tissue samples were used as controls.

2.9 Transmission electron microscopy

Differentiated cells, as detailed below, were prepared for TEM analysis as described previously.28 Serial sections were cut, mounted onto copper grids, and viewed using a Phillips EM 410 TEM (FEI, Hillsboro, OR, USA). Images were captured using a SIS Megaview III CCD (Lakewood, CO, USA).

2.10 Statistical analysis

All analyses were performed in triplicate for n = 3 at least. One-way ANOVA with the Bonferroni post hoc test were performed to determine significance using GraphPad Prism 4.02. (GraphPad Software, Inc., La Jolla, CA, USA). Significance levels were set at *P < 0.05, **P < 0.01, and ***P < 0.001. All graphical data are reported ±SEM.

3. Results

3.1 Long-term culture in high serum with PDGF-BB and TGFβ1 induces synthetic phenotype

Our previous studies established a simple step-wise differentiation protocol, in which we differentiated hPSCs in monolayers supplemented with PDGF-BB and TGF-β1, resulting in highly purified cultures of SMLCs.5,24 The current study ultimately aimed to mature these SMLCs to contractile phenotype vSMCs. Two principal strategies for the maturation of SMLCs (Day 12 of differentiation) were examined: continuous culture in differentiation medium and the effect of deprivation of serum and growth factors during the culture period. The molecular analysis of ECM, cytoskeleton, and contractile proteins enabled the monitoring of the various stages of the maturation process. The aortic vSMC line, which exhibited high expression levels of the contractile proteins, was chosen as the control for mature human vSMCs (Supplementary material online, Figure S1).

In the first stage, we examined the effect of long-term culture using the differentiation medium. SMLCs (Day 12 of differentiation) were cultured for an additional 18 days in differentiation medium containing 10% serum, 10 ng/mL PDGF-BB, and 1 ng/mL TGF-β1. Interestingly, the 30-day differentiated cells took on a synthetic vSMC (Syn-vSMC) phenotype compared with SMLCs, including (i) a decrease in calponin mRNA expression, no significant difference in SMA and SM22 mRNA expression, and a decrease in the mRNA expression of SMMHC (Figure 1A and B); (ii) an increased expression and production of collagen I and fibronectin and a decreased expression of elastin (Figure 1C and D); (iii) and an increased expression of membrane type 1 matrix metalloproteinase (MT1-MMP), MMP1, and MMP2 (Figure 1E). These data proved consistent among the different hPSC lines examined. Because it was suggested that PDGF-BB interferes with vSMC maturation,3134 we attempted to induce maturation by the exposure of Day 12 differentiated SMLCs to TGF-β1 or to no growth factors at all. We found some detectable increase in the expression of SMMHC and elastin, especially when no growth factors were added (Supplementary material online, Figure S2). Although these culture conditions did not induce significantly improved contractility, they suggest that a lessening of signalling activation may induce contractile maturation.

Figure 1

SMLCs take on a synthetic phenotype. On Day 12 of culturing SMLCs, they were moved to differentiation medium containing 10% serum with PDGF-BB and TGF-β1 for an additional 18 days. (A) Quantitative RT–PCR revealed no significant changes in the expression of SMA and SM22 and showed a decrease in calponin and SMMHC between SMLCs and Syn-vSMCs. (B) Western blot analysis confirmed the protein expression of SMA, calponin, and SM22 in SMLCs, Syn-vSMCs, and aortic vSMCs. We found (C and D) increased expression of ECM proteins, collagen I, and fibronectin and a decrease in the expression of elastin in Syn-vSMCs compared with SMLCs. (E) Up-regulation in the expression of MT-1 MMP, MMP1, and MMP2 in Syn-vSMCs. Scale bars are 100 μm. *P < 0.05, **P < 0.01, and ***P < 0.001.

3.2 Serum starvation and PDGF-BB deprivation induce contractile phenotype

The proposed association of quiescence with the contractile phenotype of vSMCs after birth29,30 led us to examine the effects of serum starvation and growth factor depletion during the differentiation of SMLCs. At first, we tested the Syn-vSMC derivatives for an additional 6 days in culture in a medium containing 10% serum plus TGF-β1 or 0.5% serum plus TGF-β1. We could not detect up-regulation in the expression of contractile proteins, specifically SMMHC and elastin, under either of the conditions (Figure 2A). This implied that Syn-vSMC derivatives had already committed to the synthetic phenotype. Thus, we attempted to mature the SMLCs (Day 12) in the same conditions. Indeed, in medium containing 0.5% serum plus TGF-β1, matured SMLCs (mSMLCs) were detected with significantly up-regulated expressions of the contractile proteins SMMHC (∼40-fold) and elastin (approximately eight-fold) and with no significant change in the expression of early cytoskeleton markers (i.e. SMA, calponin, and SM22) and ECM proteins (i.e. collagen and fibronectin; Figure 2B). The mSMLCs began to acquire a more filamentous cytoskeleton organization, as observed with F-actin, SMA, calponin, SM22, and occasionally SMMHC; they also began to produce elastin (Figure 2C). These data were consistent among the different hPSC lines examined. Culturing the mSMLCs in media containing high concentrations of serum for 6 days resulted in down-regulation of both elastin and SMMHC (data not shown). It should be noted that attempts to differentiate SMLCs in medium without serum (0% serum) could not support cell growth, resulting in extensive cell death after 6 days. Re-adding serum to the mSMLCs for another 6 days resulted in down-regulation of SM-MHC and elastin (data not shown).

Figure 2

Serum starvation and PDGF deprivation induce contractile maturation. (A) Syn-vSMCs and (B) SMLCs were cultured for six additional days in media containing 10% serum with PDGF-BB and TGF-β1, 10% serum with TGF-β1, and 0.5% with TGF-β1; they were analysed for the mRNA expression of relevant cytoskeleton and ECM genes. (C) Immunofluorescence analysis of mSMLCs demonstrates the expression and organization of various cytoskeleton proteins, including calponin, SMA, SM22, phalloidin, and SMMHC, as well as the expression of elastin. Scale bars are 100 μm. *P < 0.05, **P < 0.01, and *** P < 0.001.

3.3 Long-term serum starvation and TGF-β1 supplementation induce contractile maturation

To achieve the maturation of Con-vSMCs from PSCs at levels comparable with those in the body, we examined the effect of short-term (6 days) and long-term (12 days) culture in media containing 0.5% serum with and without TGF-β1. First, as expected, we noticed that the growth rate decreased along the culture period in low serum. The continuous differentiation of mSMLCs for an additional 6 days in either set of conditions was not sufficient to induce maturation (Figure 3A). Continuous differentiation of the mSMLCs in low-serum medium for 12 days (a total of 30 days of differentiation) induced Con-vSMC maturation, namely the up-regulation of SMMHC and elastin, with slightly different responses to the addition of TGF-β1 among different hPSC lines (Figure 3B; Supplementary material online, Figure S3). Nonetheless, we found that levels of SMMHC expressed in Con-vSMCs were slightly higher than those of aortic vSMCs, while elastin levels were inconsistent in the cell lines but were, overall, higher than in the Con-vSMCs (Supplementary material online, Figure S4). Notably, culturing these Con-vSMCs in low-serum medium with TGF-β1 for up to 18 days maintained high levels of SMMHC and elastin expression with decreasing proliferation rates, whereas culturing them in high-serum medium reduced the levels of SMMHC and elastin expression with increasing proliferation rates (data not shown).

Figure 3

Mature SMLCs acquire a contractile phenotype, with or without the addition of TGF-β1. H9-hESC-derived mSMLCs were cultured in medium containing 0.5% serum, with and without TGF-β1, for (A) an additional 6 days and (B) an additional 12 days and analyzed for the mRNA expression of relevant cytoskeleton and ECM genes. (C) Comparison of the mRNA expression of pathway regulators during the stages of differentiation and maturation. *P < 0.05, **P < 0.01, and ***P < 0.001.

An up-regulation in the expression of myocardin, a serum response factor (SRF) coactivator, through ERK was found to correlate with Con-vSMCs maturation (Figure 3C). The activation of the pathway proved more prominent in the hESCs than in the integration-free hiPSCs (Supplementary material online, Figure S5). Both SMAD3 and Yap1 were up-regulated in the Syn-vSMCs compared with Con-vSMCs. Interestingly, these data also suggest that TGF-β1 is imperative for the proper regulation of those pathways (Figure 3C).

3.4 Characterization of Syn-vSMC and Con-vSMC derivatives

We continued by characterizing the Con-vSMCs and Syn-vSMCs. Both are spindle shaped, with the Syn-vSMCs more elongated (Figure 4A). Filamentous cytoskeleton organization of F-actin, SMA, calponin, SM22, and SMMHC was observed in the Con-vSMC, but to a lesser extent in the Syn-vSMCs (Figure 4A). The production of elastin was detected in the Con-vSMCs but not in the Syn-vSMCs (Figure 4A), and the assembly of elastin was further detected after several days in culture (Figure 4B). Con-vSMCs proliferated slower than Syn-vSMCs (14.15 ± 4.20 vs. 83.19 ± 10.22%; Figure 4C). Finally, TEM analysis revealed that the Con-vSMCs have more (and more active) caveolae than the Syn-vSMCs, which have fewer caveolae (Figure 4Di-ii). The Con-vSMC has a larger endoplasmic reticulum (ER) than the Syn-vSMC (Figure 4Dii; Supplementary material online, Figure S6), while plentiful actin stress fibres (with occasion bundles) were observed in the Con-vSMCs (Figure 4Diii; Supplementary material online, Figure S6).

Figure 4

Characterization of Con-vSMCs and Syn-vSMCs. (A) The two phenotypes were characterized for morphology, using light microscopy (LM), and the organization of various cytoskeleton proteins, including calponin, SMA, SMMHC, SM22, and phalloidin, as well as the expression of elastin, all using immunofluorescence staining. (B) Con-vSMCs cultured for 6 days were analysed for elastin assembly using immunofluorescence staining. (C) Proliferative cells were detected using immunofluorescence staining of Ki-67 in both cell phenotypes. Scale bars are 100 μm. (D) TEM analysis showed (i) caviolae (arrowheads) with occasional fusion (asterisk); (ii) ER; and (iii) stress fibres (arrows). ER, endoplasmic reticulum.

3.5 Functionality

To determine functionality, we first measured contractility in vitro. Contraction studies indicated that Con-vSMCs contract significantly better than Syn-vSMCs; aortic vSMCs and Syn-vSMCs contract better than SMLCs; and Con-vSMCs contract similarly to the human aortic vSMC line (Figure 5A). Our earlier studies demonstrated that vSMC derivatives of human PSCs, which were synthetic by nature, migrate towards and support vasculature both in vitro5 and in vivo.31 Here, we examined Syn-vSMC and Con-vSMC interaction with newly formed functional blood vessels. We observed transplanted Syn-vSMCs and Con-vSMCs migrating to the vasculature and locating in the outer layers of the mouse blood vessels that penetrated into the matrigel plug (Figure 5Bi; Supplementary material online, Figure S6). In the case of Con-vSMCs, human elastin was further detected around some of the smaller mouse blood vessels that penetrated into the matrigel plug (Figure 5Bii-iii). On some occasions, the human Con-vSMCs were found to wrap the smaller mouse vasculature circumferentially (Figure 5C), narrowing the endothelial tube (Figure 5D). These were not observed with the Syn-vSMCs.

Figure 5

Functionality of hPSC derivatives. (A) Contractility rates of SMLCs (n = 9), Syn-vSMCs (n = 12), Con-vSMCs (n = 9), and human aortic vSMCs (n = 12). (B) Subcutaneously transplanted Con-vSMCs migrated to newly formed blood vessels within the Matrigel, as indicated by representative (i) confocal images; and (ii–iii) light microscopy images of human elastin-stained sections. Scale bars are 50 μm. (C) Occasionally, Con-vSMCs were found wrapping around the functional vasculature. The high-magnification inset shows the boxed area. (D) (i) 3D reconstruction of confocal microscopy images revealed an event of tube narrowing of small vessels by the transplanted Con-vSMCs further demonstrated by (ii) measurements of adjusted areas within the vessel with and without Con-vSMC wrapping. Scale bars are 20 μm. For confocal images, Con-vSMCs in r­­ed (some indicated with white arrows); mouse endothelial cells in green; and nuclei in blue. *P < 0.05, **P < 0.01, and ***P < 0.001.

Hence, to achieve the contractile or synthetic maturation of differentiating hPSCs, we propose the use of a stage-specific differentiation practice, with appropriate concentrations of factors known to control these developmental steps in the early embryo and in adulthood (Supplementary material online, Figure S7). Moreover, individual hPSC lines require the optimized administration of TGF-β1 for efficient maturation of contractile vSMCs. Such an approach enables the acquisition of the morphological features, cytoskeleton expression, and contractility typical for the contractile phenotype.

4. Discussion

Our previous studies demonstrated that we could derive vascular lineages from hESCs by administering angiogenic growth factors using a 2D monolayer differentiation protocol or by isolating vascular progenitor cells or CD34+ cells from 10-day-old EBs, followed by selective induction into either endothelial-like cells [using vascular endothelial growth factor (VEGF)] or SMLCs (using PDGF-BB).25,29 More recently, building on these initial studies, we established a simple step-wise differentiation protocol that cultured hPSCs in monolayers and supplemented them with PDGF-BB and TGF-β1, resulting in highly purified cultures of SMLCs.5 These SMLCs were >98% positive for SMA, calponin, and SM22 and ∼50% positive for SMMHC. They produced collagen and fibronectin, and they contracted in response to carbachol. Further in vitro tubulogenesis assays revealed that these hPSC-derived SMLCs interacted with human endothelial progenitor cells to support and augment the formation of cord-like structures.5 The current study sought to determine how these SMLCs make the synthetic vs. contractile phenotype decision.

Synthetic-vSMCs produce ECM proteins, such as collagen and fibronectin, as well as MMP proteins, in order to aid in cell migration.32 We first demonstrated that long-term (up to 30-day) cultures of the differentiated SMLCs in high serum with PDGF-BB and TGF-β1 resulted in maturation towards a synthetic phenotype, reducing the expression of contractile proteins and increasing the expression of ECM proteins, collagen, fibronectin, and MMPs. Indeed, both of these growth factors were suggested in early stages of differentiation.5,25,29 Attempts to eliminate only PDGF-BB or both growth factors from the culture media somewhat increased synthetic phenotype characteristics (i.e. SMMHC and elastin expression), suggesting that this strategy may prove useful for guiding the contractile phenotype. Nonetheless, after their long-term exposure to PDGF-BB and TGF-β1, these Syn-vSMCs seemed unable to acquire a contractile phenotype when deprived of serum and growth factor, suggesting a terminal synthetic phenotype.

To mimic the native state of vSMCs in vessels, we wanted to switch to a quiescent and contractile state.29 Quiescence is marked by the reduction of the proliferative capacity of a cell. Vascular SMCs in vessel walls replicate at the low frequency of 0.047% per day.33 In this low proliferative state, the vSMC becomes committed to its contractile function.11,30 Growth factors, as well as foetal calf serum, drive the proliferative capacity of vSMCs.34 However, we still do not know how the proliferative state of native vSMCs becomes suppressed. Moreover, it has been suggested that PDGF-BB interferes with vSMC maturation.3134 SMMHC has a high specificity for SMCs and is also considered a mature marker indicating a contractile phenotype.11 The ECM protein elastin also gets expressed in the contractile state.35,36 In adult vSMCs, elastin acts as an autocrine regulator and also determines mechanical responsiveness.37 Indeed, when SMLCs were matured in media containing low concentrations of serum and supplemented with TGF-β1, we saw the up-regulation of SMMHC and elastin in the mSMLCs. These mSMLCs seem to retain plasticity, as indicated by down-regulation of the contractile proteins SMMHC and elastin when differentiated in media containing high concentrations of serum.

Continued quiescence of mSMLCs in media containing low concentrations of serum and supplemented with or without TGF-β1 induced additional up-regulation in the expression of contractile proteins. These Con-vSMCs maintained their contractile phenotype when cultured in low-serum conditions; they exhibited plasticity with the down-regulated expression of contractile protein when cultured in high-serum concentrations.

Myocardin, a potent SRF coactivator expressed exclusively in vSMCs and cardiomyocytes,38 reportedly promoted SMC differentiation through transcriptional stimulation of SRF-dependent smooth muscle genes, including SMMHC.39,40 A recent study demonstrated that myocardin−/− mouse ESCs differentiate to vSMCs, suggesting the dispensability of myocardin for the development of vascular SMCs.41 In support of this observation, we report that deprivation of TGF-β1 seems to affect the activation of the different pathways, although up-regulation of contractile proteins was observed. Overall, our data using hPSCs show that up-regulating the myocardin pathway was not necessarily associated with the contractile state of the differentiating vSMCs. Finally, both Yap-1 and SMAD3 have been suggested as regulators important for inducing the synthetic phenotype in vSMCs.42,43 Our data suggest that these also get up-regulated during the synthetic phenotype maturation of hPSC derivatives. Here as well, deprivation of TGF-β1 seems to affect the activation of these pathways in contractile maturation. Additional studies to delineate the specific mechanism underlying these observations in hPSCs are required and will become the focus of our future investigations.

Comparing Con-vSMC and Syn-vSMC derivatives, we observed that both acquire a more spindle-shaped morphology than SMLCs. More prominent filamentous organization of the various cytoskeleton proteins was found in Con-vSMC than in Syn-vSMC derivatives. Interestingly, both cell derivatives showed increases in contractility: Syn-vSMCs showed some increased contractility, which may be attributed to the needed optimization of the culture period and to cell confluence; Con-vSMCs exhibit a rather greater increase in contractility than human aortic vSMCs, most likely due to higher SMMHC expression. Reducing the serum concentrations in media of SMLCs markedly decreased the proliferation rates of the cells and was accompanied by an increase in the contractile phenotype. Indeed, the Con-vSMC phenotype was marked by a reduced proliferative capacity, unlike the Syn-vSMC phenotype, which exhibited a high proliferative capacity. Finally, high-resolution analysis further revealed profound differences previously observed between the two phenotypes.44 Unlike Syn-vSMCs, Con-vSMCs exhibited numerous and active caveolae with enlarged ER and abundant stress fibres and bundles, underlining the distinctive shift between two major differentiated states with distinct morphological and functional properties.

Researchers envision human iPSCs—which can be derived directly from a patient, thereby reducing the risk of immunogenicity upon transplantationas dramatically revolutionizing cell-based therapies for regenerative medicine. Since Takahashi and Yamanaka's pioneering discovery,45 hiPSC technology has evolved rapidly. While the hiPSC technologies initially reported have several obvious shortcomings, many of these have recently been overcome. This study tested MR31—a hiPSC clone derived from the IMR90 line, which was derived from normal foetal lung fibroblasts using a lentivirus to deliver three reprogramming factors (Oct-4, Sox2, and Klf4)25—and BC1, which was induced using CD34+ blood cells from bone marrow using plasmids encoding all four reprogramming factors.26,27 We have shown that hiPSCs respond to the differentiation protocol similarly to hESCs and can mature into the synthetic and contractile phenotypes of vSMCs. The mSMLCs derived from all the hPSCs examined exhibited comparable expression levels of both SMMHC and elastin. We observed some differences during their long-term exposure to serum starvation with and without TGF-β1. Specifically, when culturing mSMLCs derived from MR31 in a low concentration of serum, with or without TGF-β1, we detected up-regulated elastin expression and down-regulated expression of SMMHCs. The derivation of vSMCs from the BC1 line, an integration-free-induced PSC line,26,27 offers a practical approach for using this clinically relevant technology for vascular regeneration. Thus, it seems apparent that hiPSCs have immense potential for providing effective treatments or cures for vascular diseases, which warrants further investigations and improvements.

Previous studies suggested that vSMCs wrap circumferentially rather than longitudinally around blood vessels.46,47 Some have suggested that this wrapping improves the mechanical properties29,48 of the vessel wall while also managing proper vasoactive activity.29 In early studies, we demonstrated the contribution of vSMC derivatives of a synthetic nature to growing vasculature.5,29 The current study tested whether the Con-vSMCs could still migrate towards a growing vessel, as well as begin wrapping. Utilizing a subcutaneous transplantation model assay, we have shown that Con-vSMCs encapsulated in Matrigel plugs migrate to sites near newly grown functional vasculature where they produce elastin that stabilizes those vasculatures. Moreover, the Con-vSMCs were sometimes found wrapping and even narrowing the host vessels. Such Con-vSMCs offer opportunities to use such derivatives to enhance the stabilization and maturation of new blood vessels in regenerating tissues.

In summary, the findings reported here demonstrate fate decisions in vascular smooth muscle phenotypes during the differentiation of hPSCs. By monitoring the expression of SMMHC and elastin, we demonstrate the possibility of generating synthetic or contractile phenotypes from different hPSC lines with appropriate concentrations of factors known to control these developmental steps in the early embryo and in adulthood. These findings highlight the importance of designing stage-specific differentiation strategies that follow key developmental steps to exploit cellular plasticity for vSMC phenotypic decisions. Finally, contractile hPSC-vSMCs derived from the integration-free hiPSC line BC1 may prove useful for regenerative therapy involving blood vessel differentiation and stabilization.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Conflict of interest: none declared.

Funding

This work was supported by an American Heart Association-Scientist Development Grant and the National Institutes of Health grant R01HL107938 (both to S.G.). M.W. is an IGERT trainee.

Acknowledgements

We thank Dr Michael McCaffery from the Integrated Imaging Center at JHU for assistance with TEM and confocal microscope imaging, Prof. Charles Steenbergen and Karen Fox-Talbot for immunohistochemistry of elastin, Yu-I (Tom) Shen and Donny Hanjaya-Putra for helping with animal studies, Prof. Linzhao Cheng for providing hiPSC lines, and Ying Wang for BC1 expansion.

References

View Abstract