Introduction

The ability to respond to mechanical stresses, such as stretching tension and frictional force, is a property of most of the cells in the human body. While some mechanical stresses, including gravity, originate in the external environment, others are generated inside the body. The fluid mechanical forces in blood vessels are generated by flowing blood and blood pressure, and when vascular cells are exposed to fluid mechanical forces, they actively respond to them by altering their morphology, function, and gene expression [1]. Vascular cell responses to fluid mechanical forces play important roles in sustaining the homeostasis of the circulatory system, and various vascular diseases develop when vascular cell responses are impaired. The same holds true for other vessels, such as the tubules and collecting ducts in the kidney, the airway, and the intestines [2–5]. Cells that make up such structures are known to control the physiological functions of these tissues and organs in response to fluid mechanical forces.

It has recently become clear that fluid mechanical forces affect immature and undifferentiated cells as well as adult cells. For example, fluid mechanical forces have been shown to control embryonic development and organogenesis: intracardiac fluid forces are essential to the formation of a functional heart in zebrafish embryos, and the direction of fluid flow on the node of mouse embryos determines the left-right asymmetry of their body plan [6, 7]. Moreover, immature cells, such as endothelial progenitor cells (EPCs) and embryonic stem (ES) cells, have been found to differentiate into vascular cell lineage in response to fluid mechanical forces. This review focuses on the differentiation of stem/progenitor cells into vascular cells in response to fluid mechanical forces.

Shear stress and endothelial cells

The endothelial cells (ECs) that line blood vessels are in direct contact with flowing blood and thus are constantly exposed to shear stress, a flow-generated mechanical force. The intensity of shear stress (τ) can be calculated by using the formula τ = 4 μQ/πr3, where μ is blood viscosity, Q is blood flow volume, r is the radius of the blood vessel, and π is the ratio of the circumference of a circle to its diameter. Under physiological conditions, arterial ECs are exposed to shear stress ranging from 10 to 20 dynes/cm2 and venous ECs to a shear stress ranging from 1.5 to 6 dynes/cm2 [8].

Various types of flow-loading apparatus, including a cone-plate type, a parallel-plate type, a silicone-tube type, and a microfluidic type, have been developed to apply controlled levels of shear stress to cultured cells. When cultured ECs are exposed to laminar shear stress in a flow-loading apparatus, they become elongated with their long axis oriented in the direction of flow [9]. Shear stress also alters endothelial cell functions: it stimulates the production of vasodilators, such as nitric oxide (NO) and prostacyclin, and the cell-surface expression of an anti-thrombotic molecule, thrombomodulin [10–12].

Many of the alterations of EC functions induced by shear stress are accompanied by changes in the expression of related genes. DNA microarray analysis of cultured ECs has revealed that approximately 3% of all EC genes respond to a laminar shear stress of 15 dynes/cm2 for 24 h by significantly increasing or decreasing their level of expression [13]. Assuming that ECs express about 20,000 genes, that would mean that approximately 600 genes are shear stress-responsive. Shear stress regulates endothelial gene expression transcriptionally and/or posttranscriptionally [14]. The transcriptional regulation is achieved by activation of transcription factors, such as AP-1, NFκB, Egr-1, SP-1, GATA 6, and KLF2, which bind to shear stress-response elements in the gene promoter that regulates transcription of the gene. The posttranscriptional regulation is achieved by activating RNA-binding proteins that bind to mRNA and alter its stability. These EC responses to shear stress are thought to play important roles in blood-flow-dependent phenomena, such as angiogenesis, vascular remodeling, and atherogenesis, as well as in cardiovascular homeostasis.

Shear stress and endothelial progenitor cells

New blood vessels form in cardiac and skeletal muscle during development and physical training, and they form in tissues under pathological conditions, such as tumorigenesis, wound healing, and ischemia. New blood vessels in the adult were thought to be produced only by angiogenesis, that is, by migration and proliferation of ECs sprouting from a preexisting vascular wall. Recently, however, bone-marrow-derived EPCs circulating in peripheral blood have been shown to home to sites of active neovascularization, where they differentiate and proliferate, thereby contributing to the formation of new vessels [15, 16]. During the process of EPC incorporation into tissues and neovascularization, EPCs are exposed to shear stress generated by blood flow or interstitial fluid flow.

Endothelial progenitor cells isolated from human peripheral blood exhibited no particular orientation under static conditions, but when they were exposed to controlled levels of laminar shear stress for 24 h in a cone-plate type flow-loading apparatus, they became elongated with their long axes oriented in the direction of flow (Fig. 1a) [17]. EPCs also altered their functions in response to shear stress. Under static culture conditions, the cell density of EPCs was found to increase with time, but to level off after day 10, whereas exposure to shear stress resulted in an increase in cell density to a much higher level than in static cell culture (Fig. 1b), indicating that laminar shear stress stimulates EPC proliferation. Recent studies have demonstrated that shear stress stimulates the production of NO, tissue-type plasminogen activator (tPA), plasminogen activator inhibitor-1 (PAI-1), prostacyclin, and Cu/Zn superoxide dismutase (SOD) by human EPCs [18–20].

Fig. 1
figure 1

Effects of shear stress on EPCs. a EPC morphology. Phase-contrast photomicrographs of EPCs cultured under static conditions (Static) and EPCs exposed to shear stress (Shear, 2 dynes/cm2 for 24 h). The EPCs became elongated and aligned in response to shear stress. The direction of flow is left to right. b Cell growth. Photomicrographs obtained on culture days 4, 6, 8, 10, and 14 were used to measure cell density. Values shown are mean ± SD of data from five separate samples. *P < 0.001 versus the static control. c Cell differentiation. EPCs cultured under static conditions (Static) and EPCs exposed to shear stress were labeled with fluorescent antibodies to the EC-specific markers VEGFR2, VEGFR1, and VE-cadherin. The fluorescence intensity of 20,000-cell samples was measured by FACS. The vertical axis represents mean fluorescence. Shear stress accelerated increases in the expression of these EC-specific markers, indicating that shear stress promotes EPC differentiation into ECs. Values shown are mean ± SD of data from four separate samples. *P < 0.01 versus the static control

Full size image

The patterns of cell surface expression of various EC-specific markers were examined in EPCs when cultured under static conditions and when subjected to shear stress [17]. Flow cytometric analysis showed a time-dependent increase in expression of the receptors for vascular endothelial growth factor, Flk-1 (VEGFR2) and Flt-1 (VEGFR1), and of the cell-to-cell adhesion molecule VE-cadherin in EPCs cultured under static conditions (Fig. 1c). Shear stress resulted in a marked increase in these EC markers, and in EPCs exposed to shear stress the level of VEGFR2, VEGFR1, and VE-cadherin expression on day 8, day 10, and day 8, respectively, was almost identical to its level on day 21 in the EPCs cultured under static conditions. Shear stress also enhanced the ability of EPCs to form capillary-like tubes in collagen gels (Fig. 2a). These findings indicate that shear stress accelerates EPC differentiation into vascular ECs and that shear stress may promote new vessel formation in vivo by stimulating EPC proliferation and differentiation. A recent study showed that shear stress significantly promoted endothelial differentiation by EPCs, particularly when the EPCs were cocultured with vascular smooth muscle cells (SMCs), suggesting that EPC differentiation is influenced by SMCs [21]. The combined effect of shear stress and SMCs on EPC differentiation has been demonstrated to be mediated by the Akt signal pathway.

Fig. 2
figure 2

Effect of shear stress on tube formation in collagen gel. a EPCs incubated under static conditions (Static) and EPCs exposed to shear stress (Shear, 0.1–2.5 dynes/cm2 for 24 h) were seeded onto collagen gels produced with Biocoat Matrigel (BD Biosciences). After 4 days, clear tube-like structures had been formed by the shear-stressed cells, but not by the static cells. b ES-cell-derived VEGFR2+ cells that had been incubated under static conditions (Static) and that had been exposed to shear stress (1.5 dynes/cm2) for 24 h were seeded onto collagen gels. No tube-like structures had formed by 24 h after seeding of the static cells, but they had formed after seeding of the shear-stressed cells, clearly indicating that advance exposure to shear stress had promoted the formation of tube-like structures by ES-cell-derived VEGFR2+ cells

Full size image

Placenta-derived multipotent cells (PDMCs), a population of CD34−/CD133−/VEGFR2− cells at the time they are isolated from a human term placenta, are known to be capable of multilineage differentiation. Culture of PDMCs in medium containing EC growth factors under static conditions resulted in significant increases in expression of VEGFR1 and VEGFR2, and application of shear stress to these cells led to a marked increase in expression of von Willebrand factor and platelet endothelial cell adhesion molecule-1 (PECAM-1), and in uptake of acetylated low-density lipoproteins, and in the formation of tube-like structures on Matrigel [22]. These findings suggest that exposure to a combination of biochemical and mechanical stimuli has a synergistic effect in promoting the endothelial differentiation of PDMCs. A similar synergistic effect of biochemical and mechanical stimuli on endothelial differentiation was observed in human adipose tissue-derived stem cells [23].

Shear stress and embryonic stem cells

Embryonic stem cell lines have been established from the inner cell mass of mouse blastocysts and have the potential to differentiate into all embryonic cell lineages [24]. ES cells are now attracting interest as a promising source of cells for use in regenerative medicine. They have already been shown to be capable of being induced to develop into a variety of cell types, including neural cells, cardiomyocytes, blood cells, ECs, and chondrocytes [25–28], but much remains to be elucidated in regard to the molecular mechanisms of ES cell differentiation and methods of inducing ES cells to differentiate into various specialized cells in vitro.

A novel method of inducing selective differentiation of ES cells into both vascular ECs and mural cells (pericytes and SMCs) has recently been developed in which undifferentiated ES cells are cultured on type-IV collagen-coated dishes, and VEGFR2+ cells are purified by flow cytometry sorting [29]. Addition of VEGF to VEGFR2+ cells promotes endothelial differentiation, whereas differentiation into mural cells is induced when platelet-derived growth factor BB (PDGF-BB) is added.

During embryonic development ES cells are exposed to tissue fluid flow or blood flow generated by the beating heart, and recent studies on the hemodynamics of the mammalian embryo have revealed shear stress levels of between 0 and 5.5 dynes/cm2 in embryos from 8.5 to 10.5 dpc [30]. In order to investigate whether ES cells respond to shear stress, murine ES-cell-derived VEGFR2+ cells were subjected to controlled levels of shear stress in a flow-loading apparatus [31]. Under static conditions, cell density increased with time and plateaued on day 4, and addition of a maximally effective concentration of VEGF165 (50 ng/ml) markedly increased the cell density of static cells (Fig. 3a). Application of shear stress induced a much greater increase in cell density than occurred under static conditions, and the increase induced by shear stress was significantly larger than the increase induced by VEGF165. These findings indicate that shear stress stimulates VEGFR2+ cell proliferation.

Fig. 3
figure 3

Effects of shear stress on ES-cell-derived VEGFR2+ cells. a Cell density, as a function of time, of VEGFR2+ cells cultured under static conditions (Static) or exposed to shear stress (Shear). Under static conditions, VEGFR2+ cells proliferated, and their proliferation was significantly promoted by the addition of VEGF165 (50 ng/ml, R&D Systems). When exposed to a shear stress of 5 dynes/cm2, the VEGFR2+ cells proliferated much more than the static cells, exceeding the level of proliferation induced by VEGF165. Values shown are mean ± SD of data from five separate samples. *P < 0.001 versus the static control. b Cell differentiation. VEGFR2+ cells were incubated under static conditions (Static) or subjected to shear stress (5 dynes/cm2), and after 24 h they were immunostained with antibodies against an EC-specific marker, PECAM-1, and a mural-cell-specific marker, SM-α-actin. Most of the static cells were SM-α-actin-positive (brown), and few cells were PECAM-1 positive (purple). Exposure to shear stress caused marked expansion of the PECAM-1-positive cell sheets. The direction of flow is from left to right. c EC-specific marker expression. VEGFR2+ cells cultured under static conditions (Static) and VEGFR2+ cells exposed to shear stress (5 dynes/cm2) were labeled with fluorescent antibodies to VEGFR2, VEGFR1, and VE-cadherin, and the fluorescence intensity of 20,000-cell samples was measured by FACS. Cells were continuously exposed to shear stress for 24–96 h starting at culture day 3, and they were assayed every 24 h. Shear stress accelerated the increases in expression of VEGFR2, VEGFR1, and VE-cadherin, indicating that shear stress promotes VEGFR2+ cell differentiation into ECs. Values shown are mean ± SD of data from four samples. *P < 0.01 versus the static control

Full size image

VEGFR2+ cells that had either been cultured under static conditions or exposed to shear stress (5 dynes/cm2) for 24 h were immunostained for an EC marker, PECAM-1, and a mural cell marker, smooth muscle α-actin (SM-α-actin) (Fig. 3b). Under static conditions, more than 90% of the cells were SM-α-actin-positive (brown), and there were few PECAM-1-positive cells (purple), whereas when exposed to shear stress for 24 h, most of the culture consisted of PECAM-1-positive cells. The effect of shear stress on cell differentiation was confirmed by the time course of expression of various cell lineage markers as determined by flow cytometry (Fig. 3c). Under static conditions, expression of VEGFR2, VEGFR1, and VE-cadherin had increased at day 4, but declined thereafter. Although expression of these EC markers increased markedly when the cells were exposed to shear stress, there were no marked differences from the static controls in expression of SM-α-actin, the blood cell marker CD3, or the epithelial cell marker keratin. Taken together, the above results indicate that shear stress promotes differentiation of murine ES-cell-derived VEGFR2+ cells into the vascular EC lineage.

VEGFR2+ cells cultured under static conditions and VEGFR2+ cells exposed to shear stress for 24 h were seeded in collagen gels, and the gels were microscopically examined for tube formation (Fig. 2b). Control cells cultured under static conditions exhibited no tube-like structures at 24 h, whereas the shear-stressed VEGFR2+ cells had formed an extensive tubular network, indicating that shear stress enhances the ability of VEGFR2+ cells to form tube-like structures in collagen gel.

When exposed to shear stress, murine ES-cell-derived VEGFR2+ cells became elongated and aligned parallel to the direction of flow, which closely resembled the flow-induced morphological changes in adult murine aortic ECs [32]. Human ES-cell-derived ECs also elongated and became aligned in the direction of flow, and their gene expression profile changed: expression of the gene encoded cyclooxygenase 2 (COX2) and membrane metalloproteinase 1 (MMP1) became upregulated in response to shear stress, whereas expression of the gene encoding monocyte chemoatractant protein 1 (MCP1), vascular adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and tPA became downregulated [33]. Similar effects of shear stress on cell differentiation have been observed in embryonic mesenchymal progenitor cell line C3H/10T1/2, which has the ability to differentiate into osteocytes, chondrocytes, adipocytes, and SMCs [34]. When the cells were exposed to a shear stress of 15 dynes/cm2 for 12 h, expression of PECAM-1, VE-cadherin, and von Willebrand factor increased at both the protein and mRNA levels. Uptake of acetylated low-density lipoprotein and tube-like structure formation in collagen gel both increased in the cells exposed to shear stress. Thus, shear stress may be an important factor in the differentiation of cardiovascular precursors during maturation of the embryonal vasculature.

Shear stress and arterio-venous differentiation

The development of arteries and veins in the embryo has been assumed to depend on the influence of physiological conditions in the environment, such as oxygenation, blood pressure, and shear stress [35, 36], but its cellular and molecular mechanisms are still poorly understood. Recent studies have shown that arteries and veins can be distinguished on the basis of the presence or absence of several molecules that are specifically expressed by the ECs of arteries or veins alone. For example, ephrinB2, an Eph family transmembrane ligand, marks arterial ECs, whereas EphB4, a receptor for ephrinB2, marks venous ECs [37–39]. When murine ES-cell-derived VEGFR2+ cells were exposed to shear stress, expression of ephrinB2 increased dose-dependently (Fig. 4a) [40]. The ephrinB2 mRNA levels also increased in response to shear stress, whereas the mRNA levels of EphB4 decreased. These findings suggest that shear stress plays a role in the artery and vein EC specification of ES cells.

Fig. 4
figure 4

Effect of shear stress on arterio-venous differentiation of ES-cell-derived VEGFR2+ cells. a Western blot of the arterial EC marker ephrinB2 in VEGFR2+ cells cultured under static conditions (Static) and VEGFR2+ cells exposed to shear stress for 48 h. β-Actin was used as a protein loading control. The lower panel shows the results of a quantitative analysis by densitometry. EphrinB2 protein increased dose-dependently in response to shear stress. Values are mean ± SD of data from five separate samples. *P < 0.01 versus the static control. b Western blot of cleaved Notch1 in the nucleus. Shear stress caused Notch cleavage and translocation of the NICD into the nucleus in a time-dependent manner. The VEGFR2 kinase inhibitor SU1498 (10  μM), γ-secretase inhibitors DAPT (2.5 μM) and L685,458 (0.1 μM), and recombinant extracellular domain of the Notch ligand DLL4 (rmDLL4, 1 μg/mL) abolished the NICD translocation that occurred in the cells exposed to shear stress for 1 h, indicating that VEGFR, the notch ligand DLL4, and γ-secretase are involved in the Notch activation by shear stress. Values shown are mean ± SD of data from six samples. *P < 0.01 versus the static control

Full size image

Notch signaling has been shown to be involved in the effect of shear stress on ephrinB2 expression. Notch cleavage and translocation of the Notch intracellular domain (NICD) into the nucleus occurred as early as 30 min after the start of exposure to shear stress, and γ-secretase inhibitors (DAPT and L685,458) that block Notch cleavage, the recombinant extracellular domain of the Notch ligand DLL4 (mDLL4), and the VEGFR2 kinase inhibitor (SU1498) abolished the shear stress-induced NICD translocation (Fig. 4b); that, in turn, blocked the shear stress-induced up-regulation of ephrinB2 expression. These findings suggest that shear stress increases expression of ephrinB2 via the VEGF-Notch signaling pathways. In vivo, the differentiation of arteries and veins in the chick embryo has been found to be governed by hemodynamic forces, and ephrinB2 expression has been demonstrated to be controlled by blood flow [41].

A similar effect of shear stress on ephrinB2 expression has been observed in EPCs [42]. When cultured human EPCs were exposed to shear stress in a flow-loading device, the ephrinB2 mRNA levels increased, but the EphB4 mRNA levels decreased. EphrinB2 protein increased in the shear-stressed EPCs, and the increase in ephrinB2 expression was due to activation of gene transcription. Shear stress markedly increased binding of transcription factor Sp1 to its consensus sequence in the ephrinB2 gene promoter, and a mutation in its consensus sequence abolished shear stress-induced ephrinB2 transcription. These results indicate that shear stress induces differentiation of EPCs into arterial ECs by increasing ephrinB2 expression through Sp1 activation. Thus, fluid mechanical forces, as well as the genetic program, appear to play a critical role in artery-vein specification.

Shear stress mechanotransduction

The fact that shear stress affects the proliferation and differentiation of ES cells suggests that ES cells recognize shear stress and transmit signals into the cell interior; however, the mechanism of the shear stress mechanotransduction in ES cells is not yet fully understood. VEGFR has been shown to be involved in the shear-stress mechanotransduction of ES cells, and the shear stress-induced proliferation and differentiation of ES cells is almost completely blocked by SU1498, which specifically inhibits the enzymatic activity of VEGFR2 kinase and downstream events [31]. Also, shear stress causes tyrosine phosphorylation of VEGFR2 in a dose-dependent manner that is not blocked by a specific VEGF-neutralizing antibody. Based on all of these findings, shear stress appears to activate VEGFR2 in a ligand-independent manner, which is consistent with the findings observed in adult ECs [43]. Shear stress causes ligand-independent activation of VEGFR2 in bovine aortic ECs that results in phosphorylation of NO synthase and activation of IκB, which regulates transcription factor NFκB. The mechanism of the VEGFR2 activation by shear stress, however, remains to be elucidated. A recent study showed that shear stress activates histone deacetylase 3 (HDAC3) in murine ES cells through the VEGFR2-PI3K-Akt pathway, and that the HDAC3 in turn deacetylates p53, leading to p21 activation [44]. The VEGFR2-PI3K-Akt-HDAC3-p53-p21 pathway has been shown to be crucial to shear stress-induced ES cell differentiation into ECs. Another recent study demonstrated that shear stress modulates histone acetylation and promotes chromatin remodeling in undifferentiated ES cells, and that it activates transcription factors, such as myocyte enhancer factor-2C (MEF-2C) and Smad4, in a complex form with CBP/p300 histone acetyltransferases (HATs), thereby providing a molecular basis for its promotion of the cardiovascular differentiation of ES cells [45].

In adult ECs, on the other hand, shear stress is known to activate multiple signal transduction pathways involving ion channels, G-protein coupling receptors, adhesion molecules, such as PECAM-1 and integrins, the cytoskeleton, the glycocalyx, and primary cilia, but which pathways are primary and which are secondary remains unclear, nor has the initial sensor that recognizes shear stress been identified [1]. Shear stress may activate several pathways simultaneously. When ECs are exposed to shear stress, there is an immediate, dose-dependent influx of extracellular Ca2+ across the cell membrane [46]. The ATP-gated P2X4 ion channel expressed on ECs plays a key role in the shear stress-dependent Ca2+ influx. Mice lacking the P2X4 gene do not exhibit normal EC responses to shear stress, such as a Ca2+ influx and subsequent production of the potent vasodilator NO, and as a result their shear-dependent control of vascular tone and remodeling is impaired [47]. The ATP released by ECs has recently been found to be involved in the shear stress-induced activation of P2X4 channels [48]. Many types of cells, including mammary gland cells, bladder epithelial cells, and ECs, are known to release ATP in response to mechanical stimuli, and the ATP released mediates mechanotransduction in an autocrine or paracrine manner [49]. Thus, mechano-ATP coupling may play a role in shear stress mechanotransduction in EPCs and ES cells.

Cyclic strain and embryonic stem cells

Cyclic strain affects the phenotype of adult vascular cells, including ECs and SMCs [50], and it has recently become clear that cyclic strain also regulates the differentiation of bone marrow mesenchymal progenitor cells (BMPCs) and ES cells. When rat BMPCs were exposed to cyclic strain (19%, 1 Hz) for 7 days, the cells became aligned perpendicular to the direction of strain, and their proliferation was markedly inhibited [51]. The expression of smooth muscle markers, such as SM-α-actin and hi-calponin, increased in rat BMPCs in response to cyclic strain, indicating that cyclic strain induces the differentiation of BMPCs into the SMC lineage. Similarly, cyclic strain has been shown to increase gene expression of the smooth muscle markers SM-α-actin and SM22-α in BMPCs within 1 day after the start of exposure [52].

Embryonic stem cells have been shown to respond to cyclic strain by differentiating into the cardiovascular cell lineage. Application of cyclic strain to murine ES cells by the Flexercell system increases gene expression of VEGF and hypoxia-inducible factor-1 (HIF-1), which play an important role in neovascularization, and gene expression of such transcription factors as MEF-2C and GATA4, which are involved in cardiomyogenesis [53]. These effects of cyclic strain are mediated by reactive oxygen species produced in ES cells through activation of NADPH oxidase.

When murine ES-cell-derived VEGFR2+ cells seeded on flexible silicone membranes were exposed to cyclic strain (4–12% strain, 1 Hz, 24 h), expression of SM-α-actin and SM-myosin heavy chain (SM-MHC) significantly increased in a dose-dependent manner, whereas expression of VEGFR2 decreased, and there were no changes in VEGFR1, VE-cadherin, PECAM-1, CD3, or keratin [54]. These findings suggest that cyclic strain induces selective differentiation of VEGFR2+ cells into the SMC lineage and not into other cell lineages. Upregulation of SMC markers by cyclic strain has been observed in other immature cell lines, such as rat bone marrow progenitor cells [51], human bone marrow mesenchymal stem cells [52], and murine embryonic mesenchymal progenitor cells [55]. In contrast, there is a report indicating that cyclic strain inhibits ES cell differentiation [56]. Exposure of ES cells to cyclic strain (14%, 10 cycles/min) for 14 days was found to increase the percentage of cells expressing marker proteins for undifferentiated cells, such as SSEA-4 (stage-specific embryonic antigen-4) and Oct4 (a POU domain transcription factor), in comparison with ES cells cultured under static conditions.

Interestingly, cyclic strain immediately caused PDGF receptor (PDGFR) phosphorylation in murine ES-cell-derived VEGFR2+ cells in a ligand-independent manner [54], and PDGFR phosphorylation plays a critical role in SMC differentiation by these cells. Thus, cyclic strain and shear stress may act via a common mechanism in which growth factor receptors are activated by mechanical forces without ligand binding. Since ES cells seem to be exposed to both cyclic strain and shear stress in vivo, in order to understand the roles of fluid mechanical forces in cardiovascular differentiation and development in embryo, it will be necessary to know not only how ES cells respond to cyclic strain alone and to shear stress alone, but how they respond to combinations of the two.

Applications to tissue engineering

Knowledge of the effects of shear stress and cyclic strain on adult cells and immature cells opens up new possibilities for application of fluid mechanical forces to tissue engineering as a means of manipulating cell functions. The development of cell-incorporated engineered blood vessel grafts has been progressing ever since Weinberg and Bell [57] produced the first tissue-engineered vessel. Several groups have used mechanical conditioning of cells to develop tissue-engineered vascular grafts. Exposing cells cultured on distensible substrata to shear stress and cyclic strain has been shown to stimulate tissue growth by enhancing ECM synthesis and increasing cell proliferation, as well as to restore cells to their normal in vivo phenotypes [58, 59]. The histological appearance of these tissue-engineered vessels is similar to that of native arteries, and they have the ability to contract in response to pharmacological agents.

Investigators have recently explored the use of EPCs and ES cells to develop artificial arteries. When EPCs obtained from human peripheral blood were seeded on the gelatin-coated luminal surface of microporous polyurethane tubes (1.5 mm in diameter) and exposed to a shear stress of 30 dynes/cm2 for 12 h, the EPCs formed a confluent monolayer of cells that covered the inner surface of the tubes and were aligned in the direction of the shear stress [60]. Also, small diameter vessels (4 mm in diameter) were created by seeding EPCs isolated from the peripheral blood of sheep onto decellularized porcine iliac arteries. When the EPC-explanted grafts were preconditioned over 2 days by gradually increasing the shear stress from low (1 dynes/cm2) to high (25 dynes/cm2), they began to exhibit contractile activity and NO-mediated vascular relaxation, similar to native arteries. The grafts were used to replace a segment of the carotid artery in lambs by creating two end-to-end anastomoses. The EPC-seeded grafts remained patent for 130 days, whereas most of the non-seeded grafts were occluded by thrombus formation within 5 days [61]. In addition, human EPCs seeded with human SMCs on biodegradable polymer scaffolds have been shown to form capillary-like structures [62]. These findings indicate that EPCs are a suitable cell source for vascular tissue engineering.

Fluid mechanical forces have been shown to simultaneously induce ES cell differentiation into ECs and SMCs on a three-dimensional scaffold [63]. Murine ES cells were seeded on compliant microporous polyurethane tubes coated with fibronectin and incubated under static conditions, first in the presence of VEGF for 2 days and then in the absence of VEGF for 2 days. Layering of the cells that had grown, mostly SM-α-actin-positive cells, was observed only on the luminal surface of the tube, and the cells were flat, polygonal, and randomly oriented. After 2 days of incubation under shear stress and circumferential strain in the absence of VEGF, on the other hand, the cells in the superficial layer were regularly oriented in the direction of the flow and positive for PECMA-1, whereas the cells growing into the interstices deeper in the tube wall were positive for SM-α-actin. These findings suggest that differentiation into two different cell types and segregation of incorporated ES cells are induced by fluid mechanical forces, and that preconditioning of ES cells by exposure to fluid mechanical forces may be useful in developing hierarchically structured hybrid blood vessel grafts composed of several vascular wall cell types.

There is also evidence that ES cells can be used as a source of seed cells for cardiac tissue engineering [64]. Embryoid bodies obtained from murine ES cells were induced to differentiate into cardiomyocytes in medium containing ascorbic acid, and cardiomyocytes were enriched and mixed with type I collagen to construct engineered cardiac tissue. After in vitro stretching for 7 days, the engineered cardiac tissue beat synchronously and responded to physical and chemical stimulation, and it resembled neonatal native cardiac muscle both structurally and functionally.

Conclusions

Shear stress and cyclic strain activate vascular cells to release a variety of bioactive mediators, and they play an important role in the control of circulatory functions. It has recently become clear that fluid mechanical forces modulate the functions of immature cells as well as adult cells. ES cells and bone marrow-derived EPCs proliferate and differentiate into adult vascular cells in response to shear stress and cyclic strain, suggesting a possible role of fluid mechanical forces in the formation of new blood vessels. Elucidation of their role is very important to understanding the complex mechanisms of angiogenesis and vasculogenesis in vivo. Fluid mechanical forces also appear to be useful as a means of cell manipulation in regenerative medicine. For example, cell therapy with autologous EPCs is currently being used to treat patients with leg arteries occluded by atherosclerosis. Preconditioning EPCs with fluid mechanical forces prior to injection into a patient’s leg tissue can expand the EPC population and accelerate EPC differentiation, which may lead to better results. As stated in this review, application of fluid mechanical forces to EPCs and ES cells is already being used to develop tissue-engineered blood vessel grafts and cardiac tissue. The creation of targeted tissues and organs for use in regenerative medicine by using mechanical forces in combination with chemical mediators, such as cell growth and differentiating factors, will continue to advance in the near future.