Extracellular Matrix Differentiating Good Flow Versus Bad Flow
See related article, pages 995–1003
Mounting evidence indicates that shear stress, the tangential component of blood flow acting on the vessel wall, is crucial for endothelial functions and pathophysiology. From the vascular biology point of view, arterial areas under laminar flow, with high mean shear stress and little bidirectional oscillation, are atheroprotective, whereas areas under disturbed flow, with oscillation, reverse flow, and low mean shear stress, are atheroprone. This principle holds true for vessel specimens from both human patients and various animal models. To assess the effect of distinct flow patterns at the molecular level, immunocytochemical methods are commonly used to detect the expression of genes in areas experiencing laminar versus disturbed flow. For example, the expression of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1, indicative of vascular inflammation, is markedly enhanced in regions prone to atherosclerosis in low-density lipoprotein receptor-null mice fed an atherogenic diet.1
From the perspective of fluid mechanics, laminar flow has low Reynolds number and is characterized by smooth, constant fluid motion. In contrast, turbulent flow has high Reynolds number and tends to cause eddies, vortices, and other flow disturbances. Flow channels have been used as in vitro models to investigate the mechanisms by which shear stress modulates endothelial functions, because they permit close control over the fluid mechanics imposed on cultured vascular endothelial cells (ECs). Advances in molecular and cell biology, in conjunction with flow channel experiments, have facilitated investigations into the mechanotransduction mechanism by which shear stress modulates gene expression in cultured ECs. Many second-messenger systems in ECs have been shown to be mechanosensitive. More than a decade ago, several research groups reported that mitogen-activated protein kinases (MAPKs), including c-Jun N-terminal kinase (JNK), can be activated by shear stress: exposure to a step change of the applied shear stress causes a transient activation of JNK.2,3 JNK, a kinase that binds and phosphorylates c-Jun, responds to environmental stress stimuli such as inflammatory cytokines, heat shock, irradiation, and osmotic shock.4 In ECs, JNK is involved in the induction of inflammation-induced adhesion molecules, including vascular cell adhesion molecule-1, intercellular adhesion molecule-1, and selectins (P-, E-, and L-selectins). In vivo, the administration of SP600125, a pharmacological inhibitor of JNK, decreased lesion development in atheroprone areas of apolipoprotein E knockout (ApoE−/−) mice.5
In this issue of Circulation Research, Hahn et al6 report that the onset of laminar flow and/or long-term oscillatory flow can activate JNK. Apparently, the JNK activation is extracellular matrix (ECM)-dependent, with enhanced activity in ECs seeded on fibronectin rather than basement membrane protein or collagen. These results suggest that ECM is mechanosensitive. The ECM types in the subendothelial space would determine the responsiveness of ECs to the imposed flow patterns in terms of the temporal activation of JNK. Indeed, the authors found a higher level of fibronectin and phosphorylated JNK in atheroprone than in atheroprotective areas of C57BL and ApoE−/− mice. Atherosclerosis in the vascular wall is associated with changes in ECM deposition and modification. One such event is the assembly of newly deposited fibronectin into a fibrillar network.7 Thus, fibronectin deposited in the atheroprone areas may be particularly sensitive to the “bad flow” namely, unsteady, disturbed flow patterns.
ECs rely on a specific ECM–integrin interaction to mediate both “inside-out” and “outside-in” signaling communications with ECM. The integrins α1β1 and α2β1 interact with collagen, whereas α5β1 and αvβ3 bind to fibronectin. In human umbilical vein ECs plated on fibronectin and in which αvβ3 integrin is blocked by LM609 anti-αvβ3 monoclonal antibody, shear stress–activated JNK is attenuated.8 In the present study, Hahn et al6 applied 16G3 monoclonal antibody to block the integrin-binding sites on fibronectin. Consequently, JNK activation by shear stress was abolished. These complementary approaches reinforce the model that a dynamic interaction between integrins and their cognate ECM proteins is a very early event in mechanotransduction in ECs responding to shear stress.9
MAPK family members are phosphorylated by specific MAPK kinases (ie, MAP2K). MKK4 and MKK7 are 2 MAP2Ks known to phosphorylate and hence activate JNK. By using small interfering RNA against MKK4 or MKK7, Hahn et al6 demonstrated that MKK4 but not MKK7 is essential for the shear stress-activated JNK. Conceptually, a sudden onset of flow would cause the dynamic interaction of fibronectin with α5β1 and/or αvβ3 integrins, which in turn activates the MKK4-JNK pathway. This mechanosensitive signaling would be constitutively active for ECs exposed to disturbed flow patterns. However, this proinflammatory pathway would be quiescent for cells under prolonged laminar flow. Apparently, the duration of this mechanosensitive mechanism is crucial for regulating the expression of adhesion molecules and inflammatory cytokines that exacerbate endothelial dysfunction. A transient activation of MKK4-JNK may represent an immediate response of ECs to sudden changes in the mechanical environment, regardless of the applied flow patterns. Under laminar flow, the activation of MKK4-JNK quickly subsides. Somehow, disturbed flow patterns sustain the activation of this pathway, which results in inflammatory responses in ECs.
Some issues remain to be investigated. Although the synthesis, processing, and deposition of some ECMs such as fibronectin are associated with atherogenesis, whether shear stress plays any role in the change of ECM in the atheroprone areas is unknown. During their change from a contractile to a synthetic phenotype, vascular smooth muscle cells express a distinct ECM profile.10 Shear stress may also regulate the expression of ECM by ECs, depending on the imposed flow patterns. In addition, shear stress may alter the ECM composition simply through the mechanical shearing forces. JNK has several isoforms: JNK1, JNK2, and JNK3. Knockout of JNK2 but not JNK1 decreases lesion development in mice with an ApoE−/− background.1 However, tumor necrosis factor α–mediated cellular processes seem to be controlled mainly by JNK1 and not JNK2.11 Whether any of the JNK isoforms in ECs is uniquely activated by shear stress is unclear. If they are, of interest is the molecular basis underlying the unique mechanical response. The specificity of MAP2K should also be addressed. Presumably, the activation of MKK4-JNK is at the downstream of fibronectin interacting with α5β1 and αvβ3 integrins that resulted from shear stress stimulation. Other MAPKs, including extracellular signal-regulated kinases and p38, are also activated in ECs by shear stress. The corresponding MAP2Ks could be activated upstream, leading to the activation of these MAPKs. Thus, the comprehensiveness of the mechanosensitive ECM–integrin interaction and the consequent MAPK cascade in response to different flow patterns deserves further study. These experiments will provide new insights into the integrative responses of ECs in health and disease under various mechanical environments.
Hahn et al6 suggest further that an assembly of platelet endothelial cell adhesion molecule (PECAM)-1, vascular endothelial (VE)-cadherin, and vascular endothelial growth factor receptor (VEGFR)-2 at the endothelial junctions is the upstream mechanosensing complex, leading to the ECM-integrin activation. Through this complex, shear stress activates phosphatidylinositol (PI)3-kinase, and the increased level of PI3 lipids then changes the affinity of integrins such as α5β1 and/or αvβ3. Although PECAM, VE-cadherin, and VEGFR-2 can all be activated by shear stress in a rapid manner, the question remains of the selectivity of this complex. Additional issues are whether other shear stress-sensitive kinases can be regulated in a similar manner, and if they can, what is the involved ECM-integrin interaction.
In summary, the article by Hahn et al6 unveils evidence that ECM in the subendothelial space is material-smart and mechanosensitive. The type of ECM (eg, collagen) underneath the endothelium in atheroprotective regions would be inert in terms of eliciting proinflammatory signaling. After chronic exposure to various atherogenic risk factors, the ECM (eg, fibronectin) deposited in vessels is known to become atheroprone, particularly in the bends and bifurcations of aging vessels. Through the engagement of cognate integrins, disturbed flow patterns can exacerbate the inflammatory responses in ECs (Figure 8 in the article by Hahn et al6), causing the dysfunctional endothelium. In addition, the deposited ECM would change the compliance of the vessel wall. An atheroprone environment is thus formed to facilitate the plaque initiation and development.
Sources of Funding
Supported in part by NIH grants HL77448 and HL89940.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
Ricci R, Sumara G, Sumara I, Rozenberg I, Kurrer M, Akhmedov A, Hersberger M, Eriksson U, Eberli FR, Becher B, Borén J, Chen M, Cybulsky MI, Moore KJ, Freeman MW, Wagner EF, Matter CM, Lüscher TF. Requirement of JNK2 for scavenger receptor A-mediated foam cell formation in atherogenesis. Science. 2004; 306: 1558–1561.
Li YS, Shyy JY, Li S, Lee JD, Su B, Karin M, Chien S. The Ras/JNK pathway is involved in the shear-induced gene expression. Mol Cell Biol. 1996; 16: 5947–5954.
Iiyama K, Hajra L, Iiyama M, Li H, DiChiara M, Medoff BD, Cybulsky MI. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ Res. 1999; 85: 199–207.
Hahn C, Orr AW, Sanders JM, Jhaveri KA, Schwartz MA. The subendothelial extracellular matrix modulates JNK activation by flow. Circ Res. 2009; 104: 995–1003.
Jalali S, del Pozo MA, Chen KD, Miao H, Schwartz MA, Shyy JY, Chien S. Integrin-mediated mechanotransduction requires its dynamic interaction with the ECM ligand. Proc Natl Acad Sci U S A. 2001; 98: 1042–1046.
Shyy JY, Chien S. Role of integrins in endothelial mechanosensing of shear stress. Circ Res. 2002; 91: 769–775.