This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Vascular endothelium is approximately 0.2-0.3 µm thick. The neighbouring cells touch against each other tightly to form a dynamic monolayer between the bloodstream and arterial wall.1 Virchow first noted 150 years ago who viewed that endothelial cells (EC) changed shape to different patterns of flow.2 Laminar and rapid blood flow where there is high shear stress the endothelial cells are polygonal-shaped aligned in direction of flow but where blood flow is turbulent or slow such as around curvatures and bifurcations the shear stress is low, the endothelial cells had lost orientation and visually look rounder.3 This fact illustrates that cells respond to shear stress by sensing and transmitting signal inside the cell where an appropriate response is formulated. However endothelial cells not only change morphology in response to shear stress they are also involved in many other important signalling pathways. This shows that the mechanisms responsible for transmission and transduction process are important in understanding endothelial dysfunction which will help to understand many cardiovascular pathologies. Over the last 2 decades research with help of new techniques the mechanotransduction mechanism has been deeply studied in vivo and in vitro. I aim to review these mechanisms.
Overview of Hemodynamic forces acting on the endothelium
Before we continue to understand how mechanotransduction works, we have to look at the mechanical forces which endothelial cells undergo. There are primarily two forces acting on the endothelium; compressive forces acting perpendicular to blood flow as a result of blood pressure, this is primarily borne on vascular smooth muscle cells and tangential frictional force (shear stress) acting parallel to the wall on the surface of endothelium.2,4,5
Blood flow varies through elastic and muscular, small and large arteries, arterioles and pre-capillary vessels. This produces range of stresses ranging in size, frequency and magnitude.2, 4
In large arteries, mean wall shear stress range is between 10-40 dyncm-2 where vessels are straight and away from bifurcations or joining of vessels furthermore with respect to different junctures of the cardiac cycle it produces different shear stress gradients. However these regions have high relatively constant shear stress with respect to temporal and spatial gradients.2, 4, and 5
At curvatures, bifurcations or where vessels unite steady laminar flow changes to complex flow such as vortices or recirculation that may appear or disappear, shorten or lengthen with cardiac cycle.4 These regions where mean wall shear stress is low are sites atherosclerotic deposition. 4
Spatial shear stress is the difference of shear stress between two close points at the same point in time whereas temporal shear stress is the change in flow so higher the temporal gradient the gradient the change in flow.5 Spatial gradients are generated within recirculation zones, and given that blood flow is pulsatile temporal gradients are normally generated however more so in such regions.5
In addition to these flow characteristics, perpendicular to principle flow, secondary flow characteristics may also modify primary flow hence influencing shear stress on the wall. In vortice regions shear stress ranges from negative values to zero to 40-50dyncm-2. During high cardiac output or high blood pressure it increases well above 100 dyncm-2. 4
The complex flow profiles of vortices should not be confused with turbulent flow. Turbulent flow implies random movement of blood constituents in flow; it will also be unsteady so there is likely to be range of shear stress endothelium experiences. Given that turbulent flow accounts for a small fraction of the total vasculature it is unclear if it plays a role in atherosclerotic plaque development. 5
As discussed earlier that shear stress results in morphological change in endothelial cell which seems to reduce the spatial stress gradient, high shear stress also seems to be involved in survival of endothelial cell and releasing atheroprotective factors such as inhibit coagulation, leucocyte migration and smooth muscle cell proliferation as this disrupts the monolayer integrity. Investigators have shown it is required for regeneration of injured cell.
Shear stress distribution across the endothelium?
Shear stress distribution is not only dependant on the force acting on the cell, it is also depends on the geometry of cell, which as discussed changes. Satcher et al used computational method which with refinement by using Atomic Force Microscopy.4,6 It showed a 3D view of endothelial cells which gave a smooth streamlined cell surface and revealed heterogeneities in surface shear stress among different cells. This may help explain the variety in responses within different endothelial cells.4,6
This takes into account regarding the location of mechanosensors responsible for transducing into biochemical signals. Davies et al proposed two mechanisms, centralised and decentralised model.4
Fig 1. Adapted from Davies et al 4
A - Centralised Mechanotransduction
B - Decentralised Mechanotransduction Centralised model (A) favours mechanotransducers being local to the stimuli shear stress so likely to be at the surface where the mechanical sensor likely to have a specialised structure is physically displaced after which electrophysiological pathways control responses.4,8
Decentralized model (B) where shear stress transduction occurs at multiple sites mechanically linked with cytoskeletons that are distant from the site of stimulus. This takes into account for all the mechanotransducers distant from the cell surface.4,8
Fig 2. Adapted from Ando et al 3
Showing the possible routes taken which results in response to change in flow
The mechanisms that recognise shear stress have not been completely identified but possible candidates have been identified.
Ion channels: Several ion channels sensitive to shear stress were found more than 10 years ago, among them are shear stress K+ channel, Cl- and Ca2+ P2X4 channel and a stretch activated channel. 2,3
First reported by Olesen et al is one of the most rapid responses to date; where flow was imposed it activated inward rectifying K+ channels using whole-cell recordings. This led to cell membrane hyperpolarisation initiated rapidly at 0.1 dyncm-2, attained half maximal activation at 0.7dyncm-2 and activity reached plateau at 20dyncm-2. The current did not desensitize quickly and inactivated when flow was stopped. This was also only specific to endothelial cells.4, 6
Cooke et al and Ohno et al demonstrated blocking of inward K+ channels using extracellular barium, cesium and tetraethyl ammonium which stopped hyperpolarisation and led to depolarisation, inhibition NO release and expression of eNOS and TGF-β showing that these channels modulate responses to flow. 2,4,6
Flow-induced depolarisation was inhibited by blocking Cl- channels suggesting further involvement of shear-sensitive chloride channels; this did not change the cell to hyperpolarized state.2, 6
Some Ca2+ channels are also shown to be shear stress responsive such as P2X4 purinoreceptor and transient receptor both of which mediate calcium influx to the cell which subsequently triggers downstream signalling pathways. The P2X4 gene knockout (KO) mice study revealed that failed to exhibit normal responses to shear stress, these mice showed weak NO induced vasodilatation and also had high blood pressure this resembled to eNOS KO mice suggesting P2X4 plays a special role in blood flow dependent vasodilatation and vascular remodelling. 3
Furthermore, a different vector of forces has also been shown to activate ion channels. Lansmen et al observed during suction of 1-20mmHg pressure using micropipette on the endothelial cell membrane to make a tight seal. The stretch-sensitive transmembrane cation channels activated and resulted in calcium influx in turn caused depolarisation of cell. However little is known about mechanisms by which hemodynamic forces may control these stretch-activated ion channels. It appears unlikely that lipid bilayer transfers force to ion channel instead cytoskeleton which may be stretched might activate the channel. 4
Recently Na+ channels SCN4 and SCN89 were also identified that mediate the activation of ERK1/2 in response to shear stress. 2
Tyrosine Kinase Receptors: Vascular Endothelial Growth Factor Receptor 2 (VEGFR2) and Tie-2 (angiopoietin receptor) are also thought to be shear-sensitive and ligand-independent; that is activation can occur in absence of VEGF or angiopoietin. Therefore may have a dual role as biochemical and biomechanical responses. 2,3
When endothelial cells are exposed to flow VEGFR2 either spatially distribute or change shape which forms clusters to bind to Shc (an Src adaptor protein) phosphorylating the receptor and activates various protein kinases ERK, JNK PI3-kinase and Akt which results in eNOS activation and inhibition of apoptosis. 2,3,7
Flk-1 a tyrosine kinase receptor is also activated by shear-stress that leads to activation of Shc-dependent pathway which subsequently activates Ras in turn activating ERK and JNK. 7
Fig 3. Adapted from Chen et al 7
Proposed mechanism in response to shear stress by Tyrosine Kinase Receptors (RTK) and Integrins in ECs.
G-Proteins itself or G-proteins coupled with Receptors(GPCR): These have been hypothesized to have a role in shear stress sensing. In sensory transduction systems, seven transmembrane domain G-protein linked receptors (serpentine receptors) have been commonly involved.4, 8 On EC's they are well distributed across the whole cell surface.4, 8 When the G-protein receptor is activated, it transmits a signal via its cytoplasmic COOH- terminal of the receptor to interact with G-protein resulting in guanosine diphosphate analogue disassociation from the α subunit. That subunit binds to GTP causing conformational change resulting in dissociation of βγ and α subunit. Resulting in activation of various second messengers.4, 8 Through real time molecular imaging it was demonstrated that when shear stress was applied on EC's Bradykinin B2 GPCR changed shape that led to activation of receptors.3 Also Jo et al showed EC's treated with pertussis toxin prevented shear stress mediated ERK1/2 activation. Gudi et al also demonstrated that G-proteins inserted into liposomes without receptors showed an increase activity in response to shear stress. This activity could be modulated if more cholesterol was added to lipid bilayer making it rigid and decreasing response. This suggests that caveolae and G-proteins may act as couple mechanosensors. 2
Further evidence of G-proteins and GPCRs are involved is its consequence of PLC activation common to several flow-initiated endothelial response. The regulation for PDGF gene expression by shear stress via PKC dependent mechanism and also phospholipase A2 activity regulated with arachidonic acid release linked to GPCRs, resulting in prostacyclin release by shear stress. These finding suggest a definite role in mechanotransduction.4
Fig 3. Adapted from Frank et al 9
Shows the difference in wall thickness and lumen diameter Caveolae are 50-100nm invaginations in the plasma membrane characterised by its high cholesterol content and maker caveolin-1. 2,9Caveolae are shown to be important in several signalling pathways and endothelial permeability.3,8,9 These are increased in ECs when exposed to laminar shear stress. 9Caveolae and Cav-1 move towards the luminal edge of ECs where they are linked with Ca2+ wave propagation.3,9 The Ca2+ increase occurring in caveolae results in eNOS to be released which is catalysed in the cytoplasm to NO.3,9 Yu et al demonstrated Caveolae and Cav-1 role by using Cav-1 Knockout mice which showed overexpression of Cav-1 in Cav-1 KO mice. They then ligated the left carotid artery for 14 days which in control showed reduced lumen diameter but not in Cav-1KO mice where wall thickness increased showing impaired vessel remodelling. Also acetylcholine-mediated arteriodilatation was also increased excessively in Cav-1 KO suggesting Caveolae and Cav-1 are involved in control of vascular function.9
Fig 4. Adapted from Frank et al 9
The mechanism involving calveolae that leads to release of NO
Fig 4. (Left Image) Adapted from Davies et al 4 Shows how cytoskeleton connects to integrin which connects at Focal Adhesion site.Integrins: are membrane glycoproteins made up of α and β subunits. Each subunit has a large extracellular region, a transmembrane spanning region and a short cytoplasmic region.3,10 The extracellular component interacts with ECM whereas the cytoplasmic domain interacts with signalling molecules such as FAK, Src Family protein kinases (SH2 and SH3), Fyn and cytoskeletal proteins (α-actinin, vinculin, talin, tensin and paxilin).2,3 The integrin-ECM complex aggregate at specific locations are known as focal adhesions. Also using tandem scanning microscopy with image analysis and reconstruction it demonstrated at unidirectional shear stress focal adhesion sites remodelled and became directional which led to progressive alignment of the sites in direction flow.4,6 Wang et al demonstrated shear stress activated integrins by using magnetic microbeads coated with antibodies against integrins, and resulted in cytoskeletal filaments becoming reoriented with a force dependent cell stiffening response occured.4,6,8 This showed integrins as a possible mechanosensor and transmitting signals to the cytoskeleton consistent with tensegrity model. Integrin-mediated signalling, results in phosphorylation of proteins such as FAK that mediate downstream signalling of Ras-ERK, Rho pathways, cell motility and apoptosis. 8, 10
Fig 5. Adapted from Chien et al 10 Shows the mechanism in response to shear stress involving integrin.
PECAM-1: (Platelet endothelial cell adhesion molecule-1) is diffusely spread in solitary cells but localises at endothelial cell-cell adhesion site, where it mediates leukocyte extravasation during inflammation, reorganisation of stress fibres and focal adhesions. When the ECs are exposed to shear stress PECAM-1 has been shown to be tyrosine phosphorylated within 30 seconds resulting in Ras signalling leading to ERK activation. 2,3,5 When magnetic beads coated with PECAM-1 antibodies were applied it mimicked mechanical pulling which lead to PECAM-1 phosphorylation implying it as a mechanosensor.2,5 PECAM-1 may also play role in discriminating between temporal shear stress gradients and steady flow NO related vasodilatation. Mice with PECAM-1 KO showed impaired regulation of arteriolar dilatation in response to sudden changes in fluid flow compared to wild type mice. However shear-independent NO release was similar suggesting that PECAM-1 knockout mice are less sensitive to temporal shear stress gradients.,5
VE-Cadherin: are the major proteins of the adherens junctions that mediate cell-cell adhesion specifically in endothelial cells. Cadherin's cytoplasmic component are linked to cytoskeletal actin via β catenin and plakoglobin. It is thought to form a mechanosensory complex with PECAM-1 and VEGFR2. Deletion of VE-cadherin led to impaired remodelling and maturation of the vascular network and also stopped intracellular signalling via VEGFR2 in result to shear stress. This suggests PECAM-1 transduces forces that activates VEGFR2 which inturn activates P13 kinase mediating integrin activation where VE-Cadherin acts as the adaptor protein. 2, 3
Tensegrity: Cytoskeleton stabilises cells structure and shape by interconnected network of cytoskeletal elements (F-actin stress fibers, intermediate filaments, and microtubules).2,3,4,8 The precise role of 3 components is not fully known in mechanotransduction. As earlier discussed some structures with in the endothelial cell that are not directly exposed to mechanical forces so how do they sense a mechanical force and transduce it? Hence Tensegrity model is proposed where transmission of mechanical forces from one part of cell to another. This would allow mechanotransduction occurring at distant site. When shear stress is applied to this model the cytoskeletal elements undergoes organisation without any cell shape disruption. Cytoskeleton has been shown to play an important role in shear stress induced NO release and ICAM-1 gene expression. Therefore its role is more in providing a transmission of force to intracellular components involved in mechanotransduction.2,3,4,6,8 This will be discussed further later in transmission.
Glycocalyx: a negatively charged membrane-bound macromolecules. Due to its location, it must be considered as a possible mechanosensor of shear stress. Involvement of glycocalyx has been shown by degradation of heparin sulphate, hyaluronic acid and sialic acid that lead to inhibition of shear stress induced NO production. This can be due to two possible mechanisms: Heparan sulphate proteoglycan undergoes morphological change from a coil to a filament structure in flow conditions and is accompanied by increase Na+ ion binding sites which may result in intracellular signalling.3,11 It is also possible that shear stress is transmitted via cytoskeleton to other transduction sites. Shear induced PGI2 production was not inhibited when any of the glycocalyx molecules degraded suggesting that transduction pathway occurs in a distinct location.11
Primary Cilia are rod-like non-motile structures on the luminal surface of ECs (Aortic, Embryonic, Human Umbilical veins) that are connected to cytoskeletal microtubules. Its displacement transmits shear stress to cell. It has been recently shown polycystin-1 and -2 which have a long extracellular domain are involved in shear stress sensing. Knocking out its genes resulted in unable to respond to shear stress by NO related vasodilatation.3
Adaptation and signal filtering
Endothelial cells being extremely close are continuously stimulated by hemodynamic forces. Consequently, there must be mechanisms to cope with overstimulation. The mechanisms likely to be present are adaptation and filtering of signal.4
Adaptation where sustained stimulation results in feedback inhibition of mechanotransduction. Examples include desensitisation of shear sensitive ion channels, adjustment of cell surface adenine nucleotide concentrations, intracellular renormalization of calcium levels, cell alignment under flow and its reversal several hours after cessation of flow, streamlining of cell surface to reduce the shear stress gradients.4
The other process is signal filtering when mechanical stimulus is filtered through structural components that only allow certain range or type of mechanical stress to activate mechanotransduction. E.g. Pacinian corpuscle which only allows high frequency stimulus to be transmitted. Similarly cytoskeleton may also filter hemodynamic stimuli in endothelial cells, frequency of flow mediated Ca2+ ion oscillations may also be a selective filter. Endothelial cell do not respond at certain frequencies of pulsatile flow. This may be because the stimulus needs to be sustained such as pinocytosis rates that increase with steady shear stress were unaffected by 1 Hz oscillations of same mean value of shear stress. 4
Endothelin-1 RNA that downregulates at 15 dyncm-2 failed to change when back and forth flow was applied to the cell. Helminger et al has reported that endothelial cells are very selective such that thy can discriminate different types of flow, those being oscillatory and steady as well as patterns of oscillatory. This group also reported that Ca2+ levels in endothelial cells were strongly dependent on characteristics of flow. 4
Shear stress acting on the luminal surface causes internal stresses within the cell that are transmitted to abluminal sites or neighbouring cells. Furthermore, blood pressure in turn creates compressive forces on these cells. 4
Yet, endothelial cells are in state of tension; it is thought that cytoskeleton cooperate with areas of cell to maintain cell shape. When such forces are loaded on the cells, internal cellular tension changes to equalise the external force. 4
From this it can be concluded that shear stress induced mechanotransduction in endothelial cells may happen by first local displacement of sensors at cell membrane followed by transmission of this force through the cell via cytoskeleton and finally force transduction of transmitted stress at mechanotransduction sites. 4
The prominent reorganisation of F-actin stress fibers and intermediate filaments, and microtubules to external force links it as principle force transmission element in endothelial cells. The F-actin filaments linked with transmembrane integrins seem to be principal method of transmission forces also of interest, is its implication in transduction when the filaments are disrupted it alters a number of primary and secondary responses. There is also change of stretch of activated ion channel activity in response to membrane deformation and endothelial cell shape change, realignment to flow, focal adhesion remodelling and gene regulation when microfilament turnover is inhibited. 4 , 6
Overview about Mass-Agonist Transport
We have thoroughly considered that endothelial mechanotransduction mechanism occurs by physical displacement of cell structure(s) but we have not considered about flow affecting concentration of solutes near the endothelial surface.
The local concentration of a plasma solute in unstirred boundary layer at cell surface is similar to its concentration in the bulk fluid if removal rate is slow. For most ligands, removal is receptor mediated endocytosis which takes several minutes. However when flow rate is high it can alter solute concentrations and hence the agonist-receptor interactions can modulate EC response.4
Some agonists that are rapidly removed by enzyme degradation at cell surface are also influenced by flow. If removal rate exceeds agonist replenishment rate from bulk fluid then there is a chemical gradient. If an agonist is secreted by endothelial cell which can interact with the receptor, the concentration gradient near the surface will also be affected by flow and therefore influencing autocrine interactions. 4
This mechanism has been studied in endothelial cell [Ca2+]i. Using Adenine nucleotides as the agonist and [Ca2+]i mobilisation as the response. Investigators have demonstrated imbalance between mass transport delivery of ATP/ADP to the cell surface (which is increased by flow) and degradation by extonucleotidases. Small changes in the convection rate of ATP from bulk fluid or degradation of ATP at its receptor significantly influenced ATP Boundary layer concentrations. Calculations showed change in concentration within 10 sec when flow rate is changes and is consistent with on-off calcium responses. 4
Vascular endothelial cell is a very sensitive cell that responds to hemodynamic forces by blood flow. The mechanisms are not fully understood but advances have been made which are increasingly making us understand how the endothelial cell responds at all levels. These mechanotransduction mechanisms are likely to be working all together rather than be mutually exclusive. The studies in vivo and in vitro have provided us with detailed analysis of EC responses to shear stress. The responses of EC to change in flow are: ECs lining blood flow that is rapid and unidirectional are spindle shaped and parallel where as where blood is stagnant are rounder in shape.3 When blood flow increases, NO mediated vasodilatation occurs as well as upregulation of its gene, on the other hand, potent vasoconstrictors, endothelin and ACE that generate Angiotensin II decreases in response to shear stress.3 Shear stress enhances antithrombin activity by expressing thrombomodulin, a membrane glycoprotein that inhibits thrombin (Coagulant) and activates protein C (anticoagulant). Shear stress also results in increased production of PDGF, HB-EGF, bFGF, TGF-B, IL-1 and IL-6 and GM-CSF. Shear stress also modulates adhesion of leucocytes.3 Shear stress also influences reactive oxygen species.3 Oscillatory shear stress increases ROS production resulting in atherosclerosis whereas laminar shear stress increases superoxide dismutase (SOD) which inactivates ROS. Shear stress also effects the expression of genes.3