| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,1
Department of Biomedical Engineering and Division of Cardiovascular Medicine, USC School of Engineering and Keck School of Medicine, Los Angeles;
Department of Mechanical and Aerospace Engineering, UCLA School of Engineering and Applied Sciences, Los Angeles;
* Division of Cardiology, Department of Medicine, UCLA School of Medicine, Los Angeles; and
Department of Molecular Pharmacology & Toxicology, USC School of Pharmacy, Los Angeles, California, USA
1Correspondence: University of Southern California, Olin Hall of Engineering 500, Los Angeles, CA 90089-1451, USA. E-mail: hsiai{at}usc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
) at ± 2.6 dyn/cm2 at a time-averaged shear stress (ave)=0 and 0.5 Hz, individual monocytes displayed unique to-and-fro trajectories undergoing rolling, binding, and dissociation with other monocyte, followed by solid adhesion on EC. Our study quantified individual monocyte/EC binding kinetics in terms of displacement and velocity profiles. Oscillatory flow induces up-regulation of adhesion molecules and cytokines to mediate monocyte/EC interactions over a dynamic range of shear stress ± 2.6 dyn/cm2 (P=0.50, n=10).—Hsiai, T. K., Cho, S. K., Wong, P. K., Ing, M., Salazar, A., Sevanian, A., Navab, M., Demer, L. L., Ho, C.-M. Monocyte recruitment to endothelial cells in response to oscillatory shear stress.
Key Words: micro-electromechanical systems (MEMS) • cell tracking velocimetry • shear stress sensors • endothelial cells • monocytes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
, 2)
. Shear stress, the tangential drag force of blood passing along the surface of the endothelium (3)
, imparts profound effects on endothelial cell (EC) function (4
5
6)
. Around arterial bends and branches in which the inflammatory responses prevail, the fluid mechanical environment is distinct from the laminar pulsatile environment present in the long, straight sections of the vessel wall. Oscillatory flow with a time-averaged shear stress of 0, characteristic of reattachment points in the arterial branches, modulates the biological activities of EC (7
8
9)
. Leukocyte rolling, adhesion, and transmigration have been observed in response to inflammatory stimuli at the microvascular levels (10)
. What is warranted is a dynamic, high spatial and temporal resolution view of these events in a more complex physiologic flow environment. Therefore, we have developed specialized micro-electromechanical systems (MEMS) sensors and cell tracking velocimetry to link spatial and temporal resolution of real-time shear stress with the monocyte/EC binding kinetics, simulating reattachment points at arterial bifurcations (11)
.
No direct measurement has been achieved to resolve the spatial and temporal variations, albeit evidence exists for direct correlations of shear stress with distribution of focal atherosclerotic lesions (12)
and the computational fluid dynamic (CFD) determinations of shear stress in the arterial circulation (13
, 14)
. MEMS, which are derived from semiconductor technology, enable the realization of highly miniaturized mechanical devices such as sensors and actuators. We developed and fabricated MEMS shear stress sensors, nearly the size of an elongated EC, based on heat transfer principles. Heat transfer from a resistively heated element of the sensor to the flowing fluid was measured, allowing determinations of shear stress level from a linear relation between V2 (V = voltage) and 1/3 (15)
. By simulating oscillatory flow at the unsteady and moving separation point of arterial bifurcations (16)
, we were able to link real-time shear stress with the trajectory patterns of monocyte/EC binding kinetics. We found that individual cell–cell interactions occurred over a range of shear stress of ± 2.6 dyn/cm2. While both biochemical and biomechanical forces are synergistic in the regulation of cell–cell binding kinetics, the relative levels of these two forces may be dependent on the geometric configuration of arterial circulation and the local flow milieu.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. The change in temperature of the local flow milieu leads to the change in resistivity of the sensors.
The dynamic performance of the sensors is characterized by a three-layer structure (Fig. 1
a): 1) the sensing element or the film layer is the top layer; 2) the silicon nitride diaphragm in the middle is an insulation layer, and 3) the silicon substrate at the bottom serves as heat sink. From Fig. 1a
, the energy balance equation is expressed as (18)
:
 | (1) |
are negligible. Thus, Eq. 1
can be simplified to set power dissipation to equal to the convective heat transfer term (from the heated sensor to the ambient fluid). The convective heat transfer coefficient is expressed as h (u), where the shear velocity u is related to wall shear stress as w = ur2 
, and is the fluid density. The heating power for the sensor is related to the wall shear stress as (17)
:
 | (2) |
 | (3) |
is the temperature coefficient of resistance (TCR). An important feature governing the thermal shear stress sensor operation is the temperature over-heat ratio, T, defined as the relative change of sensor temperature compared with the ambient temperature:
 | (4) |
R, which reflects the relative change of sensor resistance compared with the resistance at the ambient temperature,
 | (5) |
|
Given p =V2/R, where V is the voltage across the sensor, Eq. 4
becomes
 | (6) |
, Eq. 3
) is dependent on the concentration of boron doped into the polysilicon strip. The measured TCR (temperature coefficient of resistivity) of the sensor was 
0.1%/°C for high doping level at 1016/cm2 and 60 keV of energy (18)
. In a fully developed channel flow, the flow shear stress determines the rate of heat transfer from a heated resistive element to the surrounding fluid field. The heating power is proportional to convective heat transfers. Thus, a linear relation between V2 and w1/3 is established (Fig. 1b
).
MEMS shear stress sensor design
The schematic diagram of the sensor (Fig. 2
) features the polysilicon as the heating and sensing element at the center of a cavity diaphragm. The polysilicon, measuring 2 µm wide, 0.5 µm high, and 80 µm long, was uniformly doped with boron to a low sheet resistance value of 50 
/cm2 to a typical resistance of 1.25 5 k at room temperature. After doping, the wafer was annealed at 1000°C to activate the dopant and to reduce the intrinsic stress on the polysilicon. The resistance of the sensing element could be adjusted by changing the doping levels. The optimal resistance for the MEMS shear stress sensor ranged from 1 to 10 k, higher than that of the traditional metal sensor (5 to 50 ). This unique resistance level gave rise to a high-frequency response.
|
An array of resistors was fabricated on a single chip while each individual resistor was lying across the diaphragm (Fig. 2f
). The diaphragm was separated from the bottom of the cavity by an 2 µm gap while the pressure inside the cavity was estimated at 300 torr. Thermal evaporation of the aluminum metallization (300 nm in height and 10 µm in width) forms the leads connecting the polysilicon resistor or the sensing element to the external electronics. The novel feature of the vacuum cavity was to generate effective thermal isolation between the heated element and the substrate, thereby minimizing the heat conduction from the diaphragm to the substrate through the gap (18)
.
MEMS sensor fabrication
The process flow involved both surface and bulk micromachining. Silicon nitride 100 nm in thickness was first deposited by low-pressure chemical vapor deposition (LPCVD) on a 4 inch wafer (Fig. 2a
). LPCVD of a 1.4 µm sacrificial polysilicon layer was deposited and patterned (Fig. 2b
), followed by thermal oxidation of the sacrificial layer at 800 nm (the first layer of diaphragm) (Fig. 2c
). Deposition of 350 nm silicon nitride as the second diaphragm material was established by LPCVD (Fig. 2d
). Next, etching holes were opened by removing two layers: silicon nitride by reactive ion etching (RIE) and the thermal oxide by buffered oxide etchant (BOE) (Fig. 2e
). The sacrificial polysilicon was then removed by tetramethyl ammonium hydroxide (TMAH) at 80°C for 3 h. An oxide layer (1.2 µm) and silicon-nitride layer (300 nm) were deposited at 300 mTorr (0.04 Pa) to seal the cavity under vacuum, followed by patterning using both RIE and BOE (Fig. 2f
).
Doping the polysilicon was done by ion implantation with boron at a total dose of 1 x 1016/cm2 (Fig. 2g
). The wafer was then annealed at 1050°C for 1 h to activate the boron ions and to reduce the intrinsic stress of the polysilicon. With sputtering, aluminum metallization formed the leads (Fig. 2h
). A 1.5 µm layer of LPCVD oxide was deposited for passivation of the polysilicon sensing element to prevent resistance drift from spontaneous oxidation. Since the cavity was held under vacuum, the diaphragm was bent down by the external atmospheric pressure, giving rise to an optical interference patterns referred as Newton rings in Fig. 2
.
MEMS sensor calibration
Calibration was performed to establish a linear relation between V2 and w1/3. In the case of an unsteady state, we generated four individual pulsatile flows with different mean flow rates, Qn. The averaged voltage output signals (Vn) from the MEMS sensor corresponded to the individual mean flow rates at 37± 0.15°C (data not shown). The shear stress values corresponding to these four individual flow rates were calculated using Eq. 7
, and a linear relation between V2 and 1/3 was established (data not shown).
 | (7) |
Integrating MEMS sensors with a pulsatile flow channel
A novel pulsatile flow system delivered well-defined flow profiles simulating the flow conditions in the arterial circulation (19)
(Fig. 3
). This unique configuration ensured velocity uniformity and absence of flow separation across the width of the channel during flow reversal. Due to the symmetry of the rectangular flow channel, we were able to flush-mount the sensor opposite to the EC monolayers, which were seeded on the bottom parallel plate. This approach circumvents the local flow disturbance introduced by conventional probes.
|
Validation of real-time shear stress
Theoretical shear stress was obtained by using the Bernoulli’s theorem with an orifice located upstream from the flow channel. The pressure signals were measured with a piezoelectric sensor at 10 Hz (SenSym Model 143SC03D). The signals were phase-averaged, then converted to the pulsatile flows. The experimental measurements by the MEMS sensors were validated against the theoretical formulation, which can be accessed at http://atvb.ahajournals.org (20)
.
Monocyte tracking velocimetry
We developed a cell tracking velocimetry algorithm (21)
to reconstruct the trajectories of monocyte/EC interactions in response to oscillatory shear stress. The time-dependent positions of monocyte/monocyte and monocyte/EC binding kinetics were recorded to videotape by a CCD camera and digitalized with a video capture card (Truevision TARGA 1000). The digital images were then analyzed by the velocimetry algorithms to capture precise monocyte locomotion in relation to the EC monolayers. The algorithm initiates the identification of monocytes by dividing the fields of interest into interrogation areas that are comparable to the size of individual monocytes. The fields of interest were sampled such that 50% the interrogation areas were overlapped to satisfy the Nyquist sampling criteria (22)
. Both the presence of monocytes and the center position of the cells in the interrogation area were determined by the convolution of the interrogation area with a typical monocyte image (Fig. 4
a). The displacement of the monocytes was extracted by 2-dimensional cross-correlation of the image area with the corresponding area in the next digital image. The peaks of the 2-dimensional cross-correlation signal provide displacement information of monocytes between different digital images (Fig. 4b
).
|
Endothelial cell culture
Bovine aortic endothelial cells (BAEC) between passages 5 and 9 were seeded on Cell-Tak cell adhesive (Becton Dickson Labware, Bedford, MA, USA) and Vitrogen (Cohesion, Palo Alto, CA, USA; RC 0701) -coated glass slides (1 cmx5 cm) at 3 x 106 cells per slide. BAEC were then grown to confluent monolayers in DMEM (Dulbecco’s modified Eagle’s medium) supplemented with 20% fetal bovine serum, and 0.05% amphotericin B, and 100 U/mL streptomycin for 48 h in 5% CO2 at 37°C.
Experimental protocols
Glass slides containing confluent BAEC were placed in the flow channel and exposed to one of the two flow conditions at 1 Hz for 4 h: 1) pulsatile flow at a shear stress slew rate (
/t) of 293 dyne/(cm2s), with a time-averaged shear stress (ave) of 50 dynes/cm2 (Fig. 5
a),2) oscillating flow (±2.6 dynes/cm2 at 0.5 Hz) with ave = 0 dynes/cm2 at 0.5 Hz (Fig. 5b
). For oscillating flow, minimal forward flow at a mean shear stress of 0.2 dynes/cm2 was provided every hour to deliver nutrients and remove waste products from the cells.
|
Real-time monocyte/endothelial cell interactions in response to oscillatory flow
Monocytes were isolated using a modification of the Recalde method as described previously (23)
from normal volunteers with institutional review board approval. Freshly isolated monocytes (105 monocytes/cm3) were introduced into the testing channel under oscillatory flow with parameters as described in Experimental Protocols. The flow channel was mounted on an inverted light microscope for a real-time monitoring of cell–cell interactions. Five percent CO2 was delivered to the circulating culture medium. Monocytes and BAEC interactions in response to flow conditions were captured by phase contrast microscope (Nikon Eclipse TE 200), and transmitted by a CCD camera (SONY CCD-IRIS/RGB, Model DXC 151A) to a TV monitor. Imaging data collection spanned 12 h to visualize different stages of cell–cell interactions and tight association of cell–cell interaction throughout the observation period. Real-time shear stress at which different stages of cell–cell interactions occurred was linked with monocyte-monocyte binding and separation, monocyte tethering, and adhesion with the EC monolayers. At a given shear stress, the velocity of the flow field, which was calculated from Eq. 7
, was compared with the individual monocyte velocities.
Monocyte adhesion assay
After 4 h of flow pulsatile vs. oscillatory flow conditions as described in Experimental Protocols, BAEC monolayers on the cover slides were rinsed with DMEM and assayed for monocyte binding activities. Endothelial monolayers were incubated with freshly isolated monocytes (105 monocytes/cm2) for 10 min at 37°C under static conditions. Nonadherent monocytes were washed with DMEM. The adherent monocytes were fixed with 1% glutaraldehyde. Adherent monocytes were counted in a total of 20 high-power (400x) standardized fields.
Quantitative real-time RT-PCR
After BAEC were exposed to the flow conditions, total RNA was isolated using RNeasy kit (Qiagen, Chatsworth, CA, USA). Real-time RT-PCR was performed according to the recommendations of PE Biosystems TaqMan PCR Core Reagent Kit (24)
. Equal amounts of RNA at 0.5 µg/µL were reverse-transcribed to bring the mixed solution to a final concentration of 1x TaqMan buffer, 5 mmol/L MgCl2, 200 µmol/L dATP/dCTP/dGTP, 400 µmol/L dUTP, 100 nmol/L probe, 400 nmol/L primers, 0.01 U/µL AmpErase, and 0.025 U/µL AmpliTag Cold DNA polymerase. Total RNA at 0.4 ng/µL in 5 µL was then transferred to the 96-well plate. PCR was performed at 50°C for 2 min and at 95°C for 10 min, then run for 40 cycles at 95°C for 15 s and 60°C for 1 min on the real-time RT-PCR Engine (MJ Research Opticon®, Watertown, MA, U SA). Ct, which is the threshold cycle number at which the initial amplification becomes detectable by fluorescence. 
Rn, normalizes fluorescence. TaqMan probes (25)
were used for added specificity and sensitivity. Assuming amplification is 100% efficient, we used the difference in Ct values for various flow conditions vs. control to mathematically determine the relative difference in the level of P-selectin, intercellular adhesion molecule (ICAM-1), and MCP-1 mRNA expression (26)
. The primers (Table 1
) were produced on an automated synthesizer (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s protocol. For each gene, quantitative RT-PCR was conducted in duplicate. To ensure the quality of the measurements, negative and positive controls were systematically included in duplicate in each plate. The statistical analysis of the quantitative RT-PCR results was done using the Ct value (Ctgene of interest-CtGAPDH). Relative gene expression was obtained by Ct methods (Ctsample-CtGAPDH) using the control group as a calibrator for comparison of every unknown sample gene expression level. The conversion between Ct and relative gene expression levels is fold induction = 2-Ct (26)
.
|
Statistics
The values of shear stress at which monocyte-monocyte binding and separation, monocyte-endothelial cell tethering, adhesion and detachment were expressed as mean values ± standard deviation. A nonparametric (Wilcoxon rank sum test) comparison of the groups (means of five cell–cell binding events) was applied. Pvalue, which defined the 95% confident interval, was considered statistically significant at < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
). The oscillatory pressure profile at an arbitrary periodicity of 0.5 Hz simulated the flow pattern seen at the reattachment points (Fig. 5b
). This profile reflected the absolute pressure signals fluctuating between +2.6 and -2.6 dyn/cm2 with ave = 0. The time-averaged shear stress (ave) was 50 dyn/cm2. The flow reversal was recorded as upward deflection by the sensor. We validated the real-time shear stress measurement using the Bernoulli’s Principle with an orifice (19)
.
Nonlinear displacement of monocyte locomotion in response to oscillatory flow
The paths followed by monocyte locomotion were traced by cell tracking velocimetry, revealing diverse modes of cell–cell interactions in response to oscillating flow. Image sequences show the meandering paths of a monocyte at fluctuating velocity while undergoing to-and-fro rolling and firm attachment to the EC (Fig. 6
a). Under oscillatory shear stress, the irregularities of monocyte trajectory (Fig. 6b
) may be due to the undulating surface topology of the EC monolayers and to the time variation in integrin–ligand bond formation and breakage. Two monocytes undergoing attachment, separation, short-lived reattachment, and reseparation with one another were captured (Fig. 7
a–e). Both monocytes briefly established anchorage on EC (Fig. 7b
). Monocyte 2 then separated from monocyte 1, which eventually established anchorage on EC as illustrated at 0 reference point on the y-axis (Fig. 7d
). The corresponding trajectories of these two individual monocytes were captured using cell tracking velocimetry (Fig. 7f
). Monocyte 1 was observed to undergo tethering, characterized as abrupt halt in locomotion alternating with resumption of to-and-for motion. The trajectory of the nonviable monocytes that were buoyant above the EC monolayer contrasted with those of monocytes 1 and 2 (Fig. 7f
). In the presence of oscillatory shear stress, the absolute displacement, velocities, and direction of monocytes were nonlinear, and nonrandom, implicating the dynamics of molecular interactions underlying the cell–cell binding kinetics. The rolling velocities of individual monocytes were distinct from the hydrodynamic velocity of the flow field near the wall of the channel and that of the nonviable monocyte (Fig. 7g
). Unlike the velocity profile of nonviable monocyte, individual monocyte motion was retarded when they are in contact with the EC monolayers. Individual monocytes were observed to display binding and separation with other monocytes, transient tethers and solid attachment to the EC monolayers over a dynamic range of wall shear stress (±2.6 dyn/cm2 at 0.5 Hz; Fig. 8
). The mean shear stress values corresponding to the individual events were statically insignificant (n=10 events, P=0.50).
|
|
|
P-selectin and ICAM-1 mRNA expression in the oscillatory flow milieu
Expression of adhesion molecules and the counter receptors mediate the cell–cell interactions for the initial capturing of monocytes from the flow field. We demonstrated that P-selectin mRNA expression was up-regulated by 3.4-fold (5.1±0.3; 1.5±0.1, P<0.05, n=5) while ICAM-1 was up-regulated by 10.5-fold in response to oscillatory vs. pulsatile shear stress (26.5±3.3; 2.5±0.7, P<0.05, n=5) (Fig. 9
) However, the nonviable monocyte, which failed to harbor counter receptors, displayed trajectory and velocity profiles similar to the direction and periodicity of those of the flow field. Therefore, the relative levels of biochemical and biomechanical forces in the regulation of cell–cell binding kinetics may be determined by the arterial geometry in which pulsatile vs. oscillatory flows develop.
|
Effects of pulsatile vs. oscillating shear stress on monocyte/EC binding and MCP-1 mRNA expression
Oscillatory flow, which is representative of expected flow at a typical reattachment point in the arterial bifurcation, induced a significant increase in the distribution and probability of monocytes binding to EC (static control=5±2 monocytes/HPF, oscillatory flow=36 ± 4, P<0.05) (Fig. 10
b). In contrast, unidirectional pulsatile flow, which is representative of expected flow at a typical long straight portion of artery, significantly attenuated monocyte binding to the EC (4±1 monocytes/HPF, P<0.05) (Fig. 10a
). Oscillatory shear stress up-regulated of MCP-1 mRNA expression by 4.8-fold compared with pulsatile flow conditions (8.47±0.63; 1.75±0.63; P<0.05, n=5; Fig. 10d
). Thus, the relative MCP-1 mRNA expression correlated with the extent of monocyte bound (Fig. 10d
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
ave=0) affected the duration of transient tethering and subsequent attachment. The increase in residence time for the individual monocyte under oscillatory flow is conducive to receptors binding to sufficient ligand over the duration of intercellular contact against shear stress induced detachment. We demonstrated the unique behavior of cell–cell interactions in which a sequence of bond formation breakage events occurred causing the monocytes to undergo to-and-fro rolling, transient tethering before firm reattachment on the undulating EC surface.
Shear environment in the vasculature provides the dynamics and molecular constituents necessary to mediate leukocyte-EC adhesion. Not only the magnitude, but also the temporal component of shear stress, is critical in regulating the adhesive interactions. The binding kinetics, which was mediated by adhesion molecules such as selectin and integrin, was reported to range from 4 to 7 dyn/cm2 (27)
. Shear stress has been implicated in modulating the kinetics and receptor specificity of polymorphonuclear cell (PMN) tumor cell interactions (28)
. Frequency of tethering formation in shear flow of neutrophiles to P-selectin was demonstrated to be related to shear rate (s–1) rather than to shear stress (dyne/cm2) by varying viscosity (29)
. Furthermore, the kinetics of PMN transmigration through endothelial cell junctions was significantly faster under shear flow than that of PMNs placed under static conditions (30)
.
We demonstrated a dynamic range of oscillating shear stress over which monocyte/EC binding kinetics occurred (Fig. 8)
. The nature of oscillatory flow, specifically, low mean shear stress and high shear stress gradient, up-regulated the expression of P-selectin, ICAM-1, and MCP-1mRNA in parallel with increasing number of monocytes bound to EC. In vivo studies have demonstrated the presence of redundant mechanisms in leukocyte recruitment. L and P-selectins appear to be primarily responsible for the initial capture of cells from the flowing blood, E and P-selectins for mediating rolling, and ß2 integrins, ICAM-1, and PECAM-1 for adhesion and transmigration (31
, 32)
. The product of arachidonate and complement cascades and chemokines, a large family of small peptides that activate leukocytes via binding to G-protein-coupled receptors, also seem to play fundamental roles in the recruitment process (33
, 34)
. Ramos et al. (1)
demonstrated that P-selectin- and VCMA-1-dependent mononuclear cell rolling in ApoE-/- mice was significantly inhibited by mononuclear antibody blockade of P-selectin or its leukocyte ligand P-selectin glycoprotein ligand 1. They also observed that rolling velocities increased after monoclonal antibody blockade of mononuclear cell 4-integrin or VCAM-1.
The characteristic of oscillatory shear stress has been linked to elevated the oxidative stress in EC by increasing NADH oxidase activity, reactive oxygen species production, and heme oxygenase-1 mRNA expression (35
, 36)
. Therefore, oscillatory flow induces a focal flow milieu in which the expression of cytokines and the activation of receptors/ligands appear be dominant over the dynamic range of shear stress applied in regulating cell–cell interactions.
The present model cannot perfectly replicate the in vivo biochemical milieu at the arterial bifurcation. However, the dynamic visualization displays the nonlinear locomotion of monocytes mediated by the resident time, topology of EC monolayers, integrin-ligand engagement, and expression of cytokines. The constant formation and retraction of filopodial processes may be responsible for the dynamics changes in monocyte morphology, motility patterns, and tethering (37)
. High-magnification images with high depth of focus (z-axis increments) will be required to allow resolution of fine subcellular structures. The use of vitronectin-coated glass rather than more physiologic substrates may affect cellular responses, since the normal artery wall is compliant and has a more complex extracellular matrix. However, since EC produce their own basement membrane in culture, the substrate in this model is reasonably physiologic. The lack of erythrocytes in the medium may also influence the leukocyte–endothelial interactions. Rolling red cells may collide with, enhance binding of, or displace the leukocytes interacting with EC 38
(38, 39)
. These effects are likely to be small and of second order. Overall, this modeling provides previously unachievable quantitative analysis. MEMS technology and cell tracking velocimetry demonstrate monocyte /EC focal contact dynamics and variation in binding kinetics in real-time under realistic shear stress fields. These processes are fundamental to the initiation of inflammatory responses.
The microshear stress sensor offers three particularly novel design characteristics: 1) the heat insulation features a free-standing diaphragm on a vacuum cavity, which minimizes heat loss to the substrate; 2) the semiconductor material serves as a high-performance heating/sensing element, providing a much higher resistivity than the metal used in conventional thermal sensors; and 3) the high sensitivity of sensing element allows real-time (frequency >15 kHz) shear stress measurements and a fine spatial resolution comparable to the length of an elongated endothelial cell (<100 µm).
The shear stress sensor is most sensitive when the direction of shear is perpendicular to the width dimension. When the shear stress is at an angle to the longitudinal direction of the sensing element, the sensitivity of the sensor will be proportional to the sine of that angle. The measured sensitivity was 52 uV(ac)/Pa in a gaseous medium and 13.7 uV/V-kPa in a liquid medium (15)
. The length of the resistive element determines the thermal conduction to the substrate. Longer elements transfer more heat to the substrate but shorter elements confer less sensitivity.
Under oscillatory flow, the shear stress corresponding to flow reversal was recorded in upward deflection from a single sensor. The thermal shear stress operated on heat transfer principle. To better estimate the directionality of shear stress, cross-correlation using a pair of sensors as time dependent events will be indicated.
The optimum resistance range of the micromachined sensor is from 1 to 10 k, much higher than that of the traditional sensor (5 to 50 ). Previously, Liu et al. reported the measured time constant to be 72 µs for the MEMS sensors (15)
. According to the approximate relation between the time constant tc and cut-off frequency fc at constant temperature, fc =1/(1.5 tc), the cut-off frequency is estimated as 9 kHz. This property of the MEMS shear stress sensors contributes to their achieving high-frequency responses.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 4, 2002. Accepted for publication September 1, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Ai, M. Rouhanizadeh, J. C. Wu, W. Takabe, H. Yu, M. Alavi, R. Li, Y. Chu, J. Miller, D. D. Heistad, et al. Shear stress influences spatial variations in vascular Mn-SOD expression: implication for LDL nitration Am J Physiol Cell Physiol, June 1, 2008; 294(6): C1576 - C1585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Newcomer, C. L. Sauder, N. T. Kuipers, M. H. Laughlin, and C. A. Ray Effects of posture on shear rates in human brachial and superficial femoral arteries Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1833 - H1839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Laughlin, S. C. Newcomer, and S. B. Bender Importance of hemodynamic forces as signals for exercise-induced changes in endothelial cell phenotype J Appl Physiol, March 1, 2008; 104(3): 588 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Himburg, S. E. Dowd, and M. H. Friedman Frequency-dependent response of the vascular endothelium to pulsatile shear stress Am J Physiol Heart Circ Physiol, July 1, 2007; 293(1): H645 - H653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. S. Chatzizisis, A. U. Coskun, M. Jonas, E. R. Edelman, C. L. Feldman, and P. H. Stone Role of Endothelial Shear Stress in the Natural History of Coronary Atherosclerosis and Vascular Remodeling: Molecular, Cellular, and Vascular Behavior J. Am. Coll. Cardiol., June 26, 2007; 49(25): 2379 - 2393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chien Mechanotransduction and endothelial cell homeostasis: the wisdom of the cell Am J Physiol Heart Circ Physiol, March 1, 2007; 292(3): H1209 - H1224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. V. Pereverzev and O. V. Prezhdo Dissociation of Biological Catch-Bond by Periodic Perturbation Biophys. J., July 15, 2006; 91(2): L19 - L21. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hwang, M. H. Ing, A. Salazar, B. Lassegue, K. Griendling, M. Navab, A. Sevanian, and T. K. Hsiai Pulsatile Versus Oscillatory Shear Stress Regulates NADPH Oxidase Subunit Expression: Implication for Native LDL Oxidation Circ. Res., December 12, 2003; 93(12): 1225 - 1232. [Abstract] [Full Text] [PDF]
|
|
|
|
