Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation Research
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents
Circulation Research. 2003;93:e136-e142
Published online before print October 16, 2003, doi: 10.1161/01.RES.0000101744.47866.D5
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
93/10/e136    most recent
01.RES.0000101744.47866.D5v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Florian, J. A.
Right arrow Articles by Tarbell, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Florian, J. A.
Right arrow Articles by Tarbell, J. M.
Related Collections
Right arrow Endothelium/vascular type/nitric oxide
(Circulation Research. 2003;93:e136.)
© 2003 American Heart Association, Inc.

UltraRapid Communications

Heparan Sulfate Proteoglycan Is a Mechanosensor on Endothelial Cells

Jeffry A. Florian, Jason R. Kosky, Kristy Ainslie, Zhengyu Pang, Randal O. Dull, John M. Tarbell

From the Biomolecular Transport Dynamics Laboratory (J.A.F., J.R.K., K.A., Z.P., J.M.T.), Departments of Chemical Engineering and Bioengineering, The Pennsylvania State University, University Park, Pa; and the Department of Anesthesiology and Critical Care Medicine (R.O.D.), The Johns Hopkins University School of Medicine, Baltimore, Md.

Correspondence to John M. Tarbell, Department of Biomedical Engineering, Steinman Hall, 2nd Floor, The City College of New York/CUNY, Convent Ave at 140th St, New York, NY 10031. E-mail tarbell{at}ccny.cuny.edu


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMaterials and Methods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
The objective of this study was to test whether a glycosaminoglycan component of the surface glycocalyx layer is a fluid shear stress sensor on endothelial cells (ECs). Because enhanced nitric oxide (NO) production in response to fluid shear stress is a characteristic and physiologically important response of ECs, we evaluated NOx (NO2- and NO3-) production in response to fluid shear stress after enzymatic removal of heparan sulfate, the dominant glycosaminoglycan of the EC glycocalyx, from cultured ECs. The significant NOx production induced by steady shear stress (20 dyne/cm2) was inhibited completely by pretreatment with 15 mU/mL heparinase III (E.C.4.2.2.8) for 2 hours. Oscillatory shear stress (10±15 dyne/cm2) induced an even greater NOx production than steady shear stress that was completely inhibited by pretreatment with heparinase III. Addition of bradykinin (BK) induced significant NOx production that was not inhibited by heparinase pretreatment, demonstrating that the cells were still able to produce abundant NO after heparinase treatment. Fluorescent imaging with a heparan sulfate antibody revealed that heparinase III treatments removed a substantial fraction of the heparan sulfate bound to the surfaces of ECs. In summary, these experiments demonstrate that a heparan sulfate component of the EC glycocalyx participates in mechanosensing that mediates NO production in response to shear stress. The full text of this article is available online at http://www.circresaha.org.


Key Words: shear stress • endothelial cells • heparan sulfate • nitric oxide • glycocalyx


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMaterials and Methods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
The inner surfaces of blood vessels are lined with a monolayer of endothelial cells (ECs) that is continually exposed to the mechanical shearing forces (stresses) of blood flow. Variations in shear stress magnitude as well as temporal and spatial distribution have been shown to induce alterations in endothelial permeability and hydraulic conductivity,1–3 cytoskeletal structure,4–7 surface adhesion molecule expression,8 and gene expression.9,10 In addition, the exposure of endothelial cells to shear (both steady and oscillatory) has been shown to alter the production of vasoregulating agents of which nitric oxide (NO) is perhaps the most notable.11–13

NO is a vasodilator produced by the conversion of L-arginine to L-citrulline that is catalyzed by endothelial nitric oxide synthase (eNOS). NO modulates vascular tone by eliciting relaxation of smooth muscle cells while inhibiting smooth muscle cell growth.14 NO production responds to changes in shear stress in a biphasic manner in human umbilical vein endothelial cells (HUVECs) and bovine aortic endothelial cells (BAECs).11,12 There is an initial rapid NO production phase that is G protein and Ca2+-dependent and is influenced by rate of change of shear and not the shear level per se. The subsequent phase is characterized by a lower rate of NO production rate that is G protein and Ca2+-independent but is shear level dependent.12,15 Both phases of the NO response are mediated by mechanotransduction, or the transmission of the fluid shear force to the cellular surface.16 The hypothesis of this study is that a specific component of the plasma membrane surface layer, termed the glycocalyx, is in fact the mechanosensor for the NO response.

The surface of endothelial cells is decorated with a wide variety of membrane-bound macromolecules, including glycoproteins, proteoglycans, and their associated glycosaminoglycan (GAG) side chains that constitute the glycocalyx. Heparan sulfates are the most common endothelial cell-surface GAG, comprising some 50% to 90% of the GAG pool,17 and they have been the most extensively studied GAG in regard to their ability to function as signal transduction molecules.18 Heparan sulfate proteoglycans (HSPGs) can be divided into three different groups based on their protein backbones (syndecans, glypicans, and perlecans). The least common HSPG, perlecan, consists of only one variant with a total of five GAG linkages and is located within the EC basement membrane. Syndecans form the largest group of HSPGs on the endothelial surface and are distinguished by highly variable extracellular domains19 and by their status as the only HSPG that penetrates the cytoplasm, allowing for interaction with the cytoskeleton.20 Glypicans form the second most common HSPG group, and possess structural similarity, typically differing only in the number of GAG attachment sites. These molecules are bound to the surface through a COOH-terminal GPI (glycosyl phosphatidyl inositol) anchor.21

There are several possible mechanisms by which the glycocalyx might serve as a mechanotransducer. First, the glycocalyx acts as a transport boundary between blood flow and the endothelial surface, and as such, it is privy to differing concentrations of agonists.22–24 Alterations in flow could affect the distribution of agonists in this layer, indirectly transmitting flow conditions to the cell. Alternatively, the membrane-attached proteoglycans may be physically displaced when exposed to shear. This proteoglycan displacement could be transmitted to the cellular surface, provoking an intracellular response.16 To efficiently transmit shear stress from flowing blood by this mechanism, the membrane-bound proteins should possess an extended extracellular domain for shear sensing along with an intracellular, cytoplasmic sequence for direct transmittance.25 Such an arrangement is typical of syndecans on the cellular surface,19 which have an extensive GAG component to sense shear and a cytoplasmic domain that is linked with known signaling elements such as G-protein receptors, including those that form a cytoplasmic bond with eNOS,16,25 and cytoskeletal elements like actin that can transduce physical forces throughout the cell26–28 to affect cell function. However, it has not yet been conclusively demonstrated that there is a dominant apical distribution of syndecans on endothelial cells.

In order to assess the role of the glycocalyx in shear-induced NO production, we employed an engineered shearing device to expose monolayers of BAECs to defined shear stress (steady and oscillatory). Heparinase III was used to selectively deplete the glycocalyx of the heparan sulfate GAG component. The NO production rate in response to shearing conditions was determined for heparinase treated and untreated monolayers, and it was observed that heparinase treatment completely inhibited shear-induced NO production.


*    Materials and Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Materials and Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Chemicals
The following chemicals were obtained from Sigma Chemical Co: bovine serum albumin (BSA, Fraction V, 30% solution), minimal essential media (MEM), fetal bovine serum (FBS), L-glutamine, penicillin-streptomycin, fibronectin, trypsin-EDTA, Dulbecco’s phosphate-buffered saline (DPBS) without Ca2+ and Mg2+, Hank’s balanced salt solution (HBSS), heparinase III from Flavobacterium heparanum, sodium nitrate, sodium nitrite, ß-nicotinamide adenine dinucleotide phosphate (ß-NADPH), N-(1-naphthyl)ethylenediamine (NED), flavin adenine dinucleotide (FAD), sulfanilamide, nitrate reductase, 2,3-diaminonaphthalene (DAN), Bandeiraea simplicifolia biotinylated lectin, bradykinin (BK), and histamine.

Polycarbonate Transwell filters (Costar, 0.4-µm pore size, 24.5-mm diameter) were obtained from Fisher. Heparan sulfate primary antibody (HepSS-1 mouse anti-mouse IgM) was obtained from US Biological. Alexa Fluor 488-labeled secondary antibodies (goat anti-mouse IgM) and NeutriAvidin Alexa-Fluor 350 avidin conjugate were obtained from Molecular Probes.

Cell Culture
Bovine aortic endothelial cells (BAECs) were harvested from bovine thoracic aortas as described by Sill et al.30 Briefly, after removal, the aorta was transported in ice-cold HBSS containing 1% penicillin-streptomyocin. For cell harvesting, the aorta was rinsed twice with HBSS while the fat was separated from the aorta. Subsequently the aorta was split longitudinally between the intercostal arteries and formed into a trough. A solution of 60 mL of complete media (MEM w/ 10% FBS, 1% L-glutamine, and 1% penicillin-streptomyocin) was added to 1 mL of Liberase Blendzyme 2 (Roche Diagnostics Corporation), and subsequently, 10 mL of this solution was added to the luminal surface. This solution was allowed to incubate at room temperature for 40 minutes, then the solution was collected, centrifuged, and the pellet was resuspended and plated on a T-25 flask. Five more washings involving 10 mL of the Blendzyme cocktail were repeated, with incubations of 5 minutes at room temperature. Each time, the solution was pipetted off the aorta, centrifuged, and the resulting pellet was resuspended and plated on a T-25 flask.

Cells on flasks were grown in MEM supplemented with 10% FBS, penicillin (50 U/mL), streptomyocin (50 µg/mL), and L-glutamine (2 mmol/L) at 37°C in 5% CO2. To examine the effect of shear stress on cellular response, BAECs (passage 5 to 9) were seeded onto fibronectin-coated slides (30 µg/slide) at a cell density of 0.25x106 cells/cm2. Before cell seeding, however, the slides were glued to the bottom of a 24.5 mm diameter Transwell filter holder (after removal of the filter) with a silicone elastomer (Sylgard). The cylindrical filter holder could then be used to support the cells when defined shear stress was applied using the rotating disk apparatus described below. These slides mounted on the filter holder were used in experiments 3 to 4 days after cell seeding.

Shear Apparatus
A detailed description of the shear apparatus used in these experiments can be found in Sill et al.2 Briefly, a cylindrical disk (radius=12 mm) was supported above the monolayer at a distance of approximately 251 µm. This disk was rotated by a motor drive to produce a defined shear stress that varies linearly from zero at the center to a maximum value at the edge (mean value is two-thirds the maximum value). All subsequent values of shear stress will be reported as the maximum value. To obtain oscillatory shear stress, we varied the rotational speed of the shear disk by adding a sinusoidal waveform to the steady rotational motion. Steady shear experiments were conducted at 20 dyne/cm2. Oscillatory shear was produced by using a steady shear component of 10 dyne/cm2 and an oscillatory component of 15 dyne/cm2 (equivalent to a shear stress of 10±15 dyne/cm2 at 1 Hz) as described in Hillsley and Tarbell.29

Just before a shear experiment, the monolayer was rinsed twice with the experimental media consisting of phenol red free MEM (PRF-MEM), supplemented with penicillin, streptomyocin, L-glutamine, and 1% BSA, and after washing, 2 mL of experimental media were added to the top of the monolayer that had a typical cell density of 0.25x106 cells/cm2. The slide on the cylindrical filter holder was positioned beneath the shear apparatus and raised into position using an adjustable stage. After shear was initiated, 300 µL samples were taken at 0, 5, 15, 30, 60, 120, and 180 minutes after application of shear and replaced with an equal volume of fresh media. Due to this dilution of media, the measured values of concentration were adjusted by the dilution factor (1.7/2.0) and reported as the undiluted (cumulative) values.

Nitrite/Nitrate Determination
In this procedure, samples were first reduced with nitrate reductase, converting free NO3- to NO2-. Four microliters of the enzymatic mixture, containing 50 µmol/L ß-NADPH and 5 mmol/L FAD, was mixed with 5 µL of KH2PO4 buffer (pH 7.5) and 50 µL of sample and incubated at 37°C for 3 hours. A fluorometric assay with accuracy down to 10 nmol/L was then used.30 This quantification is based on the reaction of nitrite with 2,3-diaminonaphthalene (DAN), which produces the fluorescent compound, 1-(H)-naphthotriazole. After the reduction step, 10 µL of fresh DAN solution (0.05 mg/mL in 0.62 mol/L HCl) was added to each well and refrigerated at 20°C for 10 minutes. Next, 10 µL of 2.8 N NaOH was added to terminate the reaction. Fluorescent readings were obtained using a Packard FluoroCount fluorometer with excitation at 360 nm and emission at 425 nm (corresponding to the fluorescent spectra peaks of 1-(H)-naphthotriazole) and Plate Reader Version 3.0 software. Nitrite standards using NaNO3 in PRF-MEM with 1% BSA were run over the range of 12 nmol/L to 40 µmol/L. To convert measured sample concentrations in molar units (moles/103 mL) to the normalized units used in presenting the results (moles/106 cells), we multiplied by the volume of media on top of the cell monolayer (2 mL) to determine the number of moles present and next divided by the area of the monolayer (4.52 cm2) and finally by cell density (0.25x106 cells/cm2). Using these conversion factors, the concentration at the upper end of the standard curve (40 µmol/L) becomes 70.8 nmol/106 cells.

Heparinase Treatment
Heparinase III (heparinase; E.C.4.2.2.8) is an enzyme selective for heparan sulfate within the glycocalyx. For heparinase treatment, 2 mL of a solution containing 15 mU/mL of heparinase in experimental media were placed on the monolayer and incubated for 2 hours before experimentation. After exposure to heparinase, the monolayer was washed twice in experimental media and then exposed to shear stress as described above. The solution of heparinase was reconstituted from aliquots prepared within 2 months of the experiment, 15 minutes before application.

Protease Activity
Heparinase was assayed for protease activity using a RediPlate 96-EnzCheck Protease Assay kit from Molecular Probes. The protease assay kit is capable of detecting a broad variety of proteases including metallo-, serine, acid and sulfhydryl proteases. The assay was performed in accordance with manufacturer’s instructions. The protease activity of the 15-mU/mL heparinase III was compared against known concentrations of chymotrypsin and pronase at 1, 3, and 24 hours. All samples of chymotrypsin and pronase were run in triplicate.

Bradykinin and Histamine Treatments
Bradykinin (BK) and histamine were utilized to test the viability of the endothelial cells’ NO production machinery after enzymatic treatments. For experiments involving BK, the 2-mL starting volume contained BK at a concentration of 5 nmol/L. Similarly, the media used to replace samples in these experiments also contained BK at a concentration of 5 nmol/L. Related experiments utilized a 2-mL starting volume containing 100 µmol/L histamine with replacement media at the same concentration.

Fluorescent Labeling of Heparan Sulfate and Chondroitin Sulfate
In order to assess the effectiveness of heparinase in removing heparan sulfate from the glycocalyx and leaving chondroitin sulfate unaffected, fluorescent antibody imaging was used. BAECs were grown to confluence on slides (usually 3 to 5 days). For heparan sulfate imaging, slides were incubated with 2 mL of a 15-mU/mL heparinase solution for 2 hours. Control slides were left untreated. The monolayers were then washed with DPBS and a solution of 2 µL HepSS-131 was diluted to 1 mL using DPBS and applied to the monolayers. After a 15-minute incubation period, 2 µL Alexa Fluor 488-labeled secondary antibody was diluted to 1 mL with DPBS and applied to the monolayers for another 15-minute period. For chondroitin sulfate imaging, slides were incubated with 2 mL of a 15-mU/mL heparinase solution for 2 hours. Control slides were left untreated. The monolayers were then washed with DPBS and a solution of 0.005 mg/mL of biotinylated lectin in DPBS was applied. This lectin binds specifically to the N-acetyl-D-galactosamine residue present on chondroitin sulfate. After a 15-minute incubation period, a solution of 0.005 mg/mL fluorescent-labeled avidin conjugate in DPBS was applied to the monolayers for another 15-minute period. Then the monolayers were washed twice with DPBS and imaged using wide field fluorescence at 40x on an Olympus BX60 digital microscope with an exposure of 750 ms. Overall fluorescence intensity was calculated for a series of image fields, using Image Pro Plus. Comparison of the fluorescence intensity of the untreated and treated image fields provided an estimate of the degree of heparan sulfate and chondroitin sulfate removal associated with the heparinase treatment.

Data Presentation and Statistical Analysis
Significant differences between group means were evaluated by repeated measures analysis where cumulative NOx concentration was analyzed by a two-way (time and concentration) ANOVA using Minitab. NOx concentrations were reported as mean±SE. A level of P<0.05 was considered significant for statistical analysis.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 *Results
down arrowDiscussion
 down arrowReferences
 
Steady Shear Experiments
EC monolayers were exposed to 20 dyne/cm2 steady shear stress with and without heparinase pretreatment, and stationary controls with and without heparinase pretreatment were run concurrently (Figure 1). Henceforth the conditions will be abbreviated as follows: stationary control (C), control with heparinase pretreatment (CH), steady shear of 20 dyne/cm2 (SS), and steady shear of 20 dyne/cm2 with heparinase pretreatment (SSH).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 1. Effect of steady shear stress on NOx production after heparinase treatment. NOx production induced by 20 dyne/cm2 steady shear stress (n=7) was significantly greater than control (*P<0.05, n=14), control with heparinase exposure (+P<0.05, n=13), and steady shear with heparinase treatment (#P<0.05, n=7). Treatment with heparinase substantially inhibited overall NOx production. Data are represented as mean±SE.

SS NOx production was biphasic with an elevated production rate in the first 5 minutes that tapered off to a sustained rate by 60 minutes. SS NOx concentration was elevated significantly relative to C NOx concentration (P<0.05 after 30 minutes) and was 3-fold higher at 180 minutes.

The SSH response was dramatically inhibited, reaching a final concentration at 180 minutes that was not significantly different from C (P>0.98) or CH (P>0.35). The SSH response was significantly lower than the SS response after 60 minutes (P<0.05). C and CH concentrations were not significantly different from each other (P>0.95 at all times).

Oscillatory Shear Experiments
EC monolayers were exposed to oscillatory shear stress with a mean component of 10 dyne/cm2 and a sinusoidal component of 15 dyne/cm2 (10±15 dyne/cm2) so that shear stress reversal (shear in the direction opposite the mean shear) would occur over a portion of the cycle (Figure 2). The abbreviations OS and OSH will be used to denote experiments with oscillatory shear and oscillatory shear with heparinase pretreatment, respectively.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. Effect of oscillatory shear stress on NOx production after heparinase treatment. NOx production induced by 10±15 dyne/cm2 oscillatory shear stress (n=7) was significantly elevated above control (*P<0.01, n=14), control with heparinase exposure (+P<0.01, n=13), and oscillatory shear with heparinase exposure (#P<0.01, n=7). Treatment with heparinase greatly inhibited NOx production. Data are presented as mean±SE.

OS experiments were characterized by rapid NOx production in the first 5 minutes with a lower but sustained rate at later times. The final NOx concentration at 180 minutes (9.07±0.60) was higher than in the SS experiments (6.54±0.42) and 4.1-fold higher than in C experiments. OS NOx concentration was significantly higher than C NOx concentration at all times (P<0.01).

Treatment with heparinase (OSH) completely eliminated the NOx response to shear stress (Figure 2). The OSH response was significantly lower than the OS response at all times (P<0.01) and was reduced by 94% at 180 minutes.

Bradykinin and Histamine Experiments
EC monolayers were exposed to 5 nmol/L BK, and significant NOx production was induced over a 3-hour period (Figure 3). The NOx produced after 3 hours of exposure to BK was similar to the NOx produced after 3 hours of exposure to steady or oscillatory shear stress (cf, Figures 1 and 2Up). When ECs were pretreated with heparinase, the NOx produced in response to BK exposure was nearly identical to that of untreated cells (P>0.99 for all times). The BK response was significantly greater than the control (no BK) for both heparinase pretreated (P<0.05) and untreated (P<0.05) monolayers.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Effect of bradykinin on NOx production after heparinase treatment. After application of 5 nmol/L BK, controls (n=5) displayed a significant increase in NOx production in comparison to controls not exposed to BK (*P<0.05, n=14). After heparinase treatment, cells exposed to 5 nmol/L BK (n=5) displayed a significant increase in comparison to heparinase treated cells without BK (#P<0.05, n=7). Controls exposed to BK and heparinase treated cells exposed to BK had the same NOx production (P>0.99). Data are presented as mean±SE.

NOx production in response to histamine (100 µmol/L) without heparinase reached a concentration of 4.50±0.36 (n=3) nmol NOx/million cells after 3 hours of exposure, whereas the 3-hour response to the same dose of histamine after heparinase treatment was 3.95±0.71 (n=3) nmol NOx/million cells. These responses were not significantly different from each other (P>0.23), but they were significantly different from the control response (no histamine or heparinase) of 1.45±0.34 (n=3) nmol NOx/million cells (P<0.002).

Verification of Heparan Sulfate Removal by Heparinase III
Previously, Haldenby et al32 reported that heparinase treatments were capable of removing a substantial fraction of heparan sulfate proteoglycans. To verify the removal associated with our heparinase pretreatment, both treated and untreated EC monolayers were labeled with an antibody specific to heparan sulfate and visualized with a fluorescently labeled secondary antibody before and after pretreatment. Representative images demonstrating this process are displayed in Figure 4. Clearly there was a substantial reduction of heparan sulfate after heparinase treatment that was quantified as a 45.9±13.8% (n=3) reduction in fluorescence intensity relative to the untreated controls. To verify that the heparinase pretreatment did not remove chondroitin sulfate proteoglycans, both treated and untreated EC monolayers were labeled with a biotinylated lectin that binds specifically to the N-acetyl-D-galactosamine residue present on chondroitin sulfate and visualized with a fluorescent labeled avidin conjugate. Representative images demonstrating this process are displayed in Figure 5. There was a negligible reduction of chondroitin sulfate after heparinase treatment that was quantified as a 1.0±1.4% (n=4) reduction in fluorescence intensity.



View larger version (81K):
[in this window]
[in a new window]
  		Figure 4. Images of fluorescent antibody to heparan sulfate without (top) and with (bottom) treatment with 15 mU/mL of heparinase for 2 hours.



View larger version (79K):
[in this window]
[in a new window]
  		Figure 5. Images of fluorescent avidin conjugate bound to chondroitin sulfate without (top) and with (bottom) treatment with 15 mU/mL of heparinase for 2 hours.

Protease Activity of Heparinase III
There was no detectable protease activity in the heparinase (15mU/mL) at any time interval tested. Specifically, after 3 hours of incubation in the protease assay, chymotrypsin protease activity varied from 214 to 420 fluorescence units as the concentration increased from 0.16 to 1.25 µmol/L. For the same concentration range, pronase activity varied between 135 and 520 fluorescence units. These data demonstrate the lower and upper limits of sensitivity of the assay (135 to 520 fluorescence units). Five replications of heparinase (15 mU/mL) incubation (3 hours) in the protease assay produced activity readings between 104 and 121 fluorescence units, demonstrating that no detectable protease activity was present.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 up arrowResults
 *Discussion
down arrowReferences
 
A number of studies using cultured ECs have reported shear-induced NOx production. Kuchan and Frangos12 using human umbilical vein endothelial cells (HUVECs) exposed to steady shear observed a biphasic response characterized by an initial burst of NOx production within minutes followed by a more gradual NOx release over hours. It was also demonstrated that the initial response was G protein and Ca2+-dependent, but independent of shear magnitude in the range of 6 to 25 dyne/cm2.12,15

In contrast, the secondary response was G protein and Ca2+-independent and shear level dependent. Frangos et al33 also demonstrated that the initial elevated response was a result of the high rate of change of shear stress associated with the step change from zero to elevated (steady) shear stress. The initial response could be eliminated by ramping up the shear to the same steady level in a gradual fashion.33 Chang et al11 exposed BAECs to step changes in shear stress and observed a biphasic response of NOx production similar to that reported for HUVECs.12 In the present study, NOx production in response to a step change to steady shear stress displayed the characteristic biphasic response (Figure 1) that has been described in earlier studies. NOx production under oscillatory shear conditions was also characterized by distinct biphasic behavior with an elevated rate in the first 5 minutes followed by a more gradual production out to 3 hours (Figure 2). The mechanisms associated with the two phases of the oscillatory response have not been investigated, but it seems likely that the initial burst of production is associated with the sudden start up of the shear stress process.

Previous experiments have demonstrated that heparinase, an enzyme specific for heparan sulfates in the glycocalyx, has the ability to alter the thickness34,35 and barrier function of the glycocalyx.36 Dull et al37 used the same concentration of heparinase as in the present study, and observed 67% removal of labeled sulfate (35SO4) in bovine lung microvascular endothelial cells. Haldenby et al32 demonstrated that heparinase treatments were able to reduce fluorescence intensity by 27% to 51% using fluorescence-labeled wheat germ agglutinin in vitro. Results from the present experiments (Figure 4) indicated a reduction of 45.9±13.8% in the fluorescence intensity associated with a heparan sulfate antibody, which is consistent with previous studies. The specificity of the heparinase enzyme was further verified by demonstrating a negligible degradation of chondroitin sulfate (Figure 5) and undetectable protease activity. The effects of other glycoprotein-specific enzymes on the structure and function of the glycocalyx have also been examined. Desjardins and Duling34 reported that neuraminidase, papain, pronase E, and clostripain had no effect on capillary tube hematocrit, a marker of glycocalyx thickness. Henry and Duling38 observed that hyaluronidase increased access of 70- and 145-kDa dextrans to the endothelial surface layer but not larger dextrans or red blood cells.

Having demonstrated that our EC monolayers produced NOx in response to shear stress in a manner consistent with previous reports in the literature and that we were able to substantially reduce the heparan sulfate GAG component of the glycocalyx through heparinase pretreatment, we were able to address the hypothesis that this GAG component is a shear stress sensor. After treatment with heparinase, NOx production under steady shear stress was substantially inhibited (Figure 1). Oscillatory shear stress experiments with heparinase-treated EC monolayers were characterized by complete inhibition of NOx production at all times (Figure 2). The viability of the NO production apparatus of our ECs after treatment with heparinase was also investigated by exposure to bradykinin and histamine in the absence of shear stress. We observed that BK (Figure 3) and histamine induced significant NOx production that was not altered by pretreatment with heparinase. This indicates that heparinase treatment did not have any inhibitory effects on NOx production unless flow (shear) was present. These experiments support our hypothesis that heparan sulfate is the shear sensor for NO production by EC.

An earlier study used the enzyme neuraminidase (0.2 U/mL) to remove sialic acid residues from saline-perfused rabbit mesenteric arteries and observed that flow-dependent vasodilation was abolished by a 30-minute pretreatment with the enzyme.39 Because flow-dependent vasodilation is mediated by NO release in many arteries, and neuraminidase degrades the surface glycocalyx, this study is consistent with our observations. However, the interpretation of this whole artery study is less clear. The enzyme may have also affected the smooth muscle cells in the artery and thus the effect cannot be uniquely associated with the endothelium. In addition, the protease activity of neuraminidase was not assessed in that study as it was for heparinase in the present work.

There are also important differences in the structural alterations generated by various enzymes and their subsequent impact on flow-mediated signals that may be transduced into a biological response. As mentioned previously, neuraminidase digestion of the capillary surface was not associated with changes in capillary tube hematocrit and thus not likely to have caused marked changes in glycocalyx thickness. Heparinase digestion, however, was clearly associated with changes in glycocalyx thickness and glycocalyx barrier function. We believe these two observations are complimentary. Constituents of the glycocalyx form an entangled matrix covering the endothelial surface, and it is highly likely that mechanical properties of this surface layer are imparted not only by the scaffolding components, such as heparan sulfates, but also by associated macromolecules including sialic acid-containing glycoproteins, proteoglycans, and oligosaccharides that are degraded by neuraminidase. We have demonstrated that heparan sulfates are involved in flow-mediated mechanotransduction, but we do not exclude the possibility that other components of the glycocalyx participate in this process.

It is interesting to note that by removing only 45% of cell surface heparan sulfate, we could completely abolish shear-induced NO production. The ability of 45% heparan sulfate removal to totally abolish NO production suggests that a threshold of transmitted stress must be exceeded to stimulate NO production. Many signaling pathways require a "threshold level " of stimulation that must be attained in order to provide activation. We believe that the mechanotransduction system is no different. Partial removal of the mechanosensor may reduce the transmitted signal below the threshold for response. This is of course speculative because we do not yet know the details of the signal transduction mechanism downstream of the mechanosensor.

The underlying model associated with our hypothesis that heparan sulfate is a shear sensor is the following: as blood flow interacts with proteoglycans extending from the endothelial surface, a shearing force is applied to the outstretched heparan sulfate GAG component, displacing the core proteins from their unsheared orientation. This displacement is transmitted into the cell provoking a response.

Of particular interest in regard to NO production are G protein–linked receptors, which interact with G protein complexes. G proteins, in turn, are believed to be involved in eNOS activation. In HUVECs, the first phase of the NO response to a step change in shear stress is both G protein and Ca2+-dependent, whereas the later, sustained phase does not depend on G proteins or Ca2+.15 We have observed in BAECs that the sustained phase is also inhibited by removal of the heparan sulfate shear sensor, suggesting that linkages in addition to G proteins are affected. A possible explanation lies in the myristoylation site at the N-terminus of eNOS, which may anchor itself to the plasma membrane, or possibly even to proteoglycans, which penetrate the plasma membrane.40 The orientation of this linkage would likely be independent of both Ca2+ and G proteins. Further studies are required to evaluate this hypothesis.

A unifying hypothesis of mechanotransduction would likely require the involvement of cytoskeletal elements in series with the cell surface shear sensor. The arrangement of heparan sulfate as a shear sensor with subsequent signaling through the cytoskeleton is attractive for several reasons. First, heparan sulfates are abundant on the endothelial surface and, at least one class of HSPG, ie, syndecans, is known to associate with cytoskeletal elements including actin, either directly or through associated actin-binding proteins. Secondly, the NO pathway in endothelial cells has been shown to be regulated by the state of local actin filaments,41 thus providing a mechanistic framework to link cell-surface heparan sulfates to the nitric oxide pathway. Support for this notion can be derived from studies on FGF-mediated signaling, where FGF-induced vasodilatation is mediated by NOS activation and requires heparan sulfates as a coligand; over expression of syndecan-4 enhances FGF-mediated NOS activity, as does the addition of exogenous heparan sulfate,42 demonstrating the importance of heparan sulfate as a cofactor in NOS activation. Thus, physical association, whether direct or indirect, of heparan sulfates with actin and NOS provide a strong conceptual framework to evaluate mechanotransduction mechanisms. Studies are underway to evaluate the structural mechanisms underlying heparan sulfate-mediated mechanotransduction. To this end, we have mutants of syndecan-1 that have the heparan sulfate attachment sites deleted and a mutant with the cytoplasmic domain truncated. These mutant syndecans should allow us to further define the role of heparan sulfates, proteoglycan core proteins, and cytoskeletal interactions in endothelial mechanotransduction.


*    Acknowledgments
 
Acknowledgments

This work was supported by NIH National Heart, Lung, and Blood Institute Grants HL35549 and HL57093 to J.M.T. and by a Johns Hopkins University Clinician-Scientist Award and NIH NHLBI Grant HL680683 to R.O.D.


*    Footnotes
 
Original received October 17, 2002; first resubmission received June 11, 2003; second resubmission received July 11, 2003; third resubmission received October 6, 2003; accepted October 6, 2003.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMaterials and Methods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Lever MJ, Tarbell JM, Caro CG. The effect of luminal flow in rabbit carotid artery on transmural fluid transport. Exp Physiol. 1992; 77: 553–563.[Abstract]
  2. Sill HW, Chang YS, Artman JR, Frangos JA, Hollis TM, Tarbell JM. Shear stress increases hydraulic conductivity of cultured endothelial monolayers. Am J Physiol. 1995; 268: H535–H543.[Medline] [Order article via Infotrieve]
  3. Williams DA. Network assessment of capillary hydraulic conductivity after abrupt changes in shear stress. Microvasc Res. 1999; 57: 107–117.[CrossRef][Medline] [Order article via Infotrieve]
  4. Dewey CF Jr, Bussolari SR, Gimbrone MA Jr, Davies PF. The dynamic response of vascular endothelial cells to fluid shear stress. J Biomech Eng. 1981; 103: 177–185.[Medline] [Order article via Infotrieve]
  5. Eskin SG, Ives CL, McIntire LV, Navarro LT. Response of cultured endothelial cells to steady flow. Microvasc Res. 1984; 28: 87–94.[CrossRef][Medline] [Order article via Infotrieve]
  6. Franke RP, Grafe M, Schnittler H, Seiffge D, Mittermayer C, Drenckhahn D. Induction of human vascular endothelial stress fibres by fluid shear stress. Nature. 1984; 307: 648–649.[CrossRef][Medline] [Order article via Infotrieve]
  7. Levesque MJ, Nerem RM. The study of rheological effects on vascular endothelial cells in culture. Biorheology. 1989; 26: 345–357.[Medline] [Order article via Infotrieve]
  8. Frangos JA, Eskin SG, McIntire LV, Ives CL. Flow effects on prostacyclin production by cultured human endothelial cells. Science. 1985; 227: 1477–1479.[Abstract/Free Full Text]
  9. Hsieh HJ, Li NQ, Frangos JA. Shear-induced platelet-derived growth factor gene expression in human endothelial cells is mediated by protein kinase C. J Cell Physiol. 1992; 150: 552–558.[CrossRef][Medline] [Order article via Infotrieve]
  10. Lan Q, Mercurius KO, Davies PF. Stimulation of transcription factors NF {kappa}B and AP1 in endothelial cells subjected to shear stress. Biochem Biophys Res Commun. 1994; 201: 950–956.[CrossRef][Medline] [Order article via Infotrieve]
  11. Chang YS, Yaccino JA, Lakshminarayanan S, Frangos JA, Tarbell JM. Shear-induced increase in hydraulic conductivity in endothelial cells is mediated by a nitric oxide-dependent mechanism. Arterioscler Thromb Vasc Biol. 2000; 20: 35–42.[Abstract/Free Full Text]
  12. Kuchan MJ, Frangos JA. Role of calcium and calmodulin in flow-induced nitric oxide production in endothelial cells. Am J Physiol. 1994; 266: C628–C636.[Medline] [Order article via Infotrieve]
  13. Rubanyi GM, Romero JC, Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol. 1986; 250: H1145–H1149.[Medline] [Order article via Infotrieve]
  14. Ziegler T, Bouzourene K, Harrison VJ, Brunner HR, Hayoz D. Influence of oscillatory and unidirectional flow environments on the expression of endothelin and nitric oxide synthase in cultured endothelial cells. Arterioscler Thromb Vasc Biol. 1998; 18: 686–692.[Abstract/Free Full Text]
  15. Kuchan MJ, Jo H, Frangos JA. Role of G proteins in shear stress-mediated nitric oxide production by endothelial cells. Am J Physiol. 1994; 267: C753–C758.[Medline] [Order article via Infotrieve]
  16. Davies PF, Mundel T, Barbee KA. A mechanism for heterogeneous endothelial responses to flow in vivo and in vitro. J Biomech. 1995; 28: 1553–1560.[CrossRef][Medline] [Order article via Infotrieve]
  17. Ihrcke NS, Wrenshall LE, Lindman BJ, Platt JL. Role of heparan sulfate in immune system-blood vessel interactions. Immunol Today. 1993; 14: 500–505.[CrossRef][Medline] [Order article via Infotrieve]
  18. Turnbull J, Powell A, Guimond S. Heparan sulfate: decoding a dynamic multifunctional cell regulator. Trends Cell Biol. 2001; 11: 75–82.[CrossRef][Medline] [Order article via Infotrieve]
  19. Bernfield M, Kokenyesi R, Kato M, Hinkes MT, Spring J, Gallo RL, Lose EJ. Biology of the syndecans: a family of transmembrane heparan sulfate proteoglycans. Annu Rev Cell Biol. 1992; 8: 365–393.[CrossRef][Medline] [Order article via Infotrieve]
  20. Asundi VK, Carey DJ. Self-association of N-syndecan (syndecan-3) core protein is mediated by a novel structural motif in the transmembrane domain and ectodomain flanking region. J Biol Chem. 1995; 270: 26404–26410.[Abstract/Free Full Text]
  21. Weksberg R, Squire JA, Templeton DM. Glypicans: a growing trend. Nat Genet. 1996; 12: 225–227.[CrossRef][Medline] [Order article via Infotrieve]
  22. Dull RO, Davies PF. Flow modulation of agonist (ATP)-response (Ca2+) coupling in vascular endothelial cells. Am J Physiol. 1991; 261: H149–H154.[Medline] [Order article via Infotrieve]
  23. Dull RO, Tarbell JM, Davies PF. Mechanisms of flow-mediated signal transduction in endothelial cells: kinetics of ATP surface concentrations. J Vasc Res. 1992; 29: 410–419.[Medline] [Order article via Infotrieve]
  24. Vink H, Duling BR. Capillary endothelial surface layer selectively reduces plasma solute distribution volume. Am J Physiol Heart Circ Physiol. 2000; 278: H285–H289.[Abstract/Free Full Text]
  25. Pries AR, Secomb TW, Gaehtgens P. The endothelial surface layer. Pflugers Arch. 2000; 440: 653–666.[CrossRef][Medline] [Order article via Infotrieve]
  26. Rapraeger AC. Syndecan-regulated receptor signaling. J Cell Biol. 2000; 149: 995–998.[Abstract/Free Full Text]
  27. Zimmermann P, David G. The syndecans, tuners of transmembrane signaling. FASEB J. 1999; 13: S91–S100.[Medline] [Order article via Infotrieve]
  28. Carey DJ. Syndecans: multifunctional cell-surface co-receptors. Biochem J. 1997; 327: 1–16.[Medline] [Order article via Infotrieve]
  29. Hillsley MV, Tarbell JM. Oscillatory shear alters endothelial hydraulic conductivity and nitric oxide levels. Biochem Biophys Res Commun. 2002; 293: 1466–1471.[CrossRef][Medline] [Order article via Infotrieve]
  30. Misko TP, Schilling RJ, Salvemini D, Moore WM, Currie MG. A fluorometric assay for the measurement of nitrite in biological samples. Anal Biochem. 1993; 214: 11–16.[CrossRef][Medline] [Order article via Infotrieve]
  31. Kure S, Yoshie O. A syngeneic monoclonal antibody to murine Meth-A sarcoma (HepSS-1) recognizes heparan sulfate glycosaminoglycan (HS-GAG): cell density and transformation dependent alteration in cell surface HS-GAG defined by HepSS-1. J Immunol. 1986; 137: 3900–3908.[Abstract]
  32. Haldenby KA, Chappell DC, Winlove CP, Parker KH, Firth JA. Focal and regional variations in the composition of the glycocalyx of large vessel endothelium. J Vasc Res. 1994; 31: 2–9.[Medline] [Order article via Infotrieve]
  33. Frangos JA, Huang TY, Clark CB. Steady shear and step changes in shear stimulate endothelium via independent mechanisms–superposition of transient and sustained nitric oxide production. Biochem Biophys Res Commun. 1996; 224: 660–665.[CrossRef][Medline] [Order article via Infotrieve]
  34. Desjardins C, Duling BR. Heparinase treatment suggests a role for the endothelial cell glycocalyx in regulation of capillary hematocrit. Am J Physiol. 1990; 258: H647–H654.[Medline] [Order article via Infotrieve]
  35. Pries A, Secomb T, Jacobs H, Sperandio M, Osterloh K, Haehtgens P. Microvascular blood flow resistance: role of endothelial surface layer. Am J Physiol Heart Circ Physiol. 1997; 273: H2272–H2279.[Abstract/Free Full Text]
  36. Mulivor AW, Lipowsky HH. Role of glycocalyx in leukocyte-endothelial cell adhesion. Am J Physiol Heart Circ Physiol. 2002; 283: H1282–H1291.[Abstract/Free Full Text]
  37. Dull RO, Dinavahi R, Schwartz L, Humphries D, Berry D, Sasiskeharan R, Garcia J. Lung endothelial heparan sulfate mediate cationic peptide-induced barrier dysfunction: a new role for the glycocalyx. Am J Physiol Lung Cell Mol Physiol. 2003; 285: L986–L995.[Abstract/Free Full Text]
  38. Henry CB, and Duling BR. Permeation of the luminal capillary glycocalyx is determined by hyaluronan. Am J Physiol Heart Circ Physiol. 1999; 277: H508–H514.[Abstract/Free Full Text]
  39. Pohl U, Herlan K, Huang A, Bassenge E. EDRF-mediated shear-induced dilation opposes myogenic vasoconstriction in small rabbit arteries. Am J Physiol Heart Circ Physiol. 1991; 261: H2106–H2013.
  40. Davies P. Flow-mediated endothelial mechanotransduction. Physiol Rev. 1995; 75: 519–560.[Abstract/Free Full Text]
  41. Zharikov SI, Sigova AA, Chen S, Bubb MR, Block ER. Cytoskeletal regulation of the L-arginine/NO pathway in pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L465–L473.[Abstract/Free Full Text]
  42. Li J, Partovian C, Hampton TG, Metais C, Tkachenko E, Sellke FW, Simons M, Parovian C. Modulation of microvascular signaling by heparan sulfate matrix: studies in syndecan-4 transgenic mice. Microvasc Res. 2002; 64: 38–46.[CrossRef][Medline] [Order article via Infotrieve]



This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
M. Nieuwdorp, M. C. Meuwese, H. L. Mooij, C. Ince, L. N. Broekhuizen, J. J. P. Kastelein, E. S. G. Stroes, and H. Vink
Measuring endothelial glycocalyx dimensions in humans: a potential novel tool to monitor vascular vulnerability
J Appl Physiol, March 1, 2008; 104(3): 845 - 852.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
A. P. Stevens, V. Hlady, and R. O. Dull
Fluorescence correlation spectroscopy can probe albumin dynamics inside lung endothelial glycocalyx
Am J Physiol Lung Cell Mol Physiol, August 1, 2007; 293(2): L328 - L335.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
Y. Yao, A. Rabodzey, and C. F. Dewey Jr.
Glycocalyx modulates the motility and proliferative response of vascular endothelium to fluid shear stress
Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H1023 - H1030.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
M. D. Carattino, W. Liu, W. G. Hill, L. M. Satlin, and T. R. Kleyman
Lack of a role of membrane-protein interactions in flow-dependent activation of ENaC
Am J Physiol Renal Physiol, July 1, 2007; 293(1): F316 - F324.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
P. Cabrales, B. Y. S. Vazquez, A. G. Tsai, and M. Intaglietta
Microvascular and capillary perfusion following glycocalyx degradation
J Appl Physiol, June 1, 2007; 102(6): 2251 - 2259.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
R. O. Dull, I. Mecham, and S. McJames
Heparan sulfates mediate pressure-induced increase in lung endothelial hydraulic conductivity via nitric oxide/reactive oxygen species
Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1452 - L1458.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
M. Hirakawa, M. Oike, M. Watanabe, Y. Karashima, and Y. Ito
Pivotal role of integrin {alpha}5{beta}1 in hypotonic stress-induced responses of human endothelium
FASEB J, October 1, 2006; 20(12): 1992 - 1999.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
B. van den Berg and H. Vink
Glycocalyx perturbation: cause or consequence of damage to the vasculature?
Am J Physiol Heart Circ Physiol, June 1, 2006; 290(6): H2174 - H2175.
[Full Text] [PDF]

Home page
 Am. J. Physiol. Cell Physiol.Home page
A. Makino, E. R. Prossnitz, M. Bunemann, J. M. Wang, W. Yao, and G. W. Schmid-Schonbein
G protein-coupled receptors serve as mechanosensors for fluid shear stress in neutrophils
Am J Physiol Cell Physiol, June 1, 2006; 290(6): C1633 - C1639.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
D. Hong, D. Jaron, D. G. Buerk, and K. A. Barbee
Heterogeneous response of microvascular endothelial cells to shear stress
Am J Physiol Heart Circ Physiol, June 1, 2006; 290(6): H2498 - H2508.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
M. Nieuwdorp, T. W. van Haeften, M. C.L.G. Gouverneur, H. L. Mooij, M. H.P. van Lieshout, M. Levi, J. C.M. Meijers, F. Holleman, J. B.L. Hoekstra, H. Vink, et al.
Loss of Endothelial Glycocalyx During Acute Hyperglycemia Coincides With Endothelial Dysfunction and Coagulation Activation In Vivo
Diabetes, February 1, 2006; 55(2): 480 - 486.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
B. M. van den Berg, J. A. E. Spaan, T. M. Rolf, and H. Vink
Atherogenic region and diet diminish glycocalyx dimension and increase intima-to-media ratios at murine carotid artery bifurcation
Am J Physiol Heart Circ Physiol, February 1, 2006; 290(2): H915 - H920.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. Gouverneur, J. A. E. Spaan, H. Pannekoek, R. D. Fontijn, and H. Vink
Fluid shear stress stimulates incorporation of hyaluronan into endothelial cell glycocalyx
Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H458 - H452.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
M. B. Dancu, D. E. Berardi, J. P. Vanden Heuvel, and J. M. Tarbell
Asynchronous Shear Stress and Circumferential Strain Reduces Endothelial NO Synthase and Cyclooxygenase-2 but Induces Endothelin-1 Gene Expression in Endothelial Cells
Arterioscler.