AJP - Heart Calcium Transients and Cell-Sarcomere
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 278: H469-H476, 2000;
0363-6135/00 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (30)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Phelps, J. E.
Right arrow Articles by DePaola, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Phelps, J. E.
Right arrow Articles by DePaola, N.
Vol. 278, Issue 2, H469-H476, February 2000

Spatial variations in endothelial barrier function in disturbed flows in vitro

Jeffrey E. Phelps and Natacha DePaola

Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, New York 12180


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Hindered barrier function has been implicated in the initiation and progression of atherosclerosis, a disease of focal nature associated with altered hemodynamics. In this study, endothelial permeability to macromolecules and endothelial electrical resistance were investigated in vitro in monolayers exposed to disturbed flow fields that model spatial variations in fluid shear stress found at arterial bifurcations. After 5 h of flow, areas of high shear stress gradients showed a 5.5-fold increase in transendothelial transport of dextran (molecular weight 70,000) compared with no-flow controls. Areas of undisturbed fully developed flow, within the same monolayer, showed a 2.9-fold increase. Monolayer electrical resistance decreased with exposure to flow. The resistance measured during flow and the rate of change in monolayer resistance after removal of flow were lowest in the vicinity of flow reattachment (highest shear stress gradients). These results demonstrate that endothelial barrier function and permeability to macromolecules are regulated by spatial variations in shear stress forces in vitro.

endothelial permeability; electrical impedance; atherosclerosis; intercellular gaps


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

AT THE INTERFACE of the arterial wall and the circulating blood, the endothelium constitutes the principal barrier to transport of blood-borne substances into the arterial wall. If the permselective barrier function of the endothelium is compromised, molecules such as low-density lipoprotein (LDL) as well as monocytes are allowed to pass from the circulation into the subendothelial space, setting the stage for the initiation of atherosclerotic lesions.

Localized regions of enhanced permeability in vivo are associated with a number of functional and structural features of the endothelium, such as enhanced endothelial cell turnover, lipid accumulation and metabolism, and polygonal cell morphology, among others (20, 28). What controls endothelial permeability to macromolecules is still an unanswered question. Weinbaum and colleagues (22, 34) proposed that the main pathway for enhanced macromolecular transport across the endothelium is intercellular gaps (leaky junctions) that appear transiently when endothelial cells divide. Another known pathway for water-soluble solutes is pinocytosis. Cells that are dying or sloughing off also contribute to enhanced transendothelial macromolecule transport (21).

In their natural environment, endothelial cells are constantly exposed to physical and biochemical stimuli that can alter cell permeability. Transendothelial transport of macromolecules has been shown to be responsive to flow shear stress, hydrostatic pressure, thermal shock, and agonists such as histamine, thrombin, bradykinin, and phorbol 12-myristate 13-acetate (PMA), among others (2, 13, 14, 19, 23, 25, 33). Endothelial permeability has been investigated in vitro under uniform shear flow conditions. Davies (6) observed that step changes in shear stress increase the rate of pinocytosis. Jo et al. (19) demonstrated that albumin permeability of endothelial monolayers is shear dependent, time dependent, and reversible (higher shear stress results in higher permeability). More recently, Sill et al. (29) reported a shear-dependent increase in hydraulic conductivity in cultured endothelial monolayers. However, in vivo, areas of enhanced endothelial permeability coincide with the presence of disturbed flows (i.e., arterial bifurcations), where the shear stress magnitude is low but the spatial and temporal gradients in shear stress are large. Furthermore, the enhanced permeability characteristic of regions of disturbed flow is not transient but permanent.

We hypothesize that the shear stress gradients present in areas of disturbed flow alter the structure and function of intercellular junctions, resulting in a permanent increase in transendothelial transport of macromolecules. We have previously demonstrated that cell proliferation, gap junctional communication, gene transcription and expression of the gap junctional protein connexin43 (Cx43), cell motion, and monolayer cell density in vascular endothelial cells are dynamically regulated by macroscopic shear stress gradients in vitro (7-10, 27).

In this study we developed an in vitro transport assay to evaluate spatial variations in the active transport of macromolecules across endothelial monolayers exposed to disturbed flows. We also quantified the spatial and temporal variations in monolayer resistance to ion transport during flow exposure in an attempt to identify flow-mediated alterations in barrier function and macromolecule transport pathways that may contribute to the pathological levels of transendothelial permeability observed at arterial bifurcations in vivo.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Cell culture. Bovine aortic endothelial cells were harvested from calf thoracic aortas with the use of standard techniques (18). Cell monolayers were cultured and grown to confluence in DMEM buffered with 25 mM HEPES and containing 10% calf serum, 0.3 mg/ml L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin.

Cells between passages 12 and 26 were seeded at a density of 2 × 105 cells/cm2 on gelatin-coated polycarbonate filters (0.4-µm pore size and 24-mm diameter; Corning Costar Transwell, Cambridge, MA) or on gelatin-coated tissue culture plates (25 × 75 mm) containing eight small gold electrodes (Applied Biophysics, Troy, NY) and were exposed to fluid forces between 3 and 5 days postseeding. Before cells were seeded, each polycarbonate filter was removed from the Transwell chamber and mounted on a stainless steel frame onto which a step (0.5 × 2 × 28 mm) was fixed to generate a local flow disturbance (Fig. 1A). The electrode plates also contained local flow disturbances (Fig. 2A).


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1.   A: porous membrane assemblage for transport studies. A steel frame holds a polycarbonate filter (0.4-µm pore size) with a 2 × 28 × 0.5-mm step fixed to its surface (flow disturbance). B: transport assay chamber. The abluminal side of the porous membrane is placed in direct contact with a 22-mm-diameter agarose layer resting on a glass slide (75 × 38 mm). A bottomless well clamped on the steel frame of each sample forms an upper luminal chamber.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2.   A: electrode location for electrical impedance measurements under conditions of disturbed flow. Tissue culture plates contain 8 electrodes: 2 positioned within the recirculation region, 2 close to flow reattachment, and 4 in the region of fully developed flow. Only 4 electrodes are shown (shaded squares); the other 4 electrodes are located along a line parallel to the flow direction 2 cm apart from electrodes shown. Roman numerals indicate points of interest: I, edge of step; II, maximum shear stress in reverse flow; III, flow reattachment; IV, 84% recovered flow; and V, 95% recovered flow. B: sections of agarose gel examined in transport assay. Each section measures 1.5 × 15 × 1 mm. Regions A1-A7 are described in detail in text.

Flow chambers. Two separate flow apparatuses were used in these studies. For the permeability protocols, the steel frames holding the filters with confluent monolayers were loaded in a parallel plate flow chamber previously described by DePaola et al. (9). The frames sit in a recess on the bottom plate of the chamber so that the luminal cell surface is continuous with the channel surface. The subluminal cell surface is permanently wetted by the stagnant fluid in the recess. Shear stress forces on the endothelial cells are generated by the viscous flowing medium. The interposition of a rectangular step, with its largest dimension perpendicular to the flow direction, results in an area of flow separation and recirculation with a nonuniform shear stress distribution that models physiological flows at arterial bifurcations. The detailed shear stress distribution in the apparatus was obtained using finite-element models and was previously reported by DePaola and colleagues (7, 9). Table 1 summarizes the average shear stress and shear stress gradient in regions of interest. In all experiments reported here, perfusion rates were adjusted to generate wall shear stresses of 10 dyn/cm2 in flow regions away from the step.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Shear stress and shear stress gradients on the endothelial surface

The measurements of monolayer electrical resistance under disturbed flow were conducted using a flow chamber specially designed to be used with the electrical cell substrate impedance-sensing (ECIS) system developed by Giaever and Keese (16, 17). This parallel plate flow chamber allows perfusion of the culture plate containing the surface electrodes and flow disturbances without interfering with the ECIS measurement. A detailed description of the design and operation of this flow chamber with the ECIS system has been previously reported by DePaola et al. (11). Briefly, the flow chamber consisted of two parallel plates connected to a flow loop. A precisely machined recess (1 × 25 × 60 mm) on the top plate defined the flow path in the chamber. The top plate (Teflon) also housed the inlet and outlet ports and a glass window for light transmission and sample visualization. The bottom plate had a machined recess for sample (electrode plate with culture cells) placement with a glass window for light transmission and visualization. The electrode contacts connecting the flow chamber to the ECIS system were also held on the bottom plate. A medical-grade silicon gasket was used to seal the chamber and prevent fluid leakage while the plates were held together by four screws. Flow was imposed on the endothelial surface by connecting the parallel plate chamber to a flow circuit. The interposition of a rectangular step on the electrode plate surface (Fig. 2A) results in an area of flow separation and recirculation with a nonuniform shear stress distribution. The fluid dynamic conditions (shear stress distribution) are identical to those in the flow chamber used for the permeability studies (Table 1). The flow chamber and flow loop were kept inside a tissue culture incubator (37°C, 5% CO2 atmosphere) for the duration of the experiments. The flow chamber was transferred to the mechanical stage of an inverted-phase microscope for visualization of the monolayers during the experiments.

Transport assay. To quantify the spatial variations in permeability of an endothelial cell layer exposed to disturbed flow fields, an assay was developed whereby color-labeled macromolecules permeating the cell layer were caught in a subluminal gel. As the macromolecules came into contact with the gel, the transport was hindered, thereby creating a record of the amount of macromolecules that have permeated the cell layer at a given location in a given period of time. The intensity of color in the gel was used as an indicator of the level of permeability of the corresponding cell area. Therefore, spatial variations in transendothelial cell macromolecular transport were determined by measuring the spatial distribution of color intensity in the gel.

Confluent endothelial monolayers were exposed to disturbed flow for 5 h at the flow conditions shown in Table 1. When flow ceased, the samples were rinsed in warm PBS (GIBCO, Gaithersburg, MD) and transferred to a transport assay chamber in which the abluminal side of the polycarbonate filter was in direct contact with a 1-mm-thick, 1.75% by mass, agarose gel layer (Fig. 1B). A bottomless well with a sealing edge was lightly clamped on the steel frame of the sample to form an upper luminal chamber (3 ml). A known concentration (8 mg/ml) of FITC-dextran [molecular weight (MW) 70,000 and approximate Stokes-Einstein radius of 3-4.6 nm; Sigma, St. Louis, MO] in PBS containing 1% (wt/vol) BSA and 25 mM HEPES (pH 7.4) was added to the luminal well and allowed to permeate at room temperature for 30 min. The agarose layer was then removed from the chamber, and seven 1.5 × 15-mm strips were sectioned from selected areas (Fig. 2B). Two strips corresponded to cells in the recirculation region, another two corresponded to cells in the area of flow recovery, and the last three corresponded to cells exposed to fully recovered, undisturbed flow. Five of the seven 1.5-mm-wide strips were consecutively cut from the downstream edge of the step, whereas the last two strips were randomly cut from areas further downstream (Fig. 2B). The strips were then dissolved in warm PBS, and the resulting suspensions were analyzed in a spectrophotometer at a wavelength of 490 nm. The ensuing absorbance readings quantify the amount of tracer that permeated the selected sections of the cell layer, thus allowing the evaluation of spatial variations in monolayer permeability.

Positive controls consisted of confluent monolayers grown on polycarbonate filters and incubated in 0.5 µM PMA (Sigma, St. Louis, MO) for 90 min. Before permeability was assessed, these samples were thoroughly but gently rinsed in three separate PBS baths.

Static control monolayers were plated on polycarbonate filters at the same time as experimental samples and maintained under no-flow conditions in a standard 37°C, humidified 95% air-5% CO2 incubator for the duration of the flow experiments. Controls were processed simultaneously with corresponding flow samples by following the same procedures.

Cell layers were scrupulously examined before and after the permeability assay. Any layers showing damage caused by manipulation were discarded.

Electrical resistance measurements. The ECIS technique (16, 17) was used to monitor changes in endothelial cell barrier function under conditions of disturbed flow. In this system cells are cultured on tissue culture plates containing several small electrodes (0.6 × 10-3 cm2) and a large counter electrode (1 cm2). The electrodes are connected to a lock-in amplifier, and an approximately constant current source applies a 1 µA alternating current signal between the small and counter electrodes, using culture medium as the electrolyte. Voltage changes between the small and counter electrodes are monitored with the lock-in amplifier. The measured in-phase voltage is proportional to the total electrical resistance across the monolayer, which is reflective of focal adhesion, and the resistance between cells (17). Widening of the intercellular spacing in confluent endothelial monolayers is detected by the ECIS system as a decrease in the measured resistance (31). A personal computer controls the output of the amplifier and switches the measurement to different small electrodes. The data from each small electrode provide information from a population of ~70 cells covering the electrode.

In the studies reported here, monolayers (18 cm2) exposed to disturbed flow were grown on ECIS culture plates with eight small gold electrodes strategically located to provide information from small groups of cells subjected to different flow conditions within the same monolayer. The electrodes were distributed such that two were located in the recirculation region, two were close to flow reattachment, and four were in downstream areas where the flow is fully recovered (Fig. 2A). The shear stress and shear stress gradients in each of those regions are shown in Table 1.

Electrode plates with confluent cell monolayers were placed in the ECIS flow chamber and allowed to equilibrate, under no-flow conditions, to 37°C for 45 min. Before the onset of flow, 5 min of impedance data were collected to establish baseline resistances. All monolayers were exposed to disturbed flows for 5 h. Resistance measurements were taken every minute during exposure to flow. Postflow monolayer resistances were monitored for an additional 24-64 h in a chamber with a larger volume, designed to accommodate the ECIS culture plate and the connections to the ECIS system in a static no-flow environment and to provide cell nutrients for the extended period of the study.

The chambers with the cell monolayers and reservoirs with medium for perfusion were kept inside a cell culture incubator (37°C, 5% CO2) for the entire duration of the experiments.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Endothelial cells exposed to disturbed flows exhibited a significant increase in transendothelial macromolecular transport in areas of high shear stress gradients. Figure 3A shows the spatial distribution of dextran (MW 70,000) accumulated in the subendothelial gel of monolayers exposed to disturbed flow for 5 h. Immediately downstream from the step (within the area of flow recirculation) endothelial permeability is highest, decreasing in the downstream direction as flow recovers. The average shear stress gradient in regions A1 (flow recirculation) and A2 (at and close to flow reattachment) is 79 and 85 dyn · cm-2 · cm-1, respectively (Table 1 and Fig. 3). Downstream from flow reattachment, the shear stress gradient is an order of magnitude lower (5 dyn · cm-2 · cm-1) in regions A3 and A4 and almost negligible further downstream (regions A5-A7). Some variability was observed in the macromolecule transport values obtained for the region of flow recovery (regions A3-A5). However, the data obtained in the region of fully developed flow far downstream from the flow disturbance (regions A6 and A7, randomly chosen) showed very little variability with macromolecule transport values significantly lower than those found within the region of flow disturbance.


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 3.   Spatial variation in transendothelial macromolecule transport. Monolayers were exposed to disturbed flows for 5 h. Data plotted are mass of dextran (embedded in agarose layer) per unit area of agarose (mean ± SE) as a function of distance from the step; n = 6 independent samples. Transport was allowed for 30 min. A: average transendothelial macromolecule transport for regions A1-A7, which correspond to sections indicated in Fig. 2B. B: average transendothelial macromolecule transport in regions of high shear stress gradients (HSSG; 79-85 dyn · cm-2 · cm-1), regions of fully developed flow (FDF), monolayer regions treated with phorbol 12-myristate 13-acetate (PMA; 0.05 µM, 90 min), and static no-flow controls. * P < 0.01 compared with static no-flow control; + P < 0.05 compared with FDF (Student's t-test).

Figure 3B shows the average mass of dextran accumulated in the subendothelial gel for regions of high shear stress gradient (regions A1 and A2), fully developed downstream flow (regions A5-A7), PMA-treated cell layers, and no-flow control layers. In areas of high shear stress gradients, endothelial monolayers showed a 5.5-fold increase in the transport of dextran across the layer compared with no-flow control layers. However, areas of undisturbed fully developed flow, within the same monolayer, showed only a 2.9-fold increase. Endothelial monolayers exposed to disturbed flow for 5 h are on average 1.9 times more permeable in the area of flow recirculation than in areas exposed to undisturbed fully developed flow. These flow-induced alterations in monolayer permeability to dextran (MW 70,000) are of the same order of magnitude as those obtained with PMA treatment. Visual inspection (light microscopy) at various points during the transport assay confirmed that monolayers were not damaged by manipulation.

Spatial variations in endothelial monolayer resistance to ion transport were continuously monitored during flow exposure. Figure 4 shows the monolayer resistance in three regions of interest (within flow recirculation, close to flow reattachment, and fully developed downstream flow) during exposure to disturbed flow for 5 h. The position of the electrodes in relation to the step disturbance is shown in Fig. 2. Resistances are normalized by monolayer baseline values determined before the onset of flow. Measured baseline resistances varied from 3,200 to 5,000 Omega , with the mean being 4,200 Omega  (measured values include the 1,800-Omega average resistance of a bare electrode). Endothelial monolayer resistance followed the same pattern of change for all three regions investigated. There was a sharp increase in resistance at the onset of flow, reaching 1.5 times baseline values in ~15 min, followed by a downslope to baseline that lasted for nearly the same period of time. For the next 2 h of flow, the resistance plateaued, sometimes with an apparent slight hump or secondary peak of much lower magnitude. After this period, endothelial resistance in all three regions decreased, reaching ~50% of the baseline values within 5 h of flow. Although no statistically significant differences were found among the three regions examined within the monolayer, there was a characteristic spread of the data that was seen in all experiments. The average resistance in the recirculation zone was larger than the resistance of the region of fully developed flow in every single experiment, and the resistance measured in the vicinity of flow reattachment was always the lowest of the three. An additional 1.5-h monitoring of the monolayer resistance after cessation of flow showed a sustained decrease in resistance with no significant difference among the three regions.


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 4.   Average endothelial resistance vs. time of exposure to flow. RC, flow recirculation (n = 14 electrodes); FR, flow reattachment (n = 14 electrodes); FDF, n = 24 electrodes. Data plotted are average resistances, normalized by baseline values, with SE bars shown every 1 h ± 10 min obtained from 9 independent experiments. Five minutes of baseline data were collected before onset of flow (flow on, thin arrow), and 1.5 h of data were recorded after the removal of flow (flow off, thick arrow).

When monolayers were examined for up to 24 h after removal of flow, there were significant differences between the resistance close to flow reattachment and the resistance of the other examined regions of the monolayer (Fig. 5A). In all regions the average resistance continued to decrease for ~2 h after the removal of flow before increasing and overshooting baseline values. The minimum monolayer resistance was recorded in the region of flow reattachment, where the resistance increased at a much slower rate than in the other regions, only slightly overshooting baseline values within 24 h.


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5.   A: monolayer resistance after cessation of flow. RC, n = 6 electrodes; FR, n = 6 electrodes; FDF, n = 12 electrodes. Data plotted were obtained from 4 independent experiments. Flow was stopped at 5 h (flow stop, arrow). Values are average normalized resistance with SE bars shown every 3 ± 0.5 h. B: rate of change in monolayer resistance calculated as average slope of resistance vs. time curve for selected time periods after cessation of flow. Values are means ± SE. * P < 0.05, FR vs. FDF regions of monolayer. + P < 0.05, FR vs. RC regions of monolayer.

Figure 5B shows the rate of change in monolayer resistance in the three regions of interest. Between 5 and 7 h (first 2 h after flow removal), the resistance decreased at nearly the same rate for both regions of flow reattachment and regions of fully developed flow, whereas in the recirculation region, the resistance decreased at a much higher rate (P < 0.05). However, between 7 and 15 h, the resistance changed with a significantly different rate (P < 0.05) between regions of flow reattachment and regions of fully developed flow. Values for the recirculation region showed no significant difference compared with those for the region of fully developed flow. After 15 h, the rates were similar for all areas. In general, the rate of change of monolayer resistance near flow reattachment was lower than in the fully developed region for all time periods. After 30 h, the monolayer resistance at flow reattachment remained at the same value with little change (Fig. 6). In contrast, the resistance in the regions of recirculation and fully developed flow continued to rise and reached 2 and 2.1 times the baseline value, respectively. These elevated resistances were sustained for ~5 h, decreasing to ~1.4 times the initial value between 40 and 60 h. At 60 h, the endothelial resistance was nearly homogeneous throughout the entire layer at ~1.4 times the initial value. Monolayers examined up to 90 h after flow (not shown) did not reveal any further changes in monolayer resistance.


View larger version (31K):
[in this window]
[in a new window]
  		Fig. 6.   Changes in transendothelial resistance <= 64 h after flow removal. Values are average normalized resistance with SE bars shown every 5 ± 1 h. RC and FR, n = 4 electrodes; FDF, n = 8 electrodes. Monolayers were exposed to flow between 0 and 5 h. Data were obtained from 3 independent experiments.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

In vivo, localized regions of enhanced endothelial permeability coincide with the presence of disturbed flows. Although the effect of fluid flow on transendothelial transport has been investigated in vitro under unidirectional laminar flow conditions (19, 25, 29), to our knowledge this is the first study to quantify macromolecule and ion transport across endothelial monolayers exposed to controlled disturbed flows in vitro in which spatial variations in fluid shear stress are similar to those found at arterial bifurcations.

In this paper we report increased macromolecule transport across endothelial monolayers exposed to disturbed flows. Macromolecule transport was considerably larger in regions of flow disturbance than in adjacent regions of the same cell monolayer exposed to uniform shear flow, demonstrating a strong correlation between shear stress gradients and monolayer permeability. In areas of flow disturbance, shear stress gradients are high, as opposed to regions of fully developed flow, where shear stress gradients are negligible. The wall shear stress magnitude in regions of disturbed flow is lower than in regions of fully developed flow but, on average, is of the same order of magnitude.

Although the mechanisms that regulate endothelial permeability in vivo and in vitro are not fully understood, there is a general agreement that transendothelial macromolecular transport occurs via paracellular gaps (1, 4). Our data suggest that differential forces (shear stress gradients) among neighboring cells in the disturbed flow region may widen the intercellular junctions, locally increasing macromolecule transport. This is consistent with our previous finding that shear stress gradients regulate intercellular junction structure. Specifically, normal Cx43 gap junctions distributed at the cell periphery were severely disassembled in regions of disturbed flow after 5 h with much less disruption of the normal junctional pattern observed in regions of fully developed flow (7, 9). It is possible that intercellular gap junctional disassembly contributes to intercellular widening and, consequently, to macromolecule permeability.

Cells that are dying or sloughing off could also account for increased transport across the monolayer (21, 34). However, in these experiments, flow-induced changes in permeability were not associated with cell loss or gross damage to the cell layer.

Enhanced macromolecular permeability of endothelial cells has also been associated with cell mitosis (22). Previous studies in our laboratory have demonstrated that shear stress gradients significantly increased endothelial cell DNA synthesis in monolayers exposed to disturbed flow in vitro for 5 h (7, 8, 12), indicating that the effects of hemodynamics on the regulation of cell growth may be important in controlling endothelial permeability. However, in the present study the permeability of the cell layer was assessed immediately after 5 h of flow, long before cells committed to the cell cycle in response to shear gradients could have reached mitosis. Therefore, the early changes in macromolecule permeability reported here are most likely caused by widening of the intercellular gaps. Under prolonged exposure to flow, sustained endothelial DNA synthesis in areas of high shear gradients (8, 10) suggests that cell mitosis (and associated leaky junctions) may be continuously resulting in regions of permanently enhanced permeability.

Macromolecule permeability also increased in areas of fully developed flow. With the use of Fick's law of diffusion to evaluate flux across a semipermeable membrane between two well-stirred compartments (5), the tracer concentration in the abluminal chamber ([A]) at time t is given by [A]t = V([L]0/VAC) [1 - e-(PS/V)t], where [L]0 is the initial tracer concentration in the luminal side, P is the permeability coefficient, S is the surface area, VAC is the volume of the abluminal compartment, and V is (VACVLC)/(VAC + VLC), where VLC is the volume of the luminal compartment. If (PS/V)t is <0.1, then P can be approximated as P = VAC[A]/([L]0St) = MAC/([L]0t), where MAC is the mass of dextran embedded in the agarose layer per unit area of agarose, which is the quantity we measure in our assay. Assuming a constant concentration of the luminal chamber (8 mg/ml) and a constant rate over the 30 min of transport, we may estimate endothelial permeability for each region of the monolayer using the above equation and the values reported in Fig. 3. The permeability of the regions of fully developed flow in the monolayer (A6 and A7) is 1.1 × 10-6 cm/s, whereas the permeability in the areas of high shear stress gradients (A1 and A2) is 2 × 10-6 cm/s. These values of permeability are of the same order of magnitude as those reported by Jo et al. (19) and Ohshima and Ookawa. (25) for endothelial monolayers exposed to unidirectional laminar flows with similar shear stress forces. In vivo, endothelial permeability to macromolecules is two to three orders of magnitude lower than the values reported in in vitro studies with cell cultures grown in filter membranes (1, 4, 5). The values reported in the present study are also higher than in vivo permeability rates. However, our transport assay allowed the quantification of fine spatial differences in transendothelial transport in monolayers exposed to fluid forces characteristic of the in vivo hemodynamic conditions at arterial bifurcations. The subluminal agarose gel was able to quickly trap small amounts of tracer, allowing fine differences between small areas of the monolayer to be quantified.

Our results indicating that shear stress gradients may be a localizing factor in enhanced endothelial permeability are supported by the in vivo localization of leaky areas in arterial endothelium, also found in association with altered hemodynamics, where shear stress forces are variable (24, 28). Moreover, the permeability ratio between regions of disturbed and undisturbed (fully developed) flow reported here in vitro are of the same order as those estimated in in vivo studies of spatial variations of arterial wall permeability (3, 32). Barakat et al. (3) found that the density of sites of enhanced permeability to the macromolecule horseradish peroxidase (HRP spots) is 2 to 10 times higher in the immediate vicinity of aortic ostia than within a few millimeters of it. Truskey et al. (32) reported endothelial permeabilities for LDL ~10 times higher in high permeability regions compared with low permeability regions in the rabbit arterial wall. Similar spatial variations in permeability were found in the present studies. In regions of flow disturbance permeability was ~2 times higher than in regions of undisturbed flow and 6 times higher than in control (no-flow) monolayers. On the basis of our early studies on endothelial cell morphology and function, we may expect a larger difference between the permeabilities in the regions of disturbed and undisturbed flow under prolonged (<= 48 h) exposure to disturbed flows. Endothelial permeability in regions of undisturbed flow should decrease after 24 h, once endothelial cells have reorganized their intercellular junctions and adapted to the new flow condition (7, 9). In contrast, endothelial permeability in regions of disturbed flow is expected to remain high because of sustained localized cell proliferation and intercellular junction disassembly (7-10).

Changes in transendothelial resistance measured with the ECIS technique have been previously correlated with both changes in intercellular spacing and permeability to macromolecules (15, 26, 31). In the present study, continuous evaluation of the transendothelial resistance during exposure to flow showed that resistance decreases with flow, which is consistent with the fact that flow increases permeability. However, significant differences in electrical resistance between various regions of the monolayer were not accounted for (Fig. 4) despite the fact that the transport data demonstrated significant spatial variations in monolayer permeability to dextran (MW 70,000) (Fig. 3). This suggests that after 5 h of flow, the differences between intercellular widening in regions of disturbed and undisturbed flow may be very subtle and beyond the level of detection by the ECIS technique.

Changes in electrical resistance may represent changes in cell-cell or cell-substrate spacing (16, 17). The peak in resistance at the onset of flow (Fig. 4) may then be interpreted as a significant reduction of the cell-substrate spacing caused by the sudden increased shear and pressure forces pushing the cell layer against the substrate (11). A transient reduction in intercellular spacing caused by cell deformation also may be expected at the onset of flow. The peak disappears shortly after (15 min) because of flow-induced intercellular gap widening.

The plateau or slight hump in monolayer resistance (Figs. 4 and 5A) observed during the first 2 h of flow may be attributed to cell motion. Studies from our laboratory (12) have demonstrated that there is a significant increase in cell motion during the first 2 h of exposure to flow. Cell motion and migration involve some degree of cell spreading that may reduce the cell-substrate and/or cell-cell spacing that would in turn increase monolayer resistance. Therefore, the plateau or slight hump shown in Fig. 4 may be the result of a partial balance between flow-induced gap widening and cell spreading during motion. Cell motion was observed throughout the entire monolayer, with the highest activity reported in the region of flow recirculation (10, 12, 30), which coincidentally also showed the largest resistance peak among the three regions investigated. Moreover, the slight meandering seen in the resistance trace during this time period (Fig. 4) is consistent with resistance fluctuations detected with the ECIS system during cell motion (17).

After the cessation of flow, the monolayer resistance increased, indicating some degree of recovery that in fact resulted in resistance values that exceeded baseline conditions within 24 h. This overcompensating cell response was less pronounced in areas of the monolayer close to flow reattachment. These results demonstrate that flow-induced alteration in monolayer resistance, although reversible throughout the monolayer, is altered to a greater extent by the flow conditions at flow reattachment. The new increased baseline resistance, reached by 70 h after flow removal, may be caused by a remnant of the effects of flow or may be simply reflective of extracellular matrix alterations or higher cell packing in the monolayer caused by the extra time (3 days) in culture.

These studies demonstrate that spatial variations in shear stress forces regulate endothelial barrier function and permeability to macromolecules in vitro. Similarities between in vivo spatial variations in permeability and findings of this study suggest that shear stress gradient may be a key regulator of endothelial permeability in vivo. Although the local increase in macromolecule transport appears to be a result of widening of junctions by direct effect of differential forces among neighboring cells, variations in monolayer resistance during and after removal of flow may reflect other mechanisms by which endothelial cells respond to disturbed flow. Shear stress gradient regulation of cell growth may also be an important parameter in controlling endothelial permeability. In conclusion, these results demonstrated that shear stress gradients associated with altered hemodynamics play an important role in determining regional differences in endothelial permeability in vitro.


    ACKNOWLEDGEMENTS

We thank Applied Biophysics, Inc. (Troy, NY) for the use of the ECIS equipment and Dr. Charles Keese for valuable discussions.


    FOOTNOTES

This work was supported by National Science Foundation Grant NSF-9624991 (to N. DePaola) and a Whitaker Foundation Biomedical Engineering Research Grant (to N. DePaola).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests and other correspondence: N. DePaola, Dept. of Biomedical Engineering, Rensselaer Polytechnic Institute, 110 8th St., Troy, NY 12180.

Received 28 September 1998; accepted in final form 3 August 1999.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Albelda, S. M., P. M. Sampson, F. R. Haselton, J. M. McNiff, S. M. Mueller, S. K. Williams, A. P. Fishman, and E. M. Levine. Permeability characteristics of cultured endothelial monolayers. J. Appl. Physiol. 64: 308-322, 1988[Abstract/Free Full Text].

2.   Baetscher, M., and K. Brune. An in vitro system for measuring endothelial permeability under hydrostatic pressure. Exp. Cell Res. 148: 541-547, 1983[Web of Science][Medline].

3.   Barakat, A. I., P. A. F. Uhthoff, and C. K. Colton. Topographical mapping of sites of enhanced HRP permeability in the normal rabbit aorta. J. Biomech. Eng. 114: 283-292, 1992[Web of Science][Medline].

4.   Casnocha, S. A., S. G. Eskin, E. R. Hall, and L. V. McIntire. Permeability of human endothelial monolayers: effect of vasoactive agonists and cAMP. J. Appl. Physiol. 67: 1997-2005, 1989[Abstract/Free Full Text].

5.   Cooper, J. A., P. J. Del Vecchio, F. L. Minnear, K. E. Burhop, W. M. Selig, J. G. Garcia, and A. B. Malik. Measurement of albumin permeability across endothelial monolayers in vitro. J. Appl. Physiol. 62: 1076-1083, 1987[Abstract/Free Full Text].

6.   Davies, P. F. How do vascular endothelial cells respond to flow? News in Physiol. Sci. 4: 22-25, 1989[Abstract/Free Full Text].

7.   Davies, P. F., N. DePaola, and D. C. Polacek. Hemodynamic forces and vascular cell communication. Atherosclerosis XI 9: 943-948, 1998.

8.   DePaola, N. The effect of shear stress gradients on endothelial cell proliferation and monolayer integrity. Adv. Bioeng. 28: 321-322, 1994.

9.   DePaola, N., P. F. Davies, W. F. Pritchard, L. Florez, N. Harbeck, and D. C. Polacek. Spatial and temporal regulation of gap junction connexin43 in vascular endothelial cells exposed to controlled disturbed flows in vitro. Proc. Natl. Acad. Sci. USA 96: 3154-3159, 1999[Abstract/Free Full Text].

10.   DePaola, N., M. A. Gimbrone, Jr., P. F. Davies, and C. F. Dewey, Jr. Vascular endothelium responds to fluid shear stress gradients. Arterioscler. Thromb. 12: 1254-1257, 1992[Abstract/Free Full Text]. [Corrigenda. Arterioscler. Thromb. 13: March 1993, p. 465.]

11.  DePaola, N., J. Phelps, L. Florez, C. Keese, F. Minnear, I. Giaever, and P. Vincent. Electrical impedance of cultured endothelium under fluid flow. Ann. Biomed. Eng. In press.

12.  DePaola, N., D. Polacek, L. Florez, P. F. Davies, and J. Phelps. Shear stress gradients and endothelial cell function. Proc. Japan-US-Singapore-China Conf. Biomechanics 5th Sendai 1998 p. 48-49.

13.   Eaton, B. M., V. J. Toothill, H. A. Davies, J. D. Pearson, and G. E. Mann. Permeability of human venous endothelial cell monolayers perfused in microcarrier cultures: effects of flow rate, thrombin, and cytochalasin D. J. Cell. Physiol. 149: 88-99, 1991[Web of Science][Medline].

14.   Ehringer, W. D., M. J. Edwards, and F. N. Miller. Mechanisms of alpha -thrombin, histamine, and bradykinin induced endothelial permeability. J. Cell. Physiol. 167: 562-569, 1996[Web of Science][Medline].

15.   Garcia, J. G., K. L. Schaphorst, S. Shi, A. D. Vernin, C. M. Hart, K. S. Callahan, and C. E. Patterson. Mechanisms of ionomycin-induced endothelial cell barrier dysfunction. Am. J. Physiol. Lung Cell. Mol. Physiol. 273: L172-L184, 1997[Abstract/Free Full Text].

16.   Giaever, I., and C. R. Keese. Use of electrical fields to monitor the dynamical aspect of cell behavior in tissue culture. IEEE Trans. Biomed. Eng. 33: 242-247, 1986[Web of Science][Medline].

17.   Giaever, I., and C. R. Keese. Micromotion of mammalian cells measured electrically. Proc. Natl. Acad. Sci. 88: 7896-7900, 1991[Abstract/Free Full Text].

18.   Gimbrone, M. A. Culture of vascular endothelium. In: Progress in Hemostasis and Thrombosis. New York: Grone and Stratton, 1976, p. 1-28.

19.   Jo, H., R. O. Dull, T. M. Hollis, and J. M. Tarbell. Endothelial albumin permeability is shear dependent, time dependent, and reversible. Am. J. Physiol. Heart Circ. Physiol. 260: H1992-H1996, 1991[Abstract/Free Full Text].

20.   Kim, D. W., B. L. Langille, M. K. Wong, and A. I. Gotlieb. Patterns of endothelial microfilament distribution in the rabbit aorta in situ. Circ. Res. 64: 21-31, 1989[Abstract/Free Full Text].

21.   Lin, S. J., K. M. Jan, and S. Chien. The role of dying endothelial cells in transendothelial macromolecular transport. Arteriosclerosis 10: 703-709, 1990[Abstract/Free Full Text].

22.   Lin, S. J., K. M. Jan, G. Schuessler, S. Weinbaum, and S. Chien. Enhanced macromolecule permeability of aortic endothelial cells in association with mitosis. Arteriosclerosis 73: 223-232, 1988.

23.   Liu, S. M., and T. Sundqvist. Effects of hydrogen peroxide and phorbol myristate acetate on endothelial transport and F-actin distribution. Exp. Cell Res. 217: 1-7, 1995[Web of Science][Medline].

24.   Newman, D. L., J. R. Batten, and N. L. Bowden. Influence of experimental stenosis on uptake of albumin by abdominal aorta. Atherosclerosis 26: 195-204, 1977[Web of Science][Medline].

25.   Ohshima, N., and K. Ookawa. Effects of shear stress on cytoskeletal structure and physiological functions of cultured endothelial cells. In: Cell Mechanics and Cellular Engineering. New York: Springer-Verlag, 1994, p. 107-117.

26.   Patil, S., J. E. Kaplan, and F. L. Minnear. Protein, not adenosine or adenine nucleotides, mediates platelet decrease in endothelial permeability. Am. J. Physiol. Heart Circ. Physiol. 273: H2304-H2311, 1997[Abstract/Free Full Text].

27.   Polacek, D., N. DePaola, W. Pritchard, and P. F. Davies. Spatial up-regulation of gap junction connexin43 mRNA in endothelial cells exposed to steady laminar and disturbed flows in vitro (Abstract). FASEB J. 9: A267, 1995.

28.   Schwenke, D. C., and T. E. Carew. Initiation of atherosclerotic lesions in cholesterol-fed rabbits. II. Selective retention of LDL vs. selective increase in LDL permeability in susceptible sites of arteries. Arteriosclerosis 9: 908-918, 1989[Abstract/Free Full Text].

29.   Sill, H. W., Y. S. Chang, J. R. Artman, J. A. Frangos, T. M. Hollis, and J. M. Tarbell. Shear stress increases hydraulic conductivity of cultured endothelial monolayers. Am. J. Physiol. Heart Circ. Physiol. 268: H535-H543, 1995[Abstract/Free Full Text].

30.   Tardy, Y., N. Resnick, T. Nagel, M. A. Gimbrone, Jr., and C. F. Dewey, Jr. Shear stress gradients remodel endothelial monolayers in vitro via a cell proliferation-migration-loss cycle. Arterioscler. Thromb. Vasc. Biol. 17: 3102-3106, 1997[Abstract/Free Full Text].

31.   Tiruppathi, C., A. B. Malik, P. J. Del Vecchio, C. R. Keese, and I. Giaever. Electrical method for detection of endothelial cell shape change in real time: assessment of endothelial barrier function. Proc. Natl. Acad. Sci. 89: 7919-7923, 1992[Abstract/Free Full Text].

32.   Truskey, G. A., W. L. Roberts, R. A. Herrmann, and R. A. Malinauskas. Measurement of endothelial permeability to 125I-low density lipoproteins in rabbit arteries by use of en face preparations. Circ. Res. 71: 883-897, 1992[Abstract/Free Full Text].

33.   Waters, C. M. Flow-induced modulation of the permeability of endothelial cells cultured on microcarrier beads. J. Cell. Physiol. 168: 403-411, 1997.

34.   Weinbaum, S., G. Tzeghai, P. Gantos, R. Pfeffer, and S. Chien. Effect of cell turnover and leaky junctions on arterial macromolecular transport. Am. J. Physiol. Heart Circ. Physiol. 248: H945-H960, 1985.

Am J Physiol Heart Circ Physiol 278(2):H469-H476
0363-6135/00 $5.00 Copyright © 2000 the American Physiological Society


This article has been cited by other articles:


Home page
 Cardiovasc ResHome page
T. Miyazaki, K. Honda, and H. Ohata
m-Calpain antagonizes RhoA overactivation and endothelial barrier dysfunction under disturbed shear conditions
Cardiovasc Res, October 7, 2009; (2009) cvp311v2.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
J. A. LaMack and M. H. Friedman
Individual and combined effects of shear stress magnitude and spatial gradient on endothelial cell gene expression
Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H2853 - H2859.
[Abstract] [Full Text] [PDF]

Home page
 JCBHome page
A. W. Orr, R. Stockton, M. B. Simmers, J. M. Sanders, I. J. Sarembock, B. R. Blackman, and M. A. Schwartz
Matrix-specific p21-activated kinase activation regulates vascular permeability in atherogenesis
J. Cell Biol., February 26, 2007; 176(5): 719 - 727.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Cell Physiol.Home page
J. E. Bodmer, A. English, M. Brady, K. Blackwell, K. Haxhinasto, S. Fotedar, K. Borgman, E.-W. Bai, and A. B. Moy
Modeling error and stability of endothelial cytoskeletal membrane parameters based on modeling transendothelial impedance as resistor and capacitor in series
Am J Physiol Cell Physiol, September 1, 2005; 289(3): C735 - C747.
[Abstract] [Full Text] [PDF]

Home page
 JCBHome page
A. W. Orr, J. M. Sanders, M. Bevard, E. Coleman, I. J. Sarembock, and M. A. Schwartz
The subendothelial extracellular matrix modulates NF-{kappa}B activation by flow: a potential role in atherosclerosis
J. Cell Biol., April 11, 2005; 169(1): 191 - 202.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
T. Sokabe, K. Yamamoto, N. Ohura, H. Nakatsuka, K. Qin, S. Obi, A. Kamiya, and J. Ando
Differential regulation of urokinase-type plasminogen activator expression by fluid shear stress in human coronary artery endothelial cells
Am J Physiol Heart Circ Physiol, November 1, 2004; 287(5): H2027 - H2034.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
S. Q. Liu, D. Tang, C. Tieche, and P. K. Alkema
Pattern formation of vascular smooth muscle cells subject to nonuniform fluid shear stress: mediation by gradient of cell density
Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H1072 - H1080.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
S. Q. Liu, C. Tieche, D. Tang, and P. Alkema
Pattern formation of vascular smooth muscle cells subject to nonuniform fluid shear stress: role of PDGF-{beta} receptor and Src
Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H1081 - H1090.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
A. R. Brooks, P. I. Lelkes, and G. M. Rubanyi
Gene expression profiling of human aortic endothelial cells exposed to disturbed flow and steady laminar flow
Physiol Genomics, April 10, 2002; 9(1): 27 - 41.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
K. G. Birukov, A. A. Birukova, S. M. Dudek, A. D. Verin, M. T. Crow, X. Zhan, N. DePaola, and J. G. N. Garcia
Shear Stress-Mediated Cytoskeletal Remodeling and Cortactin Translocation in Pulmonary Endothelial Cells
Am. J. Respir. Cell Mol. Biol., April 1, 2002; 26(4): 453 - 464.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
S. M. Dudek and J. G. N. Garcia
Cytoskeletal regulation of pulmonary vascular permeability
J Appl Physiol, October 1, 2001; 91(4): 1487 - 1500.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
A. R. Brooks, P. I. Lelkes, and G. M. Rubanyi
Gene expression profiling of human aortic endothelial cells exposed to disturbed flow and steady laminar flow
Physiol Genomics, April 10, 2002; 9(1): 27 - 41.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (30)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Phelps, J. E.
Right arrow Articles by DePaola, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Phelps, J. E.
Right arrow Articles by DePaola, N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online