| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Physiology, 2Pharmacology, and 3Pathology, and 4Center for Lung Biology, University of South Alabama, Mobile, Alabama
Submitted 19 July 2005 ; accepted in final form 23 December 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
P) was controlled by the reservoir height, and the filtration rate corrected for surface area (Jv/A) was measured by timing fluid movement in a calibrated micropipette. These parameters were used to calculate hydraulic conductance (Lp) by using linear regression of Jv/A on P. Mean Lp values for newly confluent RPAEC monolayers were 22 times higher than those for RPMVEC monolayers (28.6 ± 5.6 vs. 1.30 ± 0.50 x 10–7 cm·s–1·cmH2O–1; P 
0.01). After confluence was reached, electrical resistance and Lp remained stable in RPAEC but continued to change in RPMVEC with days in culture. Both phenotypes exhibited an initial time-dependent sealing response, but Lp also had an inverse relationship to P in RPMVEC monolayers 
4 days postconfluence that was attributed to cell overgrowth rather than junctional length. In a comparison of the cadherin contents, E-cadherin was predominant in RPMVEC, but VE-cadherin was predominant in RPAEC. At a constant P of 40–45 cmH2O for 2 h, Jv/A increased 225% in RPAEC monolayers but did not change significantly in RPMVEC monolayers. Significant decreases in Lp were obtained after treatment with 5% albumin, GdCl3, or isoproterenol plus rolipram in both phenotypes. Thus lung microvascular endothelial cells exhibited a significantly lower Lp than conduit vessel endothelium, which would limit alveolar flooding relative to perivascular edema cuff formation during increased pulmonary vascular pressures. capillary permeability; pulmonary edema; VE-cadherin; gadolinium
Previous studies from investigators in our group also have demonstrated a reduced baseline permeability to solutes and a reduced permeability response to inflammatory mediators in rat pulmonary microvascular endothelial cells (RPMVEC) compared with rat pulmonary arterial endothelial cells (RPAEC) (9). The mechanism may relate to a reduced store-operated Ca2+ entry and enhanced cAMP turnover in RPMVEC (19), which uncouple permeability from Ca2+ entry (2). During mechanical stress, Ca2+ also may enter endothelial cells through stretch-activated cation channels. Blockade of these channels using GdCl3 was shown to attenuate high-airway pressure mechanical injury in intact rat lungs (15), but the contribution of the segmental endothelial cell phenotypes to this response is unknown. The purpose of the present studies was to compare Lp measurements of cultured RPAEC and RPMVEC monolayers, their responses to a sustained hydrostatic stress, and the effects of pharmacological interventions that modulate permeability.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-actin. RPMVEC were cultured as previously described (19). The heart and lungs were removed and placed in a DMEM bath containing penicillin and streptomycin. Thin strips were removed from the lung periphery, finely minced, and transferred to a 15-ml tube containing 3 ml of DMEM and 3 ml of a digestion solution containing 0.5 g of BSA, 10,000 units of type 2 collagenase (Worthington Biochemical, Lakewood, NJ), and cmf-PBS (GIBCO BRL) to make 10 ml in total volume. The digestion mixture was incubated at 37°C for 15 min and strained through an 80-mesh sieve into a sterile beaker. The sieve was washed with 5 ml of DMEM containing penicillin and streptomycin. The mixture was transferred to a 15-ml conical tube and centrifuged at 300 g for 5 min. The medium was aspirated, and the cells were resuspended in 5 ml of medium containing one part vascular conditioned medium and three parts incomplete medium [80% RPMI 1640, 20% FBS, 12.3 U/ml heparin (Elkins-Sinn, Cherry Hill, NJ), and 6.7 µg/ml Endogro (Vec Technologies, Rensselaer, NY) with 30 µg/ml penicillin and streptomycin]. Centrifugation was repeated, and the cells were resuspended in 3 ml of complete medium, incubated at 37°C for 30 min, and placed onto 35-mm culture dishes. The dishes were checked daily for contaminating cells, which were removed by scraping and aspiration. Endothelial cell colonies were isolated with cloning rings, trypsinized, resuspended in 100 µl of complete medium, and placed as a drop in the center of a T-25 flask. The cells were allowed to attach (1 h at 37°C with 5% CO2) before the addition of 5 ml of complete medium. Cultures were characterized using scanning electron microscopy, uptake of 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine-labeled LDL, and a lectin-binding panel of Helix pomatia, Griffonia simplicifolia, and Glycine max. H. pomatia is selective for RPMVEC, whereas G. simplicifolia and G. max are selective for RPAEC (11). Passages 4–13 were used for experiments.
Experimental protocol.
Experiments were performed on monolayers grown by seeding 80,000 cells/cm2 on 12-mm Transwell filters (0.4-µm pore size; Fisher Scientific, Suwanee, GA) and growing to confluence with culture times ranging from 3 to 14 days. Electrical resistance was measured using an EVOM impedance meter and an Endohm resistance well (World Precision Instruments, Sarasota, FL). Filters were then mounted on a six-well Ussing diffusion chamber (Harvard Apparatus, Holliston, MA) and perfused with Eagle's minimum essential medium containing 25 mM HEPES. Perfusate was circulated between the apical chamber and reservoir with a Gilson roller pump, and reservoirs were bubbled with 95% O2-5% CO2. Perfusion pressure (P) was controlled by adjusting reservoir height, and filtration rates (Jv) were measured by timing the meniscus movements in volumetric pipettes. Simultaneous meniscus movements were recorded for all six chambers with a Watec video camera and Panasonic video recorder. Filtration rates were corrected for filtration surface area (A). Lp (in 10–7 cm·s–1·cmH2O–1) was calculated from a regression of Jv/A vs. P at different P using
 | (1) |
P, Lp was calculated using only the P at that Jv/A.
Experimental protocols included step changes in P or a sustained P for 2 h. In selected experiments, 5% BSA, 30 µM GdCl3, or 20 µM isoproterenol plus 10 µM rolipram (Sigma Chemical, St. Louis, MO) were added to the reservoir.
Immunoblotting.
RPAEC and RPMVEC were grown to confluence, and cells were lysed with 500 µl of ice-cold lysis buffer at pH 7.4 (50 mM HEPES, 5 mM EDTA, 50 mM NaCl), 1% Triton X-100, protease inhibitors (10 µg/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin), and phosphatase inhibitors (50 mM sodium fluoride, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate). Solubilized proteins were centrifuged at 14,000 g in a microfuge (4°C) for 15 min, and supernatants were stored at –80°C. Extracted proteins were quantified using the Bradford assay. Samples prepared directly after cell lysis were separated using SDS-PAGE and were transferred to nitrocellulose membranes. Membranes were blocked overnight at 4°C with PBS containing 5% nonfat dry milk and 0.1% Tween 20. The blots were incubated for 1 h with primary antibodies in PBS containing 1% nonfat dry milk and 0.1% Tween 20. After incubation with secondary antibodies for 1 h in PBS containing 1% nonfat dry milk and 0.1% Tween 20, proteins were detected using chemiluminescence. 
-Actin and E-cadherin antibodies were purchased from BD Transduction Laboratories (San Diego, CA). VE-cadherin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The relative densities of the protein bands were quantitated using NIH Image version 1.63.
Transmission electron microscopy. Porous filters 1.2 cm in diameter were seeded with RPAEC and RPMVEC in DMEM with 20% FBS. At 1 and 4 days after confluence was reached, the monolayers were fixed in 3% glutaraldehyde in cacodylate buffer. For transmission electron microscopy, the filters were rinsed in cacodylate buffer, post-fixed for 1 h with 1% osmium tetroxide, dehydrated using a graded alcohol series, and embedded in PolyBed 812 resin (Polysciences, Warrington, PA). Thin sections (80 nm) were cut from each block with a diamond knife and then stained with uranyl acetate and Reynold's lead citrate. Sections were examined and photographed using a Philips CM 100 transmission electron microscope (FEI, Hillsboro, OR) (11). The lengths of each complete junction on 10–12 photographs were measured and scaled to the magnification to determine length. We measured 14–16 junctions for each cell type at each time period.
Statistics.
Data are expressed as means ± SE. Statistical differences between groups were determined using either Student's t-test for two groups or one-way ANOVA with a Newman-Keuls posttest for multiple groups. A P 0.05 established a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
 | (2) |
|
P] were used to calculate Lp for the two phenotypes. All regressions for monolayers during this time period were significant with mean r2 values of 0.92 ± 0.03 for RPAEC and 0.95 ± 0.01 for RPMVEC. In contrast, RPMVEC cultured for longer time periods (Fig. 1B) exhibited an unexpected pressure-induced sealing for steady-state filtration, whereby steady-state Jv/A actually decreased. Filtration rates became vanishingly small at higher pressures, and Lp could not be calculated from a conventional regression analysis. An estimate of Lp could be obtained in these postconfluent RPMVEC monolayers by dividing Jv/A by the absolute pressure increase from zero at each state. In these RPMVEC monolayers, calculated Lp would decrease with pressure rather than remain constant with pressure increases. Estimates of average Lp values obtained using this method for calculating Lp in the mature RPMVEC monolayers were 1.7 x 10–8, 2.7 x 10–9, 1.2 x 10–9, and 5 x 10–10 cm·s–1·cmH2O–1 for pressure increases of 10, 20, 30, and 40 cmH2O, respectively. Figure 2 summarizes the Lp values (means ± SE) calculated from regression analyses of each experiment in confluent monolayers described in Fig. 1B. For RPAEC and RPMVEC monolayers (n = 11 each), the Lp values averaged 28.6 ± 5.6 and 1.3 ± 0.5 x 10–7 cm·s–1·cmH2O–1, respectively. Thus the average Lp for RPAEC was significantly 22-fold greater than that for RPMVEC.
|
3–4 days and RPAEC after 4–5 days. As shown in Fig. 3, electrical resistance initially decreased for 2–3 days and then increased as cells reached confluence. However, RPMVEC resistances (n = 11) continued to increase as long as they remained in culture, whereas RPAEC (n = 10) reached a constant plateau resistance after 5 days in culture. Figure 4 indicates that the initial Lp values in monolayers cultured over this time period also exhibited a significant reduction (P 0.001) with days in culture for RPMVEC monolayers (n = 6) but remained stable with days in culture for RPAEC monolayers (n = 6). These Lp changes correlated with the monolayer electrical resistance changes with time.
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Measurements and estimates of endothelial cell Lp in situ are generally lower than those measured in cultured monolayers, but in situ estimates also suggest a much lower microvascular Lp than predicted from cultures of conduit vessel endothelial cells. In direct micropuncture measurements of Lp obtained using the split-drop technique in dog lung venules, Bhattacharya (1) calculated Lp values of 2.9 ± 0.02 x 10–7 cm·s–1·cmH2O–1. In later studies using the split drop in surface venules in rat lungs, Safdar et al. (17) calculated a baseline Lp of 4.4 x 10–7 cm·s–1·cmH2O–1, which decreased to 1.5 x 10–7 cm·s–1·cmH2O–1 after hypertonic challenge. Parker and Yoshikawa (16) reported a segmental distribution of the filtration coefficient in isolated rat lungs that was 18% arterial, 41% capillary, and 41% venous, and they used estimates of vascular surface areas based on vascular cast data to estimate Lp values for arterial, venous, and microvascular segments of 6 x 10–8, 1.3 x 10–7, and 1.8 x 10–9 cm·s–1·cmH2O–1. Note that these estimates predicted an Lp for RPAEC that was 30 times that of the Lp for RPMVEC and an Lp for venules that was 65 times that for microvascular segments. The venular Lp was approximately the value subsequently reported in isolated rat lungs by investigators using the split-drop technique after hypertonic challenge (17). Other Lp estimates can be derived from morphometric vascular surface area measurements reported by Weibel (27). He reported an average capillary surface area of 0.41 m2 in lungs of 140-g rats. Using our estimates of lung weight (1.22 g) and filtration coefficient (0.1 ml·min–1·cmH2O–1·100 g–1) for this size rat, we can calculate a vascular surface area of 3,360 cm2/g lung weight and an average Lp of 4 x 10–9 cm·s–1·cmH2O–1 from the data. Thus the Lp reported in the present study for cultured RPAEC and RPMVEC are at least an order of magnitude higher than values estimated from morphologic estimates of vascular surface area, but the 22-fold greater Lp that we observed for RPAEC compared with RPMVEC is within a range predicted from morphologic estimates. These data all indicate a much lower fluid conductance of alveolar capillaries than could be predicted from previous studies of systemic conduit vessel endothelial cells.
The marked difference in Lp between RPMVEC and RPAEC mirrors many of the significant differences reported for these segmental endothelial phenotypes. RPMVEC and RPAEC have different embryologic origins, being derived respectively by vasculogenesis and angiogenesis, and RPMVEC have a much higher growth potential than RPAEC (7, 8, 11). King et al. (11) reported that RPMVEC grown for 6 days in 10% FBS had approximately twice the cell number as RPAEC when seeded at the same cell densities. In fact, the RPMVEC tended to overgrow and actually form bilayers when cultured for long periods (Fig. 6D). The faster growth rate and tighter junctional permeability of RPMVEC suggest many epithelial-like characteristics. Epithelial monolayers generally have electrical resistances an order of magnitude higher than that of endothelial monolayers and much lower solute diffusion permeabilities (10). The fact that E-cadherin was the major adherens junction cadherin identified in RPMVEC could partially account for the lower permeability of the RPMVEC monolayers (17). VE-cadherin has been considered the major cadherin in endothelial cells from most other vascular beds and was the major cadherin present in RPAEC (4). Although RPMVEC bind distinctly different lectins than RPAEC, both phenotypes retain surface marker characteristics of endothelial cells such as acetylated LDL uptake, factor VIII expression, and the absence of immunostaining for smooth muscle actin (11). In addition to the lower baseline permeability to water and solutes, RPMVEC have an attenuated responsiveness to inflammatory mediators compared with RPAEC. Kelly et al. (9) reported significantly lower permeabilities to different-sized dextrans diffusing across monolayers of RPMVEC compared with RPAEC both at baseline and when exposed to thapsigargin, an irreversible inhibitor of intracellular Ca2+ uptake and opener of store-operated Ca2+ channels. The baseline lower Lp values reported for the first time in the present study for RPMVEC monolayers are compatible with tighter junctions and lower diffusional permeability reported by Kelly et al. (9). A higher turnover rate for intracellular cAMP also could contribute to the lower baseline permeability of RPMVEC to fluid and solutes (3). Recent reports suggest that RPMVEC lack a Ca2+-specific, store-operated channel, which could account for the apparent uncoupling of barrier permeability from intracellular Ca2+ increases (2, 21, 28). However, the reduced Lp responses to addition of 5% albumin, GdCl3, or drugs that increase intracellular cAMP were similar to the responses predicted from other cultured and in situ endothelial phenotypes and were present in both lung endothelial phenotypes (14, 15, 20).
All epithelial and endothelial monolayers "seal" to some extent with time when first exposed to pressure, and all of these experiments were begun by sealing the monolayers at 15–20 cmH2O. Both endothelial phenotypes exhibited this phenomenon (Fig. 1A). This effect is attributed to a passive compression of glycosaminoglycan and proteoglycan elements of the basement membrane, junctional pathways, and intercellular portions of the glycocalyx, in addition to active recruitment of zonula occludens-1 (ZO-1) to the tight junctions (6, 18). After initial sealing, a steady-state filtration rate is attained that is proportional to the applied pressure, and this relationship is a prerequisite for a valid calculation of Lp using a least-squares regression. These conditions were met for monolayers of both phenotypes for the first few days after confluence was reached and by RPAEC for longer periods in culture. An interesting feature of the older RPMVEC monolayers was a further "sealing" effect at progressively higher pressures in monolayers when used after more than 4–5 days after confluence was reached. That is, steady-state Jv/A was actually decreased by each pressure increase so that the conventional method for calculation of Lp could not be employed. Most previous studies of Lp in endothelial monolayers have used conduit vessel endothelial cells, and pressure-induced sealing has not been observed in conduit vessel studies or in single-capillary measurements of Lp (13). The continued growth of the RPMVEC monolayers in culture, continued increases in monolayer electrical resistances, and decreases in Lp estimates suggested that changes in junctional structure such as more extensive overlap of cell edges could account for these observed differences between RPMVEC and RPAEC. However, morphologic measurements revealed that junction lengths were not significantly different between phenotypes (Fig. 7). Rather, the RPMVEC escaped contact inhibition of growth, and the overgrowth formed superimposed monolayers. Although growth was strongly stimulated by the 20% FBS in the medium, formation of multiple layers is not observed in normal lung capillaries. The overgrowth of stimulated RPMVEC may represent a condition of exuberant endothelial growth observed in the obstructive arteriolar lesions of primary pulmonary hypertension. In this condition, the obliteration of small precapillary pulmonary arteries by proliferating endothelial cells produces fatal pulmonary hypertension (26). Hydrostatic compression of multiple cell layers could also block and compress paracellular pathway dimensions sufficient to restrict fluid movement and produce the lower Jv/A observed as pressure increased in the overgrown RPMVEC filters.
The responses to a sustained pressure gradient over time also were considerably different between endothelial phenotypes. Tarbell et al. (23) reported a 350% increase in Lp for bovine aortic endothelial cell monolayers subjected to a sustained pressure of 30 cmH2O for 4 h that was comparable to the 225% increase in Lp we observed at 2 h for the RPAEC monolayers under 40 cmH2O of sustained pressure. They proposed that the shear stress effect of filtration through junctional clefts increased junctional permeability to water flow during continuous filtration. In other studies, the Lp of bovine aortic endothelial cell monolayers was increased by surface laminar shear stress that was accompanied by a decrease in junctional ZO-1 (5). In contrast to the pressure-induced increased Lp in RPAEC, we found that the Lp increase in RPMVEC monolayers did not reach significance during the 2-h period of a sustained transmural pressure gradient. The failure of RPMVEC to respond to pressure stress supports previous observations that, unlike RPAEC, the RPMVEC do not realign under sustained shear stress or cyclical stretch (unpublished data). Tarbell et al. (23) could reverse the Lp increase due to sustained hydrostatic pressure by adding 1 mM dibutyryl cAMP, a protein kinase A agonist. We also observed that both lung endothelial phenotypes responded by a significant reduction in Lp to the combination of isoproterenol and rolipram to increase intracellular cAMP. Thus the pressure-induced increase in Lp caused by junctional filtration shear appears to be a property of conduit vessel endothelial cells that is attenuated in RPMVEC monolayers.
A low alveolar capillary Lp also would have important implications for lung fluid homeostasis. To efficiently exchange respiratory gases by diffusion, the alveolar capillary endothelium has a surface area in humans that may exceed 100 m2, and thickness of the alveolar capillary barrier is only 0.35–1.0 µm over much of this area (25). However, this architecture also facilitates fluid filtration, which in turn could interfere with gas exchange if alveoli flood with fluid. Among the adjustments in Starling tissue forces that contribute to the Starling "safety factors" against edema, an increase in lymph flow is considered to contribute only a few millimeters of Hg of protection against increased hydrostatic pressure (24). To the traditional Starling safety factors we can now add a capillary endothelial barrier that is at least 20 times more restrictive to fluid transport than previously surmised. Considering the magnitude of the measured Lp differences between RPMVEC and RPAEC, the contribution of such a low Lp for preventing edema formation across the pulmonary capillaries would quantitatively exceed the passive contribution of adjustments in all other Starling forces, because the alveolar capillaries would require at least 22-fold greater hydrostatic pressure than that required for an equivalent filtration across a comparable surface area of arterial segments (24). This low microvascular Lp could be a determinant in the well-known pattern of early hydrostatic edema formation in lungs where edema first accumulates in perivascular cuffs around small conduit vessels, which we have shown to have a considerably higher Lp (24).
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
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. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Stevens, S. Phan, M. G. Frid, D. Alvarez, E. Herzog, and K. R. Stenmark Lung Vascular Cell Heterogeneity: Endothelium, Smooth Muscle, and Fibroblasts Proceedings of the ATS, September 15, 2008; 5(7): 783 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Solodushko, J. C. Parker, and B. Fouty Pulmonary Microvascular Endothelial Cells Form a Tighter Monolayer when Grown in Chronic Hypoxia Am. J. Respir. Cell Mol. Biol., April 1, 2008; 38(4): 491 - 497. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Clark, D. F. Alvarez, M. Alexeyev, J. A. C. King, L. Huang, M. C. Yoder, and T. Stevens Regulatory role for nucleosome assembly protein-1 in the proliferative and vasculogenic phenotype of pulmonary endothelium Am J Physiol Lung Cell Mol Physiol, March 1, 2008; 294(3): L431 - L439. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. F. Alvarez, L. Huang, J. A. King, M. K. ElZarrad, M. C. Yoder, and T. Stevens Lung microvascular endothelium is enriched with progenitor cells that exhibit vasculogenic capacity Am J Physiol Lung Cell Mol Physiol, March 1, 2008; 294(3): L419 - L430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Shasby Cell-cell adhesion in lung endothelium Am J Physiol Lung Cell Mol Physiol, March 1, 2007; 292(3): L593 - L607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. C. Aird Phenotypic Heterogeneity of the Endothelium: II. Representative Vascular Beds Circ. Res., February 2, 2007; 100(2): 174 - 190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Parker Hydraulic conductance of lung endothelial phenotypes and Starling safety factors against edema Am J Physiol Lung Cell Mol Physiol, February 1, 2007; 292(2): L378 - L380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Quadri and J. Bhattacharya Resealing of endothelial junctions by focal adhesion kinase Am J Physiol Lung Cell Mol Physiol, January 1, 2007; 292(1): L334 - L342. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
; n = 11) and RPMVEC (
; n = 11) 2–3 days postconfluence used to calculate mean Lp values, and older (
; n = 10).
