Hydraulic conductance of lung endothelial phenotypes and Starling safety factors against edema

Recent permeability studies comparing endothelial cell phenotypes derived from alveolar and extra-alveolar vessels have significant implications for interpreting the mechanisms of fluid homeostasis in the intact lung. These studies indicate that confluent monolayers of rat pulmonary microvascular endothelial cells had a hydraulic conductance (L(p)) that was only 5% and a transendothelial flux rate for 72-kDa dextran only 9% of values determined for rat pulmonary artery endothelial cell monolayers. On the basis of previous studies partitioning the filtration coefficients between alveolar and extra-alveolar vascular segments in rat lungs and previous studies of lymph albumin fluxes and permeability, the contribution of the alveolar capillary segment to total albumin flux in lymph was estimated to be less than 10%. In addition, the Starling safety factors against the edema calculated for the alveolar capillaries are quite different from those estimated for whole lung. Estimates of the edema safety factor due to increased filtration across the alveolar capillary wall based on the low L(p) indicate it is quantitatively the greatest safety factor, although it would be a minor safety factor for extra-alveolar vessels. Also, a markedly higher effective protein osmotic absorptive force for plasma proteins must occur in the capillaries relative to extra-alveolar vessels. The lower L(p) for alveolar capillaries also has implications for the sequence of hydrostatic edema formation, and it also may have a role in preventing exercise-induced alveolar flooding.     

Filtration profile in isolated zone 1 and zone 3 dog lungs at constant high alveolar pressure

We measured the rate of liquid filtration in isolated dog lung lobes inflated to a constant alveolar pressure of 25 cmH2O and with all open vessels filled with plasma. We measured lung weight gain at vascular pressures ranging from 5 to 40 cmH2O relative to pleural pressure. We confirmed that under zone 1 conditions the "arterial" and "venous" extra-alveolar segments have essentially the same filtration characteristics. Using the combined extra-alveolar vascular system, we determined when recruitment of filtration surface area occurred as we increased vascular pressure from 0 to 40 cmH2O. Based on an abrupt increase in filtration rate as vascular pressure approached the zone 1/3 boundary, we infer that a sudden recruitment of exchange surface area occurred at that point. Based on the slopes of the zone 1 and zone 3 filtration profiles, we conclude that extra-alveolar vascular segments contribute approximately 25% of total to filtration in the lung under zone 3 conditions, although the exact vessels filtering under zone 1 conditions have yet to be determined. Our analysis of the data supports the concept that there is a difference in the perimicrovascular pressure around alveolar and extra-alveolar vessels, which in part may account for the apparent high filtration fraction apportioned to extra-alveolar vessels.     

Hypoxia and angiotensin II infusion redistribute lung blood flow in lambs

To assess the effects of alveolar hypoxia and angiotensin II infusion on distribution of blood flow to the lung we performed perfusion lung scans on anesthetized mechanically ventilated lambs. Scans were obtained by injecting 1-2 mCi of technetium-labeled albumin macroaggregates as the lambs were ventilated with air, with 10-14% O2 in N2, or with air while receiving angiotensin II intravenously. We found that both alveolar hypoxia and infusion of angiotensin II increased pulmonary vascular resistance and redistributed blood flow from the mid and lower lung regions towards the upper posterior region of the lung. We assessed the effects of angiotensin II infusion on filtration pressure in six lambs by measuring the rate of lung lymph flow and the protein concentration of samples of lung lymph. We found that angiotensin II infusion increased pulmonary arterial pressure 50%, lung lymph flow 90%, and decreased the concentration of protein in lymph relative to plasma. These results are identical to those seen when filtration pressure increases during alveolar hypoxia. We conclude that alveolar hypoxia and angiotensin II infusion both increase fluid filtration in the lung by increasing filtration pressure. The increase in filtration pressure may be the result of a redistribution of blood flow in the lung with relative overperfusion of vessels in some areas and transmission of the elevated pulmonary arterial pressure to fluid-exchanging sites in those vessels.     

Effect of outflow pressure on lung lymph flow in unanesthetized sheep

Studies in anesthetized animals have shown that the flow rate from lung lymphatics (QL) depends on the pressure at the outflow end of the vessels (Po). We tested this in unanesthetized sheep prepared with chronic lung lymph cannula. We measured QL with the lymph cannula held at various heights above the olecranon and calculated Po as the height + QL X cannula resistance. QL decreased with increases in Po (delta QL/delta Po = -8.2 +/- 6.4 microliter X min-1 X cmH2O-1, mean +/- SD). We increased QL by raising left atrial pressure or infusing Ringer solution or Escherichia coli endotoxin and found that QL was even more sensitive to Po (delta QL/delta Po = -32 +/- 22). Cannula resistance caused a 9-70% reduction in QL. Changes in QL caused by increasing Po were not associated with changes in lymph protein concentration for up to 330 min. This indicates that increases in Po shunt lymph away from cannulated vessels but do not substantially effect microvascular filtration rate. The shunted lymph may flow into other vessels or collect in the lung. We conclude that QL does not accurately represent microvascular filtration rate because it depends on the cannula resistance and position at which the investigator chooses to place the cannula.     

Starling forces that oppose filtration after tissue oncotic pressure is increased

We tested the hypothesis that the effective oncotic force that opposes fluid filtration across the microvessel wall is the local oncotic pressure difference across the endothelial surface glycocalyx and not the global difference between the plasma and tissue. In single frog mesenteric microvessels perfused and superfused with solutions containing 50 mg/ml albumin, the effective oncotic pressure exerted across the microvessel wall was not significantly different from that measured when the perfusate alone contained albumin at 50 mg/ml. Measurements were made during transient and steady-state filtration at capillary pressures between 10 and 35 cmH(2)O. A cellular-level model of coupled water and solute flows in the interendothelial cleft showed water flux through small breaks in the junctional strand limited back diffusion of albumin into the protected space on the tissue side of the glycocalyx. Thus oncotic forces opposing filtration are larger than those estimated from blood-to-tissue protein concentration differences, and transcapillary fluid flux is smaller than estimated from global differences in oncotic and hydrostatic pressures.     

Bicarbonate uptake by basolateral membrane vesicles from rat jejunum

Basolateral membranes from rat jejunal enterocytes have been obtained by self-orienting Percoll-gradient centrifugation. Bicarbonate and L-glucose uptake into osmotically active basolateral membrane vesicles has been studied by a rapid filtration technique. In closed vessels and at pH 8.2 the uptake kinetics of both [14C]bicarbonate and L[3H]glucose have been followed for 30 min at 18 degrees C. Bicarbonate uptake seems to be fast and in efflux experiments SITS and DIDS effect is negligible. This work demonstrates that it is possible to determine bicarbonate flux across basolateral membrane vesicles at pH and temperature values close to usual experimental conditions.     

Analysis of the Components of Ionic Flux across a Membrane

The unidirectional flux of an ionic species may occur because of several mechanisms such as active transport, passive diffusion, exchange diffusion, etc. The contribution of such mechanisms to the total unidirectional flux across a membrane cannot be determined by only measuring that flux. It is shown that if the pertinent experimental data (the opposite unidirectional fluxes and the composite phenomenological resistance coefficient of the ionic species for a given electrochemical potential difference) obey a certain inequality, then the parameters of a model consisting of parallel, independent, active transport, and passive processes may be determined. Although the existence of "additional" processes including exchange diffusion, single-file pore diffusion, isotope interaction, etc. is not disproved, their existence is unnecessary if the inequality is satisfied. Two types of violations of the inequality may occur: (a) if the upper limit is disobeyed the presence of another substance contributing to the measured resistance and/or a constant affinity of the active transport process may be indicated; (b) if the lower limit is disobeyed it is necessary to postulate the existence of an additional process. For the latter type of violation, exchange diffusion is chosen as an example. Methods are given for determining the contribution of exchange diffusion, active transport, and passive diffusion to the unidirectional flux for some special cases.     

Pulmonary fluid balance following pulmocutaneous baroreceptor denervation in the toad

Pulmonary hemodynamics and net transcapillary fluid flux (NTFF) were measured in conscious toads before and following bilateral denervation of the recurrent laryngeal nerves (rLN), which contain afferents from baroreceptors located in the pulmocutaneous arteries. Denervation caused an acute doubling of the arterial-venous pressure gradient across the lung and a threefold increase in pulmonary blood flow. Calculated pulmonary vascular resistance fell and remained below control values through the period of experimentation. NTFF increased by an order of magnitude (0.74-7.77 ml X kg-1 X min-1), as filtration increased in response to the hemodynamic changes caused by rLN denervation. There was a better correlation between NTFF and pulmonary blood flow than between NTFF and pulmonary driving pressure. Our results support the view that tonic neural input from pulmocutaneous baroreceptors protects the anuran lung from edema by restraining pulmonary driving pressure and blood flow and perhaps by reflexly maintaining vascular tone in the extrinsic pulmonary artery, therefore tending to increase the pre-to-postpulmonary capillary resistance ratio and biasing the Starling relationship in the pulmonary capillaries against filtration.     

Studies on the filtration properties of isolated renal basement membranes.

1. The filtration properties of films of renal basement membrane were studied in vitro using pressure filtration chambers. 2. Retention of cytochrome c by the films was found to be dependent upon the filtration pressure indicating that it was transferred across the films by convective as well as diffusive flow. In contrast, serum albumin was transferred by diffusive movement only. 3. When solutions containing both cytochrome c and IgG were filtered it was found that increasing the filtration pressure reduced the flux of cytochrome c across the films. A similar phenomenon occurred when serum was filtered, less protein passed through the films at high filtration pressures. These phenomena are explained by concentration-polarisation effects. 4. The flux of cytochrome c through the films was found to decrease in a non-linear manner as the films thickness was increased. With thin films, the flux of cytochrome c increased in a non-linear manner as the concentration of the protein in the overstanding solution was increased. With thicker films the flux was linearly dependent on concentration. These findings are interpreted as supporting the view that movement of cytochrome c occurs, at least in part, by convective flow.     

Indirect estimate of perimicrovascular pressure rise in edema in isolated zone 1 lung

To determine how liquid accumulation affects extra-alveolar perimicrovascular interstitial pressure, we measured filtration rate under zone 1 conditions (25 cmH2O alveolar pressure, 20 or 10 cmH2O vascular pressure) in isolated dog lung lobes in which all vessels were filled with autologous plasma. In the base-line condition, starting with normal extra-alveolar water content, filtration rate decreased by about one-half over 1 h as edema liquid slowly accumulated. We repeated each experiment after inducing edema (up to 100% lung weight gain). The absolute values and time course of filtration in the edema condition did not differ from base-line, i.e., the edema did not affect the time course of filtration. To compute the maximal initial and maximal change in extra-alveolar perimicrovascular pressure that occurred over each 1-h filtration study, we first assumed that the reflection coefficient is 0 in the Starling equation, then calculated perimicrovascular pressure and filtration coefficient from two equations with two unknowns. The mean filtration coefficient in 10 lobes is 0.063 g/(min X cmH2O X 100 g wet wt), and the initial perimicrovascular pressure is 3.9 cmH2O, rising by 4-7 cmH2O at 1 h. Finally we tested low protein perfusates and found the filtration rate was higher. We calculated an overall reflection coefficient = 0.44, a decrease in the initial perimicrovascular pressure to 1.9 cmH2O and a slightly lower increase after 1 h of edema formation, 2.2-6.6 cmH2O.     

Role of t-tubules in the control of trans-sarcolemmal ion flux and intracellular Ca2+ in a model of the rat cardiac ventricular myocyte

The t-tubules of mammalian ventricular myocytes are invaginations of the surface membrane that form a complex network within the cell, with restricted diffusion to the bulk extracellular space. The trans-sarcolemmal flux of many ions, including Ca(2+), occurs predominantly across the t-tubule membrane and thus into and out of this restricted diffusion space. It seems possible, therefore, that ion concentration changes may occur in the t-tubule lumen, which would alter ion flux across the t-tubule membrane. We have used a computer model of the ventricular myocyte, incorporating a t-tubule compartment and experimentally determined values for diffusion between the t-tubule lumen and bulk extracellular space, and ion fluxes across the t-tubule membrane, to investigate this possibility. The results show that influx and efflux of different ion species across the t-tubule membrane are similar, but not equal. Changes of ion concentration can therefore occur close to the t-tubular membrane, thereby altering trans-sarcolemmal ion flux and thus cell function, although such changes are reduced by diffusion to the bulk extracellular space. Slowing diffusion results in larger changes in luminal ion concentrations. These results provide a deeper understanding of the role of the t-tubules in normal cell function, and are a basis for understanding the changes that occur in heart failure as a result of changes in t-tubule structure and ion fluxes.     

pH-regulated chloride secretion in fetal lung epithelia

The fetal lung actively transports chloride across the airway epithelium. ClC-2, a pH-activated chloride channel, is highly expressed in the fetal lung and is located on the apical surface of the developing respiratory epithelium. Our goal was to determine whether acidic pH could stimulate chloride secretion in fetal rat distal lung epithelial cells mounted in Ussing chambers. A series of acidic solutions stimulated equivalent short-circuit current (I(eq)) from a baseline of 28 +/- 4.8 (pH 7.4) to 70 +/- 5 (pH 6.2), 114 +/- 12.8 (pH 5.0), and 164 +/- 19.2 (pH 3.8) microA/cm(2). These changes in I(eq) were inhibited by 1 mM cadmium chloride and did not result in large changes in [(3)H]mannitol paracellular flux. Immunofluorescent detection by confocal microscopy revealed that ClC-2 is expressed along the luminal surface of polarized fetal distal lung epithelial cells. These data suggest that the acidic environment of the fetal lung fluid could activate chloride channels contributing to fetal lung fluid production and that the changes in I(eq) seen in these Ussing studies may be due to stimulation of ClC-2.     

Albumin transport across pulmonary capillary-interstitial barrier in anesthetized dogs

To evaluate albumin transport across the pulmonary capillary endothelial and interstitial barriers, we simultaneously measured blood-to-tissue (QA,t) and blood-to-lymph (QA,l) clearances of 125I-radiolabeled albumin as well as endogenous albumin clearance (Qa,l) in the canine lung in vivo (n = 10). Steady-state prenodal lung lymph flows (Qw,l) and protein clearances were measured over a 2-h period at a constant capillary pressure (Pc, 13-33 cmH2O). Comparison between QA,t and QA,l as a function of Pc suggests that little of the albumin that crossed the capillary wall remained in the lung tissue, with most leaving in the lymph. Qw,l increased significantly as Pc increased, but lung tissue water was minimally affected. From the ratio of the clearance-Pc slopes for albumin and water, the albumin reflection coefficient was estimated to be 0.81 using QA,l and Qw,l and 0.56 using Qa,l and Qw,l. The permeability surface area product for the sum of blood-to-tissue and blood-to-lymph fluxes of labeled albumin (QA,t + QA,l) was 31 +/- 9 microliters/min, whereas that calculated from the blood-to-lymph flux of endogenous albumin (Qa,l) was 97 +/- 22 microliters/min. These data suggest that 1) both tissue and lymph accumulations of albumin must be considered when microvascular permeability is evaluated using protein tracers; 2) lymph clearance, but not tissue accumulation of albumin, was filtration dependent; and 3) lymph flow was an important contributor to the safety factor against edema formation over a moderate range of capillary pressures.     

Oxidant-increased endothelial permeability: prevention with phosphodiesterase inhibition vs. cAMP production.

The present objective was to determine whether hydrogen peroxide (H(2)O(2)) increases transvascular albumin clearance and lung weight in an isolated rat lung and whether posttreatment with cAMP-enhancing agents can prevent these increases. Transvascular albumin clearance was assessed by (125)I-labeled albumin clearance ((125)I-albumin flux/perfusate concentration of (125)I-albumin) at a given fluid filtration. Nonlinear regression analysis of transvascular albumin clearance vs. fluid filtration yielded values for the permeability-surface area product (PS) and the reflection coefficient (sigma). H(2)O(2) decreased sigma from a control value of 0.93 to 0.38, did not change PS, and increased lung weight. Posttreatment with isoproterenol, a beta(2)-adrenergic-receptor agonist, reduced the H(2)O(2)-induced decrease in sigma to 0.65 and augmented the increase in lung weight. Posttreatment with CP-80633, a phosphodiesterase 4 inhibitor, further reduced the H(2)O(2)-induced decrease in sigma to 0.79 and blocked the rise in lung weight. In the presence of isoproterenol or CP-80633, H(2)O(2) increased PS. Therefore, H(2)O(2) increased the convective and diffusive clearances of albumin across an intact pulmonary vasculature. Furthermore, inhibition of cAMP metabolism more effectively attenuated the H(2)O(2)-induced increases in convective albumin clearance and lung weight as compared with stimulation of cAMP production.     

New insights into saccular development and vascular formation in lung allografts under the renal capsule

The study of distal lung morphogenesis and vascular development would be greatly facilitated by an in vitro or ex vivo experimental model. In this study we show that the growth of mouse embryonic day 12.5 lung rudiments implanted underneath the kidney capsules of syngeneic or immunodeficient hosts follows closely lung development in utero. The epithelium develops extensively with both proximal and distal differentiation to the saccular stage. The vasculature also develops extensively. Large blood vessels accompany large airways and capillaries develop within the saccular walls. Interestingly, vessels in the lung grafts develop from endothelial progenitor cells endogenous to the explants and host vessels do not vascularize the grafts independently. This suggests that embryonic lungs possess mechanisms to prevent the inappropriate ingrowth of surrounding vessels. However, vessels in the lung grafts do connect to host vessels, showing that embryonic lungs have the ability to stimulate host angiogenesis and recruit host vessel connections. These data support the hypothesis that the lung vasculature develops by both vasculogenic and angiogenic processes: a vascular network develops in situ in lung mesenchyme, which is then connected to angiogenic processes from central vessels. The lung renal capsule allograft is thus an excellent model to study the development of the pulmonary vasculature and of late fetal lung development that requires a functional blood supply.     

Evaluating pulmonary vascular permeability with radiolabeled proteins: an error analysis

Using techniques of mathematical simulation, we compared two methods of evaluating pulmonary vascular permeability, i.e., transvascular protein flux. Both methods calculate a transport rate constant [pulmonary transcapillary escape rate (PTCER)] after making external radiation detection measurements of an intravenously administered radiolabeled protein. With one method, lung tissue time-activity data are acquired by positron emission tomography (PET) and are interpreted with a two-compartment model. With the other method, the time-activity data are acquired with simple detector probes and are interpreted by linear regression after normalizing for various physical factors (slope-intercept or SI method). The results show that significant errors in calculating PTCER can result from using the SI method, because it ignores the effects of back-flux on the tissue time-activity measurements. Both methods produce errors if the data analysis includes activity from vascular volumes not involved in tracer exchange with the extravascular compartment. Significant errors can also occur with the PET method, particularly when permeability is nearly normal, if pulmonary vascular volume changes significantly during the period of data collection. On balance, the PET method appears to be the method of choice for accurately evaluating pulmonary vascular permeability by protein flux measurements, although both methods may be useful in clinical applications.     

Pulmonary transport of water and solutes: functional and structural correlations

Assessment of endothelial barrier parameters is based on formulations derived from the thermodynamics of irreversible processes. These formulations contain simplifying assumptions that, with uncertainties concerning the surface area involved in the transport, preclude definitive modeling with respect to the possible pathways. In the development of edema in isolated lung preparations perfused at 37 C, changes of structural features such as increases of vesicle volume density can be associated with changes of barrier parameters such as increased filtration coefficients and increased permeability coefficients to sodium ion. However, that edema can develop at 15 C without increased vesiculation and that cuffing of extraalveolar vessels also occurs in vivo with circulatory overload without increased vesiculation do not suggest that vesicles play a determining role in the development of edema. Available evidence indicates that the exchange of water across the endothelium is at a rate approximating the flow rate of blood and therefore must involve the entire endothelial surface. Separation or noncommunality of pathways for water and solutes makes modeling of the pathways from current formations even less certain.     

Bayesian Estimation of the Probability of Asbestos Exposure from Lung Fiber Counts

Summary Asbestos exposure is a well‐known risk factor for various lung diseases, and when they occur, workmen's compensation boards need to make decisions concerning the probability the cause is work related. In the absence of a definitive work history, measures of short and long asbestos fibers as well as counts of asbestos bodies in the lung can be used as diagnostic tests for asbestos exposure. Typically, data from one or more lung samples are available to estimate the probability of asbestos exposure, often by comparing the values with those from a reference nonexposed population. As there is no gold standard measure, we explore a variety of latent class models that take into account the mixed discrete/continuous nature of the data, that each subject may provide data from more than one lung sample, and that the within‐subject results across different samples may be correlated. Our methods can be useful to compensation boards in providing individual level probabilities of exposure based on available data, to researchers who are studying the test properties for the various measures used in this area, and more generally, to other test situations with similar data structure.     

Separation of yeast by cross-flow filtration with backwashing

The following results were obtained at the separation of yeast by cross-flow filtration from fermentation broth or yeast suspensions with or without backwashing by the filtrate, using micro-porous membranes: 1) the first stage of the filtration process was described by the standard blocking filtration model, 2) the mean filtration flux in one filtration cycle was kept almost constant for at least 3 h by backwashing, 3) the mean filtration flux with backwashing was roughly estimated from the results of filtration without backwashing, and 4) the mean filtration flux at certain yeast concentration in concentrating the yeast was not so much affected by the previous concentration path, and almost agreed that in the filtration of a constant concentration at the same concentration.