Protein Kinase C Activation Does not Stimulate Lung Liquid Clearance in Anesthetized Sheep1-3
YVES BERTHIAUME,4 MARIUSZ SAPIJASZKO, JOYCE MACKENZIE, and MICHAEL P. WALSH5 With the technical assistance of Derrice Payne
A
Introduction
substantial body of evidence suggests that resolution of pulmonary edema may depend in part on the active transport of Na" from the air spaces of the lung to the interstitial space and, subsequently, to the vascular space. This evidence was obtained in many different experimental models. First, it is known that lung liquid clearance can be stimulated by beta-adrenergic agonists, and this increase in lung liquid clearance can be inhibited byamiloride, a sodium transport inhibitor (1). Sodium transport has also been clearly demonstrated in isolated perfused lungs (2) as well as in cultured alveolar type II cells (3). This Na' transport can be inhibited by amiloride (2, 3). Beta-adrenergic stimulation also increased Na" transport and lung liquid clearance in the isolated lung preparation (2,4) as well as dome formation (5) and short-circuit current (6) in cultured alveolar type II cells. Recently, some preliminary experimental data has suggested that Na" transport across the alveolar epithelium could occur either through a Na" channel (7, 8) or a nonspecific cation channel (9).Finally, recent human studies support the hypothesis that active ion transport across the alveolar epithelium is the primary mechanism for clearance of edema fluid from the air spaces of the lung (10). Although this evidence clearly supports the role of active transport of Na' in the resolution of pulmonary edema, the mechanisms regulating Na" transport are still not clearly established. Twomajor intracellular second messengers have been implicated in the regulation of ion transport. Adenosine 3',5'cyclic monophosphate (cAMP) is well known to stimulate chloride transport in the airway epithelium (11) as well as Na" transport in epithelial cells derived from the kidney (12). Recently, cAMP has also been shown to playa significant role in stimulation of Na" transport in lung liq-
SUMMARY Although active transport of ions could play an important role in the resolution process of pulmonary edema, the exact mechanism regulating this process is stili unknown. In this study, we investigated the effect of phorbol myristate acetate (PMA) on lung liquid clearance in anesthetized, ventilated sheep to evaluate the possible role of protein kinase C. To study lung liquid and protein clearance, we measured the removal of 100 ml of autologous serum from the air spaces of anesthetized sheep. Either serum alone or serum mixed with PMA (10-7 M) was instilled. After 4 h, the residual lung water was 76.8 ± 9.2 ml when serum alone was instilled and 79.5 ± 15.7 when serum with PMA (10-7 M) was instilled. The lack of effect of PMA (10-7 M) on lung liquid clearance cannot be explained by increased movement of liqUid from the vascular space to the air space since we did not have any evidence of increased pressure or increased permeability In the lung. This lack of effect of PMA (10-7 M) is not due to an absence of stimulation of protein kinase C since instillation of BSA and PMA (10-7 M) in rat lung produced a translocation of protein kinase C activity from the cytosolic fraction to the membrane fraction 2 h after the Instillation. These results were confirmed In two sheep experiments, which demonstrated clear activation of protein kinase C after 4 h. These data suggest that activation of protein kinase C does not stimulate lung liqUid clearance. However, a possible role of protein kinase C in modulating lung liquid clearance has not been excluded. AM REV RESPIR DIS 1991; 144:1085-1090
uid clearance. cAMP analogs and betaadrenergic agonists were shown to increase both dome formation (5) and short-circuit currents across a monolayer of alveolar epithelial type II cells (6, 13)by increasing sodium transport from the apical side to the serosal side (6). These effects of cAMP analogs or betaadrenergic agonist are partially inhibited by amiloride (6, 13).Moreover, cAMP analogs increase Na" transport (14) and liquid clearance (4) in isolated rat lungs. Recently, we also demonstrated that cAMP in combination with aminophylline can stimulate lung liquid clearance in sheep (15). Another second messenger system that has a significant role in chloride transport or Na" transport is protein kinase C. First, protein kinase C can modulate the activity of proteins involved in ion transport. For example, protein kinase C can activate or inactivate chloride channels depending on Ca2+ concentration (16). The sodium potassium ATPase pump is also an effector protein of protein kinase C (17). Protein kinase C activation also can influence ion transport. For example, protein kinase C stimulation with 12-0-tetradecanoylphorbol-13-
acetate (TPA) and phorboI12,13-dibutyrate enhances Na" entry into 3T3 cells (18). TPA also stimulates short-circuit currents in frog skin, probably through stimulation of a Na' channel (19). In contrast, activation of protein kinase C inhibits apical sodium transport in A 6 epithelial cells (20) as well as sodium and potassium transport in the rabbit cortical collecting tubule (21). Because activation of protein kinase
(Received in original form January 17, 1991 and in revised form June 14, 1991) 1 From the Departments of Medicine and Medical Biochemistry, University of Calgary, Calgary, Alberta, Canada. 2 Supported by Grants-in-Aid MA-10273-YB and MT-9097-MPW from The Medical Research Council of Canada. 3 Correspondence and requests for reprints should be addressed to Dr. Yves Berthiaume, Department of Medicine, University of Calgary, 3330 Hospital Drive NW., Calgary, Alberta T2N 4NI Canada. 4 Clinical Investigator of the Alberta Heritage Foundation for Medical Research and a Scholar of The Medical Research Council of Canada. 5 Recipient of a Medical Research Council of Canada Scientist Award and an Alberta Heritage Foundation for Medical Research Scholarship.
1085
1086
C through its effect on ion transport could have an important effect on lung liquid clearance, we decided to evaluate if the activation of protein kinase C with phorbol myristate acetate (PMA), a wellknown activator of the enzyme (22), can stimulate lung liquid clearance in anesthetized sheep. Methods Animal Preparation and Serum Instillation Sheep surgicalpreparation and serum instillation. Twenty yearling sheep were anesthetized intravenously with pentothal (30 mg/ kg). A tracheotomy was done, and a cuffed tracheotomy tube was inserted. The lungs were ventilated with a constant-volume ventilator (Harvard Apparatus Co. Inc., South Natick, MA) at a tidal volume of 13to 15ml/kg with a peak inspiratory pressure of 15 to 20 em H 20. The respiratory rate was adjusted to achieve a Paco, between 30 and 40 mm Hg. Anesthesia was maintained by 0.8 to 1.50/0 isofluorane mixed with supplemental oxygen (40 to 60%). The sheep were paralyzed intravenously with pancuronium bromide (1.0mg) at the time of the thoracotomy and just before the bronchoscopy. The preparation of the anesthetized sheep to measure systemic and pulmonary hemodynamics was similar to the one previously published (1).After surgery, the sheep were placed in the prone position, with head and thorax tilted up at a 5-degree angle to maintain the lower lobes in a slightly dependent position. The procedures for preparation and instillation of the serum solution are essentially the same as we have previously published (1). Briefly, on the day before the experiment, 250 ml of blood was withdrawn from the sheep. The blood was clotted and centrifuged to obtain 125 ml of serum, which was refrigerated overnight. The next morning, 30 mg of anhydrous Evans blue was added to bind to the albumin (1). 1251-labeled human serum albumin (20 Ci) (Frosst, Division of Merck Frosst Canada Inc., Kirkland, Quebec, Canada) was added to the serum as a protein tracer. Before instilling the serum, a sample of the instillate was taken for total protein measurement and wet/dry ratio measurements so that the dry weight could be subtracted from the final lung water calculations. Serum (100ml) was instilled into the right lower lobe over 5 to 10 min through a flexible fiberoptic bronchoscope (Olympus Corp., Lake Success, NY) using a three-way 'l-piece adapter. The ventilation was continued during the serum instillation. Rat surgicalpreparation and BSA instillation. Although PMA is a well-known activator of protein kinase C (22), we wanted to establish clearly that PMA at the dose used can stimulate protein kinase C in the lung. To study this question we used a rat model simulating our sheep model. The rats were anesthetized with halothane
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
(2 to 3%) mixed with 100070 oxygen. A tracheostomy was done and, using a rodent feeding tube, 0.2 to 0.4 ml of bovine serum albumin (5%) was injected in the right lower lobe. The wound was closed with Vetbound tissue adhesive (Animal Care Products; 3M, St. Paul, MN). The animal was then allowed to recover from anesthesia. The animal would be up and active in about 5 min after the end of the procedure. It was not necessary to exclude any animals from the study because of cross contamination in control lungs.
of 20 em H 20, and the lungs were removed from the thorax. Airway liquid (2 to 5 ml) was obtained by passing a catheter 3 mm in diameter into the distal airways of the seruminstilled lung. The sample was centrifuged, and the total protein concentration and radioactivity (counts per milliliter) of the supernatant were measured. Then the lungs were frozen in liquid nitrogen. The next day, each lung was examined in a cryostat at - 20 0 C to determine the distribution of the Evansblue-labeled serum and to confirm that none of the fluid had spilled over into the control lung. General Protocol Each lung was then homogenized separateThree experimental groups of sheep were studly, and the extravascular lung water was deied using the following general protocol. After surgery there was a baseline period of 1.5 termined by calculating the wet/dry weight to 2 h during which vascular pressure was sta- ratio (grams H 20/gram dry lung) (1). The exble. The serum solution with or without PMA cess water in the experimental lung was calwas then instilled. During the next 4 h, sys- culated with the same equation as described temic pulmonary and left atrial pressure and previously (1, 24). The coefficient of variation for this technique is 11.5% (1). cardiac output were measured every 15 min. Alveolar permeability. Because PMA is Blood samples were taken every 0.5 h for towell known to induce lung injury, we evalutal protein concentration measurements and 1251-albumin counts. After 4 h, the sheep were ated the permeability of the alveolar barrier as measured by the clearance of protein from killed, and the lungs were removed for samthe lung. For assessment of protein clearance, pling of the airway liquid to obtain the final 20 IlCi of 1251-human serum albumin was addairway liquid protein concentration, measureed to the instilled serum. The method for asment of the excess lung water to assess lung liquid clearance, and quantification of the 1251_ sessing the total amount of tracer recovered in homogenized lung and plasma was similar albumin left in the lung (see Measurements). to the method previously published (1). Protein kinase C activity. The rats were Specific Protocol killed with an intraperitoneal injection of 0.2 Sheep experiments. To determine the effect g/ml sodium pentobarbitone, the lungs were of PMA on lung liquid clearance, either serum removed, and the site of instillation was idenalone (sevensheep) or serum mixed with PMA tified by the presence of Evans blue. This porat 10-7 M (seven sheep) was instilled into the tion of the lung was removed, and a portion right lower lobe. To evaluate the activation of comparable size was removed from the of PKC in the sheep after PMA (i0-7 M) in- uninstilled lung. These pieces of tissue were stillation, two extra sheep were studied to homogenized at 4 0 C separately in 5 ml of evaluate protein kinase C activity. In order extraction buffer (25 mM TRIS-HCI at pH to see if a higher dose of PMA could be used 7.5; ImMEDTA; 1mMEGTA;2mMPMSF; to study this question, another group of 0.25 M sucrose; 50 ug/ml leupeptin; 10 mM animals (four sheep) was studied when serum OTT). mixed with PMA (10-5 M) was instilled. The homogenate was centrifuged at Rat experiments. To evaluate the time 11,000 g for 10min. The supernatant was then course of activation of protein kinase C after centrifuged at 100,000 g for 1 h. The pellet PMA instillation, the protein kinase C activ- from the 100,000 g centrifugation (containity was measured in rat lung 10min (four rats), ing the membrane fraction) was resuspended 1 h (four rats), 2 h (three rats), 3 h (three rats), by sonication in 0.5 ml of extraction buffer and 4 h (six rats) after instillation of PMA containing 0.1% (vol/vol) Triton X-loo and 7 (10- M) and bovine serum albumin. assayed for PKC activity. The supernatant (cell cytosol) was assayed directly for PKC Measurements activity. Hemodynamics and protein concentration. The PKC activity in the lung tissue was Left atrial, pulmonary arterial, systemic ar- measured using the mixed micellar assay (25). terial, and airway pressure, cardiac output, Assay conditions were 20 mM TRIS-HCl at and blood gases were measured as previously pH 7.5, 10 mM MgCb, 0.1 mM CaCl z, phosdescribed (1). At 30-min intervals blood sam- phatidylserine (40 ug/rnl), 1,2-diolein (0.8 ples were collected. Total protein concentra- ug/ml), histone III -S (0.2 mg/rnl), 10 IlM tion in blood was measured using the Biuret [y_32P]ATP. The time course of histone phostechnique (23). phorylation was determined by withdrawing Excess lung water measurements. After samples of reaction mixtures at 1, 1.5, 2, 2.5, each experiment, a 40-ml blood sample was 3, 3.5,4, and 4.5 min and quenching the reacobtained to measure the hemoglobin concen- tion with 25% trichloroacetic acid and 2% tration and the wet/dry weight ratio of blood sodium pyrophosphate prior to quantificafor the lung water calculation. Then the sheep tion of protein-bound [32P]phosphate as prewas killed by cross-clamping the aorta. The viously described (26). trachea was clamped at an airway pressure The activity of protein kinase C was nor-
PROTEIN KINASE C AND WNG LIQUID CLEARANCE
lung was due to the instillation of BSA. In the instilled cytosol, albumin represented a greater proportion (40010) of the total protein than in the control cytosol (20%).
100
75 Residual Lung Water (ml)
1087
Statistics
50
The data are presented as the mean ± standard deviation. The data related to liquid and protein clearance and protein kinase C activity were analyzed by a one-way analysis of variance and the Newman-Keuls multiple range test (27). The changes in hemodynamics between groups and between baseline and experimental periods within each group were analyzed by a two-way analysis of variance with repeated measures and the Newman Keuls multiple range test (27). For the baseline pel iod, data from the last hour wereused, but for the experimental period, data from all 4 h after serum instillation were averaged. Those differences with a p value less than 0.05 were considered significant.
25
Increase in Protein Concentration (g/d I)
Results
Fig. 1. Residua/lung water (Top) and increase in protein concentration of the airway liquid (Bottom) when serum alone (n = 7) or serum mixed with PMA 10-7 M (n = 7) were instilled into a sheep lung.
malized to the dry weight of the lung included in the assay for the cytosolic fraction and to protein mass included in the assay for the membrane fraction. We have normalized the cytosolic fraction to the dry weight of the lung because the protein concentration was significantly higher in the instilled lung (3.5 ± 0.7 mg/ml, n = 20) than in the control lung (2.8 ± 0.6 mg/ml, n = 20) because we were instilling a BSA solution into the lung. However, in the membrane fraction the protein concentration was similar between the instilled lung (1.1 ± 0.3 mg/ml, n = 20) and the control lung (1.4 ± 0.3 mg/ml, n = 20). The difference in protein concentration of cytosolie fractions was not due to the difference in amount of tissue used since we had similar size of tissue sample for the control and the instilled lungs. The mean total dry weight of the control lung was 43.8 ± B.O mg (n = 20), and the mean total dry weight of the instilled lung was 48.5 ± 17.4 mg (n = 20). We have further demonstrated by gel electrophoresis of cytosolic fractions that the increased protein concentration in the instilled TABLE 1
Control PMA 10-7 M
Bloodt (%)
Lung t (%)
Total t (%)
1.0 ± 0.8 1.3 ± 1.0
93.9 ± 2.6 92.3 ± 4.9
94.9 ± 5.8 97.0 ± 6.1
• Values are mean ± so. Percent of total amount of 12sl-Albumin instilled in the lung.
t
Discussion
To evaluate the role of protein kinase C in lung liquid clearance, wechose to study the effect of PMA on lung liquid clearance. PMA, a known stimulator of protein kinase C (22), should stimulate lung liquid clearance if activation of protein kinase C has a stimulating effect on lung liquid clearance. Our results demonstrate that PMA at 10-7 M does not decrease the residual lung water or increase the protein concentration of airway fluid when compared with those in a control group. These results indicate that PMA does not stimulate lung liquid clearance. The fact that PMA does not stimulate
TABLE 2 HEMODYNAMIC AND MEAN AIRWAY RESPONSE TO SERUM INSTILLATION"
Experiment
RECOVERY OF 1251-ALBUMIN TRACER INSTILLED INTO SHEEP LUNG"
Condition
The residual lung water in the serum alone group was 76.8 ± 9.2 ml, which is similar to the data published previously for the sheep (1). The presence of PMA (10-7 M) did not have a stimulating effect on lung liquid clearance (figure 1). Furthermore, there was no significant difference in the increase in protein concentration of the airway liquid between the PMA and the control group (figure 1).The permeability properties of the alveolar barrier (table 1) and the hemodynamic response (table 2) were not significantly different in the PMA (10-7 M) group compared with the control group. Because we could not demonstrate any effect of PMA (10-7 M), we carried out a series of experiments by instilling PMA (10-s M) in four sheep. However, by using this dose we could clearly demonstrate the presence of lung injury as characterized by a greater permeability of the alveolar barrier as measured by the clearance of I2sI-albumin from the lung. In this group of four sheep, there was 4.3
± 4.10,10 of 125 I-albumin in the blood and only 80.3 ± 14.50,10 was left in the lung. This increased permeability was also associated with the presence of pleural effusion in three of the four animals and a significant increase in the pulmonary arterial pressure from 20.3 ± 4.5 to 27.3 ± 5.2 em H 2 0 . These data indicated that the higher dose of PMA induces lung injury. The presence of PMA in the instilled rat lung induced a significant activation of protein kinase C (figure 2). This activation is characterized by a shift of the activity of protein kinase C from the cytosolic fraction to the membranous fraction. This activation was present 2, 3, and 4 h after PMA (mixed in BSA) instillation. Activation of PKC was also observed in two sheep studied at 4 h postinstillation. The activity in the cytosolic fraction of the sheep PM A-instilled lung was 101.1 ± 11.0 pmol/min/mg dry weight compared with 179.6 ± 13.0 pmol/min/mg dry weight in the control lung. The activity in the membrane fraction of the PMA-instilled sheep lung was 943.0 ± 96 pmol/min/mg protein compared with 430.4 ± 0.5 pmol/min/mg protein in the control lung.
Serum alone Baseline Experimental PMA 10-7 M Baseline Experimental
Pulmonary Artery Pressure (em H2O)
Left Atrial Pressure (em H2O)
Cardiac Output
24.6 ± 5.9 25.1 ± 3.1
9.4 ± 1.5 9.3 ± 2.6
3.9 ± 1.9 4.5 ± 1.9
9.1 ± 1.2 11.1 ± 1.5:j::
21.7 ± 3.2 23.8 ± 2.4
8.1 ± 3.6 8.0 ± 3.9
4.5 ± 1.6 4.4 ± 1.3
11.0 ± 1.4 13.0 ± 1.3
(Llmin)
• Values are mean ± so. Mean airway pressure = peak airway pressure/2 + positive end-expiratory pressure/2 . Significantly different from baseline.
t
*
Mean Airway Pressure" (em H2O)
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
1088
* 3.0
Ratio 2.0 Protein Kinase C activity ( Instilled/control) 1.0
0,0
10
60
120
180
Mlnut••
lung liquid clearance could be due to several different reasons. Because PMA is well known to injure the lung (28-33), there could be an increased movement of fluid from the vascular space to the air spaces caused by a change in vascular and epithelial permeability (31-33) and/or an increase in the microcirculation pressure (28-30). To evaluate the possibility that the lack of effect of PMA was due to an increased permeability of the alveolar barrier, we first evaluated the protein clearance in this experimental group. As shown in table 1, there was no evidence that the 125I-albumintracer was cleared more rapidly from the lung of the sheep that had received PMA than from the sheep that had received serum alone. This evidence would suggest that at least there were no changes in the permeability of the lung when PMA (10-7 M) was instilled. However, because the effect of PMA on edema formation has been related to an increased permeability as well as an increase in the pressure of the microcirculation of the lung (28-30), an increase in left atrial pressure or venoconstriction of the pulmonary circulation could induce an increased fluid filtration of the lung that would then mask any effect of PMA on lung liquid clearance. The changes in the pulmonary artery pressure and left atrial pressure were not significantly different between the control group and the group treated with PMA (10-7 M), making this hypothesis unlikely. In summary, the lack of effect of PMA on lung liquid clearance cannot be explained by hemodynamic-induced increased liquid filtration or change in the permeability of the lung. Was the dose of 10-7 M of PMA a sufficient dose to induce changes in lung liquid clearance? We chose the dose of 10-7 M PMA based on multiple evidence in the literature that this dose could in-
240
Fig. 2. Ratio of protein kinase C (PKC) activity in the rat lung at different time periods. Ratio of PKC activity is defined as the PKC activity in the PMA-instilled lung divided by the activity in the controllung. At 10 min, the activity of PKC in the cytosolic fraction in the control lung was 127.2pmol/min/mg dry weight, and the activity in the membrane fraction of the control lung was 344.3 pmoll min/mg protein. In the control lung, the activity of PKC in the cytosolic fraction and in the membrane fraction did not change significantly with time. Filled bars = cytosol; hatched bars = membrane. Asterisks indicate statistical difference from the values at 10 and 60 min.
fluence ion transport properties in many different tissues. First, in the airway epithelium, Barthelson and coworkers (34) have shown that TPA at a dose of 10-7 M induces an increase in short-circuit current that is secondary to chloride secretion in airway epithelial cells. Furthermore, Li and colleagues (16) have shown, using a different approach, that 10-7 M PMA also increases chloride secretion in cultured airway epithelial cells. In other tissues, doses of PMA of 1.6 to 1 x 10-7 M have also been implicated in modulating sodium transport and chloride transport across frog skin (19) as well as across rabbit cortical collecting tubules (21). Finally, PMA at 10-7 M was also shown to have multiple effects on cellular function (35). So, based on these previous results, it seemed appropriate to choose a dose of 10-7 M to study the effect of PMA on lung liquid clearance. However, because we could not demonstrate any effect of 10-7 M PMA on lung liquid clearance, we carried out a fewexperiments using 10-5 M PMA. Unfortunately, as described in RESULTS, in those animals there was clear evidence of lung injury induced by this high dose of PMA. Because of this effect of a high dose of PMA on the lung, the interpretation of these results with respect to the modulation of lung liquid clearance by PKC activation is impossible. Although the dose of 10-7 M PMA seems an appropriate dose for study of the effect of protein kinase C on cellular modulation, is this dose appropriate for study in a whole animal model since the information available regarding tile effect of PMA on stimulation of protein kinase C comes mainly from isolated cell or tissue systems? We felt that it was important to prove that the presence of PMA in the instilled lung was really stimulating the activity of protein kinase
C. To pursue this question, we used a parallel model to the sheep experiments. Bovine serum albumin mixed with PMA was instilled in the right lower lobe of rats, and the activity of protein kinase C in the instilled lung was compared with that in the noninstilled lung. Our results clearly demonstrate that the presence of PMA in the instilled serum increases the activity of protein kinase C as demonstrated by an increase of PKC activity in the membrane fraction of the lung and a decrease in the activity of protein kinase C in the cytosolic fraction. This translocation of activity from the cytosolic fraction to the membranous fraction is characteristic of activation of protein kinase C by PMA (36). The activation of PKC occurred between 1 and 2 h after the PMA instillation, leaving 2 h of activation to stimulate lung liquid clearance. Because lung liquid clearance in the rat has been demonstrated by J ayr and coworkers (37) to be much faster than in the sheep, it may be argued that using this model in parallel to our sheep model might be inappropriate. In order to evaluate this point, we measured a change in PKC activity in two sheep experiments in which PMA (10-7 M) was instilled. Again, it was clearly demonstrated in those two animals that there was translocation of activity from the cytosolic fraction to the membrane fraction after 4 h. This would suggest that our results in the rat are at least compatible with these experiments in sheep. Furthermore, although the rate of lung liquid clearance has been demonstrated to be much faster in the rat than in the sheep (37), qualitatively, the lung liquid clearance pattern is similar between the two species. First, in both species cAMP seems to be a key modulator in lung liquid clearance (4, 14, 15). Furthermore, the mechanism involved in lung liquid clearance seems to be similar in both species since amiloride can inhibit lung liquid clearance in the two species (1, 38, 39). So, although the sheep may have a slower lung liquid clearance since, qualitatively, the mechanism seems to be similar between the two species, it seems appropriate to parallel the changes in protein kinase C activity as wehave measured it in the rat to the lung liquid clearance in the sheep. Furthermore, those experiments clearly demonstrate that the time lag between the instillation of serum with PMA and the activation of protein kinase C is much longer in vivo than when PMA is applied to a cell. When PMA is directly applied to the cell, changes in
PROTEIN KINASE C AND LUNG LIQUID CLEARANCE
cellular response occur in a matter of seconds to minutes. However, in our system, it took between 1 and 2 h for PMA to influence the activity of protein kinase C. This difference probably can be explained by the experimental method. First, PMA was dissolved in either bovine serum albumin or in autologous serum. Since PMA is a hydrophobic molecule, it may bind to protein, which may delay its action on the cell. Furthermore, PMA was not applied directly to the alveolar epithelium but by an instillation of fluid into the air spaces. It will take some time for the instilled liquid to flood completely the alveolar area, which may explain some of the delay of the response to PMA. This again points to the importance of doing these parallel experiments to determine if PMA indeed affects the activity of protein kinase C in the lung in vivo. In summary, the presence of PMA at a dose of 10-7 M induces an activation of protein kinase C without stimulating lung liquid clearance. Our results add to "the diversity of effects of PKC on ion transport published in the literature. Stimulation of protein kinase C has been shown to increase Na" transport in 3T3 cells (18) and to have an effect on Na" transport in frog epithelium (19). However, in the rabbit cortical collecting tubule (21) and in A6 epithelial cells (20), activation of protein kinase C with PMA or TPA had an inhibitory effect on Na' transport. In the jejunum, phorbol ester treatment produced a stimulation of secretion of water as well as Na", CI-, and HCO-3 transluminally (40). A clear understanding of these differences is lacking. In the lung, phorbol esters are known to stimulate surfactant secretion (41) and to increase CI- secretion in airway epithelium (34). In our system, PMA did not stimulate lung liquid clearance, but we cannot comment on its direct effect on Na' or CItransport since we did not measure those variables. However, there was no effect on lung liquid clearance. Even if direct stimulation of protein kinase C does not increase lung liquid clearance, it does not mean that protein kinase C is not important in the resolution process of pulmonaryedema. Protein kinase C activation could very well modulate the action of other second messengers such as the stimulating effect of cAMP and betaadrenergic agonists. For example, in airway epithelial cells, treatment of cells with PMA for 1 h showed a significant decrease in response to stimulation of short-circuit current by isoproterenol and
1089
dibutyryl cAMP (34). In other systems such as the cortical collecting duct (42) and in the bladder (43), stimulation of protein kinase C has been shown to inhibit the vasopressin effect on water transport. In conclusion, although protein kinase C is a well-known intracellular regulator of many cellular functions, including ion transport (22), and some physiologic mediators or hormones such as the alphaadrenergic agonists, atrial natriuretic peptide, and epidermal growth factor could have been involved in stimulating protein kinase C (22, 44, 45) and stimulating lung clearance during the resolution process of pulmonary edema, our experimental results suggest that activation of protein kinase C per se is not enough to stimulate lung liquid clearance. However, we cannot exclude a possible role of protein kinase C in modulating Na' transport and lung liquid clearance through its interaction with other mediators. Further studies are needed to evaluate this possibility. Acknowledgment The writers thank Alana Williamson and Marie Plummer for secretarial assistance. References 1. Berthiaume Y, Staub NC, Matthay MA. Betaadrenergic agonists increase lung liquid clearance in anesthetized sheep. J Clin Invest 1987;79:335-43. 2. Goodman BE, Kim KJ, Crandall ED. Evidence for active sodium transport across alveolar epithelium of isolated rat lung. J Appl Physiol 1987; 62:2460-6. 3. Goodman BE, Fleischer RS, Crandall ED. Evidence for active Na' transport by cultured monolayers of pulmonary alveolar epithelial cells. Am J Physiol 1983; 245:C78-83. 4. Saumon G, Basset G, Bouchonnet F, Crone C. cAMP and beta-adrenergic stimulation of rat alveolar epithelium. Effects on fluid absorption and paracellular permeability. Pflugers Arch 1987; 410:464-70. 5. Goodman BE, Brown SES, Crandall ED. Regulation of transport across pulmonary alveolar epithelial cell monolayers. J Appl Physiol 1984; 57:703-10. 6. Cheek JM, Kim KJ, Crandall ED. Tight monolayers of rat alveolar epithelial cells: bioelectric properties and active sodium transport. Am J Physiol 1989; 256:C688-93. 7. Matalon S, Kirk K, Benos OJ. Immunofluorescent localization of Na' channel protein in type II pneumocytes (abstract). J Cell BioI 1989; 109: BOA. 8. Fischer H, Van Driessche W, Clauss W. Evidence for apical sodium channels in frog lung epithelial cells. Am J Physiol 1989; 256:C764-71. 9. Orser B, Bertlik H, Fedorko L, O'Brodovich H. Cation channels in apical membrane of fetal alveolar epithelium (abstract). Can J Anaesthesiol 1990; 37:S3. 10. Matthay MA, Wiener-Kronish JP. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis
1990; 142:1250-7. 11. Welsh MJ. Electrolyte transport by airway epithelia. Physiol Rev 1987; 67:1143-84. 12. Garty H, Benos OJ. Characteristics and regulatory mechanisms of the amiloride-blockable Na' channel. Physiol Rev 1988; 68:309-73. 13. Cott GR, Sugahara K, Mason RJ. Stimulation of net active ion transport across alveolar type II cell monolayers. Am J Physiol1986; 250:C222-7. 14. Goodman BE, Anderson JL, Clemens JW Evidence for regulation of sodium transport from airspace to vascular space by cAMP. Am J Physiol 1989; 257:L86-93. 15. Berthiaume Y. Effect of exogenous cAMP and aminophylline on alveolar and lung liquid clearance in anesthetized sheep. J Appl Physiol 1991; 70:2490-7. 16. Li M, McCann JD, Anderson MP, et 01. Regulation of chloride channels by protein kinase C in normal and cystic fibrosis airway epithelia. Science 1989; 244:1353-6. 17. Bertorello A, Aperia A. Na+-K+-ATPaseis an effector protein for protein kinase C in renal proximal tubule cells. Am J Physiol1989; 256:F370-3. 18. Dicker P, Rozengurt E. Phorbol ester stimulation of Na" influx and Na-K pump activity in swiss 3T3 cells. Biochem Biophys Res Commun 1981; 100:433-41. 19. Civan MM, Rubenstein D, Mauro T, O'Brien TG. Effects of tumor promoters on sodium ion transport across frog skin. Am J Physiol 1985; 248:C457-65. 20. Yanase M, Handler JS. Activators of protein kinase C inhibit sodium transport in A6 epithelia. Am J Physiol 1986; 250:C517-22. 21. Hays SR, Baum M, Kokko JP. Effects of protein kinase C activation on sodium, potassium, chloride and total CO 2 transport in the rabbit cortical collecting tubule. J Clin Invest 1987; 80:1561-70. 22. Nishizuka Y. Studies and perspectives of protein kinase C. Science 1986; 233:305-12. 23. Doumas BT, Bayse DD, Carter RJ, Peters T Jr, Schaffer R. A candidate reference method for determination of total protein in serum 1. Development and validation. Clin Chern 1981; 27: 1642-50. 24. Berthiaume Y, Broaddus CV, Gropper M, Tanita T, Onizuka M, Matthay MA. Alveolar liquid and protein clearance from the normal dog lung. J Appl Physiol 1988; 65:585-93. 25. Bell RM, Hannun Y, Loomis C. Mixed micelle assay of protein kinase C. Methods Enzymo11986; 124:353-9. 26. Walsh MP, Hinkins S, Dabrowska R, Hartshorne OJ. Smooth muscle myosin light chain kinase. Methods Enzymol 1983; 99:279-88. 27. Zar JH. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall, 1974; 130-81. 28. Perry M, Taylor A. Phorbol myristate acetateinduced injury of isolated perfused rat lungs: neutrophil dependence. J Appl Physiol1988; 65:2164-9. 29. Allison RC, Hernandez EM, Prasad VR, Grisham MB, Taylor AE. Protective effects of O 2 radical scavengers and adenosine in PMA-induced lung injury. J Appl Physiol 1988; 64:2175-82. 30. Johnson A. PMA-induced pulmonary edema: mechanisms of the vasoactive response. J Appl Physiol 1988; 65:2302-12. 31. Shasby DM, Vanbenthuysen KM, Tate RM, Shasby SS, McMurtry I, Repine JE. Granulocytes mediate acute edematous lung injury in rabbits and in isolated rabbit lungs perfused with phorbol myristate acetate: role of oxygen radicals. Am Rev Respir Dis 1982; 125:443-7. 32. Shasby DM, Shasby SS, Peach MV. Granulocytes and phorbol myristate acetate increase permeability to albumin of cultured endothelial mono-
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
1090 layers and isolated perfused lungs. Am Rev Respir Dis 1983; 127:72-6. 33. Dyer EL, Snapper VR. Role of circulating granulocytesin sheep lung injury produced by phorbol myristate acetate. J Appl Physiol 1986; 60: 576-89. 34. Barthelson RA, Jacoby DB, Widdicombe JH. Regulation of chloride secretion in dog tracheal epithelium by protein kinase C. Am J Physiol 1987; 253:C802-8. 35. Blumberg PM. In vitro studies on the mode of action of the phorbol esters, potent tumor promoters: Part 1. CRC Crit Rev Toxicol 1980; 153-97. 36. Kraft AS, Anderson WB. Phorbol esters increase the movement of Ca", phospholipid-dependent protein kinase associated with plasma membrane. Nature 1983; 301:621-3. 37. Jayr C, Leger N, Zeiter M, Matthay MA. AI-
veolar epithelial liquid clearance in anesthetized, ventilated rats is fast compared with studies in dogs, sheep, and rabbits (abstract). Am Rev Respir Dis 1990; 141:A300. 38. Matthay MA. Resolution of pulmonary edema. Mechanisms of liquid, protein, and cellular clearance from the lung. Clin Chest Med 1985; 6:521-45. 39. Jayr C, Leger N, Zelter M, Matthay MAo Effect of amiloride on alveolar and lung liquid clearance in ventilated rats (abstract). Am Rev Respir Dis 1991; 143:AI50. 40. WeikelCS, Sando n, Guerrant RL. Stimulation of porcine jejunal ion secretion in vivo by protein kinase C activators. J Clin Invest 1985; 76: 2430-5. 41. Sano K, VoelkerDR, Mason RJ. Involvement of protein kinase C in pulmonary surfactant secretion from alveolar type II cells. J Biol Chern 1985;
260:12725-9. 42. Ando Y, Jacobson HR, Breyer MD. Phorbol myristate acetate, dioctanoylglycerol, and phosphatidic acid inhibit the hydroosmotic effect of vasopressin on rabbit cortical collecting tubule. J Clin Invest 1987; 80:590-3. 43. Schlondorff 0, Levine SO. Inhibition of vasopressin-stimulated water flow in toad bladder by phorbol myristate acetate, dioctanoylglycerol, and RHC-80267. J Clin Invest 1985; 76:1071-8. 44. Minneman KP. ai-Adrenergic receptor subtypes, inositol phosphates, and sources of cellCa". Pharmacol Rev 1988; 40:87-119. 45. Mohrmann M, Cantiello HF, Ausiello DA. Inhibition of epithelial Na" transport byatriopeptin, protein kinase C, and pertussis toxin. Am J Physiol 1987; 253:F372-6.
YVES BERTHIAUME,4 MARIUSZ SAPIJASZKO, JOYCE MACKENZIE, and MICHAEL P. WALSH5 With the technical assistance of Derrice Payne
A
Introduction
substantial body of evidence suggests that resolution of pulmonary edema may depend in part on the active transport of Na" from the air spaces of the lung to the interstitial space and, subsequently, to the vascular space. This evidence was obtained in many different experimental models. First, it is known that lung liquid clearance can be stimulated by beta-adrenergic agonists, and this increase in lung liquid clearance can be inhibited byamiloride, a sodium transport inhibitor (1). Sodium transport has also been clearly demonstrated in isolated perfused lungs (2) as well as in cultured alveolar type II cells (3). This Na' transport can be inhibited by amiloride (2, 3). Beta-adrenergic stimulation also increased Na" transport and lung liquid clearance in the isolated lung preparation (2,4) as well as dome formation (5) and short-circuit current (6) in cultured alveolar type II cells. Recently, some preliminary experimental data has suggested that Na" transport across the alveolar epithelium could occur either through a Na" channel (7, 8) or a nonspecific cation channel (9).Finally, recent human studies support the hypothesis that active ion transport across the alveolar epithelium is the primary mechanism for clearance of edema fluid from the air spaces of the lung (10). Although this evidence clearly supports the role of active transport of Na' in the resolution of pulmonary edema, the mechanisms regulating Na" transport are still not clearly established. Twomajor intracellular second messengers have been implicated in the regulation of ion transport. Adenosine 3',5'cyclic monophosphate (cAMP) is well known to stimulate chloride transport in the airway epithelium (11) as well as Na" transport in epithelial cells derived from the kidney (12). Recently, cAMP has also been shown to playa significant role in stimulation of Na" transport in lung liq-
SUMMARY Although active transport of ions could play an important role in the resolution process of pulmonary edema, the exact mechanism regulating this process is stili unknown. In this study, we investigated the effect of phorbol myristate acetate (PMA) on lung liquid clearance in anesthetized, ventilated sheep to evaluate the possible role of protein kinase C. To study lung liquid and protein clearance, we measured the removal of 100 ml of autologous serum from the air spaces of anesthetized sheep. Either serum alone or serum mixed with PMA (10-7 M) was instilled. After 4 h, the residual lung water was 76.8 ± 9.2 ml when serum alone was instilled and 79.5 ± 15.7 when serum with PMA (10-7 M) was instilled. The lack of effect of PMA (10-7 M) on lung liquid clearance cannot be explained by increased movement of liqUid from the vascular space to the air space since we did not have any evidence of increased pressure or increased permeability In the lung. This lack of effect of PMA (10-7 M) is not due to an absence of stimulation of protein kinase C since instillation of BSA and PMA (10-7 M) in rat lung produced a translocation of protein kinase C activity from the cytosolic fraction to the membrane fraction 2 h after the Instillation. These results were confirmed In two sheep experiments, which demonstrated clear activation of protein kinase C after 4 h. These data suggest that activation of protein kinase C does not stimulate lung liqUid clearance. However, a possible role of protein kinase C in modulating lung liquid clearance has not been excluded. AM REV RESPIR DIS 1991; 144:1085-1090
uid clearance. cAMP analogs and betaadrenergic agonists were shown to increase both dome formation (5) and short-circuit currents across a monolayer of alveolar epithelial type II cells (6, 13)by increasing sodium transport from the apical side to the serosal side (6). These effects of cAMP analogs or betaadrenergic agonist are partially inhibited by amiloride (6, 13).Moreover, cAMP analogs increase Na" transport (14) and liquid clearance (4) in isolated rat lungs. Recently, we also demonstrated that cAMP in combination with aminophylline can stimulate lung liquid clearance in sheep (15). Another second messenger system that has a significant role in chloride transport or Na" transport is protein kinase C. First, protein kinase C can modulate the activity of proteins involved in ion transport. For example, protein kinase C can activate or inactivate chloride channels depending on Ca2+ concentration (16). The sodium potassium ATPase pump is also an effector protein of protein kinase C (17). Protein kinase C activation also can influence ion transport. For example, protein kinase C stimulation with 12-0-tetradecanoylphorbol-13-
acetate (TPA) and phorboI12,13-dibutyrate enhances Na" entry into 3T3 cells (18). TPA also stimulates short-circuit currents in frog skin, probably through stimulation of a Na' channel (19). In contrast, activation of protein kinase C inhibits apical sodium transport in A 6 epithelial cells (20) as well as sodium and potassium transport in the rabbit cortical collecting tubule (21). Because activation of protein kinase
(Received in original form January 17, 1991 and in revised form June 14, 1991) 1 From the Departments of Medicine and Medical Biochemistry, University of Calgary, Calgary, Alberta, Canada. 2 Supported by Grants-in-Aid MA-10273-YB and MT-9097-MPW from The Medical Research Council of Canada. 3 Correspondence and requests for reprints should be addressed to Dr. Yves Berthiaume, Department of Medicine, University of Calgary, 3330 Hospital Drive NW., Calgary, Alberta T2N 4NI Canada. 4 Clinical Investigator of the Alberta Heritage Foundation for Medical Research and a Scholar of The Medical Research Council of Canada. 5 Recipient of a Medical Research Council of Canada Scientist Award and an Alberta Heritage Foundation for Medical Research Scholarship.
1085
1086
C through its effect on ion transport could have an important effect on lung liquid clearance, we decided to evaluate if the activation of protein kinase C with phorbol myristate acetate (PMA), a wellknown activator of the enzyme (22), can stimulate lung liquid clearance in anesthetized sheep. Methods Animal Preparation and Serum Instillation Sheep surgicalpreparation and serum instillation. Twenty yearling sheep were anesthetized intravenously with pentothal (30 mg/ kg). A tracheotomy was done, and a cuffed tracheotomy tube was inserted. The lungs were ventilated with a constant-volume ventilator (Harvard Apparatus Co. Inc., South Natick, MA) at a tidal volume of 13to 15ml/kg with a peak inspiratory pressure of 15 to 20 em H 20. The respiratory rate was adjusted to achieve a Paco, between 30 and 40 mm Hg. Anesthesia was maintained by 0.8 to 1.50/0 isofluorane mixed with supplemental oxygen (40 to 60%). The sheep were paralyzed intravenously with pancuronium bromide (1.0mg) at the time of the thoracotomy and just before the bronchoscopy. The preparation of the anesthetized sheep to measure systemic and pulmonary hemodynamics was similar to the one previously published (1).After surgery, the sheep were placed in the prone position, with head and thorax tilted up at a 5-degree angle to maintain the lower lobes in a slightly dependent position. The procedures for preparation and instillation of the serum solution are essentially the same as we have previously published (1). Briefly, on the day before the experiment, 250 ml of blood was withdrawn from the sheep. The blood was clotted and centrifuged to obtain 125 ml of serum, which was refrigerated overnight. The next morning, 30 mg of anhydrous Evans blue was added to bind to the albumin (1). 1251-labeled human serum albumin (20 Ci) (Frosst, Division of Merck Frosst Canada Inc., Kirkland, Quebec, Canada) was added to the serum as a protein tracer. Before instilling the serum, a sample of the instillate was taken for total protein measurement and wet/dry ratio measurements so that the dry weight could be subtracted from the final lung water calculations. Serum (100ml) was instilled into the right lower lobe over 5 to 10 min through a flexible fiberoptic bronchoscope (Olympus Corp., Lake Success, NY) using a three-way 'l-piece adapter. The ventilation was continued during the serum instillation. Rat surgicalpreparation and BSA instillation. Although PMA is a well-known activator of protein kinase C (22), we wanted to establish clearly that PMA at the dose used can stimulate protein kinase C in the lung. To study this question we used a rat model simulating our sheep model. The rats were anesthetized with halothane
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
(2 to 3%) mixed with 100070 oxygen. A tracheostomy was done and, using a rodent feeding tube, 0.2 to 0.4 ml of bovine serum albumin (5%) was injected in the right lower lobe. The wound was closed with Vetbound tissue adhesive (Animal Care Products; 3M, St. Paul, MN). The animal was then allowed to recover from anesthesia. The animal would be up and active in about 5 min after the end of the procedure. It was not necessary to exclude any animals from the study because of cross contamination in control lungs.
of 20 em H 20, and the lungs were removed from the thorax. Airway liquid (2 to 5 ml) was obtained by passing a catheter 3 mm in diameter into the distal airways of the seruminstilled lung. The sample was centrifuged, and the total protein concentration and radioactivity (counts per milliliter) of the supernatant were measured. Then the lungs were frozen in liquid nitrogen. The next day, each lung was examined in a cryostat at - 20 0 C to determine the distribution of the Evansblue-labeled serum and to confirm that none of the fluid had spilled over into the control lung. General Protocol Each lung was then homogenized separateThree experimental groups of sheep were studly, and the extravascular lung water was deied using the following general protocol. After surgery there was a baseline period of 1.5 termined by calculating the wet/dry weight to 2 h during which vascular pressure was sta- ratio (grams H 20/gram dry lung) (1). The exble. The serum solution with or without PMA cess water in the experimental lung was calwas then instilled. During the next 4 h, sys- culated with the same equation as described temic pulmonary and left atrial pressure and previously (1, 24). The coefficient of variation for this technique is 11.5% (1). cardiac output were measured every 15 min. Alveolar permeability. Because PMA is Blood samples were taken every 0.5 h for towell known to induce lung injury, we evalutal protein concentration measurements and 1251-albumin counts. After 4 h, the sheep were ated the permeability of the alveolar barrier as measured by the clearance of protein from killed, and the lungs were removed for samthe lung. For assessment of protein clearance, pling of the airway liquid to obtain the final 20 IlCi of 1251-human serum albumin was addairway liquid protein concentration, measureed to the instilled serum. The method for asment of the excess lung water to assess lung liquid clearance, and quantification of the 1251_ sessing the total amount of tracer recovered in homogenized lung and plasma was similar albumin left in the lung (see Measurements). to the method previously published (1). Protein kinase C activity. The rats were Specific Protocol killed with an intraperitoneal injection of 0.2 Sheep experiments. To determine the effect g/ml sodium pentobarbitone, the lungs were of PMA on lung liquid clearance, either serum removed, and the site of instillation was idenalone (sevensheep) or serum mixed with PMA tified by the presence of Evans blue. This porat 10-7 M (seven sheep) was instilled into the tion of the lung was removed, and a portion right lower lobe. To evaluate the activation of comparable size was removed from the of PKC in the sheep after PMA (i0-7 M) in- uninstilled lung. These pieces of tissue were stillation, two extra sheep were studied to homogenized at 4 0 C separately in 5 ml of evaluate protein kinase C activity. In order extraction buffer (25 mM TRIS-HCI at pH to see if a higher dose of PMA could be used 7.5; ImMEDTA; 1mMEGTA;2mMPMSF; to study this question, another group of 0.25 M sucrose; 50 ug/ml leupeptin; 10 mM animals (four sheep) was studied when serum OTT). mixed with PMA (10-5 M) was instilled. The homogenate was centrifuged at Rat experiments. To evaluate the time 11,000 g for 10min. The supernatant was then course of activation of protein kinase C after centrifuged at 100,000 g for 1 h. The pellet PMA instillation, the protein kinase C activ- from the 100,000 g centrifugation (containity was measured in rat lung 10min (four rats), ing the membrane fraction) was resuspended 1 h (four rats), 2 h (three rats), 3 h (three rats), by sonication in 0.5 ml of extraction buffer and 4 h (six rats) after instillation of PMA containing 0.1% (vol/vol) Triton X-loo and 7 (10- M) and bovine serum albumin. assayed for PKC activity. The supernatant (cell cytosol) was assayed directly for PKC Measurements activity. Hemodynamics and protein concentration. The PKC activity in the lung tissue was Left atrial, pulmonary arterial, systemic ar- measured using the mixed micellar assay (25). terial, and airway pressure, cardiac output, Assay conditions were 20 mM TRIS-HCl at and blood gases were measured as previously pH 7.5, 10 mM MgCb, 0.1 mM CaCl z, phosdescribed (1). At 30-min intervals blood sam- phatidylserine (40 ug/rnl), 1,2-diolein (0.8 ples were collected. Total protein concentra- ug/ml), histone III -S (0.2 mg/rnl), 10 IlM tion in blood was measured using the Biuret [y_32P]ATP. The time course of histone phostechnique (23). phorylation was determined by withdrawing Excess lung water measurements. After samples of reaction mixtures at 1, 1.5, 2, 2.5, each experiment, a 40-ml blood sample was 3, 3.5,4, and 4.5 min and quenching the reacobtained to measure the hemoglobin concen- tion with 25% trichloroacetic acid and 2% tration and the wet/dry weight ratio of blood sodium pyrophosphate prior to quantificafor the lung water calculation. Then the sheep tion of protein-bound [32P]phosphate as prewas killed by cross-clamping the aorta. The viously described (26). trachea was clamped at an airway pressure The activity of protein kinase C was nor-
PROTEIN KINASE C AND WNG LIQUID CLEARANCE
lung was due to the instillation of BSA. In the instilled cytosol, albumin represented a greater proportion (40010) of the total protein than in the control cytosol (20%).
100
75 Residual Lung Water (ml)
1087
Statistics
50
The data are presented as the mean ± standard deviation. The data related to liquid and protein clearance and protein kinase C activity were analyzed by a one-way analysis of variance and the Newman-Keuls multiple range test (27). The changes in hemodynamics between groups and between baseline and experimental periods within each group were analyzed by a two-way analysis of variance with repeated measures and the Newman Keuls multiple range test (27). For the baseline pel iod, data from the last hour wereused, but for the experimental period, data from all 4 h after serum instillation were averaged. Those differences with a p value less than 0.05 were considered significant.
25
Increase in Protein Concentration (g/d I)
Results
Fig. 1. Residua/lung water (Top) and increase in protein concentration of the airway liquid (Bottom) when serum alone (n = 7) or serum mixed with PMA 10-7 M (n = 7) were instilled into a sheep lung.
malized to the dry weight of the lung included in the assay for the cytosolic fraction and to protein mass included in the assay for the membrane fraction. We have normalized the cytosolic fraction to the dry weight of the lung because the protein concentration was significantly higher in the instilled lung (3.5 ± 0.7 mg/ml, n = 20) than in the control lung (2.8 ± 0.6 mg/ml, n = 20) because we were instilling a BSA solution into the lung. However, in the membrane fraction the protein concentration was similar between the instilled lung (1.1 ± 0.3 mg/ml, n = 20) and the control lung (1.4 ± 0.3 mg/ml, n = 20). The difference in protein concentration of cytosolie fractions was not due to the difference in amount of tissue used since we had similar size of tissue sample for the control and the instilled lungs. The mean total dry weight of the control lung was 43.8 ± B.O mg (n = 20), and the mean total dry weight of the instilled lung was 48.5 ± 17.4 mg (n = 20). We have further demonstrated by gel electrophoresis of cytosolic fractions that the increased protein concentration in the instilled TABLE 1
Control PMA 10-7 M
Bloodt (%)
Lung t (%)
Total t (%)
1.0 ± 0.8 1.3 ± 1.0
93.9 ± 2.6 92.3 ± 4.9
94.9 ± 5.8 97.0 ± 6.1
• Values are mean ± so. Percent of total amount of 12sl-Albumin instilled in the lung.
t
Discussion
To evaluate the role of protein kinase C in lung liquid clearance, wechose to study the effect of PMA on lung liquid clearance. PMA, a known stimulator of protein kinase C (22), should stimulate lung liquid clearance if activation of protein kinase C has a stimulating effect on lung liquid clearance. Our results demonstrate that PMA at 10-7 M does not decrease the residual lung water or increase the protein concentration of airway fluid when compared with those in a control group. These results indicate that PMA does not stimulate lung liquid clearance. The fact that PMA does not stimulate
TABLE 2 HEMODYNAMIC AND MEAN AIRWAY RESPONSE TO SERUM INSTILLATION"
Experiment
RECOVERY OF 1251-ALBUMIN TRACER INSTILLED INTO SHEEP LUNG"
Condition
The residual lung water in the serum alone group was 76.8 ± 9.2 ml, which is similar to the data published previously for the sheep (1). The presence of PMA (10-7 M) did not have a stimulating effect on lung liquid clearance (figure 1). Furthermore, there was no significant difference in the increase in protein concentration of the airway liquid between the PMA and the control group (figure 1).The permeability properties of the alveolar barrier (table 1) and the hemodynamic response (table 2) were not significantly different in the PMA (10-7 M) group compared with the control group. Because we could not demonstrate any effect of PMA (10-7 M), we carried out a series of experiments by instilling PMA (10-s M) in four sheep. However, by using this dose we could clearly demonstrate the presence of lung injury as characterized by a greater permeability of the alveolar barrier as measured by the clearance of I2sI-albumin from the lung. In this group of four sheep, there was 4.3
± 4.10,10 of 125 I-albumin in the blood and only 80.3 ± 14.50,10 was left in the lung. This increased permeability was also associated with the presence of pleural effusion in three of the four animals and a significant increase in the pulmonary arterial pressure from 20.3 ± 4.5 to 27.3 ± 5.2 em H 2 0 . These data indicated that the higher dose of PMA induces lung injury. The presence of PMA in the instilled rat lung induced a significant activation of protein kinase C (figure 2). This activation is characterized by a shift of the activity of protein kinase C from the cytosolic fraction to the membranous fraction. This activation was present 2, 3, and 4 h after PMA (mixed in BSA) instillation. Activation of PKC was also observed in two sheep studied at 4 h postinstillation. The activity in the cytosolic fraction of the sheep PM A-instilled lung was 101.1 ± 11.0 pmol/min/mg dry weight compared with 179.6 ± 13.0 pmol/min/mg dry weight in the control lung. The activity in the membrane fraction of the PMA-instilled sheep lung was 943.0 ± 96 pmol/min/mg protein compared with 430.4 ± 0.5 pmol/min/mg protein in the control lung.
Serum alone Baseline Experimental PMA 10-7 M Baseline Experimental
Pulmonary Artery Pressure (em H2O)
Left Atrial Pressure (em H2O)
Cardiac Output
24.6 ± 5.9 25.1 ± 3.1
9.4 ± 1.5 9.3 ± 2.6
3.9 ± 1.9 4.5 ± 1.9
9.1 ± 1.2 11.1 ± 1.5:j::
21.7 ± 3.2 23.8 ± 2.4
8.1 ± 3.6 8.0 ± 3.9
4.5 ± 1.6 4.4 ± 1.3
11.0 ± 1.4 13.0 ± 1.3
(Llmin)
• Values are mean ± so. Mean airway pressure = peak airway pressure/2 + positive end-expiratory pressure/2 . Significantly different from baseline.
t
*
Mean Airway Pressure" (em H2O)
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
1088
* 3.0
Ratio 2.0 Protein Kinase C activity ( Instilled/control) 1.0
0,0
10
60
120
180
Mlnut••
lung liquid clearance could be due to several different reasons. Because PMA is well known to injure the lung (28-33), there could be an increased movement of fluid from the vascular space to the air spaces caused by a change in vascular and epithelial permeability (31-33) and/or an increase in the microcirculation pressure (28-30). To evaluate the possibility that the lack of effect of PMA was due to an increased permeability of the alveolar barrier, we first evaluated the protein clearance in this experimental group. As shown in table 1, there was no evidence that the 125I-albumintracer was cleared more rapidly from the lung of the sheep that had received PMA than from the sheep that had received serum alone. This evidence would suggest that at least there were no changes in the permeability of the lung when PMA (10-7 M) was instilled. However, because the effect of PMA on edema formation has been related to an increased permeability as well as an increase in the pressure of the microcirculation of the lung (28-30), an increase in left atrial pressure or venoconstriction of the pulmonary circulation could induce an increased fluid filtration of the lung that would then mask any effect of PMA on lung liquid clearance. The changes in the pulmonary artery pressure and left atrial pressure were not significantly different between the control group and the group treated with PMA (10-7 M), making this hypothesis unlikely. In summary, the lack of effect of PMA on lung liquid clearance cannot be explained by hemodynamic-induced increased liquid filtration or change in the permeability of the lung. Was the dose of 10-7 M of PMA a sufficient dose to induce changes in lung liquid clearance? We chose the dose of 10-7 M PMA based on multiple evidence in the literature that this dose could in-
240
Fig. 2. Ratio of protein kinase C (PKC) activity in the rat lung at different time periods. Ratio of PKC activity is defined as the PKC activity in the PMA-instilled lung divided by the activity in the controllung. At 10 min, the activity of PKC in the cytosolic fraction in the control lung was 127.2pmol/min/mg dry weight, and the activity in the membrane fraction of the control lung was 344.3 pmoll min/mg protein. In the control lung, the activity of PKC in the cytosolic fraction and in the membrane fraction did not change significantly with time. Filled bars = cytosol; hatched bars = membrane. Asterisks indicate statistical difference from the values at 10 and 60 min.
fluence ion transport properties in many different tissues. First, in the airway epithelium, Barthelson and coworkers (34) have shown that TPA at a dose of 10-7 M induces an increase in short-circuit current that is secondary to chloride secretion in airway epithelial cells. Furthermore, Li and colleagues (16) have shown, using a different approach, that 10-7 M PMA also increases chloride secretion in cultured airway epithelial cells. In other tissues, doses of PMA of 1.6 to 1 x 10-7 M have also been implicated in modulating sodium transport and chloride transport across frog skin (19) as well as across rabbit cortical collecting tubules (21). Finally, PMA at 10-7 M was also shown to have multiple effects on cellular function (35). So, based on these previous results, it seemed appropriate to choose a dose of 10-7 M to study the effect of PMA on lung liquid clearance. However, because we could not demonstrate any effect of 10-7 M PMA on lung liquid clearance, we carried out a fewexperiments using 10-5 M PMA. Unfortunately, as described in RESULTS, in those animals there was clear evidence of lung injury induced by this high dose of PMA. Because of this effect of a high dose of PMA on the lung, the interpretation of these results with respect to the modulation of lung liquid clearance by PKC activation is impossible. Although the dose of 10-7 M PMA seems an appropriate dose for study of the effect of protein kinase C on cellular modulation, is this dose appropriate for study in a whole animal model since the information available regarding tile effect of PMA on stimulation of protein kinase C comes mainly from isolated cell or tissue systems? We felt that it was important to prove that the presence of PMA in the instilled lung was really stimulating the activity of protein kinase
C. To pursue this question, we used a parallel model to the sheep experiments. Bovine serum albumin mixed with PMA was instilled in the right lower lobe of rats, and the activity of protein kinase C in the instilled lung was compared with that in the noninstilled lung. Our results clearly demonstrate that the presence of PMA in the instilled serum increases the activity of protein kinase C as demonstrated by an increase of PKC activity in the membrane fraction of the lung and a decrease in the activity of protein kinase C in the cytosolic fraction. This translocation of activity from the cytosolic fraction to the membranous fraction is characteristic of activation of protein kinase C by PMA (36). The activation of PKC occurred between 1 and 2 h after the PMA instillation, leaving 2 h of activation to stimulate lung liquid clearance. Because lung liquid clearance in the rat has been demonstrated by J ayr and coworkers (37) to be much faster than in the sheep, it may be argued that using this model in parallel to our sheep model might be inappropriate. In order to evaluate this point, we measured a change in PKC activity in two sheep experiments in which PMA (10-7 M) was instilled. Again, it was clearly demonstrated in those two animals that there was translocation of activity from the cytosolic fraction to the membrane fraction after 4 h. This would suggest that our results in the rat are at least compatible with these experiments in sheep. Furthermore, although the rate of lung liquid clearance has been demonstrated to be much faster in the rat than in the sheep (37), qualitatively, the lung liquid clearance pattern is similar between the two species. First, in both species cAMP seems to be a key modulator in lung liquid clearance (4, 14, 15). Furthermore, the mechanism involved in lung liquid clearance seems to be similar in both species since amiloride can inhibit lung liquid clearance in the two species (1, 38, 39). So, although the sheep may have a slower lung liquid clearance since, qualitatively, the mechanism seems to be similar between the two species, it seems appropriate to parallel the changes in protein kinase C activity as wehave measured it in the rat to the lung liquid clearance in the sheep. Furthermore, those experiments clearly demonstrate that the time lag between the instillation of serum with PMA and the activation of protein kinase C is much longer in vivo than when PMA is applied to a cell. When PMA is directly applied to the cell, changes in
PROTEIN KINASE C AND LUNG LIQUID CLEARANCE
cellular response occur in a matter of seconds to minutes. However, in our system, it took between 1 and 2 h for PMA to influence the activity of protein kinase C. This difference probably can be explained by the experimental method. First, PMA was dissolved in either bovine serum albumin or in autologous serum. Since PMA is a hydrophobic molecule, it may bind to protein, which may delay its action on the cell. Furthermore, PMA was not applied directly to the alveolar epithelium but by an instillation of fluid into the air spaces. It will take some time for the instilled liquid to flood completely the alveolar area, which may explain some of the delay of the response to PMA. This again points to the importance of doing these parallel experiments to determine if PMA indeed affects the activity of protein kinase C in the lung in vivo. In summary, the presence of PMA at a dose of 10-7 M induces an activation of protein kinase C without stimulating lung liquid clearance. Our results add to "the diversity of effects of PKC on ion transport published in the literature. Stimulation of protein kinase C has been shown to increase Na" transport in 3T3 cells (18) and to have an effect on Na" transport in frog epithelium (19). However, in the rabbit cortical collecting tubule (21) and in A6 epithelial cells (20), activation of protein kinase C with PMA or TPA had an inhibitory effect on Na' transport. In the jejunum, phorbol ester treatment produced a stimulation of secretion of water as well as Na", CI-, and HCO-3 transluminally (40). A clear understanding of these differences is lacking. In the lung, phorbol esters are known to stimulate surfactant secretion (41) and to increase CI- secretion in airway epithelium (34). In our system, PMA did not stimulate lung liquid clearance, but we cannot comment on its direct effect on Na' or CItransport since we did not measure those variables. However, there was no effect on lung liquid clearance. Even if direct stimulation of protein kinase C does not increase lung liquid clearance, it does not mean that protein kinase C is not important in the resolution process of pulmonaryedema. Protein kinase C activation could very well modulate the action of other second messengers such as the stimulating effect of cAMP and betaadrenergic agonists. For example, in airway epithelial cells, treatment of cells with PMA for 1 h showed a significant decrease in response to stimulation of short-circuit current by isoproterenol and
1089
dibutyryl cAMP (34). In other systems such as the cortical collecting duct (42) and in the bladder (43), stimulation of protein kinase C has been shown to inhibit the vasopressin effect on water transport. In conclusion, although protein kinase C is a well-known intracellular regulator of many cellular functions, including ion transport (22), and some physiologic mediators or hormones such as the alphaadrenergic agonists, atrial natriuretic peptide, and epidermal growth factor could have been involved in stimulating protein kinase C (22, 44, 45) and stimulating lung clearance during the resolution process of pulmonary edema, our experimental results suggest that activation of protein kinase C per se is not enough to stimulate lung liquid clearance. However, we cannot exclude a possible role of protein kinase C in modulating Na' transport and lung liquid clearance through its interaction with other mediators. Further studies are needed to evaluate this possibility. Acknowledgment The writers thank Alana Williamson and Marie Plummer for secretarial assistance. References 1. Berthiaume Y, Staub NC, Matthay MA. Betaadrenergic agonists increase lung liquid clearance in anesthetized sheep. J Clin Invest 1987;79:335-43. 2. Goodman BE, Kim KJ, Crandall ED. Evidence for active sodium transport across alveolar epithelium of isolated rat lung. J Appl Physiol 1987; 62:2460-6. 3. Goodman BE, Fleischer RS, Crandall ED. Evidence for active Na' transport by cultured monolayers of pulmonary alveolar epithelial cells. Am J Physiol 1983; 245:C78-83. 4. Saumon G, Basset G, Bouchonnet F, Crone C. cAMP and beta-adrenergic stimulation of rat alveolar epithelium. Effects on fluid absorption and paracellular permeability. Pflugers Arch 1987; 410:464-70. 5. Goodman BE, Brown SES, Crandall ED. Regulation of transport across pulmonary alveolar epithelial cell monolayers. J Appl Physiol 1984; 57:703-10. 6. Cheek JM, Kim KJ, Crandall ED. Tight monolayers of rat alveolar epithelial cells: bioelectric properties and active sodium transport. Am J Physiol 1989; 256:C688-93. 7. Matalon S, Kirk K, Benos OJ. Immunofluorescent localization of Na' channel protein in type II pneumocytes (abstract). J Cell BioI 1989; 109: BOA. 8. Fischer H, Van Driessche W, Clauss W. Evidence for apical sodium channels in frog lung epithelial cells. Am J Physiol 1989; 256:C764-71. 9. Orser B, Bertlik H, Fedorko L, O'Brodovich H. Cation channels in apical membrane of fetal alveolar epithelium (abstract). Can J Anaesthesiol 1990; 37:S3. 10. Matthay MA, Wiener-Kronish JP. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis
1990; 142:1250-7. 11. Welsh MJ. Electrolyte transport by airway epithelia. Physiol Rev 1987; 67:1143-84. 12. Garty H, Benos OJ. Characteristics and regulatory mechanisms of the amiloride-blockable Na' channel. Physiol Rev 1988; 68:309-73. 13. Cott GR, Sugahara K, Mason RJ. Stimulation of net active ion transport across alveolar type II cell monolayers. Am J Physiol1986; 250:C222-7. 14. Goodman BE, Anderson JL, Clemens JW Evidence for regulation of sodium transport from airspace to vascular space by cAMP. Am J Physiol 1989; 257:L86-93. 15. Berthiaume Y. Effect of exogenous cAMP and aminophylline on alveolar and lung liquid clearance in anesthetized sheep. J Appl Physiol 1991; 70:2490-7. 16. Li M, McCann JD, Anderson MP, et 01. Regulation of chloride channels by protein kinase C in normal and cystic fibrosis airway epithelia. Science 1989; 244:1353-6. 17. Bertorello A, Aperia A. Na+-K+-ATPaseis an effector protein for protein kinase C in renal proximal tubule cells. Am J Physiol1989; 256:F370-3. 18. Dicker P, Rozengurt E. Phorbol ester stimulation of Na" influx and Na-K pump activity in swiss 3T3 cells. Biochem Biophys Res Commun 1981; 100:433-41. 19. Civan MM, Rubenstein D, Mauro T, O'Brien TG. Effects of tumor promoters on sodium ion transport across frog skin. Am J Physiol 1985; 248:C457-65. 20. Yanase M, Handler JS. Activators of protein kinase C inhibit sodium transport in A6 epithelia. Am J Physiol 1986; 250:C517-22. 21. Hays SR, Baum M, Kokko JP. Effects of protein kinase C activation on sodium, potassium, chloride and total CO 2 transport in the rabbit cortical collecting tubule. J Clin Invest 1987; 80:1561-70. 22. Nishizuka Y. Studies and perspectives of protein kinase C. Science 1986; 233:305-12. 23. Doumas BT, Bayse DD, Carter RJ, Peters T Jr, Schaffer R. A candidate reference method for determination of total protein in serum 1. Development and validation. Clin Chern 1981; 27: 1642-50. 24. Berthiaume Y, Broaddus CV, Gropper M, Tanita T, Onizuka M, Matthay MA. Alveolar liquid and protein clearance from the normal dog lung. J Appl Physiol 1988; 65:585-93. 25. Bell RM, Hannun Y, Loomis C. Mixed micelle assay of protein kinase C. Methods Enzymo11986; 124:353-9. 26. Walsh MP, Hinkins S, Dabrowska R, Hartshorne OJ. Smooth muscle myosin light chain kinase. Methods Enzymol 1983; 99:279-88. 27. Zar JH. Biostatistical analysis. Englewood Cliffs, NJ: Prentice-Hall, 1974; 130-81. 28. Perry M, Taylor A. Phorbol myristate acetateinduced injury of isolated perfused rat lungs: neutrophil dependence. J Appl Physiol1988; 65:2164-9. 29. Allison RC, Hernandez EM, Prasad VR, Grisham MB, Taylor AE. Protective effects of O 2 radical scavengers and adenosine in PMA-induced lung injury. J Appl Physiol 1988; 64:2175-82. 30. Johnson A. PMA-induced pulmonary edema: mechanisms of the vasoactive response. J Appl Physiol 1988; 65:2302-12. 31. Shasby DM, Vanbenthuysen KM, Tate RM, Shasby SS, McMurtry I, Repine JE. Granulocytes mediate acute edematous lung injury in rabbits and in isolated rabbit lungs perfused with phorbol myristate acetate: role of oxygen radicals. Am Rev Respir Dis 1982; 125:443-7. 32. Shasby DM, Shasby SS, Peach MV. Granulocytes and phorbol myristate acetate increase permeability to albumin of cultured endothelial mono-
BERTHIAUME, SAPIJASZKO, MACKENZIE, AND WALSH
1090 layers and isolated perfused lungs. Am Rev Respir Dis 1983; 127:72-6. 33. Dyer EL, Snapper VR. Role of circulating granulocytesin sheep lung injury produced by phorbol myristate acetate. J Appl Physiol 1986; 60: 576-89. 34. Barthelson RA, Jacoby DB, Widdicombe JH. Regulation of chloride secretion in dog tracheal epithelium by protein kinase C. Am J Physiol 1987; 253:C802-8. 35. Blumberg PM. In vitro studies on the mode of action of the phorbol esters, potent tumor promoters: Part 1. CRC Crit Rev Toxicol 1980; 153-97. 36. Kraft AS, Anderson WB. Phorbol esters increase the movement of Ca", phospholipid-dependent protein kinase associated with plasma membrane. Nature 1983; 301:621-3. 37. Jayr C, Leger N, Zeiter M, Matthay MA. AI-
veolar epithelial liquid clearance in anesthetized, ventilated rats is fast compared with studies in dogs, sheep, and rabbits (abstract). Am Rev Respir Dis 1990; 141:A300. 38. Matthay MA. Resolution of pulmonary edema. Mechanisms of liquid, protein, and cellular clearance from the lung. Clin Chest Med 1985; 6:521-45. 39. Jayr C, Leger N, Zelter M, Matthay MAo Effect of amiloride on alveolar and lung liquid clearance in ventilated rats (abstract). Am Rev Respir Dis 1991; 143:AI50. 40. WeikelCS, Sando n, Guerrant RL. Stimulation of porcine jejunal ion secretion in vivo by protein kinase C activators. J Clin Invest 1985; 76: 2430-5. 41. Sano K, VoelkerDR, Mason RJ. Involvement of protein kinase C in pulmonary surfactant secretion from alveolar type II cells. J Biol Chern 1985;
260:12725-9. 42. Ando Y, Jacobson HR, Breyer MD. Phorbol myristate acetate, dioctanoylglycerol, and phosphatidic acid inhibit the hydroosmotic effect of vasopressin on rabbit cortical collecting tubule. J Clin Invest 1987; 80:590-3. 43. Schlondorff 0, Levine SO. Inhibition of vasopressin-stimulated water flow in toad bladder by phorbol myristate acetate, dioctanoylglycerol, and RHC-80267. J Clin Invest 1985; 76:1071-8. 44. Minneman KP. ai-Adrenergic receptor subtypes, inositol phosphates, and sources of cellCa". Pharmacol Rev 1988; 40:87-119. 45. Mohrmann M, Cantiello HF, Ausiello DA. Inhibition of epithelial Na" transport byatriopeptin, protein kinase C, and pertussis toxin. Am J Physiol 1987; 253:F372-6.
