Hemoglobin
potentiates
oxidant
injury
in isolated rat lungs
ALLAN F. SEIBERT, AUBREY E. TAYLOR, JOHN B. BASS, AND JOHNSON HAYNES, JR. Pulmonary and Critical Care Division, Departments of Medicine and Physiology, University of South Alabama, Mobile, Alabama 36617
SEIBERT, ALLAN F., AUBREY E. TAYLOR, JOHN AND JOHNSON HAYNES, JR. Hemoglobin potentiates
B. BASS, oxidant injury in isolated rat lungs. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H1980-H1984,1991.Isolated perfused rat lungs were subjected to oxidant injury induced by tert-butyl hydroperoxide (t-buOOH), which caused a significant increase in capillary permeability as assessed by the change in the capillary filtration coefficient. t-buOOH caused an increase in the change in the capillary filtration coefficient (AK,) of 0.27 t 0.05 ml min cmH20-‘o 100 g lung tissue-’ (mean t SE) that was accompanied by an increase in thiobarbituric acid reactive products of lipid peroxidation in the lung perfusate. The addition of hemoglobin to the perfusate potentiated t-buOOH-induced lung injury as evidenced by a significantly greater (P = 0.007) AKfc of 0.43 k 0.05. t-buOOH also caused hemoglobin to release large quantities of free iron in vitro. The potentiation of tbuOOH-induced lung injury by hemoglobin was prevented by apotransferrin as evidenced by a significant reduction (P = 0.001) in AK,, to 0.13 k 0.02. No statistically significant (P > 0.05) changes in segmental resistances or pulmonary vascular pressures occurred in any of the lungs injured with t-buOOH when compared with time controls. These results demonstrate that t-buOOH causes an oxidant injury in isolated rat lungs that can be potentiated by free iron released from hemoglobin. l
oxidant lung injury; tert-butyl hydroperoxide; apotransferrin; lipid peroxidation; iron
EVIDENCE
IMPLICATING
OXIDANT
injury
Fenton
reagent;
in the pathogen-
within hemoglobin (Hb) with no open coordination site (21). An open coordination site is a stringent requirement for iron to catalyze hydroxyl radical formation (8). However, at least two reports indicate that Hb can act as a biological Fenton reagent by releasing catalytic iron when exposed to oxidant stress in vitro (9, 23). The possibility that Hb could release catalytic iron and potentiate oxidant injury in an isolated-perfused lung has not, to our knowledge, been investigated. In the present paper, we investigated the effects of Hb on oxidant injury due to tert-butyl hydroperoxide (t-buOOH) in an isolated rat lung perfused with a physiological solution that was free of significant numbers of formed blood elements. tbuOOH is a lipid soluble organic hydroperoxide that can directly oxidize membrane lipids and generate toxic oxygen metabolites in the process (6). t-buOOH was chosen as the oxidant, because we have previously studied lung injury due to t-buOOH in the rat lung, and iron-dependent oxidation was a contributing mechanism (15). Based on this observation and the in vitro observation that Hb can act as a biological Fenton reagent, we hypothesized that t-buOOH would cause the release of catalytic iron from Hb, which would accentuate injury in rat lungs perfused with t-buOOH and Hb. Our results indicate that t-buOOH causes Hb to act as a biological Fenton reagent in the isolated-perfused rat lung. METHODS
esis of several disease states of the lung has grown substantially in recent years. The reactive oxygen species Materials superoxide anion (010 ), hydrogen peroxide (H202), and NaCl, KCl, MgSO,, NaHC03, KH2P04, glucose, bovine hydroxyl radical ( .OH) are thought to be the primary serum albumin (Alb), bovine Hb, human apotransferrin oxygen species responsible for cytotoxic oxidant stress (TF), thiobarbituric acid, hydrochloric acid, butylated (1, 3, 11-17). Their immediate cytotoxicity, especially hydroxytoluene, Ferrozine, ascorbic acid, trichloroacetic the more reactive @OH,lies in the ability to initiate lipid acid, ammonium acetate, neocuproine, sodium dithionperoxidation of cell membranes (1, 16, 18, 25). Many in ate, and an iron standard solution were purchased from vitro investigations have demonstrated that ferrous iron Sigma Chemical (St. Louis, MO). Spectraphor dialysis catalyzes the production of .OH from hydrogen peroxide, membrane tubing was purchased from Thomas Scientific which can promote and initiate lipid peroxidation (2, lo(Atlanta, GA). 1420). Lipid peroxidation of endothelial cell membranes in the pulmonary capillary bed by cytotoxic oxygen spe- Isolated Perfused Lung cies is thought to be a biochemical mechanism responsible for producing the permeability pulmonary edema Male Charles River CD rats (250-350 g body wt, Charles River Breeding, Wilmington, MA) were anestheseen in the adult respiratory distress syndrome (1). tized with pentobarbital sodium (20-25 mg ip), and the Mammalian tissues possess a variety of antioxidant lungs were removed for extracorporeal perfusion as predefenses designed to prevent or protect against oxidant viously described (22). A tracheostomy was performed injury. One such defense mechanism is the compartmentalization of transitional metal ions like iron that are that permitted ventilation with a Harvard rodent venticapable of catalyzing the production of OH from HZOz. lator (model 683) at 55 breaths/min, with a tidal volume Approximately two-thirds of the total body iron is bound of 2.5 ml, and a positive end-expiratory pressure of 2 l
H1980
0363-6135/91
$1.50 Copyright
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
HEMOGLOBIN
IN OXIDANT
cmH20. The inspired gas mixture was 95% air-5% COz. Subsequently, a median sternotomy was performed, heparin (100 IU) was injected into the right ventricle, and cannulas were placed in the pulmonary artery and left ventricle. The heart, lungs, and mediastinal structures were removed en bloc and suspended from a Grass force displacement transducer (model FT03) for monitoring weight changes and were then placed into a humidified acrylic chamber. The lungs were perfused by a Gilson Minipuls 2 peristaltic pump using 4 g/100 ml Alb physiological salt solution (PSS) at a constant flow of 0.03 ml. g body wt-’ .min. The PSS-albumin perfusate (PSS-Alb) contained (in mM) 119 NaCl, 4.7 KCl, 1.17 MgSO,, 22.6 NaHCOz, 1.18 KHzPOd, 3.2 CaClz, and 5.5 glucose. The initial 75 ml of PSS-Alb, which contained residual blood cells and plasma, was perfused through the lungs in a nonrecirculating manner and discarded. An additional 50 ml of PSS-Alb was then used for recirculation. Pulmonary artery (Ppa) and pulmonary venous (P,J pressure were continuously monitored with Cope pressure transducers (model 041-500-503) and were recorded on a Grass polygraph recorder (model 7E). Zone 3 conditions (Ppa > PpV> alveolar pressures) were maintained throughout all experiments. Pulmonary Capillary Pressure The pulmonary capillary pressure (P,,) was estimated using the double-occlusion method. Arterial inflow and venous outflow lines were simultaneously occluded, and the equilibrium P,, and P,, were measured. This equilibration pressure is well correlated with isogravimetric measures of P,, and also reflects the prevailing capillary pressure when the lung is not isogravimetric. With the use of Ppc,the pulmonary arterial, or precapillary, resistance (R,) and venous, or postcapillary, resistance (R,) were calculated using the following equations and total flow (Q): R, = (P,, - P&/Q and R, = (Ppc - P&/Q.
INJURY
H1981
of perfusate was reduced with 0.02% ascorbic acid for 5 min at room temperature. One-half milliliter of 11.3% trichloroacetic acid was added, mixed, and centrifuged at 1,000 g for 5 min. One milliliter of clear supernatant was removed, and 0.4 ml of 10% ammonium acetate and 0.1 ml of mixed ferroan color reagent was added. The mixed ferroan color reagent contained Ferrozine and neocuproine as previously described. Absorption was measured on a Perkin-Elmer spectrophotometer at 562 nm. Results were expressed as nanomoles per 50 ml. Measurement of Thiobarbituric Acid Reactive Products of Lipid Peroxidation Thiobarbituric acid reactive products of lipid peroxidation (TBAR) were measured with a method previously described (20). Briefly, 1 ml of perfusate from lungs injured with varying concentrations of t-buOOH (see Specific Protocols) was vortexed with 1 ml of a suspension containing trichloroacetic acid (l5%), thiobarbituric acid (0.67%), and hydrochloric acid (0.25 N). Fifteen microliters of butylated hydroxytoluene in 95% ethanol was then added and vortexed, and the sample was incubated in boiling water for 20 min. After boiling was completed, the sample was cooled and centrifuged at 2,000 g for 10 min, and the absorbance of the supernatant was read at 532 nm. Quantitation was based on the molar extinction coefficient of 1.56 X 10’. Reduction of Hb Commercially available Hb contains a mixture of oxidized and reduced Hb; therefore, all Hb was reduced by a modification of a previously described method (19). Briefly, Hb (2.5 mM) was dialized against sodium dithionate (25 mM) in Spectraphor dialysis tubing. The absence of oxidized Hb was determined spectrophotometrically by measurement of absorbance from 500 to 630 nm.
Capillary Filtration Coefficient The capillary filtration coefficient (Kfc) was measured using the method previously described (7). After an isogravimetric period in which the lung is not losing or gaining weight, the outflow pressure (P,J was rapidly elevated by 8 cmH20 for 12 min. The increase in lung weight was recorded and characterized by a rapid weight gain (vascular filling) phase (O-7 min) followed by a slower rate of weight gain (7-12 min). The rate of weight change (AW/At) during the 7- to 12-min interval was analyzed using linear regression techniques after the log,, transformation of the rate of weight change (AW) per minute (At). The initial rate of weight gain was calculated by extrapolation of AW/At to time 0 and taking the antiloglo of this AW/At intercept. The Kfc was calculated by dividing the AWlAt at time 0 by the change in P,, that occurred after increasing venous outflow pressure. The Kfc (expressed as ml. min-’ . cmHnO-‘. 100 g lung tissue-‘) was normalized using the baseline wet lung weight. Measurement of Ferrous Iron Ferrous iron was spectrophotometrically measured as described by Carter (5) using ferrocene. Briefly, 0.5 ml
Specific Protocols Determination of ferrous iron in perfusates. Ferrous iron was measured (see METHODS) in 0.5-ml aliquots from four groups (n = 8 for each group) of perfusates at 37°C in vitro. Group 1 was control [PSS + Alb (4 g/100 ml)]. Group 2 was t-buOOH challenge (PSS + Alb + 0.89 mM t-buOOH). Group 3 was Hb only (PSS + Alb + 50 PM Hb), andgroup 4 was Hb + t-buOOH challenge (PSS + Alb + 50 PM Hb + 0.89 mM t-buOOH). In perfusates containing t-buOOH, iron was measured 20 min after the addition of t-buOOH. Production of TBAR by t-buOOH-induced oxidant lung injury. Lungs were isolated from five rats and were perfused with PSS-Alb and ventilated with 95% air-5% COz. Lungs were perfused with 0.45, 0.89, 4.5, 45, or 90 mM t-buOOH. After the 30-min equilibration period, tbuOOH was added to the reservoir and allowed to recirculate for 20 min. Aliquots were removed in triplicate from the reservoir immediately after the addition of tbuOOH and 20 min later. TBAR were measured in these perfusate aliquots as described (see METHODS). Results were expressed as the difference in TBAR (nmol/50 ml perfusate) before and after perfusion with t-buOOH.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
H1982
HEMOGLOBIN
IN
Role of Hb in lung injury due to t-buOOH. Lungs were isolated from 34 rats, perfused with PSS-Alb, and ventilated with a 95% air-5% CO2 gas mixture. The lungs were divided into four groups: 1) time control (n = 8); 2) 0.89 mM t-buOOH alone (n = 10); 3) t-buOOH + 50 PM Hb (n = 10); and 4) t-buOOH + 50 PM Hb + 10 PM TF (n = 6). Hb and TF, in lungs perfused with these agents, were added to the perfusate reservoir at the onset of the 30-min equilibration period. The time control group received no drugs. After the 30-min equilibration period in each group, baseline hemodynamic profiles were obtained under isogravimetric and zone 3 conditions, and Kfc was measured. t-buOOH was then added to the perfusate reservoir in a final circulating concentration of 0.89 mM and recirculated for 20 min. The P, was adjusted to maintain an isogravimetric state, and the hemodynamic profile and Kfc were remeasured. In the time control group, a hemodynamic profile and Kfc were obtained after the 30-min period of equilibration and 20 min later. Results are expressed as the change (A) in Ppa, Ppv, Ppc, R,, R, and Kfc from baseline. Statistics Results are presented as means t SE. Statistical analysis was done using a one-way analysis of variance (ANOVA). The unpaired Student’s t test was used when ANOVA indicated a within-group statistical significance. Differences were considered to be significant when P c 0.05.
OXIDANT
INJURY
TBAR Production in t-buOOH-Induced Oxidant Lung Injury Figure 2 depicts the quantity of TBAR produced as a function of the log of t-buOOH (in mM) in the perfusate from five lungs. In lungs perfused with 0.45,0.89,4.5,45, and 90 mM t-buOOH, the measured TBAR (nmol/50 ml) was 1.31, 4.97, 7.01, 15.02, and 17.81, respectively. There was a linear relationship between products of lipid peroxidation and the log of the t-buOOH concentration in the perfusate (r = 0.97). Effect of Hb and TF on AK, in t-buOOH Lung Injury Figure 3 compares the AK, (post-t-buOOH KfC - baseline KfC) seen in time controls (-0.008 t 0.02), t-buOOH alone (0.27 t 0.05), t-buOOH + Hb pretreatment (0.43 + 0.05), and t-buOOH + Hb and TF pretreatment (0.13 7 0.02). Baseline KfC in these four groups was 0.31 t 0.04, 0.33 t 0.04, 0.26 t 0.03, and 0.33 t 0.05, respectively. A significant increase (P = 0.007) in the AKfc was observed in the t-buOOH + Hb compared with the tbuOOH alone group. A significant decrease (P = 0.001) in AKfc was observed in the t-buOOH + Hb + TF compared with the t-buOOH + Hb group. In addition, the decrease in KfC in the Hb + TF + t-buOOH was also significant (P = 0.0319) when compared with t-buOOH alone.
iY Q a t-
RESULTS
Quantity of Ferrous Iron in Perfusates
0.1
Figure 1 compares the amount of free iron measured in PSS + Alb (274.9 t 37.6 nmol/50 ml) with PSS + Alb + t-buOOH (177.1 t 31.2 nmol/50 ml), PSS + Alb + Hb (216.1 t 21.2 nmol/50 ml), and PSS + Alb + Hb + tbuOOH (1,052.5 t 51.6 nmol/50 ml). The measured iron concentration was significantly higher (P = 0.001) in the perfusate containing PSS + Alb + Hb + t-buOOH than in all other perfusates. No significant differences (P > 0.05) in the measured free iron was observed among the other perfusates.
J
-3
*
‘2ooT A
rtbuOOH1 L
2. Thiobarbituric acid reactive products (TBAR) production in isolated rat lungs perfused with t-buOOH. TBAR was measured in triplicate from 5 rats before and after t-buOOH perfusion for 20 min and depicted as quantity of TBAR produced by concentration of tbuOOH in perfusate. FIG.
,
-0.1
PSS+ ALB
PSS+ ALB+ tbuOOH
PSS+ ALB+ Hb
PSS+ ALB+Hb tbuOOH
+
FIG. 1. Free iron (Fe”) concentration in 4 different perfusates (n = 8 in each perfusate). There was a significant increase (P = 0.001) in Fe”’ in perfusate containing physiological salt solution (PSS) + albumin (Alb) + hemoglobin (Hb) that had been incubated with 0.89 mM tert-butyl hydroperoxide (t-buOOH). t-buOOH incubation time was 20 min in 2 perfusates exposed to t-buOOH. Data are means t SE. *P = 0.001 compared with all other perfusates.
1
time control n=8
tbuOOH alone n=lO
tbuOOH +Hb n=lO
tbuOOH +Hb+TF n=6
FIG. 3. Effect of Hb and apotransferrin (TF) on capillary filtration coefficient (KfC) changes due to t-buOOH. Three groups of isolated lungs were subjected to recirculating 0.89 mM t-buOOH. KfC was measured before and after 20 min of perfusion with t-buOOH and depicted as AK,,. Kfc was measu .red in time controls at end of equilibration period and 20 min later without addition of t-buOOH. Lungs pretreated with 50 PM Hb demonstrated a potentiation of injury due to t-buOOH evidenced by a significantly higher (P = 0.007) AK,, after Hb pretreatment. TF pretreatment resulted in a prevention of Hbassociated potentiation of t-buOOH injury as evidenced by significant reduction (P = 0.0001) of AK,, compared with lungs pretreated with Hb alone. Data are means k SE. *P = 0.007 compared with t-buOOH alone. 1-P = 0.007 compared with t-buOOH + Hb and P = 0.0319 compared with t-buOOH alone.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
HEMOGLOBIN
IN
OXIDANT
H1983
INJURY
Effect of Hb and TF on Hemodynamic Response to t-buOOH
observed increase in pulmonary capillary permeability is the endothelial cell membrane. Disruption of the endothelial cell membrane by lipid peroxidation would explain Figures 4 and 5 deplict the change in segmental resist- the increase in capillary filtration coefficient measured antes and pulmonary vascular pressures from time conin lungs exposed to t-buOOH. trols and lungs injured with t-buOOH. Although arterial This study also demonstrated that oxidant injury resistance tended to rise and venous resistance and cap- caused by t-buOOH was significantly potentiated by the illary pressures tended to fall in lu.ngs pretreated with presence of hemoglobin and that t-buOOH causes heHb before t-bu .OOH injury, there were no significant moglobin to release free iron. Free iron is capable of differences (P > 0.05) in any of the hemodynamic meas- catalyzing the production of *OH from HZOZ, and in our urements between any groups or when compared with experiments the perfusate containing the largest amount time controls. of free iron was associated with a significant potentiation of t-buOOH-induced lung injury. To further support the DISCUSSION hypothesis that catalytic iron released from hemoglobin was the cause of the potentiated injury, we demonstrated Our results demonstrate that t-buOOH, an organic hydroperoxide with significant oxidant potential, causes that apotransferrin prevented the hemoglobin-associated potentiation (Fig. 3). Although less significant, the cona large increase in the permeability of the pulmonary capillary bed (Fig. 3). This permeability-type injury is centration of apotransferrin used also attenuated the accompanied by evidence of lipid peroxidation in the lung injury when compared with t-buOOH alone. This latter observation implies that the iron contamination of perfusate from lungs exposed to t-buOOH. The quantity commercially available bovine albumin (40-70 nmol Fe/ of lipid peroxidation products detected is well correlated g Alb) is important in determining the degree of injury with the quantity of t-buOOH used to produce the injury due to t-buOOH. These experiments clearly show that (Fig. 2). t-buOOH apparently causes lung injury by lipid peroxidation that is manifested as an increase in the hemoglobin accentuates t-buOOH injury and that the permeability of the pulmonary capillary bed. The most most likely mechanism for this potentiation is the tlikely site of lipid peroxidation that would explain the buOOH-induced release of catalytic iron from hemoglobni . Our data have linked injury potentiation to the avail9.OE-2 ability of free iron but have not identified @OHproduction as the specific mechanism of potentiation. We have E 6.OE-2 previously shown that dimethyl sulfoxide, a *OH scav.-SI 3.OE-2 enger, has little effect on t-buOOH-induced injury in the E isolated rat lung (15). However, iron is also capable of 0 catalyzing the production of alkoxy and peroxy radicals 0.0 I? from lipid peroxides and the production these reactive E -3.OE-2 species could contribute to the observed injury (11). The 0 capability of iron to produce @OH is widely recognized, -6.OE-2 but, in fact, iron is a nonspecific catalyst that facilitates the production of more reactive species from less reactive -9.OE-2 1 species (14). Previous studies have provided evidence E’IG. 4. Effect of Hb and TF on segmental resistances in 4 groups that hydrogen peroxide, organic hydroperoxides, and suof rat lungs. Solid bar, time controls (n = 8); left diagonal, t-buOOH peroxide anion-generating systems can cause hemoglobin alone (n = 10); right diagonal, t-buOOH + Hb (n = 10); crosshatched, to release free iron (9, 23). This released iron is capable t-buOOH + Hb + TF (n = 6). Pulmonary arterial (R,)and venous of catalyzing OH production and initiating lipid peroxresistances (fiV) were calculated before and after perfusion with tbuOOH and in time controls and are expressed as change in (;1)R, and idation in vitro (2). In addition, transferrin, haptoglobin, AR,.. There were no significant differences (P > 0.05) in AP, or AR. and desferrioxamine have been shown to prevent the from any group compared with time controls. Data are means t SE. iron-catalyzed .OH formation by binding all open iron coordination sites (13, 14). An open coordination site must be available for iron to catalyze OH formation (8). There is very little iron with catalytic capability in normal mammalian tissues (16). The majority of iron is bound to hemoglobin or to other macromolecules where it is not available to act as a catalyst (21). This extensive binding or compartmentalization of iron is an important antioxidant defense mechanism in humans. Only when 1 exposed to an oxidant stress, as in the present study, is FIG. 5. Effect of Hb and TF on pulmonary vascular pressures in 4 the iron reservoir in hemoglobin available as a Fentongroups of rat lungs. Solid bar, time control (n = 8); left diagonal, ttype catalyst. huOOH alone (n = 10); right diagonal, t-buOOH + Hb (n = 10); crosshatched, t-buOOH + Hb + TF (n = 6). Pulmonary artery (P&, This reservoir of heme iron is also compartmentalized venous (PI,\), and capillary (P,,,.) pressures were measured before and within the erythrocyte, which further limits its ability to after perfusion with t-buOOH and in time controls and were expressed act as a catalyst. Erythrocytes are rich in antioxidant as change in (A)PIjB) AP1,V, and API,,. There were no significant differenzymes such as superoxide dismutase, catalase, and ences (P > 0.05) in API,,, APpx., or API,, from any group compared with time controls. Data are means t SE. glutathione peroxidase and have been shown to decrease
ARa
ARv
l
l
-0.6
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
H1984
HEMOGLOBIN
IN
H202-mediated damage to isolated rat lungs, rat hearts, and cultured bovine pulmonary artery endothelial cells (4,16,26). The protective effect of erythrocytes in H20zinduced injury may be due to the antioxidant enzymes they contain. Catalase activity is specific for hydrogen peroxide, and it is relatively ineffective at dismutating larger, organic hydroperoxides like t-buOOH (16). Glutathione peroxidase in the presence of glutathione can reduce nonmembrane-bound organic hydroperoxides to water (14, 27). These antioxidant enzymes are in intimate proximity to hemoglobin within the cytosol of the erythrocyte. Although hemoglobin can release catalytic iron when exposed to oxidant stress, the injurious potential of iron-catalyzed free radical production is limited by the cytosolic antioxidant enzymes before the free radicals can gain access to the red cell membrane or extracellular space. t-buOOH has been shown to cause erythrocyte hemolysis, which could theoretically result in the release of catalytic iron into the extracellular space in the absence of haptoglobin and transferrin (28). Erythrocytes possess antioxidant potential and compartmentalize the heme-iron reservoir, but our experiments were performed in isolated lungs that were virtually free of erythrocytes and the antioxidant enzymes they contain. In summary, our experiments provide evidence that 1) t-buOOH causes injury in isolated rat lungs accompanied by evidence of lipid peroxidation within the lung, 2) tbuOOH causes the release of catalytic iron from hemoglobin, 3) hemoglobin potentiates lung injury due to tbuOOH, and 4) apotransferrin prevents the hemoglobinassociated accentuation of lung injury due to t-buOOH. These findings are consistent with the hypothesis that free hemoglobin, when exposed to an oxidant stress, can act as a biological Fenton reagent in the isolated rat lung. We thank Sandy Mead for preparation of the manuscript and Dr. Joe McCord for technical advice. This work was supported by a Florence Foundation Research Career Development grant, National Heart, Lung, and Blood Institute Clinical Investigator Award HL-02352 (to J. Haynes, Jr.), and a University of South Alabama Intramural grant (to A. F. Seibert). Address for reprint requests: A. F. Seibert, 2451 Fillingim St., Mastin Bldg., Rm. 312, Univ. of South Alabama, Mobile, AL 36617. Received
17 September
1990; accepted
in final
form
27 February
1991.
REFERENCES 1. BERTRAND, Y. Oxygen-free radicals and lipid peroxidation in adult respiratory distress syndrome. Intensive Care Med. 11: 56-60,1985. 2. BRAUGHLER, J. M., L. A. DUNCAN, AND R. L. CHASE. The involvement of iron in lipid peroxidation. J. Biol. Chem. 261: 10282-10289, 1986. 3. BRIGHAM, K. L., AND B. MEYRICK. Interactions of granulocytes with the lungs. Circ. Res. 54: 623-635, 1984. 4. BROWN J. M., M. A. GRASSO, L. S. TERADA, L. J. BEEHLER, K. M. TOTH, G. J. WHITMAN, A. H. HARKEV, AND J. E. REPINE. Erythrocytes decrease myocardial hydrogen peroxide levels and reperfusion injury. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): H584-H588,1989. 5. CARTER, P. Spectrophotometric determination of serum iron at the submicrogram level with a new reagent (Ferrozine). Anal. Biochem. 40: 450-458, 1971.
OXIDANT
INJURY
6. CHANCE, B., H. SIES, AND A. BOVERIS. Hydroperoxide metabolism in mammalian organs. Physiol. Reu. 59: 527-605, 1979. 7. DRAKE, R., K. A. GAAR, AND A. E. TAYLOR. Estimation of the filtration coefficient of pulmonary exchange vessels. Am. J. Physiol. 234 (Heart Circ. Physiol. 3): H266-H274, 1978. 8. GRAF, E., J. R. MAHONEY, R. G. BRYANT, AND J. W. EATON. Iron catalyzed hydroxyl radical formation. Stringent requirement for the free iron coordination site. J. Biol. Chem. 259: 3620-3624,1984. 9. GUTTERIDGE, J. M. C. Iron promoters of the Fenton reaction and lipid peroxidation can be released from haemoglobin by peroxides. FEBS Lett. 201: 291-295,1986. 10. GUTTERIDGE, J. M. C., S. K. PATERSON, A. W. SEGAL, AND B. HALLIWELL. Inhibition of lipid peroxidation by the iron binding protein lactoferrin. Biochem. J. 199: 259-261, 1981. 11. HALLIWELL, B. Oxidants and human disease, some new concepts. FASEB J. 1: 358-364, 1987. 12. HALLIWELL, B., AND J. M. C. GUTTERIDGE. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219: l14, 1984. 13. HALLIWELL, B., AND J. M. C. GUTTERIDGE. The importance of free radicals and catalytic metal ions in human disease. IMoL. Aspects Med. 8: 89-193, 1985. 14. HALLIWELL, B., AND J. M. C. GUTTERIDGE. Oxygen free radicals and iron in relation to biology and medicine: some problems and concepts. Arch. Biochem. Biophys. 246: 501-514, 1986. 15. HAYNES, J., A. SEIBERT, J. BASS, AND A. TAYLOR. U74500A inhibition of oxidant-mediated lung injury. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H144-H148, 1990. 16. HEFFNER, J. E., AND J. E. REPINE. Pulmonary strategies of antioxidant defense. Am. Reu. Respir. Dis. 140: 531-554, 1989. 17. JACKSON, R. M., C. F. VEAL, B. ALEXANDER, A. L. BRANNEN, AND J. D. FULMER. Re-expansion pulmonary edema: a potential role for free radicals in its pathogenesis. Am. Reu. Respir. Dis. 137: 1165-1171,1984. 18. KURODA, M., K. MURAKAMI, AND Y. ISHIKAWA. Role of hydroxyl radicals derived from granulocytes in lung injury induced by phorbol myristate acetate. Am. Reu. Respir. Dis. 136: 1435-1444, 1987. 19. MARTIN, W., G. M. VILLANI, D. JOTHIANANDAN, AND R. F. FURCHGOTT. Selective blockage of endothelium-dependent and glyceryl trinitrate-induced relaxation by hemoglobin and by methylene blue in rabbit aorta. J. Pharmacol. Exp. Ther. 232: 708-716, 1984. 20. MINOHI, G., AND S. D. AUST. The requirement for iron (III) in the initiation of lipid peroxidation by iron (II) and hydrogen peroxide. J. Biol. Chem. 262: 1098-1104,1987. 21. NORWOOD, M. The clinical biochemistry of iron (Abstract). Semin.
Hematol. 14: 31, 1977. 22. PERRY, M., AND A. E. TAYLOR. Phorbol myristate acetate-induced injury of isolated perfused rat lungs: neutrophil dependence. J. Appl. Physiol. 65: 2164-2169, 1988. 23. SADRZADEH, S., E. GRAF, S. PANTER, P. E. HALLAWAY, AND J. W. EATON. Hemoglobin. A biologic Fenton reagent. J. Biol. Chem. 259: 14354-14356,1984. 24. SAID, S. I., AND H. D. FODA. Pharmacologic modulation of lung injury. Am. Reu. Respir. Dis. 139: 1553-1564, 1989. 25. TATE, R. M., K. M. VANBENTHUYSEN, D. M. SHASBY, I. F. MCMURTRY, AND J. E. REPINE. Oxygen-radical-mediated permeability edema and vasoconstriction in isolated perfused rabbit lungs. Am. Reu. Respir. Dis. 126: 802-806, 1982. 26. TOTH, K. M., D. P. CLIFFORD, E. M. BERGER, C. W. WHITE, AND J. E. REPINE. Intact human erythrocytes prevent hydrogen-peroxide mediated damage to isolated perfused rat lungs and cultured bovine pulmonary artery endothelial cells. J. Clin. Inuest. 74: 292295,1984. 27. TRAVIS, J. Oxidants and antioxidants in the lung. Am. Reu. Respir.
Dis. 135: 773-774,1987. 28. TROTTA, R. J., S. G. SULLIVAN, AND A. STERN. Lipid peroxidation and haemoglobin degradation in red blood cells exposed to t-butyl hydroperoxide. The relative roles of haem- and glutathione-dependent decomposition of t-butyl hydroperoxide and membrane lipid hydroperoxides in lipid peroxidation and haemolysis. Biochem. J. 212: 759-772, 1983.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
potentiates
oxidant
injury
in isolated rat lungs
ALLAN F. SEIBERT, AUBREY E. TAYLOR, JOHN B. BASS, AND JOHNSON HAYNES, JR. Pulmonary and Critical Care Division, Departments of Medicine and Physiology, University of South Alabama, Mobile, Alabama 36617
SEIBERT, ALLAN F., AUBREY E. TAYLOR, JOHN AND JOHNSON HAYNES, JR. Hemoglobin potentiates
B. BASS, oxidant injury in isolated rat lungs. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H1980-H1984,1991.Isolated perfused rat lungs were subjected to oxidant injury induced by tert-butyl hydroperoxide (t-buOOH), which caused a significant increase in capillary permeability as assessed by the change in the capillary filtration coefficient. t-buOOH caused an increase in the change in the capillary filtration coefficient (AK,) of 0.27 t 0.05 ml min cmH20-‘o 100 g lung tissue-’ (mean t SE) that was accompanied by an increase in thiobarbituric acid reactive products of lipid peroxidation in the lung perfusate. The addition of hemoglobin to the perfusate potentiated t-buOOH-induced lung injury as evidenced by a significantly greater (P = 0.007) AKfc of 0.43 k 0.05. t-buOOH also caused hemoglobin to release large quantities of free iron in vitro. The potentiation of tbuOOH-induced lung injury by hemoglobin was prevented by apotransferrin as evidenced by a significant reduction (P = 0.001) in AK,, to 0.13 k 0.02. No statistically significant (P > 0.05) changes in segmental resistances or pulmonary vascular pressures occurred in any of the lungs injured with t-buOOH when compared with time controls. These results demonstrate that t-buOOH causes an oxidant injury in isolated rat lungs that can be potentiated by free iron released from hemoglobin. l
oxidant lung injury; tert-butyl hydroperoxide; apotransferrin; lipid peroxidation; iron
EVIDENCE
IMPLICATING
OXIDANT
injury
Fenton
reagent;
in the pathogen-
within hemoglobin (Hb) with no open coordination site (21). An open coordination site is a stringent requirement for iron to catalyze hydroxyl radical formation (8). However, at least two reports indicate that Hb can act as a biological Fenton reagent by releasing catalytic iron when exposed to oxidant stress in vitro (9, 23). The possibility that Hb could release catalytic iron and potentiate oxidant injury in an isolated-perfused lung has not, to our knowledge, been investigated. In the present paper, we investigated the effects of Hb on oxidant injury due to tert-butyl hydroperoxide (t-buOOH) in an isolated rat lung perfused with a physiological solution that was free of significant numbers of formed blood elements. tbuOOH is a lipid soluble organic hydroperoxide that can directly oxidize membrane lipids and generate toxic oxygen metabolites in the process (6). t-buOOH was chosen as the oxidant, because we have previously studied lung injury due to t-buOOH in the rat lung, and iron-dependent oxidation was a contributing mechanism (15). Based on this observation and the in vitro observation that Hb can act as a biological Fenton reagent, we hypothesized that t-buOOH would cause the release of catalytic iron from Hb, which would accentuate injury in rat lungs perfused with t-buOOH and Hb. Our results indicate that t-buOOH causes Hb to act as a biological Fenton reagent in the isolated-perfused rat lung. METHODS
esis of several disease states of the lung has grown substantially in recent years. The reactive oxygen species Materials superoxide anion (010 ), hydrogen peroxide (H202), and NaCl, KCl, MgSO,, NaHC03, KH2P04, glucose, bovine hydroxyl radical ( .OH) are thought to be the primary serum albumin (Alb), bovine Hb, human apotransferrin oxygen species responsible for cytotoxic oxidant stress (TF), thiobarbituric acid, hydrochloric acid, butylated (1, 3, 11-17). Their immediate cytotoxicity, especially hydroxytoluene, Ferrozine, ascorbic acid, trichloroacetic the more reactive @OH,lies in the ability to initiate lipid acid, ammonium acetate, neocuproine, sodium dithionperoxidation of cell membranes (1, 16, 18, 25). Many in ate, and an iron standard solution were purchased from vitro investigations have demonstrated that ferrous iron Sigma Chemical (St. Louis, MO). Spectraphor dialysis catalyzes the production of .OH from hydrogen peroxide, membrane tubing was purchased from Thomas Scientific which can promote and initiate lipid peroxidation (2, lo(Atlanta, GA). 1420). Lipid peroxidation of endothelial cell membranes in the pulmonary capillary bed by cytotoxic oxygen spe- Isolated Perfused Lung cies is thought to be a biochemical mechanism responsible for producing the permeability pulmonary edema Male Charles River CD rats (250-350 g body wt, Charles River Breeding, Wilmington, MA) were anestheseen in the adult respiratory distress syndrome (1). tized with pentobarbital sodium (20-25 mg ip), and the Mammalian tissues possess a variety of antioxidant lungs were removed for extracorporeal perfusion as predefenses designed to prevent or protect against oxidant viously described (22). A tracheostomy was performed injury. One such defense mechanism is the compartmentalization of transitional metal ions like iron that are that permitted ventilation with a Harvard rodent venticapable of catalyzing the production of OH from HZOz. lator (model 683) at 55 breaths/min, with a tidal volume Approximately two-thirds of the total body iron is bound of 2.5 ml, and a positive end-expiratory pressure of 2 l
H1980
0363-6135/91
$1.50 Copyright
0 1991 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
HEMOGLOBIN
IN OXIDANT
cmH20. The inspired gas mixture was 95% air-5% COz. Subsequently, a median sternotomy was performed, heparin (100 IU) was injected into the right ventricle, and cannulas were placed in the pulmonary artery and left ventricle. The heart, lungs, and mediastinal structures were removed en bloc and suspended from a Grass force displacement transducer (model FT03) for monitoring weight changes and were then placed into a humidified acrylic chamber. The lungs were perfused by a Gilson Minipuls 2 peristaltic pump using 4 g/100 ml Alb physiological salt solution (PSS) at a constant flow of 0.03 ml. g body wt-’ .min. The PSS-albumin perfusate (PSS-Alb) contained (in mM) 119 NaCl, 4.7 KCl, 1.17 MgSO,, 22.6 NaHCOz, 1.18 KHzPOd, 3.2 CaClz, and 5.5 glucose. The initial 75 ml of PSS-Alb, which contained residual blood cells and plasma, was perfused through the lungs in a nonrecirculating manner and discarded. An additional 50 ml of PSS-Alb was then used for recirculation. Pulmonary artery (Ppa) and pulmonary venous (P,J pressure were continuously monitored with Cope pressure transducers (model 041-500-503) and were recorded on a Grass polygraph recorder (model 7E). Zone 3 conditions (Ppa > PpV> alveolar pressures) were maintained throughout all experiments. Pulmonary Capillary Pressure The pulmonary capillary pressure (P,,) was estimated using the double-occlusion method. Arterial inflow and venous outflow lines were simultaneously occluded, and the equilibrium P,, and P,, were measured. This equilibration pressure is well correlated with isogravimetric measures of P,, and also reflects the prevailing capillary pressure when the lung is not isogravimetric. With the use of Ppc,the pulmonary arterial, or precapillary, resistance (R,) and venous, or postcapillary, resistance (R,) were calculated using the following equations and total flow (Q): R, = (P,, - P&/Q and R, = (Ppc - P&/Q.
INJURY
H1981
of perfusate was reduced with 0.02% ascorbic acid for 5 min at room temperature. One-half milliliter of 11.3% trichloroacetic acid was added, mixed, and centrifuged at 1,000 g for 5 min. One milliliter of clear supernatant was removed, and 0.4 ml of 10% ammonium acetate and 0.1 ml of mixed ferroan color reagent was added. The mixed ferroan color reagent contained Ferrozine and neocuproine as previously described. Absorption was measured on a Perkin-Elmer spectrophotometer at 562 nm. Results were expressed as nanomoles per 50 ml. Measurement of Thiobarbituric Acid Reactive Products of Lipid Peroxidation Thiobarbituric acid reactive products of lipid peroxidation (TBAR) were measured with a method previously described (20). Briefly, 1 ml of perfusate from lungs injured with varying concentrations of t-buOOH (see Specific Protocols) was vortexed with 1 ml of a suspension containing trichloroacetic acid (l5%), thiobarbituric acid (0.67%), and hydrochloric acid (0.25 N). Fifteen microliters of butylated hydroxytoluene in 95% ethanol was then added and vortexed, and the sample was incubated in boiling water for 20 min. After boiling was completed, the sample was cooled and centrifuged at 2,000 g for 10 min, and the absorbance of the supernatant was read at 532 nm. Quantitation was based on the molar extinction coefficient of 1.56 X 10’. Reduction of Hb Commercially available Hb contains a mixture of oxidized and reduced Hb; therefore, all Hb was reduced by a modification of a previously described method (19). Briefly, Hb (2.5 mM) was dialized against sodium dithionate (25 mM) in Spectraphor dialysis tubing. The absence of oxidized Hb was determined spectrophotometrically by measurement of absorbance from 500 to 630 nm.
Capillary Filtration Coefficient The capillary filtration coefficient (Kfc) was measured using the method previously described (7). After an isogravimetric period in which the lung is not losing or gaining weight, the outflow pressure (P,J was rapidly elevated by 8 cmH20 for 12 min. The increase in lung weight was recorded and characterized by a rapid weight gain (vascular filling) phase (O-7 min) followed by a slower rate of weight gain (7-12 min). The rate of weight change (AW/At) during the 7- to 12-min interval was analyzed using linear regression techniques after the log,, transformation of the rate of weight change (AW) per minute (At). The initial rate of weight gain was calculated by extrapolation of AW/At to time 0 and taking the antiloglo of this AW/At intercept. The Kfc was calculated by dividing the AWlAt at time 0 by the change in P,, that occurred after increasing venous outflow pressure. The Kfc (expressed as ml. min-’ . cmHnO-‘. 100 g lung tissue-‘) was normalized using the baseline wet lung weight. Measurement of Ferrous Iron Ferrous iron was spectrophotometrically measured as described by Carter (5) using ferrocene. Briefly, 0.5 ml
Specific Protocols Determination of ferrous iron in perfusates. Ferrous iron was measured (see METHODS) in 0.5-ml aliquots from four groups (n = 8 for each group) of perfusates at 37°C in vitro. Group 1 was control [PSS + Alb (4 g/100 ml)]. Group 2 was t-buOOH challenge (PSS + Alb + 0.89 mM t-buOOH). Group 3 was Hb only (PSS + Alb + 50 PM Hb), andgroup 4 was Hb + t-buOOH challenge (PSS + Alb + 50 PM Hb + 0.89 mM t-buOOH). In perfusates containing t-buOOH, iron was measured 20 min after the addition of t-buOOH. Production of TBAR by t-buOOH-induced oxidant lung injury. Lungs were isolated from five rats and were perfused with PSS-Alb and ventilated with 95% air-5% COz. Lungs were perfused with 0.45, 0.89, 4.5, 45, or 90 mM t-buOOH. After the 30-min equilibration period, tbuOOH was added to the reservoir and allowed to recirculate for 20 min. Aliquots were removed in triplicate from the reservoir immediately after the addition of tbuOOH and 20 min later. TBAR were measured in these perfusate aliquots as described (see METHODS). Results were expressed as the difference in TBAR (nmol/50 ml perfusate) before and after perfusion with t-buOOH.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
H1982
HEMOGLOBIN
IN
Role of Hb in lung injury due to t-buOOH. Lungs were isolated from 34 rats, perfused with PSS-Alb, and ventilated with a 95% air-5% CO2 gas mixture. The lungs were divided into four groups: 1) time control (n = 8); 2) 0.89 mM t-buOOH alone (n = 10); 3) t-buOOH + 50 PM Hb (n = 10); and 4) t-buOOH + 50 PM Hb + 10 PM TF (n = 6). Hb and TF, in lungs perfused with these agents, were added to the perfusate reservoir at the onset of the 30-min equilibration period. The time control group received no drugs. After the 30-min equilibration period in each group, baseline hemodynamic profiles were obtained under isogravimetric and zone 3 conditions, and Kfc was measured. t-buOOH was then added to the perfusate reservoir in a final circulating concentration of 0.89 mM and recirculated for 20 min. The P, was adjusted to maintain an isogravimetric state, and the hemodynamic profile and Kfc were remeasured. In the time control group, a hemodynamic profile and Kfc were obtained after the 30-min period of equilibration and 20 min later. Results are expressed as the change (A) in Ppa, Ppv, Ppc, R,, R, and Kfc from baseline. Statistics Results are presented as means t SE. Statistical analysis was done using a one-way analysis of variance (ANOVA). The unpaired Student’s t test was used when ANOVA indicated a within-group statistical significance. Differences were considered to be significant when P c 0.05.
OXIDANT
INJURY
TBAR Production in t-buOOH-Induced Oxidant Lung Injury Figure 2 depicts the quantity of TBAR produced as a function of the log of t-buOOH (in mM) in the perfusate from five lungs. In lungs perfused with 0.45,0.89,4.5,45, and 90 mM t-buOOH, the measured TBAR (nmol/50 ml) was 1.31, 4.97, 7.01, 15.02, and 17.81, respectively. There was a linear relationship between products of lipid peroxidation and the log of the t-buOOH concentration in the perfusate (r = 0.97). Effect of Hb and TF on AK, in t-buOOH Lung Injury Figure 3 compares the AK, (post-t-buOOH KfC - baseline KfC) seen in time controls (-0.008 t 0.02), t-buOOH alone (0.27 t 0.05), t-buOOH + Hb pretreatment (0.43 + 0.05), and t-buOOH + Hb and TF pretreatment (0.13 7 0.02). Baseline KfC in these four groups was 0.31 t 0.04, 0.33 t 0.04, 0.26 t 0.03, and 0.33 t 0.05, respectively. A significant increase (P = 0.007) in the AKfc was observed in the t-buOOH + Hb compared with the tbuOOH alone group. A significant decrease (P = 0.001) in AKfc was observed in the t-buOOH + Hb + TF compared with the t-buOOH + Hb group. In addition, the decrease in KfC in the Hb + TF + t-buOOH was also significant (P = 0.0319) when compared with t-buOOH alone.
iY Q a t-
RESULTS
Quantity of Ferrous Iron in Perfusates
0.1
Figure 1 compares the amount of free iron measured in PSS + Alb (274.9 t 37.6 nmol/50 ml) with PSS + Alb + t-buOOH (177.1 t 31.2 nmol/50 ml), PSS + Alb + Hb (216.1 t 21.2 nmol/50 ml), and PSS + Alb + Hb + tbuOOH (1,052.5 t 51.6 nmol/50 ml). The measured iron concentration was significantly higher (P = 0.001) in the perfusate containing PSS + Alb + Hb + t-buOOH than in all other perfusates. No significant differences (P > 0.05) in the measured free iron was observed among the other perfusates.
J
-3
*
‘2ooT A
rtbuOOH1 L
2. Thiobarbituric acid reactive products (TBAR) production in isolated rat lungs perfused with t-buOOH. TBAR was measured in triplicate from 5 rats before and after t-buOOH perfusion for 20 min and depicted as quantity of TBAR produced by concentration of tbuOOH in perfusate. FIG.
,
-0.1
PSS+ ALB
PSS+ ALB+ tbuOOH
PSS+ ALB+ Hb
PSS+ ALB+Hb tbuOOH
+
FIG. 1. Free iron (Fe”) concentration in 4 different perfusates (n = 8 in each perfusate). There was a significant increase (P = 0.001) in Fe”’ in perfusate containing physiological salt solution (PSS) + albumin (Alb) + hemoglobin (Hb) that had been incubated with 0.89 mM tert-butyl hydroperoxide (t-buOOH). t-buOOH incubation time was 20 min in 2 perfusates exposed to t-buOOH. Data are means t SE. *P = 0.001 compared with all other perfusates.
1
time control n=8
tbuOOH alone n=lO
tbuOOH +Hb n=lO
tbuOOH +Hb+TF n=6
FIG. 3. Effect of Hb and apotransferrin (TF) on capillary filtration coefficient (KfC) changes due to t-buOOH. Three groups of isolated lungs were subjected to recirculating 0.89 mM t-buOOH. KfC was measured before and after 20 min of perfusion with t-buOOH and depicted as AK,,. Kfc was measu .red in time controls at end of equilibration period and 20 min later without addition of t-buOOH. Lungs pretreated with 50 PM Hb demonstrated a potentiation of injury due to t-buOOH evidenced by a significantly higher (P = 0.007) AK,, after Hb pretreatment. TF pretreatment resulted in a prevention of Hbassociated potentiation of t-buOOH injury as evidenced by significant reduction (P = 0.0001) of AK,, compared with lungs pretreated with Hb alone. Data are means k SE. *P = 0.007 compared with t-buOOH alone. 1-P = 0.007 compared with t-buOOH + Hb and P = 0.0319 compared with t-buOOH alone.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
HEMOGLOBIN
IN
OXIDANT
H1983
INJURY
Effect of Hb and TF on Hemodynamic Response to t-buOOH
observed increase in pulmonary capillary permeability is the endothelial cell membrane. Disruption of the endothelial cell membrane by lipid peroxidation would explain Figures 4 and 5 deplict the change in segmental resist- the increase in capillary filtration coefficient measured antes and pulmonary vascular pressures from time conin lungs exposed to t-buOOH. trols and lungs injured with t-buOOH. Although arterial This study also demonstrated that oxidant injury resistance tended to rise and venous resistance and cap- caused by t-buOOH was significantly potentiated by the illary pressures tended to fall in lu.ngs pretreated with presence of hemoglobin and that t-buOOH causes heHb before t-bu .OOH injury, there were no significant moglobin to release free iron. Free iron is capable of differences (P > 0.05) in any of the hemodynamic meas- catalyzing the production of *OH from HZOZ, and in our urements between any groups or when compared with experiments the perfusate containing the largest amount time controls. of free iron was associated with a significant potentiation of t-buOOH-induced lung injury. To further support the DISCUSSION hypothesis that catalytic iron released from hemoglobin was the cause of the potentiated injury, we demonstrated Our results demonstrate that t-buOOH, an organic hydroperoxide with significant oxidant potential, causes that apotransferrin prevented the hemoglobin-associated potentiation (Fig. 3). Although less significant, the cona large increase in the permeability of the pulmonary capillary bed (Fig. 3). This permeability-type injury is centration of apotransferrin used also attenuated the accompanied by evidence of lipid peroxidation in the lung injury when compared with t-buOOH alone. This latter observation implies that the iron contamination of perfusate from lungs exposed to t-buOOH. The quantity commercially available bovine albumin (40-70 nmol Fe/ of lipid peroxidation products detected is well correlated g Alb) is important in determining the degree of injury with the quantity of t-buOOH used to produce the injury due to t-buOOH. These experiments clearly show that (Fig. 2). t-buOOH apparently causes lung injury by lipid peroxidation that is manifested as an increase in the hemoglobin accentuates t-buOOH injury and that the permeability of the pulmonary capillary bed. The most most likely mechanism for this potentiation is the tlikely site of lipid peroxidation that would explain the buOOH-induced release of catalytic iron from hemoglobni . Our data have linked injury potentiation to the avail9.OE-2 ability of free iron but have not identified @OHproduction as the specific mechanism of potentiation. We have E 6.OE-2 previously shown that dimethyl sulfoxide, a *OH scav.-SI 3.OE-2 enger, has little effect on t-buOOH-induced injury in the E isolated rat lung (15). However, iron is also capable of 0 catalyzing the production of alkoxy and peroxy radicals 0.0 I? from lipid peroxides and the production these reactive E -3.OE-2 species could contribute to the observed injury (11). The 0 capability of iron to produce @OH is widely recognized, -6.OE-2 but, in fact, iron is a nonspecific catalyst that facilitates the production of more reactive species from less reactive -9.OE-2 1 species (14). Previous studies have provided evidence E’IG. 4. Effect of Hb and TF on segmental resistances in 4 groups that hydrogen peroxide, organic hydroperoxides, and suof rat lungs. Solid bar, time controls (n = 8); left diagonal, t-buOOH peroxide anion-generating systems can cause hemoglobin alone (n = 10); right diagonal, t-buOOH + Hb (n = 10); crosshatched, to release free iron (9, 23). This released iron is capable t-buOOH + Hb + TF (n = 6). Pulmonary arterial (R,)and venous of catalyzing OH production and initiating lipid peroxresistances (fiV) were calculated before and after perfusion with tbuOOH and in time controls and are expressed as change in (;1)R, and idation in vitro (2). In addition, transferrin, haptoglobin, AR,.. There were no significant differences (P > 0.05) in AP, or AR. and desferrioxamine have been shown to prevent the from any group compared with time controls. Data are means t SE. iron-catalyzed .OH formation by binding all open iron coordination sites (13, 14). An open coordination site must be available for iron to catalyze OH formation (8). There is very little iron with catalytic capability in normal mammalian tissues (16). The majority of iron is bound to hemoglobin or to other macromolecules where it is not available to act as a catalyst (21). This extensive binding or compartmentalization of iron is an important antioxidant defense mechanism in humans. Only when 1 exposed to an oxidant stress, as in the present study, is FIG. 5. Effect of Hb and TF on pulmonary vascular pressures in 4 the iron reservoir in hemoglobin available as a Fentongroups of rat lungs. Solid bar, time control (n = 8); left diagonal, ttype catalyst. huOOH alone (n = 10); right diagonal, t-buOOH + Hb (n = 10); crosshatched, t-buOOH + Hb + TF (n = 6). Pulmonary artery (P&, This reservoir of heme iron is also compartmentalized venous (PI,\), and capillary (P,,,.) pressures were measured before and within the erythrocyte, which further limits its ability to after perfusion with t-buOOH and in time controls and were expressed act as a catalyst. Erythrocytes are rich in antioxidant as change in (A)PIjB) AP1,V, and API,,. There were no significant differenzymes such as superoxide dismutase, catalase, and ences (P > 0.05) in API,,, APpx., or API,, from any group compared with time controls. Data are means t SE. glutathione peroxidase and have been shown to decrease
ARa
ARv
l
l
-0.6
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
H1984
HEMOGLOBIN
IN
H202-mediated damage to isolated rat lungs, rat hearts, and cultured bovine pulmonary artery endothelial cells (4,16,26). The protective effect of erythrocytes in H20zinduced injury may be due to the antioxidant enzymes they contain. Catalase activity is specific for hydrogen peroxide, and it is relatively ineffective at dismutating larger, organic hydroperoxides like t-buOOH (16). Glutathione peroxidase in the presence of glutathione can reduce nonmembrane-bound organic hydroperoxides to water (14, 27). These antioxidant enzymes are in intimate proximity to hemoglobin within the cytosol of the erythrocyte. Although hemoglobin can release catalytic iron when exposed to oxidant stress, the injurious potential of iron-catalyzed free radical production is limited by the cytosolic antioxidant enzymes before the free radicals can gain access to the red cell membrane or extracellular space. t-buOOH has been shown to cause erythrocyte hemolysis, which could theoretically result in the release of catalytic iron into the extracellular space in the absence of haptoglobin and transferrin (28). Erythrocytes possess antioxidant potential and compartmentalize the heme-iron reservoir, but our experiments were performed in isolated lungs that were virtually free of erythrocytes and the antioxidant enzymes they contain. In summary, our experiments provide evidence that 1) t-buOOH causes injury in isolated rat lungs accompanied by evidence of lipid peroxidation within the lung, 2) tbuOOH causes the release of catalytic iron from hemoglobin, 3) hemoglobin potentiates lung injury due to tbuOOH, and 4) apotransferrin prevents the hemoglobinassociated accentuation of lung injury due to t-buOOH. These findings are consistent with the hypothesis that free hemoglobin, when exposed to an oxidant stress, can act as a biological Fenton reagent in the isolated rat lung. We thank Sandy Mead for preparation of the manuscript and Dr. Joe McCord for technical advice. This work was supported by a Florence Foundation Research Career Development grant, National Heart, Lung, and Blood Institute Clinical Investigator Award HL-02352 (to J. Haynes, Jr.), and a University of South Alabama Intramural grant (to A. F. Seibert). Address for reprint requests: A. F. Seibert, 2451 Fillingim St., Mastin Bldg., Rm. 312, Univ. of South Alabama, Mobile, AL 36617. Received
17 September
1990; accepted
in final
form
27 February
1991.
REFERENCES 1. BERTRAND, Y. Oxygen-free radicals and lipid peroxidation in adult respiratory distress syndrome. Intensive Care Med. 11: 56-60,1985. 2. BRAUGHLER, J. M., L. A. DUNCAN, AND R. L. CHASE. The involvement of iron in lipid peroxidation. J. Biol. Chem. 261: 10282-10289, 1986. 3. BRIGHAM, K. L., AND B. MEYRICK. Interactions of granulocytes with the lungs. Circ. Res. 54: 623-635, 1984. 4. BROWN J. M., M. A. GRASSO, L. S. TERADA, L. J. BEEHLER, K. M. TOTH, G. J. WHITMAN, A. H. HARKEV, AND J. E. REPINE. Erythrocytes decrease myocardial hydrogen peroxide levels and reperfusion injury. Am. J. Physiol. 256 (Heart Circ. Physiol. 25): H584-H588,1989. 5. CARTER, P. Spectrophotometric determination of serum iron at the submicrogram level with a new reagent (Ferrozine). Anal. Biochem. 40: 450-458, 1971.
OXIDANT
INJURY
6. CHANCE, B., H. SIES, AND A. BOVERIS. Hydroperoxide metabolism in mammalian organs. Physiol. Reu. 59: 527-605, 1979. 7. DRAKE, R., K. A. GAAR, AND A. E. TAYLOR. Estimation of the filtration coefficient of pulmonary exchange vessels. Am. J. Physiol. 234 (Heart Circ. Physiol. 3): H266-H274, 1978. 8. GRAF, E., J. R. MAHONEY, R. G. BRYANT, AND J. W. EATON. Iron catalyzed hydroxyl radical formation. Stringent requirement for the free iron coordination site. J. Biol. Chem. 259: 3620-3624,1984. 9. GUTTERIDGE, J. M. C. Iron promoters of the Fenton reaction and lipid peroxidation can be released from haemoglobin by peroxides. FEBS Lett. 201: 291-295,1986. 10. GUTTERIDGE, J. M. C., S. K. PATERSON, A. W. SEGAL, AND B. HALLIWELL. Inhibition of lipid peroxidation by the iron binding protein lactoferrin. Biochem. J. 199: 259-261, 1981. 11. HALLIWELL, B. Oxidants and human disease, some new concepts. FASEB J. 1: 358-364, 1987. 12. HALLIWELL, B., AND J. M. C. GUTTERIDGE. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219: l14, 1984. 13. HALLIWELL, B., AND J. M. C. GUTTERIDGE. The importance of free radicals and catalytic metal ions in human disease. IMoL. Aspects Med. 8: 89-193, 1985. 14. HALLIWELL, B., AND J. M. C. GUTTERIDGE. Oxygen free radicals and iron in relation to biology and medicine: some problems and concepts. Arch. Biochem. Biophys. 246: 501-514, 1986. 15. HAYNES, J., A. SEIBERT, J. BASS, AND A. TAYLOR. U74500A inhibition of oxidant-mediated lung injury. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H144-H148, 1990. 16. HEFFNER, J. E., AND J. E. REPINE. Pulmonary strategies of antioxidant defense. Am. Reu. Respir. Dis. 140: 531-554, 1989. 17. JACKSON, R. M., C. F. VEAL, B. ALEXANDER, A. L. BRANNEN, AND J. D. FULMER. Re-expansion pulmonary edema: a potential role for free radicals in its pathogenesis. Am. Reu. Respir. Dis. 137: 1165-1171,1984. 18. KURODA, M., K. MURAKAMI, AND Y. ISHIKAWA. Role of hydroxyl radicals derived from granulocytes in lung injury induced by phorbol myristate acetate. Am. Reu. Respir. Dis. 136: 1435-1444, 1987. 19. MARTIN, W., G. M. VILLANI, D. JOTHIANANDAN, AND R. F. FURCHGOTT. Selective blockage of endothelium-dependent and glyceryl trinitrate-induced relaxation by hemoglobin and by methylene blue in rabbit aorta. J. Pharmacol. Exp. Ther. 232: 708-716, 1984. 20. MINOHI, G., AND S. D. AUST. The requirement for iron (III) in the initiation of lipid peroxidation by iron (II) and hydrogen peroxide. J. Biol. Chem. 262: 1098-1104,1987. 21. NORWOOD, M. The clinical biochemistry of iron (Abstract). Semin.
Hematol. 14: 31, 1977. 22. PERRY, M., AND A. E. TAYLOR. Phorbol myristate acetate-induced injury of isolated perfused rat lungs: neutrophil dependence. J. Appl. Physiol. 65: 2164-2169, 1988. 23. SADRZADEH, S., E. GRAF, S. PANTER, P. E. HALLAWAY, AND J. W. EATON. Hemoglobin. A biologic Fenton reagent. J. Biol. Chem. 259: 14354-14356,1984. 24. SAID, S. I., AND H. D. FODA. Pharmacologic modulation of lung injury. Am. Reu. Respir. Dis. 139: 1553-1564, 1989. 25. TATE, R. M., K. M. VANBENTHUYSEN, D. M. SHASBY, I. F. MCMURTRY, AND J. E. REPINE. Oxygen-radical-mediated permeability edema and vasoconstriction in isolated perfused rabbit lungs. Am. Reu. Respir. Dis. 126: 802-806, 1982. 26. TOTH, K. M., D. P. CLIFFORD, E. M. BERGER, C. W. WHITE, AND J. E. REPINE. Intact human erythrocytes prevent hydrogen-peroxide mediated damage to isolated perfused rat lungs and cultured bovine pulmonary artery endothelial cells. J. Clin. Inuest. 74: 292295,1984. 27. TRAVIS, J. Oxidants and antioxidants in the lung. Am. Reu. Respir.
Dis. 135: 773-774,1987. 28. TROTTA, R. J., S. G. SULLIVAN, AND A. STERN. Lipid peroxidation and haemoglobin degradation in red blood cells exposed to t-butyl hydroperoxide. The relative roles of haem- and glutathione-dependent decomposition of t-butyl hydroperoxide and membrane lipid hydroperoxides in lipid peroxidation and haemolysis. Biochem. J. 212: 759-772, 1983.
Downloaded from www.physiology.org/journal/ajpheart at Washington Univ (128.252.067.066) on February 13, 2019.
