- Email: [email protected]
Oxidant inhibition of epithelial active sodium transport
Free Radical Biology & Medicine, Vol. 6, pp. 557 564, 1989 Printed in the USA. All rights reserved.
0891-5849/89 $3.00+ .00 © 1989 PergamonPress plc
Original Contribution OXIDANT
INHIBITION
OF
EPITHELIAL
ACTIVE
SODIUM
TRANSPORT
SADIS MATALON,* JOSEPH S. BECKMAN,* MICHAEL E. DUFFEY,J a n d BRUCE A . FREEMAN* *Departments of Anesthesiology, Physiology and Biophysics, and Biochemistry, University of Alabama at Birmingham, Birmingham, AL 35233; tDepartment of Physiology, State University of New York at Buffalo, Buffalo, NY 14214, USA (Received 26 April 1988; Revised 3 August 1988; Accepted 13 September 1988)
Abstract--The purpose of this study was to quantify the effects of extracellularly generated partially reduced oxygen species on active sodium (Na ÷) transport across the ventral toad skin, a well-studied epithelium. Sections of skin from decapitated toads were mounted in an Ussing chamber, bathed on both sides with electrolyte solution containing 500/tM xanthine and bubbled continuously with room air. The tissues were short-circuited, and shortcircuit current (lsc) and tissue resistance (R,) were monitored continuously with an automatic voltage clamp apparatus. Fifteen mU/ml of xanthine oxidase (XO), either purchased from Calbiochem or purified from cream, were instilled in either the apical (mucosal) or basolateral (serosal) baths at t = 0 and t = 10 min. Hydrogen peroxide (H202) concentrations increased to 200/~M within the first 20 min and then decreased, reaching a value of 40/zM by 60 min. Mean [H_,O2] was 90/zM. Instillation of XO in the apical bath resulted in a large decrease in lsc and an increase in R,, their values being 43% and 160% of their corresponding controls 85 min after the first instillation. Addition of superoxide dismutase and catalase completely prevented these changes. Instillation of XO in the basolateral bath had no effect. Similar physiological responses were obtained using the Calbiochem XO or the purified XO, which contained no measurable protease activity. It was concluded that extracellularly generated partially reduced oxygen species may interfere with active Na + transport by possibly damaging apical Na + channel proteins. Keywords--Xanthine oxidase, Active Na transport, Free radical, Short-circuit current, Hydrogen peroxide, Superoxide, Epithelial
include superoxide (O2-), hydrogen peroxide (H202), and hydroxyl ions ( ' O H ) . 4 These species can be generated both inside and outside cells. Both O2- and H202 are produced by mitochondria of mammalian cells during normal metabolism and at increased rates following exposure of organisms to elevated concentrations of oxygen or xenobiotics. 4'5 Superoxide may react with intracellular targets or exit the cells via anion channels. 6 Cell membranes are freely permeable to H202. In addition, partially reduced oxygen species are released in the extracellular milieu by activated neutrophils. Epithelial cells are prime targets for partially reduced oxygen species. Previous studies have shown that exposure to normobaric (1 ATA) or hyperbaric (5 ATA) hyperoxia increased alveolar permeability to solute in rabbits 7 and decreased Na + absorption across the tight epithelium of toad bladders, s In addition, exposure to H202 increased the paracellular conductance of cultured M D C K cells. 9 However, changes in active
INTRODUCTION
Sodium (Na ÷) enters epithelial cells across the apical (mucosal) membrane, through amiloride-sensitive pathways, down a favorable electrochemical potential gradient. Once inside the cell, it is actively transported across the basolateral (serosal) membrane by the Na ÷ :K + ATPase. ~ This process is linked to the movement of water across epithelial tissue and plays an important role in the reabsorption o f lung fluid from the fetal airspace and the maintenance of fluid-free alveolar space in the adult lung. 2,3 The incomplete reduction of oxygen leads to the formation of partially reduced oxygen species, which Supported by grants from the National Institutes of Health (HL 31197), the Health Effects Institute, a Career Investigator Award by the American Lung Association, and the Alabama Affiliate of the American Heart Association. Address correspondence to Sadis Matalon, PhD, Department of Anesthesiology, University of Alabama at Birmingham, 619 Nineteenth Street South, Birmingham, AL 35233, USA.
557
558
S. MATALON, J. S. BECKMAN, M. E. DUFFEr, and B. A. FREEMAN
Na + transport following exposure of epithelial tissues to partially reduced oxygen species have not been measured. In the present study, we quantified whether the presence of increased extracellular concentrations of partially reduced oxygen species would compromise active Na + transport across a tight epithelium, the ventral toad skin. The rate of Na + transport was assessed by measuring the short-circuit current (Isc) across this epithelium when mounted in an Ussing chamber. Partially reduced oxygen species were generated by xanthine plus xanthine oxidase (XO) added to either the mucosal or serosal sides of the toad skin. We quantified rates of production and steady-state concentrations of H202 and correlated these values with changes in epithelial short-circuit current and conductance in a temporal manner. Furthermore, we assessed whether the presence of proteases in XO compounds oxidant injury to Na + transport. The results of this study allow us to draw specific conclusions about the site and mechanisms of oxidant-induced injury to Na + transport across this epithelium.
MATERIALS AND M E T H O D S
Tissue isolation and electrophysiological measurements Sections of ventral abdominal skin of decapitated toads (Bufo marinus) were washed with an electrolyte solution containing 2.5 mM KHCO~, 111 mM Na2SO4, and 1.0 mM Ca2SO4. When bubbled with room air, the pH of this solution was 8.3. The tissue was then mounted vertically as a flat sheet between two conical Plexiglas hemichambers with an aperture of 1 cm 2. Each hemichamber was connected to a bubble-lift apparatus containing 15 ml electrolyte solution, which was thoroughly aerated and uniformly mixed by continuous bubbling with room air (21% 02, 79% N2). A voltage-current clamp apparatus (Physiologic Instruments; Model VCC 600) was used for the continuous monitoring of transepithelial potential difference (V,), short-circuit current (Isc), and transepithelial resistance (R,). The transepithelial potential difference was measured between Hg +/HgC1 electrodes, which made contact with the bathing solutions via agar bridges (3% agar in glucose-free, standard electrolyte solution), positioned at 3 mm at either side of the tissue. Short-circuit current was measured by passing sufficient current through Ag +/Ag electrodes, in contact with the bathing solutions via agar bridges, to reduce the spontaneous V, to zero. R, was determined using O h m ' s law from the observed change in Isc when V, was clamped at -+5-10 mV for 60 ms; chamber fluid
resistance was automatically subtracted by the voltage clamp. Tissues were continuously short-circuited. Short-circuit current was recorded in a two-channel strip-chart recorder throughout each experiment. Measurements were preceded by an equilibration period during which steady-state values for lsc and R, were reached (Isc > 40/~A/cm2; A Isc/hr < 5 ILA/ cm2; R, > 500 Ohm.cm2). Less than 10% of the preparations did not meet these criteria within three hours of mounting and were discarded. When lsc reached a stable value, xanthine and XO were added to the bathing fluid (see below) and changes in Isc and R, were followed for at least 60 min. At this time, 15/LI of 0.1 M amiloride was added into the mucosal bathing solution (final concentration: 10 4 M) and changes in lsc and R, were recorded for an additional 15 min.
Biochemical assays and measurements Generation of partially reduced oxygen species. Superoxide, H202, and "OH were generated in the extracellular medium by adding into either the mucosal or serosal bathing fluid 800 Itl of 10 mM xanthine (2,6 dihydroxypurine, sodium salt, grade IIl; Sigma Chemical Company; final bathing concentration: 500/tM) and 3 - 5 min later, 15 mU/ml of XO. Because XO may be self-inactivated by reactive oxygen products, "~ an additional 15 m U / m l of XO was instilled ten min later. Since even small changes in the pH of the perfusing medium are known to affect the value of Isc, ~' the xanthine stock solution was dissolved in the electrolyte solution perfusing the tissue and adjusted to a pH of 8.3 by the addition of dilute sulfuric acid. Two different types of XO were used in this study: the first one was purchased commercially (grade Ill Calbiochem, dissolved in 2.3 M NH4SO~) and was showing to contain significant amounts of protease activity; the second was purified from fresh cream in our laboratory according to the method of Waud et al. ~2 and shown to be essentially protease free (see Results). Just prior to instillation into the bathing fluid, all XO samples were chromatographed to remove interference from endogenous low molecular weight inhibitors and preservatives (ammonium sulfate or choline chloride). A prepacked Sephadex G-25 column (9 x 2 cm, Pharmacia) was equilibrated with the electrolyte solution and fitted through a bored-out # 8 cork in a 50 ml polyethylene tube. After removing all fluid by centrifugation, a small volume of XO, diluted in 1 ml electrolyte solution, was added to the column, which was placed in a new tube and centrifuged at 900 x g for five min. The protein sample was collected and its XO activity was assayed by monitoring the production of uric acid at 295 nm (A~ M = !.1 × 10 4 M-~ . cm ') at 25°C. The reaction mixture consisted of 50
Oxidant injury and sodium transport
#M xanthine in electrolyte solution (pH = 8.3). One unit of XO was defined as the amount of enzyme generating 1 /tmol/min of urate. In four control experiments Cu,Zn superoxide dismutase (SOD; 4500 U) and bovine liver catalase (CAT; 9000 U) were added to the perfusing medium prior to the XO instillation. SOD was generously provided by Diagnostic Data, Inc. (Mountain View, CA) and by Grunenthal GMBH (Aachen, West Germany). Catalase was obtained from Cooper Biomedical Corp (Freehold, N J). Both enzymes were dialyzed for 48 hours against cold electrolyte solution for ionic and pH equilibration. Measurement of protease activity. Xanthine oxidase protease activity was assessed by mixing various amounts of XO with 500/~1 azocasein (Calbiochem, 1 mg/ml), incubating the mixtures at 37°C for 30 min, precipitating the unfragmented protein with 500/tl of 10% tricbloroacetic acid (TCA), and measuring hydrolyzed fragments in the supernatant at 440 nm. Xanthine oxidase associated proteases caused the hydrolysis of azocasein to acid-soluble. Measurement of H202 concentration. Concentrations of H202 in the perfusion medium were measured throughout the course of the study in five experimental and three control preparations. One hundred/zl of perfusion fluid was mixed with 400/~1 of a 100 mM KPO4 mixture (pH = 7.0) containing 2000 U of horseradish peroxidase (type II, 200 U/rag; Sigma Chemical Company), 1.5 mM 4-aminoantipyrine (Sigma), and 0.11 M phenol (Phenol ACS reagent; Sigma). The assay solution also contained the XO inhibitor allopurinol (20 /IM) to prevent further generation of H202. The concentration of H202 was calculated from the absorba n c e a t 5 1 0 n m , usingan¢M = 6 . 5 8 m M ~ . c m Statistics. All data are presented as means + SE. Significant differences among several group means were evaluated using the one-way analysis of variance. If the F value was significantly different from zero (p < 0.05), the Bonferonni modification of the t-test was applied to compare group means.~3
RESULTS
559
sorbance and amounts of commercial XO added (r 2 = 0.95; p < 0.01). This relationship was abolished when XO was preincubated with 1 mM phenylmethylsulfonyl-fluoride (a specific protease inhibitor) and eluted through a Sephadex G-200 column prior to assay. Urate and H202 production The temporal profile of the rate of urate production generated by xanthine and XO was different when it was continuously assayed in a spectrophotometer cuvette versus obtaining sequential A295readings of samples removed from the Ussing chamber. In the spectrophotometer, there was a steady decrease in the rate of urate production, its value at ten min being approximately 50% of the initial one (Fig. 2). Addition of another aliquot of XO resulted in a temporary increase in urate production, followed by a steep decrease, while addition of xanthine or bubbling the cuvette with room air did not reverse the decrease in urate production (Fig. 2). These findings are consistent with a time-dependent inactivation of XO in the spectrophotometer cuvette sitting in the ultraviolet beam. However, XO was apparently not inactivated when added to the Ussing chamber (Fig. 2). In the absence of toad skin, urate production remained constant at about 18/~M/min, increased to 37 /~M/min after the instillation of the second bolus of XO, and then decreased to about 5 ~M/min by 30 min. In this case, instillation of xanthine prevented the decrease in urate production, indicating that the observed decrease was due to substrate depletion (data not shown). Assuming that XO remained fully active, the available substrate would be exhausted by 30 min after the first instillation of XO. As predicted from the rates of urate production in the Ussing chamber with no toad skin present (Fig. 2), H202 concentration increased with time during the first 30 min (Fig. 3), and reached a plateau value of around 400/~M. In contrast, in the presence of toad skin, this variable reached a peak value of only 200/IM and then decreased to 30 ~M. Thus, the toad skin was either consuming U202 directly or releasing substances with peroxidase activity into the perfusion medium. Similar rates of urate production and H202 concentrations were obtained with either the commercial or purified XO.
Xanthine oxidase activity
Short-circuit current (lsc) and epithelial resistance (Rt)
Figure 1 shows hydrolysis of azocasein to low molecular weight TCA-soluble fragments by XO-associated proteases, as reflected by changes in absorbance at 440 nm. Absorbance was unchanged when purified XO was added to the assay mixture. On the other hand, there was a linear correlation between changes in ab-
Figure 4 shows the time course of Isc in a typical preparation before and after the instillation of XO in the perfusion medium. Stable values for Isc and R, were obtained in all preparations within 90 rain of mounting the toad skin in the Ussing chambers. Mean values in 21 preparations (+SE) were: Isc = 70 +- 5/zA/cm2;
560
S. MATALON, J. S. BECKMAN, M. E. DUFFEY, and B. A. FREEMAN
200
o 1"
o aJ o
15o.
o o
.c) o
100
.
...
....
,, ...........
...........
v~ 1
50
'
0
I
4
,
I
,
8
I
12
XO
,
I
16
,
I
20
,
I
24
(mU/ml)
Fig. 1. Changes in light absorbance in 0.5 ml supernatants of azocasein j'ollowing the addition of' XO, and precipitation of' unfragmented proteins with 10% trichloroacetic acid. Values are expressed % change from control (no XO activity). Means -+ 1 SE. The solid line represents a linear least squares fit through the indicated points (r -~ = 0.95: p < 0.01). Symbols are as follows: • = XO purified from cream (n - 5) • = Calbiochem XO (n = 4) O = Calbiochem XO incubated with PMSF (n - 4)
R, = 881 -+ 59 ~ • cm 2. Instillation of amiloride in the mucosal bath ( 1 0 - 4 M ) at the end of the study decreased Isc to below 3 / t A / c m 2 within 10 min. Addition of xanthine or xanthine oxidase alone had no effect on any of the measured variables. Instillation of XO in the mucosal bath resulted in the immediate generation of H202 (Fig. 4) and caused a gradual decline in Isc and an increase in R,, starting at 20 min after instillation (Fig. 4). Similar changes were obtained with either commercial or purified XO. Pooled mean values for Isc and R, for both type of XO (expressed as percentage of the corresponding values just prior to instillation) are shown in Tables 1 and 2. Addition of SOD and CAT in the perfusion medium prevented the formation of H 2 0 2 and obviated changes in Isc and R,, indicating the involvement of partially reduced oxygen species in the observed reduction of Na + absorption. In contrast, instillation of xanthine and XO in the serosal bath had no appreciable effect in Isc and R, (Tables 1 and 2) in spite of the fact that similar amounts of H202 were generated in the perfusion medium. DISCUSSION
Xanthine oxidase catalyzes the conversion of xanthine to urate and reduces molecular oxygen to partially reduced oxygen species. At pH 8.3, approximately 30% of these species appear as 0 2 - and the rest as H202.~4 Since 0 2 - dismutes spontaneously or by enzymatic catalysis to H202, this more stable species was quantified to reflect levels of XO-mediated stress. The
m e a n H 2 0 2 in the fluid perfusing the toad skin after the instillation of xanthine and XO was 90 ~M. Tate et al. ~5reported that the levels of free radicals produced by 20 m U / m l of XO and 2 mM purine were equivalent to those generated in vivo by activated circulating neutrophils. Thus, the imposed stress was within physiological limits. In any event, in preliminary experiments, a single dose of XO (15 mU/ml) did not have any effect on either lsc or R,. Fridovich and coworkers reported XO inactivation during the oxidation of acetaldehyde or xanthine, ~0and initiation of membrane lipid peroxidation during coincubation of this reaction system with erythrocytes or egg phosphatidylcholine-containing liposomes.~°.~6 The rate of inactivation was dependent on the initial XO concentration (half-time = 5 min for - 12 mU XO). Superoxide dismutase and catalase scavenge 02 and H202, the primary reactive products of acetaldehyde and xanthine oxidation, and when added to incubation systems, inhibited inactivation of XO and peroxidative membrane damage• t6 We also saw about a 50% inhibition of 15 m U / m l XO 10 rain after the initiation of the reaction. XO activity was increased upon addition of another aliquot of XO, with no enhancement of XO activity after provision of additional substrate in a spectrophotometer cuvette. However, when similar assessment of XO activity over time was conducted on samples removed from the Ussing chamber, no inhibition of the initial XO activity occurred in the first 10 min (Fig. 2). Addition of another aliquot
Oxidant i~ury and sodium transport
561
35 30 c o O~
25
".~_,
20
E~
15 10 5 0 0
5
10
15
20
25
30
Time (rain) Fig. 2. Urate production vs. time. Xanthine oxidase (15 mU/ml) was added either in a quartz cuvette ( 0 , ©) or in the mucosal bath of an Ussing chamber ( . ) at t = O. They both contained 500 ,uM xanthine in electrolyte solution. At t = 10 min. a bolus of 15 mU/ml XO ( 0 , m) or 500 #M xanthine (O) were instilled in either the cuvette or the Ussing chamber. Each point represents the mean of at least four measurements.
of 15 mU/ml XO caused a transient increase in the rate of urate production followed by a decrease secondary to substrate depletion. This decrease was prevented by instillation of an additional bolus of xanthine. As mentioned above, this period of XO stability did not occur in the spectrophotometric assay system (Fig. 2). It is possible that prolonged exposure of the reaction system to ultraviolet light in the spectrophotometric
study exacerbated free radical inactivation of XO, or that the more aggressive buffer oxygenation in the Ussing chamber volatilized otherwise enzyme-damaging reaction products such as H202. In none of the experiments was 02 limiting to XO activity, as evidenced by calculations of 02 consumption and the lack of stimulation of XO activity when xanthine alone was added (Fig. 2). Na ÷ movement across the toad skin is a two-step
500
400
300 o :~ ea~ k..-d
200
100
/~
\6--~
,w
J
]
0
10
20
Toad skin
i
i
30 40 T i m e (rain)
t
i
50
60
Fig. 3. Measurements of H202 concentrations in an Ussing chamber in the presence ( 0 ) or absence ( A ) of toad skin. In both cases, the chambers contained electrolyte solution with 500 ~M xanthine. XO (15 mU/ml) was added to the chamber at t = 0 and t = 10 rain. Means - 1 SE; n = 5. Corresponding time points after 20 min are significantly different from each other at p < 0.02 level.
562
S. MATALON, J. S. BECKMAN, M. E. DUFFEY, and B. A. FREEMAN
120
ea
90
°
I[11 llll
o oo 30
TrTil+ : : : : : : : : : : : : : : : : : : : : : : : : : 0 20 40 60 80 Time
(rain)
Fig. 4. Time course of [sc in a typical experiment. Vertical excursions represent Isc changes when the clamping voltage was changed by l0 mV in a bipolar fashion. Tissue resistance can be calculated from O h m ' s law. Single and double arrows indicate the time of instillation of xanthine (X) and XO, respectively.
process: Na + crosses the apical membrane due to the existing electrochemical gradient through amiloride-sensitive channels and is then actively extruded across the basolateral membrane through the Na+: K+ATPase. ~5 In addition, a small amount of Na ÷ may move through paracellular junctions. In our preparation, Isc was predominantly caused by the intracellular movement of Na + because amiloride which specifically blocks Na + channels but does not affect paracellular Na + movement t7 rapidly decreased lsc to undetectable levels. Our results indicate that instillation of xanthine and XO in the apical side of the toadskin caused a significant decrease in Isc and an increase in R, in a timedependent fashion ten min after instillation of the second aliquot of XO. This lag time may be due to cell protection by antioxidant enzymes subject to oxidant inactivation 4 or depletable low molecular weight scavengers. In addition, as in the in vivo situation, there may be a period between cellular and biochemical lesions and onset of functional alterations. Addition of SOD and catalase in the perfusion medium prevented the formation of H202 by xanthine and XO, the decrease in lsc and the increase in R,, indicating that O2-, H202 and possibly .OH were directly involved in the initiation of these changes. A possible mechanism by which partially reduced 02 species may interfere with Na ÷ transport is by oxidizing sulfhydryl groups in Na + channel proteins. Addition of the organic mercurials mersalyl, p-chloromercuribenzoate, and p-chloromercuribenzene sulfonate to the solutions bathing the mucosal side of pieces
of rabbit colon mounted in an Ussing chamber interfered with Na + transport by reacting with sulfhydryl groups located in the apical membrane.~8 Also, partially reduced oxygen species may be reacting with a number of amino acid residues which are necessary for the proper functioning of Na + channels. Wang et al. ~9 reported that chloramine-T removed sodium channel inactivation of squid giant axons by modifying one of its methionine residues. Nonspecific reactions such as lipid peroxidation and protein inactivation may also contribute to the observed decrease in Isc. For example, it has been suggested that oxidation of membrane lipids may allow them to function as calcium ionophores, thereby increasing intracellular calcium levels 2° which, in turn may decrease the permeability of the apical membrane to sodium. 2~.z2 Since addition of xanthine and XO to the serosal side of the skin did not alter either the Isc or R, (Table 2), it is unlikely that the Na + :K + ATPase was inhibited. A number of studies suggest that the Na + :K + ATPase may be resistant to partially reduced 02 species. Spragg et al. 23 reported that exposure of cultured endothelial or P388D~ murine cells to 5 mM H202, an order of magnitude higher than in our study, resulted after three min in a large decrease in cellular ATP levels. However, N a + : K + ATPase activity remained unchanged from control conditions. Maridonneau et al. 24 demonstrated that the treatment of human erythrocytes with phenazine methosulfate, an agent known to result in increased production of partially reduced oxygen species, caused an increase in passive membrane K ÷ permeability, but had no effect on the
O x i d a n t injury a n d s o d i u m t r a n s p o r t
563
T a b l e 1. C h a n g e s in S h o r t - C i r c u i t C u r r e n t (Isc) a n d Tissue R e s i s t a n c e (R,) F o l l o w i n g Instillation o f X a n t h i n e a n d X a n t h i n e O x i d a s e in the A p i c a l ( M u c o s a l ) B a t h X + SOD + CAT + XO
X+XO Isc
R,
Time (min)
# of expts
0 5 10 15 25 35 45 55 65 75 85
9 9 9 9 9 9 8 7 6 4 4
99 98 95 86 75 64 55 49 47 43
86 88 90
4 4 4
13 ± 2* 5 ± I* 3 ± 1"
Isc # of expts
(% o f c o n t r o l ) I00 ± 1 ± 1 ± 2 ± 3*t ± 3*t ± 3*t ± 4*+ ± 5*t ± 4*t ± l*t
100 103 113 133 147 155 165 165 165 160
100 ± 1 ± 2 ± 4 ± 8*t ± 8*t ± 9*t ± 19"t ± 7*t ± 8*t ± 8*t
Amiloride 3 9 0 ± 90 4 1 0 ± 100
R, (% o f c o n t r o l )
4 4 4 4 4 4 4 4 3 3 3
100 100 ± 100 ± 99 ± 99 ± 98 ± 97 ± 96 ± 95 ± 93 ± 90 ±
3 3
22 ± 1 4.5 ± 3
0 0 l 1 1 1 1 I 1 I
100 100 ± 101 ± 104 ± 106 ± 109 ± 114 ± 105 ± 108 ± 106 ± 110 ±
0 1 2 2 2 4* 2 2 2 3*
380 ± 9 0 390 ± 9 0
Note. Numbers are means -+ 1 SE; X. SOD. CAT were added to the mucosal bath at t = 0; X = xanthine (500 liM): XO = xanthine oxidase (purified or Catbiochem); 15 mU/ml at t = 0 and t - 10 min. SOD = superoxide dismutase (4500 U); CAT = catalase (9,000 U); amiloride (10 4 M) was added at t = 85 min. *p < 0.05 from the corresponding control value. tp < 0.05 from the corresponding value at the same time interval while the bath contained SOD and CAT.
Na + : K + ATPase activity. Also, lsc is limited by the rate of entry of N a + through the apical membrane channels, so injury to the Na + :K + ATPase might not result in a decrease in Isc. Previous reports indicate that proteases may act synergistically with oxygen metabolites in injuring cells and tissues. 25 In this study, we were unable to show any difference in the responses of Isc between
Table 2. C h a n g e s in S h o r t - C i r c u i t C u r r e n t , (Isc), a n d Tissue R e s i s t a n c e (R,) F o l l o w i n g Instillation o f X a n t h i n e and X a n t h i n e O x i d a s e in the B a s o l a t e r a l (Serosal) B a t h X+XO Time (min)
# of expts
0 5 10 15 25 35 45 55 65
9 5 5 5 5 4 4 3 2
Isc
R, (% o f control)
100 100 ± 100 ± 102 ± 105 ± 108 ± 109 ± 108 ± 109 ±
1 1 1 2 1 1 3 0
100 101 ± 102 ± 104 ± 105 ± 109 ± 115 ± 110 ± 117
1 2 3 5 6 7* 5
Note. Numbers are means +- 1 SE; all chemicals were added to the mucosal bath at the specified times. X = xanthine (500/tM); t = 0. XO = xanthine oxidase (purified or Calbiochem); 15 mU/ml at t = 0 and t = 10 min. *p < 0.05 from corresponding control value.
commercially available XO that contained significant levels of proteases and purified XO that did not. Active sodium transport across alveolar epithelial cells is an important mechanism of fluid reabsorption from the alveolar to the interstitial space. A number of noxious agents that may damage the alveolar epithelium produce chemotactic agents that attract neutrophils to the interstitial space and stimulate them, causing the release of O2-, H202 and other toxic species. Our previous findings indicate that these species increase the alveolar paracellular permeability to a number of lipid-insoluble proteins. 7 The results of this present study indicate that they also may interfere with active Na + transport and thus compound the accumulation of fluid in the alveolar space. In conclusion, we found that active Na + transport across the toad skin can be inhibited by the addition of xanthine and XO in the apical side of this epithelium. This inhibition is caused by the production of 02 , H202 and possibly .OH, since it is completely prevented by the addition of SOD and catalase. Finally, since addition of these substances to the serosal side had no effect, we speculate that the observed decrease in Na ÷ transport is due to the action of partially reduced 02 species on apical membrane entry sites, possibly Na + channels. REFERENCES 1. M a c K n i g h t , A. D. V.; D e b o n a , D. R.; L e a f , A. S o d i u m transport a c r o s s t o a d u r i n a r y bladder: A m o d e l " t i g h t " e p i t h e l i u m . Physiol. Rev. 6 0 : 6 1 5 - 7 1 5 ; 1980.
564
S. MATALON, J. S. BECKMAN,M. E. DUFFEY, and B. A. FREEMAN
2. Matthay, M. A.; Widdicombe, J. H.; Staub, N. C. Clearance of alveolar fluid in sheep may involve an active ion transport process. Fed. Proc. 41:1244 abstr. (1982). 3. Goodman, B. E.; Brown, S. E. S.; Crandall, E. D. Regulation of transport across pulmonary alveolar epithelial cell monolayers. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 57:703-710; 1984. 4. Freeman, B. A.; Crapo. J. D. Biology of disease: free radicals and tissue injury. Lab. Invest. 47:412-426; 1982. 5. Freeman, B. A.; Crapo, J. D. Hyperoxia increases oxygen radical production in rat lungs and lung mifochondria. J. Biol. Chem. 256:10986-10992; 1981. 6. Kellogg. E. G.; Fridovich, I. Liposome oxidation and eryrhrocyte lysis by enzymatically generated superoxide and hydrogen peroxide. J. Biol. Chem. 252:6721-6728; 1977. 7. Matalon, S.; Egan, E. A. The effects of Oz breathing on the permeability of the alveolar epithelium to solute. J. Appl. Physiol.: Respirat. Environ. Exercise Phvsiol. 50:859-863; 198 I. 8. Miller. J. H.; Mendoza, S. A. Inhibition of sodium transport by hyperbaric oxygen in the toad urinary bladder. Undersea Biomed. Res. 41333-345; 1977. 9. Welsh, M. J.; Shasby, D. M.; Husted, R. M. Oxidants increase paracellular permeability in a cultured epithelial cell line. J. Clin. Invesr. 76:1 155-I 168; 1985. 1. Autoinactivation of xanthine oxi10. Lunch. R. E.; Fridovich. dase. The role of superoxide radicals and hydrogen peroxide. B&him. Biophys. Acta 571: 195-200; 1979. 11. Fischbarg, J.; Whittenbury, G. The effect of external pH on osmotic permeability, ion and fluid transport across isolated frog skin. J. Phpsiol. London 275:403-417; 1978. 12. Waud, W. R.; Brady, F. 0.: Wiley. R. D.; Rajagopalan. K. V. A new purification procedure for bovine milk xanthine oxidase: effect of proteolysis on the subunit structure. Arch. Biochem. Bioph?;s. 169:695-701; 1975. 13. Wallenstein, S. C.; Zuckner. C. L.: Fleisa. J. F. Some statistical methods useful in circulation research. Circ. Res. 47: 1-9; 1980. aspects of the production of super14. Fridovich, 1. Quantitative
15.
16.
17. 18.
19.
20.
21.
22.
23.
34.
25.
oxide anion radical by milk xanthine oxidase. .I. Biol. Chem. 245:4053-4057; 1970. Tate. R. M.; Vanbenthuysen, K. D.; Shasby, D. M.; McMurtry, I. F.: Repine, J. E. Oxygen-radical-mediated permeability edema and vasoconstriction in isolated perfused rabbit lungs. Am. Rev. Respir. Dis. 126:802-906; 1982. Kellogg, E. W.. III; Fridovich, I. Liposome oxidation and erythrocyte lysis by enzymically generated superoxide and hydrogen peroxide. J. Biol. Chem. 252:6721-6728; 1977. Benos, D. J.: Saccomani. G.; Sarigan-Sohraby. S. The epithelial sodium channel. J. Biol. Chem. 262:10613-10618; 1987. Luger, A.; Turnheim, K. Modification of cation permeability of rabbit descending colon by sulphydryl reagents. J. Physiol. 317:49-66; 1981. Wang, G. K.; Brodwi. M. S.; Eaton, D. C. Removal of sodium channel inactivation in squid axon by the oxidant chloramineT. J. Gen. Ph,vsio/. 86:289-302: 1985. Serham, C.: Anderson. P.; Dunham, P.; Weissmann, G. Phosphatidate and oxidized fatty acids are calcium ionophores. J. Biol. Chcm. 25612736-2741: 198 I. Garty. H.; Asher, C. Ca” dependent, temperature sensitive. regulation of Na channels in tight epithelia. A study using membrane vesicles. J. Biol. Chem. 260:8330-8335; 198.5. Garty. H.; Asher. C. Ca” induced down regulation of Na’ channels in toad bladder epithelium. J. Biol. Chem. 261:74007406; 1986. Spragg. R. G.; Hinshaw. D. B.; Hyslop, P. A.; Schraufstatter. I. U.; Cochrane, C. G. Alterations in adenosine trophosphate and energy change in cultured endothelial and P388D, cells after oxidant injury. J. C/in. lnvesr. 76:1471-1476; 1985. Maridonneau, I. P.; Braquet, R. P.; Garay. R. P. Na’ and K ’ transport damage induced by oxygen free radicals in human red cell membranes. J. Biol. Chem. 258:3107-31 13; 1983. Baird, B. R.: Cheronis. J. C.: Sandhaus, R. A.; Berger, E. M.; White. C. W.: Repine, J. E. 0, metabolites and neutrophil elastase synergistically cause edematous injury in isolated rat lungs. J. Appl. Physiol. 61:224-229: 1986.
0891-5849/89 $3.00+ .00 © 1989 PergamonPress plc
Original Contribution OXIDANT
INHIBITION
OF
EPITHELIAL
ACTIVE
SODIUM
TRANSPORT
SADIS MATALON,* JOSEPH S. BECKMAN,* MICHAEL E. DUFFEY,J a n d BRUCE A . FREEMAN* *Departments of Anesthesiology, Physiology and Biophysics, and Biochemistry, University of Alabama at Birmingham, Birmingham, AL 35233; tDepartment of Physiology, State University of New York at Buffalo, Buffalo, NY 14214, USA (Received 26 April 1988; Revised 3 August 1988; Accepted 13 September 1988)
Abstract--The purpose of this study was to quantify the effects of extracellularly generated partially reduced oxygen species on active sodium (Na ÷) transport across the ventral toad skin, a well-studied epithelium. Sections of skin from decapitated toads were mounted in an Ussing chamber, bathed on both sides with electrolyte solution containing 500/tM xanthine and bubbled continuously with room air. The tissues were short-circuited, and shortcircuit current (lsc) and tissue resistance (R,) were monitored continuously with an automatic voltage clamp apparatus. Fifteen mU/ml of xanthine oxidase (XO), either purchased from Calbiochem or purified from cream, were instilled in either the apical (mucosal) or basolateral (serosal) baths at t = 0 and t = 10 min. Hydrogen peroxide (H202) concentrations increased to 200/~M within the first 20 min and then decreased, reaching a value of 40/zM by 60 min. Mean [H_,O2] was 90/zM. Instillation of XO in the apical bath resulted in a large decrease in lsc and an increase in R,, their values being 43% and 160% of their corresponding controls 85 min after the first instillation. Addition of superoxide dismutase and catalase completely prevented these changes. Instillation of XO in the basolateral bath had no effect. Similar physiological responses were obtained using the Calbiochem XO or the purified XO, which contained no measurable protease activity. It was concluded that extracellularly generated partially reduced oxygen species may interfere with active Na + transport by possibly damaging apical Na + channel proteins. Keywords--Xanthine oxidase, Active Na transport, Free radical, Short-circuit current, Hydrogen peroxide, Superoxide, Epithelial
include superoxide (O2-), hydrogen peroxide (H202), and hydroxyl ions ( ' O H ) . 4 These species can be generated both inside and outside cells. Both O2- and H202 are produced by mitochondria of mammalian cells during normal metabolism and at increased rates following exposure of organisms to elevated concentrations of oxygen or xenobiotics. 4'5 Superoxide may react with intracellular targets or exit the cells via anion channels. 6 Cell membranes are freely permeable to H202. In addition, partially reduced oxygen species are released in the extracellular milieu by activated neutrophils. Epithelial cells are prime targets for partially reduced oxygen species. Previous studies have shown that exposure to normobaric (1 ATA) or hyperbaric (5 ATA) hyperoxia increased alveolar permeability to solute in rabbits 7 and decreased Na + absorption across the tight epithelium of toad bladders, s In addition, exposure to H202 increased the paracellular conductance of cultured M D C K cells. 9 However, changes in active
INTRODUCTION
Sodium (Na ÷) enters epithelial cells across the apical (mucosal) membrane, through amiloride-sensitive pathways, down a favorable electrochemical potential gradient. Once inside the cell, it is actively transported across the basolateral (serosal) membrane by the Na ÷ :K + ATPase. ~ This process is linked to the movement of water across epithelial tissue and plays an important role in the reabsorption o f lung fluid from the fetal airspace and the maintenance of fluid-free alveolar space in the adult lung. 2,3 The incomplete reduction of oxygen leads to the formation of partially reduced oxygen species, which Supported by grants from the National Institutes of Health (HL 31197), the Health Effects Institute, a Career Investigator Award by the American Lung Association, and the Alabama Affiliate of the American Heart Association. Address correspondence to Sadis Matalon, PhD, Department of Anesthesiology, University of Alabama at Birmingham, 619 Nineteenth Street South, Birmingham, AL 35233, USA.
557
558
S. MATALON, J. S. BECKMAN, M. E. DUFFEr, and B. A. FREEMAN
Na + transport following exposure of epithelial tissues to partially reduced oxygen species have not been measured. In the present study, we quantified whether the presence of increased extracellular concentrations of partially reduced oxygen species would compromise active Na + transport across a tight epithelium, the ventral toad skin. The rate of Na + transport was assessed by measuring the short-circuit current (Isc) across this epithelium when mounted in an Ussing chamber. Partially reduced oxygen species were generated by xanthine plus xanthine oxidase (XO) added to either the mucosal or serosal sides of the toad skin. We quantified rates of production and steady-state concentrations of H202 and correlated these values with changes in epithelial short-circuit current and conductance in a temporal manner. Furthermore, we assessed whether the presence of proteases in XO compounds oxidant injury to Na + transport. The results of this study allow us to draw specific conclusions about the site and mechanisms of oxidant-induced injury to Na + transport across this epithelium.
MATERIALS AND M E T H O D S
Tissue isolation and electrophysiological measurements Sections of ventral abdominal skin of decapitated toads (Bufo marinus) were washed with an electrolyte solution containing 2.5 mM KHCO~, 111 mM Na2SO4, and 1.0 mM Ca2SO4. When bubbled with room air, the pH of this solution was 8.3. The tissue was then mounted vertically as a flat sheet between two conical Plexiglas hemichambers with an aperture of 1 cm 2. Each hemichamber was connected to a bubble-lift apparatus containing 15 ml electrolyte solution, which was thoroughly aerated and uniformly mixed by continuous bubbling with room air (21% 02, 79% N2). A voltage-current clamp apparatus (Physiologic Instruments; Model VCC 600) was used for the continuous monitoring of transepithelial potential difference (V,), short-circuit current (Isc), and transepithelial resistance (R,). The transepithelial potential difference was measured between Hg +/HgC1 electrodes, which made contact with the bathing solutions via agar bridges (3% agar in glucose-free, standard electrolyte solution), positioned at 3 mm at either side of the tissue. Short-circuit current was measured by passing sufficient current through Ag +/Ag electrodes, in contact with the bathing solutions via agar bridges, to reduce the spontaneous V, to zero. R, was determined using O h m ' s law from the observed change in Isc when V, was clamped at -+5-10 mV for 60 ms; chamber fluid
resistance was automatically subtracted by the voltage clamp. Tissues were continuously short-circuited. Short-circuit current was recorded in a two-channel strip-chart recorder throughout each experiment. Measurements were preceded by an equilibration period during which steady-state values for lsc and R, were reached (Isc > 40/~A/cm2; A Isc/hr < 5 ILA/ cm2; R, > 500 Ohm.cm2). Less than 10% of the preparations did not meet these criteria within three hours of mounting and were discarded. When lsc reached a stable value, xanthine and XO were added to the bathing fluid (see below) and changes in Isc and R, were followed for at least 60 min. At this time, 15/LI of 0.1 M amiloride was added into the mucosal bathing solution (final concentration: 10 4 M) and changes in lsc and R, were recorded for an additional 15 min.
Biochemical assays and measurements Generation of partially reduced oxygen species. Superoxide, H202, and "OH were generated in the extracellular medium by adding into either the mucosal or serosal bathing fluid 800 Itl of 10 mM xanthine (2,6 dihydroxypurine, sodium salt, grade IIl; Sigma Chemical Company; final bathing concentration: 500/tM) and 3 - 5 min later, 15 mU/ml of XO. Because XO may be self-inactivated by reactive oxygen products, "~ an additional 15 m U / m l of XO was instilled ten min later. Since even small changes in the pH of the perfusing medium are known to affect the value of Isc, ~' the xanthine stock solution was dissolved in the electrolyte solution perfusing the tissue and adjusted to a pH of 8.3 by the addition of dilute sulfuric acid. Two different types of XO were used in this study: the first one was purchased commercially (grade Ill Calbiochem, dissolved in 2.3 M NH4SO~) and was showing to contain significant amounts of protease activity; the second was purified from fresh cream in our laboratory according to the method of Waud et al. ~2 and shown to be essentially protease free (see Results). Just prior to instillation into the bathing fluid, all XO samples were chromatographed to remove interference from endogenous low molecular weight inhibitors and preservatives (ammonium sulfate or choline chloride). A prepacked Sephadex G-25 column (9 x 2 cm, Pharmacia) was equilibrated with the electrolyte solution and fitted through a bored-out # 8 cork in a 50 ml polyethylene tube. After removing all fluid by centrifugation, a small volume of XO, diluted in 1 ml electrolyte solution, was added to the column, which was placed in a new tube and centrifuged at 900 x g for five min. The protein sample was collected and its XO activity was assayed by monitoring the production of uric acid at 295 nm (A~ M = !.1 × 10 4 M-~ . cm ') at 25°C. The reaction mixture consisted of 50
Oxidant injury and sodium transport
#M xanthine in electrolyte solution (pH = 8.3). One unit of XO was defined as the amount of enzyme generating 1 /tmol/min of urate. In four control experiments Cu,Zn superoxide dismutase (SOD; 4500 U) and bovine liver catalase (CAT; 9000 U) were added to the perfusing medium prior to the XO instillation. SOD was generously provided by Diagnostic Data, Inc. (Mountain View, CA) and by Grunenthal GMBH (Aachen, West Germany). Catalase was obtained from Cooper Biomedical Corp (Freehold, N J). Both enzymes were dialyzed for 48 hours against cold electrolyte solution for ionic and pH equilibration. Measurement of protease activity. Xanthine oxidase protease activity was assessed by mixing various amounts of XO with 500/~1 azocasein (Calbiochem, 1 mg/ml), incubating the mixtures at 37°C for 30 min, precipitating the unfragmented protein with 500/tl of 10% tricbloroacetic acid (TCA), and measuring hydrolyzed fragments in the supernatant at 440 nm. Xanthine oxidase associated proteases caused the hydrolysis of azocasein to acid-soluble. Measurement of H202 concentration. Concentrations of H202 in the perfusion medium were measured throughout the course of the study in five experimental and three control preparations. One hundred/zl of perfusion fluid was mixed with 400/~1 of a 100 mM KPO4 mixture (pH = 7.0) containing 2000 U of horseradish peroxidase (type II, 200 U/rag; Sigma Chemical Company), 1.5 mM 4-aminoantipyrine (Sigma), and 0.11 M phenol (Phenol ACS reagent; Sigma). The assay solution also contained the XO inhibitor allopurinol (20 /IM) to prevent further generation of H202. The concentration of H202 was calculated from the absorba n c e a t 5 1 0 n m , usingan¢M = 6 . 5 8 m M ~ . c m Statistics. All data are presented as means + SE. Significant differences among several group means were evaluated using the one-way analysis of variance. If the F value was significantly different from zero (p < 0.05), the Bonferonni modification of the t-test was applied to compare group means.~3
RESULTS
559
sorbance and amounts of commercial XO added (r 2 = 0.95; p < 0.01). This relationship was abolished when XO was preincubated with 1 mM phenylmethylsulfonyl-fluoride (a specific protease inhibitor) and eluted through a Sephadex G-200 column prior to assay. Urate and H202 production The temporal profile of the rate of urate production generated by xanthine and XO was different when it was continuously assayed in a spectrophotometer cuvette versus obtaining sequential A295readings of samples removed from the Ussing chamber. In the spectrophotometer, there was a steady decrease in the rate of urate production, its value at ten min being approximately 50% of the initial one (Fig. 2). Addition of another aliquot of XO resulted in a temporary increase in urate production, followed by a steep decrease, while addition of xanthine or bubbling the cuvette with room air did not reverse the decrease in urate production (Fig. 2). These findings are consistent with a time-dependent inactivation of XO in the spectrophotometer cuvette sitting in the ultraviolet beam. However, XO was apparently not inactivated when added to the Ussing chamber (Fig. 2). In the absence of toad skin, urate production remained constant at about 18/~M/min, increased to 37 /~M/min after the instillation of the second bolus of XO, and then decreased to about 5 ~M/min by 30 min. In this case, instillation of xanthine prevented the decrease in urate production, indicating that the observed decrease was due to substrate depletion (data not shown). Assuming that XO remained fully active, the available substrate would be exhausted by 30 min after the first instillation of XO. As predicted from the rates of urate production in the Ussing chamber with no toad skin present (Fig. 2), H202 concentration increased with time during the first 30 min (Fig. 3), and reached a plateau value of around 400/~M. In contrast, in the presence of toad skin, this variable reached a peak value of only 200/IM and then decreased to 30 ~M. Thus, the toad skin was either consuming U202 directly or releasing substances with peroxidase activity into the perfusion medium. Similar rates of urate production and H202 concentrations were obtained with either the commercial or purified XO.
Xanthine oxidase activity
Short-circuit current (lsc) and epithelial resistance (Rt)
Figure 1 shows hydrolysis of azocasein to low molecular weight TCA-soluble fragments by XO-associated proteases, as reflected by changes in absorbance at 440 nm. Absorbance was unchanged when purified XO was added to the assay mixture. On the other hand, there was a linear correlation between changes in ab-
Figure 4 shows the time course of Isc in a typical preparation before and after the instillation of XO in the perfusion medium. Stable values for Isc and R, were obtained in all preparations within 90 rain of mounting the toad skin in the Ussing chambers. Mean values in 21 preparations (+SE) were: Isc = 70 +- 5/zA/cm2;
560
S. MATALON, J. S. BECKMAN, M. E. DUFFEY, and B. A. FREEMAN
200
o 1"
o aJ o
15o.
o o
.c) o
100
.
...
....
,, ...........
...........
v~ 1
50
'
0
I
4
,
I
,
8
I
12
XO
,
I
16
,
I
20
,
I
24
(mU/ml)
Fig. 1. Changes in light absorbance in 0.5 ml supernatants of azocasein j'ollowing the addition of' XO, and precipitation of' unfragmented proteins with 10% trichloroacetic acid. Values are expressed % change from control (no XO activity). Means -+ 1 SE. The solid line represents a linear least squares fit through the indicated points (r -~ = 0.95: p < 0.01). Symbols are as follows: • = XO purified from cream (n - 5) • = Calbiochem XO (n = 4) O = Calbiochem XO incubated with PMSF (n - 4)
R, = 881 -+ 59 ~ • cm 2. Instillation of amiloride in the mucosal bath ( 1 0 - 4 M ) at the end of the study decreased Isc to below 3 / t A / c m 2 within 10 min. Addition of xanthine or xanthine oxidase alone had no effect on any of the measured variables. Instillation of XO in the mucosal bath resulted in the immediate generation of H202 (Fig. 4) and caused a gradual decline in Isc and an increase in R,, starting at 20 min after instillation (Fig. 4). Similar changes were obtained with either commercial or purified XO. Pooled mean values for Isc and R, for both type of XO (expressed as percentage of the corresponding values just prior to instillation) are shown in Tables 1 and 2. Addition of SOD and CAT in the perfusion medium prevented the formation of H 2 0 2 and obviated changes in Isc and R,, indicating the involvement of partially reduced oxygen species in the observed reduction of Na + absorption. In contrast, instillation of xanthine and XO in the serosal bath had no appreciable effect in Isc and R, (Tables 1 and 2) in spite of the fact that similar amounts of H202 were generated in the perfusion medium. DISCUSSION
Xanthine oxidase catalyzes the conversion of xanthine to urate and reduces molecular oxygen to partially reduced oxygen species. At pH 8.3, approximately 30% of these species appear as 0 2 - and the rest as H202.~4 Since 0 2 - dismutes spontaneously or by enzymatic catalysis to H202, this more stable species was quantified to reflect levels of XO-mediated stress. The
m e a n H 2 0 2 in the fluid perfusing the toad skin after the instillation of xanthine and XO was 90 ~M. Tate et al. ~5reported that the levels of free radicals produced by 20 m U / m l of XO and 2 mM purine were equivalent to those generated in vivo by activated circulating neutrophils. Thus, the imposed stress was within physiological limits. In any event, in preliminary experiments, a single dose of XO (15 mU/ml) did not have any effect on either lsc or R,. Fridovich and coworkers reported XO inactivation during the oxidation of acetaldehyde or xanthine, ~0and initiation of membrane lipid peroxidation during coincubation of this reaction system with erythrocytes or egg phosphatidylcholine-containing liposomes.~°.~6 The rate of inactivation was dependent on the initial XO concentration (half-time = 5 min for - 12 mU XO). Superoxide dismutase and catalase scavenge 02 and H202, the primary reactive products of acetaldehyde and xanthine oxidation, and when added to incubation systems, inhibited inactivation of XO and peroxidative membrane damage• t6 We also saw about a 50% inhibition of 15 m U / m l XO 10 rain after the initiation of the reaction. XO activity was increased upon addition of another aliquot of XO, with no enhancement of XO activity after provision of additional substrate in a spectrophotometer cuvette. However, when similar assessment of XO activity over time was conducted on samples removed from the Ussing chamber, no inhibition of the initial XO activity occurred in the first 10 min (Fig. 2). Addition of another aliquot
Oxidant i~ury and sodium transport
561
35 30 c o O~
25
".~_,
20
E~
15 10 5 0 0
5
10
15
20
25
30
Time (rain) Fig. 2. Urate production vs. time. Xanthine oxidase (15 mU/ml) was added either in a quartz cuvette ( 0 , ©) or in the mucosal bath of an Ussing chamber ( . ) at t = O. They both contained 500 ,uM xanthine in electrolyte solution. At t = 10 min. a bolus of 15 mU/ml XO ( 0 , m) or 500 #M xanthine (O) were instilled in either the cuvette or the Ussing chamber. Each point represents the mean of at least four measurements.
of 15 mU/ml XO caused a transient increase in the rate of urate production followed by a decrease secondary to substrate depletion. This decrease was prevented by instillation of an additional bolus of xanthine. As mentioned above, this period of XO stability did not occur in the spectrophotometric assay system (Fig. 2). It is possible that prolonged exposure of the reaction system to ultraviolet light in the spectrophotometric
study exacerbated free radical inactivation of XO, or that the more aggressive buffer oxygenation in the Ussing chamber volatilized otherwise enzyme-damaging reaction products such as H202. In none of the experiments was 02 limiting to XO activity, as evidenced by calculations of 02 consumption and the lack of stimulation of XO activity when xanthine alone was added (Fig. 2). Na ÷ movement across the toad skin is a two-step
500
400
300 o :~ ea~ k..-d
200
100
/~
\6--~
,w
J
]
0
10
20
Toad skin
i
i
30 40 T i m e (rain)
t
i
50
60
Fig. 3. Measurements of H202 concentrations in an Ussing chamber in the presence ( 0 ) or absence ( A ) of toad skin. In both cases, the chambers contained electrolyte solution with 500 ~M xanthine. XO (15 mU/ml) was added to the chamber at t = 0 and t = 10 rain. Means - 1 SE; n = 5. Corresponding time points after 20 min are significantly different from each other at p < 0.02 level.
562
S. MATALON, J. S. BECKMAN, M. E. DUFFEY, and B. A. FREEMAN
120
ea
90
°
I[11 llll
o oo 30
TrTil+ : : : : : : : : : : : : : : : : : : : : : : : : : 0 20 40 60 80 Time
(rain)
Fig. 4. Time course of [sc in a typical experiment. Vertical excursions represent Isc changes when the clamping voltage was changed by l0 mV in a bipolar fashion. Tissue resistance can be calculated from O h m ' s law. Single and double arrows indicate the time of instillation of xanthine (X) and XO, respectively.
process: Na + crosses the apical membrane due to the existing electrochemical gradient through amiloride-sensitive channels and is then actively extruded across the basolateral membrane through the Na+: K+ATPase. ~5 In addition, a small amount of Na ÷ may move through paracellular junctions. In our preparation, Isc was predominantly caused by the intracellular movement of Na + because amiloride which specifically blocks Na + channels but does not affect paracellular Na + movement t7 rapidly decreased lsc to undetectable levels. Our results indicate that instillation of xanthine and XO in the apical side of the toadskin caused a significant decrease in Isc and an increase in R, in a timedependent fashion ten min after instillation of the second aliquot of XO. This lag time may be due to cell protection by antioxidant enzymes subject to oxidant inactivation 4 or depletable low molecular weight scavengers. In addition, as in the in vivo situation, there may be a period between cellular and biochemical lesions and onset of functional alterations. Addition of SOD and catalase in the perfusion medium prevented the formation of H202 by xanthine and XO, the decrease in lsc and the increase in R,, indicating that O2-, H202 and possibly .OH were directly involved in the initiation of these changes. A possible mechanism by which partially reduced 02 species may interfere with Na ÷ transport is by oxidizing sulfhydryl groups in Na + channel proteins. Addition of the organic mercurials mersalyl, p-chloromercuribenzoate, and p-chloromercuribenzene sulfonate to the solutions bathing the mucosal side of pieces
of rabbit colon mounted in an Ussing chamber interfered with Na + transport by reacting with sulfhydryl groups located in the apical membrane.~8 Also, partially reduced oxygen species may be reacting with a number of amino acid residues which are necessary for the proper functioning of Na + channels. Wang et al. ~9 reported that chloramine-T removed sodium channel inactivation of squid giant axons by modifying one of its methionine residues. Nonspecific reactions such as lipid peroxidation and protein inactivation may also contribute to the observed decrease in Isc. For example, it has been suggested that oxidation of membrane lipids may allow them to function as calcium ionophores, thereby increasing intracellular calcium levels 2° which, in turn may decrease the permeability of the apical membrane to sodium. 2~.z2 Since addition of xanthine and XO to the serosal side of the skin did not alter either the Isc or R, (Table 2), it is unlikely that the Na + :K + ATPase was inhibited. A number of studies suggest that the Na + :K + ATPase may be resistant to partially reduced 02 species. Spragg et al. 23 reported that exposure of cultured endothelial or P388D~ murine cells to 5 mM H202, an order of magnitude higher than in our study, resulted after three min in a large decrease in cellular ATP levels. However, N a + : K + ATPase activity remained unchanged from control conditions. Maridonneau et al. 24 demonstrated that the treatment of human erythrocytes with phenazine methosulfate, an agent known to result in increased production of partially reduced oxygen species, caused an increase in passive membrane K ÷ permeability, but had no effect on the
O x i d a n t injury a n d s o d i u m t r a n s p o r t
563
T a b l e 1. C h a n g e s in S h o r t - C i r c u i t C u r r e n t (Isc) a n d Tissue R e s i s t a n c e (R,) F o l l o w i n g Instillation o f X a n t h i n e a n d X a n t h i n e O x i d a s e in the A p i c a l ( M u c o s a l ) B a t h X + SOD + CAT + XO
X+XO Isc
R,
Time (min)
# of expts
0 5 10 15 25 35 45 55 65 75 85
9 9 9 9 9 9 8 7 6 4 4
99 98 95 86 75 64 55 49 47 43
86 88 90
4 4 4
13 ± 2* 5 ± I* 3 ± 1"
Isc # of expts
(% o f c o n t r o l ) I00 ± 1 ± 1 ± 2 ± 3*t ± 3*t ± 3*t ± 4*+ ± 5*t ± 4*t ± l*t
100 103 113 133 147 155 165 165 165 160
100 ± 1 ± 2 ± 4 ± 8*t ± 8*t ± 9*t ± 19"t ± 7*t ± 8*t ± 8*t
Amiloride 3 9 0 ± 90 4 1 0 ± 100
R, (% o f c o n t r o l )
4 4 4 4 4 4 4 4 3 3 3
100 100 ± 100 ± 99 ± 99 ± 98 ± 97 ± 96 ± 95 ± 93 ± 90 ±
3 3
22 ± 1 4.5 ± 3
0 0 l 1 1 1 1 I 1 I
100 100 ± 101 ± 104 ± 106 ± 109 ± 114 ± 105 ± 108 ± 106 ± 110 ±
0 1 2 2 2 4* 2 2 2 3*
380 ± 9 0 390 ± 9 0
Note. Numbers are means -+ 1 SE; X. SOD. CAT were added to the mucosal bath at t = 0; X = xanthine (500 liM): XO = xanthine oxidase (purified or Catbiochem); 15 mU/ml at t = 0 and t - 10 min. SOD = superoxide dismutase (4500 U); CAT = catalase (9,000 U); amiloride (10 4 M) was added at t = 85 min. *p < 0.05 from the corresponding control value. tp < 0.05 from the corresponding value at the same time interval while the bath contained SOD and CAT.
Na + : K + ATPase activity. Also, lsc is limited by the rate of entry of N a + through the apical membrane channels, so injury to the Na + :K + ATPase might not result in a decrease in Isc. Previous reports indicate that proteases may act synergistically with oxygen metabolites in injuring cells and tissues. 25 In this study, we were unable to show any difference in the responses of Isc between
Table 2. C h a n g e s in S h o r t - C i r c u i t C u r r e n t , (Isc), a n d Tissue R e s i s t a n c e (R,) F o l l o w i n g Instillation o f X a n t h i n e and X a n t h i n e O x i d a s e in the B a s o l a t e r a l (Serosal) B a t h X+XO Time (min)
# of expts
0 5 10 15 25 35 45 55 65
9 5 5 5 5 4 4 3 2
Isc
R, (% o f control)
100 100 ± 100 ± 102 ± 105 ± 108 ± 109 ± 108 ± 109 ±
1 1 1 2 1 1 3 0
100 101 ± 102 ± 104 ± 105 ± 109 ± 115 ± 110 ± 117
1 2 3 5 6 7* 5
Note. Numbers are means +- 1 SE; all chemicals were added to the mucosal bath at the specified times. X = xanthine (500/tM); t = 0. XO = xanthine oxidase (purified or Calbiochem); 15 mU/ml at t = 0 and t = 10 min. *p < 0.05 from corresponding control value.
commercially available XO that contained significant levels of proteases and purified XO that did not. Active sodium transport across alveolar epithelial cells is an important mechanism of fluid reabsorption from the alveolar to the interstitial space. A number of noxious agents that may damage the alveolar epithelium produce chemotactic agents that attract neutrophils to the interstitial space and stimulate them, causing the release of O2-, H202 and other toxic species. Our previous findings indicate that these species increase the alveolar paracellular permeability to a number of lipid-insoluble proteins. 7 The results of this present study indicate that they also may interfere with active Na + transport and thus compound the accumulation of fluid in the alveolar space. In conclusion, we found that active Na + transport across the toad skin can be inhibited by the addition of xanthine and XO in the apical side of this epithelium. This inhibition is caused by the production of 02 , H202 and possibly .OH, since it is completely prevented by the addition of SOD and catalase. Finally, since addition of these substances to the serosal side had no effect, we speculate that the observed decrease in Na ÷ transport is due to the action of partially reduced 02 species on apical membrane entry sites, possibly Na + channels. REFERENCES 1. M a c K n i g h t , A. D. V.; D e b o n a , D. R.; L e a f , A. S o d i u m transport a c r o s s t o a d u r i n a r y bladder: A m o d e l " t i g h t " e p i t h e l i u m . Physiol. Rev. 6 0 : 6 1 5 - 7 1 5 ; 1980.
564
S. MATALON, J. S. BECKMAN,M. E. DUFFEY, and B. A. FREEMAN
2. Matthay, M. A.; Widdicombe, J. H.; Staub, N. C. Clearance of alveolar fluid in sheep may involve an active ion transport process. Fed. Proc. 41:1244 abstr. (1982). 3. Goodman, B. E.; Brown, S. E. S.; Crandall, E. D. Regulation of transport across pulmonary alveolar epithelial cell monolayers. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 57:703-710; 1984. 4. Freeman, B. A.; Crapo. J. D. Biology of disease: free radicals and tissue injury. Lab. Invest. 47:412-426; 1982. 5. Freeman, B. A.; Crapo, J. D. Hyperoxia increases oxygen radical production in rat lungs and lung mifochondria. J. Biol. Chem. 256:10986-10992; 1981. 6. Kellogg. E. G.; Fridovich, I. Liposome oxidation and eryrhrocyte lysis by enzymatically generated superoxide and hydrogen peroxide. J. Biol. Chem. 252:6721-6728; 1977. 7. Matalon, S.; Egan, E. A. The effects of Oz breathing on the permeability of the alveolar epithelium to solute. J. Appl. Physiol.: Respirat. Environ. Exercise Phvsiol. 50:859-863; 198 I. 8. Miller. J. H.; Mendoza, S. A. Inhibition of sodium transport by hyperbaric oxygen in the toad urinary bladder. Undersea Biomed. Res. 41333-345; 1977. 9. Welsh, M. J.; Shasby, D. M.; Husted, R. M. Oxidants increase paracellular permeability in a cultured epithelial cell line. J. Clin. Invesr. 76:1 155-I 168; 1985. 1. Autoinactivation of xanthine oxi10. Lunch. R. E.; Fridovich. dase. The role of superoxide radicals and hydrogen peroxide. B&him. Biophys. Acta 571: 195-200; 1979. 11. Fischbarg, J.; Whittenbury, G. The effect of external pH on osmotic permeability, ion and fluid transport across isolated frog skin. J. Phpsiol. London 275:403-417; 1978. 12. Waud, W. R.; Brady, F. 0.: Wiley. R. D.; Rajagopalan. K. V. A new purification procedure for bovine milk xanthine oxidase: effect of proteolysis on the subunit structure. Arch. Biochem. Bioph?;s. 169:695-701; 1975. 13. Wallenstein, S. C.; Zuckner. C. L.: Fleisa. J. F. Some statistical methods useful in circulation research. Circ. Res. 47: 1-9; 1980. aspects of the production of super14. Fridovich, 1. Quantitative
15.
16.
17. 18.
19.
20.
21.
22.
23.
34.
25.
oxide anion radical by milk xanthine oxidase. .I. Biol. Chem. 245:4053-4057; 1970. Tate. R. M.; Vanbenthuysen, K. D.; Shasby, D. M.; McMurtry, I. F.: Repine, J. E. Oxygen-radical-mediated permeability edema and vasoconstriction in isolated perfused rabbit lungs. Am. Rev. Respir. Dis. 126:802-906; 1982. Kellogg, E. W.. III; Fridovich, I. Liposome oxidation and erythrocyte lysis by enzymically generated superoxide and hydrogen peroxide. J. Biol. Chem. 252:6721-6728; 1977. Benos, D. J.: Saccomani. G.; Sarigan-Sohraby. S. The epithelial sodium channel. J. Biol. Chem. 262:10613-10618; 1987. Luger, A.; Turnheim, K. Modification of cation permeability of rabbit descending colon by sulphydryl reagents. J. Physiol. 317:49-66; 1981. Wang, G. K.; Brodwi. M. S.; Eaton, D. C. Removal of sodium channel inactivation in squid axon by the oxidant chloramineT. J. Gen. Ph,vsio/. 86:289-302: 1985. Serham, C.: Anderson. P.; Dunham, P.; Weissmann, G. Phosphatidate and oxidized fatty acids are calcium ionophores. J. Biol. Chcm. 25612736-2741: 198 I. Garty. H.; Asher, C. Ca” dependent, temperature sensitive. regulation of Na channels in tight epithelia. A study using membrane vesicles. J. Biol. Chem. 260:8330-8335; 198.5. Garty. H.; Asher. C. Ca” induced down regulation of Na’ channels in toad bladder epithelium. J. Biol. Chem. 261:74007406; 1986. Spragg. R. G.; Hinshaw. D. B.; Hyslop, P. A.; Schraufstatter. I. U.; Cochrane, C. G. Alterations in adenosine trophosphate and energy change in cultured endothelial and P388D, cells after oxidant injury. J. C/in. lnvesr. 76:1471-1476; 1985. Maridonneau, I. P.; Braquet, R. P.; Garay. R. P. Na’ and K ’ transport damage induced by oxygen free radicals in human red cell membranes. J. Biol. Chem. 258:3107-31 13; 1983. Baird, B. R.: Cheronis. J. C.: Sandhaus, R. A.; Berger, E. M.; White. C. W.: Repine, J. E. 0, metabolites and neutrophil elastase synergistically cause edematous injury in isolated rat lungs. J. Appl. Physiol. 61:224-229: 1986.