~e;Copyright 1986 by The t|tunana Press Inc. All rights of any nature whatsoever reserved. 0163-4984/86/11004)019502.00
The Effect of Copper Ion on Giutathione and Hemolysis in Rabbit Erythrocytes JAMES M. CAFFREY, JR)'* ASOK DASMAHAPATRA~, HARRY A. SMITH ~, KARYN HEDE 2, AND EARL FRIEDEN ~ 'Dept. of Chemistry, Florida State University, Tallahassee, FL, 32306-3006; and 2Biology Dept., Stetson University, DeLand, FL 32720 Received April 4, 1986; Accepted June 22, 1986
ABSTRACT The rate of hemolysis and the decline in glutathione (GSH) in rabbit erythrocytes caused by copper (Cu) ions were determined. Prior investigations have proposed that the oxidative stress induced by Cu ion depleted the normal cell protective mechanisms. The decline in GSH has been proposed as a necessary prerequisite for hemolysis. We have observed that both GSH decline and hemolysis are Cu dependent, but are two concurrent and independent processes. We have confirmed that oxygen is a necessary reactant for hemolysis and responsible for a major portion of GSH decline. However, in the presence of Cu ion, a slow decline in GSH occurs even in a deaerated system. Index Entries: Hemolysis, in rabbit erythrocytes; rabbit erythrocytes; glutathione, effect of copper on; copper dependence for hemolysis; copper dependence for glutathione decline; copper phosphate; copper, ion, effect of on glutathione in rabbit erythrocytes; copper ion, effect of on hemolysis in rabbit erythrocytes.
INTRODUCTION The hemolysis of erythrocytes, red blood cells (RBC), induced by copper (Cu) ion [Cu(II)] is not well understood, although it is believed to "Author to w h o m all correspondence and reprint requests should be addressed.
Biological Trace Element Research
19
VoL 11, 1986
20
Caffrey et aL
result from an oxidative stress induced by Cu(II). Barnes and Frieden (1) s h o w e d that hemolysis did not occur if oxygen was displaced from the cell by either a nitrogen or carbon monoxide atmosphere. The hemolysis of RBC occurs in two phases---a prelytic induction phase, in which a minimal a m o u n t of hemolysis occurs, followed by a lytic, rapid hemolysis phase. To account for the prelytic period, the normal cell protective mechanisms are initially assumed operable, but gradually fail over the prelytic period. This leads to the rapid hemolysis phase. The role of glutathione (GSH) in converting activated oxygen species, e.g., h y d r o g e n peroxide, into water is well d o c u m e n t e d (2). Thus, the decrease in GSH level was a prime candidate for one of the failed protective mechanisms. To further elucidate the role of GSH decline, we determined GSH levels and hemolysis for rabbit erythrocytes. We also studied the effect of N2 and CO atmospheres on GSH levels. Sivertsen (3) has studied the Cu-ion induced decline in GSH levels in erythrocytes of a n u m b e r of mammalian species. The GSH level declined at different rates for each species. The zero-time GSH level for the erythrocytes in the absence of Cu ion was always slightly higher than the corresponding zero-time GSH level for a Cu-ion-containing system. Sivertsen also observed a minimal hemolysis (10% or less in 14 h). In contrast, Barnes and Frieden (1) observed almost complete hemolysis in 2 h.
MATERIALS AND METHODS
Cu Analysis The m e t h o d described by Schilt (4) was modified as follows: To a test tube, 0.500 mL of 1 mM sodium bathocuproine disulfonate (Sigma Chemical Co.), 0.500 mL of 10% hydroxylamine hydrochloride, and 1 mL of sample were added. The absorbance of the u n k n o w n at 479 nm was compared to a standard curve.
RBC Suspension A 6.0-mL total vol of RBC at 1.5~ hematocrit in buffer was used. The RBC at ice temperature were added to buffer first; then sufficient aqueous Cu sulfate (1.0 x 10 3M) was a d d e d to give the desired final concentration. A 20-mL counting vial was used. Incubation was carried out at 37~ in a swirling water bath for times from 0 to 180 min. A separate vial was prepared for each time. The data represent the average of three determinations.
RBC Preparation Whole blood from rabbits was collected in an equal vol of 40 mM citrate, 20 mM sodium p h o s p h a t e buffer, and 155 mM saline (sodium chloride). The RBC were separated by centrifugation and washed four times Biological Trace Element Research
Vol. 1 I, 1986
Erythrocyte Glutathione and Hemolysis
21
with either 20 mM phosphate and 155 mM saline buffer (PBS), pH 7.4 or 30 mM barbiturate/acetate and 155 mM saline buffer (BAS) pH 7.4. Finally, the RBC were suspended in the appropriate buffer.
Hemolysis The % hemolysis was determined by measuring the absorbance of an aliquot of an osmotically hemolyzed suspension of RBC at 410 nm and an aliquot of the supernatant after removal of the RBC. The absorbance value of supernatant divided by that of osmotically hemolyzed cells • 100 gave the % hemolysis.
Cu in RBC The RBC were centrifuged out of the reaction mixture, washed once with fresh buffer and centrifuged again. The cells were hemolyzed with 6 mL water. One milliliter of 90% trichloroacetic acid (TCA) was added to the hemolysate. The precipitated solids were removed by centrifugation. The supernatant liquid was extracted three times with ether to remove the TCA. Dissolved ether was removed by nitrogen purge. Any traces of TCA were neutralized with one drop of concentrated NH4OH. Onemilliliter samples were used in the above described Cu analysis.
GSH Determinations The RBC were separated by centrifugation, then used as is in the Buetler procedure (5). A modification of the osmotic hemolysis step was made by hemolyzing with 2 mL of I • 10 2M solution of ethylenediaminetetraacetic acid disodium salt (EDTA) to complex the Cu ion so that it would not react with the GSH.
Buffer Selection The question of a suitable buffer required special attention because we recovered an unaccounted for, bluish-white, Cu phosphate precipitate together with RBC from a suspension of RBC, 1 • 10-*M Cu(II), and 20 mM PBS, upon centrifuging the suspension. The Cu phosphate precipitate was identified by comparing its Raman spectrum with the spectrum of a known Cu phosphate. Concurrently, we observed a lower Cu ion concentration in the supernatant liquid. In prior RBC-Cu ion studies, phosphate, HEPES, and Tris buffers were used. For this study, a suitable buffer system had to meet several requirements. First, the pH needed to be maintained at 7.4. Second, the Cu ions must remain in solution. Finally, the Cu ion must induce hemolysis. Tris (10-100 mM) and BAS (30 mM) gave no precipitation with 1 • 10-4M Cu(II). Phosphate (PBS, 20 mM), HEPES (10-100 mM), and Pipes (10-100 mM) all formed precipitates with 1 • 10 4M Cu(II). (PrecipitaBiological Trace Element Research
VoL I 1, 1986
Caffrey e t aL
22
tion was observed on mixing at 0 or at 37~ after incubation for I h.) We further evaluated Tris and BAS as a m e d i u m for hemolysis. We observed no hemolysis with Tris, but hemolysis occurred with BAS. Buffers, such as Tris or HEPES, used by Sivertsen (3), have been s h o w n by Hegetschweiler (6) to complex the Cu ion strongly. Thus no hemolysis occurred. Since several investigators have used PBS to carry out Cu-induced reactions with RBC suspensions, we compared the a m o u n t of Cu ion precipitation for PBS and BAS in more detail by determining the percentage of initially a d d e d Cu ion remaining in the supernatant of each buffer after centrifuging. O n e - h u n d r e d percent of the initial Cu ion remained in solution using BAS buffer either on initial mixing at 0~ or after incubation at 37~ for 0.5 h. Conversely, w h e n using PBS buffer, 50% of the initial Cu ion precipitated on initial mixing at 0~ 100% was lost after 0.5 h of incubation at 37~ Buffers, such as phosphates, that can form insoluble precipitates with Cu ion, leave the actual soluble Cu ion concentration in the system indeterminant. Also, the insoluble Cu ion may contribute to u n w a n t e d reactions. Figure 1 illustrates an indeterminant Cu concentration (see curve b---PBS buffer) giving a longer prelytic period. Figure 2 shows a series of hemolysis vs time curves for Cu ion concentrations between 1 • 10 5 and 1 x 10-4M. We observed no hemolysis at 1 x 10 5M Cu(lI). For increasing concentrations of Cu ion, typical lysis curves with shorter prelytic periods at higher concentrations were observed. Using the hemolysis curve (b) from Fig. 2, which closely matches the hemolysis curve (b) in PBS buffer in Fig. 1, we estimate that 50% of the Cu ions are active, even though substantial total precipitation of Cu ion occurred. I00 --
(a)
d 60
-
o _
~'~"I"--~1 30
I
60 90 MINUTES
I
120
Fig. 1. Comparison of hemolysis for PBS and BAS buffers. Procedure: The reaction vol. 6.0 in a 20-mL counting vial; 1.5% hematocrit; RBC from rabbit washed four times with buffer of the experiment. Average of three determinations. Ordinate, % hemolysis; abcissa time, min; % hemolysis: (a) in BAS, (b) in PBS. Biological Trace Element Research
VoL 11, 1986
Erythrocyte Glutathione and Hemolysis I00
23
--
(a) 60
o 20
MINUTES Fig. 2. Percentage hemolysis vs time for several initial Cu concentrations using BAS buffer. Procedure: See Fig. 1 legend; ordinate, % hemolysis; abcissa time, min; % hemolysis for different Cu(II) concentrations, (a) 1 x 10- 4M; (b) 5 X 10 5M; (c) 3 x I0 5M; and (d) 1 x 10-~M. An example of an unwanted chemical reaction probably caused by precipitated Cu phosphate is an initial instantaneous drop of 50-60% in GSH level using phosphate buffer. No such drop was observed with BAS. The question that arises is, is the GSH undergoing an intracellular reaction between PBS, Cu(II), and RBC that is reducing the GSH before analysis or is the reduced GSH level an artifact occurring at the time of the hemolysis step during GSH analysis? The reduced GSH level observed in the system [PBS-RBC-1 x 10-4M Cu(II)] can be shown to be an artifact if we suppress two reactions: First, a Cu ion-GSH reaction inhibited by EDTA (modified Buetler GSH procedure); and second, a GSH-O2 reaction (Cu ion mediated) suppressed by deaerating the system with nitrogen just prior to analysis.
RESULTS AND DISCUSSION
GSH Levels and Hemolysis~Aerated Cu Solutions for Cell Suspension Barnes and Frieden (1) suggested that the Cu(II)-induced lysis may result from increased O 2 (or other activated oxygen compounds--H202, 9OH, 02 (singlet), which eventually overpowers the cell's natural protective mechanisms against oxidative attack. Maintaining GSH levels may be an important part of this protective mechanism. We have determined GSH levels and hemolysis on the same samples in an attempt to detect any link between hemolysis and GSH levels. We have correlated GSH declines for increasing incubation times at 37~ using the RBC suspension described in the materials section, with various Cu ion concentrations, as shown in Fig. 3 curves (a), (b), and (c). Biological Trace Element Research
Vol. I 1, 1986
Caffrey et aL
24
2501--
I00-o
5o-
O0
I
50
I
60
90
I 120
MINUTES Fig. 3. GSH Levels vs time for different Cu(II) concentrations in BAS. Procedure: See Fig. 1 legend; ordinate ~mol GSH/100 mL RBC; abcissa time, rain; Cu(II), (a) 1 x 10 5M; (b) 3 x 10 5M;and (c) 1 x 10 4M. The decline in GSH levels at 1 x 10- SM Cu(II) is slower than for 1 x 10 4M Cu(II). An intermediate concentration of 3 x 10 5M Cu(II) gives an intermediate rate of decline. Similarly, we have s h o w n that the onset of hemolysis requires a shorter time for increasing Cu(II) concentration, as s h o w n in Fig. 2. Although we have s h o w n above that both processes are Cu ion d e p e n d e n t , we s h o w below that hemolysis does not result from the loss of a specific a m o u n t of GSH. For example, in Fig. 3, curve (a), the GSH concentration of 100 ~mol/100 mL RBC, 1 x 10 SM initial Cu(II), occurs after 100 min. Hemolysis of a 1 x 10 -~M Cu ion suspension of RBC is only 3% after 100 min [see Fig. 2, curve (d)]. Similarly, a GSH concentration of 100 ~mol/100 mL RBC, in 1 • 10- 4 M Cu(II) occurs after 48 rain [see Fig. 3, curve(c)]. After 48 min the RBC suspension containing 1 x 10-4 M Cu(II) was 56% hemolyzed [see Fig. 2, curve (a)]. Clearly, the loss of GSH is not an absolute requirement for hemolysis. We conclude that GSH loss and hemolysis are two concurrent and indep e n d e n t processes.
GSH Levels and Hemolysis---Aerated Buffer for Suspension of CuTreated Cells Next we m a d e a study of hemolysis and GSH levels u n d e r a set of special conditions for exposing the cell to Cu. The Cu ion in contact with the cell is quickly taken up by the RBC suspension in BAS. Initially, at 1 x 10-4M Cu(II), approximately 20% of the Cu enters the cell, roughly 150 ~,mols/100 mL RBC. Over the prelytic period (-20-30 min) the concentrations increased to approximately 300 ~moV100 mL RBC. Likewise, at an initial concentration of 5 x 10 4M Cu(II), the initial Cu ion concentration in the cell was approximately 405 ~mol/100 mL RBC. Biological Trace Element Research
Vol. I 1, 1986
Erythrocyte Glutathione and Hemolysis
25
Using the above data as a guide, we modified our normal p r o c e d u r e - - i n c u b a t i o n in the presence of a Cu ion solution. In this series of experiments, we incubated the cells in fresh buffer after a short-time exposure to Cu ion. (The exposure to Cu ion was only long e n o u g h to add the Cu ion, spin the cells down, and wash and resuspend the cells in fresh buffer.) We observed that cells exposed to Cu ion for only a short period lyse differently from cells incubated in Cu ion, and the cells do show a typical linear GSH decline. First, hemolysis was compared for the short exposure to 5 x 10 -4M Cu(II) to incubation in 1 x 10 -4M Cu(II). The hemolysis induced by incubation at 1 • 10 4M Cu has a prelytic period of 20-30 rain, followed by rapid lysis [Fig. 2, curve (a)]. The hemolysis i n d u c e d by a short exposure to 5 x 10- 4M Cu(II) has no prelytic period and only a slow linear lysis, reaching 25% lysis in 60 rain [Fig. 4, curve (b)]. The rate of decline of GSH for the short exposure to 5 x 10-4M Cu(II) is slightly greater than the rate of decline for the normal incubation of 1 x 10- M Cu(II) (slopes - 2 . 8 vs -2.3). In the above data, GSH declines are similar, whereas hemolysis appears to follow different patterns. Thus, this additional evidence reinforces our conclusion that GSH decline and hemolysis are two concurrent and i n d e p e n d e n t processes.
GSH Level and Hemo!ysis~--Deaerated Solutions We examined the effect of nitrogen displacement on GSH levels vs time of incubation at 37~ In Fig. 5, we compare a Cu-ion-free system with both aerated and nitrogen or carbon monoxide deaerated systems.
oo[-
250
( • 60
150
0
I 30 MINUTES
",
5O
60
Fig. 4. Short time Cu(lI) exposure--hemolysis vs time, and GSH decline vs time for Cu(II) concentrations of 5 x 10- 4M in BAS. Procedure: Short exposure to Cu (see text). Left Ordinate--% hemolysis, right ordinate I~mol GSH/100 mL abcissa time, min; Cu -- 5 x 10 4M, (a) GSH decline (b) hemolysis. Biological Trace Element Research
Vol. I t, t 986
Caffrey et aL
26 250~
150
g,oo_ 50
O0
(c) ,
I
60
,
I
120
MINUTES
Fig. 5. GSH levels vs time--comparison of deaerated to aerated RBC. Procedure: See Fig. 1 for curves (a) and (c); in curve (b) either N2 or CO purged through sample for 20 rain. Vials capped without 02 exposure. Ordinate--p, mol GSH/100 mL RBC; abcissa time, rain; (a) Copper free; (b) Nitrogen or CO deaerated; and (c) Aerated. 1 x ] 0 - 4 M Cu(II) in (b) and (c). Aerated RBC s h o w the most rapid decline in GSH, deaerated a slower GSH decline, a n d Cu free, no GSH decline within the experimental period. The Cu-ion free a n d deaerated systems did not result in hemolysis. The rate of decline in GSH for both nitrogen- a n d CO-deaerated systems was the same. The data of Fig. 5 indicate that a C u - i n d u c e d , n o n o x y g e n d e p e n d e n t reaction occurs with GSH, resulting in its decline. The observations of Crook (2) on sulfhydryl oxidation in the absence of oxygen m a y account for the C u - i o n - i n d u c e d decline in GSH levels.
ACKNOWLEDGMENT S u p p o r t e d by NIH Grant AM 33540.
REFERENCES 1. G. Barnes and E. Frieden, Biochem. Biophys. Res. Comm. 115, 2, 680 (1983). 2. E. M. Crook, Glutathione-Biochemical Society Symposium #17, 9, University Press, Cambridge, 1959. 3. T. Sivertsen, Acta Pharmacol. Toxicol. 46, 121 (1980). 4. A. A. Schilt, Analytical Applications of I, lO-phenanthroline and Related Compounds, Pergamon, Oxford, 1969, p. 74. 5. E. Buetler, Red Cell Metabolism, Grune and Stratton, Inc., Third Ed., Orlando, FI, 1984, p. 131. 6. K. Hegetschweiler and P. Saltman, lnorg. Chem. 25, 107 (1986).
Biological Trace Element Research
Vol. 11, ] 986
The Effect of Copper Ion on Giutathione and Hemolysis in Rabbit Erythrocytes JAMES M. CAFFREY, JR)'* ASOK DASMAHAPATRA~, HARRY A. SMITH ~, KARYN HEDE 2, AND EARL FRIEDEN ~ 'Dept. of Chemistry, Florida State University, Tallahassee, FL, 32306-3006; and 2Biology Dept., Stetson University, DeLand, FL 32720 Received April 4, 1986; Accepted June 22, 1986
ABSTRACT The rate of hemolysis and the decline in glutathione (GSH) in rabbit erythrocytes caused by copper (Cu) ions were determined. Prior investigations have proposed that the oxidative stress induced by Cu ion depleted the normal cell protective mechanisms. The decline in GSH has been proposed as a necessary prerequisite for hemolysis. We have observed that both GSH decline and hemolysis are Cu dependent, but are two concurrent and independent processes. We have confirmed that oxygen is a necessary reactant for hemolysis and responsible for a major portion of GSH decline. However, in the presence of Cu ion, a slow decline in GSH occurs even in a deaerated system. Index Entries: Hemolysis, in rabbit erythrocytes; rabbit erythrocytes; glutathione, effect of copper on; copper dependence for hemolysis; copper dependence for glutathione decline; copper phosphate; copper, ion, effect of on glutathione in rabbit erythrocytes; copper ion, effect of on hemolysis in rabbit erythrocytes.
INTRODUCTION The hemolysis of erythrocytes, red blood cells (RBC), induced by copper (Cu) ion [Cu(II)] is not well understood, although it is believed to "Author to w h o m all correspondence and reprint requests should be addressed.
Biological Trace Element Research
19
VoL 11, 1986
20
Caffrey et aL
result from an oxidative stress induced by Cu(II). Barnes and Frieden (1) s h o w e d that hemolysis did not occur if oxygen was displaced from the cell by either a nitrogen or carbon monoxide atmosphere. The hemolysis of RBC occurs in two phases---a prelytic induction phase, in which a minimal a m o u n t of hemolysis occurs, followed by a lytic, rapid hemolysis phase. To account for the prelytic period, the normal cell protective mechanisms are initially assumed operable, but gradually fail over the prelytic period. This leads to the rapid hemolysis phase. The role of glutathione (GSH) in converting activated oxygen species, e.g., h y d r o g e n peroxide, into water is well d o c u m e n t e d (2). Thus, the decrease in GSH level was a prime candidate for one of the failed protective mechanisms. To further elucidate the role of GSH decline, we determined GSH levels and hemolysis for rabbit erythrocytes. We also studied the effect of N2 and CO atmospheres on GSH levels. Sivertsen (3) has studied the Cu-ion induced decline in GSH levels in erythrocytes of a n u m b e r of mammalian species. The GSH level declined at different rates for each species. The zero-time GSH level for the erythrocytes in the absence of Cu ion was always slightly higher than the corresponding zero-time GSH level for a Cu-ion-containing system. Sivertsen also observed a minimal hemolysis (10% or less in 14 h). In contrast, Barnes and Frieden (1) observed almost complete hemolysis in 2 h.
MATERIALS AND METHODS
Cu Analysis The m e t h o d described by Schilt (4) was modified as follows: To a test tube, 0.500 mL of 1 mM sodium bathocuproine disulfonate (Sigma Chemical Co.), 0.500 mL of 10% hydroxylamine hydrochloride, and 1 mL of sample were added. The absorbance of the u n k n o w n at 479 nm was compared to a standard curve.
RBC Suspension A 6.0-mL total vol of RBC at 1.5~ hematocrit in buffer was used. The RBC at ice temperature were added to buffer first; then sufficient aqueous Cu sulfate (1.0 x 10 3M) was a d d e d to give the desired final concentration. A 20-mL counting vial was used. Incubation was carried out at 37~ in a swirling water bath for times from 0 to 180 min. A separate vial was prepared for each time. The data represent the average of three determinations.
RBC Preparation Whole blood from rabbits was collected in an equal vol of 40 mM citrate, 20 mM sodium p h o s p h a t e buffer, and 155 mM saline (sodium chloride). The RBC were separated by centrifugation and washed four times Biological Trace Element Research
Vol. 1 I, 1986
Erythrocyte Glutathione and Hemolysis
21
with either 20 mM phosphate and 155 mM saline buffer (PBS), pH 7.4 or 30 mM barbiturate/acetate and 155 mM saline buffer (BAS) pH 7.4. Finally, the RBC were suspended in the appropriate buffer.
Hemolysis The % hemolysis was determined by measuring the absorbance of an aliquot of an osmotically hemolyzed suspension of RBC at 410 nm and an aliquot of the supernatant after removal of the RBC. The absorbance value of supernatant divided by that of osmotically hemolyzed cells • 100 gave the % hemolysis.
Cu in RBC The RBC were centrifuged out of the reaction mixture, washed once with fresh buffer and centrifuged again. The cells were hemolyzed with 6 mL water. One milliliter of 90% trichloroacetic acid (TCA) was added to the hemolysate. The precipitated solids were removed by centrifugation. The supernatant liquid was extracted three times with ether to remove the TCA. Dissolved ether was removed by nitrogen purge. Any traces of TCA were neutralized with one drop of concentrated NH4OH. Onemilliliter samples were used in the above described Cu analysis.
GSH Determinations The RBC were separated by centrifugation, then used as is in the Buetler procedure (5). A modification of the osmotic hemolysis step was made by hemolyzing with 2 mL of I • 10 2M solution of ethylenediaminetetraacetic acid disodium salt (EDTA) to complex the Cu ion so that it would not react with the GSH.
Buffer Selection The question of a suitable buffer required special attention because we recovered an unaccounted for, bluish-white, Cu phosphate precipitate together with RBC from a suspension of RBC, 1 • 10-*M Cu(II), and 20 mM PBS, upon centrifuging the suspension. The Cu phosphate precipitate was identified by comparing its Raman spectrum with the spectrum of a known Cu phosphate. Concurrently, we observed a lower Cu ion concentration in the supernatant liquid. In prior RBC-Cu ion studies, phosphate, HEPES, and Tris buffers were used. For this study, a suitable buffer system had to meet several requirements. First, the pH needed to be maintained at 7.4. Second, the Cu ions must remain in solution. Finally, the Cu ion must induce hemolysis. Tris (10-100 mM) and BAS (30 mM) gave no precipitation with 1 • 10-4M Cu(II). Phosphate (PBS, 20 mM), HEPES (10-100 mM), and Pipes (10-100 mM) all formed precipitates with 1 • 10 4M Cu(II). (PrecipitaBiological Trace Element Research
VoL I 1, 1986
Caffrey e t aL
22
tion was observed on mixing at 0 or at 37~ after incubation for I h.) We further evaluated Tris and BAS as a m e d i u m for hemolysis. We observed no hemolysis with Tris, but hemolysis occurred with BAS. Buffers, such as Tris or HEPES, used by Sivertsen (3), have been s h o w n by Hegetschweiler (6) to complex the Cu ion strongly. Thus no hemolysis occurred. Since several investigators have used PBS to carry out Cu-induced reactions with RBC suspensions, we compared the a m o u n t of Cu ion precipitation for PBS and BAS in more detail by determining the percentage of initially a d d e d Cu ion remaining in the supernatant of each buffer after centrifuging. O n e - h u n d r e d percent of the initial Cu ion remained in solution using BAS buffer either on initial mixing at 0~ or after incubation at 37~ for 0.5 h. Conversely, w h e n using PBS buffer, 50% of the initial Cu ion precipitated on initial mixing at 0~ 100% was lost after 0.5 h of incubation at 37~ Buffers, such as phosphates, that can form insoluble precipitates with Cu ion, leave the actual soluble Cu ion concentration in the system indeterminant. Also, the insoluble Cu ion may contribute to u n w a n t e d reactions. Figure 1 illustrates an indeterminant Cu concentration (see curve b---PBS buffer) giving a longer prelytic period. Figure 2 shows a series of hemolysis vs time curves for Cu ion concentrations between 1 • 10 5 and 1 x 10-4M. We observed no hemolysis at 1 x 10 5M Cu(lI). For increasing concentrations of Cu ion, typical lysis curves with shorter prelytic periods at higher concentrations were observed. Using the hemolysis curve (b) from Fig. 2, which closely matches the hemolysis curve (b) in PBS buffer in Fig. 1, we estimate that 50% of the Cu ions are active, even though substantial total precipitation of Cu ion occurred. I00 --
(a)
d 60
-
o _
~'~"I"--~1 30
I
60 90 MINUTES
I
120
Fig. 1. Comparison of hemolysis for PBS and BAS buffers. Procedure: The reaction vol. 6.0 in a 20-mL counting vial; 1.5% hematocrit; RBC from rabbit washed four times with buffer of the experiment. Average of three determinations. Ordinate, % hemolysis; abcissa time, min; % hemolysis: (a) in BAS, (b) in PBS. Biological Trace Element Research
VoL 11, 1986
Erythrocyte Glutathione and Hemolysis I00
23
--
(a) 60
o 20
MINUTES Fig. 2. Percentage hemolysis vs time for several initial Cu concentrations using BAS buffer. Procedure: See Fig. 1 legend; ordinate, % hemolysis; abcissa time, min; % hemolysis for different Cu(II) concentrations, (a) 1 x 10- 4M; (b) 5 X 10 5M; (c) 3 x I0 5M; and (d) 1 x 10-~M. An example of an unwanted chemical reaction probably caused by precipitated Cu phosphate is an initial instantaneous drop of 50-60% in GSH level using phosphate buffer. No such drop was observed with BAS. The question that arises is, is the GSH undergoing an intracellular reaction between PBS, Cu(II), and RBC that is reducing the GSH before analysis or is the reduced GSH level an artifact occurring at the time of the hemolysis step during GSH analysis? The reduced GSH level observed in the system [PBS-RBC-1 x 10-4M Cu(II)] can be shown to be an artifact if we suppress two reactions: First, a Cu ion-GSH reaction inhibited by EDTA (modified Buetler GSH procedure); and second, a GSH-O2 reaction (Cu ion mediated) suppressed by deaerating the system with nitrogen just prior to analysis.
RESULTS AND DISCUSSION
GSH Levels and Hemolysis~Aerated Cu Solutions for Cell Suspension Barnes and Frieden (1) suggested that the Cu(II)-induced lysis may result from increased O 2 (or other activated oxygen compounds--H202, 9OH, 02 (singlet), which eventually overpowers the cell's natural protective mechanisms against oxidative attack. Maintaining GSH levels may be an important part of this protective mechanism. We have determined GSH levels and hemolysis on the same samples in an attempt to detect any link between hemolysis and GSH levels. We have correlated GSH declines for increasing incubation times at 37~ using the RBC suspension described in the materials section, with various Cu ion concentrations, as shown in Fig. 3 curves (a), (b), and (c). Biological Trace Element Research
Vol. I 1, 1986
Caffrey et aL
24
2501--
I00-o
5o-
O0
I
50
I
60
90
I 120
MINUTES Fig. 3. GSH Levels vs time for different Cu(II) concentrations in BAS. Procedure: See Fig. 1 legend; ordinate ~mol GSH/100 mL RBC; abcissa time, rain; Cu(II), (a) 1 x 10 5M; (b) 3 x 10 5M;and (c) 1 x 10 4M. The decline in GSH levels at 1 x 10- SM Cu(II) is slower than for 1 x 10 4M Cu(II). An intermediate concentration of 3 x 10 5M Cu(II) gives an intermediate rate of decline. Similarly, we have s h o w n that the onset of hemolysis requires a shorter time for increasing Cu(II) concentration, as s h o w n in Fig. 2. Although we have s h o w n above that both processes are Cu ion d e p e n d e n t , we s h o w below that hemolysis does not result from the loss of a specific a m o u n t of GSH. For example, in Fig. 3, curve (a), the GSH concentration of 100 ~mol/100 mL RBC, 1 x 10 SM initial Cu(II), occurs after 100 min. Hemolysis of a 1 x 10 -~M Cu ion suspension of RBC is only 3% after 100 min [see Fig. 2, curve (d)]. Similarly, a GSH concentration of 100 ~mol/100 mL RBC, in 1 • 10- 4 M Cu(II) occurs after 48 rain [see Fig. 3, curve(c)]. After 48 min the RBC suspension containing 1 x 10-4 M Cu(II) was 56% hemolyzed [see Fig. 2, curve (a)]. Clearly, the loss of GSH is not an absolute requirement for hemolysis. We conclude that GSH loss and hemolysis are two concurrent and indep e n d e n t processes.
GSH Levels and Hemolysis---Aerated Buffer for Suspension of CuTreated Cells Next we m a d e a study of hemolysis and GSH levels u n d e r a set of special conditions for exposing the cell to Cu. The Cu ion in contact with the cell is quickly taken up by the RBC suspension in BAS. Initially, at 1 x 10-4M Cu(II), approximately 20% of the Cu enters the cell, roughly 150 ~,mols/100 mL RBC. Over the prelytic period (-20-30 min) the concentrations increased to approximately 300 ~moV100 mL RBC. Likewise, at an initial concentration of 5 x 10 4M Cu(II), the initial Cu ion concentration in the cell was approximately 405 ~mol/100 mL RBC. Biological Trace Element Research
Vol. I 1, 1986
Erythrocyte Glutathione and Hemolysis
25
Using the above data as a guide, we modified our normal p r o c e d u r e - - i n c u b a t i o n in the presence of a Cu ion solution. In this series of experiments, we incubated the cells in fresh buffer after a short-time exposure to Cu ion. (The exposure to Cu ion was only long e n o u g h to add the Cu ion, spin the cells down, and wash and resuspend the cells in fresh buffer.) We observed that cells exposed to Cu ion for only a short period lyse differently from cells incubated in Cu ion, and the cells do show a typical linear GSH decline. First, hemolysis was compared for the short exposure to 5 x 10 -4M Cu(II) to incubation in 1 x 10 -4M Cu(II). The hemolysis induced by incubation at 1 • 10 4M Cu has a prelytic period of 20-30 rain, followed by rapid lysis [Fig. 2, curve (a)]. The hemolysis i n d u c e d by a short exposure to 5 x 10- 4M Cu(II) has no prelytic period and only a slow linear lysis, reaching 25% lysis in 60 rain [Fig. 4, curve (b)]. The rate of decline of GSH for the short exposure to 5 x 10-4M Cu(II) is slightly greater than the rate of decline for the normal incubation of 1 x 10- M Cu(II) (slopes - 2 . 8 vs -2.3). In the above data, GSH declines are similar, whereas hemolysis appears to follow different patterns. Thus, this additional evidence reinforces our conclusion that GSH decline and hemolysis are two concurrent and i n d e p e n d e n t processes.
GSH Level and Hemo!ysis~--Deaerated Solutions We examined the effect of nitrogen displacement on GSH levels vs time of incubation at 37~ In Fig. 5, we compare a Cu-ion-free system with both aerated and nitrogen or carbon monoxide deaerated systems.
oo[-
250
( • 60
150
0
I 30 MINUTES
",
5O
60
Fig. 4. Short time Cu(lI) exposure--hemolysis vs time, and GSH decline vs time for Cu(II) concentrations of 5 x 10- 4M in BAS. Procedure: Short exposure to Cu (see text). Left Ordinate--% hemolysis, right ordinate I~mol GSH/100 mL abcissa time, min; Cu -- 5 x 10 4M, (a) GSH decline (b) hemolysis. Biological Trace Element Research
Vol. I t, t 986
Caffrey et aL
26 250~
150
g,oo_ 50
O0
(c) ,
I
60
,
I
120
MINUTES
Fig. 5. GSH levels vs time--comparison of deaerated to aerated RBC. Procedure: See Fig. 1 for curves (a) and (c); in curve (b) either N2 or CO purged through sample for 20 rain. Vials capped without 02 exposure. Ordinate--p, mol GSH/100 mL RBC; abcissa time, rain; (a) Copper free; (b) Nitrogen or CO deaerated; and (c) Aerated. 1 x ] 0 - 4 M Cu(II) in (b) and (c). Aerated RBC s h o w the most rapid decline in GSH, deaerated a slower GSH decline, a n d Cu free, no GSH decline within the experimental period. The Cu-ion free a n d deaerated systems did not result in hemolysis. The rate of decline in GSH for both nitrogen- a n d CO-deaerated systems was the same. The data of Fig. 5 indicate that a C u - i n d u c e d , n o n o x y g e n d e p e n d e n t reaction occurs with GSH, resulting in its decline. The observations of Crook (2) on sulfhydryl oxidation in the absence of oxygen m a y account for the C u - i o n - i n d u c e d decline in GSH levels.
ACKNOWLEDGMENT S u p p o r t e d by NIH Grant AM 33540.
REFERENCES 1. G. Barnes and E. Frieden, Biochem. Biophys. Res. Comm. 115, 2, 680 (1983). 2. E. M. Crook, Glutathione-Biochemical Society Symposium #17, 9, University Press, Cambridge, 1959. 3. T. Sivertsen, Acta Pharmacol. Toxicol. 46, 121 (1980). 4. A. A. Schilt, Analytical Applications of I, lO-phenanthroline and Related Compounds, Pergamon, Oxford, 1969, p. 74. 5. E. Buetler, Red Cell Metabolism, Grune and Stratton, Inc., Third Ed., Orlando, FI, 1984, p. 131. 6. K. Hegetschweiler and P. Saltman, lnorg. Chem. 25, 107 (1986).
Biological Trace Element Research
Vol. 11, ] 986
