- Email: [email protected]
Inactivation of glutathione peroxidase following entrapment of purified α or β hemoglobin chains in human erythrocytes
Clinica Chimica Acta. 217 (1993) 187-192 © 1993 Elsevier Science Publishers B.V. All rights reserved. 0009-8981/93/$06.00
187
CCA05547
"f" Inactivation of gl utathlone peroxidase following entrapment of purified, or/3 hemoglobin chains in human erythrocytes
Francesco Grelloni, Rosita Gabbianelli, Anna M a r i a Santroni and Giancarlo Falcioni Department of Molecular, Cellular and Animal Biology, University of Camerino, Camerino (Italy) (Received 20 May 1992; revision received 8 February 1993; accepted 13 March 1993) Key words: Glutathione peroxidase; Loaded erythrocytes; Human hemoglobin chains: Thalassemia
Summary Inactivation of glutathione peroxidase correlates with the rate of hemoglobin chain oxidation. The enzyme inactivation is mainly present in those conditions where the autoxidation of the oxygenated chains is followed by trasformation of the oxidized molecule into a hemichrome. Free hemoglobin chains have been encaps,dated i~' human red blood cells by a dialysis technique that involves transient hypotonic hemolysis followed by isotonic resealing Chain-loaded erythrocytes represent a good in vitro model of thalassemia. The presence of free human chains in the cell alters the intraerythrocytic glutathione peroxidase activity (oLchains are more effective in the inactivation of the enzyme with respect to the 0 chains).
Introduction Entrapment of isolated hemoglobin chains (c~ or 0) in human erythrocytes leads to the formation of an in vitro model of thalassemia. These resealed red blood cells represent a model system for studying the mechanisms involved in thalassemic pathophysiology due to the presence of free chains, without having to use patient samples where damage has already occurred in vivo. It has been reported that entrapment of purified a chains in normal human erythrocytes leads to structural and Correspondence to: G. Falcioni, Department of Molecular, Cellular and Animal Biology, University of Camerino, via Camerini n. 2, 62032 Camerino, Italy.
188
functional modifications very ~imilar to what is observed in B-thalassemia erythrocytes from patients [1,2]. In the present study, ~e report data on the correlation between glutathione peroxidase (GSH-Px) activity and the presence in the erythrocyte of free c~or B chains. It is known that glutathione peroxidase, a key enzyme in providing protection against oxidative damage of the red cell membranes, can undergo inactivation in erythrocytes subject to an increased oxidant stress [3,4]. The results reported below indicate that the entrapment of a or/3 chains in normal erythrocytes alters the steady-state flux of oxygen radicals both by increased generation of oxidants (via hemoglobin oxidation) and by a decreased capacity of the repair-antioxidative system. Materials and Methods
All reagents were of analytical grade. Glutathione peroxidase (EC !.11.19), glutathione reductase (EC 1.6.4.2), reduced glutathione and nicotinamide adenine dinucleotide phosphate (reduced form) were purchased from Sigma Chemica!Co., St. Louis, MO; CPDA (Citrate (90 mM), phosphate (15 mM), dextrose (160 mM), adenine (2 mM)) was obtained from Teruno Co., Tokyo. Carrier ampholines were obtained from LKB, Sweden. Hemoglobin A was pt,,rified by DEAE-cellulose chromatography performed in 0.2 M glycine buffer, pH 7.8. The preparation of isolated hemoglobin a and ~ chains was performed by treatment of human hemoglobin at pH 6.0 with an excess of phydr~xymercmibenzoate (PMB) according to tbe method of Bucci and Fronticelli [5]. In a typical preparation, 4 ml of a 5% (w/v) solution of oxyhemoglobin in 0.01 M phosphate buffer, pH 6.0. and 0. I M NaCI were mixed with 6-8-fold excess of PMB dissolved in dilute NaOH. The plt of the solution was then adjusted to 6-6. I with acetic acid and the mixture was left standing in the cold for 24 h. After dialysis against 0.01 M phosphate, pH 6.5, for 12 h, the solution was clarified by centrifugation and then applied to a carboxymethyl cellulose column (20 × 3 cm) equilibrated with the same buffer. The elution was performed using the pH gradient obtained by mixing I I of 0.01 M phosphate (pH 6.5) and 1 i of 0.02 M K2~IPO~. The homogeneity of the chains was tested by ekotrophoresis. The chains obtained by this method have their -SH groups blocked by PMB. The regeneration of the sulphydryl groups was carried out by treatment with mercaptoethanol on a cilromotographic column as previously reported [6]. Ferric chains were obtained by addition of ferricyanide (molar ratio 2:1) to t~e oxygenated derivative. E:~eessof oxidizing agent and ferrocyanide were removed by gel filtration through a Sephadex G-25 column. Since the ferriproteins eluted faster than the yellow excess of ferricyanide, visual observations of clear separations could be made. Carboxy chains were prepared by expost:re of a chain solution to a weak vacuum and then to pure CO. In view of the very high affinity of subunit for CO, subsequent experiments were carried out in air. The purified chains, with regenerated sulphydryl groups, were encapsulated in human erythrocytes by a dialysis technique involving transient hypotonic hemolysis followed by isotonic resealing [71. Human erythrocytes, washed three times in 10 mM Tris buffer containing 154 mM NaCI, pH 7.4, were dialyzed for 90 min at a hematocrit of 70% against hypotonic
189
buffer obtained by a 1:3 dilution with washing buffer. The dialyzed suspension (2 ml) was then mixed with the chains (10 mg dissolved in'0~5 ml of hypotonic buffer) and kept at 4°C for 30 min with gentle stirring. Restoration of isotonicity was obtained by addition of CaCI2 and ATP both a final concentration of l mM and NaCI up to 154 mM, followed by incubation at 37°C for 30 ~in. In order to remove the unloaded chains as well as the hemolysate components present in the supernarant, resealed erythrocytes were washed several times with isotonic buffer before use. The presence of frf~e chains in the cell was demonstrated oy ion focusing using an LKB column (totv,! volume: 110 ml) and carrier ampholines (1% concentration) in the pH range 6-fJ.5; samples obtained by lysis of loaded cells ( - 2 0 mg Hb) were electrofocused at 900 V for 24 h. The GSH-Px activity was assayed with a GSH reduction coupled to a NADPH oxidation by gl'atathione reductase according to Paglia and Valentine's method [8]. The reaction mixture consisted of 50 mM potassium phosphate buffer (pH 7.0) plus 5 mM EDTA, 8.4 mM NADPH, 125 mM NAN3, 150 mM GSH and 2.2 mM H202. The activity of the GSH-Px was monitored at 340 nm (at 20°C) by the disappearance of NADPH, using H202 as a substrate. The samples, when necessary, were mixed with an equal volume of double strength Drabkin's reagent to convert all hemoglobin to the stable cyanomethemoglobin form. • The activity determinations were performed as a function of incubation time (37°C) on glutathione peroxidase solutions (in the presence and absence of different chain derivatives) and on the hemolysates obtained by lysis of chain-loaded erythrocyte suspensions, respectively. Results
The incubation of a glutathione peroxidase solution at 37°C in the presence of free hemoglobin chains may be associated with inactivation of the enzyme. As can be seen in Table l, the activity loss of the enzyme (after 90 min of incubation) is present only under some conditions. In the presence of oxygenated • chains at pH 5.0, after 90 min of incubation at 37°C, the activity measured corresponds to about 14%, while at pH 6.0 it is 52.5% and at pH 7.2 inactivation is undetectable. The experiments were performed at different pH values because the behavior of the autoxidation process is strongly pH dependent. The inactivation of glutathione peroxidase is reduced in the presence of an equivalent amount of oxygenated ~ chains: after 90 min of incubation at 37°C at pH 5.0, the activity measured corresponds to 84%, while it is 100% at higher pH values. In addition, the data reported in Table I show that the activity of glutathione peroxidase is not changed after 90 min incubation at pH 5.0 if the enzyme is alone', in the presence of ¢CO under the same conditions a small decrease in activity is observed, possibly related to lack of deoxygenation during the experiment. We have also carried out experiments in the presence of ferric chains (a÷). Inhibition of glutathione peroxidase activity ~s absent at the beginning of the incubation, while it is observed following incubation at 37°C; after 30 min the activity of the enzyme is almost halved. We, therefore, undertook a series of experiments to see whether entrapment of free
190 TABLE !
Influence of different human hemoglobin chain derivatives on glutathione peroxi~ase (GSH-Px) activity after 90 min of incubation at 37°C a Incubation mixture
pH
% Activity
GSH.Px GSH.Px GSH-Px GSH-Px GSH-Px GSH-Px GSH-Px GSH-Px
5.0 5.0 5.0 6.0 7.2 5.0 6.0 7.2
100
+ cd32 + aCO + eeO2 + ed32 + ~O 2 +/~O 2 + BO2
14 80 52.5
100 84
I00 I00
SConditions: [GPx], 0.1 mg/ml; [chain], 0.5 mg/ml, temperature, 37°C; 0.1 M phosphate or acetate buffer.
chains in normal erythrocytes was capable of modifying the intraerythrocytic activity of the glutathione peroxidase. By a dialysis technique (see Materials and Methods), we entrapped free c~ or/~ chains in normal erythrocytes. The effective presence in the cell of free subunit was monitored by ion focusing (the free chain concentration was about 1% of the total hemoglobin present in the hemolysate obtained by lysis of the loaded erythrocyte). The incubation in air at 37°C of chain-loaded erythrocytes suspended in motonic medium (CPDA) is associated (Table 11) with a more marked loss of glutathione peroxidase activity with respect to controls (unloaded but opened and resealed cells). The presence of a chair,s in the ~rythrocyte is more effective in the inactivation of TABLE !i
lntraerythrocytic activity of the glutathione peroxidase of chain-loaded erythrocyte suspensions Suspensions
Incubation time (h)
a-Loaded erythrocytes
0,0 4.5 24.0
~-Loaded erythrocytes
86.5 57,6 52,6
0,0
89.2
4.5
76.2 56,f~
24,0
Unloaded erythrocytes
% Activity
0.0 4,5 24.0
100.0 96.4 66.0
After entrapment of~hains, erythrocytes were washed several times and incubated at 37°C in cPDA. This suspension (Hct -- 30%) contained 4,0 x 1012 RB cells/I.
191
the enzyme with respect to the/3 chains. Control suspension incubated in air at 37°C exhibited a measurable but clearly slower rate of glutathione peroxidase inactivation similar to what occurs following exposure of chain-loaded erythrocyte suspension to pure CO (data not shown). The initial inactivation of the enzyme (Table II) is due to the resealing procedure that foresees 30 min incubation at 37°C (i.e. time 0 is really 30 mm)
Discussion The data described above indicate that the presence of free hemoglobin chains can alter the glutathione peroxidase activity. The results, reported in Table l, obtained at different pH values with both the • and /3 chains indicate that a significant decrease of the enzyme activity is observed, particularly in those experiments where the autoxidation of the oxygenated chains of human hemoglobin is followed by transformation of the oxidized molecule (high-spin Fe 3÷) into a hemichrome, i.e. a low.spin Fe 3+ compound. (For an analysis on the formation of hemichromes in the autoxidation of the isolated c~ and B chains of human hemoglobin, see Brunori et al. [9].) These results are in agreement with foregoing evidence [10] that allowed us to propose that the formation of hemichromes, rather than the mere autoxidation of hemoglobin (or the increased flux of oxygen radicals produced during autoxidation), was responsible for the inactivation of glutathione peroxidase. • The presence of free chains (-- 1% of total hemoglobin measured in the hemolysate of the loaded cells) leads to an in vitro model of thalassemia. This thalassemic model permits the investigation of the direct effect that the entrapped chains have on normal human erythrocytes. The rate of the observed changes suggests the hypothesis that the GSH-Px damage occurs in erythropoiesis and parallel chain synthesis (this hypothesis, due to the different chain stability, is obviously better supported for with respect to 8 chains). Although the molecular mechanism whereby hemichrome formation leads to GSH-Px inactivation is still unknowr~ the reported data may be of some use in improving red cell survival in thalassemic patients. Potential therapeutic agents must be capable of reducing the o,:idative stress, taking into account the correlation between selenium concentration and selenium-dependent GSH-Px activity in blood.
Acknowledgements The authors express their thanks to Professor M. Brunori (Rome) for reading the manuscript and for critical comments. The technical assistance of Mr. S. Polzoni and S. Morosi is gratefully acknowledged. This work was supported by a grant from CNR (Progetto Finalizzato Biotecnologie e Biostrumentazioni).
References Scott MD, Rouyer-Fessard P, Lubin BH, Beuzard Y. Entrapment of purified a-hemoglobin chains in normal erythrocytes: a model for B.thalassemia. J Biol Chem 1990;265:17953-17959. Scott MD, Rouyer-Fessard P, Lubin BH, Beuzard Y. a and B hemoglobin chain induced changes in normal red cell deformability: comparison to B.thalassemia intermedia and Hb H disease. Blood 1990c;76(suppl. I):289(75a).
192 3 4 5 6 7 8 9
10
Falcioni G, Ci~,ola G, Brunori M. Giutathione peroxidase and oxidative hemolysis in trout red b~ood cells. Fed Eur Bi~hem Soc Lett 1987;221:355-358. Mavelli !, Ciriolo MR, Rotilio G. Inactivation of red cell glutathione peroxidase by divicine and its relation to the hemolysis in favism. Biochim Biophys Acta 1985:847:280-284. Bucci E, Fronticelli C. A new method for the preparation of a and ~ subunits of human haemoglobin. J Biol Chem 1965;240:551-552. Geraci G, Parkhurst l.J, Gibson QH. Preparation and properties of a- and 0-chains from human hemoglobin. J Bioi Chem 1969~244:4664-4667. Falciont G, Gabbianelli R, Concerti A, Grello~3iV, Soila L, Brunori M. Aprotinm rele;~seby loaded mouse erythrocytes. Adv Biosci 1991;81:87-91. P~Jia ED, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocytes glutathione peroxidase. J Lab Clin Med 1967;70:!58-169. Brunori M, Falcioni G, Fioretti E, Giardina B, Rotilio G. Formation of superoxide in the autoxidation of isolated a and O chains of human hemoglobin and its involvement in hemichrome precipitation. Eur J Biochem 1975;53:99-104. Grelloni F, Gabbianelli R, Falcioni G. Inactivation of glutathio~te pcr¢~xiddsetoilowing hemoglobin oxidation. Biochem lnt 1991;25(5):789-795.
187
CCA05547
"f" Inactivation of gl utathlone peroxidase following entrapment of purified, or/3 hemoglobin chains in human erythrocytes
Francesco Grelloni, Rosita Gabbianelli, Anna M a r i a Santroni and Giancarlo Falcioni Department of Molecular, Cellular and Animal Biology, University of Camerino, Camerino (Italy) (Received 20 May 1992; revision received 8 February 1993; accepted 13 March 1993) Key words: Glutathione peroxidase; Loaded erythrocytes; Human hemoglobin chains: Thalassemia
Summary Inactivation of glutathione peroxidase correlates with the rate of hemoglobin chain oxidation. The enzyme inactivation is mainly present in those conditions where the autoxidation of the oxygenated chains is followed by trasformation of the oxidized molecule into a hemichrome. Free hemoglobin chains have been encaps,dated i~' human red blood cells by a dialysis technique that involves transient hypotonic hemolysis followed by isotonic resealing Chain-loaded erythrocytes represent a good in vitro model of thalassemia. The presence of free human chains in the cell alters the intraerythrocytic glutathione peroxidase activity (oLchains are more effective in the inactivation of the enzyme with respect to the 0 chains).
Introduction Entrapment of isolated hemoglobin chains (c~ or 0) in human erythrocytes leads to the formation of an in vitro model of thalassemia. These resealed red blood cells represent a model system for studying the mechanisms involved in thalassemic pathophysiology due to the presence of free chains, without having to use patient samples where damage has already occurred in vivo. It has been reported that entrapment of purified a chains in normal human erythrocytes leads to structural and Correspondence to: G. Falcioni, Department of Molecular, Cellular and Animal Biology, University of Camerino, via Camerini n. 2, 62032 Camerino, Italy.
188
functional modifications very ~imilar to what is observed in B-thalassemia erythrocytes from patients [1,2]. In the present study, ~e report data on the correlation between glutathione peroxidase (GSH-Px) activity and the presence in the erythrocyte of free c~or B chains. It is known that glutathione peroxidase, a key enzyme in providing protection against oxidative damage of the red cell membranes, can undergo inactivation in erythrocytes subject to an increased oxidant stress [3,4]. The results reported below indicate that the entrapment of a or/3 chains in normal erythrocytes alters the steady-state flux of oxygen radicals both by increased generation of oxidants (via hemoglobin oxidation) and by a decreased capacity of the repair-antioxidative system. Materials and Methods
All reagents were of analytical grade. Glutathione peroxidase (EC !.11.19), glutathione reductase (EC 1.6.4.2), reduced glutathione and nicotinamide adenine dinucleotide phosphate (reduced form) were purchased from Sigma Chemica!Co., St. Louis, MO; CPDA (Citrate (90 mM), phosphate (15 mM), dextrose (160 mM), adenine (2 mM)) was obtained from Teruno Co., Tokyo. Carrier ampholines were obtained from LKB, Sweden. Hemoglobin A was pt,,rified by DEAE-cellulose chromatography performed in 0.2 M glycine buffer, pH 7.8. The preparation of isolated hemoglobin a and ~ chains was performed by treatment of human hemoglobin at pH 6.0 with an excess of phydr~xymercmibenzoate (PMB) according to tbe method of Bucci and Fronticelli [5]. In a typical preparation, 4 ml of a 5% (w/v) solution of oxyhemoglobin in 0.01 M phosphate buffer, pH 6.0. and 0. I M NaCI were mixed with 6-8-fold excess of PMB dissolved in dilute NaOH. The plt of the solution was then adjusted to 6-6. I with acetic acid and the mixture was left standing in the cold for 24 h. After dialysis against 0.01 M phosphate, pH 6.5, for 12 h, the solution was clarified by centrifugation and then applied to a carboxymethyl cellulose column (20 × 3 cm) equilibrated with the same buffer. The elution was performed using the pH gradient obtained by mixing I I of 0.01 M phosphate (pH 6.5) and 1 i of 0.02 M K2~IPO~. The homogeneity of the chains was tested by ekotrophoresis. The chains obtained by this method have their -SH groups blocked by PMB. The regeneration of the sulphydryl groups was carried out by treatment with mercaptoethanol on a cilromotographic column as previously reported [6]. Ferric chains were obtained by addition of ferricyanide (molar ratio 2:1) to t~e oxygenated derivative. E:~eessof oxidizing agent and ferrocyanide were removed by gel filtration through a Sephadex G-25 column. Since the ferriproteins eluted faster than the yellow excess of ferricyanide, visual observations of clear separations could be made. Carboxy chains were prepared by expost:re of a chain solution to a weak vacuum and then to pure CO. In view of the very high affinity of subunit for CO, subsequent experiments were carried out in air. The purified chains, with regenerated sulphydryl groups, were encapsulated in human erythrocytes by a dialysis technique involving transient hypotonic hemolysis followed by isotonic resealing [71. Human erythrocytes, washed three times in 10 mM Tris buffer containing 154 mM NaCI, pH 7.4, were dialyzed for 90 min at a hematocrit of 70% against hypotonic
189
buffer obtained by a 1:3 dilution with washing buffer. The dialyzed suspension (2 ml) was then mixed with the chains (10 mg dissolved in'0~5 ml of hypotonic buffer) and kept at 4°C for 30 min with gentle stirring. Restoration of isotonicity was obtained by addition of CaCI2 and ATP both a final concentration of l mM and NaCI up to 154 mM, followed by incubation at 37°C for 30 ~in. In order to remove the unloaded chains as well as the hemolysate components present in the supernarant, resealed erythrocytes were washed several times with isotonic buffer before use. The presence of frf~e chains in the cell was demonstrated oy ion focusing using an LKB column (totv,! volume: 110 ml) and carrier ampholines (1% concentration) in the pH range 6-fJ.5; samples obtained by lysis of loaded cells ( - 2 0 mg Hb) were electrofocused at 900 V for 24 h. The GSH-Px activity was assayed with a GSH reduction coupled to a NADPH oxidation by gl'atathione reductase according to Paglia and Valentine's method [8]. The reaction mixture consisted of 50 mM potassium phosphate buffer (pH 7.0) plus 5 mM EDTA, 8.4 mM NADPH, 125 mM NAN3, 150 mM GSH and 2.2 mM H202. The activity of the GSH-Px was monitored at 340 nm (at 20°C) by the disappearance of NADPH, using H202 as a substrate. The samples, when necessary, were mixed with an equal volume of double strength Drabkin's reagent to convert all hemoglobin to the stable cyanomethemoglobin form. • The activity determinations were performed as a function of incubation time (37°C) on glutathione peroxidase solutions (in the presence and absence of different chain derivatives) and on the hemolysates obtained by lysis of chain-loaded erythrocyte suspensions, respectively. Results
The incubation of a glutathione peroxidase solution at 37°C in the presence of free hemoglobin chains may be associated with inactivation of the enzyme. As can be seen in Table l, the activity loss of the enzyme (after 90 min of incubation) is present only under some conditions. In the presence of oxygenated • chains at pH 5.0, after 90 min of incubation at 37°C, the activity measured corresponds to about 14%, while at pH 6.0 it is 52.5% and at pH 7.2 inactivation is undetectable. The experiments were performed at different pH values because the behavior of the autoxidation process is strongly pH dependent. The inactivation of glutathione peroxidase is reduced in the presence of an equivalent amount of oxygenated ~ chains: after 90 min of incubation at 37°C at pH 5.0, the activity measured corresponds to 84%, while it is 100% at higher pH values. In addition, the data reported in Table I show that the activity of glutathione peroxidase is not changed after 90 min incubation at pH 5.0 if the enzyme is alone', in the presence of ¢CO under the same conditions a small decrease in activity is observed, possibly related to lack of deoxygenation during the experiment. We have also carried out experiments in the presence of ferric chains (a÷). Inhibition of glutathione peroxidase activity ~s absent at the beginning of the incubation, while it is observed following incubation at 37°C; after 30 min the activity of the enzyme is almost halved. We, therefore, undertook a series of experiments to see whether entrapment of free
190 TABLE !
Influence of different human hemoglobin chain derivatives on glutathione peroxi~ase (GSH-Px) activity after 90 min of incubation at 37°C a Incubation mixture
pH
% Activity
GSH.Px GSH.Px GSH-Px GSH-Px GSH-Px GSH-Px GSH-Px GSH-Px
5.0 5.0 5.0 6.0 7.2 5.0 6.0 7.2
100
+ cd32 + aCO + eeO2 + ed32 + ~O 2 +/~O 2 + BO2
14 80 52.5
100 84
I00 I00
SConditions: [GPx], 0.1 mg/ml; [chain], 0.5 mg/ml, temperature, 37°C; 0.1 M phosphate or acetate buffer.
chains in normal erythrocytes was capable of modifying the intraerythrocytic activity of the glutathione peroxidase. By a dialysis technique (see Materials and Methods), we entrapped free c~ or/~ chains in normal erythrocytes. The effective presence in the cell of free subunit was monitored by ion focusing (the free chain concentration was about 1% of the total hemoglobin present in the hemolysate obtained by lysis of the loaded erythrocyte). The incubation in air at 37°C of chain-loaded erythrocytes suspended in motonic medium (CPDA) is associated (Table 11) with a more marked loss of glutathione peroxidase activity with respect to controls (unloaded but opened and resealed cells). The presence of a chair,s in the ~rythrocyte is more effective in the inactivation of TABLE !i
lntraerythrocytic activity of the glutathione peroxidase of chain-loaded erythrocyte suspensions Suspensions
Incubation time (h)
a-Loaded erythrocytes
0,0 4.5 24.0
~-Loaded erythrocytes
86.5 57,6 52,6
0,0
89.2
4.5
76.2 56,f~
24,0
Unloaded erythrocytes
% Activity
0.0 4,5 24.0
100.0 96.4 66.0
After entrapment of~hains, erythrocytes were washed several times and incubated at 37°C in cPDA. This suspension (Hct -- 30%) contained 4,0 x 1012 RB cells/I.
191
the enzyme with respect to the/3 chains. Control suspension incubated in air at 37°C exhibited a measurable but clearly slower rate of glutathione peroxidase inactivation similar to what occurs following exposure of chain-loaded erythrocyte suspension to pure CO (data not shown). The initial inactivation of the enzyme (Table II) is due to the resealing procedure that foresees 30 min incubation at 37°C (i.e. time 0 is really 30 mm)
Discussion The data described above indicate that the presence of free hemoglobin chains can alter the glutathione peroxidase activity. The results, reported in Table l, obtained at different pH values with both the • and /3 chains indicate that a significant decrease of the enzyme activity is observed, particularly in those experiments where the autoxidation of the oxygenated chains of human hemoglobin is followed by transformation of the oxidized molecule (high-spin Fe 3÷) into a hemichrome, i.e. a low.spin Fe 3+ compound. (For an analysis on the formation of hemichromes in the autoxidation of the isolated c~ and B chains of human hemoglobin, see Brunori et al. [9].) These results are in agreement with foregoing evidence [10] that allowed us to propose that the formation of hemichromes, rather than the mere autoxidation of hemoglobin (or the increased flux of oxygen radicals produced during autoxidation), was responsible for the inactivation of glutathione peroxidase. • The presence of free chains (-- 1% of total hemoglobin measured in the hemolysate of the loaded cells) leads to an in vitro model of thalassemia. This thalassemic model permits the investigation of the direct effect that the entrapped chains have on normal human erythrocytes. The rate of the observed changes suggests the hypothesis that the GSH-Px damage occurs in erythropoiesis and parallel chain synthesis (this hypothesis, due to the different chain stability, is obviously better supported for with respect to 8 chains). Although the molecular mechanism whereby hemichrome formation leads to GSH-Px inactivation is still unknowr~ the reported data may be of some use in improving red cell survival in thalassemic patients. Potential therapeutic agents must be capable of reducing the o,:idative stress, taking into account the correlation between selenium concentration and selenium-dependent GSH-Px activity in blood.
Acknowledgements The authors express their thanks to Professor M. Brunori (Rome) for reading the manuscript and for critical comments. The technical assistance of Mr. S. Polzoni and S. Morosi is gratefully acknowledged. This work was supported by a grant from CNR (Progetto Finalizzato Biotecnologie e Biostrumentazioni).
References Scott MD, Rouyer-Fessard P, Lubin BH, Beuzard Y. Entrapment of purified a-hemoglobin chains in normal erythrocytes: a model for B.thalassemia. J Biol Chem 1990;265:17953-17959. Scott MD, Rouyer-Fessard P, Lubin BH, Beuzard Y. a and B hemoglobin chain induced changes in normal red cell deformability: comparison to B.thalassemia intermedia and Hb H disease. Blood 1990c;76(suppl. I):289(75a).
192 3 4 5 6 7 8 9
10
Falcioni G, Ci~,ola G, Brunori M. Giutathione peroxidase and oxidative hemolysis in trout red b~ood cells. Fed Eur Bi~hem Soc Lett 1987;221:355-358. Mavelli !, Ciriolo MR, Rotilio G. Inactivation of red cell glutathione peroxidase by divicine and its relation to the hemolysis in favism. Biochim Biophys Acta 1985:847:280-284. Bucci E, Fronticelli C. A new method for the preparation of a and ~ subunits of human haemoglobin. J Biol Chem 1965;240:551-552. Geraci G, Parkhurst l.J, Gibson QH. Preparation and properties of a- and 0-chains from human hemoglobin. J Bioi Chem 1969~244:4664-4667. Falciont G, Gabbianelli R, Concerti A, Grello~3iV, Soila L, Brunori M. Aprotinm rele;~seby loaded mouse erythrocytes. Adv Biosci 1991;81:87-91. P~Jia ED, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocytes glutathione peroxidase. J Lab Clin Med 1967;70:!58-169. Brunori M, Falcioni G, Fioretti E, Giardina B, Rotilio G. Formation of superoxide in the autoxidation of isolated a and O chains of human hemoglobin and its involvement in hemichrome precipitation. Eur J Biochem 1975;53:99-104. Grelloni F, Gabbianelli R, Falcioni G. Inactivation of glutathio~te pcr¢~xiddsetoilowing hemoglobin oxidation. Biochem lnt 1991;25(5):789-795.