The FASEB Journal • Research Communication
Increased basal oxidation of peroxiredoxin 2 and limited peroxiredoxin recycling in glucose-6-phosphate dehydrogenase-deficient erythrocytes from newborn infants Fook-Choe Cheah,* Alexander V. Peskin,‡ Fei-Liang Wong,* Azlin Ithnin,† Ainoon Othman,储 and Christine C. Winterbourn‡,§,1 *Department of Paediatrics and †Department of Pathology, Universiti Kebangsaan Malaysia Medical Centre, Kuala Lumpur, Malaysia; ‡Centre for Free Radical Research, Department of Pathology, and §Gravida National Centre for Growth and Development, University of Otago Christchurch, Christchurch, New Zealand; and 储Department of Pathology, Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia Erythrocytes require glucose-6-phosphate dehydrogenase (G6PD) to generate NADPH and protect themselves against hemolytic anemia induced by oxidative stress. Peroxiredoxin 2 (Prx2) is a major antioxidant enzyme that requires NADPH to recycle its oxidized (disulfide-bonded) form. Our aims were to determine whether Prx2 is more highly oxidized in G6PD-deficient erythrocytes and whether these cells are able to recycle oxidized Prx2 after oxidant challenge. Blood was obtained from 61 Malaysian neonates with G6PD deficiency (average 33% normal activity) and 86 controls. Prx2 redox state was analyzed by Western blotting under nonreducing conditions. Prx2 in freshly isolated blood was predominantly reduced in both groups, but the median level of oxidation was significantly higher (8 vs 3%) and the range greater for the G6PD-deficient population. When treated with reagent H2O2, the G6PD-deficient erythrocytes were severely compromised in their ability to recycle oxidized Prx2, with only 27 or 4% reduction after 1 h treatment with 0.1 or 1 mM H2O2 respectively, compared with >97% reduction in control erythrocytes. The accumulation of oxidized Prx2 in oxidatively stressed erythrocytes with common G6PD variants suggests that impaired antioxidant activity of Prx2 could contribute to the hemolysis and other complications associated with the condition.— Cheah, F.-C., Peskin, A. V., Wong, F.-L., Ithnin, A., Othman, A., Winterbourn, C. C. Increased basal oxidation of peroxiredoxin 2 and limited peroxiredoxin recycling in glucose-6-phosphate dehydrogenase deficient erythrocytes from newborn infants. FASEB J. 28, 3205–3210 (2014). www.fasebj.org ABSTRACT
Key Words: antioxidant defense 䡠 hydrogen peroxide 䡠 hemolytic anemia
Abbreviations: EA, enzyme activity; FST, fluorescent spot test; G6PD, glucose-6-phosphate dehydrogenase; IQR, interquartile range; NEM, N-ethylmaleimide; Prx2, peroxiredoxin 2 0892-6638/14/0028-3205 © FASEB
Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most commonly inherited genetically linked enzymopathy. An estimated 4 million people worldwide carry a mutation (1). Although normally asymptomatic, G6PD-deficient individuals can present with acute hemolysis associated with the ingestion of fava beans (favism), sulfur-containing and antimalarial drugs, or infection (2). Newborn infants can present with early onset and severe jaundice because of hyperbilirubinemia (1), which, if untreated, can result in bilirubin encephalopathy or kernicterus. G6PD deficiency is prevalent in many parts of Asia, including Malaysia (5–10%; ref. 3), where there is a nationwide routine cord blood screening program. The infants range from being asymptomatic to presenting with severe jaundice, and they are monitored closely with delayed discharge from the hospital. G6PD is part of the pentose phosphate pathway, which in erythrocytes is the sole source of NADPH (4). NADPH provides the reducing equivalents needed for the removal of reactive oxygen species, and clinical problems occur in G6PD deficiency when cells are exposed to increased oxidative stress. The problem is largely restricted to erythrocytes because of their dependence on the pentose phosphate pathway (1) and because hemoglobin is a major source of reactive oxidants (5). Furthermore, most G6PD mutations result in an unstable, short-lived protein (6), which cannot be replenished in mature erythrocytes. Several antioxidant pathways use NADPH as the terminal reductant. Although NADPH increases the efficiency of H2O2 removal by catalase (7), the common view is that it is mainly required for removal of peroxides via the glutathione peroxidase/glutathione 1 Correspondence: Centre for Free Radical Research, Department of Pathology, University of Otago Christchurch, PO Box 4345, Christchurch, New Zealand. E-mail: christine. [email protected] doi: 10.1096/fj.14-250050
3205
Figure 1. Scheme showing oxidation of Prx2 by H2O2. Prx2 exists as a homodimer. Oxidation occurs at a highly reactive active-site Cys residue, which condenses with the resolving Cys on the adjacent chain to form a disulfide-linked dimer. This is recycled by thioredoxin (Trx) through thioredoxin reductase (TrxR), with electrons coming from NADPH. NADPH is supplied by G6PD and the pentose phosphate pathway (PPP).
reductase pathway (8). However, another important class of NADPH-dependent antioxidant enzymes is the peroxiredoxins (9). Peroxiredoxins (there are 6 in humans) are thiol proteins that react extremely rapidly with H2O2, peroxynitrite, and other peroxides (10 –12). Peroxiredoxin 2 (Prx2) is the major peroxiredoxin in erythrocytes, where it constitutes the third most abundant protein (9, 13). Reduced Prx2 exists as a noncovalent homodimer with active-site cysteine (Cys) residues. Oxidation results in a disulfide-linked dimer that is recycled by thioredoxin/thioredoxin reductase and NADPH (Fig. 1). Therefore, the antioxidant activity of erythrocyte Prx2 requires G6PD. However, there have been no studies examining whether the peroxiredoxin pathway is compromised in G6PD deficiency or whether this could contribute to the hemolytic mechanism. Prx2 plays a major role in peroxide removal in the erythrocyte (14 –17). It becomes oxidized in erythrocytes exposed to very low concentrations of H2O2, even though they contain active catalase and glutathione peroxidase (14). It also scavenges endogenous peroxide generated through hemoglobin autoxidation (14), and erythrocytes from Prx2-knockout mice contained higher amounts of hemoglobin oxidation products (18, 19). Increased levels of erythrocyte Prx2 oxidation have been observed in a mouse model of sepsis (20). It is likely therefore that Prx2 function could be impaired in G6PD deficiency and contribute to the phenotype. To test this proposal, we compared erythrocytes from newborn infants with and without G6PD deficiency to determine whether Prx2 is more highly oxidized under basal conditions in the infants with G6PD deficiency and whether G6PD-deficient erythrocytes are compromised in their ability to regenerate reduced Prx2 after exposure to H2O2.
included. For controls, study bloods were obtained during routine sampling from non-G6PD-deficient infants who were admitted to the neonatal intensive care unit because of jaundice requiring phototherapy or considered at risk for low glucose or infection, or from healthy infants awaiting discharge from the postnatal ward. Blood for basal Prx2 analysis was collected by venous sampling at the same time as blood for other routine investigations which included enzyme activity (EA) for G6PD and bilirubin levels. For Prx2 analysis, 350 l blood was collected into an EDTA-containing microcentrifuge tube with preadded N-ethylmaleimide (NEM; 150 l, 200 mM). The median age at sampling was 3 d [interquartile range (IQR), 2–5 d]. Prx2 reacts with H2O2 very rapidly (21) and can easily be oxidized by traces of H2O2 during cell lysis (14). To prevent artifactual Prx2 oxidation, blood was added directly to a large excess of NEM to trap its existing oxidation state before lysis of the cells. Thus, no free thiols would still be present during lysis, and oxidation by H2O2 would be prevented. To check that sufficient NEM was added to study samples, two blood samples were treated with different amounts of 200 mM NEM (Fig. 2) then subjected to immunoblot analysis (as described below). At NEM:blood ratios of 1:1 and 1:4, there were similar low levels of oxidized (dimeric) Prx2, but at a 1:10 ratio, trapping with NEM became less efficient and the percentage oxidation began to increase. Thus the procedure used should prevent oxidation during lysis. To study in vitro responses to H2O2, fresh blood samples (without NEM) were also collected from a subpopulation of the control and G6PD-deficient infants.
MATERIALS AND METHODS Study population and samples This study was conducted over a period of 18 mo from July 2011 until December 2012. Ethics approval was obtained from the Universiti Kebangsaan Malaysia (UKM) Research Ethics Committee (UKM-FF-FRGS0038-2010). All infants delivered at the UKM Medical Centre were routinely screened for G6PD deficiency by the fluorescent spot test (FST) on cord blood. Infants born by spontaneous vaginal delivery or elective cesarean sections to uncomplicated pregnancies were enrolled after parental consent was obtained. Randomly selected infants who were newly diagnosed with G6PD deficiency and monitored for jaundice in the hospital were 3206
Vol. 28
July 2014
Figure 2. Effect of NEM concentration on the redox state of Prx2 in erythrocytes after lysis. Blood samples from 2 control subjects (I and II) were mixed with 200 mM NEM at various volume ratios before lysis and subjected to nonreducing SDS-PAGE. The top band of the Western blot represents oxidized Prx2 (disulfide-linked dimer) which is ⬃44 kDa; the bottom band is reduced monomer (22 kDa). Dimerization of Prx2 was inhibited with an excess of NEM (200 mM) to blood of ⱖ1:4 in ratio.
The FASEB Journal 䡠 www.fasebj.org
CHEAH ET AL.
Bilirubin and phototherapy Newborn infants with G6PD deficiency diagnosed by FST screening are routinely monitored for jaundice in the hospital for a minimum of 5 d at the UKM Medical Centre. G6PD analysis by EA is performed as further confirmation and to evaluate the category of severity in enzyme deficiency. If a baby developed jaundice, peripheral blood was taken for the measurement of serum bilirubin, and phototherapy was initiated using the guide proposed by the Ministry of Health, which was based on the reference curve developed by the American Academy of Pediatrics (22). In general, a bilirubin level that breached 255 M would be considered as significant hyperbilirubinemia requiring phototherapy intervention regardless of the age of the infant. Detection of G6PD deficiency Screening for G6PD deficiency was performed on all cord bloods using the modified FST as described by Beutler et al. (2). In this assay, a blood spot is applied immediately to filter paper and dried at room temperature. The disc of dried blood is cut out and reacted with 100 l of a mixture containing glucose-6-phosphate, GSSG, and NADP⫹. After 15 min at 37°C, 10 l of this mixture is applied to filter paper, dried, and examined under ultraviolet light. Quantitative measurement of G6PD enzyme activity was performed on 5 l of whole EDTA blood using the OSMMR-D G6PD assay kit (R & D Diagnostics, Holargos, Greece) with hemoglobin normalization, according to the supplier’s instructions. The reference values for normal, partial deficiency, and deficiency of G6PD activity were ⬎6.76, 2.25– 6.76, and ⬍2.25 U/g hemoglobin, respectively, adapted based on the ranges published by Azma et al. (23). Genotype analysis Blood spots on filter paper were processed for molecular analysis to screen for the 10 most common G6PD mutations prevalent in Southeast Asia: c.871G⬎A, c.487G⬎A, c.1388G⬎A, c.1376G⬎T, c.95A⬎G, c.1024C⬎T, c.563C⬎T, c.392G⬎T, c.1003G⬎A, and c.592C⬎T. DNA was extracted using the TaqMan Sample-to-SNP kit, and genotyping was done using the TaqMan MGB single-nucleotide polymorphism (SNP) assay (Applied Biosystems, Foster City, CA, USA). Immunoblot analysis for Prx2 Whole blood was analyzed without separation of the erythrocytes. After collection, samples were left at room temperature for 1 h, then 10 l of blood was diluted to 1 ml with PBS containing 40 mM NEM. After 30 min, 100 l was diluted with equal volumes of 200 mM NEM and 6% SDS. Samples were stored at ⫺80°C before being transported frozen to Christchurch, New Zealand, for analysis. Proteins were resolved by nonreducing 12% SDS-PAGE and transferred electrophoretically to a Hybond polyvinylidene difluoride (PVDF) membrane (14). Blocking was performed for 1 h at room temperature in 5% (w/v) nonfat dried milk in Tris-buffered saline with 0.05% Tween 20 (TBST20). Incubation with antibodies to Prx2 (Sigma Chemical Co., St Louis, MO, USA) was performed overnight at 4°C or for 1 h at room temperature in TBST20 with 5% milk. Then membranes were washed and probed with horseradish peroxidase-conjugated goat anti-rabbit antibodies (Dako, Carpinteria, CA, USA). Bands were visualized through enhanced chemiluminescence using the ECLPlus Western blotting Detection System (GE Healthcare Biosciences Corp., Piscataway, NJ, USA). ImmuPrx2 IN G6PD-DEFICIENT ERYTHROCYTES
noblots were scanned using a ChemiDoc XRS gel documentation system, and densitometry was performed using Quantity One software (Bio-Rad, Hercules, CA, USA). H2O2 oxidation of Prx2 Freshly drawn blood from randomly selected infants was used to study the efficiency of reduction of Prx2 in erythrocytes treated with H2O2. Blood samples were diluted 10-fold in Hanks buffer with 5 mM glucose, and 45 l aliquots were reacted with 5 l of either 1 mM or 10 mM H2O2 for 5 min or 1 h at room temperature. These high concentrations were used to overcome erythrocyte catalase and ensure that dimerization of the Prx2 was (near) complete so that reduction could be followed. The reaction was stopped by blocking thiols with 20 l of 200 mM NEM, then 20 l of 10% SDS was added and the samples were frozen for later immunoblotting. Statistical analysis Data were analyzed using SigmaPlot 12 (Systat Software Inc., San Jose, CA, USA). Results are expressed as means ⫾ sd if normally distributed, otherwise as medians (IQR). Comparison of 2 groups of normally distributed data was based on the t test, while nonparametric data was analyzed using the Mann-Whitney U test. Categorical data such as gender, birth mode, and phototherapy were analyzed using 2 and the relationship between 2 parameters using the Pearson correlation and linear regression. The level of statistical significance was taken when P ⬍ 0.05.
RESULTS Group clinical characteristics The characteristics of the study population are shown in Table 1. On average, erythrocytes from infants with G6PD deficiency had 33% of the enzyme activity of the control infants (Table 1). Most had activities in the 20 – 60% of normal range, with only 4 showing severe deficiency. Half the infants had the two most common mutations in the Asian community (33% c.871G⬎A and 17% c.487G⬎A), which are classified as class III (deficiency associated with 10 – 60% of residual EA) in the WHO functional definition of G6PD deficiency (24). The third most common mutation was c.1388G⬎A (11%), which is categorized as class II (deficiency ⬍10% of residual EA). There was no difference between infants with G6PD deficiency and infants with normal G6PD activity in gestational age, birth weight, or mode of delivery (Table 1). The infants were of term gestation (mean of 38 wk) and two-thirds were born by spontaneous vaginal delivery. As expected for an X-linked condition, the majority of the infants with G6PD deficiency were male. Peak bilirubin levels, severe hyperbilirubinemia (bilirubin ⬎255 M), and the proportion of infants receiving phototherapy were similar between groups. Because infants with G6PD deficiency are treated aggressively because of their “high-risk” category, by commencing phototherapy at a low serum bilirubin threshold on the reference curve, none of the infants became more severely jaundiced. 3207
TABLE 1. Characteristics of the study population Parameter
n G6PD enzyme level (U/g Hb)a Deficiency, n ⫽ 50 (U/g Hb) Range (U/g Hb) Severe deficiency, n ⫽ 4 (U/g Hb) Range (U/g Hb) Gender [n (%)] Male Female Mode of birth Spontaneous vaginal (n) Cesarean section [n (%)] Gestation (wk) Birth weight (kg) Bilirubin level (M) Peak, wk 1 of life IQR Received phototherapy [n (%)]
Normal
G6PD deficient
86 10.9 ⫾ 1.8 – – – –
61 3.6 ⫾ 1.3** 3.8 ⫾ 1.1 2.3–6.5 1.0 ⫾ 0.7 0.4–2.0
38 (44) 48 (56)
57 (93)** 4 (7)
59 27 (31) 38 ⫾ 2 2.96 ⫾ 0.51
47 14 (23) 38 ⫾ 2 3.06 ⫾ 0.45
228 ⫾ 80 160–281 30 (35)
232 ⫾ 111 189–273 23 (38)
Unless stated otherwise, values are means ⫾ sd. Comparisons were not statistically significant unless indicated. aEnzyme activity available for n ⫽ 77 normal and n ⫽ 54 G6PD-deficient infants. **P ⬍ 0.001 vs. normal group.
Basal oxidation state of Prx2 The oxidation state of Prx2 in the red cells of the infants was assessed by nonreducing SDS-PAGE and Western blotting. Under these conditions, reduced Prx2 runs as a 22-kDa monomer, and the reversibly oxidized protein, which forms a disulfide-linked dimer, runs slower at twice this molecular mass (25). As shown for selected samples in Fig. 3A, Prx2 was in most cases predominantly in the reduced form, with varying amounts of the oxidized dimer present. However, the percentage of oxidized Prx2 was significantly higher in the G6PD-deficient erythrocytes (median 8%) than in the cells with normal G6PD activity (median 3%, P⫽0.027), and the range was greater (Fig. 3B). The extent of oxidation for all infants was also negatively correlated with G6PD enzyme activity (correlation coefficient ⫺0.20, P⫽0.024, n⫽131). No association between extent of oxidation and mode of delivery was evident.
controls (99 and 97% respectively; Fig. 4B). G6PD enzyme activity was measured for all the infants included in this part of the study and showed a positive
Impaired ability of G6PD-deficient erythrocytes to reduce Prx2 When normal erythrocytes are treated with H2O2, the Prx2 is initially oxidized to the disulfide and then recycled to the reduced monomer once the peroxide has been consumed (14). To test whether the recycling mechanism is impaired in G6PD deficiency, H2O2 was added to fresh blood samples, and the oxidation state of Prx2 was monitored at 5 and 60 min. As shown for the examples in Fig. 4A, all or almost all of the Prx2 was dimerized after 5 min treatment with 0.1 or 1 mM H2O2, regardless of G6PD status. However, after 1 h, there was a clear distinction, with only limited recovery in reduced Prx2 in G6PD-deficient erythrocytes (median 27% with 0.1 mM and 4% with 1 mM H2O2) compared with almost complete reduction for the 3208
Vol. 28
July 2014
Figure 3. A) Representative Western blots of Prx2 in freshly isolated erythrocytes from newborn infants who had normal (N) G6PD activity or were deficient (D) in G6PD. Electrophoretic conditions and band identity are as in Fig. 2. B) Quantification of basal oxidized Prx2 in blood drawn from normal and G6PD-deficient infants. Western blots were run as in A and relative band intensities in each lane were quantified by densitometry. Box plots show medians and IQRs with whiskers representing 10 and 90% values and circles representing 5 and 95% values. G6PD-deficient erythrocytes showed almost 3-fold higher oxidized Prx2 than erythrocytes with normal activity [median (IQR): 8% (1–26%) vs. 3%, (1– 10%), respectively; P ⫽ 0.027].
The FASEB Journal 䡠 www.fasebj.org
CHEAH ET AL.
Figure 4. A) Representative Western blots of Prx2 in erythrocytes from newborn infants with normal or deficient G6PD enzyme activity, after challenge with H2O2. Samples were exposed to 0.1 or 1 mM H2O2 for 5 or 60 min, followed by addition of NEM and electrophoresis as in Fig. 2. B) Comparison of the extent of regeneration of reduced Prx2 at 60 min after challenge with H2O2 between normal and G6PD-deficient erythrocytes. Recovery was almost complete in normal erythrocytes [median (IQR): 99% (73– 100%) and 97% (56 –100%)] as compared with those that were G6PD deficient [median (IQR): 29% (10 –72%) and 4% (0 –7%)]. Box plots show medians and IQRs with whiskers representing 10 and 90% values and circles representing 5 and 95% values. Differences between normal (n⫽29) and G6PD-deficient erythrocytes (n⫽19) were statistically significant (P⬍0.001) for 0.1 and 1 mM H2O2.
correlation with the extent of Prx2 reduction (correlation coefficient 0.66 or 0.83 at 0.1 or 1 mM H2O2 respectively, P ⬍ 0.001, n⫽48).
DISCUSSION G6PD deficiency presents as hemolytic anemia when erythrocytes are unable to generate sufficient NADPH for their antioxidant systems to handle oxidative stress. Although Prx2 is a major antioxidant protein that requires NADPH for its activity (Fig. 1), impaired Prx2 function has received little attention as a contributor to the hemolytic process, and, to our knowledge, there are no previous reports describing the redox properties of Prx2 in G6PD-deficient erythrocytes. We examined erythrocytes from an infant population that is routinely screened for G6PD deficiency and monitored because of the risk of developing jaundice. We found that there was significantly more oxidized Prx2 present in freshly isolated G6PD-deficient erythrocytes compared with controls, with this basal Prx2 oxidation negatively correlated with G6PD enzyme activity. This indicates that the G6PD-deficient cells were in a more oxidized state. It is well established that erythrocytes continuously generate H2O2 (e.g., through autoxidation of hemoglobin), and they also scavenge H2O2 when it is present in their surroundings (26). This reacts with Prx2, so that if recycling is prevented by inhibiting thioredoxin reductase, oxidized Prx2 rapidly accumulates (14). In normal erythrocytes, recycling by thioredoxin and thioredoxin reductase is sufficient to handle basal oxidant generation (14). However, with increased oxidant exposure, such as in the presence of stimulated neutrophils, the rate of oxidation exceeds reduction, and oxidized Prx2 accumulates (20). For most of the G6PD-deficient erythrocyte samples we examined, a majority of the Prx2 was reduced, indicating that the low level of G6PD activity in these cells (mean 33% normal) is sufficient Prx2 IN G6PD-DEFICIENT ERYTHROCYTES
for the Prx2 antioxidant system to cope in circulation under basal conditions, albeit less efficiently. Erythrocytes from a minority of G6PD-deficient infants, and a few controls, showed high levels of Prx2 oxidation, suggesting that they could have been under increased oxidative stress. We have observed that erythrocyte Prx2 oxidation is increased in a mouse model of endotoxemia (20), and it is possible that infection, inflammation, or other conditions giving rise to oxidative stress could have been responsible for the high levels in the infants. However, no obvious cause could be identified from their clinical profiles, and further investigation is required to determine what influences Prx2 redox state. Stressing the cells with H2O2 revealed a major defect in capacity to regenerate reduced Prx2. As previously reported (14), normal erythrocytes regenerate reduced Prx2 within 1 h of H2O2 treatment. In contrast, very little regeneration was seen in the G6PD-deficient cells over that time, especially at the higher H2O2 dose. The extent of regeneration correlated positively with G6PD enzyme activity. On this basis, it would be expected that the Prx2 system would be severely compromised in circulating erythrocytes of G6PD-deficient individuals if they are exposed to the oxidative conditions that give rise to hemolysis. Furthermore, the classical presentation of drug-induced hemolysis in G6PD deficiency is intraerythrocytic formation of Heinz bodies, and Heinz body hemolytic anemia was the most striking phenotype of Prx2-knockout mice (18). It seems likely, therefore, that diminished ability of reduced Prx2 to remove peroxides contributes to hemolysis in G6PD deficiency. These observations also make it clear that G6PD deficiency is a condition in which the antioxidant function not only of the Prx2 system is impaired and not just that of glutathione peroxidase and catalase. Further investigation comparing the redox transformations of Prx2 and GSH in oxidatively stressed G6PD-deficient cells 3209
should help determine which of these systems plays a more critical role in the hemolytic process. One of the more common problems associated with G6PD deficiency is the development of jaundice shortly after birth (1, 27). As severe neonatal hyperbilirubinemia is potentially associated with grave consequences, such as bilirubin encephalopathy or kernicterus, preventive measures taken in some countries include programs for community and in-hospital monitoring of bilirubin and early access to phototherapy. With increased disease awareness, education, and newborn screening programs, acute hemolytic crises are becoming rarer. Furthermore, there is emerging and cumulative evidence that a nonhemolytic process contributes to neonatal jaundice in G6PD deficiency (2, 27). Whether this mechanism involves Prx2 is also unclear. Our study was not set up to detect whether jaundice is associated with Prx2 oxidation, in part because we intervene aggressively with phototherapy for the deficient infants in our institution to avoid the risk of severe hyperbilirubinemia. Also, a substantial number of our control infants were in the neonatal unit because they developed jaundice for other reasons. However, it might be expected that an increase in Prx2 oxidation could be a marker of oxidative stress and development of jaundice, and further investigation is warranted to explore this possibility. This work was supported by the Fundamental Research Grant Scheme (UKM-FF-03-FRGS0038-2010) from the Ministry of Higher Education of Malaysia and the Health Research Council of New Zealand. The authors thank Ms. Angeline Loh Mee-Ling, Dr. Chow Soon-Ken, and the nurses, doctors, and staff in the postnatal wards, neonatal intensive care unit, and molecular hematology laboratory of Universiti Kebangsaan Malaysia Medical Centre for their technical assistance.
2. 3.
4. 5. 6. 7.
8. 9.
3210
Cappellini, M. D., and Fiorelli, G. (2008) Glucose-6-phosphate dehydrogenase deficiency. Lancet 371, 64 –74 Beutler, E. (2008) Glucose-6-phosphate dehydrogenase deficiency: a historical perspective. Blood 111, 16 –24 Nkhoma, E. T., Poole, C., Vannappagari, V., Hall, S. A., and Beutler, E. (2009) The global prevalence of glucose-6-phosphate dehydrogenase deficiency: a systematic review and metaanalysis. Blood Cells Mol. Dis. 42, 267–278 Stanton, R. C. (2012) Glucose-6-phosphate dehydrogenase, NADPH, and cell survival. IUBMB Life 64, 362–369 Winterbourn, C. C. (1985) Free-radical production and oxidative reactions of hemoglobin. Environ. Health Perspect. 64, 321– 330 Mason, P. J., Bautista, J. M., and Gilsanz, F. (2007) G6PD deficiency: the genotype-phenotype association. Blood Rev. 21, 267–283 Gaetani, G. F., Ferraris, A. M., Rolfo, M., Mangerini, R., Arena, S., and Kirkman, H. N. (1996) Predominant role of catalase in the disposal of hydrogen peroxide within human erythrocytes. Blood 87, 1595–1599 Hochstein, P. (1988) Perspectives on hydrogen peroxide and drug-induced hemolytic anemia in glucose-6-phosphate dehydrogenase deficiency. Free Radic. Biol. Med. 5, 387–392 Low, F. M., Hampton, M. B., and Winterbourn, C. C. (2008) Peroxiredoxin 2 and peroxide metabolism in the erythrocyte. Antioxid. Redox Signal. 10, 1621–1630
Vol. 28
July 2014
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
REFERENCES 1.
10.
22. 23.
24. 25. 26.
27.
Rhee, S. G., and Woo, H. A. (2011) Multiple functions of peroxiredoxins: peroxidases, sensors and regulators of the intracellular messenger HO, and protein chaperones. Antioxid. Redox Signal. 15, 781–794 Hall, A., Karplus, P. A., and Poole, L. B. (2009) Typical 2-Cys peroxiredoxins–structures, mechanisms and functions. FEBS J. 276, 2469 –2477 Peskin, A. V., Cox, A. G., Nagy, P., Morgan, P. E., Hampton, M. B., Davies, M. J., and Winterbourn, C. C. (2010) Removal of amino acid, peptide and protein hydroperoxides by reaction with peroxiredoxins 2 and 3. Biochem. J. 432, 313–321 Moore, R., Mankad, M., Shriver, S., Mankad, V., and Plishker, G. (1991) Reconstitution of Ca(2⫹)-dependent K⫹ transport in erythrocyte membrane vesicles requires a cytoplasmic protein. J. Biol. Chem. 266, 18964 –18968 Low, F. M., Hampton, M. B., Peskin, A. V., and Winterbourn, C. C. (2007) Peroxiredoxin 2 functions as a noncatalytic scavenger of low-level hydrogen peroxide in the erythrocyte. Blood 109, 2611–2617 Johnson, R. M., Goyette, G., Jr., Ravindranath, Y., and Ho, Y. S. (2005) Hemoglobin autoxidation and regulation of endogenous H2O2 levels in erythrocytes. Free Radic. Biol. Med. 39, 1407–1417 Johnson, R. M., Ho, Y. S., Yu, D. Y., Kuypers, F. A., Ravindranath, Y., and Goyette, G. W. (2010) The effects of disruption of genes for peroxiredoxin-2, glutathione peroxidase-1, and catalase on erythrocyte oxidative metabolism. Free Radic. Biol. Med. 48, 519 –525 Cho, C. S., Lee, S., Lee, G. T., Woo, H. A., Choi, E. J., and Rhee, S. G. (2010) Irreversible inactivation of glutathione peroxidase 1 and reversible inactivation of peroxiredoxin II by H2O2 in red blood cells. Antioxid. Redox Sig. 12, 1235–1246 Lee, T. H., Kim, S. U., Yu, S. L., Kim, S. H., Park, D. S., Moon, H. B., Dho, S. H., Kwon, K. S., Kwon, H. J., Han, Y. H., Jeong, S., Kang, S. W., Shin, H. S., Lee, K. K., Rhee, S. G., and Yu, D. Y. (2003) Peroxiredoxin II is essential for sustaining life span of erythrocytes in mice. Blood 101, 5033–5038 Nagababu, E., Mohanty, J. G., Friedman, J. S., and Rifkind, J. M. (2013) Role of peroxiredoxin-2 in protecting RBCs from hydrogen peroxide-induced oxidative stress. Free. Radic. Res. 47, 164 –171 Bayer, S. B., Maghzal, G., Stocker, R., Hampton, M. B., and Winterbourn, C. C. (2013) Neutrophil-mediated oxidation of erythrocyte peroxiredoxin 2 as a potential marker of oxidative stress in inflammation. FASEB J. 27, 3315–3322 Peskin, A. V., Low, F. M., Paton, L. N., Maghzal, G. J., Hampton, M. B., and Winterbourn, C. C. (2007) The high reactivity of peroxiredoxin 2 with H(2)O(2) is not reflected in its reaction with other oxidants and thiol reagents. J. Biol. Chem. 282, 11885–11892 American Academy of Pediatrics (2004) Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 114, 297–316 Azma, R. Z., Hidayati, N., Farisah, N. R., Hamidah, N. H., and Ainoon, O. (2010) G6PD enzyme activity in normal term Malaysian neonates and adults using a OSMMR2000-D kit with Hb normalization. Southeast Asian J. Trop. Med. Publ. Health 41, 982–988 WHO Working Group. (1989) Glucose-6-phosphate dehydrogenase deficiency. Bull. WHO 67, 601–611 Poynton, R. A., and Hampton, M. B. (2014) Peroxiredoxins as biomarkers of oxidative stress. Biochim. Biophys. Acta 1840, 906 –912 Winterbourn, C. C., and Stern, A. (1987) Human red cells scavenge extracellular hydrogen peroxide and inhibit formation of hypochlorous acid and hydroxyl radical. J. Clin. Invest. 80, 1486 –1491 Seidman, D. S., Shiloh, M., Stevenson, D. K., Vreman, H. J., and Gale, R. (1995) Role of hemolysis in neonatal jaundice associated with glucose-6 phosphate dehydrogenase deficiency. J. Pediatr. 127, 804 –806
The FASEB Journal 䡠 www.fasebj.org
Received for publication January 19, 2014. Accepted for publication March 10, 2014.
CHEAH ET AL.
Increased basal oxidation of peroxiredoxin 2 and limited peroxiredoxin recycling in glucose-6-phosphate dehydrogenase-deficient erythrocytes from newborn infants Fook-Choe Cheah,* Alexander V. Peskin,‡ Fei-Liang Wong,* Azlin Ithnin,† Ainoon Othman,储 and Christine C. Winterbourn‡,§,1 *Department of Paediatrics and †Department of Pathology, Universiti Kebangsaan Malaysia Medical Centre, Kuala Lumpur, Malaysia; ‡Centre for Free Radical Research, Department of Pathology, and §Gravida National Centre for Growth and Development, University of Otago Christchurch, Christchurch, New Zealand; and 储Department of Pathology, Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia Erythrocytes require glucose-6-phosphate dehydrogenase (G6PD) to generate NADPH and protect themselves against hemolytic anemia induced by oxidative stress. Peroxiredoxin 2 (Prx2) is a major antioxidant enzyme that requires NADPH to recycle its oxidized (disulfide-bonded) form. Our aims were to determine whether Prx2 is more highly oxidized in G6PD-deficient erythrocytes and whether these cells are able to recycle oxidized Prx2 after oxidant challenge. Blood was obtained from 61 Malaysian neonates with G6PD deficiency (average 33% normal activity) and 86 controls. Prx2 redox state was analyzed by Western blotting under nonreducing conditions. Prx2 in freshly isolated blood was predominantly reduced in both groups, but the median level of oxidation was significantly higher (8 vs 3%) and the range greater for the G6PD-deficient population. When treated with reagent H2O2, the G6PD-deficient erythrocytes were severely compromised in their ability to recycle oxidized Prx2, with only 27 or 4% reduction after 1 h treatment with 0.1 or 1 mM H2O2 respectively, compared with >97% reduction in control erythrocytes. The accumulation of oxidized Prx2 in oxidatively stressed erythrocytes with common G6PD variants suggests that impaired antioxidant activity of Prx2 could contribute to the hemolysis and other complications associated with the condition.— Cheah, F.-C., Peskin, A. V., Wong, F.-L., Ithnin, A., Othman, A., Winterbourn, C. C. Increased basal oxidation of peroxiredoxin 2 and limited peroxiredoxin recycling in glucose-6-phosphate dehydrogenase deficient erythrocytes from newborn infants. FASEB J. 28, 3205–3210 (2014). www.fasebj.org ABSTRACT
Key Words: antioxidant defense 䡠 hydrogen peroxide 䡠 hemolytic anemia
Abbreviations: EA, enzyme activity; FST, fluorescent spot test; G6PD, glucose-6-phosphate dehydrogenase; IQR, interquartile range; NEM, N-ethylmaleimide; Prx2, peroxiredoxin 2 0892-6638/14/0028-3205 © FASEB
Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most commonly inherited genetically linked enzymopathy. An estimated 4 million people worldwide carry a mutation (1). Although normally asymptomatic, G6PD-deficient individuals can present with acute hemolysis associated with the ingestion of fava beans (favism), sulfur-containing and antimalarial drugs, or infection (2). Newborn infants can present with early onset and severe jaundice because of hyperbilirubinemia (1), which, if untreated, can result in bilirubin encephalopathy or kernicterus. G6PD deficiency is prevalent in many parts of Asia, including Malaysia (5–10%; ref. 3), where there is a nationwide routine cord blood screening program. The infants range from being asymptomatic to presenting with severe jaundice, and they are monitored closely with delayed discharge from the hospital. G6PD is part of the pentose phosphate pathway, which in erythrocytes is the sole source of NADPH (4). NADPH provides the reducing equivalents needed for the removal of reactive oxygen species, and clinical problems occur in G6PD deficiency when cells are exposed to increased oxidative stress. The problem is largely restricted to erythrocytes because of their dependence on the pentose phosphate pathway (1) and because hemoglobin is a major source of reactive oxidants (5). Furthermore, most G6PD mutations result in an unstable, short-lived protein (6), which cannot be replenished in mature erythrocytes. Several antioxidant pathways use NADPH as the terminal reductant. Although NADPH increases the efficiency of H2O2 removal by catalase (7), the common view is that it is mainly required for removal of peroxides via the glutathione peroxidase/glutathione 1 Correspondence: Centre for Free Radical Research, Department of Pathology, University of Otago Christchurch, PO Box 4345, Christchurch, New Zealand. E-mail: christine. [email protected] doi: 10.1096/fj.14-250050
3205
Figure 1. Scheme showing oxidation of Prx2 by H2O2. Prx2 exists as a homodimer. Oxidation occurs at a highly reactive active-site Cys residue, which condenses with the resolving Cys on the adjacent chain to form a disulfide-linked dimer. This is recycled by thioredoxin (Trx) through thioredoxin reductase (TrxR), with electrons coming from NADPH. NADPH is supplied by G6PD and the pentose phosphate pathway (PPP).
reductase pathway (8). However, another important class of NADPH-dependent antioxidant enzymes is the peroxiredoxins (9). Peroxiredoxins (there are 6 in humans) are thiol proteins that react extremely rapidly with H2O2, peroxynitrite, and other peroxides (10 –12). Peroxiredoxin 2 (Prx2) is the major peroxiredoxin in erythrocytes, where it constitutes the third most abundant protein (9, 13). Reduced Prx2 exists as a noncovalent homodimer with active-site cysteine (Cys) residues. Oxidation results in a disulfide-linked dimer that is recycled by thioredoxin/thioredoxin reductase and NADPH (Fig. 1). Therefore, the antioxidant activity of erythrocyte Prx2 requires G6PD. However, there have been no studies examining whether the peroxiredoxin pathway is compromised in G6PD deficiency or whether this could contribute to the hemolytic mechanism. Prx2 plays a major role in peroxide removal in the erythrocyte (14 –17). It becomes oxidized in erythrocytes exposed to very low concentrations of H2O2, even though they contain active catalase and glutathione peroxidase (14). It also scavenges endogenous peroxide generated through hemoglobin autoxidation (14), and erythrocytes from Prx2-knockout mice contained higher amounts of hemoglobin oxidation products (18, 19). Increased levels of erythrocyte Prx2 oxidation have been observed in a mouse model of sepsis (20). It is likely therefore that Prx2 function could be impaired in G6PD deficiency and contribute to the phenotype. To test this proposal, we compared erythrocytes from newborn infants with and without G6PD deficiency to determine whether Prx2 is more highly oxidized under basal conditions in the infants with G6PD deficiency and whether G6PD-deficient erythrocytes are compromised in their ability to regenerate reduced Prx2 after exposure to H2O2.
included. For controls, study bloods were obtained during routine sampling from non-G6PD-deficient infants who were admitted to the neonatal intensive care unit because of jaundice requiring phototherapy or considered at risk for low glucose or infection, or from healthy infants awaiting discharge from the postnatal ward. Blood for basal Prx2 analysis was collected by venous sampling at the same time as blood for other routine investigations which included enzyme activity (EA) for G6PD and bilirubin levels. For Prx2 analysis, 350 l blood was collected into an EDTA-containing microcentrifuge tube with preadded N-ethylmaleimide (NEM; 150 l, 200 mM). The median age at sampling was 3 d [interquartile range (IQR), 2–5 d]. Prx2 reacts with H2O2 very rapidly (21) and can easily be oxidized by traces of H2O2 during cell lysis (14). To prevent artifactual Prx2 oxidation, blood was added directly to a large excess of NEM to trap its existing oxidation state before lysis of the cells. Thus, no free thiols would still be present during lysis, and oxidation by H2O2 would be prevented. To check that sufficient NEM was added to study samples, two blood samples were treated with different amounts of 200 mM NEM (Fig. 2) then subjected to immunoblot analysis (as described below). At NEM:blood ratios of 1:1 and 1:4, there were similar low levels of oxidized (dimeric) Prx2, but at a 1:10 ratio, trapping with NEM became less efficient and the percentage oxidation began to increase. Thus the procedure used should prevent oxidation during lysis. To study in vitro responses to H2O2, fresh blood samples (without NEM) were also collected from a subpopulation of the control and G6PD-deficient infants.
MATERIALS AND METHODS Study population and samples This study was conducted over a period of 18 mo from July 2011 until December 2012. Ethics approval was obtained from the Universiti Kebangsaan Malaysia (UKM) Research Ethics Committee (UKM-FF-FRGS0038-2010). All infants delivered at the UKM Medical Centre were routinely screened for G6PD deficiency by the fluorescent spot test (FST) on cord blood. Infants born by spontaneous vaginal delivery or elective cesarean sections to uncomplicated pregnancies were enrolled after parental consent was obtained. Randomly selected infants who were newly diagnosed with G6PD deficiency and monitored for jaundice in the hospital were 3206
Vol. 28
July 2014
Figure 2. Effect of NEM concentration on the redox state of Prx2 in erythrocytes after lysis. Blood samples from 2 control subjects (I and II) were mixed with 200 mM NEM at various volume ratios before lysis and subjected to nonreducing SDS-PAGE. The top band of the Western blot represents oxidized Prx2 (disulfide-linked dimer) which is ⬃44 kDa; the bottom band is reduced monomer (22 kDa). Dimerization of Prx2 was inhibited with an excess of NEM (200 mM) to blood of ⱖ1:4 in ratio.
The FASEB Journal 䡠 www.fasebj.org
CHEAH ET AL.
Bilirubin and phototherapy Newborn infants with G6PD deficiency diagnosed by FST screening are routinely monitored for jaundice in the hospital for a minimum of 5 d at the UKM Medical Centre. G6PD analysis by EA is performed as further confirmation and to evaluate the category of severity in enzyme deficiency. If a baby developed jaundice, peripheral blood was taken for the measurement of serum bilirubin, and phototherapy was initiated using the guide proposed by the Ministry of Health, which was based on the reference curve developed by the American Academy of Pediatrics (22). In general, a bilirubin level that breached 255 M would be considered as significant hyperbilirubinemia requiring phototherapy intervention regardless of the age of the infant. Detection of G6PD deficiency Screening for G6PD deficiency was performed on all cord bloods using the modified FST as described by Beutler et al. (2). In this assay, a blood spot is applied immediately to filter paper and dried at room temperature. The disc of dried blood is cut out and reacted with 100 l of a mixture containing glucose-6-phosphate, GSSG, and NADP⫹. After 15 min at 37°C, 10 l of this mixture is applied to filter paper, dried, and examined under ultraviolet light. Quantitative measurement of G6PD enzyme activity was performed on 5 l of whole EDTA blood using the OSMMR-D G6PD assay kit (R & D Diagnostics, Holargos, Greece) with hemoglobin normalization, according to the supplier’s instructions. The reference values for normal, partial deficiency, and deficiency of G6PD activity were ⬎6.76, 2.25– 6.76, and ⬍2.25 U/g hemoglobin, respectively, adapted based on the ranges published by Azma et al. (23). Genotype analysis Blood spots on filter paper were processed for molecular analysis to screen for the 10 most common G6PD mutations prevalent in Southeast Asia: c.871G⬎A, c.487G⬎A, c.1388G⬎A, c.1376G⬎T, c.95A⬎G, c.1024C⬎T, c.563C⬎T, c.392G⬎T, c.1003G⬎A, and c.592C⬎T. DNA was extracted using the TaqMan Sample-to-SNP kit, and genotyping was done using the TaqMan MGB single-nucleotide polymorphism (SNP) assay (Applied Biosystems, Foster City, CA, USA). Immunoblot analysis for Prx2 Whole blood was analyzed without separation of the erythrocytes. After collection, samples were left at room temperature for 1 h, then 10 l of blood was diluted to 1 ml with PBS containing 40 mM NEM. After 30 min, 100 l was diluted with equal volumes of 200 mM NEM and 6% SDS. Samples were stored at ⫺80°C before being transported frozen to Christchurch, New Zealand, for analysis. Proteins were resolved by nonreducing 12% SDS-PAGE and transferred electrophoretically to a Hybond polyvinylidene difluoride (PVDF) membrane (14). Blocking was performed for 1 h at room temperature in 5% (w/v) nonfat dried milk in Tris-buffered saline with 0.05% Tween 20 (TBST20). Incubation with antibodies to Prx2 (Sigma Chemical Co., St Louis, MO, USA) was performed overnight at 4°C or for 1 h at room temperature in TBST20 with 5% milk. Then membranes were washed and probed with horseradish peroxidase-conjugated goat anti-rabbit antibodies (Dako, Carpinteria, CA, USA). Bands were visualized through enhanced chemiluminescence using the ECLPlus Western blotting Detection System (GE Healthcare Biosciences Corp., Piscataway, NJ, USA). ImmuPrx2 IN G6PD-DEFICIENT ERYTHROCYTES
noblots were scanned using a ChemiDoc XRS gel documentation system, and densitometry was performed using Quantity One software (Bio-Rad, Hercules, CA, USA). H2O2 oxidation of Prx2 Freshly drawn blood from randomly selected infants was used to study the efficiency of reduction of Prx2 in erythrocytes treated with H2O2. Blood samples were diluted 10-fold in Hanks buffer with 5 mM glucose, and 45 l aliquots were reacted with 5 l of either 1 mM or 10 mM H2O2 for 5 min or 1 h at room temperature. These high concentrations were used to overcome erythrocyte catalase and ensure that dimerization of the Prx2 was (near) complete so that reduction could be followed. The reaction was stopped by blocking thiols with 20 l of 200 mM NEM, then 20 l of 10% SDS was added and the samples were frozen for later immunoblotting. Statistical analysis Data were analyzed using SigmaPlot 12 (Systat Software Inc., San Jose, CA, USA). Results are expressed as means ⫾ sd if normally distributed, otherwise as medians (IQR). Comparison of 2 groups of normally distributed data was based on the t test, while nonparametric data was analyzed using the Mann-Whitney U test. Categorical data such as gender, birth mode, and phototherapy were analyzed using 2 and the relationship between 2 parameters using the Pearson correlation and linear regression. The level of statistical significance was taken when P ⬍ 0.05.
RESULTS Group clinical characteristics The characteristics of the study population are shown in Table 1. On average, erythrocytes from infants with G6PD deficiency had 33% of the enzyme activity of the control infants (Table 1). Most had activities in the 20 – 60% of normal range, with only 4 showing severe deficiency. Half the infants had the two most common mutations in the Asian community (33% c.871G⬎A and 17% c.487G⬎A), which are classified as class III (deficiency associated with 10 – 60% of residual EA) in the WHO functional definition of G6PD deficiency (24). The third most common mutation was c.1388G⬎A (11%), which is categorized as class II (deficiency ⬍10% of residual EA). There was no difference between infants with G6PD deficiency and infants with normal G6PD activity in gestational age, birth weight, or mode of delivery (Table 1). The infants were of term gestation (mean of 38 wk) and two-thirds were born by spontaneous vaginal delivery. As expected for an X-linked condition, the majority of the infants with G6PD deficiency were male. Peak bilirubin levels, severe hyperbilirubinemia (bilirubin ⬎255 M), and the proportion of infants receiving phototherapy were similar between groups. Because infants with G6PD deficiency are treated aggressively because of their “high-risk” category, by commencing phototherapy at a low serum bilirubin threshold on the reference curve, none of the infants became more severely jaundiced. 3207
TABLE 1. Characteristics of the study population Parameter
n G6PD enzyme level (U/g Hb)a Deficiency, n ⫽ 50 (U/g Hb) Range (U/g Hb) Severe deficiency, n ⫽ 4 (U/g Hb) Range (U/g Hb) Gender [n (%)] Male Female Mode of birth Spontaneous vaginal (n) Cesarean section [n (%)] Gestation (wk) Birth weight (kg) Bilirubin level (M) Peak, wk 1 of life IQR Received phototherapy [n (%)]
Normal
G6PD deficient
86 10.9 ⫾ 1.8 – – – –
61 3.6 ⫾ 1.3** 3.8 ⫾ 1.1 2.3–6.5 1.0 ⫾ 0.7 0.4–2.0
38 (44) 48 (56)
57 (93)** 4 (7)
59 27 (31) 38 ⫾ 2 2.96 ⫾ 0.51
47 14 (23) 38 ⫾ 2 3.06 ⫾ 0.45
228 ⫾ 80 160–281 30 (35)
232 ⫾ 111 189–273 23 (38)
Unless stated otherwise, values are means ⫾ sd. Comparisons were not statistically significant unless indicated. aEnzyme activity available for n ⫽ 77 normal and n ⫽ 54 G6PD-deficient infants. **P ⬍ 0.001 vs. normal group.
Basal oxidation state of Prx2 The oxidation state of Prx2 in the red cells of the infants was assessed by nonreducing SDS-PAGE and Western blotting. Under these conditions, reduced Prx2 runs as a 22-kDa monomer, and the reversibly oxidized protein, which forms a disulfide-linked dimer, runs slower at twice this molecular mass (25). As shown for selected samples in Fig. 3A, Prx2 was in most cases predominantly in the reduced form, with varying amounts of the oxidized dimer present. However, the percentage of oxidized Prx2 was significantly higher in the G6PD-deficient erythrocytes (median 8%) than in the cells with normal G6PD activity (median 3%, P⫽0.027), and the range was greater (Fig. 3B). The extent of oxidation for all infants was also negatively correlated with G6PD enzyme activity (correlation coefficient ⫺0.20, P⫽0.024, n⫽131). No association between extent of oxidation and mode of delivery was evident.
controls (99 and 97% respectively; Fig. 4B). G6PD enzyme activity was measured for all the infants included in this part of the study and showed a positive
Impaired ability of G6PD-deficient erythrocytes to reduce Prx2 When normal erythrocytes are treated with H2O2, the Prx2 is initially oxidized to the disulfide and then recycled to the reduced monomer once the peroxide has been consumed (14). To test whether the recycling mechanism is impaired in G6PD deficiency, H2O2 was added to fresh blood samples, and the oxidation state of Prx2 was monitored at 5 and 60 min. As shown for the examples in Fig. 4A, all or almost all of the Prx2 was dimerized after 5 min treatment with 0.1 or 1 mM H2O2, regardless of G6PD status. However, after 1 h, there was a clear distinction, with only limited recovery in reduced Prx2 in G6PD-deficient erythrocytes (median 27% with 0.1 mM and 4% with 1 mM H2O2) compared with almost complete reduction for the 3208
Vol. 28
July 2014
Figure 3. A) Representative Western blots of Prx2 in freshly isolated erythrocytes from newborn infants who had normal (N) G6PD activity or were deficient (D) in G6PD. Electrophoretic conditions and band identity are as in Fig. 2. B) Quantification of basal oxidized Prx2 in blood drawn from normal and G6PD-deficient infants. Western blots were run as in A and relative band intensities in each lane were quantified by densitometry. Box plots show medians and IQRs with whiskers representing 10 and 90% values and circles representing 5 and 95% values. G6PD-deficient erythrocytes showed almost 3-fold higher oxidized Prx2 than erythrocytes with normal activity [median (IQR): 8% (1–26%) vs. 3%, (1– 10%), respectively; P ⫽ 0.027].
The FASEB Journal 䡠 www.fasebj.org
CHEAH ET AL.
Figure 4. A) Representative Western blots of Prx2 in erythrocytes from newborn infants with normal or deficient G6PD enzyme activity, after challenge with H2O2. Samples were exposed to 0.1 or 1 mM H2O2 for 5 or 60 min, followed by addition of NEM and electrophoresis as in Fig. 2. B) Comparison of the extent of regeneration of reduced Prx2 at 60 min after challenge with H2O2 between normal and G6PD-deficient erythrocytes. Recovery was almost complete in normal erythrocytes [median (IQR): 99% (73– 100%) and 97% (56 –100%)] as compared with those that were G6PD deficient [median (IQR): 29% (10 –72%) and 4% (0 –7%)]. Box plots show medians and IQRs with whiskers representing 10 and 90% values and circles representing 5 and 95% values. Differences between normal (n⫽29) and G6PD-deficient erythrocytes (n⫽19) were statistically significant (P⬍0.001) for 0.1 and 1 mM H2O2.
correlation with the extent of Prx2 reduction (correlation coefficient 0.66 or 0.83 at 0.1 or 1 mM H2O2 respectively, P ⬍ 0.001, n⫽48).
DISCUSSION G6PD deficiency presents as hemolytic anemia when erythrocytes are unable to generate sufficient NADPH for their antioxidant systems to handle oxidative stress. Although Prx2 is a major antioxidant protein that requires NADPH for its activity (Fig. 1), impaired Prx2 function has received little attention as a contributor to the hemolytic process, and, to our knowledge, there are no previous reports describing the redox properties of Prx2 in G6PD-deficient erythrocytes. We examined erythrocytes from an infant population that is routinely screened for G6PD deficiency and monitored because of the risk of developing jaundice. We found that there was significantly more oxidized Prx2 present in freshly isolated G6PD-deficient erythrocytes compared with controls, with this basal Prx2 oxidation negatively correlated with G6PD enzyme activity. This indicates that the G6PD-deficient cells were in a more oxidized state. It is well established that erythrocytes continuously generate H2O2 (e.g., through autoxidation of hemoglobin), and they also scavenge H2O2 when it is present in their surroundings (26). This reacts with Prx2, so that if recycling is prevented by inhibiting thioredoxin reductase, oxidized Prx2 rapidly accumulates (14). In normal erythrocytes, recycling by thioredoxin and thioredoxin reductase is sufficient to handle basal oxidant generation (14). However, with increased oxidant exposure, such as in the presence of stimulated neutrophils, the rate of oxidation exceeds reduction, and oxidized Prx2 accumulates (20). For most of the G6PD-deficient erythrocyte samples we examined, a majority of the Prx2 was reduced, indicating that the low level of G6PD activity in these cells (mean 33% normal) is sufficient Prx2 IN G6PD-DEFICIENT ERYTHROCYTES
for the Prx2 antioxidant system to cope in circulation under basal conditions, albeit less efficiently. Erythrocytes from a minority of G6PD-deficient infants, and a few controls, showed high levels of Prx2 oxidation, suggesting that they could have been under increased oxidative stress. We have observed that erythrocyte Prx2 oxidation is increased in a mouse model of endotoxemia (20), and it is possible that infection, inflammation, or other conditions giving rise to oxidative stress could have been responsible for the high levels in the infants. However, no obvious cause could be identified from their clinical profiles, and further investigation is required to determine what influences Prx2 redox state. Stressing the cells with H2O2 revealed a major defect in capacity to regenerate reduced Prx2. As previously reported (14), normal erythrocytes regenerate reduced Prx2 within 1 h of H2O2 treatment. In contrast, very little regeneration was seen in the G6PD-deficient cells over that time, especially at the higher H2O2 dose. The extent of regeneration correlated positively with G6PD enzyme activity. On this basis, it would be expected that the Prx2 system would be severely compromised in circulating erythrocytes of G6PD-deficient individuals if they are exposed to the oxidative conditions that give rise to hemolysis. Furthermore, the classical presentation of drug-induced hemolysis in G6PD deficiency is intraerythrocytic formation of Heinz bodies, and Heinz body hemolytic anemia was the most striking phenotype of Prx2-knockout mice (18). It seems likely, therefore, that diminished ability of reduced Prx2 to remove peroxides contributes to hemolysis in G6PD deficiency. These observations also make it clear that G6PD deficiency is a condition in which the antioxidant function not only of the Prx2 system is impaired and not just that of glutathione peroxidase and catalase. Further investigation comparing the redox transformations of Prx2 and GSH in oxidatively stressed G6PD-deficient cells 3209
should help determine which of these systems plays a more critical role in the hemolytic process. One of the more common problems associated with G6PD deficiency is the development of jaundice shortly after birth (1, 27). As severe neonatal hyperbilirubinemia is potentially associated with grave consequences, such as bilirubin encephalopathy or kernicterus, preventive measures taken in some countries include programs for community and in-hospital monitoring of bilirubin and early access to phototherapy. With increased disease awareness, education, and newborn screening programs, acute hemolytic crises are becoming rarer. Furthermore, there is emerging and cumulative evidence that a nonhemolytic process contributes to neonatal jaundice in G6PD deficiency (2, 27). Whether this mechanism involves Prx2 is also unclear. Our study was not set up to detect whether jaundice is associated with Prx2 oxidation, in part because we intervene aggressively with phototherapy for the deficient infants in our institution to avoid the risk of severe hyperbilirubinemia. Also, a substantial number of our control infants were in the neonatal unit because they developed jaundice for other reasons. However, it might be expected that an increase in Prx2 oxidation could be a marker of oxidative stress and development of jaundice, and further investigation is warranted to explore this possibility. This work was supported by the Fundamental Research Grant Scheme (UKM-FF-03-FRGS0038-2010) from the Ministry of Higher Education of Malaysia and the Health Research Council of New Zealand. The authors thank Ms. Angeline Loh Mee-Ling, Dr. Chow Soon-Ken, and the nurses, doctors, and staff in the postnatal wards, neonatal intensive care unit, and molecular hematology laboratory of Universiti Kebangsaan Malaysia Medical Centre for their technical assistance.
2. 3.
4. 5. 6. 7.
8. 9.
3210
Cappellini, M. D., and Fiorelli, G. (2008) Glucose-6-phosphate dehydrogenase deficiency. Lancet 371, 64 –74 Beutler, E. (2008) Glucose-6-phosphate dehydrogenase deficiency: a historical perspective. Blood 111, 16 –24 Nkhoma, E. T., Poole, C., Vannappagari, V., Hall, S. A., and Beutler, E. (2009) The global prevalence of glucose-6-phosphate dehydrogenase deficiency: a systematic review and metaanalysis. Blood Cells Mol. Dis. 42, 267–278 Stanton, R. C. (2012) Glucose-6-phosphate dehydrogenase, NADPH, and cell survival. IUBMB Life 64, 362–369 Winterbourn, C. C. (1985) Free-radical production and oxidative reactions of hemoglobin. Environ. Health Perspect. 64, 321– 330 Mason, P. J., Bautista, J. M., and Gilsanz, F. (2007) G6PD deficiency: the genotype-phenotype association. Blood Rev. 21, 267–283 Gaetani, G. F., Ferraris, A. M., Rolfo, M., Mangerini, R., Arena, S., and Kirkman, H. N. (1996) Predominant role of catalase in the disposal of hydrogen peroxide within human erythrocytes. Blood 87, 1595–1599 Hochstein, P. (1988) Perspectives on hydrogen peroxide and drug-induced hemolytic anemia in glucose-6-phosphate dehydrogenase deficiency. Free Radic. Biol. Med. 5, 387–392 Low, F. M., Hampton, M. B., and Winterbourn, C. C. (2008) Peroxiredoxin 2 and peroxide metabolism in the erythrocyte. Antioxid. Redox Signal. 10, 1621–1630
Vol. 28
July 2014
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
REFERENCES 1.
10.
22. 23.
24. 25. 26.
27.
Rhee, S. G., and Woo, H. A. (2011) Multiple functions of peroxiredoxins: peroxidases, sensors and regulators of the intracellular messenger HO, and protein chaperones. Antioxid. Redox Signal. 15, 781–794 Hall, A., Karplus, P. A., and Poole, L. B. (2009) Typical 2-Cys peroxiredoxins–structures, mechanisms and functions. FEBS J. 276, 2469 –2477 Peskin, A. V., Cox, A. G., Nagy, P., Morgan, P. E., Hampton, M. B., Davies, M. J., and Winterbourn, C. C. (2010) Removal of amino acid, peptide and protein hydroperoxides by reaction with peroxiredoxins 2 and 3. Biochem. J. 432, 313–321 Moore, R., Mankad, M., Shriver, S., Mankad, V., and Plishker, G. (1991) Reconstitution of Ca(2⫹)-dependent K⫹ transport in erythrocyte membrane vesicles requires a cytoplasmic protein. J. Biol. Chem. 266, 18964 –18968 Low, F. M., Hampton, M. B., Peskin, A. V., and Winterbourn, C. C. (2007) Peroxiredoxin 2 functions as a noncatalytic scavenger of low-level hydrogen peroxide in the erythrocyte. Blood 109, 2611–2617 Johnson, R. M., Goyette, G., Jr., Ravindranath, Y., and Ho, Y. S. (2005) Hemoglobin autoxidation and regulation of endogenous H2O2 levels in erythrocytes. Free Radic. Biol. Med. 39, 1407–1417 Johnson, R. M., Ho, Y. S., Yu, D. Y., Kuypers, F. A., Ravindranath, Y., and Goyette, G. W. (2010) The effects of disruption of genes for peroxiredoxin-2, glutathione peroxidase-1, and catalase on erythrocyte oxidative metabolism. Free Radic. Biol. Med. 48, 519 –525 Cho, C. S., Lee, S., Lee, G. T., Woo, H. A., Choi, E. J., and Rhee, S. G. (2010) Irreversible inactivation of glutathione peroxidase 1 and reversible inactivation of peroxiredoxin II by H2O2 in red blood cells. Antioxid. Redox Sig. 12, 1235–1246 Lee, T. H., Kim, S. U., Yu, S. L., Kim, S. H., Park, D. S., Moon, H. B., Dho, S. H., Kwon, K. S., Kwon, H. J., Han, Y. H., Jeong, S., Kang, S. W., Shin, H. S., Lee, K. K., Rhee, S. G., and Yu, D. Y. (2003) Peroxiredoxin II is essential for sustaining life span of erythrocytes in mice. Blood 101, 5033–5038 Nagababu, E., Mohanty, J. G., Friedman, J. S., and Rifkind, J. M. (2013) Role of peroxiredoxin-2 in protecting RBCs from hydrogen peroxide-induced oxidative stress. Free. Radic. Res. 47, 164 –171 Bayer, S. B., Maghzal, G., Stocker, R., Hampton, M. B., and Winterbourn, C. C. (2013) Neutrophil-mediated oxidation of erythrocyte peroxiredoxin 2 as a potential marker of oxidative stress in inflammation. FASEB J. 27, 3315–3322 Peskin, A. V., Low, F. M., Paton, L. N., Maghzal, G. J., Hampton, M. B., and Winterbourn, C. C. (2007) The high reactivity of peroxiredoxin 2 with H(2)O(2) is not reflected in its reaction with other oxidants and thiol reagents. J. Biol. Chem. 282, 11885–11892 American Academy of Pediatrics (2004) Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 114, 297–316 Azma, R. Z., Hidayati, N., Farisah, N. R., Hamidah, N. H., and Ainoon, O. (2010) G6PD enzyme activity in normal term Malaysian neonates and adults using a OSMMR2000-D kit with Hb normalization. Southeast Asian J. Trop. Med. Publ. Health 41, 982–988 WHO Working Group. (1989) Glucose-6-phosphate dehydrogenase deficiency. Bull. WHO 67, 601–611 Poynton, R. A., and Hampton, M. B. (2014) Peroxiredoxins as biomarkers of oxidative stress. Biochim. Biophys. Acta 1840, 906 –912 Winterbourn, C. C., and Stern, A. (1987) Human red cells scavenge extracellular hydrogen peroxide and inhibit formation of hypochlorous acid and hydroxyl radical. J. Clin. Invest. 80, 1486 –1491 Seidman, D. S., Shiloh, M., Stevenson, D. K., Vreman, H. J., and Gale, R. (1995) Role of hemolysis in neonatal jaundice associated with glucose-6 phosphate dehydrogenase deficiency. J. Pediatr. 127, 804 –806
The FASEB Journal 䡠 www.fasebj.org
Received for publication January 19, 2014. Accepted for publication March 10, 2014.
CHEAH ET AL.
