Eur J Appl Physiol (2012) 112:2549–2556 DOI 10.1007/s00421-011-2229-0
O R I G I N A L A R T I CL E
Is glucose-6-phosphate dehydrogenase deWciency a risk factor for hyperbaric oxygen exposure? Mirit Eynan · Dimitry Tsitlovsky · Liron Batit · Ayala Hochman · Nitzan Krinsky · Amir Abramovich
Received: 23 December 2010 / Accepted: 27 October 2011 / Published online: 11 November 2011 © Springer-Verlag 2011
Abstract Divers and patients lacking glucose-6-phosphate dehydrogenase (G6PD) may face a serious threat of central nervous system oxygen toxicity (CNS-OT) during exposure to hyperbaric oxygen (HBO), due to the important part played by G6PD in cellular redox balance. Our objective was to investigate G6PD deWciency as a risk factor for CNS-OT. We exposed G6PD-deWcient (G6PDdef) and wild type (WT) mice to HBO at 405 kPa. Latency to CNS-OT was measured by observing the animal and monitoring the time to appearance of convulsions. Changes in glutathione peroxidase (GPx) and catalase activity were measured in red blood cells, and levels of endothelial and neuronal nitric oxide synthase (eNOS and nNOS) and 3-nitrotyrosine (NT) were measured in extracts of whole brain tissue by Western blot analysis. Unexpectedly, latency to CNS-OT was more than twice as long in G6PDdef mice compared with WT (36.9 § 15.4 and 15.6 § 13.2 min, respectively, P < 0.005). No signiWcant diVerences were found in GPx and catalase activity or in protein levels of eNOS. However, nNOS and NT levels were lower in G6PDdef mice compared with WT (50.6%, P < 0.01 and 52.8%, P < 0.05, respectively). Our results suggest that the enhanced resistance of G6PDdef mice to HBO is due in part to a reduction in nNOS and NT
Communicated by Dag Linnarsson. M. Eynan (&) · D. Tsitlovsky · L. Batit · N. Krinsky · A. Abramovich Israel Naval Medical Institute, IDF Medical Corps, P.O. Box 8040, 31080 Haifa, Israel e-mail:
[email protected] A. Hochman Department of Biochemistry, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel
levels in the brain. We conclude that G6PD deWciency at the level of the animals in the present study may not be a risk factor for developing CSN-OT, but this remains to be veriWed for human subjects. Keywords Central nervous system oxygen toxicity · G6PD deWciency · Hyperbaric oxygen
Introduction Exposure to hyperbaric oxygen in divers breathing a high partial pressure of O2 or patients undergoing hyperbaric oxygen therapy was shown to cause central nervous system oxygen toxicity (CNS-OT), which is characterized by convulsions similar to epileptic seizures and sudden loss of consciousness, with or without any warning signs (Arieli et al. 2006). It has been shown that reactive oxygen (ROS) play a role in the etiology of CNS-OT (Ohtsuki et al. 1992; Torbati et al. 1992); other investigators have demonstrated the importance of nitric oxide (NO) and related nitrogen species (RNS) in the pathophysiology of oxygen toxicity (Bitterman and Bitterman 1998; Elayan et al. 2000; Oury et al. 1992, 1994). Glucose-6-phosphate dehydrogenasedeWcient (G6PDdef) individuals, because of their apparent reduced ability to neutralize ROS, are therefore expected to be more susceptible to CNS-OT. G6PD deWciency, an X-linked disorder, is the most common genetic disorder in humans. It aVects more than 400 million persons worldwide, mostly throughout Africa, Asia, the Mediterranean and the Middle East, as well as cases of several gene mutations in all populations (Beutler 1994). The G6PD-deWcient variants are grouped into diVerent classes which match the disease’s severity. Class 1 is nonspherocytic hemolytic anemia in the presence of normal
123
2550
erythrocyte function. This class is severe but uncommon. Class 2 is severe, characterized by less than 10 percent of normal activity. This variation is common in Asia and in the Mediterranean basin. Class 3 is moderate, characterized by 10–60 percent of normal activity. This variation is common in Africa and among 10 percent of black males in the United States (Frank 2005). In Israel, G6PD deWciency is found mostly in Sephardic Jews whose families originated in Asia Minor. The Mediterranean variant, in which the activity of the enzyme is only 3% of the normal (Oppenheim et al. 1993), is common in this population. Individuals who have this disorder exhibit symptoms only when they consume certain foods, such as Vicia faba, or when they take certain drugs which render them susceptible to hemolytic crisis (Beutler 1994; Cappellini and Fiorelli 2008; Luzzatto and Mehta 1995). G6PD acts to reduce nicotinamide-adenine dinucleotide phosphate (NADP) to NADPH, which is essential for the reduction of glutathione from its oxidized form (GSSG) to the reduced form (GSH) by the antioxidant system. GSH is a low molecular antioxidant. It also plays a role in the homeostatic maintenance of cellular reduced thiols, and is a cofactor of the enzymes glutathione peroxidase (GPx) and glutathione transferase. Consequently, maintenance of an adequate steady state of GSH/GSSG is essential for protection against ROS in general, and in particular to detoxify hydrogen peroxide and lipid hydroperoxides. To the best of our knowledge, there are no publications on the eVect of G6PD deWciency on sensitivity to CNS-OT. It was therefore the purpose of the present study, using G6PDdef mice as a model, to investigate whether reduced activity of G6PD is a risk factor for CNS-OT when diving with a high partial pressure of oxygen or during hyperbaric oxygen therapy. We compared latency to CNS-OT during exposure to hyperbaric oxygen in G6PDdef mice and the wild type (WT). We also compared the activity of catalase and GPx in the red blood cells (RBC) of the two strains, to examine whether there is compensation for the alleged reduction in antioxidant capacity. We Wnally compared levels of endothelial and neuronal nitric oxide synthase (eNOS and nNOS) and 3-nitrotyrosine (NT) in the brain, because of the involvement of nitric oxide (NO) in the etiology of CNS-OT.
Materials and methods Animals Breeding pairs of G6PD-mutant C3H mice and WT were purchased from the Medical Research Council of England (Harwell, UK). The Animal Care Committee of the Israel Ministry of Defense approved the experimental procedure,
123
Eur J Appl Physiol (2012) 112:2549–2556
and the animals were handled and surgical procedures performed in accordance with internationally accepted humane standards. This G6PDdef mouse line was originally created by Pretsch et al. (1988), and exhibits decreased G6PD protein expression and enzyme activity caused by a single mutation in the untranslated region of the X-linked G6PD gene (Sanders et al. 1997). To establish a breeding colony, one mutant female and one mutant male or one WT female and one WT male were placed in each cage. Laboratory rodent chow and tap water were provided ad libitum. When it was seen that a female had become pregnant, the male was removed to a diVerent cage. After delivery, pups were left with the mother for 21 days. The number, sex and G6PD genotype of weaned pups were recorded. G6PD genotyping DNA was prepared from the animals’ tails, and a 269-bp fragment of the G6PD gene was ampliWed by polymerase chain reaction (PCR) using the forward primer GGAAACTGGCTGTGCGCTAC and reverse primer TCAGCTCCGGCTCTCTTCTG (Sanders et al. 1997). The PCR fragment was then digested with Ddel restriction endonuclease (New England Biolabs, Ipswich, MA, USA), and the restriction products were separated on 2.5% agarose gel. Ddel enzyme produced bands of 55 and 214 bp in the WT mice, but did not cleave the mutant G6PD sequence in G6PDdef mice. Experimental system and procedure We exposed 11 G6PDdef and 15 WT mice one by one to hyperbaric oxygen. The mouse was put in the experimental cage, which was placed in a 150-l hyperbaric chamber (Roberto Galeazzi, La Spezia, Italy) (Arieli et al. 2006). The animal was unrestrained, and could move about freely inside the cage. The pressure in the chamber was raised at a rate of 180 kPa/min until the desired pressure of 405 kPa was reached. The Xow of gas through the cage was controlled by two needle valves. A small portion of the outgoing gas was directed out of the pressure chamber (controlled by another needle valve), passed through a Xowmeter, and was sampled by an O2 analyzer (Servomex, Crowborough, Sussex, UK) which monitored the concentration of O2 in the experimental cage throughout the exposure. The temperature in the cage was maintained within a thermoneutral range (28 § 1°C) to avoid any eVect of temperature on CNS-OT (Arieli and Ertracht 1999). When the desired pressure of 405 kPa was reached, a period of 20 min was allowed for acclimation to the experimental conditions, during which air Xowed through the cage at approximately 8 l/min. At the end of the acclimation period, the Xow of air was immediately replaced by pure
Eur J Appl Physiol (2012) 112:2549–2556
oxygen at a similar Xow rate for fast replacement of the cage’s atmosphere. When the oxygen level reached 95%, oxygen Xow into the experimental cage was reduced to 100 ml/min. The exposure was terminated immediately on appearance of the Wrst convulsions, as determined by the observer. The latency was noted, and decompression was commenced at 180 kPa/min. On removal from the pressure chamber at the end of the exposure, the mouse was immediately anesthetized with xylazine (20 mg/ml) and ketamine HCl (50 mg/ml) and killed. Blood and brains were removed from both the transgenic and WT mice as soon as possible after killing (2 § 1 min). The brains were washed in 10 mM phosphate buVer saline (PBS, pH 7.4), placed in liquid nitrogen, and stored at ¡80°C for later analysis. For the activity of G6PD, GPx and catalase, whole blood was diluted 1:1 with 5% glucose and centrifuged at 16,300£g for 15 min at a temperature of 4°C. RBC in the pellet were hemolyzed by the addition of four volumes of double distilled water, and after stirring were incubated on ice for 20 min followed by centrifugation at 16,300£g for 15 min at a temperature of 4°C. The supernatant was divided into aliquots and stored at ¡80°C for subsequent analysis. Serum was produced for analysis of total glutathione (GST), GSH, GSSG and haptoglobin levels. Biochemical assays Assessment of antioxidant enzyme activity G6PD activity was assayed spectrophotometrically in the 15 G6PDdef and 11 WT mice using a microplate reader (SpectraMax 190, Molecular Devices, Sunnyvale, CA, USA) in 96-well microtiter plates at 30°C, by monitoring the reduction of NADP+ at 340 nm according to the procedure reported by Ben-Bassat and Goldberg (1980). The reaction mixture contained 4 mM glucose 6-phosphate, 40 mM Tris–HCl (pH 8.2), 5 mM MgCl2, 0.3 mM NADP+, and 2–8 g of sample protein in a total volume of 200 l. One unit of enzyme activity was deWned as the amount of enzyme that catalyzes the reduction of 1 mol NADP+ per min. GPx activity was assayed spectrophotometrically in the 15 G6PDdef and 11 WT mice using a microplate reader (Spectramax 190, Molecular Devices, Sunnyvale, CA, USA) in 96-well microtiter plates at 30°C, by monitoring the oxidation of NADPH at 340 nm according to the procedure reported by Flohé and Günzler (1984). The reaction mixture contained 0.15 mM NADPH, 50 mM potassium phosphate buVer (pH 7), 1 mM EDTA, 1 mM GSH, 0.25 units glutathione reductase, 1.2 mM t-butyl-hydroxyperoxidase and 5–20 g of sample protein in a total volume of 200 l. One unit of enzyme activity was deWned as the
2551
amount of enzyme that catalyzes the oxidation of 1 mol of NADPH per min. Catalase activity was assayed polarographically in 8 of the G6PDdef and 6 of the WT mice using a Clark-type oxygen electrode (Yellow Springs Instrument Co., Yellow Springs, OH, USA), by monitoring the initial linear rate of oxygen production from H2O2 according to the procedure reported by Goldberg and Hochman (1989). The assay was performed at 30°C in a reaction mixture containing 100 mM potassium phosphate buVer (pH 6.3) and 20 mM H2O2. One unit of enzyme activity was deWned as the amount of enzyme that catalyzes the decomposition of 1 mol of H2O2/min at an initial concentration of 20 mM H 2O 2 . GST, GSH and GSSG levels Serum levels of GST, GSH and GSSG were determined in six of the G6PDdef and Wve of the WT mice using a commercially available kit (BioVision, Mountain View, CA, USA). Haptoglobin measurement To ensure that no hemolytic process occurred following exposure to HBO, we measured the level of haptoglobin in serum from blood taken at the end of the HBO exposure. Levels of haptoglobin in serum were assessed with the Mouse Haptoglobin ELISA Kit (Immunology Consultants Laboratory, Newberg, OR, USA) in six of the G6PDdef and Wve of the WT mice. Western blot analysis of eNOS, nNOS and NT This analysis was conducted on Wve of the G6PDdef and Wve of the WT mice. The brain of each animal was thawed and homogenized with SDS buVer (20% glycerol and 6% SDS in 0.12 M Tris buVer with a pH of 6.8), centrifuged at 13,000£g for 20 min at a temperature of 4°C, and boiled for 10 min. The protein concentration of the brain specimens was quantiWed by the Bradford method (Bio-Rad Laboratories, Richmond, CA, USA). Fifty micrograms of total protein was loaded in each gel well. After blotting, the membranes were incubated for 1 h in blocking solution containing 5% skimmed milk in tris-buVered saline Tween20 (TBST). The membranes were then washed brieXy in TBST and incubated overnight at 4°C with the polyclonal IgG cross-reactive to the antibodies of eNOS and nNOS (Cell Signaling Technology, Beverly, MA, USA) and NT (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1:1,000. After three repeated washings in TBST, the membranes were incubated at room temperature for 1 h with horseradish peroxidase-conjugated goat anti-rabbit IgG in a
123
2552
Eur J Appl Physiol (2012) 112:2549–2556
1:2,000 dilution (Cell Signaling Technology, Beverly, MA, USA). The membranes were then developed to enhance detection by chemiluminescence (Pierce Biotechnology, Inc., Rockford, IL, USA) and exposed to X-ray Wlm. Levels of eNOS, nNOS and NT were measured by scanning the Wlms with a laser densitometer (Image Master VDS-CL, Amersham Pharmacia Biotech, Uppsala, Sweden). Brain homogenate from a speciWc WT animal was used as a sample reference.
a
WT
G6PDdef
269 bp
214 bp
Results G6PD genotyping and activity To determine the eVect of G6PD deWciency on susceptibility to CNS-OT, we used G6PDdef mice. These mice carry a mutation at the 5⬘ untranslated sequence of the X-linked G6PD gene, and through a splicing defect exhibit decreased G6PD expression (Sanders et al. 1997). G6PDdef mice and WT genotypes were veriWed using PCR analysis (Fig. 1a). G6PD activity was 16% in G6PDdef mice compared with the WT (0.09 § 0.03 and 0.56 § 0.15 units/mg Hgb/min, respectively, P < 0.001; Fig. 1b). Since there was a similar degree of G6PD activity in the RBC of male hemizygous and female homozygous G6PDdef mice, they were grouped together and compared with WT littermate controls. Latency to CNS-OT; GST, GSH and GSSG concentrations; glutathione peroxidase and catalase activity To determine whether G6PD deWciency aVects latency to CNS-OT, we measured the time to appearance of the Wrst convulsions by observation of the animals during the exposure. As shown in Fig. 2, latency to CNS-OT was found to be more than twice as long in mutant mice compared with the WT (P < 0.005). GST, GSH, and GSSG were measured in the RBC. These cellular thiols are sensitive markers of the cellular redox state. No signiWcant diVerences were found in GST, GSH or GSSG levels, indicating that there was no oxidative deWcit in the G6PDdef mice compared with the WT animals after HBO exposure. GST levels were 1.21 § 0.27 and 1.17 § 0.27 g/well, GSH levels 1.01 § 0.28 and 1.11 § 0.30 g/well, and GSSG levels 0.05 § 0.01 and 0.05 § 0.02 g/well, in the G6PDdef mice (n = 6) and the WT (n = 5), respectively.
123
140 120 100 80 60 40
*
20 0 WT
G6PDdef
Fig. 1 a PCR products after Ddel enzyme restriction. b G6PD activity in G6PDdef (n = 15) and WT mice (n = 11). G6PD activity in G6PDdef mice was 16% compared with the WT (*P < 0.005). Data are expressed as mean § SD
Laten ncy to C NS oxygen toxicity ( min)
Statistical signiWcance was evaluated by the Student t test. All data are expressed as mean § SD. The level of signiWcance was P < 0.05.
b G6PD acctivity (% of o WT)
Statistical analysis
60
* 50 40 30 20 10 0 WT
G6PDdef
Fig. 2 CNS-OT in the G6PDdef (n = 15) and WT mice (n = 11). Latency to CNS-OT after compression to 405 kPa was found to be more than twice as long in G6PDdef mice compared with the WT (*P < 0.005). Data are expressed as mean § SD
Because lower G6PD activity may reduce the antioxidant capacity of the G6PDdef mice, we checked for a compensation eVect by an increase in the activity of the antioxidant enzymes GPx and catalase. However, there were no signiWcant diVerences in the activity of these two enzymes in RBC. GPx was 6.9 § 1.7 versus 6.5 § 1.1 and
Eur J Appl Physiol (2012) 112:2549–2556 WT
2553
G6PDdef
250 kDa 150 kDa 100 kDa 75 kDa
1.2 1
50 kDa
0.8
37 kDa
*
0.6
Fig. 4 Representative Western blot of cerebral homogenates stained with polyclonal antibody reactive against nitrotyrosine. Two major bands of molecular weight (75 kDa and »65 kDa) were found
0.4 0.2 0 WT
WT
G6PDdef
Fig. 3 nNOS levels in the brain tissue of G6PDdef (n = 5) and WT mice (n = 5). Results are compared with WT mice for each trial, and are also shown as representative bands produced by Western blot analysis. nNOS levels were found to be signiWcantly lower in the G6PDdef mice compared with the WT (*P < 0.01). Data are expressed as mean § SD
catalase 0.27 § 0.03 versus 0.25 § 0.07 units/mg Hgb/min in G6PDdef and WT mice, respectively. Haptoglobin measurement No signiWcant diVerences were found between the G6PDdef and WT mice (not shown). eNOS, nNOS, and 3-NT immunoblotting A lower level of nNOS (50.6%) was found in the brain tissue of G6PDdef mice compared with the WT (P < 0.01; Fig. 3). However, there were no diVerences in eNOS levels between WT and G6PDdef animals (not shown). Peroxynitrite, formed by the interaction of superoxide and nitric oxide, is a strongly reactive oxidant. It was shown to interact with proteins and cause nitrosative modiWcation of tyrosine residues to form NT. We used Western blot analysis to measure NT levels, and found two major bands of molecular weight (75 kDa and »65 kDa) (Fig. 4). No signiWcant diVerence was found between WT and G6PDdef mice at 75 kDa, but at 65 kDa a lower level of NT (52.8%) was found in the brain tissue of G6PDdef mice compared with the WT (P < 0.05; Fig. 5).
Discussion In the present study, we demonstrate the novel Wnding that G6PD deWciency with a level of activity 16% of the WT is most likely not a risk factor for the development of CNSOT during exposure to hyperbaric oxygen, although this requires to be veriWed for human subjects. We showed that the latency to convulsions was twice as long in
G6PDdef ~65kDa
65 kDa protein nitration level (simple/ marker pixel ratio)
nNO S Level (sample / marker pixel ratio) r
160kD
2.5 2 1.5
*
1 0.5 0
WT
G6PDdef
Fig. 5 Nitrotyrosine levels in the brain tissue of G6PDdef (n = 7) and WT mice (n = 8). Results are compared with WT mice for each trial, and are also shown as representative bands produced by Western blot analysis. Nitrotyrosine levels were found to be signiWcantly lower in the G6PDdef mice compared with the WT (*P < 0.05). Data are expressed as mean § SD
G6PDdef mice compared with the WT, suggesting that G6PD deWciency may in fact provide protection against CNS-OT. NADPH, a product of the activity of the enzyme G6PD, is a major cellular reductant. G6PD deWciency is characterized by decreased concentrations of NADPH, reduced glutathione, and high levels of markers of lipid peroxidation (Jain et al. 2004; Nicol et al. 2000; Stockham et al. 1994; Xu et al. 2010), as well as increased accumulation of cellular ROS and enhanced susceptibility to oxidative stress (PandolW et al. 1995). Hence, it is generally accepted that this enhanced oxidative stress and oxidant sensitivity is a direct consequence of the decrease in GSH that results from the decline in NADPH, which serves as a substrate for the activity of glutathione reductase. Since exposure to hyperbaric oxygen induces oxidative stress (Demchenko et al. 2001, 2003), it is expected that G6PD deWciency will enhance susceptibility to the resulting CNS-OT not only in experimental animals but also in divers breathing a high partial pressure of oxygen or patients in the hyperbaric chamber. However, we showed that latency to convulsions was twice as long in G6PDdef mice compared with the WT, suggesting that G6PD deWciency at the level of the
123
2554
animals in the present study (16% of the WT) may in fact provide protection against CNS-OT. In explanation of our Wndings, we propose that NADPHassociated cellular activity other than the reduction of GSH may be related to another cellular function of NADPH, which serves as a substrate for NADPH oxidase. This enzyme catalyzes the production of superoxide by oxidation of NADPH. Hence, G6PD deWciency may reduce oxidative stress, which may explain the prolongation of latency to CNS-OT. In addition, a reduction in NADPH oxidase activity in the G6PDdef mice may decrease the production of peroxynitrite (ONOO¡), which is generated by superoxide and NO, and this might also help improve resistance to CNS-OT. In the present study, we found downregulation of nNOS. To establish grounds for our suggestion that a reduction in nNOS decreases the toxic eVect of HBO, we measured the level of the protein tyrosine residues (3-nitrotyrosine, which serves as a marker for peroxynitrite). We found that NT levels were also lower in the brain of G6PDdef mice compared with the WT. A number of studies have shown that NO plays an important role in the development of CNS-OT. NO is Wrst and foremost a powerful vasodilator, leading to increased cerebral blood Xow (Allen et al. 2009; Demchenko et al. 2001, 2005). NO is highly diVusible, and may also increase cytotoxicity by promoting the formation of peroxynitrite, a potent oxidant that induces oxidation of cellular targets such as nucleic acids, lipids, and proteins (Beckman and Koppenol 1996). Nitrotyrosine has been used as a peroxynitrite marker for a number of pathologies, such as the development of hyperbaric seizures, traumatic brain injury, and neurodegenerative disorders (Chavko et al. 2003; Mésenge et al. 1998; Torreilles et al. 1999). Chavko et al. (2003) found that a speciWc nNOS inhibitor reduced the increases in seizure-induced protein nitrotyrosine and protein carbonyl, and signiWcantly postponed the onset of HBO convulsions. This supports our results indicating a correlation between prolonged latency to CNS oxygen toxicity and a reduction of nNOS and protein NT in the brain of G6PDdef mice. Other studies have also demonstrated a connection between HBO-induced convulsions and NOS. Atochin et al. (2003) showed that late HBO-induced vasodilation depends upon both eNOS- and nNOS-derived NO. Demchenko et al. (2003) found that latency to convulsions was signiWcantly longer in nNOS and eNOS knockout mice compared with the WT strain. The authors suggest that eNOS-derived NO may be involved in cerebral vasoconstriction, whereas nNOS-derived NO may mediate the toxic eVect of HBO, mainly by its reaction with superoxide radicals to generate peroxynitrite. In the present study, the reduced level of nNOS, but not eNOS, in the brain tissue of
123
Eur J Appl Physiol (2012) 112:2549–2556
G6PDdef mice suggests that their resistance to CNS-OT might be due more to a reduction in the toxic eVect of HBO rather than to attenuation of the cerebral vasodilatation that occurs prior to the appearance of convulsions. The contention that G6PD deWciency might be of beneWt in coping with oxidative stress is supported by additional studies, some of which found an association with changes in the level of NO or NOS (Leopold et al. 2001, 2003; Matsui et al. 2006; Tsai et al. 1998). Tsai et al. (1998) found that G6PD deWciency causes a decrease in the production of NO, superoxide and H2O2 in granulocytes. It was shown that increased G6PD activity stimulated oxidative stress in adipocytes by upregulation of NADPH oxidase and iNOS (Park et al. 2006), whereas G6PD deWciency reduced vascular superoxide in mice (Matsui et al. 2006). Gupte et al. (2007) compared heart tissue taken from healthy subjects with that taken from subjects who suVered from congestive heart failure. They found a signiWcant positive correlation between G6PD activity, NADPH oxidase activity, and the oxidative stress level. Gupte (2008) later suggested that these Wndings warrant the research and development of new G6PD inhibitors to deal more eVectively with heart failure, vascular dysfunction, and pulmonary hypertension. G6PD deWciency was also reported to have a protective eVect against coronary heart disease (Meloni et al. 2008). The authors suggested that this may be related to downregulation of NADPH oxidase activity, which leads to a reduction in superoxide. Our Wnding that G6PD deWciency is probably not a risk factor for the development of CNS-OT following treatment by hyperbaric oxygen may have explanations that are independent of the “antioxidant” function of NADPH, mediated by GSH. Indeed, a number of studies have questioned the hypothesis that the cellular eVects of G6PD deWciency are mediated by GSH. For example, Kelman et al. (1982) demonstrated that primaquine toxicity, which is enhanced in G6PDdef individuals, was independent of GSH status. The action of glutathione peroxidase as an H2O2 scavenger was shown to be shared equally by catalase (Gaetani et al. 1989). Our original hypothesis was that the possible decrease in cellular activity of GPx in the G6PDdef mice, due to an alleged lack of GSH, would increase the activity of the H2O2 scavenging enzymes. However, we found that the in vitro activity of both GPx and catalase was similar in the G6PDdef mice and the WT, and found no reduction in GSH levels in the G6PDdef mice compared with the WT. It was shown in yet further studies that catalase activity in the RBC of G6PDdef human volunteers was 80% of the WT normal (Nikolaidis et al. 2006; Scott et al. 1993). It was suggested that the decline in catalase activity was due to lower cellular levels of NADPH, which functions as a catalase activator (Kirkman and Gaetani 1984). Similarly, it may be that in our study there was an increase in the protein
Eur J Appl Physiol (2012) 112:2549–2556
of catalase, but this was not expressed in elevated activity due to depletion of NADPH. In conclusion, we have demonstrated for the Wrst time that G6PD deWciency at the level of the animals in the present study (16% of the WT) is probably not a risk factor for CNS-OT. Moreover, G6PD deWciency may even have a protective eVect against CNS-OT, in that seizure latency was found to be more than twice as long in G6PDdef mice compared with the WT; this resistance was associated with a decrease in nNOS and NT levels in brain tissue. Acknowledgments This study was supported by a research grant from the IDF Medical Corps and the Israel MOD. The authors thank Mr. R. Lincoln for skillful editing. The opinions and assertions contained herein are the private ones of the authors, and are not to be construed as oYcial or as reXecting the views of the Israel Naval Medical Institute.
References Allen BW, Demchenko IT, Piantadosi CA (2009) Two faces of nitric oxide: implications for cellular mechanisms of oxygen toxicity. J Appl Physiol 106:662–667 Arieli R, Ertracht O (1999) Latency to CNS oxygen toxicity in rats as a function of PCO2 and PO2. Eur J Appl Physiol 80:598–603 Arieli R, Arieli Y, Daskalovic Y, Eynan M, Abramovich A (2006) CNS oxygen toxicity in closed-circuit diving: signs and symptoms before loss of consciousness. Aviat Space Environ Med 77:1153–1157 Atochin DN, Demchenko IT, Astern J, Boso AE, Piantadosi CA, Huang PL (2003) Contributions of endothelial and neuronal nitric oxide synthases to cerebrovascular responses to hyperoxia. J Cereb Blood Flow Metab 23:1219–1226 Beckman JS, Koppenol WH (1996) Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and the ugly. Am J Physiol 271:C1424–C1437 Ben-Bassat A, Goldberg I (1980) PuriWcation and properties of glucose-6-phosphate dehydrogenase (NADP+/NAD+) and 6-phosphogluconate dehydrogenase (NADP+/NAD+) from methanolgrown Pseudomonas C. Biochim Biophys Acta 611:1–10 Beutler E (1994) G6PD deWciency. Blood 84:3613–3636 Bitterman N, Bitterman H (1998) L-Arginine-NO pathway and CNS oxygen toxicity. J Appl Physiol 84:1633–1638 Cappellini MD, Fiorelli G (2008) Glucose-6-phosphate dehydrogenase deWciency. Lancet 371:64–74 Chavko M, Auker CR, McCarron RM (2003) Relationship between protein nitration and oxidation and development of hyperoxic seizures. Nitric Oxide 9:18–23 Demchenko IT, Boso AE, Whorton AR, Piantadosi CA (2001) Nitric oxide production is enhanced in rat brain before oxygen-induced convulsions. Brain Res 917:253–261 Demchenko IT, Atochin DN, Boso AE, Astern J, Huang PL, Piantadosi CA (2003) Oxygen seizure latency and peroxynitrite formation in mice lacking neuronal or endothelial nitric oxide synthases. Neurosci Lett 344:53–56 Demchenko IT, Luchakov YI, Moskvin AN, Gutsaeva DR, Allen BW, Thalmann ED, Piantadosi CA (2005) Cerebral blood Xow and brain oxygenation in rats breathing oxygen under pressure. J Cereb Blood Flow Metab 25:1288–1300 Elayan IM, Axley MJ, Prasad PV, Ahlers ST, Auker CR (2000) EVect of hyperbaric oxygen treatment on nitric oxide and oxygen free radicals in rat brain. J Neurophysiol 83:2022–2029
2555 Flohé L, Günzler WA (1984) Assays of glutathione peroxidase. Methods Enzymol 105:114–121 Frank JE (2005) Diagnosis and management of G6PD deWciency. Am Fam Physician 72:1277–1282 Gaetani GF, Galiano S, Canepa L, Ferraris AM, Kirkman HN (1989) Catalase and glutathione peroxidase are equally active in detoxiWcation of hydrogen peroxide in human erythrocytes. Blood 73:334–339 Goldberg I, Hochman A (1989) PuriWcation and characterization of a novel type of catalase from the bacterium Klebsiella pneumoniae. Biochim Biophys Acta 991:330–336 Gupte SA (2008) Glucose-6-phosphate dehydrogenase: a novel therapeutic target in cardiovascular diseases. Curr Opin Investig Drugs 9:993–1000 Gupte RS, Vijay V, Marks B, Levine RJ, Sabbah HN, Wolin MS, Recchia FA, Gupte SA (2007) Upregulation of glucose-6-phosphate dehydrogenase and NAD(P)H oxidase activity increases oxidative stress in failing human heart. J Card Fail 13:497–506 Jain M, Cui L, Brenner DA, Wang B, Handy DE, Leopold JA, Loscalzo J, Apstein CS, Liao R (2004) Increased myocardial dysfunction after ischemia-reperfusion in mice lacking glucose-6phosphate dehydrogenase. Circulation 109:898–903 Kelman SN, Sullivan SG, Stern A (1982) Primaquine-mediated oxidative metabolism in the human red cell. Lack of dependence on oxyhemoglobin, H2O2 formation, or glutathione turnover. Biochem Pharmacol 31:2409–2414 Kirkman HN, Gaetani GF (1984) Catalase: a tetrameric enzyme with four tightly bound molecules of NADPH. Proc Natl Acad Sci USA 81:4343–4347 Leopold JA, Cap A, Scribner AW, Stanton RC, Loscalzo J (2001) Glucose-6-phosphate dehydrogenase deWciency promotes endothelial oxidant stress and decreases endothelial nitric oxide bioavailability. FASEB J 15:1771–1773 Leopold JA, Walker J, Scribner AW, Voetsch B, Zhang YY, Loscalzo AJ, Stanton RC, Loscalzo J (2003) Glucose-6-phosphate dehydrogenase modulates vascular endothelial growth factor-mediated angiogenesis. J Biol Chem 278:32100–32106 Luzzatto L, Mehta A (1995) Glucose-6-phosphate dehydrogenase deWciency. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds) The metabolic and molecular bases of inherited disease, vol 3, 7th edn edn. McGraw Hill, New York, pp 3367–3398 Matsui R, Xu S, Maitland KA, Mastroianni R, Leopold JA, Handy DE, Loscalzo J, Cohen RA (2006) Glucose-6-phosphate dehydrogenase deWciency decreases vascular superoxide and atherosclerotic lesions in apolipoprotein E¡/¡ mice. Arterioscler Thromb Vasc Biol 26:910–916 Meloni L, Manca MR, Loddo I, Cioglia G, Cocco P, Schwartz A, Muntoni S, Muntoni S (2008) Glucose-6-phosphate dehydrogenase deWciency protects against coronary heart disease. J Inherit Metab Dis 31:412–417 Mésenge C, Charriaut-Marlangue C, Verrecchia C, Allix M, Boulu RR, Plotkine M (1998) Reduction of tyrosine nitration after Nnitro-L-arginine-methylester treatment of mice with traumatic brain injury. Eur J Pharmacol 353:53–57 Nicol CJ, Zielenski J, Tsui LC, Wells PG (2000) An embryoprotective role for glucose-6-phosphate dehydrogenase in developmental oxidative stress and chemical teratogenesis. FASEB J 14:111– 127 Nikolaidis MG, Jamurtas AZ, Paschalis V, Kostaropoulos IA, KladiSkandali A, Balamitsi V, Koutedakis Y, Kouretas D (2006) Exercise-induced oxidative stress in G6PD-deWcient individuals. Med Sci Sports Exerc 38:1443–1450 Ohtsuki T, Matsumoto M, Kuwabara K, Kitagawa K, Suzuki K, Taniguchi N, Kamada T (1992) InXuence of oxidative stress on induced tolerance to ischemia in gerbil hippocampal neurons. Brain Res 599:246–252
123
2556 Oppenheim A, Jury CL, Rund D, Vulliamy TJ, Luzzatto L (1993) G6PD Mediterranean accounts for the high prevalence of G6PD deWciency in Kurdish Jews. Hum Genet 91:293–294 Oury TD, Ho YS, Piantadosi CA, Crapo JD (1992) Extracellular superoxide dismutase, nitric oxide, and central nervous system O2 toxicity. Proc Natl Acad Sci USA 89:9715–9719 Oury TD, Crapo JD, Piantadosi CA (1994) The nitric oxide synthesis inhibitor N--nitro-L-arginine decreases mortality and delays seizure onset in mice exposed to hyperbaric oxygen. In: Bennett PB, Marquis RE (eds) Basic and applied high pressure biology. University of Rochester Press, Rochester, pp 437–444 PandolW PP, Sonati F, Rivi R, Mason P, Grosveld F, Luzzatto L (1995) Targeted disruption of the housekeeping gene encoding glucose 6-phosphate dehydrogenase (G6PD): G6PD is dispensable for pentose synthesis but essential for defense against oxidative stress. EMBO J 14:5209–5215 Park J, Choe SS, Choi AH, Kim KH, Yoon MJ, Suganami T, Ogawa Y, Kim JB (2006) Increase in glucose-6-phosphate dehydrogenase in adipocytes stimulates oxidative stress and inXammatory signals. Diabetes 55:2939–2949 Pretsch W, Charles DJ, Merkle S (1988) X-linked glucose-6-phosphate dehydrogenase deWciency in Mus musculus. Biochem Genet 26:89–103 Sanders S, Smith DP, Thomas GA, Williams ED (1997) A glucose-6phosphate dehydrogenase (G6PD) splice site consensus sequence
123
Eur J Appl Physiol (2012) 112:2549–2556 mutation associated with G6PD enzyme deWciency. Mutat Res 374:79–87 Scott MD, Wagner TC, Chiu DT-Y (1993) Decreased catalase activity is the underlying mechanism of oxidant susceptibility in glucose6-phosphate dehydrogenase-deWcient erythrocytes. Biochim Biophys Acta 1181:163–168 Stockham SL, Harvey JW, Kinden DA (1994) Equine glucose-6-phosphate dehydrogenase deWciency. Vet Pathol 31:518–527 Torbati D, Church DF, Keller JM, Pryor WA (1992) Free radical generation in the brain precedes hyperbaric oxygen-induced convulsions. Free Radic Biol Med 13:101–106 Torreilles F, Salman-Tabcheh S, Guérin M-C, Torreilles J (1999) Neurodegenerative disorders: the role of peroxynitrite. Brain Res Brain Res Rev 30:153–163 Tsai K-J, Hung I-J, Chow CK, Stern A, Chao SS, Chiu DT-Y (1998) Impaired production of nitric oxide, superoxide, and hydrogen peroxide in glucose 6-phosphate-dehydrogenase-deWcient granulocytes. FEBS Lett 436:411–414 Xu Y, Zhang Z, Hu J, Stillman IE, Leopold JA, Handy DE, Loscalzo J, Stanton RC (2010) Glucose-6-phosphate dehydrogenase-deWcient mice have increased renal oxidative stress and increased albuminuria. FASEB J 24:609–616
