It is well known that after blunt trauma to the skeletal muscle, intracellular metabolites are liberated into the extracellular space and thus the circulation (1, 2). Unlike specific enzymes (e.g. creatine kinase and glutamic oxalacetic transaminase, which serve as clinical markers of muscle damage), many of the liberated substances are known to exert substantial damage to other tissues when released from injured cells. In addition to ions such as potassium and phosphate, this is especially true for the skeletal muscle oxygen storage protein myoglobin (1, 3), which is capable of exerting functional damage to the kidney via its redox-active iron-containing central heme moiety (4).
By virtue of its high content of myoglobin, skeletal muscle-containing 300 mg of the total 3 to 5 g-is one of the body's iron-rich tissues (5). Because, as a consequence, substantial synthesis of iron-containing proteins must occur, an intermediate pool of chelatable iron should exist in skeletal myocytes. Chelatable or low-molecular-weight iron is complexed to low-molecular-weight organic ligands, for example, citrate, adenosine triphosphate, and other nucleotides or loosely associated to proteins or membrane lipids (6, 7), and only constitutes a small part (0.2%-3%) of the total intracellular iron. This "chelatable" (because its detection is based on the use of complex-forming ligands) or "labile" iron is, despite its physiological need, potentially capable of exerting oxidative damage by acting as a catalyst of the Fenton reaction. Thus, if chelatable iron exists in skeletal myocytes, it should be released subsequent to muscle crush trauma as with other ions. This is pathologically important because after trauma, superoxide, and thus hydrogen peroxide, the substrate for the Fenton reaction is generated at myocyte membranes (8) that is, in close vicinity to the regions where chelatable iron is released.
The aim of this study is to examine the hypothesis that severe trauma results in the release of chelatable iron from muscle cells into the local tissue and the circulation, and that this iron constitutes an oxidative hazard owing to its chemical properties. Therefore, we examined in vitro the release of chelatable iron into the supernatant of destroyed muscle tissue and characterized its ligands, redox properties, and its ability to bind to transferrin. To further explore the potential pathological significance of iron liberation from damaged muscle, we investigated whether non-transferrin-bound iron is liberated into the blood in vivo and whether parameters of oxidative damage in the blood serum and in the injured tissue itself are altered after trauma.
MATERIALS AND METHODS
Male Wistar rats (350-500 g) were obtained from the central animal unit of the Essen University Hospital. Animals were kept under standardized conditions of temperature (22 ± 1°C), humidity (55% ± 5%), and 12-h/12-h light/dark cycles with free access to food and water. All animals received humane care according the standards of the Federation of European Laboratory Animal Science Association.
Apotransferrin, holotransferrin, ferritin (from horse spleen), Chelex 100 (chelating resin), 1,10-phenanthroline, ascorbic acid, ammonium iron(II) sulfate, myoglobin (from horse skeletal muscle), 2,2′-dipyridyl (2,2′-DPD), butylated hydroxytoluene (BHT), hydrogen peroxide, and 1,1,3,3-tetramethoxypropane were obtained from Sigma-Aldrich (Taufkirchen, Germany). Phen green SK dipotassium salt (PGSK) was purchased from Molecular Probes, Inc. (Eugene, Ore), proteinase K from Roche Applied Science (Mannheim, Germany), and the Halt protease inhibitor cocktail from Pierce Biotechnology (Rockford, Ill). Triton X-100, trichloroacetic acid, 2-thiobarbituric acid, sulfosalicylic acid, nitric acid (65%), and buffer salts were obtained from Merck (Darmstadt, Germany). All chemicals were of the highest purity commercially available.
Destruction of rat Musculus gastrocnemius
Subsequent to a short perfusion with calcium- and EDTA-free Krebs-Henseleit buffer, rat gastrocnemius muscle was excised, minced into small pieces, and suspended 1:10 (w/v) in warm saline (0.9% NaCl) supplemented with 20 mM HEPES (pH 7.5; 37°C) that had been treated overnight with Chelex 100 (5 g/100 mL) to remove any contaminating iron. The buffer was left without additions or supplemented with protease inhibitors (10 μL/mL of the commercially available cocktail), myoglobin (0.315 mg/mL), or with the iron chelators 1,10-phenanthroline (1 mM) and 2,2′-DPD (1 mM), respectively, before adding the muscle tissue. The mixture was homogenized using an Ultra-Turrax homogenizer (IKA, Staufen, Germany) at 17,500 U/min for approximately 30 s until only some connective tissue was remaining, and the resulting homogenate was placed in a water bath (37°C).
After distinct times of incubation at 37°C, aliquots of the warm muscle homogenate were centrifuged at 16,000 g (10 min, 22°C) to subsequently determine chelatable iron in the resulting supernatant. In experiments studying iron binding to transferrin, additionally, sodium hydrogen carbonate (1.2 mg/mL) and apotransferrin or holotransferrin (2.5 mg/mL each) were mixed into the supernatant 5 min before the first iron detection. To stabilize pH, these samples were stored in an incubator at 5% CO2 during further treatment.
In another set of experiments, aliquots of the supernatant were left untreated or incubated with proteinase K (100 mU/mL; 30 min, 60°C) or 1% Triton X-100 (30 min, 37°C) and subsequently filtered through a 30-kDa cutoff filter (Vivascience, Hannover, Germany) at 16,000 g (10 min, 22°C). Controls were performed in buffer without muscle tissue for all variations described.
Determination of chelatable iron
The resulting samples (and buffer controls) of all previously discussed incubations were mixed with the reducing compound ascorbic acid (1 mM) for 20 min to transfer the readily accessible iron to its ferrous state. Subsequently, the samples were immediately incubated with the fluorescent iron indicator phen green SK (50 μM) for 2 min, and chelatable iron was determined as described (9). In brief, the fluorescence of the dye PGSK is quenched by chelatable iron complexed to its phenanthroline moiety. Measurements of the PGSK fluorescence in the samples in relation to PGSK-supplemented homogenization buffer were performed on a laser scanning microscope (LSM 510; Zeiss, Oberkochen, Germany) using λexc = 488 nm, λem = 505-530 nm. Because of the known 3:1 stoichiometry of the PGSK:iron(II) complex (9), the concentration of chelatable iron was calculated directly from the different PGSK fluorescence within the buffer (without any contaminant iron) and the muscle supernatant. As a second method, the amount of chelatable iron was calculated from the decrease in PGSK fluorescence subsequent to the addition of 5 μM ammonium iron(II) sulfate (from a stock solution of 1 mM ammonium iron(II) sulfate and 20 mM ascorbic acid) to the samples. This alternative method yielded the same results as using the stoichiometrical calculation method (data not shown).
Determination of the total iron content
The total iron content of the gastrocnemius muscle was determined by atomic absorption spectroscopy using a modified standard method (10). Aliquots of homogenate and supernatant (see above) were heated in the presence of proteinase K (100 mU/mL; 90 min, 60°C) and incubated overnight with 2 volumes of nitric acid (14.4 N). The iron content was determined using an atomic absorption spectrometer (M Series; Thermo Electron, Oberhausen, Germany). A muscle-free buffer sample was prepared the same way and used as a blank.
Determination of myoglobin
The myoglobin content of gastrocnemius muscle and plasma was measured using a commercially available rat myoglobin enzyme-linked immunosorbent assay kit (Life Diagnostics Inc., West Chester, Pa) after homogenization and/or centrifugation as previously discussed.
In vivo model of closed muscle trauma
Rats were anesthetized with isoflurane (1.0%-1.5% in 100% medical O2 at 0.5-1.0 L/min) through face masks connected to a vaporizer (Dräger Medical, Lübeck, Germany) and received ketamine (80 mg/kg body weight, s.c.) into the right chest wall for analgesia. After local lidocaine application (5 mg/kg body weight, s.c.), a median skin-deep incision was made along the throat and a Portex catheter placed within the left carotid artery (for monitoring of arterial blood pressure and blood sampling) and the right jugular vein (for volume substitution with sterile 0.9% NaCl solution; 5 mL/kg body weight per hour). After insertion of the catheters, no further interventions were performed for 10 min, allowing the animals to adapt. Throughout the experiment, isoflurane anesthesia was sustained, and body temperature was maintained at 37°C using a temperature-controlled heating pad.
For bilateral mechanical trauma, both legs of the rat were positioned in a retainer to avoid breakage of the bones and hind limb movement during the impact. Trauma was applied to both gastrocnemius muscles using a standardized "weight-drop" device. Muscles were impacted laterally by a 1-kg weight (impact area, 2 × 2 cm) falling from 40 cm height, thus yielding an impact energy of 3.92 J per leg. Sham controls underwent the complete surgical procedure but were not traumatized. At the end of the experiment, animals were killed by cardiac incision under deep isoflurane anesthesia. The experimental protocol was approved by the regional government based on the local animal protection act.
Two series of animal experiments were conducted to determine (A) non-transferrin-bound iron (NTBI) and (B) protein carbonyls, thiobarbituric acid-reactive substances (TBARS), myoglobin, and creatine kinase in serum and plasma, respectively. The experimental flow is summarized schematically as follows:
Before and 5, 30 min, and 1, 2, 3, and 4 h after trauma, 0.5 mL of blood was taken from the arterial catheter to obtain plasma or serum. Part of the plasma samples was analyzed immediately for creatine kinase activity, and another part was stored at −20°C until analysis of its myoglobin content. Serum was frozen at −80°C until analysis of NTBI or used immediately for the determination of serum protein carbonyls and formation of TBARS.
In each series, four animals were subjected to muscle trauma; two (A) or four (B) animals served as sham controls.
Determination of NTBI
Non-transferrin-bound iron was assessed as described earlier (11). Briefly, the thawed serum was supplemented with Tris-carbonatocobaltate(III) trihydrate and incubated at 37°C for 1 h. To capture NTBI without chelating transferrin- or ferritin-bound iron, serum samples were subsequently mixed with nitrilotriacetic acid to a final concentration of 80 mM and allowed to stand for 30 min at room temperature. The solution was then ultrafiltered using a Centricon-30 filter at 3,000 g for 1 h. The sample ultrafiltrate was analyzed for the iron-nitrilotriacetic acid complex by using high-performance liquid chromatography (Waters Bio-System 625, nonmetallic gradient module with 911 photodiode array detector). The iron levels in the ultrafiltrate were determined from a standard curve.
Determination of serum protein carbonyls
Protein carbonyls were determined in fresh serum from traumatized and sham control rats. Serum samples were 1) left untreated, 2) supplemented with 2 mM hydrogen peroxide or 3) with the iron chelator 1,10-phenanthroline (5 mM) and then with hydrogen peroxide and incubated at 37°C for 60 min. At the end of the incubation time, oxidation was stopped in all samples by adding the antioxidant BHT (40 μM). Subsequently, protein carbonyl concentrations were determined using a commercially available enzyme-linked immunosorbent assay kit (BioCell, Papatoetoe, Auckland, New Zealand).
Determination of TBARS formation
For in vitro experiments, muscle homogenates were preincubated with proteinase K (100 mU, 60°C, 30 min) or not and incubated in a water bath (37°C) under continuous shaking to ensure that the oxygen content in the suspension was not limited (A). Samples were incubated in the presence of protease inhibitor, 1,10-phenanthroline or 2,2′-DPD (see above), or left untreated. At time points indicated, 1-mL aliquots of the whole homogenate were incubated for 5 min with 0.5 mL trichloroacetic acid (30%) and subsequently heated in the presence of 2-thiobarbituric acid (23 mM; 95°C, 60 min). After centrifuging at 5,000 g (10 min, 22°C), samples were measured photometrically at 532 nm. For in vivo experiments, muscle samples were homogenized in the presence of the antioxidant BHT (40 μM) as previously described (B). Serum samples were treated as outlined under protein carbonyl determination (C). Aliquots of the respective samples (B, C) were incubated for 10 min with 0.1 vol. of 27.5% sulfosalicylic acid. Subsequently, the samples were heated in the presence of 2-thiobarbituric acid (23 mM; 95°C, 60 min). After centrifuging at 5,000 g (10 min, 22°C), samples were measured with a fluorometer (Shimadzu, Kyoto, Japan) at 525 nm for excitation and 547 nm for emission. The amount of TBARS formed was expressed as malondialdehyde equivalents using 1,1,3,3-tetramethoxypropane as a standard.
In vitro experiments were performed with homogenates obtained from at least three different animals. In vivo experiments were repeated four times. Data are expressed as mean ± SEM. Data obtained from two groups were compared by means of Student t test. Comparisons among multiple groups were performed using one-way analysis of variance with Student-Newman-Keuls comparisons. Repeated measurements were compared by two-way analysis of variance, followed by Bonferroni post hoc tests. A P value of less than 0.05 was considered significant.
In vitro experiments
In the supernatants of crushed rat gastrocnemius muscles, approximately 5 μM iron could be detected by using the fluorescent indicator PGSK (Fig. 1). This chelatable iron was released immediately during muscle homogenization and remained detectable over a period of 24 h. As determined by atomic absorption spectroscopy, the total iron concentration in the whole muscle homogenate was 17 μM; 11 μM of this iron (i.e. 2/3 of the myocyte total iron) was found in the supernatant after centrifugation. The detected 5 μM PGSK-chelatable iron thus accounted for 45% of the supernatant iron. Furthermore, 3.0 ± 0.6 μM (27%) of iron within the supernatant of crushed muscle homogenate was found to be myoglobin iron.
When the supernatant from crushed muscle was passed through a 30-kDa cutoff filter, chelatable iron was barely detectable in the filtrate using PGSK (0.4 ± 0.2 μM; Fig. 2). When the supernatant of the muscle homogenate was preincubated with Triton X-100, to disintegrate lipid structures, only 1.1 ± 0.5 μM chelatable iron was found after the subsequent 30-kDa filtration; proteinase K pretreatment-to break down proteins-however, increased the PGSK-detectable iron concentration of the filtrate to 10.4 ± 1.2 μM (Fig. 2). No difference to controls was detected when 18.5 μM myoglobin iron was added before muscle homogenization (5.5 ± 1.3 μM chelatable iron in the presence and 5.2 ± 0.8 μM in the absence of myoglobin, respectively).
As a marker of iron-induced oxidative damage, the peroxidative decomposition of polyunsaturated fatty acids was determined in the muscle homogenate. After muscle crushing, the amount of lipid peroxidation products, that is, TBARS, increased to 10.8 ± 1.9 μM during 6 h of incubation (Fig. 3). A similar kinetic pattern was obtained when the homogenate was incubated in the presence of protease inhibitors (data not shown), thus excluding the possibility that additional iron, capable of participating in TBARS formation, was released from storage proteins by proteolytic digestion during the incubation period. In contrast, preincubation of the muscle homogenate with proteinase K more than doubled TBARS formation, whereas incubation with 1 mM iron chelator solutions (1,10-phenanthroline and 2,2′-DPD, respectively), inhibited the formation of TBARS almost completely (Fig. 3).
After addition of 30 μM (iron-free) apotransferrin to the muscle supernatant, the PGSK-detectable iron concentration within the supernatant decreased nonsignificantly from 5.6 ± 0.5 (without apotransferrin) to 3.8 ± 0.3 μM (Fig. 4). Control incubations with (iron-saturated) holotransferrin yielded-as expected-an amount of 5.4 ± 0.5 μM chelatable iron, which is comparable with the results in the absence of transferrin.
In vivo Experiments
To account for the in vivo relevance of the data found under in vitro conditions, a model of bilateral trauma to the rat gastrocnemius muscle was applied. The traumatic impact caused an increase in the marker proteins of skeletal muscle damage, creatine kinase, and myoglobin within the blood plasma (Fig. 5). Plasma levels of both metabolites increased within 30 min after the induction of trauma, reached the maximum level after 60 min, and stayed elevated over the rest of the 4-h period that was monitored after trauma.
In the animal model of skeletal muscle trauma, serum samples were taken before, 5, 30 min, and 1, 2, and 4 h after injury and analyzed for serum NTBI. However, all values were less than the detection limit of the method (averaged iron levels of traumatized animals and sham controls, 0.29 and 0.30 μM; detection limit ∼0.5 μM). Consequently, no significant differences in the serum levels of NTBI were detectable.
To search for the existence of redox-active iron via its oxidative effect on lipids and proteins, serum and muscle tissue of traumatized rats were analyzed for the formation of TBARS and protein carbonyls. Serum content of TBARS was found only modestly higher in traumatized animals as in sham controls (Fig. 6A). Serum protein carbonyls likewise were not significantly elevated in serum of trauma animals in comparison to sham controls (Fig. 6B). In addition, after in vitro addition of 2 mM hydrogen peroxide (to generate the highly reactive hydroxyl radical by the iron-catalyzed Fenton reaction) to serum obtained from trauma animals 4 h after injury, no further increases in TBARS (from 57.4 ± 6.2 to 58.1 ± 4.6 nmol/g protein) and protein carbonyls (from 237 ± 20 to 257 ± 26 nmol/g protein) were found. Iron chelation by 1,10-phenanthroline accordingly did not significantly diminish this hydrogen peroxide-induced oxidation (data not shown).
In the injured muscle tissue itself, 4 h after trauma, TBARS was significantly (P = 0.002) elevated in comparison to the muscle tissue of sham controls (Fig. 6C).
In this study, we demonstrated in vitro that chelatable iron is released from disrupted muscle tissue, and that the released chelatable iron is redox-active and responsible for lipid peroxidation within the muscle homogenate. In vivo, we confirmed the occurrence of oxidative processes within the damaged muscle tissue. However, the liberation of chelatable iron from the injured muscle into the circulation and its contribution to oxidative alterations of serum lipids and proteins could not be verified.
Iron is the most abundant transition metal in the body and plays a crucial role in vital biochemical processes such as oxygen sensing and transport, electron transfer, and catalysis (12). Thus, reasonably, all tissues contain iron in diverse forms, for example, incorporated in cytochromes, myoglobin, ferritin, and as chelatable iron. Chelatable iron thus far was detected and quantified directly in intact cultured cells (9, 13-15) and-after destruction and homogenization-in organ homogenates (16-22). We here demonstrated the permanent existence of approximately 5 μM chelatable iron in 1:10 homogenates of skeletal muscle; a result well in line with the findings of Ibrahim and Chow (21) who detected approximately 40 to 50 nmol/g chelatable iron in 1:10 rat skeletal muscle homogenates but more than 15-fold higher than the 1 μM iron found in 1:3 homogenates by Jenkins et al. (22). Importantly, the chelatable iron determined in our experiments was hardly bound to low-molecular-weight ligands but most likely associated to proteins greater than 30 kDa. We, however, do neither know whether it had been associated to other ligands in the intact cells-potentially a major redistribution to proteins occurs at the moment of disruption-nor whether the amount of 5 μM iron, as present in muscle tissue homogenate, equals the intracellular pool of healthy muscle. The concentration of iron found using disrupting methods might naturally be higher than the original intracellular concentration because of the potential destruction of iron-binding proteins and other structures during tissue processing (6, 7). Unfortunately, no direct, intracellular determination of the physiological chelatable iron pool in the living muscle is possible because the PGSK method (and also other existing methods) cannot be applied in viable tissues. However, even if the amount of chelatable iron found after muscle disruption is larger than the physiological intracellular iron pool, this should also be true under in vivo trauma conditions.
The contribution of iron to oxidative damage after muscle injury thus far has been completely centered on the role of myoglobin; it was proposed that toxic chelatable iron is released from myoglobin heme after trauma or that redox cycling of heme iron of intact myoglobin displays the decisive oxidative event (4, 23). However, according to our in vitro results, the amounts of liberated myoglobin and of chelatable iron were in the same range and lipid peroxidation, that is, the formation of TBARS in the homogenate, was catalyzed exclusively by chelatable iron and not by myoglobin iron, which cannot be deactivated by iron chelators. Lipid peroxidation in skeletal muscle homogenates has already been reported (24); however, the authors claimed that not muscle iron, but iron contaminations within the used chemicals were responsible for lipid peroxidation-a source ruled out in our experiments by Chelex treatment of the buffers.
Enhanced lipid peroxidation not only occurred in skeletal muscle homogenates in vitro but could also be demonstrated in vivo. In injured rat gastrocnemius muscle, we found enhanced lipid peroxidation 4 h after the traumatic incident. Although the amount of TBARS generated was significantly higher than in muscle tissue of sham animals, its formation may have been limited by the ischemic conditions within the destructed tissue. Besides this local pro-oxidative action, iron should be liberated from the damaged muscle tissue into the circulation such as myoglobin and other intracellular metabolites (1) after traumatic injury. However, despite a considerable increase in plasma creatine kinase activity and myoglobin concentration after bilateral trauma, the concentration of myoglobin (∼2.4 nM = 40 ng/mL; Fig. 5) accounts for only a very small fraction of the myoglobin present in both muscles (which would yield to micromolar plasma levels when totally liberated). Chelatable iron emerging in the circulation should either be immediately scavenged by the iron transport protein transferrin to prevent its pro-oxidative activity (25, 26) or-if not-contribute to the serum NTBI pool. Our data demonstrate that only a minor part of the iron from crushed muscle was scavenged by transferrin, which is most probably for sterical reasons, that is, because of its association to proteins greater than 30 kDa. Therefore, under in vivo conditions, iron liberated from injured muscle tissue might be expected to expand the NTBI pool, which is known to exist also in various other pathologies (27-32). Its exact chemical nature is still unknown, but it seems to be redox-active (27, 29), chelatable by artificial ligands and composed predominantly of protein-associated iron (26, 33, 34). However, we were not able to reliably detect an increase in NTBI within the blood serum after trauma-most likely because only a very small amount of the chelatable iron was liberated into the circulation. Because we detected approximately 30 μM of myoglobin and 50 μM of PGSK-chelatable iron within the undiluted muscle homogenate and because the plasma concentration of myoglobin after bilateral trauma was approximately 2.4 nM (see above), it can be concluded that the serum concentration of chelatable iron might reaches a maximum of 4 nM after bilateral trauma in our trauma model, assuming that both myoglobin and chelatable iron are liberated to the same degree. This value is clearly less than the detection limit of the method used (and also of all other published methods), explaining the lack of obvious results in our experiments. Under these conditions, serum markers of iron-mediated oxidative damage only increased marginally. Because antioxidants are present within the blood serum, it thus seems that the iron liberated-if at all-is not sufficiently active or present to overcome the antioxidative defense of the serum. Thus, at least in our model, chelatable iron does not seem to promote systemic oxidative damage after muscle trauma; however, its participation in a more large-scale trauma remains to be elucidated. Nevertheless, under trauma conditions, the major part of oxidative stress should initially occur within the destructed tissue itself. Herein, liberated chelatable iron clearly takes part in the local oxidative tissue destruction (even more so as it catalyses the formation of the most toxic hydroxyl radical from ROS originating from other sources) and thus should contribute to the aggravation of muscle destruction after blunt trauma.
The authors thank Veronika Hiltenkamp for performing the atomic absorption spectroscopic measurements.
1. Allison RC, Bedsole DL: The other medical causes of rhabdomyolysis. Am J Med Sci
2. Mohri T, Tanaka H, Tajima G, Kajino K, Sonoi H, Hosotsubo H, Ogura H, Kuwagata Y, Shimazu T, Sugimoto H: Synergistic effects of recombinant human soluble thrombomodulin and fluid-volume resuscitation in a rat lethal crush injury model. Shock
3. Emig U, Schmidt G, Hellige G, Vetterlein F: Contribution of myoglobin
-induced increases in vascular resistance to shock decompensation in experimental crush-syndrome in anesthetized rats. Shock
4. Holt S, Moore K: Pathogenesis of renal failure in rhabdomyolysis: the role of myoglobin
. Exp Nephrol
5. Papanikolaou G, Pantopoulos K: Iron metabolism and toxicity. Toxicol Appl Pharmacol
6. Petrat F, de Groot H, Sustmann R, Rauen U: The chelatable iron
pool in living cells: a methodically defined quantity. Biol Chem
7. Kakhlon O, Cabantchik ZI: The labile iron pool: characterization, measurement, and participation in cellular processes. Free Radic Biol Med
8. Kerkweg U, Petrat F, Korth HG, de Groot H: Disruption of skeletal myocytes initiates superoxide release: contribution of NADPH oxidase. Shock
9. Petrat F, de Groot H, Rauen U: Determination of the chelatable iron
pool of single intact cells by laser scanning microscopy. Arch Biochem Biophys
10. Alexander NM: Iron. In: Seiler HG, Sigel A, Sigel H, (eds.): Handbook on Metals in Clinical and Analytical Chemistry
. New York, NY: Marcel Dekker, Inc, 411-421, 1994.
11. Gosriwatana I, Loreal O, Lu S, Brissot P, Porter J, Hider RC: Quantification of non-transferrin
-bound iron in the presence of unsaturated transferrin
. Anal Biochem
12. Aisen P, Enns C, Wessling-Resnick M: Chemistry and biology of eukaryotic iron metabolism. Int J Biochem Cell Biol
13. Epsztejn S, Kakhlon O, Glickstein H, Breuer W, Cabantchik I: Fluorescence analysis of the labile iron pool of mammalian cells. Anal Biochem
14. Lipinski P, Drapier JC, Oliveira L, Retmanska H, Sochanowicz B, Kruszewski M: Intracellular iron status as a hallmark of mammalian cell susceptibility to oxidative stress: a study of L5178Y mouse lymphoma cell lines differentially sensitive to H2
15. Ramachandran A, Ceaser E, Darley-Usmar VM: Chronic exposure to nitric oxide alters the free iron pool in endothelial cells: role of mitochondrial respiratory complexes and heat shock proteins. Proc Natl Acad Sci U S A
16. Al-Mehdi AB, Zhao G, Tozawa K, Fisher AB: Depolarization-associated iron release with abrupt reduction in pulmonary endothelial shear stress in situ. Antioxid Redox Signal
17. Kozlov AV, Yegorov D, Vladimirov YA, Azizova OA: Intracellular free iron in liver tissue and liver homogenate: studies with electron paramagnetic resonance on the formation of paramagnetic complexes with desferal and nitric oxide. Free Radic Biol Med
18. Sohal RS, Wennberg-Kirch E, Jaiswal K, Kwong LK, Forster MJ: Effect of age and caloric restriction on bleomycin-chelatable and nonheme iron in different tissues of C57BL/6 mice. Free Radic Biol Med
19. Bralet J, Schreiber L, Bouvier C: Effect of acidosis and anoxia on iron delocalization from brain homogenates. Biochem Pharmacol
20. Miura T, Muraoka S, Fujimoto Y: Temperature-dependent lipid peroxidation of rat brain homogenate. Res Commun Mol Pathol Pharmacol
21. Ibrahim W, Chow CK: Dietary vitamin E reduces labile iron in rat tissues. J Biochem Mol Toxicol
22. Jenkins RR, Krause K, Schofield LS: Influence of exercise on clearance of oxidant stress products and loosely bound iron. Med Sci Sports Exerc
23. Reeder BJ, Wilson MT: Hemoglobin and myoglobin
associated oxidative stress: from molecular mechanisms to disease states. Curr Med Chem
24. Jackson MJ, Jones DA, Edwards RH: Lipid peroxidation of skeletal muscle: an in vitro study. Biosci Rep
25. Halliwell B, Gutteridge JM: The antioxidants of human extracellular fluids. Arch Biochem Biophys
26. Anderson GJ: Non-transferrin
-bound iron and cellular toxicity. J Gastroenterol Hepatol
27. Esposito BP, Breuer W, Sirankapracha P, Pootrakul P, Hershko C, Cabantchik ZI: Labile plasma iron in iron overload: redox activity and susceptibility to chelation. Blood
28. al-Refaie FN, Wickens DG, Wonke B, Kontoghiorghes GJ, Hoffbrand AV: Serum non-transferrin
-bound iron in beta-thalassaemia major patients treated with desferrioxamine and L1. Br J Haematol
29. Le Lan C, Loréal O, Cohen T, Ropert M, Glickstein H, Lainé F, Pouchard M, Deugnier Y, Le Treut A, Breuer W, et al.: Redox active plasma iron in C282Y/C282Y hemochromatosis. Blood
30. de Valk B, Addicks MA, Gosriwatana I, Lu S, Hider RC, Marx JJ: Non-transferrin
-bound iron is present in serum of hereditary haemochromatosis heterozygotes. Eur J Clin Invest
31. Buonocore G, Perrone S, Longini M, Paffetti P, Vezzosi P, Gatti MG, Bracci R: Non protein bound iron as early predictive marker of neonatal brain damage. Brain
32. Gutteridge JM, Quinlan GJ, Evans TW: Transient iron overload with bleomycin detectable iron in the plasma of patients with adult respiratory distress syndrome. Thorax
33. Hider RC: Nature of nontransferrin-bound iron. Eur J Clin Invest
34. Evans RW, Rafique R, Zarea A, Rapisarda C, Cammack R, Evans PJ, Porter JB, Hider RC: Nature of non-transferrin
-bound iron: studies on iron citrate complexes and thalassemic sera. J Biol Inorg Chem