Neurochem Res DOI 10.1007/s11064-014-1256-8
ORIGINAL PAPER
Phenanthrolines Protect Astrocytes from Hemin Without Chelating Iron Jessica E. Owen • Glenda M. Bishop Stephen R. Robinson
•
Received: 16 December 2013 / Revised: 2 February 2014 / Accepted: 5 February 2014 Ó Springer Science+Business Media New York 2014
Abstract Hemin, the degradation product of hemoglobin, contributes to the neurodegeneration that occurs in the weeks following a hemorrhagic stroke. The breakdown of hemin in cells releases redox-active iron that can facilitate the production of toxic hydroxyl radicals. The present study used 3-week old primary cultures of mouse astrocytes to compare the toxicity of 33 lM hemin in the presence of the iron chelator 1,10-phenanthroline or its nonchelating analogue, 4,7-phenanthroline. This concentration of hemin killed approximately 75 % of astrocytes within 24 h. Both isoforms of phenanthroline significantly decreased the toxicity of hemin, with the non-chelating analogue providing complete protection at concentrations of 33 lM and above. The decrease in toxicity was associated with less cellular accumulation of hemin. Approximately 90 % of the hemin accumulated was not degraded, irrespective of treatment condition. These observations indicate that chelatable iron is not the cause of hemin toxicity. Cell-free experiments demonstrated that hemin can inactivate a molar excess of hydrogen peroxide (H2O2), and that the rate of inactivation is halved in the presence of either isoform of phenanthroline. We conclude that phenanthrolines may protect astrocytes by limiting hemin uptake and by impairing the capacity of intact hemin to interact with endogenous H2O2. Keywords Brain Fenton reaction Heme oxygenase Hydrogen peroxide Oxidative stress
J. E. Owen G. M. Bishop S. R. Robinson (&) School of Health Sciences, RMIT University, PO Box 71, Bundoora, VIC 3083, Australia e-mail: [email protected]
Introduction Hemorrhagic strokes account for only 10–15 % of all strokes, however they are generally more severe than ischemic strokes and are associated with a higher risk of mortality [1]. It has become evident that hemin, a breakdown product of hemoglobin (in red blood cells), is responsible for much of the secondary brain damage that follows a hemorrhagic stroke [2]. The fact that neurodegeneration continues while the patient is in clinical care provides the potential to intervene with therapeutic treatments to prevent this injury [2]. Surgical removal of the hematoma is generally not performed, due to the high risk of damaging healthy brain tissue or causing further bleeding, so other forms of intervention are needed. Hemin is taken up by brain cells via the heme carrier protein-1 (HCP1) [3], or by binding to hemopexin, such that the resultant complex is endocytosed [4]. Once inside the cell, hemin has the potential to be toxic via a variety of mechanisms, including peroxidation of lipid membranes [5], production of free radicals through interaction with endogenous hydrogen peroxide (H2O2) [6] or by the enzymatic metabolism of hemin by heme oxygenase, producing free iron, carbon monoxide and biliverdin [7, 8]. The free iron that is released from hemin is presumed to be redox-active and thus able to participate in the Fenton reaction [9], producing cytotoxic hydroxyl radicals that can lead to cell death [10, 11]. The brain tightly regulates surplus iron in order to prevent it from cycling in the Fenton reaction [12]. Normally, when the levels of iron in the brain increase, ferritin is synthesized so that the excess iron can be stored as Fe3? and cannot participate in the Fenton reaction [13], but ferritin synthesis does not seem to occur when iron is released from hemin [14]. Researchers have attempted to
123
Neurochem Res
prevent hemin toxicity by inducing ferritin expression in cells, however it has been found that the upregulation of ferritin is not sufficient to protect against hemin toxicity, and instead genetic overexpression of ferritin is required [15], which is impractical in a clinical setting. Chelators that bind metal ions usually prevent them from being redox-active [16]. Desferrioxamine, a widely-used iron chelator, decreases the accumulation of free iron by cells, thereby limiting the production of reactive oxygen species and reducing the extent of neuronal death [17–19]. However, desferrioxamine does not readily permeate cell membranes, making it less effective at chelating intracellular iron [20]. An alternate iron chelator, 1,10-phenanthroline, is lipid soluble and can readily penetrate the cell membrane. A series of important studies have demonstrated that co-incubation with 1,10-phenanthroline can protect cultured neurons and astrocytes from the toxicity of hemoglobin [21, 22] and hemin [23, 24]. These results support the conclusion that the breakdown of hemin by heme oxygenase leads to the release of redox-active iron, and that chelation of this iron by 1,10-phenanthroline attenuates the toxicity. During the course of experiments in our laboratory that had been intended to build on previous findings, we used a non-chelating analogue (4,7-phenanthroline) as a negative experimental control. It was discovered that this analogue is also effective at protecting cultured astrocytes from the toxicity of hemin. The present paper describes these findings and the subsequent investigations that were undertaken into the basis of this unexpected effect. These new findings have stimulated a reassessment of the role of redox-active iron in the toxicity of hemin.
Materials and Method Primary Cell Cultures The use of animals to obtain cell cultures was approved by the Monash University Animal Ethics Committee. Primary astrocyte cell cultures were prepared from newborn C57BL/ 6JAsmu mice (\24 h old), as described previously [25]. Cells were plated in 24-well culture plates at 300,000 cells/well. All cells were incubated in culture medium [90 % Dulbecco’s modified Eagle medium (DMEM), 10 % fetal calf serum, 20 U/mL penicillin G, and 20 lg/mL streptomycin sulphate] and maintained at 37 °C in 95 % humidity with 10 % CO2. Culture media were changed every seventh day and cultures were used for experiments after 21–23 days in vitro. Incubation Protocol The culture medium was removed and the cells were washed twice with 1 mL 37 °C DMEM, after which the
123
pre-incubation solution was applied for 1 h. Pre-incubation solutions contained DMEM and 0–100 lM of 1,10-phenanthroline or 4,7-phenanthroline. After 1 h, the preincubation solution was removed and 1 mL of incubation solution was applied for 24 h. Incubation solutions contained DMEM and the same concentration of either 1,10phenanthroline or 4,7-phenanthroline that was used during the pre-incubation, plus 0 or 33 lM hemin. Hemin was dissolved as a 1 mM stock solution in 10 mM NaOH and then diluted to a final concentration of 33 lM in the incubation solution. The 0 lM hemin condition contained an identical concentration of NaOH as the 33 lM hemin condition. At the conclusion of the incubation period, media samples were collected into Eppendorf tubes and then the cells were washed twice with 1 mL of ice cold 10 mM potassium phosphate-buffered saline (K–PBS, pH 7.4) and stored at -20 °C. Cell Viability Cell viability was determined by measuring the amount of lactate dehydrogenase (LDH) released into the media during the incubation [26]. The activity of LDH was determined by measuring the rate of NADH decomposition as pyruvate was reduced to lactate. The LDH activity of samples was compared to a 100 % LDH release condition represented by untreated cells that were lysed with 1 % Triton X-100 to release all LDH present. Iron Content The total iron content of the sample was determined colorimetrically [27]. The amount of non-heme iron in the sample was determined with the same procedure with the exception that potassium permanganate was excluded from the reaction mixture [28]. As well as determining the iron content of the cells, the iron content of media samples was measured before and after the incubation. The protein content of equivalently-treated cell cultures was used to standardize iron content. Protein content was determined with the Lowry Protein Estimation assay [29]. Hydrogen Peroxide Assay To investigate the capacity of hemin, alone or in combination with phenanthroline, to react with H2O2 in a cellfree assay, incubation solutions were prepared as described above, except that they were made using pyruvate-free DMEM. In addition, H2O2 was added to each solution at a final concentration of 40 or 100 lM. Samples of each solution were collected at specific time points for up to 90 min, allowing H2O2 detoxification to be determined, as previously described [30].
Neurochem Res
protective at all higher concentrations, causing a significant decrease in LDH release (p \ 0.05). Similarly, 4,7-phenanthroline was not protective at 3.3 lM when compared to the 0 lM condition, whereas it was protective at all higher concentrations (p \ 0.05). Comparisons were made between 1,10-phenanthroline and 4,7-phenanthroline within each specific concentration, revealing that 4,7-phenanthroline provided significantly more protection against hemin toxicity than 1,10-phenanthroline at 33 lM (t(4) = 3.6, p \ 0.05) and 100 lM (t(4) = 3.8, p \ 0.05). Effect of Phenanthrolines on Cellular iron Accumulation and Metabolism of Hemin Fig. 1 Cell viability as measured by LDH release after 1 h preincubation with 0–100 lM 1,10-phenanthroline or 4,7-phenanthroline followed by 24 h incubation with 33 lM hemin plus 0–100 lM 1,10phenanthroline or 4,7-phenanthroline. *p \ 0.05 when compared to 0 lM concentration within each phenanthroline condition. ^p \ 0.05 when compared to the equivalent concentration of 1,10phenanthroline
Statistical Analysis All experiments were performed with three independent cultures, with triplicate conditions within each culture. Analyses were performed using the mean ± SD of each triplicate for each of the independent cultures, giving n = 3. Linear regression, independent t tests and planned comparison ANOVA with post hoc Dunnett’s tests were performed with significance set at p \ 0.05.
Results Effect of Phenanthrolines on Cell Viability After 24 h, cell cultures that had been incubated with 33 lM hemin had experienced considerably more cell death (75.5 ± 7.5 % LDH release) than those that had been incubated with 0 lM hemin (3.8 ± 1.3 % LDH release; p \ 0.05). When cells were incubated with either 1,10phenanthroline or 4,7-phenanthroline and 0 lM hemin, less than 7 % LDH release was observed, indicating that neither analogue of phenanthroline is toxic to astrocytes over a 24 h period. When astrocytes were co-incubated with 33 lM hemin plus up to 100 lM of either of the phenanthrolines, a dosedependent effect was observed, such that concentrations C10 lM produced a significant decrease in LDH release (Fig. 1). Linear regression analysis revealed significant effects of concentration for both 1,10-phenanthroline (r = 0.91, p \ 0.05), and 4,7-phenanthroline (r = 0.96, p \ 0.05). A one-way ANOVA showed that 1,10-phenanthroline was not protective at 3.3 lM whereas it was
Cellular iron content was measured and then standardized to the protein content of equivalently-treated cultures to account for cell death. Astrocytes incubated for 24 h with 0 lM hemin contained 3.5 ± 2.5 nmol total iron/mg protein, while cells incubated with 33 lM hemin accumulated 117.1 ± 17.9 nmol total iron/mg protein (Fig. 2a). Linear regression analysis was used to compare the concentration of total iron in astrocytes that had been incubated with hemin plus 0–100 lM of either phenanthroline. A significant decrease in iron accumulation was found with increasing concentrations of both 1,10-phenanthroline (r = 0.71, p \ 0.05), and 4,7-phenanthroline (r = 0.89, p \ 0.05). A one-way ANOVA revealed significantly less iron accumulation for cells incubated with 10 lM 1,10phenanthroline (p \ 0.05) and 100 lM 1,10-phenanthroline (p \ 0.05). Similarly, less iron accumulated in cells treated with 10 lM 4,7-phenanthroline (p \ 0.05), 33 lM 4,7-phenanthroline (p \ 0.05) and 100 lM 4,7-phenanthroline (p \ 0.05). Comparisons between 1,10-phenanthroline and 4,7-phenathroline at equal concentrations showed that cells incubated with 4,7-phenanthroline accumulated significantly less iron at 33 lM (t(4) = 4.2, p \ 0.05) and at 100 lM (t(4) = 3.3, p \ 0.05). Astrocytes incubated for 24 h with 0 lM hemin contained 1.7 ± 1.2 nmol non-heme iron/mg protein, while cells incubated with 33 lM hemin accumulated 12.4 ± 2.4 nmol nonheme iron/mg protein (Fig. 2b). Significantly less non-heme iron was found with increasing concentrations of 1,10-phenanthroline (r = .84, p \ 0.05) and 4,7-phenanthroline (r = .78, p \ 0.05). Significantly less non-heme iron was found for all concentrations of 1,10-phenanthroline (p \ 0.05). Significantly less non-heme iron was also accumulated in the 4,7-phenanthroline condition at 33 and 100 lM (p \ 0.05). No significant differences were found between 1,10-phenanthroline and 4,7-phenanthroline at equivalent concentrations. The non-heme iron content of the extracellular media was measured after 24 h incubation, and no significant differences were found between 0 lM phenanthroline and either 100 lM 1,10-phenanthroline or 100 lM 4,7phenanthroline.
123
Neurochem Res
To confirm whether cells metabolized hemin after incubation, and whether 1,10-phenanthroline or 4,7-phenanthroline affected this process, the amount of non-heme iron was expressed as a percentage of the amount of total iron (Fig. 2c). Astrocytes incubated for 24 h with 33 lM hemin contained 10.6 ± 0.7 % non-heme iron. A significant decrease in the percentage of non-heme iron was found with increasing concentrations of 1,10-phenanthroline (r = 0.65, p \ 0.05). No relationship was found for 4,7-phenanthroline. Each concentration of phenanthroline was then compared to the respective 0 lM condition. Three concentrations of 1,10-phenanthroline resulted in a significantly lower percentage of non-heme iron when compared to 0 lM (3.3 lM; 33 lM; 100 lM, p \ 0.05). No differences were found for 4,7-phenanthroline. When 1,10-phenanthroline and 4,7-phenanthroline were compared at equivalent concentrations, it was revealed that there was a significantly higher percentage of non-heme iron after incubation with 4,7-phenanthroline compared to 1,10phenanthroline at the 33 lM concentration (t(4) = 8.7, p \ 0.05) and at the 100 lM concentration (t(4) = 3.7, p \ 0.05). These results indicate that at higher concentrations of phenanthroline, less non-heme iron is available when incubated with 1,10-phenanthroline than when incubated with 4,7-phenanthroline. Capacity of Hemin, Iron and Phenanthrolines to Inactivate H2O2
Fig. 2 Cellular iron content after 1 h pre-incubation with 0–100 lM 1,10-phenanthroline or 4,7-phenanthroline followed by 24 h incubation with 33 lM hemin plus 0–100 lM 1,10-phenanthroline or 4,7phenanthroline. a Total cellular iron accumulation following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 3.5 ± 2.5 nmol total iron/mg protein. b Cellular non-heme iron accumulation following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 1.7 ± 1.2 nmol non-heme iron/mg protein. c Percentage of non-heme iron accumulated by cells following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 76.5 ± 17.4 % of non-heme iron (as a % of total iron). *p \ 0.05 when compared to 0 lM concentration within each phenanthroline condition. ^p \ 0.05 when compared to the equivalent concentration of 1,10-phenanthroline
123
The above results show that while the phenanthrolines were able to protect astrocytes from hemin toxicity, they had little or no effect on hemin metabolism, particularly the non-chelating analogue. This finding raised the possibility that the phenanthrolines could be interacting with intact hemin in order to provide protection, possibly by preventing direct interaction with H2O2. An assay that measures the concentration of H2O2 over time was used to examine whether hemin directly interacts with H2O2 and whether either of the phenanthrolines are able to interfere with this reaction. When no hemin or phenanthroline were present in the solution, the amount of H2O2 remained relatively stable for the 90 min time period, indicating that H2O2 does not spontaneously degrade (Fig. 3a). In contrast, when the incubation solution contained hemin at varying concentrations, the H2O2 was removed from the solution in a concentration-dependent manner, indicating that the hemin reacted with H2O2 and caused it to degrade (Fig. 3a). When equivalent iron concentrations to those in hemin were applied in the form of ferric ammonium citrate, a non-heme iron source, the rate of degradation of H2O2 was much slower and reached a plateau at 25 lM H2O2 after 30 min (Fig. 3b). By contrast, hemin resulted in a
Neurochem Res
when 100 lM of either 1,10-phenanthroline or 4,7-phenanthroline were added to the hemin incubation solution, the H2O2 degradation was slowed substantially. By 30 min, the 100 lM H2O2 had degraded to a concentration of approximately 90 lM in the presence of either phenanthroline, to a concentration of 70 lM in the presence of either phenanthroline plus hemin, and had degraded to just 25 lM in the presence of hemin alone (Fig. 3c). These results indicate that both phenanthrolines interfere with the reaction between hemin and H2O2 and they slow the degradation of H2O2 by approximately 70 % during the first 30 min.
Discussion
Fig. 3 Cell-free experiments showing degradation of H2O2 over 90 min. a 0–40 lM hemin incubated with 40 lM H2O2. b 0–40 lM FAC incubated with 40 lM H2O2. c 33 lM hemin with or without 100 lM 1,10-phenanthroline or 4,7-phenanthroline with 100 lM H2O2
progressive degradation of H2O2 that continued for at least 90 min in a concentration-dependent manner, so that by 90 min the 40 lM H2O2 had been completely inactivated by both 20 and 40 lM hemin. The addition of 100 lM of either 1,10-phenanthroline or 4,7-phenanthroline to a 100 lM solution of H2O2 had no effect on the degradation of H2O2, indicating that neither of the phenanthrolines react with H2O2 (Fig. 3c). Addition of 33 lM hemin to the 100 lM solution of H2O2 resulted in the complete inactivation of H2O2 within 90 min, however
The present study investigated the effect of phenanthrolines on the toxicity of hemin to astrocytes. The results confirmed that the iron-chelating isoform (1,10-phenanthroline) is protective against hemin toxicity, however the non-chelating isoform (4,7-phenanthroline) was just as protective, and in some cases more so. This latter result suggests that the protection afforded by 1,10-phenanthroline is unlikely to be due to its capacity to chelate iron. These findings are discussed below. It has previously been reported that hemin is extremely toxic to many cell types due to the release of redox-active iron [5, 31], and that metal chelators such as 1,10-phenanthroline can attenuate this toxicity [23]. The present study confirmed that hemin causes substantial toxicity to primary astrocyte cultures, and the extent of cell loss after 24 h was consistent with previous reports [28, 32]. Furthermore, the cell-permeable iron chelator 1,10-phenanthroline protected astrocytes in a concentration-dependent manner, which is consistent with previous research [23, 24]. Interestingly, the present results contrast with our previous findings that 1,10-phenanthroline did not protect cultured rat astrocytes from hemin toxicity [28]. While that study was similar to the present one, it used 2-week old cultures of rat astrocytes whereas the present study and those of other research groups used 3-week old cultures of mouse astrocytes. This difference may be due to the increased presence of the connexin gap junction proteins that protect astrocytes from zinc toxicity by dissipating toxic zinc from cells [33]. Such proteins are expressed more abundantly as astrocyte cultures become confluent, and consequently they will be more abundant after 3 weeks of culture than after 2 weeks. Further investigations will be required to determine whether cell maturation or species differences account for this difference in protection from 1,10-phenanthroline. The present study found that the non-chelating analogue, 4,7-phenanthroline, also protects astrocytes from
123
Neurochem Res
hemin toxicity, with 33 and 100 lM 4,7-phenanthroline protecting cells to a greater extent than equivalent concentrations of 1,10-phenanthroline. Furthermore, the amount of LDH released during incubation of cells with 33 lM hemin plus 100 lM 4,7-phenanthroline did not differ from the 0 lM hemin condition, indicating that 100 lM 4,7-phenanthroline provided complete protection. This was an unexpected result as 4,7-phenanthroline has no capacity to chelate iron, and its original purpose in the present study was to act as a negative control for the effects of 1,10-phenanthroline. Astrocytes accumulate hemin in a time- and concentration-dependent manner, and the total iron content provides a reliable measure of the amount of hemin accumulated by cells [3]. The present study found that the amount of total iron accumulated by astrocytes decreased as the concentration of the phenanthrolines increased. This unexpected result indicates that both of the phenanthrolines restricted the accumulation of hemin by astrocytes, and this may have contributed to their protective effect. Interestingly, the lowest levels of cellular iron correspond to the lowest levels of LDH, suggesting that the lower the amount of hemin uptake, the lower the resulting toxicity of hemin. A potential explanation of this effect is that phenathrolines promote the intracellular breakdown of hemin and the subsequent export of non-heme iron. This possibility predicts that increased concentrations of non-heme iron will be present in the culture media after 24 h. However, analysis of the composition of the cell culture media revealed no significant differences in the amount of non-heme iron after incubation with 0 lM phenanthroline or 100 lMof either 1,10-phenanthroline or 4,7-phenanthroline. Thus, this possibility is not supported. The non-heme iron content of cells and extracellular media confirms that there was no increase in the amount of non-heme iron present therefore providing an accurate indication of how much hemin has been metabolized [28]. Both phenanthrolines demonstrated a concentrationdependent effect, with the highest concentrations resulting in the lowest accumulation of non-heme iron/mg protein. This result is not surprising, since these cells also contained less total iron, and therefore less hemin. However, when the amount of non-hemin iron was expressed as a percentage of total iron, a difference emerged between the phenanthrolines. For 1,10-phenanthroline, there was a significant concentration-dependent decrease in the percentage of non-heme iron, whereas for 4,7-phenanthroline no concentration-dependent effect was seen. The present study found that after incubating astrocytes with hemin for 24 h, approximately 10 % of total intracellular iron was present as non-heme iron, indicating that 90 % of the iron remained within intact hemin. This result is consistent with our earlier finding that rat astrocytes
123
metabolize only a small fraction of the hemin accumulated during the first 24 h [28]. Since the vast majority of hemin taken up by cells remains intact, it seems probable that intact hemin is the main cause of hemin toxicity, as suggested by previous studies [6, 9]. This conclusion is supported by the present finding that the non-chelating form of phenanthroline is even more effective at preventing hemin toxicity than the chelating form and that 4,7-phenanthroline had no apparent effect on hemin metabolism after uptake, yet it does decrease hemin accumulation. These observations led us to consider whether the phenanthrolines are able to interact directly with hemin to prevent it from being toxic. Since the toxicity of intact hemin is speculated to be due to redox-cycling in the presence of H2O2 [9], cellfree experiments were conducted to examine whether hemin reacts with H2O2 and, if so, whether the phenanthrolines interfere with this reaction. The H2O2 assay confirmed that hemin is capable of inactivating H2O2 in a concentration-dependent manner. Importantly, hemin was found to inactivate amounts of H2O2 that were present in several-fold molar excess, and this inactivation continued in a curvilinear fashion for at least 90 min. This pattern of H2O2 inactivation was very different from that seen with ferric ammonium citrate, and is consistent with hemin undergoing redox cycling in the presence of H2O2 and ambient oxygen in a Fenton-like reaction [6]. The present study found that 1,10-phenanthroline and 4,7-phenanthroline greatly slow the degradation of H2O2 by hemin. These results may provide an insight into why phenanthrolines are protective. The pyridine groups on phenanthrolines have the capacity to bind covalently at the two remaining binding sites of iron in hemin, in much the same way that oxygen binds to iron in heme [34]. We speculate that such binding will enable the phenathrolines to limit the capacity of hemin to participate in redox reactions, as the occluded iron centre of hemin will have less access to substrates such as H2O2. Another potential mechanism underlying the protective effect of the phenanthrolines relates to the HCP1, which we have previously shown to contribute to hemin uptake by astrocytes [3]. The present data do not exclude the possibility that the phenanthrolines block the active site on HCP1, slowing the rate of hemin uptake into astrocytes. The results of the present study suggest that intact hemin is the main source of hemin toxicity in astrocytes, rather than redox-active iron that is released following hemin metabolism. The protective effects of 4,7-phenanthroline suggest that both phenanthrolines may be able to interact with intact hemin to prevent cellular toxicity by a mechanism other then iron chelation. Further investigations are needed to determine whether 4,7-phenanthroline (or a structurally-related compound) could be used therapeutically to reduce the extent of brain damage that occurs in the
Neurochem Res
penumbra in the weeks following a hemorrhagic stroke. An advantage of non-chelating compounds is that they will not interfere with normal cellular functions that require iron. Acknowledgments We would like to acknowledge the School of Psychology and Psychiatry at Monash University, Melbourne Australia as some data for this project was collected at the Clayton Campus. We would like to thank Hania Czerwinska for her technical assistance with some aspects of this project.
References 1. Andersen KK, Olsen TS, Dehlendorff C, Kammersgaard LP (2009) Hemorrhagic and ischemic strokes compared: stroke severity, mortality, and risk factors. Stroke 40:2068–2072 2. Qureshi AI, Mendelow AD, Hanley DF (2009) Intracerebral haemorrhage. Lancet 373:1632–1644 3. Dang TN, Bishop GM, Dringen R, Robinson SR (2010) The putative heme transporter HCP1 is expressed in cultured astrocytes and contributes to the uptake of hemin. Glia 58:55–65 4. Kumar S, Bandyopadhyay U (2005) Free heme toxicity and its detoxification systems in human. Toxicol Lett 157:175–188 5. Balla G, Jacob HS, Eaton JW, Belcher JD, Vercellotti GM (1991) Hemin: a possible physiological mediator of low density lipoprotein oxidation and endothelial injury. Arterioscler Thromb 11:1700–1711 6. Huffman LJ, Miles PR, Shi X, Bowman L (2000) Hemoglobin potentiates the production of reactive oxygen species by alveolar macrophages. Exp Lung Res 26:203–217 7. Nagababu E, Rifkind JM (2004) Heme degradation by reactive oxygen species. Antioxid Redox Signal 6:967–978 8. Wagner KR, Sharp FR, Ardizzone TD, Lu A, Clark JF (2003) Heme and iron metabolism: role in cerebral hemorrhage. J Cereb Blood Flow Metab 23:629–652 9. Robinson SR, Dang TN, Dringen R, Bishop GM (2009) Hemin toxicity: a preventable source of brain damage following hemorrhagic stroke. Redox Rep 14:228–235 10. Halliwell B, Gutteridge J (2007) Free radicals in biology and medicine. Oxford University Press, New York 11. Gaasch JA, Lockman PR, Geldenhuys WJ, Allen DD, Van Der Schyf CJ (2007) Brain iron toxicity: differential responses of astrocytes, neurons, and endothelial cells. Neurochem Res 32:1196–1208 12. Moos T, Nielsen TR, Skjørringe T, Morgan EH (2007) Iron trafficking inside the brain. J Neurochem 103:1730–1740 13. Balla J, Vercellotti GM, Jeney V, Yachie A, Varga Z, Jacob HS, Eaton JW, Balla G (2007) Heme, heme oxygenase, and ferritin: how the vascular endothelium survives (and dies) in an iron-rich environment. Antioxid Redox Signal 9:2119–2137 14. Ke Y, Qian ZM (2007) Brain iron metabolism: neurobiology and neurochemistry. Prog Neurobiol 83:149–173 15. Li Z, Chen-Roetling J, Regan RF (2009) Increasing expression of H- or L-ferritin protects cortical astrocytes from hemin toxicity. Free Radic Res 43:613–621
16. Sigel A, Sigel H (2004) Metal ions and their complexes in medication. Marcel Dekker, New York 17. Masuda T, Hida H, Kanda Y, Aihara N, Ohta K, Yamada K, Nishino H (2007) Oral administration of metal chelator ameliorates motor dysfunction after a small hemorrhage near the internal capsule in rat. J Neurosci Res 85:213–222 18. Hoepken HH, Korten T, Robinson SR, Dringen R (2004) Iron accumulation, iron-mediated toxicity and altered levels of ferritin and transferrin receptor in cultured astrocytes during incubation with ferric ammonium citrate. J Neurochem 88:1194–1202 19. Wu H, Wu T, Xu X, Wang J (2010) Iron toxicity in mice with collagenase-induced intracerebral hemorrhage. J Cereb Blood Flow Metab 31:1243–1250 20. Halliwell B (1989) Protection against tissue damage in vivo by desferrioxamine: What is its mechanism of action? Free Radic Biol Med 7:645–651 21. Regan RF, Rogers B (2003) Delayed treatment of hemoglobin neurotoxicity. J Neurotrauma 20:111–120 22. Vanderveldt GM, Regan RF (2004) The neurotoxic effect of sickle cell hemoglobin. Free Radic Res 38:431–437 23. Regan RF, Wang Y, Xin M, Chong A, Guo Y (2001) Activation of extracellular signal-regulated kinases potentiates hemin toxicity in astrocyte cultures. J Neurochem 79:545–555 24. Regan RF, Chen J, Benvenisti-Zarom L (2004) Heme oxygenase2 gene deletion attenuates oxidative stress in neurons exposed to extracellular hemin. BMC Neurosci 5:34 25. Hamprecht B, Loffler F (1985) Primary glial cultures as a model for studying hormone action. Methods Enzymol 109:341–345 26. Dringen R, Kussmaul L, Hamprecht B (1998) Detoxification of exogenous hydrogen peroxide and organic hydroperoxides by cultured astroglial cells assessed by microtiter plate assay. Brain Res Protoc 2:223–228 27. Riemer J, Hoepken HH, Czerwinska H, Robinson SR, Dringen R (2004) Colorimetric ferrozine-based assay for the quantitation of iron in cultured cells. Anal Biochem 331:370–375 28. Dang TN, Bishop GM, Dringen R, Robinson SR (2011) The metabolism and toxicity of hemin in astrocytes. Glia 59:1540–1550 29. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275 30. Liddell JR, Hoepken HH, Crack PJ, Robinson SR, Dringen R (2006) Glutathione peroxidase 1 and glutathione are required to protect mouse astrocytes from iron-mediated hydrogen peroxide toxicity. J Neurosci Res 84:578–586 31. Liu Y, Ortiz De Montellano PR (2000) Reaction intermediates and single turnover rate constants for the oxidation of heme by human heme oxygenase-1. J Biol Chem 275:5297–5307 32. Laird MD, Wakade C, Alleyne CH Jr, Dhandapani KM (2008) Hemin-induced necroptosis involves glutathione depletion in mouse astrocytes. Free Radic Biol Med 45:1103–1114 33. Lee J, Yim YS, Ko SJ, Kim DG, Kim CH (2011) Gap junctions contribute to astrocytic resistance against zinc toxicity. Brain Res Bull 86:314–318 34. Steed JW, Atwood JL (2009) Supramolecular chemistry. Wiley, West Sussex, United Kingdom
123
ORIGINAL PAPER
Phenanthrolines Protect Astrocytes from Hemin Without Chelating Iron Jessica E. Owen • Glenda M. Bishop Stephen R. Robinson
•
Received: 16 December 2013 / Revised: 2 February 2014 / Accepted: 5 February 2014 Ó Springer Science+Business Media New York 2014
Abstract Hemin, the degradation product of hemoglobin, contributes to the neurodegeneration that occurs in the weeks following a hemorrhagic stroke. The breakdown of hemin in cells releases redox-active iron that can facilitate the production of toxic hydroxyl radicals. The present study used 3-week old primary cultures of mouse astrocytes to compare the toxicity of 33 lM hemin in the presence of the iron chelator 1,10-phenanthroline or its nonchelating analogue, 4,7-phenanthroline. This concentration of hemin killed approximately 75 % of astrocytes within 24 h. Both isoforms of phenanthroline significantly decreased the toxicity of hemin, with the non-chelating analogue providing complete protection at concentrations of 33 lM and above. The decrease in toxicity was associated with less cellular accumulation of hemin. Approximately 90 % of the hemin accumulated was not degraded, irrespective of treatment condition. These observations indicate that chelatable iron is not the cause of hemin toxicity. Cell-free experiments demonstrated that hemin can inactivate a molar excess of hydrogen peroxide (H2O2), and that the rate of inactivation is halved in the presence of either isoform of phenanthroline. We conclude that phenanthrolines may protect astrocytes by limiting hemin uptake and by impairing the capacity of intact hemin to interact with endogenous H2O2. Keywords Brain Fenton reaction Heme oxygenase Hydrogen peroxide Oxidative stress
J. E. Owen G. M. Bishop S. R. Robinson (&) School of Health Sciences, RMIT University, PO Box 71, Bundoora, VIC 3083, Australia e-mail: [email protected]
Introduction Hemorrhagic strokes account for only 10–15 % of all strokes, however they are generally more severe than ischemic strokes and are associated with a higher risk of mortality [1]. It has become evident that hemin, a breakdown product of hemoglobin (in red blood cells), is responsible for much of the secondary brain damage that follows a hemorrhagic stroke [2]. The fact that neurodegeneration continues while the patient is in clinical care provides the potential to intervene with therapeutic treatments to prevent this injury [2]. Surgical removal of the hematoma is generally not performed, due to the high risk of damaging healthy brain tissue or causing further bleeding, so other forms of intervention are needed. Hemin is taken up by brain cells via the heme carrier protein-1 (HCP1) [3], or by binding to hemopexin, such that the resultant complex is endocytosed [4]. Once inside the cell, hemin has the potential to be toxic via a variety of mechanisms, including peroxidation of lipid membranes [5], production of free radicals through interaction with endogenous hydrogen peroxide (H2O2) [6] or by the enzymatic metabolism of hemin by heme oxygenase, producing free iron, carbon monoxide and biliverdin [7, 8]. The free iron that is released from hemin is presumed to be redox-active and thus able to participate in the Fenton reaction [9], producing cytotoxic hydroxyl radicals that can lead to cell death [10, 11]. The brain tightly regulates surplus iron in order to prevent it from cycling in the Fenton reaction [12]. Normally, when the levels of iron in the brain increase, ferritin is synthesized so that the excess iron can be stored as Fe3? and cannot participate in the Fenton reaction [13], but ferritin synthesis does not seem to occur when iron is released from hemin [14]. Researchers have attempted to
123
Neurochem Res
prevent hemin toxicity by inducing ferritin expression in cells, however it has been found that the upregulation of ferritin is not sufficient to protect against hemin toxicity, and instead genetic overexpression of ferritin is required [15], which is impractical in a clinical setting. Chelators that bind metal ions usually prevent them from being redox-active [16]. Desferrioxamine, a widely-used iron chelator, decreases the accumulation of free iron by cells, thereby limiting the production of reactive oxygen species and reducing the extent of neuronal death [17–19]. However, desferrioxamine does not readily permeate cell membranes, making it less effective at chelating intracellular iron [20]. An alternate iron chelator, 1,10-phenanthroline, is lipid soluble and can readily penetrate the cell membrane. A series of important studies have demonstrated that co-incubation with 1,10-phenanthroline can protect cultured neurons and astrocytes from the toxicity of hemoglobin [21, 22] and hemin [23, 24]. These results support the conclusion that the breakdown of hemin by heme oxygenase leads to the release of redox-active iron, and that chelation of this iron by 1,10-phenanthroline attenuates the toxicity. During the course of experiments in our laboratory that had been intended to build on previous findings, we used a non-chelating analogue (4,7-phenanthroline) as a negative experimental control. It was discovered that this analogue is also effective at protecting cultured astrocytes from the toxicity of hemin. The present paper describes these findings and the subsequent investigations that were undertaken into the basis of this unexpected effect. These new findings have stimulated a reassessment of the role of redox-active iron in the toxicity of hemin.
Materials and Method Primary Cell Cultures The use of animals to obtain cell cultures was approved by the Monash University Animal Ethics Committee. Primary astrocyte cell cultures were prepared from newborn C57BL/ 6JAsmu mice (\24 h old), as described previously [25]. Cells were plated in 24-well culture plates at 300,000 cells/well. All cells were incubated in culture medium [90 % Dulbecco’s modified Eagle medium (DMEM), 10 % fetal calf serum, 20 U/mL penicillin G, and 20 lg/mL streptomycin sulphate] and maintained at 37 °C in 95 % humidity with 10 % CO2. Culture media were changed every seventh day and cultures were used for experiments after 21–23 days in vitro. Incubation Protocol The culture medium was removed and the cells were washed twice with 1 mL 37 °C DMEM, after which the
123
pre-incubation solution was applied for 1 h. Pre-incubation solutions contained DMEM and 0–100 lM of 1,10-phenanthroline or 4,7-phenanthroline. After 1 h, the preincubation solution was removed and 1 mL of incubation solution was applied for 24 h. Incubation solutions contained DMEM and the same concentration of either 1,10phenanthroline or 4,7-phenanthroline that was used during the pre-incubation, plus 0 or 33 lM hemin. Hemin was dissolved as a 1 mM stock solution in 10 mM NaOH and then diluted to a final concentration of 33 lM in the incubation solution. The 0 lM hemin condition contained an identical concentration of NaOH as the 33 lM hemin condition. At the conclusion of the incubation period, media samples were collected into Eppendorf tubes and then the cells were washed twice with 1 mL of ice cold 10 mM potassium phosphate-buffered saline (K–PBS, pH 7.4) and stored at -20 °C. Cell Viability Cell viability was determined by measuring the amount of lactate dehydrogenase (LDH) released into the media during the incubation [26]. The activity of LDH was determined by measuring the rate of NADH decomposition as pyruvate was reduced to lactate. The LDH activity of samples was compared to a 100 % LDH release condition represented by untreated cells that were lysed with 1 % Triton X-100 to release all LDH present. Iron Content The total iron content of the sample was determined colorimetrically [27]. The amount of non-heme iron in the sample was determined with the same procedure with the exception that potassium permanganate was excluded from the reaction mixture [28]. As well as determining the iron content of the cells, the iron content of media samples was measured before and after the incubation. The protein content of equivalently-treated cell cultures was used to standardize iron content. Protein content was determined with the Lowry Protein Estimation assay [29]. Hydrogen Peroxide Assay To investigate the capacity of hemin, alone or in combination with phenanthroline, to react with H2O2 in a cellfree assay, incubation solutions were prepared as described above, except that they were made using pyruvate-free DMEM. In addition, H2O2 was added to each solution at a final concentration of 40 or 100 lM. Samples of each solution were collected at specific time points for up to 90 min, allowing H2O2 detoxification to be determined, as previously described [30].
Neurochem Res
protective at all higher concentrations, causing a significant decrease in LDH release (p \ 0.05). Similarly, 4,7-phenanthroline was not protective at 3.3 lM when compared to the 0 lM condition, whereas it was protective at all higher concentrations (p \ 0.05). Comparisons were made between 1,10-phenanthroline and 4,7-phenanthroline within each specific concentration, revealing that 4,7-phenanthroline provided significantly more protection against hemin toxicity than 1,10-phenanthroline at 33 lM (t(4) = 3.6, p \ 0.05) and 100 lM (t(4) = 3.8, p \ 0.05). Effect of Phenanthrolines on Cellular iron Accumulation and Metabolism of Hemin Fig. 1 Cell viability as measured by LDH release after 1 h preincubation with 0–100 lM 1,10-phenanthroline or 4,7-phenanthroline followed by 24 h incubation with 33 lM hemin plus 0–100 lM 1,10phenanthroline or 4,7-phenanthroline. *p \ 0.05 when compared to 0 lM concentration within each phenanthroline condition. ^p \ 0.05 when compared to the equivalent concentration of 1,10phenanthroline
Statistical Analysis All experiments were performed with three independent cultures, with triplicate conditions within each culture. Analyses were performed using the mean ± SD of each triplicate for each of the independent cultures, giving n = 3. Linear regression, independent t tests and planned comparison ANOVA with post hoc Dunnett’s tests were performed with significance set at p \ 0.05.
Results Effect of Phenanthrolines on Cell Viability After 24 h, cell cultures that had been incubated with 33 lM hemin had experienced considerably more cell death (75.5 ± 7.5 % LDH release) than those that had been incubated with 0 lM hemin (3.8 ± 1.3 % LDH release; p \ 0.05). When cells were incubated with either 1,10phenanthroline or 4,7-phenanthroline and 0 lM hemin, less than 7 % LDH release was observed, indicating that neither analogue of phenanthroline is toxic to astrocytes over a 24 h period. When astrocytes were co-incubated with 33 lM hemin plus up to 100 lM of either of the phenanthrolines, a dosedependent effect was observed, such that concentrations C10 lM produced a significant decrease in LDH release (Fig. 1). Linear regression analysis revealed significant effects of concentration for both 1,10-phenanthroline (r = 0.91, p \ 0.05), and 4,7-phenanthroline (r = 0.96, p \ 0.05). A one-way ANOVA showed that 1,10-phenanthroline was not protective at 3.3 lM whereas it was
Cellular iron content was measured and then standardized to the protein content of equivalently-treated cultures to account for cell death. Astrocytes incubated for 24 h with 0 lM hemin contained 3.5 ± 2.5 nmol total iron/mg protein, while cells incubated with 33 lM hemin accumulated 117.1 ± 17.9 nmol total iron/mg protein (Fig. 2a). Linear regression analysis was used to compare the concentration of total iron in astrocytes that had been incubated with hemin plus 0–100 lM of either phenanthroline. A significant decrease in iron accumulation was found with increasing concentrations of both 1,10-phenanthroline (r = 0.71, p \ 0.05), and 4,7-phenanthroline (r = 0.89, p \ 0.05). A one-way ANOVA revealed significantly less iron accumulation for cells incubated with 10 lM 1,10phenanthroline (p \ 0.05) and 100 lM 1,10-phenanthroline (p \ 0.05). Similarly, less iron accumulated in cells treated with 10 lM 4,7-phenanthroline (p \ 0.05), 33 lM 4,7-phenanthroline (p \ 0.05) and 100 lM 4,7-phenanthroline (p \ 0.05). Comparisons between 1,10-phenanthroline and 4,7-phenathroline at equal concentrations showed that cells incubated with 4,7-phenanthroline accumulated significantly less iron at 33 lM (t(4) = 4.2, p \ 0.05) and at 100 lM (t(4) = 3.3, p \ 0.05). Astrocytes incubated for 24 h with 0 lM hemin contained 1.7 ± 1.2 nmol non-heme iron/mg protein, while cells incubated with 33 lM hemin accumulated 12.4 ± 2.4 nmol nonheme iron/mg protein (Fig. 2b). Significantly less non-heme iron was found with increasing concentrations of 1,10-phenanthroline (r = .84, p \ 0.05) and 4,7-phenanthroline (r = .78, p \ 0.05). Significantly less non-heme iron was found for all concentrations of 1,10-phenanthroline (p \ 0.05). Significantly less non-heme iron was also accumulated in the 4,7-phenanthroline condition at 33 and 100 lM (p \ 0.05). No significant differences were found between 1,10-phenanthroline and 4,7-phenanthroline at equivalent concentrations. The non-heme iron content of the extracellular media was measured after 24 h incubation, and no significant differences were found between 0 lM phenanthroline and either 100 lM 1,10-phenanthroline or 100 lM 4,7phenanthroline.
123
Neurochem Res
To confirm whether cells metabolized hemin after incubation, and whether 1,10-phenanthroline or 4,7-phenanthroline affected this process, the amount of non-heme iron was expressed as a percentage of the amount of total iron (Fig. 2c). Astrocytes incubated for 24 h with 33 lM hemin contained 10.6 ± 0.7 % non-heme iron. A significant decrease in the percentage of non-heme iron was found with increasing concentrations of 1,10-phenanthroline (r = 0.65, p \ 0.05). No relationship was found for 4,7-phenanthroline. Each concentration of phenanthroline was then compared to the respective 0 lM condition. Three concentrations of 1,10-phenanthroline resulted in a significantly lower percentage of non-heme iron when compared to 0 lM (3.3 lM; 33 lM; 100 lM, p \ 0.05). No differences were found for 4,7-phenanthroline. When 1,10-phenanthroline and 4,7-phenanthroline were compared at equivalent concentrations, it was revealed that there was a significantly higher percentage of non-heme iron after incubation with 4,7-phenanthroline compared to 1,10phenanthroline at the 33 lM concentration (t(4) = 8.7, p \ 0.05) and at the 100 lM concentration (t(4) = 3.7, p \ 0.05). These results indicate that at higher concentrations of phenanthroline, less non-heme iron is available when incubated with 1,10-phenanthroline than when incubated with 4,7-phenanthroline. Capacity of Hemin, Iron and Phenanthrolines to Inactivate H2O2
Fig. 2 Cellular iron content after 1 h pre-incubation with 0–100 lM 1,10-phenanthroline or 4,7-phenanthroline followed by 24 h incubation with 33 lM hemin plus 0–100 lM 1,10-phenanthroline or 4,7phenanthroline. a Total cellular iron accumulation following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 3.5 ± 2.5 nmol total iron/mg protein. b Cellular non-heme iron accumulation following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 1.7 ± 1.2 nmol non-heme iron/mg protein. c Percentage of non-heme iron accumulated by cells following incubation with 33 lM hemin. Incubation with 0 lM hemin resulted in 76.5 ± 17.4 % of non-heme iron (as a % of total iron). *p \ 0.05 when compared to 0 lM concentration within each phenanthroline condition. ^p \ 0.05 when compared to the equivalent concentration of 1,10-phenanthroline
123
The above results show that while the phenanthrolines were able to protect astrocytes from hemin toxicity, they had little or no effect on hemin metabolism, particularly the non-chelating analogue. This finding raised the possibility that the phenanthrolines could be interacting with intact hemin in order to provide protection, possibly by preventing direct interaction with H2O2. An assay that measures the concentration of H2O2 over time was used to examine whether hemin directly interacts with H2O2 and whether either of the phenanthrolines are able to interfere with this reaction. When no hemin or phenanthroline were present in the solution, the amount of H2O2 remained relatively stable for the 90 min time period, indicating that H2O2 does not spontaneously degrade (Fig. 3a). In contrast, when the incubation solution contained hemin at varying concentrations, the H2O2 was removed from the solution in a concentration-dependent manner, indicating that the hemin reacted with H2O2 and caused it to degrade (Fig. 3a). When equivalent iron concentrations to those in hemin were applied in the form of ferric ammonium citrate, a non-heme iron source, the rate of degradation of H2O2 was much slower and reached a plateau at 25 lM H2O2 after 30 min (Fig. 3b). By contrast, hemin resulted in a
Neurochem Res
when 100 lM of either 1,10-phenanthroline or 4,7-phenanthroline were added to the hemin incubation solution, the H2O2 degradation was slowed substantially. By 30 min, the 100 lM H2O2 had degraded to a concentration of approximately 90 lM in the presence of either phenanthroline, to a concentration of 70 lM in the presence of either phenanthroline plus hemin, and had degraded to just 25 lM in the presence of hemin alone (Fig. 3c). These results indicate that both phenanthrolines interfere with the reaction between hemin and H2O2 and they slow the degradation of H2O2 by approximately 70 % during the first 30 min.
Discussion
Fig. 3 Cell-free experiments showing degradation of H2O2 over 90 min. a 0–40 lM hemin incubated with 40 lM H2O2. b 0–40 lM FAC incubated with 40 lM H2O2. c 33 lM hemin with or without 100 lM 1,10-phenanthroline or 4,7-phenanthroline with 100 lM H2O2
progressive degradation of H2O2 that continued for at least 90 min in a concentration-dependent manner, so that by 90 min the 40 lM H2O2 had been completely inactivated by both 20 and 40 lM hemin. The addition of 100 lM of either 1,10-phenanthroline or 4,7-phenanthroline to a 100 lM solution of H2O2 had no effect on the degradation of H2O2, indicating that neither of the phenanthrolines react with H2O2 (Fig. 3c). Addition of 33 lM hemin to the 100 lM solution of H2O2 resulted in the complete inactivation of H2O2 within 90 min, however
The present study investigated the effect of phenanthrolines on the toxicity of hemin to astrocytes. The results confirmed that the iron-chelating isoform (1,10-phenanthroline) is protective against hemin toxicity, however the non-chelating isoform (4,7-phenanthroline) was just as protective, and in some cases more so. This latter result suggests that the protection afforded by 1,10-phenanthroline is unlikely to be due to its capacity to chelate iron. These findings are discussed below. It has previously been reported that hemin is extremely toxic to many cell types due to the release of redox-active iron [5, 31], and that metal chelators such as 1,10-phenanthroline can attenuate this toxicity [23]. The present study confirmed that hemin causes substantial toxicity to primary astrocyte cultures, and the extent of cell loss after 24 h was consistent with previous reports [28, 32]. Furthermore, the cell-permeable iron chelator 1,10-phenanthroline protected astrocytes in a concentration-dependent manner, which is consistent with previous research [23, 24]. Interestingly, the present results contrast with our previous findings that 1,10-phenanthroline did not protect cultured rat astrocytes from hemin toxicity [28]. While that study was similar to the present one, it used 2-week old cultures of rat astrocytes whereas the present study and those of other research groups used 3-week old cultures of mouse astrocytes. This difference may be due to the increased presence of the connexin gap junction proteins that protect astrocytes from zinc toxicity by dissipating toxic zinc from cells [33]. Such proteins are expressed more abundantly as astrocyte cultures become confluent, and consequently they will be more abundant after 3 weeks of culture than after 2 weeks. Further investigations will be required to determine whether cell maturation or species differences account for this difference in protection from 1,10-phenanthroline. The present study found that the non-chelating analogue, 4,7-phenanthroline, also protects astrocytes from
123
Neurochem Res
hemin toxicity, with 33 and 100 lM 4,7-phenanthroline protecting cells to a greater extent than equivalent concentrations of 1,10-phenanthroline. Furthermore, the amount of LDH released during incubation of cells with 33 lM hemin plus 100 lM 4,7-phenanthroline did not differ from the 0 lM hemin condition, indicating that 100 lM 4,7-phenanthroline provided complete protection. This was an unexpected result as 4,7-phenanthroline has no capacity to chelate iron, and its original purpose in the present study was to act as a negative control for the effects of 1,10-phenanthroline. Astrocytes accumulate hemin in a time- and concentration-dependent manner, and the total iron content provides a reliable measure of the amount of hemin accumulated by cells [3]. The present study found that the amount of total iron accumulated by astrocytes decreased as the concentration of the phenanthrolines increased. This unexpected result indicates that both of the phenanthrolines restricted the accumulation of hemin by astrocytes, and this may have contributed to their protective effect. Interestingly, the lowest levels of cellular iron correspond to the lowest levels of LDH, suggesting that the lower the amount of hemin uptake, the lower the resulting toxicity of hemin. A potential explanation of this effect is that phenathrolines promote the intracellular breakdown of hemin and the subsequent export of non-heme iron. This possibility predicts that increased concentrations of non-heme iron will be present in the culture media after 24 h. However, analysis of the composition of the cell culture media revealed no significant differences in the amount of non-heme iron after incubation with 0 lM phenanthroline or 100 lMof either 1,10-phenanthroline or 4,7-phenanthroline. Thus, this possibility is not supported. The non-heme iron content of cells and extracellular media confirms that there was no increase in the amount of non-heme iron present therefore providing an accurate indication of how much hemin has been metabolized [28]. Both phenanthrolines demonstrated a concentrationdependent effect, with the highest concentrations resulting in the lowest accumulation of non-heme iron/mg protein. This result is not surprising, since these cells also contained less total iron, and therefore less hemin. However, when the amount of non-hemin iron was expressed as a percentage of total iron, a difference emerged between the phenanthrolines. For 1,10-phenanthroline, there was a significant concentration-dependent decrease in the percentage of non-heme iron, whereas for 4,7-phenanthroline no concentration-dependent effect was seen. The present study found that after incubating astrocytes with hemin for 24 h, approximately 10 % of total intracellular iron was present as non-heme iron, indicating that 90 % of the iron remained within intact hemin. This result is consistent with our earlier finding that rat astrocytes
123
metabolize only a small fraction of the hemin accumulated during the first 24 h [28]. Since the vast majority of hemin taken up by cells remains intact, it seems probable that intact hemin is the main cause of hemin toxicity, as suggested by previous studies [6, 9]. This conclusion is supported by the present finding that the non-chelating form of phenanthroline is even more effective at preventing hemin toxicity than the chelating form and that 4,7-phenanthroline had no apparent effect on hemin metabolism after uptake, yet it does decrease hemin accumulation. These observations led us to consider whether the phenanthrolines are able to interact directly with hemin to prevent it from being toxic. Since the toxicity of intact hemin is speculated to be due to redox-cycling in the presence of H2O2 [9], cellfree experiments were conducted to examine whether hemin reacts with H2O2 and, if so, whether the phenanthrolines interfere with this reaction. The H2O2 assay confirmed that hemin is capable of inactivating H2O2 in a concentration-dependent manner. Importantly, hemin was found to inactivate amounts of H2O2 that were present in several-fold molar excess, and this inactivation continued in a curvilinear fashion for at least 90 min. This pattern of H2O2 inactivation was very different from that seen with ferric ammonium citrate, and is consistent with hemin undergoing redox cycling in the presence of H2O2 and ambient oxygen in a Fenton-like reaction [6]. The present study found that 1,10-phenanthroline and 4,7-phenanthroline greatly slow the degradation of H2O2 by hemin. These results may provide an insight into why phenanthrolines are protective. The pyridine groups on phenanthrolines have the capacity to bind covalently at the two remaining binding sites of iron in hemin, in much the same way that oxygen binds to iron in heme [34]. We speculate that such binding will enable the phenathrolines to limit the capacity of hemin to participate in redox reactions, as the occluded iron centre of hemin will have less access to substrates such as H2O2. Another potential mechanism underlying the protective effect of the phenanthrolines relates to the HCP1, which we have previously shown to contribute to hemin uptake by astrocytes [3]. The present data do not exclude the possibility that the phenanthrolines block the active site on HCP1, slowing the rate of hemin uptake into astrocytes. The results of the present study suggest that intact hemin is the main source of hemin toxicity in astrocytes, rather than redox-active iron that is released following hemin metabolism. The protective effects of 4,7-phenanthroline suggest that both phenanthrolines may be able to interact with intact hemin to prevent cellular toxicity by a mechanism other then iron chelation. Further investigations are needed to determine whether 4,7-phenanthroline (or a structurally-related compound) could be used therapeutically to reduce the extent of brain damage that occurs in the
Neurochem Res
penumbra in the weeks following a hemorrhagic stroke. An advantage of non-chelating compounds is that they will not interfere with normal cellular functions that require iron. Acknowledgments We would like to acknowledge the School of Psychology and Psychiatry at Monash University, Melbourne Australia as some data for this project was collected at the Clayton Campus. We would like to thank Hania Czerwinska for her technical assistance with some aspects of this project.
References 1. Andersen KK, Olsen TS, Dehlendorff C, Kammersgaard LP (2009) Hemorrhagic and ischemic strokes compared: stroke severity, mortality, and risk factors. Stroke 40:2068–2072 2. Qureshi AI, Mendelow AD, Hanley DF (2009) Intracerebral haemorrhage. Lancet 373:1632–1644 3. Dang TN, Bishop GM, Dringen R, Robinson SR (2010) The putative heme transporter HCP1 is expressed in cultured astrocytes and contributes to the uptake of hemin. Glia 58:55–65 4. Kumar S, Bandyopadhyay U (2005) Free heme toxicity and its detoxification systems in human. Toxicol Lett 157:175–188 5. Balla G, Jacob HS, Eaton JW, Belcher JD, Vercellotti GM (1991) Hemin: a possible physiological mediator of low density lipoprotein oxidation and endothelial injury. Arterioscler Thromb 11:1700–1711 6. Huffman LJ, Miles PR, Shi X, Bowman L (2000) Hemoglobin potentiates the production of reactive oxygen species by alveolar macrophages. Exp Lung Res 26:203–217 7. Nagababu E, Rifkind JM (2004) Heme degradation by reactive oxygen species. Antioxid Redox Signal 6:967–978 8. Wagner KR, Sharp FR, Ardizzone TD, Lu A, Clark JF (2003) Heme and iron metabolism: role in cerebral hemorrhage. J Cereb Blood Flow Metab 23:629–652 9. Robinson SR, Dang TN, Dringen R, Bishop GM (2009) Hemin toxicity: a preventable source of brain damage following hemorrhagic stroke. Redox Rep 14:228–235 10. Halliwell B, Gutteridge J (2007) Free radicals in biology and medicine. Oxford University Press, New York 11. Gaasch JA, Lockman PR, Geldenhuys WJ, Allen DD, Van Der Schyf CJ (2007) Brain iron toxicity: differential responses of astrocytes, neurons, and endothelial cells. Neurochem Res 32:1196–1208 12. Moos T, Nielsen TR, Skjørringe T, Morgan EH (2007) Iron trafficking inside the brain. J Neurochem 103:1730–1740 13. Balla J, Vercellotti GM, Jeney V, Yachie A, Varga Z, Jacob HS, Eaton JW, Balla G (2007) Heme, heme oxygenase, and ferritin: how the vascular endothelium survives (and dies) in an iron-rich environment. Antioxid Redox Signal 9:2119–2137 14. Ke Y, Qian ZM (2007) Brain iron metabolism: neurobiology and neurochemistry. Prog Neurobiol 83:149–173 15. Li Z, Chen-Roetling J, Regan RF (2009) Increasing expression of H- or L-ferritin protects cortical astrocytes from hemin toxicity. Free Radic Res 43:613–621
16. Sigel A, Sigel H (2004) Metal ions and their complexes in medication. Marcel Dekker, New York 17. Masuda T, Hida H, Kanda Y, Aihara N, Ohta K, Yamada K, Nishino H (2007) Oral administration of metal chelator ameliorates motor dysfunction after a small hemorrhage near the internal capsule in rat. J Neurosci Res 85:213–222 18. Hoepken HH, Korten T, Robinson SR, Dringen R (2004) Iron accumulation, iron-mediated toxicity and altered levels of ferritin and transferrin receptor in cultured astrocytes during incubation with ferric ammonium citrate. J Neurochem 88:1194–1202 19. Wu H, Wu T, Xu X, Wang J (2010) Iron toxicity in mice with collagenase-induced intracerebral hemorrhage. J Cereb Blood Flow Metab 31:1243–1250 20. Halliwell B (1989) Protection against tissue damage in vivo by desferrioxamine: What is its mechanism of action? Free Radic Biol Med 7:645–651 21. Regan RF, Rogers B (2003) Delayed treatment of hemoglobin neurotoxicity. J Neurotrauma 20:111–120 22. Vanderveldt GM, Regan RF (2004) The neurotoxic effect of sickle cell hemoglobin. Free Radic Res 38:431–437 23. Regan RF, Wang Y, Xin M, Chong A, Guo Y (2001) Activation of extracellular signal-regulated kinases potentiates hemin toxicity in astrocyte cultures. J Neurochem 79:545–555 24. Regan RF, Chen J, Benvenisti-Zarom L (2004) Heme oxygenase2 gene deletion attenuates oxidative stress in neurons exposed to extracellular hemin. BMC Neurosci 5:34 25. Hamprecht B, Loffler F (1985) Primary glial cultures as a model for studying hormone action. Methods Enzymol 109:341–345 26. Dringen R, Kussmaul L, Hamprecht B (1998) Detoxification of exogenous hydrogen peroxide and organic hydroperoxides by cultured astroglial cells assessed by microtiter plate assay. Brain Res Protoc 2:223–228 27. Riemer J, Hoepken HH, Czerwinska H, Robinson SR, Dringen R (2004) Colorimetric ferrozine-based assay for the quantitation of iron in cultured cells. Anal Biochem 331:370–375 28. Dang TN, Bishop GM, Dringen R, Robinson SR (2011) The metabolism and toxicity of hemin in astrocytes. Glia 59:1540–1550 29. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275 30. Liddell JR, Hoepken HH, Crack PJ, Robinson SR, Dringen R (2006) Glutathione peroxidase 1 and glutathione are required to protect mouse astrocytes from iron-mediated hydrogen peroxide toxicity. J Neurosci Res 84:578–586 31. Liu Y, Ortiz De Montellano PR (2000) Reaction intermediates and single turnover rate constants for the oxidation of heme by human heme oxygenase-1. J Biol Chem 275:5297–5307 32. Laird MD, Wakade C, Alleyne CH Jr, Dhandapani KM (2008) Hemin-induced necroptosis involves glutathione depletion in mouse astrocytes. Free Radic Biol Med 45:1103–1114 33. Lee J, Yim YS, Ko SJ, Kim DG, Kim CH (2011) Gap junctions contribute to astrocytic resistance against zinc toxicity. Brain Res Bull 86:314–318 34. Steed JW, Atwood JL (2009) Supramolecular chemistry. Wiley, West Sussex, United Kingdom
123
