Humic acid mediates iron release from ferritin and promotes lipid peroxidation in vitro: a possible mechanism for humic acid-induced cytotoxicity

Humic acid (HA) has been shown to be a toxic factor for many mammalian cells, however the specific mechanism of the cytotoxicity induced by HA remains unclear. From the assessment of its redox properties, HA has been shown to be capable of reducing iron(III) to iron(II) in aqueous conditions over a broad range of pH values (from 4.0 to 9.0). By using thiobarbituric acid-reactive substances as an index, the presence of HA was noted to increase the extent of lipid peroxidation both for linoleic acids and within rat liver microsomes. In addition, the increase in HA-induced lipid peroxidation is partially inhibited by sodium azide (a singlet oxygen scavenger) or disodium 4,5-dihydroxy-1,3-benzene-disulfonic acid (a superoxide scavenger), reflecting the involvement of singlet oxygen and superoxide in the process of lipid peroxidation. The addition of HA into a reaction system has been shown to generate superoxide in a dose-dependent manner by the superoxide dismutase-inhibitable cytochrome c reduction assay. In addition, HA is able to reduce and release iron from ferritin, but this process is partially inhibited by superoxide scavengers. Subsequently, the iron released from ferritin was shown to accelerate the HA-induced lipid peroxidation. From our results we conclude that HA has the ability to reduce and release iron from ferritin storage as well as to promote lipid peroxidation. Therefore, HA coupled with released iron can disturb the redox balance and elicit oxidative stress within a biological system. This may be one of the most important mechanisms for HA-induced cytotoxicity.     

Anthracyclines induce accumulation of iron in ferritin in myocardial and neoplastic cells: inhibition of the ferritin iron mobilization pathway

Anthracyclines are potent antitumor agents that cause cardiotoxicity at high cumulative doses. Because anthracycline cardiotoxicity is attributed to their ability to avidly bind iron (Fe), we examined the effect of anthracyclines on intracellular Fe trafficking in neoplastic cells and differentiated cardiomyocytes. In both cell types, incubation with doxorubicin (DOX) resulted in a significant (p < 0.004) accumulation of Fe in the storage protein, ferritin. Pulse-chase experiments using control cells demonstrated that within 6 h, the majority of (59)Fe donated from transferrin was incorporated into ferritin. Over longer incubation periods up to 18 to 24 h, (59)Fe was subsequently mobilized from ferritin into other compartments in control cells. However, anthracyclines inhibited ferritin-(59)Fe redistribution during the 18- to 24-h period, resulting in a significant (p < 0.0003) 3- to 5-fold accumulation of ferritin-(59)Fe compared with control cells. The increase in ferritin-(59)Fe after a 24-h incubation with DOX could not be correlated with increased ferritin expression, suggesting that (59)Fe accumulation occurred in pre-existing ferritin. In addition to DOX, other redox-cycling agents (i.e., menadione and paraquat) also increased ferritin-(59)Fe levels. Moreover, the intracellular superoxide scavenger, Mn(III) tetrakis(4-benzoic acid)-porphyrin complex, partially prevented the ability of DOX and menadione at inducing this effect. Hence, superoxide generation by these compounds could play a role in causing ferritin-(59)Fe accumulation. This study is the first to demonstrate the effect of anthracyclines at inhibiting Fe mobilization from ferritin, resulting in marked Fe accumulation within the molecule. This response may have consequences in terms of the cytotoxic effects of anthracyclines.     

Development and optimization of the drug release determination of iron from iron prolonged release tablets

The aim of this research was to develop and optimize the drug release determination of iron from iron prolonged release tablets. First, solubility characteristics of ferrous sulfate in different aqueous media with a pH in the range of 1 to 8 were studied. According to the results obtained different physicochemical conditions that influence drug release of iron from iron prolonged release tablets were checked. Various aqueous media with a pH in the range of 1 to 7.4, different volumes of dissolution media, various rotation speeds of stirring elements, different dissolution apparatus (apparatus 1/basket apparatus and apparatus 2/paddle apparatus according to USP/Ph. Eur.) were verified. For evaluation of dissolved iron two different methods were checked: atomic absorption spectrometry and redox titration. Redox titration was proved to be a reproducible, discriminatory, selective and cost effective method for evaluation of dissolved iron from iron prolonged release tablets and can be applied in the quality control of solid dosage forms containing iron compounds.     

6-Hydroxydopamine releases iron from ferritin and promotes ferritin-dependent lipid peroxidation

Iron was released from ferritin by the catecholamine analog, 6-hydroxydopamine. Iron release was more efficient under nitrogen than in air, suggesting that the hydroquinone has the major role in the process. Superoxide dismutase, alone or in combination with catalase, strongly inhibited 6-hydroxydopamine oxidation and greatly enhanced the amount of ferritin iron release. Catalase alone had a similar, but lesser effect. Iron released from ferritin accelerated the autoxidation of 6-hydroxydopamine. This occurred by a mechanism that was inhibited by a combination of catalase and a chelator, and to a lesser extent by superoxide dismutase. 6-Hydroxydopamine was a good promoter of metal-catalysed lipid peroxidation, and ferritin-iron participated in the process. Superoxide dismutase, and to a lesser extent catalase, stimulated peroxidation catalysed by adventitious levels of iron, but in the presence of ferritin, each enzyme was inhibitory. It appears that the greatly enhanced iron release seen under these conditions accelerated the autoxidation of 6-hydroxydopamine so that less was available to participate in peroxidative reactions. However, when 6-hydroxydopamine autoxidation was prevented by a combination of superoxide dismutase and catalase, lipid peroxidation was also inhibited, suggesting that some intermediate of autoxidation is a further requirement for the process.     

Release of iron from ferritin by aqueous extracts of cigarette smoke.

This study demonstrates the ability of aqueous extracts of cigarette smoke to reduce iron and cause its release from ferritin. Superoxide dismutase (SOD) increases the rates of iron release with the less filtered smoke extracts, but has no effect on the rate of iron release caused by aqueous extracts of well-filtered gas-phase cigarette smoke. Faster rates of iron release are observed under anaerobic conditions, and the reducing power of the cigarette smoke extracts is prolonged when incubated in argon. Hydroquinone and catechol, two of the major polyhydroxybenzenes in cigarette smoke, increase in concentration in the smoke extracts as these are subjected to less filtration, and higher concentrations of polyhydroxybenzenes correlate with higher rates of iron release from ferritin. Concomitant with iron release, depletions of amino acids in ferritin are observed. Depletion of histidine is partially prevented by bathophenanthrolinedisulfonate and mannitol, while lysine and arginine depletions remain unaffected. These observations suggest that cigarette smoke components react directly with these amino acid residues in ferritin. Cigarette smoke induced release of iron could alter iron metabolism in the lungs of chronic smokers and contribute to the increase in the total oxidative burden on the lungs of smokers.     

Genetic variation of basal iron status, ferritin and iron regulatory protein in mice: potential for modulation of oxidative stress

Toxic and carcinogenic free radical processes induced by drugs and other chemicals are probably modulated by the participation of available iron. To see whether endogenous iron was genetically variable in normal mice, the common strains C57BL/10ScSn, C57BL/6J, BALB/c, DBA/2, and SWR were examined for major differences in their hepatic non-heme iron contents. Levels in SWR mice were 3- to 5-fold higher than in the two C57BL strains, with intermediate levels in DBA/2 and BALB/c mice. Concentrations in kidney, lung, and especially spleen of SWR mice were also greater than those in C57BL mice. Non-denaturing PAGE of hepatic ferritin from all strains showed a major holoferritin band at approximately 600 kDa, with SWR mice having > 3-fold higher levels than C57BL strains. SDS PAGE showed a band of 22 kDa, mainly representing L-ferritin subunits. A trace of a subunit at 18 kDa was also detected in ferritin from SWR mice. The 18 kDa subunit and a 500 kDa holoferritin from which it originates were observed in all strains after parenteral iron overload, and there was no major variation in ferritin patterns. Although iron uptake studies showed no evidence for differential duodenal absorption between strains to explain the variation in basal iron levels, acquisition of absorbed iron by the liver was significantly higher in SWR mice than C57BL/6J. As with iron and ferritin contents, total iron regulatory protein (IRP-1) binding capacity for mRNA iron responsive element (IRE) and actual IRE/IRP binding in the liver were significantly greater in SWR than C57BL/6J mice. Cytosolic aconitase activity, representing unbound IRP-1, tended to be lower in the former strain. SWR mice were more susceptible than C57BL/10ScSn mice to the toxic action of diquat, which is thought to involve iron catalysis. If extrapolated to humans, the findings could suggest that some people might have the propensity for greater basal hepatic iron stores than others, which might make them more susceptible to iron-catalysed toxicity caused by oxidants.     

Synergistic effect of nifedipine and propranolol on adenosine (catabolite) release from ischemic rat heart

Both nifedipine a calcium antagonist, and propranolol a β-adrenergic blocker, are used as protective agents of the ischemic myocardium. In the clinical setting, the combination of the two drugs is used succesfully although several case reports indicate potential dangers of the combination. For this reason we decided to study the combined effect of nifedipine and DL-propranolol in the isolated rat heart made ischemic for a short period of time. Apex displacement was taken as a measure of contractility. Release of the AMP catabolites adenosine, inosine, (hypo)xanthine and uric acid was used as a marker of ATP breakdown. Contractility during ischemia was not affected by the drugs. DL-Propranolol (30 or 150 μg/1) had no effect on ischemic myocardial purine release, while nifedipine (15 μg/1) reduced purine release during ischemia by 33% (P < 0.02). The combination of 15 μg/1 nifedipine and 150 μg/1 DL-propranolol decreased purine release by 53% (P < 0.005 vs. nifedipine). We conclude from these results that propranolol has a synergistic effect, adding to the beneficial action of nifedipine on ischemic myocardium.     

Biology of ferritin in mammals: an update on iron storage,oxidative damage and neurodegeneration

Iron is an abundant transition metal that is essential for life, being associated with many enzyme and oxygen carrier proteins involved in a variety of fundamental cellular processes. At the same time, the metal is potentially toxic due to its capacity to engage in the catalytic production of noxious reactive oxygen species. The control of iron availability in the cells is largely dependent on ferritins, ubiquitous proteins with storage and detoxification capacity. In mammals, cytosolic ferritins are composed of two types of subunits, the H and the L chain, assembled to form a 24-mer spherical cage. Ferritin is present also in mitochondria, in the form of a complex with 24 identical chains. Even though the proteins have been known for a long time, their study is a very active and interesting field yet. In this review, we will focus our attention to mammalian cytosolic and mitochondrial ferritins, describing the most recent advancement regarding their storage and antioxidant function, the effects of their genetic mutations in human pathology, and also the possible involvement in non-iron-related activities. We will also discuss recent evidence connecting ferritins and the toxicity of iron in a set of neurodegenerative disorder characterized by focal cerebral siderosis.     

非酒精性脂肪肝患者血清铁、转铁蛋白、血清铁蛋白水平测定

目前研究发现,铁作为一种独立的危险因素,在非酒精性脂肪肝(NAFL)的发生发展过程中起着一定的作用[1]:启动脂质过氧化反应,导致机体组织细胞损伤。但国内研究相对较少,本文就不同程度的NAFL和血清铁的关系加以研究,旨在为本病诊断提供重要指标。1材料和方法2001年6月~2004年3月     

Release of iron from ferritin by metabolites of benzene and superoxide radical generating agents.

The release of iron from ferritin in the presence of benzene metabolites, viz. phenol (P), catechol (CT), hydroquinone (HQ) and superoxide radical generating compounds, viz. pyrogallol (PL), phloroglucinol (PG), phenylhydrazine (PH) or phenylenediamine (PD) was studied in acetate buffer, pH 5.6. Monitoring the formation of the iron-ferrozine complex quantitated the release of iron from ferritin. The presence of P (125 microM) did not result in the release of iron from ferritin, whereas the same concentration of CT, HQ, PL, PH or PD resulted in the release of significant amounts of iron from ferritin and a marginal amount of iron in the presence of PG, CT, HQ, PL, PH or PD concentration and time-dependent increase in iron release from ferritin were observed although the increase was not linear as a function of time and concentration of the compounds studied. The presence of superoxide dismutase inhibited significantly the release of iron from ferritin by CT, HQ, PL, PH or PD. The iron released from ferritin by CT, HQ, PL, PH or PD enhanced lipid peroxidation in rat brain homogenate and released aldehydic products from bleomycin-dependent degradation of DNA and also caused single strand nicks to pUC18 DNA. These studies indicate that CT and HQ, the two principal polyphenolic metabolites of benzene and PL, PH or PD, the superoxide radical generating compounds were capable of reducing ferric iron from ferritin and also mobilizing and releasing iron from ferritin core. The release of iron from ferritin by these compounds is a result of direct reduction of ferritin iron by electron transfer and also reduction via superoxide radical. The release of iron from ferritin by CT and HQ may have toxicological implications in relation to benzene toxicity. The release of iron by superoxide radical generating agents suggests that oxidative stress may play a role as this could lead to disruption of intracellular iron homeostasis.     

Biosynthesis of all-trans-retinoic acid from all-trans-retinol: catalysis of all-trans-retinol oxidation by human P-450 cytochromes.

Oxidative conversion of all-trans-retinol (t-ROH) to all-trans-retinal (t-RAL) is recognized as the rate-limiting step for biosynthesis of all-trans-retinoic acid from t-ROH in mammalian hepatic tissues. The purpose of this study was to investigate the role of human cytochrome P-450 (CYP)-dependent monooxygenation in the conversion of t-ROH to t-RAL. Adult human liver microsomes (HLMS) were incubated with t-ROH, and retinoids generated were identified and quantified by liquid chromatography-mass spectroscopy, HPLC, and other methods. HLMS-catalyzed generation of t-RAL from t-ROH was primarily NADPH-dependent and was strongly inhibited by carbon monoxide. Rates of reactions increased linearly with time and concentrations of HLMS, and exhibited classical substrate saturation. These observations strongly indicated that the reaction proceeded via CYP-catalyzed monooxygenation. On the basis of responses to selective chemical inhibitors, isoforms from CYP family 1 and the CYP3A subfamily appeared to be very active. Members of the CYP2C subfamily and CYP2D6 exhibited lesser activities and CYP2A6, CYP2B6, and CYP2E1 were virtually inactive. cDNA-expressed human CYP enzymes (CYP SUPERSOMES) also were used to assess the capacity of individual CYP enzymes to catalyze the reaction. Based on responses to selective chemical inhibitors, specific activities, and levels present in adult human hepatic tissues, CYP1A2 and CYP3A4 strongly appeared to be the major CYP enzymes catalyzing hepatic oxidative conversion of t-ROH to t-RAL in the adult human liver. CYP1A1 and CYP1B1 SUPERSOMES both exhibited exceptionally high activities, and in extrahepatic tissues, these isoforms could play important roles in biosynthesis of all-trans-retinoic acid from t-ROH.     

Cellular pool of transient ferric iron, chelatable by deferoxamine and distinct from ferritin, that is involved in oxidative cell injury.

A cellular pool of transient ferric iron that is chelatable by deferoxamine, distinct from ferritin, and required for oxidative cell injury has been identified in cultured rat hepatocytes labeled with 59FeCl3. Pretreatment of hepatocytes with deferoxamine depleted the cellular pool of chelatable iron and protected the cells from an oxidative injury. Incubation of deferoxamine-pretreated hepatocytes in serum-free medium restored both the chelatable iron pool and the susceptibility to oxidative injury. Furthermore, inhibition of protein degradation with chymostatin prevented the restoration of both the chelatable pool and susceptibility to oxidative injury. The deferoxamine-chelatable iron pool was distinguished kinetically and immunochemically from the larger cellular pool of ferritin iron. The labeled iron in the deferoxamine-chelatable pool was transient, unlike either the total cellular uptake of 59Fe or its incorporation into ferritin, both of which increased with time of labeling. With pulse-chase labeling, the percentage of the total uptake of 59Fe that was represented by the deferoxamine-chelatable pool decreased. At the same time, the percentage represented by radioactivity immunoprecipitable as ferritin increased. Furthermore, immunoprecipitation of ferritin from the labeled lysates enriched the resulting immunosupernatants in deferoxamine-chelatable iron. The degree of enrichment for chelatable iron correlated with the percentage of the cellular label that was immunoprecipitable as ferritin. The deferoxamine-chelatable iron appears to represent a metabolically common pool of iron that is rapidly in transit through the cell. Extracellular iron entering the pool can be utilized for heme synthesis or stored in ferritin, whereas protein degradation releases storage iron into this pool.     

An animal model of iron overload and its application to study hepatic ferritin iron mobilization by chelators

Administration of 3,5,5-trimethylhexanoyl ferrocene in the diet of male Wistar rats results in a substantial increase in hepatic ferritin protein (greater than 2-fold) and of ferritin iron (4-8-fold). The iron-loading in liver, under the conditions used, appears to be essentially in parenchymal cells rather than in reticulo-endothelial cells. It is suggested that the model represents a useful system for the study of the potential efficacy of new iron chelators for the mobilization of hepatic storage iron. The ability of desferal (DFO) and of a new siderophore, desferrithiocin (DFT), to mobilize hepatic ferritin iron is observed in this model of iron overload. Desferrithiocin stimulates ferritin iron mobilization, when administered either by gavage or by intraperitoneal injection, whereas desferal is active intraperitoneally but inactive orally. Our studies lead to the conclusion that DFT merits further examinations, for its activity as an orally active iron chelator.     

缺铁性贫血患者血清红细胞生成素 铁蛋白等相关指标及T淋巴细胞水平的检测及分析

目的分析缺铁性贫血(IDA)患者血清红细胞生成素、铁蛋白、叶酸、维生素B12及T淋巴细胞水平的变化,为该病的诊断、治疗提供临床依据。方法收集2015年7月至2017年7月我院收治的缺铁性贫血患者80例为IDA组;同期慢性疾病性贫血患者80例为ACD组;同期于我院体检的健康志愿者80名为健康对照组。检测各组患者血清中红细胞生成素、铁蛋白、叶酸、维生素B12及T淋巴细胞水平。结果 IDA组及慢性疾病性贫血(ACD)组患者血清中红细胞生成素水平显著高于健康对照组,且IDA组患者血清中红细胞生成素水平显著高于ACD组,差异均有统计学意义(P<0.05)。IDA组患者血清铁蛋白水平明显低于健康对照组,ACD组患者血清铁蛋白水平明显高于健康对照组,差异有统计学意义(P<0.05)。IDA组及ACD组患者血清叶酸、维生素B12、CD3+、CD4+、CD4+/CD8+水平显著低于对照组,CD8+水平明显高于健康对照组,差异均有统计学意义(P<0.05)。IDA组患者血清叶酸、维生素B12水平显著低于ACD组,差异有统计学意义(P<0.05)。结论缺铁性贫血会导致患者红细胞生成及铁代谢障碍,维生素水平显著降低、细胞免疫紊乱。血清铁蛋白含量可作为区分缺铁性贫血及慢性疾病性贫血的指标。     

血清铁和铁蛋白水平在海人酸诱导的大鼠癫痫模型中的意义

目的 探讨海人酸(KA)诱导的癫痫大鼠血液中铁和铁蛋白水平的变化。方法 30只SD大鼠通过随机数字表法分为模型组和对照组(每组各15只),模型组采用一次性腹腔注射KA的方式建立癫痫大鼠模型,对照组采用腹腔注射等量生理盐水,分别于给药前1周和给药后2个月取静脉血,应用原子吸收光谱仪来测定血清铁,采用放射性标记酶联免疫分析法测定血清铁蛋白,分别观察两组大鼠血清铁和血清铁蛋白的变化。结果 模型组血清铁、血清铁蛋白含量分别为(3.502±0.475) mg/L、(29.229±1.912) μg/L,均明显高于对照组的(2.408±0.760) mg/L、(23.449±1.711) μg/L(t=1.328、19.169,P=0.018 3、0.026 7)。血清铁、血清铁蛋白与癫痫发作具有较强相关性,相关系数分别为0.633、0.732(P值分别为 0.035 0、0.013 9)。结论 KA诱导的大鼠癫痫模型血液中铁和铁蛋白水平升高,可能与海人酸诱导的癫痫发作有关。     

Relationship between serum ferritin, hepatic iron staining, diabetes mellitus and fibrosis progression in patients with chronic hepatitis C

BACKGROUND: Chronic infection with the hepatitis C virus affects over 170 million individuals worldwide and 20% of patients develop cirrhosis after 20 years. Increased iron stores and hepatic iron content have been suggested to be important in fibrosis progression. The increased prevalence of diabetes mellitus has been associated with increased iron deposits in patients with chronic hepatitis C. AIM: To assess the potential relationship between serum ferritin and hepatic iron staining and liver fibrosis in patients with chronic hepatitis C virus infection and whether these factors are increased in diabetic patients with hepatitis C virus. METHODS: This was a cross-sectional, multi-centre study involving hospitals in the north-east of London between 1992 and 2003. Chronic hepatitis C patients with a liver biopsy and data concerning age, sex, basal metabolic index, diabetes mellitus or impaired glucose tolerance, alcohol intake, serum ferritin level and ethnicity were enrolled. Each biopsy was scored for fibrosis and stained for hepatic iron. RESULTS: Three hundred and thirty nine patients (200 Caucasian; 139 Asian) were enrolled. Fifty three patients had no fibrosis, 131 had mild fibrosis (stage one to two Modified Ishak), 68 moderate fibrosis (stage three to four) and 87 cirrhosis (stage five to six). 4.4% of patients had elevations in serum ferritin, whilst 11% had increased hepatic iron staining. The serum ferritin and hepatic iron staining were unrelated to the degree of fibrosis. Serum ferritin was significantly higher in patients with diabetes or impaired glucose tolerance compared to non-diabetics. No association was seen between diabetes and hepatic iron staining. CONCLUSIONS: Many patients with chronic hepatitis C virus infection may have elevated serum ferritin and/or iron deposition within the liver. However, both played no significant role in the progression of hepatitis C virus related liver injury. The association between chronic hepatitis C virus infection and type II diabetes mellitus exists, however the biological mechanism of this association still remains to be elucidated.     

Effect of dietary aluminum on tissue nonheme iron and ferritin levels in the chick

Aluminum toxicity is well documented but the mechanism of action is poorly understood. In renal failure patients with aluminum overload, disturbances in iron metabolism leading to anemia are apparent. Few animal models, however, have been used to study the effects of dietary aluminum on iron metabolism. The purpose of this study was to determine if dietary aluminum exposure alters tissue iron and ferritin concentrations in the chick, as has been found in cultured human cells exposed to aluminum. Groups of day-old chicks were fed purified diets containing one of two levels of iron (control or high iron), and one of three levels of aluminum chloride in a 2 x 3 factorial design. Diets were consumed ad libitum for 1 week, then pair-feeding was initiated for 2 more weeks. A seventh group consumed a low iron diet ad libitum for comparative purposes. After the 3-week feeding period, samples of kidney, liver, and intestinal mucosa were analyzed for nonheme iron and ferritin concentrations by a colorimetric assay and SDS-PAGE, respectively. Results showed that dietary aluminum intake reduced iron stores in liver and intestine, but had no effect on nonheme iron levels in the kidney. Ferritin levels were reduced by aluminum intake in all tissues studied. The decreases in tissue ferritin levels were proportionately more than the decreases in tissue nonheme iron levels. This resulted in increased nonheme iron to ferritin ratios that amounted to as much as 140 and 525% in kidney and intestine, respectively. These findings are consistent with the interpretation that, in the growing chick, dietary aluminum can inhibit iron absorption, disrupt the regulation of tissue ferritin levels by iron, and potentially alter the compartmentalization and protective sequestration of iron within cells.     

Salsolinol, a catechol neurotoxin, induces modification of ferritin: Protection by histidine dipeptide

1-Methyl-6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline (salsolinol), an endogenous neurotoxin present in the mammalian brain, is known to perform a role in the pathogenesis of Parkinson's disease. In this study, we evaluated oxidative modifications of ferritin occurring after incubation with salsolinol. When ferritin was incubated with salsolinol, protein aggregation increased in a time-dependent manner. Free radical scavengers inhibited this salsolinol-mediated ferritin modification. The exposure of ferritin to salsolinol also results in the generation of protein carbonyl compounds and the formation of dityrosine. The results of this study show that free radicals may perform a pivotal role in salsolinol-mediated ferritin modification. Histidine dipeptides, such as carnosine, have been proposed to function as antioxidant agents in vivo. In this study, we also attempted to determine whether the histidine dipeptides, carnosine and N-acetyl-carnosine, could prevent salsolinol-mediated oxidative modification of ferritin. Our results showed that both carnosine and N-acetyl-carnosine significantly reduced ferritin aggregation. Both compounds effectively inhibited the formation of both carbonyl compounds and dityrosine. These results suggest that carnosine derivatives can, indeed, protect against salsolinol-mediated ferritin modification, as the consequence of free radical-scavenging activity.