A Mössbauer spectroscopic study of the form of iron in iron overload

There are major differences in the temperature dependence of the Mössbauer spectra of ferritin and haemosiderin extracted from the organs of humans suffering from transfusional iron overload. Iron overload can also occur in animal systems as a result of artificial treatments or dietary factors. None of the animal systems which were investigated in the present study showed evidence in their Mössbauer spectra for the presence of the haemosiderin found in transfusional iron overload in humans. This suggests that the haemosiderin which occurs in the case of human transfusional iron overload may be specific to that situation.     

M?ssbauer spectroscopic studies of deproteinised, sub-fractionated and reconstituted ferritins: the relationship between haemosiderin and ferritin

In a previous study of human haemosiderin and ferritin by a combination of M?ssbauer spectroscopy and electron microscopy, it was observed that the M?ssbauer spectra of haemosiderin showed a very different temperature dependence to those of ferritin. These differences were related to the superparamagnetic behaviour of small particles of a magnetic material and suggested that the magnetic anisotropy constant of the haemosiderin was considerably larger than that of the ferritin. In the present work, samples of ferritin have been examined by M?ssbauer spectroscopy following partial deproteinisation, subfractionation, and reconstitution with and without phosphate, in order to investigate whether these procedures lead to changes in the magnetic anisotropy constant of the iron-containing cores. There is no evidence from the present data that changes in the protein shell, in the size of the iron-containing cores of ferritin, or in the phosphate content lead to any significant changes in the magnetic anisotropy constant, as obtained from the temperature dependence of the M?ssbauer spectra. These results indicate that the different magnetic anisotropy constant observed in the case of human haemosiderin resulting from transfusional iron overload must arise from other significant differences in the composition or structure of the iron-containing cores.     

Mössbauer spectroscopic studies of deproteinised,sub-fractionated and reconstituted ferritins: the relationship between haemosiderin and ferritin

In a previous study of human haemosiderin and ferritin by a combination of Mössbauer spectroscopy and electron microscopy, it was observed that the Mössbauer spectra of haemosiderin showed a very different temperature dependence to those of ferritin. These differences were related to the superparamagnetic behaviour of small particles of a magnetic material and suggested that the magnetic anisotropy constant of the haemosiderin was considerably larger than that of the ferritin. In the present work, samples of ferritin have been examined by Mössbauer spectroscopy following partial deproteinisation, subfractionation, and reconstitution with and without phosphate, in order to investigate whether these procedures lead to changes in the magnetic anisotropy constant of the iron-containing cores. There is no evidence from the present data that changes in the protein shell, in the size of the iron-containing cores of ferritin, or in the phosphate content lead to any significant changes in the magnetic anisotropy constant, as obtained from the temperature dependence of the Mössbauer spectra. These results indicate that the different magnetic anisotropy constant observed in the case of human haemosiderin resulting from transfusional iron overload must arise from other significant differences in the composition or structure of the iron-containing cores.     

Electron spin resonance studies of splenic ferritin and haemosiderin

Preparations of haemosiderin and ferritin isolated from iron-loaded human spleens were studied by electron spin resonance (ESR) spectroscopy at X-band (approx. 9.2 GHz). The spectra were mainly composed of two overlapping, broad features, one extremely anisotropic with its major component occurring at 0.1-0.2 T (feature A), the other nearly isotropic and occurring at around g = 2 (feature B). There is relatively more feature A and less feature B in ferritin than in haemosiderin. Both features originate from the iron oxyhydroxide crystallites of these iron proteins which, due to their small size, are superparamagnetic. Feature B is maximal in small cores or at high temperatures, where superparamagnetic fluctuations average out anisotropic magnetic interactions; feature A is greatest at low temperatures or in large cores, for which such fluctuations are blocked and an ESR spectrum characteristic of a magnetically ordered system is observed. It is concluded that there is no evidence in the ESR spectra for 'loose' protein-bound Fe3+ in ferritin or haemosiderin, and that haemosiderin cores are on average smaller than those of ferritin. The relationship of the ESR spectra between these two proteins supports the view that haemosiderin is derived from ferritin.     

Biochemical studies on the isolation and characterization of human spleen haemosiderin.

Haemosiderin was isolated from thalassaemic human spleens by centrifugation through concentrated KI solutions. A method for solubilizing haemosiderin was developed which leaves the iron oxyhydroxide cores and constituent polypeptides intact, facilitating further purification and analysis. Purified haemosiderin contained no detectable haem, trace amounts of carbohydrate, and iron and phosphorus in a molar ration of 6:1; much of the phosphate may be present as core-adsorbed. Several lipids were present, but it is not certain whether these are contaminants or components of the haemosiderin granules. In all preparations examined, a characteristic group of six to seven peptides of apparent Mr 12 900-17 800 were found, with a major band at Mr 14 500 and, in addition, a minor component of Mr 42 000; these peptides co-chromatographed with the cores. Negatively stained electron micrographs suggest that these peptides form an incomplete shell about the cores, consistent with the view that haemosiderin is a proteolytic product of ferritin.     

M?ssbauer spectroscopy, electron microscopy and electron diffraction studies of the iron cores in various human and animal haemosiderins

M?ssbauer spectroscopy has indicated significant differences in the iron-containing cores of various haemosiderins. In the present study, haemosiderin was isolated from a number of animal species including man. In addition, haemosiderin was isolated from patients with primary idiopathic haemochromatosis or with secondary (transfusional) iron-overload. The iron cores of the animal and normal human haemosiderin appear to be very similar by M?ssbauer spectroscopy, and the electron diffraction data indicate a ferrihydrite structure similar to that of ferritin cores. The haemosiderin isolated from secondary iron-overload shows anomalous behaviour in its temperature-dependent M?ssbauer spectra. This can be understood in terms of the microcrystalline goethite structure of the cores as indicated by electron diffraction. The haemosiderin cores obtained in the case of primary haemochromatosis have an amorphous Fe(III) oxide structure and show M?ssbauer spectra characteristic of a magnetically disordered material, which only orders at very low temperatures.     

Biochemical and biophysical investigations of the ferrocene-iron-loaded rat. An animal model of primary haemochromatosis.

Male Wistar rats fed with ferrocene had high hepatic iron loading (7.24 +/- 1.97 mg Fe/g tissue) after 6 weeks, principally located in lysosomes, which was comparable to the levels and distribution determined in human haemochromatosis. The two iron-storage proteins, ferritin and haemosiderin were isolated from the livers of the ferrocene-loaded rats and their iron cores were investigated by M?ssbauer spectroscopy and inductively coupled plasma-emission spectrometry. Ferrihydrite was the predominant form of iron present in both ferritin and haemosiderin, while haemosiderin contained higher amounts of phosphorus, magnesium, calcium and barium, then either normal or ferrocene-loaded ferritin. Free-radical-mediated damage in the iron-loaded livers was inferred by the significant depletion of alpha-tocopherol in both the livers and subcellular hepatic lysosomal fraction, which inversely correlated with the increasing iron content (r = -0.61; P less than 0.05) and was associated with increased fragility of the lysosomal membranes.     

Transport of siderotic Kupffer cells to the lung and bronchial excretion of iron in experimental iron-overload.

In 3,5,5-trimethylhexanoylferrocene-induced iron overload of rats, three different types of iron-loaded macrophages and derivatives thereof were found in the lungs. On the basis of their localization and of their pattern of iron load it was possible to distinguish: (1) Resident macrophages, showing an alveolar localization and a moderate iron content represented by lysosomal ferritin and haemosiderin. (2) Liver-derived macrophages and giant cells, as well as fragments of them. They showed an exclusive localization in capillaries and alveolar septa, and high concentrations of free ferritin molecules in addition to polymorphous ferritin- and haemosiderin-containing siderosomes. (3) Monocyte-derived intravascular pulmonary macrophages. Initially, they contained iron only as lysosomal aggregates of ferritin and haemosiderin, as a result of phagocytosis of liver-derived macrophageal cell fragments. Later in iron overload, they also showed free ferritin molecules in the cytosol and fused intrapulmonarily to giant cells. The resident as well as the liver-derived siderotic pulmonary macrophages provide a way for iron excretion through the airways.     

Exceptional iron concentrations in larval lampreys (Geotria australis) and the activities of superoxide radical detoxifying enzymes.

This study aimed to elucidate the way in which larvae of the lamprey Geotria australis counteract the potential problems of the very high concentrations of non-haem iron they contain and thereby avoid the deleterious effects associated with iron overload in other vertebrates. Particular attention has been paid to ascertaining whether increasing concentrations of iron are accompanied by (i) change to a less readily available form of iron and (ii) an increase in the activity of those detoxifying enzymes responsible for minimizing the production of harmful hydroxyl radicals via the Haber-Weiss reaction. The mean concentrations of haemosiderin and ferritin in larval G. australis were each far higher in the nephric fold than in either the liver or intestine, but all these concentrations were much greater than those in rat liver. Since haemosiderin releases iron far more slowly than ferritin, the iron it contains is much less readily available to catalyse the Haber-Weiss reaction. It is thus relevant that (i) non-haem iron in the nephric fold occurred to a greater extent as large dense haemosiderin granules than as ferritin molecules and (ii) the proportion of iron in the form of haemosiderin rose with increasing concentration of total non-haem iron. A strong correlation was also recorded between the activity of superoxide dismutase in the nephric fold and the concentrations of total non-haem iron and its haemosiderin and ferritin components. This demonstrates that enzyme detoxification of O2.- rises with increasing amounts of iron. The exceptional iron concentrations in the nephric fold were not reflected by a greater measured activity of superoxide dismutase than that found in other tissues. However, the nephric fold was shown to contain an augmentation factor which is presumed to enhance the activity of this enzyme in vivo. The activity of catalase and glutathione peroxidase, which catalyse the breakdown of H2O2 to O2 and water, were each significantly correlated with the concentration of ferritin.     

Formation of hydroxyl radicals in the presence of ferritin and haemosiderin. Is haemosiderin formation a biological protective mechanism?

Horse spleen and human spleen ferritins increase the formation of hydroxyl radicals (OH) at both pH 4.5 and pH 7.4 in reaction mixtures containing ascorbic acid and H2O2. The generation of OH is inhibited by the chelator desferrioxamine. Human spleen haemosiderin also accelerates OH generation in identical reaction mixtures, but is far less effective (on a unit iron basis) than ferritin under all reaction conditions. It is proposed that conversion of ferritin into haemosiderin in iron overload is biologically advantageous in that it decreases the ability of iron to promote oxygen-radical reactions.     

Studies on haemosiderin and ferritin from iron-loaded rat liver

Summary Haemosiderin has been isolated from siderosomes and ferritin from the cytosol of livers of rats iron-loaded by intraperitoneal injections of iron-dextran. Siderosomal haermosiderin, like ferritin, was shown by electron diffraction to contain iron mainly in the form of small particles of ferrihydrite (5Fe₂O₃ · 9H₂O), with average particle diameter of 5.36±1.31 nm (SD), less than that of ferritin iron-cores (6.14±1.18 nm). Mössbauer spectra of both iron-storage complexes are also similar, except that the blocking temperature,T _B, for haemosiderin (23 K) is lower than that of ferritin (35 K). These values are consistent with their differences in particle volumes assuming identical magnetic anisotropy constants. Measurements of P/Fe ratios by electron probe microanalysis showed the presence of phosphorus in rat liver haemosiderin, but much of it was lost on extensive dialysis. The presence of peptides reacting with anti-ferritin antisera and the similarities in the structures of their iron components are consistent with the view that rat liver haemosiderin arises by degradation of ferritin polypeptides, but its peptide pattern is different from that found in human-thalassaemia haemosiderin. The blocking temperature, 35 K, for rat liver ferritin is near to that reported, 40 K, for human-thalassaemia spleen ferritin. However, the haemosiderin isolated from this tissue, in contrast to that from rat liver, had aT _B higher than that of ferritin. The iron availability of haemosiderins from rat liver and human-thalassaemic spleen to a hydroxypyridinone chelator also differed. That from rat liver was equal to or greater, and that from human spleen was markedly less, than the iron availability from either of the associated ferritins, which were equivalent. The differences in properties of the two types of haemosiderin may reflect their origins from primary or secondary iron overload and differences in the duration of the overload.     

Sulfide silver amplification of ferritin iron cores in blood and bone marrow cells. Methods, adaptations to microphysical analyses, and the impact of advanced iron overload

A short exposure of cell suspensions to gaseous hydrogen sulfide, appropriate fixations, and subsequent physical development of silver shells around sulfidated insoluble metals were used to amplify ferritin iron cores in blood and bone marrow cells. The methods described are suitable for both light microscopy and transmission electron microscopy. These techniques made it possible to visualize Prussian Blue stainable ferritin and haemosiderin, as well as a large variety of isoferritin iron and other smaller particles beyond the sensitivity of Prussian Blue staining. Admixtures of sulfidatible zinc and traces of other heavy metals had to be taken into consideration. For further research, adaptations of sulfide silver staining to microphysical analyses were developed. However, conventional energy dispersive X-ray analysis was not sensitive enough to signalize the presence of Fe in sulfide silver amplified iron cores of a single or a few ferritin molecule(s). Proton-induced X-ray emission was used to measure Fe and Zn down to 1 fg/single cell in unstained or sulfide silver stained smears on thin foils. However, multielement analysis of homogeneous cell concentrates was much easier to perform and far more sensitive. In advanced iron overload, highly increased sulfide silver staining was found in peripheral blood cells including lymphocytes, monocytes, eosinophils, basophils, and--in extreme cases--also in neutrophils and platelets.     

Synthetic methods and in vitro iron binding studies of the novel 1-alkyl-2-ethyl-3-hydroxypyrid-4-one iron chelators

Three novel iron chelators namely the 1-methyl-, 1-ethyl- and 1-propyl-2-ethyl-3-hydroxypyrid-4-ones were prepared in high yields from ethyl maltol and the related alkylamine in a one step reaction. These chelators formed 3 chelator:1 iron stable, coloured, neutral complexes at physiological pH and mobilise iron from transferrin, ferritin and haemosiderin. The rate of iron mobilisation from these proteins was of the order transferrin > haemosiderin > ferritin. The cheap synthesis and strong iron binding properties of the 1-alkyl-2-ethyl-3-hydroxypyrid-4-ones at physiological pH requires the need for further investigation and development of these compounds and their homologues, for the treatment of iron overload and other diseases of iron imbalance and toxicity.     

Iron release from haemosiderin and ferritin by therapeutic and physiological chelators.

Iron release from both human and horse spleen haemosiderin to desferrioxamine was substantially less than that released from ferritin samples. This finding contradicts a previous report [Kontoghiorges, Chambers & Hoffbrand (1987) Biochem. J. 241, 87-92]. Differences in phosphate content of cores and in core size between haemosiderin and ferritin did not account for the different iron-release rates. Iron released to acetate was found to stimulate lipid peroxidation in liposomes, whereas that released to stronger chelators such as citrate and desferal did not. Absorption spectra and gel-filtration studies suggest that the acetate-solubilized iron was in the form of low-molecular-mass (less than 5 kDa) ferrihydrite fragments.     

Depot iron in the rat liver after hypertransfusion with rat erythrocytes

In the rat liver the deposition of iron was measured after hypertransfusion with rat erythrocytes. The liver iron fractions were studied during four weeks after the hypertransfusions. In the first week the haemosiderin iron fraction increased together with the ferritin iron fraction. Most iron was deposited as ferritin iron. In the last week of the experiments, while the ferritin iron fraction still increased, the haemosiderin iron fraction decreased. At the same time plasma iron was utilized when erythropoiesis, which had been suppressed by the hypertransfusion, recommenced. It is suggest that, under these experimental conditions, liver haemosiderin iron is used in haemoglobin synthesis.     

Iron overload of the bone marrow by trimethylhexanoyl-ferrocene in rats.

Iron-deficient female Wistar rats were fed a diet which contained 0.5% 3,5,5-trimethylhexanoyl (TMH)-ferrocene over a 57-week period. The state of iron deficiency was characterized by means of the absence of stainable iron in the bone marrow. After the first days on the iron-enriched diet, ferritin-containing siderosomes were found, in numerous erythroblasts up to orthochromatic normoblasts and in reticulocytes, i.e. the dispensed iron was used for haemoglobin synthesis. After 1 week the first macrophages showed a positive Perls' Prussian blue reaction. In the cytoplasm they stored the iron in the form of free ferritin molecules and lysosomally as aggregated ferritin and/or haemosiderin. The iron loading of the macrophages increased in both of the storage qualities proportionally with duration of the feeding period and reached a maximum after 38 weeks. Final stages showed extremely iron-loaded macrophages with high concentrations of free ferritin molecules and large siderosomes, partially flowing together to still greater units. Iron deposits within endothelial cells of bone marrow sinusoids can be observed for the first time after 4 weeks. In these cells the iron is stored as ferritin in siderosomes of relatively small and uniform size; free ferritin molecules in the cytosol were of only slight concentration. The TMH-ferrocene model of iron overload shows in the bone marrow: (1) an unimpeded utilization of the iron component for erythropoiesis, (2) development of excessive iron overload of the bone marrow in macrophages and endothelial cells of sinusoids and (3) a pattern of distribution of iron as seen in secondary haemochromatosis.     

Changes in the amount of nonhaem iron in the plasma, whole body, and selected organs during the postlarval life of the lamprey Geotria australis

The concentration of plasma nonhaem iron and the concentration and weight of all nonhaem iron in the whole body and selected organs, together with its partitioning into ferritin and haemosiderin iron, have been measured during the metamorphosis and upstream spawning migration of the Southern Hemisphere lamprey Geotria australis. Some nonhaem iron was lost from the animal during metamorphosis. However, the concentration and weight of nonhaem iron in the liver rose sharply at this time, following its release from important storage sites in adipose tissue and the degradation of larval haemoglobins. The nephric fold of larval and metamorphosing stages contained over 40% of all nonhaem iron in the body at the commencement of metamorphosis. This was predominantly in the form of haemosiderin. While the rise in liver iron during the transition from larva to adult primarily reflected an increase in the weight of ferritin iron, the amount of iron stored as haemosiderin rose conspicuously towards the end of metamorphosis. The rise in ferritin iron in the liver was accompanied by a decrease in ferritin iron in the plasma, which implies that changes in the liver during metamorphosis result in a greater filtering of circulating ferritin. Such a process would account for the very much lower plasma nonhaem iron concentrations which characterise later adult stages. The weight of nonhaem iron increased markedly in the liver and adult opisthonephros and in the whole animal during the nontrophic upstream spawning migration. This was primarily due to a marked rise in ferritin which in turn could be related to the degradation of adult haemoglobins.(ABSTRACT TRUNCATED AT 250 WORDS)     

Iron K-edge absorption spectroscopic investigations of the cores of ferritin and haemosiderins.

The extended X-ray absorption fine structure (EXAFS) associated with the iron K-edge has been measured and interpreted for ferritin and haemosiderin extracted from horse spleen, and haemosiderin extracted from the livers of humans with treated primary haemochromatosis, and from the spleens of humans with treated secondary haemochromatosis. For ferritin, the data are consistent with, on average, each iron atom being in an environment comprised of approx. six oxygen atoms at 1.93 +/- 0.02 A, approx. 1.5 iron atoms at 2.95 +/- 0.02 A and approx. 1.1 iron atoms at 3.39 +/- 0.02 A, with a further shell of oxygens at approx. 3.6 A. Iron in horse spleen haemosiderin is in an essentially identical local environment to that in horse spleen ferritin. In contrast, the EXAFS data for primary haemochromatosis haemosiderin indicate that the iron-oxide core is amorphous; only a single shell of approx. six oxygen atoms at approx. 1.94 +/- 0.02 A being apparent. Secondary haemochromatosis haemosiderin shows an ordered structure with approx. 1.4 iron atoms at both 2.97 +/- 0.02 and 3.34 +/- 0.02 A. This arrangement of iron atoms is similar to that in horse spleen haemosiderin, but the first oxygen shell is split with approx. 2.9 atoms at 1.90 +/- 0.02 A and approx. 2.7 at 2.03 +/- 0.02 A, indicative of substantial structural differences between secondary haemochromatosis haemosiderin and horse spleen haemosiderin.     

The enhancement of iron-dependent luminol peroxidation by 2,2′-Dipyridyl and nitrilotriacetate

The generation of radicals from luminol and H₂O₂, in the presence of iron and iron chelates was monitored by measuring the chemiluminescence produced by further oxidation of these radicals. 2,2′-Dipyridyl enhanced the production of chemiluminescence in the presence of FeSO₄, farritin and haemosiderin but not FeCI₃ or horseance of both FeSO₄ and FeCI₃ but not ferritin or haemosiderin. The enhancement of chemiluminescence by iron chelation may have analytical applications and the process by which these iron chelates are able to generate radicals from the nitrogenous base luminol may be similar to that responsible for their toxic effects on DNA.