Abstract
Background/Aim: Our recent studies have indicated that trace copper co-existed with iron in hemosiderin particles of human genetic iron overload. To understand this phenomenon, we analyzed hemosiderin particles in iron-overloaded rat liver by using scanning transmission electron microscopy - energy-dispersive X-ray (STEM-EDX) spectroscopy. Materials and Methods: Samples for STEM-EDX spectroscopy were prepared from the liver of rats administered an intraperitoneal injection of dextran iron. Results: The micro-domain analysis with STEM-EDX spectroscopy showed that dense bodies contained high levels of iron and trace copper. Quantitative analysis of copper levels in the liver specimen using atomic spectrophotometry showed that copper concentration in the liver was not increased by iron overload. These findings suggest that the overload of iron induced distribution of trace copper to hemosiderin particles without changing cellular copper levels. Conclusion: Co-existence of copper with iron was observed in hemosiderin particles of the liver of an experimental model of iron overload, suggesting that iron overload induced distribution of trace copper into hemosiderin particles.
Iron is an essential element for the body and adults have approximately 3-4 g. About half of the body iron accumulates in the liver for inducing hematopoiesis, while its entry into the body is strictly regulated by the iron-regulatory system including hepcidin, which is mainly released from the liver. Daily excretion of iron from the body is estimated to be approximately 1 mg, and therefore, the excess administration of iron results in iron-overload (1). If iron overload persists, oxidative damage to various organs including the pancreas, heart, and liver may occur by reactive oxygen species produced by the Haber-Weiss and Fenton-reactions (2).
In cells, ferritins form a spherical cage made of 24 molecules and incorporate the excess of cellular ferrous iron in inner space to be stored as ferric hydrate, and thus it protects cells from the strong toxicity of ferrous iron (3, 4). The number of ferritin molecules is increased depending upon biosynthesis in response to iron concentration. However, ferric iron in ferritin can be mobilized as soluble ferrous iron in either iron-replete or -depleted condition by the peculiar action of ferritin or digestion by mechanisms including autophagy (5, 6). Thus, ferritin iron plays important roles in iron homeostasis in the whole body.
Under iron-overload condition, iron-rich ferritin particles accumulate in lysosomes and form large deposits, called hemosiderin. The iron-rich deposits are observed in classical hereditary hemochromatosis and secondary hemochromatosis (e.g., sideroblastic anemia treated with frequent transfusion). In those cases, hemosiderin is observed primarily in liver parenchymal cells, while in the case of ferroportin disease caused by a mutation of ferroportin, an exporter of ferrous iron from the cells, hemosiderin is mostly found in macrophages and macrophage-like cells. In the later disease no remarkable histopathological change is observed except hemosiderin deposited in Kupffer cells (1, 7). Ferritin cage sequesters excess iron from the cytosol and also it is reportedly known that ferritin can bind small amounts of other divalent cations, such as Zn2+, Cd2+ and Cu2+(8). We recently reported that in the liver of patients with hereditary hemochromatosis, ferroportin disease, or secondary iron overload by aceruloplasminemia, hemosiderin particles contained low levels of copper (1, 9-11).
To clarify whether the iron excess condition induces copper mobilization in hemosiderin particles, in this study, we developed an iron overload rat model by intraperitoneal administration of iron dextran as already reported (12-14). In this model, we investigated copper distribution in macrophage-like cell hemosiderin particles using the STEM-EDX spectrometry system and visualized the distribution of iron and copper in hemosiderin particles.
Materials and Methods
Animals. Male Wistar rats (8 weeks; Japan SLC, Hamamatsu, Japan) had free access to water and chow (CE-2, CLEA Japan, Tokyo, Japan). Dextran iron (Asuka Pharmaceuticals, Tokyo, Japan) was diluted with distilled water and administrated intraperitoneally on Monday, Wednesday, and Friday at 10 mg/kg body weight/day 6 times and then given 20 mg/kg body weight/day 5 times. After 6 months from the last administration, the liver was removed from the rats under anesthesia with pentobarbital. Liver specimen was cut into small sections, which were washed in saline and prepared for histochemistry, non-heme iron assay, copper assay, and STEM-EDX spectrometry.
This study was approved by the institutional animal experiment committee (Nagoya University) and conducted accordance with the institutional guidelines for animal experiment.
Assay of non-heme iron and copper. Liver specimen (about 2 mm3) was dried in an oven at 80°C for over 4 h and weighed after cooling down to room temperature. For non-heme iron assay, the specimen was immersed in 1 ml of acid solution containing 10% trichloroacetic acid and 3 M hydrochloride and the acid soluble non-heme iron was extracted at 65°C for 20 h under shaking. Non-heme iron in the extract solution was assayed using chelate assay with bathophenanthroline sulfate (BPS, Sigma Japan, Tokyo, Japan), as reported previously (15).
The copper assay was performed according to previous reports (16, 17) using acid digested liver tissue. Briefly, a few mg of dried liver sample was weighed into a PT-25 PTFE container (San-ai Kagaku, Nagoya, Japan) washed with 6 N HCl and rinsed with Milli-Q water (Merck, Tokyo, Japan). An acid mixture consisted with 3 ml of nitric acid, 0.3 ml of HCl, 0.3 ml of perchloric acid and 0.15 ml of hydrogen fluoride was added to the container, which was put into outer PTFE vessel, and 1.5 ml of 1.0 M NaOH was added to the vessel. The PTFE vessel was tightly sealed in a polypropylene jacket (PP-25, San-ai Kagaku) by tightening with a torque wrench (torque setting: 18N·m), and was heated in a beaker containing 50 ml of water for 5 min in a microwave oven at 200 W. After removal of the beaker, the sample vessel was heated further 3.5 min. The acid digested sample in a PT-25 container was completely dried on a hot-plate at 85°C overnight in a fume chamber avoiding contamination with room air. The residue was dissolved in 1.0 ml of 0.1 M perchloric acid. The copper concentration in the sample was measured using atomic absorption spectrometer (Agilent AA280Z, Agilent Technology, Tokyo, Japan).
Histochemical study. Liver specimen was fixed with 4% paraformaldehyde and embedded in paraffin blocks for standard histology. Hematoxylin eosin (H-E) staining was used for morphological observation, Berlin blue staining for iron detection and rhodanic acid staining for copper detection. These were conducted by The Tohkai Cytopathology Institute (Gifu, Japan).
Electron microscopic study and X-ray microanalysis. About 1 mm3 liver specimen was fixed with 2% glutaraldehyde in 0.1 M phosphate buffer (PBS), washed with PBS, dehydrated with ethanol, which was then replaced by propylene oxide and also dehydrated with propylene oxide, followed by embedding in epoxy resin (TAAB 812, TABB Laboratories Equipment, Berkshire, UK). Ultrathin sections were mounted on gold TEM grid (HR24 200 mesh, Pyser-SGI, Edenbridge, UK) as previously reported (18). Uranyl stained sections were examined under transmission electron microscopy (TEM), and unstained sections were analyzed using STEM-EDX system (JEM-1400PLUS, JOEL, Japan). Two dimensional image analysis was performed using Image J (National Institute of Mental Health, Bethesda, MD, USA) (19). The specific Kα X-rays of iron and copper were identified using an auto-analysis system of TEM-EDX.
Statistical analysis. Student’s t-test was performed using statistical function of Excel 2010 for Windows (Microsoft Co., Seattle, WA, USA). Values are expressed as mean±SD.
Results
Histochemical and electron-microscopic studies of the rat liver. After 6 months of the final intraperitoneal administration of iron-dextran in rats, iron toxicity to rat hepatocytes was not observed by histochemical study with H-E staining (Figure 1). Iron, visualized with Berlin blue staining, was observed as hemosiderin iron in the sinusoide Kupffer cells but not in hepatocytes. Copper staining using rhodanic acid was negative in both cells.
Histological studies on rat liver. Six months after iron loading, liver tissues were observed using H-E staining, Berlin blue staining for iron, and Rhodanine staining for copper. (A) H-E staining, (B) Berlin blue staining, (C) Rhodanine staining.
TEM images of the rat liver stained with uranyl acetate revealed the presence of dense body primarily in the sinusoidal region and not in hepatocytes in two specimens from an iron-dextran administrated rat as shown in Figure 2. These indicated that most of the administrated iron was not taken up by hepatocytes but by macrophage-like cells, Kupffer cells.
Transmission electron microscopy (TEM) images of the rat liver. Ultrathin sections of rat liver were stained with uranyl acetate, and sinusoidal Kupffer cells and surrounding liver tissues were observed using TEM.
Concentrations of non-hem iron and copper in the liver. Metal contents in the liver were analyzed using liver specimens removed 6 months after the final administration of iron dextran. Total administration weight of iron was 160 mg per head. As shown in Table I, non-heme iron content in control rats was 16.3±2.5 μg/100 mg dry weight. That of iron-loaded rats increased about 13-fold to 210.3±47.0 μg/100 mg dry weight.
Concentration of non-heme iron and copper in the rat liver.
Copper contents in control rats and iron-loaded rat livers were 4.13±2.85 and 1.71±1.31 μg/100 mg dry weight. The deviations of copper concentration were relatively large, and there was no statistically significant difference in copper concentrations between control and iron-loaded rats.
X-ray microanalysis of hemosiderin particles. Two-dimensional EDX analyses along with scanning TEM of dense bodies revealed that they consisted of large amount of iron, sulfur and phosphorus, and trace of copper (Figure 3). These indicated copper coexisted with iron in hemosiderin particles. However, whether iron and copper exist as compounds containing sulfur or phosphate is unclear.
Energy-dispersive X-ray images of rat hepatic hemosiderin 6 months after iron-dextran loading. Two-dimensional distributions of iron (Fe K), phosphorus (P K), oxygen (O K), copper (Cu K) and sulfur (S K) are shown. Dense bodies contained a large amount of iron, sulfur and phosphorus and trace amount of copper. BF is a bright field image of scanning transmission electron microscope (STEM). K is the Kα line.
Regional analyses with Image J of two-dimensional EDX images of hemosiderin particles showed heterogeneity of iron concentration among hemosiderin particles (Figure 4). Copper levels in hemosiderin area were low, but tended to be higher than those in non-hemosiderin area. Changes of copper concentration might be smaller than those of iron between hemosiderin particles, indicating the proportion of iron to copper in hemosiderin particles is heterogeneous (Figure 4). Microanalysis also indicated this and none of other heavy metals were detected using X-ray spectrometry microanalysis of hemosiderin (Figure 5).
Analysis with Image J of energy-dispersive X-ray (EDX) image of hemosiderin particles. Two dimensional EDX images of hemosiderin particles were analyzed using Image J at 8 bit 256 gradations. Upper: The area surrounded by a yellow frame, indicated by an arrow, was analyzed using Image J. Panel C shows area measured to obtain a background value. Lower: Results are expressed as the mean±S.D. of gray value distribution. *p<0.01, **p<0.001.
Scanning transmission electron microscopy-BF(frame 1) (STEM-BF) images of hemosiderin and micro region elemental analysis. Upper: STEM-BF image of hemosiderin was shown. A region shown by an arrow in the figure was taken as an analysis target, and a micro region element analysis was carried out. Panel C shows the area measured to obtain a background value. Table: Results of micro domain element analysis in the hemosiderin region (A, B) and non-hemosiderin region (C) are shown.
Occasionally an inclusion, which is a large aggregate of hemosiderin particles, was observed as shown in Figure 6. As seen on the bright field (BF) panel, scanning TEM revealed inclusions consisting of a high-density core region and low density surroundings, as reported previously in hepatocyte nuclei of mice (20). EDX analyses of the hemosiderin particles forming the inclusions indicated that the core region was an aggregate of hemosiderin containing high levels of iron and small amount of copper. However, the peripheral area appeared to consist of a low density aggregation of small hemosiderin particles, which contained moderate levels of iron and small amount of copper (Figure 7). Thus, the iron concentration was higher in the core region than in the periphery, while copper concentration appeared almost at the same level.
Two-dimensional energy-dispersive X-ray analyses of inclusions observed in the rat liver. Two dimensional distribution of iron (Fe K), phosphorus (P K), oxygen (O K), copper (Cu K) and sulfur (S K) is shown. The ratio of iron and copper is different between the core region and the surrounding region. BF is a bright field image of scanning transmission electron microscope (STEM). K is the Kα line.
Regional analyses using Image J of energy-dispersive X-ray (EDX) images of inclusions observed in the rat liver. Two dimensional EDX images of hemosiderin particles were analyzed using Image J at 8 bit 256 gradations. Upper: The area surrounded by a yellow frame, indicated by an arrow, was analyzed using Image J. Panel C shows area measured to obtain the background value. Table: Results are expressed as the mean±S.D. of gray value distribution. *p<0.01, **p<0.001.
Discussion
In this study, parenteral administration of iron dextran induced iron overload in macrophage like cells, Kupffer cells, in the rat liver, while liver parenchymal cells did not show iron overload. This agrees with the results of a previous report by Carthew et al. (13), although a large dosage of iron reportedly induced iron overload in hepatocytes (14).
Kupffer cell dominant iron-overload in the liver is a characteristic of ferroportin disease (1). Small amount of copper accumulated in hemosiderin particles of Kupffer cells in patients with ferroportin disease, in liver parenchymal cells of hereditary hemochromatosis including hemojuvelin-hemochromatosis and transferrin receptor 2-hemochromatosis, or aceruloplasminemia (9, 21).
In this study, Kupffer cells in iron-administered rats showed many hemosiderin particles containing large amount of iron and small amount of copper. Thus, hemosiderin particles containing large amount of iron appeared to acquire small amount of copper. Hemosiderin is formed in lysosome where ferritins incorporate excess iron forming a ferric hydrate core in the inner space of the cage. Several previous reports (22-24) have reported that ferritin contained copper and zinc without injection of external metal ions (Pb, Cu, Cd, Zn and Be) or by in vitro co-incubation with those metals. Binding experiments indicated both ferritin and apoferritin can bind almost similar amounts of copper, although zinc is bound to ferritin more than to apoferritin (5, 24). Thus, ferritin protein can bind copper ion independently of iron, while zinc binding is promoted by iron. However, their precise binding sites are not identified and it is not clear whether they are the outside or inside of ferritin cages. X-ray spectrometry analysis of hemosiderin did not detect zinc specific signal in this study, meaning that iron does not induce zinc binding to ferritin in vivo, although it has been indicated that excess dietary zinc inhibited deposition of iron into the cellular ferritin in a dose-dependent manner and that ferritin protected zinc toxicity (23, 25, 26).
However, no obvious evidence of copper accumulation in iron-injected animals can be found in the literature. This novel evidence of copper/iron co-existence in hemosiderin particles in iron-injected animals raises new question about its roles in the function of ferritin and effects on iron metabolism. Ferritin iron can be released as soluble ferrous ion by its catalytic action and by digestion in lysosomes through autophagy and/or ferritinophagy (6, 27). Considering the fact that hemosiderin is formed in lysosomes, hemosiderin appears to be affected by lysosome digestive enzymes as in the latter cases. As seen in Figure 4 and Figure 5, heterogenicity of hemosiderin iron/copper proportion and the difference in that proportion and iron density between the core and periphery in inclusion bodies (Figure 6 and Figure 7) seems to reflect the difference in the iron and copper content of ferritin taken up by lysosomes.
In the cytosol, iron exists as toxic ferrous ion, and protection is provided by its binding to iron chaperons PCBP1 and PCBP2 and delivery to ferritin (28) and membrane iron transporter proteins, DMT1 and ferroportin (29, 30). Thus, in case of mobilization of iron in cells, ferric ion is reduced to ferrous ion.
Copper is a transition metal element similar to iron. Owing to this characteristic, copper functions as a cofactor in some redox enzymes including superoxide dismutase 1 (SOD1; Cu, Zn-SOD), and hephaestin and ceruloplasmin (ferroxidase), which are involved in iron metabolism, and/or removal of superoxide by binding to ferritin as reported previously (31).
Although the amount of copper in dense bodies was small, copper may play an important role, like hephaestin and ceruloplasmin that have ferroxidase activity, upon incorporation of iron into ferritin or release of iron from ferritin. However, to clarify if such relate to copper distribution, protein analysis of dense bodies is required.
In summary, accumulation of copper in lysosomal hemosiderin was ascertained in a rat iron overload model by using STEM-EDX spectroscopy. There was a difference in the proportion of copper and iron in hemosiderins and within the inclusion bodies, suggesting that there is mobilization of copper into lysosome hemosiderin under iron-overload condition.
Footnotes
Authors’ Contributions
Ryoji Koide contributed X-ray microanalysis and preparation of this manuscript. Ryota Shigemasa contributed X-ray microanalysis. Katsunori Hashimoto contributed to the histochemical study. Yasuaki Tatsumi and Hisao Hayashi contributed to the atomic spectrometric analysis of copper. Takayoshi Suzuki contributed the analysis of non-heme iron. Shinya Wakusawa contributed the analysis of non-heme iron and preparation of this manuscript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflicts of Interest
All Authors declare that there are no conflicts of interest in relation to this study.
- Received August 3, 2023.
- Revision received September 9, 2023.
- Accepted September 15, 2023.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.