Iron storage in human disease: Fractionation of hepatic and splenic iron into ferritin and haemosiderin with histochemical correlations

Biopsy-based optimization and calibration of a signal-intensity-ratio-based MRI method (1.5 Tesla) in a dextran-iron loaded mini-pig model, enabling estimation of very high liver iron concentrations

Objective

Magnetic resonance imaging (MRI)-based techniques for non-invasive assessing liver iron concentration (LIC) in patients with iron overload have a limited upper measuring range around 35 mg/g dry weight, caused by signal loss from accelerated T1-, T2-, T2* shortening with increasing LIC. Expansion of this range is necessary to allow evaluation of patients with very high LIC.

Aim

To assess measuring range of a gradient-echo R2* method and a T1-weighted spin-echo (SE), signal intensity ratio (SIR)-based method (TE = 25 ms, TR = 560 ms), and to extend the upper measuring range of the SIR method by optimizing echo time (TE) and repetition time (TR) in iron-loaded minipigs.

Methods

Thirteen mini pigs were followed up during dextran-iron loading with repeated percutaneous liver biopsies for chemical LIC measurement and MRIs for parallel non-invasive estimation of LIC (81 examinations) using different TEs and TRs.

Results

SIR and R2* method had similar upper measuring range around 34 mg/g and similar method agreement. Using TE = 12 ms and TR = 1200 ms extended the upper measuring range to 115 mg/g and yielded good method of agreement.

Discussion

The wider measuring range is likely caused by lesser sensitivity of the SE sequence to iron, due to shorter TE, leading to later signal loss at high LIC, allowing evaluation of most severe hepatic iron overload. Validation in iron-loaded patients is necessary.

Hepcidin-Ferroportin Interaction Controls Systemic Iron Homeostasis

Despite its abundance in the environment, iron is poorly bioavailable and subject to strict conservation and internal recycling by most organisms. In vertebrates, the stability of iron concentration in plasma and extracellular fluid, and the total body iron content are maintained by the interaction of the iron-regulatory peptide hormone hepcidin with its receptor and cellular iron exporter ferroportin (SLC40a1). Ferroportin exports iron from duodenal enterocytes that absorb dietary iron, from iron-recycling macrophages in the spleen and the liver, and from iron-storing hepatocytes. Hepcidin blocks iron export through ferroportin, causing hypoferremia. During iron deficiency or after hemorrhage, hepcidin decreases to allow iron delivery to plasma through ferroportin, thus promoting compensatory erythropoiesis. As a host defense mediator, hepcidin increases in response to infection and inflammation, blocking iron delivery through ferroportin to blood plasma, thus limiting iron availability to invading microbes. Genetic diseases that decrease hepcidin synthesis or disrupt hepcidin binding to ferroportin cause the iron overload disorder hereditary hemochromatosis. The opposite phenotype, iron restriction or iron deficiency, can result from genetic or inflammatory overproduction of hepcidin.

Iron-oxide minerals in the human tissues

Iron is critically important and highly regulated trace metal in the human body. However, in its free ion form, it is known to be cytotoxic; therefore, it is bound to iron storing protein, ferritin. Ferritin is a key regulator of body iron homeostasis able to form various types of minerals depending on the tissue environment. Each mineral, e.g. magnetite, maghemite, goethite, akaganeite or hematite, present in the ferritin core carry different characteristics possibly affecting cells in the tissue. In specific cases, it can lead to disease development. Widely studied connection with neurodegenerative conditions is widely studied, including Alzheimer disease. Although the exact ferritin structure and its distribution throughout a human body are still not fully known, many studies have attempted to elucidate the mechanisms involved in its regulation and pathogenesis. In this review, we try to summarize the iron uptake into the body. Next, we discuss the known occurrence of ferritin in human tissues. Lastly, we also examine the formation of iron oxides and their involvement in brain functions.

Magnetic mapping of iron in rodent spleen

Evaluation of iron distribution and density in biological tissues is important to understand the pathogenesis of a variety of diseases and the fate of exogenously administered iron-based carriers and contrast agents. Iron distribution in tissues is typically characterized via histochemical (Perl's) stains or immunohistochemistry for ferritin, the major iron storage protein. A more accurate mapping of iron can be achieved via ultrastructural transmission electron microscopy (TEM) based techniques, which involve stringent sample preparation conditions. In this study, we elucidate the capability of magnetic force microscopy (MFM) as a label-free technique to map iron at the nanoscale level in rodent spleen tissue. We complemented and compared our MFM results with those obtained using Perl's staining and TEM. Our results show how MFM mapping corresponded to sizes of iron-rich lysosomes at a resolution comparable to that of TEM. In addition MFM is compatible with tissue sections commonly prepared for routine histology.

Estimating tissue iron burden: current status and future prospects

Iron overload is becoming an increasing problem as haemoglobinopathy patients gain greater access to good medical care and as therapies for myelodysplastic syndromes improve. Therapeutic options for iron chelation therapy have increased and many patients now receive combination therapies. However, optimal utilization of iron chelation therapy requires knowledge not only of the total body iron burden but the relative iron distribution among the different organs. The physiological basis for extrahepatic iron deposition is presented in order to help identify patients at highest risk for cardiac and endocrine complications. This manuscript reviews the current state of the art for monitoring global iron overload status as well as its compartmentalization. Plasma markers, computerized tomography, liver biopsy, magnetic susceptibility devices and magnetic resonance imaging (MRI) techniques are all discussed but MRI has come to dominate clinical practice. The potential impact of recent pancreatic and pituitary MRI studies on clinical practice are discussed as well as other works-in-progress. Clinical protocols are derived from experience in haemoglobinopathies but may provide useful guiding principles for other iron overload disorders, such as myelodysplastic syndromes.

DIFFERENTIAL FERRIOXAMINE TEST FOR MEASURING CHELATABLE BODY IRON

The differential ferrioxamine test is a simple method for the measurement of chelation of body iron by desferrioxamine. A single six-hour specimen of urine is obtained after intravenous Desferal, accompanied by (59)Fe-ferrioxamine. Two values are measured: F(d), the excretion of ferrioxamine derived from body iron by chelation, and F(ex), the proportion of ferrioxamine excreted from a known intravenous dose. The data enables F(v), chelation of iron in vivo, to be calculated by simple proportion. Desferrioxamine chelation proceeds for about half an hour after injection. The results in normal subjects, in cases with known high iron stores, and in cases of iron-deficiency anaemia are described. High, normal, and low body iron states have been differentiated. F(v) values in the higher ranges obtained in iron-storage diseases and in haemolytic states are differentiated by the pattern of excretion, high F(d) values and low F(ex) values respectively. IT IS SUGGESTED THAT THERE ARE TWO MAIN SOURCES OF CHELATABLE BODY IRON: as ferritin-haemosiderin and as iron newly released from haem in a more readily chelatable form. The significance of variable chelation susceptibility in iron metabolism is briefly discussed. It is suggested that variable chelatability of different sources of body iron may explain the preferential utilization of iron released from red cells or absorbed from the intestine, rather than storage iron, in the biosynthesis of haem.

Non-transferrin-bound-iron in serum and low-molecular-weight-iron in the liver of dietary iron-loaded rats

1. The feeding of 0.5% (3,5,5-trimethylhexanoyl)ferrocene (TMH-ferrocene) in rats resulted in a severe and progressive liver siderosis (total liver iron, 30 mg/g liver wet weight, after 30 weeks). 2. High concentrations of an iron-rich ferritin (up to 250 mg/l) were detected in serum of heavily iron-loaded rats forming a large fraction of non-transferrin-bound-iron (5000 micrograms/dl in maximum). 3. Ferritin and not haemosiderin was the major iron storage protein in the liver. 4. The total liver iron concentration (from 0.4 to > 30 mg Fe/g wet wt) but not the cytosolic low-molecular-weight-iron fraction (from 0.5 to 2.5 microM) was extremely increased during iron-loading.

Iron stores and the international variation in mortality from coronary artery disease

Possible roles for iron in coronary artery disease (CAD) have emerged, including contributions to atherogenesis and/or the vulnerability of the myocardium to ischemia/reperfusion events. The value of hepatic storage iron as a potential risk factor for CAD was evaluated independently and in combination with various lipoprotein indices using CAD mortality data from 11 countries along with available data on liver iron stores. CAD mortality rates were found to be best correlated with the liver iron-serum cholesterol product in both men (r = 0.72) and, more importantly, in both genders combined (r = 0.74). It was also found that estimated CAD incidence could be related in a non-linear fashion to iron-cholesterol values in a simple normal distribution model where all subjects above a threshold value of iron-cholesterol were assumed to have CAD. Hepatic iron values thus appear to be useful in describing the differences in CAD due to both diet (and/or culture) and sex.

Coincidental hemochromatosis and viral hepatitis

A 35-year-old woman presented with liver failure, hepatic iron overload, and secondary amenorrhea due to hypogonadotropic hypogonadism. She had chronic inflammatory hepatitis which was considered to be due to post-transfusional viral hepatitis. Her hepatic iron overload was considered to be due to hemochromatosis. Her premature menopause was thought to be due to the severity of her liver disease, but her iron overload also could have contributed to gonadotrophin deficiency. She underwent liver transplantation and 5 months later, she experienced return of menstrual function. The distinction between hepatitis as a cause of iron loading, hemochromatosis as a cause of hepatic inflammation, the small influence of alcohol on increased iron stores, and other features of her history, physical examination, and laboratory evaluation are discussed.

Liver iron stores and hepatitis B antigen status

Higher serum iron and ferritin levels noted in hepatitis B antigen (HBAg) carriers than in noncarriers suggests that virus might actively replicate in hepatocytes, stimulate ferritin synthesis, and result in increased liver iron stores. A comparative semiquantitative study of immunohistochemical ferritin (0-12) and hemosiderin (0-9) was performed on 54 normal, 13 cirrhotic, and 70 nonneoplastic livers from patients with hepatocellular carcinoma, in each group, comparing amounts in HBAg-positive and HBAg-negative patients. Mean scores for ferritin and hemosiderin were high in all three groups, normal livers averaging 8.3 and 6, respectively, cirrhotic livers, 8.5 and 7.4, respectively, and carcinoma livers, 5.6 and 6.1, respectively. In each group, there was no significant difference in ferritin and hemosiderin mean scores in HBAg-positive and HBAg-negative patients. The large liver iron stores do not appear to be a consequence of hepatitis B virus infection alone. Their role in the development of hepatocellular carcinoma is still to be elucidated.

Depletion of excessive liver iron stores with desferrioxamine

To determine the therapeutic effect of long-term, intensive iron chelation therapy, we studied liver iron content and histology in four children with thalassaemia major during 52-83 months of intensive therapy with desferrioxamine. The initial biopsies obtained prior to or within 21 months after beginning chelation therapy had Grade IV iron staining, with heavy iron deposition present in parenchymal and reticuloendothelial cells. Subsequent biopsies, obtained when serum ferritin levels had fallen to 71-246 micrograms/l, contained Grade 0 or Grade I stainable iron. Little or no iron was present in parenchymal or reticuloendothelial cells. The liver iron concentration, measured by magnetic susceptibility, returned to normal or nearly normal levels. Hepatic fibrosis did not progress during treatment with desferrioxamine. These findings demonstrate that intensive and sustained chelation therapy with desferrioxamine will remove excessive liver iron and preserve hepatocellular structure.

Immunohistochemical ferritin in hepatocellular carcinoma

Serum ferritin concentrations are elevated in 35% to 100% of patients with hepatocellular carcinoma (HCC). With an immunoperoxidase technique, ferritin was demonstrated in tumor tissue from 32 of 74 (43%) black southern African patients, and from 12 of 19 (63%) American patients with HCC (P greater than 0.1). Ferritin was present in nonneoplastic liver in 82% of African and 100% of American patients (P greater than 0.1). Moderate to large amounts of stainable hepatic storage iron (hemosiderin) were present in 76% of African and 67% of American patients (P greater than 0.1). Fifty-two (70%) African patients had macronodular cirrhosis. In the literature, 80% to 90% of American patients with HCC have cirrhosis. High serum ferritin levels in patients with HCC may be due to ferritin production by the tumor, or related to the associated iron overload and/or cirrhosis.

The amount of ferritin and hemosiderin in the livers of patients with iron-loading diseases

Depot iron, ferritin iron, and ferritin protein were measured in 34 liver needle biopsy specimens obtained from patients with primary and secondary iron-loading diseases. The patients were classified as idiopathic hemochromatosis (14), porphyria cutanea tarda (4), and iron-loading anemias (16). With accumulation of depot iron, the amount of liver ferritin protein increased, however, the ratio of ferritin protein to depot iron fell at concentrations of depot iron in excess of 1,000 micrograms per gm of liver. The liver ferritin protein concentration was not influenced by the specific kind of the iron storage disorder. The mean iron content of ferritin molecules increased about 50% in profound iron overload. In low grade iron overload, the bulk of depot iron was present as ferritin; however, in subjects with heavy iron overload, depot iron consisted of approximately equal amounts of hemosiderin (nonferritin iron) and ferritin iron.

Human nails and body iron

The iron content of human nails has been measured by atomic absorption spectrophotometry and compared with other measurements of iron status including the bone marrow. Four groups of individuals were studied: 40 healthy laboratory staff, five iron-deficient subjects before and during iron therapy, four patients at various stages of treatment with iron, and 15 postmortem cases. The iron status of the individual was reflected by the amount of iron present in nail samples. Nail sampling is proposed as a cheap, noninvasive method of assessing the iron status of the individual.

A method for measurement of liver iron fractions in needle biopsy specimens and some results in acute liver disease

Methods are described for measurement of total tissue iron, ferritin iron, haem iron and ferritin protein in approx. 15 mg of tissue obtained by liver biopsy. The validity of these methods is examined by comparison with the values observed in larger samples of the same post-mortem derived liver tissue. Correlation coefficients vary between 0.80 and 0.99 (n = 11--16). It appears that in post-mortem liver tissue the haem iron concentration is higher than in biopsy specimens from patients. Analysis of liver biopsy specimens from patients with hepatitis showed a large variation in the mean iron content of the liver ferritin molecules. Also, the non-ferritin depot iron concentration and ferritin protein concentration is quite variable. It is suggested that in cases of advanced ferritin catabolism during hepatitis the mean percentage of iron in ferritin molecules often increases while at the same time the non-ferritin depot iron fraction decreases, probably because of iron release from the liver.

On the quantitative determination of liver ferritin in the rat

The measurement of liver ferritin is usually performed after homogenization of liver, heating the homogenate and centrifugation. Because ferritin is stable to 80 degrees this protein and its iron are recovered in the supernatant. We found that this procedure resulted in losses of ferritin so we developed a method to measure ferritin protein in the unheated homogenate. Total liver ferritin iron could be calculated with use of the ferritin protein and ferritin iron values as measured in the supernatant after heating the liver homogenate.

Liver iron concentration, stainable iron, and total body storage iron

Liver iron concentration has been determined chemically in 154 liver biopsies and the findings compared with the routine histological assessment of stainable parenchymal iron, performed by an independent observer. There was a significant correlation between liver iron concentration and histochemical grading but the relationship did not have a normal linear form. Absence of stainable iron corresponded to liver iron concentrations below the mean value for control male subjects (77 mug/100 mg dry liver). In general grade 1 siderosis corresponded to liver iron concentrations in the upper part of the control range and grade 2 siderosis to marginally elevated values. The transition from grade 2 to grade 3 (submaximal) siderosis represented a sharp increase in liver iron concentration and as grade 3 siderosis corresponded to a wide range of chemical values it is also the most difficult histochemical grade to interpret in quantitative terms. Grade 4 siderosis invariably indicated heavy iron excess.There was a close correlation between liver iron concentration and measurements of total body storage iron obtained by quantitative phlebotomy in patients with idiopathic haemochromatosis and by determination of DTPA-chelatable body iron in a variety of iron-loading disorders.

Cobalt excretion test for the assessment of body iron stores

Iron absorption is under delicate control and the level of absorption is adjusted to comply with the body's need for iron. To measure the intestinal setting for iron absorption, and thereby indirectly assess body iron requirements, cobaltous chloride labelled with (57)Co or (60)Co was given by mouth and the percentage of the test dose excreted in the urine in 24 hours was measured in a gamma counter. Seventeen control subjects with normal iron stores excreted 18% (9-23%) of the dose. Increased excretion, 31% (23-42%), was found in 10 patients with iron deficiency anemia and in 15 patients with depleted iron stores in the absence of anemia. In contrast, 12 patients with anemia due to causes other than iron deficiency excreted amounts of radiocobalt within the normal control range. In patients with iron deficiency, replenishment of iron stores by either oral or parenteral iron caused the previously high results to return to normal.Excretion of the test dose was normal in portal cirrhosis with normal iron stores but it was markedly increased in patients with cirrhosis complicated by either iron deficiency or endogenous iron overload. It was also raised in primary hemochromatosis. Excretion of the dose was reduced in gluten-sensitive enteropathy. Gastrointestinal surgery and inflammatory disease of the lower small intestine had no effect on the results except that some patients with steatorrhea had diminished excretion.The cobalt excretion test provides the clinician with a tool for the assessment of iron absorption, the detection of a reduction in body iron stores below the level that is normal for the subject in question, the differentiation of iron deficiency anemia from anemia due to other causes, and the investigation of patients with iron-loading disorders.

Measurement of iron stores in cirrhosis using diethylenetriamine penta-acetic acid

The chelating agent diethylenetriamine penta-acetic acid was used to measure iron stores in 83 patients with chronic liver disease. Iron chelation was normal in patients with chronic cholestasis. Chelation was increased above the control range in 14 out of 26 patients with alcoholic cirrhosis, in nine out of 28 patients with non-alcoholic cirrhosis, and in 11 out of 15 cirrhotics with a portacaval anastomosis. Iron stores in excess of 1.5 g were predicted from the results in 24 subjects; however, in only three were the values in the range found in propositi with untreated idiopathic haemochromatosis. Increased chelation did not correlate with hepatocellular impairment per se but was associated in 18 cases with surgical or large spontaneous portal systemic shunts. Exogenous factors for excess iron were present in three cases with alcoholic cirrhosis and portal systemic collaterals in one, but no special factor apart from alcoholism was apparent in the remainder. The correlation between chelatable iron and stainable liver iron content was not close and was better in haemochromatosis than in other forms of cirrhosis; in some cases considerable siderosis was present with normal or only slightly increased chelation values.

Storage iron in "muscle"

In Bantu subjects with iron overload iron is visible in skeletal muscle cells and in the tissue histiocytes which lie between these cells. In the present study the concentrations of ;muscle' iron were measured chemically in subjects with varying hepatic storage iron concentrations. The results indicate that the concentrations of storage iron in ;muscle' are much lower than those in the liver. However, the muscle mass is so large that the total amount of iron present is at least equal to that in the liver in subjects with normal body stores. The concentrations of iron in ;muscle' are raised in subjects with iron overload but the degree to which they rise is far less than occurs in the liver; a thirtyfold increase in hepatic iron concentrations is associated with only a sixfold increase in ;muscle' iron. Experiments in rats revealed that storage iron in ;muscle' represents a relatively non-miscible pool which responds very little to acute changes in the iron environment.