METAL-ION BINDING OF HUMAN TRANSFERRIN

Disposition, toxicity, and intestinal absorption of cobaltous chloride in male Fischer 344 rats

The absorption and disposition of inorganic cobalt salts after oral administration have not been well characterized. The objectives of this study were to compare in vivo results with cobalt transport through the in vitro everted small intestine and to relate the disposition results to a biochemical indicator of cobalt toxicity. Cobalt chloride was given to male Fischer 344 rats orally at 33.3 mg Co(II)/kg or intravenously at 4.16 mg Co(II)/kg. By 36 h, 74.5% of the oral dose was eliminated in the feces. The liver, kidney, and heart accumulated cobalt to the greatest extent. Following the single oral dose, the blood cobalt concentration-time curve was triphasic, peaked at 3.2 h, and had an absorptive half-life of 0.9 h, an elimination phase half-life of 3.9 h, and a terminal elimination half-life of 22.9 h. Following intravenous administration, 10.1% of the dose was excreted in the feces, indicating that cobalt can be secreted in the bile. Following a single intravenous injection, the concentration-time curve displayed three segments. The first segment, which occurred during the first 4 h, had a rapid half-life of 1.3 h. The second phase, from 4 to 12 h, demonstrated a slower clearance rate with a half-life of 4.3 h. The final and slowest phase, from 12 to 36 h, had a half-life of 19 h. Intestinal jejunal ring experiments indicated that cobalt transport has both active and passive components; however, cobalt transport through the in vitro rat everted duodenum indicated that cobalt transport had almost exclusively passive components with facilitated diffusion. The finding that uptake was saturable may explain the small extent of absorption following oral dosing. Heme oxygenase studies following subcutaneous and intravenous administration resulted in an increase in activity (twofold) over controls, while oral administration did not. We concluded that the extent of cobalt absorption across the gastrointestinal tract is incomplete, and that the concentration administered and the route of exposure may determine its systemic toxicity.

X.a.f.s. studies of chicken dicupric ovotransferrin

A comparison of Cu K-edge x.a.f.s. spectra with that of the equivalent Fe K-edge for chicken ovotransferrin (COT) indicates that the metal ions occupy essentially the same binding sites in the protein. However, in the case of the Cu2+ complex the metal appears to have reduced co-ordination. Changes are observed in the x.a.f.s. of 90%-saturated COT (Cu1.8COT) on freeze-drying. The three-dimensional X-ray structures of rabbit serum transferrin and human lactoferrin have shown that the ferric cations are co-ordinated by four protein ligands and a bidentate carbonate anion in a distorted octahedral arrangement [Anderson, Baker, Dodson, Norris, Rumball, Waters & Baker (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 1768-1774; Anderson, Baker, Norris, Rice and Baker (1989) J. Mol. Biol. 209, 711-734; Bailey, Evans, Garratt, Gorinsky, Hasnain, Horsburgh, Jhoti, Lindley, Mydin, Sarra & Watson (1988) Biochemistry 27, 5804-5812]. This structural information, together with the differences in e.x.a.f.s. spectra for solution and freeze-dried samples of diferric COT [Hasnain, Evans, Garratt & Lindley (1987) Biochem. J. 247, 369-375] suggests that the synergistic carbonate anion may be capable of behaving as a unidentate linkage to the Cu2+ in the dicupric complex. Data for Cu1.8COT are consistent with only three protein ligands bound to Cu2+, monodentate binding of the synergistic anion in one lobe and its bidentate binding in the other lobe. Such flexibility in the anion co-ordination may be a requirement for the uptake and release of metals by the transferrins.

Iodogen-catalyzed iodination of transferrin

Transferrin (human, rabbit) labels at low efficiency (1%-10%) with 125I when reaction of 0.5-0.7 ng of I- (8-10 microCi) with 20 micrograms of the protein is catalyzed by iodogen in a constant volume of 0.1 ml. Microiodination by this technique was therefore analyzed with regard to the relative proportions of the reactants, oxidant requirement, and timing. In vials giving a reaction volume-to-active surface ratio of 0.88, efficiency was independent of the amount of iodogen in the range from 1 microgram to 15 micrograms, and prolongation of the reaction beyond 1 min failed to improve yields. In contrast, the amount of I- present was decisive. Butanol/NH4OH chromatograms of iodination reactions carried out with 0.6 ng or 20 ng of I- showed 3-4 radioactivity peaks, the relative proportions of which markedly depended on the amount of I- present originally. A link was established between labeling efficiency and chromatographic profile of the I- derivatives formed during oxidation. Dual-label experiments in rats showed that transferrin (20 micrograms) can be labeled using iodogen (1-5 micrograms, 1 min) to behave indistinguishably from its IC1-labeled counterpart. However, prolonged exposure to more oxidant progressively damaged the protein. The damage was independent of substituting I and it manifested itself in increased protein binding to the anion exchange resin, Dowex 1-X8. Over 99.5% of the labeled residues in iodotransferrin were mono- and diiodotyrosines (MIT, DIT). DIT content of the protein increased linearly with the number of I atoms substituted. At comparable levels of substitution, more label was present as MIT after using iodogen than after using IC1. Electrophoretic data are presented regarding homogeneity of the label as obtained after iodinating transferrin by different methods and to varying extents.

The relative affinity of transferrin and albumin for zinc

The relative affinity of transferrin and albumin for zinc has been measured by competitive dialysis at a low zinc concentration in 0.15 M NaCl, 50 mM HEPES, 0.1 mM trisodium citrate (pH 7.2). There were small differences between albumins and larger ones between transferrin preparations, but all albumins bound zinc more firmly than any transferrin did. It is known that transferrin is largely responsible for the uptake of zinc from an intestinal membrane in rats, but much of the metal is subsequently transferred to albumin. The current results show that both in humans and in rats (a) no special mechanism is needed to provide energy for this transfer, and (b) full equilibration would lead to virtually complete transfer in contrast with what actually occurs in vivo.

[On the complex formation of proteins with cu ions under acidic conditions]

In order to determine the influence of copper additives on the protein digestibility, the complex formation of proteins with Cu ions under acidic conditions (pH=2.2) was investigated. Isolated soja-protein (300 mg) resp. pepsin was mixed with 10 mumol (2.49 mg) CuSO4-5H2O and dialysed against low molecular complexing agents with increasing affinity towards copper ions. The experiments with soja-protein showed, that most of the Cu ions dialysed (97% of total Cu2+ content) without any addition of complexing agents, pepsin, however, had a lower percentage (83% of total Cu2+ content). The remaining copper ions in pepsin could only be removed by alanin and ammonium-tetramethylendithiocarbamate. Therefore it is concluded, that in the acidic range Cu ions only form complexes with pepsin, but not with soja-protein. This effect is due to the very low isoelectric point of pepsin. The determination of iron, nickel, and zinc in native proteins revealed, that at pH=2.2 iron and nickel are stronger linked to pepsin than to soja-protein; only zinc was loosely bound to both proteins.

Integrity of the iron transport process in mice with X-linked anaemia

The defect in iron (Fe) absorption in X linked anaemia (sla) remains an enigma; absorption of a tracer dose of Fe is impaired in mice raised on an iron-containing cube diet but not in those raised on an iron-deficient diet. Because cobalt (Co) shares a similar intestinal transport pathway with Fe, a study was made of the effect of iron deficient diet on Co absorption. The duodenum of sla and genetically normal mice was perfused for 30 min with labelled solutions containing Co or Fe. Co uptake and transfer were similar in sla and normals fed cubes whereas Fe uptake and transfer were less in sla than in normals. The iron deficient diet caused an increase in the uptake and transfer of Co and Fe in sla and normals. When Co and Fe were perfused together in sla fed deficient diet, the uptake and transfer of each metal was less than when performed alone. The distribution of Fe and Co in subcellular mucosal fractions was determined by a differential centrifugation technique. Deficient diet resulted in a directionally similar change in the subcellular distribution of Co and Fe in sla and normals. The increase in Co as well as Fe absorption in the sla on an iron deficient diet to the same high level found in genetically normal animals, and the inhibitory effect of each metal on the absorption of the other suggests that the absorption defect in sla is unlikely to be due to a primary defect in the function of the transport carrier.

An enquiry into the role of cobalt in the heart disease of chronic beer drinkers

The natural history of a new disease entity in chronic beer drinkers consisting of pericardial effusion, a low cardiac output, and in about half of the cases of polycythemia is presented. Normalization of the heart volume, the hemoglobin value, and the hemodynamic state was obtained in patients who stopped drinking. Further evidence of the importance of cobalt in the development of the disease is presented. All patients drank beer to which cobalt was added and the disease was not seen in chronic alcoholics drinking wine or other alcoholic beverages. In view of the rather small quantity of cobalt consumed, an hypothesis is formulated which attributes the cobalt toxicity in chronic alcoholics, to a dietary deficiency of sulfhydryl-groups containing amino acids and a dietary deficiency of protein.

A method for the separation of peptides and alpha-amino acids

1. Peptides and alpha-amino acids, occurring in mixtures from various sources, can be separated into one fraction containing the amino acids and several peptide fractions. This is achieved by chelation of the mixture with Cu(2+) ions and subsequent chromatography of these chelates over the acetate form of diethylaminoethylcellulose or triethylaminoethylcellulose. 2. The amino acid fraction is obtained by elution with 0.01m-collidine-acetate buffer, pH8.0. 3. Peptide fractions are eluted with 0.01m-collidine-acetate buffer, pH4.5, 0.17n-acetic acid and 0.1n-hydrochloric acid respectively. 4. With the exception of aspartic acid and glutamic acid, which are partly found in the acidic peptide fraction, the amino acids are completely separated from the peptides. 5. Contamination of the acidic peptide fraction with glutamic acid and aspartic acid can be largely avoided by previous addition of an excess of arginine. 6. Copper is removed from the eluates by extraction with 8-hydroxyquinoline in chloroform.