Zinc concentration in patients with iron overload receiving oral iron chelator 1,2-dimethyl-3-hydroxypyrid-4-one or desferrioxamine.

AIMS--To determine the changes in serum zinc concentration and the extent of urinary zinc excretion in patients with iron overload receiving the oral iron chelator 1,2-dimethyl-3-hydroxypyrid-4-one (L1) or desferrioxamine (DFX), and to correlate these results with blood glucose concentration. METHODS--Serum zinc and ferritin concentrations, urinary zinc and iron excretion were regularly assayed in 39 patients and the glucose tolerance test (GTT) was performed in each patient. Patients were segregated according to their GTT into normal, diabetic, and those with an abnormal GTT. The mean of L1- or DFX associated urinary zinc excretion for each group was determined and compared with the other two groups and with normal value. L1 associated urinary zinc excretion was also compared with L1 dose, serum ferritin values, and urinary iron excretion. RESULTS--Both DFX and L1 were associated with a significantly increased urinary zinc excretion (15.1 (7.3) mumol/24 hours, 11.1 (6.0) mumol/24 hours, respectively) compared with normal subjects. In patients receiving DFX this increase only occurred in patients with diabetes mellitus. Both diabetic and non-diabetic patients receiving L1 treatment excreted more zinc than normal. Diabetic patients receiving L1 or DFX excreted more zinc than non-diabetics receiving the same treatment. No correlation was found between urinary zinc excretion and L1 dose or patients' serum ferritin concentrations. In seven patients receiving long term L1 treatment a fall in serum zinc was observed from an initial 13.6 (1.6) mumol/l to a final 9.6 (0.8) mumol/l. In one patient this was associated with symptoms of dry skin and itchy skin patches requiring treatment with oral zinc sulphate. CONCLUSIONS--In contrast to DFX, L1 treatment is associated with increased zinc loss. This, however, is modest and does not lead in most patients to subnormal serum zinc concentrations. In a few patients whose negative zinc balance may give rise to symptoms, zinc supplementation rapidly corrects the deficit.     

In vitro effects of three iron chelators on mitogen-activated lymphocytes: identification of differences in their mechanisms of action.

The effects of three iron chelators (ADR-529/ICRF-187; omadine/pyrithione; and a newly synthesized pyridoxal-based iron chelator, SAG-15) on cultured BALB/c murine lymph node cells stimulated with phorbol myristate acetate and ionomycin have been investigated. All three agents were found to inhibit [3H]-thymidine incorporation after 66-72 h incubation. Pretreatment of ADR-529 and omadine with Fe(III) or Fe(II) ions did not prevent their inhibitory effects. However, pretreatment of SAG-15 with Fe(II) or Fe(III) ions led to a significant increase in the ID50. Time-course studies of cell viability and thymidine incorporation demonstrated that the inhibitory effect of omadine was attributable to cell killing while for ADR-529 and SAG-15 there were both cytostatic and cytotoxic effects. Cell cycle analysis showed that treatment of cells with ADR-529 led to arrest in G2/M while treatment with SAG-15 led to a G0/G1 arrest. Iron has an obligatory role in T-lymphocyte activation that may be related to the formation of reactive oxygen species. SAG-15 is a new iron chelator that will help in the elucidation of the precise role of iron in lymphoproliferation. Since SAG-15 is an extremely effective iron chelator in vivo it has potential as an immunosuppressive agent.     

Cell cycle synchronisation by 3-hydroxypyridin-4-one iron chelators

We describe a protocol for the synchronisation of normal and tumour cells grown in suspension cultures using 3-hydroxypyridin-4-one iron chelators. These compounds inhibit ribonucleotide reductase, one of the rate limiting enzymes in DNA synthesis, and so block the cell cycle in late G₁ phase. After removal of the chelator or repletion of cellular iron, cells progress through the cycle and remain synchronised for at least one full cell cycle. Cell viability is unaffected for at least 72 hours post-incubation and chelator treatment has no effect on RNA and protein synthesis. This method of synchronisation has been successful with all cell lines tested including normal and leukaemic human cell lines.     

Bioverfügbarkeit des in einem oralen Eisenpräparat enthaltenen Eisen(III)-Hydroxid-Dextrin-Komplexes

Summary An iron(III)-hydroxide-polymaltose complex used in the FRG for oral iron therapy was labeled with⁵⁹Fe and investigated for bioavailability in subjects with normal and depleted iron stores. Following the oral application of a 100 mg Fe(III)-equivalent of the⁵⁹Fe(III)-hydroxide polymaltose complex the postabsorptive serum increase after 3 h as well as the whole body retention and erythrocyte incorporation of absorbed⁵⁹Fe after 2 weeks were measured by intraindividual comparison with a 100 mg Fe(II)-equivalent in⁵⁹Fe(II)-ascorbate.Subjects with normal iron stores absorbed and retained in the body 7.95±1.53 (_a±SD) mg Fe from the oral dose from⁵⁹Fe(II)ascorbate but only 0.49±0.42 mg Fe from Fe(III)hydroxide polymaltose. Due to their depleted iron stores subjects in prelatent/latent iron deficiency absorbed and retained increased amounts of 14.6±2.7 mg Fe from⁵⁹Fe(II)ascorbate but only 1.34±0.61 mg Fe from⁵⁹Fe(III)-hydroxide-polymaltose. Because of the existing close correlation (r=0.93) between the whole-body retention and erythrocyte incorporation of absorbed⁵⁹Fe 5.85±1.1 mg Fe from the⁵⁹Fe(II)ascorbate iron and only 0.46±0.45 mg Fe from the⁵⁹Fe(III)-hydroxide polymaltose complex iron were incorporated into erythrocytes by the subjects with normal iron stores after 2 weeks. The erythrocyte-⁵⁹Fe-incorporation was increased to 11.9±2.51 mg Fe from the⁵⁹Fe(II)ascorbate but only 1.42±0.80 mg Fe from the⁵⁹Fe(III)-hydroxide polymaltose iron in subjects with depleted iron stores.If compared with Fe(II)ascorbate or sulfate (=100%) relative bioavailability values of 10% and 13% respectively, for a polymere iron(III)-citrate complex and 6% and 9%, respectively, for the iron(III)-hydroxide polymaltose complex were estimated for subjects with normal and depleted iron stores. The trivalent iron in the polymaltose complex has the lowest bioavailability of all oral iron preparations investigated with reliable methods up to now and does not represent an exception from the 50-year-old rule that all trivalent iron containing oral iron preparations are so poorly absorbable that they are ineffective in oral iron therapy. As in all other countries, trivalent iron containing oral iron preparations should be climinated from iron therapy also in the FRG.

Herrn Professor Fritz Reimann, Istanbul, in Verehrung gewidmet     

Role of antibody and enterobactin in controlling growth of Escherichia coli in human milk and acquisition of lactoferrin- and transferrin-bound iron by Escherichia coli.

Growth of Escherichia coli NCTC 8623 in human milk was slow during the first 10 h of incubation, but this bacteriostatic effect had disappeared by 24 h. The bacteriostatic phase could be abolished by adding sufficient iron to saturate the lactoferrin in human milk, and also by adding supernatant from a 24-h milk culture or by adding enterobactin, an enterobacterial iron chelator. Growth in the presence of enterobactin was even more rapid than in the presence of excess iron. Partial loss of bacteriostatic activity could be achieved by absorbing the milk with bacterial antigens, but no clear correlation with removal of antibodies to O, K, or H antigens was apparent. When E. coli was grown in human serum trace-labeled with 59Fe, the organisms acquired iron from transferrin during growth. Cultivation of E. coli in a minimal medium supplemented with transferrin or lactoferrin doubly labeled with 125I and 59Fe showed that iron acquisition occurred without either assimilation or degradation of the iron-binding proteins.     

Ferritin and intestinal iron absorption: pancreatic enzymes and free iron.

Rat intestinal mucosa gave low yields of ferritin purified by standard procedures. The resulting ferritin had less protein relative to iron and migrated faster electrophoretically than ferritin from other rat tissues. Pancreatic duct ligation reduced these differences, suggesting digestive enzyme attack during ferritin isolation. Even in ligated rats, ferritin accounted for only 5-10% of mucosal iron. However, shortly after giving 59FeCl3 orally, 50% of mucosal radioactivity occurred in cell sap, about equally distributed between ferritin and low-molecular-weight (chelated?) iron. No other cell sap components were 59Fe labeled. Iron may thus be transported as a chelate with which ferritin is in rapid equilibrium. Mucosal ferritin content increased with age and iron treatment and decreased with iron deficiency. The iron-deficient rats showed accelerated 59Fe uptake into blood with little mucosal retention. One day after administering parenteral iron to deficient rats, 59Fe transfer to blood became retarded but 59Fe now accumulated excessively in the mucosa, suggesting that iron status affects transport more rapidly at the serosal than at the mucosal cell surface. A scheme for control of iron absorption is presented.     

Desferri-Exochelin, a lipid-soluble, hexadentate iron chelator, effectively removes tissue iron

Chronic iron-overload is damaging to the heart, liver, and other organs. Better iron chelators are needed to treat this serious medical condition. The uptake and distribution of the lipid-soluble, hexadentate iron chelator desferri-Exochelin 772SM (D-Exo) is studied and its efficacy in removing iron from tissue in rodent models is evaluated. After an intravenous bolus of tritiated D-Exo to rats, counts rapidly disappeared from the blood and rapidly appeared in 15 organs studied, usually peaking within 15 min. There was considerable uptake in the heart and liver, 2 organs especially susceptible to damage from clinical iron overload. To assess actual decreases in cardiac and hepatic iron in response to D-Exo, mice loaded with 42 mg of iron dextran (2100 mg/kg) were studied. Untreated, iron-loaded mice sacrificed 9 weeks later had a 4-fold increase in cardiac iron and a 20-fold increase in hepatic iron compared with controls that were not iron-loaded. In iron-loaded mice treated with 7 mg of D-Exo intraperitoneally (i.p.) 4 days/week for 8 weeks (total 224 mg), tissue iron, measured by atomic absorption, was reduced by 20% in the liver and 25% in the heart (P < 0.01 for each organ). During the first 8 h after a D-Exo dose, iron was excreted in the urine. Mice treated with D-Exo gained weight normally and showed no evidence of toxicity. In conclusion, in this iron-overload mouse model, D-Exo administered intravenously or i.p. rapidly diffuses into multiple organs, including the heart and liver, and effectively removes iron without apparent toxicity.     

Binding and accumulation of hemin in Neisseria gonorrhoeae.

The ability to utilize hemin and hemin-containing compounds for nutritional iron (Fe) uptake has been documented for several pathogenic bacteria. Neisseria gonorrhoeae can utilize free hemin as a source of Fe for growth; however, little is known concerning the mechanisms involved in hemin transport. In this study we have characterized the binding and accumulation of hemin by N. gonorrhoeae and defined the specificity of the gonococcal hemin receptor. N. gonorrhoeae F62 was grown in a chemically defined medium containing the iron chelator Desferal, and hemin transport was initiated by the addition of [59Fe]hemin (4.0 or 8.0 microM; specific activity, 7.0 Ci/mol). 59Fe uptake from radiolabeled hemin by N. gonorrhoeae was energy dependent, and 59Fe was shown to accumulate in the cell at a constant rate during logarithmic growth. However, we observed a decrease in the uptake of 59Fe from radiolabeled hemin when inorganic iron was present in the growth medium. Binding of 59Fe from radiolabeled hemin was inhibited by the addition of either cold hemin, hematoporphyrin, or hemoglobin, but not by ferric citrate. Although [14C]hemin was found to support the growth of N. gonorrhoeae, we did not detect the uptake of 14C from radiolabeled hemin. Extraction of the gonococcal periplasmic ferric binding protein (Fbp) from cultures grown with [59Fe]hemin indicated that a majority of the 59Fe was associated with the Fbp. Taken together, the results presented here indicate that hemin binds to a gonococcal outer membrane receptor through the protoporphyrin portion of the molecule and that following binding, iron is removed and transported into the cell, where it is associated with the gonococcal periplasmic ferric binding protein, Fbp.     

Increased susceptibility to oxidative stress in scrapie-infected neuroblastoma cells is associated with intracellular iron status

The molecular mechanism of neurodegeneration in prion diseases remains largely uncertain, but one of the features of infected cells is higher sensitivity to induced oxidative stress. In this study, we have investigated the role of iron in hydrogen peroxide (H(2)O(2))-induced toxicity in scrapie-infected mouse neuroblastoma N2a (ScN 2 a) cells. ScN 2 a cells were significantly more susceptible to H(2)O(2) toxicity than N2a cells as revealed by cell viability (MTT) assay. After 2h exposure, significant decrease in cell viability in ScN 2 a cells was observed at low concentrations of extracellular H(2)O(2) (5-10 microM), whereas N2a cells were not affected. The increased H(2)O(2) toxicity in ScN 2 a cells may be related to intracellular iron status since ferrous iron (Fe(2+)) chelator 2,2'-bipyridyl (BIP) prevented H(2)O(2)-induced decrease in cell viability. Further, the level of calcein-sensitive labile iron pool (LIP) was significantly increased in ScN 2 a cells after H(2)O(2) treatment. Finally, the production of reactive oxygen species (ROS) was inhibited by 30% by iron chelators desferrioxamine (DFO) and BIP in ScN 2 a cells, whereas no significant effect of iron chelators on basal ROS production was observed in N2a cells. This study indicates that cellular resistance to oxidative stress in ScN 2 a cells is associated with intracellular status of reactive iron.     

Radical-promoting "free" iron level in the serum of rats treated with ferric nitrilotriacetate: comparison with other iron chelate complexes.

Iron plays a critical role in the production of activated oxygen species and the activity of chelated iron in the biological system depends on the chemical forms of the chelators. In the present study, we used ferric nitrolotriacetate (Fe-NTA, molar ratio of iron to chelators = 1:3), ferric ethylenediaminetetraacetate (Fe-EDTA, 1:3 complex) and ferric Desferal (Fe-Des, 1:1.1 complex) to see their "free" iron content in aqueous solutions in vitro and in the serum obtained after a single intraperitoneal injection of the chelates to rats (7.5 mg of iron/kg). "Free" iron was measured by the bleomycin-assay system. When Fe-NTA was dissolved in water, "free" iron increased linearly with total iron concentration up to 10 microM, whereas Fe-EDTA and Fe-Des showed no "free" iron with corresponding iron concentrations. When these three ferric chelates were dissolved in normal rat serum, "free" iron in Fe-NTA increased abruptly between 40 microM and 60 microM iron concentrations, then increased slowly up to 100 microM. Fe-Des did not show any "free" iron at comparable iron concentrations. Fe-EDTA had an intermediate "free" iron level in the serum. Among the ferric chelate complexes, Fe-NTA showed a much faster increase of and a higher content of "free" iron in the serum than the other two complexes after a single injection of the chelates into rats.(ABSTRACT TRUNCATED AT 250 WORDS)     

Whole body59Fe-elimination rates and corresponding blood losses in patients with factitious anemia induced by self-blood letting

Summary Factitious anemia induced by self-bloodletting is a rare cause of chronic haemorrhagic iron deficiency anemia. Within a year two cases from Southern Germany were identified and quantitated for blood losses by whole body counting of retained iron. Whereas the 2 patients did not lose much blood (6–10 ml blood/day equivalent to 0.12–0.34%⁵⁹Fe-whole body elimination/day) during their observation within the hospital their blood losses increased to 41–187 ml/day (=whole body⁵⁹Fe-elimination rates of 0.8 to 6.1%/day) upon release from the hospital. Although they lost about 5 liter blood in 50 days faecal occult blood tests were always negative and the daily faecal and urinary⁵⁹Fe-excretion was with 0.018–0.031%/day within the normal range. Haemoglobin and erythrocyte levels were stable during the periods of clinical observation when the whole body⁵⁹Fe-elimination rates were only just above the normal range (0.12–0.21%/d). These values dropped however to very low values during or following periods of extremely elevated whole body⁵⁹Fe-elimination rates (1–6%/d=46–187 ml blood loss/d) immediately after release of the patients from the hospital. One patient received more than 400 blood and erythrocyte infusions (84 g iron) and absorbed about 8 g food iron within 10 years. Nevertheless she was always severely anemic (5 g% haemoglobin) and had no storage iron in the tissues. She must have lost about 92 g iron in 10 years (= 529 liter blood) or 25 mg Fe/d (= 145 ml blood/d) by self bloodletting avoiding the otherwise extreme generalized transfusion haemosiderosis. Blood losses of up to 145 ml/d (= 50 mg Fe/d) can be compensated by oral ferrous iron therapy with 4×50=200 mg Fe²⁺/d. The patients did however refuse to take oral or parenteral iron and insisted in one case on regular blood transfusions twice per week.Two more probable cases of factitious anemia from Northern Germany were Subsequently observed and partially investigated (cases 3 and 4).With the technical assistance of Fatima Iagi     

Nanoparticle–chelator conjugates as inhibitors of amyloid-β aggregation and neurotoxicity: A novel therapeutic approach for Alzheimer disease

Oxidative stress and amyloid-β are considered major etiological and pathological factors in the initiation and promotion of neurodegeneration in Alzheimer disease (AD). Insomuch as causes of such oxidative stress, transition metals, such as iron and copper, which are found in high concentrations in the brains of AD patients and accumulate specifically in the pathological lesions, are viewed as key contributors to the altered redox state. Likewise, the aggregation and toxicity of amyloid-β is dependent upon transition metals. As such, chelating agents that selectively bind to and remove and/or “redox silence” transition metals have long been considered as attractive therapies for AD. However, the blood–brain barrier and neurotoxicity of many traditional metal chelators has limited their utility in AD or other neurodegenerative disorders. To circumvent this, we previously suggested that nanoparticles conjugated to iron chelators may have the potential to deliver chelators into the brain and overcome such issues as chelator bioavailability and toxic side-effects. In this study, we synthesized a prototype nanoparticle–chelator conjugate (Nano-N2PY) and demonstrated its ability to protect human cortical neurons from amyloid-β-associated oxidative toxicity. Furthermore, Nano-N2PY nanoparticle–chelator conjugates effectively inhibited amyloid-β aggregate formation. Overall, this study indicates that Nano-N2PY, or other nanoparticles conjugated to metal chelators, may provide a novel therapeutic strategy for AD and other neurodegenerative diseases associated with excess transition metals.     

Reversion of age-related recognition memory impairment by iron chelation in rats

It is now generally accepted that iron accumulates in the brain during the ageing process. Increasing evidence demonstrate that iron accumulation in selective regions of the brain may generate free radicals, thereby possessing implications for the etiology of neurodegenerative disorders. In a previous study we have reported that aged rats present recognition memory deficits. The aim of the present study was to evaluate the effect of desferoxamine (DFO), an iron chelator agent, on age-induced memory impairment. Aged Wistar rats received intraperitoneal injections of saline or DFO (300mg/kg) for 2 weeks. The animals were submitted to a novel object recognition task 24h after the last injection. DFO-treated rats showed normal recognition memory while the saline group showed long-term recognition memory deficits. The results show that DFO is able to reverse age-induced recognition memory deficits. We also demonstrated that DFO reduced the oxidative damage to proteins in cortex and hippocampus. Thus, the present findings provide the first evidence that iron chelators might prevent age-related memory dysfunction.     

The effect of consumption of the pyrrolizidine alkaloid-containing plant Senecio jacobaea on iron and copper metabolism in the rat

The effect of dietary tansy ragwort (Senecio jacobaea), a pyrrolizidine alkaloid (PA)-containing plant, on mineral metabolism in rats was studied. In experiment 1, rats were fed a dietary level of 5% tansy ragwort. At intervals of 1, 2, 4, 6 and 8 weeks animals were killed and tissue mineral levels determined. As compared to comparable controls, rats fed tansy ragwort showed by 6 weeks elevated liver and spleen copper and iron levels. Experiment 2 was a 3 X 4 factorial experiment with added dietary copper levels of 0, 50 and 250 ppm, and tansy ragwort levels of 0, 1, 2.5 and 5%. Liver copper levels increased markedly with increasing levels of tansy ragwort; at 0, 50 and 250 ppm added copper, liver copper levels were 4, 18 and 21 times greater in rats fed 5% tansy ragwort as compared to those with no tansy ragwort. Increases in liver iron and spleen copper were noted with consumption of tansy ragwort. Higher liver copper levels were observed when a casein-based diet rather than a soybean meal diet was used, suggesting an effect of phytate in soybean meal in reducing copper absorption. In the last experiment, 59Fe was administered to rats fed diets with or without tansy ragwort. After 5 weeks on tansy ragwort, rats showed very low levels of 59Fe in erythrocytes, tibia and liver, and elevated levels in spleen and kidney, suggesting either an impairment of hematopoesis or accelerated erythrocyte destruction as a result of PA consumption. These results indicate that PA's cause increased liver copper content, and disturbances in iron metabolism.     

Antioxidant activity of nasunin, an anthocyanin in eggplant

Delphinidine-3-(p-coumaroylrutinoside)-5-glucoside (nasunin), an anthocyanin was isolated as purple colored crystals from eggplant peels, Solanum melongena L. 'Chouja'. Using an electron spin resonance spectrometer and 5,5-dimethyl-1-pyrroline-N-oxide (DMPO), spin trapping, hydroxyl (.OH) or superoxide anion radicals (02*-) generated by the Fenton reaction or the hypoxanthine-xanthine oxidase system were measured as DMPO-OH or DMPO-OOH spin adducts. L-Ascorbic acid 2-[3,4-dihydro-2,5,7,8-tetra-methyl-2-(4,8,12-trimethyltridecyl)-2 H-1-benzopyran-6yl-hydrogen phosphate] potassium salt (EPC-K1) and bovine erythrocyte superoxide dismutase (SOD) were used as standards for .OH and O2*-, respectively. Nasunin directly scavenged O2*- with a potency of 143+/-8 SOD-equivalent units/mg), and inhibited formation of DMPO-OH (0.65+/-0.07 EPC-K1 micromol/mg). A spectrophotometric study showed that nasunin formed an iron complex with a molar ratio of nasunin : Fe3+ of 2 : 1. Therefore, hydroxyl radical scavenging by nasunin is not due to direct radical scavenging but inhibition of .OH generation by chelating iron. Nasunin (1 microM) significantly protected against lipid peroxidation of brain homogenates (p<0.001) as measured by malonaldehyde and 4-hydroxyalkenals. These findings demonstrate that nasunin is a potent O2*- scavenger and iron chelator which can protect against lipid peroxidation.     

Reassessment of the use of desferrioxamine B in iron overload.

Treatment of transfusion-induced iron overload by daily intramuscular injection of the chelator desferrioxamine has not produced impressive urinary iron excretion. We attempted to augment net iron excretion by altering both the route and quantity of chelator administered. Two ascorbic acid-replete patients excreted a mean of 14.5 mg of iron per 24 hours after a single intramuscular injection of 750 mg of desferrioxamine. Excretion increased to a mean of 44.9 mg when this dose was delivered by a continuous 24-hour intravenous infusion. When the intravenous dose of chelator was increased incrementally to as high as 16,000 mg per 24 hours, iron excretion increased up to 180 mg per day. At these high-dose levels, efficiency of binding of iron to chelator was compromised. Constant exposure of the labile iron pool to a chelating agent markedly enhances net iron excretion in splenectomized transfusion-dependent patients.     

Correlation of postabsorptive serum iron increase and erythrocyte-59Fe-incorporation with the whole body retention of absorbed59Fe

Summary Following the oral application of 100 mg⁵⁹Fe-equivalents in different ferrous iron preparations the whole body retention (R) and erythrocyte incorporation (EI) of absorbed⁵⁹Fe as well as the postabsorptive serum increase SII after 3 h were estimated by intraindividual and simultaneous comparison in healthy subjects with normal and depleted iron stores. The regression equationR=(0.0±0.4)+(1.16±0.04)×EI and a Spearman rank correlation coefficientr=0.93 were found to describe the correlation between whole body-⁵⁹Fe-retentionR (mg Fe) and erythrocyte-⁵⁹Fe-incorporation EI (mg Fe) after 2 weeks. The correlation between the postabsorptive serum iron increase SII (µmol/l) after 3 h and the whole body-⁵⁹Fe-retentionR after 2 weeks is characterized by the regression equationR=(0.0±0.8)+(0.36±0.03)×SII and a correlation coefficientr=0.88. Whole body counting of absorbed⁵⁹Fe represents the most sensitive, reliable and only quantitative method for measuring iron absorption or bioavailability in man. If however this method is not available postabsorptive serum increase and it's conversion into iron absorption using the described regression equation is a useful substitute for the at least semiquamtitative estimation of iron bioavailability from oral iron preparations.     

Structural and cellular adaptation of duodenal iron uptake in rats maintained on an iron-deficient diet

Iron deficiency induced in rats maintained on a commercial diet with a low iron content has been used to investigate adaptive mechanisms that enhance duodenal iron uptake. These adaptive changes have been divided into those that result from changes in villus surface area (structural adaptation) and those that reflect changes in the way individual enterocytes express iron transport function (cellular adaptation). Cellular adaptation was assessed by carrying out microdensitometry of autoradiographs prepared from duodenal tissue previously incubated for 5 min in 200 micromol/l 59Fe2+-ascorbate. Structural adaptation was studied by performing image analysis of microdissected and sectioned villi. Cellular adaptation involved increased iron uptake by enterocytes present in the lower villus. Thus iron deficiency resulted in a threefold enhanced expression of uptake in the lower 100 microm villus (3.9+/-2.4 versus 12.6+/-1.5 arbitrary units, P<0.001). Maximal uptake was reached in the upper region of both control and iron-deficient villi, but iron deficiency had no effect on cellular uptake at this part of the villus. Structural adaptation involved the lengthening (+16%, P<0.05) and broadening (+14%) of villi in the duodenum of iron-deficient rats. The resultant expansion in villus area caused a further increase in uptake that was mostly expressed in the upper villus. Maximal uptake corrected for structure occurred in the middle third of villi from control and iron-deficient rats. Cellular plus structural adaptation produced a twofold increase in iron uptake. More than half of this effect was caused by changes in villus structure. [3H]Thymidine labelling experiments revealed a slightly earlier expression of enterocyte iron uptake in iron deficiency.