Zinc absorption in adult humans: the effect of protein sources added to liquid test meals

The influence of different protein sources on Zn absorption was evaluated in healthy adults by radioisotopic labelling of single meals, followed by whole-body retention measurements 14 d after intake. Semi-synthetic liquid diets were used for the evaluation of different animal-protein sources and dephytinized soyabean-protein isolate (< 0.01 g phytic acid/kg). Zn absorption was measured in the same subjects from identical test meals containing no added protein. No statistically significant differences were found in the Zn absorption from test meals containing bovine whey, casein or egg albumen when compared with test meals without added protein. Bovine serum albumin (BSA) and soyabean-protein isolate (< 0.01 g phytic acid/kg) significantly reduced the mean absorption of Zn from 45-49% (no added protein) to 38.0 (SD 10.9) (BSA, P < 0.05) and 33.9 (SD 12.6)% (soyabean-protein isolate < 0.01 g phytic acid/kg, P < 0.01). These results demonstrate that Zn absorption is inhibited by certain protein sources, such as BSA and dephytinized soyabean-protein isolate, while other proteins have little or no effect.

Zinc and calcium apparent absorption from an infant cereal: a stable isotope study in healthy infants

Fractional apparent absorption of Zn and Ca from a wheat-milk-based infant cereal was studied in six healthy infants (18-30 weeks old). Mineral absorption was measured by a stable-isotope technique based on faecal excretion of the isotopes. Each test meal (40 g cereal) was extrinsically labelled with 70Zn and 42Ca before intake. All faecal material passed during the 21 d following intake of the labelled test meal was collected on trace-element-free nappies. Individual stool samples were analysed for their content of 70Zn and 42Ca by thermal ionization mass spectrometry. Apparent absorption was calculated as intake minus total faecal excretion of the isotopes over 68-92 h after administration. The fractional apparent absorption values for Zn and Ca were 33.9 (SD 16.4) % (range 19.2-63.9%) and 53.5 (SD 12.6) % (range 36.7-71.7%) respectively. Re-excretion of absorbed 70Zn (> 68-92 h to 21 d after intake of the labelled meal) was 0.44 (SD 0.38) % of administered dose while only one infant re-excreted detectable amounts of 42Ca (1.74% of administered dose). The analysis of individual stool samples confirmed that 72 h is a sufficient time period for complete collections of non-absorbed isotopes in faecal material from infants during the weaning period and that re-excretion of initially absorbed 70Zn and 42Ca (> 68-92 h to 21 d after intake of the labelled meal) is negligible.

Dietary fiber in weaning cereals: a study of the effect on stool characteristics and absorption of energy, nitrogen, and minerals in healthy infants

We evaluated the effect of increased dietary fiber (DF) content in weaning cereals based on wheat/soy (8.0 and 1.8% DF) and wheat/milk (5.3 and 2.0% DF) in healthy, formula-fed infants 7-17 weeks old. The study had a cross-over design, each infant acting as his or her own control. Stool characteristics and anthropometry were monitored over 4-week periods in groups of 34 (wheat/soy) and 23 (wheat/milk) infants. Absorption of zinc (Zn) and calcium (Ca) was studied by measuring the fecal excretion of stable isotopes during 72 h (70Zn and 42Ca) in a subgroup of the infants consuming wheat/soy cereals. Iron (Fe) bioavailability was evaluated by analysis of the incorporation of 58Fe into erythrocytes 14 days after administration. Fractional absorption (X +/- SD: 8.0 versus 1.8% DF) was 45.3 +/- 27.5 versus 41.2 +/- 19.4% of 70Zn and 63.4 +/- 15.8 versus 64.4 +/- 10.6% of 42Ca. Bioavailability of 58Fe varied between 1.0% and 5.4% (8.0% DF) and from <0.9% to 9.1% (1.8% DF). No significant difference in energy (95.3 +/- 2.0% versus 95.7 +/- 1.2%) or nitrogen (92.6 +/- 2.3% versus 93.0 +/- 1.6%) apparent absorption from the total diet was found during consumption of cereal with 8.0 and 1.8% DF. The intake of cereal decreased with higher DF content in the wheat/soy product: 34 +/- 23 g/d (8.0% DF) versus 42 +/- 23 g/d (1.8% DF), p < 0.01. While consuming the 8.0% DF product, 11 infants were reported to have "gritty stools"; no other differences were observed between different groups in stool characteristics or anthropometry. These results demonstrate no negative effect on the absorption of energy and nutrients with higher dietary fiber intake in primarily formula-fed infants. The impact of increased dietary fiber levels remains unknown in less well-nourished infants.

Manganese absorption in humans: the effect of phytic acid and ascorbic acid in soy formula

The absorption of manganese from soy formula was studied in adult volunteers by extrinsic labeling of test meals with 54Mn, followed by whole-body retention measurements for approximately 30 d after intake. Eight subjects participated twice in each of the two studies, acting as his or her own control. Soy formula containing the native content of phytic acid was compared with a similar dephytinized formula: geometric mean manganese absorption increased 2.3-fold from 0.7% (range: 0.2-1.1%) to 1.6% (range: 1.0-7.2%) (P < 0.01) with the dephytinized formula. In addition, the effect of the ascorbic acid content of the phytic acid-containing formula was investigated. Manganese absorption was not influenced by an increase in the ascorbic acid from 625 mumol/L (110 mg/L) to 1250 mumol/L (220 mg/L): the geometric mean manganese absorption was 0.6% (range: 0.3-1.0%) and 0.6% (range: 0.3-1.1%), respectively. In conclusion, fractional manganese absorption was approximately doubled by the dephytinization of soy formula but was not influenced by an increase in the ascorbic acid content of a soy formula containing the native amount of phytic acid.

Zinc absorption in adult humans: the effect of iron fortification

The effect of Fe fortification on the absorption of Zn was studied by radioisotopic labelling of single meals, followed by measurements of whole-body retention of 65Zn at 14 d after intake. Healthy adult volunteers participated in the study. Weaning cereal, wheat bread and infant formula, foods that are all frequently Fe-fortified, were evaluated in the study. The amounts of Fe added as FeSO4 were similar to the levels in commercial products in Europe and the USA, and were 200 or 500 mg Fe/kg (weaning cereal), 65 mg Fe/kg (white wheat flour) and 12 mg Fe/l (infant formula). For comparison, Zn absorption was measured in the same subjects, from identical test meals containing no added Fe. No statistically significant differences were found when Zn absorption from the Fe-fortified test meals was compared with that from non-Fe-fortified test meals. Fractional Zn-absorption values from Fe-fortified v. non-fortified meals were 31.1 (SD 11.9) v. 30.7 (SD 7.0)% (weaning cereal; 200 mg Fe/kg), 37.7 (SD 16.6) v. 30.2 (SD 9.9)% (weaning cereal; 500 mg Fe/kg), 36.5 (SD 14.4) v. 38.2 (SD 18.1)% (bread; 65 mg Fe/kg flour) and 41.6 (SD 8.1) v. 38.9 (SD 14.5)% (infant formula; 12 mg Fe/l). The addition of Fe to foods at the currently used fortification levels was thus not associated with impaired absorption of Zn and the consumption of these Fe-fortified foods would not be expected to have a negative effect on Zn nutrition.

Iron bioavailability studied in infants: the influence of phytic acid and ascorbic acid in infant formulas based on soy isolate

The influence of phytic acid and ascorbic acid content of soy formula on iron (Fe) bioavailability was investigated in infants by analysis of the incorporation of stable isotopes of Fe into red blood cells 14 d after administration using a double stable isotope technique. Paired comparisons were made with each infant acting as his or her own control. The geometric mean fractional Fe incorporation into red blood cells increased from 5.5 to 6.8% (p < 0.05) when soy formula with the native content of phytic acid was compared with a 83% dephytinized formula. A more pronounced effect was shown with soy formula containing no phytic acid; the mean fractional Fe incorporation increased from 3.9 (native phytic acid) to 8.7% (zero phytic acid; p < 0.001). A significant (p < 0.01) effect was also demonstrated when the Fe:ascorbic acid molar ratio in the native phytate-containing formula was increased from 1:2.1 to 1:4.2; mean fractional Fe incorporation increased from 5.9 to 9.6%. These results demonstrate that the Fe bioavailability from soy-based infant formulas can be similarly increased by either removing phytic acid or increasing the ascorbic acid content.

Sodium iron EDTA [NaFe(III)EDTA] as a food fortificant: the effect on the absorption and retention of zinc and calcium in women

The iron fortificant NaFeEDTA could have a potential negative effect on the metabolism of other minerals. We have used stable isotopes to monitor zinc and calcium metabolism in 10 women consuming a single meal of high-extraction wheat rolls (100 g flour) fortified with 5 mg Fe as either FeSO4 or NaFeEDTA. Six-day chemical balances were made simultaneously to study apparent zinc and calcium retention from the complete diet containing the differently iron-fortified breads (200 g flour; 10 mg added Fe/d). Mean 70Zn absorption from the bread meal increased from 20.9% with FeSO4 to 33.5% with NaFeEDTA (P < 0.05) whereas mean 44Ca absorption was 53.3% from both breads. When NaFeEDTA-fortified bread was consumed, there was a small but significant increase in urinary excretion of 70Zn and 44Ca. There was a similar small increase in urinary zinc excretion during the 6-d balance, although the apparent retention of zinc and calcium was not different. Thus, we found no negative overall effect of NaFeEDTA consumption on the metabolism of zinc and calcium. In contrast, the results suggest that NaFeEDTA added to low-bioavailability diets might increase zinc absorption as well as provide iron with high bioavailability.

Minerals and trace elements in infant nutrition

Little information is available on absorption and metabolism of minerals and trace elements during infancy. The lack of data is related to the methodological problems involved in these studies. By using stable isotopes as labels, studies can be conducted in infants without introducing exposure to radiation, or any other risk, and studies on bioavailability of minerals and trace elements during early life can therefore be performed. This paper discusses results from studies of trace element/mineral absorption and metabolism in infants, based on stable isotope techniques.

A double stable isotope technique for measuring iron absorption in infants

A stable isotope technique has been developed which uses 57Fe and 58Fe as labels and which enables the simultaneous measurement of Fe absorption from two test meals in infants. The method was evaluated by measuring Fe absorption from a commercial whey-adjusted infant formula in nine healthy infants aged 13-25 weeks. Each infant was fed 210 ml formula, labelled with either 57Fe or 58Fe, on four consecutive mornings, in random order. The total Fe content in each feed was 2.5 mg Fe; either as 2.5 mg 57Fe, or 0.6 mg 58Fe plus 1.9 mg Fe with normal isotopic composition. Isotopic enrichment of Fe in erythrocytes was measured by thermal ionization mass spectrometry 14 d after the last administration, and Fe absorption was calculated based on isotope ratio shifts, total circulating Fe and intake of each isotope. Geometric mean absorption for the 57Fe and 58Fe labels was 6.72 and 6.58% respectively, and the absorption of the two isotopes was not significantly different (Student's paired t test). By this technique, paired comparisons of Fe absorption can be obtained and systematic studies of the influence of dietary factors on Fe absorption during infancy can be conducted.

NaFe3+EDTA as a food fortificant: influence on zinc, calcium and copper metabolism in the rat

The general acceptance of NaFe3+EDTA for food Fe fortification has been partly restricted by concern over the influence of EDTA on the metabolism of other nutritionally important trace elements and minerals. We have investigated the influence of NaFe3+EDTA, and of increasing dietary levels of Na2EDTA, on Zn, Cu and Ca metabolism in rats fed on Zn-sufficient and Zn-deficient soya-bean-isolate-based diets. With the Zn-deficient diets, changing the dietary Fe compound from FeSO4 to NaFe(3+)-EDTA significantly (P < 0.05) increased mean apparent Zn absorption from 50.2 to 67.4%, urinary Zn excretion from 2.0 to 4.0% of intake, and Zn retention from 48.2 to 63.4%. Increasing the dietary EDTA level to 1000 mg/kg further increased Zn absorption to 78.1%, urinary Zn excretion to 15.6% of intake and Zn retention to 62.5%. Increased Zn retention was accompanied by a significant increase in weight gain indicating that the extra Zn was available for normal metabolic processes. With rats fed on the Zn-sufficient diet, NaFe3+EDTA and Na2EDTA similarly increased the absorption, urinary excretion and retention of Zn but to a lesser extent. NaFe3+EDTA, however, had no influence on the absorption, urinary excretion and retention of Cu and Ca, and additional Na2EDTA caused only minor increases in Cu absorption and retention and in the urinary excretion of Ca. We conclude that using NaFe3+EDTA as a food fortificant would have no detrimental effect on the metabolism of Zn, Cu and Ca and, in some situations, could improve Zn absorption and retention from low-bioavailability diets.

Influence of lactoferrin on iron absorption from human milk in infants

Lactoferrin (Lf) is a major iron (Fe)-binding protein in human milk and has been proposed to facilitate Fe absorption. The potential effect of Lf on Fe absorption was investigated by measuring Fe absorption in infants fed breast milk (with its native content of Lf) and the same milk from which Lf had been removed (> 97%) by treatment with heparin-Sepharose. Eight breast-fed infants (2-10 mo; mean age 5 mo) were fed 700 to 1000 g of each milk in a randomized, cross-over design with each child acting as his/her own control. The milk was labeled with 8.6 mumol (0.5 mg) of 58Fe and Fe absorption was measured by quantifying the incorporation of the isotope into red blood cells 14 d after intake using thermal ionization mass spectrometry. Fractional Fe absorption was significantly lower (p < 0.05) from breast milk than from Lf-free breast milk. The geometric mean (range) was 11.8% (3.4-37.4%) for breast milk and 19.8% (8.4-72.8%) for Lf-free breast milk. These results do not support a direct role for Lf in the enhancement of Fe absorption from human milk at this age. In addition, Fe absorption (11.8%) from human milk fed over several feeds was lower than that previously reported for single feed studies.

Modified EPR spectra of the tyrosineD radical in photosystem II in site-directed mutants of Synechocystis sp. PCC 6803: identification of side chains in the immediate vicinity of tyrosineD on the D2 protein

The oxidizing side of photosystem II contains two redox-active tyrosyl side chains, TyrZ and TyrD, and a cluster of Mn atoms involved in water oxidation. The structural environment of these components is unknown, and with computer-assisted modeling we have created a three-dimensional model for the structures around TyrZ and TyrD [Svensson et al. (1990) EMBO J. 9, 2051-2059]. Both tyrosines are proposed to form hydrogen bonds to nearby histidine residues (for Synechocystis 6803, these are His190 on the D1 and His189 on the D2 proteins). We have tested this proposal by electron paramagnetic resonance (EPR) spectroscopy of TyrDox in mutants of the cyanobacterium Synechocystis 6803 carrying site-directed mutations in the D2 protein. In two mutants, where His189 of the D2 protein is changed to either Tyr or Leu, the normal EPR spectrum from TyrDox is replaced by narrow, structureless radical signals with g-values similar to that of TyrDox (g approximately 2.0050). The new radicals copurify with photosystem II, are dark-stable, destabilized by elevated pH, and light-inducible, and originate from radicals formed by oxidation. These properties are similar to those of normal TyrDox, and we assign the new spectra to TyrDox in an altered environment induced by the point mutation in His189. In a third mutant, where Gln164 of the D2 protein was mutated to Leu, we also observed a modified EPR spectrum from TyrDox. This is also consistent with the model in which this residue is found in the immediate vicinity of TyrDox. Thus the results provide experimental evidence supporting essential aspects of the structural model.

The effect of individual dietary components on manganese absorption in humans

Our knowledge of dietary factors affecting manganese absorption in man is very limited. In this study we used a recently developed radionuclide technique to measure manganese absorption in human adults. Using paired observations, we explored the effects of adding calcium and manganese to human milk on manganese absorption. Furthermore, the effects of adding phytate, phosphate, and ascorbic acid to infant formula as well as iron and magnesium to wheat bread were evaluated. Addition of calcium to human milk resulted in a significant decrease in manganese absorption whereas no significant differences in manganese absorption were observed as a result of the other test meals administered with and without the evaluated dietary component, respectively. Thus, manganese absorption was not significantly affected by most dietary factors evaluated in this study, except for the addition of calcium to human milk.

Selenium status and absorption of zinc (65Zn), selenium (75Se) and manganese (54Mn) in patients with short bowel syndrome

Selenium level and activity of glutathione peroxidase in plasma were studied in seven patients with extensive short bowel resection due to Crohn's disease, before and during 27-54 weeks of intake of a vitamin and trace element supplement containing 50 micrograms of selenium as sodium selenite. Initial levels of selenium were normal in all except one of the patients. The supplementation had no or minor effects on plasma selenium levels and glutathione peroxidase activity. The absorption of zinc, manganese and selenium was measured with a radionuclide technique before and/or after the supplementation period in five of the patients. The absorption of zinc and manganese was similar to that observed earlier in healthy subjects, while the absorption of selenium was significantly lower. The results indicate that a higher selenium intake or a different form of selenium is needed in these patients to compensate for the impaired bowel function.

Effect of long-term trace element supplementation on blood trace element levels and absorption of (75Se), (54Mn) and (65Zn)

A vitamin and trace element supplement containing recommended dietary amounts or "safe and adequate" levels was given to ten healthy subjects for 12 to 35 weeks. Plasma levels of selenium and zinc, activity of glutathione peroxidase in plasma and platelets, whole blood manganese, activity of superoxide dismutase in hemolysate, activity of alkaline phosphatase in serum, iron status indices and urinary excretion of zinc and selenium were measured. A small but significant change in plasma selenium from 1.01 +/- 0.14 mumol/L to 1.08 +/- 0.10 mumol/L was observed after two weeks. However, at the end of the supplementation plasma selenium levels did not differ from the initial levels. Plasma glutathione peroxidase levels showed a similar trend and changes in glutathione peroxidase activity in platelets were also transient. A small increase in serum zinc values was observed after 30 weeks of supplementation. No significant changes were observed in the other blood and urine parameters studied. In seven of the subjects absorption of zinc, manganese and selenium was measured after 30-31 weeks of supplementation by a radionuclide technique. The absorption of selenium and manganese after long term supplementation was 30-50% lower than observed previously in non-supplemented subjects. In conclusion, present available indices of trace element status are only to a limited extent affected by 30 weeks of a doubling of the normal dietary intake.

Identification of transferrin as the major plasma carrier protein for manganese introduced orally or intravenously or after in vitro addition in the rat

It is known that the metabolic handling of manganese (Mn) introduced via the diet or by intravenous injection is quite different. We hypothesized that this difference could be due in part to different proteins carrying Mn in plasma that could affect tissue uptake and retention. To test this idea, 54Mn was administered orally or intravenously to rats, and blood samples were taken by cardiac puncture at various time points postdosing. Plasma proteins were separated using fast protein liquid chromatography with a combination of anion exchange and gel filtration columns. Using these methods, independent of the route of 54Mn administration, transferrin was identified as the major Mn-binding protein in plasma. The identity was further confirmed by SDS-polyacrylamide gradient gel electrophoresis and Western blotting. These results conclusively show that 54Mn in plasma is carried by transferrin, regardless of route of administration and time postdosing.

Manganese absorption from human milk, cow's milk, and infant formulas in humans

Manganese absorption from human milk, cow's milk, and infant formulas was studied in humans by using extrinsic labeling of the diets with manganese 54 or manganese 52 and whole-body retention measurements. The fractional manganese absorption from human milk (8.2% +/- 2.9%) was significantly different when compared with cow's milk (2.4% +/- 1.7%), soy formula (0.7% +/- 0.2%), and whey-preponderant cow's milk formula with 12 mg/L of iron (1.7% +/- 1.0%) and without iron fortification (2 mg/L of iron) (3.1% +/- 2.8%), while no significant difference was observed between a whey-preponderant cow's milk formula with 7 mg/L of iron (5.9% +/- 4.8%) and human milk. The total amount of absorbed manganese was significantly higher from the non--iron-fortified cow's milk formula (2 mg/L of iron) as compared with human milk, while no significant differences were observed for the other milks and formulas.

Manganese retention in man: a method for estimating manganese absorption in man

Whole-body retention of 54Mn was studied in man by measurements in a sensitive whole-body counter after intake of an extrinsically labeled infant formula. Reproducible retention figures at day 10 were observed after repeated administrations to six subjects; 2.3 +/- 1.1, 3.3 +/- 3.1, and 2.4 +/- 1.4% (means +/- SD) for three separate occasions. Interindividual variation of manganese retention after intake of the same labeled diet was, however, shown to be substantial. Retention at day 10 was 2.9 +/- 1.8% (means +/- SD) and varied from 0.6 to 9.2% when measured in 14 healthy subjects. Large interindividual variations in rate of excretion were observed. The retention measurements for days 10-30, however, could be closely fitted to a single exponential function for each individual. The results regarding intraindividual and interindividual variation in Mn retention indicate that factors influencing Mn absorption can be identified only by repeated administrations using each subject as his/her own control.

Intrinsic and extrinsic labeling for studies of manganese absorption in humans

A dual-radioisotope method was used to simultaneously study whole-body manganese retention from a chicken liver based meal intrinsically labeled with 54Mn and extrinsically labeled with 52Mn. Manganese retention was monitored in a sensitive whole-body counter during approximately 30 d in six young adult women. Both radioisotopes were retained to a similar degree and excreted at identical rates. Retention at d 5 was 14.4 +/- 10.3 and 14.0 +/- 9.9% while retention at d 10 was 5.0 +/- 3.1 and 5.0 +/- 3.0% (X +/- SD) for 54Mn and 52Mn, respectively. From these results we conclude that the intrinsic and extrinsic Mn isotopes did form a common pool before absorption. The results can therefore be regarded as a direct validation of the use of extrinsic labeling for studies of Mn retention for estimating Mn absorption in man.

Retention of selenium (75Se), Zinc (65Zn) and manganese (54Mn) in humans after intake of a labelled vitamin and mineral supplement

The whole body retention of 75Se, 65Zn, and 54Mn after intake of a labelled vitamin and mineral supplement was followed in 12 healthy volunteers. The supplement had a vitamin and mineral content according to recommended dietary allowances or the so-called "safe and adequate levels" for trace elements, including 15 mg of zinc as zinc citrate, 50 micrograms of selenium as sodium selenite and 2.5 mg of manganese as manganese sulphate. The supplement was taken either in the fasting state or together with a light meal. Retention day 14 was 48 +/- 6%, 33 +/- 6% and 5 +/- 2% (mean +/- SD) for selenium, zinc and manganese, respectively, when the supplement was taken fasting and 45 +/- 3%, 8 +/- 1% and 1.0 +/- 0.2% when it was taken with food. During day 1-14, 27%, 1% and less than 0.01% of the administered selenium, zinc and manganese radionuclides, respectively, were excreted in the urine. Based on the rate of turnover of the radionuclides and the urinary losses of 75Se, the absorption of selenium, zinc and manganese from the supplement was estimated to be 89 +/- 5%, 38 +/- 7%, 9 +/- 3% (mean +/- SD) in the fasting state and 87 +/- 4%, 10 +/- 2% and 2 +/- 1% with food. These results indicate that when a supplement is taken with food the minerals are absorbed and metabolized in the same way as are native minerals in food. When the supplement is taken in the fasting state, the absorption of zinc and manganese can be substantially higher.

The effect of vegetables and beet fibre on the absorption of zinc in humans from composite meals

1. The absorption of zinc in humans from composite meals, was determined by extrinsic labelling of the meals with 65Zn and measurement of the whole-body retention of the radioisotope. 2. Low-Zn (mean 25 mumol) chicken meals with 150 g white bread or 225 g potatoes, carrots, turnips, cabbage or green peas were studied. The effect of a beet-pulp-fibre preparation used as a breakfast cereal, in bread and as a meat extender on Zn absorption was also studied. 3. The mean percentage absorption from the chicken meals with white bread, carrots and cabbage was significantly different from the meals with potatoes, turnips and green peas. When the amount of Zn in the meals was taken into account a slightly higher absorption was observed from the white-bread meal compared with the meals with potatoes and cabbage, while no differences were seen between the vegetable meals. 4. The beet-pulp-fibre preparation did not affect the extent of Zn absorption when used as a meat extender. The absorption of Zn was higher when the beet fibre was included in bread than when used as müesli. 5. The results obtained suggest that, besides the low-Zn content in vegetables, a large intake of vegetables or a pure-vegetable-fibre preparation has no significant effect on Zn availability from animal-protein-based meals.

Zinc status and dark adaptation in patients subjected to total gastrectomy: effect of zinc supplementation

Totally gastrectomized patients could be regarded at risk for development of nutritional deficiencies including trace elements. Sensitive indices for early detection of these deficiencies are lacking. In order to evaluate the nutritional status of trace elements in patients previously subjected to a total gastrectomy we studied the zinc status in 10 patients by determining serum zinc, urinary zinc excretion and also, by a simplified dark adaptation test, the effect on these parameters of 4 weeks of zinc supplementation. The serum zinc level but not the 24-h urinary zinc excretion was lower in gastrectomized patients compared to age-matched controls. A slower dark adaptation was observed in the former patients when measured in the non-fasting state but not when fasted. Dark adaptation was slower in the patients as well as the age-matched controls compared to younger healthy subjects. Zinc supplementation increased the serum zinc levels and the urinary zinc excretion in the gastrectomized patients but had no effect on dark adaptation.

Oral iron, dietary ligands and zinc absorption

The effect of iron on zinc absorption in humans was investigated by using 65Zn and whole-body counting after 2 wk. Increasing the molar ratio of ferrous iron (with ascorbic acid) to zinc from 1:1 to 2.5:1 did not affect absorption of zinc from water when given in a fasting state; 59 and 58% was absorbed, respectively. However, at an Fe:Zn ratio of 25:1, zinc absorption from water decreased significantly to 34%. When oral iron in the same ratio to zinc was given with a meal, no inhibitory effect was observed (25, 23 and 22%, respectively). Addition of the zinc ligand, histidine, to the water solution decreased the inhibitory effect of the higher dose of iron, resulting in a zinc absorption of 47%. Two weeks of iron preloading did not affect zinc absorption from water. The results demonstrate that when a multimineral supplement is taken on an empty stomach, excessive iron levels can negatively affect zinc absorption. Intake of the supplement with a meal or with a zinc ligand (such as histidine) may overcome this inhibitory effect.