A moderately high intake compared to a low intake of zinc depresses magnesium balance and alters indices of bone turnover in postmenopausal women


Objective: To determine whether moderately high or low intakes of zinc adversely affect the copper status of postmenopausal women to result in unfavorable changes in calcium and magnesium metabolism and other indicators of bone turnover.

Design: After a 10-day equilibration period in which the diet provided 31.5 μmol (2 mg) Cu and 137.7 μmol (9 mg) Zn/8.4 MJ (2000 kcal), the subjects were randomly divided into two groups, with one group fed the basal diet supplemented to provide15.7 μmol (1 mg) Cu/8.4 MJ, and the other group fed the same diet supplemented to provide 47.2 μmol (3 mg) Cu/8.4 MJ. After equilibration, both groups were fed the basal diet with no zinc supplemented (provided 45.9 μmol [3 mg] Zn/8.4 MJ) for 90 days; this was followed by another 10-day equilibration period before the basal diet was supplemented with zinc to provide 811 μmol (53 mg)/8.4 MJ for 90 days.

Setting: The metabolic unit of the Grand Forks Human Nutrition Research Center, Grand Forks, ND, USA.

Subjects: A total of 28 postmenopausal women recruited by advertisement throughout the United States of America. Among them, 25 women (64.9+6.7 y) completed the study; 21 as designed.

Results: The moderately high intake compared to the low intake of zinc increased the excretion of magnesium in the feces and urine, which resulted in a decreased magnesium balance. In the women fed low dietary copper, plasma osteocalcin was higher during the low-zinc than high-zinc dietary period. The urinary excretion of N-telopeptides was increased and the serum calcitonin concentration was decreased by high dietary zinc regardless of dietary copper.

Conclusions: A moderately high intake of zinc (811 μmol/day; 53 mg/day) did not induce changes in copper metabolism that resulted in unfavorable changes in bone or mineral metabolism. However, low dietary zinc (45.9 μmol/day; 3 mg/day) apparently resulted in undesirable changes in circulating calcitonin and osteocalcin. As a moderately high intake of zinc decreased magnesium balance, further study of the possibility that a high intake of zinc is a health concern for individuals consuming less than the recommended amounts of magnesium is warranted.


Both copper and zinc have biochemical roles important for normal bone formation and turnover (Danks, 1988; Strain, 1998; Yamaguchi, 1998). For example, copper is a cofactor for lysyl oxidase, an enzyme involved in the maturation or crosslinking of collagen to form stable fibrils upon which calcium is deposited to form bone. Among the zinc-requiring enzymes involved in bone turnover are alkaline phosphatase, which has a role in bone calcification, and collagenase, which is involved in bone resorption. Because of these important roles, both copper and zinc deficiencies result in bone abnormalities. Copper deficiency in animals, including sheep, cattle, pigs, dogs, chickens, and rats results in fragile brittle bones (Danks, 1988). Copper deficiency in preterm infants, and children with Menke's syndrome results in skeletal defects including osteoporosis, cupping and flaring of the metaphyses of long bones and spontaneous fractures (Danks, 1988; Uauy et al, 1998). Zinc deficiency in animals, including chicks, cows, pigs, monkeys, and rats results in delayed or defective mineralization, widened growth plates, indistinct zones of provisional calcification, and malformed bones (Hurley, 1981; Yamaguchi, 1998). Both copper and zinc supplementations have been shown to inhibit bone loss in adult humans (Strause et al, 1994; Eaton-Evans et al, 1996).

Although zinc supplementation has been shown to be beneficial to bone growth, development and maintenance, it has been sometimes shown to affect adversely calcium metabolism in humans (Spencer et al, 1987) and bone characteristics in rats (Song et al, 1986; Kenny & McCoy, 1997). For example, increasing dietary zinc to a moderately high concentration of 1.1 mmol (72 mg)/kg diet decreased the bone strength of rats fed a low-calcium diet (Kenny & McCoy, 1997). When a moderately high intake of zinc has detrimental effects, it is often attributed to causing a copper (Walsh et al, 1994; Sandstead, 1995) or iron (Aggett et al, 1983; Donangelo et al, 2002) deficiency or altered utilization in humans. This occurs for copper because extremely high amounts of zinc have been shown to interfere with the uptake and metabolism of copper. For example, zinc supplementation supplying 2.3 mmol (150 mg) Zn/day for an adult has been used as a therapeutic agent for the treatment of Wilson's disease, an inherited, autosomal recessive disease of copper accumulation (Brewer, 2000). Sandstead (1995) has reviewed the evidence that suggests a moderately high intake of zinc, even those close to the recommended dietary allowance (RDA), should be considered a toxic risk because of the possibility of inducing copper deficiency pathology in humans. The RDAs for zinc in the United States at the time of this review were 12 and15 mg (184 and 229 μmol)/day for adult women and men, respectively (Food and Nutrition Board, 1989), but these were recently lowered to 8 and 11 mg (122 and 168 μmol)/day (Food and Nutrition Board, 2001). Although there is some compelling evidence for this suggestion, some reported findings challenge its unequivocal acceptance. These include findings reported by us earlier (Milne et al, 2001), which showed that an inadequate intake of zinc (45.9 μmol/day or 3 mg/day) was more effective than a moderately high intake of zinc (811 μmol/day or 53 mg/day) in inducing changes associated with a decreased copper status in postmenopausal women. Those findings were consistent with the lack of a report showing that specific copper-deprivation signs such as decreased serum ceruloplasmin and platelet cytochrome-c oxidase are induced by moderate zinc supplementation in humans. Thus, although it is clear that low and moderately high intakes of zinc affect mineral and bone metabolism, it has not been definitively established that this occurs with high intakes by changing copper status or utilization.

The work presented here was part of a larger study that was performed to determine the effects of moderately deficient and excessive intakes of zinc on copper metabolism and use in humans fed low and plentifully adequate amounts of copper. The specific objective of the component presented in the following was to establish whether high dietary zinc adversely affects mineral (calcium and magnesium) metabolism and bone status indicators more markedly when dietary copper is relatively low. At the time of the study, the United States tolerable upper intake level (UL) of 40 mg (612 μmol)/day for zinc had not appeared. Thus, the basis for making the higher intake of zinc 53 mg (811 μmol)/day was that it was near the amount used in early studies (Fischer et al, 1984; Yadrick et al, 1989), suggesting that a moderately high intake of zinc decreased copper status. In the study, the low or marginal intake of copper (about 15.7 μmol/8.4 MJ or 1 mg/2000 kcal) was considered so because it was less than the lower limit (23.6 μmol/day or 1.5 mg/day) of the estimated safe and adequate daily dietary intake (ESADDI) established by the Food and Nutrition Board (1989) at the time of the study. The adequate intake (47.2 μmol/day or 3 mg/day) was the upper limit of the ESADDI. The upper limit was used to ensure the prevention of the potential induction of copper deficiency by high dietary zinc. After the present study was performed, a new United States copper RDA of 0.9 mg (14.2 μmol)/day for adults was established (Food and Nutrition Board, 2001). The zinc:copper ratios ranged from 1 to 53 in the experimental diets. Postmenopausal women were chosen to be subjects in the study because they were employed in studies showing an effect of copper (Eaton-Evans et al, 1996) and zinc (Strause et al, 1994) on bone loss, and because of their susceptibility to osteoporosis.

Subjects and methods

A total of 28 postmenopausal women were recruited for the study after they had been informed in detail both verbally and in writing of the nature of the research and associated risks, and after medical, psychological, and nutritional evaluation had established that they had no underlying disease and were emotionally suited for the project. One volunteer was dismissed shortly after starting the study because of high blood pressure that was not present during recruitment; another volunteer was dismissed because of diet incompatibility; and one volunteer left the study for personal reasons. The experimental protocol was approved by the Institutional Review Boards of the University of North Dakota and US Department of Agriculture and followed the Guidelines of the Department of Health and Human Services and the Helsinki Declaration regarding the use of human subjects.

The study was conducted at two different times, with about half of the women participating each time. The 25 women who completed the study were between the ages of 50 and 76 y (64.9±6.7 y, mean±s.d.) at entry. They were 159.6±7.6 cm tall and weighed 65.1±9.5 kg at the beginning of the study. All subjects were white and did not smoke during the study. In all, 21 subjects completed the study as designed.

The women were maintained in a metabolic unit under close 24-h supervision for 200 days. They were fed a constant, weighed basal diet of conventional foods that was low in copper (9.6 μmol/8.4 MJ or 0.6 mg/2000 kcal) and zinc (45.9 μmol/8.4 MJ or 3 mg/2000 kcal) on a 3-day menu rotation. The average energy intake of the subjects was 8.4 MJ (2000 kcal)/day. Details of the diet have been described (Davis et al, 2000). The diet was adequate in all other known nutrients with 12.3 mmol (492 mg) of the total of about 20 mmol (800 mg) daily intake of calcium, the United States RDA at the time of the study (Food and Nutrition Board, 1989), supplemented as calcium carbonate. The RDA was changed to an adequate intake (AI) level of 1200 mg (29.94 mmol)/day in 1997 (Food and Nutrition Board, 1997). Also, 7.4 mmol (180 mg) of the total of about 12.3 mmol (300 mg) daily intake of magnesium was supplemented as magnesium gluconate. Calcium and magnesium had to be supplemented because making the basal diet low in copper and zinc resulted in the exclusion of foods high in these minerals. The scheme of the experimental design is shown in Figure 1. This scheme shows that during an initial 10-day equilibration period, the subjects were fed the basal diet supplemented with 22 μmol (1.4 mg) of copper per day (31.5 μmol or 2 mg total) and 91.8 μmol (6 mg) of zinc per day (137.7 μmol or 9 mg total). After equilibration and initial testing, the women were assigned to two groups; one group was fed the basal diet supplemented with 63.4 μmol Cu/8.4 MJ (0.4 mg Cu/2000 kcal) (total of 15.7 μmol/8.4 MJ or 1 mg/2000 kcal) and the other group was fed the basal diet supplemented with 37.8 μmol (2.4 mg) of Cu/day (a total of about 47.2 μmol/8.4 MJ or 3.0 mg/2000 kcal). The remaining 190 days of the study were divided into two 90-day dietary periods for both groups. The basal diet (45.9 μmol Zn/8.4 MJ or 3 mg Zn/2000 kcal) with no zinc supplement was fed for the first 90-day period, and the diet supplemented with 765 μmol (50 mg) Zn/day was fed for the second 90-day period. The order of zinc supplementation was not randomized because of the concern that the high zinc intake would result in increased stores that would inhibit a response within 90 days to the deficient zinc intake by adults. The two 90-day periods were separated by a second equilibration period of 10 days, during which the basal diet was supplemented with 22 μmol (1.4 mg) of copper and 91.8 μmol (6 mg) of zinc per day. The purpose of the second equilibration period was to alleviate the concern that the response to a sudden excess of dietary zinc after being fed a deficient amount for an extended period of time might not be reflective of the alterations that would occur if subjects had been maintained on a normal diet. Zinc was supplemented as zinc gluconate and copper was supplemented as cupric sulfate in beverages served with the meals. All other aspects of the diet remained constant throughout the study. All water for drinking and food preparation was deionized (Super Q, Millipore, Bedford, MA, USA).

Figure 1

Scheme of the experimental design.

Foods were weighed to an accuracy of 1% during preparation in the metabolic kitchen and eaten quantitatively with the aid of spatulas and rinse bottles by the women. The dietary intake of each woman was based on energy needs as calculated by the Harris–Benedict (1919) equation, plus an additional 60% of basal energy expenditure for normal activity so that individual body weights were maintained within 2% of admission weights; the amount of basal diet fed was adjusted as necessary in 0.84 MJ (200 kcal) increments by proportionally changing the amounts of all foods.

Urine and feces were collected continuously during the last 78 days of each 90-day dietary period with precautions to avoid mineral contamination. Duplicate diets of 8.4 MJ (2000 kcal) were prepared daily for analysis and blended in a plastic blender with stainless-steel blades. Adjustments for differences in individual energy intakes were calculated proportionally.

The calcium and magnesium contents of 6-day composites for diets and feces were measured by inductively coupled argon plasma emission spectroscopy (Perkin-Elmer, Norwalk, CT, USA) after wet digestion of aliquots of freeze-dried material with nitric and perchloric acids (Analytical Methods Committee, 1990; Sims et al, 1990). Urinary calcium and magnesium were determined by analysis with inductively coupled argon plasma emission spectroscopy of an aliquot. Concurrent replicate analysis of Standard Reference Material (SRM) 1577B (bovine liver; National Institute of Standards and Technology, Gaithersburg, MD, USA) yielded values of 115±4 and 595±13 μg/g (n=7) during diet analysis, and 116±4 and 588±16 μg/g (n=26) during fecal analysis for calcium and magnesium, respectively; certified values were 116±4 and 601±28 μg/g. Concurrent replicate analysis of SRM 1548A (typical diet; National Institute of Standards and Technology, Gaithersburg, MD, USA) yielded values of 1914±92 and 568±11 μg/g (n=4) for calcium and magnesium, respectively; certified values were 1967±113 and 580±27 μg/g.

Holter electrocardiograms were performed twice during initial equilibration and then during weeks 4, 7, 9, 11, and 12 of each 90-day dietary period on all subjects. A four-fold increase from equilibration baseline in ventricular premature discharges while on the low copper diet prompted the supplementation with copper.

Blood was drawn at weekly intervals into plastic syringes from antecubital veins that had been distended by temporary use of a tourniquet after the subjects had fasted for 12 h. An average of no more than 235 ml/month was drawn. Aliquots for plasma ionized calcium and magnesium were mixed with lithium heparin and determinations made by using ion-selective electrodes (Nova Biomedical, Waltham, MA, USA) within 90 min of the time blood was drawn. Serum creatine kinase was determined by using a Cobas Fara II Centrifugal analyzer (Roche Diagnostic Systems, Inc., Somerville, NJ, USA). To assess the effect of the dietary manipulations on mineral metabolism and bone turnover, serum and urine biochemical markers determined were those in general use at the time the study protocol was formulated for institutional review. Commercially available radioimmunoassay kits (Incstar Corporation, Stillwater, MN, USA) were used to determine calcitonin, osteocalcin, and 25-hydroxycholecalciferol. As it was thought that zinc and copper would not have much of an effect on serum parathyroid hormone concentrations, it was not determined. Enzyme-linked immunoassay methods were used to determine serum bone-specific alkaline phosphatase (ALPHASE-B; Metra Biosystems, Inc., Mountain View, CA, USA) and urine N-telopeptides (Ostex, Seattle, WA, USA). Urine N-telopeptides were determined instead of pyridinoline because of variability reasons, and instead of hydroxyproline because of greater bone specificity.

Changes in some of the biochemical variables appeared only toward the end of each 90-day period. Thus, to ensure that changes were recognized, only the last two measurements for each of the variables were used in the statistical analysis. The last six 6-day balance periods were statistically analyzed. The data were analyzed by two-way repeated measures analysis of variance with a SAS general linear model program (SAS version 8.02, SAS Institute, Cary, NC, USA). Tukey's contrasts differentiated among means for variables that had been significantly affected by the treatments. A P≤0.05 was considered significant. Variances were expressed as a pooled standard deviation, calculated as the square root of the mean square error from the analysis of variance.


Two of the women fed the diet marginal in copper (15.7 μmol/day or 1 mg/day) exhibited changes in heart rhythm that resulted in the obligatory supplementation with copper midway through the study; data from these two subjects were not included in the reported results. After the study, it was discovered that two women were using a dental adhesive containing extremely high amounts of zinc; their data also are not included in the reported values. Thus, a total of 21 women completed the study as designed.

Table 1 shows that the dietary manipulations affected some indices of bone turnover. In the women fed low dietary copper, plasma osteocalcin was 38% higher during the low zinc than during the high zinc dietary period. In the women fed high dietary copper, dietary zinc did not affect plasma osteocalcin. The excretion of N-telopeptides was significantly increased and the serum calcitonin concentration was significantly decreased by high dietary zinc regardless of dietary copper. Plasma bone-specific alkaline phosphatase also tended to be increased by high dietary zinc (P<0.07). Serum 25-hydroxycholecalciferol was not affected by the dietary manipulations.

Table 1 Effect of zinc and copper intake on indicators of bone turnovera

Table 2 shows that dietary zinc affected magnesium metabolism. Percentages of dietary magnesium excreted in the feces and urine are reported in Table 2 because this calculation overcomes the problem of slight differences in dietary magnesium affecting the amount found in the feces without truly reflecting the actual effect on apparent absorption. The percentages of dietary magnesium that appeared in the feces and urine were significantly higher when dietary zinc was high than when low; this resulted in a significantly decreased magnesium balance when high dietary zinc was fed. Table 3 shows that the dietary manipulations did not significantly affect calcium metabolism. Although it appeared that high dietary zinc was having a similar effect on calcium balance as it was on magnesium balance, data variation resulted in no significant differences in the calcium metabolism variables.

Table 2 Effect of zinc and copper intakes on magnesium balance during the last 36 days of each 90-day dietary period
Table 3 Effect of zinc and copper intakes on calcium balance during the last 36 days of each 90-day dietary period

Table 4 shows the results of other magnesium- and calcium-related variables that were measured in the study. Neither plasma ionized calcium nor ionized magnesium was affected by dietary zinc. An apparent significant copper effect on ionized calcium occurred; however, this cross-sectional comparison most likely was significant because the women on the low-copper diet had a lower initial (equilibrium) value than the women on the high-copper diet. Creatine kinase, a magnesium-activated enzyme, was not significantly affected by zinc or copper.

Table 4 Effect of zinc and copper intake on indicators of calcium and magnesium statusa


A surprising finding in this relatively long term, tightly controlled feeding study was that a moderately high dietary intake of zinc (811 μmol/day or 53 mg/day) compared to a low dietary intake of zinc (45.9 μmol/day or 3 mg/day) increased the excretion of magnesium in the feces and urine, thus resulting in a decreased magnesium balance. As reported earlier (Milne et al, 2001), the moderately high dietary intake of zinc in the present study did not decrease copper balance (−1.6 compared to −1.4 μmol/day with low dietary zinc) when dietary copper was low (15.7 μmol/day or 1 mg/day), and significantly enhanced copper balance (+2.8 compared to −0.8 μmol/day with low dietary zinc) when dietary copper was unquestionably adequate (47.2 μmol/day or 3 mg/day). Moreover, the moderately high dietary intake of zinc did not decrease serum ceruloplasmin or platelet cytochrome c oxidase activity. These findings suggest that some of the changes considered undesirable by a moderately high-zinc intake reported in the past might have been the result of an antagonistic effect on magnesium metabolism, not copper metabolism. Although the attention to the antagonism between copper and zinc apparently has constrained investigation into the possibility that high dietary zinc is detrimental to the metabolism or utilization of another mineral such as magnesium, there are a few reports indicating that an antagonistic relationship could occur between magnesium and zinc. Fox et al (1988) found that magnesium deficiency signs of mortality, increased spleen size and anemia were greater in Japanese quail when dietary zinc was 100 mg/kg than when it was 20 mg/kg. In some phosphorylation reactions, ZnATP can replace, but is less effective, than MgATP (Saris et al, 2001). A Mg2+ transport protein named AtMHX has been identified in Arabidoposis (Shaul et al, 1999). This Mg2+/H+ exchanger also mediates the flux of Zn2+ and Fe2+ in exchange for H+, but a physiologically relevant role of AtMHX in the transport of Zn2+ seems unlikely because of the poor affinity of this ion for the transporter. It has been speculated that Mg2+/H+ exchange transport is present in the apical membrane of intestinal epithelial cells (Leonhard, 1999). Thus, it is possible that, when dietary zinc is high, it competes with magnesium in its exchange transport.

Perhaps some of the inconsistencies in finding that a moderately high-zinc intake affects some nonspecific signs of copper deprivation such as plasma total and HDL cholesterol and copper–zinc superoxide dismutase, as discussed previously (Milne et al, 2001), can be attributed to the subjects having a different magnesium status. Magnesium deficiency depresses HDL cholesterol in rats (McCoy et al, 1992). Magnesium deficiency also alters indicators of oxidative metabolism. For example, in rats, magnesium deficiency decreases heart rate, increases plasma, and does not affect erythrocyte copper–zinc superoxide dismutase activity (Uehara et al, 2001). These findings and the finding that high dietary zinc affects magnesium balance suggest that one must exercise caution in assuming that changes in blood cholesterol and oxidative status indicators caused by a moderately high intake of zinc are the result of an induced copper deficiency.

As the change in magnesium balance was unexpected, plasma ionized magnesium was the only specific magnesium status indicator determined in the present study; it was unaffected by the dietary manipulations. Creatine kinase, a magnesium enzyme that uses MgATP as a substrate, was measured for health status purposes; the change that approached significance in this enzyme was in the direction that one would not expect with a low magnesium status. As dietary magnesium was provided in an amount near the recently established United States RDA of 320 mg (13.17 mmol)/day for adult women (Food and Nutrition Board, 1997), and magnesium balance was not negative, it is likely that the decreased magnesium balance induced by the moderately high intake of zinc did not induce a relative magnesium deficiency. Thus, it is questionable whether a change in magnesium status contributed to the changes in bone turnover indices.

The changes in bone status indices as the result of the dietary manipulations are difficult to interpret. The highest value for plasma osteocalcin was found when the women fed the low dietary copper consumed deficient dietary zinc. This could be considered unfavorable because an increased concentration of serum total osteocalcin has been associated with increased hip bone loss in humans (Bauer, 2001) and with suppressed bone formation and stimulated bone resorption in rats (Tanimoto et al, 1991). High dietary zinc tended to increase bone-specific alkaline phosphatase, especially in the women fed low dietary copper. The alkaline phosphatase finding is strengthened by other reports showing that moderately high intakes of zinc increased plasma bone-specific alkaline phosphatase (Dimai et al, 1998; Peretz et al, 2000), which was considered desirable for bone mineralization and maintenance. The finding that urinary N-telopeptide was highest when dietary zinc was high is not supportive of a favorable effect on bone maintenance. The method for urinary N-telopeptide measurement is quite specific for the degradation product of type l collagen in bone during the resorption process (Christenson, 1997). However, the increased N-telopeptide excretion with high zinc intake might be indicating that bone collagenase activity was depressed and bone turnover was decreased during the low dietary zinc period. In zinc-deficient chicks, half-turnover time for tibia collagen was almost tripled and tibia collagenase activity was reduced by 40–80% when compared to zinc-adequate chicks (Starcher et al, 1980). These defects in collagen metabolism were suggested as being responsible for the leg deformities of zinc-deficient chicks. Thus, the decreased N-telopeptide excretion with low dietary zinc found in the present study might not necessarily indicate a beneficial effect on bone maintenance, especially in light of the plasma osteocalcin results. Suggesting that the finding of decreased circulating calcitonin when dietary zinc was relatively high is indicative of desirable change for bone maintenance also could be questioned. However, although pharmacologic amounts of calcitonin prevent bone loss associated with osteoporosis, low circulating calcitonin is not a characteristic of osteoporosis. The opposite is apparently true: excessive skeletal calcium release such as that occurring with osteoporosis apparently stimulates calcitonin secretion and increases the concentration of calcitonin in plasma (Tiegs et al, 1985; McCarroll, 1993). Thus, the finding of decreased serum calcitonin with relatively high dietary zinc probably should not be considered undesirable.

In summary, the present findings indicate that a moderately high intake of zinc does not induce changes in copper metabolism that result in a detrimental effect on bone or mineral metabolism. In contrast, low dietary zinc might be a nutritional stressor of bone maintenance in postmenopausal women. Although a moderately high intake of zinc did not induce a copper deficiency, it had an apparently unfavorable effect on magnesium balance. Further study is needed to ascertain whether the decreased magnesium balance resulting from a moderately high intake of zinc is a health concern, including bone health, for individuals consuming less than recommended amounts of magnesium.


  1. Aggett PJ, Crofton RW, Khin C, Gvozdanovic S & Gvozdanovic D (1983): The mutual inhibitory effects on their bioavailability of inorganic zinc and iron. In Zinc Deficiency in Human Subjects, eds AS Prasad, A Cadvar, GJ Brewer & PJ Aggett, pp 117–124. New York: Alan R, Liss.

    Google Scholar 

  2. Analytical Methods Committee (1990): Methods of destruction of organic matter. Analyst 85, 642–656.

  3. Bauer DC (2001): Biochemical markers of bone turnover: the study of osteoporotic fracture. In Bone Markers: Biochemical and Clinical Perspectives (Proc Meeting 2000), ed. R Eastell, pp 219–223. London: Martin Dunitz.

    Google Scholar 

  4. Brewer GJ (2000): Raulin Award Lecture: Wilson's disease therapy with zinc and tetrathiomolybdate. J. Trace Elem. Exp. Med. 13, 51–61.

    CAS  Article  Google Scholar 

  5. Christenson RH (1997): Biochemical markers of bone metabolism: an overview. Clin. Biochem. 30, 573–593.

    CAS  Article  Google Scholar 

  6. Danks DM (1988): Copper deficiency in humans. Ann. Rev. Nutr. 8, 235–257.

    CAS  Article  Google Scholar 

  7. Davis CD, Milne DB & Nielsen FH (2000): Changes in dietary zinc and copper affect zinc-status indicators of postmenopausal women, notably extracellular superoxide dismutase and amyloid precursor proteins. Am. J. Clin. Nutr. 71, 781–788.

    CAS  Article  Google Scholar 

  8. Dimai HP, Hall SL, Stilt-Coffing B & Farley JR (1998): Skeletal response to dietary zinc in adult female mice. Calcif. Tissue Int. 62, 309–315.

    CAS  Article  Google Scholar 

  9. Donangelo CM, Woodhouse LR, King SM, Viteri FE & King JC (2002): Supplemental zinc lowers measures of iron status in young women with low iron reserves. J. Nutr. 132, 1860–1864.

    CAS  Article  Google Scholar 

  10. Eaton-Evans J, McIlrath EM, Jackson WE, McCartney H & Strain JJ (1996): Copper supplementation and the maintenance of bone mineral density in middle-aged women. J. Trace Elem. Exp. Med. 9, 87–94.

    CAS  Article  Google Scholar 

  11. Fischer PWF, Giroux A & L’Abbe MR (1984): Effect of zinc supplementation on copper status in adult man. Am. J. Clin. Nutr. 40, 743–746.

    CAS  Article  Google Scholar 

  12. Food and Nutrition Board (1989): Recommended Dietary Allowances, 10th edition. Washington, DC: National Academy Press.

  13. Food and Nutrition Board (1997): Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. Washington, DC: National Academy Press.

  14. Food and Nutrition Board (2001): Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Washington, DC: National Academy Press.

  15. Fox MRS, Tao S-H, Fry Jr BE & Lee YH (1988): Production of Mg deficiency anemia by Zn and phytate in young Japanese quail. In Trace Elements in Man and Animals, Vol 6, eds LS Hurley, CL Keen, B Lonnerdal & RB Rucker, pp 575–576. New York: Plenum Press.

    Google Scholar 

  16. Harris JA & Benedict FG (1919): A Biometric Study of Basal Metabolism in Men. Carnegie Publication No. 279. Philadelphia: JB Lippincott.

    Google Scholar 

  17. Hurley LS (1981): Teratogenic aspects of manganese, zinc, and copper nutrition. Physiol. Rev. 61, 249–295.

    CAS  Article  Google Scholar 

  18. Kenny MA & McCoy H (1997): Adding zinc reduces bone strength of rats fed a low-calcium diet. Biol. Trace Elem. Res. 58, 35–41.

    Article  Google Scholar 

  19. Leonhard MS (1999): Do forestomach epithelia exhibit a Mg2+/2H+-exchanger? Magnesium Res. 12, 99–108.

    Google Scholar 

  20. McCarroll NA (1993): Bone disorders (osteoporosis). Anal. Chem. 65, 388R–395R.

    CAS  Article  Google Scholar 

  21. McCoy H, Kenney MA & Williams L (1992): Low magnesium diet depresses HDL cholesterol in rats. FASEB J. 6, A1666.

    Google Scholar 

  22. Milne DB, Davis CD & Nielsen FH (2001): Low dietary zinc alters indices of copper function and status in postmenopausal women. Nutrition 17, 701–708.

    CAS  Article  Google Scholar 

  23. Peretz A, Bergmann P, Papadopoulos T, Siderova V & Neve J (2000): Effect of zinc supplementation on biological parameters of bone turnover in healthy men. In Trace Elements in Man and Animals, Vol. 10, eds AM Roussel, RA Anderson & AE Favier, pp 1009–1012. New York: Kluwer Academic/Plenum Publishers.

    Google Scholar 

  24. Sandstead HH (1995): Requirements and toxicity of essential trace elements, illustrated by zinc and copper. Am. J. Clin. Nutr. 61 (Suppl), 621S–624S.

    CAS  Article  Google Scholar 

  25. Saris N-EL, Krestinina OV, Azarashvili TS, Odinokova IV, Tyynela J & Evtodienko YuV (2001): Regulation of ATP synthase by Ca2+ and Mg2+-dependent phosphorylation of subunit c. In Advances in Magnesium Research. Nutrition and Health, eds Y Rayssiguier, A Mazur & J Durlach, pp 101–106. Eastleigh, England: John Libbey.

    Google Scholar 

  26. Shaul O, Hilgemann DW, Almeida-Engler J, Van Montagu M, Inze D & Galili G (1999): Cloning and characterization of a novel Mg2+/H+ exchanger. EMBO J. 18, 3973–3980.

    CAS  Article  Google Scholar 

  27. Sims RL, Mullen LM & Milne DB (1990): Application of inductively coupled plasma emission spectroscopy to multielement analysis of foodstuffs used in metabolic studies. J. Food Compos. Anal. 3, 27–37.

    CAS  Article  Google Scholar 

  28. Song MK, Adham NF & Ament ME (1986): Levels and distribution of zinc, copper, magnesium, and calcium in rats fed different levels of dietary zinc. Biol. Trace Elem. Res. 11, 75–88.

    CAS  Article  Google Scholar 

  29. Spencer H, Rubio N, Kramer L, Norris C & Osis D (1987): Effect of zinc supplements on the intestinal absorption of calcium. J. Am. Coll. Nutr. 6, 47–51.

    CAS  Article  Google Scholar 

  30. Starcher BC, Hill CH & Madaras JG (1980): Effect of zinc deficiency on bone collagenase and collagen turnover. J. Nutr. 110, 2095–2102.

    CAS  Article  Google Scholar 

  31. Strain JJ (1998): Copper and postmenopausal osteoporosis. In Copper and Zinc in Inflammatory and Degenerative Diseases, eds KD Rainsford, R Milanino, JRJ Sorenson & GP Velo, pp 173–178. Lancaster: Kluwer Academic Publishers.

    Google Scholar 

  32. Strause L, Saltman P, Smith KT, Bracker M & Andon MB (1994): Spinal bone loss in postmenopausal women supplemented with calcium and trace minerals. J. Nutr. 124, 1060–1064.

    CAS  Article  Google Scholar 

  33. Tanimoto H, Lau K-HW, Nishimoto SK, Wergedal JE & Baylink DJ (1991): Evaluation of the usefulness of serum phosphatases and osteocalcin as serum markers in a calcium depletion-repletion rat model. Calcif. Tissue Int. 48, 101–110.

    CAS  Article  Google Scholar 

  34. Tiegs RD, Body JJ, Wahner HW, Barta J, Riggs BL & Heath III H (1985): Calcitonin secretion in postmenopausal osteoporosis. N. Engl. J. Med. 312, 1097–1100.

    CAS  Article  Google Scholar 

  35. Uauy R, Olivares M & Gonzalez M (1998): Essentiality of copper in humans. Am. J. Clin. Nutr. 67 (Suppl), 952S–959S.

    CAS  Article  Google Scholar 

  36. Uehara M, Chiba H, Fujii A, Matsuzaki H, Masuyama R & Suzuki K (2001): Induction of phospholipid hydroperoxides in relation to change of tissue mineral distribution caused by magnesium-deficiency in rats. In Advances in Magnesium Research. Nutrition and Health, eds Y Rayssiguier, A Mazur & J Durlach, pp 291–296. Eastleigh, England: John Libbey.

    Google Scholar 

  37. Walsh CT, Sandstead HH, Prasad AS, Newberne PM & Fraker PJ (1994): Zinc: health effects and research priorities for the 1990s. Environ. Health Perspect. 102 (Suppl. 2), 5–46.

    CAS  PubMed  PubMed Central  Google Scholar 

  38. Yadrick MK, Kenny MA & Winterfeldt EA (1989): Iron, copper, and zinc status: response to supplementation with zinc or zinc and iron in adult females. Am. J. Clin. Nutr. 49, 145–150.

    CAS  Article  Google Scholar 

  39. Yamaguchi M (1998): Role of zinc in bone formation and bone resorption. J. Trace Elem. Exp. Med. 11, 119–135.

    CAS  Article  Google Scholar 

Download references


We thank Dr Craig Boreiko and the International Lead Zinc Research Organization for financial support, LuAnn Johnson for statistical support, and Mary Rydell for manuscript preparation and handling. We also thank the members of the Grand Forks Human Nutrition Research Center clinical staff whose special talents and skills made this study possible: Leslie Klevay (medical supervision), James Penland (psychological supervision), Henry Lukaski (exercise physiology), Bonita Hoverson and staff (dietary), Sandra Gallagher and staff (clinical chemistry and metabolic unit), and Terrence Shuler and staff (mineral analysis).

Author information




The US Department of Agriculture, Agricultural Research Service, Northern Plains Area, is an equal opportunity/affirmative action employer, and all agency services are available without discrimination. Mention of a trademark or proprietary product does not constitute a guarantee or warranty by the US Department of Agriculture and does not imply its approval to the exclusion of other products that also might be suitable. This study was supported in part by the International Lead Zinc Organization. Portions of these data were presented at the VIth Conference of the International Society for Trace Element Research In Humans, Quebec City, Quebec, Canada, and published in abstract form (J. Trace Elem. Exp. Med. 14, 286–287, 2001).

Contributors: The study was designed and supervised by DBM and FHN. The paper was written by FHN.

Guarantor: FH Nielsen.

Corresponding author

Correspondence to F H Nielsen.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Nielsen, F., Milne, D. A moderately high intake compared to a low intake of zinc depresses magnesium balance and alters indices of bone turnover in postmenopausal women. Eur J Clin Nutr 58, 703–710 (2004). https://doi.org/10.1038/sj.ejcn.1601867

Download citation


  • zinc
  • copper
  • magnesium
  • bone
  • trace elements
  • calcium

Further reading