The Alteration of Magnesium, Calcium and Phosphorus Metabolism by Dietary Magnesium Deprivation in Postmenopausal Women Is Not Affected by Dietary Boron Deprivation

Home , Calcium metabolism

Magnesium Research (2004) 17, 3, 197-210 REVIEW ARTICLE Clinical paper The alteration of magnesium, calcium and phosphorus metabolism by dietary magnesium deprivation in postmenopausal women is not affected by dietary boron deprivation.

Forrest H. Nielsen United States Department offlgri culture, Agricultural Research Service, Grand Forks Human Nutrition Research Center, Grand For/cs, ND, 58202-9034, USA

Summary: A study with human volunteers was conducted to test the hypothesis that naturally occurring inadequate intakes of magnesium induce negative magnesium balance and undesirable changes in calcium metabolism variables, and that these changes are influenced by dietary boron. Diets composed of ordinary Western foods providing approximately 118 and 318 mg Mgid and approximately 0.25 and 3.25 mg Bid were fed in a double -blind Latin square design to 13 healthy, post menopausal Caucasian women (aged 50-78 years) living in a metabolic unit. Magnesium balance, which was positive when dietary magnesium was 318 mg/d, became negative when dietary magnesium was 118 rng/d. Magnesium deprivation decreased urinary calcium excretion, and significantly increased calcium balance when balance data analyzed came from all collections during the 42-day periods. Urinary phosphorus excretion was increased, but fecal phosphorus excretion was decreased, thus phosphorus balance was not significantly affected by magnesium deprivation. Magnesium deprivation did, not affect manganese or zinc balance. The balance data indicated that 700 mg of calcium, 1.0 mg of manganese, and 10 nig of zinc were adequate for post menopausal women. Magnesium deprivation increased serum 25-hydroxycholecalciferol and decreased serum total cholesterol concentrations. Boron deprivation increased but magnesium deprivation decreased urinary potassium excretion. Boron supplementation decreased serum 173- estradiol and progesterone when dietary magnesium was low. The dietary treatments did not affect serum calcitonin, parathyroid hormone, osteocalcin or alkaline phosphatase concentrations. One woman placed on consecutive magnesium-low dietary periods exhibited heart ventricular ectopy after consuming the magnesium-low diet for 72 days; the ectopy disappeared upon consuming the magnesium-adequate diet. The findings indicated that consuming an ordinary diet deficient in magnesium, resulting in negative magnesium balance, can affect calcium, potassium, and cholesterol metabolism. Dietary boron did not have an obvious effect on the response to magnesium deprivation. Key words: magnesium, boron, calcium, phosphorus, potassium, cholesterol, manganese, zinc

Introduction ous dysfunctional states that result in its malab- sorption or excessive excretion [1]. Whether inad- Magnesium deficiency is generally recognized as a equate intakes of magnesium by individuals not contributor to morbidity and mortality in numer- exposed to negative modifiers of magnesium

197 F.H. NIELSEN, ET AL. homeostasis is a practical nutritional concern is rhythm [17] and altered plasma lipid concentra- controversial. One expert group stated that a tions [17]. "dietary deficiency of magnesium of a severity Controlled human studies suggest that magne- sufficient to provoke pathologic changes is rare sium status alters the response to deficient [2]". Other magnesium experts have concluded intakes of boron, an element that may be essential that chronic primary magnesium deficiency for optimal bone health, brain function, and occurs frequently and has various pathological immune function [18, 19]. Boron deprivation exac- consequences including cardiovascular, bone and erbated heightened plasma calcitonin concentra- psychiatric disorders [3-6]. This controversy is tions, and decreased serum enzymatic ceruloplas- fueled by a lack of a good magnesium status indi- mm in magnesium-deprived post menopausal cator to use in well-controlled definitive studies to women not on estrogen therapy, but not when show that a diet low in magnesium provokes they were magnesium-adequate [18). When pathologic changes. This lack also has made it dietary magnesium was adequate, boron depriva- difficult to derive estimates of magnesium tion of post menopausal women increased eryth- requirements. For example, the United States rocyte superoxide dismnutase and depressed the Institute of Medicine Food and Nutrition Board 171 serum 170-estradiol increase induced by estrogen cited only one balance study to use for the deter- therapy; when dietary magnesium was inad- mination of the estimated average requirement equate, boron deprivation did not have significant (EAR) for women between the ages of 31 and effects [18]. In another study, boron deprivation of 50 years, and found no balance studies to use for post menopausal women increased the percent- women between the ages of 51 to 70 years. Thus, age of calcium lost in urine when dietary magne- the recommended dietary allowance (RDA) for sium was inadequate, but decreased the percent- women of 320 mg/d was based on usual dietary age when dietary magnesium was adequate 1201. intakes that had not been associated with any This report describes an experiment performed adverse consequences and one balance study that to test the hypothesis that naturally occurring did not control magnesium intakes [8] (self- inadequate intakes of magnesium induce negative selected diets). The French Society for Nutrition magnesium balance and undesirable changes in also set a RDA of 330 mg/d for women [9] . How- post menopausal women, and that these changes ever, the Scientific Committee for Food of the are influenced by dietary boron. European Communities did not set population reference intakes for adults [10] because they decided the data to do so were inadequate. Subjects and methodsmethods Instead, they gave an acceptable range of intakes Subjects for adults of 150 to 500 mg/d. The range in estimates of acceptable intakes of magnesium reflect Fifteen post-menopausal women were recruited reports by various research groups presenting for the study after they had been informed in data suggesting that the magnesium requirement detail both verbally and in writing of the nature of of women is met by intakes of 100 mg/d, [11] the research and associated risks, and after medi- 176 mg/d, [121 205 mg/d, [13] and 330 mg/d [14). cal, psychological, and nutritional evaluation had The preceding indicates the need for basic established that they had no underlying disease metabolic unit magnesium deficiency studies of and were emotionally suited for the project. One healthy people that associate a decrease in magne- subject was dismissed shortly after the study sium status with undesirable biochemical and began because it was determined she was not, in functional changes. Because of the paucity of suit- fact, post menopausal. Another subject was able biochemical indicators of primary magne- recruited to restore the 14-bed metabolic unit to sium deficiency, balance still is a useful measure full occupancy; the replacement subject did not of a change in magnesium status provided there is participate in an adaptation period. Another sub- a suitable period of adaptation to a change in ject was dismissed late in the study because of a magnesium intake that is determined by direct physicians recommendation that she receive analysis. Possible physiological and biochemical treatment for an asymptomatic carotid bruit that changes sensitive to dietary magnesium depriva- would have interfered with the study design; she tion that could lead to suboptimal function and was not replaced. The thirteen Caucasian women well-being include altered brain excitability, [151 that completed the study were not on estrogen impaired energy metabolism, [16] abnormal heart therapy, did not smoke, and ranged in age from 198 ALTERATION OF MAGNESIUM. CALCIUM AND PHOSPHORUS METABOLISM

50 to 78 years, in height from 155.1 to 179.7 cm, was performed three times weekly to maintain and in weight from 52.4 to 84 kg. The subjects body composition and physical work capacity. resided for the entire study in the metabolic Energy intake was adjusted in 200 kcal incre- research unit of the Grand Forks Human Nutrition ments during the course of the study to maintain Research Center that provided a common environ- body weight (determined daily) within ± 2% of ment for strict control of food consumption, admission weight. weight, exercise and data collection. Subjects Upon arrival in the metabolic unit, all women consumed only food and beverages provided by except one completed an adaptation period of the dietary staff and were chaperoned on all out- 21 days in which the basal diet was supplemented ings from the metabolic unit to ensure compliance with 200 mg Mg/d as magnesium gluconate cap- with the study protocol. stiles, and 3 mg B/d as sodium borate capsules. The participants gave their written informed After adaptation, each woman was subjected for consents to participate in the experimental proto- 42 days each to all four supplement combinations col that was approved by the Institutional Review created by the factorial crossing of 0 (placebo) Board of the University of North Dakota and the and 200 mg magnesium, and 0 (placebo) and 3 mg Human Studies Review Committee of the United boron daily. The placebos contained lactose pow- State Department of Agriculture, and followed the der. Supplements were administered in a double- guidelines of the Department of Health and blind, Latin square design, which resulted in dif- Human Services and the Helsinki Doctrine regard- ferent lengths of time of magnesium and boron ing the use of human subjects. deprivations. Seven women consumed the magnesium-deficient diet for two successive Experimental Protocol 42-day periods while dietary boron was being varied; the other six women had their 42-day The diet based on ordinary Western foods has magnesium-deficient dietary periods alternated been described [211. Briefly, the diet, presented in with the 42-day magnesium-adequate dietary peri- a 3-d rotating menu cycle, included chicken, beef, ods. Six women were fed the boron-low diet for pork, potatoes, rice, bread and milk, but was low two successive 42-day periods while dietary mag- in vegetables and fruits. The basal diet supplied nesium was being varied; the other seven women approximately 115 mg magnesium and 0.23 mg had their 42-day boron-low dietary periods alter- boron per 2 000 kcal. The diet energy distribution nated with the 42-day boron-adequate dietary peri- was 11% protein, 35% fat and 54% carbohydrate ods. and provided approximately (mg/2 000 kcal): For balance determinations, duplicate diets of potassium, 1 800 (630 added as potassium chlo- 2 000 kcal were prepared daily for analysis and ride to the food); calcium, 700 (135 as a calcium blended in a plastic blender with stainless steel gluconate supplement); iron, 28 (18 as a ferrous blades. Urine and feces were collected completely gluconate supplement); zinc, 9.5 (4 as zinc sulfate in plastic containers and bags, respectively, to supplemented to a lunch beverage); copper, 1.6 avoid mineral contamination. Venous blood (lim- (1.2 as copper sulfate supplemented to a lunch ited to < 250 mllmo), collected in plastic syringes beverage); manganese, 1.0; thiamin, 1.2; ribofla- from antecubital veins that had been distended by vin, 1.2; niacin, 14.3; ascorbic acid, 204; vitamin the temporary use of a tourniquet after the sub- 136, 2.0; vitamin B 12, 0.0036; folic acid, 0.467; vita- jects had fasted for 12 hr, was obtained weekly for min D, 0.005; a-tocopherol, 4.6, and (in Ill) vitamin routine health assessment and cholesterol deter- A, 6950. Dietary iron was provided in excess of the mination; additional blood was obtained every RDA to mitigate the decline in iron status as a second week for the determination of other result of phlebotomy during the experiment. All experimental variables. food was weighed with an accuracy of one-tenth Because potentially harmful electrocardio- of a gram during preparation in the metabolic graphic changes have been found in people kitchen and was completely consumed by the sub- depleted in magnesium, Holter electrocardio- jects with the aid of spatulas and rinse bottles. grams were performed during the initial adapta- Deionized water was consumed ad libitum. The tion period and about week 4 in each dietary initial energy requirement for each subject was period on each subject. Tapes obtained were determined by using the Harris and Benedict equa- scanned for signs of abnormal function (Model tion [22] and adding 50% to compensate for normal 363, Accuplus, Del Mar Avionics, Irvine, CA, USA) daily activities. Individually prescribed exercise by trained nurses. A significant increase from 199

F.H. NIELSEN, ET AL. Table I. Me 490 ± 80 for phosphorus, and 0.65 ± 0.28 and adaptation baseline in ventricular premature dis- 2.4 ± 0.8 for zinc. Balances were calculated as the Dietary tre charges while on a low magnesium diet prompted difference between intake and excretion (feces a premature entry into a high magnesium dietary plus urine; feces only for manganese) for the total Low Mg - L period. 42 days and last 24 days of each dietary period. At the time Holter electrocardiogramswere per- The 24-d data are presented because they corre- Low Mg - H formed, standard neuromuscular magnesium defi- spond to the time biochemical variables were High Mg - l ciency tests, checking for Trousseaus and measured. The balance or retention calculations High Mg - I Chvosteks signs (reflexes) were done. The test did not include surface or phlebotomy losses. Pooled SD for Trousseaus sign involves inflating a sphygmo- Blood was processed within 90 min to obtain Analysis o manometer cuff to 10 mm above systolic blood serum or plasma. ionized calcium concentrations MagneSiur pressure and holding this pressure for two min- Boron eff€ S after were measured in plasma obtained with lithium utes. Carpal spasm with relaxation 5 to 10 heparin and by using an ion selective electrode Magnesiux deflation is considered positive. The test for (Ionized Calcium 2, Radiometer America, inc., Low Mg = Chvosteks sign involves tapping the facial nerve Westlake, OH, USA); unfortunately, an ion selec- just anterior to the ear lobe and just below the tive electrode was not available for the determina- Data An zygornatic arch (or between the arch and the cor- tion of ionized magnesium. Serum calcium and ner of the mouth). A positive response ranges magnesium concentrations were determined by All balar from a simple twitching of the corner of the mouth flame atomic absorption spectrometry (Perkin to a twitching of all facial muscles on the stimu- during tI Elmer 5000, Perkin Elmer, Shelton, CT, USA) after used in lated side. dilution with a 0.5% solution of lanthanum oxide in analyze deionized water 125. Concurrent analysis of Sera of varia Chem controls (Fisher Scientific, Orangeberg, NY, grain (S Laboratory methods manganese USA) yielded values (mean ± standard deviation) 0.05 wa of 9.8 ± 0.2 and2.2 ± O.lmg/dLc0mParedw1tt but < 0. The magnesium, calcium, Ph values (mean(mean ± confidence values) of d composites of diets an d tified ium and magnesium can ce. and zinc content of 6- e experiment 9.5 ± 2.2 and 2.2 ± 0.6 for calc feces were determined throughout the pooled Plasm emission square by inductively j23] coupled (Jarrell-Ash argon Atom 1 Comp 1140, respectively Serum alkaline lipoproteinphosPhata5e ( HDL)total py cholesterol, high-density analysi spectrosco Thermodigestion Elemental, of lyophilized Franklin, blended MA, USA)samples after with wet cholesterol, tTiglycerides, sodium and potassium nitric and perchlOrdeterminedbywere standard determifle methods using the rinary minerals Cobas Fara centrifugal Analyzer (Roche Diagnos- Result USA). The sodium argon tics Systems, SommerSommerville, NJ, Spe- were determined by inductivelypY ofcoupled a diluted ali- and potassium determinations utilized an ion Durint plasma emission spec period quot. For diets and feces, concurrentconcurrent replicate analysis of a standard reference material NIST cific electrode. Very lowdeflSitY lipoprotein chang dietar and1577a Technology,Gaithersburg, bovine liver (National InstituteMD, USA) of yieldedStandards cholesterol (VLDL) were and calculated lowdensitY by lipoproteinusing the triglyCer LDL- of 132 ± 13 for ide values and by difference, respectively. Corn- and bi jØimmunbassay kits were boron means (g/g) ± standard deviation s ercially available worna magnesium, 10827 ± 117 for used to determined serum calcitonin, osteocalcin, calcium , 591 ± 15 for j, 5jhydroxy0le siuin phosphorus, 9.3 ± 0.3 for manganese, and appe 117 ± 3 for zinc compared with certified values 25yroxychOliferOl alecafer0l (means ± confidence values) of 120 ± 7 for cal-mone cac, (incstar arid mid-molecule Corporation, Stillwater, parathyroid MN, USA);hor- Trou N Bi ment cium, 600 ± 159.9 for ± magnesium,0.8 for manganese, thOU ± 400and for 123 8 estradiol (IComedicals, Inc., Carson, CA, Tat phosphorus,for zinc. For urine, concurrent replicate analysis USA);nostic and Products, progesterone Los Angeles, and aldosteroneCA, USA). (Diag- tive Urin of UriChem 1 & 2 (Fisher Scientific, Orangeburg, Urinary potassium content in samples collected decr NY, USA) yielded means (uglmL) ± standard during a 24-h period was determined by using the the 1 deviations respectively of 86 ± 6 and 70 ± 7 for methods described above for calcium and magne- vent calcium, 55 ± 3 and 188 ± 14 for magfleSium sium. Urinary hydroxyproline excretion was mea- aiic€ 622 ± 51 and 541 ± 34 for Phosphorus,and sured but values obtained were extremely vari- 118 0.53 ± 0.04 and 2.63 ± 0.03 for zinc compared with able and no significant treatment effects were nifu certified values of 82 ± 18 and 75 ± 18 for calcium, obtained; thus they are not reported here. 59 ± 7 and 184 ± 44 for magnesium, 650 ± 90 and

200 ALTERATION OF MAGNESIUM, CALCIUM AND PHOSPHORUS METABOLISM

Table 1. Mean magnesium balance during the last 24 d of each 42-d dietary periods

Dietary treatment n Diet, mg/d Feces, Feces, Urine, Urine, Balance, mg/d % intake mgld % intake mg/d Low Mg- Low 13 116 60 52.2 74 63.4 -18 LowMg - HighB 12 119 51 43.3 74 62.1 -6 HighMg - LowB 13 321 177 55.1 141 43.9 +3 High Mg - High B 13 317 170 53.7 137 43.0 + 10 Pooled SD 25 13.7 8 8 21 Analysis of Variance - P Values Magnesium effect 0.0001 0.08 0.0001 0.0001 0.003 Boron effect 0.22 0.15 0.58 0.29 0.11 Magnesium x boron 0.88 0.35 0.33 0.77 0.81 Low Mg = 118 mg/d; high Mg = 318 mg/d; low B = 0.25 mg/d; high B = 3.25 mg/d.

Data Analysis Magnesium deficiency significantly decreased urinary calcium excretion (table II). This did not All balance and biochemical determinations made translate into a significant increase in calcium during the last 24 days in each dietary period were balance during the last 24 days of each dietary used in the statistical analyses. The data were period although the values looked higher when analyzed by two-way repeated measures analysis magnesium was low. However, if the comparison of variance with a SAS general linear model pro- used data from the complete 42-day periods, gram (SAS 8.02, SAS Institute, Cary, NC, USA). P !^ which reduced the effects of variability, magne- 0.05 was considered significant, and P> 0.05 sium deprivation significantly increased calcium but < 0.10 were considered approaching signifi- balance (33 compared to 10 mg/d; P < 0.04). Also, cance. Variances in the data are expressed as a although the diet supplied only about 700 mg pooled standard deviation, calculated as the CaJd, the average calcium balance was not nega- square root of the mean square error from the tive during the 179-d study. Dietary boron did not analysis of variance. significantly affect any measures of calcium metabolism. Results Magnesium affected the manner in which phosphorus was metabolized but did not significantly During the low magnesium - high boron dietary affect balance (table III). The percent of intake period, one woman exhibited heart rhythm and amount of phosphorus excreted in the feces changes that dictated an early entry into a high was decreased by magnesium deficiency. On the dietary magnesium period. As a result, balance other hand, the percent of intake and mg of phos- and biochemical data in the low magnesium - high phorus excreted daily in the urine was higher boron period did not include values from this during magnesium deprivation than magnesium woman. Three weeks after increasing her magne- repletion. Dietary boron did not affect any mea- sium intake the increased ventricular ectopy dis- sures of phosphorus metabolism. appeared. None of the women exhibited positive The dietary treatments did not significantly Trousseaus or Chvosteks signs during the experi- affect either manganese or zinc balance (table f). ment. The balance data in table IV also show that con- Table I shows that magnesium balance was posi- suming about 1 mg Mn/d and about 10 mg ZnJd for tive when the diet provided about 318 mg Mg/d. 179 days did not result in negative balance for Urinary magnesium excretion was significantly either element. decreased with dietary magnesium restriction, but Table V shows that the magnesium deprivation the homeostatic response was not enough to pre- did not markedly affect serum indicators of cal- vent an eventually non-positive magnesium bal- cium and magnesium status. Serum magnesium ance while consuming a diet providing about and calcium were not significantly affected by the 118 mg Mg/2000 kcal. Dietary boron did not sig- dietary treatments. Ionized calcium expressed as nificantly affect magnesium balance. per cent of total was significantly decreased by

201

F.H. NIELSEN, ET AL.

Table V. Efj Table JI. Mean calcium balance during the last 24 d of each 42-d dietary period and potass Balance, Dietary treatment n - Diet, mgfd Feces, Feces, Urine, Urine, mgld % intake mg/d % intake mgld 161 23.0 + 10 Dietary Low Mg - Low B 13 695 524 75.6 treatment 23.2 +25 Low Mg - High B 12 708 518 73.1 165 25.0 ± 4 High Mg - Low B 13 715 532 74.2 180 25.5 +3 Low Mg - Lo High Mg - High B 13 702 520 73.9 180 2.0 63 Low Mg - Hi Pooled SD 70 9.0 15 High Mg - L Analysis of variance - P values 0.0003 0.0003 0.45 High Mg - H Magnesium effect 0.84 0.84 0.82 0.69 0.90 Pooled SD Boron effect 0.88 0.84 0.49 0.98 0.81 Analysis of Magnesium x boron 0.78 0.79 Magnesium Low Mg = 118 mg/d; high Mg = 318 mg/d; low B = 0.25 mg/d; high B = 3.25 mg/U. Boron effec Mg x B effe. Table III. Mean phosphorus balance during the last 24 d of each 42-d dietary period Low Mg - 1 Balance, Dietary Treatment n Diet, mgld Feces, Feces, Urine, Urine, mg/d % intake mg/d mg/d % intake metabolisi 65.4 +31 increased Low Mg - Low B 13 876 272 31.3 574 +41 nary excr Low Mg - High B 12 894 266 30.0 587 65.6 61.7 +42 significarU High Mg - Low B 13 908 305 33.7 561 61.8 4-40 potassium High Mg - High B 13 886 299 33.8 547 2.8 44 tion affect Pooled SD 41 4.4 34 Analysis of variance - P values Table VI 0.0001 0.75 Magnesium effect 0.005 0.01 0.01 no effect 0.69 Boron effect 0.61 0.63 0.96 0.85 parathyro 0.47 Magnesium x boron 0.84 0.40 0.12 0.96 alkaline resulted i mg/d. Low Mg = 118 mg/d; high Mg =318 mg/U; low B = 0.25 mg/d; high B = 3.25 serum 25-

Table [V. Manganese and zinc mean balances during the last 24 d of each 42-ct dietary period Table VI. 1 to bone tu Dietary Treatment n Manganese, mgld Zinc, mg/d Diet Feces Balance Diet Feces Urine Balance Dietary treatment 0.26 Low Mg - Low B 13 0.96 0.97 -0.01 9.69 9.13 0.30 9.85 8.69 0.31 0.85 Low Mg - High B 12 0.97 0.94 +0.03 Low Mg - L 9.86 8.92 0.30 0.64 High Mg - Low B 13 1.00 1.00 0.00 Low Mg -11 +0.01 977 8.76 0.31 0.70 High Mg - High B 13 0.96 0.95 High Mg - I 0.12 0.11 1.27 0.03 1.24 Pooled SD High Mg - I I Analysis of variance - P values Pooled SD 0.88 0.85 0.31 0.72 Magnesium effect 0.55 Analysis 0 0.43 0.50 0.78 0.47 Boron effect 0.39 Magnesiun 0.87 0.53 0.73 0.81 0.50 Magnesium x boron Boron effe Low Mg = 118 mg/U; high Mg = 318 mg/d; low B = 0.25 mg/d; high B = 3.25 mg/d Magriesiun Low Mg magnesium deprivation, but when expressed as tended (P = 0.06) to decrease serum ionized cal- Abbreviati nig/dL, the decrease only approached statistical cium. Table Vindicates that the dietary treatments 1,25-dihyd I. significance (P = 0.08); boron deprivation also had a mild effect on sodium and potassium Units werE

202

ALTERATION OF MAGNESIUM, CALCIUM AND PHOSPHORUS METABOLISM

Table V Effects of dietary boron and magnesium on indicators (means) of magnesium, calcium, sodium and potassium status tee, a Serum Urinary -. Potassium, Dietary a Magnesium, Calcium, Ionized Ionized Sodium, Potassium, g124 hr treatment mg/dL mg/dL Calcium, Calcium, mg/dL mg/dL mg/dL %total LowMg - LowB 13 1.99 9.30 4.97 53.5 148.6 4.48 1.24 Low Mg - High B 12 1.99 9.30 5.01 53.9 148.4 .4.45 1.11 High Mg - Low B 13 2.01 9.27 4.99 53.9 148.0 4.44 1.52 High Mg - High 13 13 2.02 9.28 5.04 55.3 147.8 4.52 1.32 Pooled SD 0.05 0.11 0.08 0.9 1.1 0.14 0.29 Analysis of variance - P values Magnesium effect 0.10 0.31 0.08 0.02 0.05 0.51 0.002 Boron effect 0.66 0.68 0.06 0.13 0.57 0.52 0.05 Mg x B effect 0.61 0.69 0.85 0.96 0.81 0.11 0.98 Low Mg 118 mg/d; high Mg :- 318 mg/d; low B = 0.25 mg/d; high B = 3.25 ing!d. Ice,

metabolism. Magnesium deprivation significantly tended to decrease (P = 0.06) 1,25-dihydro- increased serum sodium and decreased the un- xycholecalciferol. nary excretion of potassium. Boron deprivation significantly increased the urinary excretion of Magnesium deprivation decreased serum total potassium. Neither boron nor magnesium depriva- cholesterol concentration with the decrease tion affected serum potassium concentration, occurring mainly in the LDL-cholesterol fraction 7ble VI shows that the dietary treatments had (table VII). Boron deprivation did not affect serum no effect on the serum calcitonin, mid-molecule cholesterol or its fractions. Serum 173-estradio1 parathyroid hormone (MM-PTH), osteocalcin and was slightly, but significantly higher during boron alkaline phosphalase. Magnesium deprivation deprivation than boron supplementation. Also, resulted in a small but significant increase in when dietary magnesium was low, boron supple- serum 25-hydroxycholecaJcjferol (25-OH-D.), and rnental,ion decreased serum progesterone. Nei-

Table VI. Effects of dietary magnesium and boron on mean serum indices of calcium metabolism related to bone turnover flee Dietary n Calcitonin, MM-PTH, 25-OH-D3, 1, 25-011- Alkaline Osteocalcin, treatment pg/mL pmol/L ng/mL D3, pgfmL phosphatase, ng/mL U/L Low Mg - Low B 13 60 41.8 24.8 23.7 82 4.46 Low Mg - High B 12 60 39.4 29.0 24.9 78 4.04 High Mg - Low B 13 62 45.3 23.3 27.7 85 4.08 High Mg- High B 13 65 43.1 22.7 24.6 82 4.08 Pooled SD 10 12.9 5.7 4.1 6 0.92 Analysis of variance - P values Magnesium effect 0.22 0.61 0.01 0.06 0.13 0.61 Boron effect 0.66 0.82 0.32 0.21 0.10 0.35 Magnesium x boron 0.53 0.79 0.07 0.22 0.70 0.11 cal- Low Mg = 118 mg/d; high Mg 318 mg/d; low B = 0.25 mg/d; high B 3.25 mg/d. Abbreviations: MM-PTI1 - mid-molecule parathyroid hormone; 25 OH-D D 3-25••hydroxycholecalciferol; 1,25-OH-D3 - ents I ,25-dihydroxycholecalciferol. iuIn Units were imoles of p-nitrophenyl phosphate hydrolyzed/mm.

203

F.H. NIELSEN, ET AL.

Although U Table VII. Effects of dietary magnesium and boron on mean serum sterol concentrations negative mat cantly decre Cholesterol, mg/dL in serum. Thi HDL VLDL 17-EstradiOl, Progesterone, Aldosterone, ment that se) Dietary treatment n Total LDL pg/mL ng/mL ngldL useful to ass as that appa 0.25 14.27 LowMg - LowB 13 248 163 59 26 20.5 This judgmei 0.19 12.48 12 248 165 58 27 18.4 body is able LowMg - HighB 0.24 12.75 HighMg - LowB 13 256 171 59 26 19.8 concentratio 14.6 0.24 12.75 High Mg - High B 13 255 171 58 26 deprivation i 5.1 0.06 3.37 does this by Pooled SD 8 10 4 3 magnesium Analysis of variance P values 0.09 0.68 0.47 0.09 0.27 0.54 ellular skelel Magnesium effect 0.02 0.31 0.60 0.41 0.53 0.37 0.03 0.22 lular fluid 13 Boron effect 0.23 0.32 0.45 0.41 0.94 0.53 0.05 One likely Magnesium x boron obtained in mg/d; high B = 3.25 mg/d. Low Mg = 118 mg/d; high Mg = 318 mg/d; low B = 0.25 mia. One wo l/d at baseli ther dietary boron nor magnesium affected serum Although the experimental design may have after 72 day aldostcrofle concentration. subdued the response to magnesium deprivation, about 118 r the balance data indicate that a dietary magne- induced by sium intake of 118 mg/d is not adequate for post- consuming Discussion menopausal women because this intake for 6 to three weeks 12 weeks resulted in a negative balance. On the to baseline. As noted in the Materials and Methods section, the other hand, an intake of 318 mgid is apparaently with repor, Latin square design of the experiment resulted in adequate because it resulted in a slightly positive induce vent different lengths of time of magnesium depriva- magnesium balance. Thus the biochemical and ectopy in f tion, which probably influenced the magnitude of functional changes found in the women fed consuming. the response to changes in dietary magnesium. 118 mg/d for at least six weeks probably can be the previou The order of treatment also probably influenced considered signs of a primary magnesium defi- rnent, the si the magnitude of the changes seen because ciency induced by a diet whose composition Trousseaus another experiment with post menopausal women would not be considered atypical. Based upon none were of a cross-over design showed that feeding a urinary excretion, about 43% of the 318 mg/d diac nluscl( magnesium-adequate diet for 81 days before feed- intake of magnesium was absorbed; this is consis- sium depri ing a magnesium-deficient diet (99 mg12 000 kcal), tent with reports indicating that approximately The chan in contrast to feeding the magnesium-deficient 30% to 50% of the 200 to 350 mg of magnesium considered diet first, depressed the magnesium balance, provided by the average American diet daily is women can serum magnesium, and serum cholesterol absorbed [29, 301. When the magnesium intake providing responses to magnesium deprivation 1261. The was only 118 mg/d, the percent of intake absorbed decreased order and length of time consuming a boron-low increased (based on urinary excretion) to about. increased diet also probably contributed to the lack of 63%, a 50% increase above that when dietary mag- desirable f response of calcium variables to boron depriva- nesium was adequate. The apparent difference in susceptiblE tion. In a previous study, boron deprivation for absorption is consistent with the finding that probably sI 119 days resulted in changes in magnesium and intestinal magnesium absorption is inversely pro- are. Decre calcium apparent absorption, retention and excre- I portional to the amount ingested 11]. This inverse retention C tion 1201 but in a subsequent study, boron depriva- relationship apparently occurs because magne- ciency in a tion of 42 days apparently did not [27, 281. In with incre: human experiments in which boron deprivation sium absorption has two components; one that can be saturated (active) and another that does especially lasted 63 days, it took at least four weeks after a nephrocal( change from high to low dietary boron before not (passive) [31]. The active transport compo- nent would result in a lower percentage of ciency cxi: I . boron responsive variables either stabilized or calcium is began to change [27, 281. Clear changes in calcium ingested magnesium being absorbed as the com- ponent approached or reached the saturation calcium ni status indicators took about. 45 to 50 days of con- cial for bc suming a diet providing about 0.25 mg Bid. point.

204 ALTERATION OF MAGNESIUM, CALCIUM AND PHOSPHORUS METABOLISM

Although the magnesium deprivation induced a lism findings in the present study, however, con- negative magnesium balance, it did not signifi- trast with those found in severely magnesium- cantly decrease the concentration of magnesium deficient subjects. They exhibit hypocalcemia [40] in serum. This finding is consistent with the judg- and increased urinary calcium excretion [41, 421, ment that serum magnesium concentration is not apparently caused by decreased secretion of, and useful to assess mild magnesium depletion, such end-organ resistance to, parathyroid hormone 11, as that apparently achieved in the present study. 40] which does not occur early when magnesium This judgment is based on the knowledge that the intake is severely limited or with mild magnesium body is able to maintain extracellular magnesium deprivation. The magnesium deficiency in the concentrations within narrow limits if magnesium present experiment apparently was only mild to deprivation is not severe or prolonged. The body moderate because it did not decrease serum par- does this by adjusting the urinary excretion of athyroid hormone, nor did it.induce hypomag- magnesium and by redistributing bone and intrac- nesmia and hypocalcemia. ellular skeletal muscle magnesium to the extracel- The changes in phosphorus metabolim in the lular fluid [32]. present study can not be compared to changes One likely sign of dietary magnesium deficiency that occur in magnesium deficiency studies with obtained in the present study was heart arrhyth- rats because rats apparently respond differently mia. One woman showed a marked increase (from front dogs and humans [43]. The changes occur- lid at baseline to 1341d) in ventricular discharges ring with the reduced dietary magnesium in the after 72 days of consuming the diet providing present study were consistent with those found in about 118 mg Mg/d. This increase probably was other studies of human magnesium deficiency. A induced by the low dietary magnesium because substantial decrease in phosphorus absorption consuming the diet providing 318 mg Mgid for occurred when dietary magnesium was increased three weeks decreased the ventricular discharges [42, 441, but phosphorus balance apparently was to baseline. Moreover, the literature is replete not affected [44]. The apparent increase in phos- with reports that magnesium deficiency can phorus absorption indicated by decreased fecal induce ventricular ectopy, including one showing phosphorus and increased urinary phosphorus ectopy in post menopausal women induced by when the dietary magnesium was 118 mg/d might consuming about 100 mg Mgid for 81 days [33). In be reflecting changes in high-energy phosphate the previous study [33] and the present experi- metabolism. This suggestion is supported by ment, the subjects were tested for Chvosteks and reports that magnesium deficiency increased Trousseaus signs during magnesium deprivation; energy needs of postmenopausal women [45] and none were found. This finding suggests that car- magnesium supplementation of athletes improved diac muscle is more sensitive to dietary magne- energy utilization [16]. sium deprivation than skeletal muscle. Much evidence exists showing that magnesium The changes in calcium metabolism also can be has a regulatory role in electrolyte transport, cel- considered evidence that post menopausal lular distribution, and intracellular concentration, women can become magnesium-deficient on diets especially for potassium [46]. Magnesium defi- providing only 118 mg/d. The findings of ciency inhibits the response of potassium- decreased urinary excretion of calcium and deficient individuals to potassium supplementa- increased calcium balance could be considered tion [47]. Magnesium activates the Na, K-ATPase desirable for pest menopausal women who are pump that has a major role in regulating Na and susceptible to bone loss and osteoporosis, but one K transport 46). Thus, the finding of decreased probably should be cautious in assuming that they urinary potassium and increased serum sodium are. Decreased urinary excretion and increased with a low dietary intake of magnesium in the retention of calcium are signs of magnesium defi- present study might be a further indication that ciency in animals and they have been associated the women were becoming magnesium deficient / I with increased soft tissue calcium accumulation, on an intake of 118 mg/d. However, it should be especially in the kidney [34-38). Case reports of noted that in humans made severely magnesium- nephrocalcinosis as the result of magnesium defi- deficient (6 to 10 mg/d), the urinary excretion of ciency exist [39]. The decrease in serum ionized potassium increased [41). The increase is believed calcium is further evidence that the changes in to occur through a stimulation of aldosterone calcium metabolism are not particularly benefi- secretion that increases urinary potassium excre- cial for bone maintenance. The calcium metabo- tion [48]. Hereto, perhaps the severity of the mag-.

205 F.H. NIELSEN, ET AL. nesiuni deficiency was the reason for the contrast- be of limited usefulness as indicators of mild mag- ing results. Support for this suggestion is the nesium deprivation. finding that serum aldosterone concentrations Carpenter [52] discussed conflicting findings on were not significantly affected by the dietary the effect of magnesium deficiency on circulating treatments in the present experiment. Perhaps the 25-hydroxy-vitamin D concentration. He noted that mild magnesium deficiency mostly affected vari- several studies indicating that circulating ables first to respond to low dietary magnesium, 25-hydroxyvitamin D was decreased in patients such as a reduction in the magnesium-blockade of with hypomagnesmia showed concomitant malab- the N-methyl-D-aspartate receptor [491 and/or a sorption or low intakes of vitamin D. In change in the regulation of the NaJCa exchange magnesium-deficient individuals receiving vita- [50], which results in an increased intracellular min D prior to magnesium supplementation, nor- calcium concentration. Increased intracellular mal or elevated circulating 25-hydroxyvitamin D calcium may increase the potassium transport concentrations were found [53]. Carpenter [52] into cells, which would result in less potassium concluded that in magnesium-deficient humans available for excretion in the urine. the absorption of vitamin D and its conversion to Magnesium has long been considered a risk fac- 25-hydroxyvitamin D is intact, but that target tis- tor for osteoporosis. Rude [51] reviewed the evi- sues are refractory to active forms of vitamin D. dence from animal studies, which show that mag- This refractoriness possibly results in a feedback nesium deficiency reduces serum osteocalcin and to absorb more vitamin D, which may be the alkaline phosphatase, and from epidemiological explanation for the small but significant increase studies which link dietary magnesium intake to in circulating 25-hydroxyvitamin D 3 found when bone mass in adults. Rude [51] pointed out that the the dietary intake of magnesium was low in the mechanism/s for a magnesium deficiency-induced present experiment. Carpenter [52] also suggested osteoporosis is unknown but suggested that it that in magnesium-deficient humans the conver- may involve impaired parathyroid hormone and sion of 25-hydroxyvitamin D to 1,25- 1,25-dihydroxyvitamin D 3 production. The present dihydroxyvitamin D may be impaired and noted study, however, found that magnesium depriva- findings indicating that some other unidentified tion did not affect the circulating parathyroid hor- factor(s) may have a role in this impairment. It is mone concentration, and the decrease in circulat- possible that boron may be such a factor, as there ing 1,25-dihydroxyvitamin D3 only approached is evidence that the element can enhance the significance (P < 0.06). It is possible that the action of vitamin D [54]. effects of magnesium deprivation observed in ani- The finding that magnesium deficiency exacer- mals (e.g., reduced serum osteocalcin and alka- bated the increased serum cholesterol and athero- line phosphatase) did not occur in these subjects genesis in rabbits fed a high cholesterol diet [55] because the degree of magnesium deficiency was stimulated the further study of the effect of mag- not severe enough to induce a change in circulat- nesium deficiency on lipid metabolism in other ing parathyroid hormone. As indicated by others species including humans. However, in order to [40], the severity or length of time of magnesium show that magnesium deficiency affects serum deprivation determined the effect on circulating cholesterol in rats, experiments had to be long- parathyroid hormone. Fatemi [40] found that 20 of term, use pair-feeding techniques with severe defi- 26 normal human subjects fed a severely ciency, or required high dietary cholesterol [56]. magnesium-deficient diet (< 12 mg/day) for three No convincing reports have indicated increased weeks had decreases in serum parathyroid hor- serum total cholesterol as a primary sign of mag- mone concentration, but it was elevated in the nesium deficiency in humans. Recent studies sug- remaining six subjects. These authors suggested gest that magnesium deficiency can promote that such heterogeneous results may be expected atherogenesis through mechanisms other (e.g., because, as hypomagnesmia develops, the parath- LDL-cholesterol oxidation, formation of pro- ryoid gland will react initially by increasing hor- inflammatory agents) than by increasing circulat- mone secretion, and, as intracellular magnesium ing cholesterol [57, 581. The present findings indi- depletion develops, its ability to secrete parathy- cate that a relatively moderate magnesium roid hormone is impaired. The inconsistency of deprivation slightly but significantly decreased the changes with magnesium deprivation reported serum cholesterol. This may not necessarily be by others and the findings here indicate that beneficial, as it could mean a change in membrane changes in circulating bone turnover indices may lipid composition. Lipid components of the eryth- 206 ALTERATION OF MAGNESIUM, CALCIUM AND PHOSPHORUS METABOLISM

rocyte membrane exchange rapidly with plasma the same subjects [21], results in increased brain lipoproteins, and cellular membrane composition cortical excitability as measured by an electroen- and fluidity is altered by magnesium deficiency in cephalogram. The present experiment indicates rats [51]. The magnesium deprivation in the that short-term balance studies do not accurately present study did not affect the concentration of reflect magnesium requirements because homeo- serum triglycerides (VLDL; table VII). This was not static mechanisms resulting in decreased urinary surprising because magnesium-deficient adult excretion and tissue redistribution apparently are rats are not very susceptible to developing an able to maintain balance for a short duration. altered lipid metabolism. The finding that magne- Thus, reports based on periods lasting from 5 to sium deprivation induces hyperlipemia comes 15 days should not be used to determine magne- from experiments with young growing rats fed a sium requirements. Finally, because no evidence severely magnesium-deficient diet [56]. was found showing that magnesium deprivation Serum progesterone was the only variable facilitates a response to low dietary boron, examined that was affected by a significant inter- reports that magnesium status affects the action between magnesium and boron. Consider- response to boron probably reflect an indirect ing the large number variables examined, this may relationship between the two elements. have been an statistical anomaly. Further experi- mentation is needed to confirm that adequate compared to low dietary boron decreases serum Acknowledgement progesterone when dietary magnesium is low, but not when dietary magnesium is adequate. The author wishes to thank the members of the The present study also yielded results relevant Grand Forks Human Nutrition Research Center to considerations of mineral requirements for post clinical staff whose special talents and skills made menopausal women. The finding that mean cal- this study possible: Leslie Klevay (medical super- cium balance did not become negative during the vision), James Penland (psychological supervi- 179-day study suggests that 700 mg of calcium sion), Henry Lukaski (exercise physiology), daily is adequate for Caucasian post menopausal Loanne Mullen and staff (dietary), Sandra Gal- women. The balance data are consistent, with lagher and the analytical biochemistry staff (clini- those found with elderly Japanese people [59] arid cal chemistry), Betty Vetter, Donna Neese and the with the recommended intake by some expert nursing staff (metabolic unit care), and Roger groups including the FAO/WHO Expert Consulta- Sims and analytical chemistry staff (mineral tion on Human and Vitamin and Mineral Require- analysis). The author also thank LuAnn Johnson ments [60). The positive balance for zinc when the for help with the experimental design and for the dietary intake was about 9.8 mg/day supports the statistical analysis of the data, and Christine recently established zinc RDA for women older Bogenreif for manuscript processing. than 31 years of 8 mg/day by the United States Institute of Medicine Food and Nutrition Board References [61]. This Board also defined an adequate intake for manganese of 1.8 mg/day for women older than 31 years [62]. The present study indicates 1. Rude RK. Magnesium deficiency in parathyroid function. In: JP Bilezikian, R Marcus, MA Levine, that this intake is more than adequate because an eds. The parathyroids, second edition. San Diego: intake of 1.0 mg Mnlday for 179 days did not result Academic Press, 2001, PP. 763-77. in a negative manganese balance. 2. FAO/WHO Expert Consultation. Magnesium. In: Human vitamin and mineral requirements. Geneva: Food and Agriculture Organization/World Conclusion Health Organization, 2002, pp. 223-33. The present experiment shows that a magnesium 3. Durlach J. Magnesium in clinical practice. London: John Libbey Eurotext, 1988. intake of about 118 mg/day is not adequate for healthy post menopausal women because it 4. Altura BM, Altura BT. Magnesium metabolism, induces changes in calcium, phosphorus, and cho- atherogenesis and cardiovascular pathobiology. In: H. Smetana, ed. Advances in magnesium research: lesterol metabolism which are likely to he con- magnesium in cardiology: proceedings of the 5" strued as undesirable. This intake may also lead to European Congress on Magnesium. London: John heart arrhythmia and, as reported previously for Libbey, 1998, pp. 28.37. 207 F.H. NIELSEN, ET AL.

5. Seelig M. Epidemiologic data on magnesium and humans. In: HE Goldbach, B Rerkasem, MA deficiency-associated cardiovascular disease and Wimmer, PH Brown, M Thellier, RW Bell, eds. osteoporosis: consideration of risks of current rec- Boron in plant and animal nutrition. New York: ommendations for high calcium intakes. In: Y Ray- Kluwer Academic/Plenum Publishers, 2002, pp. ssiguier, A Mazur, J Durlach, eds. Advances in mag- 37-49. nesium research: nutrition and health. Eastleigh: 20. Hunt CD, Herbel JL, Nielsen FF1. Metabolic John Libbey, 2001, 177-90. PP. responses of postmenopausal women to supple- 6. Johnson M. The multifaceted and widespread mental dietary boron and aluminum during usual pathology of magnesium deficiency. Medical and low magnesium intake: boron, calcium, and IIot.hesis 2001; 56: 163-70. magnesium absorption and retention and blood mineral concentrations. Am .1 Gun NuIr 1997; 65: 7. Food and Nutrition Board, Institute of Medicine. Magnesium. In: Dietary reference intakes for cal- 803-13. cium, phosphorus, magnesium., vitamin D and fluo- 21. Penland JG. Quantitative analysis of EEG effects ride. Washington, D.C.: National Academy Press, following experimental marginal magnesium and 1997, pp. 190-249. boron deprivation. Magnesium Res 1995; 8:341-58. 8. Lakshmanan LF, Rao RB, Kim WW, Kelsay JL. Mag- 22. Harris IA, Benedict FG. A biometric study of basal nesium intakes, balances, and blood levels of metabolism in men. Carnegie Publication no. 279. adults consuming self-selected diets. Am J GUn Philiadelphia: Lippincott .JB, 1919. 1984; 40: 1380-9. NuIr 23. Sims EL, Mullen LM, Mime DB. Application of 9. Apports nutritionnels conseillés pour la population inductively coupled plasma emission spectroscopy Francaise. A. Martin, coordonnateur. Paris: Tee. to multielement analysis of foodstuffs used in Doc., Lavoisier, 2000. metabolic studies. JFood Camp Anal 1990; 3:27-37. 10. Scientific Committee for Foods. Nutrient and 24. Analytical Methods Committee. Methods of energy intakes for the European Community. destruction of organic matter. Analyst 1960; 85: Report of the Scientific Committee for Food, thirty- 643-56. first series, Brussels: European Commission, 1993. 25. Wills MR, Sunderman FW, Savory J. Methods for 11. Marshall DII, Nordin BEC, Speed R. Calcium, phos- the estimation of serum magnesium in clinical phorus and magnesium requirement. Proc NutrSoc laboratories. Magnesium. 1986; 5: 317-27. 1976; 35: 163-73. 26. Nielsen PH, Mime DB. Some magnesium status 12. Andon MB, flich JZ, Tzagournis MA, Matkovic V. indicators and oxidative metabolism responses to Magnesium balance in adolescent females consum- low-dietary magnesium are affected by dietary ing a low- or high-calcium diet. Am J Gun iWar copper in postmenopausal women. Nutrition 2003; 1996; 63: 950-3. 19: 617-26. 13. Vormann J, Anke M. Dietary magnesium: supply, 27 Nielsen FH, Mullen LM, Gallagher SK. Effect of requirements and recommendations - results from boron depletion and repletion on blood indicators duplicate and balance studies in man. J Clin Basic of calcium status in humans fed a magnesium-low Cardiol 2002; 5: 49-53. diet. J 71-ace Elem Ex-p Med 1990; 3: 45-54. 14. Durlach J. Recommended dietary amounts of mag- 28 Nielsen PH, Gallagher SK, Johnson LK, Nielsen EJ. nesium: Mg RDA. Magnesium Res 1989; 2: 195-203. Boron enhances and mimics some effects of estrogen therapy in postmenopausal women. J 71-ace 15. Durlach J, Bac P. Mechanisms of action on the Stem Exp Med 1992; 5:237-46. nervous system in magnesium deficiency and dementia. In: M Yasui, MJ Strong, K Ota, MA Verity, 29. Rude RK. Physiology of magnesium metabolism eds. Mineral and metal neurotoxicology. Boca and the important role of magnesium in potassium deficiency. Raton: CRC Press, 1997, pp. 201-10. AmJCardiot 1989; 63: 31G-4G. 16. Golf SW. Biochemistry of magnesium in man. In: 30. Sabatier M, Keyes WR, Pont F, Arnaud MJ, Turn- S Golf, D Dralle, L Vecchiet, eds. Magnesium 1992. land JR. Comparison of stable-isotope-tracer meth- Vecchiet. London: John Libbey & Company, 1994, ods for the determination of magnesium absorp- pp. 31-41. tion in humans. AmJGlinNutr2003; 77:1206-12. 17. Altura BM, Altura BT. Cardiovascular risk factors 31. Hardwick LL, Jones ME, Brauthar N, Lee DBN. and magnesium: relationships to atherosclerosis, Magnesium absorption: mechanisms and the influ- ischernic heart disease and hypertension. Magne- ence of vitamin D, calcium and phosphate. J Nutr sium lraceJJlem 1919-1992; 10: 182-92. 1991; 121: 13-23. 18. Nielsen FF1. Evidence for the nutritional essential- 32. Wallach S. Magnesium exchangeability and bio- ity of boron. .1 Prace Eleni Erp Med 1996; 9: 215-29. availability in magnesium deficiency. In: BM Altura, J Durlach, MS Seelig, eds. Magnesium in 19. Nielsen FF1. The nutritional importance and phar cellular processes and medicine. Basel: Karger, rnacological potential of boron for higher animals 1987, pp. 27-49. 208 ALTERATION OF MAGNESIUM, CALCIUM AND PHOSPHORUS METABOLISM

33. Klevay LM, Mime DB Low dietary magnesium 47. Schroll A. Importance of magnesium for the elec- increases supraventricular ectopy. Am J (JUn Nutr trolyte homeostasis - an overview. In: R Smetana, 2002; 75: 550-4. ed. Advances in magnesium research: magnesium in cardiology: proceedings 34. Berthelot A, Esposito J. Effects of dietary magne- of the 515 European Con sium on the development of hypertension in the gress on Magnesium. London: John Libbey, Com- spontaneously hypertensive rat. J Am Coll Nutr pany, 1998, pp. 463-72. 1983; 4: 343-53. 48. Ryan MP, Whang R. Interrelationships between 35. McElroy ST, Link JE, Dowdy RP, Zinn KR, Eller- potassium and magnesium. In: H Whang, JR sieck MR. Influence of age and magnesium on cal- Aikawa, eds. Potassium: its biologic significance. cium metabolism in rats. JNutr 1991; 121: 492-7. Boca Raton: CRC Press, 1983, pp. 97-107. 36. Bergstra AE, Lemmens AG, Beynen AC. Dietary 49. McIntosh TK. Novel pharmacologic therapies in fructose vs. glucose stimulates nephrocalcinogen- the treatment of experimental traumatic brain esis in female rats. JNutr 1993; 123: 1320-7. injury: a review. JNeurotrauma 1993; 10: 215-61. 37. Planells E, Sanchez-Morito N, Montellario MA, 50. Reynolds IJ. Intracellular calcium and magnesium: Aranda A, Llopis J. Effect of magnesium deficiency critical determination of excitotoxicity? Prog on enterocyte Ca, Fe, Cu, Zn, Mn and Se content. J Brain Res 1998; 116:225-43. Physiol Biochem 2000; 56: 217-22. 51. 38. Uehara M, Chiba H, Fujii A, Matsuzaki H, Rude RK. Magnesium deficiency: a possible risk Masuyama R, Suzuki K. Induction of phospholipid factor for osteoporosis. In: Y Rassiguier, A Mazur, J hydroperoxides in relation to change in tissue min- Durlach, eds. Advances in magnesium research: eral distribution caused by magnesium-deficiency nutrition and health. Eastleigh: John Libbey, Com- pany, 2001, in rats. In: Y Rayssiguier, A Mazur, J Durlach, eds. pp. 391-8. Advances in magnesium research: nutrition and 52. Carpenter TO. Disturbances of vitamin D metabo- health. Eastleigh: John Libbey & Company, 2001, lism and action during clinical and experimental pp. 291-6. magnesium deficiency. Magnesium Res 1988; 1: 39. Rapado A, Castrillo ,JM. Nephrocalcinosis and 131-9. hypomagnesemia. In: M Cantin, MS Seelig, eds. 53. Magnesium and health and disease. Medalle R, Waterhouse C, Hahn TJ. Vitamin D resis- New York: SP tance in magnesium deficiency. Am. J Gun Nutr Medical & Scientific Books, 1980, pp. 485-97. 1976; 29: 854-8. 40. Fatemi S, Ryzen E, Flores J, Endres DB, Rude RK. Effect of experimental human magnesium deple- 54. Hunt CD, Barbel JL, Idso JP. Dietary boron modi- tion on parathyroid hormone secretion and 1,25- fies the effects of vitamin D3 nutrition on indices of dihydroxyvitamin D metabolism. J Clin Endocri not energy substrate utilization and mineral metabo- Metab 1991; 73: 1067-72. lism in the chick. J Bane Min Res 1994; 9: 171-82. 41. Shils ME. Experimental human magnesium deple- 55. Altura BT, Brust M, Bloom 5, Barbour RL, Stempak tion. Medicine 1969; 48: 61-85. JG, Altura BM. Magnesium dietary intake modu- lates blood lipid levels and atherogenesis. Proc 42. Fine KD, Santa Ma CA, Porter JL, Fordtran JS. NaIl AcadSvi USA 1990; 87: 1840-4. Intestinal absorption of magnesium from food and supplements. J GUn. Invest 1991; 88: 396-402. 56. Rayssiguier Y. Magnesium and lipid metabolism. 43. Massry SG, Brauthar N. The interrelationship In: H Sigel, A Sigel, ads. Metal ions in biological between phosphate and magnesium metabolism. systems, vol. 26. Compendium on magnesium and In: SG Massry, E Ritz, H Jahn, eds. Phosphate and its role in biology, nutrition, and physiology. New minerals in healthhealth and disease. New York: Plenum York: Marcel Dekker, 1990, pp. 341-58. Press, 1980, pp. 51-66. 57. Altura BM, Altura BT. Magnesium and altherogen- 44. Briscoe AM, Ragan C. Effect of magnesium on esis. In: T Theophanides, J Anastassopoulou, eds. calcium metabolism in man. Am J GUn Nutr 1966; Magnesium: current status and new developments. 19: 296-306. Dordrecht: Kluwer Academic Publishers 1997, pp. 383-95. 45. Lukaski HC, Nielsen FIl. Dietary magnesium depletion affects metabolic responses during submaxi- 58. Sherer Y, Bitzur R, Cohen 11, Shaish A, Varon D, mal exercise in postmenopausal women. J iVutr Shoenfeld Y, Harats D. Mechanisms of action of the 2002; 132: 930-5. anti-atherogenic effect of magnesium: lessons from a mouse model. Magnesium Res 2001; 14: 46. Altura BM, Altura BT. Magnesium and the cardio- 173-9. vascular system: experimental and clinical aspects updated. In: H Sigel, A Sigel, eds. Metal -ions in 59. Uenishi K, Ishida 11, Kamei A, Shiraki M, Ezawa 1, biological systems, vol 26. Compendium on magne- Goto 5, Fukuoka H, l-losoi T, Orimo H. Calcium sium and its rote in biology, nutrition, and physiol- requirement estimated by balance study in elderly ogy. New York: Marcel Dekker, 1990, pp. 359-416. Japanese people. Osteporosis mt 2001; 12: 858-63. 209 F.H. NIELSEN, ET AL. Magnesium

60. FAO/WHO Expert Consultation. Calcium. In: iodine, iron, manganese, molybdenum, nickel, sili- Human vitamin and mineral requirement. Geneva: con, vanadium, and zinc. Washington, D.C.: Food and Agriculture Organization/World Health National Academy Press, 2001, 442-501. Organization, 2002, pp. 151-79. pp. 62 Food and Nutrition Board, Institute of Medicine Manganese. In: Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, cop- 61. Food and Nutrition Board, Institute of Medicine. per, iodine, iron, manganese, molybdenum, nickel, Zinc. In: Dietary reference intakes for vitamin A, silicon, vanadium, and zinc. Washington, D.C.: vitamin K, arsenic, boron, chromium, coyypei National Academy Press, 2001, pp. 394-419.

210