Content uploaded by Forrest H Nielsen
Author content
All content in this area was uploaded by Forrest H Nielsen on Feb 17, 2015
Content may be subject to copyright.
Am J Clii, Nicir 1997:65:803-13. Printed in USA. © 1997 American Society for Clinical Nutrition
803
Metabolic responses of postmenopausal women to
supplemental dietary boron and aluminum during usual and
low magnesium intake: boron, calcium, and magnesium
absorption and retention and blood mineral
concentrations1 4
Curtiss D Hunt, Jo Lavne Herbel, and Forrest H Nielsen
ABSTRACT Findings from animal studies indicate that di-
etary boron affects several aspects of mineral metabolism, espe-
cially when animals are subjected to nutritional stressors. Eleven
postmenopausal volunteers living on a metabolic wand for I 67 d
(one 23-d equilibration period and six 24-d treatment periods)
were fed a conventional basal diet that supplied a daily average
intake of0.36 mg B, 109 mg Mg, and < 0. 10 mg Al/8400 kJ. They
were given supplements of 0 (BB) or 3 mg B (SB, last two periods
only), 0 (BMg) or 200 mg Mg (SMg) (with magnesium supple-
ments held constant during the last two periods), or 0 (BA1) or
1000 mg Al (SAI)/d. The SB treatment, compared with the BB
treatment, provided a 9.0-fold increase in dietary boron but yielded
only a I .5-fold increase in plasma boron concentrations. Regard-
less of boron dietary treatment, fecal plus urinary excretion of
boron accounted for nearly 100% of dietary boron intake with no
evidence of boron accumulation over time. Lack of boron accu-
mulation and relatively small changes in blood boron values dur-
ing a substantial increase in dietary boron support the concept of
boron homeostasis. In subjects fed BMg, SB decreased the per-
centage of dietary calcium lost in the urine but increased that
percentage in volunteers fed SMg, a relation that may be important
in understanding metabolic mineral disorders that perturb calcium
balance. Reduced calcium absorption during SAl suggests that
aluminum supplementation should be limited or at least monitored
in postmenopausal women prone to excessive calcium loss. De-
creased total urinary oxalate during SB in BMg subjects indicates
a possible role for boron in the control of unolithiasis during
low-magnesium nutriture. Am J C/in Nutr 1997:65:803-13.
KEY WORDS Boron, aluminum, calcium, magnesium,
postmenopause, humans, blood pressure, boron absorption,
boron metabolism, urinary boron, blood urea nitrogen, oxalate,
plasma boron, red blood cell boron
INTRODUCTION
In I 98 1, Hunt and Nielsen ( 1) reported that vitamin D3
(cholecalciferol)-deficient chicks responded to boron supple-
mentation with improved growth and reduction in abnormally
elevated plasma alkaline phosphatase activity. The findings
suggested that boron positively affected calcium metabolism
and prompted further investigation into the influence of dietary
boron on bone and mineral metabolism. A subsequent group of
studies showed that boron affected tibial growth plate morphol-
ogy in these chicks (2) and bone magnesium concentrations
in rats (3). The effects were more pronounced when the diets
were manipulated to cause nutritional stress. Those findings
prompted the first known investigation of the effects of dietary
boron in women with a typical physiologic stressor (low cm-
culating estrogen, postmenopause) compounded by concurrent
nutritional stressors (low dietary magnesium and high dietary
aluminum) for calcium metabolism.
Initial findings from the investigation with postmenopausal
women were reported in 1987 (4). The findings indicated that
urinary calcium and magnesium excretion were decreased in
the women when their low-boron diet (0.36 mg B/d) was
supplemented with boron (3 mg Bid). The decrease seemed
more marked when dietary magnesium was low.
Complete boron, calcium, and magnesium balance data from
the investigation are reported here with the successful devel-
opment and validation of an analytical method for concurrent
boron analysis (5). Urinary calcium and magnesium excretion
data reported earlier are corrected here for the influence of
fluctuating energy intake over the 6-mo course of the study.
Thus, this report provides an in-depth analysis of boron, cal-
cium, and magnesium absorption and retention in postmeno-
I From the United States Department of Agriculture. Agricultural Re-
search Service, Grand Forks Human Nutrition Research Center, Grand
Forks, ND.
2 Presented in part at the 1987 Joint Meeting ofthe Minnesota and North
Dakota Academies of Science, Moorhead. MN, April 24, 1987: at the
Trace Element in Man and Animal-6 Symposium. Pacific Grove. CA. June
3. 1987: and at the 1994 annual meeting of the Federation of American
Societies ftr Experimental Biology. Anaheim. CA. April 26. 1994.
3 Mention of a trademark or proprietary product does not constitute a
guarantee or warranty of the product by the United States Department of
Agriculture and does not imply its approval to the exclusion of other
products that may also be suitable.
4 Address reprint requests to CD Hunt. USDA. ARS, GFHNRC. P0 Box
9034. Grand Forks. ND 58202-9034. E-mail: chunt@badlands.nodak.edu.
Received December 1 1, 1995.
Accepted for publication October 7. 1996.
by guest on July 21, 2011www.ajcn.orgDownloaded from
804 HUNT ET AL
pausal women maintained in a highly controlled metabolic
unit.
SUBJECTS AND METHODS
Volunteer selection
Healthy postmenopausal women were selected on the basis
of medical (normal bone, kidney, and liver function; normal
blood pressure; no chronic medication; negative lung scan),
psychologic [free of psychopathology as determined by the
Minnesota Multiphasic Personality Inventory (NCS Assess-
ment, Minneapolis), an extensive in-house psychologic history
questionnaire and clinical interview], and nutritional (no per-
tinent food allergies or refusal to eat required foods) data.
After selection, volunteers were admitted to the study after
being informed of its purpose and associated risks. The project
was approved by the Institutional Review Board of the Uni-
versity of North Dakota and the Human Studies Committee of
the US Department of Agriculture, Agricultural Research Ser-
vice. Informed consent and experimental procedures were con-
sistent with the Declaration of Helsinki. All subjects were
chaperoned when they left the metabolic unit to prevent inges-
tion of unauthorized foods or loss of excreta samples.
Fifteen postmenopausal women began the study initially. Two
subjects left the study for personal reasons; one replacement
volunteer did not complete all experimental phases. Data from one
volunteer on estrogen therapy were not included because previous
findings from this study indicated that boron supplementation
affects serum l7f3-estradiol concentrations (4).
The 1 1 subjects (not receiving estrogen therapy) who com-
pleted the study were aged 61 .4 ± 9.7 y ( ± SD) (range:
48-82 y); all were white. They were 165 ± 7 cm tall and
weighed 66.3 ± 8.4 kg at the beginning of the study. All
subjects had plasma calcium and magnesium concentrations
within the reference range of healthy subjects for our labora-
tory at admittance into the study (6). All subjects completed
3-d food diaries before admission into the study. On the basis
of dietary interview and computerized nutrient-intake calcula-
tions (GRAND, Grand Forks Research Analysis of Nutrient
Data; USDA/ARS Grand Forks Human Nutrition Research
Center, Grand Forks, ND) from the food dietary data (7-9),
average calcium and magnesium intakes for these 1 1 volun-
teens were 800 ± 288 and 280 ± 94 mg/d, respectively, before
their entry into the study. Smoking, alcohol consumption, and
drug use were prohibited and random screening was done to
monitor compliance. Various initial (cannabinoid, cocaine,
phencyclidine, and methadone) and monthly (opiate, barbitu-
rate, amphetamine, benzodiazepine, and alcohol) urine drug
screens were all negative.
Experimental design
The experimental design is summarized in Table 1. Subjects
were fed a basal diet that supplied an average of 0.36 mg B,
I 09 mg Mg, and < 0. 10 mg Al/8400 Id. The experimental
treatments were daily supplements (described below) of 0 or 3
mgB,Oor200mgMg,and0or 1000 mg Al.
The volunteers lived on a metabolic unit under close super-
vision for 167 d (divided into dietary periods that were subdi-
vided into 6-d excreta collection periods). After an equilibra-
tion period of 23 d (basal diet supplemented with 200 mg
Mg/d), all women participated in four 24-d dietary periods: 1)
basal diet only, 2) basal diet supplemented with 1000 mg Al/d,
3) basal diet supplemented with 200 mg Mg/d, and 4) basal diet
supplemented with 1000 mg Al and 200 mg Mg/d. The treat-
ments were arranged in a Latin-square design.
Completion of these four 24-d periods and the equilibration
period meant that the volunteers were fed a diet low in boron
for 1 19 d. After completing this phase of the study, all volun-
teens participated in two additional 24-d dietary periods in
which the basal diet was supplemented with 3 mg B/d. Six
women were fed the boron basal diet only and the boron basal
diet supplemented with 1000 mg Al/d; thus, these six women
were fed a diet low in magnesium for the full 48 d. The other
five women were fed the boron basal diet supplemented with
200 mg Mg/d and the boron basal diet supplemented with 200
mg Mg and 1000 mg Al/d. All dietary supplements throughout
the study were fed in a double-blind fashion (except for boron)
and given in divided doses at mealtimes.
Diet
The experimental treatments were daily supplements of 0 or
3 mg B (as Na2B4O7-10H20; JT Baker Inc, Phillipsburg, NJ) in
gelatin capsules that each provided 1 .0 ± 0.02 mg B/capsule);
0 or 200 mg Mg (as C12H26MgO16; Freeda Vitamins, Inc, New
York) in gelatin capsules that each provided 25 ± 0.5 mg
Mg/capsule, or 0 or 1000 mg Al [as Al(OH)3; Fisher Scientific,
Pittsburgh] in gelatin capsules that each provided 167 ± 3.3
TABLE 1
Experimental design’
. .
Dietary penod
and length
Treatments
B Mg, Al
Mg, Al Mg, Al
Mg, Al
ing/d mg/d
0, 23 d 0.36
309, <0.1 [3]2 309 <0.1 [3] 309, <0.1 [3]
309, <0.1 [3]
1, 24 d
0.36 109, <0.1 [3]
109, 1000 [3] 309, <0.1 [3] 309, 1000 [2]
2. 24 d 0.36 309, <0.1 [3]
309, 1000 [3]
109, 1000 [3] 109, <0.1 [2]
3, 24 d 0.36
109, 1000 [3] 109, <0.1 [3] 309, 1000 [3] 309, <0.1 [2]
4, 24 d 0.36 309, 1000 [3] 309, <0.1 [3]
109, <0.1 [3] 109, 1000 [2]
5, 24 d 3.23
109, <0.1 [2] 109, 1000 [4]
309, <0.1 [2] 309, 1000 [3]
6, 24 d
3.23 109, 1000 [2] 109, <0.1 [4]
309, 1000 [2] 309, <0.1 [3]
‘ Period 0 was an equilibration period. For periods 1-4, the magnesium and aluminum treatments were arranged in a Latin-square design. For periods
5-6, the magnesium treatment remained constant and was the reverse of that fed in period 4.
2 , in brackets.
by guest on July 21, 2011www.ajcn.orgDownloaded from
BORON AND ALUMINUM NUTRITURE IN POSTMENOPAUSAL WOMEN
805
mg Al/capsule (analyzed value: 176.5 mg Al/capsule). Place-
bos were indistinguishable from the paired supplements.
The basal diet was composed of conventional foods but low
in fruit and vegetables to minimize dietary sources of boron
and was planned on a 3-d notating menu cycle (Table 2). In
addition to the basal diet, volunteers were allowed to consume
limited quantities of several low-energy foods containing no
significant amounts of boron, magnesium, or aluminum. Salt,
pepper, and coffee, if selected individually by each volunteer,
were served in constant amounts throughout the study. Re-
corded quantities of sugar-free lemonade and tropical punch
(sweetened with aspartame) were consumed ad libitum but
limited to 0.96 L/d. Except for the lemonade and punch drinks,
all dietary ingredients were weighed to within 1% accuracy.
Unrecorded quantities of deionized water (= 18.0 Mf - cm;
Super Q system, Millipore Corp. Bedford, MA) were con-
sumed ad libitum.
The basal diet was planned as a weighed metabolic diet that
provided 6700-I 0 000 kJ/d (1 600-2400 kcal/d) at 830-kJ (200-
kcal) intervals. The 8400-kJ (2000-kcal) diet was the baseline
from which other energy levels were derived by varying the
amounts of all foods but not the amounts of specific vitamin
and mineral supplements described below (Table 3). The en-
ergy intake of each volunteer was based on her energy needs as
calculated with the Harris-Benedict equation (10) plus an ad-
ditional 50% of basal energy expenditure for normal activity.
Initial body weight was maintained (± 2%) by adjusting en-
ergy intake.
To ensure adequacy, the menu was supplemented with some
nutrients (Table 3) in constant amounts: 630 mg K/d as 1.2 g
KC1 (USP grade; JT Baker Chemical Co); an average of 135
mg Ca/d given as one on two tablets on alternate days as
calcium gluconate (90 mg Ca/tablet; Eli Lilly & Co. Indianap-
ohs), one at lunch and one at the evening meal; 0.8 mg Cu/d as
CuSO4 (USP grade; JT Baker Chemical Co) solution prepared
on site, dispensed into a breakfast beverage on day 1, an
evening-meal beverage on day 2, and a noon-meal beverage on
day 3; an average of 18 mg Fe/d as one tablet of ferrous
gluconate (36 mg Fe/tablet, Fergon; Winthrop Consumer Prod-
ucts, New York) at breakfast on alternate days; an average of
200 p.g folic acid/d given as one tablet (400 p.g folic acid/
tablet; Nature’s Bounty, Bohemia, NY) on alternate days at
breakfast; and 400 IU (10 g) cholecalcifenol/d as one tablet
(Natures’s Bounty) at the noon meal. Cholecalciferol was sup-
plemented in amounts greater than the recommended dietary
allowance (RDA; 1 1) to approximate typical US consumption.
Total calcium intake was provided in amounts less than the
RDA to help magnify the potential effects of dietary boron on
calcium and to maximize the effect of excess aluminum sup-
plementation on calcium absorption. Total folic acid intake
reflected partial fulfillment of the 1980 RDA (12).
Blood sampling
After a 10-h fast, blood samples were obtained from the
volunteers between 0600 and 0700 on days 7 (60 mL), 16 (60
mL), and 24 (90 mL) in each 24-d dietary period for analyses
that included mineral concentrations. To diminish carryover
artifact between treatment periods, all reported blood variables
represent data obtained from the third draw of each 24-d
dietary period only. Blood was drawn from the cubital vein
with a butterfly needle into 20-mL polypropylene syringes and
an aliquot for serum calcium and magnesium analyses was
immediately transferred to evacuated glass tubes with no ad-
ditive (Vacutainer; Becton Dickinson and Co. Lincoln Park,
NJ) and allowed to clot. For plasma copper, iron, and zinc
analyses, another aliquot was transferred to polypropylene
tubes containing 0.2 mL 10% potassium oxalate (Certified;
Fisher Scientific, Fain Lawn, NJ). Both samples were centni-
fuged (1800 X g for 10 mm at 4 #{176}C)and the serum or plasma
was removed by plastic transfer pipette and stored at 0 #{176}Cin
sealed 5-mL polypropylene tubes (Becton Dickinson and Co).
For plasma boron analysis, blood was drawn into an ultraclean
polypropylene syringe (Sarstedt, Inc. Newton, NC) that con-
tamed heparin (20 000 U hepanin in normal salinefL whole
TABLE 2
Three-day rotating menu
Day 1 Day2
Day3
Breakfast Orange drink mix
Pork sausage
White bread
Strawberry jelly
Margarine
French toast
Syrup
Margarine
2%-fat milk
Corn flakes
Sugar
2%-fat milk
Coffee cake with topping
Margarine
Dinner Barbecued beef
Steamed rice
Lettuce
French dressing
Pound cake
Breaded pork
Parsley potatoes
White bread
Margarine
Shortbread cookies
Orange drink mix
Hamburger-cheese casserole
White bread
Margarine
Vanilla wafers
Supper
Vegetable beef stew
White bread
Cheese
Margarine
Peach gelatin
Vanilla wafers
Lemon-lime carbonated beverage
Chicken rice soup
Crackers
Angel food cake
Cherries
Crispy pork
Steamed rice
Lime gelatin with peaches
Shortbread cookies
Snack Shortbread cookies
2%-fat milk
Cherry gelatin with pears
Vanilla wafers
Pound cake with lemon glaze
by guest on July 21, 2011www.ajcn.orgDownloaded from
806
HUNT ET AL
808 HUNT ET AL
TABLE 4
Effects of dietary boron, dietary aluminum, and their interaction on urinary and fecal mineral excretion in postmenopausal women fed different
amounts of magnesium’
Mineral excretion
Treatments
P< 0.1 mg Al/d 1000 m g Al/d
0.36 mg B/d 3.23 mg B/d B Al B X Al RMSE30.36 mg B/d 3.23 mg B/d
Basal magnesium diet
(109 mg Mg/d total)
Boron (% of intake)
Urinary 102 [6l 89 [5] 108 [6] 87 [6] NS NS NS 30
Fecal 12 [61 3 [5] 9 [6] 2 [6] 0.0001 0.03 NS 2
Calcium (% of intake)
Urinary 18.7 [61
17.5 [5] 19.1 [6] 17.6 [6] 0.05
NS NS 1.19
Fecal (%) 55.6 [61 48.6 [51 55.9 [61 55.2 [6] NS
NS NS 7.67
Magnesium, (% of intake)
Urinary 61.2 161 59.6 [51 62.0 [6] 61.5 [6] NS NS NS 2.46
Fecal 31.1 161 29.8 [5] 28.0 [61 26.6 [6] NS NS
NS 4.53
Supplemental magnesium diet
(340 mg Mg/d total)5
Boron (% of intake)
Urinary I 1 1 [5] 89 [5] 104 [51 89 [5] NS NS NS 38
Fecal 17 [5] 5 [5]
24 [5] 5 [5] 0.004 NS NS 10
Calcium (% of intake)
Urinary 23.4 [5] 28.7 [51 24.5 [51 26.3 [5] 0.05 NS NS 3.63
Fecal 67.7 [5] 65.4 [51 69.6 [5] 82.1 [5] NS 0.03 NS
8.15
Magnesium (% of intake)
Urinary
36.3 [5] 36.3 [51
34.8 [5] 35.9 [5] NS NS NS
1.64
Fecal 45.4 [5] 42.1 15] 42.6 151
48.7 [5] NS NS 0.04 4.51
, Urinary calcium and magnesium excretion data from the present study were published earlier (4) without corrections for changes in daily intake of
calcium and magnesium and on the basis of individual daily urine samples instead of composites.
2 Analyzed by repeated-measures ANOVA.
3 Root mean square error.
4 Group mean: a in brackets.
5 Mean daily amount of dietary magnesium: includes a 200-mg supplement of magnesium (per 8400 Id) as magnesium gluconate.
percentage of dietary calcium lost in the urine in volunteers fed
supplemental magnesium, which indicates that the effect of boron
is modified by magnesium nutriture.
Supplemental aluminum substantially decreased apparent
calcium absorption in volunteers fed supplemental magnesium
(Table 4). The mean daily calcium intakes in volunteers fed no
supplemental magnesium or supplemental magnesium were
580 ± 53 and 616 ± 69 mg/d, respectively; respective mean
daily calcium urinary losses were 121 ± 49 and 165 ± 108
mg/d and mean daily calcium fecal losses were 3 14 ± 127 and
430 ± 148 mg/d.
Magnesium
An interaction between dietary boron and aluminum affected
fecal magnesium in volunteers fed supplemental magnesium (Ta-
ble 4). Supplemental boron decreased apparent magnesium ab-
sorption when the diet contained supplemental aluminum. The
mean daily magnesium intakes in volunteers fed no supplemental
magnesium or supplemental magnesium were 109 ± 15 and
340 ± 19 mg/d, respectively; respective mean daily urinary mag-
nesium losses were 70 ± 12 and 123 ± 33 mg/d and mean daily
fecal magnesium losses were 32 ± 13 and 150 ± 51 mgld.
Blood mineral content
Boron
RBC boron concentrations were not affected by boron or
aluminum intake nor by an interaction between boron and
aluminum (Table 5). The effect of the dietary treatments on
plasma boron concentrations could not be assessed because
there were too few complete data sets, the result of randomly
insufficient plasma volumes. However, by collapsing the mag-
nesium and aluminum treatments, it was determined (by paired
t test) that the 9.0-fold increase in dietary boron increased
plasma boron concentrations only 1 .5-fold (8.79 ± 5. 18 corn-
pared with 5.92 ± 4.16 tmol B/U; P < 0.025). Plasma con-
centrations of copper, iron, and zinc were not affected by
dietary boron, dietary aluminum, or an interaction between
those two treatments regardless of magnesium nutniture (data
not shown).
Calcium, magnesium, and potassium
In volunteers fed no supplemental magnesium, supplemental
boron decreased serum magnesium concentrations (Table 5).
However, in volunteers fed supplemental magnesium, supple-
mental boron tended to increase RBC magnesium concentra-
tions (P < 0.07). Dietary boron did not affect serum calcium
(Table 5) or potassium (data not shown) concentrations in
volunteers fed either diets without or with supplemental
magnesium.
Electrocardiograms
The data collected from the 12-lead ECGs indicated that for
all volunteers for all measurement intervals, there was no
by guest on July 21, 2011www.ajcn.orgDownloaded from
BORON AND ALUMINUM NUTRITURE IN POSTMENOPAUSAL WOMEN 809
TABLES
Effects of dietary boron, dietary aluminum, and their interaction on blood mineral concentrations in postmenopausal women fed different amounts of
magnesium
Blood mineral
Treatments
P’< 0.1
mg Al/d
1000 m
g Al/d
0.36 mg B/d 3.23 mg B/d 0.36 mg B/d 3.23 mg B/d B Al B
X Al RMSE2
Basal magnesium diet
(109 mg Mg/d total)
Boron, red blood cell
(jtmollkg dry wt)
Calcium, serum (mmolIL)
Magnesium, serum (mmol/L)
Magnesium, red blood cell
(mmol/kg dry wt)
Supplemental magnesium diet
(340 mg Mg/d total)4
Boron, red blood cell
(pmolIkg dry wt)
Calcium, serum (mmolIL)
Magnesium, serum (mmollL)
Magnesium, red blood cell
(mmollkg dry wt)
14.6 [5]3
2.44 [5]
0.87 [5]
5.30 141
18.2 [4]
2.42 [4]
0.91 [41
5.26 [4]
16.1 [5]
2.42 [5]
0.83 [5]
5.59 [4]
16.2 [4]
2.44 [4]
0.86 [4]
5.47 [4]
16.9 [5]
2.48 [5]
0.90 [5]
5.47 [4]
29.8 [4]
2.40 [4]
0.88 [4]
5.38 [4]
18.3 [5]
2.48 [51
0.86 [5]
5.51 [41
18.8 [4]
2.46 [4]
0.86 [4]
5.92 [41
NS
NS
0.02
NS
NS
NS
NS
0.07
NS
NS
0.07
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
5.8
0.08
0.03
0.34
13.2
0.09
0.06
0.36
I Analyzed by repeated-measures ANOVA.
2 Root mean square error.
3 Group mean: n in brackets.
4 Mean daily amount of dietary magnesium; includes a 200-mg supplement of magnesium (per 8400 Id) as magnesium gluconate.
interruption of normal sinus rhythm. Supplemental boron de- to be prolonged when it lasts > 0.08 s (17). By this criterion,
creased the width of the QRS complex in volunteers fed no supplemental boron slightly improved electrical transmission.
supplemental magnesium but not in those fed supplemental Supplemental aluminum increased the width of the QRS com-
magnesium (Table 6). In general, a QRS duration is considered plex in volunteers fed supplemental magnesium but not in
TABLE 6
Effects of dietary boron. dietary aluminum, and their interaction on variables associated with circulatory functions in postmenopausal women fed
different amounts of magnesium
Circulatory function
Treatments
P’
< 0. 1 mg Al/d 1000 m g Al/d
0.36 mg B/d 3.23 mg B/d 0.36 mg B/d 3.23 mg B/d B Al B
X Al RMSE2
Basal magnesium diet
(109 mg Mg/d total)
Blood pressure (mm Hg)
Diastolic
Systolic
QRS complex (S)
Lead 1
Lead 2
Lead 3
Supplemental magnesium diet
(340 mg Mg/d total)4
Blood pressure (mm Hg)
Diastolic
Systolic
QRS complex (5)
Lead I
Lead 2
Lead 3
69 [6]
1 14 16]
0.078 [4]
0.088 [41
0.085 [4]
69 [5]
ll3[5]
0.072 [5]
0.080 [51
0.084 [5]
76 [6]
122 [6]
0.075 [4]
0.075 [41
0.075 [41
73 [5]
1l9[5]
0.068 [5]
0.084 [5]
0.084 [51
69 [6]
1 13 16]
0.085 [4]
0.085 [41
0.083 [4]
71 [51
ll2[5]
0.080 [51
0.094 [5]
0.096 [5]
75 [6J
120 [6J
0.073 [4]
0.080 [4]
0.080 141
72 [5]
1l9[5]
0.076 [5]
0.092 [5]
0.088 [5]
0.0001
0.0005
0.05
0.03
NS
0.03
0.03
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.05
0.01
0.01
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
2
4.20
0.007
0.007
0.007
2
5.56
0.008
0.008
0.006
I Analyzed by repeated-measures ANOVA.
2 Root mean square error.
3 Group mean; n in brackets.
.1 Mean daily amount of dietary magnesium; includes a 200-mg supplement of magnesium (per 8400 LI) as magnesium gluconate.
by guest on July 21, 2011www.ajcn.orgDownloaded from
810
HUNT ET AL
those fed no supplemental magnesium. Thus, the data suggest
that supplemental aluminum delayed conduction of the electni-
cal impulse through the ventricles.
Blood pressure
Supplemental boron induced a modest increase in systolic pres-
sure in volunteers fed either the no supplemental magnesium or
supplemental magnesium diets (Table 6). Because the volunteers
received boron supplementation during the last two treatment
periods only (Table 1), we examined whether the apparent effects
of boron on systolic pressure were the result of time. In comparing
individual systolic blood pressures with the day of experiment
(data not shown), the slope of the regression line changed between
the periods without and with supplemental boron in the plots of
three of the six volunteers fed no supplemental magnesium (P <
0.0006, 0.04, 0.04). This finding indicates that boron supplemen-
tation affected blood pressure in at least three of the six volunteers
fed basal amounts of magnesium. One of five volunteers fed
supplemental magnesium had a change (P < 0.0001) in slope
between the no supplemental boron and supplemental boron pe-
riods and three exhibited a trend toward (P < 0.06, 0.06), or a
significant increase in (P < 0.007), systolic pressure as a function
of time oven the entire length of the study. Thus, the apparent
effect of boron on systolic blood pressure in volunteers fed sup-
plemental amounts of magnesium was probably more the result of
time.
The observed effect of supplemental boron on diastolic
blood pressure may have been an artifact of time. For example,
only 1 volunteer in the group of 6 fed no supplemental mag-
nesium had a slope change between the no supplemental boron
and supplemental boron periods (P < 0.02); 5 of I 1 volunteers
fed either no supplemental magnesium or supplemental mag-
nesium showed trends toward, or significant increases in, dia-
stolic blood pressure as a function of time over the entire length
of the study (no supplemental magnesium: P < 0.07, 0.03,
0.09; supplemental magnesium: P < 0.02, 0. 1), or a trend
toward decreased diastolic blood pressure (no supplemental
magnesium: P < 0.06).
Kidney-related variables
Supplemental boron decreased blood urea nitrogen (BUN) in
volunteers consuming no supplemental magnesium (Table 7).
However, supplemental boron did not affect BUN in volunteers
consuming supplemental magnesium. Supplemental boron de-
creased urinary oxalate excretion in volunteers consuming no
supplemental magnesium but had no effect on oxalate excre-
tion in volunteers consuming supplemental magnesium. Urine
volume was not affected by supplemental boron during either
magnesium regimen. Two determinations of serum creatinine
oven the course of the study for purposes of general health
monitoring were insufficient for appropriate statistical analysis
of that variable.
DISCUSSION
Analyses of food and personal care products (I 8-20) mdi-
cate that usual adult human dietary consumption of boron in the
United States is in the range of 1-2 mg (0.092-0. 185 mmolld).
Increased consumption of specific foods with high boron con-
tent increases boron intake significantly; one serving of avo-
cado provides 1.1 1 mg (0.102 mmol) B (18). The findings from
the present study indicate that a low-boron diet [0.36 mg (0.033
mmol)/d] supplemented with boron in amounts [3.00 mg
(0.277 mmol)/d] equivalent to those found in diets with plenty
of fruit, vegetables, and nuts is sufficient to affect several
aspects of human physiology. Therefore, the findings on the
effects of dietary boron on human physiology are relevant to at
least postmenopausal woman and need to be considered when
mineral status assessments are conducted.
Boron absorption and excretion
All mineral excretion data were presented as a percentage of
mineral intake because those intakes fluctuated as energy in-
takes were adjusted to maintain body weight, a common con-
cern in human metabolic studies (2 1, 22). The current report
includes the first summary of boron balance data collected
from humans maintained in a highly controlled environment.
TABLE 7
Effects of dietary boron, dietary aluminum. and their interaction on kidney-related variables in postmenopausal women fed different amounts of
magnesium
Kidney-related variables
Treatments
P’< 0.1 mg Al/d 1000 m g Al/d
0.36 mg B/d 3.23 mg B/d 0.36 mg B/d 3.23 mg B/d B Al B X Al RMSE2
Basal magnesium diet
(109 mg Mg/d total)
Oxalate, urine (mol/d)
Urea nitrogen, serum (mmolIL urea)
Urine volume (Lid)
Supplemental magnesium diet
(340 mg Mg/d total)4
Oxalate, urine (pmol/d)
Urea nitrogen, serum (mmolIL urea)
Urine volume (Lid)
130 [6l
6.36 [51
2.31 [6]
100 [5]
5.28 [41
2.89 [5]
89 [6]
5.30 151
2.29 [61
104 151
5.88 [4]
3.04 [5]
104 [61
6.55 15]
2.40 [6]
93 [5]
5.58 [41
2.86 [5]
87 16]
5.72 [5]
2.31 [61
89 [5]
5.75 [41
2.88 [5]
0.03
0.003
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
29
0.58
0.22
15
0.83
0.41
‘ Analyzed by repeated-measures ANOVA.
2 Root mean square error.
3 Group mean: a in brackets.
4 Mean daily amount of dietary magnesium: includes a 200-mg supplement of magnesium (per 8400 LI) as magnesium gluconate.
by guest on July 21, 2011www.ajcn.orgDownloaded from
BORON AND ALUMINUM NUTRITURE IN POSTMENOPAUSAL WOMEN
811
The boron excretion data do not indicate that a dietary intake of
3.23 mg B/d causes abnormal boron accumulation. Further-
more, the data suggest that postmenopausal women ingesting
very low amounts of boron (0.36 mg/d) are prone to net boron
loss.
The boron excretion data support earlier findings that boron
is highly absorbable and excreted primarily by the kidneys.
Urinary excretion data collected from rats indicated that the
absorption of an intrinsically labeled ‘#{176}Bdose from broccoli
was 100% (23). In heifers (24) on sheep (25), only 30% on 41%,
respectively, of dietary boron was excreted in the urine. How-
ever, the boron load (from natural typical foodstuffs) on a body
weight basis was much higher in those animal studies (0.715
mg B in the heifens and 0.667 mg B/kg body wt in the sheep)
than in the present study (0.005 mg B in the no supplemental
boron period and 0.048 mg B/kg body wt in the supplemental
boron period). Other boron excretion data from the present
study also support earlier findings (23) that the rate of boron
excretion is extremely rapid. For example, within 24 h after
initiation of the boron supplementation regimen, urinary boron
rose to amounts similar to those present in the supplement (data
not shown).
A range of boron concentrations in whole blood of appar-
ently healthy humans with unknown dietary histories was re-
ported (26). It was the opinion of the investigators that the
range was much narrower than that expected for a nonessential
ultratrace element. A finding from a study with yearling beef
heifers indicated that as the amount of filtered boron increased,
the percentage of filtered boron that was reabsorbed decreased
(27). Also, in another animal study, female rats consuming
water high in boron (100 mgIL) for 21 d showed increased
plasma boron concentrations although some mechanism con-
currently eliminated any excess boron from the liver and brain
against their own concentration gradients (28). A finding from
the present study indicates that there is an obligatory boron
loss. During the dietary regimen providing 340 mg Mg/d, 89%
of dietary boron was excreted in the urine in volunteers fed
supplemental boron, a percentage that increased to 1 1 1% in
volunteers deprived of boron. Further study with concurrent
measures of urinary creatinine and plasma boron concentra-
tions is needed to determine whether increased renal boron
clearance during boron supplementation reflects decreased tu-
bular reabsorption and, therefore, homeostatic control of boron.
Effects on calcium and magnesium utilization
A previous report on the present study concluded that boron
supplementation reduced urinary calcium and magnesium loss
(4). That report did not take into account adjustments in energy
intake as described above. As now reported, boron supplemen-
tation did not affect urinary magnesium excretion but did
induce a minor decrease or modest increase in the percentage
of dietary calcium lost in the urine as magnesium intake
changed from amounts considered appreciably lower than, to
slightly more than, the current RDA for magnesium, nespec-
tively (I I). Similar findings were reported from an animal
study in which boron supplementation (2.46 compared with
< 0.06 mg B/kg diet) increased total 24-h urinary calcium
(0.91 ± 0.38 compared with 0.72 ± 0.22 mg) in 8-wk-old,
cholecalciferol-depnived rats (exhibiting few signs of cholecal-
ciferol deficiency) fed adequate magnesium (29). However, the
findings from two separate studies of female volunteers differ
from those of the present study. In a I 0-mo study of premeno-
pausal women consuming 83 mg Mg and 680 mg Ca/d,
supplemental boron did not affect urinary calcium loss (30). In
a short 6-wk study of postmenopausal volunteers consuming
298 mg Mg and 927 mg Ca/d, supplemental dietary boron
did not affect urinary excretion of either calcium or magnesium
(3 1). However, the composition of the basal diet seemed to
elevate urinary excretion of calcium, which may have inhibited
or obscured any effect of boron.
Assuming a magnesium sweat loss of 15 mg/d (32, 33), a
dietary intake of 109 mg Mg/d resulted in essentially zero
magnesium balance in the postmenopausal women. Earlier
findings indicate that a magnesium intake of I 00 mg/d for
14 d is sufficient to maintain positive magnesium balance
(34). Whether zero magnesium balance can be maintained for
periods longer than 24 d in postmenopausal women fed 109 mg
Mg/d remains unknown.
Other findings from this study indicate that dietary alumi-
num decreases calcium absorption. In a different study with
adult men, small doses of aluminum-containing antacids (for
example, between 90 and 450 mL antacid/d containing be-
tween 1530 and 7650 mg Al/d) increased fecal calcium when
they received an average of 252 mg Ca/d (35). During a
calcium intake of 800 mg/d, these doses of antacids did not
affect calcium excretion. Therefore, the present findings mdi-
cate that aluminum supplementation should probably be lim-
ited or at least monitored in postmenopausal women prone to
excessive calcium loss and who are consuming low amounts of
calcium.
Blood mineral content
Boron
The mean plasma boron concentrations reported in the cur-
rent study are higher than limited published values. The newest
study available reported a median plasma boron concentration
of2.3l j.mol/L (range: 1.30-3.61 p.molIL) for 12 subjects with
detectable boron concentrations but unknown dietary histories
(36). Boron volatizes at high temperatures (5) and it is un-
known whether boron was lost during vessel venting, a neces-
sary step in the microwave digestion method used in that study.
The method used to analyze for boron in the present study
provides satisfactory recovery of boron from spiked samples
(99.7 ± 0.5%) (5). In a different study, the plasma boron
concentration (determined by neutron activation and mass
spectrometny) of one individual with an unknown dietary his-
tory was reported as 3.03 ± 0. 15 .amolIL (26). Other investi-
gatons reported a median value of 2.06 jamol/L and a range of
0.77-4.45 j.tmollL for serum boron (37). The amount of boron
(18.6 p.mollkg dry wt) found in washed, dried RBCs in the
present study was approximately twice that reported in whole,
dried blood analyzed by neutron irradiation (9.0 p.mol/kg dry
wt) (38).
Although boron supplementation did increase plasma boron
concentrations, the data do not indicate that plasma boron
content is a sensitive indicator of dietary boron intake. In the
present study, the supplemental boron regimen provided a
9.0-fold increase in dietary boron, yet this increase caused only
a 1 .5-fold increase in plasma boron concentrations in post-
menopausal women. In magnesium-adequate chicks, a ninefold
increase in dietary boron (1.58 compared with 0.18 mg/kg)
by guest on July 21, 2011www.ajcn.orgDownloaded from
812 HUNT ET AL
increased plasma boron concentrations by only twofold (10.8
compared with 5.3 .amoUL). In another study with cholecal-
ciferol-deficient chicks only, a 7.5-fold increase in dietary
boron (from 0.465 to 3.465 mg/kg) yielded only a 2.0-fold
increase (from 7. 12 to I 4. 1 molJL) in plasma boron concen-
trations (39).
Magnesium
The mechanism by which dietary boron supplementation
reduced serum magnesium concentrations in postmenopausal
women fed no supplemental magnesium in this study on similar
amounts in a different study (40) remains unknown. Inorganic
boron in concentrations found in blood or urine of normal pH
(7.35-7.45 or 4.5-8, respectively) would exist almost entirely
in the mononuclear uncharged species B(OH)3 (41) and there-
fore would not complex with the portion of ultrafilterable
plasma magnesium that complexes with anions (42). Further-
more, the disproportionate molar ratio between magnesium and
boron in plasma (0.87:0.01) or urine (5.00:0.27) would pre-
dude any significant direct chemical interaction. Thus, boron
may influence magnesium metabolism through intermediate or
parallel molecular mechanisms.
Blood pressure and kidney-related variables
The finding that boron supplementation increased systolic
blood pressure in postmenopausal women fed basal amounts of
magnesium is reported because of no compelling reason to
assume it to be artifactual. The finding was unexpected because
there is strong evidence that lactoovovegetarian diets (high in
boron content), compared with omnivorous diets (typically
lower in boron), depress blood pressure (43, 44); although fruit
and vegetables are important sources of many nutrients, they
also have a relatively high boron content (typically 2-3 mg
B/kg wet wt) (19). The total amount of boron consumed by the
boron-supplemented women in the present study was similar to
that typically consumed by the adult US population. Therefore,
it seems highly unlikely that the affect of boron on blood
pressure was pharmacologic or toxicologic in nature. Most
importantly, supplemental boron did not increase systolic pres-
sure to values outside of the normal range.
In a different study, BUN concentrations were reported to be
lower at the end of a boron-repletion period of 49 d (4. 1 mmol
urea/L) compared with those determined at the end of the
preceding boron depletion period of 63 d (4.8 mmol urea/L)
(45). That finding was replicated in the present study. BUN
concentrations, in the absence of disease states, reflect the
degree of protein catabolism. whether produced by a high-
protein diet or by factors that result in the mobilization of
protein for energy purposes (46). There is previous evidence
for a role of boron in energy substrate metabolism. In the
cholecalciferol-deficient chick nutrition model, physiologic
supplements of boron alleviated perturbation in plasma glucose
and tniacylglycerol concentrations and substantially improved
food consumption (47). Furthermore, findings from a human
study conducted after the one reported here indicated that
dietary boron supplementation decreased serum glucose con-
centrations in postmenopausal women fed 1 15 mg Mg/d (45).
Thus, the modest influence of boron on BUN concentrations
may reflect an uncharactenized role for boron in energy sub-
strate metabolism.
Characterization of the mechanism through which boron
affects oxalate excretion may reveal a relation between boron
nutriture and the formation of oxalate stones, one type of
urolithiasis responsible for considerable pain, discomfort, and
medical expense in the human population. Oxalate in the body
is derived from dietary sources (rhubarb, spinach) and from
glycine and ascorbic acid metabolism (48). It is thought that a
decrease in urinary oxalate excretion would reduce the degree
of urine supersaturation with respect to calcium oxalate and so
diminish the tendency to form oxalate stones (49). Because
supplemental boron reduced total urine oxalate and calcium in
volunteers fed no supplemental magnesium (but not in those
fed supplemental magnesium), postmenopausal women con-
suming diets low in magnesium may benefit from dietary boron
in amounts normally found in diets with ample quantities of
fruit and vegetables.
In summary, the findings indicate that volunteers fed 0.36 or
3.23 mg B/d did not accumulate boron. The boron supplement
was within the range of normal dietary boron intake. Obliga-
tory boron loss and relatively small changes in boron blood
values during substantial increases in dietary boron support the
concept of boron homeostasis and, therefore, a possible bio-
logical function for boron. Dietary boron directly affects mag-
nesium metabolism because supplemental boron decreased Se-
rum magnesium concentrations in volunteers fed no
supplemental magnesium. Decreased total urine oxalate during
supplemental boron treatment in volunteers receiving no sup-
plemental magnesium indicates a possible role for boron in the
control of urolithiasis. The change in urinary calcium excretion
as a function of boron and magnesium nutriture indicates a
close interrelation among boron, calcium, and magnesium, a
relation that may be important in understanding better the
mineral metabolism disorders that perturb long-term calcium
balance. Finally, aluminum supplementation should be limited
or at least monitored in postmenopausal women prone to ex-
cessive calcium loss. U
We thank members of the Grand Forks Human Nutrition Research
Center staff whose special talents and skills made this study possible.
Members of this staff include Leslie Klevay (medical); Henry Lukaski
(physiologic): James Penland (psychologic): David Milne and Sandra
Gallagher (clinical chemistry); Karin Nunley, Shiela Massie, Pat Willey,
Susan Fleck-Sheppard, and Lane Cunningham (sample processing);
Terrance Shuler (analytic); LuAnn Johnson and Maria Siu (statistical
analysis); Betty Vetter and Donna Neese (nursing): Janet Greger and Janet
Hunt (special planning comments); and LoAnne Mullen (dietary).
REFERENCES
1. Hunt CD, Nielsen FH. Interaction between boron and cholecalciferol
in the chick. In: Gawthorne I, White C, eds. Trace element metabolism
in man and animals-4. Canberra, Australia: Australian Academy of
Science, 1981:597-600.
2. Hunt CD, Nielsen FH. Interactions among dietary boron, magnesium,
and cholecalciferol in the chick. Proc N D Acad Sci 1987;41:50.
3. Nielsen F, Shuler TR, Zimmerman TI, Uthus EO. Magnesium and
methionine deprivation affect the response of rats to boron deprivation.
Biol Trace Elem Res 1988:17:91-107.
4. Nielsen FH, Hunt CD, Mullen L, Hunt JR. Effect of dietary boron on
mineral, estrogen, and testosterone metabolism in postmenopausal
women. FASEB I I 987; 1:394-7.
5. Hunt CD, Shuler TR. Open-vessel. wet-ash, low-temperature digestion
of biological materials for inductively coupled argon plasma spectros-
by guest on July 21, 2011www.ajcn.orgDownloaded from
BORON AND ALUMINUM NUTRITURE IN POSTMENOPAUSAL WOMEN
813
copy (ICAP) analysis of boron and other elements. I Micronutrient
Anal 1989:6: 16 1-74.
6. Milne DB. Trace elements. In: Burtis C, Ashwood E, eds. Tietz
textbook of clinical chemistry. Philadelphia: WB Saunders, 1994:
I317-53.
7. Anonymous. US Department of Agriculture handbooks 8.1-14 and 16:
composition offoods. Hyattsville, MD: US Department of Agriculture,
1976-1986.
8. Anonymous. US Department of Agriculture Consumer Nutrition Cen-
ter Data Set 456-3. Hyattsville. MD: US Department of Agriculture,
1977.
9. Anonymous. US Department of Agriculture provisional table on the
nutrient content of bakery foods and related items. Hyattsville, MD:
US Department of Agriculture, 1981.
10. Harris IA, Benedict FG. A biometric study of basal metabolism in
man. Philadelphia: Lippincott, I 919.
1 1. National Research Council. Recommended dietary allowances. Wash-
ington, DC: National Academy Press. 1989.
I 2. National Research Council. Recommended dietary allowances. Wash-
ington, DC: National Academy of Sciences, 1980.
13. Bates B. A guide to physical examination. Philadelphia: lB Lippincott
Co, 1974.
14. Trudeau DL, Freier EF. Determination of calcium in serum by atomic
absorption spectrophotometry (AAS). Clin Chem 1967;12: 101-14.
15. Milne DB. Canfield WK, Mahalko JR. Sandstead HH. Effect of oral
folic acid supplement on zinc, copper, and iron absorption and excre-
tion. Am I Clin Nutr 1984:39:535-9.
16. Sims RL, Mullen LM, Milne DB. Application of inductively coupled
plasma emission spectroscopy to multielement analysis of foodstuffs
used in metabolic studies. I Food Comp Anal 1990:3:27-37.
17. Guyton AC. Textbook of medical physiology. Philadelphia: WB Saun-
dersCo, 1971.
18. Anderson DL, Cunningham WC, Lindstrom TR. Concentrations and
intakes of H, B, 5, K, Na, Cl, and NaCI in foods. I Food Comp Anal
1994:7:59-82.
19. Hunt CD, Shuler TR, Mullen LM. Concentration of boron and other
elements in human foods and personal-care products. I Am Diet Assoc
l99l;9l :558-68.
20. Iyengar GV, Clarke WB, Downing RG. Determination of boron and
lithium in diverse biological matrices using neutron activation-mass
spectrometry (NA-MS). Fresenuis I Anal Chem 1990:338:562-6.
21. Wardlaw GM, Snook IT, Lin M-C, Puangco MA, Kwon IS. Serum
lipid and apolipoprotein concentrations in healthy men on diets en-
niched in either canola oil or safflower oil. Am I Clin Nutr
1991:54:104-10.
22. Coburn SP, Ziegler P1, Costill DL, et al. Response of vitamin B-6
content of muscle to changes in vitamin B-6 intake in men. Am I Cliii
Nutr 1991:53:1436-42.
23. Vanderpool RA, Hoff D, Johnson PE. Use of inductively coupled
plasma-mass spectrometry in boron- I 0 stable isotope experiments
with plants, rats, and humans. Environ Health Perspect 1994: 102(suppl
7): 13-20.
24. Green GH, Weeth HI. Responses of heifers ingesting boron in water.
I Anim Sci 1977:46:812-8.
25. Brown TF, McCormick ME, Moms DR. Zeringue LK. Effects of
dietary boron on mineral balance in sheep. Nutr Res 1989:9:503-12.
26. Clarke WB, Koekebakker M, Barr RD. Downing RG, Fleming RF.
Analysis of ultratrace lithium and boron by neutron activation and
mass-spectrometric measurement of 3He and 4He. Int I Rad AppI
Instrum [A] 1987:38:735-43.
27. Weeth HI, Speth CF. Hanks DR. Boron content of plasma and urine as
indicators of boron intake in cattle. Am I Vet Res 1981:42:474-7.
28. Magour S. Schramel P. Ovcar I, Maser H. Uptake and distribution of
boron in rats: interaction with ethanol and hexobarbital in the brain.
Arch Environ Contam Toxicol 1982:11:521-5.
29. Hunt CD, Herbel I. Effects of dietary boron on calcium and mineral
metabolism in the streptozotocin-injected, vitamin D-deprived rat.
Magnes Trace Elem 1991-1992:10:387-408.
30. Meacham SL. Taper LI, Volpe SL. Effect of boron supplementation on
blood and urinary calcium, magnesium, and phosphorus, and urinary
boron in athletic and sedentary women. Am I Clin Nutr
1995:61:341-5.
3 1 . Beattie IH, Peace HS. The influence of a low-boron diet and boron
supplementation on bone, major mineral and sex steroid metabolism in
postmenopausal women. Br I Nutr 1993:69:871-84.
32. Consolazio CF. Matoush LO, Nelson RA, Harding RS, Canham IE.
Excretion of sodium, potassium, magnesium and iron in human sweat
and the relation of each to balance and requirements. I Nutr
1963:79:407-15.
33. Seelig MS. Magnesium requirements in human nutrition. Magnes Bull
1981:3:26-47.
34. Marshall DH. Nordin BEC, Speed R. Calcium, phosphorus, and mag-
nesium requirement. Proc Nutr Soc 1976:35:163-73.
35. Spencer H, Kramer L. Norris C, Osis D. Effect of small doses of
aluminum-containing antacids on calcium and phosphorus metabo-
lism. Am I Clin Nutr 1982:36:32-40.
36. Ferrando AA, Green NR, Barnes KW, Woodward B. Microwave
digestion preparation and ICP determination of boron in human
plasma. Biol Trace Elem Res 1993:37:17-25.
37. Abou-Shakra FR, Havercroft I, Ward N. Lithium and boron in bio-
logical tissues and fluids. Trace Elem Med 1989:6:142-6.
38. Clarke WB, Webber CE, Koekebakker M, Barr RD. Lithium and
boron in human blood. I Lab Clin Med 1987:109:155-8.
39. Hunt CD. Dietary boron modified the effects of magnesium and
molybdenum on mineral metabolism in the cholecalciferol-deficient
chick. Biol Trace Elem Res 1989:22:201-20.
40. Nielsen FH, Mullen LM, Gallagher 5K. Effect of boron depletion and
repletion on blood indicators of calcium status in humans fed a
magnesium-low diet. I Trace Elem Exp Med 1990:3:45-54.
41. Kakihana H, Kotaka M, Satoh 5, Nomura M, Okamoto M. Fundamen-
tat studies on the ion-exchange separation of boron isotopes. Bull
Chem Soc lpn 1977:50:158-63.
42. Walser M. Magnesium metabolism. Ergeb Physiol 1967:59:185-296.
43. Rouse IL, Beilin U, Armstrong BK, Vandongen R. Blood pressure
lowering effect of a vegetarian diet: a controlled trial in normotensive
subjects. Lancet 1983:1:5-10.
44. Margetts BM, Beilin LI, Vandongen R, Armstrong BK. Vegetarian
diet in mild hypertension: a randomized controlled trial. Br Med I
1986:293:1468-71.
45. Nielsen FH. Dietary boron affects variables associated with copper
metabolism in humans. In: Anke M. Baumann W, Brhunlich H,
BrUckner C, Groppel B, Grain M. eds. 6th International Trace Element
Symposium 1989, Vol 4. lena, Germany: Karl-Marx-Universitat,
Leipzig and Friedrich-Schiller-Universitat, 1989: 1 106-1 1.
46. Kaplan A. Szabo LL. Clinical chemistry: interpretations and tech-
niques. Philadelphia: Lea & Febiger, 1983.
47. Hunt CD, Herbel IL, Idso IP. Dietary boron modifies the effects of
vitamin D3 nutriture on indices of energy substrate utilization and
mineral metabolism in the chick. I Bone Miner Res 1994:9:171-81.
48. Cornatzer WE. Clinical application of laboratory tests. Springfield, IL:
Charles C Thomas, 1986.
49. Anonymous. Diet and urolithiasis. Dairy Council Digest 1983;
54:25-30.
by guest on July 21, 2011www.ajcn.orgDownloaded from
