Bioavailability of Nickel in Man: Effects of Foods and Chemically-Defined Dietary Constituents on the Absorption of Inorganic Nickel3

Abstract

By serial determination of the change in plasma nickel concentration following a standard dose of 22.4 mg of nickel sulfate hexahydrate containing 5 mg of elemental nickel, the bioavailability of nickel was estimated in human subjects. Plasma nickel concentration was stable in the fasting state and after an unlabeled test meal, but after the standard dose of nickel in water was elevated 48.8, 73.0, 80.0, and 53.3 microgram/1, respectively, at hours 1, 2, 3, and 4. Plasma nickel did not rise above fasting levels when 5 mg of nickel was added to two standard meals: a typical Guatemalan meal and a North American breakfast. When 5 mg of nickel was added to five beverages-whole cow milk, coffee, tea, orange juice, and Coca Cola-the rise in plasma nickel was significantly suppressed with all but Coca Cola. Response to nickel also was suppressed in the presence of 1 g of ascorbic acid. Phytic acid in a 2:1 molar ratio with nickel, however, did not affect the rise in plasma nickel. The chelate of iron and ethylenediaminetetraacetate, NaFeEDTA, an iron-fortifying agent suggested for application in Central America, slightly but not significantly depressed plasma nickel rise at 2 hours, whereas disodium EDTA depressed plasma nickel levels significantly below the fasting nickel curve at 3 and 4 hours postdose. These studies suggest that the differential responses of inorganic nickel to distinct foods, beverages, and chemically-defined dietary constituents could be important to human nutrition.

Full text

Bioavailability of Nickel in Man: Effects of Foods and
Chemically-Defined Dietary Constituents on the Absorption
of Inorganic Nickel1'5
NOEL W. SOLOMONS,0 1 FERNANDO VITERI.t
TERRENCE R. SHULERf ANDFORREST H. NIELSEN}
'Massachusetts Institute of Technology, Cambridge, MA
02139; the \Division of Human Nutrition and Biology,
Institute of Nutrition of Central America and Panama,
Guatemala City, Guatemala, Apartado 11-88; and \U. S.
Department of Agriculture, Science and Education
Administration, Grand Forks Human Nutrition Research
Center, Grand Forks, ND, 58202.
ABSTRACT By serial determination of the change in plasma nickel concentration
following a standard dose of 22.4 mg of nickel sulfate hexahydrate condtaining 5 mg
of elemental nickel, the bioavailability of nickel was estimated in human subjects. Plasma
nickel concentration was stable in the fasting state and after an unlabeled test meal, but
after the standard dose of nickel in water was elevated 48.8, 73.0, 80.0, and 53.3 pg/
1, respectively, at hours 1, 2, 3, and 4. Plasma nickel did not rise above fasting levels
when 5 mg of nickel was added to two standard meals: a typical Guatemalan meal and
a North American breakfast. When 5 mg of nickel was added to five beverages-whole
cowmilk, coffee, tea, orange juice, and Coca Cola®-the rise in plasma nickel was sig
nificantly suppressed with all but Coca Cola®.Response to nickel also was suppressed
in the presence of l g of ascorbic acid. Phytic acid in a 2:1 molar ratio with nickel,
however, did not affect the rise in plasma nickel. The chelate of iron and ethylenedi-
aminetetraacetate, NaFeEDTA, an iron-fortifying agent suggested for application in
Central America, slightly but significantly depressed plasma nickel rise at 2 hours,
whereas disodium EDTA depressed plasma nickel levels significantly below the fasting
nickel curve at 3 and 4 hours postdose. These studies suggest that the differential
responses of inorganic nickel to distinct foods, beverages, and chemically-defined dietary
constitutents could be important to human nutrition. J. Nutr. 112: 39-50, 1982.
INDEXING KEY WORDS nickel •ascorbic acid •phytic acid •EDTA
Nickel has been recognized recently as The nickel contents of common foods from
essential for certain vertebrate species in- the United States (15, 16), Great Britain (17,
eluding rats (1-3), chicks (4), swine (5, 6),
goats (6), and sheep (7). Biologically-impor
tant interactions between nickel and other -
trace minerals have also been discovered (8- ?!â„¢lwdj""""""T 8^ â„¢*
r l l t i_ 111 Presented in part at the Western Hemisphere Nutrition Congress VI,
11). In tissue or healthy humans, nickel has LOSAngeles,August,i9so,p.&4(abstracts).
been consistently found in concentrations .
Tanging from 0.04 tO 2.8 Mg/g °n a dry basis ll, Guatemala City, Guatemala, Central America, Apartado 11-88.
/IO l O\ ICU i J XT /i A\ *Mention of a trademark or proprietary product does not constitute a
(12, 13), and Scnroeder and Nason (14) es- guarante<,orwarrantyo(theproductbytheu s.DepartmentofAgriculture
timated that the human body Contains 10 mg and does not imP'y its »PP"»alto the exclusion of other products that may
Of nickel, OÕ which 18% ¡S in the skin. It iS 5SA'bbret¡at¡onSUSed;NickeUulfate=NiS0..6H,0;ethylenedi»nimc-tet-
that nickel iS metabolically active, if raacetic acid =, EI?TA; sodi,um jron ethylenediamineletraacetic acid
.. J = NaFeEDTA; disodium elhylenediaminetetraacetic acid = N32EDTA.
not nutritionally essential, in humans as well. 2H,o
39
by guest on July 14, 2011jn.nutrition.orgDownloaded from

40 SOLOMONS ET AL.
18), Sweden (19), and Holland (20) are avail
able. Dry beans, cocoa products, baking soda,
and some nuts contain high levels of nickel
(>2.0 Mg/g); wheat and wheat products,
shellfish, processed meats and many vege
tables contain intermediate levels (0.2-2.0
Mg/g); and whole and dried milk, fresh fruits,
meat, eggs and Coca Cola®contain low levels
of nickel (<0.2 ng/g). Thomas et al. (17) sug
gested that contact between food and ma
chinery or cans can contribute to dietary
nickel, as processed and canned vegetables
contain more than fresh. In some areas of
Europe, where nickel is used as a catalyst in
hydrogénation, margarine can be a substan
tial dietary source of nickel (21).
In Sweden, the estimated daily nickel in
take ranged from 200 to 4460 /ig and aver
aged 750 ng (19). Early estimates of daily
nickel consumption in the U. S. ranged from
300 to 600 jug (15). The average nickel con
tent of nine institutional diets in North Da
kota was 168 ±11 Hg, or 75/1000 kcal (22).
Based on extrapolation from animal data, the
hypothetical human requirement for nickel
would be 16 to 25 ¿tg/1000 kcal or about 75
Hg of elemental nickel per day (23). Most
balanced diets probably exceed that amount
of nickel. The issue of nickel in the human
diet, however, has further ramifications.
Low-nickel diets have been advocated in
the management of nickel-sensitivity der
matitis (21, 24) which apparently is exacer
bated by orally-ingested nickel (25). The pos
sibility was advanced that a strict nickel-
elimination diet might have implications for
human health (23).
Despite broad, general knowledge about
customary dietary intakes, little is known
about the chemical form of nickel in foods
or the factors that affect its biological avail
ability. Concern about the biological avail
ability of trace elements in food is justified
by extensive experience with dietary iron
(26-28) and dietary zinc (29-32). Vohra et
al. (33) showed that nickel was among a host
of elements that formed stable complexes
with phytic acid in vitro; possibly such com
plexes would explain interference of unre
fined cereal foods with the absorption of
nickel (34). In the only study of its kind,
Horak and Sunderman (35) have estimated
from nickel balance experiments that about
10% of the nickel in a normal diet is ab
sorbed.
Techniques to pursue a detailed explora
tion of nickel bioavailability are limited. In
laboratory animals, Onkelinx et al. (36, 37)
used a radioisotope, ^Ni, as a tracer; the ra
dioactive half-life (92 years) of this isotope,
however, precluded its use in man. A stable
isotope of nickel, MNi, could theoretically be
used as a tracer because its natural abun
dance is 1.16% and it can be measured by
thermal neutron activation analysis. Tech
nological aspects of this approach, however,
await further development. Spruits and Bon-
gaarts (38) reported that plasma nickel con
centration rose by about 54 /ig/1 within two
hours of the ingestion of 5 mg of elemental
nickel as 22.4 mg of nickel sulfate in a single
healthy individual. As the change in plasma
zinc concentration has recently been ex
ploited as an index of zinc absorption (34, 35,
39-44), we thought that an analagous ap
proach could be applied to study the effects
of meals, beverages, and chemically-defined
dietary constituents on the absorption of in
organic nickel.
MATERIALS AND METHODS
Subjects. The subjects were adult volun
teers, male and female, in apparent good
health and without known or suspected gas
trointestinal disease. They agreed to partic
ipate after the nature and purpose of the
study were carefully explained. No individ
ual with a history of reactions related to
nickel sensitivity was included in the study.
Experiments were performed in accordance
with the Declaration of Helsinki and the pro
tocol were approved by the Committee on
the Use of Humans as Experimental Subjects
of the Massachusetts Institute of Technology.
Plasma nickel determinations. Blood sam
ples (6 ml) were drawn before and hourly for
4 hours after administration of the dose of
nickel. Venous blood was sampled with plas
tic syringes and stainless-steel needles, and
collected in plastic tubes (Falcon®tubes, Di
vision of Becton, Dickinson & Co., Oxnard,
CA) containing 50 /ul of 20% potassium ox-
alate. The whole blood was centrifuged and
by guest on July 14, 2011jn.nutrition.orgDownloaded from

BIOAVAILABILITY OF NICKEL IN MAN 41
the plasma was frozen until it was analyzed
for content. The plasma samples were thawed
at room temperature, then mixed on a vortex-
type rotary mixer. One ml of plasma was
transferred to a 12 X 100 mm borosilicate
glass culture tube (Kimax®,borosilicate glass
culture tubes, Kimble Division, Owens-Illi
nois, Toledo, OH) to which 0.6 ml of nitric
acid (double distilled in Vycor, G. Frederick
Smith, Columbus, OH) was added. The tubes
were tightly capped with teflon-lined screw
caps and placed into an 80°water bath for
16 hours. After digestion, the samples were
stored in 12 X 75 mm polyethylene tubes
(Falcon® tubes) until analysis. Throughout
the analytic process, duplicate tubes of con
trol serum (Wellcomtrol® One BC01 Quality
Control Serum, Burroughs Wellcome Co.,
Greenville, NY), nickel standard (50 /tl/1)
and blanks were subjected to the same pro
cedures used for the plasma samples.
Nickel was quantitated by its absorbance
at 232.3 nm with an atomic absorption
spectrophotometer (Model 503, HGA-2100
Graphite Furnace, Perkin-Elmer Corp., In
struments Division, Norwalk, CT) equipped
with electrothermal atomizer, nickel "inten-
sitron" hollow cathode lamp, control unit
with a temperature-programming accessory
for "ramp mode," deuterium arc background
corrector, and model 56 recorder. Of the
plasma sample, 50 ¿ilwas mixed with 50 n\
of concentrated ammonium hydroxide (an
alyzed, J.T. Baker Co., Phillipsburg, NJ) in
side the pyrolytically-coated graphite tube.
The analytical settings were: lamp current,
20 ma; slit width, 3 (0.2 nm); recorder range,
10 mV full scale; chart speed, 5 mm/min.
The electrothermal atomizer was pro
grammed for the following cycle: drying,
ramped for 30 seconds from 25° to 125°,
which was maintained for 80 seconds; ashing,
ramped for 80 seconds from 125°to 1000°,
which was maintained for 10 seconds; at-
omization, 2650°for 8 seconds. The system
was purged with argon at a flow rate of 25
cc/min. Simultaneous deuterium arc back
ground was used throughout. The nickel con
tent of each sample was calculated from a
standard curve. The reliability of the ana
lytical procedure was ascertained with or
chard grass leaves (National Bureau of Stan
dards, SRM-1571) certified to contain 1.3 ±
2 Mg of nickel/g. Our value, 1.22 ±0.16
H/g compared favorably with the certified
value.
Nickel absorption from aqueous solutions.
In all tests of nickel absorption from aqueous
solutions, 22.4 mg of nickel sulfate hexahy-
drate (NiSO4-6H2O, M.W. = 262.86, Fischer
Scientific Co., Fair Lawn, NJ), equivalent to
5 mg of elemental nickel, was dissolved in
100 ml of water. This was administered to
the fasting subjects alone, or with one of the
following chemically-defined substances; 112
mg of phytic acid (J. T. Baker Co., Phillips-
burg, NJ), l g of ascorbic acid (Redoxon®,F.
Hoffman LaRoche & Co., Ltd., Basel, Swit
zerland), 40 mg of NaFeEDTA (Hampshire
Organic Chemicals, Nashua, NH), or 40 mg
of Na2EDTA-2H2O (Sigma Chemical Co.,
St. Louis, MO).
Nickel absorption with beverages and
meals. The 22.4 mg of nickel sulfate was
added to 250 ml of five beverages: whole
cowmilk (Sunnyhurst Farms, Stoughton, MA);
orange juice (Del Sol, Alimentos Naturales,
S. A., San JoséPinula, Guatemala); tea (Su
perior Tea and Coffee Co., Chicago, IL) with
lemon and sugar; black coffee (CaféIncasa-
fuerte, Industrias de Café,S.A., Guatemala
City, Guatemala) with sugar; and Coca Cola®
(Coca Cola Bottling Co., Atlanta, GA). The
same dose of nickel was also added to two
test meals. One was a traditional rural Gua
temalan meal composed of 120 g of black
beans (Phaseolus vulgaris) as a gruel (Ducal,
Alimentos Kern de Guatemala, S.A., Guate
mala City, Guatemala), 120 g of corn tortillas
(obtained from a single indigenous vendor
of "homemade" tortillas in a local market
place) prepared from lime-soaked corn (Zea
mays), and 250 ml of coffee prepared from
3 g of coffee powder (CaféIncasa-fuerte) and
sweetened with 15 g of sucrose. This is similar
to the meal used in previous studies on min
eral absorption (31, 34). The other meal was
an American style breakfast, consisting of 2
pieces of bacon (Toledo, Empacadora To
ledo, S. A., Guatemala City, Guatemala), 2
slices of white bread (Iberia, Panificadora
Iberia, Guatemala City, Guatemala) toast
with margarine (Blue Bonnet, Standard
Brands, N. Y.), 120 g of scrambled eggs and
by guest on July 14, 2011jn.nutrition.orgDownloaded from

42 SOLOMONS ET AL.
250 ml of coffee prepared with 3 g of coffee
powder (CaféIncasa-instantaneo, Industrias
de Café,S. A., Guatemala City, Guatemala),
lightened with non-dairy creamer (Coffee-
mate®,Carnation Co., Los Angeles, CA), and
sweetened with 10 g of sucrose.
Data analysis. To compare the mean rise
in plasma nickel concentration at a given in
terval following the dose of nickel, we used
the Student t test. Relative standard deviation
was calculated as the mean divided by the
standard deviation times 100.
RESULTS
Plasma nickel concentration during fast
ing and after nickel in aqueous solution. The
responses of plasma nickel after ingestion of
5 mg of elemental nickel, as 22.4 mg of
NiSO4-6H2O in water, was monitored over
4 hours in fasting subjects (fig. 1), and the
means changes with respect to the zero time
nickel level were calculated (fig. 2). The sta
bility of plasma nickel concentration in the
fasting state was analyzed during 4 hours of
continuous fasting after an overnight fast in
another group of individuals. For the 1-hour
interval, a cluster of samples was contami
nated and the data were invalid. Otherwise,
plasma nickel concentrations were stable (fig.
2). In a single subject, studied over a 24-hour
g «-
TIME IN HOURS
TIME IN HOURS
Fig. 1 Individual curves of plasma nickel concentra
tion over 4 hours following a dose of 22.4 mg of NiSO< •
6H2O (5 mg of elemental nickel) in 100 ml of water.
Fig. 2 Change in plasma nickel concentration (mean
±SE) following overnight and continued fasting (open
bars) and following ingestion of 22.4 mg of NiSO4-6H2O
(5 mg of elemental nickel) in 100 ml of water (open
bars). The number of uncontaminated specimens (J) at
1 hour was insufficient for inclusion of a mean. Each bar
represents five subjects.
period, we found that the excursion of plasma
nickel and the kinetics of concentration
change were nearly identical (fig. 3) to those
in the previous report by Spruits and Bon-
gaarts (38).
Effects of meals on nickel absorption. We
measured the effects of two test meals on the
absorption of nickel. One meal represented
the typical rural diet consumed in Guatemala
and the other was a typical North American
breakfast. The 22.4 mg of NiSO4-6H2O were
added to the beans and the scrambled eggs
in the respective meals. Curves of change in
plasma nickel concentration were flat after
both meals (fig. 4). The change in plasma
nickel at any interval after either meal was
not significantly different from the change
in fasting subjects. At 1 hour after consump
tion of the nickel-labeled Guatemalan break
fast, however, insufficient samples were
available for inclusion of a mean value.
In five other subjects, plasma nickel con
centrations were monitored over the first 4
postprandial hours after consumption of a
typical Guatemalan meal without added
by guest on July 14, 2011jn.nutrition.orgDownloaded from

BIOAVAILABILITY OF NICKEL IN MAN 43
TIME IN HOURS
Fig. 3 Plasma nickel concentration at intervals over 24 hours in a single subject following a dose of 22.4 mg of
NiSO,-6H2O (5 mg of elemental nickel) in water. The dotted line represents the subject studied by Spruits and
Bongaarts (38). The solid line represents a subject studied in the present study.
nickel. The changes were again similar to nickel response to the Guatemalan meal, with
those during simple fasting (fig. 4) and no or without added nickel.
difference was detected between the plasma Effect of common beverages on nickel
20-
-10-
I Guatemalan meal alone
I Guatemalan meal plus Ni
üf North American breakfast plus Ni
1 HOUR 2 HOURS 3 HOURS 4 HOURS
TIME IN HOURS
Fig. 4 Change in plasma nickel concentration (means ±SE) at 4 hourly intervals following consumption of a
standard Guatemalan meal, with or without 5 mg of nickel, and North American breakfast with 5 mg of nickel.
The number of uncontaminated specimens (J) at 1 hour was insufficient for inclusion of a mean. Dotted line
represents the mean of the change in fasting plasma nickel levels. Each bar represents the mean of five subjects.
by guest on July 14, 2011jn.nutrition.orgDownloaded from

44 SOLOMONS ET AL.
absorption. When the nickel sulfate was
mixed with each of five beverages, plasma
nickel levels rose significantly above fasting
concentrations at hours 2, 3 and 4 post-inges
tion (fig. 5). with respect to the rise in plasma
nickel after nickel sulfate in water, tea re
duced the elevation at 3 hours (P < 0.05),
orange juice and coffee at 2 hours (P < 0.05)
and 3 hours (P < 0.05), and milk at 2 hours
(P< 0.025) and 3 hours (P < 0.05). With
Coca Cola®,on the other hand, the rise in
plasma nickel was not significantly lower at
any time interval that the increment pro
duced by the aqueous solution of nickel
alone.
Effect of phytic acid and ascorbic acid on
nickel absorption. Two chemically-defined
dietary substances that affect the absorption
of a number of trace elements are phytic acid
and ascorbic acid. We tested phytate and
nickel in a 2:1 molar ratio by adding 112 mg
of phytic acid and 5 mg of nickel to 100 ml
of water and allowing it to stand overnight.
Phytic acid did not significantly affect plasma
nickel (fig. 6). When l g of ascorbic acid was
added to the standard dose (fig. 6), the rise
in plasma nickel was significantly suppressed
(P < 0.05) at 2 hours and 3 hours, compared
to nickel in water.
Effects of NaFeEDTA and Na2EDTA-
2H2O on nickel absorption. Uncertainty re
mains regarding the differential biological
behavior of sodium EDTA, a commonly used
food preservative, and sodium iron EDTA,
a substance proposed for iron fortification of
foods for humans. When table sugar was for
tified with NaFeEDTA at 1 ppm in field
trials in four Guatemalan villages, daily con
sumption was 40 mg (Viteri, F. E. Unpub
lished datum). For either form of EDTA, 40
1 HOUR 2 HOURS 4 HOURS
TIME IN HOURS
Fig. 5 Change in plasma nickel concentration (means ±SE) at 4 hourly intervals following ingestion of 22.4 mg
of NiSO4-6H2O (5 mg of elemental nickel) in 250 ml of 5 beverages: cow milk; orange juice; tea; coffee; and Coca
Cola®.The lower border of the shaded area represents the mean of the changes in plasma nickel level after ingestion
of 5 mg of nickel in water. Each bar represents the mean of five subjects.
by guest on July 14, 2011jn.nutrition.orgDownloaded from

BIOAVAILABILITY OF NICKEL IN MAN 45
TIME IN HOURS
Fig. 6 Change in plasma nickel concentration (means
±SE) at 4 hourly intervals following ingestion of 22.4
mg of NiSO4-6H2O (5 mg of elemental nickel) with 1
g of ascorbic acid or 112 mg of phytic acid. The lower
border of the shaded area represents the mean of the
changes in plamsa nickel level after ingestion of 5 mg
of nickel in water. Each bar represents the mean of five
subjects.
mg represents an EDTA:nickel molar ratio
of 1.27:1. We tested each combination in five
subjects. The rise in plasma nickel was re
duced significantly by NaFeEDTA at the 2-
hour interval (P < 0.05) and at all hourly in
tervals by Na2EDTA. At the third and fourth
hours after ingestion of the nickel and
Na2EDTA solution, mean plasma nickel lev
els were actually significantly lower than the
corresponding fast ing levels (P < 0.05).
DISCUSSION
Since the biohazards from radioisotopes of
nickel are prohibitive, and the stable isotope
tracer technology is not yet established, we
exploited the observations of Spruits and
Bongaarts (38)-that a 5 mg oral dose of nickel
produces a detectable rise in plasma concen-
tration-to develop preliminary observations
on factors affecting nickel absorption. This
approach is safe and acceptable; none of our
subjects exhibited allergic or other adverse
reactions or side-effects.
The change-in-plasma-concentration ap
proach was used in early investigations of
iron absorption in humans (45-48), but the
nutritional requirement of the host for iron
was a major determinant variable (49, 50).
The same experimental approach has been
used more recently in studies of zinc absorp
tion (34, 34, 39-44). In table 1, we compared
the variance of the mineral concentrations
in plasma between our data for nickel and
each of six sets of published data for zinc
(24, 29, 42, 43). In terms of relative inter-
individual variance, the change-in-plasma-
concentration method apparently is not more
variable for nickel than for zinc. The ana
lytical determination of plasma nickel, how
ever, is more tedious and complicated than
that for other trace elements, because nickel
is present at relatively low levels and is much
more susceptible to exogenous contamination
during handling and processing. In the single
TIME IN HOURS
Fig. 7 Change in plasma nickel concentration (means
±SE) at 4 hourly intervals following ingestion of 22.4
mg of NiSOj-6H2O (5 mg of elemental nickel) with 40
mg of NaFeEDTA or 40 mg of Na2EDTA-2H2O in 100
ml of water. The lower border of the shaded area rep
resents the mean of the changes in plasma nickel level
after ingestion of 5 mg of nickel in water. The dotted
line represents the mean of the changes in fasting plasma
nickel level. Each experiment represents the mean of
five subjects.
by guest on July 14, 2011jn.nutrition.orgDownloaded from

46 SOLOMONS ET AL.
TABLE 1
Comparative variance in circulating mineral concentrations after aqueous doses
Aqueous
mineral
dose5
mgNi++25
mgZn++25
mgZn+*25
mgZn++50
mgZn*+50
mgZn++50
mg Zn+*2
hours107.0
±19.4'
(18.3%)151.4
±19.4b(12.8%)212
±42(19.8%)—280
±45
(16.1%)336
±20(6.0%)—Time
post-dose3
hours115.0
±25
(21.7%)135.6
±4.8(3.5%)—195
±25(12.8%)——239
±58(24.3%)4
hours104.0
±21.4
(20.5%)120.2
±13.1
(10.8%)——221
±68(30.1%)——No.
of
subjects55681668Reference
sourcePresent
studyréf.
34réf.
39réf.
42réf.
43réf.
39réf.
42
*Means ±SD for nickel concentration in Mg/1. bMeans ±SD for zinc concentration in Mg/dl. The number
in parentheses are the relative standard déviations.
subject whose plasma nickel was followed for
24 hours in response to a 5 mg dose of nickel,
(fig. 3) the curve resembled that recorded by
Spruits and Bongaarts (38). Plasma clearance,
however, apparently differs between nickel
and zinc. Whereas both minerals reached
their maximum plasma increments between
2 and 3 hours postingestion, zinc generally
returned to baseline values by 6 hours (34)
while nickel required more than 24 hours
(fig. 3).
The same qualifications previously noted
for zinc (34, 44) apply to our data for nickel.
These reservations are a) the tracer dose is
outside of the physiological range; b) nickel
clearance and enterohepatic circulation or its
internal redistribution may affect the curves
in unknown ways; and c) the change-in-
plasma-concentration method probably mea
sures the rate of nickel uptake, rather than
net absorption. Nonetheless, consistent and
differential information was produced in dif
ferent experimental situations with this
method. We consider it reasonable to assume
for nickel, as well as for zinc (34), that the
area under the discontinuous curve of plasma
concentration reflects intestinal absorption of
the element.
Background data were not available on
nickel bioavailability in man, so we devised
treatments based on experience with other
divalent trace minerals-iron, zinc, and cop-
per-for which findings regarding intestinal
absorption have recently proliferated. The
absorption of the aforementioned elements
can be depressed by binding- or chelating-
substances, competitive inhibitors, or redox
reagents; absorption is often enhanced by
substances that improve pH, solubility, or
oxidation, or by such chelating agents (e.g.
amino acids) which are actively absorbed.
Those compounds are found, alone or in com
bination, in the various beverages and meals
tested in the present study.
Orange juice contains three potential fac
tors that might affect trace mineral absorp
tion: ascorbic acid, citric acid, and pectins.
Tea, and to a lesser extent coffee, contain
tannins. Tea is a potent inhibitor of iron ab
sorption (51, 52) and coffee inhibits absorp
tion of iron (31) and zinc (34, 36, 42). Cow
milk inhibits the absorption of iron (53) and
zinc (39, 54); the constituents of bovine milk
that were implicated as inhibitors of trace
element absorption are calcium, inorganic
phosphates, and proteins. In the present
by guest on July 14, 2011jn.nutrition.orgDownloaded from

BIOAVAILABILITY OF NICKEL IN MAN 47
study, all four of these beverages limited
nickel uptake by roughly the same amount.
Finally, Coca Cola®,despite its high content
of carbonates and phosphates, apparently
does not affect the absorption of iron or zinc,
and has, in fact, been used as a vehicle for
the administration of tracer doses of these
minerals in human experiments (28, 55).
Coca Cola®was the only one beverage stud
ied that did not inhibit the absorption of
nickel.
On the basis of in vitro chemical studies
(27), Nielsen (28) postulated that phytic acid
might inhibit nickel absorption. The Guate
malan diet is extraordinarily rich in phytate
as well as dietary fiber and calcium, all po
tential inhibitors of the absorption of divalent
cations (34). The test meal used in the present
study is identical to that used previously (31,
34), except that the 40 g of sweet roll was
eliminated. Like iron (31) and zinc (34, 35),
nickel was markedly inhibited in its intestinal
absorption by the typical, rural, Guatemalan
diet.
Two additional observations suggest that
dietary components other than phytates are
responsible for the reduction in nickel ab
sorption. First, the representative North
American breakfast of eggs, bacon, toast,
margarine and coffee-relatively poor in fi
ber, phytates, and calcium, but rich in lipids
and nitrates-dramatically reduced the incre
ment in plasma nickel. Second, in marked
contrast to the experience of Pecoud et al.
(39) with zinc and phytic acid in aqueous
solution, in which a 1:6 phytate:mineral ratio
produced a 40% reduction in plasma zinc
elevation, phytate:nickel molar ratio of 2:1
in aqueous solution here had no effect on
nickel uptake. In fact, phytic acid proved to
be the only chemically-defined dietary con
stituent without inhibitory effects. Several
investigators (34, 39, 56-58) showed that cir
culating zinc concentration declined pro
gressively after meals, but we found no ev
idence for a similar postprandial fall in
plasma nickel levels after a standard, unla-
beled meal.
Ascorbic acid (AA), a natural dietary re
ducing- and acidify ing-agent, profoundly
elevated the absorption of inorganic (non-
heme) iron (59, 60), but depressed the ab
sorption of dietary copper (61). AA did not
affect the absorption of inorganic zinc or co
balt (44). We found a depression of nickel
absorption in the presence of l g of AA. The
Ni: AA molar ratio in the present study is 1.5
X 10~3. If we assume that an ordinary break
fast contains between 100 and 200 fig of
nickel (15), and that 8 oz. of orange juice
contains 100 mg of AA (62), the Ni:AA ratio
in our experimental meal would exceed that
in a routine breakfast. The nickel:ascorbate
interaction should be examined throughout
a wide range of Ni:AA molar ratios to de
termine the nutritional significance of di
etary AA on the bioavailability of nickel.
Our interest in NaFeEDTA stems from the
proposal for its use as an iron-fortifying agent
for the people of Central America (31, 34).
The EDTA, itself, reduced the absorption of
iron (63). When added to animal feeds, how
ever, EDTA improved the bioavailability of
zinc for poultry (64-68) and rats (69).
NaFeEDTA is an effective agent for iron
fortification (31, 70-72) and, in doses of 15
and 40 mg, did not interact with a 25 mg
dose of zinc (34). When we tested the general
safety of NaFeEDTA for man by adding 40
mg to the 5 mg dose of nickel, nickel ab
sorption was suppressed only at 2 hours. The
Na2EDTA-2H2O, however, not only com
pletely attenuated plasma nickel increments,
but also caused a net decrement in circulat
ing nickel levels at 3 and 4 hours. Although
EDTA is commonly used as a preservative
in foods, the precise relevance of this finding
is not immediately evident. The fact that the
competitive action of NaFeEDTA is appar
ently distinct from that of the iron-free che-
late, however, reinforces our notion of the
relative safety of NaFeEDTA as an iron-for
tifying agent for humans.
Since chemical analyses of human diets
consumed in the U. S. have indicated that
the total nickel content approximates the the
oretical requirement, the major determinant
of its nutritional adequacy would be its bio
availability. Additional questions remain re
garding other dietary constituents and pos
sible competitive interactions with other
trace minerals. The present research, how
ever, permitted a preliminary investigation
of dietary factors influencing the absorption
of inorganic nickel by man, and set the stage
for a more complete elucidation of factors
by guest on July 14, 2011jn.nutrition.orgDownloaded from

48 SOLOMONS ET AL.
affecting the bioavailability of nickel in the
human diet. Moreover, the present study was
conducted with pharmacological doses of in
organic nickel. Experimental evaluation of
the absorption of physiological doses of
nickel must await developments in stable
isotope technology. Little is known about the
actual chemical forms of nickel in various
foods or whether dietary nickel has distinct
"organic" forms with enhanced bioavailabil
ity analogous to those of iron (73) and chro
mium (74). If so, the interpretation of the
results requires additional qualifications.
LITERATURE CITED
1. Nielsen, F. H., Myron, D. R., Givand, S. H.,
Zimmerman, T.J. & Ollerich, D. A. (1975) Nickel
deficiency in rats. J. Nutr. Õ05,1620-1630.
2. Schnegg, A. & Kirchgessner, M. (1975) Zur Es-
sentialitätvon Nickel fürTierische Wachstum. Z.
Tierphysiol. Tierernährgu Futtermittelkde 36, 63-
74.
3. Schnegg, A. & Kirchgessner, M. (1975) Verän
derungen des Hamoglobingehaltes, der Erythrozy-
tensahl und des Hamatokrits bei Nickelmangel.
Nutr. Metab. J9, 268-278.
4. Nielsen, F. H., Myron, D. R., Givand, S. H. & Ol
lerich, D. A. (1975) Nickel deficiency and nickel-
rhodium interaction in chicks. J. Nutr. 105, 1607-
1619.
5. Anke, M., Grün,M., Dittrich, G., Groppel, B. &
Hennig, A. (1974) Low nickel rations for growth
and reproduction in pigs. In: Trace Element Metab
olism in Animals (Hoekstra, W. G., Suttie, J. W.,
Ganther, H. E. & Mertz, W., eds.) Vol. 2, pp. 715-
718, University Park Press, Baltimore.
6. Anke, M., Hennig, A., Grün,M., Partschefeld, M.,
Groppel, B., & Ludke, H. (1977) Nickel-ein es-
sentialles Spurenelement. Arch. Tierernährung 27,
25-38.
7. Spears, J. W., Hatfield, E. E., Forbes, R. M. & Ko-
enig, S. E. (1978) Studies on the role of nickel in
the ruminant, J. Nutr. JOS,313-320.
8. Nielsen, F. H., Zimmerman, T. J., Collings, M. E.
& Myron, D. R. (1979) Nickel deprivation in rats:
nickel-iron interaction. J. Nutr. 109, 1623-1632.
9. Schnegg, A. & Kirchgessner, M. (1976) Zur In
teraktion von Nickel mit Eisen, Kupfer und Zink.
Arch. Tierernährung26, 543-549.
10. Spears, J. W., Hatfield, E. E. & Forbes, R. M. (1978)
Interrelationship between nickel and zinc in the rat.
J. Nutr. 108, 307-312.
11. Spears,]. W., Hatfield, E. E. & Forbes. R. M. (1977)
Nickel-copper interrelationship in the rat. Proc. Soc.
Exp. Biol. Med. 156, 140-143.
12. Sunderman, F. W., Jr., Nomoto, S. & Nechay, M.
(1971) Nickel metabolism in myocardial infarction.
2. Measurement of nickel in human tissues. In: Trace
Substances in Environmental Health. IV. Hemphill,
D. D., ed.) Vol. 4, pp. 352-356, University of Mis
souri Press, Columbia.
13. Casey, C. E. & Robinson, M. F. (1978) Copper,
manganese, zinc, nickel, cadmium, and lead in hu
man foetal tissues. Br. J. Nutr. 39, 639-646.
14. Schroeder, H. A. & Nason, A. P. (1971) Trace-
element analysis in clinical chemistry. Clin. Chem.
17, 461-474.
15. Schroeder, H. A., Balassa, J. J. & Tipton, I. H.
(1962) Abnormal trace metals in man: nickel, j.
Chronic Dis. J5, 51-65.
16. Furr, A. K., MacDaniels, L. H., St. John, L. E. Jr.,
Gutenmann, W. H., Pakkala, 1. S. & Lisk, D. J.
(1979) Elemental composition of tree nuts. Bull.
Environm. Contam. Toxicol. 21, 392-396.
17. Thomas, B., Roughan, J. A. &Watters, E. D. (1974)
Cobalt, chromium, and nickel content of some veg
etable foodstuffs. J. Sci. Food Agr. 25, 771-776.
18. Evans, W. H., Read, 1. & Lucas, B. E. (1978)
Evaluation of a method for determination of total
cadmium, lead, and nickel in foodstuffs using
measurement by flame atomic absorption
spectrophotometry. Analyst 103, 580-594.
19. Dencker, I., Fristedt, B. & Ovrum, P. (1970) Jam,
kopper, krom, nickel, bly och kobolt i föda.Läk-
artidningen 68, 4027-4040.
20. Ellen, G., Van den Bosch-Tibbesma, G. & Douma,
F.F. (1978) Nickel content of various Dutch food
stuffs. Z. Lebensm. Unters. Fursh. J66, 145-147.
21. Rudzki, E. & Grzywa, Z. (1977) Exacerbation of
nickel dermatitis by margarine. Contact Derm. 3,
344.
22. Myron, D. R., Zimmerman, T. J., Shuler, T. R., Kle-
vay, L. M., Lee, D. E. & Nielsen, F. H. (1978)
Intake of nickel and vanadium by humans. A survey
of selected diets. Am. J. Clin. Nutr. 31, 527-531.
23. Nielsen, F. H. (1980) Evidence for the essentiality
of arsenic, nickel, and vanadium and their possible
nutritional significance. In: Advances in Nutritional
Research (Draper, H. H., ed.) vol. 3, pp. 137-172,
Plenum, New York.
24. Kaaber, K., Veien, & Tjell, J. C. (1978) Low nickel
diet in the treatment of patients with chronic nickel
dermatitis. Brit. J. Dermatol. 98, 197-201.
25. Christensen, O. B. & Möller,H. (1975) External
and internal exposure to the antigen in the hand
eczema of nickel allergy. Contact Derm. 1, 136-141.
26. Horak, S. E. & Sunderman, F. W. (1972) Fecal
nickel excretion by healthy adults. Clin. Chem. 19,
429-430.
27. Vohra, P., Gray, G. A. & Kratzer, F. H. (1965)
Phytate-acid-metal complex. Proc. Soc. Exp. Biol.
Med. 120, 447-449.
28. Nielsen, F. H. (1976) Newer trace elements and
possible application in man. In: Trace Elements in
Human Health and Disease. Essential and Toxic
Elements. (Prasad, A. S., ed.) vol. 2, pp. 379-399,
Academic Press, New York.
29. Sayer, M. H., Lynch, S. R., Jacobs, P., Charlton,
R. W., Bothwell, T. H., Walker, R. B. & Mayet, F.
(1973) The effects of ascorbic acid supplementation
on the absorption of iron in maize, wheat, and soya.
Br. J. Haematol. 24, 209-218.
30. Cook, J. D. (1977) Absorption of food iron. Fed
eration Proc. 36, 2028-2032.
31. Viteri, F. E., GarcÃ-a-Ibáñez,R. & Torún,B. (1978)
by guest on July 14, 2011jn.nutrition.orgDownloaded from

BIOAVAILABILITY OF NICKEL IN MAN 49
Sodium iron NaFeEDTA as an iron fortification
compound in Central America. Absorption studies.
Am. J. Clin. Nutr. 31, 961-971.
32. Reinhold, J. G., Faraji, B., Abadi, P. & Ismail-Beigi,
F. (1976) Decreased absorption of calcium, mag
nesium, zinc and phosphorus by humans due to in
creased fiber and phosphorus consumption as wheat
bread. J. Nutr. 106, 493-503.
33. Ismail-Beigi, F., Reinhold, J. G., Faraji, B. & Abadi,
P. (1977) The effect of cellulose added to diets of
low and high fiber content upon the metabolism of
calcium, magnesium, zinc and phosphorus by man.
J. Nutr. 107, 510-518.
34. Solomons, N. W., Jacob, R. A., Pineda, O. & Viteri,
F. E. (1979) Studies on the bioavailability of zinc
in man. Effects of the Guatemalan rural diet and
of the iron-fortifying agent, NaFeEDTA. J. Nutr.
109, 1519-1528.
35. Solomons, N. W., Jacob, R. A., Pineda, O. & Viteri,
F. E. (1979) Studies on the bioavailability of zinc
in man. 2. Absorption of zinc from organic and in
organic sources. J. Lab. Clin. Med. 94, 335-343.
36. Onkelinx, C., Becker, J. & Sunderman, F. W. Jr.
(1973) Compartmental analysis of the metabolism
of ^Niill) in rats and rabbits. Res. Commun. Chem.
Pathol. Pharmacol. 6, 663-675.
37. Onkelinx, C. (1977) Whole-body kinetics of metal
salts in rats. In: Clinical Chemistry and Chemical
Toxicology of Metals (Brown, S. S., éd.)pp. 37-40,
Elsevier, North Holland, Amsterdam.
38. Spruits, D. & Bongaarts, P. J. M. (1977) Nickel
content of plasma, urine, and hair in contact der
matitis. Dermatologica 154, 291-300.
39. Pecoud, A., Donzel, P. & Schelling, J. L. (1975)
Effects of foodstuffs on the absorption of zinc sulfate.
Clin. Pharmacol. Ther. 17, 469-474.
40. Schelling, J. L., Muller-Hess, S. & Thonney, F.
(1973) The effect of food on zinc absorption. Lan
cet 2, 968-969.
41. Andersson, K. E., Bratt, L., Dencker, H. & Lanner,
E. (1976) Some aspects of intestinal absorption of
zinc in man. Eur. J. Clin. Pharmacol. 9, 423-428.
42. Oelshlegel, F. J. & Brewer, G. J. (1977) Absorption
of pharmacological doses of zinc. In: Zinc Metabo
lism: Current Aspects in Health and Disease (Brewer,
G. J. & Prasad, A. S., eds.), pp. 299-311, Alan R.
Liss, Inc., New York.
43. Sullivan, J. F., Jetton, M. & Burch, R. E. (1978)
A zinc tolerance test. J. Lab. Clin. Med. 93, 485-
492.
44. Solomons, N. W., Jacob, R. A., Pineda, O. & Viteri,
F. E. (1979) Studies on the bioavailability of zinc
in man. 3. Effects of ascorbic acid on zinc absorption.
Am. J. Clin. Nutr. 32, 2495-2499.
45. Crawley.J. (1952) Iron absorption test in anaemia:
the use of intravenous iron preparations. Edinburgh
Med. J. 59, 478-497.
46. Henley, E. D., Christenson, W. N., Grace, W. J. &
Wolff, H. (1956) Absorption of iron from the gas
trointestinal tract. Am. J. Clin. Nutr. 4, 609-618.
47. Joseph, H. (1958) Absorption of iron as a problem
in human physiology. Blood 13, 1-54.
48. Hallberg, L. & Sovell, L. (1960) Absorption of a
single dose of iron in man. Acta Med. Scand. (Suppl.
358) 168, 19-42.
49. Gross, S. J., Stuart, M. J., Swender, P. T. & Oski,
F. A. (1976) Malabsorption of iron in children
with iron deficiency. J. Pediat. 88, 795-799.
50. Massa, E., MacLean, W. C. Jr., de Romana, G. L.,
de Martinez, Y. & Graham, G. G. (1978) Oral iron
absorption in infantile protein-energy malnutrition.
J. Pediat. 93, 1045-1049.
51. Disler, P. B., Lynch, S. R., Charlton, R. W., Terr
anee, J. D., Bothwell, T. W., Walker, R. B. & Mayet,
F. (1975) The effect of tea on iron absorption. Gut
16, 193-200.
52. de Alarcon, P. A., Donavan, M. E., Forbes, G. B.,
Landaw, S. A. & Stockman, J. A. HI. (1979) Iron
absorption in the thalassemia syndrome and its in
hibition by tea. N. Engl. J. Med. 300, 5-8.
53. Schulz, J. & Smith, N. J. (1958) A quantitative
study of the absorption of food iron in infants and
children. Am. J. Dis. Child. 45, 109-119.
55. Solomons, N. W. & Jacob, R. A. (1981) Studies on
the bioavailability of zinc in man. Effect of heme
and nonheme iron on the absorption of inorganic
zinc. Am. J. Clin. Nutr. 34, 475-481.
56. McBean, L. & Halsted.J. A. (1969) Fasting versus
postprandial plasma zinc level. J. Clin. Pathol. 22,
623.
57. Hetland, O. & Brubakk, E. (1973) Diurnal vari
ation in serum zinc concentration. Scand. J. Clin.
Lab. Invest. 32, 225-226.
58. Walker, B. E., Dawson, J. B., Keller, J. & Losowsky,
S. (1973) Plasma and urinary zinc in patients with
malabsorption syndrome and hepatic cirrhosis. Gut
14, 943-948.
59. Brise, H. & Hallberg, L. (1962) Effect of ascorbic
acid on iron absorption. Acta Med. Scand. (Suppl
376) 171, 51-58.
60. Monsen, E. R., Hallberg, L., Layrisse, M., Hegsted,
D. M.,Cook,J. D.,Mertz,W. & Finch, C. A. (1978)
Estimation of available dietary iron. Am. J. Clin.
Nutr. 31, 134-141.
61. Van Campen, D. & Gross, E. (1968) Influence of
ascorbic acid on the absorption of copper by rats.
J. Nutr. 95, 617-622.
61. Tablas de Composición de Alimentos por Uso en
America Latina (1961) p. 16, Institute of Nutrition
of Central America and Panama, Guatemala City.
63. Cook, J. D. & Monsen, E. R. (1976) Food iron
absorption in man. 2. The effect of EDTA on ab
sorption of dietary nonheme iron. Am. J. Clin. Nutr.
29, 614-620.
64. Morrison, A. B. & Sarrett, H. P. (1958) Studies on
zinc deficiency in the chick. J. Nutr. 65, 267-280.
65. Savage, J. E., Yohe, J. M., Pickett, E. E. & O'Dell,
B. L. (1964) Zinc metabolism in the growing
chick: tissue concentration and effect of phytate on
absorption. Poultry Sci. 43, 420-426.
66. O'Dell, B. L., Yohe, J. M. & Savage, J. E. (1964)
Zinc availability in the chick as affected by phytate,
calcium and ethylenediamine-tertraacetate. Poultry
Sci. 43, 415-419.
67. Kratzer, F. H., Allred, J. B., Davis, P. N., Marshall,
by guest on July 14, 2011jn.nutrition.orgDownloaded from

50 SOLOMONS ET AL.
B. J. & Vohra, P. (1959) The effect of autoclaving
soybean protein and the addition of EDTA acid on
the biological availability of dietary zinc for turkey
poults. J. Nutr. 68, 313-322.
68. Davis, R. M, Norris, L. C. & Kratzer, F. H. (1962)
Interference of soybean protein with the utilization
of trace minerals. J. Nutr. 77, 217-223.
69. Forbes, R. M. & Yohe, J. M. (1960) Zinc require
ments and balance studies with the rat. J. Nutr. 70,
53-57.
70. Garby, L. & Areekul, S. (1974) Iron supplemen
tation in Thai fish-sauce. Ann. Trop. Med. Parásito!.
68, 467-476.
71. Layrisse, M. & Martinez-Torres, C. (1977) Fe(HI)-
EDTA complex as iron fortification. Am. J. Clin.
Nutr. 30, 1166-1174.
72. Martinez-Torres, C., Romanao, E. L., Renzi, M. &
Layrisse, M. (1979) Fe(III)-EDTA complex as iron
fortification. Further studies. Am. J. Clin. Nutr. 32,
809-816.
73. Hallberg, L. & Björn-Rasmussen, E. (1972) De
termination of iron absorption from whole diet—A
new two-pool model using two radioiron isotopes
given as haem and nonhaem iron. Scand. J. Hae-
matol. 9, 193-197.
74. Mertz, W., Toepfer, E. W., Roginski, E. E. & Po-
lansky, M. M. (1974) Present knowledge of the
role of chromium. Federation Proc. 33, 2275-2280.
by guest on July 14, 2011jn.nutrition.orgDownloaded from