Influence of Fish on the Bioavailability of

Plant Iron in the Anemic Rat1'2

LUCIA S. CHAO AND DENNIS T. GORDON34
Seafoods Laboratory, Oregon State University,
250-36th Street, Astoria, OR 97103

level at time of dosing. Diet composition did not appear to have the same effect on the
percentage of 59Fe retained after 110 hours by the rat as compared to levels of he
moglobin regeneration (i.e., RBV). A "meat factor" effect was not shown by substituting
fish for casein in the diets containing
113: 1643-1652, 1983.

plant iron sources fed anemic rats.

INDEXING KEY WORDS

iron bioavailability
bran spinach
soy protein isolate

Numerous studies have demonstrated differences in bioavailability of iron in food and

iron supplements. Heme iron in animal flesh
is highly assimilated (1, 2), compared to nonheme iron in plants (3-5). The low utilization
of plant iron has been attributed, in part, to
the presence of various plant components
such as phytate (6), fiber (7) and tannic acid
(8). Absorption of nonheme iron is improved

meat factor

fish

J. Nutr.
wheat

health-related reasons (14), the latest to obtain more dietary w3 fatty acids (15). Substitution of fish for red meat in the diet would
reduce intake of both heme and total iron
(16, 17). However, decreased intake of iron
might be offset by enhancement of absorption of nonheme iron from the diet by fish
protein. Limited studies suggest this may be
true in humans (18, 19).

by ascorbic acid (9) and animal protein (10,

11). The latter enhancing
effect has been

For long-term

objectives

dealing

with the

demonstrated only in humans, although the

mechanism
,
.
,
Of the
factors

'f3
American Institute of Nutrition Received for publication U
December 1982
-Presented in par. at the meeting of the American Institute of Nutrition,

S not
that

absorption

could

anemia

known.
.
enhance

An

Understanding
and

inhibit

help eliminate

products
r r

do

not

iron

iron dfi- Anaheim,

CA,Apri,
i9soFedProc39;1044

, . ~>
(1).

* This work was supported in part by US. Department of Congress, Na|ona|Oceanic and Atmospheric Administration. Institutional Sea Grant 04-

OCCUpy
I

large

proportion of the American diet (13). neeommendations have been made to increase
the

Consumption

O fish

for

number

1643

7-158-440SS.the NationalFisheriesInstitute,Ine and a grant to one of the

authors (D.T.G.) from the Nutrition Research Instituir, Oregon Slate Uni-

H',rTrJSJ^S2
Hall. University of Missouri-Columbia, Columbia, MO 65211
'To whom correspondence should be addiessed.

Downloaded from jn.nutrition.org by guest on September 10, 2014

ABSTRACT
The effect of fish protein and fish oil on the utilization of endogenous
iron in wheat bran, spinach and soy protein isolate was investigated by using the anemic
rat as an animal model. Marine products were substituted for casein and corn oil in the
diets of these animals. Hemoglobin regeneration was one criteria used to measure iron
uptake. Relative biological values (RBV) were computed from a regression equation
obtained from control animals receiving graded levels of FeSO4
7H2O. The RBV of
iron from plant sources provided in diets containing casein-corn oil versus fish-fish oil
were: wheat bran, 123 vs. 111%; spinach, 53 vs. 49%; and soybean isolate, 84 vs. 67%;
RBV FeSO, = 100%. These changes were not significant. The decreases in iron absorption
from diets containing marine products was attributed to the fish oil. Absorption of
exogenous iron (59Fe) was measured in the same animals after the 14-day repletion
period. Assimilation of the 59Fe was highly correlated (r2 = 0.958) with hemoglobin

1644

CHAO AND GORDON

mechanism

of action of the "meat

factor"

effect, the rat was evaluated as an animal

model. This study reports the effect of sub
stituting fish and fish oil for casein and corn
011on the bioavailability of endogenous iron
from wheat bran, spinach and soy protein
isolate by anemic rats. Absorption of exoge
nous iron (59Fe) from animal groups consum
ing these diets was also determined.
MATERIALS AND METHODS

AACC Certified Food Grade Whe.1 Bran R07-3691. letter AACC dated
December 7. 1976.

Downloaded from jn.nutrition.org by guest on September 10, 2014

Dietary components.
American Associa
tion Cereal Chemists (AACC) wheat bran5
(20) was ground in a Thomas-Wiley
Labo
ratory Mill (Model 4, Arthur H. Thomas Co.,
Philadelphia, PA) to pass a 1.00-mm screen.
Freeze-dried
spinach was a commercial
product provided by Oregon Freeze Dry
Foods, Inc., Albany, OR, and ground in the
same manner as the wheat bran. Soy protein
isolate (SUPRO 610) was provided by the
Ralston Purina Co., St. Louis, MO. Turbot,
Atheresthes stomias, obtained fresh, was fil
leted, freeze-dried,
stored under N2> and
ground prior to incorporation into test diets.
Because of the residual oil in the freeze-dried
fillets, additional fish oil was added to obtain
uniform lipid levels and composition in the
marine product diets. Additional fillets were
cooked in a steam-jacketed
kettle with dis
tilled water. Butylated hydroxytoluene (BHT)
and butylated hydroxyanisole
(BHA) were
added during cooking to give a concentration
for each of 0.005% in the oil based on theo
retical yield. The oil was removed, washed
once with distilled water and centrifuged. An
additional 0.005% of each antioxidant was
added to the finished oil. Compositional in
formation about the main dietary ingredients
is given in table 1.
Animals
and dietsrepletion
studies.
Weanling male Long-Evans rats, 23 days old,
weight 39 6 g, were obtained from Charles
River Breeding Laboratories
(North Wil
mington, MA). Animals were housed in stain
less-steel cages with wire-mesh floors. Light
ing was regulated automatically
to provide
12 hours of light and 12 hours of darkness.
Food and glass-distilled water were provided
ad libitum. These animals were maintained
on a low iron diet (7 jug Fe/g) for 33 days.
Rats were then divided into groups (seven

animals per group) having uniform hemo

globin levels (4.6 1.0 g/dl), and body weight
was recorded. Blood taken from the tail of
the rat was analyzed for hemoglobin by the
cyanmethemoglobin
method (21). After a 14day repletion period on test diets, hemoglo
bin level and body weight were again mea
sured. A negative control group (diet No. 1)
of animals was maintained through the re
pletion period, and a positive control group
(diet No. 13) of rats (35 Mg Fe/g diet; FeSO4)
was maintained through the depletion and
repletion periods. An additional group of an
imals was given a fish diet without added
iron (diet No. 12). The hemoglobin level in
the positive control group of rats (diet No.
13) at the start of the repletion period was
13.3 0.4 g/dl.
Composition of the control diet (diet No.
1) and substitutions made in preparing test
diets are reported in table 2. Test diets Nos.
5-9 were first formulated to provide 25 ng
endogenous plant iron per gram diet. Casein
or freeze-dried fish were added to provide
17.1% dietary protein (Kjeldahl N X 6.25).
Casein was selected as the control protein
because of its high nutritional quality, and
it was considered to be an appropriate stand
ard to compare marine protein in meeting
the objectives of this study. Corn oil or fish
oil was added to appropriate diets to provide
a final concentration of 5%. Diets Nos. 10 and
11 contained one-half the soy protein in diet
No. 7 and had a calculated iron content of
12.5 Mg/g- Adjustments to provide 17.1%
protein and 5% lipid in these two diets were
as previously described. All diets were sup
plemented with vitamins D, E and K as rec
ommended
(22). Animals were provided
fresh diet daily.
Relative biological values (RBV). Two cri
teria were used to calculate the RBV of en
dogenous iron in test diets. These were, 1)
change in hemoglobin and 2) the difference
in the products, final hemoglobin times final
weight minus initial hemoglobin times initial
weight during the repletion period and are
reported as RBV-1 and RBV-2, respectively.
Using these two measured response param
eters in animals fed control diets with in-

1645

FISH AND PLANT IRON BIOAVAILABILITY

TABLE 1
Protein, lipids, oxalate, phytate and iron concentration of major dietary test ingredients

bran2Spinach,
Wheat
driedSoy freeze
isolate3Fish,
protein
freeze dried4Protein115.132.890.173.1Lipid4.58.60.326.9Oxalate%014.00Phytate3.410.221.28IronV-glg11916217022
1 Kjeldahl N X 6.25. 2 AACC wheat bran.
arrowtooth flounder (Atheresthes stomias).

3 Supro 610, Ralston Purina Co., St. Louis, MO.

RBVs. The response of each animal was used

in calculating mean RBV and for analysis of
variance.
Analytical methods. Dietary ingredients
and diets were wet digested in HNO3-HClO4
then analyzed for iron by atomic absorption
spectrophotometry (16). Protein was deter
mined by the micro-Kjeldahl method (23).
Lipid was gravimetrically quantified after

2Diet
composition1Freeze-

Diet
no.

DietD-Glucose

monoCasein
hydrateTABLE

12345678910111213Control-negativeControl
FeControl+ 6.25 ppm
FeControl+ 12.5 ppm
FeWheat4- 25.0 ppm
caseinWheat
bran +
fishSpinach
bran +
caseinSpinach
+
fishSoy +
isolateSoy
protein
50%+
protein
50%Soy
casein
50%+
protein
50%FishControl
fish

Corn
oil

dried
fishFish oilWheatbran

protein
SpinachSoy isolate

21.0023.44

Downloaded from jn.nutrition.org by guest on September 10, 2014

creasing levels of iron (i.e., 6.25; 12.5 and

25.0 itg Fe/g diet) supplied as FeSO4
7H2O,
regression equations were calculated by the
method of least squares. Each response cri
teria observed in animals fed a test diet was
substituted in the appropriate regression
equation and the corresponding iron level
was determined. Dividing this value by the
added iron level in the diet X 100 provided

4 Turbot:

21.00

15.0016.71
15.000.50

14.39

23.44 1.15

14.707.357.35
+ 35.0 ppm Fe68.50685068.5068.5052.1049.0761.0461.2969.2668.8870.6270.0768.5020.0020.0020.0020.0016.3514.264.5512.2

1Diets contained in addition: mineral mix 5.0% (salts from J. T. Baker Chemical Co., Phillipsburg, NJ) containing
in grams/100 g diet: CaHPO4, 1.765; NaCl, 0.259; K-citrate, 0.734; K2SO4, 0.174; MgSO4, 0.248; glucose H2O,
1.787; in milligrams/100 g diet; ZnSO4
7H2O, 13.193; MnSO4
H2O, 16.616; CuSO4, 2.358; KIO3, 0.034; Na2SeO3,
0.022; CrK(SO4)-12H2O, 1.090; and miamin premii 0.70% (obtained from U.S. Biochemical Corp, Cleveland,
OH), containing in milligrams/100 g diet (except as noted): thiamin
HC1, 0.7; riboflavin, 0.7; pyridoxine, 0.7;
nicotinic acid, 3.15; D-calcium pantothenate, 2.1; folie acid, 0.063; D-biotin, 0.014, cyanocobalamin, 0.945 jig; retinyl
palmitate, 630 ID;calciferol, 70 lU; menaquinone, 1.575 g; diets further supplemented with: cholecalciferol, 30 IU;
DL-o-tocopherol acetate, 1.5 IU; menadione (vitamin K oil soluble); 3.425 /jg; (all obtained from U.S. Biochemical
Corp.); choline-CI 0.50% and DL-methionine 0.30%.

1646

CHAO AND GORDON

portion of the experiment, six animals from

seven groups (diets Nos. 5, 6, 7, 8, 10, 11, 14)
were maintained on their test diets for an
other 3 days. Subsequently, after a 12-hour
fast, rats were given 7 g of diet containing
1.4 /iCi 59Fe (Ferrous sulfate, 20 tCi/ml in
0.1 M suif uric acid, specific activity 26.4
mCi/mg,
New England Nuclear, Boston,
MA). One millilitri- of the 59Fe solution was
added to 100 g of each diet, the latter being
thoroughly mixed. Whole-body radioactivity
was measured with a model 446 Armac
Whole-Body
Liquid Scintillation
Detector
assembled by Packard Instrument Co. (Des
Plaines, IL). Counting was initiated within
2 hours after consumption of the 59Fe and
repeated at approximately
24-hour intervals
until a plateau level was reached (110 hours).
The experiment was terminated and hemo
globin, liver weight and liver 59Fe levels were
determined.
RESULTS
Sources of plant iron were selected on the
basis of their high content of this element
(table 1). Animals consuming wheat bran had
the highest weight gain and total food con
sumption (table 3) compared to other groups.
Weight gain and food consumption were ap
proximately
equal among control animals
receiving graded levels of FeSO4 and animals
consuming spinach and soy protein diets. The
average increase in hemoglobin level was also
highest in animals on wheat bran diets. The
lowest increase in body weight and food con
sumption was observed in the control group,

with no added Fe (diet No. 1) and rats main

tained on a fish diet (diet No. 12). Although
the fish provided a small amount of addi
tional iron to the diet above the control, its
contribution to hemoglobin regeneration
in
the test diets was considered negligible. Sub
stituting fish and fish oil in the plant product
diets for casein and corn oil did not appear
to cause a palatability problem based on food
consumption figures (table 3).
A regression equation (Y = 0.256
[Fe]
1.723; where [Fe] is added iron in micrograms/gram;
r2 = 0.814) was obtained from
the change in hemoglobin concentration ver
sus iron content in control diets (diets Nos.
2, 3, 4). From this equation, the RBV-1 (table
4) of endogenous plant iron was calculated.
The RBV-1 for wheat bran iron was signif
icantly higher than the reference, FeSO4.
The RBV-1 for spinach and soy protein was
significantly lower. For each plant iron source,
substituting fish for casein in the diets re
sulted in a small but significant decrease in
RBV-1. Decreases in RBV-1 when fish and
fish oil were substituted for casein and corn
oil in diets containing wheat bran, spinach
and soy protein were 10, 8 and 20%, respec
tively.
Hemoglobin
concentrations
are also af
fected by body weight (29). This parameter
was considered in calculating RBV-2. The
regression equation of final hemoglobin times
final weight minus initial hemoglobin times
initial weight for the repletion period versus
iron content (Y = 18.90
[Fe] - 135.55; where
[Fe] is added iron in micrograms/gram)
for
control diets (diets Nos. 2, 3, 4) also had a
high correlation coefficient (r2 = 0.876). With
the wheat bran diet containing fish and fish
oil (diet No. 6), the RBV-2 of wheat bran
iron was 20% higher compared to the value
obtained by considering change in hemoglo
bin only (i.e., RBV-1). Similar increases were
not seen with the other test diets upon cal
culating the RBV-2. Wheat bran and spinach
diets containing fish and fish oil had slightly
higher RBV-2 compared to the plant diets
with casein and corn oil. Fish and fish oil
decreased bioavailability
of iron from soy
protein.
The RBV-2 of iron in soy protein when fed
at a dietary level of 14.7% (diet No. 9) was
slightly higher than when incorporated
at

Downloaded from jn.nutrition.org by guest on September 10, 2014

extraction by the method of Folch et al. (24).

The procedure outlined by Franco and Krinitz (25) was used to measure oxalate. Phytate
was extracted as described by de Boland et
al. (26). Phytic acid was calculated assuming
it contains 28.2% phosphorus.
Phosphorus
concentration
was determined
colorimetrically by the method of Fiske and SubbaRow
(27) after sample digestion in HNO3-HC1O4
(16). Oxalate and phytic acid levels were
determined
as part of a companion study
dealing with effect of these compounds on
iron absorption (Gordon, D. T. and Chao,
L. S., unpublished data).
Absorption and retention of radioactive
iron (S9Fe). Upon completion of the repletion

1647

FISH AND PLANT IRON BIOAVAILABILITY

TABLE 3
Dietary iron concentration, initial body weight, weight gain, hemoglobin change and total food
consumption during 14-day iron repletion period1
weight4Initiala118
concentrationDietControlControlControlControlWheat
Diet
no.12345678910111213Iron

changeg/dl-2.0

food
consumptiong130

1Values are means SDfor seven animals.

2 Values not followed by the same superscript letter are significantly
different (P < 0.05). 3 Mean value of duplicate determinations.
4 At start and completion of repletion period,
day 56 and 70 (animal age), respectively.
5 Not included in statistical analysis.

one-half this amount (diet No. 10). The dif

ference was not significant. The average
RBV-2 of iron found in wheat bran (diets
Nos. 5, 6), spinach (Diets Nos. 7, 8) and soy
TABLE 4
Relative biological value (RBV) of plant iron to the
anemic rat with different sources of
dietary protein and lipid1-2
Diet
no.5

13124

67

bran + casein
fishSpinach
Wheat bran +

89

+ casein
Spinach
fishSoy +

protein
Soy protein + casein
10
Soy protein + fishRBV11DietWheat

19"
111
14"53

15b>
14b-92
49 *

SOD
ISI
23"45

9"""48

lld"86

26b
19b81

84 44'b
35*
70 34bcd'
67 39b<RBV-23%125

1 Means so of seven animals per group.

2 Values not fol

lowed by the same superscript letter are significantly different

(P < 0.05).
3 RBV values are based on those observed in an
imals consuming a casein, cellulose-free diet supplemented with
FeSO<: RBV FeSO4 = 100%. RBV-1, change in hemoglobin
(Hb); RBV-2, final Hb X final wt - initial Hb X initial
wt.
* Significantly different from 100%.

protein (diets Nos. 10, 11) were 118, 51 and

76%, respectively. The correlation between
the two methods of calculating RBV was r2
= 0.904 (P < 0.0001). In the rat, true iron
absorption is estimated to be approximately
one-half that reported as RBV (29).
Animals on wheat bran diets (diets Nos. 5,
6) continued to grow more rapidly during
the second part of the experiment during
which extrinsically labeled iron (59Fe) was
administered (table 5). Food consumption
was higher for animals consuming wheat
bran. Hemoglobin levels did not change from
day 70, the end of the repletion period, to
day 78, the termination of the experiment
for animals on the wheat bran diets.
The percent 59Fe retained by animals con
suming seven different diets is illustrated in
figure 1. Except for the soy protein fish diet
(diet No. 11) and control + 35 ppm Fe (diet
No. 13), 59Fe retention from the other diets
began to plateau after 86 hours. Whole-body
retention of 59Fe after 110 hours was used as
the quantitative measurement of iron ab
sorption. These values ranged from 54 to
80%. The absorption of 59Fe was highly cor
related (r2 = 0.958) with hemoglobin levels
at time of dosing (fig. 2). These results do not

Downloaded from jn.nutrition.org by guest on September 10, 2014

12d175

0.8f-0.3

llf52
13ab125

25C172

13de55

0.8e1.8

25ab109

19207

ig01*675

l.lc4.6

16ab101

26b250

l.lb6.2

22b131

llab73

22a238

SO"1*90

1.2"5.3

26a121

caseinWheat
bran +
12a171

0.9a151.7

6a49

22ab111

fishSpinach
bran +
l.O"11.4

35C171

17de64

27ab105

caseinSpinach
+
35193
+
llbcd71

O.y^4.2

14ab107

fishSoy +
31cb179

22bc55

1.7b1.0

20ab108

proteinSoy
28cb186

lA^0.4

29"b102

IO01"866

caseinSoy
protein +
13cb144

1.3de-2.5

15b111

9M43

fishFishControl5AddedVg/g06.2512.525.025.025.025.025.025.012.512.55.235.0Actual3diet7.012.718.030.332.033.925.028.126.214.816.39.445.
protein +
1.0*1.3

18d246

15ab202

7e46
35
6Gainr20 17Hemoglobin 0.8Total

1648

CHAO AND GORDON

TABLE 5
Body weight, hemoglobin and food consumption prior to and after ^Fe administration1'2
wtDay

Diet
no.

DietBody

703Day

784HemoglobinDay
703Day

food
consumption5
784Total
8 days

g/100 ml
10.8 0.5
10.5 0.9a
9.8 0.3b
9.9 0.8"

5
6

Wheat bran + casein

Wheat bran + fish

207 29a
211 25"

246 24"
252 + 25"

7
8

Spinach + casein
Spinach + fish

161 33b
170 17b

199 26b
187 + 22b

6.3 0.3
6.1 0.7C

6.4 0.6b
6.0 0.8b

129 23b
120 llb

10
11

Soy protein + casein

Soy protein + fish

166 29b
166 llb

197 26b
196 13b

5.6 0.91
5.1 0.4d

4.9 1.0e
4.4 0.8

116 18b
126 20b

13

Control + 35 ppm Fe6

254 14

279 15

14.6 0.8

14.1 0.6

168 16"
168 25"

141 14

agree with RBVs determined by hemoglobin

regeneration (table 4). Animals on test diets
having the highest hemoglobin (i.e., wheat
bran) had the lowest retention of 59Fe. The
control group of animals having 35 ppm iron
in their diet throughout the study (55 days),
retained 62% of ingested 59Fe. This level was
not significantly different than the absorption
levels observed with the wheat bran diets.
The amount of 59Fe retained in the whole
body of the rats did correspond with the lev
els of 59Fe determined in livers (table 6).
DISCUSSION

The utilization of iron from the three foods

examined in this study has been previously
reported by other investigators with slightly
different RBV. Morris and Ellis found an
RBV of 98% for iron in wheat bran (30). In
a previous study (31) by the same authors,
the monoferric phytate isolated from wheat
bran had an RBV equivalent to that of the
reference standard. The earlier study (31)
appears to be the first report that suggests
that a phytic acid complex of iron does not
materially affect absorption of this element.
It is still debatable to what extent phytic acid
may interfere with iron absorption (32). The
high RBV of iron in wheat bran observed in
the authors study was attributed to growth
rate. Use of the repletion method for mea
suring the bioavailability of endogenous iron

in foods versus exogenous iron supplements

for which the hemoglobin regeneration assay
was originally designed (33) does present
some problem of interpretation when the
growth rate differs. There appears to be
something unique about wheat bran regard
ing growth. Fiber has been shown to stimu
late growth in different animals (34-36).
Feeding spinach containing intrinsically
labeled 59Fe to nonanemic rats, Van Campen
and Welch (37) found ca. 74% of the ad
ministered dose to be retained. Similar levels
of 59Fe were absorbed by animals given 59Fe
oxalate or 59Fe chloride. These high absorp
tion values may be due to the use of radio
active iron (i.e., extrinsic label technique)
employed to test bioavailability. Iron bioavailability from spinach is considered to be
low (4) and this study confirms these obser
vations.
Steinke and Hopkins (38), evaluated the
same protein isolate (i.e., Supro 610) and
found an RBV of 67% for its endogenous iron
by the slope ratio method. Casein was added
at the expense of isolated soy protein to ob
tain the desired iron levels. The RBV-1 of
92% observed in the present study for iron
in the soy protein isolate was reduced to 84
and 67% when it was diluted one-half in diets
with casein and fish, respectively. That dif
ferent protein and or amino acids have an
effect on iron absorption has been reported

Downloaded from jn.nutrition.org by guest on September 10, 2014

1Values are means SDfor six animals.

2 Values not followed by the same superscript are significantly different
(P < 0.05). 3 Day repletion portion of experiment stopped; day refers to animal's age. 4 Day experiment was
completed and whole-body counting stopped.
5 The administration of 7 g of diet containing 59Fe on day 73 was
not included.
'' Not included in statistical analysis.

1649

FISH AND PLANT IRON BIOAVAILABILITY

100

so
20

40

60

80

TOO

120

TIME, hr
Fig.

Percent MFe retained

by rats fed test diets

with and without fish during a period of 110 hours. Diets

were: No. 5, wheat bran and casein; No. 6, wheat bran
and fish; No. 7, spinach and casein; No. 8, spinach and
fish; No. 10, soy protein and casein; No. 11, soy protein
and fish; and No. 13 control diet plus 35 ppm Fe. Vari
ations indicated by vertical bars are SEM,but only one
side illustrated to avoid overlap.

K50-

SOPBOTEtT"1-*^
CASEIN 10 4^-sj

UEIN
FISH FtSH
11 8
SPINACH
SPtNACH CASEIN 7

CONTROL 35 PPM F. 13

r
WHEAT BRANCASEIN 5
FISH WHEAT BRAN 6 .

10

12

14

HEMOGLOBIN LEVEL, g.rtl

UPON CONSUMPTION OF"Fe

Fig. 2 Relationship between percent wFe retained in

rats after 110 hours and hemoglobin concentration at
time of ^Fe consumption. Regression determined by
method of least squares Y = -4.5 g Hb/dl + 102.4 (r2
= 0.958). Data obtained from animals consuming diet
No. 13, control + 35 ppm Fe, was not included in regres
sion calculations.

sources of plant iron with and without fish.

The anticipated enhancing effect of fish on
plant iron utilization was not observed. In
corporation of marine lipids into the diet
could have caused a decrease in palatability
because of oxidation. However, this was dis
counted because weight gain and food intake
were similar among rats fed different plant
iron sources. The problem of oxidation may
be responsible for results observed. The de
gree of unsaturation between the corn oil and
turbot oil used in this study was found to be
135 versus 152 double bonds, respectively,
per weight percent of fatty acid determined
by gas-liquid chromatography.6 Marine oils
oxidize faster than other oils because of the
long-chain highly unsaturated fatty acids
they contain (42). Increased levels of dietary
lipid unsaturation have been implicated as
being able to lower iron bioavailability
(43, 44).
Results of this study did not clearly dem
onstrate use of the rat as an animal model
to further investigate the "meat factor" ef
fect. However, recent observations with rats
(45) have demonstrated that beef protein in* As methyl estersseparatedon 2 m X 2 mm (id.) column containing 10%
diethylene glycol succinate (DECS) on acid-washed Chromosorb W, (Supelco Inc.. Bellefonle. PA) 80/100 mesh at 195V (isothermal)

Downloaded from jn.nutrition.org by guest on September 10, 2014

(10, 19, 39, 40). Casein is a phosphorus-con

taining protein, and the phosphoproteins
may contribute to lowered absorption of iron.
Rotruck and Luhrsen (41) found that the
RBV of iron in soy protein isolate and a pro
cessed soy protein isolate ranged from 82
to 100%.
The lower RBV of iron in spinach and soy
protein compared to wheat bran is attributed
to lower growth rates with spinach and soy
protein isolate diets. However, the growth
rate was generally equivalent among animal
groups consuming spinach, soy protein and
control diets with three graded levels of iron
(i.e., 6.25, 12.5 and 25 fig Fe/g diet).
Growth rate and food consumption were
similar in groups of animals fed the different

1650

CHAO AND GORDON

TABLE 6
Liver weight and percent ^Fe per gram of liver1'2

Downloaded from jn.nutrition.org by guest on September 10, 2014

trinsic iron from the same diet cannot be

compared. This is due to the varying iron
status of the animals when administered 59Fe.
accumulation
Diet
Iron status (i.e., hemoglobin level) of the an
per gliver3K0.29b0.28b0.510.550.61'0.57a0.31
of
wlK9.8a10.57.4b6.7-6.8k7.5b10.4MFe
no5678101113DietWheat
imals appears to override any effect on 59Fe
absorption that may have been caused by
diet. Numerous studies indicate a high cor
caseinWheat
bran +
relation between iron bioavailability from
fishSpinach
bran
the same food(s) by using extrinsically and
caseinSpinach
+
intrinsically labeled iron (46, 48, 49). There
fishSoy +
has been one report (50) that states the two
caseinSoy
protein +
sources of radioiron may not give comparable
bioavailability results. More extrinsic 59Fe
fishControl
protein +
was retained by rats than intrinsic 59Fe pro
+ 35 ppm Fe4Liver
vided by turnip greens (50). Amine and
1 Mean values of six animals.
2 Values not followed by
Hegsted (51) dosed rats with 59Fe, which had
the same superscript are significantly different (P < 0.05).
previously been used to determine the RBV
3 Based on total counts in whole animals upon terminating ex
periment
4 Not included in statistical analysis.
in food products by the hemoglobin regen
eration method. Hemoglobin levels at the
time of administering the 59Fe were stated
creases the bioavailability of iron in rice-pro
to be uniform within each group, ranging
tein bran.
from 8 to 10 g Hb/dl. However, they ob
A second and independent method was served a significant negative correlation be
used to ascertain if fish could enhance iron tween the retention of the isotope iron and
absorption. It was also planned to determine
the results of the bioassay. In 30 trials, the
iron assimilation by animals that were not average absorption of 59Fe was 162% of RBV
severely anemic. This was accomplished by determined against FeSO4. Johnson and
measuring the retention (i.e., absorption) of Evans (52) suggest that the extrinsic level
59Fe in the same animals after the repletion
technique may not always be applicable for
portion of the bioavailability study. Based on iron because the iron added extrinsically may
previous reports (1, 46-48), this extrinsically
be more susceptible to a change in oxidation
labeled iron was expected to mix in a com
state. Change in iron valence prior to being
mon pool with the endogenous iron provided
consumed or in the alimentary tract may
by the plant foods evaluated. Rather than
have the most dramatic influence on iron
being absorbed and retained in the ratios as bioavailability. Although differences were
indicated by the RBV, the retention of the observed in this study between methods, no
extrinsic iron from the test diets was highly significant enhancing effect on the absorption
of extrinsically labeled iron added to diets
correlated with the animals hemoglobin lev
els at time of dosing. Efficiency of 59Fe ab
containing fish was observed. The close sim
ilarity of hemoglobin levels between animal
sorption and subsequent retention appeared
groups consuming the two diets containing
to be clearly related to iron status of the an
imals and not to diet composition. The iron a single plant iron source, with or without
in soy protein fed with fish had a low RBV fish, appear to support this conclusion.
(67%), and the rats had a low hemoglobin
(5.1 g/dl) at the time they received the 59Fe.
ACKNOWLEDGMENT
However, this group of animals retained the
highest amount of 59Fe(80%) after 110 hours.
We appreciate the help of Dr. William C.
Mickelberry,
Oregon Freeze Dry Foods, Inc.,
The opposite results were observed in ani
Albany,
OR,
for the freeze-drying of fish.
mals consuming the wheat bran with fish
diet (RBV, 111%; Hb, 10.0 g/dl; 59Fe re
LITERATURE CITED
tained, 54%.)
Because of the protocol of this study, the
1. Bjrn-Rasmussen,E., Hallberg, L., Isaksson, B. &
Arvidsson, B. (1974) Food iron absorption in man.
bioavailability of endogenous iron and ex

FISH AND PLANT IRON BIOAVAILABILITY

absorption in human subjects. III. Comparison of the

effect of animal proteins on nonheme iron absorp
tion. Am. J. Clin. Nutr. 29, 859-867.
20. Mullen, J. D. (1978) Dietary fiber sources for hu
man studies. Am. J. Clin. Nutr. 31, S103-S106.
21. Crosby, W. H., Munn, J. I. & Furth, F. W. (1954)
Standardizing a method for clinical hemoglobinometry. U.S. Armed Forces Med. J. 5, 693-703.
22. American Institute of Nutrition (1977) Report of
the AIN Ad Hoc Committee on standards for nu
tritional studies. J. Nutr. 107, 1340-1348.
23. Association of Official Analytical Chemists (1975)
Official Methods of Analysis, 12th ed., See. 2.049,
AOAC, Washington, DC.
24. Folch, J., Lees, M. & Sloane Stanley, G. H. (1957)
A simple method for the isolation and purification
of total lipides from animal tissues. J. Biol. Chem.
226, 497-509.
25. Franco, V. & Krinitz, B. (1973) Determination of
oxalic acid in foods. J. Assoc. Off. Anal. Chem. 56,
164-169.
26. de Boland, A. R., Garner, G. B. & O'Dell, B. L.
(1975) Identification and properties of "phytate"
in cereal grains and oilseed products. J. Agrie. Food
Chem. 23, 1186-1189.
27. Fiske, C. H. & SubbaRow, Y. (1925) The calori
metrie determination of phosphorus. J. Biol. Chem.
66, 375-385.
28. Snedecor, G. W. & Cochran, W. G. (1976) Statis
tical Methods, 6th ed, pp. 147 and 258-267, The
Iowa State University Press, Ames, Iowa.
29. Mahoney, A. W., VanOrden, C. C. & Hendricks,
D. G. (1974) Efficiency of converting food iron
into hemoglobin by the anemic rat. Nutr. Metab. 17,
223-230.
30. Morris, E. R. & Ellis, R. (1980) Bioavailability to
rats of iron and zinc in wheat bran: response to low
phytate bran and effect of the phytate/zinc molar
ratio. J. Nutr. 110, 2000-2010.
31. Morris, E. R. & Ellis, R. (1976) Isolation of monoferric phytate from wheat bran and its biological
value as an iron source to the rat. J. Nutr. 106, 753760.
32. Reinhold, J. G., Ismail-Beigi, F. & Faradji, B. (1975)
Fiber vs. phytate as determinant of the availability
of calcium, zinc and iron of breadstuffs. Nutr. Rept.
Int. 12, 75-85.
33. Pia, G. W. & Fritz, J. C. (1970) Availability of
iron. J. Assoc. Off. Anal. Chem. 53, 791-800.
34. Fahey, G. C., Williams, J. E. & McLaren, G. A.
(1976) Influence of molasses lignin-hemicellulose
fractions in rat nutrition. J. Nutr. 106, 1447-1451.
35. Cheeke, P. R. & Patton, N. M. (1978) Effect of
alfalfa and dietary fiber on the growth performance
of weanling rabbits. Lab Anim. Sci. 28, 167-172.
36. Knehans, A. W., Kincaid, R. L., Regan, W. O. &
O'Dell, B. L. (1979) An unrecognized dietary fac
tor for guinea pigs associated with the fibrous frac
tion of plant products. J. Nutr. 109, 418-425.
37. Van Campen, D. R. & Welch, R. M. (1980) Avail
ability to rats of iron from spinach: effects of oxalic
acid. J. Nutr. 110, 1618-1621.
38. Steinke, F. H. & Hopkins, D. T. (1978) Biological
availability to the rat of intrinsic and extrinsic iron
with soybean protein isolates. J. Nutr. 108, 481-489.

Downloaded from jn.nutrition.org by guest on September 10, 2014

Applications of the two-pool extrinsic tag method

to measure heme and non-heme iron absorption
from the whole diet. J. Clin. Invest. 53, 247-255.
2. Conrad, M. E., Burton, I. B., Williams, H. L. & Foy,
A. L. (1967) Human absorption of hemoglobiniron. Gastroenterology 53, 5-10.
3. Layrisse, M. (1975) Dietary iron absorption. In:
Iron Metabolism and its Disorders (Ruf, H., ed.), p.
25, Excerpta Medica, Amsterdam.
4. Layrisse, M., Cook, J. D., Roche, M., Khn,I. N.,
Walder, R. B. & Finch, C. A. (1969) Food iron
absorption: a comparison of vegetable and animal
foods. Blood 33, 430-443.
5. Fritz, J. C. & Pia, G. W. (1972) Application of the
animal hemoglobin repletion test to measurement
of iron availability in foods. J. Assoc. Off. Anal.
Chem. 55, 1128-1132.
6. Davies, N. T. & Nightingale, R. (1975) The effects
of phytate on intestinal absorption and secretion of
Zn and whole body retention of Zn, copper, iron and
manganese in rats. Brit. J. Nutr. 34, 243-258.
7. Bjrn-Rasmussen,E. (1974) Iron absorption from
wheat bread. Influence of various amounts of bran.
Nutr. Metab. J6, 101-110.
8. Disler, P. B., Lynch, S. R., Charlton, R. W., Torrance, J. D., Bothwell, T. H., Walker, R. B. & Mayet,
F. (1975) The effect of tea on iron absorption. Gut
16, 193-200.
9. Sayers, 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.
Brit. J. Haematol. 24, 209-218.
10. Bjrn-Rasmussen,E. & Hallberg, L. (1979) Effect
of animal proteins on the absorption of food iron in
man. Nutr. Metab. 23, 192-202.
11. Cook, J. D. & Monsen, E. R. (1975) Food iron
absorption. I. Use of a subsynthetic diet to study
absorption of nonheme iron. Am. J. Clin. Nutr. 28,
1289-1295.
12. Food and Nutrition Board, National Research Coun
cil (1971) Workshop on extent and meanings of
iron deficiency in the U.S., 113 pp.. National Acad
emy of Sciences, Washington, DC.
13. National Marine Fisheries Service (1975) Fisheries
of the United States 1974, Current Fishery Statistics
No. 6700, pp. 64-65, U.S. Department of Com
merce.
14. U.S. Senate Select Committee on Nutrition and Hu
man Needs (1977) Dietary Goals for the United
States, 95th Congress, 1st Session, p. 13, U.S. Gov
ernment Printing Office, Washington, DC.
15. Anonymous (1979) Does eicosapentenoic acid pre
vent thrombosis and atherosclerosis? Nutr. Rev. 37,
316-317.
16. Gordon, D. T. & Roberts, G. L. (1977) Mineral
and proximate composition of Pacific Coast fish. J.
Agrie. Food Chem. 25, 1262-1268.
17. Martinez-Torres, C., Leets, I. & Layrisse, M. (1975)
Iron absorption by humans from fish. Arch. Latinoam. Nutr. 25, 199-210.
18. Layrisse, M., Martinez-Torres, C. & Roche, M.
(1968) Effect of interaction of various foods on iron
absorption. Am. J. Clin. Nutr. 21, 1175-1183.
19. Cook, J. D. & Monsen, E. R. (1976) Food iron

1651

1652

CHAO AND GORDON

39. Miller, J. (1974) Metabolism of iron by rats fed

wheat diets supplemented
with lysine and methionine. Nutr. Rep. Int. 10, 1-8.
40. Monsen, E. R. & Cook, J. D. (1979) Food iron
absorption in human subjects. V. Effects of the major
dietary constituents of a semisynthetic
meal. 32,
804-808.
41. Rotruck, J. T & Luhrsen, K. R. (1979) A com
parative study in rats of iron bioavaiiability
from
cooked beef and soybean protein. J. Agrie. Food
Chem. 27, 27-33.
42. Olcott, H. S. (1962) Marine products. In: Sym
posium on Foods: Lipids and Their Oxidation
(Schultz, H. W., Day, E. A. & Sinnhuber, R. O.,
eds.), pp. 173-189, AVI Pubi. Co., Westport, CT.
43. Amine, E. K. & Hegsted, D. M. (1975) Effect of
dietary carbohydrates
and fats on inorganic iron
absorption. J. Agrie. Food Chem. 23, 204-208.

47.

48.

49.

50.

51.

May 1983, Chicago, IL. J. Am. Oil Chem. Soc. (abs.).

60, 739-740.
45. Gordon, D. T., Godber, S. J. & Hess, R L. (1982)
The enhancement
of nonheme iron uptake by beef
protein in the rat. Institute of Food Technologists

52.

Downloaded from jn.nutrition.org by guest on September 10, 2014

44. Gordon, D. T. (1983) Iron absorption by the ane

mic rat as influenced by beef tallow, corn oil, salmon
oil, vitamin E and butylated
hydroxy toluene.
American Oil Chemists' Society 74th Ann. Meeting,

46.

Program 42nd Ann. Meeting, June 1982, Las Vegas,

NV, p. 139 (abs.), IFT, Chicago, IL.
Monson, E. R. (1974) Validation of an extrinsic
iron label in monitoring absorption of nonheme food
iron in normal and iron-deficient rats. J. Nutr. 104,
1490-1495.
Bjrn-Rasmussen, E., Hallberg, L. & Walker, R. B.
(1972) Food iron absorption in man. 1. Isotopie ex
change between food iron and inorganic iron salt
added to food: studies on maize, wheat and eggs.
Am. J. Clin. Nutr. 25, 317-323.
Layrisse, M., Martinez-Torres,
C., Cook, J. D.,
Walker, R. & Finch, C. A. (1973) Iron fortification
of food: its measurement by the extrinsic tag method.
Blood 41, 333-352.
Hallberg, L. (1974) The pool concept in food iron
absorption and some of its implications. Proc. Nutr.
Soc. 33, 285-291.
Wien, E. M., Van Campen, D. R. & Rivers, J. M.
(1975) Factors affecting the concentration and bioavailability of iron in turnip greens to rats. J. Nutr.
105, 459-466.
Amine, E. K. & Hegsted, D. M. (1974) Biological
assessment of available iron in food products. J.
Agrie. Food Chem. 22, 470-475.
Johnson, P. E. & Evans, G. W. (1978) Iron ab
sorption by rats from nonprescription
dietary iron
supplements. J. Agrie. Food Chem. 26, 844-847.