- Email: [email protected]
Mineral interactions in rats fed AIN-76A diets with excess calcium
Pergamon
0278-6915(93)E0011-W
Fd Chem. Toxic. Vol. 32, No. 3, pp. 255-263. 1994 Elsevier Science Ltd. Printed in Great Britain 0278-6915/94 $7.00 + 0.00
MINERAL INTERACTIONS IN RATS FED AIN-76A DIETS WITH EXCESS CALCIUM M. E. SHACKELFORD*, T. F. X. COLLINS,T. N. BLACK, M. J. AMES,S. DOLAN, N. S. SHEIKH, R. K. CHi and M. W. O'DONNELL Center for Food Safety and Applied Nutrition, US Food and Drug Administration, Washington, DC 20204, USA
(Accepted 23 August 1993) Abstract--The effects of moderate increases in dietary calcium on maternal and foetal mineral interactions were studied in Charles River CD/VAF Plus rats. Female rats were given 0.50, 0.75, 1.00or 1.25% dietary calcium as calcium carbonate in AIN-76A diets for 6 wk before mating, during mating and for 20 days of gestation. Inductively coupled argon plasma-atomic emission spectrometry was used to determine mineral levels in the tissues of non-pregnant rats after 42 days on the diets, in the tissues of pregnant rats on day 20 of gestation and in the whole body of day-20 foetuses. The femurs of the non-pregnant and pregnant rats had a dose-related linear increase in calcium content, in livers of the non-pregnant rats, dose-related linear increases in the phosphorus, zinc and magnesium content were observed, but there was a dose-related decrease in the iron content. There were dose-related linear decreases in the iron and copper contents of the kidneys from the non-pregnant rats. in pregnant rats dose-related linear decreases were observed in the iron content of the liver and in the zinc, iron and magnesium contents of the kidney. The foetuses from rats given a moderate increase in dietary calcium had dose-related decreases in the whole-body contents of phosphorus, iron, copper and magnesium.
INTRODUCTION
In 1959 the National Institutes of Health sponsored a symposium on the effects of high calcium intake because of concern that the level of dietary calcium had increased to the extent that it might be a nutritional health hazard (Davis, 1967). One of the purposes of the symposium was to evaluate the effects of calcium intake on the utilization of other nutrients (Davis, 1959). In more recent years, public health concern over the prevention of osteoporosis has led to increased fortification of foods with calcium and sales of calcium supplements have increased from $47 million in 1983 to $200 million in 1987 (Consumer Reports, 1988). Since the symposium in 1959, mineral balance experiments have been conducted to study the effects of dietary calcium and/or phytic acid on the utilization of zinc, magnesium and phosphorus (Forbes, 1963 and 1964; Likuski and Forbes, 1965). Likuski and Forbes (1965) did an 18-day balance study with male Sprague-Dawley rats fed nutritionally adequate casein-based, semi-purified diets with or without 2% phytic acid. Calcium, as CaCO 3 was added to give levels of 0.4, 0.8 and i.2%. Dietary calcium decreased zinc absorption from diets containing phytic acid,
*To whom correspondence and reprint requests should be addressed at: US Food and Drug Administration (HFS507). 8301 Muirkirk Road, Laurel, MD 20708, USA. Abbreviations: ANOVA = analysis of variance: LSD test = least significant difference test.
but did not significantly lower the amount of zinc absorbed from diets without phytic acid. Absorption of magnesium and phosphorus was decreased in a dose-related manner in the presence or absence of phytic acid. A relationship between dietary calcium content and iron absorption has been postulated for many years, it is well documented that in rats high calcium intake diminishes the absorption of inorganic iron (Barton et al., 1983). By using isolated intestinal loops it was shown that increasing the calcium concentration (1 to 100mM CaCI2) decreased the entry of radiolabelled iron into the microvilli of intestinal epithelial cells. To test the significance of this in animals, male Wistar rats were fed diets containing 1.55-1.75% calcium (high calcium) or 0.63-0.68% calcium (control) for 4wk. The diets contained 0.45% phosphorus and casein as a protein source; they were either iron replete (85 mg/kg) or iron deficient (21 mg/kg). In rats fed the iron-replete diets, those receiving the high-calcium diets developed decreased marrow iron stores, but had no significant change in microhaematocrit, serum iron concentration or total iron-binding capacity of the serum. Among rats fed the mildly iron-deficient diets, those fed the high-calcium diets had significantly lower serum iron concentrations and decreased marrow iron stores. This condition was shown to result from a calcium-related inhibition of iron absorption but not from an increased gastro-intestinal transit. The study demonstrated the potential of a high-calcium diet to induce or accelerate iron deficiency.
255
M.E. SHACKELFORDet al.
256
Although it is known that excess calcium intake can interfere with mineral utilization, no animal studies were found in which the added physiological stress of pregnancy on these calcium-mineral interactions was carefully examined. In the study reported here, the effects of feeding 0.50 (control), 0.75, 1.00 or 1.25% dietary calcium on foetal and adult mineral levels were investigated to determine whether excess dietary calcium could perturb tissue mineral levels. All diets were nutritionally adequate AIN-76A diets with casein as the protein source and without dietary phytic acid. MATERIALS AND METHODS
Experimental design. Female and male CD-VAF/ Plus Charles River rats were obtained from Charles River Laboratories, Inc. (Wilmington, MA, USA); when received, the females were 52 days old and the males were 44 days old. At the end of a 6-day quarantine period, 15 of the female rats were randomly selected for evaluation of tissue mineral status and histopathology before assignment to the dietary groups, and 276 female rats were assigned to the control or test diets by a stratified random method according to body weight. The rats in the four dietary groups received 0.50 (control), 0.75, 1.00 or 1.25% calcium in powdered AIN-76A for 6wk before mating, throughout mating and for 20 days of gestation. Males were fed Purina Chow 5002 (Purina Mills, inc., Richmond, IN, USA) except during mating, when they also received the control or test diets. The AIN-76A diets were prepared by BIOSERV (Frenchtown, N J, USA). Homogeneity and concentrations of calcium, phosphorus, magnesium, manganese, zinc, iron and copper were assayed by the New Jersey Feed Laboratory, Inc. (Trenton, N J, USA) using the methods of the Association of Official Analytical Chemists (Padmore, 1990). Full details of the study design, including housing, food and water consumption, diets, mating, and performance of caesarean sections have been described in the previous paper (Shackelford et al., 1993). The female rats that were evaluated for tissue mineral status and histopathology before assignment to the dietary groups are referred to below as baseline animals. After 6 wk on the control or test diets, 60 females (15 per group) were randomly selected for evaluation of tissue mineral status and histopathology. The remaining females were mated and the sperm-positive females were continued on the appropriate diets for 20 additional days. 15 pregnant rats per dietary group were randomly selected for removal of tissues when caesarean sections were performed on day 20 of gestation. From these 15 pregnant rats, 12 were randomly selected for evaluation of the mineral status of the foetuses. Two foetuses per litter (one from each uterine horn) were randomly selected.
Removal of tissues. The rats were killed by carbon dioxide asphyxiation, and then exsanguinated through the posterior vena cava to remove blood from the organs and tissues. For pregnant rats from which the foetuses were removed, the uterine arteries were clamped before exsanguination through the posterior vena cava to prevent withdrawal of the blood supply to the foetuses before they were removed from the uterus. All organs and tissues were examined grossly, and any abnormal findings were recorded. The organs and tissues (liver, left kidney and both femurs) and foetuses were placed in polyethylene bags and stored at - 4 0 " C . Terminal body weights and selected organ weights were obtained at the time of autopsy. At the time of autopsy, heart, liver and kidneys were weighed (left and fight kidneys were weighed separately). Histological methods. The following organs and tissues from non-pregnant and pregnant rats were placed in 10% neutral buffered formalin: heart, a part of the liver, the fight kidney and bone (right humerus and sternum). The tissues were later processed, embedded in paraffin, sectioned at 5 pro, stained with haematoxylin and eosin and examined histopathologically. The bone specimens were decalcified in a mixture of formic acid, hydrochloric acid and distilled water and subsequently processed as above. Special stains included the following: (I) Periodic Acid Schiff reaction with and without diastase for detecting the presence of glycogen on selected liver sections; (2) Von Koss's stain for calcium on selected liver, bone and kidney sections; (3) osmium tetroxide stain for fat on selected liver sections. The lesions were graded as minimal (Grade 1), mild (Grade 2), moderate (Grade 3) or severe (Grade 4). Mineral analysis. Determinations of mineral levels in specific tissues were made before dietary treatments (baseline values), after 6 wk on the diets and on day 20 of gestation. Calcium, phosphorus, zinc, iron, copper, magnesium and manganese were determined in the livers and left kidneys from non-pregnant and pregnant rats, and in the foetuses. Calcium, phosphorus and magnesium were determined in the left femurs from the adult animals. A portion of the liver, left kidney and femur, and whole pups, were weighed and wet-digested in mixtures of nitric-perchloric or nitric-perchloricsulfuric acids (Rader et al., 1984). Final acid concentrations were 10% perchloric acid for femurs and 10% sulfuric acid for livers, kidneys and whole pups. Tissue digests were analysed by inductively coupled argon plasma-atomic emission spectrometry on the ARL 3580 (Applied Research Laboratories, Dearborn, MI, USA) as described previously (Dolan et al., 1991). The following analytical wavelengths (A) were used for analysis: Ca, 3933.7; Cu, 3247.5; Fe, 2599.4; Mg, 3838.3; Mn, 2576.1; P, 1782.9; and Zn, 2138.6. Calcium was determined in foetuses and diluted femurs at 3179.3 ~,. Portions of National Bureau of Standards and Technology (formerly National
257
Calcium carbonate-mineral interaction Table I. Organ weights and final body weights for female rats fed diets containing excess calcium Weight (g) in rats fed calcium at: Tissue
0.50% (control)
Liver Right kidney Left kidney Heart
Finalbody weight.,. Liver Right kidney Left kidney Heart
Finat body weight...
0.75%
12.11 _+0.41 I. 10 -+ 0.03 1.09 4- 0.03 1.09 ± 0.03 311.6 ± 8.3
Nora-pregnant rats 11,98+0.51 1.10 -+ 0.02 1.07 -+ 0.02 1.04 ± 0.02 311.3_+8.7
17.22 _+ 0,52 1.19 _+ 0.04 1.17 + 0.05 1.15 _+0.02 434.2 ± 8.1
Pregnant rats 17.02 _+ 0.33 1.15 4- 0.0l 1.13 ± 0.02 1.19 _+0.03 446.1 -+ 7.6
1,00%
1,25%
I 1.39 -+ 0,47" 1.09 ± 0.04 1.05 4- 0.03 1.08 -+ 0.03 313.3+8.6
10.82-+0.28 .'° 1.06 ± 0.02 1.04 4- 0.02 1.05 _+0.02 306,5 -+ 5.8
16.30 _+0.56 1.10 -+ 0.03 1.07 _+0.03 1.17 + 0.03 429.5+12.5
16.58 _+0.47 1.15 -+ 0.02 1.13 _+0.03 1.20 _+0.02 443.6 -+ 8.2
Values are means ± SEM for groups of 15 rats. Asterisks indicate significant differences between control and test values ( ' P ~<0,05: **P ~ 0.01). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
Bureau of Standards) Standard Reference Material Bovine Liver 1577 were digested and analysed for the same elements. Values fell within 100 +_ 2% of certified values for all elements of interest. Statistical analysis. All statistical analyses were performed by the Division of Mathematics, US Food and Drug Administration. The terminal body weights were analysed by analysis of variance (ANOVA) and tissue weights were analysed by analysis of covariance using an adjustment for terminal body weight. This was followed by a protected least significant difference (LSD) test (two-tail) for pairwise comparison of treated groups with the control group. Means and standard deviations were calculated for levels of the elements in the rat tissues and foetuses. Statistical analysis involved ANOVA followed by the LSD test (two-tail) for pairwise comparison of the treated groups with the control group (Sokal and Rohlf, 1981). if the variances were homogeneous, a dose-response regression was performed. A dose-related linear trend was said to be present if the P value of the lincarity component was <0.05 and, at the
same time, the departure P value was />0.05. If a trend was present, the direction of the trend (increase or decrease) was noted. RESULTS
Organ weights are summarized in Table I. When the organ weights of the pregnant animals were adjusted for the terminal body weights, there were no significant differences between the controls and treated groups in any organ weight. For the organs from the non-pregnant animals, the liver weight of rats in the two highest dose groups was significantly lower than that of the controls. The average daily food consumption of non-pregnant animals between days 0 and 42 increased significantly for the !.00 and 1.25% dietary groups in comparison with the control groups, but there was no significant difference in weight gain over this same time interval (Shackelford et al., 1993). Thus, the decrease in the adjusted liver weight is not attributable to a decrease in food consumption or weight gain.
Table 2. Summary incidence of female rats with renal mineralization Category of ratsf Baseline Non-pregnant Non-pregnant Non-pregnant Non-pregnant Pregnant Pregnant Pregnant Pregnant
Dose level (% calcium in diet)
Renal mineralization* Incidence
Mean severity
NA 0.50 0.75 1.00 1.25 0.50 0.75 1.00 1.25
6/15 (40%) 9/I 5 (60.0%) I/I 5 (7.0%) 8il 5 (53.3%) 7/I 5 (47.0%) I 0/15 (67.0%) 5/I 5 (33.3%) 7/15 (47.0%) 5/15 (33.3%)
1.3 1.6 1.0 1.4 I. I 1.5 1.6 1.4 1.0
NA - not applicable "The values for renal mineralization are expressed as the incidence (the number of animals affected out of the total number of animals examined, parcentage in parentheses), and the mean severity (the average grade/number of tissues affected). tBaselin¢ rats wcrc killed for analysis before the feeding study began. Non-pregnant rats were fed the control or calcium enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
258
M.E. SHACKELFORD et al.
Table 3. Mineral levels for non-pregnant animals before the feeding study was begun 1baseline rats) Mineral levels* in the: Mineral Liver Kidney Femur Ca 32.8+2.1 79.0±22.1 144.34-6.8 P 3303.94- 149.1 2834.6_+156.4 72.2_+3.1 Zn 27.0 _4-2.3 21.8 4. 1.6 ND Fe 173.6 4. 37.9 76.8 4- 17.0 ND Cu 4.0 4- 0.3 9.2 __.3.8 ND Mg 207.9 _+8.5 190.1 4- 10.7 3.0 4-0.2 Mn 1.94-0.2 0.64-0.1 ND ND = not done *The mineral content is expressed as pg, g fresh tissue weight for the kidney and liver, and as mg/g fresh tissue weight for the femur. Values are means _4-SD of singledeterminations of 15 specimens.
The histopathological findings included periportal hepatocellular vacuolation in the liver, mineralization in the kidney a n d chronic myocarditis in the heart. Hepatocellular vacuolation was frequently seen in the n o n - p r e g n a n t a n d p r e g n a n t animals, but it was not observed in the baseline rats. T h e incidence o f rats with hepatocellular vacuolation and the mean severity (mild to moderate) o f the lesion were generally similar in control and treated groups. The s u m m a r y incidence o f rats with renal mineralization is presented in T a b l e 2. T h e mineralization was observed p r e d o m i n a n t l y in the renal t u b u l a r epithelial cells at the corticomcdullary j u n c t i o n a n d in the inner stripe o f the outer medulla. Myocarditis, characterized by focal to multifocal myocardial degeneration, fibrosis and m o n o n u c l c a r cell infiltration, was observed in one bascline rat, one n o n - p r e g n a n t rat on the i . 0 0 % calcium dict, three p r e g n a n t rats on the control diet, a n d one p r e g n a n t rat in each o f two highest calcium dose groups. There was no observable histopathological c h a n g e in the bone or bone m a r r o w in any o f the rats. The mineral c o n t e n t s o f the kidney, liver a n d femur (Table 3) o f o u r baseline animals were similar to those reported by R a d e r et al. (1984). However, o u r
Mineral Ca P Zn Fe Cu
Mg Mn
results for the iron c o n t e n t o f the livers from the baseline rats (174 l~g/g fresh weight) were higher than those o f R a d e r et al. (1984), who reported liver iron values o f 4 0 - 5 9 / z g / g fresh weight. The data of Rader et al. (1984) were o b t a i n e d from 22- to 26-day-old male L o n g - E v a n s rats that had been b o r n to dams fed the N I H - 3 ! diet, whereas o u r data were o b t a i n e d from 58-day-old Charles River rats ( C D - V A F / P l u s ) that had been b o r n to females fed Purina Lab Chow. Thus, the female animals used in o u r study had normal tissue mineral c o n t e n t s at the onset o f the study. T h e mineral levels in the livers o f the rats fed diets with excess calcium are s h o w n in Table 4. In the n o n - p r e g n a n t rats, after 6 wk on the diets the copper a n d m a n g a n e s e c o n t e n t s o f the livers from the groups fed the high-calcium diets were not significantly different from those o f the control group. A d o s e - r e s p o n s e regression for p h o s p h o r u s , zinc and m a g n e s i u m showed a dose-related linear increase. F o r calcium, the d o s e - r e s p o n s e regression had borderline significance ( P = 0.0571). in contrast, a dose-related linear decrease in the iron content of the liver was observed. In the p r e g n a n t rats, the calcium, p h o s p h o r u s , copper, magnesium a n d m a n g a n e s e contents o f the livers o f those fed high-calcium diets did not differ signiticantly from the levels o f the 0.50% calcium (control) g r o u p (Table 4). The zinc c o n t e n t o f the liver was significantly decreased at thc 1.00% calcium level, but was similar to the control value at 0.75 and 1.25% calcium. T h e r e was no dose-related trend. The iron c o n t e n t o f the livers o f the p r e g n a n t rats, like that of the livers o f the n o n - p r e g n a n t rats, was significantly decreased in rats fed 0.75, 1.00 a n d !.25% dietary calcium. The trend of the dose-response regression for iron showed a dose-related linear decrease. Table 5 lists the mineral levels in the kidneys o f the rats fed diets with excess calcium. In the nonp r e g n a n t animals, after 6 wk on the diets, no signifi-
Table 4. Liver mineral levels for female rats fed diets with excess calcium Mineral levels (pg/g fresh weight of liver)'l"in rats fed calcium at: 0.50% (control) 0.75% 1.00% 1.25% 31.8 _4-3.8 2952.5 +_ 221.8 22.6 + 1.8 164.0 ± 37.0 4.3 + 0.5
193.3 ± 13.5 2.0+0.3
Non-pregnant rats 32.1 _+ 1.5 2965.5 ± 146.1 23.2 + 1.4 138.4 + 29.3* 4.3 ± 0.5
192,5 +9.1 2.1 +0.2
Zn
40.4 ± 2.6 3139.2 ± I00.I 23.2 ± I.I
Pregnant rats 42.5 _ 6.5 3134.8 _+.99.1 22.8 +_ 1.3
Fe
99.1 ± 26.6
78.1 .:l: 15.7 °0
Ca
P Cu Mg Mn
4.2 + 0.2 223.2 + 9.3 2.9 -t- 0.3
4.2 "1-0.3 222.3 4. 5.8 2.9 + 0.2
34.6 ± 4.2" 3099.4 ± 216.7" 23.9 + 2.2" 132.2 + 38.6 °" 4.6 __.0.7
203.0 + 12.3" 2.2+0.3 39.9 ± 3.1 3122.8 _+ 138.8 22.3 ± 0.9* 63.7 + 9 . 7 °**
4.2 + 0.3 222,1 + 9.1 2.9 + 0.3
33.7 ± 4.3 3102.4 _+ 120.9" 23.7 ± 0.9 122.4 +_ 24.5"** 4.4 ± O,2 21)1.9 ± 8.6*
2.1 _.+0,3 41.0 ± 3,6
3114.5 ± 171.7 23.4 ± 1,6 65.9 ± 16.0 **0
4.2 + 0.4 220.3 ± 12.5 2.7 ± 0.4
tValucs are means _+SD of single determinations of 15 specimens. Asterisks indicate significant differences between control and test values (*P ~ 0.05; * ' P ~0,01; * * ' P ~ 0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6wk. then mated and continued on the appropriate diets until day 20 of gestation.
Calcium carbonate-mineral interaction
259
Table 5. Kidney mineral levels for female rats fed diets with excess calcium Mineral levels (mg/g fresh weight of kidney)* in rats fed calcium at: Mineral
0.5000 (control)
No. ~'~ec'imens. Ca P
Zn Fe Cu Mg Mn No. of specimens.
15 (14)
76.6 __. 19.2 2733.2 ± 212.7 21.5 ± 2.1 101.3 + 19.7 8.3 + 1.7 176.6 + 14.0 0.8 ±0.1 15(14)
84.8 ± 30.5 2587.8 ± 173.0 20.8 _+ 1.3 77.4 ± 10.8 7.3 ± 3.0 166.7 ± 11.2 0.9 +0.1
Ca P Zn Fe Cu Mg Mn
0.75*:, Non-pregnant rtts 15
67.2 + 4.0 2736.84- 146.7 21.4 ± 1.4 94.2 + 17.4 8.0 ± 1.8 176.1 + 9.2 0.8 ±0.1 Pregnant rats 15
82.1 _+34.8 2528.9_ 69.9 20.2 ± 1.4 70.3 _+9.6* 6.5 + 1.9 164.2 ± 5.0 0.8 ±0.1
1.00%
1.25%
15 (14)
70.9 ± 4.7 2777.5± 150.1 21.6__. 1.6 81.0 + 16.1"* 7.6 ___1.8 178.6 + 9.4 0.8 ±0.1
/5 (14)
77.8 + 18.2 2841.0± 255.4 21.7 + 2.0 90.7 4- 19.6 6.6 + 1.3"* 182.6 + 15.8 0.8 ± 0.0
15
15
75.0 ± 9.5 2501.0 ± 78.3 19.6 ± 1.1" 62.9 _+6.8"** 6.4 ± 2.4 159.6 +_4.8* 0.8 ±0.1
75.3 ± 5.8 2508.6 4- 129.0 19.6 ± 1.1" 62.6 4- 7.4*** 6.6 ± 3.2 158.6 ± 8.7** 0.8 4- 0.1
tValues are means 4- SD of single determinations of the number of specimens shown (where outliers in calcium analysis reduced N values these reduced N values are given in parenthesis). Asterisks indicate significant differences between control and test values. (*P ~<0.05; **P ~<0.01; "'*P ~<0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk. then mated and continued on the appropriate diets until day 20 of gestation.
T h e m i n e r a l levels in the f e m u r s o f the rats are s h o w n in T a b l e 6. In p r e g n a n t rats, a l t h o u g h the decrease in the m a g n e s i u m c o n t e n t w a s significant in the g r o u p fed 0 . 7 5 % calcium, there w a s no doserelated trend. T h e r e were n o signilicant effects o n the m a g n e s i u m c o n t e n t s o f the f e m u r s f r o m the n o n p r e g n a n t rats. F o r the calcium c o n t e n t o f the f e m u r s f r o m the n o n - p r e g n a n t a n d p r e g n a n t a n i m a l s , the t r e n d o f the d o s e - r e s p o n s e regression s h o w e d a doserelated linear increase. T h e r e w a s no effect o n the p h o s p h o r u s c o n t e n t o f the femurs. T h e m i n e r a l c o n t e n t s o f f o e t u s e s f r o m d a m s fed excess dietary c a l c i u m are s u m m a r i z e d in T a b l e 7. C a l c i u m , zinc a n d m a n g a n e s e c o n c e n t r a t i o n s in the foetuses were similar in the c o n t r o l a n d h i g h - c a l c i u m g r o u p s . Both the p h o s p h o r u s a n d m a g n e s i u m c o n t e n t s were signilicantly decreased in the 1.25% g r o u p . A l t h o u g h the decrease in c o p p e r c o n t e n t o f the f o e t u s e s w a s small, it w a s significant in the g r o u p s fed 1.00 o r 1.25% calcium. Foetal iron c o n t e n t
c a n t differences were seen in the calcium, p h o s p h o r u s , zinc, m a g n e s i u m o r m a n g a n e s e c o n t e n t s o f the kidneys f r o m g r o u p s fed h i g h - c a l c i u m diets, in c o m p a r i s o n with levels in the g r o u p fed 0 . 5 0 % calcium. T h e decrease in the iron c o n t e n t w a s significant at the 1.00% level a n d the decrease in the c o p p e r c o n t e n t w a s significant at the 1.25% c a l c i u m level. T h e trend o f the d o s e - r e s p o n s e r e g r e s s i o n for these t w o m i n e r a l s s h o w e d a d o s e - r e l a t e d linear decrease. In the p r e g n a n t rats, n o significant differences were o b s e r v e d in the calcium, p h o s p h o r u s , c o p p e r a n d m a n g a n e s e c o n t e n t s o f the kidneys f r o m a n i m a l s fed h i g h - c a l c i u m diets in c o m p a r i s o n with the c o n t r o l s . T h e zinc a n d m a g n e s i u m c o n t e n t s o f the kidney were decreased significantly in the g r o u p s led the t w o highest d o s e s o f calcium, a n d the iron c o n t e n t w a s significantly decreased in the p r e g n a n t rats fed 0.75, 1.00 o r 1.25% dietary calcium. T h e trend o f the d o s e - r e s p o n s e regression for these three m i n e r a l s s h o w e d a d o s e - r e l a t e d linear decrease.
Table 6. Femur mineral levels for female rats fed diets with excess calcium Mineral levels (mg/g fresh weight of femur)* in rats fed calcium at: Mineral No. of specimens,..
0.50°/, (control) 15
Ca
165.3 ± 7.2
P Mg
80.7 + 3.2
3,1 -t-0.2
0.75*/0 Non~prc~nan! rats 14
166,0 ± 5.2 80.8 +_2.7 3.2+0.2
Pll'e~nan| No. of specimen... Ca P Mg
15 168.4 + 4.0 82.0 + 2.2 2.9 ± O.I
1.00% 15
170.3 ± 6.1" 82.5 ± 3.2
3.2±0.2
1.25% 15
169.6 4- 4.7" 81.9 + 2,8 3.0±0.3
rats
14
171.1 + 4.6 83.6 + 2.5 3.1 ± 0.2**
15
171.4 + 4.2 83.0 ± 2.7 3.0 4- 0.2
15
173.8 ± 4.2*** 83.4 ± 2.4 2.9 4- 0.3
tValues are means :1: SD o f single determinations of the number of specimens shown. Asterisks indicate significant differences between control and test values ("P ~ 0.05; P ~ 0.01; " ' * P ~<0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
260
M. E. SHACKELFORD et al. Table 7. Mineral levels in foetuses from dams fed diets with excess calcium Mineral levels in rats fed calcium at: Mineral
0.50'% (control)
0.75%
1.00%
1.25%
No. o~ spet'imt'ns... Ca P Zn Fe Cu Mg Mn
24 1877.8 ± 147.9 2567.2 + 92.5 17.1 + 1.4 48.9 -+ 8.9 2.4 + 0.3 178.0 _+6.6 0.2 4-_0.0
22 1907.5 + 147.5 2551.0 + 135.6 16.6_+ 1.2 42.3 -+ 9.7"* 2.4 -+ 0.2 177.9 + 8.0 0.2 _+0.0
24 1888.8 + 192.4 2541.8 + 145.8 17.1 4- I.I 38.6 -+ 5.6*** 2.2 _+0.3* 175.1 _+ 7.3 0.2 + 0.0
24 1844.8 + 1870 2493.7 + 125.6" 17.0_+ 1.5 31.9 _+_+6.3"** 2.1 _+0.2** 168.2 + 9.3"** 0.2 + 0.0
~'The mineral content is expressed as ug/g fresh weight of foetuses. Values are means + SD of single determinations of the number of specimens shown. Asterisks indicate significant differences between control and test values (*P ~<0.05; **P ~<0.01: ***P ~<0.001).
was significantly decreased at all three dose levels in comparison with the controls. There was a dose-related negative linear trend for all four of these minerals (phosphorus, magnesium, copper and iron). DISCUSSION
Excess calcium intake can interfere with mineral utilization, but early reviews did not examine the added physiological stress of pregnancy (Davis. 1959 and 1967). We were interested in calcium-nutrient interactions because foetal development can be affected by deficiencies of minerals such as iron, magnesium and zinc (tturlcy, 1977), The purpose of the present study was to provide information on the maternal and foetal mineral interactions in rats fed a moderate increase in dietary calcium. The discussion is limited to a comparison of the data obtained in our study with the published literature on calciummineral interactions in rats. Where previously published data are available, the discussion includes a comparison between mineral intake, tissue mineral levels and prenatal development in animals. In our study we observed significantly lower iron levels in the maternal liver and kidney and in the foetuses from rats fed calcium-enriched diets. The iron requirement of a normal pregnancy is distributed to the foetus, placenta and expanded cell mass, but the mechanism by which the body and the mucosal cells achieve regulation of the body iron concentration is not adequately understood (Cook. 1990). The increased nutritional demand of pregnancy and the reduction of iron bioavailability due to the calcium may have contributed to the reduction in iron levels in the maternal and foetal tissues. We have found only one published study (AIt et al., 1938) in which liver iron Ic',els arc comparcd ~ith haemoglobin levels in rats and their offspring following feeding of the dams with iron-deficient diets to induce iron deficiency. In this one-generation. two-litter study, 21 days after both the first and second pregnancy, the maternal liver iron content had decreased to 20% of normal. Although the haemoglobin content was only slightly less than normal after the first pregnancy, after the second pregnancy it had decreased from 14.4mg/100ml
in the controls to 9.1 mg/100 ml in the iron-deficient rats. In the newborn offspring of the first litters, the total-body iron content was reduced by approximately 50°/. and the haemoglobin content was normal, but in the newborn offspring of the second litters the total-body iron content had decreased to 25*/0 of normal and there was a 50% reduction (6.6 l'. 13.1 mg/100ml) in haemoglobin c o n t e n t a level indicative of anaemia. To summarize, AIt et al. (1938) had shown that the impact of an iron-deficient diet on offspring could be greatest in the second litter, in which iron-deficient anaemia was observed. In our study, the foetuses from dams fed diets containing 0.75, 1.00 or 1.25"/o calcium had 87, 79 and 65%, respectively, of the iron content of foetuses from dams on the control dict. and the maternal livers of female rats on excess calcium had 84, 80 and 74%, respectively, of the iron content of females on the control diet. It is conceivable that if rats are fed excess calcium in nutritionally adequate diets for multiple generations, an iron-deficient anaemia could eventually develop in the offspring. Dietary calcium has been known to affect magnesium absorption in male (Likuski and Forbes, 1965) and female rats (Hock et aL, 1988). Recently, Hock et al. (1988) studied the effects in female Riv-TOX rats fed a semi-purified diet containing 0.40% phosphorus for 28 days. An increase in dietary calcium from 0.50 to 0.75% was associated with a significant increase in the faecal excretion of magnesium and a decrease in the elticiency of magnesium absorption, but had no effect on urinary magnesium output or magnesium retention. It is not clear from the literature (Elin, 1988: Maclntyre, 1967) what actual mechanism is involved and whether calcium ir~tcrfcrcs with magnesium absorption because of common transport carrier proteins. Although a calcium-magnesium interaction has been shown to exist in non-pregnant rats, no effects on tissue magnesium levels have been observed unless the rats were fed excessive amounts of dietary calcium. In a 28-day balance study with male Sprague-Dawley rats, Forbes (1964) showed that there was no effect on the magnesium content of the kidney when calcium was increased from 0.39
Calcium carbonate-mineral interaction to 0.78% (0.36% phosphorus) in an egg-protein, semi-purified diet. In another 28-day balance study, McAleese and Forbes (1961) fed male SpragueDawley rats a casein-based, semi-purified diet with a normal magnesium content of 500 ppm. When the calcium content was increased from 0.40 to 0.80%, there was no observed effect on the levels of magnesium in blood serum or bone ash or in the magnesium concentration in the kidney. Bodgen et al. (1992) fed male Sprague-Dawley rats AIN-76A diets containing 0.50 or 2.50% calcium for I yr. The excessive level of dietary calcium resulted in a significant decrease in femur and plasma magnesium but had no effect on the magnesium content of the kidney. Our data are consistent with the previously reported effects of moderate increases in dietary calcium on the magnesium contents of the femur and kidney of non-pregnant rats. Although the significant increase in the magnesium content of the liver at the 1.00 and 1.25% calcium levels (Table 4) was dosedependent, it is not readily explained from the current literature. The effects of the calcium-magnesium interaction were more pronounced in pregnant rats than in non-pregnant rats. Our data show a significant decrease in the maternal kidney content at 1.0 and 1.25% dietary calcium (Table 5) and in the wholebody magnesium of the foetuses at 1.25% (Table 7). it is of interest to compare our results with those of a study by Gunther et aL (1981), who determined the reproductive ctl'ccts of feeding varying levels of dietary magnesium to pregnant rats. Decreasing the dietary magnesium content from 360 ppm (control) to 260, 160, 110 or 40 ppm resulted in dose-dependent increases in the percentage of resorptions and in decreases in foetal weight. In our study, a moderate increase in dietary calcium did affect the levels of magnesium in the maternal kidney and the foetuses, but these effects were not associated with an increase in foetal rcsorptions or a decrease in foetal weight (Shackelford et al., 1993). Other calcium-nutrient interactions that have been studied include effects of calcium on the utilization of calcium and phosphorus. In the study by H o e k e t al. (1988) with female rats, increasing dietary calcium from 0.50 to 0.75% significantly increased the faecal excretion of calcium and phosphorus and strongly inhibited the urinary excretion of phosphorus without affecting the urinary calcium excretion. The group mean whole-body retention of calcium was increased, but that of phosphorus was lowered. Our data indicate that retention of calcium, but not of phosphorus, increased in the femurs of non-pregnant and pregnant rats (Table 6), but that the calcium and phosphorus contents of the kidneys were not affected (Table 5). Our findings for mineral levels in the femur are similar to those of Bodgen et al. (1992), who reported a significant increase in femur calcium when male rats were fed 2.5% calcium (v. controls, 0.50% calcium) for l yr. However, Hock et aL (1988)
261
reported a non-significant decrease in the calcium and phosphorus contents of the kidney, and Bodgen et al. (1992) reported a non-significant decrease in the calcium content of the kidney. The reason for the dose-dependent increase in the phosphorus content of the livers of non-pregnant rats which was observed in our study (Table 4) is not clear. Numerous experimental approaches have been developed to study the effects of excess dietary calcium on zinc absorption in casein-based, semipurified diets. In a study with male Sprague-Dawley rats, Forbes and Yohe (1960) failed to show a specific effect of calcium on zinc absorption or the zinc requirement. Heth et al. (1966) reported that in male Holtzman rats fed i.45% calcium in the diet (in comparison with 0.70% for the controls) for I month the absorption of dietary ~SZn was significantly decreased when the phosphorus content was increased to i%, but was not affected at concentrations of 0.3-0.5% phosphorus. Huber and Gershoff (1970) reported that 1.3% calcium in the diet fed to male Charles River rats significantly decreased 6~Zn absorption if the diet was zincdeficient, but had no effect with a control diet (16.3 ppm zinc). The addition of 2% phytic acid to the casein-based, semi-purified diet fed to male Sprague-Dawley rats for 18 days was associated with a dose-dependent decrease in the amount of zinc absorbed as the calcium level was increased from 0.4 to 0.8 or 1.2% (Likuski and Forbes, 1965); the addition of extra calcium to diets without phytic acid (control diets) did not significantly decrease the percentage of zinc absorbed. Forbes (1964) observed a decrease in zinc absorption resulting from increased calcium levels (0.40-0.80%) when soyabean protein, semi-purified diets were fed to male Spraguc-Dawley rats. in contrast extra calcium had no effect on zinc absorption when rats were fed semi-purified diets containing whole egg white protein which, unlike soyabean protein, does not contain phytic acid. Oberleas et al. (1966) proposed the formation of a Zn-Ca-phytic acid complex which removes zinc from solution in the intestinal lumen. Thus, both the protein source and dietary calcium concentration can affect zinc absorption. Although the results of previously published studies have indicated that excess calcium in a caseinbased, semi-purified diet does not affect zinc absorption, the results of our study may indicate a possible effect of increased dietary calcium in a casein-based, semi-purified diet on some aspect of zinc metabolism. The dose-related increase in the liver zinc content of non-pregnant rats is not consistent with current literature reports on additional dietary calcium. A dose-related negative linear trend was found in the zinc content of the kidney of pregnant rats without a concurrent effect on the foetal zinc content. This increased tissue catabolism in the kidney of pregnant rats may have resulted from the increased nutritional
262
M.E. SR^Crd~LFORDet al.
demands of the mother and foetus that are associated with a reduced zinc availability due to increased dietary calcium. The mechanism of reduced zinc availability is not clear from the literature, because zinc and calcium do not share common absorption paths (Cousins, 1985). However, in the study of Rogers et al. (1985), under simulated conditions of feeding pregnant rats reduced levels of dietary zinc, the maternal kidney and whole-body foetus had significantly reduced zinc contents, whereas the liver zinc content was not affected. The decreased zinc content in the maternal kidney (Table 5) and the normal levels of zinc in the maternal liver (Table 4) that were observed in our study appear to be consistent with a pattern of maternal tissue catabolism associated with reduced zinc availability due to increased dietary calcium. There are no published studies that have investigated the effects of additional dietary calcium on manganese or copper absorption. In our study the manganese levels were not affected. Unexpected findings were the significant decreases in the copper content of the kidneys of non-pregnant rats (Table 5) and the slight but significant decrease in the wholebody foetal copper content (Table 7) at the two highest dose levels. Our data may indicate the presence of a calcium-copper antagonism. The microscopic lesions diagnosed in the liver, heart and kidney were considered spontaneous in nature and/or incidental. The feeding of a purified or semi-purified diet to rodents has been shown to induce spontaneous lesions that are not found in young rodents fed diets composed of natural ingredients. These spontaneous lesions include myocardial damage (Mitchell et al., 1989), periportal fatty change (Medinsky et al., 1982) and renal mineralization (Woodard, 1971a,b; Woodard and Jee, 1984a,b). in our study the distribution and degree of severity of the renal mineralization (synonyms include ncphrocalcinosis, nephrolithiasis and calculosis) appear to be analogous to the kidney lesions that have been reported to have been caused by feeding rodents purified or semi-purified diets. There are two findings in the incidence of renal mineralization for which there was no obvious explanation. The first is that the baseline female rats had the lesion (Table 2). Since these animals did not eat the AIN-76A diet, other dietary factors may have contributed to the development of this lesion. The second is that the incidence was lowered in the non-pregnant rats fed the 0.75% calcium diet. Hock et al. (1988) reported that renal mineralization was essentially absent in female rats on a semi-purified diet containing 0.75% calcium, although it did occur in those fed the same diet with 0.50% calcium. Published data have indicated that whole-kidney function is largely unaffected by renal mineralization (AI-Modhefer et al., 1986: Stonard et at., 1984) because of the vast reserve capacity of the kidney. Thus, we would not expect the renal mineralization
to have contributed to the tissue mineral findings in our study. In summary, the most significant findings of this study were in the mineral analyses of the foetuses. The dose-related decreases in the foetal iron content were accompanied by dose-related decreases in the liver and kidney iron contents of non-pregnant and pregnant rats. The dose-related decrease in the magnesium content of the foetuses was accompanied by a dose-related decrease in the magnesium content of the maternal kidney, but not in that of the livers of the pregnant rats. A dose-related decrease in phosphorus was observed in the foetuses, but there was no effect on the phosphorus content of the livers and kidneys of the pregnant rats. A well known effect of additional calcium in the diet is a decreased absorption of iron. magnesium and phosphorus. Our results indicate that additional calcium in the diet decreased the bioavailability of iron, magnesium, phosphorus and copper. The increased nutritional demand of pregnancy and the reduced mineral bioavailability due to the effects of calcium resulted in effects on mineral levels in the foetuses, and, in some cases, also on those of the maternal tissues. Acknowledgements --We express o u r appreciation to Catherine Nelson and James I. Rorie for their excellent technical support. Thanks are also due to the Technical Editing Branch, Food and Drug Administration, for assistance in preparing the manuscript.
REFERENCES AI-Modhefer A. K. J., Atherton J. C., Garland H. O., Singh H. J. and Walker J. (1986) Kidney function in rats with corticomedullary nephrocalcinosis: effects of alterations in dietary calcium and magnesium. Journal of Physiology 380, 405-414. Alt H. L. (1938) Iron deliciency in pregnant rats. American Journal of DLseases of ChiMren 56, 975 -984. Barton J. C., Conrad M. E. and Parmley R. T. (1983) Calcium inhibition of inorganic iron absorption in rats. Gastroenterology 84, 90 10l. Bogden J. D., Gertner S. B., Christakos S., Kemp F. W., Yang Z., Katz S. R. and Chu C. (1992) Dietary calcium modifies concentrations of lead and other metals and renal calbindin in rats. Journal o[" Nutrition 122, 1351-1360. Consumer Reports (1988) The truth about calcium. Consumer Reports May, 288-291. Cook J. D. (1990) Adaptation in iron metabolism. American Journal of Clinical Nutrition 51, 301 ~308. Cousins R. J. 0985) Absorption. transport, and hepatic metabolism of copper and zinc: special reference to metallothionein and ceruloplasmin. PhysiohJgical Reviews 65, 238-309. Davis G. K. 09591 Effects of high calcium intakes on the absorption of other nutrients. Federation Proceedings 18, 1119-1124. Davis G. K. 0967) Excess minerals in the diet. Journal of the American Dietetic Association 51, 46-50. Dolan S. P.. Sinex S. A.. Capar S. G., Montaser A. and Clifford R. H. (1991) On-line preconcentration and volatilization of iodine for inductively coupled plasma atomic emission spectrometry. Analytical Chemistry 63, 2539-2542.
Calcium carbonate-mineral interaction Elin R. J. (1988) Magnesium metabolism in health and disease. Disease-a-Month 34 (4), 161-219. Forbes R. M. (1963) Mineral utilization in the rat. I. Effects of var)ing dietary ratios of calcium, magnesium, and phosphorus. Journal of Nutrition 80, 321-326. Forbes R. M. (1964) Mineral utilization in the rat. 11. Effects of calcium, phosphorus, lactose, and source of protein in zinc-deficient and in zinc-adequate diets. Journal of Nutrition 83, 225-233. Forbes R. M. and Yohe M. (1960) Zinc requirement and balance studies with the rat. Journal of Nutrition 70, 53-57. Gunther T.. lsing H.. Mohr-Nawroth F+. Chahoud 1. and Merker H. J. (1981) Embryotoxic effects of magnesium deficiency and stress on rats and mice. Teratology 24, 225-233. Heth D. A., Becket W. M. and Hoekstra W. G. (1966) Effect of calcium, phosphorus, and zinc on zinc-65 absorption and turnover in rats fed semipurified diets. Journal of Nutrition 88, 331-337. Hoek A. C., Lemmens A. G., Mullink J. W. M. A. and Beynen A. C. (1988) Influence of dietary calcium: phosphorus ratio on mineral excretion and nephroealcinosis in female rats. JournalofNutrition 118, 1210-1216. Huber A. M. and Gershoff S. N. (1970) Effects of dietary zinc and calcium on the retention and distribution of zinc in rats fed semipurified diets. Journal of Nutrition !00, 949 +954. tlurley U S. (1977) Nutritional deficiencies and excesses. In thmdbook +~1"Teratoh~g)'. Edited by J. G. Wilson and F. C. Fraser. Vol. I. pp. 261-308. Plenum Press, New York. Likuski I1. J. A. and Forbes R. M. (1965) Mineral utilization in the rat. IV. Effects of calcium and phytic acid on the utilization of dietary zinc. Journal of Nutrition 85, 230 234. McAIcese D. M. and Forbes R. M. (1961)The requirement and tissue distribution of magnesium in the rat as influenced by environmental temperature and dietary calcium. Journal o[ Nutrithm 114, 94 106. Maclntyre I+ (1967) Magnesium metabolism. Ad~,ances in blternal Medicine 13, 143 154. Medinsky M. A., Dua P. N., Jenkins M. G. and Grundel E. (1982) Development of hepatic lesions in male Fischer 344 rats fed AIN-76A purified diet. Toxicoh~gy and Applied PharnlacohJ,g;r 62, 11 I 120.
263
Mitchell G. V., Dua P. N., Jenkins M. G. and Grundel E. (1989) Nutritional and pathological changes in male and female rats fed modifications of the AIN-76A diet. Food and Chemical Toxicology 27, 185-191. Oberleas D., Muhrer M. E. and O'Dell B. L. (1966) Dietary metal--complexing agents and zinc availability in the rat. Journal of Nutrition °AI, 56-62, Padmore J. M. (1990) Animal feed. In Official Methods of
Analysis of the Association of Official Analytical Chemists. Edited by K, Helrich. 15th Ed. pp. 84-88. AOAC, Arlington, VA. Rader J. i., Wolnik K. A., Gaston C. M.. Celesk E. M., Peeler J. T., Fox M, R. S. and Fricke F. L. (1984)Trace element studies in weanling rats: maternal diets and baseline tissue mineral values. Journal of Nutrition 114, 1946-1954. Rogers J. M., Keen C. L. and Hurley L. S. (1985) Zinc deficiency in pregnant Long-Evans Hooded rats: teratogenicity and tissue trace elements. Teratology 31, 89-100. Shackelford M. E., Collins T. F. X.. Welsh J. J.. Black T. N., Ames M. J.. Chi R. K. and O'Donnell M. W. (1993) Foetal development in rats fed AIN-76A diets supplemented with excess calcium. Food and Chemh'al Toxicology 31,953 961. Sokal R. R. and Rohlf F. J. (1981) Biometry, pp. 179-245. W. H. Freeman and Company. San Francisco. CA. Stonard M. D., Samuels D. M. and Lock E. A. (1984) The pathogenesis and effect on renal function of nephrocalcinosis induced by different diets in female rats. Food and Chemk'al Toxh'ohJgy 22, 139 146. Woodard J. C. (1971a) A morphologic and biochemical study of nutritional nephrocalcinosis in feln;tle rats fed semipurified diets. American Journal of Pathohlg)" 65, 253 -268. Woodard J. C. (1971b) Relationship between the ingredients of semipurilied diets and nutritional nephrocalcinosis of rats. Amerh'an Journal of PathobJgy 65, 269 -278. Woodard J. C. and Jee W. S. S. (1984a) Effects of dietary calcium, phosphorus, and magnesium on intranephronic calculosis in rats. Jourmd of Nutrition 114, 2331 -2338. Woodard J. C. and Jee W. S, S. (1984b) Effect of diet and intranephronic calculosis on bone modeling and parathyroid volume in rats. Journal oj" Nutrition 114, 2339 2352.
0278-6915(93)E0011-W
Fd Chem. Toxic. Vol. 32, No. 3, pp. 255-263. 1994 Elsevier Science Ltd. Printed in Great Britain 0278-6915/94 $7.00 + 0.00
MINERAL INTERACTIONS IN RATS FED AIN-76A DIETS WITH EXCESS CALCIUM M. E. SHACKELFORD*, T. F. X. COLLINS,T. N. BLACK, M. J. AMES,S. DOLAN, N. S. SHEIKH, R. K. CHi and M. W. O'DONNELL Center for Food Safety and Applied Nutrition, US Food and Drug Administration, Washington, DC 20204, USA
(Accepted 23 August 1993) Abstract--The effects of moderate increases in dietary calcium on maternal and foetal mineral interactions were studied in Charles River CD/VAF Plus rats. Female rats were given 0.50, 0.75, 1.00or 1.25% dietary calcium as calcium carbonate in AIN-76A diets for 6 wk before mating, during mating and for 20 days of gestation. Inductively coupled argon plasma-atomic emission spectrometry was used to determine mineral levels in the tissues of non-pregnant rats after 42 days on the diets, in the tissues of pregnant rats on day 20 of gestation and in the whole body of day-20 foetuses. The femurs of the non-pregnant and pregnant rats had a dose-related linear increase in calcium content, in livers of the non-pregnant rats, dose-related linear increases in the phosphorus, zinc and magnesium content were observed, but there was a dose-related decrease in the iron content. There were dose-related linear decreases in the iron and copper contents of the kidneys from the non-pregnant rats. in pregnant rats dose-related linear decreases were observed in the iron content of the liver and in the zinc, iron and magnesium contents of the kidney. The foetuses from rats given a moderate increase in dietary calcium had dose-related decreases in the whole-body contents of phosphorus, iron, copper and magnesium.
INTRODUCTION
In 1959 the National Institutes of Health sponsored a symposium on the effects of high calcium intake because of concern that the level of dietary calcium had increased to the extent that it might be a nutritional health hazard (Davis, 1967). One of the purposes of the symposium was to evaluate the effects of calcium intake on the utilization of other nutrients (Davis, 1959). In more recent years, public health concern over the prevention of osteoporosis has led to increased fortification of foods with calcium and sales of calcium supplements have increased from $47 million in 1983 to $200 million in 1987 (Consumer Reports, 1988). Since the symposium in 1959, mineral balance experiments have been conducted to study the effects of dietary calcium and/or phytic acid on the utilization of zinc, magnesium and phosphorus (Forbes, 1963 and 1964; Likuski and Forbes, 1965). Likuski and Forbes (1965) did an 18-day balance study with male Sprague-Dawley rats fed nutritionally adequate casein-based, semi-purified diets with or without 2% phytic acid. Calcium, as CaCO 3 was added to give levels of 0.4, 0.8 and i.2%. Dietary calcium decreased zinc absorption from diets containing phytic acid,
*To whom correspondence and reprint requests should be addressed at: US Food and Drug Administration (HFS507). 8301 Muirkirk Road, Laurel, MD 20708, USA. Abbreviations: ANOVA = analysis of variance: LSD test = least significant difference test.
but did not significantly lower the amount of zinc absorbed from diets without phytic acid. Absorption of magnesium and phosphorus was decreased in a dose-related manner in the presence or absence of phytic acid. A relationship between dietary calcium content and iron absorption has been postulated for many years, it is well documented that in rats high calcium intake diminishes the absorption of inorganic iron (Barton et al., 1983). By using isolated intestinal loops it was shown that increasing the calcium concentration (1 to 100mM CaCI2) decreased the entry of radiolabelled iron into the microvilli of intestinal epithelial cells. To test the significance of this in animals, male Wistar rats were fed diets containing 1.55-1.75% calcium (high calcium) or 0.63-0.68% calcium (control) for 4wk. The diets contained 0.45% phosphorus and casein as a protein source; they were either iron replete (85 mg/kg) or iron deficient (21 mg/kg). In rats fed the iron-replete diets, those receiving the high-calcium diets developed decreased marrow iron stores, but had no significant change in microhaematocrit, serum iron concentration or total iron-binding capacity of the serum. Among rats fed the mildly iron-deficient diets, those fed the high-calcium diets had significantly lower serum iron concentrations and decreased marrow iron stores. This condition was shown to result from a calcium-related inhibition of iron absorption but not from an increased gastro-intestinal transit. The study demonstrated the potential of a high-calcium diet to induce or accelerate iron deficiency.
255
M.E. SHACKELFORDet al.
256
Although it is known that excess calcium intake can interfere with mineral utilization, no animal studies were found in which the added physiological stress of pregnancy on these calcium-mineral interactions was carefully examined. In the study reported here, the effects of feeding 0.50 (control), 0.75, 1.00 or 1.25% dietary calcium on foetal and adult mineral levels were investigated to determine whether excess dietary calcium could perturb tissue mineral levels. All diets were nutritionally adequate AIN-76A diets with casein as the protein source and without dietary phytic acid. MATERIALS AND METHODS
Experimental design. Female and male CD-VAF/ Plus Charles River rats were obtained from Charles River Laboratories, Inc. (Wilmington, MA, USA); when received, the females were 52 days old and the males were 44 days old. At the end of a 6-day quarantine period, 15 of the female rats were randomly selected for evaluation of tissue mineral status and histopathology before assignment to the dietary groups, and 276 female rats were assigned to the control or test diets by a stratified random method according to body weight. The rats in the four dietary groups received 0.50 (control), 0.75, 1.00 or 1.25% calcium in powdered AIN-76A for 6wk before mating, throughout mating and for 20 days of gestation. Males were fed Purina Chow 5002 (Purina Mills, inc., Richmond, IN, USA) except during mating, when they also received the control or test diets. The AIN-76A diets were prepared by BIOSERV (Frenchtown, N J, USA). Homogeneity and concentrations of calcium, phosphorus, magnesium, manganese, zinc, iron and copper were assayed by the New Jersey Feed Laboratory, Inc. (Trenton, N J, USA) using the methods of the Association of Official Analytical Chemists (Padmore, 1990). Full details of the study design, including housing, food and water consumption, diets, mating, and performance of caesarean sections have been described in the previous paper (Shackelford et al., 1993). The female rats that were evaluated for tissue mineral status and histopathology before assignment to the dietary groups are referred to below as baseline animals. After 6 wk on the control or test diets, 60 females (15 per group) were randomly selected for evaluation of tissue mineral status and histopathology. The remaining females were mated and the sperm-positive females were continued on the appropriate diets for 20 additional days. 15 pregnant rats per dietary group were randomly selected for removal of tissues when caesarean sections were performed on day 20 of gestation. From these 15 pregnant rats, 12 were randomly selected for evaluation of the mineral status of the foetuses. Two foetuses per litter (one from each uterine horn) were randomly selected.
Removal of tissues. The rats were killed by carbon dioxide asphyxiation, and then exsanguinated through the posterior vena cava to remove blood from the organs and tissues. For pregnant rats from which the foetuses were removed, the uterine arteries were clamped before exsanguination through the posterior vena cava to prevent withdrawal of the blood supply to the foetuses before they were removed from the uterus. All organs and tissues were examined grossly, and any abnormal findings were recorded. The organs and tissues (liver, left kidney and both femurs) and foetuses were placed in polyethylene bags and stored at - 4 0 " C . Terminal body weights and selected organ weights were obtained at the time of autopsy. At the time of autopsy, heart, liver and kidneys were weighed (left and fight kidneys were weighed separately). Histological methods. The following organs and tissues from non-pregnant and pregnant rats were placed in 10% neutral buffered formalin: heart, a part of the liver, the fight kidney and bone (right humerus and sternum). The tissues were later processed, embedded in paraffin, sectioned at 5 pro, stained with haematoxylin and eosin and examined histopathologically. The bone specimens were decalcified in a mixture of formic acid, hydrochloric acid and distilled water and subsequently processed as above. Special stains included the following: (I) Periodic Acid Schiff reaction with and without diastase for detecting the presence of glycogen on selected liver sections; (2) Von Koss's stain for calcium on selected liver, bone and kidney sections; (3) osmium tetroxide stain for fat on selected liver sections. The lesions were graded as minimal (Grade 1), mild (Grade 2), moderate (Grade 3) or severe (Grade 4). Mineral analysis. Determinations of mineral levels in specific tissues were made before dietary treatments (baseline values), after 6 wk on the diets and on day 20 of gestation. Calcium, phosphorus, zinc, iron, copper, magnesium and manganese were determined in the livers and left kidneys from non-pregnant and pregnant rats, and in the foetuses. Calcium, phosphorus and magnesium were determined in the left femurs from the adult animals. A portion of the liver, left kidney and femur, and whole pups, were weighed and wet-digested in mixtures of nitric-perchloric or nitric-perchloricsulfuric acids (Rader et al., 1984). Final acid concentrations were 10% perchloric acid for femurs and 10% sulfuric acid for livers, kidneys and whole pups. Tissue digests were analysed by inductively coupled argon plasma-atomic emission spectrometry on the ARL 3580 (Applied Research Laboratories, Dearborn, MI, USA) as described previously (Dolan et al., 1991). The following analytical wavelengths (A) were used for analysis: Ca, 3933.7; Cu, 3247.5; Fe, 2599.4; Mg, 3838.3; Mn, 2576.1; P, 1782.9; and Zn, 2138.6. Calcium was determined in foetuses and diluted femurs at 3179.3 ~,. Portions of National Bureau of Standards and Technology (formerly National
257
Calcium carbonate-mineral interaction Table I. Organ weights and final body weights for female rats fed diets containing excess calcium Weight (g) in rats fed calcium at: Tissue
0.50% (control)
Liver Right kidney Left kidney Heart
Finalbody weight.,. Liver Right kidney Left kidney Heart
Finat body weight...
0.75%
12.11 _+0.41 I. 10 -+ 0.03 1.09 4- 0.03 1.09 ± 0.03 311.6 ± 8.3
Nora-pregnant rats 11,98+0.51 1.10 -+ 0.02 1.07 -+ 0.02 1.04 ± 0.02 311.3_+8.7
17.22 _+ 0,52 1.19 _+ 0.04 1.17 + 0.05 1.15 _+0.02 434.2 ± 8.1
Pregnant rats 17.02 _+ 0.33 1.15 4- 0.0l 1.13 ± 0.02 1.19 _+0.03 446.1 -+ 7.6
1,00%
1,25%
I 1.39 -+ 0,47" 1.09 ± 0.04 1.05 4- 0.03 1.08 -+ 0.03 313.3+8.6
10.82-+0.28 .'° 1.06 ± 0.02 1.04 4- 0.02 1.05 _+0.02 306,5 -+ 5.8
16.30 _+0.56 1.10 -+ 0.03 1.07 _+0.03 1.17 + 0.03 429.5+12.5
16.58 _+0.47 1.15 -+ 0.02 1.13 _+0.03 1.20 _+0.02 443.6 -+ 8.2
Values are means ± SEM for groups of 15 rats. Asterisks indicate significant differences between control and test values ( ' P ~<0,05: **P ~ 0.01). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
Bureau of Standards) Standard Reference Material Bovine Liver 1577 were digested and analysed for the same elements. Values fell within 100 +_ 2% of certified values for all elements of interest. Statistical analysis. All statistical analyses were performed by the Division of Mathematics, US Food and Drug Administration. The terminal body weights were analysed by analysis of variance (ANOVA) and tissue weights were analysed by analysis of covariance using an adjustment for terminal body weight. This was followed by a protected least significant difference (LSD) test (two-tail) for pairwise comparison of treated groups with the control group. Means and standard deviations were calculated for levels of the elements in the rat tissues and foetuses. Statistical analysis involved ANOVA followed by the LSD test (two-tail) for pairwise comparison of the treated groups with the control group (Sokal and Rohlf, 1981). if the variances were homogeneous, a dose-response regression was performed. A dose-related linear trend was said to be present if the P value of the lincarity component was <0.05 and, at the
same time, the departure P value was />0.05. If a trend was present, the direction of the trend (increase or decrease) was noted. RESULTS
Organ weights are summarized in Table I. When the organ weights of the pregnant animals were adjusted for the terminal body weights, there were no significant differences between the controls and treated groups in any organ weight. For the organs from the non-pregnant animals, the liver weight of rats in the two highest dose groups was significantly lower than that of the controls. The average daily food consumption of non-pregnant animals between days 0 and 42 increased significantly for the !.00 and 1.25% dietary groups in comparison with the control groups, but there was no significant difference in weight gain over this same time interval (Shackelford et al., 1993). Thus, the decrease in the adjusted liver weight is not attributable to a decrease in food consumption or weight gain.
Table 2. Summary incidence of female rats with renal mineralization Category of ratsf Baseline Non-pregnant Non-pregnant Non-pregnant Non-pregnant Pregnant Pregnant Pregnant Pregnant
Dose level (% calcium in diet)
Renal mineralization* Incidence
Mean severity
NA 0.50 0.75 1.00 1.25 0.50 0.75 1.00 1.25
6/15 (40%) 9/I 5 (60.0%) I/I 5 (7.0%) 8il 5 (53.3%) 7/I 5 (47.0%) I 0/15 (67.0%) 5/I 5 (33.3%) 7/15 (47.0%) 5/15 (33.3%)
1.3 1.6 1.0 1.4 I. I 1.5 1.6 1.4 1.0
NA - not applicable "The values for renal mineralization are expressed as the incidence (the number of animals affected out of the total number of animals examined, parcentage in parentheses), and the mean severity (the average grade/number of tissues affected). tBaselin¢ rats wcrc killed for analysis before the feeding study began. Non-pregnant rats were fed the control or calcium enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
258
M.E. SHACKELFORD et al.
Table 3. Mineral levels for non-pregnant animals before the feeding study was begun 1baseline rats) Mineral levels* in the: Mineral Liver Kidney Femur Ca 32.8+2.1 79.0±22.1 144.34-6.8 P 3303.94- 149.1 2834.6_+156.4 72.2_+3.1 Zn 27.0 _4-2.3 21.8 4. 1.6 ND Fe 173.6 4. 37.9 76.8 4- 17.0 ND Cu 4.0 4- 0.3 9.2 __.3.8 ND Mg 207.9 _+8.5 190.1 4- 10.7 3.0 4-0.2 Mn 1.94-0.2 0.64-0.1 ND ND = not done *The mineral content is expressed as pg, g fresh tissue weight for the kidney and liver, and as mg/g fresh tissue weight for the femur. Values are means _4-SD of singledeterminations of 15 specimens.
The histopathological findings included periportal hepatocellular vacuolation in the liver, mineralization in the kidney a n d chronic myocarditis in the heart. Hepatocellular vacuolation was frequently seen in the n o n - p r e g n a n t a n d p r e g n a n t animals, but it was not observed in the baseline rats. T h e incidence o f rats with hepatocellular vacuolation and the mean severity (mild to moderate) o f the lesion were generally similar in control and treated groups. The s u m m a r y incidence o f rats with renal mineralization is presented in T a b l e 2. T h e mineralization was observed p r e d o m i n a n t l y in the renal t u b u l a r epithelial cells at the corticomcdullary j u n c t i o n a n d in the inner stripe o f the outer medulla. Myocarditis, characterized by focal to multifocal myocardial degeneration, fibrosis and m o n o n u c l c a r cell infiltration, was observed in one bascline rat, one n o n - p r e g n a n t rat on the i . 0 0 % calcium dict, three p r e g n a n t rats on the control diet, a n d one p r e g n a n t rat in each o f two highest calcium dose groups. There was no observable histopathological c h a n g e in the bone or bone m a r r o w in any o f the rats. The mineral c o n t e n t s o f the kidney, liver a n d femur (Table 3) o f o u r baseline animals were similar to those reported by R a d e r et al. (1984). However, o u r
Mineral Ca P Zn Fe Cu
Mg Mn
results for the iron c o n t e n t o f the livers from the baseline rats (174 l~g/g fresh weight) were higher than those o f R a d e r et al. (1984), who reported liver iron values o f 4 0 - 5 9 / z g / g fresh weight. The data of Rader et al. (1984) were o b t a i n e d from 22- to 26-day-old male L o n g - E v a n s rats that had been b o r n to dams fed the N I H - 3 ! diet, whereas o u r data were o b t a i n e d from 58-day-old Charles River rats ( C D - V A F / P l u s ) that had been b o r n to females fed Purina Lab Chow. Thus, the female animals used in o u r study had normal tissue mineral c o n t e n t s at the onset o f the study. T h e mineral levels in the livers o f the rats fed diets with excess calcium are s h o w n in Table 4. In the n o n - p r e g n a n t rats, after 6 wk on the diets the copper a n d m a n g a n e s e c o n t e n t s o f the livers from the groups fed the high-calcium diets were not significantly different from those o f the control group. A d o s e - r e s p o n s e regression for p h o s p h o r u s , zinc and m a g n e s i u m showed a dose-related linear increase. F o r calcium, the d o s e - r e s p o n s e regression had borderline significance ( P = 0.0571). in contrast, a dose-related linear decrease in the iron content of the liver was observed. In the p r e g n a n t rats, the calcium, p h o s p h o r u s , copper, magnesium a n d m a n g a n e s e contents o f the livers o f those fed high-calcium diets did not differ signiticantly from the levels o f the 0.50% calcium (control) g r o u p (Table 4). The zinc c o n t e n t o f the liver was significantly decreased at thc 1.00% calcium level, but was similar to the control value at 0.75 and 1.25% calcium. T h e r e was no dose-related trend. The iron c o n t e n t o f the livers o f the p r e g n a n t rats, like that of the livers o f the n o n - p r e g n a n t rats, was significantly decreased in rats fed 0.75, 1.00 a n d !.25% dietary calcium. The trend of the dose-response regression for iron showed a dose-related linear decrease. Table 5 lists the mineral levels in the kidneys o f the rats fed diets with excess calcium. In the nonp r e g n a n t animals, after 6 wk on the diets, no signifi-
Table 4. Liver mineral levels for female rats fed diets with excess calcium Mineral levels (pg/g fresh weight of liver)'l"in rats fed calcium at: 0.50% (control) 0.75% 1.00% 1.25% 31.8 _4-3.8 2952.5 +_ 221.8 22.6 + 1.8 164.0 ± 37.0 4.3 + 0.5
193.3 ± 13.5 2.0+0.3
Non-pregnant rats 32.1 _+ 1.5 2965.5 ± 146.1 23.2 + 1.4 138.4 + 29.3* 4.3 ± 0.5
192,5 +9.1 2.1 +0.2
Zn
40.4 ± 2.6 3139.2 ± I00.I 23.2 ± I.I
Pregnant rats 42.5 _ 6.5 3134.8 _+.99.1 22.8 +_ 1.3
Fe
99.1 ± 26.6
78.1 .:l: 15.7 °0
Ca
P Cu Mg Mn
4.2 + 0.2 223.2 + 9.3 2.9 -t- 0.3
4.2 "1-0.3 222.3 4. 5.8 2.9 + 0.2
34.6 ± 4.2" 3099.4 ± 216.7" 23.9 + 2.2" 132.2 + 38.6 °" 4.6 __.0.7
203.0 + 12.3" 2.2+0.3 39.9 ± 3.1 3122.8 _+ 138.8 22.3 ± 0.9* 63.7 + 9 . 7 °**
4.2 + 0.3 222,1 + 9.1 2.9 + 0.3
33.7 ± 4.3 3102.4 _+ 120.9" 23.7 ± 0.9 122.4 +_ 24.5"** 4.4 ± O,2 21)1.9 ± 8.6*
2.1 _.+0,3 41.0 ± 3,6
3114.5 ± 171.7 23.4 ± 1,6 65.9 ± 16.0 **0
4.2 + 0.4 220.3 ± 12.5 2.7 ± 0.4
tValucs are means _+SD of single determinations of 15 specimens. Asterisks indicate significant differences between control and test values (*P ~ 0.05; * ' P ~0,01; * * ' P ~ 0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6wk. then mated and continued on the appropriate diets until day 20 of gestation.
Calcium carbonate-mineral interaction
259
Table 5. Kidney mineral levels for female rats fed diets with excess calcium Mineral levels (mg/g fresh weight of kidney)* in rats fed calcium at: Mineral
0.5000 (control)
No. ~'~ec'imens. Ca P
Zn Fe Cu Mg Mn No. of specimens.
15 (14)
76.6 __. 19.2 2733.2 ± 212.7 21.5 ± 2.1 101.3 + 19.7 8.3 + 1.7 176.6 + 14.0 0.8 ±0.1 15(14)
84.8 ± 30.5 2587.8 ± 173.0 20.8 _+ 1.3 77.4 ± 10.8 7.3 ± 3.0 166.7 ± 11.2 0.9 +0.1
Ca P Zn Fe Cu Mg Mn
0.75*:, Non-pregnant rtts 15
67.2 + 4.0 2736.84- 146.7 21.4 ± 1.4 94.2 + 17.4 8.0 ± 1.8 176.1 + 9.2 0.8 ±0.1 Pregnant rats 15
82.1 _+34.8 2528.9_ 69.9 20.2 ± 1.4 70.3 _+9.6* 6.5 + 1.9 164.2 ± 5.0 0.8 ±0.1
1.00%
1.25%
15 (14)
70.9 ± 4.7 2777.5± 150.1 21.6__. 1.6 81.0 + 16.1"* 7.6 ___1.8 178.6 + 9.4 0.8 ±0.1
/5 (14)
77.8 + 18.2 2841.0± 255.4 21.7 + 2.0 90.7 4- 19.6 6.6 + 1.3"* 182.6 + 15.8 0.8 ± 0.0
15
15
75.0 ± 9.5 2501.0 ± 78.3 19.6 ± 1.1" 62.9 _+6.8"** 6.4 ± 2.4 159.6 +_4.8* 0.8 ±0.1
75.3 ± 5.8 2508.6 4- 129.0 19.6 ± 1.1" 62.6 4- 7.4*** 6.6 ± 3.2 158.6 ± 8.7** 0.8 4- 0.1
tValues are means 4- SD of single determinations of the number of specimens shown (where outliers in calcium analysis reduced N values these reduced N values are given in parenthesis). Asterisks indicate significant differences between control and test values. (*P ~<0.05; **P ~<0.01; "'*P ~<0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk. then mated and continued on the appropriate diets until day 20 of gestation.
T h e m i n e r a l levels in the f e m u r s o f the rats are s h o w n in T a b l e 6. In p r e g n a n t rats, a l t h o u g h the decrease in the m a g n e s i u m c o n t e n t w a s significant in the g r o u p fed 0 . 7 5 % calcium, there w a s no doserelated trend. T h e r e were n o signilicant effects o n the m a g n e s i u m c o n t e n t s o f the f e m u r s f r o m the n o n p r e g n a n t rats. F o r the calcium c o n t e n t o f the f e m u r s f r o m the n o n - p r e g n a n t a n d p r e g n a n t a n i m a l s , the t r e n d o f the d o s e - r e s p o n s e regression s h o w e d a doserelated linear increase. T h e r e w a s no effect o n the p h o s p h o r u s c o n t e n t o f the femurs. T h e m i n e r a l c o n t e n t s o f f o e t u s e s f r o m d a m s fed excess dietary c a l c i u m are s u m m a r i z e d in T a b l e 7. C a l c i u m , zinc a n d m a n g a n e s e c o n c e n t r a t i o n s in the foetuses were similar in the c o n t r o l a n d h i g h - c a l c i u m g r o u p s . Both the p h o s p h o r u s a n d m a g n e s i u m c o n t e n t s were signilicantly decreased in the 1.25% g r o u p . A l t h o u g h the decrease in c o p p e r c o n t e n t o f the f o e t u s e s w a s small, it w a s significant in the g r o u p s fed 1.00 o r 1.25% calcium. Foetal iron c o n t e n t
c a n t differences were seen in the calcium, p h o s p h o r u s , zinc, m a g n e s i u m o r m a n g a n e s e c o n t e n t s o f the kidneys f r o m g r o u p s fed h i g h - c a l c i u m diets, in c o m p a r i s o n with levels in the g r o u p fed 0 . 5 0 % calcium. T h e decrease in the iron c o n t e n t w a s significant at the 1.00% level a n d the decrease in the c o p p e r c o n t e n t w a s significant at the 1.25% c a l c i u m level. T h e trend o f the d o s e - r e s p o n s e r e g r e s s i o n for these t w o m i n e r a l s s h o w e d a d o s e - r e l a t e d linear decrease. In the p r e g n a n t rats, n o significant differences were o b s e r v e d in the calcium, p h o s p h o r u s , c o p p e r a n d m a n g a n e s e c o n t e n t s o f the kidneys f r o m a n i m a l s fed h i g h - c a l c i u m diets in c o m p a r i s o n with the c o n t r o l s . T h e zinc a n d m a g n e s i u m c o n t e n t s o f the kidney were decreased significantly in the g r o u p s led the t w o highest d o s e s o f calcium, a n d the iron c o n t e n t w a s significantly decreased in the p r e g n a n t rats fed 0.75, 1.00 o r 1.25% dietary calcium. T h e trend o f the d o s e - r e s p o n s e regression for these three m i n e r a l s s h o w e d a d o s e - r e l a t e d linear decrease.
Table 6. Femur mineral levels for female rats fed diets with excess calcium Mineral levels (mg/g fresh weight of femur)* in rats fed calcium at: Mineral No. of specimens,..
0.50°/, (control) 15
Ca
165.3 ± 7.2
P Mg
80.7 + 3.2
3,1 -t-0.2
0.75*/0 Non~prc~nan! rats 14
166,0 ± 5.2 80.8 +_2.7 3.2+0.2
Pll'e~nan| No. of specimen... Ca P Mg
15 168.4 + 4.0 82.0 + 2.2 2.9 ± O.I
1.00% 15
170.3 ± 6.1" 82.5 ± 3.2
3.2±0.2
1.25% 15
169.6 4- 4.7" 81.9 + 2,8 3.0±0.3
rats
14
171.1 + 4.6 83.6 + 2.5 3.1 ± 0.2**
15
171.4 + 4.2 83.0 ± 2.7 3.0 4- 0.2
15
173.8 ± 4.2*** 83.4 ± 2.4 2.9 4- 0.3
tValues are means :1: SD o f single determinations of the number of specimens shown. Asterisks indicate significant differences between control and test values ("P ~ 0.05; P ~ 0.01; " ' * P ~<0.001). Non-pregnant rats were fed the control or calcium-enriched diets for 6 wk and then killed for tissue analysis. Pregnant rats were fed the test diets for 6 wk, then mated and continued on the appropriate diets until day 20 of gestation.
260
M. E. SHACKELFORD et al. Table 7. Mineral levels in foetuses from dams fed diets with excess calcium Mineral levels in rats fed calcium at: Mineral
0.50'% (control)
0.75%
1.00%
1.25%
No. o~ spet'imt'ns... Ca P Zn Fe Cu Mg Mn
24 1877.8 ± 147.9 2567.2 + 92.5 17.1 + 1.4 48.9 -+ 8.9 2.4 + 0.3 178.0 _+6.6 0.2 4-_0.0
22 1907.5 + 147.5 2551.0 + 135.6 16.6_+ 1.2 42.3 -+ 9.7"* 2.4 -+ 0.2 177.9 + 8.0 0.2 _+0.0
24 1888.8 + 192.4 2541.8 + 145.8 17.1 4- I.I 38.6 -+ 5.6*** 2.2 _+0.3* 175.1 _+ 7.3 0.2 + 0.0
24 1844.8 + 1870 2493.7 + 125.6" 17.0_+ 1.5 31.9 _+_+6.3"** 2.1 _+0.2** 168.2 + 9.3"** 0.2 + 0.0
~'The mineral content is expressed as ug/g fresh weight of foetuses. Values are means + SD of single determinations of the number of specimens shown. Asterisks indicate significant differences between control and test values (*P ~<0.05; **P ~<0.01: ***P ~<0.001).
was significantly decreased at all three dose levels in comparison with the controls. There was a dose-related negative linear trend for all four of these minerals (phosphorus, magnesium, copper and iron). DISCUSSION
Excess calcium intake can interfere with mineral utilization, but early reviews did not examine the added physiological stress of pregnancy (Davis. 1959 and 1967). We were interested in calcium-nutrient interactions because foetal development can be affected by deficiencies of minerals such as iron, magnesium and zinc (tturlcy, 1977), The purpose of the present study was to provide information on the maternal and foetal mineral interactions in rats fed a moderate increase in dietary calcium. The discussion is limited to a comparison of the data obtained in our study with the published literature on calciummineral interactions in rats. Where previously published data are available, the discussion includes a comparison between mineral intake, tissue mineral levels and prenatal development in animals. In our study we observed significantly lower iron levels in the maternal liver and kidney and in the foetuses from rats fed calcium-enriched diets. The iron requirement of a normal pregnancy is distributed to the foetus, placenta and expanded cell mass, but the mechanism by which the body and the mucosal cells achieve regulation of the body iron concentration is not adequately understood (Cook. 1990). The increased nutritional demand of pregnancy and the reduction of iron bioavailability due to the calcium may have contributed to the reduction in iron levels in the maternal and foetal tissues. We have found only one published study (AIt et al., 1938) in which liver iron Ic',els arc comparcd ~ith haemoglobin levels in rats and their offspring following feeding of the dams with iron-deficient diets to induce iron deficiency. In this one-generation. two-litter study, 21 days after both the first and second pregnancy, the maternal liver iron content had decreased to 20% of normal. Although the haemoglobin content was only slightly less than normal after the first pregnancy, after the second pregnancy it had decreased from 14.4mg/100ml
in the controls to 9.1 mg/100 ml in the iron-deficient rats. In the newborn offspring of the first litters, the total-body iron content was reduced by approximately 50°/. and the haemoglobin content was normal, but in the newborn offspring of the second litters the total-body iron content had decreased to 25*/0 of normal and there was a 50% reduction (6.6 l'. 13.1 mg/100ml) in haemoglobin c o n t e n t a level indicative of anaemia. To summarize, AIt et al. (1938) had shown that the impact of an iron-deficient diet on offspring could be greatest in the second litter, in which iron-deficient anaemia was observed. In our study, the foetuses from dams fed diets containing 0.75, 1.00 or 1.25"/o calcium had 87, 79 and 65%, respectively, of the iron content of foetuses from dams on the control dict. and the maternal livers of female rats on excess calcium had 84, 80 and 74%, respectively, of the iron content of females on the control diet. It is conceivable that if rats are fed excess calcium in nutritionally adequate diets for multiple generations, an iron-deficient anaemia could eventually develop in the offspring. Dietary calcium has been known to affect magnesium absorption in male (Likuski and Forbes, 1965) and female rats (Hock et aL, 1988). Recently, Hock et al. (1988) studied the effects in female Riv-TOX rats fed a semi-purified diet containing 0.40% phosphorus for 28 days. An increase in dietary calcium from 0.50 to 0.75% was associated with a significant increase in the faecal excretion of magnesium and a decrease in the elticiency of magnesium absorption, but had no effect on urinary magnesium output or magnesium retention. It is not clear from the literature (Elin, 1988: Maclntyre, 1967) what actual mechanism is involved and whether calcium ir~tcrfcrcs with magnesium absorption because of common transport carrier proteins. Although a calcium-magnesium interaction has been shown to exist in non-pregnant rats, no effects on tissue magnesium levels have been observed unless the rats were fed excessive amounts of dietary calcium. In a 28-day balance study with male Sprague-Dawley rats, Forbes (1964) showed that there was no effect on the magnesium content of the kidney when calcium was increased from 0.39
Calcium carbonate-mineral interaction to 0.78% (0.36% phosphorus) in an egg-protein, semi-purified diet. In another 28-day balance study, McAleese and Forbes (1961) fed male SpragueDawley rats a casein-based, semi-purified diet with a normal magnesium content of 500 ppm. When the calcium content was increased from 0.40 to 0.80%, there was no observed effect on the levels of magnesium in blood serum or bone ash or in the magnesium concentration in the kidney. Bodgen et al. (1992) fed male Sprague-Dawley rats AIN-76A diets containing 0.50 or 2.50% calcium for I yr. The excessive level of dietary calcium resulted in a significant decrease in femur and plasma magnesium but had no effect on the magnesium content of the kidney. Our data are consistent with the previously reported effects of moderate increases in dietary calcium on the magnesium contents of the femur and kidney of non-pregnant rats. Although the significant increase in the magnesium content of the liver at the 1.00 and 1.25% calcium levels (Table 4) was dosedependent, it is not readily explained from the current literature. The effects of the calcium-magnesium interaction were more pronounced in pregnant rats than in non-pregnant rats. Our data show a significant decrease in the maternal kidney content at 1.0 and 1.25% dietary calcium (Table 5) and in the wholebody magnesium of the foetuses at 1.25% (Table 7). it is of interest to compare our results with those of a study by Gunther et aL (1981), who determined the reproductive ctl'ccts of feeding varying levels of dietary magnesium to pregnant rats. Decreasing the dietary magnesium content from 360 ppm (control) to 260, 160, 110 or 40 ppm resulted in dose-dependent increases in the percentage of resorptions and in decreases in foetal weight. In our study, a moderate increase in dietary calcium did affect the levels of magnesium in the maternal kidney and the foetuses, but these effects were not associated with an increase in foetal rcsorptions or a decrease in foetal weight (Shackelford et al., 1993). Other calcium-nutrient interactions that have been studied include effects of calcium on the utilization of calcium and phosphorus. In the study by H o e k e t al. (1988) with female rats, increasing dietary calcium from 0.50 to 0.75% significantly increased the faecal excretion of calcium and phosphorus and strongly inhibited the urinary excretion of phosphorus without affecting the urinary calcium excretion. The group mean whole-body retention of calcium was increased, but that of phosphorus was lowered. Our data indicate that retention of calcium, but not of phosphorus, increased in the femurs of non-pregnant and pregnant rats (Table 6), but that the calcium and phosphorus contents of the kidneys were not affected (Table 5). Our findings for mineral levels in the femur are similar to those of Bodgen et al. (1992), who reported a significant increase in femur calcium when male rats were fed 2.5% calcium (v. controls, 0.50% calcium) for l yr. However, Hock et aL (1988)
261
reported a non-significant decrease in the calcium and phosphorus contents of the kidney, and Bodgen et al. (1992) reported a non-significant decrease in the calcium content of the kidney. The reason for the dose-dependent increase in the phosphorus content of the livers of non-pregnant rats which was observed in our study (Table 4) is not clear. Numerous experimental approaches have been developed to study the effects of excess dietary calcium on zinc absorption in casein-based, semipurified diets. In a study with male Sprague-Dawley rats, Forbes and Yohe (1960) failed to show a specific effect of calcium on zinc absorption or the zinc requirement. Heth et al. (1966) reported that in male Holtzman rats fed i.45% calcium in the diet (in comparison with 0.70% for the controls) for I month the absorption of dietary ~SZn was significantly decreased when the phosphorus content was increased to i%, but was not affected at concentrations of 0.3-0.5% phosphorus. Huber and Gershoff (1970) reported that 1.3% calcium in the diet fed to male Charles River rats significantly decreased 6~Zn absorption if the diet was zincdeficient, but had no effect with a control diet (16.3 ppm zinc). The addition of 2% phytic acid to the casein-based, semi-purified diet fed to male Sprague-Dawley rats for 18 days was associated with a dose-dependent decrease in the amount of zinc absorbed as the calcium level was increased from 0.4 to 0.8 or 1.2% (Likuski and Forbes, 1965); the addition of extra calcium to diets without phytic acid (control diets) did not significantly decrease the percentage of zinc absorbed. Forbes (1964) observed a decrease in zinc absorption resulting from increased calcium levels (0.40-0.80%) when soyabean protein, semi-purified diets were fed to male Spraguc-Dawley rats. in contrast extra calcium had no effect on zinc absorption when rats were fed semi-purified diets containing whole egg white protein which, unlike soyabean protein, does not contain phytic acid. Oberleas et al. (1966) proposed the formation of a Zn-Ca-phytic acid complex which removes zinc from solution in the intestinal lumen. Thus, both the protein source and dietary calcium concentration can affect zinc absorption. Although the results of previously published studies have indicated that excess calcium in a caseinbased, semi-purified diet does not affect zinc absorption, the results of our study may indicate a possible effect of increased dietary calcium in a casein-based, semi-purified diet on some aspect of zinc metabolism. The dose-related increase in the liver zinc content of non-pregnant rats is not consistent with current literature reports on additional dietary calcium. A dose-related negative linear trend was found in the zinc content of the kidney of pregnant rats without a concurrent effect on the foetal zinc content. This increased tissue catabolism in the kidney of pregnant rats may have resulted from the increased nutritional
262
M.E. SR^Crd~LFORDet al.
demands of the mother and foetus that are associated with a reduced zinc availability due to increased dietary calcium. The mechanism of reduced zinc availability is not clear from the literature, because zinc and calcium do not share common absorption paths (Cousins, 1985). However, in the study of Rogers et al. (1985), under simulated conditions of feeding pregnant rats reduced levels of dietary zinc, the maternal kidney and whole-body foetus had significantly reduced zinc contents, whereas the liver zinc content was not affected. The decreased zinc content in the maternal kidney (Table 5) and the normal levels of zinc in the maternal liver (Table 4) that were observed in our study appear to be consistent with a pattern of maternal tissue catabolism associated with reduced zinc availability due to increased dietary calcium. There are no published studies that have investigated the effects of additional dietary calcium on manganese or copper absorption. In our study the manganese levels were not affected. Unexpected findings were the significant decreases in the copper content of the kidneys of non-pregnant rats (Table 5) and the slight but significant decrease in the wholebody foetal copper content (Table 7) at the two highest dose levels. Our data may indicate the presence of a calcium-copper antagonism. The microscopic lesions diagnosed in the liver, heart and kidney were considered spontaneous in nature and/or incidental. The feeding of a purified or semi-purified diet to rodents has been shown to induce spontaneous lesions that are not found in young rodents fed diets composed of natural ingredients. These spontaneous lesions include myocardial damage (Mitchell et al., 1989), periportal fatty change (Medinsky et al., 1982) and renal mineralization (Woodard, 1971a,b; Woodard and Jee, 1984a,b). in our study the distribution and degree of severity of the renal mineralization (synonyms include ncphrocalcinosis, nephrolithiasis and calculosis) appear to be analogous to the kidney lesions that have been reported to have been caused by feeding rodents purified or semi-purified diets. There are two findings in the incidence of renal mineralization for which there was no obvious explanation. The first is that the baseline female rats had the lesion (Table 2). Since these animals did not eat the AIN-76A diet, other dietary factors may have contributed to the development of this lesion. The second is that the incidence was lowered in the non-pregnant rats fed the 0.75% calcium diet. Hock et al. (1988) reported that renal mineralization was essentially absent in female rats on a semi-purified diet containing 0.75% calcium, although it did occur in those fed the same diet with 0.50% calcium. Published data have indicated that whole-kidney function is largely unaffected by renal mineralization (AI-Modhefer et al., 1986: Stonard et at., 1984) because of the vast reserve capacity of the kidney. Thus, we would not expect the renal mineralization
to have contributed to the tissue mineral findings in our study. In summary, the most significant findings of this study were in the mineral analyses of the foetuses. The dose-related decreases in the foetal iron content were accompanied by dose-related decreases in the liver and kidney iron contents of non-pregnant and pregnant rats. The dose-related decrease in the magnesium content of the foetuses was accompanied by a dose-related decrease in the magnesium content of the maternal kidney, but not in that of the livers of the pregnant rats. A dose-related decrease in phosphorus was observed in the foetuses, but there was no effect on the phosphorus content of the livers and kidneys of the pregnant rats. A well known effect of additional calcium in the diet is a decreased absorption of iron. magnesium and phosphorus. Our results indicate that additional calcium in the diet decreased the bioavailability of iron, magnesium, phosphorus and copper. The increased nutritional demand of pregnancy and the reduced mineral bioavailability due to the effects of calcium resulted in effects on mineral levels in the foetuses, and, in some cases, also on those of the maternal tissues. Acknowledgements --We express o u r appreciation to Catherine Nelson and James I. Rorie for their excellent technical support. Thanks are also due to the Technical Editing Branch, Food and Drug Administration, for assistance in preparing the manuscript.
REFERENCES AI-Modhefer A. K. J., Atherton J. C., Garland H. O., Singh H. J. and Walker J. (1986) Kidney function in rats with corticomedullary nephrocalcinosis: effects of alterations in dietary calcium and magnesium. Journal of Physiology 380, 405-414. Alt H. L. (1938) Iron deliciency in pregnant rats. American Journal of DLseases of ChiMren 56, 975 -984. Barton J. C., Conrad M. E. and Parmley R. T. (1983) Calcium inhibition of inorganic iron absorption in rats. Gastroenterology 84, 90 10l. Bogden J. D., Gertner S. B., Christakos S., Kemp F. W., Yang Z., Katz S. R. and Chu C. (1992) Dietary calcium modifies concentrations of lead and other metals and renal calbindin in rats. Journal o[" Nutrition 122, 1351-1360. Consumer Reports (1988) The truth about calcium. Consumer Reports May, 288-291. Cook J. D. (1990) Adaptation in iron metabolism. American Journal of Clinical Nutrition 51, 301 ~308. Cousins R. J. 0985) Absorption. transport, and hepatic metabolism of copper and zinc: special reference to metallothionein and ceruloplasmin. PhysiohJgical Reviews 65, 238-309. Davis G. K. 09591 Effects of high calcium intakes on the absorption of other nutrients. Federation Proceedings 18, 1119-1124. Davis G. K. 0967) Excess minerals in the diet. Journal of the American Dietetic Association 51, 46-50. Dolan S. P.. Sinex S. A.. Capar S. G., Montaser A. and Clifford R. H. (1991) On-line preconcentration and volatilization of iodine for inductively coupled plasma atomic emission spectrometry. Analytical Chemistry 63, 2539-2542.
Calcium carbonate-mineral interaction Elin R. J. (1988) Magnesium metabolism in health and disease. Disease-a-Month 34 (4), 161-219. Forbes R. M. (1963) Mineral utilization in the rat. I. Effects of var)ing dietary ratios of calcium, magnesium, and phosphorus. Journal of Nutrition 80, 321-326. Forbes R. M. (1964) Mineral utilization in the rat. 11. Effects of calcium, phosphorus, lactose, and source of protein in zinc-deficient and in zinc-adequate diets. Journal of Nutrition 83, 225-233. Forbes R. M. and Yohe M. (1960) Zinc requirement and balance studies with the rat. Journal of Nutrition 70, 53-57. Gunther T.. lsing H.. Mohr-Nawroth F+. Chahoud 1. and Merker H. J. (1981) Embryotoxic effects of magnesium deficiency and stress on rats and mice. Teratology 24, 225-233. Heth D. A., Becket W. M. and Hoekstra W. G. (1966) Effect of calcium, phosphorus, and zinc on zinc-65 absorption and turnover in rats fed semipurified diets. Journal of Nutrition 88, 331-337. Hoek A. C., Lemmens A. G., Mullink J. W. M. A. and Beynen A. C. (1988) Influence of dietary calcium: phosphorus ratio on mineral excretion and nephroealcinosis in female rats. JournalofNutrition 118, 1210-1216. Huber A. M. and Gershoff S. N. (1970) Effects of dietary zinc and calcium on the retention and distribution of zinc in rats fed semipurified diets. Journal of Nutrition !00, 949 +954. tlurley U S. (1977) Nutritional deficiencies and excesses. In thmdbook +~1"Teratoh~g)'. Edited by J. G. Wilson and F. C. Fraser. Vol. I. pp. 261-308. Plenum Press, New York. Likuski I1. J. A. and Forbes R. M. (1965) Mineral utilization in the rat. IV. Effects of calcium and phytic acid on the utilization of dietary zinc. Journal of Nutrition 85, 230 234. McAIcese D. M. and Forbes R. M. (1961)The requirement and tissue distribution of magnesium in the rat as influenced by environmental temperature and dietary calcium. Journal o[ Nutrithm 114, 94 106. Maclntyre I+ (1967) Magnesium metabolism. Ad~,ances in blternal Medicine 13, 143 154. Medinsky M. A., Dua P. N., Jenkins M. G. and Grundel E. (1982) Development of hepatic lesions in male Fischer 344 rats fed AIN-76A purified diet. Toxicoh~gy and Applied PharnlacohJ,g;r 62, 11 I 120.
263
Mitchell G. V., Dua P. N., Jenkins M. G. and Grundel E. (1989) Nutritional and pathological changes in male and female rats fed modifications of the AIN-76A diet. Food and Chemical Toxicology 27, 185-191. Oberleas D., Muhrer M. E. and O'Dell B. L. (1966) Dietary metal--complexing agents and zinc availability in the rat. Journal of Nutrition °AI, 56-62, Padmore J. M. (1990) Animal feed. In Official Methods of
Analysis of the Association of Official Analytical Chemists. Edited by K, Helrich. 15th Ed. pp. 84-88. AOAC, Arlington, VA. Rader J. i., Wolnik K. A., Gaston C. M.. Celesk E. M., Peeler J. T., Fox M, R. S. and Fricke F. L. (1984)Trace element studies in weanling rats: maternal diets and baseline tissue mineral values. Journal of Nutrition 114, 1946-1954. Rogers J. M., Keen C. L. and Hurley L. S. (1985) Zinc deficiency in pregnant Long-Evans Hooded rats: teratogenicity and tissue trace elements. Teratology 31, 89-100. Shackelford M. E., Collins T. F. X.. Welsh J. J.. Black T. N., Ames M. J.. Chi R. K. and O'Donnell M. W. (1993) Foetal development in rats fed AIN-76A diets supplemented with excess calcium. Food and Chemh'al Toxicology 31,953 961. Sokal R. R. and Rohlf F. J. (1981) Biometry, pp. 179-245. W. H. Freeman and Company. San Francisco. CA. Stonard M. D., Samuels D. M. and Lock E. A. (1984) The pathogenesis and effect on renal function of nephrocalcinosis induced by different diets in female rats. Food and Chemk'al Toxh'ohJgy 22, 139 146. Woodard J. C. (1971a) A morphologic and biochemical study of nutritional nephrocalcinosis in feln;tle rats fed semipurified diets. American Journal of Pathohlg)" 65, 253 -268. Woodard J. C. (1971b) Relationship between the ingredients of semipurilied diets and nutritional nephrocalcinosis of rats. Amerh'an Journal of PathobJgy 65, 269 -278. Woodard J. C. and Jee W. S. S. (1984a) Effects of dietary calcium, phosphorus, and magnesium on intranephronic calculosis in rats. Jourmd of Nutrition 114, 2331 -2338. Woodard J. C. and Jee W. S, S. (1984b) Effect of diet and intranephronic calculosis on bone modeling and parathyroid volume in rats. Journal oj" Nutrition 114, 2339 2352.