An experimental study on the long-term effect of cadmium in mice fed cadmium-polluted rice with special reference to the effect of repeated reproductive cycles

ENVIRONMENTALRESEARCH 40, 25-46 (1986)

An Experimental Study on the Long-Term Effect of Cadmium in Mice Fed Cadmium-Polluted Rice with Special Reference to the Effect of Repeated Reproductive Cycles MASAO WATANABE,* KAZUKO SHIROISHI,t HARUMI NISHINO,~ T E T S U O SHINMURA,'~ HITOSHI M U R A S E , t TOSHIO S H O J I , t YUCHI N A R U S E , * AND SADANOBU KAGAMIMORI*

*Department of Community Medicine, Faculty of Medicine, Toyama Medical & Pharmaceutical University, and #Toyama Institute of Health, Toyama, Japan Received May 17, 1983 Long-term biological effects of cadmium-polluted rice and the effect of repeated reproductive cycles on them were examined. Female SLC-B6D2F mice (female C57BL/6, male DBA/2) were fed a rice diet containing 65% unpolished rice for about 2 years from 7 weeks of age. The unpolished rice preparations used were commercially available rice (non-Cdpolluted) and Cd-poUuted rice (over 1.0 ppm). Average Cd contents in each diet class were 0.12, 0.48, 1.78, 1.75, and 47.1 ppm (50 ppm Cd as CdC12 added). Some experimental mice were subjected to repeated reproductive cycles (parity group). Hematological, biochemical, and pathological examinations of urine, blood, and tissues, including Cd measurement, were carried out. Results after statistical analysis indicate Cd toxicities such as anemia and disturbances of Ca metabolism. These Cd effects were found to be enhanced by the reproductive cycles. Soft X-ray radiograms showed osteoporosis in the parity groups, especially in the groups with diets of higher Cd content. However, we could not find any sign of disturbance of renal function under our experimental conditions. © 1986AcademicPress, Inc.

INTRODUCTION Although the mechanism of pathogenesis of the so-called Itai-Itai disease in Toyama Prefecture, Japan, is not yet completely elucidated, it is considered that Cd intake via rice and drinking water may play a role in the manifestation of this disease (Friberg et al., 1974; Tsuchiya, 1978). From these points of view, many animal experiments on the toxicity of Cd were carried out using inorganic Cd salt solutions of relatively high concentration, as cited in the above-mentioned reviews. On the other hand, Cd was reported to exist in rice in a form(s) bound to the rice protein such as glutelin (Matsue et al., 1971; Kobashi et al., 1978). The toxicity of this bound form of Cd by oral administration has not yet been compared sufficiently with that of the inorganic salt form. In the present study, the long-term effects of Gd at low concentration were investigated in mice fed a diet containing Cd-polluted rice as the main ingredient, and, in addition, the effect of pregnancy and lactation on the Cd toxicity was examined, as most Itai-Itai disease patients were multiparous (Itai-Itai Disease Research Committee, 1967). MATERIALS AND METHODS Animals

Female SLC-B6D2F mice, 4 weeks of age of (female C57BL/6, male DBA/2), 25 0013-9351/86 $3.00 Copyright© 1986by AcademicPress, Inc. All rightsof reproductionin any formreserved.

26

WATANABE ET AL.

were obtained commercially (Shizuoka-ken Animal Research Institute, K.K.) and were used for the experiment after 3 weeks observation.

Diet Commercial diet (Com. D.) was obtained from Funahashi Farm, K.K., and rice diets were prepared by the same company so as to contain 65% unpolished rice, nonpolluted or polluted with Cd (Table 1). In this study, the term "Cd-polluted rice" is used for rice containing over 0.4 ppm cadmium according to the criteria for environmental control recommended by the Japanese government's Committee for the Study of Trace Metals (1970). The same committee also recommended 1.0 ppm Cd as a criterion for safety control (1970). Cd-polluted rice used in this experiment was derived from a test pilot rice field of the Toyama Prefectural Station for Agricultural Research. Commercially available rice was used as nonpolluted rice. One of the diets (V) used in this experiment was prepared by adding 50 ppm Cd as CdC12 to nonpolluted rice diet IF. Diets III and 1V were found to have almost the same Cd content in spite of different Cd concentrations in the unpolished rice as shown in Table 1. The reason for this is not clear, but may be due to the variation in Cd content of diet components other than rice.

Schedule Ten experimental groups were set up as shown in Table 2. In four groups, mice were studied at several stages of the reproductive cycle, including mating, pregnancy, delivery, and lactation (hereafter, these groups are designated as parity groups) in the following manner. The reproductive cycle started at 10 weeks of age; one cycle lasted 40 days, including 1 week for mating, about 3 weeks for pregnancy, and 10 days for lactation. The cycle was repeated eight times during the total experimental period. The average number of times of delivery per group in the whole period was calculated from the total number of times of delivery and the mating times and was found to be 5.5-6.3. The average number of offspring per delivery was 7-10. Biochemical and pathological examinations were carried out in the 1st, 4th, 10th, and 16th months. Test materials were taken at or before autopsy. In the parity groups, autopsy was carried out after the completion of a reproductive cycle. In the nonparity groups, diet intake, feces, and urine were measured daily.

Biochemical and Hematological Examinations Methods of examination are listed in Table 3. These examinations were carried out at the start of the experiment and in the 1st, 4th, 10th, and 16th months of the experiment, a total of five times, parallel with the pathological examinations. Test samples were taken from four to six mice per group. For examination of Cd or Ca content tissue samples of liver, kidney, and femur were taken from two mice (three mice at the start of the experiment) per group at autopsy.

Pathological and Other Examinations Pathological examinations were carried out five times during the whole experi-

E F F E C T OF C A D M I U M - P O L L U T E D RICE

r,,:

~D

÷1 o,

[..,

"6e.~

~3 ,...;

.< r..) ~D

+1 -7.

"'

÷1 ~. -q

"2

"

27

28

WATANABE ET AL. TABLE 2 EXPERIMENTALGROUPS

Group

Diet

Rice added

Reproductive cycle

Number of mice

1 1'

I

Nonpolluted

+

30 24

2 2'

II

Nonpolluted

+

35 24

3 3'

III

Cd-polluted

+

35 24

4

IV

Cd-polluted

-

35

5 5'

V

Nonpolluted

+

24 24

Reference

Com.D.

--

-

34

mental period, parallel with the examinations of urine and blood. Six mice were autopsied before the start of the experiment, and thereafter four mice per group every time. Half of them, selected at random, were supplied for the pathological examination. Histological examinations were carried out after formalin fixation and hematoxylin-eosin staining by the usual techniques. PAS staining was sometimes used if necessary. Radiography of bones was carried out with a soft X-ray apparatus (Sofron, SRO-505 Soken Co. Ltd., Tokyo) after fixation. The length TABLE 3 METHODS OF EXAMINATION Urine

Sugar, qualitative Sugar, quantitative Protein, qualitative Protein, quantitative Lysozyme Amino acids Inorganic phosphorus Ca

Paper method (Tes-Tape, Lilly Co.) o-Toluidine-boric acid method Paper method (Albustick) Tsuchiya-Biuret method Lyso plate Trinitrobenzenesulfonate method Fiske-SubbaRow method Atomic absorption

Blood RBC volume Hemoglobin RBC Hemogram Total blood protein Albumin Alkaline phosphatase Inorganic phosphorus Ca GOT

Hematocrit Cyanmethemoglobin method Erythrocytometer May-Grfinwald-Giemsa staining Biuret method 2-(4-Hydroxybenzene-azo)-benzoic acid method Bessey Lowry method Fiske- SubbaRow method Atomic absorption Reitman Frankel method

Cd

Atomic absorption with dithizone-chloroform extraction

EFFECT OF CADMIUM-POLLUTEDRICE

29

and diameter of a bone and the thickness of its cortex at the middle of the bone diaphysis were measured after the photograph was enlarged 20 times with a multipurpose projector (Olympus).

Statistical Analysis Data from the examinations were analyzed by an analysis of variance with three factors: diet (I, II, III, and V), parity, and experimental duration. RESULTS

Food Intake, Excretion of Urine and Feces, and Growth The average daily intake of food was 2.5-3.2 g per mouse in the nonparity groups; there were no significant differences among the diets or among the experimental durations. The amount of food intake measured in the breeding cage was found to be 1.2-1.4 times higher than that in the sampling cage, perhaps because of food loss in the breeding cage. Because of this difference in the type of cage, food intakes in the parity groups were estimated by using the breeding cage and by correcting with a factor of the ratio between the intakes in the two kinds of cages, 0.77. In the parity groups, the level of food intake increased steeply with lactation. Thus the total diet intake during one cycle of reproduction (40 days) was calculated to be about 300 g, which was about 2.9 times that in the nonparity groups, 111 g. The average amounts of feces (means _+ SD) were 0.9 + 0.09, 0.6 + 0.16, and 0.9 + 0.34 g/day at 11, 24, and 76 weeks of age, respectively, in the nonparity groups, and 1.3 _+ 0.61, 1.2 _ 0.28, and 1.8 _+ 0.32 g/day, respectively, in the Com. D. groups. There was no significant difference among the experimental rice diet groups (data not shown). Daily urine excretion in the nonparity, rice diet groups was found to be 0.7 _+ 0.35, 0.6 +_ 0.21, and 0.9 + 0.47 g/day at 11, 24, and 76 weeks of age, respectively; values for the Com. D. groups were 0.7 +- 0.24, 0.8 + 0.24, and 1.I _ 0.41, respectively. The growth curves between 7 and 110 weeks of age are shown in Fig. 1. All groups showed marked growth until 40 weeks of age and slow growth thereafter. Parity groups exhibited a higher growth rate for the whole period than the nonparity groups. No consistent effect of diet on growth rate was found.

Urine Urine examination results are shown in Fig. 2. Urinary protein varied significantly with diet and with experimental duration (data not shown). However, this did not suggest any relationship to the Cd content of the diet. Urinary Ca was significantly higher in those experimental groups fed higher Cd-containing diets (III and IV), and it increased with experimental duration. These results for the nonparity groups suggest an increase in urinary Ca excretion parallel to the increase in Cd intake. On the other hand, urinary Ca excretion was significantly lowered (P < 0.05) in the parity groups (Fig. 3). Urinary inorganic

30

W A T A N A B E ET A L .

d

L L y/

j 9 ~r C"

~,

v

v 0,.I

9

/ ! i l

•

o I

I

,,...;

E6

I

I

O rO

od

v

~q6 tam ~po~

' O

$1----.-

EFFECT OF CADMIUM-POLLUTED RICE

31

Protein

2ood

e~

~i000~ E

20O I

'~1600J"

Glucose

2 o ~ ~ 8 o o ~ ~

.-"

~400 L

r Amino acid _l L . - "®'.

4OOl% 2 0 0 jE

A

-

mg/dl I

5of Calcium

.Z ~

2

1

0~

mg/df _Phosphorus

OIcalcium/phosphorus ratio

008[

°.°2F~ - - : o

01

~

:__-~_

4 10 16 Experimental duration(Months)

FIO. 2. Results of urinary examinations. For explanation of symbols, refer to Fig. l.

phosphorus and the ratio of Ca to P concentration varied significantly with a tendency similar to that of urine Ca. Urine lysozyme, which is known to be an index for excretion of proteins of low

32

WATANABE ET AL.

(mg/dl) 1,412" ~_)

e= -r--

I08o

i°

6-

I0

4-

T

0

:

Non-parity group

•

:

Parity group

l

I

I

I

II

Ill

I

V

Diet class FIG. 3. Urinary Ca (average amount per mouse). *Experimental duration in months.

molecular weight because of renal tubular disturbance, did not increase (data not shown).

Blood Hematological examination. Results are illustrated in Fig. 4. A statistical analysis of variance showed a significant variance of all indices concerned with red blood cells. Hb concentration (Fig. 5) and Ht decreased significantly in the groups fed diets III and V (P < 0.01) and in the later stage of the experiment in the parity groups (P < 0.05). Mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCC) all varied significantly, and were lowered in the group fed diet V, in parity groups, and in groups in the 10th to 16th months, respectively. Biochemical Examination of Serum Biochemical examination results are shown in Fig. 6. Serum Ca concentration was significantly affected by all three factors; it was particularly low in the 16th month in the parity group fed diet V (P < 0.05). There were significant interactions between diet and parity in a comparison of diet V to other diets (P < 0.05), suggesting a synergistic effect of diet Cd and reproductive cycle on the decrease in serum Ca. Similar interactions between diet and experimental duration, and among the three factors, were observed. Alkaline phosphatase activity and serum inorganic phosphorus were affected by diet and experimental duration, but no meaningful tendency was found.

EFFECT OF CADMIUM-POLLUTED RICE

% [~ Hematocrit

15

]04/mm~

1200 t

""

Red blood c e l l count

/~w-...

. .®

1ooo[. Mean corpuscular hemoglobin (MCH)

pm3

Mean corpuscular volume (MCV)

45 40

~-"

Mean corpuscular hemoglobin

3( 2E 0

I

I

4

10

Experimental duration

|

16 (months)

FIG. 4. Results of blood examinations. For examin~ion of symbols, r e a r to Fig. 1.

33

34

WATANABE ET AL. (g/dl) 15.5

=

15.0 j..--o

4

%

~

14.5 I0 16 0

14.0

!

I

!

I'I

I ~I V Diet class FIG. 5. Blood hemoglobin (average amount per mouse). For explanation of symbols and numbers of curves, refer to Fig. 3.

Morphological Findings Various morphological abnormalities were observed in red blood cells (RBC), mainly in the experimental groups fed diet V, suggesting that the diet containing the largest amount of Cd had a toxic effect. However, there might not exist any significant effect of Cd in rice because there is no significant tendency among these groups of diets.

Ca in Bone Calcium and phosphorus in bone were measured as shown in Table 4. Variances were significant among three factors (P < 0.01). Bone Ca was significantly lowered in the group fed diet V, in the parity group, and in the groups at 10 and 16 months (Fig. 7). Interaction was significant between diet and parity in the difference between diet II and diets containing larger amounts of Cd (P < 0.01). The difference between the groups fed diet II and III was significant in the parity group (P < 0.01). Interactions among the three factors and between parity and experimental duration were also significant (P < 0.05). These results suggest that Ca in bone was lost by Cd accumulation in tissues in the parity groups.

Intake and excretion of Cd Cd intake per day was calculated on the basis of the amount of food intake and Cd content of diet as shown in Tables 5 and 6. In the parity groups, it increased steeply after delivery and reached 960 rag/day in the groups fed diet V on the 8th to 10th days. Cd excretion in feces in group 5 was much higher than in the other groups and remained at almost the same level for the duration of the experiment (Table 7). In the groups fed diet V, urinary excretion of Cd increased up to the 10th month and thereafter showed the same level. In the parity groups, samples were taken between reproductive cycles, so a level of Cd excretion similar to that

35

EFFECT OF CADMIUM-POLLUTED RICE

mg/dl g/

_ 1

g

~

l

~

15

!6~1 Phosphorus

-"~-

Albumin/globulin r a t i o 1.4~

1.2 1.0 0.5

El

01

i

I

I

4

10

16

(months) Experimental duration

0.6

0.4 "

Alkaline phosphatase

B

_ -_--.T I o'-

U

Iransaminase (G.O.f.)

60

, ~-~-,

40

"

01

_

.

,o

","

4.

!

!

10

16

Experimental duration

(months)

FIG. 6. Results of biochemical examinations of serum. For explanation of symbols, refer to Fig. 1.

WATANABE ET AL.

36

TABLE 4 CONCENTRATION OF CALCIUM AND PHOSPHORUS IN FEMORAL BONE

Experimental duration (months) Group

0

4

10

Ca (rag/g) 121 129

142 121

115 112

110 109

148 143

130 136

120 128

2t

139 125

125 121

3

131 142

1

134

1'

115 125

1

118 123 135 127

2

113 3'

135 121

122 4

117

139 135

5 5' Com.D.

128 147 212 133

16

0

1

4

10

P (mg/g) 58 63

70 62

59 57

56 57

75 73

72 72

65 68

129 129

66 61

69 68

68 68

149 146

137 140

66 81

78 75

75 74

127 133

117 109

131 108

62 67

67 59

74 58

172 176

151 133

-145

45 47

74 71

-73

131 123

132 125

131 135

64 60

74 69

71 73

120 115

106 103

110 107

62 56

60 60

60 60

150 117

138 129

122 --

79 61

81 76

66 --

63 58 65

54 61 64 65

50

66 65

57 64

64 64

65 69 31 65

16

in the nonparity groups indicates that Cd excretion was not affected by the preceding reproductive cycle, although Cd excretion is assumed to increase during the reproductive cycle because of increased diet intake. In general, urinary Cd is assumed to be an indicator for Cd exposure, but the present experiment showed (mg/g)

15°1

iiit

,0-

o ,Ol Ioo

]

.

.

.

.

0 : Non-parity group • : Parity group

]0

T II '

I'iI

V'

Diet class

FIG. 7. B o n e calcium. *Experimental duration in m o n t h s . For explanation of symbols, refer to Fig. 3.

37

EFFECT OF CADMIUM-POLLUTED RICE TABLE 5 DALLYINTAKEOF CADMIUM1N NONPARITYGROUPSa Group

Age (weeks)

1

2

3

4

5

Reference

11 24 76

0.32 0.30 0.38

1.30 1.13 1.34

4.79 4.40 5.62

5.45 4.61 6.14

127 116 149

0.28 0.30 0.40

a

In micrograms.

that urinary Cd, was not affected by exposure to amounts of cadmium as small as those in diets I-IV. Cd concentration in milk was also estimated by using the pooled stomach contents of suckling mice 1 week after birth. It was higher in the groups fed diet V (average 24.7 ng/g) than in the other groups. A short time after birth (several hours) in groups fed low-Cd diets, Cd content was under the technical limit, but in the groups fed diet V it was 2.0 ng/g (second delivery) and 7.1 ng/g (seventh delivery). These data indicate the possibility of transfer of Cd from mother to baby via milk. However, excretion in milk was very |ow and may not have a significant effect on total excretion. TABLE 6 INTAKE OF CADMIUMDURINGPREGNANCYAND LACTATION

Group

Stage

Food intake ~ (g)

1'

Early, b per day Middle, c per day One cycle, e per 40 days (per day)

3.93 4.35 319.5

0.47 0.52 38.4 (0.96)

2'

Early, per day Middle, per day One cycle, per 40 days (per day)

2.49 3.61 --

1.20 1.73 153e (3.84)

3'

Early, per day Middle, per day One cycle, per 40 days (per day)

2.35 3.21 --

4.18 5.71 569e (14.2)

5'

Early, per day Middle, per day One cycle, per 40 days (per day)

2.28 3.70 --

107 174 15,052 e (376)

" Calculated amount of food intake in breeding cages multiplied by 0.77. b 7--9 days after mating. c 14-16 days after mating. a Total cadmium intake during a cycle (40 days) from one mating to the next. e Calculated daily diet intake in group 1'.

Cd intake (~g)

38

WATANABE ET AL. TABLE 7 CONCENTRATION OF CADMIUMIN LIVER, KIDNEY, FECES, AND URINE Experimental duration (months)

Group

l

0

1

<0.01 0.01

1'

0.01

0.12 0.09

0.17 0.22

0.07 0.09

0.11 0.14

0.22 0.23

0.25 0.26

0.44 0.41

0.40 0.43

0.10

0.36 0.27

0.70 0.66

0.94 0.73

0.14 0.14

0.29 0.27

0.52 0.58

0.65 0.75

3

0.02 4

0.10

0.02 5 6.76 5'

--

0.02 0.02

Com.D.

1

1'

0.35

0.02 0.01

0.07 0.08

0.01

0.20

3'

16

0.03 0.03

0.08

0,02

Liver (~g/g) 0.01 0.01

10

0.02 0.02

2 2'

4

14.0 13.7

28.1 30.8

30.0 32.8

28.1 26.4

63.2 72.8

74.1 81.2

0.04 0.06

0.03 0.03

0

0.01 0.01

0.02

0.01

0.01

0.03

16

Kidney 0xg/g) 0.03 0.04 0.03 0.04

0.13 0,14

0.29 0.29

0.5l 0.63

0.26 0.30

0.53 0.60

0.78 0.92

0.51 0.55

0.98 0.95

1.74 1.64

0.15

0.96 0.76

2.77 2.41

3.18 2.43

0.24 0.24

0.69 0.60

1.60 1.60

2.13 2.36

0.16

0.04 12.0 10.2 0.04 0.03

0.6

10

0.14 0.12

0.24

0.04

4

0.05 0.07

0.15

0.03 0.03

Feces (mean, t~g/g)a 0.37 0.51 0.40 0.54 0.52

1

0.6

26.6 26.9 50.1 56.2 0.12 0,13

58.2 66.3 116 129 0.11 0.08

68.2 78.5 117 143 0.11 0,09

Urine (¢g/l) ° 1.1 0.9 1,1 0.8

2 2'

2.2

2.3 2.1

2.5 2.3

2.1 2.1

1.1

1.3 0.6

0.5 0.6

0,1 0.3

3 3'

7.6

8.0 7.5

6.9 7.3

7.0 6.9

0.9

1.0 1.5

0.6 0.5

0.3 0.2

7.5

9.3

8.5

8.l

0.9

1.2

0.2

0.5

1.2

6.1 8.6

23.4 50.3

36.5 36.9

1.4

0.4

0.3

0.4

4 5 5' Com,D.

0.42

200 0.50

183 203 0.47

167 229 0.48

215 224 0.43

2.8

a Average concentrations with 4 or 6 mice. b Data for pooled sample from 4 or 6 mice.

Cd Concentrations in Liver and Kidney C d c o n c e n t r a t i o n s i n t i s s u e s a r e s h o w n i n T a b l e 7. C d a c c u m u l a t i o n i n l i v e r a n d kidney increased generally with Cd concentration of diet, experimental duration, and reproductive cycle. It may be noted that time-course pattterns of Cd accu-

EFFECT OF CADMIUM-POLLUTEDRICE

39

mulation in liver and kidney were almost comparable among experimental groups fed diets V, diet III or IV (Cd-polluted rice diet), and even nonpolluted rice, and that Cd concentration in kidney was always higher than that in liver with a ratio 1.4-4.3. Correlations between total Cd intake and Cd content in kidney are clearly demonstrated in Fig. 8. Accordingly, the higher Cd accumulation in the parity groups might be explained in part by the higher intake of Cd via diet. Although diets III and IV contained the same concentration of Cd, groups fed these diets exhibited a difference in liver and kidney Cd concentrations. This difference might correspond to a difference in the Cd concentration of the Cdpolluted unpolished rice, and can be assumed to be due to the difference in the efficiency of Cd absorption from diet components other than rice.

Pathological Examination Macroscopic findings. In all the mice used, epilation in part of the face was observed 1 month after the start of the experiment. The epilation was an irregular shaped spot about 5 × 5 mm, but the skin had no lesion, and an examination for parasites and fungi had negative results. The frequency was relatively low in groups 5 and 5', but there was no significant difference among the experimental groups. There were slight signs of pneumonia in about a half of the mice, and a few cases of hydronephrosis, chronic nephritis, and malignant tumor were found in the 16th month. Histologicalfindings: Liver. Solitary or sporadic cell nets were found in about half of the mice, but their nature was not apparent. In a few cases, lymphocyte (pg) o : Non-parity • :Parity

iO0,O-

o

o o

iO,O -

z

QO QO

8 1.0oo

82: % 0.I"

t

i

t

10 3

10 4

10 5 (pg/mouse)

Total Cd intake

FIO. 8. Correlationbetweentotal Cd intake and Cd retention in kidneys.

40

W A T A N A B E ET AL.

infiltration into the periphery of the interlobular ductules and a sporadic centrilobular fatty change were found. But these findings did not differ significantly among the experimental groups. Kidney. Slight chronic nephropyelitis or chronic pyelonephritis was found in about 40% or more (after the 10th month), and a localized mesangial proliferative glomerulonephritis was found in about a half. In the renal tubules, there was almost no abnormal change. Lung. Slight alveolitis or bronchial pneumonia was found in about 80% of the mice used. In the other organs, the following were found in one or a few cases: arteriolosclerosis in spleen, localized purulent stomatitis, stomach ulcer, adenoma in the stomach, adenoma or metastatic tumor in duodenum, localized pancreatitis, peripheral inflamation of pancreatic duct, metastatic tumor, abnormal differentiation of the thyroid, localized pancreatitis, metastatic tumor in the thyroid, parathyroiditis. Most of those histological findings were of a very slight grade, and there was no characteristic difference among the experimental groups.

Bone Measurement Bone length and diameter at the middle of the femur diaphysis were measured with soft X-ray radiography. Although no variation among the experimental groups was found, both measurements showed a slight tendency to increase with experimental time, especially in the parity groups (Table 8). These findings may

TABLE 8 DIAMETER OF THE FEMORAL DIAPHYSISa

Experimental duration (months) Group

1

1

4

10

16

161

168

159 155

l'

163

2

149 ¢ 161 b

2'

167

178

185

3

165b

172

175

160 3' 4

148 ¢

5

156

174

178

165

163

165

163

166

166

161

178

177

160

174

169

156 5' Com.D.

159

a Transverse distance at the middle of the diaphysis, expressed in 10 2 mm. Each value is an average value for four bones except as noted. b Average of three bones. c Average of two bones.

EFFECT OF CADMIUM-POLLUTEDRICE

41

indicate growth of bone during the experimental period. The ratio of the thickness of bone cortex to diameter was measured as shown in Table 9. It was significantly lowered in the parity groups and decreased with time. To determine whether the decrease in the ratio of the thickness of bone cortex to diameter in the parity groups was due to the increase in the diameter of bone or not, the following calculations were carried out. The rate of increase of bone diameter (rate 1) in the parity group above that in the nonparity group and the rate of decrease of the ratio of cortex to diameter (rate 2) in the parity group below that of the nonparity group were calculated using the data in Tables 8 and 9, respectively. If rate 2 is greater than rate 1, the excess indicates the actual reduction in cortical thickness caused by reproductive cycles. This excess in the rates is designated as the attributable parity effect on cortex (APEC) in this paper. As shown in Table 10, the variation of APEC with diet is clear. It is always higher in diet V than in diet II. APEC is at the same level in diets II and III until the 10th month, but it increases in diet III in the 16th month. These findings for APEC may indicate enhancement with a dose effect of the parity effect on bone cortex by a Cd-containing rice diet. Soft X-ray radiograms also indicate a low intensity of bone in the parity groups after 4 months, and this weakness in the intensity became more obvious with experimental duration (Fig. 9). These changes in the intensity of the radiogram of bone appeared to be much emphasized in parity groups fed diet V.

TABLE 9 RATIO BETWEEN CORTICALTHICKNESSAND DIAMETEROF THE FEMORAL DIAPHYSIS(%)a

Experimental duration (months) Group

1

1

4

10

16

31.3 C*

28.7**

23.9*

24.8*

20.1"*

20.2*

30.9 b

26.3**

24.1"

30.1

21.3"*

19.7"

31.8

28.8

27.0

30.6*

28.5**

26.1"

25.2*

17.8"*

20.0*

32.4

29.0

25.7

29.9* 26.5

1'

24.9*

2 27.3 b 2' 3 29.5 3' 4

29.7 C

5 28.4 5' Com.D.

29.3

a Each ratio was calculated as the sum of the thickness of the right and left cortices divided by the diameter of the diaphysis. b Average of three bones. c Average of two bones. * 0.01 < P < 0.05. ** P < 0.01.

WATANABEET AL.

42

TABLE 10 ATTRIBUTABLEPARITYEFFECT FOR BONE CORTEX

Experimental duration(months) Diet

4

10

16

iI III V

0.087 ~ 0.080 0.188

0.194 0.178 0.303

0.051 0.162 0.167

a See text. Correlation with Cd Concentration in Tissues

Table 11 shows the correlation coefficients, in which urinary P and Ca/P, bone Ca, ratio of cortical thickness, Hb, Ht, MCV, and MCH decreased significantly because of kidney or liver Cd concentration. These findings showed good agreement with the results of the analysis of variance, indicating a disturbance of Ca metabolism and anemic changes produced by Cd-containing rice diets. So the effects of reproductive cycles and experimental duration might be explained in part as a response to Cd accumulation in kidney and liver. DISCUSSION Although diets were prepared according to our specifications, nutrient analysis indicated that our goals were not always satisfied, especially with respect to the Group 2

3

2~

3~

4

5

Ref.

5'

FIG. 9. Soft X-ray photographs of right femora of two animals from each group.

43

E F F E C T OF C A D M I U M - P O L L U T E D R I C E TABLE 11 CORRELATION COEFFICIENTSBETWEEN EACH EXAMINATIONRESULT AND CD ACCUMULATION IN KIDNEY AND LIVER

Kidney

Liver

Examination (unit)

N

Correlation coefficient

Significance

N

Correlation coefficient

Significance

Urinary Ca (mg/dl) Urinary P (mg/dl) Urinary Ca/P (%) Serum Ca (mg/dl) Bone Ca (mg/g) Bone cortex ratio (%)

41 41 41 41 41 41 41 41 41 41 41 41

0.029 - 0.432 0.451 -0.254 -0.495 -0.335 0.306 -0.391 -0.478 0.062 -0.561 -0.489

N.S. ~ ** ** N.S. ** * N.S. * ** N.S. *** **

38 38 38 38 38 38 38 38 38 38 38 38

-0.003 - 0.426 0.443 -0.224 -0.520 -0.274 0.257 -0.403 -0.478 0.064 -0.510 -0.442

N.S. ** *** N.S. *** N.S. N.S. ** ** N.S. *** **

RBC (104/ram 3) Hb (g/dl) Ht (%) MCC (%) MCV (ixm3) MCH (10 -12 g) N.S., not significant. * 0.01 < P < 0.05. ** 0.001 < P < 0.01. *** P < 0.001.

concentrations of Cd, Ca, and inorganic phosphorus. It is known that the Ca/P ratio of the diet can influence the absorption of Ca from the intestinal mucosa. So the evaluation of diet effects appeared to be very complicated and a statistical analysis seemed to be necessary. After all the data were fully evaluated, we drew the following conclusions from the present study. Urine

Urine examination did not show any renal dysfunction due to cadmium intake. Indices of urine examination showing a significant variance among the diets by the analysis of variance were protein, amino acids, Ca, inorganic phosphorus, and the Ca/P ratio. However, urinary protein and amino acids did not always appear to correlate with Cd content of diet, and they decreased significantly in the later stage of the experiment. In mice subjected to reproductive cycles, the urinary protein sometimes increased, but not meaningfully. As proteinuria, glucosuria, and aminoaciduria are indices for injured renal function due to Cd poisoning, the results of the urine examination in the present experiment do no indicate the existence of a severe renal disturbance in mice fed a Cd-polluted rice diet. This lack of renal dysfunction as indicated by urine examination is also supported by the fact that no specific pathological conditions in kidney were observed, and it may be due to the very low accumulation of Cd in kidney, which did not even at its maximum the critical concentration, 200 ~g, mentioned by Friberg (1974) in kidney cortex.

44

WATANABE ET AL.

Anemia

Hematological examination demonstrated a hypochromic and microcytic anemia in mice as a function of diet and reproductive cycle. Especially in the group fed diet III (Cd-polluted rice diet) the hemoglobin value decreased significantly compared to the experimental groups fed the nonpolluted rice diet. A significant interaction for anemic change among the three factors suggests an effect of Cd accumulation, in addition to the effect of the reproductive cycle. Significant correlations between indices in hematological examinations and Cd accumulation in kidney and liver indicate further support for the effect of Cd in diet. Therefore, the effect of reproductive cycle can be explained in part by the much greater food intake during lactation periods. Disturbance o f Ca Metabolism

In the groups fed diets containing larger amounts of Cd, there was an increase in urine Ca and decrease in serum and bone Ca with experimental time. In parity groups, Ca always decreased in urine, serum, and bone. The decrease in bone Ca, especially in the parity groups, is supported by soft X-ray radiography of bone and measurements of cortex thickness. Soft X-ray radiograms showed a decrease in bone density in the parity groups; the relative cortical thickness of bone decreased as a function of reproductive cycle and experimental duration. APEC calculated as an index of the effect of the reproductive cycle on decreasing bone cortex thickness indicates an enhancement by Cd-containing diets of the parity effect on bone cortex. These results seem to suggest a loss of Ca from bone via serum and urine as a function of Cd accumulation. The decrease in urine Ca in parity groups might indicate a transfer of Ca from bone to fetus. The disturbance of Ca metabolism found in mice fed Cd-polluted rice diets appeared to be enhanced by the reproductive cycle. In groups with CdC12 added to the diet (diet V), Cd seems to have disturbed Ca metabolism more than in the groups fed other diets, but there was no characteristic difference. In addition to these results, significant correlations between bone Ca or relative cortical thickness and Cd accumulation in tissues indicate that the disturbance of Ca metabolism in the parity groups and in the later stages of the experiment may be at least in part a response to the increased Cd accumulation due to increased food intake or Cd absorption, in addition to the specific physiological changes. Diets III and I were found to have much more serum Ca and P than the other diets, and this difference was assumed to play some role in the variation of examination data, because dietary Ca has been reported to inhibit the disturbance of Ca metabolism due to Cd accumulation and chronic poisoning in animals (Ishizaki et al., 1974; Hamilton et al., 1978; Kello et a1., 1979). Disturbances of Ca metabolism produced in animal experiments using CdCI2 or other Cd salts have been already reported by other workers as shown in the above-mentioned reviews, and our results using Cd-polluted rice diets were comparable to their reports in principle. As to the serum level of Ca, there seemed to be some discrepancies in the data of other workers on the effects of Cd administration or of the reproductive cycle (Takizawa et al., 1981). However, our data seem to indicate more consistently the existence of a disturbance of Ca metabolism by Cd accumulation and especially

EFFECT OF CADMIUM-POLLUTEDRICE

45

its enhancement by reproductive cycles as a whole. On the other hand, we have already discussed the lack of renal dysfunction. Thus our experiments indicate that the disturbance of Ca metabolism occurred without evidence of renal dysfunction. Similar results have also been reported by other workers (Nomiyama et al., 1976). However, enhancement of the disturbance of Ca metabolism by repeated pregnancy and lactation was first reported in our present study, and seemed to correspond to the fact that most Itai-Itai disease patients were multiparous. The mechanism of enhancement by the reproductive cycle should be investigated further. Alkaline phosphatase (A1-P) and inorganic P in serum showed a significant variation with diet or with experimental duration, but there was no characteristic tendency to correlate Cd content of the diet and Cd accumulation in tissues, so that there was no sign of osteomalacia in the serum examination data. The possibility of manifestation of osteomalacia by Ca intake is also a problem for future study of chronic Cd intoxication. Correlations between Examination Indices and Cd Accumulation in Tissues

Cd excretion in feces and urine and Cd accumulation in liver and kidney clearly correlated with Cd content of diet, and this correlation was further enhanced by the reproductive cycle. This Cd accumulation was also shown to correlate with Cd intake estimated by the amounts of food intake and excretion in feces and urine. So it is clear that Cd accumulation in tissues is a result of long-term intake of Cd from diet. However, even in groups with highest Cd intake (diet V), the Cd content of the kidney did not reach the so-called critical concentration reported by Friberg. This result is comparable to data of experiments with low Cd administration reported in the review article (Friberg, 1974; Tsuchiya, 1978), and is also the case in the greater accumulation of Cd in kidney than in liver as an effect of long-term Cd intake. Although the Cd accumulation was relatively small, correlation of examination indices with Cd accumulation in tissues was clearly demonstrated as shown in Table 9. This fact suggests strongly that most of indices examined here were affected by Cd accumulation. SUMMARY

The results may be summarized as follows: 1. Urinary protein, amino acids, and sugar were not shown to have any relationship to the Cd content of the diet. These results suggest that renal dysfunction due to Cd toxicity did not occur in this experiment. 2. Urinary Ca was significantly higher in groups fed diets containing larger amounts of Cd and lower in the parity groups. 3. The results of hematological examination suggest strongly that Cd-polluted rice produces an anemic change. 4. Lower serum Ca in the parity groups in the later stage of the experiment and a significant interaction between diet class and reproductive cycle were shown. These results suggest a loss of Ca from the body in the nonparity groups and a change in Ca metabolism in the parity groups in accordance with the results of urine examination. 5. Alkaline phosphatase and inorganic phosphorus in serum showed no rela-

46

WATANABE ET AL.

tionship to diet and other factors, suggesting no tendency to osteomelacia due to Cd toxicity in the present experiment. 6. B o n e Ca was significantly lowered in groups fed diets containing larger a m o u n t s o f Cd, especially in the parity groups. Soft X-ray radiograms also showed a reduction in the relative thickness of bone cortex in the parity groups, and this reduction was found to be enhanced by Cd-containing diets. 7. Cd accumulation in liver and kidney was very low, but appeared to increase with Cd content o f the diet and experimental duration, especially in the parity groups. 8. Urinary phosphorus, (Ca/P) ratio hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, and bone Ca were found to correlate inversely with Cd concentration in kidney and liver; in addition, the ratio of the thickness of bone cortex to diameter correlated with liver Cd. In conclusion, the present animal experiment with Cd-polluted rice demonstrates the possibility or production of a Cd-dependent anemia and disturbance of Ca metabolism. Furthermore, these Cd effects were enhanced by reproductive activities such as p r e g n a n c y and lactation. However, we could not find any sign of disturbance of renal function in urinary and histological examinations, perhaps because of the low accumulation of Cd in the kidneys, or any sign of osteomalacia in serum examinations. In a comparison between the Cd-polluted rice diet and the CdC12-supplemented nonpolluted rice diet, these biological effects of Cd in the polluted rice seem to be comparable in principle to those of the inorganic Cd salt.

ACKNOWLEDGMENTS This study was supported by Department of Health and Welfare of ToyamaPrefecture. The authors are grateful for the gift of material from the Department of Agriculture and Fishery of ToyamaPrefecture.

REFERENCES Friberg, L., Piscator, M., Nordberg, G., and Kjellstrom, T. (Eds.) (1974). "Cadmium in the Environment." CRC Press, Cleveland, Ohio. Hamilton, D. I., and Smith, M. W. (1978). Inhibition of intestinal calcium uptake by cadmium and the effect of a low calcium diet on cadmium retention. Environ. Res. 15, 175-184. Ishizaki, A., Fukushima, M., Kobayashi, E., and Kurachi, T. (1974). On the retention of orally administered cadmium in rat liver and kidney. Hokuriku J. Public Health 1, 9-15. Itai-Itai Disease Research Committee (1967). "Research about the Cause of So-Called Itai-ltai Disease." Toyama Prefecture Authorities, Toyama, Japan. [in Japanese] Kello, D., Dekamic, D., and Kostial, K. (1979). Influence of sex and dietary calcium on intestinal cadmium absorption in rats. Arch. Environ. Health 34, 30-33. Kobashi, K., Nakai, N., Hase, J., Miyahara, T., Kozuka, H., and Fujii, M. (1978). Chemical forms of cadmium in cadmium-polluted rice. I. Binding properties of glutelin-cadmium complex. Eisei Kagaku 24, 314-321. Matsue, R., Fukuda, H., Honda, S., Hayashi, M., and Kubota, K. (1971). Distribution of cadmium in the rice seed in cadmium polluted districts. Med. Biol. 83, 239-243. Nomiyama, K., et al. (1976). "Effect of Cadmium on Human Health--A Review on Studies Mainly Performed in Japan--(Summary)." Japan Public Health Assoc., Tokyo. Takizawa, Y., Nakamura, I., Kurayama, T., Hirasawa, E, Watanuki, T., and Kawai, K. (1981). Effects of pregnancy on cadmium-treated rats. In "Itai-Itai Disease and Cadmium Toxicity," pp. 6-11. Japan Public Health Assoc. Tokyo. Tsuchiya, K. (Ed.) (1978). "Cadmium Studies in Japan: A Review." Kodansha, Tokyo; Elsevier/ North-Holland, Amsterdam.