THE JOURNAL OF VITAMINOLOGY 18, 159-154 (1972)

Effect of Thiamine on Growth, Tissue Magnesium-, Thiamine Levels and Transketolase Activity in Magnesium Deficient Rats

YOSHINORI ITOKAWA, KIKUKO INOUE, YASUKO NATORI KIMIHIKO OKAZAKI AND MOTONORI FUJIWARA1

Department of Hygiene, Faculty of Medicine, Kyoto University, Kyoto (Post No. 606)

(Received June, 21 1972)

Six groups of Wistar rats were used to demonstrate effects of magnesium deficiencies on thiamine metabolism. Group I was fed both a thiamine and magnesium deficient diet, Group II a thiamine deficient and magnesium sufficient diet, Group III a thiamine adequate and magnesium deficient diet, Group IV a thiamine adequate and magnesium sufficient diet (control), Group V an excess thiamine and magnesium deficient diet and Group VI an excess thiamine and magnesium sufficient diet. Animals were sacrificed after four weeks and magne sium, thiamine and transketolase activities were determined in various tissues. Magnesium concentration in blood cells, plasma and bone decreased markedly in the thiamine supplemented-magnesium deficient groups (Group Ill and V) compared to that in the group deficient in both thiamine and magnesium (Group I). This finding suggests that the thiamine deficiency inhibited utilization of magnesium. Thiamine content of the sciatic nerve, liver and kidney of the magnesium deficient rats were lower than in the magnesium sufficient rats. This suggests that magnesium plays a considerable role in the maintenance of thiamine in tissues. No decrease of thiamine was detected in the central nervous system of magnesium deficient rats. Tissue transketolase activity decreased markedly in thiamine deficient and thiamine-magnesium deficient rats (Group I and II). Addition of thiamine pyrophosphate to tissue homogenates of thiamine deficient rats (Group II) resulted in recovery of transketolase activity, however this pheno menon did not occur when rats were also deficient in magnesium (Group I).

From many biochemical, physiological and mine-deficient rats resulted in rapid recovery of clinical investigations, it is a generally accepted body weight. In thiamine and magnesium view that metabolism of thiamine and magne deficient rats however, thiamine treatment failed sium are interdependent. Evidence is summa to have a restoring effect (4). rized as follows: (c) Chronic alcoholics with manifestations (a) Thiamine dependent enzymes (pyruvate of thiamine deficiency sometimes failed to dehydrogenase, a-ketoglutarate dehydrogenase respond to parenteral injection of thiamine. and transketolase) require magnesium (1-3). Magnesium depletion is also common in these (b) Thiamine treatment on moribund thia patients. Thus there is a possibility that magne

1 糸 川嘉 則 ,井 上 喜 久子,名 取 恭 子,岡 崎 公 彦,藤 原元 典

159 160 ITOKAWA ET AL. sium deficiency interfered with the thiamine less steel cages with a raised wire bottom. response (5, 6). Table 1 shows the composition of the diets fed (d) The activity of transketolase, a thiamine during the experiments. Diets were given ad dependent enzyme, was shown to be depressed libitum and food consumption determined daily. in magnesium deficiency and the recovery Diets were refrigerated at 0-5•Ž until needed. response of this enzyme to thiamine in the Analytical Methods thiamine-depleted animal was incomplete in the After four weeks on these dietary regimens presence of magnesium deficiency (7). animals were sacrificed, blood, brain, spinal cord, (e) Blood serotonin levels increased signifi sciatic nerve, liver, kidney and bone (femur) cantly in the thiamine excess and magnesium were removed. Portions of these tissues were deficient rats rather than the both thiamine and used for determination of magnesium, thiamine magnesium deficient rats (8). or transketolase activity. Red blood cells and

The inter-relationship between thiamine and plasma were centrifugally separated, blood cells magnesium is as yet in the early stages of washed twice with approximately equal volumes investigation. The purpose of the present study of isotonic saline. is to clarify the effect of either excess thiamine Determination of Magnesium: Magne or thiamine deficiency in the magnesium defi sium was determined using atomic absorption cient rats. spectrophotometer after wet oxidation of tissues

(9, 10). MATERIALSAND METHODS Determination of Thiamine: Free or total thiamine was assayed by the thiochrome method Animals and Diets of Fujiwara and Matsui (11). Male rats of the Wistar strain, weighing Transketolase Activity: Tissue transke 80-110 grams, were housed individually in stain tolase was assayed by the method of Brin et al.

TABLE 1 Composition of diets1

1 Zinc values were found to be approx . 150ƒÊg in 100g diet (atomic absorption spectrophotometry) 2 Purchased from Nutritional Biochemical Corporation , Cleveland, Ohio, U. S. A. 3 The Mg free salt mixture contained: (milligrams in 100g diet) NaCl, 170; Na2HPO4•EH2O, 340; K2HPO4, 950; CaH4(PO4)2•EH2O, 530; Fe(C6H5O7) 5H2O, 120; Ca(C3H5O3(2 5H2O, 1290. 4 The thiamine free vitamin mixture contained: (micrograms in 100g diet)

riboflavin, 750; nicotinic acid, 5,000; pyridoxine, 500; cyanocobalamin, 5;

pantothenic acid, 2,500; folic acid, 250; biotin, 40; ascorbic acid, 18,750; ƒ¿-tocopherol, 500; retinyl palmitate, 1 ,250I.U.; Vitamin D2, 100I.U. INTER-RELATIONSHIP OF THIAMINE AND MAGNESIUM 161

(12) with and without in vitro addition of thia TABLE 2 mine pyrophosphate (TPP). Food consumption, magnesium and thiamine intake Protein Determination: Protein was determined by the method of Lowry et al. (13). Statistical Treatment: Significant diffe rences were considered based on student's "t" test (14).

RESULTS

1. Growth 1 Mean•}S .E. of 5rats. The effects on growth of either thiamine 2 Significant difference at p<0.05 compared with

excess or thiamine deficient rats, with and No. IV without magnesium deficiencies are shown on Fig. 1. In thiamine adequate-magnesium suffi reached a peak in about 15 days followed by a cient and thiamine excess-magnesium sufficient progressive loss of weight. rats (Group IV and VI), good growth was Daily intakes of food, magnesium and thia observed, while growth in the thiamine sufficient mine in each experimental group are shown in magnesium deficient groups (Group III and V) Table 2. was slightly depressed. Body weights of the As described in a previous paper (8), after thiamine deficient (Group II) and the both three weeks on these dietary regimens, rats thiamine and magnesium deficient (Group I) rats fed a thiamine containing magnesium deficient diet (Group III and V) developed erythema in the ear and nose as the first manifestation of experimental magnesium deficiency. However, rats that were both thiamine and magnesium deficient showed no such erythema (Group I). 2. Magnesium The magnesium concentration of the plasma and femur decreased markedly in magnesium deficient groups and it is noteworthy that the decrease of magnesium was greater in the thiamine sufficient groups than in the thiamine deficient groups. Table 3 shows the magnesium concentration in each organ. 3. Thiamine Table 4 shows the concentration of thiamine

FIG. 1 Body growth curves of animals (5rat in the various organs. The thiamine concentra average) tion in liver and kidney in the thiamine The terminal weight (mean •} SE. of 5rats) adequate-magnesium deficient rats (Group III) I Thiamine deficient magnesium deficient 111•}172 was lower than that in the magnesium sufficient II Thiamine deficient magnesium sufficient 122•}122 rats (Group IV). On the other hand, in central III Thiamine adequate magnesium deficient 145•}151 IV Thiamine adequate magnesium sufficient 219•}25 nervous tissue, i.e, brain and spinal cord, no V Thiamine excess magnesium deficient 153•}141 decrease was observed in the thiamine content VI Thiamine excess magnesium sufficient 177•}9 of the magnesium deficient groups. Such a 1 Significant difference at p<0.05 compared with tendency was observed also in the brain, spinal No. IV cord, liver and kidney of thiamine deficient 2 Significant difference at p<0.01 compared with No. IV groups (Groups I and II) although the difference 162 ITOKAWA ET AL.

TABLE 3 Magnesium concentration in tissue

1 Mean•}S . E. of 5-8 rats. 2 Significant difference at p<0 .05 compared with No. IV 3 Significant difference at p<0.01 compared with No. IV

TABLE 4 Thiamine concentration in tissues

1 Mean•}S .E. of 5rats. 2 Significant difference at p<0 .05 compared with No. IV 3 Significant difference at p<0.01 compared with No. IV

TABLE 5 Free and total thiamine content and ratio o f free thiamine to total thiamine in brain and liver

was less significant. The ratio of free thiamine 4. Transketolase Activity to total thiamine in the brain and liver revealed Table 6 shows the activity of transketolase no difference among the groups as shown in and the effect of the addition of thiamine pyro Table 5. phosphate. Transketolase activity in red blood INTER-RELATIONSHIP OF THIAMINE AND MAGNESIUM 163

TABLE 6 Transketolase activity

1 Mean•}S .E. of 5rats. 2 Significant difference at p<0 .01 compared with (a) in each group. 3 Significant difference at p<0.05 compared with No. IV 4 Significant difference at p<0 .01 compared with No. IV

cells, brain, spinal cord and liver of rats serotonin. As magnesium is required for the decreased markedly in the thiamine deficient maintenance of the mast cell, these results may rats. No difference in activity between the be interpreted to suggest that excess thiamine thiamine adequate and excess groups was ob may liberate magnesium from the mast cell and served. Regardless of the organ, in vitro as a consequence, serotonin is liberated into the addition of thiamine pyrophosphate to tissue blood. In the present finding, where magnesium homogenates from thiamine deficient rats result concentration was decreased markedly in blood ed in a marked recovery of the transketolase and bone of the thiamine supplemented-magne activity. This was not observed however when sium deficient group, it is apparent that excess rats were deficient in thiamine and magnesium thiamine, in some unknown manner, accelerates

(Group I). These transketolase assays were the magnesium deficiency. On the other hand, carried out in magnesium containing buffer thiamine deficiency does not deplete magnesium. medium (0.7-1.2•@ƒÊmoles/ml) as described by Brin It is also noteworthy that the thiamine et al. (12). To eliminate the effect of magne content in the peripheral tissues of the thiamine sium ion in vitro, magnesium free medium was adequate-magnesium deficient group was lower used to determine transketolase activity in red than that of the thiamine adequate and magne blood cells and liver. There was no marked sium sufficient group. It is possible that magne difference on transketolase in each experimental sium plays a role in the maintenance of thia group when the magnesium ion was omitted mine in cells, presumably binding thiamine to from the reaction mixture. protein in tissues. There was no change in thiamine content in central nervous tissues in DISCUSSION the case of magnesium deficiency, thus suggesting the existence of some barrier system between Our previous observations (8) showed that brain and blood. blood serotonin levels increased significantly in Our observation that the in vitro addition rats fed a thiamine excess and magnesium of thiamine pyrophosphate to tissue homogenates deficient diet while a diet deficient in both of both thiamine and magnesium deficient rats thiamine and magnesium did not elevate blood failed to restore transketolase activity is curious. 164 ITOKAA ET AL.

Similar results were obtained by Zieve et al. elucidate this problem. (7) in blood and liver transketolase of rats. Fennelly et al. (15) reported that the addition ACKNOWLEDGEMENT of thiamine pyrophosphate to red blood cell hemolysates increased the reduced transketolase This work was supported by Research Grants activity in malnourished alcoholic patients with No. 8423 of the Ministry of Education, Japan. a normal liver, but frequently this effect was not seen on transketolase activity in alcoholics REFERENCES with a cirrhotic liver. Of great interest is the fact that this latter condition has a striking 1. Datta, A. G., and Racker, E., J. Biol. Chem., similarity with respect to transketolase activity 236, 624 (1961). seen in the both thiamine and magnesium 2. Koike, M., J. Vitaminol., 14, 86 (1968). deficient rats. 3. Lenn, T. C., Pettit, F. H., and Reed, L. J., Proc. Natl. Acad. Sci. U.S.A., 62, 234 (1969). Amyotropic lateral sclerosis, a degenerative 4. Zieve, L., Ann. N.Y. Acad. Sci., 162, 732 (1969). nerve disease, is prevalent in the Muro district, 5. McCollistar, R.J., Funk, E.B., and Doe, R.P., J. Wakayama Prefecture of Japan. According to Lab. Gun. Med., 55, 98 (1960). the study of Kimura (16) the mineral contents, 6. Zieve, L., and Hill, E., Am. J. Clin. Nutr., 13, i.e. magnesium, manganese and calcium in rivers 312 (1963). as well as the drinking water from this area 7. Zieve, L., Doizaki, W. M., and Stenross, L. E., were lower than in other rivers in Japan. J. Lab. Gun. Med., 72, 268 (1968). Moreover, the mineral content in hair and nails 8. Itokawa, Y., Tanaka, C., and Kimura, M., of these patients was lower than that of the Metabolism, 21, 375 (1972). 9. MacIntyre, I., and Davidsson, D., Biochem. J., normal Japanese. In addition, he reported 70, 456 (1958). many cases cured by the simultaneous admini 10. Alcock, N. W., and MacIntyre, I., Methods of stration of magnesium and thiamine propyl Biochemical Analysis. (D. Glick, Ed.) Inter disulfide, a lipotropic thiamine derivative discov science Publishers, Inc. p. 1. (1966). ered by Fujiwara (17). Thiamine metabolism 11. Fujiwara, M., and Matsui, K., Anal. Chem., 25, of these patients with amyotrophic lateral scle 810 (1953). 12. Brin, M., Tai, M., Ostashever, A. S., and rosis has been investigated by our co-workers(1). Kalinsky, H., J. Nutr., 71, 273 (1960). The results are summarized as follows: blood 13. Lowry, O. H., Rosebrough, N. J., Farr. A. L., and urine thiamine levels are low and when and Randall, R.J., J. Biol. Chem., 193, 265 (1951). 10mg of thiamine was administered to these 14. Snedecor, G.W., Statistical Methods., The Iowa State College Press., p. 76 (1957). patients, more than 80% of the dose was excreted in the urine over a 24 hour period. 15. Fennelly, J., Frank, O., Baker, H., and Leevy, CM., Am. J. Clin. Nutr., 20, 946 (1967). These observations suggest that these patients 16. Kimura, K., Wakayama Med. Reports, 9, 177 lost the ability to utilize thiamine. It is postu (1965). lated that a thiamine deficiency coupled with a 17. Fujiwara, M., Watanabe, H., and Matsui, K., mineral deficiency may be one of the causes of J. Biochem., 41, 29 (1953). such a disease in the nervous tissues. 18. Itokawa, Y., and Cooper, J. R., Biochem. Phar macol., 18, 545 (1968). Recently studies of Itokawa and Cooper 19. Itokawa, Y., and Cooper, J. R., Science 166, concerning thiamine and nervous tissues has 759 (1969). been reported (18-22), which suggest that thia 20. Itokawa, Y., and Cooper, J. R., Biochem. Phar mine has a non-coenzymatic function on the macol., 19, 985 (1970). nervous system. Further investigations on the 21. Itokawa, Y., and Cooper, JR., Biochim. Biophys. Acta, 196, 274 (1970). relationship between thiamine and minerals, 22. Itokawa, Y., Schulz, R. A., and Cooper, J. R., especially in the nervous systems should further Biochim. Biophys. Acta; 266, 293 (1972).

(1) A preliminary report of some of these findings has been made: Fujiwara , M., Sasagawa, S., Itokawa, Y., and Okazaki, K., Japanese J. of Hygiene 17:10 (1962) Abstract.