Diabetes Research and Clinical Practice, 9 (1990) 109-l 14 Elsevier DIABET
109
00386
Changes on levels of B, vitamin and aminotransferase liver of diabetic animals S. Nanbara’,
K. Tanaka’,
H. Koide2,
T. Tanaka2
and T. Hayashi
in the ’
‘Department of Clinical Pathology and ‘Department of The Third Internal Medicine, Osaka Medical College, Takatsuki.Japan (Received 2 August 1989) (Revision received 17 November 1989) (Accepted 23 December 1989)
Summary We measured aminotransferase activity and vitamin B, content in the livers of diabetic mice. Two different types of mice were used for the measurements, spontaneously non-obese diabetic (NOD) or alloxan-induced diabetic (Allo) mice, and control mice were either non-diabetic NOD or Institute of Cancer Research (ICR). The liver of diabetic mice had more aspartate aminotransferase (AST) activity than those of normal mice. The diabetic livers also had more vitamin B, than did normal livers, and pyridoxamine (PM) levels were particularly high but pyridoxal (PL) levels were not. ICR livers showed hepatic alanine aminotransferase activities inversely correlated with blood glucose concentrations, while diabetic livers did not. The abundance of AST and B, in the diabetic liver is consistent with the great need for gluconeogenic substrate there. This is understandable in that most aminotransferases require B, vitamins, and especially the correlation between s-AST and PM levels was recognized in the diabetic liver. Conversely, the AST and PM levels were negatively correlated in normal mice. A metabolic shift towards gluconeogenesis apparently produces more B, and PM while it induced holo-AST synthesis. Key words: Vitamin B 6; Pyridoxamine; (ALT); Liver, diabetic
Aspartate
Introduction Diabetes shifts metabolism towards gluconeogenesis in the liver. Protein catabolism accelerates in the peripheral tissues, particularly the muscles, and hepatocytes take up the resulting amino acids [l-3]. We have reported that the two different Address for correspondence: ment of Clinical Pathology, Takatsuki 569, Japan. 0168-8227/90/$03.50
T. Hayashi, M.D., DepartOsaka Medical College,
0 1990 Elsevier Science Publishers
aminotransferase
(AST);
Alanine aminotransferase
types of diabetic animals, spontaneously nonobese diabetic (NOD) mice and alloxan-induced diabetic (Allo) mice, have unusually high aspartate aminotransferase (AST) activity in the cytosolic fractions of their livers, but not such high levels of alanine aminotransferase (ALT) activity [4]. Most aminotransferases require B, vitamins, using pyridoxal phosphate as a co-enzyme. Previously, working from a different perspective, Kotake and Inada [5], Lepkovsky et al. [6] and Beaton et al. [7] induced the diabetic state by
B.V. (Biomedical
Division)
110
administering xanthurenic acid or withholding B, from the diet. In this report we examine some relationships between aminotransferases and B, levels in the livers of diabetic and normal mice.
Materials and methods NOD mice established at the Shionogi Pharmaceutical Co. Laboratory in Japan [8,9] were used as a diabetic animal model for comparison with Al10 mice. Al10 mice were prepared from ICR mice by injecting alloxan (55.5 mg/kg) into the tail vein, and those that survived 10 days after the injection were compared with diabetic NOD mice. Normal ICR mice at comparable ages were used as controls. Since spontaneous diabetes develops more frequently in female NOD mice r01 --I.. c---l_” __.^_^..,,A \0J, or11y l~ttlki.lt;s WGlt; uszu. Without anesthetics, the animals were rendered unconscious by a hard blow to the head, and their hearts and livers were immediately removed. Soon afterwards the whole organs were washed in ice-cold physiological saline, weighed, and homo-
genized by the method of Solar0 et al. using Triton X-100 [lo]. The activities of AST and ALT in the homogenates were routinely measured by ultravioletabsorbance methods using the HITACHI 736 automatic analyzer (Japan). Vitamin B5 was determined with high-performance liquid chromatography (HPLC), differentiating pyridoxine (PN), pyridoxal (PL), pyridoxamine (PM) from B, group [ 11,121. AST isoenzymes were electrophoretically separated into soluble (s-AST) and mitochondrial (m-AST) fractions by the method of Sakakibara et al. [ 131. The apo-enzymes of AST and ALT were determined according to the method recommended by the International Federation of Clinical Chemistry (IFCC), adding pyridoxal-5’-phosphate to the homogenate, and expressed as the activating rate. The protein con,&lr, nc+l..a ~.-.r,....+..:--~i2c;c;“IU_ ,.-?.,-.-A CGillL “I CUGh~_..,.,,,C,” I‘“III”~GIIQl.GJI.._” w*> US;~CIIIllUCU ing to the method of Lowry et al. [ 141, and the activity of each enzyme and the contents of the vitamin B, group were expressed per gram protein. The blood glucose level was determined with
TABLE 1 B, vitamins and the ratios of apo-aminotransferase normoglycemic and hyperglycemic mice Normoglycemic
V-B; PM” PL” PN” APO/T-ASTb APO/T-ALTb Blood-Gl” PM, pyridoxamine;
activities to total aminotransferase
activities in the liver homogenates
Hyperglycemic
mice
of the
mice
ICR (n = 18)
Non-diabetic NOD (n = 16)
Diabetic NOD (n = 25)
Alloxan (n = 11)
42.1 + 12.2 + 29.8 + < 1.0 4.1 f 10.1 f 97.8 k
45.6 f 10.1 18.2 f 6.2 27.4 + 4.1 < 1.0 4.2 + 2.1 11.0 _+ 3.1 82.4 f 18.4
82.2 f 31.4 44.1 k 22.9 38.0 f. 12.4 < 1.0 4.6 + 1.9 15.9 * 5.7 308 f 111
69.9 f 24.4 33.6 + 15.5 36.3 +_ 14.3 < 1.0 5.8 + 3.4 11.4 f 3.5 355 f 46
PL, pyridoxal;
14.1 6.3 8.1 1.6 3.7 20.8
PN, pyridoxine; T, total; Gl, glucose. Significant differences in V-B,, PM, and PL are and hyperglycemic groups. Significant difference in ape/T-ALT is P < 0.005 between the diabetic NOD group and others. No significant difference is found with ape/T-AST between normoglycemic and hyperglycemic group. a pg/g protein; b %; ’ mg/lOO ml.
P < 0.005 between the normoglycemic
111 Correlation
Diabetic
between
AST
and ALT
NOD mice
Yz-747f036X 04,
’ 1600
2400
3000 AST
I 3600
I 4200
1
1300
IU /g-protein
1900 AST
IU /g -protein
Fig. 1. Relationship between AST and ALT in the liver homogenates of diabetic NOD mice and ICR mice. The coefficient for ICR mice (r = 0.699) was statistically significant (P < 0.05), and that ofnon-diabetic NOD (r = 0.834) was also significant, while those of diabetic NOD and Al10 mice were not significant.
glucose test paper using the TOECHO (Kyoto Daiichi Kagaku), with maximum measurable level of 600 mg/lOO ml. Statistical analyses used the F- and t-tests. P values less than 0.05 were regarded as significant. 700
1
1
Results
OL
60
60
xx
36-2.05X
120
140
.
l
g :
Vitamin B, levels in the liver Table 1 shows all the values from the hyperglycemic mice and the normoglycemic mice. The two control subgroups did not differ in total liver B, or PL contents significantly; the B, of the non-diabetic NOD liver contained about 60 y0 PL while that of ICR livers was about 70 %. On the other hand, the hypergiycemic mice had much more B,, PM and PL in their livers than did each control mouse, and especially increases of PM were noticeable. Thus they had lower PL/total B, ratios than controls. This was 46.2% and 51.9% in diabetic NOD and Al10 mice, respectively. PN was at trace levels in the livers of both control and diabetic mice.
Yz795
40
Dlobetlc
NOD mice
b . . . .
.
.
. .
Fig. 2. Relationship between blood glucose levels and hepatic ALT activities of ICR and diabetic NOD mice. The coefficient between hepatic ALT and blood glucose was significant only in ICR mice (r = - 0.538, P < 0.05). Those between hepatic AST and the glucose were not significant at all.
112
AST-ALT correlation in the liver AST and ALT activities were signifkantly correlated in the livers of ICR mice (r = 0.699). NOD m-ice do have such (Y &____a_ correlation _______..____ \. = 0,834) before becoming diabetic, in normoglycemic stage. On the other hand, the correlations were not significant in diabetic livers of diabetic NOD and Al10 mice (r = 0.327 and r = - 0.258, respectively).
o.
Dlabetlc
NOD
Fig. 1 illustrates the correlation between hepatic AST and ALT in diabetic NOD mice and ICR mice. Correlation between hepatic ALT activities and blood glucose levels ICR mice had a significant negative correlation between hepatic ALT activity and blood glucose
Correlotlon
between
V-B,
r=
-0.435
and
s-
AST
ICR mice
mice
I r=0.542
28OOC
0 700
0 i
Y.892.07+8.79x 5 h
or
t
’ 40
*
’ 80
*
’ 120
*
’ 160
B
1 V-B5
b.
Dlobetlc
NOD
)-1g /g-proteu-
Correlation
between
PM
ond
mice
s- AST
ICR r:
2800
mice
-0.405 l .
700 !
le
I,, , , , 8
l
20
40
1
Y ~986.51
60
+14.23X t
80 PM
Fig. 3. Relationship hnmnoc=nster ..“...“~-..U””
statistically
between vitamin B,, pyridoxamine
nf “1 Aicahc=tic. . . ..“I.._ Nnll ..VY
&r-s. 1.11”” snrl -1..
TCR .“I.
mire= L.llWl.
IJ~ /g-protein
(PM), and cytosolic aspartate The 111w rnrffirientr -“I..II.~u.“,
r z
“.542
an,4 -1..
aminotransferase
(s-AST) in the liver
r c l-l /;ACl ;n .Gnh\nt;,. I - “.“7”, 111 UlUVIbLI
Nnn I.VU
mica ~XIPIP 11,1w\1nrlr
significant (P < 0.05), while r = - 0.435 and r = 0.405 were not. The coefficient between total AST and B, or PM were also significant (P < 0.05) in diabetic NOD mice (r = 0.507 and r = 0.571, respectively).
113
level (Y = - 0.538). Neither group in the nondiabetic and diabetic NOD mice nor the Al10 mice showed any relationship between hepatic AST and blood glucose (r = - 0.087 and r = - 0.055, respectively). Fig. 2 illustrates the correlation between hepatic ALT and blood glucose in ICR and diabetic NOD mice. AST and ALT apo-enzymes in the liver Table 1 also shows that both hyperglycemic and normoglycemic mice had lower apo-AST/total AST ratios than apo-ALT/total ALT in the liver. No group significantly differed in apo-AST ratios. In apo-ALT ratios, however, only diabetic NOD mice had a significantly high ratio against the others. Correlation between aminotransferases and B, vita-^:-- Ilf :.. trsx +L- uver I.‘...., III&J ALT activity correlated with none of the B, vitamins in the liver. AST activity did correlate with B, and PM in diabetic NOD livers (Y = 0.507 and 0.571, respectively) but not in control nondiabetic NOD livers (r = 0.07 1 and 0.184, respectively). The isoenzyme s-AST showed clear correlations with B, and PM in diabetic NOD liver. AST in the Allo liver also showed correlations with B, and PM (r = 0.795 and 0.963, respectively). On the other hand, control ICR livers showed no correlation between AST and B, or PM (r = 0.085 and 0.109, respectively), but tended towards negative correlations between s-AST and B, or PM. Fig. 3 illustrates the correlation only in diabetic NOD mice and ICR mice. PL in the liver correlated with none of the aminotransferases.
[ 171 separately determined the isoenzymes of AST in the liver, and detected the marked induction of AST isoenzymes in the soluble fraction when the metabolic condition shifted towards gluconeogenesis, but observed few changes in those localized in mitochondria. We have also recognized that AST activity in the liver increased more markedly in the hyperglycemic mice, the diabetic NOD and Allo mice, than in the normoglycemic mice, the ICR and non-diabetic NOD mice [4]. Abnormally high AST activity was seen in the cytosolic fraction but not in the mitochondrial fraction [ 41. These aminotransferases require pyridoxal phosphate of the B, vitamins to turn into the holotype. A relationship between the diabetic state and B, deficiency has been already reported by Kotake et al. [5-71. From a different _.:_-.--Z-r n _-c-l__1:_-v1ewpuu11 __.^ we _&._AZ-l sLuuItxl Dg IIltmlouIIsIIl in aie diabetic liver using the HPLC method of B, determination [ 11,121 using two different types of diabetic animals. Not only the increase of AST activity but also that of B, content was distinctly recognized in the diabetic liver of NOD and Al10 mice. This change of the total B,v in ~~~the -~~ diabetic liver depends on the increase of the PM fraction but not the PL fraction. ALT functions in the glucose-alanine interorgan cycle that regulates the formation of glucose from pyruvate [2,3]. This gluconeogenesis can stem from various sources : starvation or obesity, from exercise, pregnancy, neonatal development, insulin deficiency, and glucocorticoid excess [ 21. In this study the control mice ICR had hepatic ALT activities that inversely correlated with their blood glucose levels. This fact means that ALT will play an important role in gluconeogenesis in tha nnrmnl CI1td ll”lIxlclJ
Discussion NOD mice were recently used as a model of human IDDM [8,9], and the increase of gluconeogenic key enzymes and the decrease of glycolytic key enzymes have been reported in the livers of NOD mice as human diabetics [ 151. Katsunuma and Katsunuma [ 161 and Pestana
l&m,.. I,“GI.
c\m +ha ,.+I.,, “I1 CIICZ “LIIGII
I.,...A UCulU
A..- ,a:_~_,,+:, “I.&I UIil”Cl,I~
did not. In the diabetic liver, oxaloacetate and aspartate would more often be substrates for the gluconeogenesis than would be for pyruvate or alanine. Although diabetic and control mice did not differ in their apo-AST ratios, the diabetic NOD mice did have significantly more apo-ALT. During the formation of holoenzymes, AST was preferentially activated with the co-enzyme pyrimice
114
doxal phosphate in the diabetic NOD liver while ALT was left unfinished. A difference observed in the apo-ALT ratio between NOD and Al10 liver would result from the cause diabetes, whether spontaneously or chemically. Aminotransferases can transform amino acids to carbohydrate and fat and vice versa. Diabetic livers of NOD and Allo mice had more B, as well as more AST, owing to more abundant PM. Some of this extra B, may have come from deconstrutted muscle [3,18]. In these diabetic mice, PM usurped PL’s position as the dominant fraction of the hepatic B, content in normoglycemic mice. Hyperglycemia caused the loss of the clear correlation between AST and ALT in normoglycemic livers; conversely, diabetes caused a new correlation in the liver to emerge, i.e., that between AST activity and B, content, particularly s-AST and PM. Diabetes would necessitate more gluconeogenesis, more AST activity in the cytosolic fraction, and more PM produced by B, metabolism in the liver. This requirement in the diabetic liver might be met by salvaged vitamins from muscles in the same manner as amino acids [3,18], and thereby B, distribution on the whole would be localized.
References 1 Felig, P., Marliss, E., Ohman, J.L. and Cahill, G.F. (1970) Plasma amino acid levels in diabetic ketoacidosis. Diabetes 19, 723-729. 2 Exton, J.H. (1972) Progress in endocrinology and metabolism: gluconeogenesis. Metabolism 10, 945-990. 3 Bender, D.A. (1985) Nitrogen balance and protein turnover. In: D.A. Bender (Ed.), Aminoacid Metabolism. John Wiley and Sons, New York, pp. 56-62. 4 Tanaka, K., Nanbara, S., Tanaka, T., Koide, H. and Hayashi, T. (1988) Aminotransferase activity in the liver of diabetic mice. Diab. Res. Clin. Pratt. 5, 71-75. 5 Kotake, Y. and Inada, T. (1953) Studies on xanthurenic acid, I, II. J. Biochem. 40, 287-294.
6 Lepkovsky, S., Roboz, E. and Haggen-Smit, A.J. (1943) Xanthurenic acid and its role in the tryptophan metabolism of pyridoxine-deficient rats. J. Biol. Chem. 149, 195-201. 7 Beaton, G.H., Haufschild, A.M. and Mchenry, E.W. (1956) Prevention of inanition in vitamin B,-deprived rats by insulin treatment. J. Nutr. 60, 455-462. 8 Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K. and Tochino, Y. (1980) Breeding of a nonobese diabetic strain of mice. Exp. Anim. 29, 1-13. 9 Fujita, T., Yui, R., Kusumoto, Y., Serizawa, Y., Makino, S. and Tochino, Y. Lymphocytic insulitis in a non-obese diabetic (NOD) strain of mice. Biomed. Res. 3,429-443. 10 Solaro, R.J., Pang, D.C. and Biggs, F.N. (1971) The purification of cardiac myofibrils with Triton X-100. Biochim. Biophys. Acta 245, 259-262. 11 Yoshida, T., Yunoki, N., Nakajima, Y., Kaito, T. and Anmo, T. (1978) Simultaneous determination of vitamin B, group in blood by high-performance liquid chromatography (HPLC). Yakugaku Zasshi 98, 1319-1326 (in Japanese). 12 Tani, Y. (1983) Assay methods of vitamin B, 1. Separation and determination of total B,. Vitamins 57, 263-271 (in Japanese). 13 Sakakibara, S., Shiomi, K., Kobayashi, S., Inai, S. and Kagamiyama, H. (1983) A convenient and sensitive method for the determination of serum aspartate aminotransferase isozymes after electrophoresis. Clin. Chim. Acta 133, 119-123. 14 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. 15 Tochino, Y. (1986) Metabolic disturbances in the NOD mouse. In: Tarui, S., Tochino, Y. and Nonaka, K. (Eds.), Insulin and Type 1 Diabetes: Lessons from the NOD Mouse. Academic Press, Tokyo, pp. 217-224. 16 Katsunuma, N. and Katsunuma, T. (1965) The relationship between intracellular localization and function of transaminase isozyme. Jpn. Clin. Sci. 1, 803-811 (in Japanese). 17 Pestana, A. (1969) Dietary and hormonal control of enzymes of amino acid catabolism in liver. Eur. J. Biochem. 11,400-404. 18 Leklem, J.E. (1984) Physical activity and vitamin B, metabolism in men and women. In: Reynolds, R.D. and Leklem, J.E. (Eds.), Current Topics in Nutrition and Disease, Vol. 13, Vitamin B,. Alan R. Liss, New York, pp. 221-241.
109
00386
Changes on levels of B, vitamin and aminotransferase liver of diabetic animals S. Nanbara’,
K. Tanaka’,
H. Koide2,
T. Tanaka2
and T. Hayashi
in the ’
‘Department of Clinical Pathology and ‘Department of The Third Internal Medicine, Osaka Medical College, Takatsuki.Japan (Received 2 August 1989) (Revision received 17 November 1989) (Accepted 23 December 1989)
Summary We measured aminotransferase activity and vitamin B, content in the livers of diabetic mice. Two different types of mice were used for the measurements, spontaneously non-obese diabetic (NOD) or alloxan-induced diabetic (Allo) mice, and control mice were either non-diabetic NOD or Institute of Cancer Research (ICR). The liver of diabetic mice had more aspartate aminotransferase (AST) activity than those of normal mice. The diabetic livers also had more vitamin B, than did normal livers, and pyridoxamine (PM) levels were particularly high but pyridoxal (PL) levels were not. ICR livers showed hepatic alanine aminotransferase activities inversely correlated with blood glucose concentrations, while diabetic livers did not. The abundance of AST and B, in the diabetic liver is consistent with the great need for gluconeogenic substrate there. This is understandable in that most aminotransferases require B, vitamins, and especially the correlation between s-AST and PM levels was recognized in the diabetic liver. Conversely, the AST and PM levels were negatively correlated in normal mice. A metabolic shift towards gluconeogenesis apparently produces more B, and PM while it induced holo-AST synthesis. Key words: Vitamin B 6; Pyridoxamine; (ALT); Liver, diabetic
Aspartate
Introduction Diabetes shifts metabolism towards gluconeogenesis in the liver. Protein catabolism accelerates in the peripheral tissues, particularly the muscles, and hepatocytes take up the resulting amino acids [l-3]. We have reported that the two different Address for correspondence: ment of Clinical Pathology, Takatsuki 569, Japan. 0168-8227/90/$03.50
T. Hayashi, M.D., DepartOsaka Medical College,
0 1990 Elsevier Science Publishers
aminotransferase
(AST);
Alanine aminotransferase
types of diabetic animals, spontaneously nonobese diabetic (NOD) mice and alloxan-induced diabetic (Allo) mice, have unusually high aspartate aminotransferase (AST) activity in the cytosolic fractions of their livers, but not such high levels of alanine aminotransferase (ALT) activity [4]. Most aminotransferases require B, vitamins, using pyridoxal phosphate as a co-enzyme. Previously, working from a different perspective, Kotake and Inada [5], Lepkovsky et al. [6] and Beaton et al. [7] induced the diabetic state by
B.V. (Biomedical
Division)
110
administering xanthurenic acid or withholding B, from the diet. In this report we examine some relationships between aminotransferases and B, levels in the livers of diabetic and normal mice.
Materials and methods NOD mice established at the Shionogi Pharmaceutical Co. Laboratory in Japan [8,9] were used as a diabetic animal model for comparison with Al10 mice. Al10 mice were prepared from ICR mice by injecting alloxan (55.5 mg/kg) into the tail vein, and those that survived 10 days after the injection were compared with diabetic NOD mice. Normal ICR mice at comparable ages were used as controls. Since spontaneous diabetes develops more frequently in female NOD mice r01 --I.. c---l_” __.^_^..,,A \0J, or11y l~ttlki.lt;s WGlt; uszu. Without anesthetics, the animals were rendered unconscious by a hard blow to the head, and their hearts and livers were immediately removed. Soon afterwards the whole organs were washed in ice-cold physiological saline, weighed, and homo-
genized by the method of Solar0 et al. using Triton X-100 [lo]. The activities of AST and ALT in the homogenates were routinely measured by ultravioletabsorbance methods using the HITACHI 736 automatic analyzer (Japan). Vitamin B5 was determined with high-performance liquid chromatography (HPLC), differentiating pyridoxine (PN), pyridoxal (PL), pyridoxamine (PM) from B, group [ 11,121. AST isoenzymes were electrophoretically separated into soluble (s-AST) and mitochondrial (m-AST) fractions by the method of Sakakibara et al. [ 131. The apo-enzymes of AST and ALT were determined according to the method recommended by the International Federation of Clinical Chemistry (IFCC), adding pyridoxal-5’-phosphate to the homogenate, and expressed as the activating rate. The protein con,&lr, nc+l..a ~.-.r,....+..:--~i2c;c;“IU_ ,.-?.,-.-A CGillL “I CUGh~_..,.,,,C,” I‘“III”~GIIQl.GJI.._” w*> US;~CIIIllUCU ing to the method of Lowry et al. [ 141, and the activity of each enzyme and the contents of the vitamin B, group were expressed per gram protein. The blood glucose level was determined with
TABLE 1 B, vitamins and the ratios of apo-aminotransferase normoglycemic and hyperglycemic mice Normoglycemic
V-B; PM” PL” PN” APO/T-ASTb APO/T-ALTb Blood-Gl” PM, pyridoxamine;
activities to total aminotransferase
activities in the liver homogenates
Hyperglycemic
mice
of the
mice
ICR (n = 18)
Non-diabetic NOD (n = 16)
Diabetic NOD (n = 25)
Alloxan (n = 11)
42.1 + 12.2 + 29.8 + < 1.0 4.1 f 10.1 f 97.8 k
45.6 f 10.1 18.2 f 6.2 27.4 + 4.1 < 1.0 4.2 + 2.1 11.0 _+ 3.1 82.4 f 18.4
82.2 f 31.4 44.1 k 22.9 38.0 f. 12.4 < 1.0 4.6 + 1.9 15.9 * 5.7 308 f 111
69.9 f 24.4 33.6 + 15.5 36.3 +_ 14.3 < 1.0 5.8 + 3.4 11.4 f 3.5 355 f 46
PL, pyridoxal;
14.1 6.3 8.1 1.6 3.7 20.8
PN, pyridoxine; T, total; Gl, glucose. Significant differences in V-B,, PM, and PL are and hyperglycemic groups. Significant difference in ape/T-ALT is P < 0.005 between the diabetic NOD group and others. No significant difference is found with ape/T-AST between normoglycemic and hyperglycemic group. a pg/g protein; b %; ’ mg/lOO ml.
P < 0.005 between the normoglycemic
111 Correlation
Diabetic
between
AST
and ALT
NOD mice
Yz-747f036X 04,
’ 1600
2400
3000 AST
I 3600
I 4200
1
1300
IU /g-protein
1900 AST
IU /g -protein
Fig. 1. Relationship between AST and ALT in the liver homogenates of diabetic NOD mice and ICR mice. The coefficient for ICR mice (r = 0.699) was statistically significant (P < 0.05), and that ofnon-diabetic NOD (r = 0.834) was also significant, while those of diabetic NOD and Al10 mice were not significant.
glucose test paper using the TOECHO (Kyoto Daiichi Kagaku), with maximum measurable level of 600 mg/lOO ml. Statistical analyses used the F- and t-tests. P values less than 0.05 were regarded as significant. 700
1
1
Results
OL
60
60
xx
36-2.05X
120
140
.
l
g :
Vitamin B, levels in the liver Table 1 shows all the values from the hyperglycemic mice and the normoglycemic mice. The two control subgroups did not differ in total liver B, or PL contents significantly; the B, of the non-diabetic NOD liver contained about 60 y0 PL while that of ICR livers was about 70 %. On the other hand, the hypergiycemic mice had much more B,, PM and PL in their livers than did each control mouse, and especially increases of PM were noticeable. Thus they had lower PL/total B, ratios than controls. This was 46.2% and 51.9% in diabetic NOD and Al10 mice, respectively. PN was at trace levels in the livers of both control and diabetic mice.
Yz795
40
Dlobetlc
NOD mice
b . . . .
.
.
. .
Fig. 2. Relationship between blood glucose levels and hepatic ALT activities of ICR and diabetic NOD mice. The coefficient between hepatic ALT and blood glucose was significant only in ICR mice (r = - 0.538, P < 0.05). Those between hepatic AST and the glucose were not significant at all.
112
AST-ALT correlation in the liver AST and ALT activities were signifkantly correlated in the livers of ICR mice (r = 0.699). NOD m-ice do have such (Y &____a_ correlation _______..____ \. = 0,834) before becoming diabetic, in normoglycemic stage. On the other hand, the correlations were not significant in diabetic livers of diabetic NOD and Al10 mice (r = 0.327 and r = - 0.258, respectively).
o.
Dlabetlc
NOD
Fig. 1 illustrates the correlation between hepatic AST and ALT in diabetic NOD mice and ICR mice. Correlation between hepatic ALT activities and blood glucose levels ICR mice had a significant negative correlation between hepatic ALT activity and blood glucose
Correlotlon
between
V-B,
r=
-0.435
and
s-
AST
ICR mice
mice
I r=0.542
28OOC
0 700
0 i
Y.892.07+8.79x 5 h
or
t
’ 40
*
’ 80
*
’ 120
*
’ 160
B
1 V-B5
b.
Dlobetlc
NOD
)-1g /g-proteu-
Correlation
between
PM
ond
mice
s- AST
ICR r:
2800
mice
-0.405 l .
700 !
le
I,, , , , 8
l
20
40
1
Y ~986.51
60
+14.23X t
80 PM
Fig. 3. Relationship hnmnoc=nster ..“...“~-..U””
statistically
between vitamin B,, pyridoxamine
nf “1 Aicahc=tic. . . ..“I.._ Nnll ..VY
&r-s. 1.11”” snrl -1..
TCR .“I.
mire= L.llWl.
IJ~ /g-protein
(PM), and cytosolic aspartate The 111w rnrffirientr -“I..II.~u.“,
r z
“.542
an,4 -1..
aminotransferase
(s-AST) in the liver
r c l-l /;ACl ;n .Gnh\nt;,. I - “.“7”, 111 UlUVIbLI
Nnn I.VU
mica ~XIPIP 11,1w\1nrlr
significant (P < 0.05), while r = - 0.435 and r = 0.405 were not. The coefficient between total AST and B, or PM were also significant (P < 0.05) in diabetic NOD mice (r = 0.507 and r = 0.571, respectively).
113
level (Y = - 0.538). Neither group in the nondiabetic and diabetic NOD mice nor the Al10 mice showed any relationship between hepatic AST and blood glucose (r = - 0.087 and r = - 0.055, respectively). Fig. 2 illustrates the correlation between hepatic ALT and blood glucose in ICR and diabetic NOD mice. AST and ALT apo-enzymes in the liver Table 1 also shows that both hyperglycemic and normoglycemic mice had lower apo-AST/total AST ratios than apo-ALT/total ALT in the liver. No group significantly differed in apo-AST ratios. In apo-ALT ratios, however, only diabetic NOD mice had a significantly high ratio against the others. Correlation between aminotransferases and B, vita-^:-- Ilf :.. trsx +L- uver I.‘...., III&J ALT activity correlated with none of the B, vitamins in the liver. AST activity did correlate with B, and PM in diabetic NOD livers (Y = 0.507 and 0.571, respectively) but not in control nondiabetic NOD livers (r = 0.07 1 and 0.184, respectively). The isoenzyme s-AST showed clear correlations with B, and PM in diabetic NOD liver. AST in the Allo liver also showed correlations with B, and PM (r = 0.795 and 0.963, respectively). On the other hand, control ICR livers showed no correlation between AST and B, or PM (r = 0.085 and 0.109, respectively), but tended towards negative correlations between s-AST and B, or PM. Fig. 3 illustrates the correlation only in diabetic NOD mice and ICR mice. PL in the liver correlated with none of the aminotransferases.
[ 171 separately determined the isoenzymes of AST in the liver, and detected the marked induction of AST isoenzymes in the soluble fraction when the metabolic condition shifted towards gluconeogenesis, but observed few changes in those localized in mitochondria. We have also recognized that AST activity in the liver increased more markedly in the hyperglycemic mice, the diabetic NOD and Allo mice, than in the normoglycemic mice, the ICR and non-diabetic NOD mice [4]. Abnormally high AST activity was seen in the cytosolic fraction but not in the mitochondrial fraction [ 41. These aminotransferases require pyridoxal phosphate of the B, vitamins to turn into the holotype. A relationship between the diabetic state and B, deficiency has been already reported by Kotake et al. [5-71. From a different _.:_-.--Z-r n _-c-l__1:_-v1ewpuu11 __.^ we _&._AZ-l sLuuItxl Dg IIltmlouIIsIIl in aie diabetic liver using the HPLC method of B, determination [ 11,121 using two different types of diabetic animals. Not only the increase of AST activity but also that of B, content was distinctly recognized in the diabetic liver of NOD and Al10 mice. This change of the total B,v in ~~~the -~~ diabetic liver depends on the increase of the PM fraction but not the PL fraction. ALT functions in the glucose-alanine interorgan cycle that regulates the formation of glucose from pyruvate [2,3]. This gluconeogenesis can stem from various sources : starvation or obesity, from exercise, pregnancy, neonatal development, insulin deficiency, and glucocorticoid excess [ 21. In this study the control mice ICR had hepatic ALT activities that inversely correlated with their blood glucose levels. This fact means that ALT will play an important role in gluconeogenesis in tha nnrmnl CI1td ll”lIxlclJ
Discussion NOD mice were recently used as a model of human IDDM [8,9], and the increase of gluconeogenic key enzymes and the decrease of glycolytic key enzymes have been reported in the livers of NOD mice as human diabetics [ 151. Katsunuma and Katsunuma [ 161 and Pestana
l&m,.. I,“GI.
c\m +ha ,.+I.,, “I1 CIICZ “LIIGII
I.,...A UCulU
A..- ,a:_~_,,+:, “I.&I UIil”Cl,I~
did not. In the diabetic liver, oxaloacetate and aspartate would more often be substrates for the gluconeogenesis than would be for pyruvate or alanine. Although diabetic and control mice did not differ in their apo-AST ratios, the diabetic NOD mice did have significantly more apo-ALT. During the formation of holoenzymes, AST was preferentially activated with the co-enzyme pyrimice
114
doxal phosphate in the diabetic NOD liver while ALT was left unfinished. A difference observed in the apo-ALT ratio between NOD and Al10 liver would result from the cause diabetes, whether spontaneously or chemically. Aminotransferases can transform amino acids to carbohydrate and fat and vice versa. Diabetic livers of NOD and Allo mice had more B, as well as more AST, owing to more abundant PM. Some of this extra B, may have come from deconstrutted muscle [3,18]. In these diabetic mice, PM usurped PL’s position as the dominant fraction of the hepatic B, content in normoglycemic mice. Hyperglycemia caused the loss of the clear correlation between AST and ALT in normoglycemic livers; conversely, diabetes caused a new correlation in the liver to emerge, i.e., that between AST activity and B, content, particularly s-AST and PM. Diabetes would necessitate more gluconeogenesis, more AST activity in the cytosolic fraction, and more PM produced by B, metabolism in the liver. This requirement in the diabetic liver might be met by salvaged vitamins from muscles in the same manner as amino acids [3,18], and thereby B, distribution on the whole would be localized.
References 1 Felig, P., Marliss, E., Ohman, J.L. and Cahill, G.F. (1970) Plasma amino acid levels in diabetic ketoacidosis. Diabetes 19, 723-729. 2 Exton, J.H. (1972) Progress in endocrinology and metabolism: gluconeogenesis. Metabolism 10, 945-990. 3 Bender, D.A. (1985) Nitrogen balance and protein turnover. In: D.A. Bender (Ed.), Aminoacid Metabolism. John Wiley and Sons, New York, pp. 56-62. 4 Tanaka, K., Nanbara, S., Tanaka, T., Koide, H. and Hayashi, T. (1988) Aminotransferase activity in the liver of diabetic mice. Diab. Res. Clin. Pratt. 5, 71-75. 5 Kotake, Y. and Inada, T. (1953) Studies on xanthurenic acid, I, II. J. Biochem. 40, 287-294.
6 Lepkovsky, S., Roboz, E. and Haggen-Smit, A.J. (1943) Xanthurenic acid and its role in the tryptophan metabolism of pyridoxine-deficient rats. J. Biol. Chem. 149, 195-201. 7 Beaton, G.H., Haufschild, A.M. and Mchenry, E.W. (1956) Prevention of inanition in vitamin B,-deprived rats by insulin treatment. J. Nutr. 60, 455-462. 8 Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K. and Tochino, Y. (1980) Breeding of a nonobese diabetic strain of mice. Exp. Anim. 29, 1-13. 9 Fujita, T., Yui, R., Kusumoto, Y., Serizawa, Y., Makino, S. and Tochino, Y. Lymphocytic insulitis in a non-obese diabetic (NOD) strain of mice. Biomed. Res. 3,429-443. 10 Solaro, R.J., Pang, D.C. and Biggs, F.N. (1971) The purification of cardiac myofibrils with Triton X-100. Biochim. Biophys. Acta 245, 259-262. 11 Yoshida, T., Yunoki, N., Nakajima, Y., Kaito, T. and Anmo, T. (1978) Simultaneous determination of vitamin B, group in blood by high-performance liquid chromatography (HPLC). Yakugaku Zasshi 98, 1319-1326 (in Japanese). 12 Tani, Y. (1983) Assay methods of vitamin B, 1. Separation and determination of total B,. Vitamins 57, 263-271 (in Japanese). 13 Sakakibara, S., Shiomi, K., Kobayashi, S., Inai, S. and Kagamiyama, H. (1983) A convenient and sensitive method for the determination of serum aspartate aminotransferase isozymes after electrophoresis. Clin. Chim. Acta 133, 119-123. 14 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. 15 Tochino, Y. (1986) Metabolic disturbances in the NOD mouse. In: Tarui, S., Tochino, Y. and Nonaka, K. (Eds.), Insulin and Type 1 Diabetes: Lessons from the NOD Mouse. Academic Press, Tokyo, pp. 217-224. 16 Katsunuma, N. and Katsunuma, T. (1965) The relationship between intracellular localization and function of transaminase isozyme. Jpn. Clin. Sci. 1, 803-811 (in Japanese). 17 Pestana, A. (1969) Dietary and hormonal control of enzymes of amino acid catabolism in liver. Eur. J. Biochem. 11,400-404. 18 Leklem, J.E. (1984) Physical activity and vitamin B, metabolism in men and women. In: Reynolds, R.D. and Leklem, J.E. (Eds.), Current Topics in Nutrition and Disease, Vol. 13, Vitamin B,. Alan R. Liss, New York, pp. 221-241.
