- Email: [email protected]
Experimental diabetes causes mitochondrial loss and cytoplasmic enrichment of pyridoxal phosphate and aspartate aminotransferase activity
BIOCHEMICAL
MEDICINE
AND
METABOLIC
BIOLOGY
36, 91-97
(1986)
Experimental Diabetes Causes Mitochondrial Loss and Cytoplasmic Enrichment of Pyridoxal Phosphate and Aspartate Aminotransferase Activity KENNETH Department
S. ROGERS, EDWIN S. HIGGINS, AND EDWARD S. KLINE
of Biochemistry,
Medical College Richmond, Received
of Virginia, Virginia Virginia 23298
Commonwealth
University,
June 3, 1985
Streptozotocin diabetic rats excrete abnormal quantities of kynurenine metabolites after a tryptophan load (1) and this indicated to us that they may be vitamin B6 deficient. Lack of vitamin B, in normal rats causes 3-hydroxykynurenine to accumulate and be excreted (2,3) because pyridoxal phosphate is then unavailable as the coenzyme for kynureninase activity (4). Moreover, excretion of aspartate aminotransferase (EC 2.6.1. l), which contains pyridoxal phosphate (5), is accelerated in streptozotocin diabetes (6). Additional evidence for an altered vitamin Bs metabolism in the diabetic comes from the work of Davis et al. (7) who reported in a study of 518 diabetic patients of varying age and sex that 25% of these patients had decreased serum pyridoxal levels when compared to 371 healthy volunteers. We have selected the streptozotocin diabetic rat as a model to study how insulin deficiency alters vitamin B6 utilization by focusing on pyridoxal phosphate levels and aspartate aminotransferase activities in liver tissues. Lui and coworkers (8,9) reported that 20% of normal rat liver vitamin B6 is transported into and stored by the mitochondria which do not contain any enzymes for the vitamin’s synthesis or degradation. Liver also contains two isozymes of aspartate aminotransferase which are electrophoretically distinct; the cytoplasmic enzyme migrates anodally and the mitochondrial enzyme migrates cathodally (IO, 1 I). We report that streptozotocin diabetic rats show plasma deficiencies of pyridoxal phosphate accompanied by less storage in the liver of mitochondrial pyridoxal phosphate and aspartate aminotransferase. This is the first time to our knowledge that pyridoxal phosphate levels have been determined in the diabetic rat. MATERIALS AND METHODS Sprague-Dawley male rats were obtained from Flow Laboratories (Dublin, Va.). They were individually housed, placed on a light regime of 12 hr on/12 hr off, and given water and Purina rat chow pellets ad libitum. They were made diabetic after a 24-hr fast by injecting into the tail vein streptozotocin, 70 mg/kg 91 088545OY86
$3.00
Copyright Q 1986 by Academic Press, Inc. All rights of reproduction in any form reserved.
92
ROGERS,
HIGGINS,
AND
KLINE
body wt (12). After 15 weeks, the rats were decapitated, blood was collected, and the liver was excised immediately, weighed, and homogenized in ice-cold 0.25 M sucrose-O.5 mM disodium ethylenediaminetetraacetate (EDTA) solution, pH 7.4. Mitochondria were obtained from the liver as previously described (13) with the exception that the nuclear pellet was not extracted for residual mitochondria. The supernatant from the first mitochondrial collection was considered to represent liver cytoplasm and was saved for analyses. Mitochondrial and cytoplasmic protein was estimated in the presence of 1% sodium deoxycholate by a biuret method (14) with crystalline bovine sernm albumin as a standard. Pyridoxal phosphate analysis. Pyridoxal phosphate was extracted from plasma, liver mitochondria, and cytoplasm following the procedures of Lumeng and coworkers (8,9) which used trichloroacetic acid (TCA) and heating to 50” to precipitate protein and release the coenzyme, centrifugation, and extraction of the supernates with peroxide-free diethyl ether to provide a TCA- and protein-free sample of pH 7 for analysis. Pyridoxal phosphate was.assayed in samples incubated at 30” for 1 hr with mitochondrial aposerine hydroxymethyltransferase (EC 2.12.1) prepared (15,16) for us through the courtesy of Dr. LaVerne Schirch, Department of Biochemistry, MCV. The incubation medium contained 1.8 ml of 20 mM potassium phosphate, 2 mM EDTA, pH 7.0; 20 ~1 of water or 6 PM pyridoxal phosphate; 0.5 ml of water; and 10 ~1 of aposerine hydroxymethyltransferase (4.2 mg protein/ml) in phosphate-EDTA buffer. After incubation, 40 ~1 of NADH-alcohol dehydrogenase mixture (12 mg NADH, 8 mg crystalline yeast alcohol dehydrogenase in 2 ml of phosphate-EDTA buffer) was added. The enzyme reaction was started by addition of 0.1 ml DL-allothreonine (296 mg DL-allothreonine in 5 ml of water). This coupled enzyme assay (15) pemitted rapid quantitation of serine hydroxymethyltransferase activity and thereby pyridoxal phosphate content since activity of the enzyme was proportional to the amount of pyridoxal phosphate in the tissue samples. A typical standard curve of enzyme velocity (change in absorbance at 340 nm per minute) was linear over the concentration range of pyridoxal phosphate measured (0 to 500 pmole). Aspartute aminotransferase. Aspartate aminotransferase was assayed in liver cytoplasm and mitochondria following a slightly modified procedure of Bergmeyer and Bernt (5). The coupled enzyme reaction mixtures were adjusted to 0.1% Triton X-100 concentration in order to solubilize particulate matter. This concentration of detergent did not inhibit the involved enzymes. Endogenous absorbance changes due to liver glutamate dehydrogenase activity were corrected by subtracting appropriate controls containing no aspartate in the reaction mixtures. Plasma aspartate aminotransferase and glucose levels were determined through the courtesy of Dr. Hanns-Dieter Gruemer and Dr. Gregory Miller on the SMAC (sequential multiple analyzer and computer system (17)) in the Division of Clinical Pathology at MCV. Separation
of mitochondrial
and cytoplasmic
aspartate
aminotransferases.
The isozymes of aspartate aminotransferase were separated electrophoretically with the Corning electrophoresis system (Corning Medical and Scientific, Palo Alto, Calif.) The separation medium was agarose (Universal electrophoresis film
DIABETES
CAUSES
PLASMA
PYRIDOXAL
PHOSPHATE
LOSS
93
number 470100 containing 1% agarose, 5% sucrose, and 0.035% EDTA in 0.065 barbital buffer, pH 8.6). Barbital was used for the electrode buffer (Universal PHAB buffer powder number 470180 containing 0.05 M barbital, 0.035% EDTA, 0.05% sodium chloride, and 0.001% sucrose-octaacetate). Electrophoresis was carried out at constant voltage (90 V) for approximately 40 min. The isozymes were visualized with a developing mixture (18) containing 1.O ml of L-aspartate (28.6 mg/ml), 1.0 ml of cr-ketoglutarate (15.7 mg/ml), 0.06 ml pyridoxal phosphate (10 mg/ml), and 1.5 mg fast violet B salt (19). L-Aspartic acid, a-ketoglutarate, and pyridoxal phosphate were dissolved separately in 0.1 M Tris-chloride buffer, pH 8.0, and the pH of the aspartate and a-ketoglutarate were readjusted to 8 with potassium hydroxide. Fast violet B salt, aspartic acid, and a-ketoglutarate were obtained from Sigma Chemical Company (St. Louis, MO.). Pyridoxal phosphate was obtained from Boehringer Chemical Company (Indianapolis, Ind.). Components of developer were mixed just before use and the developer was spread on the surface of the gels following electrophoresis. Gels were incubated in the dark at room temperature for varying times. Following development, gels were washed briefly with distilled water and dried at room temperature. The dried gels were used for photography and densitometric tracings. The isozymes were detected by the appearance of a pink band at the site of their catalytic activity. Srurisrical analyses. A two-tailed t test was used to ascertain levels of significance (20). Probability values of 0.05 or less were considered significant. M
RESULTS
Streptozotocin diabetic rats were glucosuric 3 days post-treatment. After 15 weeks, the diabetic rats were usually blind with cataracts in both eyes (cf. (21)), had grossly distended abdomens with very little musculature, grew rough hair coats that had a slightly pinkish color, and were approximately one-half the weight of controls (Table 1). Livers, as a function of body weight, were 1.8 times larger than in controls, but not different in terms of absolute weight. Plasma glucose was increased 3.9-fold. Diabetes lowered plasma pyridoxal phosphate concentrations to 14% of control (Table 1). Concomitant with this was a 16% reduction in total (mitochondrial plus cytoplasm) liver pyridoxal phosphate. Separation into mitochondrial and cytoplasmic fractions demonstrated that diabetes caused a 43% diminution in mitochondrial pyridoxal phosphate per gram of liver while the cytoplasmic fraction was not significantly enhanced (7%). The mitochondrial loss could not be accounted for in terms of the lower mitochondrial protein content because mitochondrial protein was decreased by only 22% in the diabetic. Thus there was a 25% decrease in the amount of pyridoxal phosphate per milligram of mitochondrial protein in the diabetic rat (diabetic mitochondria contained 0.27 and control mitochondria contained 0.36 nmole of pyridoxal phosphate per milligram of protein). This change in the ratio of pyridoxal phosphate to protein was not observed in the cytoplasm of the diabetic rat. We measured the activity of aspartate aminotransferase to see if the change in diabetic mitochondrial pyridoxal phosphate might be accompanied by a similar
94
ROGERS, HIGGINS,
AND KLINE
TABLE 1 Pyridoxal Phosphate and Aspartate Aminotransferase
in the Diabetic Rat
Control Number of rats Initial rat wt, g Final rat wt, g Liver wt, g Plasma glucose, mg/dl Plasma pyridoxal phosphate, pmole/ml Plasma aspartate aminotransferase, U/literd Mitochondrial pyridoxal phosphate, nmole/g liver Cytoplasmic pyridoxal phosphate, nmole/g liver Mitochondrial protein, mg/g liver Cytoplasmic protein, mg/g liver Mitochondrial aspartate aminotransferase, U/g liver“ Cytoplasmic aspartate aminotransferase, U/g liverd
I 77 448 15.22 142 480 23 6.75 7.81 18.6 91 31.6 37.2
Diabetic
k4 2 34 f 1.34 24 t 51 2 1
10 71 220 13.55 551 68 28
z-4 f 11* + .63 + 156 k 18* k 3
k f k k t f
3.85 8.33 14.5 93 14.8 125.0
+ + f f f 2
.58 .30 .6 1 .9 2.7
.50 .31 .8 2 1.56 5.P
’ Mean + standard error of the mean. b P < 0.001. c P < 0.005. d Unit of activity (U) = 1 pmole aspartate converted to oxaloacetate per minute.
change in a pyridoxal phosphate dependent protein. The activity of this mitochondrial enzyme was diminished 53% per gram of diabetic liver and 40% per milligram of diabetic mitochondrial protein when compared to control values. Incubation of the enzyme samples with excess pyridoxal phosphate for 1 hr at 37” did not provide more activity; therefore, the activity of the holoenzyme had been determined. In contrast to the mitochondrial results, cytoplasmic aspartate aminotransferase activity was elevated 3.4-fold in diabetic rats. It seemed a remote possibility that this elevation of activity of the cytoplasmic enzyme might be due to mitochondrial damage or leakage of enzyme from mitochondria. However, complete destruction of mitochondria (which certainly did not occur) would account for only 25% of the cytoplasmic activity observed in the diabetic rat. Furthermore, no glutamate dehydrogenase activity (normally found in mitochondria) was detectable in the cytoplasm. The isozymes of aspartate aminotransferase were separated by agarose gel electrophoresis (Fig. 1). In normal liver, the isozymic forms were present as evidenced by visualization of the cathodal band in mitochondria and both cathodal and anodal bands in cytoplasm. The liver mitochondria from diabetic rats also contained the cathodal band but diminished in activity, whereas cytoplasm from these animals contained elevated activities in both bands. The cytoplasmic cathodal isozyme/anodal isozyme ratio has not been quantitated because we are unable to do radioimmunoassays for the different enzyme forms. DISCUSSION
Chronic diabetes induced by streptozotocin caused a large depletion in plasma of pyridoxal phosphate and moderate liver mitochondrial loss of the coenzyme.
DIABETES
CAUSES PLASMA
1
2
3
PYRIDOXAL
4
5
6
PHOSPHATE
7
LOSS
95
6
FIG. 1. Electrophoretic comparisons of aspartate aminotransferase activities in mitochondrial and cytoplasmic compartments from livers of diabetic and control rats. Lanes 1, 3, and 5 correspond to 6.3, 12.6, and 50.4 pg of control cytoplasmic proteins. respectively, and lanes 2, 4, and 6 correspond to 6.4, 12.6, and 50.6 pg of diabetic cytoplasmic proteins, respectively, in the presence of complete aspartate aminotransferase developing mixture (see Materials and Methods). Lanes 7 and 8 correspond to 23.6 and 23.6 pg of control mitochondrial and diabetic mitochondrial proteins, respectively, in the presence of the enzyme developing mixture. The top represents the cathode, the bottom represents the anode, and zero represents the origin. Note that the diabetic cytoplasmic cathodal and anodal activities are greater than the corresponding controls.
These changes are indicative of pyridoxal phosphate deficiency and are in accord with the observation of other workers (1) that streptozotocin-treated rats secreted elevated levels of kynurenine metabolites in urine after a tryptophan load. We have also found that long term streptozotocin induced diabetes caused liver mitochondrial loss and liver cytoplasmic enhancement of aspartate aminotransferase activity. Perhaps the loss of pyridoxal phosphate from the mitochondria may have facilitated turnover of the mitochondrial aspartate aminotransferase since the susceptibility of pyridoxal phosphate dependent enzymes to proteolytic attack is greatly enhanced by removal of pyridoxal phosphate (22-24). Enhancement of aspartate aminotransferase activity in the diabetic cytoplasmic fraction may have been induced through either increased dietary protein intake or glucagon excess, both of which have been shown to induce pyridoxal phosphate dependent enzymes in vitamin B6 deficient rats (25). Our streptozotocin diabetic rats were hyperphagic and had high glucagon/insulin ratios (data not presented). We wish to speculate on a possible mechanism for the increased appearance of the mitochondrial form of aspartate aminotransferase in the diabetic cytoplasm. Most of the mitochondrial proteins are encoded for by nuclear DNA, translated on cytoplasmic ribosomes, and transported into the mitochondrial organelle by highly specific processes (26). A protein destined to be imported into mitochondria always seems to be synthesized as a precursor form which binds to a mitochondrial receptor prior to transport into its designated compartment (27). As with a number of other mitochondrial enzymes, the mitochondrial matrix isozyme of chicken aspartate aminotransferase is synthesized as a precursor of molecular weight higher than that of the native enzyme (28,29). The directed import of proteins into mitochondria conceivably may be defective in various pathological states. The markedly decreased mitochondrial aspartate aminotransferase activity seen in the diabetic animals may have resulted from such a defect rather than from one of the better understood processes which control enzyme activity, e.g., synthesis, degradation, or inhibition. Moreover, the accumulation of a cathodal isozyme in the diabetic cytoplasm may have also resulted from a defect in the
96
ROGERS, HIGGINS,
AND KLINE
mitochondrial recognition site for transport of the enzyme through the mitochondrial membrane. Modification of the recognition site through extra glycosylation as occurs with blood proteins in diabetes (30) or through extensive lipid modification of the mitochondrial membrane as occurs through diet manipulation and metabolism (31) may have altered or destroyed the recognition site for the import of mitochondrial aspartate aminotransferase. The diminution of pyridoxal phosphate in diabetic serum may have resulted from interference in the phosphorylation process for pyridoxal. Bessman (32) has pointed out that various metabolic processes which require ATP may be reduced by an impairment of mitochondrial ATP formation in hypoinsulinism. Alternatively, the altered carbohydrate and amino acid metabolism in the diabetic may have increased the utilization and thereby catabolism of pyridoxal phosphate. Leklem (33) reported that an acute glucose load administered to normal individuals significantly decreased plasma pyridoxal phosphate and elevated urinary Cpyridoxic acid. SUMMARY
The streptozotocin diabetic rat was selected as a model to study how insulin deficiency alters vitamin B6 utilization by focusing on pyridoxal phosphate levels and aspartate aminotransferase activities in liver tissues. Diabetes of 15 weeks’ duration lowered plasma pyridoxal phosphate levels by 84%. Normal plasma pyridoxal phosphate was 480 pmole/ml. Fractionation of liver into mitochondrial and extramitochondrial compartments demonstrated that diabetes caused a 43% diminution in mitochondrial pyridoxal phosphate per gram of liver. There was no cytoplasmic change in these diabetic rats. Mitochondrial aspartate aminotransferase activity was decreased 53% per gram of diabetic liver and cytoplasmic aspartate aminotransferase activity was elevated 3.4-fold. Damage to diabetic mitochondria during preparation procedures could not account for the rise in cytoplasmic aspartate aminotransferase activity. Electrophoresis showed that in the diabetic cytoplasm both cathodal and anodal forms of the enzyme were elevated. Speculations concerning mitochondrial loss and cytoplasmic gain of enzyme activity as well as those on the reduction of plasma pyridoxal phosphate in the diabetic rat are presented. REFERENCES 1. Akarte, N., and Shastri, N., J. Nutr. Sci. Vit. 22, 175 (1976). 2. Walsh, M. P., Howorth, P. J. N., and Marks, V., Amer. J. Clin. Nutr. 19, 379 (1966). 3. Sauberlich, H. E., Skala, J. H., and Dowdy, R. P., “Laboratory Tests for the Assessment of Nutritional Status, Vitamin B6,” pp. 37-49. CRC Press, Cleveland, 1974. 4. Wolf, H., &and. J. Clin. Invest. Suppl. 136, (1974). 5. Bergmeyer, H. II., and Bemt, E., “Methods of Enzymatic Analysis,” Vol. 2, pp. 727-733. Academic Press, New York, 1974. 6. Kuwahara, M. D., Lyons, S. A., Rosenblit, P. D., and Metzger, R. P., Proc. Sot. Exp. Biol. Med. 153, 305 (1976). 7. Davis, R. E., Calder, J. S., and Cumow, D. H., Pathology 8, 151 (1976). 8. Lui, A., Lumeng, L., and Li, T. K., J. Biol. Chem. 256, 6041 (1981). 9. Lumeng, L., Lui, A., and Li, T. K., J. Clin. Invest. 66, 681 (1980). 10. Fleisher, G. A., Potter, C. S., and Wakim, K. G., Proc. Sot. Exp. Biol. Med. 103, 229 (1960).
DIABETES Il. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.
CAUSES PLASMA
PYRIDOXAL
PHOSPHATE
LOSS
97
Michuda, C. M., and Martinez-Canion, M., J. Biol. Chem. 245, 262 (1970). Junod, A., Lambert, A. E., Stauffacher, W., and Renold, A. E., .I. Clin. Invest. 48, 2129 (1969). Rogers, K. S., and Higgins, E. S., J. Biol. Chem. 248, 7142 (1973). Gomall, A. E., Bardawill, C. J., and David, M., J. Biol. Chem. 177, 751 (1949). Schirch, L., and Peterson, D., J. Biol. Chem. 255, 7801 (1980). Schirch, L. G., and Mason, M., J. Biol. Chem. 238, 1032 (1963). “Advanced Automated Analyses,” pp. l-34. Technicon Instruments, Tarrytown, N. Y., 1972. Harris, H., and Hopkinson, D. A., “Handbook of Enzyme Electrophoresis in Human Genetics,” pp. l-3. North-Holland, New York, 1976. to Isozyme Techniques,” p. 132. Academic Press, New York, Brewer, G. J., “Introduction 1970. Edwards, A. L., “Statistical Analysis,” p. 127. Rinehart, New York, 1958. Dulin, W. E., Gerritsen, G. C., and Chang, A. Y., “Diabetes Mellitus, Theory and Practice (M. Ellenberg and H. R&in, Eds.), 3rd ed., pp. 361-395. Med. Exam. Publ. Co., New York, 1983. Bond, J. S., Biochem. Biophys. Res. Commun. 43, 333 (1971). Matsuzawa, T., “Intracellular Protein Turnover” (R. T. Schimke and N. Katunuma, Eds.), pp. 205-206. Academic Press, New York, 1975. Katunuma, N., Kominami, E., Kobayashi, K., Hamaguchi, Y., Banno, Y., Chichibu, K., Katsunuma, T., and Shiotani, T., “Intracellular Protein Turnover” (R. T. Schimke and N. Katunuma, Eds.), pp. 187-206. Academic Press, New York, 1975. Hunter, J. E., and Harper, A. E., J. Nutr. 107, 235 (1977). Neupert, W., and Schatz, G., Trends Biochem. Sci. 5, 1 (1981). Schatz, G., and Butow, R. A., Cell 32, 216 (1983). Sonderegger, P., Taussi, R., and Christen, P., Biochem. Biophys. Res. Commun. 94, 1256 (1980). Sonderegger, P., Taussi, R., Christen, P., and Gehring, H., J. Biol. Chem. 257, 3339 (1982). Bunn, H. F., Gabbay, K. H., and Gallop, P. M., Science 200, 21 (1978). Robblee, N. M., and Clandinin, M. T., J. Nurr. 114, 263 (1984). Bessman, S. P., Zsr. J. Med. Sci. 8, 344 (1972). Leklem, J. E., Fed. Proc. 43, No. 4104, 987 (1984).
MEDICINE
AND
METABOLIC
BIOLOGY
36, 91-97
(1986)
Experimental Diabetes Causes Mitochondrial Loss and Cytoplasmic Enrichment of Pyridoxal Phosphate and Aspartate Aminotransferase Activity KENNETH Department
S. ROGERS, EDWIN S. HIGGINS, AND EDWARD S. KLINE
of Biochemistry,
Medical College Richmond, Received
of Virginia, Virginia Virginia 23298
Commonwealth
University,
June 3, 1985
Streptozotocin diabetic rats excrete abnormal quantities of kynurenine metabolites after a tryptophan load (1) and this indicated to us that they may be vitamin B6 deficient. Lack of vitamin B, in normal rats causes 3-hydroxykynurenine to accumulate and be excreted (2,3) because pyridoxal phosphate is then unavailable as the coenzyme for kynureninase activity (4). Moreover, excretion of aspartate aminotransferase (EC 2.6.1. l), which contains pyridoxal phosphate (5), is accelerated in streptozotocin diabetes (6). Additional evidence for an altered vitamin Bs metabolism in the diabetic comes from the work of Davis et al. (7) who reported in a study of 518 diabetic patients of varying age and sex that 25% of these patients had decreased serum pyridoxal levels when compared to 371 healthy volunteers. We have selected the streptozotocin diabetic rat as a model to study how insulin deficiency alters vitamin B6 utilization by focusing on pyridoxal phosphate levels and aspartate aminotransferase activities in liver tissues. Lui and coworkers (8,9) reported that 20% of normal rat liver vitamin B6 is transported into and stored by the mitochondria which do not contain any enzymes for the vitamin’s synthesis or degradation. Liver also contains two isozymes of aspartate aminotransferase which are electrophoretically distinct; the cytoplasmic enzyme migrates anodally and the mitochondrial enzyme migrates cathodally (IO, 1 I). We report that streptozotocin diabetic rats show plasma deficiencies of pyridoxal phosphate accompanied by less storage in the liver of mitochondrial pyridoxal phosphate and aspartate aminotransferase. This is the first time to our knowledge that pyridoxal phosphate levels have been determined in the diabetic rat. MATERIALS AND METHODS Sprague-Dawley male rats were obtained from Flow Laboratories (Dublin, Va.). They were individually housed, placed on a light regime of 12 hr on/12 hr off, and given water and Purina rat chow pellets ad libitum. They were made diabetic after a 24-hr fast by injecting into the tail vein streptozotocin, 70 mg/kg 91 088545OY86
$3.00
Copyright Q 1986 by Academic Press, Inc. All rights of reproduction in any form reserved.
92
ROGERS,
HIGGINS,
AND
KLINE
body wt (12). After 15 weeks, the rats were decapitated, blood was collected, and the liver was excised immediately, weighed, and homogenized in ice-cold 0.25 M sucrose-O.5 mM disodium ethylenediaminetetraacetate (EDTA) solution, pH 7.4. Mitochondria were obtained from the liver as previously described (13) with the exception that the nuclear pellet was not extracted for residual mitochondria. The supernatant from the first mitochondrial collection was considered to represent liver cytoplasm and was saved for analyses. Mitochondrial and cytoplasmic protein was estimated in the presence of 1% sodium deoxycholate by a biuret method (14) with crystalline bovine sernm albumin as a standard. Pyridoxal phosphate analysis. Pyridoxal phosphate was extracted from plasma, liver mitochondria, and cytoplasm following the procedures of Lumeng and coworkers (8,9) which used trichloroacetic acid (TCA) and heating to 50” to precipitate protein and release the coenzyme, centrifugation, and extraction of the supernates with peroxide-free diethyl ether to provide a TCA- and protein-free sample of pH 7 for analysis. Pyridoxal phosphate was.assayed in samples incubated at 30” for 1 hr with mitochondrial aposerine hydroxymethyltransferase (EC 2.12.1) prepared (15,16) for us through the courtesy of Dr. LaVerne Schirch, Department of Biochemistry, MCV. The incubation medium contained 1.8 ml of 20 mM potassium phosphate, 2 mM EDTA, pH 7.0; 20 ~1 of water or 6 PM pyridoxal phosphate; 0.5 ml of water; and 10 ~1 of aposerine hydroxymethyltransferase (4.2 mg protein/ml) in phosphate-EDTA buffer. After incubation, 40 ~1 of NADH-alcohol dehydrogenase mixture (12 mg NADH, 8 mg crystalline yeast alcohol dehydrogenase in 2 ml of phosphate-EDTA buffer) was added. The enzyme reaction was started by addition of 0.1 ml DL-allothreonine (296 mg DL-allothreonine in 5 ml of water). This coupled enzyme assay (15) pemitted rapid quantitation of serine hydroxymethyltransferase activity and thereby pyridoxal phosphate content since activity of the enzyme was proportional to the amount of pyridoxal phosphate in the tissue samples. A typical standard curve of enzyme velocity (change in absorbance at 340 nm per minute) was linear over the concentration range of pyridoxal phosphate measured (0 to 500 pmole). Aspartute aminotransferase. Aspartate aminotransferase was assayed in liver cytoplasm and mitochondria following a slightly modified procedure of Bergmeyer and Bernt (5). The coupled enzyme reaction mixtures were adjusted to 0.1% Triton X-100 concentration in order to solubilize particulate matter. This concentration of detergent did not inhibit the involved enzymes. Endogenous absorbance changes due to liver glutamate dehydrogenase activity were corrected by subtracting appropriate controls containing no aspartate in the reaction mixtures. Plasma aspartate aminotransferase and glucose levels were determined through the courtesy of Dr. Hanns-Dieter Gruemer and Dr. Gregory Miller on the SMAC (sequential multiple analyzer and computer system (17)) in the Division of Clinical Pathology at MCV. Separation
of mitochondrial
and cytoplasmic
aspartate
aminotransferases.
The isozymes of aspartate aminotransferase were separated electrophoretically with the Corning electrophoresis system (Corning Medical and Scientific, Palo Alto, Calif.) The separation medium was agarose (Universal electrophoresis film
DIABETES
CAUSES
PLASMA
PYRIDOXAL
PHOSPHATE
LOSS
93
number 470100 containing 1% agarose, 5% sucrose, and 0.035% EDTA in 0.065 barbital buffer, pH 8.6). Barbital was used for the electrode buffer (Universal PHAB buffer powder number 470180 containing 0.05 M barbital, 0.035% EDTA, 0.05% sodium chloride, and 0.001% sucrose-octaacetate). Electrophoresis was carried out at constant voltage (90 V) for approximately 40 min. The isozymes were visualized with a developing mixture (18) containing 1.O ml of L-aspartate (28.6 mg/ml), 1.0 ml of cr-ketoglutarate (15.7 mg/ml), 0.06 ml pyridoxal phosphate (10 mg/ml), and 1.5 mg fast violet B salt (19). L-Aspartic acid, a-ketoglutarate, and pyridoxal phosphate were dissolved separately in 0.1 M Tris-chloride buffer, pH 8.0, and the pH of the aspartate and a-ketoglutarate were readjusted to 8 with potassium hydroxide. Fast violet B salt, aspartic acid, and a-ketoglutarate were obtained from Sigma Chemical Company (St. Louis, MO.). Pyridoxal phosphate was obtained from Boehringer Chemical Company (Indianapolis, Ind.). Components of developer were mixed just before use and the developer was spread on the surface of the gels following electrophoresis. Gels were incubated in the dark at room temperature for varying times. Following development, gels were washed briefly with distilled water and dried at room temperature. The dried gels were used for photography and densitometric tracings. The isozymes were detected by the appearance of a pink band at the site of their catalytic activity. Srurisrical analyses. A two-tailed t test was used to ascertain levels of significance (20). Probability values of 0.05 or less were considered significant. M
RESULTS
Streptozotocin diabetic rats were glucosuric 3 days post-treatment. After 15 weeks, the diabetic rats were usually blind with cataracts in both eyes (cf. (21)), had grossly distended abdomens with very little musculature, grew rough hair coats that had a slightly pinkish color, and were approximately one-half the weight of controls (Table 1). Livers, as a function of body weight, were 1.8 times larger than in controls, but not different in terms of absolute weight. Plasma glucose was increased 3.9-fold. Diabetes lowered plasma pyridoxal phosphate concentrations to 14% of control (Table 1). Concomitant with this was a 16% reduction in total (mitochondrial plus cytoplasm) liver pyridoxal phosphate. Separation into mitochondrial and cytoplasmic fractions demonstrated that diabetes caused a 43% diminution in mitochondrial pyridoxal phosphate per gram of liver while the cytoplasmic fraction was not significantly enhanced (7%). The mitochondrial loss could not be accounted for in terms of the lower mitochondrial protein content because mitochondrial protein was decreased by only 22% in the diabetic. Thus there was a 25% decrease in the amount of pyridoxal phosphate per milligram of mitochondrial protein in the diabetic rat (diabetic mitochondria contained 0.27 and control mitochondria contained 0.36 nmole of pyridoxal phosphate per milligram of protein). This change in the ratio of pyridoxal phosphate to protein was not observed in the cytoplasm of the diabetic rat. We measured the activity of aspartate aminotransferase to see if the change in diabetic mitochondrial pyridoxal phosphate might be accompanied by a similar
94
ROGERS, HIGGINS,
AND KLINE
TABLE 1 Pyridoxal Phosphate and Aspartate Aminotransferase
in the Diabetic Rat
Control Number of rats Initial rat wt, g Final rat wt, g Liver wt, g Plasma glucose, mg/dl Plasma pyridoxal phosphate, pmole/ml Plasma aspartate aminotransferase, U/literd Mitochondrial pyridoxal phosphate, nmole/g liver Cytoplasmic pyridoxal phosphate, nmole/g liver Mitochondrial protein, mg/g liver Cytoplasmic protein, mg/g liver Mitochondrial aspartate aminotransferase, U/g liver“ Cytoplasmic aspartate aminotransferase, U/g liverd
I 77 448 15.22 142 480 23 6.75 7.81 18.6 91 31.6 37.2
Diabetic
k4 2 34 f 1.34 24 t 51 2 1
10 71 220 13.55 551 68 28
z-4 f 11* + .63 + 156 k 18* k 3
k f k k t f
3.85 8.33 14.5 93 14.8 125.0
+ + f f f 2
.58 .30 .6 1 .9 2.7
.50 .31 .8 2 1.56 5.P
’ Mean + standard error of the mean. b P < 0.001. c P < 0.005. d Unit of activity (U) = 1 pmole aspartate converted to oxaloacetate per minute.
change in a pyridoxal phosphate dependent protein. The activity of this mitochondrial enzyme was diminished 53% per gram of diabetic liver and 40% per milligram of diabetic mitochondrial protein when compared to control values. Incubation of the enzyme samples with excess pyridoxal phosphate for 1 hr at 37” did not provide more activity; therefore, the activity of the holoenzyme had been determined. In contrast to the mitochondrial results, cytoplasmic aspartate aminotransferase activity was elevated 3.4-fold in diabetic rats. It seemed a remote possibility that this elevation of activity of the cytoplasmic enzyme might be due to mitochondrial damage or leakage of enzyme from mitochondria. However, complete destruction of mitochondria (which certainly did not occur) would account for only 25% of the cytoplasmic activity observed in the diabetic rat. Furthermore, no glutamate dehydrogenase activity (normally found in mitochondria) was detectable in the cytoplasm. The isozymes of aspartate aminotransferase were separated by agarose gel electrophoresis (Fig. 1). In normal liver, the isozymic forms were present as evidenced by visualization of the cathodal band in mitochondria and both cathodal and anodal bands in cytoplasm. The liver mitochondria from diabetic rats also contained the cathodal band but diminished in activity, whereas cytoplasm from these animals contained elevated activities in both bands. The cytoplasmic cathodal isozyme/anodal isozyme ratio has not been quantitated because we are unable to do radioimmunoassays for the different enzyme forms. DISCUSSION
Chronic diabetes induced by streptozotocin caused a large depletion in plasma of pyridoxal phosphate and moderate liver mitochondrial loss of the coenzyme.
DIABETES
CAUSES PLASMA
1
2
3
PYRIDOXAL
4
5
6
PHOSPHATE
7
LOSS
95
6
FIG. 1. Electrophoretic comparisons of aspartate aminotransferase activities in mitochondrial and cytoplasmic compartments from livers of diabetic and control rats. Lanes 1, 3, and 5 correspond to 6.3, 12.6, and 50.4 pg of control cytoplasmic proteins. respectively, and lanes 2, 4, and 6 correspond to 6.4, 12.6, and 50.6 pg of diabetic cytoplasmic proteins, respectively, in the presence of complete aspartate aminotransferase developing mixture (see Materials and Methods). Lanes 7 and 8 correspond to 23.6 and 23.6 pg of control mitochondrial and diabetic mitochondrial proteins, respectively, in the presence of the enzyme developing mixture. The top represents the cathode, the bottom represents the anode, and zero represents the origin. Note that the diabetic cytoplasmic cathodal and anodal activities are greater than the corresponding controls.
These changes are indicative of pyridoxal phosphate deficiency and are in accord with the observation of other workers (1) that streptozotocin-treated rats secreted elevated levels of kynurenine metabolites in urine after a tryptophan load. We have also found that long term streptozotocin induced diabetes caused liver mitochondrial loss and liver cytoplasmic enhancement of aspartate aminotransferase activity. Perhaps the loss of pyridoxal phosphate from the mitochondria may have facilitated turnover of the mitochondrial aspartate aminotransferase since the susceptibility of pyridoxal phosphate dependent enzymes to proteolytic attack is greatly enhanced by removal of pyridoxal phosphate (22-24). Enhancement of aspartate aminotransferase activity in the diabetic cytoplasmic fraction may have been induced through either increased dietary protein intake or glucagon excess, both of which have been shown to induce pyridoxal phosphate dependent enzymes in vitamin B6 deficient rats (25). Our streptozotocin diabetic rats were hyperphagic and had high glucagon/insulin ratios (data not presented). We wish to speculate on a possible mechanism for the increased appearance of the mitochondrial form of aspartate aminotransferase in the diabetic cytoplasm. Most of the mitochondrial proteins are encoded for by nuclear DNA, translated on cytoplasmic ribosomes, and transported into the mitochondrial organelle by highly specific processes (26). A protein destined to be imported into mitochondria always seems to be synthesized as a precursor form which binds to a mitochondrial receptor prior to transport into its designated compartment (27). As with a number of other mitochondrial enzymes, the mitochondrial matrix isozyme of chicken aspartate aminotransferase is synthesized as a precursor of molecular weight higher than that of the native enzyme (28,29). The directed import of proteins into mitochondria conceivably may be defective in various pathological states. The markedly decreased mitochondrial aspartate aminotransferase activity seen in the diabetic animals may have resulted from such a defect rather than from one of the better understood processes which control enzyme activity, e.g., synthesis, degradation, or inhibition. Moreover, the accumulation of a cathodal isozyme in the diabetic cytoplasm may have also resulted from a defect in the
96
ROGERS, HIGGINS,
AND KLINE
mitochondrial recognition site for transport of the enzyme through the mitochondrial membrane. Modification of the recognition site through extra glycosylation as occurs with blood proteins in diabetes (30) or through extensive lipid modification of the mitochondrial membrane as occurs through diet manipulation and metabolism (31) may have altered or destroyed the recognition site for the import of mitochondrial aspartate aminotransferase. The diminution of pyridoxal phosphate in diabetic serum may have resulted from interference in the phosphorylation process for pyridoxal. Bessman (32) has pointed out that various metabolic processes which require ATP may be reduced by an impairment of mitochondrial ATP formation in hypoinsulinism. Alternatively, the altered carbohydrate and amino acid metabolism in the diabetic may have increased the utilization and thereby catabolism of pyridoxal phosphate. Leklem (33) reported that an acute glucose load administered to normal individuals significantly decreased plasma pyridoxal phosphate and elevated urinary Cpyridoxic acid. SUMMARY
The streptozotocin diabetic rat was selected as a model to study how insulin deficiency alters vitamin B6 utilization by focusing on pyridoxal phosphate levels and aspartate aminotransferase activities in liver tissues. Diabetes of 15 weeks’ duration lowered plasma pyridoxal phosphate levels by 84%. Normal plasma pyridoxal phosphate was 480 pmole/ml. Fractionation of liver into mitochondrial and extramitochondrial compartments demonstrated that diabetes caused a 43% diminution in mitochondrial pyridoxal phosphate per gram of liver. There was no cytoplasmic change in these diabetic rats. Mitochondrial aspartate aminotransferase activity was decreased 53% per gram of diabetic liver and cytoplasmic aspartate aminotransferase activity was elevated 3.4-fold. Damage to diabetic mitochondria during preparation procedures could not account for the rise in cytoplasmic aspartate aminotransferase activity. Electrophoresis showed that in the diabetic cytoplasm both cathodal and anodal forms of the enzyme were elevated. Speculations concerning mitochondrial loss and cytoplasmic gain of enzyme activity as well as those on the reduction of plasma pyridoxal phosphate in the diabetic rat are presented. REFERENCES 1. Akarte, N., and Shastri, N., J. Nutr. Sci. Vit. 22, 175 (1976). 2. Walsh, M. P., Howorth, P. J. N., and Marks, V., Amer. J. Clin. Nutr. 19, 379 (1966). 3. Sauberlich, H. E., Skala, J. H., and Dowdy, R. P., “Laboratory Tests for the Assessment of Nutritional Status, Vitamin B6,” pp. 37-49. CRC Press, Cleveland, 1974. 4. Wolf, H., &and. J. Clin. Invest. Suppl. 136, (1974). 5. Bergmeyer, H. II., and Bemt, E., “Methods of Enzymatic Analysis,” Vol. 2, pp. 727-733. Academic Press, New York, 1974. 6. Kuwahara, M. D., Lyons, S. A., Rosenblit, P. D., and Metzger, R. P., Proc. Sot. Exp. Biol. Med. 153, 305 (1976). 7. Davis, R. E., Calder, J. S., and Cumow, D. H., Pathology 8, 151 (1976). 8. Lui, A., Lumeng, L., and Li, T. K., J. Biol. Chem. 256, 6041 (1981). 9. Lumeng, L., Lui, A., and Li, T. K., J. Clin. Invest. 66, 681 (1980). 10. Fleisher, G. A., Potter, C. S., and Wakim, K. G., Proc. Sot. Exp. Biol. Med. 103, 229 (1960).
DIABETES Il. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.
CAUSES PLASMA
PYRIDOXAL
PHOSPHATE
LOSS
97
Michuda, C. M., and Martinez-Canion, M., J. Biol. Chem. 245, 262 (1970). Junod, A., Lambert, A. E., Stauffacher, W., and Renold, A. E., .I. Clin. Invest. 48, 2129 (1969). Rogers, K. S., and Higgins, E. S., J. Biol. Chem. 248, 7142 (1973). Gomall, A. E., Bardawill, C. J., and David, M., J. Biol. Chem. 177, 751 (1949). Schirch, L., and Peterson, D., J. Biol. Chem. 255, 7801 (1980). Schirch, L. G., and Mason, M., J. Biol. Chem. 238, 1032 (1963). “Advanced Automated Analyses,” pp. l-34. Technicon Instruments, Tarrytown, N. Y., 1972. Harris, H., and Hopkinson, D. A., “Handbook of Enzyme Electrophoresis in Human Genetics,” pp. l-3. North-Holland, New York, 1976. to Isozyme Techniques,” p. 132. Academic Press, New York, Brewer, G. J., “Introduction 1970. Edwards, A. L., “Statistical Analysis,” p. 127. Rinehart, New York, 1958. Dulin, W. E., Gerritsen, G. C., and Chang, A. Y., “Diabetes Mellitus, Theory and Practice (M. Ellenberg and H. R&in, Eds.), 3rd ed., pp. 361-395. Med. Exam. Publ. Co., New York, 1983. Bond, J. S., Biochem. Biophys. Res. Commun. 43, 333 (1971). Matsuzawa, T., “Intracellular Protein Turnover” (R. T. Schimke and N. Katunuma, Eds.), pp. 205-206. Academic Press, New York, 1975. Katunuma, N., Kominami, E., Kobayashi, K., Hamaguchi, Y., Banno, Y., Chichibu, K., Katsunuma, T., and Shiotani, T., “Intracellular Protein Turnover” (R. T. Schimke and N. Katunuma, Eds.), pp. 187-206. Academic Press, New York, 1975. Hunter, J. E., and Harper, A. E., J. Nutr. 107, 235 (1977). Neupert, W., and Schatz, G., Trends Biochem. Sci. 5, 1 (1981). Schatz, G., and Butow, R. A., Cell 32, 216 (1983). Sonderegger, P., Taussi, R., and Christen, P., Biochem. Biophys. Res. Commun. 94, 1256 (1980). Sonderegger, P., Taussi, R., Christen, P., and Gehring, H., J. Biol. Chem. 257, 3339 (1982). Bunn, H. F., Gabbay, K. H., and Gallop, P. M., Science 200, 21 (1978). Robblee, N. M., and Clandinin, M. T., J. Nurr. 114, 263 (1984). Bessman, S. P., Zsr. J. Med. Sci. 8, 344 (1972). Leklem, J. E., Fed. Proc. 43, No. 4104, 987 (1984).