- Email: [email protected]
db) mice
BIOCHEMICAL
MEDKINE
AND
METABOLIC
BIOLOGY
37, 42-50
(1987)
Effects of Oxytetracycline Treatment on Enzymes of Hepatic Glycogen Metabolism in Genetically Diabetic (db/db) Mice CAMILLOA. BENZO Department
of Anatomy
and Cell Biology, State University of New York, Syracuse, Syracuse, New York 13210 Received
May
Health
Science
Center
at
28, 1985
The genetically diabetic mouse C57BL/KsJ db (db/db) displays many metabolic disturbances that resemble those found in maturity-onset diabetes in humans. These derangements include obesity, hyperglycemia, extremely elevated levels of circulating insulin and glucagon, and resistance to the hypoglycemic effects of exogenous insulin (1,2). Several investigators have demonstrated that the administration of the antibiotic oxytetracycline (OTC) to animals either with experimental diabetes or displaying genetic diabetes-obesity syndromes resulted in the amelioration of many of their diabetic symptoms. It was found that treatment of these animals with OTC markedly reduced their hyperglycemia and hyperinsulinemia, and virtually normalized their food intake, glucose tolerance, and insulin resistance (3-9). The results of these studies suggest that OTC appears to act by increasing target cell sensitivity to the catalytic actions of insulin, perhaps with a direct effect of the drug on hepatic carbohydrate metabolism (5 7). Because our own and other recent studies (10,ll) have shown that severe structural and chemical alterations associated with hepatic glycogen metabolism in db/db mice not only accompany the diabetic syndrome in these animals, but may be contributory, we decided to investigate whether chronic treatment of db/db mice with OTC would result in an improvement of their diabetic state, particularly with regard to liver glycogen metabolism. The present report documents the effects of 7 days of OTC administration on the activities of the rate-limiting enzymes in hepatic glycogen metabolism, glycogen synthase and phosphorylase, and on the plasma levels of glucose and insulin in genetically diabetic mice. In addition, because we have observed that the alterations in liver glycogen metabolism which occur in diabetic-obese mice are accompanied by changes in hepatic lipid metabolism (12,13), the effects of the drug on the concentrations of liver glycogen, free fatty acid, and triglyceride were monitored and are reported here as well.
0885-4505187 Copyright Au rights
$3.00
0 1987 by Academic Press. Inc. of reproduction in any form reserved.
42
EFFECTS OFOTCONGLYCOGEN
MATERIALS
ENZYMES
43
AND METHODS
Animals Genetically diabetic mice (C57BL/KsJdb) and their lean littermate controls were purchased from the Jackson Laboratory, Bar Harbor, Maine. Diabetic mice used in this study were 8, 12, and 15 weeks of age, three representative intervals in the advanced stage of the diabetic syndrome in these animals. The animals were housed in plastic cages with free access to food (Purina Mouse Chow) and water. Mice were maintained on a cycle of 12 hr of light (0600 to 1800 hr) and 12 hr of dark (1800 to 0600 hr). In each experimental group, db/db mice and their lean littermates received daily intraperitoneal injections of either oxytetracycline (Terramycin, Pfizer Laboratories; 100 mg/kg diluted in olive oil to give a final injection volume of 0.2 ml) or drug diluent over a 7-day period. The injections were given between 0900 and 1100 hr. Twenty-four hours after the final injection, the mice were killed by cervical dislocation between 0900 and 1100 hr, thereby minimizing differences due to diurnal variations in liver glycogen and lipid levels. The livers were frozen in situ with liquid nitrogen, then removed rapidly with liquid nitrogen-cooled aluminum clamps and stored at -80°C until they were assayed for glycogen synthase and phosphorylase activities, and for hepatic concentrations of glycogen, triglycerides and free fatty acids. Blood for immunoreactive insulin (IRI) and glucose determinations was collected in chilled, heparinized capillary tubes. Plasma was obtained immediately by centrifugation and stored for not more than 2 weeks at -80°C. Assay Methods For determination of enzymatic activities, tissue extracts were prepared by Polytron homogenization of 100 mg of frozen liver in 1 ml 50 mM KF, 10 mM EDTA, 0.5 M sucrose, and 62.5 mM glycylglycine, pH 7.4, at one-half speed for 10 set at 2°C. the homogenates were centrifuged at 2°C for 10 min in a Sorvall SS34 rotor at 12,OOOg, and the supernatants were used for both glycogen synthase and phosphorylase assays. Glycogen synthase (EC 2.4.1.11) activity was determined on aliquots of the supernatants according to the filter paper method of Thomas et al. (14) in the presence (total activity; I + D forms) or absence (I-form activity) of 10 mM glucose 6-phosphate. The assay is based upon the transfer of glucose to glycogen, using UDPG-[‘4C]glucose as substrate, followed by the isolation and counting of the radioactive glycogen. Glycogen phosphorylase (EC 2.4.1.1) activity was determined on aliquots of the supernatants which were diluted 1:3 with 50 mu KF, 60 mu 2-mercaptoethanol, and 50 mM ME& pH 6.1, immediately prior to the assay, following the filter paper method described by Gilboe et al. (15). Total phosphorylase (a + b) activity was assayed in the presence of 3 mM 5’-AMP, and phosphorylase a-form activity was determined in the absence of this nucleotide, but in the presence of 0.5 mM caffeine. This assay is based also upon the incorporation of glucose into glycogen, but using [‘4C]glucose l-phosphate as substrate. Both enzyme assays were linear with respect to time and to the amount of tissue extract added. The results of each assay are presented as units of enzyme
44
CAMILLOA.BENZO
activity per gram of tissue (wet wt), and as the percentage of the active form (-G-6-P/ + G-6-P x 100 or -AMP/ + AMP x 100). One unit of enzyme activity is the amount causing the transfer of 1 ,umole of glucose to glycogen in 1 min at 30°C. For determination of triglycerides and free fatty acids, portions of frozen liver were weighed rapidly and homogenized immediately in chloroform:methanol(2: 1) (16). On the following day, aliquots of the chloroform layer were evaporated to dryness under a stream of nitrogen. The residue was dissolved in isopropyl alcohol, and the triglycerides were extracted and assayed according to Tri-chol procedures (Oxford Reagent Set, Oxford Laboratories, Foster City, Calif.). Liver triglyceride levels are expressed as milligrams per gram of tissue (wet wt). The remaining portion of the chloroform layer was analyzed for free fatty acids by the method of Falholt er al. (17). Free fatty acid levels are expressed as micromoles per gram of tissue (wet wt). Hepatic glycogen was determined using the anthrone reagent according to the method of Templeton (18) and it is expressed as milligrams per gram of tissue (wet wt). Plasma IRI levels were determined by the procedure of Hales and Randle (19) utilizing a kit prepared by the Amersham Corporation (Arlington Heights, Ill.). Hormone samples were assayed in duplicate, using 50 ~1 of plasma for each determination. Plasma glucose was determined by a glucose oxidase method using a Beckman glucose analyzer. Values obtained from biochemical determinations are presented as means ? SEM. Mean values were compared by Student’s I test. Differences with P values < 0.05 were considered significant. RESULTS
The effects of OTC treatment on body weight, plasma glucose, and plasma IRI in lean and db/db mice are presented in Table 1. All lean mice showed significant weight loss following OTC administration, whereas db/db mice, regardless of age, displayed no substantial changes in body weight following drug treatment. In all age groups, db/db mice maintained much greater body weights than did their lean controls during the experiment (untreated db/db vs untreated lean, P < 0.001, Student’s t test). A significant effect of OTC in reducing plasma glucose levels was observed in all lean and db/db mice following drug treatment (Table 1). Similarly, except in severely hyperinsulinemic &week-old db/db mice, OTC administration markedly lowered plasma IRI levels in both lean and db/db mice, regardless of age. Throughout the experiment, all db/db mice maintained much higher plasma levels of glucose (P < 0.005) and IRI (P < 0.001) than did their appropriate lean controls, and the concentrations of both blood sugar and circulating insulin in all OTC-treated db/db mice remained significantly higher than those of either treated or untreated lean control mice. Table 2 summarizes the effects of OTC treatment on the glycogen content, free fatty acid, and triglyceride levels in liver from lean and db/db mice. During the study, no significant differences were found in the amount of glycogen present
EFFECTS
OF OTC ON GLYCOGEN
TABLE 1 Body Weight, Plasma Glucose, and Plasma IRI Levels in Oxytetracycline-Treated and Their Controls Age and group 8 Weeks Lean Lean-OTC dbldb db/db-OTC
12 Weeks Lean Lean-OTC db/db db/db-OTC
15 Weeks Lean Lean-OTC dbldb dbjdb-OTC
Body wt. k) 20 2 1 16 2 1
45
ENZYMES
Plasma glucose (mddl) 141 -t
10
Mice Plasma IRI W/ml) 26 2 I 10 + 1
(P < 0.025) 35 4 1
97-t 10 (P < 0.025) 859 ‘c 84
(P < 0.001) 236 2 5
37 2 1 VW
228 2 55 (P < 0.005)
232 2 8 (NW
22 2 1 18 ? 1 (P < 0.05) 52 ‘- 2 50 ” 1 W)
131 t 10 902 14 (P < 0.05) 617 t 124 303 t 69 (P < 0.05)
21 5 2 13 5 1 (P < 0.025) 156 + 3 88 4 2 (P < 0.001)
25 2 1 21 2 1 (P < 0.05) 52 f 2 50 2 2 WS)
166 t 17 99% 21 (P < 0.05) 793 t 73 316 t- 21 (P < 0.005)
20 i 1 10 2 1 (P < 0.001) 107 ? 4 86 ” 2 (P < 0.01)
Note. Values are presented as means +- SEM from at least three mice in each group. Probabilities were derived from Student’s t test of differences between means. NS, not significant at P < 0.05.
in the livers of db/db mice when compared to their appropriate lean controls. OTC treatment was ineffective in altering the hepatic glycogen content in both lean and db/db mice in all age groups. Similarly, except in drug-treated S-weekold lean animals, OTC had no effect on the free fatty acid levels in the livers of either lean or db/db mice. The livers of all db/db mice contained considerably more triglyceride (P < 0.05) than that found in their lean littermates, but no consistent effect of OTC administration in altering hepatic triglyceride levels was observed for either phenotype (Table 2). The data suggest a tendency for OTC to increase the hepatic triglyceride content in both lean and db/db mice. Drug treatment, however, produced significant increases in liver triglyceride content in g-week-old db/db mice and in 12- and U-week-old lean mice. Although a marginal effect (0.05 < P < 0.1) of the drug in raising the hepatic triglyceride level in 15-week-old db/db mice is suggested, OTC was ineffective in altering the liver triglyceride concentration in the remaining groups of lean or db/db mice. The effects of OTC administration on the activities of hepatic glycogen synthase and phosphorylase in lean and db/db mice are presented in Table 3. Regardless of age, all db/db mice contained substantially more total liver glycogen synthase activity (P < 0.025) than did their lean littermates. However, the levels of the
46
CAMILLO
A. BENZ0
TABLE 2 Glycogen, Free Fatty Acid, and Triglyceride Levels in Liver from Oxytetracycline-Treated and Their Controls Age and group 8 weeks Lean Lean-OTC db/db dbldb-OTC 12 Weeks Lean Lean-OTC db/db db/db-OTC 15 Weeks Lean Lean-OTC db/db db/db-OTC
Glycogen (mg/g tissue) 44.2 +43.3 +(NS) 49.7 -’ 45.1 f (W
2.6 3.0
53.6 f 53.0 2 (W 60.1 * 52.1 t (NS)
5.3 5.1
5.7 7.6
5.9 6.7
54.9 + 3.6 54.0 k 6.0 (NS) 47.4 k 10.0 42.2 e 6.4 (NS)
Mice
Free fatty acid (pmole/g tissue)
Triglyceride (mg/g tissue)
2.95 k 0.15 2.22 + 0.29 (P < 0.05) 2.04 2 0.18 2.05 + 0.13 (N.9
11.3 2 2.2 18.8 4 5.0 (W 30.4 k 7.9 48.5 k 2.7 (P < 0.05)
2.05 k 0.16 1.96 2 0.09 (W 2.31 t 0.23 2.06 k 0.29 PJS)
14.5 2 2.4 27.4 k 4.8 (P < 0.05) 52.3 5 0.3 55.7 2 4.4 tN.9
2.08 k 0.48 2.05 iz 0.07 (NW 2.25 f 0.27 2.48 f 0.18 tN.9
11.9 * 0.1 47.0 2 13.5 (P < 0.05) 53.2 e 13.8 82.6 k 7.2 W)
Note. Values are presented as means a SEM from at least three mice in each group. Probabilities were derived from Student’s t tests of differences between means. NS, not significant at P < 0.05.
physiologically active I form of the enzyme in livers from dbjdb mice were not different from those found in control livers. Unlike glycogen synthase, both total and active glycogen phosphorylase activities were consistently greater (P < 0.01 and P < 0.005, respectively) in the livers of all db/db mice when compared with their appropriate lean controls. OTC was ineffective in altering either total or I-form glycogen synthase activity when administered to 8- and IZweek-old mice, regardless of phenotype (Table 3). OTC administration to H-week-old lean mice resulted in a decrease in total glycogen synthase activity and a proportional increase in I-form activity. When 15week-old db/db mice were given OTC, no change in total liver synthase activity was observed, but these animals showed an almost threefold increase in I-form activity following drug treatment. Both total and active glycogen phosphorylase activities were lowered substantially in all drug-treated lean mice (Table 3). OTC was without effect on the activities of either form of phosphorylase when administered to 8- and 12-week-old db/db mice. Following OTC treatment, however, Sweek-old db/db mice showed significant decreases in the activities of both total and active phosphorylase.
EFFECTS
OF OTC ON GLYCOGEN
47
ENZYMES
TABLE 3 Glycogen Synthase and Phosphorylase Activities in Liver from Oxytetracycline-Treated Their Controls
Glycogen phosphorylase (units/g tissue)
Glycogen synthase (units/g tissue) Age and group 8 Weeks Lean Lean-OTC dbjdb dbldb-OTC 12 Weeks Lean Lean-OTC db/db dbldb-OTC I5 Weeks Lean Lean-OTC db/db dbldb-OTC
I+D
Mice and
I
%I
a +b
a
%a
0.069 2 0.003 0.083 + 0.007 (NV 0.133 2 0.23 0.135 2 0.005 (NS)
0.018 * 0.002 0.021 k 0.001 (NS) 0.026 2 0.006 0.019 2 0.007 WS)
26.1 25.3
3.52 ‘- 0.25 2.46 k 0.33 (P < 0.05) 6.45 k 0.33 5.62 2 0.52 (NS)
2.92 f 0.16 2.03 k 0.12 (P < 0.01) 5.28 k 0.19 4.72 k 0.52 (NS)
83.0 82.5
0.082 f 0.013 0.074 ” 0.010 (NS) 0.148 2 0.011 0.144 ” 0.009 VW
0.020 2 0.003 0.029 5 0.006 (NW 0.022 f 0.006 0.028 + 0.001 VW
24.4 39.2
3.81 * 0.37 2.08 2 0.21 (P < 0.01) 8.83 2 1.19 6.89 k 1.66 (NS)
2.73 k 0.16 1.39 + 0.15 (P < 0.005) 6.55 2 0.71 5.47 k 1.23 PW
71.7 66.8
0.067 2 0.001 0.056 ” 0.003 (P < 0.025) 0.119 ” 0.005 0.129 2 0.019 N-3
0.016 * 0.001 0.026 k 0.001 (P < 0.005) 0.018 k 0.003 0.060 f 0.016 (P < 0.05)
23.9 46.4
3.21 2 0.27 2.15 ” 0.19 (P < 0.025) 6.75 ” 0.22 5.59 2 0.42 (P < 0.05)
2.40 -+ 0.19 1.59 2 0.29 (P < 0.05) 6.57 * 0.23 3.74 2 0.15 (P < 0.001)
74.8 74.0
19.5 14.1
14.9 19.4
15.1 46.5
81.9 84.0
74.2 79.4
97.3 66.9
Note. Values are presented as means ? SEM from at least three mice in each group. Probabilities were derived from Student’s t tests of differences between means. NS, not significant at P < 0.05.
DISCUSSION
Previous reports on the effects of oxytetracycline in a related strain, the genetically obese (&lob) mouse, have shown that drug treatment (a) restored almost completely the binding of insulin to plasma membranes of the mutant liver cells, (b) reversed the insensitivity to insulin of the ob/ob mouse diaphragm both in viva and in vitro, (c) led to the regranulation of B cells in ob/ob pancreatic islet tissue, and (d) normalized the insulin secretory response in these animals (3,4,8,9). These findings prompted the suggestion that the effectiveness of oxytetracycline in the ob/ob mouse requires the prior development of insulin resistance and glucose intolerance (3,4). However, other studies, including the present one, have noted that substantially similar drug-induced metabolic changes also occur in lean mice following chronic oxytetracycline administration (5,6). Thus, while certain effects of oxytetracycline in either ob/ob or db/db mice might be mediated through a reversal of the insulin sensitivity of specific tissues, such consequences of drug treatment would not be expected to occur in lean mice. The present data tend to support the suggestion that oxytetracycline may have at least two discrete effects on the in vivo metabolism of obese-hyperglycemic
48
CAMILLOA.BENZO
animals (5,6). One effect may well be on the pancreatic B cells, improving abnormal insulin secretion and insulin resistance (8,9). Another effect, common to both phenotypes might alter some aspect of glucose metabolism that would be related to a diminution of glycogenolysis in the liver in response to certain endogenous agents (5,6). Oxytetracycline reacts with membranes, and it has a great avidity for divalent cations, particularly Zn2 + and Ca2+ (20,21). Because of its metal-chelating properties, oxytetracycline is thought to be able to form complexes with Zn2+ and insulin, and it is believed that the insulin present in such complexes may have an increased affinity for its receptor in target tissues (8,9). Since the liver of the db/db mouse, like its ob/ob relative, is known to exhibit defective insuln binding (22,23), the interaction of oxytetracycline, Zn2+, and insulin at the membrane level could potentiate the action of insulin in these animals. Although speculative, at best, it is possible that enhanced insulin binding would result in an antagonism of glucagon-stimulated phosphorylase activation and concomitant activation of glycogen synthase either by an inhibition of a protein kinase or by a stimulation of its activating enzyme, synthase phosphatase (24,25). Alternatively, oxytetracycline might interfere with the glucose-mobilizing effects of catecholamines on the liver cell. By complexing with Ca*+, the drug could inhibit the activation of glycogen phosphorylase by calcium-mediated, ar-adrenergic stimulation (7,26,27). Either mechanism would involve the interaction of oxytetracycline and divalent cations at the liver cell membrane (7-9), and could form a basis for the insulinpotentiating effects of the drug on those aspects of hepatic glycogen metabolism presently reported. Despite the apparent therapeutic aspects of oxytetracycline treatment on carbohydrate metabolism in db/db mice, the present findings that drug-treated lean mice displayed substantial weight loss and increased hepatic lipid content in two of the three age groups examined are disturbing, and might present severe restrictions to the drug’s overall effectiveness. These findings support the observations of Dubuc et al. (6) that oxytetracycline treatment of lean littermates of ob/ob mice caused depressed skeletal growth and severe weight loss, while increasing hepatic lipids. Taken together, these studies are consistent with previous reports on the adverse effects of chronic administration of tetracycline or its derivates on protein and lipid metabolism in normal laboratory rodents and in man (2831). Ironically, it appears that the insulin-potentiating effects of oxytetracycline on carbohydrate metabolism may be offset by the drug’s ability to inhibit other positive effects of insulin, particularly on protein synthesis. Insulin plays a major role in stimulating protein biosynthesis. Manchester (32) showed that insulin administration results in an increased number of ribosomes and enhanced attachment of ribosomes to messenger RNA, and Bessman (3335) reported that insulin has a direct effect on protein synthesis, acting to promote energy-delivering processes for synthesis. Tetracycline and its derivatives have been shown to inhibit protein synthesis in mammalian cells (30). These antibiotics block the uptake of aminoacyl-transfer RNA on 30 S ribosomal units, and this appears to be the important step, resulting eventually in impaired protein synthesis (30). Furthermore, congeners of tetracycline, including oxytetracycline, localize
EFFECTS
OF OTC ON GLYCOGEN
ENZYMES
49
in mitochondria and inhibit mitochondrial protein synthesis and impair oxidative phosphorylation (30,31), thereby depressing the cellular content of ATP which could readily modify the rate of protein anabolism. Our observed accumulation of hepatic lipids following oxytetracycline administration to lean mice may be the consequence of impaired release of triglyceride from the liver (30). This steatogenic effect of oxytetracycline is thought to be another reflection of druginduced inhibition of protein (the apoprotein moiety of VLDL) synthesis in the liver (30). Thus, although the present data confirm and extend some of the conclusions of previous studies by demonstrating that oxytetracycline may be an effective agent in reducing the severity of some of the metabolic abnormalities associated with carbohydrate metabolism in the db/db mouse, it must be emphasized that these therapeutic aspects of drug administration are accompanied and, perhaps, overshadowed, by untoward effects on other critical aspects of hepatic metabolism. SUMMARY
The effects of daily oxytetracycline treatment on the activities of hepatic glycogen synthase, glycogen phosphorylase, plasma glucose, and insulin, and on liver glycogen, free fatty acid, and triglyceride levels were examined in 8- to 15 week-old genetically diabetic and lean mice. Oxytetracycline administration resulted in substantial reductions in the plasma glucose and immunoreactive-insulin levels in both diabetic and lean mice. The drug had no significant effect on the liver glycogen content in either phenotype, regardless of age, but it increased hepatic lipids and depressed body weights in lean animals. The most prominent effect of the drug was in markedly altering the activities of both glycogen synthase and phosphorylase in the liver of older diabetic mice. Oxytetracycline treatment produced a three-fold increase in the percentage of glycogen synthase I activity and reduced by one-third the percentage of glycogen phosphorylase a activity in 15week-old diabetic mice. In age-matched lean mice treated with oxytetracycline, the percentage of glycogen synthase I activity increased significantly, but the percentage of phosphorylase a activity was unchanged. These data suggest that the drug may alter an aspect of hepatic glycogen metabolism which might lead to an inhibition of glycogenolysis and subsequent diminution of blood sugar levels in the diabetic. The present results show that, while oxytetracycline may be effective in reducing the severity of some of the diabetic symptoms associated with carbohydrate metabolism in this animal model of maturity-onset diabetes, the drug may have adverse effects on aspects of protein and lipid metabolism in these animals. ACKNOWLEDGMENTS I thank Dr. Eugene M. Weiss (Terramycin), and Dr. Susan B. was supported by a Biomedical Resources, United States Public
of the Pfizer Laboratories for generously providing oxytetracycline Stearns for her excellent technical assistance and advice. This work Research Support Grant RR05402 fom the Division of Research Health Service.
REFERENCES 1. Coleman, D. L., and Hummel, K. P., Diabetologia 2. Steams, S. B., and Benzo, C. A., Horm. Metab.
3, 238 Res. 10,
(1967). 20 (1978).
50 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
CAMILLO
A. BENZ0
Begin-Heick, N., Bowcarsa, M., and Heick, H. M. C., Biochem. J. 142, 465 (1974). Begin-Heick, N., and Heick, H. M. C., Diuberologiu 12, 35 (1976). Dubuc, P. U., and Willis, P. L., Diubetologia 14, 129 (1978). Dubuc, P. U., Keith, L. D., Oey, H., and Mobley, P. W., J. N&r. 108, 874 (1978). DalpC-Scott, M., Heick, H. M. C., and Begin-Heick, N., Diabetes 31, 53 (1982). Dalp&Scott, M., Heick, H. M. C., and Begin-Heick, N., Diabetes 32, 932 (1983). Begin-Heick, N., Heick, H. M. C., and Norman, M. G., Diuberes 28, 65 (1979). Steams, S. B., and Benzo, C. A., Lub. Invesr. 37, 180 (1977). Roesler, W. J., and Khandelwal, R. L., Diaberes 34, 395 (1985). Steams, S. B., and Benzo, C. A., Biochem. Med. 25, 114 (1981). Benzo, C. A., and Steams, S. B., Biochem. Med. 26, 395 (1981). Thomas, J. A., Schlender, K. K., and Lamer, J., Anal. Biochem. 25, 486 (1968). Gilboe, D. P., Larson, K. L., and Nuttal, F. Q., Anal. Biochem. 47, 20 (1972). Froberg, S. O., Acru Med. &and. 193, 463 (1973). Falholt, K., Lund, B., and Falholt, W., Clin. Chim. Acra 46, 105 (1973). Templeton, M., J. Hisrochem. Cyrochem. 9, 670 (1961). Hales, C. N., and Randle, P. J., Biochem. J. 88, 137 (1963). Albert, A., and Rees, C. W., Nurure (London) 177, 433 (1956). Weinbrg, E. D., J. Infect. Dis. 95, 291 (1954). Chang, K. J., Huang, D., and Cuatrecasas, P., Biochem. Biophys. Res. Commun. 64,566 (1975). Soll, A. H., Kahn, C. R., Neville, D. M., and Roth, J., J. Clin. Znvesr. 56, 769 (1975). Miller, T. B., Jr., Gamache, A., and Cruz, J., .I. Biol. Chem. 259, 12470 (1984). Shen, L. C., Villar-Palasi, C., and Lamer, J., Physiol. Chem. Phys. 2, 536 (1970). Hiatt, N., and Bonorris, G., Diabetes 19, 307 (1970). Exton, J. H., Amer. J. Physiol. 238, E3 (1980). Breen, K. J., Schenker, S., and Heimberg, M., Gusrroenrerology 69, 714 (1975). Combes, B., Whalley, P. J., and Adams, R. H., Prog. Liver Dis. 4, 589 (1972). Zimmerman, H. J., “Hepatotoxicity.” Appleton-Century-Crofts, New York, 1978. Van Den Bogert, C., Menno, L., Majet, M., and Kroon, A. M., Biochim. Biophys. Acru 722, 393 (1983). Manchester, K. L., in “Biochemical Actions of Hormones” (G. Litwack, Ed.), Vol. 1, p. 267. Academic Press, New York, 1970. Bessman, S. P., J. Pediarr. 56, 191 (1960). Bessman, S. P., Amer. J. Med. 40, 740 (1966). Bessman, S. P., in “Cellular Regulation and Malignant Growth” (S. Ebashi, Ed.), pp. 276. Springer-Verlag, Berlin, 1985.
MEDKINE
AND
METABOLIC
BIOLOGY
37, 42-50
(1987)
Effects of Oxytetracycline Treatment on Enzymes of Hepatic Glycogen Metabolism in Genetically Diabetic (db/db) Mice CAMILLOA. BENZO Department
of Anatomy
and Cell Biology, State University of New York, Syracuse, Syracuse, New York 13210 Received
May
Health
Science
Center
at
28, 1985
The genetically diabetic mouse C57BL/KsJ db (db/db) displays many metabolic disturbances that resemble those found in maturity-onset diabetes in humans. These derangements include obesity, hyperglycemia, extremely elevated levels of circulating insulin and glucagon, and resistance to the hypoglycemic effects of exogenous insulin (1,2). Several investigators have demonstrated that the administration of the antibiotic oxytetracycline (OTC) to animals either with experimental diabetes or displaying genetic diabetes-obesity syndromes resulted in the amelioration of many of their diabetic symptoms. It was found that treatment of these animals with OTC markedly reduced their hyperglycemia and hyperinsulinemia, and virtually normalized their food intake, glucose tolerance, and insulin resistance (3-9). The results of these studies suggest that OTC appears to act by increasing target cell sensitivity to the catalytic actions of insulin, perhaps with a direct effect of the drug on hepatic carbohydrate metabolism (5 7). Because our own and other recent studies (10,ll) have shown that severe structural and chemical alterations associated with hepatic glycogen metabolism in db/db mice not only accompany the diabetic syndrome in these animals, but may be contributory, we decided to investigate whether chronic treatment of db/db mice with OTC would result in an improvement of their diabetic state, particularly with regard to liver glycogen metabolism. The present report documents the effects of 7 days of OTC administration on the activities of the rate-limiting enzymes in hepatic glycogen metabolism, glycogen synthase and phosphorylase, and on the plasma levels of glucose and insulin in genetically diabetic mice. In addition, because we have observed that the alterations in liver glycogen metabolism which occur in diabetic-obese mice are accompanied by changes in hepatic lipid metabolism (12,13), the effects of the drug on the concentrations of liver glycogen, free fatty acid, and triglyceride were monitored and are reported here as well.
0885-4505187 Copyright Au rights
$3.00
0 1987 by Academic Press. Inc. of reproduction in any form reserved.
42
EFFECTS OFOTCONGLYCOGEN
MATERIALS
ENZYMES
43
AND METHODS
Animals Genetically diabetic mice (C57BL/KsJdb) and their lean littermate controls were purchased from the Jackson Laboratory, Bar Harbor, Maine. Diabetic mice used in this study were 8, 12, and 15 weeks of age, three representative intervals in the advanced stage of the diabetic syndrome in these animals. The animals were housed in plastic cages with free access to food (Purina Mouse Chow) and water. Mice were maintained on a cycle of 12 hr of light (0600 to 1800 hr) and 12 hr of dark (1800 to 0600 hr). In each experimental group, db/db mice and their lean littermates received daily intraperitoneal injections of either oxytetracycline (Terramycin, Pfizer Laboratories; 100 mg/kg diluted in olive oil to give a final injection volume of 0.2 ml) or drug diluent over a 7-day period. The injections were given between 0900 and 1100 hr. Twenty-four hours after the final injection, the mice were killed by cervical dislocation between 0900 and 1100 hr, thereby minimizing differences due to diurnal variations in liver glycogen and lipid levels. The livers were frozen in situ with liquid nitrogen, then removed rapidly with liquid nitrogen-cooled aluminum clamps and stored at -80°C until they were assayed for glycogen synthase and phosphorylase activities, and for hepatic concentrations of glycogen, triglycerides and free fatty acids. Blood for immunoreactive insulin (IRI) and glucose determinations was collected in chilled, heparinized capillary tubes. Plasma was obtained immediately by centrifugation and stored for not more than 2 weeks at -80°C. Assay Methods For determination of enzymatic activities, tissue extracts were prepared by Polytron homogenization of 100 mg of frozen liver in 1 ml 50 mM KF, 10 mM EDTA, 0.5 M sucrose, and 62.5 mM glycylglycine, pH 7.4, at one-half speed for 10 set at 2°C. the homogenates were centrifuged at 2°C for 10 min in a Sorvall SS34 rotor at 12,OOOg, and the supernatants were used for both glycogen synthase and phosphorylase assays. Glycogen synthase (EC 2.4.1.11) activity was determined on aliquots of the supernatants according to the filter paper method of Thomas et al. (14) in the presence (total activity; I + D forms) or absence (I-form activity) of 10 mM glucose 6-phosphate. The assay is based upon the transfer of glucose to glycogen, using UDPG-[‘4C]glucose as substrate, followed by the isolation and counting of the radioactive glycogen. Glycogen phosphorylase (EC 2.4.1.1) activity was determined on aliquots of the supernatants which were diluted 1:3 with 50 mu KF, 60 mu 2-mercaptoethanol, and 50 mM ME& pH 6.1, immediately prior to the assay, following the filter paper method described by Gilboe et al. (15). Total phosphorylase (a + b) activity was assayed in the presence of 3 mM 5’-AMP, and phosphorylase a-form activity was determined in the absence of this nucleotide, but in the presence of 0.5 mM caffeine. This assay is based also upon the incorporation of glucose into glycogen, but using [‘4C]glucose l-phosphate as substrate. Both enzyme assays were linear with respect to time and to the amount of tissue extract added. The results of each assay are presented as units of enzyme
44
CAMILLOA.BENZO
activity per gram of tissue (wet wt), and as the percentage of the active form (-G-6-P/ + G-6-P x 100 or -AMP/ + AMP x 100). One unit of enzyme activity is the amount causing the transfer of 1 ,umole of glucose to glycogen in 1 min at 30°C. For determination of triglycerides and free fatty acids, portions of frozen liver were weighed rapidly and homogenized immediately in chloroform:methanol(2: 1) (16). On the following day, aliquots of the chloroform layer were evaporated to dryness under a stream of nitrogen. The residue was dissolved in isopropyl alcohol, and the triglycerides were extracted and assayed according to Tri-chol procedures (Oxford Reagent Set, Oxford Laboratories, Foster City, Calif.). Liver triglyceride levels are expressed as milligrams per gram of tissue (wet wt). The remaining portion of the chloroform layer was analyzed for free fatty acids by the method of Falholt er al. (17). Free fatty acid levels are expressed as micromoles per gram of tissue (wet wt). Hepatic glycogen was determined using the anthrone reagent according to the method of Templeton (18) and it is expressed as milligrams per gram of tissue (wet wt). Plasma IRI levels were determined by the procedure of Hales and Randle (19) utilizing a kit prepared by the Amersham Corporation (Arlington Heights, Ill.). Hormone samples were assayed in duplicate, using 50 ~1 of plasma for each determination. Plasma glucose was determined by a glucose oxidase method using a Beckman glucose analyzer. Values obtained from biochemical determinations are presented as means ? SEM. Mean values were compared by Student’s I test. Differences with P values < 0.05 were considered significant. RESULTS
The effects of OTC treatment on body weight, plasma glucose, and plasma IRI in lean and db/db mice are presented in Table 1. All lean mice showed significant weight loss following OTC administration, whereas db/db mice, regardless of age, displayed no substantial changes in body weight following drug treatment. In all age groups, db/db mice maintained much greater body weights than did their lean controls during the experiment (untreated db/db vs untreated lean, P < 0.001, Student’s t test). A significant effect of OTC in reducing plasma glucose levels was observed in all lean and db/db mice following drug treatment (Table 1). Similarly, except in severely hyperinsulinemic &week-old db/db mice, OTC administration markedly lowered plasma IRI levels in both lean and db/db mice, regardless of age. Throughout the experiment, all db/db mice maintained much higher plasma levels of glucose (P < 0.005) and IRI (P < 0.001) than did their appropriate lean controls, and the concentrations of both blood sugar and circulating insulin in all OTC-treated db/db mice remained significantly higher than those of either treated or untreated lean control mice. Table 2 summarizes the effects of OTC treatment on the glycogen content, free fatty acid, and triglyceride levels in liver from lean and db/db mice. During the study, no significant differences were found in the amount of glycogen present
EFFECTS
OF OTC ON GLYCOGEN
TABLE 1 Body Weight, Plasma Glucose, and Plasma IRI Levels in Oxytetracycline-Treated and Their Controls Age and group 8 Weeks Lean Lean-OTC dbldb db/db-OTC
12 Weeks Lean Lean-OTC db/db db/db-OTC
15 Weeks Lean Lean-OTC dbldb dbjdb-OTC
Body wt. k) 20 2 1 16 2 1
45
ENZYMES
Plasma glucose (mddl) 141 -t
10
Mice Plasma IRI W/ml) 26 2 I 10 + 1
(P < 0.025) 35 4 1
97-t 10 (P < 0.025) 859 ‘c 84
(P < 0.001) 236 2 5
37 2 1 VW
228 2 55 (P < 0.005)
232 2 8 (NW
22 2 1 18 ? 1 (P < 0.05) 52 ‘- 2 50 ” 1 W)
131 t 10 902 14 (P < 0.05) 617 t 124 303 t 69 (P < 0.05)
21 5 2 13 5 1 (P < 0.025) 156 + 3 88 4 2 (P < 0.001)
25 2 1 21 2 1 (P < 0.05) 52 f 2 50 2 2 WS)
166 t 17 99% 21 (P < 0.05) 793 t 73 316 t- 21 (P < 0.005)
20 i 1 10 2 1 (P < 0.001) 107 ? 4 86 ” 2 (P < 0.01)
Note. Values are presented as means +- SEM from at least three mice in each group. Probabilities were derived from Student’s t test of differences between means. NS, not significant at P < 0.05.
in the livers of db/db mice when compared to their appropriate lean controls. OTC treatment was ineffective in altering the hepatic glycogen content in both lean and db/db mice in all age groups. Similarly, except in drug-treated S-weekold lean animals, OTC had no effect on the free fatty acid levels in the livers of either lean or db/db mice. The livers of all db/db mice contained considerably more triglyceride (P < 0.05) than that found in their lean littermates, but no consistent effect of OTC administration in altering hepatic triglyceride levels was observed for either phenotype (Table 2). The data suggest a tendency for OTC to increase the hepatic triglyceride content in both lean and db/db mice. Drug treatment, however, produced significant increases in liver triglyceride content in g-week-old db/db mice and in 12- and U-week-old lean mice. Although a marginal effect (0.05 < P < 0.1) of the drug in raising the hepatic triglyceride level in 15-week-old db/db mice is suggested, OTC was ineffective in altering the liver triglyceride concentration in the remaining groups of lean or db/db mice. The effects of OTC administration on the activities of hepatic glycogen synthase and phosphorylase in lean and db/db mice are presented in Table 3. Regardless of age, all db/db mice contained substantially more total liver glycogen synthase activity (P < 0.025) than did their lean littermates. However, the levels of the
46
CAMILLO
A. BENZ0
TABLE 2 Glycogen, Free Fatty Acid, and Triglyceride Levels in Liver from Oxytetracycline-Treated and Their Controls Age and group 8 weeks Lean Lean-OTC db/db dbldb-OTC 12 Weeks Lean Lean-OTC db/db db/db-OTC 15 Weeks Lean Lean-OTC db/db db/db-OTC
Glycogen (mg/g tissue) 44.2 +43.3 +(NS) 49.7 -’ 45.1 f (W
2.6 3.0
53.6 f 53.0 2 (W 60.1 * 52.1 t (NS)
5.3 5.1
5.7 7.6
5.9 6.7
54.9 + 3.6 54.0 k 6.0 (NS) 47.4 k 10.0 42.2 e 6.4 (NS)
Mice
Free fatty acid (pmole/g tissue)
Triglyceride (mg/g tissue)
2.95 k 0.15 2.22 + 0.29 (P < 0.05) 2.04 2 0.18 2.05 + 0.13 (N.9
11.3 2 2.2 18.8 4 5.0 (W 30.4 k 7.9 48.5 k 2.7 (P < 0.05)
2.05 k 0.16 1.96 2 0.09 (W 2.31 t 0.23 2.06 k 0.29 PJS)
14.5 2 2.4 27.4 k 4.8 (P < 0.05) 52.3 5 0.3 55.7 2 4.4 tN.9
2.08 k 0.48 2.05 iz 0.07 (NW 2.25 f 0.27 2.48 f 0.18 tN.9
11.9 * 0.1 47.0 2 13.5 (P < 0.05) 53.2 e 13.8 82.6 k 7.2 W)
Note. Values are presented as means a SEM from at least three mice in each group. Probabilities were derived from Student’s t tests of differences between means. NS, not significant at P < 0.05.
physiologically active I form of the enzyme in livers from dbjdb mice were not different from those found in control livers. Unlike glycogen synthase, both total and active glycogen phosphorylase activities were consistently greater (P < 0.01 and P < 0.005, respectively) in the livers of all db/db mice when compared with their appropriate lean controls. OTC was ineffective in altering either total or I-form glycogen synthase activity when administered to 8- and IZweek-old mice, regardless of phenotype (Table 3). OTC administration to H-week-old lean mice resulted in a decrease in total glycogen synthase activity and a proportional increase in I-form activity. When 15week-old db/db mice were given OTC, no change in total liver synthase activity was observed, but these animals showed an almost threefold increase in I-form activity following drug treatment. Both total and active glycogen phosphorylase activities were lowered substantially in all drug-treated lean mice (Table 3). OTC was without effect on the activities of either form of phosphorylase when administered to 8- and 12-week-old db/db mice. Following OTC treatment, however, Sweek-old db/db mice showed significant decreases in the activities of both total and active phosphorylase.
EFFECTS
OF OTC ON GLYCOGEN
47
ENZYMES
TABLE 3 Glycogen Synthase and Phosphorylase Activities in Liver from Oxytetracycline-Treated Their Controls
Glycogen phosphorylase (units/g tissue)
Glycogen synthase (units/g tissue) Age and group 8 Weeks Lean Lean-OTC dbjdb dbldb-OTC 12 Weeks Lean Lean-OTC db/db dbldb-OTC I5 Weeks Lean Lean-OTC db/db dbldb-OTC
I+D
Mice and
I
%I
a +b
a
%a
0.069 2 0.003 0.083 + 0.007 (NV 0.133 2 0.23 0.135 2 0.005 (NS)
0.018 * 0.002 0.021 k 0.001 (NS) 0.026 2 0.006 0.019 2 0.007 WS)
26.1 25.3
3.52 ‘- 0.25 2.46 k 0.33 (P < 0.05) 6.45 k 0.33 5.62 2 0.52 (NS)
2.92 f 0.16 2.03 k 0.12 (P < 0.01) 5.28 k 0.19 4.72 k 0.52 (NS)
83.0 82.5
0.082 f 0.013 0.074 ” 0.010 (NS) 0.148 2 0.011 0.144 ” 0.009 VW
0.020 2 0.003 0.029 5 0.006 (NW 0.022 f 0.006 0.028 + 0.001 VW
24.4 39.2
3.81 * 0.37 2.08 2 0.21 (P < 0.01) 8.83 2 1.19 6.89 k 1.66 (NS)
2.73 k 0.16 1.39 + 0.15 (P < 0.005) 6.55 2 0.71 5.47 k 1.23 PW
71.7 66.8
0.067 2 0.001 0.056 ” 0.003 (P < 0.025) 0.119 ” 0.005 0.129 2 0.019 N-3
0.016 * 0.001 0.026 k 0.001 (P < 0.005) 0.018 k 0.003 0.060 f 0.016 (P < 0.05)
23.9 46.4
3.21 2 0.27 2.15 ” 0.19 (P < 0.025) 6.75 ” 0.22 5.59 2 0.42 (P < 0.05)
2.40 -+ 0.19 1.59 2 0.29 (P < 0.05) 6.57 * 0.23 3.74 2 0.15 (P < 0.001)
74.8 74.0
19.5 14.1
14.9 19.4
15.1 46.5
81.9 84.0
74.2 79.4
97.3 66.9
Note. Values are presented as means ? SEM from at least three mice in each group. Probabilities were derived from Student’s t tests of differences between means. NS, not significant at P < 0.05.
DISCUSSION
Previous reports on the effects of oxytetracycline in a related strain, the genetically obese (&lob) mouse, have shown that drug treatment (a) restored almost completely the binding of insulin to plasma membranes of the mutant liver cells, (b) reversed the insensitivity to insulin of the ob/ob mouse diaphragm both in viva and in vitro, (c) led to the regranulation of B cells in ob/ob pancreatic islet tissue, and (d) normalized the insulin secretory response in these animals (3,4,8,9). These findings prompted the suggestion that the effectiveness of oxytetracycline in the ob/ob mouse requires the prior development of insulin resistance and glucose intolerance (3,4). However, other studies, including the present one, have noted that substantially similar drug-induced metabolic changes also occur in lean mice following chronic oxytetracycline administration (5,6). Thus, while certain effects of oxytetracycline in either ob/ob or db/db mice might be mediated through a reversal of the insulin sensitivity of specific tissues, such consequences of drug treatment would not be expected to occur in lean mice. The present data tend to support the suggestion that oxytetracycline may have at least two discrete effects on the in vivo metabolism of obese-hyperglycemic
48
CAMILLOA.BENZO
animals (5,6). One effect may well be on the pancreatic B cells, improving abnormal insulin secretion and insulin resistance (8,9). Another effect, common to both phenotypes might alter some aspect of glucose metabolism that would be related to a diminution of glycogenolysis in the liver in response to certain endogenous agents (5,6). Oxytetracycline reacts with membranes, and it has a great avidity for divalent cations, particularly Zn2 + and Ca2+ (20,21). Because of its metal-chelating properties, oxytetracycline is thought to be able to form complexes with Zn2+ and insulin, and it is believed that the insulin present in such complexes may have an increased affinity for its receptor in target tissues (8,9). Since the liver of the db/db mouse, like its ob/ob relative, is known to exhibit defective insuln binding (22,23), the interaction of oxytetracycline, Zn2+, and insulin at the membrane level could potentiate the action of insulin in these animals. Although speculative, at best, it is possible that enhanced insulin binding would result in an antagonism of glucagon-stimulated phosphorylase activation and concomitant activation of glycogen synthase either by an inhibition of a protein kinase or by a stimulation of its activating enzyme, synthase phosphatase (24,25). Alternatively, oxytetracycline might interfere with the glucose-mobilizing effects of catecholamines on the liver cell. By complexing with Ca*+, the drug could inhibit the activation of glycogen phosphorylase by calcium-mediated, ar-adrenergic stimulation (7,26,27). Either mechanism would involve the interaction of oxytetracycline and divalent cations at the liver cell membrane (7-9), and could form a basis for the insulinpotentiating effects of the drug on those aspects of hepatic glycogen metabolism presently reported. Despite the apparent therapeutic aspects of oxytetracycline treatment on carbohydrate metabolism in db/db mice, the present findings that drug-treated lean mice displayed substantial weight loss and increased hepatic lipid content in two of the three age groups examined are disturbing, and might present severe restrictions to the drug’s overall effectiveness. These findings support the observations of Dubuc et al. (6) that oxytetracycline treatment of lean littermates of ob/ob mice caused depressed skeletal growth and severe weight loss, while increasing hepatic lipids. Taken together, these studies are consistent with previous reports on the adverse effects of chronic administration of tetracycline or its derivates on protein and lipid metabolism in normal laboratory rodents and in man (2831). Ironically, it appears that the insulin-potentiating effects of oxytetracycline on carbohydrate metabolism may be offset by the drug’s ability to inhibit other positive effects of insulin, particularly on protein synthesis. Insulin plays a major role in stimulating protein biosynthesis. Manchester (32) showed that insulin administration results in an increased number of ribosomes and enhanced attachment of ribosomes to messenger RNA, and Bessman (3335) reported that insulin has a direct effect on protein synthesis, acting to promote energy-delivering processes for synthesis. Tetracycline and its derivatives have been shown to inhibit protein synthesis in mammalian cells (30). These antibiotics block the uptake of aminoacyl-transfer RNA on 30 S ribosomal units, and this appears to be the important step, resulting eventually in impaired protein synthesis (30). Furthermore, congeners of tetracycline, including oxytetracycline, localize
EFFECTS
OF OTC ON GLYCOGEN
ENZYMES
49
in mitochondria and inhibit mitochondrial protein synthesis and impair oxidative phosphorylation (30,31), thereby depressing the cellular content of ATP which could readily modify the rate of protein anabolism. Our observed accumulation of hepatic lipids following oxytetracycline administration to lean mice may be the consequence of impaired release of triglyceride from the liver (30). This steatogenic effect of oxytetracycline is thought to be another reflection of druginduced inhibition of protein (the apoprotein moiety of VLDL) synthesis in the liver (30). Thus, although the present data confirm and extend some of the conclusions of previous studies by demonstrating that oxytetracycline may be an effective agent in reducing the severity of some of the metabolic abnormalities associated with carbohydrate metabolism in the db/db mouse, it must be emphasized that these therapeutic aspects of drug administration are accompanied and, perhaps, overshadowed, by untoward effects on other critical aspects of hepatic metabolism. SUMMARY
The effects of daily oxytetracycline treatment on the activities of hepatic glycogen synthase, glycogen phosphorylase, plasma glucose, and insulin, and on liver glycogen, free fatty acid, and triglyceride levels were examined in 8- to 15 week-old genetically diabetic and lean mice. Oxytetracycline administration resulted in substantial reductions in the plasma glucose and immunoreactive-insulin levels in both diabetic and lean mice. The drug had no significant effect on the liver glycogen content in either phenotype, regardless of age, but it increased hepatic lipids and depressed body weights in lean animals. The most prominent effect of the drug was in markedly altering the activities of both glycogen synthase and phosphorylase in the liver of older diabetic mice. Oxytetracycline treatment produced a three-fold increase in the percentage of glycogen synthase I activity and reduced by one-third the percentage of glycogen phosphorylase a activity in 15week-old diabetic mice. In age-matched lean mice treated with oxytetracycline, the percentage of glycogen synthase I activity increased significantly, but the percentage of phosphorylase a activity was unchanged. These data suggest that the drug may alter an aspect of hepatic glycogen metabolism which might lead to an inhibition of glycogenolysis and subsequent diminution of blood sugar levels in the diabetic. The present results show that, while oxytetracycline may be effective in reducing the severity of some of the diabetic symptoms associated with carbohydrate metabolism in this animal model of maturity-onset diabetes, the drug may have adverse effects on aspects of protein and lipid metabolism in these animals. ACKNOWLEDGMENTS I thank Dr. Eugene M. Weiss (Terramycin), and Dr. Susan B. was supported by a Biomedical Resources, United States Public
of the Pfizer Laboratories for generously providing oxytetracycline Stearns for her excellent technical assistance and advice. This work Research Support Grant RR05402 fom the Division of Research Health Service.
REFERENCES 1. Coleman, D. L., and Hummel, K. P., Diabetologia 2. Steams, S. B., and Benzo, C. A., Horm. Metab.
3, 238 Res. 10,
(1967). 20 (1978).
50 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
CAMILLO
A. BENZ0
Begin-Heick, N., Bowcarsa, M., and Heick, H. M. C., Biochem. J. 142, 465 (1974). Begin-Heick, N., and Heick, H. M. C., Diuberologiu 12, 35 (1976). Dubuc, P. U., and Willis, P. L., Diubetologia 14, 129 (1978). Dubuc, P. U., Keith, L. D., Oey, H., and Mobley, P. W., J. N&r. 108, 874 (1978). DalpC-Scott, M., Heick, H. M. C., and Begin-Heick, N., Diabetes 31, 53 (1982). Dalp&Scott, M., Heick, H. M. C., and Begin-Heick, N., Diabetes 32, 932 (1983). Begin-Heick, N., Heick, H. M. C., and Norman, M. G., Diuberes 28, 65 (1979). Steams, S. B., and Benzo, C. A., Lub. Invesr. 37, 180 (1977). Roesler, W. J., and Khandelwal, R. L., Diaberes 34, 395 (1985). Steams, S. B., and Benzo, C. A., Biochem. Med. 25, 114 (1981). Benzo, C. A., and Steams, S. B., Biochem. Med. 26, 395 (1981). Thomas, J. A., Schlender, K. K., and Lamer, J., Anal. Biochem. 25, 486 (1968). Gilboe, D. P., Larson, K. L., and Nuttal, F. Q., Anal. Biochem. 47, 20 (1972). Froberg, S. O., Acru Med. &and. 193, 463 (1973). Falholt, K., Lund, B., and Falholt, W., Clin. Chim. Acra 46, 105 (1973). Templeton, M., J. Hisrochem. Cyrochem. 9, 670 (1961). Hales, C. N., and Randle, P. J., Biochem. J. 88, 137 (1963). Albert, A., and Rees, C. W., Nurure (London) 177, 433 (1956). Weinbrg, E. D., J. Infect. Dis. 95, 291 (1954). Chang, K. J., Huang, D., and Cuatrecasas, P., Biochem. Biophys. Res. Commun. 64,566 (1975). Soll, A. H., Kahn, C. R., Neville, D. M., and Roth, J., J. Clin. Znvesr. 56, 769 (1975). Miller, T. B., Jr., Gamache, A., and Cruz, J., .I. Biol. Chem. 259, 12470 (1984). Shen, L. C., Villar-Palasi, C., and Lamer, J., Physiol. Chem. Phys. 2, 536 (1970). Hiatt, N., and Bonorris, G., Diabetes 19, 307 (1970). Exton, J. H., Amer. J. Physiol. 238, E3 (1980). Breen, K. J., Schenker, S., and Heimberg, M., Gusrroenrerology 69, 714 (1975). Combes, B., Whalley, P. J., and Adams, R. H., Prog. Liver Dis. 4, 589 (1972). Zimmerman, H. J., “Hepatotoxicity.” Appleton-Century-Crofts, New York, 1978. Van Den Bogert, C., Menno, L., Majet, M., and Kroon, A. M., Biochim. Biophys. Acru 722, 393 (1983). Manchester, K. L., in “Biochemical Actions of Hormones” (G. Litwack, Ed.), Vol. 1, p. 267. Academic Press, New York, 1970. Bessman, S. P., J. Pediarr. 56, 191 (1960). Bessman, S. P., Amer. J. Med. 40, 740 (1966). Bessman, S. P., in “Cellular Regulation and Malignant Growth” (S. Ebashi, Ed.), pp. 276. Springer-Verlag, Berlin, 1985.