Physiology& Behavior, Vol. 49, pp. 41--46. ©Pergamon Press plc, 1991. Printed in the U.S.A.
0031-9384/91 $3.00 + .00
The Effect of Alloxan-Induced Diabetes on Triamine Lesions in the Ventromedial Hypothalamus D A N L E Y F. B R O W N , J O S E P H P. M c G U I R K , S T E P H E N P. L A R S E N A N D S H A W N D. M I N T E R
Department o f Anatomy, University of Health Sciences 2105 Independence Blvd., Kansas City, MO 64124 R e c e i v e d 10 January 1989
BROWN, D. F., J. P. McGUIRK, S. P. LARSEN AND S. D. MINTER. The effect of alloxan-induced diabetes on triamine lesions in the ventromedial hypothalamus. PHYSIOL BEHAV 49(1) 41-46, 1991.--It has been demonstrated that the ventromedial hypothalamus (VMH) of alloxan-induced diabetic mice is protected from subsequent gold thioglucose (GTG)-induced lesions. Another compound, 3,3'-methyliminobis-(N-methylpropylamine) (MIMPA), a triamine structurally unrelated to GTG, has been shown to cause similar VMH lesions in mice. We chose to investigate the effect of alloxan-induced diabetes on VMH lesion formation in MIMPA-treated mice. In this study CF-1 female mice were made diabetic by a simple intravenous (IV) injection of alloxan and subsequently treated with MIMPA by subcutaneous injection (SC). Contrary to studies which showed that GTG-induced VMH lesions are insulin dependent, an insulin deficiency did not inhibit MIMPA-induced lesions in the VMH of mice. Our data suggests, albeit GTG is suspected to induce VMH necrosis by attaching to glucoreceptors and insulin sensitive neurons, MIMPA works by a different and as yet unknown mechanism. We conclude that MIMPA-induced lesions in the VMH of mice are not insulin dependent. Gold thioglucose Aliphatic triamine
Ventromedial hypothalamus
Alloxan diabetes
LEVINE and colleagues (13) have demonstrated that the aliphatic triamine, 3,3'-methyliminobis-(N-methylpropylamine) (MIMPA), produces lesions in the ventromedial hypothalamus (VMH) of mice that are histologically similar to the VMH damage resulting from gold thioglucose (GTG) (11). Since MIMPA and GTG are structurally unrelated, and both produce similar VMH lesions, these authors suggested the possibility that MIMPA and GTG caused VMH damage primarily by the same mechanism. This mechanism is thought to be a penetration of the central nervous system areas where the blood-brain barrier (BBB) is weak (13) and that GTG and MIMPA act as nonspecific neurotoxins which induce neural destruction. Moreover, Levine and colleagues state that their findings cast doubt on the hypothesis that GTG causes lesions by binding to hypothalamic glucoreceptors (13). It has been previously accepted that GTG-induced brain damage is nonspecific, and produced by vascular destruction with subsequent ischemia in areas possessing a weak BBB (14). This concept, however, overlooks the fact that GTG, as well as MIMPA, consistently spared the median eminence (an area of high BBB permeability) from necrosis (I 1). Furthermore, it has been observed that although gold is uniformly distributed throughout the brain following GTG administration, GTG-induced damage is highly focused and selective (10,11). It has been previously demonstrated (10,11) that neural elements in the VMH are the primary target sites for GTG action, contrary to earlier reports which suggested a nonspecific toxic effect of GTG on the capillary vasculature in the area. More recent ultrastructural studies have con-
3,3 '-Methyliminobis-(N-methylpropylamine)
firmed these observations (4,5). The effect of 5-thioglucose (5TG) on MIMPA- and GTG-induced pathology in the VMH has been previously documented (1). These data indicated that 5TG did not affect MIMPA lesion formation in the hypothalamus, but inhibited GTG-induced necrosis. These observations suggested that the action of GTG on the VMH is linked to a specific cell uptake or transport mechanism for glucose entry and not to a nonspecific toxic effect of the compound as has been postulated (13). Furthermore, other researchers have found that GTG-induced lesions of the VMH were absent in diabetic mice and were restored by either insulin, adrenalectomy or hypophysectomy (7, 8, 10, 11). These procedures are known to ameliorate glucose uptake by dependent cells in diabetic animals. It has also been observed that intrahypothalamic injections of insulin restored the sensitivity of diabetic mice to GTG-induced necrosis of the VMH (8). The objective of our study was to determine if MIMPA-induced VMH lesion formation was insulin dependent. This information would optimally help in determining the role of MIMPA in the production of VMH-induced necrosis. This role could be nonspecific and solely dependent on a more permeable BBB, although this hypothesis is in conflict with the sparing of the median eminance which is an area of blood-brain barrier weakness. Furthermore, MIMPA-induced necrosis may be specifically related to an affinity for glucoreceptors as has been postulated for GTG, or associated with an unknown mechanism.
41
42
BROVvN ET A~.
FIG. l. A cross-section of the VMH at the level (31"the median eminence from i~ normal mouse. The bar represents 0.12 m m (65 ,
METHOD
Twenty-seven female CF-I mice, weighing 18 to 22 grams, were used in this study. These animals were maintained at 23°C with a 12-hour light: 12-hour dark photoperiod and given Purina rat and mouse laboratory chow and tap water ad lib. VMH lesions were demonstrated by decapitation, removing the whole brain, fixation in Bouin's solution and excising coronal slabs which contained the hypothalamus. These slabs were embedded in paraffin, serially sectioned at 7 microns, mounted on glass slides, and stained with hematoxylin and eosin. GTG (Sigma) was added to distilled H20 to yield a concentration of 25 mg/ml. Control mice were administered GTG 500 mg/kg intraperito-
TABLE 1 SUMMARY OF RESULTS: EFFECT OF ALLOXAN-INDUCED DIABETES ON TRIAMINE LESIONS IN THE VMH
Treatment
G T G only M I M P A only G T G and alloxan
MIMPA and altoxan
Number of Animals
Animals With V M H Lesion %
1 6 7
1 6 2
100 100 29
13
12
92
neally (IP), sacrificed 24 hours later, and the brains examined. This insured the efficacy of our system in producing VMH lesions and that this strain and particular batch of mice were sensitive to VMH lesion production (N = 1 ). Additional control mice were pretreated with alloxan (Sigma) 100 mg/kg IV, tested with a Glucometer II Model 5550 (Ames) and shown to possess blood glucose levels greater than 400 mg%. These animals were considered to be diabetic, and 72 hours after the alloxan injection were treated with GTG 500 mg/kg IP, sacrificed 24 hours later, and the brains examined for GTG-induced VMH lesions (N = 7). The MIMPA (Aldrich) was neutralized with 10% acetic acid and diluted with saline to yield a concentration of 87 mg/ml. Animals receiving MIMPA included 2 groups. Group I was treated with MIMPA 870 mg/kg SC (N = 6). Group lI was pretreated with alloxan 100 mg/kg IV, tested and judged to be diabetic, and 72 hours later were treated with MIMPA 870 mg/kg SC (N = 13), Twenty-four hours after being injected with MIMPA the animals in Groups I and II were sacrificed and the VMH examined for histopathologic lesions. RESULTS
Figure 1 is a normal untreated mouse brain displaying the VMH. Figure 2 demonstrates a typical VMH lesion in a mouse treated with GTG only (N = 1). Figure 3 is representative of VMH tissue from mice treated with MIMPA only (N = 6). The typical bilateral necrosis seen on
ALLOXAN AND TRIAMINE LESION IN HYPOTHALAMUS
43
ii~'iiiiiiiii~!ii!i~i~!ii! ~!iiiiiii! !¸ •
):
FIG. 2. A cross-section of the VMH at the level of the median eminence from a mouse treated with 500 mg/kg GTG and sacrificed 24 hours later. Typical bilateral GTG-induced VMH lesions are seen at the tip of the arrows on either side of the third ventricle. The bar represents 0.12 mm (65 × ).
either side of the 3rd ventrical occurred in 100% of these mice. Figure 4 represents the brain from a mouse made diabetic by alloxan pretreatment and subsequently challenged with GTG 72 hours later (N = 7). Seventy-one percent of the samples examined did not have VMH lesions. Figure 5 is representative of VMH tissue from alloxan diabetic mice that were subsequently treated with MIMPA 72 hours later ( N = 13). Ninety-two pecent of these mice sustained bilateral VMH necrosis. Table 1 summarizes the occurance of VMH lesions in animals treated with GTG and MIMPA only and animals made diabetic with alloxan and subsequently treated with GTG and MIMPA. DISCUSSION
To our knowledge no specific effect on insulin-dependent glucose uptake in the VMH has been reported to date, however, previous work has clearly demonstrated that insulin is involved in the sensitivity of the hypothalamus to a GTG challenge (7, 8, 10, 11). Because of the histological similarity of GTG- and MIMPAinduced lesions in the VMH, some investigators have suggested that both compounds induce necrosis by a mechanism which is dependent primarily on a weaker blood-brain barrier and not specific cell receptors (13). The lesion produced by the triamine MIMPA has not been as extensively studied as the GTG lesion. Only ultrastructural changes induced by MIMPA have been described (13). Also, it does not appear that hormonal or metabolic studies have been conducted to
assess the effects of MIMPA-induced necrosis in the VMH. Our goal in this paper was to determine if MIMPA lesion formation is insulin dependent. As seen in Table 1, alloxan-induced diabetes does not appear to prevent VMH lesion formation by MIMPA, but does interfere with the production of GTG lesions. Our observations suggest no empirically detectable difference in VMH lesion size in mice treated with MIMPA only as compared to the alloxan-induced diabetic mice treated with MIMPA. However, we have observed that MIMPA-induced lesions are empirically more extensive than those produced by treatment with GTG. While the reason for the difference in size between GTG and MIMPA lesions is unclear, the empirically more extensive MIMPAinduced necrosis in the VMH and lack of insulin dependence may indicate that the MIMPA lesion is mechanically dependent on the higher BBB permeability. In contrast, GTG lesion formation is empirically less extensive, dependent on insulin (8, 10, 11) and adrenal hormones (2, 6, 11, 12), is blocked by beta-thioglucose (3), 5-thioglucose (1), 2-deoxy-D-glucose (15), 2-amino-2-deoxyD-glucose (15) (glucose analogs), and phlorizin a glucose transport inhibitor (9, 11, 14). These observations lead one to suspect a more specific mechanism for GTG-induced necrosis in the VMH rather than a nonspecific neurotoxic effect due to blood-brain bartier permeability changes in the area. ACKNOWLEDGEMENTS The authors wish to express their appreciation to H. A. Scalzi, Ph.D., P. M. Clevenger and V. D. Hoback for their technical assistance.
4-~
B R O W N Eq A [
FIG. 3. A cross-section of the VMH at the level of the median eminence from a mouse treated with 87(} mg/kg MIMPA aad sacrificed 24 hours later. Typical bilateral MIMPA-induced VMH lesions are seen at the tip of the arrows on either side of the third ventricle. The bar represents 0.12 mm (65 ~ ~.
REFERENCES 1. Brown, D. F. 5-Thioglucose prevents goldthioglucose lesions but not triamine lesions in the ventromedial hypothalamus. Neuroscience:669; 1983. 2. Brown, D. F. Inhibition of goldthioglucose lesion formation in the ventromedial hypothalamus by a glucocorticoid. Physiol. Behav. 37: 459--463; 1986. 3. Brown, D. F.; McGuirk, J. P.; Larsen, S. P.; Minter, S. D. Betathioglucose inhibits gold thioglucose lesions in the ventromedial hypothalamus. Physiol. Behav. 46:369-372; 1989. 4. Brown, D. F.; Viles, J. M. Systemic phlorizin prevents goldthioglucose necrosis in the ventromedial hypothalamus. Brain Res. Bull. 8: 347-351; 1982. 5. Brown, D. F.; Viles, J. M. An ultrastructural study of goldthioglucose of lesion formation in the ventromedial hypothalamus subnecrotic goldthioglucose dose. J. Neuropathol. Exp. Neurol. 45:20-27; 1986. 6. Bruce, B. K.; King, B. M.; Phelps, G. R.; Veitia, M. C. Effects of adrenalectomy and corticosterone administration on hypothalamic obesity in rats. Am. J. Physiol. 243:E152-E157; 1982. 7. Debons, A. F.; Krimsky, I.; From, A. A direct action of insulin on the hypothalamic satiety center. Am. J. Physiol. 219:938-943; 1970. 8. Debons, A. F.; Krimsky, I.; From, A.; Cloutier, R. J~ Rapid effects of insulin on the hypothalamic satiety center. Am. J. Physiol. 217:
1114-1118; 1969. 9. Debons, A. F.; Krimsky, I.; From, A.; Pattinian, H. Phlorizin inhibition of hypothalamic necrosis induced by gold thioglucose. Am J. Physiol. 226:574-578; 1974. 10. Debons, A. F.; Krimsky, l.; Likuski, H. J.; From, A.; Cloutier, R. J. Gold thioglucose damage to the satiety center inhibition in diabetes. Am. J. Physiol. 214:652-658; 1968. 11. Debons, A. F.; Krimsky, I.; Maayan, M. L.; Fani, K.; Jimenez, F. A. Gold thioglucose obesity syndrome. Fed. Proc. 36:143-147; 1977. 12. Debons, A. F.; Siclari, E.; Das, K. C.; Fuhr, B. Goldthioglucoseinduced hypothalamic damage hyperphagia, and obesity: Dependence on the adrenal gland. Endocrinology 110:2024-2029; 1982. 13. Levine, S.; Sowinski, R. Hypothalamic and medullary lesions caused by an aliphatic triamine unrelated to goldthioglucose. J. Neuropathol. Exp. Neurol. 41:54-66; 1982. 14. Liebelt, R. A.; Perry, J. H. Actions of gold thioglucose on the central nervous system. In: Code, C. F., ed. Handbook of physiology alimentary canal, vol. 1. Washington, DC: American Physiology Society; 1967:271-285. 15. Likuski, H. J.; Debons, A. E.; Cloutier, R. J. Inhibition of goldthioglucose-induced hypothalamic obesity by glucose analogues. Am. J. Physiol. 212:669-676; 1967.
A L L O X A N A N D T R I A M I N E L E S I O N IN H Y P O T H A L A M U S
FIG. 4. A cross-section of the VMH at the level of the median eminence from a mouse pretreated with 100 mg/kg alloxan, challenged 72 hours
later with 500 mg/kg GTG and sacrificed after 24 hours. No damage to the VMH is seen. The bar represents 0.12 m m (65 x ).
45
46
BROWN
E I h~
FIG. 5. A-cross section of the V M H at the level o f the median e m m e n c e l}om a mouse pretreated w~th IO0 mg/kg alloxan, challenged 72 hou~ later with 870 m g / k g M I M P A and sacrificed alter 24 hours. Typical bilateral M I M P A - i n d u c e d VMH lesions are seen at the tip ~f the arrm~. on either side of the third ventricle. The bar represents 0 . 1 2 m m ~65 x
0031-9384/91 $3.00 + .00
The Effect of Alloxan-Induced Diabetes on Triamine Lesions in the Ventromedial Hypothalamus D A N L E Y F. B R O W N , J O S E P H P. M c G U I R K , S T E P H E N P. L A R S E N A N D S H A W N D. M I N T E R
Department o f Anatomy, University of Health Sciences 2105 Independence Blvd., Kansas City, MO 64124 R e c e i v e d 10 January 1989
BROWN, D. F., J. P. McGUIRK, S. P. LARSEN AND S. D. MINTER. The effect of alloxan-induced diabetes on triamine lesions in the ventromedial hypothalamus. PHYSIOL BEHAV 49(1) 41-46, 1991.--It has been demonstrated that the ventromedial hypothalamus (VMH) of alloxan-induced diabetic mice is protected from subsequent gold thioglucose (GTG)-induced lesions. Another compound, 3,3'-methyliminobis-(N-methylpropylamine) (MIMPA), a triamine structurally unrelated to GTG, has been shown to cause similar VMH lesions in mice. We chose to investigate the effect of alloxan-induced diabetes on VMH lesion formation in MIMPA-treated mice. In this study CF-1 female mice were made diabetic by a simple intravenous (IV) injection of alloxan and subsequently treated with MIMPA by subcutaneous injection (SC). Contrary to studies which showed that GTG-induced VMH lesions are insulin dependent, an insulin deficiency did not inhibit MIMPA-induced lesions in the VMH of mice. Our data suggests, albeit GTG is suspected to induce VMH necrosis by attaching to glucoreceptors and insulin sensitive neurons, MIMPA works by a different and as yet unknown mechanism. We conclude that MIMPA-induced lesions in the VMH of mice are not insulin dependent. Gold thioglucose Aliphatic triamine
Ventromedial hypothalamus
Alloxan diabetes
LEVINE and colleagues (13) have demonstrated that the aliphatic triamine, 3,3'-methyliminobis-(N-methylpropylamine) (MIMPA), produces lesions in the ventromedial hypothalamus (VMH) of mice that are histologically similar to the VMH damage resulting from gold thioglucose (GTG) (11). Since MIMPA and GTG are structurally unrelated, and both produce similar VMH lesions, these authors suggested the possibility that MIMPA and GTG caused VMH damage primarily by the same mechanism. This mechanism is thought to be a penetration of the central nervous system areas where the blood-brain barrier (BBB) is weak (13) and that GTG and MIMPA act as nonspecific neurotoxins which induce neural destruction. Moreover, Levine and colleagues state that their findings cast doubt on the hypothesis that GTG causes lesions by binding to hypothalamic glucoreceptors (13). It has been previously accepted that GTG-induced brain damage is nonspecific, and produced by vascular destruction with subsequent ischemia in areas possessing a weak BBB (14). This concept, however, overlooks the fact that GTG, as well as MIMPA, consistently spared the median eminence (an area of high BBB permeability) from necrosis (I 1). Furthermore, it has been observed that although gold is uniformly distributed throughout the brain following GTG administration, GTG-induced damage is highly focused and selective (10,11). It has been previously demonstrated (10,11) that neural elements in the VMH are the primary target sites for GTG action, contrary to earlier reports which suggested a nonspecific toxic effect of GTG on the capillary vasculature in the area. More recent ultrastructural studies have con-
3,3 '-Methyliminobis-(N-methylpropylamine)
firmed these observations (4,5). The effect of 5-thioglucose (5TG) on MIMPA- and GTG-induced pathology in the VMH has been previously documented (1). These data indicated that 5TG did not affect MIMPA lesion formation in the hypothalamus, but inhibited GTG-induced necrosis. These observations suggested that the action of GTG on the VMH is linked to a specific cell uptake or transport mechanism for glucose entry and not to a nonspecific toxic effect of the compound as has been postulated (13). Furthermore, other researchers have found that GTG-induced lesions of the VMH were absent in diabetic mice and were restored by either insulin, adrenalectomy or hypophysectomy (7, 8, 10, 11). These procedures are known to ameliorate glucose uptake by dependent cells in diabetic animals. It has also been observed that intrahypothalamic injections of insulin restored the sensitivity of diabetic mice to GTG-induced necrosis of the VMH (8). The objective of our study was to determine if MIMPA-induced VMH lesion formation was insulin dependent. This information would optimally help in determining the role of MIMPA in the production of VMH-induced necrosis. This role could be nonspecific and solely dependent on a more permeable BBB, although this hypothesis is in conflict with the sparing of the median eminance which is an area of blood-brain barrier weakness. Furthermore, MIMPA-induced necrosis may be specifically related to an affinity for glucoreceptors as has been postulated for GTG, or associated with an unknown mechanism.
41
42
BROVvN ET A~.
FIG. l. A cross-section of the VMH at the level (31"the median eminence from i~ normal mouse. The bar represents 0.12 m m (65 ,
METHOD
Twenty-seven female CF-I mice, weighing 18 to 22 grams, were used in this study. These animals were maintained at 23°C with a 12-hour light: 12-hour dark photoperiod and given Purina rat and mouse laboratory chow and tap water ad lib. VMH lesions were demonstrated by decapitation, removing the whole brain, fixation in Bouin's solution and excising coronal slabs which contained the hypothalamus. These slabs were embedded in paraffin, serially sectioned at 7 microns, mounted on glass slides, and stained with hematoxylin and eosin. GTG (Sigma) was added to distilled H20 to yield a concentration of 25 mg/ml. Control mice were administered GTG 500 mg/kg intraperito-
TABLE 1 SUMMARY OF RESULTS: EFFECT OF ALLOXAN-INDUCED DIABETES ON TRIAMINE LESIONS IN THE VMH
Treatment
G T G only M I M P A only G T G and alloxan
MIMPA and altoxan
Number of Animals
Animals With V M H Lesion %
1 6 7
1 6 2
100 100 29
13
12
92
neally (IP), sacrificed 24 hours later, and the brains examined. This insured the efficacy of our system in producing VMH lesions and that this strain and particular batch of mice were sensitive to VMH lesion production (N = 1 ). Additional control mice were pretreated with alloxan (Sigma) 100 mg/kg IV, tested with a Glucometer II Model 5550 (Ames) and shown to possess blood glucose levels greater than 400 mg%. These animals were considered to be diabetic, and 72 hours after the alloxan injection were treated with GTG 500 mg/kg IP, sacrificed 24 hours later, and the brains examined for GTG-induced VMH lesions (N = 7). The MIMPA (Aldrich) was neutralized with 10% acetic acid and diluted with saline to yield a concentration of 87 mg/ml. Animals receiving MIMPA included 2 groups. Group I was treated with MIMPA 870 mg/kg SC (N = 6). Group lI was pretreated with alloxan 100 mg/kg IV, tested and judged to be diabetic, and 72 hours later were treated with MIMPA 870 mg/kg SC (N = 13), Twenty-four hours after being injected with MIMPA the animals in Groups I and II were sacrificed and the VMH examined for histopathologic lesions. RESULTS
Figure 1 is a normal untreated mouse brain displaying the VMH. Figure 2 demonstrates a typical VMH lesion in a mouse treated with GTG only (N = 1). Figure 3 is representative of VMH tissue from mice treated with MIMPA only (N = 6). The typical bilateral necrosis seen on
ALLOXAN AND TRIAMINE LESION IN HYPOTHALAMUS
43
ii~'iiiiiiiii~!ii!i~i~!ii! ~!iiiiiii! !¸ •
):
FIG. 2. A cross-section of the VMH at the level of the median eminence from a mouse treated with 500 mg/kg GTG and sacrificed 24 hours later. Typical bilateral GTG-induced VMH lesions are seen at the tip of the arrows on either side of the third ventricle. The bar represents 0.12 mm (65 × ).
either side of the 3rd ventrical occurred in 100% of these mice. Figure 4 represents the brain from a mouse made diabetic by alloxan pretreatment and subsequently challenged with GTG 72 hours later (N = 7). Seventy-one percent of the samples examined did not have VMH lesions. Figure 5 is representative of VMH tissue from alloxan diabetic mice that were subsequently treated with MIMPA 72 hours later ( N = 13). Ninety-two pecent of these mice sustained bilateral VMH necrosis. Table 1 summarizes the occurance of VMH lesions in animals treated with GTG and MIMPA only and animals made diabetic with alloxan and subsequently treated with GTG and MIMPA. DISCUSSION
To our knowledge no specific effect on insulin-dependent glucose uptake in the VMH has been reported to date, however, previous work has clearly demonstrated that insulin is involved in the sensitivity of the hypothalamus to a GTG challenge (7, 8, 10, 11). Because of the histological similarity of GTG- and MIMPAinduced lesions in the VMH, some investigators have suggested that both compounds induce necrosis by a mechanism which is dependent primarily on a weaker blood-brain barrier and not specific cell receptors (13). The lesion produced by the triamine MIMPA has not been as extensively studied as the GTG lesion. Only ultrastructural changes induced by MIMPA have been described (13). Also, it does not appear that hormonal or metabolic studies have been conducted to
assess the effects of MIMPA-induced necrosis in the VMH. Our goal in this paper was to determine if MIMPA lesion formation is insulin dependent. As seen in Table 1, alloxan-induced diabetes does not appear to prevent VMH lesion formation by MIMPA, but does interfere with the production of GTG lesions. Our observations suggest no empirically detectable difference in VMH lesion size in mice treated with MIMPA only as compared to the alloxan-induced diabetic mice treated with MIMPA. However, we have observed that MIMPA-induced lesions are empirically more extensive than those produced by treatment with GTG. While the reason for the difference in size between GTG and MIMPA lesions is unclear, the empirically more extensive MIMPAinduced necrosis in the VMH and lack of insulin dependence may indicate that the MIMPA lesion is mechanically dependent on the higher BBB permeability. In contrast, GTG lesion formation is empirically less extensive, dependent on insulin (8, 10, 11) and adrenal hormones (2, 6, 11, 12), is blocked by beta-thioglucose (3), 5-thioglucose (1), 2-deoxy-D-glucose (15), 2-amino-2-deoxyD-glucose (15) (glucose analogs), and phlorizin a glucose transport inhibitor (9, 11, 14). These observations lead one to suspect a more specific mechanism for GTG-induced necrosis in the VMH rather than a nonspecific neurotoxic effect due to blood-brain bartier permeability changes in the area. ACKNOWLEDGEMENTS The authors wish to express their appreciation to H. A. Scalzi, Ph.D., P. M. Clevenger and V. D. Hoback for their technical assistance.
4-~
B R O W N Eq A [
FIG. 3. A cross-section of the VMH at the level of the median eminence from a mouse treated with 87(} mg/kg MIMPA aad sacrificed 24 hours later. Typical bilateral MIMPA-induced VMH lesions are seen at the tip of the arrows on either side of the third ventricle. The bar represents 0.12 mm (65 ~ ~.
REFERENCES 1. Brown, D. F. 5-Thioglucose prevents goldthioglucose lesions but not triamine lesions in the ventromedial hypothalamus. Neuroscience:669; 1983. 2. Brown, D. F. Inhibition of goldthioglucose lesion formation in the ventromedial hypothalamus by a glucocorticoid. Physiol. Behav. 37: 459--463; 1986. 3. Brown, D. F.; McGuirk, J. P.; Larsen, S. P.; Minter, S. D. Betathioglucose inhibits gold thioglucose lesions in the ventromedial hypothalamus. Physiol. Behav. 46:369-372; 1989. 4. Brown, D. F.; Viles, J. M. Systemic phlorizin prevents goldthioglucose necrosis in the ventromedial hypothalamus. Brain Res. Bull. 8: 347-351; 1982. 5. Brown, D. F.; Viles, J. M. An ultrastructural study of goldthioglucose of lesion formation in the ventromedial hypothalamus subnecrotic goldthioglucose dose. J. Neuropathol. Exp. Neurol. 45:20-27; 1986. 6. Bruce, B. K.; King, B. M.; Phelps, G. R.; Veitia, M. C. Effects of adrenalectomy and corticosterone administration on hypothalamic obesity in rats. Am. J. Physiol. 243:E152-E157; 1982. 7. Debons, A. F.; Krimsky, I.; From, A. A direct action of insulin on the hypothalamic satiety center. Am. J. Physiol. 219:938-943; 1970. 8. Debons, A. F.; Krimsky, I.; From, A.; Cloutier, R. J~ Rapid effects of insulin on the hypothalamic satiety center. Am. J. Physiol. 217:
1114-1118; 1969. 9. Debons, A. F.; Krimsky, I.; From, A.; Pattinian, H. Phlorizin inhibition of hypothalamic necrosis induced by gold thioglucose. Am J. Physiol. 226:574-578; 1974. 10. Debons, A. F.; Krimsky, l.; Likuski, H. J.; From, A.; Cloutier, R. J. Gold thioglucose damage to the satiety center inhibition in diabetes. Am. J. Physiol. 214:652-658; 1968. 11. Debons, A. F.; Krimsky, I.; Maayan, M. L.; Fani, K.; Jimenez, F. A. Gold thioglucose obesity syndrome. Fed. Proc. 36:143-147; 1977. 12. Debons, A. F.; Siclari, E.; Das, K. C.; Fuhr, B. Goldthioglucoseinduced hypothalamic damage hyperphagia, and obesity: Dependence on the adrenal gland. Endocrinology 110:2024-2029; 1982. 13. Levine, S.; Sowinski, R. Hypothalamic and medullary lesions caused by an aliphatic triamine unrelated to goldthioglucose. J. Neuropathol. Exp. Neurol. 41:54-66; 1982. 14. Liebelt, R. A.; Perry, J. H. Actions of gold thioglucose on the central nervous system. In: Code, C. F., ed. Handbook of physiology alimentary canal, vol. 1. Washington, DC: American Physiology Society; 1967:271-285. 15. Likuski, H. J.; Debons, A. E.; Cloutier, R. J. Inhibition of goldthioglucose-induced hypothalamic obesity by glucose analogues. Am. J. Physiol. 212:669-676; 1967.
A L L O X A N A N D T R I A M I N E L E S I O N IN H Y P O T H A L A M U S
FIG. 4. A cross-section of the VMH at the level of the median eminence from a mouse pretreated with 100 mg/kg alloxan, challenged 72 hours
later with 500 mg/kg GTG and sacrificed after 24 hours. No damage to the VMH is seen. The bar represents 0.12 m m (65 x ).
45
46
BROWN
E I h~
FIG. 5. A-cross section of the V M H at the level o f the median e m m e n c e l}om a mouse pretreated w~th IO0 mg/kg alloxan, challenged 72 hou~ later with 870 m g / k g M I M P A and sacrificed alter 24 hours. Typical bilateral M I M P A - i n d u c e d VMH lesions are seen at the tip ~f the arrm~. on either side of the third ventricle. The bar represents 0 . 1 2 m m ~65 x
