0013-7227/91/l291-0361$03.00/0 Endocrinology Copyright ^ 1991 by The Endocrine Society
Vol. 129, No. 1 Printed in U.S.A.
Triiodothyronine-Receptor Complex in Rat Brain: Effects of Thyroidectomy, Fasting, Food Restriction, and Diabetes* BELEN SANCHEZ AND TRINIDAD JOLIN Institute de Investigacion.es Biomedicas, Consejo Superior de Inuestigaciones Cientificas, 28029-Madrid, Spain
ABSTRACT. In vitro saturation analysis combined with quantification of T3) by an isotopic equilibrium technique or RIA, were used to examine the effects of thyroidectomy, fasting, diabetes, and food restriction on T 3 concentration and specific binding in cerebral cortex and cerebellum. Fasting and food restriction did not affect the T3 binding parameters in the brain areas studied. Both thyroidectomy and diabetes were accompanied by a reduction in T3 content in nuclei from both cerebral cortex and cerebellum, but a decrease in T 3 binding sites was only observed in both brain areas of diabetic animals. No significant differences in the binding affinity values among the experimental groups were seen. The diabetes-induced decrease in T 3
T
content and receptor number were completely reversed by insulin treatment. Studies with fractionated nuclei from cerebral cortex and cerebellum showed that diabetes resulted in a reduction in T3 content and the number of receptors in glial nuclei from both brain areas. Although T3 content was also decreased in neuronal nuclei, the receptor concentration in these nuclear preparations did not change in concentration or affinity under the same conditions. These observations indicate that glial cells, not only have T3-binding characteristics similar to those of neuronal cells, but the T 3 receptor number is decreased in the diabetes state. (Endocrinology 129: 361-367, 1991)
nonthyroidal disease (16). In the present study we have used in vitro techniques to examine and compare in the rat the effects of starvation, food restriction, diabetes, and thyroidectomy on cerebral cortex and cerebellum nuclear T 3 receptor levels and nuclear T 3 content. This is clearly of interest, as it might provide insights into the pathophysiology of the euthyroid sick syndrome.
HYROID hormone initiates its biological effects on cellular metabolism by binding to a specific chromatin-associated receptor protein (1, 2). Specific nuclear binding sites have been identified in numerous tissues including the brain (3-5). However, the receptor concentration varies from tissue to tissue (4), suggesting that cells are programmed to respond differently to the same circulating levels of thyroid hormone. Changes in brain nuclear T 3 receptor concentration have been reported during development (6-9) and in neonatal hypothyroidism (7, 9), suggesting that they play an important role in the developing brain. However, modulation of the nuclear receptor by T 3 has not been described in the adult rat brain. Although studies in hyper- and hypothyroid rats suggest that the hepatic nuclear binding capacity does not appear to be regulated by thyroid hormone (10, 11), a decrease in the number of hepatic nuclear T 3 receptors occurs in different rat models of nonthyroidal disease (12-15). Only limited data are available, however, concerning the brain nuclear T 3 receptors in rat models of
Materials and Methods Animals Male Wistar rats, bred in our colony, weighing 95-116 g at the onset of the experiments were used. They were housed in a temperature-regulated room (22 ± 2 C) with a regulated lightdark cycle and received food and water ad libitum. Rats were made diabetic with streptozotocin (7.5 mg/100 g body wt), as previously described (17). One group of diabetic (D) rats received twice daily (0800 h and 1700 h) sc injections of 4 IU insulin/100 g body wt (Novo Laboratories, Copenhagen, Denmark) (D + I), and were killed 8 h after the last insulin dose. Groups of normal rats were fed ad libitum (C), limited to 5060% of that consumed by the C group (FR), or deprived of food for 4 days (F). D, D + I and FR rats were used 18-20 days after the onset of treatments. A group of normal rats was surgically thyroidectomized (Tx) and received a Remington low iodine diet and water supplemented with 0.9% CaCl2. One week after thyroidectomy, rats were injected with 100 >nCi 131I, ip. Rats were killed by exanguination through the abdominal aorta
Received November 29,1990. Address all correspondence and requests for reprints to: Dr. T. Jolin, Instituto de Investigaciones Biomedicas, Consejo Superior de Investigaciones Cientificas, c/ Arturo Duperier 4, 28029-Madrid, Spain. * This research was supported by from Fondo de Investigaciones Sanitarias de la Seguridad Social Grant 88/1453 and Comision Interministerial de Ciencia y Tecnologia Grant PM88-0015. 361
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
362
LOW T3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
under light ether anesthesia and were perfused with 30 ml cold 0.9% NaCl containing 3 U heparin and 10~4 M propylthiouracil (18) through the carotid arteria. The cerebral cortex and cerebellum were dissected out, weighed, and either rapidly frozen in liquid nitrogen and stored at —70 C, or used immediately. The plasma was separated by centrifugation and stored at —20 C. Isolation of nuclei
For each nuclear preparation the cerebral cortex and cerebellum from three to five rats were combined. Total nonfractionated nuclei were obtained as previously described (4, 5). Briefly, the tissues were homogenized in 0.32 M sucrose-1 mM MgCl2 solution at a final dilution of 1:15 (wt/vol). The nuclei were sedimented by centrifugation and were repeatedly washed initially with and then without 0.5% Triton X-100. The crude nuclear pellet was resuspended in 5 vol 0.32 M sucrose-1 mM MgCl2) mixed with 25 vol 2.4 M sucrose-1 mM MgCl2, and
centrifuged for 45 min at 110,000 X g. DNA recovery was 4050% for cortex and 68-75% for cerebellum. The nuclear pellet was either used immediately or resuspended in 5 vol 2.2 M sucrose-1 mM MgCl2 for fractionation into neuronal and glial nuclei by the method of Thompson (19), as modified by Gullo et al. (20). Shortly, the nuclear suspension was placed in centrifuge tubes on top of a 2.4 M sucrose mattress, and this suspension was overlayered with 2.2 M sucrose-1 mM MgCl2 and centrifuged for 30 min at 110,000 x g. Discontinuous sucrose gradient centrifugation of the crude nuclear fraction yielded four layers. The layer between 2.2-2.4 M sucrose comprised neuronal fraction (21, 22). The pellet at the base of the 2.4 M sucrose layer consisted mostly of oligodendroglia (21). According to morphological criteria, the purity of neuronal and glial nuclei-rich fractions was 92 ± 6% and 95 ± 2%, respectively. When estimating the number of binding sites in glial and neuronal nuclei the apparent cross-contamination of the two layers was taken into account. Binding assay Nonfractionated nuclei and neuronal and glial nuclei were suspended in 0.32 M sucrose-3 mM MgCl2-20 mM Tris-HCl, pH 7.5 (binding buffer) containing 2 mM dithiothreitol. Replicative samples (100 /x\) of the suspended nuclei (25-40 /A DNA/tube) were added to 400 n\ binding buffer and were incubated with five or six different concentrations of [125I]T3 (0.05-1.6 nM; New England Nuclear, Boston, MA; 1200 mCi/mg). Incubations were for 45 min at 37 C in a shaking water bath and were stopped by placing the tubes on ice. After the addition of 200 ix\ 2% Triton X-100, nuclei were pelleted by centrifugation (4 C, 10 min at 10,000 X g), and washed twice with incubation buffer containing 1% Triton X-100. Bound T3 was determined in the washed nuclear pellet and in the incubation medium after trapping labeled T 3 by a Dowex 1 x 8 anion exchange resin (23). More than 93% of the total hormone in the supernate was in the free form. Nonspecific binding, determined in the presence of stable T 3 (10~7 M), was substracted in all cases. Maximum binding capacity (MBC) and association constant (Ka) of the binding of T 3 to nuclei were determined by Scatchard analysis (24). The time course of in vitro association of T 3 to isolated
Endo«1991 Vol 129 • No I
nuclei, the rate of both T 3 dissociation and loss of receptor sites during in vitro incubation, and the endogenous T3 bound to receptors at the end of incubation period were determined as described previously (25, 26) and used to assess the actual maximal binding capacity, as outlined by Kolodny et al. (25). Nuclear T3 content Nuclear T 3 and T4 concentrations were determined by an isotopic equilibrium technique as described previously in detail (13, 27). Briefly, rats were injected daily ip with a 125I + 127I mixture (6 /iCi + 5 /ig/100 g body wt). Thirty days after initiating the treatment, the animals were divided into the above-mentioned groups. The injections were continued for the next 20 days. For the determination of plasma and nuclearlabeled iodothyronines, the samples were extracted twice in ethanol 10"4 M propylthiouracil (18). The extracts were dried, counted, and redissolved in 100 /A methanol-ammonia (9:1, vol/ vol) containing carrier T4 and T3 and subsequently submitted to TLC in 25% ammonia-methanol-chloroform (3:20:40, vol/ vol) according to the procedure previously described (28). The amounts of T4 and T3 were calculated from the specific activity of 125I. The specific activity was determined from the 125I content in aliquots of thyroid homogenates and their 127I, as determined chemically (29). In some cases, the T3 concentration in isolated nuclei as determined by the isotopic equilibrium technique was compared with values obtained by RIA (30), previously described in detail (31). T3 and T4 contents in samples from Tx rats were determined by RIA. [125I]T3 and [125I]T4 from New England Nuclear and antisera from Henning (West Berlin, West Germany) were used in the RIA techniques. Protein and DNA concentrations were determined by the methods of Lowry et al. (32) and Burton (33), respectively. Plasma glucose was measured by the glucose oxidase method (34). Statistical analysis Results have been expressed as means ± SD. Statistical comparisons were performed by analysis of variance followed by Tukey's test. The number of receptors and affinity constants in Scatchard plots were calculated by linear regression analysis.
Results The general characteristics of the experimental groups are shown in Fig. 1. Tx, F, FR, and D rats showed the expected decrease in weight and circulating levels of T3. A decrease in T4 was also observed in Tx, F, and D animals, whereas normal values were found in FR rats. D rats were hyperglycemic, whereas glucose levels in Tx, F, and FR rats were lower than those in C animals. D rats treated with insulin had similar weight, plasma glucose, and T4 and T 3 levels as C rats. In preliminary time-course experiments we found that T 3 binding to cerebrocortical and cerebellar nuclei of C, Tx, F, and D rats reached a plateau by 10 min, remaining stable up to 60 min (Fig. 2A, continuous lines). Rats from different groups had essentially the same kinetic of [125I]
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T3 RECEPTOR 600
363
Cerebral cortex
100 80 60
Cerebellum
40
6 300
20
J8
V •a
100
I
80
g
40
o to
•M-
60
20
6 0.6
100 80 60
EJFR
•c
D+J
FIG. 1. Final body wt, and plasma glucose, T4 and T 3 concentrations in C, Tx, F, FR, D, and D+I rats. F animals were deprived of food for 4 days. FR rats were limited to 50-60% of the food consumption for the C group. FR, D, and D+I rats were killed 18-20 days after the onset of treatment. D+I rats received twice daily sc injections of 4 IU/100 g body wt for the last 15-16 days and were killed 12 h after last injection. Tx animals were killed 30 days after surgery. Data are means ± SD of 8-10 rats per group. Asterisks indicates P < 0.05 or less vs. C group.
T 3 binding to both preparations of nuclei. However, for each group the percentage of bound T 3 in cerebellar nuclei was lower than that for cerebrocortical nuclei. In the presence of stable T 3 (10~7 M), the [125I]T3 dissociated from both preparations of nuclei in a time-dependent fashion (Fig. 2A, dashed lines). The disappearance rate of [12fiI]T3 from in vitro labeled cerebrocortical and cerebellar nuclei was described by an apparent constant rate (means ± SD for rats from four groups) of 0.71 ± 0.13 h"1 and 0.64 ± 0.10 h"1, respectively. The [125I]T3 disappearance rate from both preparations of nuclei of isotopically equilibrated rats (Fig. 2B) was identical in the four groups and similar to that of in vitro labeled nuclei. The results in Fig. 2C, indicating a time-dependent reduction in receptor concentration during incubation, suggest a fractional loss of receptor (means ± SD for rats from all groups) of 0.22 ± 0.05 h"1 and 0.11 ± 0.04 h"1 for nuclei from cerebral cortex and cerebellum, respectively. Identical binding pattern for the rates of T 3 association/ dissociation and loss of receptor sites were obtained for FR and D + I groups (data not shown). From these results, a 45-min incubation period at 37 C was subsequently chosen for all the experiments. The observed disappearance rates for endogenous nuclear T 3 and the loss of receptor sites were then applied as correction factors in actual MBC calculation. Scatchard analysis of saturation studies with cerebrocortical and cerebellar nuclei from C, Tx, F, and D groups are shown in Fig. 3. Figures 4 and 5 show the estimated total nuclear receptor capacity, nuclear T 3 content, and
oTx .F
20
20
40
60 0
40
20
60
Minutes
FIG. 2. A, Time course of in vitro association/dissociation of [125I]T3 added in vitro with isolated cerebrocortical and cerebellar nuclei. Nuclei were incubated with [125I]T3 (1 x 10"10 M) in binding buffer at 37 C. At 20 min (arrow) the incubation mixture was divided into two parts, to one of which was added stable T3 (final concentration 8 X 10"7 M, dashed lines) and to the other the solvent alone (continuous lines). B, Dissociation rate of endogenously bound [125I]T3 at 37 C. Isolated nuclei from both brain areas of isotopically equilibrated rats were incubated with either stable T3 (final concentration 8 x 10"7 M) or the solvent alone (not illustrated). C, In vitro MBC (fmol T3/mg DNA) of cerebrocortical and cerebellar nuclear T3 receptors determined after 45 mih incubation at 37 C, after preincubation at 37 C for 20, 40, and 60 min. Each experimental point is the mean of triplicate determinations. Individual experimental values were within 10-15% of each other. Cerebellum
cortex
Cerebral 8
1.2
6
0.9
V
z Q
? ree.
• Tx
4
u_ \
Bound
DC
Si
40
2
n
0.6
\
\
\
•c
X
oTx 0.3 *F
\
\
.
500
\
.
1000
*• \
0*
50
\ 100
150
Bound (fmol T 3 /mg DNA)
FIG. 3. Scatchard analysis of T 3 binding to isolated cerebrocortical and cerebellar nuclei from C, Tx, F, and D rats. Saturation analysis was performed at 37 C for 45 min. Each experimental point is the mean of triplicate determinations. Individual experimental values were within 10% of each other.
receptor saturation values for all experimental groups. In each group, both binding parameters were much higher in cerebrocortical than in cerebellar nuclei. Thy-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
364 1200
600
60
Cerebral cortex 60
O E 03 N.
600
300
T
1
Endo • 1991 Voll29-Nol
Cerebellum 16
30 30
D C0
• Tx mF m FR m0 • 0*1
• c
FlG. 4. T3 binding parameters of cerebrocortical nuclei in C, Tx, F, FR, D, and D+I rats. The receptor concentration values are corrected for receptor losses and nondissociated endogenous T3 during incubation at 37 C for 45 min. Data are means ± SD of four to six nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for nuclear T3 content per group. Asterisks indicate P < 0.05 or less vs. C group. For more details see the legend to Fig. 1.
m
N
DTX
BF
E3FR B D
Bo.I
FIG. 6. Concentrations of T4 in nuclei from cerebral cortex and cerebellum. Data are mean ± SD of four to six isotopically equilibrated rats for C, F, FR, D, and D+I rats. T4 content in nuclei of Tx animals was lower (nd) than limit of sensitivity of T4 RIA. Asterisks indicate P < 0.05 or less us. C group. For more details see legend to Fig. 1.
200
60
100
30
Cerebral cortex
Cerebellum
too 80 60 40 20 (00 80 60 40
E
Deo
20
D Tx 20
40
60 0
20
40
60
Minutes
D*J FIG. 5. T3 binding parameters of cerebellar nuclei in C, Tx, F, FR, D, and D+I rats. The receptor concentration values are corrected for receptor losses and nondissociated endogenous T3 during incubation at 37 C for 45 min. Data are the means ± SD of four to six nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for nuclear T3 content per group. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
roidectomy and diabetes resulted in a decrease in nuclear T 3 content, but a reduction in the MBC was only observed in D animals, in both brain areas. As expected from the marked reduction in the nuclear T 3 content, the receptor occupancy in Tx and D rats was lower than in C animals. The Ka did not significantly differ among the experimental groups, in either the cerebral cortex or the cerebellum. Insulin treatment completely reversed the diabetes-induced decrease in T 3 receptor and nuclear T 3 content. In addition, the T4 concentration in nuclei from cerebral cortex and cerebellum in D rats were lower than the corresponding values in C animals (Fig. 6). In Tx rats the T4 values in nuclei from both brain areas were too low for detection by RIA. To determine whether the decrease in the number of
FlG. 7. A, Time course of in uitro association/dissociation of [125I]T3 added in vitro with isolated neuronal and glial nuclei from cerebral cortex and cerebellum of C and D rats. Nuclei were incubated with [125I]T3 (1 x 10"10 M) in binding buffer at 37 C. At 20 min (arrow) the incubation mixture was divided into two parts, to one of which was added stable T3 (final concentration 9 x 10"7 M, dashed lines) and to the other the solvent alone (continuous lines). B, Dissociation rate of endogenously bound [125I]T3 at 37 C. Isolated neuronal and glial nuclei from both brain areas of isotopically equilibrated rats were incubated with stable T3 (final concentration 9 X 10"7 M) or the solvent alone (not illustrated).
T 3 receptors observed in D rats was a consequence of reduced receptor levels in neuronal and/or glial nuclei, studies with fractionated nuclei from cerebral cortex and cerebellum were performed. Neuronal and glial nuclei from both brain areas of C and D rats bound [125I]T3 in a time-dependent fashion; an apparent equilibrium was found after 10-60 min at 37 C (Fig. 7). The loss of endogenously bound [125I]T3 over the 60 min from the different preparations of nuclei was approximately 45% from neuronal nuclei and 40% for glial nuclei in either cerebral cortex and cerebellum (Fig. 7, A and B). These findings indicate that the binding of [125I]T3 to neuronal and glial nuclei was highly specific, rapid, and reversible, fulfilling the principal criteria for a true hormone-receptor interaction.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
365
LOW T3 SYNDROME AND NEURONAL AND GLIAL T., RECEPTOR Scatchard plots of the binding data for both nuclear preparations from each tissue are shown in Fig. 8. Figures 9 and 10 show the estimated binding parameters in both C and D rats. In each group, MBC and nuclear T 3 values were higher in neuronal and glial nuclei from cerebral cortex than that of corresponding values in each nuclear preparation from cerebellum. Diabetes produced a deCerebrol cortex
*
\
1200
150
300
Glial nuclei \.O
(No
250
50
1000
60
o E 500
30
fli •C • 0
FIG. 10. T 3 binding parameters in neuronal and glial nuclei from cerebellum in C and D rats. Receptor concentration values are corrected for nondissociated endogenous T 3 and dissociated T3-receptor complex during incubation at 37 C for 45 min. The MBC values were also corrected for cross-contamination of the two nuclear preparations. Data are mean ± SD of four to five nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for endogenous T 3 content per group. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
Discussion
FIG. 8. Scatchard analysis of T 3 binding to isolated neuronal and glial nuclei from cerebral cortex and cerebellum in C and D rats. Saturation analysis was performed at 37 C for 45 min. Each point is the mean of triplicate determinations. Individual experimental values were within 10-14% of each other.
S "3 E
2
100
Bound (fmol T 3 /mg DNA)
»
40
crease in nuclear T 3 content in the four nuclear preparations, but a reduction in the MBC was only seen in glial nuclei from both brain areas. The Kn values of binding in neuronal and glial nuclei from cerebral cortex and cerebellum were not different between C and D rats.
0.5 v
CQ v 1000
Q
s i1 too
£
\
2000
80
Dc
\
125
_
200
nuclei
3
600
20Q
Cerebellum
Neuronol
\
400
I •
FlG. 9. T 3 binding parameters in neuronal and glial nuclei from cerebral cortex in C and D rats. Receptor concentration values are corrected for nondissociated endogenous T 3 and dissociated T3-receptor complex during incubation at 37 C for 45 min. The MBC values were also corrected for cross-contamination of the two nuclear preparations. Data are means ± SD of four to five nuclear preparations for MBC, Ka, and DNA and isotopically equilibrated rats for nuclear T 3 content. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
The present study indicates that diabetes in rats results in a decrease in T 3 receptors in cerebral cortex and cerebellum. In contrast, T3-binding sites in both brain areas of Tx, F, and FR animals did not significantly change in concentration or affinity. Both thyroidectomy and diabetes were accompanied by a decrease in nuclear T 3 content and a marked reduction in receptor occupancy in both cerebral cortex and cerebellum, whereas fasting or food restriction produced no changes in T 3 concentration in any brain areas studied. Preservation of the T 3 content in both brain areas of F and FR rats might be related to the finding that in the hypothyroid rat cerebral cortex the efficiency of T4 to T 3 conversion is increased, and in both cerebral cortex and cerebellum the fractional removal rate of T 3 is slow (35, 36). Thus, the metabolic conditions resulting from a diabetes period of 20 days are probably not comparable to those resulting from 4 days of complete food withdrawal or 20 days of food restriction. The results from these studies also suggest that specific T3-binding in cerebral cortex and cerebellum may be capable of down-regulation in response to a metabolic change associated with diabetes and low circulating T 3 levels. Besides, the observation that these effects of diabetes can be prevented by the concomitant administration of insulin suggests that these receptors
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
366
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
are conceivably regulated by insulin. However, these preliminary experiments did not reveal whether the changes in T 3 receptor binding were directly or indirectly related to the action of insulin. The MBC we estimated for cerebral cortex and cerebellum nuclei from C rats agrees with that reported by others using in vitro techniques (5, 20, 25, 37), but it is higher (3.2-fold for cortex and 1.4-2.1-fold for cerebellum) than those determined by in vivo saturation techniques (4, 38, 39). Similar discrepancies have been noted between data obtained in vivo and in vitro in euthyroid and hypothyroid rat pituitaries (40). The T 3 content found in nuclei from both brain areas in this study is similar to that previously reported (38, 39). The differences between in vivo and in vitro measurements have been attributed to the existence of nuclear T3-binding sites not available for in vivo binding (38) or to the possibility that the methodological assumption inherent in the in vivo techniques (41) results in the determination of only a fraction of the total nuclear receptors (40). Irrespective of the causes for these differences, the large discrepancies between the MBC determined in vivo and in vitro and the similarity of nuclear T 3 content found in both situations may account for the discrepancies between the endogenous cerebral cortex and cerebellum receptor occupancy estimated in vivo and in vitro. A saturable T 3 binding was found in both neuronal and glial nuclei from cerebral cortex and cerebellum of C and D rats, and all had a similar Ka, very close to that for total nonfractionated cerebrocortical and cerebellar nuclei. Moreover, T 3 binding sites appear to be more numerous in neuronal than in glial nuclei in both brain areas. This reflects the findings of Gullo et al. (20, 37), who reported a lower number of T 3 binding sites in glial than in neuronal nuclei, at least in cerebral cortex. Recent reports suggest an absence of specific T 3 receptors in glial nuclei from cerebral cortex (25) and cerebellum (37) from normal rats. A possible explanation is that an incomplete dissociation of endogenous T3, dissociation of T3-receptor complex, or receptor instability during the incubation period may have resulted in serious misinterpretation of the in vitro binding data in glial nuclei. However, Yusta et al. (42) have clearly demonstrated the presence of nuclear high affinity T 3 receptors in pure secondary cultures of oligodendrocytes. Our findings show that diabetes is accompanied by a reduction of T 3 content, T3-binding sites, and receptor occupancy in glial nuclei from cerebral cortex and cerebellum. In neuronal nuclei, a different situation is seen. Thus, although nuclear T 3 levels are also decreased, the T 3 receptor sites in neuronal nuclei from both brain areas did not significantly change in concentration or affinity under the same conditions. These findings suggest that diabetes can produce receptor changes that are localized
Endo-1991 Voll29-Nol
in specific cells, leaving others unaffected. This represents, to our knowledge, the first observation that glial T3-binding can be altered in vivo. The mechanism for the decrease in glial nuclei T 3 receptors in D rats could involve degradation or inactivation rate, decreased biosynthesis rate, or both. However, our results suggest that the decreased binding in the glial nuclei of D rats is not due to down-regulation by the presence of elevated circulating T 3 levels. The diabetes-induced decrease in specific T 3 binding sites in glial nuclei is a descriptive finding, and neither the causes nor the consequences of these changes are clarified by these experiments. The decrease in binding sites may lead to alterations in the expression of certain T 3 effects. Glial cells, oligodendrocytes, are involved in myelin synthesis, and T 3 is known to influence the synthesis and activity of the enzymes
associated with myelinogenesis (43, 44). Therefore, it is possible that diabetes-mediated decrease in T 3 receptors may result in a situation where myelin synthesis is depressed. Moreover, it is conceivable that metabolic factors associated with diabetes may play a role. Regarding the last possibility, previous results have clearly demonstrated that insulin stimulates glucose uptake in glial cells from rat brain, and the effect is mediated by insulin-specific receptors (45). The importance of these findings is that a major component of the central nervous system, the glial cell, not only has T3-binding characteristics similar to those of neuronal cell, but the number of T 3 receptors is decreased in the diabetic state.
Acknowledgments The authors thank Margarita Gonzalez for excellent technical assistance. The provision of Streptozotocin by Upjohn Research Laboratories (Kalamazoo, MI) is gratefully acknowledged.
References 1. Oppenheimer JH 1979 Thyroid hormone action at the cellular level. Science 203:971-979 2. Oppenheimer JH 1983 The nuclear receptor-triiodothyronine complex: relationship to thyroid hormone distribution, metabolism and biological action. In: Oppenheimer JH, Samuels HH (eds) Molecular Basis of Thyroid Hormone Action. Academic Press, New York, pl-35 3. Eberhardt N, Valcana T, Timiras PS 1978 Triiodothyronine nuclear receptors: an in vitro comparison of the binding of triiodothyronine to nuclei of adult rat liver, cerebral hemisphere, and anterior pituitary. Endocrinology 102:556-561 4. Oppenheimer JH, Schwartz HL, Surks MI 1974 Tissue differences in the concentration of triiodothyronine nuclear binding sites in the rat liver, kidney, pituitary, heart, brain, spleen, and testis. Endocrinology 95:897-903 5. Schwartz HL, Oppenheimer JH 1978 Nuclear triiodothyronine receptor sites in brain: probable identity with hepatic receptors and regional distribution. Endocrinology 103:267-273 6. Dozin B, De Nayer P 1979 Nuclear triiodothyronine receptors in rat brain during maturation. Brain Res 177:551-554 7. Ishiguro K, Suzuki Y, Sato T 1980 Effect of neonatal hypothyroidism on maturation of nuclear triiodothyronine (T3) receptors in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T3 RECEPTOR developing rat brain. Acta Endocrinol (Copenh) 95:495-504 8. Schwartz HL, Oppenheimer JH 1978 Ontogenesis of 3,5,3'-triiodothyronine receptors in neonatal rat brain: dissociation between receptor concentration and stimulation of oxygen consumption by 3,5,3'-triiodothyronine. Endocrinology 103:943-948 9. Valcana T, Timiras PS 1978 Nuclear triiodothyronine receptors in the developing rat brain. Mol Cell Endocrinol 11:31-41 10. Silva JE, Larsen PR 1978 Contributions of plasma triiodothyronine and local thyroxine monodeiodination to triiodothyronine and nuclear triiodothyronine receptor saturation in pituitary, liver and kidney of hypothyroid rats. Further evidence relating saturation of pituitary nuclear triiodothyronine receptors and acute inhibition of thyroid-stimulating hormone release. J Clin Invest 61:1247-1259 11. Spindler BJ, MacLeod KM, Ring J, Baxter JD 1975 Thyroid hormone receptors: binding characteristics and lack of hormonal dependency for nuclear localization. J Biol Chem 250:4113-4119 12. Burman KD, Lukes Y, Wright FD, Wartofsky L 1977 Reduction in hepatic triiodothyronine binding capacity by fasting. Endocrinology 101:1331-1334 13. Jolin T 1987 Diabetes decreases liver and kidney nuclear 3,5,3'triiodothyronine receptors in rats. Endocrinology 120:2144-2151 14. Lim VS, Passo C, Murata Y, Ferrari E, Nakamura H, Refetoff S 1984 Reduced triiodothyronine in liver but not pituitary of the uremic rat model: demonstration of changes compatible with thyroid hormone deficiency in liver only. Endocrinology 114:280-289 15. Tibaldi JM, Surks MI 1985 Animal models of nonthyroidal diseases. Endocr Rev 6:87-101 16. Thrall C 1983 Effects of various pathological conditions on the nuclear T3 receptors of rat cerebral cortex: comparison with liver. Brain Res 279:177-183 17. Ortiz-Caro J, Obregon MJ, Pascual A, Jolin T 1985 Decreased T4 to T3 conversion in tissues of streptozotocin-diabetic rats. Acta Endocrinol (Copenh) 106:86-91 18. Morreale de Escobar G, Llorente P, Jolin T, Escobar del Rey F 1963 The "transient instability" of thyroxine and its biochemical applications. Biochem J 88:526-530 19. Thompson RJ 1973 Studies of RNA synthesis in two populations of nuclei from the mammalian cerebral cortex. J Neurochem 21:1940 20. Gullo D, Sinha AK, Woods R, Pervin K, Ekins RP 1987 Triiodothyronine binding in adult rat brain: compartmentation of receptor populations in purified neuronal and glial nuclei. Endocrinology 120:325-331 21. L0vtrup-Rein H, McEwen BS 1966 Isolation and fractionation of rat brain nuclei. J Cell Biol 30:405-414 22. Rose SPR, Sinha AK 1969 Some properties of isolated neuronal cell fractions. J Neurochem 16:1319-1328 23. Torresani J, DeGroot LJ 1975 Triiodothyronine binding to liver nuclear solubilized proteins in vitro. Endocrinology 96:1201-1209 24. Scatchard G 1949 The attraction of proteins for small molecules and ions. Ann NY Acad Sci 51:660-672 25. Kolodny JM, Larsen PR, Silva JE 1985 In vitro 3,5,3'-triiodothyronine binding to rat cerebrocortical neuronal and glial nuclei suggests the presence of binding sites unavailable in vivo. Endocrinology 116:2019-2028 26. Jolin T 1988 Response of hepatic mitochondrial a-glycerophosphate dehydrogenase and malic enzyme to 3,5,3'-triiodothyronine in streptozotocin-diabetic rats. Endocrinology 123:248-257 27. Surks MI, Oppenheimer JH 1977 Concentration of L-thyroxine and L-triiodothyronine specifically bound to nuclear receptors in rat liver and kidney. Quantitative evidence favoring a major role
367
of T3 in thyroid hormone action. J Clin Invest 60:555-562 28. Heider JG, Bronk JR 1965 A rapid separation of thyroxine and some analogues by thin-layer chromatography. Biochim Biophys Acta 96:353-355 29. Benotti J, Benotti N 1963 Protein bound iodine, total iodine and butanol extractable iodine by partial automation. Clin Chem 9:408410 30. Weeke J, 0rskov H 1975 Ultrasensitive radioimmunoassay for direct determination of free triiodothyronine concentration in serum. J Clin Invest 35:237-244 31. Weeke J, 0rskov H 1978 Evaluation of thyroid function. In: Alberti KGM (ed) Recent Advance in Clinical Biochemistry. Churchill Levingstone, London, vol 1:111-128 32. Lowry OH, Rosebrough NJ, Farr AL, Randle RJ 1951 Protein measurements with Folin phenol reagent. J Biol Chem 193:265269 33. Burton K 1956 A study of conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem J 62:315-323 34. Hugget ASG, Nixon DA 1957 Use of glucose-oxidase, peroxidase and o-dianisidine in determination of blood and urine glucose. Lancet 2:368-370 35. Silva JE, Larsen PR 1982 Comparison of iodothyronine 5'-deiodinase and other thyroid-hormone dependent enzyme activities in the cerebral cortex of hypothyroid neonatal rat. Evidence for adaptation to hypothyroidism. J Clin Invest 70:1110-1123 36. Silva JE, Matthews PS 1984 Production rates and turnover of triiodothyronine in rat-developing cerebral cortex and cerebellum. Response to hypothyroidism. J Clin Invest 74:1035-1049 37. Gullo D, Sinha AK, Bashir A, Hubank M, Ekins RP 1987 Differences in nuclear triiodothyronine binding in rat brain cells suggest phylogenetic specialization of neuronal fractions. Endocrinology 120:2398-2403 38. Crantz FR, Silva JE, Larsen PR 1982 An analysis of the sources and quantity of 3,5,3'-triiodothyronine specifically bound to nuclear receptors in rat cerebral cortex and cerebellum. Endocrinology 110:367-375 39. Oppenheimer JH, Schwartz HL 1985 Stereospecific transport of triiodothyronine from plasma to cytosol and from cytosol to nucleus in rat liver, kidney, brain, and heart. J Clin Invest 75:147154 40. St Germain DL, Galton VA 1985 Comparative study of pituitarythyroid hormone economy in fasting and hypothyroid rats. J Clin Invest 75:679-688 41. Pearson JD, Veall N, Vetter H 1958 A practical method for plasma albumin turnover studies. Strahlentherapie 38:290-297 42. Yusta B, Besnard F, Ortiz-Caro J, Pascual A, Aranda A, Sarlieva L 1988 Evidence for the presence of nuclear 3,5,3'-triiodothyronine receptors in secondary cultures of pure rat oligodendrocytes. Endocrinology 122:2278-2284 43. Sarlieve L, Bouchon R, Koehl C, Neskovic MM 1983 Cerebroside and sulfatide biosynthesis in the brain of Snell Dwarf mouse: effects of thyroxine and growth hormone in the early postnatal period. J Neurochem 40:1058-1062 44. Bhat NR, Sarlieve L, Subba Rao G, Pieringer RA 1979 Investigations on myelination in vitro. Regulation by thyroid hormone in cultures of dissociated brain cells from embryonic mice. J Biol Chem 245:9342-9344 45. Clarke DW, Boyd Jr FT, Kappy MS, Ralzada MK 1984 Insulin binds to specific receptors and stimulates 2-deoxy-D-glucose uptake in cultures glial cells from rat brain. J Biol Chem 259:11672-11675
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
Vol. 129, No. 1 Printed in U.S.A.
Triiodothyronine-Receptor Complex in Rat Brain: Effects of Thyroidectomy, Fasting, Food Restriction, and Diabetes* BELEN SANCHEZ AND TRINIDAD JOLIN Institute de Investigacion.es Biomedicas, Consejo Superior de Inuestigaciones Cientificas, 28029-Madrid, Spain
ABSTRACT. In vitro saturation analysis combined with quantification of T3) by an isotopic equilibrium technique or RIA, were used to examine the effects of thyroidectomy, fasting, diabetes, and food restriction on T 3 concentration and specific binding in cerebral cortex and cerebellum. Fasting and food restriction did not affect the T3 binding parameters in the brain areas studied. Both thyroidectomy and diabetes were accompanied by a reduction in T3 content in nuclei from both cerebral cortex and cerebellum, but a decrease in T 3 binding sites was only observed in both brain areas of diabetic animals. No significant differences in the binding affinity values among the experimental groups were seen. The diabetes-induced decrease in T 3
T
content and receptor number were completely reversed by insulin treatment. Studies with fractionated nuclei from cerebral cortex and cerebellum showed that diabetes resulted in a reduction in T3 content and the number of receptors in glial nuclei from both brain areas. Although T3 content was also decreased in neuronal nuclei, the receptor concentration in these nuclear preparations did not change in concentration or affinity under the same conditions. These observations indicate that glial cells, not only have T3-binding characteristics similar to those of neuronal cells, but the T 3 receptor number is decreased in the diabetes state. (Endocrinology 129: 361-367, 1991)
nonthyroidal disease (16). In the present study we have used in vitro techniques to examine and compare in the rat the effects of starvation, food restriction, diabetes, and thyroidectomy on cerebral cortex and cerebellum nuclear T 3 receptor levels and nuclear T 3 content. This is clearly of interest, as it might provide insights into the pathophysiology of the euthyroid sick syndrome.
HYROID hormone initiates its biological effects on cellular metabolism by binding to a specific chromatin-associated receptor protein (1, 2). Specific nuclear binding sites have been identified in numerous tissues including the brain (3-5). However, the receptor concentration varies from tissue to tissue (4), suggesting that cells are programmed to respond differently to the same circulating levels of thyroid hormone. Changes in brain nuclear T 3 receptor concentration have been reported during development (6-9) and in neonatal hypothyroidism (7, 9), suggesting that they play an important role in the developing brain. However, modulation of the nuclear receptor by T 3 has not been described in the adult rat brain. Although studies in hyper- and hypothyroid rats suggest that the hepatic nuclear binding capacity does not appear to be regulated by thyroid hormone (10, 11), a decrease in the number of hepatic nuclear T 3 receptors occurs in different rat models of nonthyroidal disease (12-15). Only limited data are available, however, concerning the brain nuclear T 3 receptors in rat models of
Materials and Methods Animals Male Wistar rats, bred in our colony, weighing 95-116 g at the onset of the experiments were used. They were housed in a temperature-regulated room (22 ± 2 C) with a regulated lightdark cycle and received food and water ad libitum. Rats were made diabetic with streptozotocin (7.5 mg/100 g body wt), as previously described (17). One group of diabetic (D) rats received twice daily (0800 h and 1700 h) sc injections of 4 IU insulin/100 g body wt (Novo Laboratories, Copenhagen, Denmark) (D + I), and were killed 8 h after the last insulin dose. Groups of normal rats were fed ad libitum (C), limited to 5060% of that consumed by the C group (FR), or deprived of food for 4 days (F). D, D + I and FR rats were used 18-20 days after the onset of treatments. A group of normal rats was surgically thyroidectomized (Tx) and received a Remington low iodine diet and water supplemented with 0.9% CaCl2. One week after thyroidectomy, rats were injected with 100 >nCi 131I, ip. Rats were killed by exanguination through the abdominal aorta
Received November 29,1990. Address all correspondence and requests for reprints to: Dr. T. Jolin, Instituto de Investigaciones Biomedicas, Consejo Superior de Investigaciones Cientificas, c/ Arturo Duperier 4, 28029-Madrid, Spain. * This research was supported by from Fondo de Investigaciones Sanitarias de la Seguridad Social Grant 88/1453 and Comision Interministerial de Ciencia y Tecnologia Grant PM88-0015. 361
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
362
LOW T3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
under light ether anesthesia and were perfused with 30 ml cold 0.9% NaCl containing 3 U heparin and 10~4 M propylthiouracil (18) through the carotid arteria. The cerebral cortex and cerebellum were dissected out, weighed, and either rapidly frozen in liquid nitrogen and stored at —70 C, or used immediately. The plasma was separated by centrifugation and stored at —20 C. Isolation of nuclei
For each nuclear preparation the cerebral cortex and cerebellum from three to five rats were combined. Total nonfractionated nuclei were obtained as previously described (4, 5). Briefly, the tissues were homogenized in 0.32 M sucrose-1 mM MgCl2 solution at a final dilution of 1:15 (wt/vol). The nuclei were sedimented by centrifugation and were repeatedly washed initially with and then without 0.5% Triton X-100. The crude nuclear pellet was resuspended in 5 vol 0.32 M sucrose-1 mM MgCl2) mixed with 25 vol 2.4 M sucrose-1 mM MgCl2, and
centrifuged for 45 min at 110,000 X g. DNA recovery was 4050% for cortex and 68-75% for cerebellum. The nuclear pellet was either used immediately or resuspended in 5 vol 2.2 M sucrose-1 mM MgCl2 for fractionation into neuronal and glial nuclei by the method of Thompson (19), as modified by Gullo et al. (20). Shortly, the nuclear suspension was placed in centrifuge tubes on top of a 2.4 M sucrose mattress, and this suspension was overlayered with 2.2 M sucrose-1 mM MgCl2 and centrifuged for 30 min at 110,000 x g. Discontinuous sucrose gradient centrifugation of the crude nuclear fraction yielded four layers. The layer between 2.2-2.4 M sucrose comprised neuronal fraction (21, 22). The pellet at the base of the 2.4 M sucrose layer consisted mostly of oligodendroglia (21). According to morphological criteria, the purity of neuronal and glial nuclei-rich fractions was 92 ± 6% and 95 ± 2%, respectively. When estimating the number of binding sites in glial and neuronal nuclei the apparent cross-contamination of the two layers was taken into account. Binding assay Nonfractionated nuclei and neuronal and glial nuclei were suspended in 0.32 M sucrose-3 mM MgCl2-20 mM Tris-HCl, pH 7.5 (binding buffer) containing 2 mM dithiothreitol. Replicative samples (100 /x\) of the suspended nuclei (25-40 /A DNA/tube) were added to 400 n\ binding buffer and were incubated with five or six different concentrations of [125I]T3 (0.05-1.6 nM; New England Nuclear, Boston, MA; 1200 mCi/mg). Incubations were for 45 min at 37 C in a shaking water bath and were stopped by placing the tubes on ice. After the addition of 200 ix\ 2% Triton X-100, nuclei were pelleted by centrifugation (4 C, 10 min at 10,000 X g), and washed twice with incubation buffer containing 1% Triton X-100. Bound T3 was determined in the washed nuclear pellet and in the incubation medium after trapping labeled T 3 by a Dowex 1 x 8 anion exchange resin (23). More than 93% of the total hormone in the supernate was in the free form. Nonspecific binding, determined in the presence of stable T 3 (10~7 M), was substracted in all cases. Maximum binding capacity (MBC) and association constant (Ka) of the binding of T 3 to nuclei were determined by Scatchard analysis (24). The time course of in vitro association of T 3 to isolated
Endo«1991 Vol 129 • No I
nuclei, the rate of both T 3 dissociation and loss of receptor sites during in vitro incubation, and the endogenous T3 bound to receptors at the end of incubation period were determined as described previously (25, 26) and used to assess the actual maximal binding capacity, as outlined by Kolodny et al. (25). Nuclear T3 content Nuclear T 3 and T4 concentrations were determined by an isotopic equilibrium technique as described previously in detail (13, 27). Briefly, rats were injected daily ip with a 125I + 127I mixture (6 /iCi + 5 /ig/100 g body wt). Thirty days after initiating the treatment, the animals were divided into the above-mentioned groups. The injections were continued for the next 20 days. For the determination of plasma and nuclearlabeled iodothyronines, the samples were extracted twice in ethanol 10"4 M propylthiouracil (18). The extracts were dried, counted, and redissolved in 100 /A methanol-ammonia (9:1, vol/ vol) containing carrier T4 and T3 and subsequently submitted to TLC in 25% ammonia-methanol-chloroform (3:20:40, vol/ vol) according to the procedure previously described (28). The amounts of T4 and T3 were calculated from the specific activity of 125I. The specific activity was determined from the 125I content in aliquots of thyroid homogenates and their 127I, as determined chemically (29). In some cases, the T3 concentration in isolated nuclei as determined by the isotopic equilibrium technique was compared with values obtained by RIA (30), previously described in detail (31). T3 and T4 contents in samples from Tx rats were determined by RIA. [125I]T3 and [125I]T4 from New England Nuclear and antisera from Henning (West Berlin, West Germany) were used in the RIA techniques. Protein and DNA concentrations were determined by the methods of Lowry et al. (32) and Burton (33), respectively. Plasma glucose was measured by the glucose oxidase method (34). Statistical analysis Results have been expressed as means ± SD. Statistical comparisons were performed by analysis of variance followed by Tukey's test. The number of receptors and affinity constants in Scatchard plots were calculated by linear regression analysis.
Results The general characteristics of the experimental groups are shown in Fig. 1. Tx, F, FR, and D rats showed the expected decrease in weight and circulating levels of T3. A decrease in T4 was also observed in Tx, F, and D animals, whereas normal values were found in FR rats. D rats were hyperglycemic, whereas glucose levels in Tx, F, and FR rats were lower than those in C animals. D rats treated with insulin had similar weight, plasma glucose, and T4 and T 3 levels as C rats. In preliminary time-course experiments we found that T 3 binding to cerebrocortical and cerebellar nuclei of C, Tx, F, and D rats reached a plateau by 10 min, remaining stable up to 60 min (Fig. 2A, continuous lines). Rats from different groups had essentially the same kinetic of [125I]
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T3 RECEPTOR 600
363
Cerebral cortex
100 80 60
Cerebellum
40
6 300
20
J8
V •a
100
I
80
g
40
o to
•M-
60
20
6 0.6
100 80 60
EJFR
•c
D+J
FIG. 1. Final body wt, and plasma glucose, T4 and T 3 concentrations in C, Tx, F, FR, D, and D+I rats. F animals were deprived of food for 4 days. FR rats were limited to 50-60% of the food consumption for the C group. FR, D, and D+I rats were killed 18-20 days after the onset of treatment. D+I rats received twice daily sc injections of 4 IU/100 g body wt for the last 15-16 days and were killed 12 h after last injection. Tx animals were killed 30 days after surgery. Data are means ± SD of 8-10 rats per group. Asterisks indicates P < 0.05 or less vs. C group.
T 3 binding to both preparations of nuclei. However, for each group the percentage of bound T 3 in cerebellar nuclei was lower than that for cerebrocortical nuclei. In the presence of stable T 3 (10~7 M), the [125I]T3 dissociated from both preparations of nuclei in a time-dependent fashion (Fig. 2A, dashed lines). The disappearance rate of [12fiI]T3 from in vitro labeled cerebrocortical and cerebellar nuclei was described by an apparent constant rate (means ± SD for rats from four groups) of 0.71 ± 0.13 h"1 and 0.64 ± 0.10 h"1, respectively. The [125I]T3 disappearance rate from both preparations of nuclei of isotopically equilibrated rats (Fig. 2B) was identical in the four groups and similar to that of in vitro labeled nuclei. The results in Fig. 2C, indicating a time-dependent reduction in receptor concentration during incubation, suggest a fractional loss of receptor (means ± SD for rats from all groups) of 0.22 ± 0.05 h"1 and 0.11 ± 0.04 h"1 for nuclei from cerebral cortex and cerebellum, respectively. Identical binding pattern for the rates of T 3 association/ dissociation and loss of receptor sites were obtained for FR and D + I groups (data not shown). From these results, a 45-min incubation period at 37 C was subsequently chosen for all the experiments. The observed disappearance rates for endogenous nuclear T 3 and the loss of receptor sites were then applied as correction factors in actual MBC calculation. Scatchard analysis of saturation studies with cerebrocortical and cerebellar nuclei from C, Tx, F, and D groups are shown in Fig. 3. Figures 4 and 5 show the estimated total nuclear receptor capacity, nuclear T 3 content, and
oTx .F
20
20
40
60 0
40
20
60
Minutes
FIG. 2. A, Time course of in vitro association/dissociation of [125I]T3 added in vitro with isolated cerebrocortical and cerebellar nuclei. Nuclei were incubated with [125I]T3 (1 x 10"10 M) in binding buffer at 37 C. At 20 min (arrow) the incubation mixture was divided into two parts, to one of which was added stable T3 (final concentration 8 X 10"7 M, dashed lines) and to the other the solvent alone (continuous lines). B, Dissociation rate of endogenously bound [125I]T3 at 37 C. Isolated nuclei from both brain areas of isotopically equilibrated rats were incubated with either stable T3 (final concentration 8 x 10"7 M) or the solvent alone (not illustrated). C, In vitro MBC (fmol T3/mg DNA) of cerebrocortical and cerebellar nuclear T3 receptors determined after 45 mih incubation at 37 C, after preincubation at 37 C for 20, 40, and 60 min. Each experimental point is the mean of triplicate determinations. Individual experimental values were within 10-15% of each other. Cerebellum
cortex
Cerebral 8
1.2
6
0.9
V
z Q
? ree.
• Tx
4
u_ \
Bound
DC
Si
40
2
n
0.6
\
\
\
•c
X
oTx 0.3 *F
\
\
.
500
\
.
1000
*• \
0*
50
\ 100
150
Bound (fmol T 3 /mg DNA)
FIG. 3. Scatchard analysis of T 3 binding to isolated cerebrocortical and cerebellar nuclei from C, Tx, F, and D rats. Saturation analysis was performed at 37 C for 45 min. Each experimental point is the mean of triplicate determinations. Individual experimental values were within 10% of each other.
receptor saturation values for all experimental groups. In each group, both binding parameters were much higher in cerebrocortical than in cerebellar nuclei. Thy-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
364 1200
600
60
Cerebral cortex 60
O E 03 N.
600
300
T
1
Endo • 1991 Voll29-Nol
Cerebellum 16
30 30
D C0
• Tx mF m FR m0 • 0*1
• c
FlG. 4. T3 binding parameters of cerebrocortical nuclei in C, Tx, F, FR, D, and D+I rats. The receptor concentration values are corrected for receptor losses and nondissociated endogenous T3 during incubation at 37 C for 45 min. Data are means ± SD of four to six nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for nuclear T3 content per group. Asterisks indicate P < 0.05 or less vs. C group. For more details see the legend to Fig. 1.
m
N
DTX
BF
E3FR B D
Bo.I
FIG. 6. Concentrations of T4 in nuclei from cerebral cortex and cerebellum. Data are mean ± SD of four to six isotopically equilibrated rats for C, F, FR, D, and D+I rats. T4 content in nuclei of Tx animals was lower (nd) than limit of sensitivity of T4 RIA. Asterisks indicate P < 0.05 or less us. C group. For more details see legend to Fig. 1.
200
60
100
30
Cerebral cortex
Cerebellum
too 80 60 40 20 (00 80 60 40
E
Deo
20
D Tx 20
40
60 0
20
40
60
Minutes
D*J FIG. 5. T3 binding parameters of cerebellar nuclei in C, Tx, F, FR, D, and D+I rats. The receptor concentration values are corrected for receptor losses and nondissociated endogenous T3 during incubation at 37 C for 45 min. Data are the means ± SD of four to six nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for nuclear T3 content per group. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
roidectomy and diabetes resulted in a decrease in nuclear T 3 content, but a reduction in the MBC was only observed in D animals, in both brain areas. As expected from the marked reduction in the nuclear T 3 content, the receptor occupancy in Tx and D rats was lower than in C animals. The Ka did not significantly differ among the experimental groups, in either the cerebral cortex or the cerebellum. Insulin treatment completely reversed the diabetes-induced decrease in T 3 receptor and nuclear T 3 content. In addition, the T4 concentration in nuclei from cerebral cortex and cerebellum in D rats were lower than the corresponding values in C animals (Fig. 6). In Tx rats the T4 values in nuclei from both brain areas were too low for detection by RIA. To determine whether the decrease in the number of
FlG. 7. A, Time course of in uitro association/dissociation of [125I]T3 added in vitro with isolated neuronal and glial nuclei from cerebral cortex and cerebellum of C and D rats. Nuclei were incubated with [125I]T3 (1 x 10"10 M) in binding buffer at 37 C. At 20 min (arrow) the incubation mixture was divided into two parts, to one of which was added stable T3 (final concentration 9 x 10"7 M, dashed lines) and to the other the solvent alone (continuous lines). B, Dissociation rate of endogenously bound [125I]T3 at 37 C. Isolated neuronal and glial nuclei from both brain areas of isotopically equilibrated rats were incubated with stable T3 (final concentration 9 X 10"7 M) or the solvent alone (not illustrated).
T 3 receptors observed in D rats was a consequence of reduced receptor levels in neuronal and/or glial nuclei, studies with fractionated nuclei from cerebral cortex and cerebellum were performed. Neuronal and glial nuclei from both brain areas of C and D rats bound [125I]T3 in a time-dependent fashion; an apparent equilibrium was found after 10-60 min at 37 C (Fig. 7). The loss of endogenously bound [125I]T3 over the 60 min from the different preparations of nuclei was approximately 45% from neuronal nuclei and 40% for glial nuclei in either cerebral cortex and cerebellum (Fig. 7, A and B). These findings indicate that the binding of [125I]T3 to neuronal and glial nuclei was highly specific, rapid, and reversible, fulfilling the principal criteria for a true hormone-receptor interaction.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
365
LOW T3 SYNDROME AND NEURONAL AND GLIAL T., RECEPTOR Scatchard plots of the binding data for both nuclear preparations from each tissue are shown in Fig. 8. Figures 9 and 10 show the estimated binding parameters in both C and D rats. In each group, MBC and nuclear T 3 values were higher in neuronal and glial nuclei from cerebral cortex than that of corresponding values in each nuclear preparation from cerebellum. Diabetes produced a deCerebrol cortex
*
\
1200
150
300
Glial nuclei \.O
(No
250
50
1000
60
o E 500
30
fli •C • 0
FIG. 10. T 3 binding parameters in neuronal and glial nuclei from cerebellum in C and D rats. Receptor concentration values are corrected for nondissociated endogenous T 3 and dissociated T3-receptor complex during incubation at 37 C for 45 min. The MBC values were also corrected for cross-contamination of the two nuclear preparations. Data are mean ± SD of four to five nuclear preparations for MBC, Ka, and DNA, and isotopically equilibrated rats for endogenous T 3 content per group. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
Discussion
FIG. 8. Scatchard analysis of T 3 binding to isolated neuronal and glial nuclei from cerebral cortex and cerebellum in C and D rats. Saturation analysis was performed at 37 C for 45 min. Each point is the mean of triplicate determinations. Individual experimental values were within 10-14% of each other.
S "3 E
2
100
Bound (fmol T 3 /mg DNA)
»
40
crease in nuclear T 3 content in the four nuclear preparations, but a reduction in the MBC was only seen in glial nuclei from both brain areas. The Kn values of binding in neuronal and glial nuclei from cerebral cortex and cerebellum were not different between C and D rats.
0.5 v
CQ v 1000
Q
s i1 too
£
\
2000
80
Dc
\
125
_
200
nuclei
3
600
20Q
Cerebellum
Neuronol
\
400
I •
FlG. 9. T 3 binding parameters in neuronal and glial nuclei from cerebral cortex in C and D rats. Receptor concentration values are corrected for nondissociated endogenous T 3 and dissociated T3-receptor complex during incubation at 37 C for 45 min. The MBC values were also corrected for cross-contamination of the two nuclear preparations. Data are means ± SD of four to five nuclear preparations for MBC, Ka, and DNA and isotopically equilibrated rats for nuclear T 3 content. Asterisks indicate P < 0.05 or less us. C group. For more details see the legend to Fig. 1.
The present study indicates that diabetes in rats results in a decrease in T 3 receptors in cerebral cortex and cerebellum. In contrast, T3-binding sites in both brain areas of Tx, F, and FR animals did not significantly change in concentration or affinity. Both thyroidectomy and diabetes were accompanied by a decrease in nuclear T 3 content and a marked reduction in receptor occupancy in both cerebral cortex and cerebellum, whereas fasting or food restriction produced no changes in T 3 concentration in any brain areas studied. Preservation of the T 3 content in both brain areas of F and FR rats might be related to the finding that in the hypothyroid rat cerebral cortex the efficiency of T4 to T 3 conversion is increased, and in both cerebral cortex and cerebellum the fractional removal rate of T 3 is slow (35, 36). Thus, the metabolic conditions resulting from a diabetes period of 20 days are probably not comparable to those resulting from 4 days of complete food withdrawal or 20 days of food restriction. The results from these studies also suggest that specific T3-binding in cerebral cortex and cerebellum may be capable of down-regulation in response to a metabolic change associated with diabetes and low circulating T 3 levels. Besides, the observation that these effects of diabetes can be prevented by the concomitant administration of insulin suggests that these receptors
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
366
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T 3 RECEPTOR
are conceivably regulated by insulin. However, these preliminary experiments did not reveal whether the changes in T 3 receptor binding were directly or indirectly related to the action of insulin. The MBC we estimated for cerebral cortex and cerebellum nuclei from C rats agrees with that reported by others using in vitro techniques (5, 20, 25, 37), but it is higher (3.2-fold for cortex and 1.4-2.1-fold for cerebellum) than those determined by in vivo saturation techniques (4, 38, 39). Similar discrepancies have been noted between data obtained in vivo and in vitro in euthyroid and hypothyroid rat pituitaries (40). The T 3 content found in nuclei from both brain areas in this study is similar to that previously reported (38, 39). The differences between in vivo and in vitro measurements have been attributed to the existence of nuclear T3-binding sites not available for in vivo binding (38) or to the possibility that the methodological assumption inherent in the in vivo techniques (41) results in the determination of only a fraction of the total nuclear receptors (40). Irrespective of the causes for these differences, the large discrepancies between the MBC determined in vivo and in vitro and the similarity of nuclear T 3 content found in both situations may account for the discrepancies between the endogenous cerebral cortex and cerebellum receptor occupancy estimated in vivo and in vitro. A saturable T 3 binding was found in both neuronal and glial nuclei from cerebral cortex and cerebellum of C and D rats, and all had a similar Ka, very close to that for total nonfractionated cerebrocortical and cerebellar nuclei. Moreover, T 3 binding sites appear to be more numerous in neuronal than in glial nuclei in both brain areas. This reflects the findings of Gullo et al. (20, 37), who reported a lower number of T 3 binding sites in glial than in neuronal nuclei, at least in cerebral cortex. Recent reports suggest an absence of specific T 3 receptors in glial nuclei from cerebral cortex (25) and cerebellum (37) from normal rats. A possible explanation is that an incomplete dissociation of endogenous T3, dissociation of T3-receptor complex, or receptor instability during the incubation period may have resulted in serious misinterpretation of the in vitro binding data in glial nuclei. However, Yusta et al. (42) have clearly demonstrated the presence of nuclear high affinity T 3 receptors in pure secondary cultures of oligodendrocytes. Our findings show that diabetes is accompanied by a reduction of T 3 content, T3-binding sites, and receptor occupancy in glial nuclei from cerebral cortex and cerebellum. In neuronal nuclei, a different situation is seen. Thus, although nuclear T 3 levels are also decreased, the T 3 receptor sites in neuronal nuclei from both brain areas did not significantly change in concentration or affinity under the same conditions. These findings suggest that diabetes can produce receptor changes that are localized
Endo-1991 Voll29-Nol
in specific cells, leaving others unaffected. This represents, to our knowledge, the first observation that glial T3-binding can be altered in vivo. The mechanism for the decrease in glial nuclei T 3 receptors in D rats could involve degradation or inactivation rate, decreased biosynthesis rate, or both. However, our results suggest that the decreased binding in the glial nuclei of D rats is not due to down-regulation by the presence of elevated circulating T 3 levels. The diabetes-induced decrease in specific T 3 binding sites in glial nuclei is a descriptive finding, and neither the causes nor the consequences of these changes are clarified by these experiments. The decrease in binding sites may lead to alterations in the expression of certain T 3 effects. Glial cells, oligodendrocytes, are involved in myelin synthesis, and T 3 is known to influence the synthesis and activity of the enzymes
associated with myelinogenesis (43, 44). Therefore, it is possible that diabetes-mediated decrease in T 3 receptors may result in a situation where myelin synthesis is depressed. Moreover, it is conceivable that metabolic factors associated with diabetes may play a role. Regarding the last possibility, previous results have clearly demonstrated that insulin stimulates glucose uptake in glial cells from rat brain, and the effect is mediated by insulin-specific receptors (45). The importance of these findings is that a major component of the central nervous system, the glial cell, not only has T3-binding characteristics similar to those of neuronal cell, but the number of T 3 receptors is decreased in the diabetic state.
Acknowledgments The authors thank Margarita Gonzalez for excellent technical assistance. The provision of Streptozotocin by Upjohn Research Laboratories (Kalamazoo, MI) is gratefully acknowledged.
References 1. Oppenheimer JH 1979 Thyroid hormone action at the cellular level. Science 203:971-979 2. Oppenheimer JH 1983 The nuclear receptor-triiodothyronine complex: relationship to thyroid hormone distribution, metabolism and biological action. In: Oppenheimer JH, Samuels HH (eds) Molecular Basis of Thyroid Hormone Action. Academic Press, New York, pl-35 3. Eberhardt N, Valcana T, Timiras PS 1978 Triiodothyronine nuclear receptors: an in vitro comparison of the binding of triiodothyronine to nuclei of adult rat liver, cerebral hemisphere, and anterior pituitary. Endocrinology 102:556-561 4. Oppenheimer JH, Schwartz HL, Surks MI 1974 Tissue differences in the concentration of triiodothyronine nuclear binding sites in the rat liver, kidney, pituitary, heart, brain, spleen, and testis. Endocrinology 95:897-903 5. Schwartz HL, Oppenheimer JH 1978 Nuclear triiodothyronine receptor sites in brain: probable identity with hepatic receptors and regional distribution. Endocrinology 103:267-273 6. Dozin B, De Nayer P 1979 Nuclear triiodothyronine receptors in rat brain during maturation. Brain Res 177:551-554 7. Ishiguro K, Suzuki Y, Sato T 1980 Effect of neonatal hypothyroidism on maturation of nuclear triiodothyronine (T3) receptors in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
LOW T 3 SYNDROME AND NEURONAL AND GLIAL T3 RECEPTOR developing rat brain. Acta Endocrinol (Copenh) 95:495-504 8. Schwartz HL, Oppenheimer JH 1978 Ontogenesis of 3,5,3'-triiodothyronine receptors in neonatal rat brain: dissociation between receptor concentration and stimulation of oxygen consumption by 3,5,3'-triiodothyronine. Endocrinology 103:943-948 9. Valcana T, Timiras PS 1978 Nuclear triiodothyronine receptors in the developing rat brain. Mol Cell Endocrinol 11:31-41 10. Silva JE, Larsen PR 1978 Contributions of plasma triiodothyronine and local thyroxine monodeiodination to triiodothyronine and nuclear triiodothyronine receptor saturation in pituitary, liver and kidney of hypothyroid rats. Further evidence relating saturation of pituitary nuclear triiodothyronine receptors and acute inhibition of thyroid-stimulating hormone release. J Clin Invest 61:1247-1259 11. Spindler BJ, MacLeod KM, Ring J, Baxter JD 1975 Thyroid hormone receptors: binding characteristics and lack of hormonal dependency for nuclear localization. J Biol Chem 250:4113-4119 12. Burman KD, Lukes Y, Wright FD, Wartofsky L 1977 Reduction in hepatic triiodothyronine binding capacity by fasting. Endocrinology 101:1331-1334 13. Jolin T 1987 Diabetes decreases liver and kidney nuclear 3,5,3'triiodothyronine receptors in rats. Endocrinology 120:2144-2151 14. Lim VS, Passo C, Murata Y, Ferrari E, Nakamura H, Refetoff S 1984 Reduced triiodothyronine in liver but not pituitary of the uremic rat model: demonstration of changes compatible with thyroid hormone deficiency in liver only. Endocrinology 114:280-289 15. Tibaldi JM, Surks MI 1985 Animal models of nonthyroidal diseases. Endocr Rev 6:87-101 16. Thrall C 1983 Effects of various pathological conditions on the nuclear T3 receptors of rat cerebral cortex: comparison with liver. Brain Res 279:177-183 17. Ortiz-Caro J, Obregon MJ, Pascual A, Jolin T 1985 Decreased T4 to T3 conversion in tissues of streptozotocin-diabetic rats. Acta Endocrinol (Copenh) 106:86-91 18. Morreale de Escobar G, Llorente P, Jolin T, Escobar del Rey F 1963 The "transient instability" of thyroxine and its biochemical applications. Biochem J 88:526-530 19. Thompson RJ 1973 Studies of RNA synthesis in two populations of nuclei from the mammalian cerebral cortex. J Neurochem 21:1940 20. Gullo D, Sinha AK, Woods R, Pervin K, Ekins RP 1987 Triiodothyronine binding in adult rat brain: compartmentation of receptor populations in purified neuronal and glial nuclei. Endocrinology 120:325-331 21. L0vtrup-Rein H, McEwen BS 1966 Isolation and fractionation of rat brain nuclei. J Cell Biol 30:405-414 22. Rose SPR, Sinha AK 1969 Some properties of isolated neuronal cell fractions. J Neurochem 16:1319-1328 23. Torresani J, DeGroot LJ 1975 Triiodothyronine binding to liver nuclear solubilized proteins in vitro. Endocrinology 96:1201-1209 24. Scatchard G 1949 The attraction of proteins for small molecules and ions. Ann NY Acad Sci 51:660-672 25. Kolodny JM, Larsen PR, Silva JE 1985 In vitro 3,5,3'-triiodothyronine binding to rat cerebrocortical neuronal and glial nuclei suggests the presence of binding sites unavailable in vivo. Endocrinology 116:2019-2028 26. Jolin T 1988 Response of hepatic mitochondrial a-glycerophosphate dehydrogenase and malic enzyme to 3,5,3'-triiodothyronine in streptozotocin-diabetic rats. Endocrinology 123:248-257 27. Surks MI, Oppenheimer JH 1977 Concentration of L-thyroxine and L-triiodothyronine specifically bound to nuclear receptors in rat liver and kidney. Quantitative evidence favoring a major role
367
of T3 in thyroid hormone action. J Clin Invest 60:555-562 28. Heider JG, Bronk JR 1965 A rapid separation of thyroxine and some analogues by thin-layer chromatography. Biochim Biophys Acta 96:353-355 29. Benotti J, Benotti N 1963 Protein bound iodine, total iodine and butanol extractable iodine by partial automation. Clin Chem 9:408410 30. Weeke J, 0rskov H 1975 Ultrasensitive radioimmunoassay for direct determination of free triiodothyronine concentration in serum. J Clin Invest 35:237-244 31. Weeke J, 0rskov H 1978 Evaluation of thyroid function. In: Alberti KGM (ed) Recent Advance in Clinical Biochemistry. Churchill Levingstone, London, vol 1:111-128 32. Lowry OH, Rosebrough NJ, Farr AL, Randle RJ 1951 Protein measurements with Folin phenol reagent. J Biol Chem 193:265269 33. Burton K 1956 A study of conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem J 62:315-323 34. Hugget ASG, Nixon DA 1957 Use of glucose-oxidase, peroxidase and o-dianisidine in determination of blood and urine glucose. Lancet 2:368-370 35. Silva JE, Larsen PR 1982 Comparison of iodothyronine 5'-deiodinase and other thyroid-hormone dependent enzyme activities in the cerebral cortex of hypothyroid neonatal rat. Evidence for adaptation to hypothyroidism. J Clin Invest 70:1110-1123 36. Silva JE, Matthews PS 1984 Production rates and turnover of triiodothyronine in rat-developing cerebral cortex and cerebellum. Response to hypothyroidism. J Clin Invest 74:1035-1049 37. Gullo D, Sinha AK, Bashir A, Hubank M, Ekins RP 1987 Differences in nuclear triiodothyronine binding in rat brain cells suggest phylogenetic specialization of neuronal fractions. Endocrinology 120:2398-2403 38. Crantz FR, Silva JE, Larsen PR 1982 An analysis of the sources and quantity of 3,5,3'-triiodothyronine specifically bound to nuclear receptors in rat cerebral cortex and cerebellum. Endocrinology 110:367-375 39. Oppenheimer JH, Schwartz HL 1985 Stereospecific transport of triiodothyronine from plasma to cytosol and from cytosol to nucleus in rat liver, kidney, brain, and heart. J Clin Invest 75:147154 40. St Germain DL, Galton VA 1985 Comparative study of pituitarythyroid hormone economy in fasting and hypothyroid rats. J Clin Invest 75:679-688 41. Pearson JD, Veall N, Vetter H 1958 A practical method for plasma albumin turnover studies. Strahlentherapie 38:290-297 42. Yusta B, Besnard F, Ortiz-Caro J, Pascual A, Aranda A, Sarlieva L 1988 Evidence for the presence of nuclear 3,5,3'-triiodothyronine receptors in secondary cultures of pure rat oligodendrocytes. Endocrinology 122:2278-2284 43. Sarlieve L, Bouchon R, Koehl C, Neskovic MM 1983 Cerebroside and sulfatide biosynthesis in the brain of Snell Dwarf mouse: effects of thyroxine and growth hormone in the early postnatal period. J Neurochem 40:1058-1062 44. Bhat NR, Sarlieve L, Subba Rao G, Pieringer RA 1979 Investigations on myelination in vitro. Regulation by thyroid hormone in cultures of dissociated brain cells from embryonic mice. J Biol Chem 245:9342-9344 45. Clarke DW, Boyd Jr FT, Kappy MS, Ralzada MK 1984 Insulin binds to specific receptors and stimulates 2-deoxy-D-glucose uptake in cultures glial cells from rat brain. J Biol Chem 259:11672-11675
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 02:40 For personal use only. No other uses without permission. . All rights reserved.
