Search

Type 2 deiodinase (D2) and type 3 deiodinase (D3) locally achieve the determination of the concentration of T3, which binds to the thyroid hormone receptor with high affinity. D2 converts T4 into T3, and D3 degrades T4 and T3. Neurons take up T3 released by astrocytes, the main cerebral site for the D2 expression. Because oxidative stress is believed to be involved in several neurological disorders, we explored the effects of oxidative stress on D3 and D2 in primary culture of rat astrocytes. H2O2 (250 microm) increased D3 activity with maximal effects around 8 h. Stimulation of D3 activity by H2O2 was synergistic with T4, phorbol ester, and also cAMP. H2O2 (250 microm) did not affect basal D2 activity but inhibited the stimulation of D2 activity by cAMP and factors implicating cAMP-independent pathways in astrocytes, TSH, and phorbol ester. N-Acetyl cysteine and selenium repletion, which respectively increase intracellular glutathione and glutathione peroxidase, inhibited D2 and D3 regulation by H2O2, whereas L-buthionine sulfoximine, which decreases intracellular glutathione, mimicked H2O2 effects. Oxidative stress up-regulated D3 and inhibited cAMP-stimulated D2 by transcriptional mechanisms. A decrease in cAMP by oxidative stress could contribute to the inhibition of cAMP-stimulated D2. Using specific inhibitors of signaling pathways, we show that the ERK pathway was required in D2 and D3 regulation by oxidative stress and that the p38 MAPK pathway was implicated in H2O2-induced D3. We suggest that the expected decrease in T3 might modulate the cellular injury of oxidative stress in some pathological brain conditions.

T(4) activation into T(3) is catalyzed by type 2 deiodinase (D2) in the brain. The rapid induction of D2 in astrocytes by transient brain ischemia has prompted us to explore the effects of hypoxia on D2 in cultures of astrocytes. Hypoxia (2.5% O(2)) of cultured astrocytes increased D2 activity, alone or in association with agents stimulating the cAMP pathway. Hypoxia had no effect on D2 mRNA accumulation. Cycloheximide did not block the effect of hypoxia on D2 activity and D2 half-life was enhanced under hypoxia demonstrating a posttranslational action of hypoxia. Furthermore, the D2 activity increase by hypoxia was not additive with the increase promoted by the proteasome inhibitor carbobenzoxy-L-leucyl-L-leucyl-L-leucinal (MG132). This strongly suggests that hypoxia leads to stabilization of D2 by slowing its degradation by the proteasome pathway. Hypoxia, in contrast to MG132, did not block the T(4)-induced D2 inactivation. A contribution of prolyl hydroxylase to the hypoxia effects on D2 was also suggested on the basis of increased D2 activity after addition of different prolyl hydroxylase inhibitors (cobalt chloride, desferrioxamine, dimethyloxalylglycine, dimethylsuccinate). Specific inhibitors of ERK, p38 MAPK, or phosphatidylinositol 3-kinase pathways were without any effect on hypoxia-increased D2 activity, eliminating their role in the effects of hypoxia. Interestingly, diphenyleneiodonium, an inhibitor of nicotinamide adenine dinucleotide phosphate oxidase inhibited the hypoxia-increased D2 indicating a role for some reactive oxygen species in the mechanism of D2 increase. Further studies are required to clarify the precise molecular mechanisms involved in the D2 stabilization by hypoxia.

Thyroid hormones are essential for the development and repair of the peripheral nervous system. The type 2 deiodinase, which is responsible for the activation of T(4) into T(3), is induced in injured sciatic nerve. To obtain information on the type 3 deiodinase (D3) responsible for the degradation of thyroid hormones, we looked for its expression (mRNA and activity) in the sciatic nerve after injury. D3 was undetectable in the intact sciatic nerve of adult rats, but was rapidly and highly increased in the distal and proximal segments after nerve lesion. After cryolesion, D3 up-regulation disappeared after 3 d in the proximal segment, whereas it was sustained for 10 d in the distal segment, then declined to reach basal levels after 28 d, when functional recovery was completed. After a transsection preventing the nerve regeneration, up-regulation of D3 persisted up to 28 d at high levels in the distal segment. D3 was expressed in peripheral connective sheaths and in the internal endoneural compartment. D3 mRNA was inducible by 12-O-tetradecanoylphorbol-13-acetate in cultured fibroblasts or Schwann cells. In conclusion, induction of D3 in the peripheral nervous system after injury may play an important role during the regeneration process by adjusting intracellular T(3) levels.

The type 3 iodothyronine deiodinase (D3) metabolizes thyroid hormones to inactive metabolites in many tissues, including the brain. In the present studies, we have examined the mechanisms by which T3 (T3), retinoic acid, 12-O-tetradecanoyl phorbol 13-acetate (TPA), and basic fibroblast growth factor (bFGF) induce D3 expression in primary cultures of neonatal rat astrocytes. In untreated cells, D3 messenger RNA (mRNA) was essentially undetectable by Northern analysis and RT-PCR. However, all four agents induced expression of a 2.4-kb D3 transcript as well as D3 activity. Induction of D3 by TPA and bFGF was more rapid than that by T3 and retinoic acid, and T3 potentiated the stimulatory effects of TPA and bFGF. D3 induction by TPA was blocked by GF 109203X, an inhibitor of protein kinase C. In addition, the effects of TPA and bFGF were partially prevented by PD 98059, a specific inhibitor of MEK and the Erk signaling cascade. These studies demonstrate that multiple growth factors and hormones regulate D3 activity in cultured astrocytes by inducing D3 mRNA expression. In addition, the stimulatory effects of TPA and bFGF on D3 mRNA and activity appear to be mediated at least in part by activation of the MEK/Erk signaling cascade.

The effect of an activator of protein kinase C, 12-O-tetradecanoylphorbol-13-acetate (TPA), on the 5'-deiodinase (5'D) activity was studied in rat astroglial cells cultured in chemically defined medium. TPA promoted a large increase in the type II 5'D activity, which was maximal 5-10 h after addition of TPA and then declined to the basal level at 24 h. The optimal TPA concentration was 10(-7) M. TPA and 8-Br-cAMP, an other inducer of 5'D activity were antagonist. Otherwise, TPA stimulated 5'D activity only in the presence of glucocorticoids at concentrations from 10(-8) M to 10(-4) M. Glucocorticoids alone induced a slight increase in 5'D activity. These data indicate that protein kinase C contributes to the control of 5'D activity in astroglial cells and that its action is dependent on glucocorticoids.

Adult rats who had neonatal hyperthyroidism (NH) have reduced BW and serum T4, T3, and TSH concentrations. Pituitary TSH responses to TRH administration under basal, T4-suppressed, and propylthiouracil-stimulated conditions suggest that the thyrotroph of these animals is more sensitive to the feedback effects of thyroid hormones. These studies were undertaken to examine, with the use of various thyroid hormone-responsive variables, thyroid status of adult NH rats. NH was induced by 12 daily sc injections of T4 (0.4 microgram/g BW) to neonatal male Sprague-Dawley rats. Adult NH and control rats were then studied at 120 days of age. Adult NH rats had significantly decreased mean BW (P less than 0.001), and serum T4 (P less than 0.005), T3 (P less than 0.001), and TSH (P less than 0.001) concentrations. The percent decreases were 12% for T4, 20% for T3, and 27% for TSH in adult NH rats. Mean pituitary GH concentration and hepatic alpha-glycerophosphate dehydrogenase and malic enzyme activities were significantly decreased in adult NH rats to 54% (P less than 0.005), 52% (P less than 0.025), and 39% (P less than 0.001), respectively, of control values. Mean pituitary TSH concentrations were similar in adult NH and control rats. Mean hepatic T4 5'-deiodinase activity of adult NH rats [200 +/- (SE) 23 fmol T3/min X mg protein] was significantly decreased to 56% of control levels (355 +/- 31 fmol T3/min . mg protein; P less than 0.005). Mean pituitary T4 5'-deiodinase activity of adult NH rats (32.8 +/- 3.3 fmol T3/min . mg protein) was significantly increased compared with that of control rats (21.2 +/- 1.7 fmol T3/min . mg protein; P less than 0.025). These changes are consistent with a hypothyroid state in adult NH rats. The observation of decreased serum TSH concentration and enhanced thyrotroph sensitivity to thyroid hormones in the face of increased pituitary T4 5'-deiodinase activity suggests that increased thyrotroph monodeiodination of T4 may be the central biochemical aberration responsible for the hypothyroid state in adult NH rats.

Cultured astroglial cells were found to contain a type II 5'-deiodinase (5'D) activity which was increased by 10(-3) M (Bu)2cAMP but not by 2 X 10(-3) M n-butyrate. 8-Bromo-cAMP (8-Br-cAMP) (10(-3) M) also increased this enzyme activity. Cycloheximide (2 micrograms/ml) inhibited the 8-Br-cAMP effect on 5'D activity. Forskolin (10(-5) M), cholera toxin (5 micrograms/ml), 10(-5) M isoproterenol, and 3 X 10(-6) M norephinephrine also increased the 5'D activity of astroglial cells. After a 4-h incubation these agents or cAMP analogs had maximal effect, and enzyme activities were 6- to 14-fold above control value. The stimulatory effects of isoproterenol and norepinephrine were almost completely reversed after 8 h incubation. The induction of 5'D activity by isoproterenol or norepinephrine was inhibited by the beta-adrenergic antagonist alprenolol (5 X 10(-6) M). The effect of norepinephrine was not significantly affected by the alpha 1-adrenergic antagonist, prazosin (10(-5) M). Thus, 5'D activity is controlled by agents increasing cAMP in astroglial cells, and in particular by the neurotransmitter, norephinephrine, via a beta-adrenergic mechanism.

Bovine anterior pituitary glands were fractionated by differential centrifugation. T4 5'-monodeiodination to T3 was found predominantly in microsomal fractions (M2; 105,000 . g pellet) enriched in glucose-6-phosphatase and 5'-nucleotidase activities. T4 5'-deiodinase activity in M2 fraction was 85.2 fmol T3/min X mg protein and represented an 8.5-fold enrichment over homogenate specific activity (10.6 fmol T3/min . mg protein). Further subcellular localization of the T4 5'-deiodinase was effected by discontinuous sucrose density gradient centrifugation. Maximum T4 5'-deiodinase activity was found in fraction P5 at the interface of densities 1.18/1.20 (200 fmol T3/min . mg protein) and correlated with the profile of glucose-6-phosphatase and not with that of 5'-nucleotidase, the maximum activity of which was recovered in fraction P1 at the interface of densities 1.03/1.12. Electron microscopic examination of the fractions confirmed that P5 contained in excess of 90% rough membranes in contrast to 10% or less in P1. Characterization of T4 5'-deiodinase activity was carried out in M2 preparations. The reaction was thiol dependent, requiring the presence of 50 mM dithiothreitol or more (Km, 38 mM), with a maximum velocity of 55-150 fmol T3/min . mg protein (n = 8). Enzyme activity was substrate dependent, with a Km for T4 between 35-70 nM. 5'-Monodeiodination of T4 was abolished by heating to 70 C for 30 min and was unaffected by EDTA. Propylthiouracil and methimazole did not inhibit T3 generation. Iopanoic acid, on the other hand, was a competitive inhibitor of the 5'-monodeiodination reaction, abolishing T3 production in a dose-dependent manner with a Ki of 3 microM. These data indicate that the bovine anterior pituitary contains significant T4 5'-deiodinase activity, which shares many properties of the type II 5'-deiodinase of the rat. Bovine anterior pituitary T4 5'-deiodinase appears to be predominantly localized in the rough endoplasmic reticulum.